COMMON FUND FOR COMMODITIES Technical Paper No. 5 Banana Improvement Project COMMON FUND FOR COMMODITIES Technical Paper No. 5 Banana Improvement Project Common Fund for Commodities Stadhouderskade 55, 1072 AB Amsterdam, The Netherlands Postal Address: P.O. Box 74656, 1070 BR Amsterdam, The Netherlands Tel: (31 20) 5754949 Telex 12331 cfc nl Fax: (31 20) 6760231 e-mail:Managing.Director@common-fund.org Copyright - Common Fund for Commodities 2001 The contents of this report may not be reproduced, stored in a data retrieval system or transmitted in any form or by any means without prior written permission of the Common Fund for Commodities; except that reasonable extracts may be made for the purpose of comment or review provided the Common Fund for Commodities is acknowledged as the source. Table of Contents Foreword…………………………………... ………………………………………………1 Acknowledgements………………..……………………………………………………...3 Executive Summary……………………………………………………………………….5 Part I: Banana Improvement Project Completion Report Introduction .............................................................................................................. 15 Project Structure ...................................................................................................... 15 Research Portfolio .................................................................................................... 18 Project Management ................................................................................................ 40 Public/Private Sector Collaboration ......................................................................... .42 Impact Assessment .................................................................................................. 42 Conclusion ............................................................................................................... 44 Part II: Project Evaluation and Detailed Research Findings Introduction .............................................................................................................. 47 Background .............................................................................................................. 49 Review of the Project ............................................................................................... 54 Field and Laboratory Evaluation of Diploid Bananas for their Use in Breeding Schemes (BIP 01) ..................................................................... 54 Field Crosses for Understanding the Inheritance of Black Leaf Streak Resistance to Bananas (BIP 02) ........................................................ 60 Collection, Characterisation and Evaluation of Nendran Banana in India (BIP 03) ............................................................................................... 65 Collection, Evaluation and Characterisation of Genetic Resources and Improved Banana Crop Plants in Vietnam (BIP 04) ............................... 70 Banana Breeding in Brazil (BIP 05) ............................................................... 76 Banana Biotechnology Consortium: Novel Genes for Fungal Resistance and Post-Harvest Quality (BIP 08) .............................................. 85 i Development of Transgenic Bananas with Resistance to Banana Bunchy Top and Banana Mosaic Viruses (BIP 09) ........................................ 92 Genetic Transformation of Prototype Bananas for Black Sigatoka Resistance (BIP 10) ....................................................................................... 99 Genetic Engineering of Ethylene Biosynthesis in Bananas (BIP 11) ........... 104 Use of Biotechnology to Produce Transgenic Bananas Resistant to Banana Bunchy Top Virus Infection (BIP 12) .............................................. 108 Elimination of Banana Streak Badnavirus (BSN) from Improved Musa Germplasm and related Studies on Transmission of Host Plant/Virus/Vector Interactions (BIP 14) .................................................................................... 112 Variability and Relationships within Populations of Fusarium Oxysporum FSP Cubense for its Centre of Origin (BIP 15) ............................................ 117 Origin and Distribution of Fungicide-Resistant Strains of Mycosphaerella Fijensis in Banana Plantations in Costa Rica (BIP 17) ...... 122 Study of Resistance of Banana and Plantains to Nematodes (BIP 18) ....... 128 Identification of Durable Nematode Resistance Sources in Banana and Plantain (BIP 19) ......................................................................................... 132 Identification of Durable Nematode Resistance Sources in Banana and Plantain (BIP 20) ......................................................................................... 136 Identification of Durable Nematode Resistance Sources in Banana and Plantain (BIP 21) ......................................................................................... 141 Breeding Hybrid Musaceae with Resistance to Multiple Diseases, Especially Black Sigatoka and Panama Disease (BIP 22) .......................... 143 Conclusions and Recommendations ...................................................................... 146 Annex 1: Original Terms of Reference of the BIP .................................................. 153 Annex 2: List of Acronyms ...................................................................................... 154 Part III: Impact Assessment of the Banana Improvement Project Executive Summary, Impact Assessment of Banana Improvement Project prepared by P.D. Chudleigh for the World Bank .................................................... 159 Introduction ............................................................................................................ 164 Description of Banana Improvement Programme .................................................. 164 ii Summary of Principal Outputs of the Programme .................................................. 169 Summary of Actual and Potential Outcomes and Benefits ..................................... 172 Investment Analysis ............................................................................................... 179 Conclusion ............................................................................................................. 188 References ............................................................................................................. 189 Annex 1: Terms of Reference ................................................................................ 190 Annex 2: Summary and Impacts of Individual Projects .......................................... 191 iii iv FOREWORD Bananas (Musa spp.) are cultivated in over 100 countries in the tropical and subtropical regions of the world, where they constitute a major staple food crop for over 120 million people as well as providing a valued source of income through local and international trade. Musa is the developing world’s fourth most important global food crop after rice, wheat and maize in terms of gross value of production. The vast majority of producers are small-scale farmers growing the crop either for home consumption or for local markets. Average productivity of bananas in small-scale production of 8 tons per hectare per year is very low, compared to large-scale plantations of 40 tons per hectare per year. The low productivity is mainly due to the adverse effects of pests and diseases, which are most severe for small-scale farmers, who do not have the economic or the technical capacity to manage such problems. Chemical control methods employed by large plantations are beyond the means of the majority of smallholder banana farmers and, moreover, the intensive use of pesticides is likely to have a detrimental effect on the environment and human health. The Banana Improvement Project (BIP) was conceived to address this critical problem of the banana industry and to develop improved varieties for commercial as well as for staple food production. Eighteen research institutions were involved in the BIP funded research and development work carried out from January 1994 to December 1998. The BIP was among the most important projects financed by the Common Fund for Commodities (CFC), both in terms of financial exposure as well as in terms of the scope and geographical spread of the work. It was also a unique project which was based on a collaborative effort of the CFC as financier, the World Bank as the Project Executing Agency (PEA) and the Food and Agriculture Organisation (FAO) as the initiator and supervisor of the project. The BIP was launched with the intention of “…channelling of funds to precisely targeted research objectives…” with a view to attaining new knowledge with strategic goals and new profitable products. The specific objectives that the project aspired to achieve were: (i) to develop and evaluate a range of improved banana varieties with export potential, incorporating increased productivity with disease resistance; and (ii) to develop more efficient and integrated disease management practices for Black Sigatoka disease, based on an improved understanding of disease epidemiology and pathogen variability in different areas and over different seasons. While the attainment of these objectives proved to be too ambitious, the BIP has produced many scientific advances and achieved excellent return on investment. The results obtained from the project activities can be summarised as follows: (1) New knowledge has been produced in the following areas: • Collection and characterisation of banana germplasm that will allow increased effectiveness of the banana breeding programs around the world. 1 • Additional information gathered through use of genetic markers, particularly for disease resistance to Sigatoka, Panama disease, and nematodes, will be valuable in future work. • Improved and refined technology for genetic manipulation, including genetic transformation. Use of this knowledge will enhance banana and plantain production whether for cooking or for dessert banana exports. (2) Applied outputs with immediate value to banana producers: • Black Sigatoka-resistant clones further developed are being used to help control an outbreak of the disease in the Amazonas region of Brazil. • Strains of the Fusarium Wilt pathogen in a given location can now be typed using new molecular and genetic techniques. This will be particularly beneficial to new banana enterprises in determining where and where not to plant in relation to damaging races of panama disease. Although the real value of the BIP can only be assessed over a longer period in terms of the strategic outputs of the project, the above outputs are significant. Using the strategic outputs, it will not be too long before diverse banana varieties which are disease-resistant and have desirable properties for commercial production are produced. Furthermore, present and future successes will have positive environmental and health benefits. Among other benefits that resulted from the project is the identification of research projects through an internationally competitive research grants scheme carried out by a Scientific Advisory Panel on the basis of merit-based criteria. Furthermore, BIP fostered an informal network of the research work being conducted under the BIP, which will hopefully be sustained after the termination of the project. It is our strong belief that the project produced and the applied outputs, beneficiaries of the BIP, namely the countries around the globe. The CFC objective. results, both the strategic new knowledge will generate benefits for the intended smallholder producers in more than 100 will continue with its efforts to realise this Rolf W. Boehnke Managing Director Common Fund for Commodities 2 ACKNOWLEDGEMENTS The three parts of this report have been prepared by the World Bank, with whose kind permission this report is being presented, and by consultants employed by the World Bank. The work performed, which is reflected in this report, has been largely financed by the Common Fund for Commodities. The Common Fund for Commodities greatly values the important international collaboration effort of this project, firstly with the World Bank and the Food and Agriculture Organisation, secondly with the principal investigators and their collaborators for their valuable contributions to the success of the project. The Common Fund for Commodities also pays tribute to the Scientific Advisory Panel (SAP) and the consultants who evaluated the project; namely: - Dr. Edmond de Langhe - University of Leuven – Belgium - Professor Günther Kahl – Johann Wolfgang Goethe University – Germany - Dr. Franklin Rosales – the International Network for the Improvement of Bananas and Plantains (INIBAP); and - Dr. P. D. Chudleigh – Agtrans Research, Queensland, Australia for his Impact Assessment of the project under the sponsorship of the World Bank 3 4 EXECUTIVE SUMMARY 1. Introduction “Careful channelling of funds to precisely targeted research objectives with the right programs will lead to new, useful knowledge and to new profitable products. New innovative breeding approaches combined with advances in plant biotechnology offer really new dimensions to banana improvement”. The Banana Improvement Project (BIP) has in large measure met the challenge above, quoted from the 1994 background paper on the status of global research on banana and plantain. This paper, which was commissioned by the World Bank, provided the backdrop for the 5 year BIP program that brought together some of the best research talent world-wide to work on bananas. BIP operated throughout 1994 to 1998 and this project completion report documents the results of the 18 individual research projects, plus the details of an independent economic impact assessment that reported a return on investment of 20-33% The uniquely conceived, and innovatively run, program has achieved its main goals. The structure of the project, its management, and the competitive grant style of selecting and funding individual research projects, are described in full detail in this report. 2. Background Bananas and plantain are grown mainly by smallholder farmers in developing countries. Bananas are of major importance to food security in the tropics, and their production provides a vital source of income through local and international trade. Bananas and plantains are the fourth most important global food crop and export commodity. They rank first as a fruit. Only in very recent years has research been able to improve banana breeding techniques, thus overcoming some of the barriers to genetic improvement of this crop. Producers are still facing serious pest and disease problems, which are controlled in large measure by the use of chemical sprays. Resource-poor farmers often cannot afford expensive pesticides, so they suffer damaging losses. The use of pesticides is a major health concern in developing countries, so the development of disease- and pest-resistant varieties is viewed as urgent to protect a vital industry and the health of banana producers. A significant contribution to this global effort has been through BIP, an international 5-year research program that used some innovative scientific and management approaches, notably a competitive grants structure, to harness the talents of outstanding researchers. BIP started in January 1994, and contracted research concluded in December 1998, with some of the individual research projects continuing their activities into 1999 .for completion of their work. The Project is co-sponsored by the Common Fund for Commodities (CFC), the FAO Inter-Governmental Group on Bananas (FAO/IGB), and the World Bank. The CFC committed funding of SDR 2,500,000 (approximately US$3.5m) to the project portfolio and management of the program. The World Bank, as the Project Executing 5 Agency (PEA), has prepared this completion report covering the management and research activities resulting from this investment. BIP had two main objectives: (1) (2) To develop and evaluate improved banana varieties with export potential, which would incorporate increased productivity and durable disease resistance through conventional and non-conventional breeding techniques. To develop more efficient and integrated disease management practices, especially for black Sigatoka disease. BIP funded 18 research projects in four technical areas: (1) Biotechnology, (2) Germplasm collection and characterisation, (3) Plant improvement, and (4) Pest and disease management. The intended beneficiaries were producers of dessert bananas, particularly smallholders. The potential spillovers to producers of plantain for subsistence were increasingly recognised during the course of the Project. 3. Project Structure Management: The Cosponsors and their technical representatives constituted the governing body of the Project; with the FAO/IGB as the Supervisory Body in accordance with CFC normal practices. In order to meet their obligations to the CFC, FAO convened a Technical Advisory Committee (TAC) to assess project progress. The World Bank acted as the Project Executing Agency, handling the daily management of the project, and convened a Scientific Advisory Panel (SAP) to assist in the planning, selection, and monitoring of research activities. The Bank reported to both the CFC and the FAO/IGB on project activities by way of twice-yearly Progress and Annual Reports. Annual Meetings of the Cosponsors were convened to discuss project progress, and met at the same time and venue as the Annual Scientific Meetings. Research: The eighteen projects were chosen through a merit-based process of internationally competitive research grants. Proposals were reviewed by the SAP against a set of merit-based criteria, and were selected to form a cohesive and complementary portfolio to work in parallel toward the project objectives. Projects were monitored by the SAP on a six-month basis through a review of Progress and Annual Reports. SAP recommendations were submitted to the PEA for acceptance, and Principal Investigators were notified of the review’s comments and outcome. Three Annual Scientific Meetings permitted Principal Investigators to present their work to the Cosponsors, their technical panels and scientific peers. 6 Reporting: FAO also produced reports of the Technical Advisory Committee and these were provided separately to the CFC. A BIP Newsletter was also circulated to report on progress to interested parties. 4. Research Highlights Results of research activities are set out below against the two BIP objectives, followed by results against technically related activities. 4.1 Results against BIP Objectives: a. To develop and evaluate improved banana varieties with export potential, which would incorporate increased productivity and durable disease resistance through conventional and non-conventional breeding techniques. i. ii. Refined and utilised two methods of genetic transformation. Genetically engineered expression of resistance to two viruses, BBTV and BBMV, into Cavendish and other varieties. Collected and preserved valuable germplasm, and biotypes of nematodes and pathogens. Reversed serious decline in conventional breeding capacities in two regions. Contributed to the safe international transportation of important genetic stocks. Supplied breeding programs with useful diploid lines as parental stocks. iii. iv. v. vi. b. To develop more efficient and integrated disease management practices, especially for black Sigatoka disease. i. Expanded options for pest management choices by: • Identifying important components of durable host plant resistance; and, • Adding understanding to the consequences of pesticide use. ii. iii. iv. Identified sources of resistance in germplasm collections. Mapped genetic resistance to the black Sigatoka pathogen. Completed a global map of geographical distribution of Fusarium races for risk assessment. Developed an early screening protocol for nematode-resistant varieties. v. 4.2 Results against BIP technically-related activities. AREA I: Biotechnology 7 BIP succeeded in its strategy to develop a molecular toolbox. Numerous methods for reliably transforming Musa are now a reality. By working in a network of partners, scientists supported by BIP have exchanged transformation protocols, and verified results. We now know much more about the molecular manipulation of Musa including: • • • • Receptivity of different types of plant materials; Dependability of DNA introduction methods; Gene expression; Time necessary to transform Musa. As a consequence of transformation technology, BIP-sponsored research has produced banana clones with resistance to BBTV and BBMV that are undergoing glasshouse evaluation and are ready for field testing. AREA 2: Germplasm Collection and Evaluation Valuable genetic stocks of wild and domesticated Musa species have been placed in accessible collections in India and Vietnam through the sponsorship of BIP. The future value of these preserved stocks will only be fully realised as scientists evaluate their genetic attributes and combine traits using conventional and molecular tools. Had these stocks not been preserved they would likely have been lost to future generations. AREA 3: Plant Improvement The projected demise of two critically important banana breeding programs in the Americas was reversed by the timely awarding of BIP funding. These grants permitted plant breeders to continue to generate improved progeny and to evaluate genetic selections for industry and subsistence farmers alike. The practical implication of this outcome is the critical need for conventional genetics programs to help find the solutions to production and protection problems faced by banana producers. AREA 4: Pest and Disease Management BIP’s dual strategy of applying genetic transformation and DNA selection marker methods for adding host plant resistance to banana clones for black Sigatoka resistance made significant progress. Transformation technologies (see Area 1) are now available to introduce single gene resistance should suitable candidate genes be found. Secondly, marker assistance selection technologies for multigenic resistance for black Sigatoka appear promising. The value of this strategy is in the expectation that multigenic forms of resistance should be more durable. Strategies for managing black Sigatoka pest resistance in M. fijiensis proved unreliable in those spray-free periods failed to let the pathogen rebound to sensitive forms. This research outcome removes pesticides as a sustainable solution to black Sigatoka and increases the need for nonpesticide-based IPM programs. Resistance to nematodes is a high priority for bananas in as much as chemical controls are costly and hazardous. The sources of resistance identified, as well as 8 the screening methods developed through BIP-sponsored research are a significant success. 5. Impact Assessment The World Bank commissioned an economic impact study of BIP in late 1998. The key comments and findings below have been extracted from that report. 1. New knowledge has been produced in the following areas: • • • • 2. Collection and characterisation of banana germplasm that will allow increased effectiveness of the banana breeding programs around the world. Increased efficiency of breeding programs through improved testing, screening and typing methods, and development of protocols. Additional information gathered through use of genetic markers, particularly for disease resistance to Sigatoka, Panama Disease and nematodes, will also be valuable in future work. Improved and refined technology for genetic manipulation, including transformation processes for banana and improved knowledge of disease resistance mechanisms in banana. Use of this knowledge will enhance banana and plantain production, whether for cooking, domestic consumption as dessert bananas, or for export. Applying this knowledge in future to provide real improvements for producers and exporters will require further R&D investment. Applied outputs of some of the projects will have immediate value to and impact on industry. The following stand out: • • Black Sigatoka-resistant clones further developed in BIP projects are being used to help control an outbreak in the Amazonas region of Brazil. Strains of Fusarium in a given location can now be typed using BIPgenerated techniques. This will be particularly beneficial to new banana enterprises in determining where and where not they can be planted in relation to damaging races of Panama disease. Although these outcomes are significant, the real value of BIP-supported work will be the long-term strategic knowledge flowing from the BIP projects. Landmark events using some of this new knowledge might be the development of a Cavendish or alternative export banana type that is disease resistant. Other developments that might result include socio-economic, environmental, and health benefits. The economic benefits would include reduced production costs (few chemicals required); the environment would be improved through less chemical usage; and banana producers would suffer fewer health problems by not having to handle as much toxic chemical material (organ damage to an estimated 10% of workers according to Swennen, 1997). Improvements would lead to better health of workers, higher productivity, arid an overall increase in the quality of life of communities in bananaproducing areas. The BIP-supported research has been complemented by other significant investments in R&D. 9 3. Leverage of Funds BIP also leveraged substantial other investments in banana improvement: • • • • 4. Building on the original CFC investment, further funds have been attracted to BIP, both within the projects and outside: i.e. WOB and BDC, Belgium for the nematode consortium personnel positioned around the world in various breeding programs. Funding from the New Zealand Government for post-harvest practices in Vietnam. Funding from the University of Hong Kong and the Governors Fund in Hawaii. Conservative estimate of funds contributed is US$ 2 million. Increased Collaboration A longer-term benefit to industry will be the enhanced scientific co-operation in banana improvement work as a result of the creation of a Biotechnology Consortium (Katholeike Universiteit Leuven, Belgium, Boyce Thompson Institute, USA, and Queensland University of Technology, Australia). This grouping resulted in staff exchanges and the transfer and refinement of particle bombardment and Agrobacterium transformation and regeneration technology. In a similar way BIP fostered the creation of a "nematode consortium”, which has resulted in improved screening techniques leading to a better understanding of nematode resistance. • • • • • • • • Formation of consortium in biotechnology enabling the refinement of both particle bombardment and Agrobacterium transformation techniques. Formation of nematode consortium strengthened during the project with funding from VVOB and BDC that will continue past the completion of BIP. Collaboration between Australian and Vietnamese scientists, exemplified in the training of Vietnamese scientists by institutes in Australia. Collaboration between Australian and CIRAD scientists in screening for Fusarium Wilt on CIRD breeding diploids. Collaboration between Costa Rica and Brazil in screening breeding materials. Attraction of new research providers into banana improvement. Additional permanent staff position created in CIRAD focused on banana improvement. Continued generic resources provided by Belgium through nematode consortium. The impact assessment study indicated that BIP should probably provide a rate of return on investment of at least 20% to 33%. Conservative assumptions were used in the analyses, and additional analyses showed robustness of the results. This demonstrates that further investments in banana improvement research will be well justified. 10 6. Conclusion BIP results have provided methods for transformation and regeneration of bananas, and have moved world-wide banana producers closer to a disease and pestresistant transgenic banana being a reality. Considerable progress has been made in identifying new germplasm suitable for breeding, and ways have been developed to improve the breeding BIP has, therefore, made an important contribution in both the provisions of research results, and catalysing collaborative activities and co-financing through the innovative process of utilising a competitive research grants scheme. This enabled the identification of research providers of excellent quality, and the investment of resources in a portfolio of research projects of direct relevance to the objectives of the program. The World Bank’s overarching commitment is to the alleviation of poverty. The contribution this Project has made toward the development of resistant varieties, and new control methods of banana and plantain production, is a positive step in reaching this goal. 11 12 PART I Banana Improvement Project Completion Report 13 14 1. Introduction The Banana Improvement Project (BIP) commenced on 1 January 1994, with contracted research activities reaching completion on the 31st December 1998. The Project is cosponsored by the Common Fund for Commodities (CFC), the FAO InterGovernmental Group on Bananas (FAO/IGB), and the World Bank (IBRD). The CFC committed funding of SDR 2,500,000 (approximately US$3.5 million) toward the research portfolio and management of the program. This report, prepared by the World Bank, documents the project management structure, research activities under a competitive grants program, and results achieved from this investment. Objectives BIP was conceived and specially structured to make a significant contribution toward the improvement and productivity of banana, by using higher-yielding, diseaseresistant varieties, and seeking ways to reduce the cost of production, especially the cost of pesticide application. A two-pronged approach was taken, based on commissioning research concerned with: (1) Genetic improvement, by collecting, developing, and evaluating new banana varieties with export potential, incorporating higher yields and durable disease resistance. (2) Improved and integrated pest management practices to reduce pesticide use. The intended beneficiaries of the Project were producers of dessert bananas, particularly smallholders, and subsistence farmers growing plantain and banana for their own or local consumption. Other targeted beneficiaries were plantation workers, because of the focus on improved pest management practices, and the public at large, by decreasing environmental degradation through a reduction in pesticide use and establishment of plantations in soils and regions that are not optimal to support banana production. Consumers world-wide will benefit through access to a selection of banana varieties grown under more environmentally neutral conditions. 2. Project Structure 2.1 Management: The Cosponsors and their technical representatives constituted the governing body of the Project, with the FAO/IGB as the Supervisory Body in accordance with CFC normal practices. The FAO convened a Technical Advisory Committee (TAC) to assess overall Project progress. The World Bank acted as the Project Executing Agency, handling the daily management of the Project, and convened a Scientific Advisory Panel (SAP) to assist in the planning, selection, and monitoring of the research activities. The Bank provided twice-yearly progress and annual reports, to the CFC and the FAO/IGB on Project activities. Annual Meetings of the Cosponsors were held to discuss progress at the time of the Annual Scientific Meeting. 15 2.2 Project Planning and Implementation: 2.2.1 Planning: In the early 1980s, FAO and its Intergovernmental Group on Bananas identified the need for greater investment in banana research in response to the increasing cost of production and falling returns to growers. More frequent fungicide applications were required because the major disease, black Sigatoka, continued to spread into banana- and plantain-growing areas. At the same time, fungicide-resistant strains were appearing. Consumers were becoming more concerned about environmental degradation resulting from the increased spraying. Nematicides in particular were being more widely used in an effort to control this major nematode pest, and new strains of Fusarium Wilt were emerging that were completely destroying new plantations in the tropics and subtropics. New virus diseases were also being recognised, and banana streak virus in particular was threatening the viability of the international exchange of Musa germplasm to breeding programs. At the same time, markets were expanding, and presenting an opportunity to producing countries to increase their export trade. This was of particular importance to the smaller exporting countries, particularly the ACP countries, which traded banana as their most significant export commodity. The World Bank supports banana and plantain research through its contributions to the Consultative Group on International Agricultural Research (CGIAR). The Bank had undertaken a review of international support to banana and plantain research by participating in a CGIAR Task Force whose recommendations were complementary to the plans of FAO. The FAO/IGB’s overriding interest in Musa was focused on the banana export crop, but they also shared the interest of the World Bank in addressing the problems of growers of banana and plantains for subsistence, and the importance of both domestic and export production for poverty alleviation. The banana and plantain export crop accounts for approximately 10% of total production, the remainder being a subsistence crop grown widely in Africa and the Asia/Pacific region. The two organisations therefore collaborated in designing and presenting a project proposal to the 12th Banana Improvement Project, Completion Report, December 1998 CFC, with the resultant Project Agreement being signed during International Centres Week of the CGIAR in Washington, D.C., in October 1993. 2.2.2 Implementation: The BIP Secretariat was located in the Bank within the Agricultural Research and Extension Group (ESDAR). A Technical Manager was appointed with responsibility for Project implementation, supported by a Project Co-ordinator. Initial activities of the Bank in the first half of 1994 were the preparation and dissemination of a call for proposals, and the identification of members of a Scientific Advisory Panel (SAP). A review of current banana and plantain improvement programs was also commissioned by the World Bank and a 16 report was prepared on the activities of the current banana improvement programs, and opportunities and innovative ideas identified that could be useful to exploit within the new Project. This was of particular importance due to the lack of success in breeding a disease-resistant commercial dessert banana over the past 70 years. The Bank consulted with representatives of the private sector and the donor community during the initiation of BIP. Consultative workshops were held immediately preceding meetings of the Scientific Advisory Panel (SAP) in June and December 1994. The June meeting in Washington, D.C., confirmed the relevance of the aims of BIP, with black Sigatoka being the major constraint, and the use of pesticides, involving increasing costs, environmental degradation, and worker health and safety concerns. Discussion also focused on the newly available techniques of biotechnology, with industry indicating that a genetically engineered Cavendish would be of great interest to them. The use of biotechnology was the focus at the second meeting held in Houston, Texas, where the issue of the likelihood of success in using transformation techniques with subsequent regeneration was discussed. 2.3 Research Strategy: As a result of these activities, the research strategies considered best able to address the objectives of BIP were agreed as follows: i. New banana varieties: • • • ii. Biotechnology-based breeding to genetically modify presently cultivated clones (especially Cavendish types) for a few key characteristics such as black Sigatoka resistance, resistance to banana bunchy top virus (BBTV), and delayed ripening to extend shelf life. Conventional breeding based on the crossing of improved diploid to generate new hybrids. Novel breeding schemes based on triploid resynthesis, to develop new banana types with export potential, suitable for niche markets. Better disease control practices: • • Identification of the variation in the major pathogens causing black Sigatoka disease and Fusarium Wilt, to underpin conventional and non-conventional breeding. Epidemiology and ecology of black Sigatoka disease, as a basis for better management of the disease aimed at reducing the frequency of fungicidal applications, while maintaining effective control. The Project used the innovative mechanism of an internationally competitive research grants program to identify research projects. Two calls for proposals resulted in over 100 submissions from leading research institutes world-wide. Scientists from national, regional, and international research institutes were invited to submit proposals, and combined proposals were particularly 17 encouraged. Proposals were received from institutes in both banana importing and exporting countries, and the public and private sector. 3. Research Portfolio Proposals were reviewed by the SAP against a set of merit-based criteria, and were selected in order to form a cohesive and complementary portfolio to work in parallel toward Project objectives. A brief outline of individual project results is given below, organised in technically related areas of activity. A number of institutes will be continuing BIP experiments into 1999 due to the need for materials to mature prior to analysis. Field testing of transgenic plants produced during BIP is currently under final negotiation to ensure that regulatory procedures are addressed. Further results will therefore become available in mid-1999. AREA I: Biotechnology PN8: Biotechnology Consortium: Novel genes for fungal resistance and post-harvest quality. Boyce Thompson Institute for Plant Research, USA Activity: The researchers sought to create a “molecular toolbox” in collaboration with PN9 and PNl0. The ‘toolbox’ is for use in isolating and manipulating new genetic elements from Musa SPP. The toolbox will include Grand Nain genomic libraries, cDNA libraries representing mRNAs present in a variety of banana tissues at different developmental stages or under pathogen attack, and a variety of Agrobacterium strains for Musa transformation. Work at BTI was related to genes associated with fruit ripening and fungal resistance. Background: Development of the toolbox will allow the creation of new banana germplasm, with resistance to virus and fungal diseases that now limit production and increase production costs. It will also provide a better fundamental understanding of the mechanisms controlling qualitative and quantitative changes in gene expression that occur during banana fruit development or under pathogen attack. Results: • • • • • Identification and characterisation of eleven distinct classes of differentially expressed cDNAs from the pulp of ripening Grand Nain banana fruit. Efforts continue toward the isolation of regulatory elements governing the expression of mRNAs. Transformation cassettes incorporating these regulatory elements were built. Identification of previously uncharacterised mRNAs from banana roots and leaves. Development of a transient assay system based on protoplasts derived from readily available banana pulp. cDNA libraries were generated from leaves of Gros Michel, Malaccensis, and a Cavendish variety that were challenged by Mycosphaerella fijiensis. 18 • Rapidly doubling Grand Nain embryonic cell suspensions have been developed. Outcome: This project was completed in February 1997. Since then further work has been carried out in the various areas above. Cloned genes have been distributed to other laboratories both within the consortium and to the wider research community. Further refinement of the Agrobacterium-mediated transformation system has been undertaken by identification of alternative sources of starting materials, the aim being to reduce the number of chimeric plants obtained. Several banana-derived promoter elements are being isolated and characterised that will allow for the tissue- or developmental-specific expression of foreign genes in transgenic banana plants. The project also gained a better understanding of the molecular events that occur during banana fruit development and ripening, and those events that are associated with pathogen challenge in fungal-resistant and non-resistant banana varieties. The ‘molecular toolbox’ has provided many laboratories with molecular biology materials. Impact: The major impact of this project has been the additional knowledge generated on ethylene control, and the further development of the Agrobacterium method of transformation. The project contributed to strengthening the collaborative activities of the BTI group as a member of the biotechnology consortium, with visiting scientists, and through sharing of technologies. PN9: Biotechnology Consortium: The development of transgenic bananas with resistance to banana bunchy top virus (BBTV) and banana bract mosaic virus (BBMV). Queensland University of Technology (QUT), Australia Activity: In collaboration with other members of the Biotechnology Consortium (KUL, Leuven; BTI, USA); and with the University of Hawaii, (USA), the scientists attempted to control banana bunchy top virus (BBTV) and banana bract mosaic viruses (BBMV) through the development of multivirus resistant transgenic bananas. The team utilised current and emerging knowledge of BBTV resistance, and characterisation of the BBMV genome, to develop transformation cassettes. Transformation proceeded initially on single resistance transgenes, and ultimately to multiple resistance transgenes. Transformed material was screened for resistance, and promising transgenic lines prepared for field trials. Background: BBTV is largely controlled in Australia by removal of infected plants, a costly undertaking. Cavendish is particularly susceptible to the disease, and can reduce yields by up to l00%, but averages at about 30%. BBMV creates losses of between 20% and 40%. A range of cassettes has been generated that potentially confer resistance to BBTV and BBMV in transgenic bananas, and have been used in the transformation and regeneration of Bluggoe and Cavendish embryos. Three different BBTV, and one BBMV, resistance cassette were used, resulting in regenerated plantlets that are ready for virus challenge in the field. A range of promoters for transgene expression continues to be analysed for use in generating virus resistance. 19 Results: • • • • • • Identification of potentially three different BBTV resistance genes, and one BBMV gene. Cavendish and Goldflnger plants have been regenerated from embryonic cell suspensions, using the particle bombardment transformation method. Cavendish “Grand Nain” embryos have been transformed and regenerated with GUS, GFP, two different BBTV resistance cassettes, and one BBMV resistance cassette. Additional embryonic cell suspensions have been generated, and new virus resistance constructs have been designed that are expected to be more efficient in generating resistance. Planning for initial greenhouse evaluation and field trials is progressing to ensure that biosafety regulations are addressed, and to minimise the possibility of loss of material. Three unexpected problems arose during the project that have delayed the realisation of final results: Cavendish transformation and regeneration is slower and less efficient than Bluggoe; the erratic rate of multiplication of transgenic lines for virus challenge, and identification of a suitable screening site for BBMV proved more difficult than expected, but this has now been overcome. Outcome: The technology developed under the project is directly applicable to the generation of transgenic resistance to other diseases. The development of an efficient transformation and regeneration protocol for Cavendish, and identification of promoters to drive high level transgene expression, can be utilised together to express novel genes in banana. It is expected that of the current transgenic lines produced, 50% will test resistant. This will increase as the technique is further refined. Impact: Technology to generate potentially virus-resistant Musa varieties, including transformation, regeneration, and expression of new genes, has been demonstrated. Confirmation of resistance has yet to be tested in the field, and it is expected that resistance breakdown will be in the order of 5-10% within a period of 10 years of release of resistant types. Technologies developed are relevant not only to viral resistance, but also to other diseases of banana and plantain types. Access to BIP influenced the QUT approach, therefore playing a positive role in the successful results. PNlO: Biotechnology Consortium: Genetic transformation of prototype bananas for black Sigatoka and Fusarium resistance. Katholeike University, Leuven, Belgium Activity: In collaboration with PNS and PN9, the main focus of the project was the investigation and optimisation of a reliable and transferable transformation system. Agrobucterium-mediated transformation and particle bombardment techniques were tested for their suitability and efficiency for large-scale applications. Other project activities were the isolation of a strong banana promoter, high molecular weight DNA for large insert genomic libraries; the preparation of a leaf cDNA library; and the 20 construction of expression vectors with selectable marker genes for banana transformation. Background: The impetus for this project came with the significant progress that had been made in in-vitro manipulation of banana in the early 1990s. Highly regenerable embryonic cell suspensions (ECSs) were established at KUL from vegetative meristematic tissue. The material enabled the demonstration, for the first time, of banana transformation at the level of transient gene expression in electroporated protoplasts. At the same time, regenerable ECSs also showed a great potential for direct generation of transgenic banana plants by particle bombardment. The BIP project, therefore, built on this previous work to develop or adapt these new biotechnological tools for genetic manipulation of banana and plantain, and apply them to genetic improvement. Results: • • • • • • • Particle bombardment of embryonic cell suspensions proven to be the superior transformation method. Plant expression vectors containing chimeric selectable marker genes for the two best selection systems of banana cells were constructed. Several heterologous promoters driving high gene expression in banana were identified. New assays and bacterial test strains were developed to allow for the study of interaction between banana and Agobacterium tumefaciens. cDNA library was prepared from banana leaves for the isolation and characterisation of specifically expressed genes. Partial isolation and characterisation of an actin gene from a Grand Nain genomic library was carried out. A protocol was described for the isolation of high molecular weight DNA banana leaf nuclei, resulting in library-quality DNA for the preparation of large insert genomic libraries. Outcome: The particle bombardment method was used by collaborative projects within BIP, and made available for molecular breeding of banana in independent laboratories world-wide. Optimal conditions are now defined for the most critical steps of a technology for particle bombardment of embryonic banana and plantain cell suspensions. This has many practical applications, the paramount one being the production of material resistant to major pests and diseases. The Agro bacteriummediated transformation system needs significant improvement, although work on this activity has supplied some useful information. New transformation vectors have been designed that are useful for sequential transformations to introduce multiple genes, and also provide a readily available backbone to insert any potentially interesting selectable marker gene for testing in banana. These vectors enable more versatile and routine applications for molecular breeding of banana. Impact: BIP accelerated the ability to transform banana by refining the current systems, and widening the parameters of exploration. BIP assisted in knowledge generation of the techniques enabling other laboratories to utilise the technologies. This BIP project was the catalyst for the collaborative link between the participating laboratories, which extended to the University of Hawaii (PN12). Through this 21 project, and others in the biotechnology consortium, BIP has played a role in the acceptance of potential biotechnology solutions to the major problems of banana production by the private sector. PN11: Genetic engineering of ethylene biosynthesis in bananas. Hong Kong University of Science & Technology, Hong Kong Activity: The aim of this project was to clone genes involved in ethylene biosynthesis in banana, the genetic transformation of banana with reduced ethylene production capacity, and the evaluation of transformed plants in relation to reduced disease development and fruit quality. Due to time constraints on the project, only the first objective was addressed. Background: The plant hormone ethylene is produced by the banana plant and fruit, and plays an important role in disease susceptibility and fruit ripening. Ethylene production is induced by pathogen attacks resulting in accelerated leaf senescence and yellowing, and can also cause premature ripening of the fruit resulting in reduced storage life and poor quality. It has been established that 1aminosyclopropane-1-darboxylic acid (ACC) synthase and ACC oxidase genes participate in the regulation of and catalyse the last two steps in the pathway of ethylene biosynthesis. While cloned in a number of other species, cloning in banana had not been achieved. The recent advances in genetic engineering of higher plants presented the possibility of using biotechnology to genetically engineer crop plants with endogenous ethylene production in fruits resulting in prolonged storage life. Results: • • • • • Construction of a banana genomic library and identification of some ACC synthase (ACS) and ACC oxidase (ACO) clones. Characterisation of the clones by restriction enzyme digestion and partial DNA sequencing. Complete sequencing of one ACS and two ACO genomic DNA clones. Elucidation of the regulation of ACS and ACO gene expression by developmental and environmental factors by performing an RNA gel blot analysis. Delivery of an antisense ACS and ACO gene construct to KUL for investigation. Outcome: This project has generated substantial new knowledge on the issue of ethylene production, and highlights the potential that, with further investment, postharvest losses during transportation and storage can be significantly reduced. KUL have advised that genetic transformation was successfully performed on the gene construct, but a mite problem on the resultant material has delayed work. A rehabilitation program is now underway. Further support is required for this activity to build on the results in linking the gene structure to their functions in disease resistance and fruit ripening. Impact: The increased knowledge of the genetic information associated with ethylene biosynthesis will be useful in future research aimed at post-harvest qualities of fruit, as well as disease control. The study has laid the foundation for further 22 analysis of the fruit ripening-specific promoter, and provided the basis for genetic engineering of ethylene biosynthesis and metabolic pathways in banana. PN12: Use of biotechnology to produce transgenic bananas that are resistant to banana bunchy top virus (BBTV) infection. University of Hawaii, U.S.A. Activity: The objective of this project was to produce transgenic banana resistance to BBTV by introducing viral genes into the genome. The project utilised and modified both systems developed in the “molecular toolbox” of the consortium, particle bombardment, or “gene gun” technology, and the Agrobucterium methods. Although not a member of the Biotechnology Consortium, the UH group has worked closely with its members. Background: BBTV is the most devastating virus disease of bananas in many producing areas, including Asia, Africa, and South Pacific. Since 1989 the Hawaii Dept. of Agriculture has been attempting to eradicate the virus, an approach that worked well in Australia. This approach appears insufficient in Hawaii, indicating that additional control strategies and long-term solutions to the problem must be found. Development of resistant varieties based on breeding efforts has been complicated by ploidy level differences and sterility in the commercial clones. Recent developments in biotechnology offered a new avenue. Results: • • • • • Two banana transformation and regeneration systems were evaluated for development of transgenic banana plants. Several transformation experiments were conducted with various BBTV-gene constructs. Transformed Cavendish ‘Williams’ bananas containing the BBTV gene constructs were obtained, using a modified version of the Agrobucteriummediated transformation system developed by the BTI group. Transgenic banana plants were screened in the greenhouse, resulting in the identification of BBTV-resistant material. Evaluation in field trials is planned. However, the BIP-funded research is now completed, and funds are required to undertake field experiments. Outcome: This project has provided evidence that pathogen-derived resistance works for BBTV-like viruses, and, with further support, varieties can be developed for banana growers in Asia, Africa and the South Pacific, where BBTV is a major problem at production levels. The modified Agrobacterium system will also be useful for other researchers working on resistance to other diseases, including black Sigatoka. Sixty-five transgenic plants were evaluated in the greenhouse for BBTV resistance, and of these; three plants did not develop any BBTV-specific symptoms, and registered negative in ELISA tests. The project has also succeeded in refining a protocol to identify transgenic cells. Gene constructs from this project have been made available to KUL and CATIE in Costa Rica, and are available to other researchers on request. 23 Impact: Field experiments remain to be conducted to challenge the transformed plants against BBTV. Further work needs to be undertaken using BBTV virus groups from other areas where the disease occurs. Current results, however, are highly significant in that it has been demonstrated that viral-resistant plants can be produced, and that further refinement of the technique could translate into considerable reduction in the losses caused by the virus in Asia, Africa, and South Pacific countries. AREA 2: Germplasm Collection & Evaluation PN3: Developing improved banana varieties with pest and disease resistance, postharvest superiority and maximum export potential. Kerala Agricultural University, India Activity: The collection, characterisation, and evaluation of Nendran banana in India undertaken in close collaboration with the national banana program (NRCB), to establish reference data for Nendran for use by future researchers. Selected clones from the collection were evaluated for dwarfness, short cycling, and disease resistance, particularly for yellow Sigatoka, and to a lesser extent weevils and nematodes. Background: The project concentrated on collection and characterisation of the Nendran type AAB plantain banana and wild relatives. Nendran varieties are the most important type grown in the South Indian peninsula, and are exclusively cultivated on commercial lines, but to date have not been systematically evaluated for optimal breeding characteristics. Characterisation of the Nendran-type germplasm is based on morphological and isozyme markers. Results: • • • • • • • • 125 accessions available for characterisation. 111 Nendran, and three wild types were collected from eight different locations. Fourteen accessions from the germplasm bank at the Banana Research Station, Kannara, were included in the study. The collection was established in the field at the Banana Research Station, Kannara, and the passport data for each of the collections have been prepared. Morphological characterisation was carried out using the INIBAP descriptors. Differences have been observed in colour, height, and aspect of pseudostem, inflorescence, and bunch characters. Tentative identification of 14 types has been made, and a key developed. A database has been established containing information on the collection. Procedures for isozyme polymorphism studies have been standardised, with five enzyme systems surveyed in various accessions. Results indicated a variation in the banding pattern as well as intensity of individual bands. Evaluation was also undertaken using a combination of enzyme systems in an attempt to ensure classification into distinct groups. The reaction of accessions to yellow Sigatoka was undertaken and reported, with further trials underway. The reaction of accessions to nematode attack was assessed and reported. 24 Outcome: Variations in Nendran types collected are being documented on a reference database. This will enable identification of varieties with superior agronomic traits, such as tolerance to drought and water logging, which is of particular interest in Southern India. Accessions collected and established at Kannara will be available for future collaborative programs. Impact: Southern Indian breeding programs are now able to select material better suited to the various farming systems. The possibility of identifying a short statured variety has been increased. This will assist in lowering production costs since propping would not be required. Germplasm has been collected that could be useful for breeding programs world-wide. PN4: Collection, evaluation and characterisation of genetic resources and improved banana crop plants in Vietnam. Institute for Agricultural Genetics, Hanoi, Vietnam Activity: The major focus of this project was the collection of banana germplasm, and its evaluation for productivity and disease resistance to enable selection of varieties with high export quality. The project also focused on micropropagation techniques for elimination of virus diseases. Cultivars displaying desirable agronomic and fruit qualities are being selected for use in the Vietnamese breeding program. Background: Located in the centre of origin of banana, the Vietnamese crop is highly diversified with many low-yielding and disease-susceptible cultivars being grown, resulting in a low annual yield of approximately 15 tons/hectare and below standard quality of export fruit. Even so, banana is reported as being the most important fruit crop in Vietnam, contributing significantly to the income of millions of farm households. Studies have shown that genotypes in the collection are not represented in existing collections anywhere in the world. This presents a valuable source of parent material for breeding programs. The collection has been established in vitro. The project was strengthened through training visits by selected scientists to laboratories in Queensland, Australia, to learn techniques of identification of banana diseases, and banana production technology. A further positive activity was the convening of an international seminar in 1995 that attracted the participation of a wide range of Vietnamese officials. This culminated in the formation of an internal banana research network, with close contacts between AGI and the Fruit Research Centre in Phu Ho, Vinh Phu, for banana genetic resource studies. Results: • • • A large number of banana cultivars, clones, and wild species were collected, preserved, and characterised. Material displaying the most important agrobiological characters was released for production. Guidelines for field management based on individual agrobiological traits of each cultivar were recommended to farmers in an effort to achieve the highest economic returns. Improved banana micropropagation technologies were attained, aimed at a reduction in production costs of micropropagated plants. The technology and 25 • • disease-free materials were transferred to a number of production laboratories in Vietnam, enabling them to supply large quantities of high-quality microplants for production. Information gained on levels of disease susceptibility and pathogens of banana genotypes will enable the study of biocontrol and genetic improvement of disease resistance in banana. The addition of genetic resources from introduced material, and the addition of tetraploids to the Vietnamese collection by in vitro polyploidation are providing valuable new materials for future banana breeding programs. Outcome: Higher economic returns for farmers are expected from improved understanding of production practices. Genetic resource activities are providing information to enable selection of potentially useful cultivars for breeding programs both in Vietnam and internationally. Many of the accessions being. analysed have not previously been collected, therefore this project is providing a valuable source of material with potential for disease and pest resistance and quality characteristics. The project provided the impetus required in Vietnam to focus on this important commodity, including the training of scientists, which is a valuable resource for future activities on banana improvement in Vietnam. Impact: The project has allowed scientists in Vietnam to address issues of banana research for the first time. New material has been collected that is potentially useful for world-wide breeding programs. An internal network has been established with other Musa research institutes, and strong linkages have been formed with Australia as a result of a scientific training course arranged with PN 15 at the Queensland Department of Primary Industries (QDPI). AREA 3: Plant Improvement Banana genetic improvement schemes that have been selected by BIP rely on either conventional or molecular approaches to crop breeding (or both). The BIP Research Portfolio consists of four projects addressing the availability and improvement of naturally occurring accessions for use in breeding programs. PN 1: Field and laboratory evaluation of diploid bananas for their use in breeding schemes. Centre de co-operation internationale en recherche agronomique pour le développement (CIRAWFLHOR), Guadeloupe Activity: The objective of this project is to supply breeding programs with reliable information on diploid lines with desirable characteristics for use as parent stock. Agronomic and laboratory characterisation methods are being used, as are wild and cultivated clones. Background: Collaborative arrangements for evaluation were undertaken between CIRAD-FLHOR Guadeloupe, Montpellier, and New Caledonia; WIBDECO, Saint Lucia; and the Queensland Department of Primary Industries (QDPI, PN15), Australia. Evaluations focused on agronomic and morphological traits, molecular characterisation, and resistance to black and yellow Sigatoka and Fusarium Wilt 26 Race 4. Accessions were obtained from the CIRAD-FLHOR collection and the INIBAP Transit Centre. Results: • • • • Indexed and virus-free accessions have been multiplied and sent to prepare field sites for disease resistance evaluation. Evaluation protocols have been devised for New Caledonia (black Sigatoka resistance trials) and Saint Lucia (yellow Sigatoka resistance trials). Accessions are screened for Fusarium Wilt disease in Australia at QDPI. Accessions are tested for their agronomic expression in Guadeloupe due to the absence of most major diseases. Isozyme, RFLP, and microsatellite analyses have been performed on all existing accessions in Guadeloupe, and morphotaxonomic descriptions carried out. Complete characterisation of the diploid clones is being undertaken using all available markers. Accessions new to the Guadeloupe collection have been assessed for their ploidy level by flow cytometry methods, and are currently being evaluated using all available markers. This work will result in the classification of clones by genotypic and phenotypic characters, and will assist in the identification of relations between wild and cultivated diploids. Indexation results are being awaited from Australia to enable the fmal batch of accessions to be dispatched for Fusarium Wilt and yellow Sigatoka screening trials. Outcome: This project is providing helpful information on diploid material used by all breeding institutes, and is a valuable addition to the existing information on banana accessions held in genebanks around the world. This rational basis is important to identify clearly the diploid material of the greatest potential within the available germplasm, natural or not, in order to optimise the use of resources available to breeding programs. Impact: Information has been supplied on diploid materials that are useful within all breeding programs. This has been of immediate use in the CIRAD and CRBP programs, and is also available to others. It is expected that use of this information will result in a shorter timeframe to produce suitable hybrids with the required traits. CIRAD has conservatively estimated that this new knowledge will result in a ten-fold increase in the number of dessert bananas produced. PN2: Field cross experiments for the purpose of understanding the mechanism governing the inheritance of Black Leaf Streak resistant characters in banana. Centre de co-operation intemationale en recherche agronomique pour le développement (CIRAD-FLHOR), Guadeloupe/Montpellier Activity: All breeding programs are concerned with the development of varieties showing resistance to black leaf streak disease (BLSD) in order to lessen the use of fungicides, and therefore the cost of production, environmental damage and human health concerns. Field resistance screenings against BLSD led to the identification of two forms of resistance reported as highly resistant (HR) and partially resistant (PR). Within the frame of this project, experimental crosses were undertaken with three wild Musa acuminata accessions expressing the HR, PR, and susceptible (S) 27 reactions against BLSD. Moreover, the project aimed at establishing linkages between field observations and molecular studies in order to identify quantitative trait loci (QTL), and to localise the main resistance genes on a saturated genetic map of bananas. Background: Ongoing field test screening for resistance indicates that the environmental effect on gene expression is significant, and that controlled environment testing is necessary for a precise analysis of the inheritance of the HR and PR resistance forms. Therefore, research on the PR form of resistance was dropped because it was found that field evaluation procedures routinely used were not precise enough in this case for genetic analysis. The project focus on observation of resistance pattern of hybrids arose from crosses involving the HR resistant parent and their correlation with molecular data for the localisation of the main resistance genes. Results: • • • The phenotype of most of the 153 clones of the F2 progeny from the HR x S cross has been confirmed. Questions remain about the phenotype of some hybrids (HR versus highly PR). This was undertaken by field evaluation on at least two cycles in two locations, Njombe and Ekona, CRBP, Cameroon. Molecular characterisation of the F2 progeny has been undertaken in CIRAD where the genotype has been characterised for 39 polymorphic codominant markers (RFLP I microsatellites) and 110 polymorphic dominant AFLP markers. Results showed significant allelic distorted segregations (for 58 markers) in the F2 progeny that highlight the use of molecular study for the inheritance analysis. A first mapping was concluded to join the markers in 11 linkage groups. Correlation made with field observations ranked in six classes of resistance, and a second RFLP marker mapped on a different linkage group with a lower significance level. Observations made on the F-progeny from the HR x PR cross-concluded unequivocally that the PR phenotype is dominant over the HR phenotype in bananas. Outcome: For a precise identification of resistance genes, protocols for glasshouse trials are being developed by CIRAD. This activity is designed first to confirm, in controlled conditions, the resistance phenotype of the hybrids, and second to identify all different genes involved in the resistance, even minor genes, for a durable resistance. The main genes/QTLs involved localised on the genetic molecular map, will enable the early screening for BLS resistance plants or hybrids to be done directly on vitroplants with molecular markers (marker assisted selection). The banana genetic molecular map first developed by CIRAD is being completed, and many codominant markers have been mapped and information is available for any gene / QTLs localisation and identification. This work contributes to the breeding of BLS-resistant plants that are economical and ecological necessities. Impact: Molecular marker assistance in the identification of resistance genes for black Sigatoka will initially be useful for conferring partial resistance to black Sigatoka through breeding programs. The assumption is that this will decrease the spraying frequency for black Sigatoka by 50% or more. The understanding and 28 identification of genetic resistance mechanisms will be a critical component of future strategies to reduce the economic and environmental impact of existing chemical control methods. PN5: Banana breeding in Brazil Empresa Brasileira de Pesquisa Agropecusiria (EMBRAPA), Brazil Activity: The aim of this project was to breed high-yielding varieties of banana resistant or tolerant to the main pests and diseases, with agronomic qualities of low plant height, short production cycle, and optimal fruit bunch and fruit size. This was attempted by the production of superior diploid and tetraploid cultivars and identification of new varieties; and the development of techniques for in vitro fertilisation and for propagation with reduced somoclonal variation. With the arrival of black Sigatoka into the Amazon region in 1998, the objectives of this project became even more critical than previously. Background: Banana production is an important agricultural activity in Brazil, with high economic and social benefits. The cultivated area covers about 516,000 hectares. The industry is based on smallholder farmers and is an important source of employment. BIP enabled the Brazilian program to be recognised internally and internationally. Valuable diploids were evaluated and selected as breeding parents to create diploid and tetraploid cultivars, as well as AAB dessert cultivars. This material was evaluated for resistance to rhizome weevil borer, Fusarium Wilt, Moko disease, and yellow and black Sigatoka. In tests to evaluate for nematode and insect resistance, researchers have found one hybrid that is tolerant to the rhizome weevil (Cosmopolites sordidus), and two hybrids that are tolerant to the nematode Radopholus similis. Results: • • • • • • • Diploid hybrids are being obtained by introducing pre-selected genotypes and/or new material from other breeding programs. Diploid and tetraploid hybrids were evaluated for resistance to black and yellow Sigatoka, Fusarium Wilt, and nematodes, as well as for agronomic characteristics and fruit quality. The resultant hybrids were evaluated for 23 desirable characteristics. 28 diploid hybrids and 13 tetraploids (12 Prata and one Maca type) were selected. The tetraploid (AAAB) hybrid PV-03-44 and the cultivar Caipira were recommended for commercial planting in the Amazon region. Techniques have been refined for in vitro fertilisation, somatic embyrogenesis, micro-propagation, somaclonal variation, molecular characterisation of germplasm, and somatic hybridisation. Generation of AAAB seeds and seedlings from crosses of Maca AAB and AA diploid PA Musore 2. Development of a device to measure the resistance to fruit detachment for use during germplasm evaluation. Outcome: The arrival of black Sigatoka in Northern and Northwest Brazil made this project one of considerable urgency. Two new varieties, a cultivar Caipira and hybrid 29 PV03-44, partly developed under BIP and resistant to black Sigatoka, are being used as a barrier in an attempt to avoid the spread of the disease to the plantation growing areas in Southern Brazil. These arid other materials that are being developed under BIP will provide valuable breeding lines for international breeding programs. Viable seeds and seedlings of AAAB types are considered an important result of the project, since no Maca type tetraploid hybrid had been generated previously. Impact: The project resulted in improved selection of breeding lines for EMBRAPA programs for resistance to black Sigatoka, and other pests and diseases of banana and plantain. BIP enabled the Brazilian breeding program to accelerate trials in various ecosystems, assisting in the development of a Sigatoka-resistant clone. This clone is currently being planted in the Amazon as part of a program to prevent further distribution within Brazil of this newly arrived disease, which is potentially devastating for the industry. PN22: Breeding hybrid Musacea with resistance to multiple diseases, especially black Sigatoka and Panama disease. Fundacion Honduretia de Investigation Agricola (FHIA), Honduras Activity: This project builds on the existing breeding program to obtain hybrid bananas with resistance to multiple diseases, including black Sigatoka and Panama disease, by conventional breeding. Background: Pollinations were made onto established and new plantings of the seed-fertile Lowgate (the shortest dwarf mutant of Gros Michel) and Dwarf Prata triploid parental lines for the production of tetraploid hybrids. The two main diploid pollen parents used in these 3x x 2x cross-pollinations were the burrowing nematode-resistant SH-3142 and the Race 4 of Panama disease-resistant SH-3362. The SH-3444 tetraploid was also pollinated in 4x x 2x crosses to produce tiploid hybrids. It has previously been verified that the 4x x 2x approach is a highly effective way of breeding new cooking bananas. It is therefore anticipated that breeding for a triploid export dessert banana will also be successful. Results: • • • Of the 2200 Lowgate bunches harvested, 16 seed were produced and 10 had embryos that were cultured for germination. Four of these 10 embryos were viable, and three of the plants produced have been evaluated in the field. These three mature tetraploid plants with Lowgate parentage had the desirable Grand Nain-type plant height, but were susceptible to black Sigatoka and had mediocre bunches. These results illustrate that much greater numbers of Lowgate bunches should be pollinated to produce larger populations from which selections could be made. Such larger segregating populations would greatly increase the possibility of selecting a hybrid with black Sigatoka tolerance and bunch qualities of FHIA-23, but which would also have the desired shorter plant height. FHIA-23, which was derived from Highgate x SH-3362, is tolerant to black Sigatoka and has excellent bunch features. From the 450 Dwarf Prata bunches harvested, a total of 685 seeds were produced and 250 had embryos. From these 250 cultured embryos, 123 30 • germinated and most of these hybrid plantlets were transplanted to the field for subsequent evaluation. The 600 bunches of SH-3444 yielded 84 seeds and 20 cultured embryos. From these 20 cultured embryos, 17 germinated. Two of these 17 plants have the gene for dwarfness and only these dwarf plants will be transplanted to the field for further evaluation. Outcome: The results of this project demonstrated that dwarf tetraploid export bananas could be bred. However, larger segregating populations must be produced to expect subsequent selection of dwarf plants with the desired bunch qualities and resistance to diseases. In view of the proven effectiveness of both the 3x x 2x and 4x x 2x breeding schemes for producing tetraploid and triploid commercial hybrids, it is critical that these activities be continued. Impact: The ability of FHIA to produce varieties displaying the desirable traits of pest and disease resistance, short stature, and good fruit quality was assisted by BIP, and will have a major impact on banana production when achieved. Field trials will be continuing on the various crosses undertaken by the project. AREA 4: Pest and Disease Management 4.1 Disease Control: Fusarium Wilt (Panama disease) PN15: Variability and relationships within populations of Fusarium oxysporum f.sp. cubense from its centre of origin. Queensland Department of Primary Industries, Australia Activity: This project was intended to provide information on the genetic diversity within populations of Fusarium oxysporum f.sp. cubense (Foe) in various areas in the Asia/Pacific region, and the risk this poses to banana production. The reliability of using tissue culture-derived plants as opposed to conventional planting materials when assessing pathogen resistance was also addressed. The project will enable breeding and selection programs to screen genotypes against a more complete representation of the pathogen. Isolates lyophilised during the study will also be available for use in virulence and pathogenicity tests in countries that are able to test nonendemic strains. Background: There are no control mechanisms for Panama disease, meaning that once infected the plantation has to be abandoned, with the pathogen remaining in the soil for long periods. Studies have been undertaken on genetic diversity of Fusarium as well as on field evaluations of potential breeding materials. Specific parts of the ribosomal DNA within the Fusarium Wilt fungus were characterised and sequenced, and this information was used to identify genetic variation among the different races of the fungus. Associated activities included the planning and execution of a six-month training course for four scientists from the BIP-funded project in Vietnam, and collaboration with CIRAD-FLHOR, Guadeloupe, in screening diploid breeding materials for Fusarium Wilt. 31 Results: • • • • • • • • Establishment of a world-wide collection of over 2000 isolates of Foe (Africa, South and Southeast Asia, North and South America, Australia) that have been lyophilised and will serve as a reference collection. Diversity of Foe determined through the use of vegetative compatibility groups (VCGs), DNA fingerprinting, and volatile groups. Determined the global distribution of strains of Foe. Conducted in-depth studies on the pathogens from the presumed centre of co-evolution. Tropical Race 4 populations (VCGs 012 13-01216), which are Cavendish-competent in tropical regions, were restricted to Southeast Asia (Taiwan, Indonesia, Malaysia, and Irian Jaya) and far northern Australia. Identified strain-specific PCR primers for development of PCR-based detection systems for Foe. Developed a library of DNA banding patterns. Characterised and sequenced the internal transcribed space (ITS) region and the intergenic spacer region (IGS) of the ribosomal DNA (rDNA) genes of Foe. Identified genetic variation among different races and VCGs based on IGS haplotype. Constructed a phylogeny for the genotypes of Foe based on DNA fingerprinting and IGS sequence information. Developed protocols for field evaluation of hybrids and diploids for resistance to Fusarium Wilt. Provided information on their performance to major pathogen groups. Demonstrated the greater susceptibility of plants derived from tissue culture to Fusarium Wilt when compared to conventional seed pieces. Studies have commenced to clarify the cause and to increase their resistance using rhizosphere bacteria, endophytes, mycorrhizae and abiotic inducers. Fusarium Wilt-suppressive soils have been identified to assist in this study. Evaluated a Muss acuminata ssp. population that is segregating for resistance to subtropical Race 4. Outcome: Information has been provided on the pathogenic and genetic diversity of Foe in various regions, enabling breeding and selection programs to screen Musa genotypes against a more complete representation of the pathogen. The project established that maximum variability was found in the Asia region, considered the centre of origin for Foe. Until this study, Fusarium Wilt in Cavendish cultivars was thought to be confined to the subtropics due to cool winter temperatures. Since no predisposing factors have been recognised in Southeast Asia and northern Australia, isolates in VCGs 0 12 13 and 0 12 16 appear to constitute a tropical Race 4 that has the capacity to threaten the world export trades. The only way to control Fusarium is through host resistance. Therefore disease-resistant germplasm is essential to any breeding program, together with knowledge of the genetic nature of resistance. Consequently CIRAD-FLHOR have sent their breeding diploids to Australia for field evaluation. Work on the identification of soil-borne resistance organisms will enable inoculation regimes to be designed for micropropagated plants prior to field planting. Impact: The collection of various strains of Foe will enable the more efficient identification of the pathogen, being especially useful in the identification of suitable plantation locations. The increased knowledge on genetic diversity will be useful to breeding programs to screen against a more complete profile of the pathogen, as will the protocols for field evaluation of hybrids and diploid materials. Quarantine or 32 exclusion procedures need to be quickly implemented in the Asia/Pacific region to confine tropical Race 4 populations to their existing areas of distribution. The finding of a Muss acuminata ssp. population segregating for resistance to subtropical Race 4 should be useful in developing molecular markers leading to the identification of host resistance genes. BIP has developed and strengthened professional links between the project and the world’s major breeding programs, particularly FHIA, CIIUD and EMBRAPA. Sigatoka Disease (Black Leaf Streak) PN17: Origin and distribution of fungicide-resistant strains of M. fijiensis in banana plantations in Costa Rica. Corporacion Bananera National (CORBANA), Costa Rica Activity: This project sought to increase knowledge on the dynamics of isolates of M. fijiensis (the causal agent of black Sigatoka) resistant to fungicides, to allow the design of better strategies for use of fungicides that will enhance the efficacy of control strategies. The project also investigated the biology and epidemiology of resistance to two major fungicides (benomyl and propiconazole) and less sensitive strains to improve the understanding of the effect that genes for resistance have on some important components of fitness of M. fijiensis. Background: Results to date have shown that the genes controlling resistance to fungicides in the pathogen do not reduce its aggressiveness. It has also been confirmed that two fungicides, propiconazole and tridemorph, are synergistic in their effect on the fungus, and thus may give better control in combination. Most importantly, it is now known that the previously proposed strategies for managing fungicide resistance to the more important pesticides will not work. The practice of rotating pesticides to allow a decline in the frequency of resistance to favoured pesticides will only result in failure, as the expected reduction does not occur. Apparently the expected selection forces that would cause the change to occur do not function as expected. Once resistance to a pesticide has occurred there is little hope of it once again becoming effective. This has serious consequences for banana producer’s world-wide. Results: • • • The sensitivity of progenies from 28 crosses of isolates of M. fijiensis with different sensitivities to propiconazole has been determined. Acquisition of genes for resistance to chemicals does not have an effect on the fitness of the pathogen populations. Resistance patterns remain although fungicide use is stopped. A greater understanding of the relationship between acquisition of resistance and the ability of isolates of M. fijiensis less sensitive to propiconazole to produce conidia in culture. Outcome: Results from conidial sporulation between propiconazole less sensitive, and propiconazole-sensitive strains of M. fijiensis, are important to understand the effect that gaining genes for resistance has on fitnesses of the pathogen. This would help to explain why some strategies, and what strategies, would help prevent or 33 diminish the resistance population, thereby improving the control of the disease. Results from crosses and inheritance of the resistance to propiconazole will help understand the potential for triazole-resistant genes to flow in the population and, in conjunction with other aspects of fitness, will help clarify the efficacy of strategies for resistance management in the population of M. fijiensis. Impact: Project activities have confirmed that a chemical-free period does not lessen the impact of populations of pathogens resistant to fungicides. Although this was a negative finding in some respects, it did confirm that strategies for improved use of chemicals to control the black Sigatoka pathogen are not easy to develop, and that continuing development of new chemicals, together with associated costs, is the principal avenue by which some retention of chemical control can be managed. Banana Streak Virus (BSV) PN14: Elimination of banana streak badnavirus (BSV) from improved Muss germplasm and related studies on transmission and host plant/virus/vector interactions. International Institute of Tropical Agriculture, Nigeria Activity: This project first aimed to develop reliable therapeutic methods to eliminate BSV from improved Musa germplasm. This objective was later altered to the identification of Musa plant tissues that are free of both genomic and episomal BSV, when it was found that the virus had the ability to integrate into the Musa genome. The project focused heavily on the need for a greater understanding of the host plant/virus/vector relationships, and BSV transmission and identification of its vectors, when it was realised that the virus was far more insidious than originally thought. The project also undertook to provide a protocol for the elimination of BSV from infected genotypes. Background: IITA undertook to seek more reliable methods of detection of BSV in naturally infected plants. BSV is a difficult virus inasmuch as it has the capacity to insert itself into the host genome, thus making detection difficult. IITA researchers and their collaborating partners have confirmed that the IC-PCR method is detecting viral (episomal) DNA, and they are continuing to work toward refining this detection technique. They are also continuing to gather more information about the insect vectors of the virus, and the best protocols for evaluating disease symptoms. Results: • • • • Diagnostic techniques were developed for BSV detection in infected. banana and plantain. BSV transmission through suckers was proven, and transmission through true seed implicated. It was also found that materials propagated through in vitro culture were highly susceptible to the virus. Several mealybug species that are potential vectors were identified and colonies initiated, but transmission was not confirmed. The researchers developed a better understanding of the symptom expression of BSV-infected improved IITA hybrids and local landraces, under natural and controlled-environment conditions. 34 • The project added significantly to the knowledge base of screening for resistance and characterising field resistance and/or tolerance. Outcome: This project resulted in an increased understanding of BSV: its etiology, epidemiology, and possibly control. As a direct result of the activities undertaken, informed decisions regarding the safe movement of germplasm can now be taken. The project has also generated knowledge that will greatly assist future breeding efforts and quarantine activities. Impact: The project has produced increased knowledge and understanding of BSV. Information relevant to the safe movement of germplasm and appropriate quarantine activities was produced, although the virus remains a serious constraint to the movement of germplasm due to its incorporation into the genome. BSV is now also increasingly recognised as a production constraint in certain banana-growing areas. 4.2 Pest Control: Nematode Resistance PN18: Study of tolerance/resistance of bananas to nematodes. Centre Regional Bananiers et Plantains (CRBP), Cameroon Activity: This project attempted to identify sources of resistance to the major nematode species of banana and plantain from the Musa germplasm collection at CRBP, and the provision of information on the susceptibility levels of plantain and other cooking bananas, as well as identifying less tolerant plantains that can be recommended to farmers. The project will assist the CRBP breeding program, by providing information on the susceptibility levels of parental materials and their progenies to two of the most important nematodes of bananas and plantains (Radopholus similis and Pratylenchus goodeyi). Background: Banana and plantain are important staple foods for a significant proportion of the Cameroon population, and are an important source of export revenue. Radopholus similis is a major pest in commercial plantations at low altitudes, whereas Pratylenchus goodeyi is dominant in smallholder plantations in highland areas with cool temperatures. No control of nematodes is made in smallholder plantations, whereas in commercial plantations the use of nematicides is one of the main costs of production. This project has a close association with the CIRAD Musa improvement program, and established a collaborative linkage with the nematode consortium spearheaded by the KULeuven Project, PN19. Results: • • Data collection continues in the trial concerning the verification of field resistance of accessions selected during pot experiments for resistance to Radopholus similis on two sites (Njombe and Mbalmao). The sites have different soil and climatic characteristics, but R. similis is the dominant nematode pest species found in the two locations. Results of the analysis of root samples collected nine months after planting confirmed the resistance of four accessions to R. similis, comparable to that of Yangambi Km5. 35 Outcome: Thirteen plantain hybrids and nine diploid clones have been compared for resistance to R. similis. All but one was susceptible, with one diploid showing a low level of resistance. Tests to compare resistance in tissue culture versus conventional planting material to the nematode P. goodeyi showed that there was no difference between the two. The project is contributing to the pool of knowledge on the resistance of Musa to nematodes, the development of new procedures for the selection of Musa clones for resistance to lesion nematodes, and in the identification of new sources of resistance to R. similis. Sources of identified resistance are available to breeding programs. Impact: As well as contributing to our overall knowledge on resistance, results of this project will impact mainly on the breeding program at CRBP. This will enable the selection of hybrid parents displaying resistance to nematodes, therefore increasing the rate of genetic gain with respect to resistance. A number of the selected clones also display resistance to other banana pests and pathogens, enhancing the possibility that multiple resistance can be achieved. The development of an early screening method will help in the rapid selection of resistant material. PN19: Identification of durable nematode resistance sources in banana and plantain. Katholieke Universiteit Leuven (KUL), Belgium Activity: This project was aimed at the identification of durable nematode resistance sources in banana and plantain leading to improved nematode management beneficial to both the producer and the environment through the increase in yields; decreasing production costs; and a reduction in the number of applications of nematicides. The intention is to provide a nematological component to existing breeding programmes, and to screen banana and plantain genotypes for nematode resistance based on the development of early, rapid and reliable screening methods. Background: Nematodes constitute an important limiting factor of banana and plantain throughout the world. Both the high cost and toxicity of nematicides underlie the urgent need for the identification of nematode-resistant sources and an early and rapid screening method for use in breeding programs. The principal investigator of this project was active in obtaining co-financing through the Belgium Administration for Development Co-operation (BADC) and the Flemish Association for Development Co-operation and Technical Assistance (VVOB) to fund nematological research associates at breeding programs in FHIA Honduras, IITA in Nigeria and Uganda, VASI in Vietnam, and recently CORBANA, Costa Rica. The project also worked closely with PN21 in Spain during the initial year of activities. More recently, a memorandum of understanding was signed for closer collaboration with CRBP in Cameroon. Results: • • 6 Radopholus similis, 3 Praiylenchus coffeae and 6 Meloidogyne spp. populations were initially established for use in screening experiments. Approximately 30 greenhouse experiments were conducted at KUL, and 75 Musa genotypes were evaluated for resistance to these populations. 36 • • • • • The capacity to undertake large-scale nematode resistance/tolerance screening of Musa germplasm and comparison screening methods was provided at FHIA and IITA. Host plant responses obtained by early greenhouse screening (in soil in pots) have been compared with host plant responses under field conditions. Host plant responses of in vitro micropropagated plants screened in the greenhouse were compared with the host response of suckers. Preliminary experiments demonstrated that resistance can be observed under in vitro conditions, but due to contamination of the nematode population, this screening method has been compromised. Multiple nematode species resistance screening of Musa genotypes from various sources was undertaken with no new sources of resistance found. Several genotypes were identified that are much less susceptible to R. similis compared with a susceptible reference genotype. Outcome: This work will continue following the completion of BIP, with information being generated on the host response of Musa genotypes to infection at different levels of nematode screening. New techniques are currently being investigated to obtain contamination-free R. similis for the routine in vitro screening of Musa germplasm for nematode resistance. Work also continues at KUL on the host response of R. similis of a series of diploids and triploids from the Philippines. Although work continues, the data collected to date already enable a more reliable interpretation of the host response of Musa genotypes to nematode infection. Impact: A major impact is the establishment of the Musa Nematologists’ Consortium, which will continue to function following the completion of BIP. It is envisaged that additional partners from South America, the Caribbean, Africa, Asia, and Australia will be members in the near future. This is a unique consortium in the field of nematological research. For banana production, the consortium provides a reliable nematological evaluation under different agroecological conditions of existing or new varieties. Under the project, KUL was able to assist FHIA in the evaluation of the genotypes used or produced in their Musa breeding program. This is the first time that nematode resistance of these genotypes has been reliably and thoroughly documented, allowing the identification of the most interesting parent lines for crossings aimed at nematode resistance. Valuable information has also been obtained on the host response of Musa genotypes to nematode infection at the different levels of nematode screening: in vitro, under greenhouse conditions, and in the field. PN20: Identification of durable pest and disease resistance sources in banana and plantain. Fundacion Hondurena de Investigation Agricola (FHIA), Honduras Activity: In collaboration with KUL, Leuven, the researchers sought to improve nematode management on banana and plantain to benefit both the producer and the environment through yield increases, production cost decreases, and a reduction in the use of nematicides. Background: Nineteen genotypes of Musa have been screened for resistance to R. similis and P. coffeae, the two major nematode pests. Some genotypes with a high 37 degree of resistance to one or both were identified for further resistance breeding. Results showed that tissue culture plants could be used to differentiate between different degrees of resistance and tolerance in the genotypes tested. Results: • • • • • Carrot dish cultures of R. similis and P. coffeae are being maintained in the laboratory for a continuous supply of nematodes for inocula to infest pests in the shadehouse trials. Six shadehouse trials were completed, and four further trials are underway that include 29 genotypes. Data collection and management of two field trials established in 1997 continued with 17 test genotypes. Data from the completed shadehouse and ongoing field trials were entered into a computerised database and statistical analysis is in progress. The multiplication of 22 additional genotypes from FHIA’s collection was started in the tissue culture laboratory for use in the final phase of the project. Outcome: Technicians were trained to evaluate nematode resistance in subsequent hybrid developed in the ongoing breeding program. Most of the genotypes available at FHIA have been evaluated for resistance and tolerance to R. similis and P. coffeae, and are undergoing re-evaluation because results varied with type of planting material, and the test was not performed in optimal conditions. Field trials should provide the necessary information to determine if early testing in shadehouse conditions is a reliable tool for nematode resistance screening. This knowledge will assist in the selection of improved genotypes for breeding programs, resulting in varieties with increased yield and lower production costs due to reduced use of nematicides, and should also be beneficial to the environment. Impact: Prior to BIP there had been no validation of resistant material in breeding programs. The information from this project provided knowledge of resistance when selecting parents for crossing. New sources of resistance to nematodes were found, which have the potential to increase knowledge on the mechanism for resistance. This could have considerable value if a biotechnological approach is considered for resistance. Evaluations will continue, linked within the nematology consortium. PN21: Identification of durable nematode resistance sources in banana and plantain. Institut de Recerca i Tecnoligia Agroalimentaries (IRTA), Spain. Activity: The specific aim of this project was to establish a live collection of migratory endoparasitic and root knot nematodes of the main nematode species that attack bananas and plantains. A second activity was the preparation of banana plant material for resistance evaluation and the initiation of screening trials under greenhouse conditions. Background: The project was initiated as a single-year component of the nematode consortium with the same objectives. 38 Results: • • A total of 27 isolates comprising 3 genera and 6 important Musa-attacking nematode species (Radopholus similis, Pratylenchus goodeyi, P. coffeae, and Meloidogyne spp.) were isolated and established in monoxenic cultures, with the exception of Meloidogyne spp., which were established and maintained on alternate hosts. Plant material and screening tests of Musa germplasm of interest to the Canary Islands against Pratylenchus goodeyi and Meloidogyne spp. under greenhouse conditions were organised. Outcome: A pure, abundant, and highly infective inoculum was made available for plant material evaluation and disseminated to consortium researchers for use in screening trials. Impact: As part of the nematode consortium activities, the germplasm screening for resistance will enable more effective breeding programs to be initiated through selection of resistant material. Such development should reduce costs of nematode control and lower the use of nematicides. Research Achievements In summing up at the Third Scientific Meeting, the SAP Chairman observed that the BIP-funded research projects have made exciting discoveries, contributed considerable useful new knowledge, and established strong collaboration and linkages that will continue into the future. “Important advances have been made in the understanding of the genetics of resistance to black Sigatoka. We now have a much greater understanding of what we face with the Fusarium threat. Molecular tools to transform bananas have been developed, and key banana breeding programs have been strengthened as an investment in problem solving. Resistance to nematodes is now better understood, and we have in hand the means of controlling some significant virus diseases of banana. BIP has also demonstrated that chemical control strategies aimed at managing resistance to fungicides of black Sigatoka will not work. We also now know that tissue culture apparently predisposes plants to Fusarium and nematodes. BIP has fostered an informal but solid network of scientists that should continue for many years. Areas of research that need to be addressed within any future framework are listed below in random order: 1. 2. 3. Field testing of transformed Musa materials must begin as soon as possible. The scientists involved are encouraged to make sure that this happens. There is a need to address the emerging biosafety issues regarding field-testing of the transformed material. This requires immediate attention. All of the institutions involved should address the intellectual property rights questions that revolve around the discoveries associated with BIP-sponsored research. 39 4. 5. 6. 7. 8. 9. 10. We need to develop a better understanding of industry’s resistance to cultivar replacement, because the banana research community needs industry people as partners. The Musa research community needs to develop a strategy to address consumer acceptance; i.e., issues associated with transformed cultivars. Ignoring this issue could doom some otherwise successful research. There needs to be an investment in understanding the genetics of the pathogen of black Sigatoka (M. fijiensis). This research should parallel studies of the genetics of host resistance and the development of quantitative trait loci (QTL) probes to enable breeders to breed for durable resistance to black Sigatoka.” There is a need to develop an understanding of how to use antifungal proteins (AFP) as a part of a black Sigatoka resistance strategy. Will they function as single gene resistance mechanisms and therefore represent a genetic system susceptible to being overcome by pathogen race variation? There is a need to merge the excellent Fusarium research work with molecular approaches to resistance to banana disease. This applies as well to nematode research. Investment is required in biological control strategy research, given the tantalising results with Fusarium and nematodes. Maintenance of conventional breeding capacity is crucial, as it will be needed to put together all of the above. BIP, through the project researchers, has delivered many research projects, with many significant scientific advances. Important new methods for enabling others to do research have been devised. A community of Musa scholars has been formed, which has generated a renewed interest in what can be done in Musa improvement that can no longer be ignored. The work has produced an excellent return on investments, and all principal investigators and collaborators are to be commended for their outstanding work. Substantial advances have been achieved in Musa research, but we must build on these successes by generating new funds for both medium and long-term programs. The programs need to focus on the most promising results and pressing issues to maximise benefits of the research, which should be based on a market-driven approach. The diversification of products to provide choices for diversified demand patterns should be available, and should meet the criteria of consumer markets. The need for software and materials from BIP was highlighted as a response to the urgent need in the Latin American and Caribbean area for improving the livelihood of growers, both in cash value and nutritional value.” 4. Project Management 4.1 BIP Research Management The World Bank convened a Scientific Advisory Panel (SAP) to assist in the planning, selection and monitoring of research activities. 40 The 18 projects were identified through an internationally competitive research grants scheme. Proposals were reviewed by SAP against a set of merit-based criteria, and were selected to form a cohesive and complementary portfolio to work in parallel toward the project objectives. Projects were monitored by the SAP against progress and annual reports, and following acceptance of their recommendations by the PEA, principal investigators were notified of the review outcome. Continuation of financial support was based on acceptance of reports. Annual Scientific Meetings were also held to enable Principal Investigators to present their work to the cosponsors, their technical panels, and scientific peers. The members of the Scientific Advisory Panel over the course of the Project are listed in Annex I. 4.2 BIP Annual Meetings The First Cosponsors meeting was held in Washington, D.C., in March 1995, with further meetings held in Belgium in February 1996, and in Guadeloupe in March 1997. Reports of these meetings have been provided previously. The convening of AnnuaI Scientific Meetings was extremely beneficial for information exchange for the Cosponsors and Technical and Scientific Advisory Panels, and they enhanced the monitoring process. The principal investigators were able to present and discuss their work with their peers. The meetings were also an obvious venue for discussing and planning current project activities within the two consortiums, and for interacting with other collaborators within the BIP portfolio. This helped establish new and productive external collaborative and networking opportunities. The Fourth Cosponsors and Third Scientific Meetings were held in Douala, Cameroon, on 5-6 November 1998, followed by a field trip to the CRBP facilities in Njombe. The meeting was timed to coincide with the impending completion of project activities. As in previous years, the meeting venue was selected to take advantage of the presence of Musa researchers and the development community who were participating in a ProMusa Workshop and International Seminar sponsored by INIBAP, CIRAD and CRBP immediately following the BIP meetings. 4.3 Scientific Advisory Panel (SAP) A meeting of the Scientific Advisory Panel was convened on 4-5 August 1998. This was the final meeting in a series over the life of BIP, where the Panel convened on a biannual basis to monitor the progress of contracted research by reviewing the Progress and Annual Reports supplied by BIP principal investigators. In addition to reviewing the research portfolio, members of SAP also discussed the methodology to be followed for the impact assessment study, and the planned annual meetings. Membership of SAP is contained in Annex I. The original Panel consisted of six scientists, with the Program Manager serving as scientific secretary to the Panel. Their breadth of experience and expertise covered a wide range of scientific disciplines and knowledge of the banana industry. Following the portfolio selection phase of BIP, the original Panel was reconstituted to comprise three members, with additional technical expertise being provided by observers from the CFC and FAO/IGB, and a representative of the International Network for the Improvement of 41 Banana and Plantain (INIBAP). INIBAP is an international group within the International Plant Genetic Resources Institute (IPGRI) of the Consultative Group on International Agricultural Research (CGIAR). The members of SAP gave generously of their time and expertise during the life of the Project. The World Bank would like to acknowledge that the success of BIP is due in no small part to their dedication and input. 4.4 Financial Statement The Bank received the final disbursement from the CFC during September 1998, amounting to US$ 245,669.56 (equivalent SDR l79,120.81). Funding received over the life of the project amounted to US$ 3,325,205.84, of which US$ 2,659,876.00 was invested in research activities. The balance of funds (US$ 665,329) supported project management, annual meetings, and research monitoring activities. Expenditures in the last semester, June - December 1998, amounted to US$ 181,353.81. The balance of funding is held against commitments to the annual meetings, a final publication, and project management costs. On closure of the Trust Fund a final financial report will be provided, and an external audit will be carried out and results forwarded to the CFC. Any funds remaining in the Trust Fund will be returned to the CFC. A copy of the financial statement and signed auditors report from the 1998 financial year is included in Section 2 of this report. 4. Public/Private Sector Collaboration The World Bank established an on-going dialogue with independent producers and multinational companies in the export banana industry from the start of the BIP. The purpose was to obtain input from the industry as to the constraints to sustainable banana production and industry R&D needs. At recent meetings, the potential applications of biotechnology to solve the major disease problems were discussed. As a result of recent wider consultations with a broad range of companies, producer co-operatives and other industry representatives, the Bank is examining the feasibility of establishing a banana consortium, “BioBanana”, to jointly fund further applications of biotechnology to solve the major disease problems, especially black Sigatoka disease. 6. Impact Assessment 1. Strategic Knowledge: Collection and characterisation of a wide range of banana germplasm for use in breeding programs world-wide. • Increased efficiency of breeding programs through improved testing, screening, and typing methods, development of protocols; additional information on genetic markers for resistance to nematodes/Fusarium Wilt/black and yellow Sigatoka. 42 • • • 2. Identification of resistance genes for BBTV and I&IV. Improved and refined technology for genetic manipulation, including transformation systems (Agrobacterium-mediated and particle bombardment). Greater understanding of BSV vectors and etiology. Applied Outputs Beneficial to Producers in the Short Term: • • • • • 3. Transgenic Cavendish and Bluggoe plants going into glasshouse and field trials. Support to breeding programs, i.e. enabling EMBRAPA to refine a clone with black Sigatoka resistance, currently being planted in the Amazonas (NW) region. Continuation of work at FHIA Honduras on disease-resistant, shortstatured varieties. Evaluation of Indian Nendran varieties has identified more desirable types for Southern Indian conditions. Ability to type strains of Fusarium Wilt in given locations. Information regarding the continued resistance to chemicals once fungicides are no longer applied. Leverage of Funds: • • • • 4. Building on the original CFC investment, further funds have been attracted to BIP, both within the projects, and outside: i.e. WOB and BDC, Belgium for the nematode consortium personnel positioned around the world in various breeding programs. Funding from the New Zealand Government for post-harvest practices in Vietnam. Funding from the University of Hong Kong and the Governors Fund in Hawaii. Conservative estimate of funds contributed is US$ 2 million. Increased Collaboration • • • • • • • • Formation of consortium in biotechnology enabling the refinement of both particle bombardment and Agrobacterium transformation techniques. Formation of nematode consortium strengthened during the project with funding from VVOB and BDC that will continue past the completion of BIP. Collaboration between Australian and Vietnamese scientists, exemplified in the training of Vietnamese scientists by institutes in Australia. Collaboration between Australian and CIRAD scientists in screening for Fusarium Wilt on CIRAD breeding diploids. Collaboration between Costa Rica and Brazil in screening breeding materials. Attraction of new research providers into banana improvement. Additional permanent staff position created in CIRAD focused on banana improvement and, Continued generic resources provided by Belgium through nematode consortium. Although these outcomes are significant, the real value of BIP-supported work will be the long-term strategic knowledge flowing from the BIP projects. Landmark events 43 using some of this new knowledge might be the development of a Cavendish or alternative export banana type that is disease resistant. Other developments that might result at least in part from BIP research include socio-economic, environmental, and health benefits. The economic benefits would include reduced production costs (few chemicals required); the environment would be improved through less chemical usage; and banana producers would suffer fewer health problems by not having to handle as much toxic chemical material (organ damage to an estimated 1 % of workers according to Swennen 1997). Improvements would lead to better health of workers, higher productivity, and an overall increase in the quality of life of communities in banana-producing areas. A longer-term benefit to industry will be the enhanced collaboration in banana improvement work as a result of the creation of a Biotechnology Consortium (Katholeike Universiteit Leuven, Belgium, Boyce Thompson Institute, USA, and Queensland University of Technology, Australia). This grouping resulted in staff exchanges and the transfer and refinement of particle bombardment and Agrobacterium transformation and regeneration technology. In a similar way, BIP fostered the creation of a “nematode consortium”, which has resulted in improved screening techniques leading to a better understanding of nematode resistance. The impact assessment study indicated that BIP should probably provide a rate of return on investment of at least 20% to 33%. Conservative assumptions were used in the analyses, and additional analyses showed robustness of the results. This demonstrates that further investments in banana improvement research will be well justified. 7. Conclusion BIP results have provided methods for transformation and regeneration of bananas, and have world-wide banana producers closer to a disease and pest-resistant transgenic banana being a reality. Considerable progress has been made in identifying new germplasm suitable for breeding, and ways have been developed to improve the breeding programs. The independent impact assessment clearly indicates that the most important output from BIP is strategic new knowledge that can be used for future research and development projects. BIP has, therefore, made an important contribution in both the provisions of research results, and in catalysing collaborative activities and co-financing through the innovative process of utilising a competitive research grants program. This enabled the identification of research providers of excellent quality, and investing resources in a portfolio of research projects of direct relevance to the objectives of the program. The World Bank’s overarching commitment is to the alleviation of poverty. The contribution this Project has made toward the development of resistant varieties and new control methods of banana and plantain production is a positive step in reaching this goal. 44 PART II Project Evaluation and Detailed Research Findings 45 46 1. Introduction In 1993, the COMMON FUND FOR COMMODITIES (CFC) agreed to finance a fiveyear project, entitled “Banana Improvement Project” (BIP), for a total of US$ 3,566,000. The Agreement was signed between CFC, the World Bank as the “Project Executing Agency” (PEA), and the “Intergovernmental Group on Bananas” (FAO) as the Supervisory Body. The BIP officially started on January 1, 1994. Its objectives were: 1. To develop and evaluate, over a five-year period, a range of improved banana varieties with export potential, incorporating increased productivity with durable disease resistance, using conventional and non-conventional breeding techniques; 2. To develop more efficient and integrated disease management practices for black Sigatoka disease based on an improved understanding of disease epidemiology and pathogen variability in different areas and over different seasons. The expected outputs and desired activities are reported in Annex 1. During the preparatory phase to the Agreement, the Intergovernmental Group (IGG) had agreed on the proposal for this project, with the same objectives, and had even indicated a number of leading research institutes (such as the FHIA) and laboratories for possible implementation. The Agreement was manifestly focussed on (1) export bananas and (2) the Sigatoka diseases, but left some flexibility for interpretation. Terms such as ‘new Musa genetic material’ (Output 1), ‘related pests and pathogens’ (Output 2), and some terms in the activity titles allowed to a certain extent for other research objectives. The PEA had the difficult task of distributing the operational funds among a number of selected research bodies, in the most efficient way, but still maintaining the agreed objectives. The PEA took two major decisions: • • In 1994, it commissioned a “review of banana research needs, banana improvement programs and banana research laboratories”. The report of that review (October 1994) ended with a number of recommendations representing a research spectrum far exceeding the scope of the Agreement. The recommendations were nevertheless accepted. It addressed a “Call for Applications” in two rounds to probably the widest forum of bodies, actually or potentially involved in Musa research, ever approached for possible research involvement. The call referred to the objectives of the Agreement. On the basis of the commissioned review’s recommendations, the PEA eventually selected 18 of the 95 proposals received. These became the projects reviewed in the present document. The initiative has since been called a “Competitive Research Grant System”. 47 A cursory glance through the project titles makes clear that a number of research subjects do not match the original objectives and expected outputs: virus diseases (Projects BIP 09, 12 and 14), nematode pests (Projects BIP 18, 19, 20 and 21), and Fusarium Wilt (Project BIP 15). The only project addressing Objective (2) of the Agreement is, in a remote sense, Project BIP 17. This important change in the orientation of the BIP was implicitly accepted by the sponsors, probably in the first place for the following major practical reason: In the long period between the initiation of the concept (in 1989 within the IGG) and its shaping at the end of 1994, a change had taken place with respect to diseases and pests encountered by banana growers. It was felt that the black Sigatoka disease, while still a menace, could cause a less serious problem ‘in the end’ than the killing Fusarium Wilt and virus diseases, for which no sources of resistance had been detected for (conventional) resistance breeding. There was also the growing and increasingly shared perception that the nematode pests could be responsible to a great extent for the decline in banana production in many areas throughout the world. The actual operational phase of the BIP started by March 1995 for the majority of the 18 projects with the contracts signed and funds received soon afterwards. A major consequence of this late start was that only three to three and a half years were left for implementation. Another consequence of the revised concept of the BIP was the wide spectrum of research subjects covered by the 18 projects. A challenge was how to group the 18 topics with due respect to the original objectives and outputs? The PEA eventually used the following classification: A. BREEDING AND GENETIC RESOURCES: projects 01 to 05 and 22 B. BIOTECHNOLOGY: projects 07 to 12 C. PEST AND DISEASE CONTROL Disease Control: projects 14, 15 and 17 Nematode Control: projects 18 to 21. After due consideration, the present Review Team found that the following grouping is more practical for discussion, especially with respect to the Chapters “Background” and “Conclusions and Lessons Learned”: I II GENETIC IMPROVEMENT: including: Conventional breeding: projects 02, 05 and 22 Non-conventional breeding: projects 08 to 12 Genetic resources and resistance sources: projects 01, 03, 04, 18 to 21. PEST AND DISEASE CONTROL: projects 14, 15 and 17 48 It should be noted that this classification still recognises that projects 14 and 15 do 1 have implications for genetic improvement in their results . 2. Background This chapter provides information on the state of the art in Musa research at the time the BIP was shaped. Genetic Improvement Conventional Breeding The major goal of the current banana/plantain breeding programs is to produce disease/pest-resistant hybrids with good performance in the field and meeting local, regional or even international preferences, so as to be market competitive. The terms “durable resistance” and “multiple disease/pest resistance” can be a source of confusion. Any breeder naturally wishes to produce hybrids with a longstanding resistance to more than one disease and/or pest, but is also aware of the practical limits. The most practical goal is to obtain hybrids that could be satisfactory alternatives for certain regions/preferences in the immediate future. During the distribution-selectionintegration process of some such initial hybrids, the breeder can in the meantime work further on the construction of better performing hybrids, or hybrids that would meet the requirements of other regions, or resistance to other diseases or strains and so on. Such repetitive processes, if maintained for a long time, would hopefully meet the various requirements over the world and meanwhile tackle new strains, diseases, and pests. The actual goal of the breeder is to produce a range of hybrids, allowing for selection at any place according to the local/regional requirements. Not many breeding schemes are feasible in the case of Musa. When it comes to improving dessert bananas, the possibilities are even more restricted due to the long list of post-harvest and consumption criteria to be met. The program at FHIA (Honduras) has a long history in banana breeding, and its current efforts are based on the considerable capital in intermediate hybrids as well as in expertise, accumulated over three decades. It is widely accepted that this Institute offers the best opportunities for creating a resistant alternative to the famous ‘Gros Michel’ dessert cultivar, and Project BIP 22 was shaped with that purpose in mind. The EMBRAPA (Brazil) embarked on a banana-breeding program during the eighties with the aim to improve two popular dessert varieties in the country: ‘Prata’ and ‘Maça’. Both cultivars belong to a group (AAB) genetically quite distant from the 1 The ‘missing’ project numbers need comment. Projects 06, 13, and 16 were originally to be implemented, but were not contracted for various practical reasons (but see also Project 15 regarding Project 16). Project 07 was to deal with the ‘Biotechnology Consortium’ (Projects 08 to 10) but its budget implications were incorporated within the latter projects. 49 ‘Cavendish’ or the ‘Gros Michel’. Very different, and previously not used, genetic material had to be explored, and specific intermediary hybrids created. Disease resistance is focussed on Fusarium Wilt and the Sigatokas. At the time project BIP 05 was launched at EMBRAPA, breeding expertise had been built around the promoter Dr. Ken Shepherd, who had been a banana breeder of international repute at the Banana Board in Jamaica for many years. The breeding program at Guadeloupe (CIRAD-FHLOR) started in the seventies and is based on an entirely new strategy. While the programs at FHIA and EMBRAPA are built around the key crosses (3nx2n 4n; 4nx2n 3n; i.e. starting from existing triploid cultivars), the Guadeloupe program aims at de novo construction of the desired triploid hybrids with disease resistance. Suitable edible diploids are selected for the purpose, and manipulated (colchicine treatment) so as to obtain tetraploids which are then crossed for triploid obtention. Crosses with resistance sources are implemented during this procedure. The program proved to be valid, but the access to rather homozygous resistance sources is of paramount importance. Hence, Project BIP 02 was initiated to focus on the genetics of black Sigatoka-resistance. Non-Conventional Breeding The term ”non-conventional breeding” encompasses both molecular marker-assisted selection of cultivars with desirable agronomic traits, and the transfer of foreign genes into selected genotypes with a favourable genetic background, so as to engineer new traits. As such, non-conventional breeding has been a reality for a series of crops for a long time. However, the genus Musa remained largely outside the mainstream of biotechnology, and, in genetic terms, was an orphan crop at the outset of the BIP. It is beyond the scope of this review to explain this obvious deficiency, but certainly the general biotechnological interest focussed on crops of moderate climates, and on crops with a good regenerative potential (e.g. potatoes, rape seed), thus mostly excluding bananas and plantains. This situation has changed since 1990; a series of projects on the non-conventional improvement of bananas is now funded by various donors, and a network of collaborating institutions exists world-wide. The BIP with its limited funds and time has contributed to this improved situation, which contrasts to the pre-BIP situation in several aspects. Though there had been experiments on the transformation of banana cell cultures effected at Texas A & M University and KULeuven, no reliable transformation and regeneration protocols were at hand, nor were meaningful genes or promoters available. The general strategy was aimed at transgenic bananas producing edible vaccines. Doubtless, the development of the ”molecular toolbox” by Projects BIP 08, 09, and 10 was to catalyse progress in this important area. There was, and still is, no meaningful natural source of virus resistance in the germplasm. Again, basic research into genetically engineered virus resistance had started before BIP (e.g. at QUT, aiming at a coat protein protection strategy). This research would have continued without BIP, though at a lower speed and with less international co-operation. The same holds for Project BIP 12, and for the BIPenhanced research into virus resistance so that transgenic plants resistant to BBTV 50 and BMV can be expected soon in the field, a major aim of Projects BIP 09 and BIP 12. Although transgenic bananas expressing ectopic resistance genes under the control of regulated promoters were discussed before the BIP, none of the tools were developed: no resistance genes and no inducible promoters were available, and even the transformation process was far from being optimised, not to speak of reproducible regeneration of transformed cells or tissues. Project BIP 10, with some loose co-operation with Projects BIP 08 and 09, has changed this situation. Some time before the BIP started, research intensified in the area of fruit ripening and its control. A plethora of scientific papers appeared on the biosynthesis of ethylene, its regulation, and its function(s) before and amidst the fruit ripening. The corresponding genes were isolated and sequenced, and their expression in the ripening process tested. However, the genus Musa was largely excluded from this progress. Yet the rapid increase in the knowledge of ripening in plants simultaneously stimulated several groups to work on bananas, amongst them BTI and HKUST. At the onset of BIP, banana genes (e.g. ACC synthase) were already isolated and sequenced, and experimentation into the molecular processes underlying banana fruit ripening was underway. Project BIP 08 therefore enhanced an otherwise already established research initiative. Genetic Resources and Resistance Sources Unexplored resources It can be stated that by the year 1990, and under the auspices of international and co-ordinated national organisations, a core Musa germplasm of the whole tropical world had been collected and investigated, with a few exceptions. But the few exceptions remained important gaps because they were dealing with germplasm generated within the wild Musa area. Germplasm in Vietnam, Cambodia Laos and Myanmar had never been thoroughly studied. In India, and despite earlier important work achieved in different research stations, the status of the “Nendran-complex”, which is most important for food supply in South-India, was still unclear. They mainly belong to the AAB-Plantain group (Simmonds), but there are indications that the complex contains yet unknown cultivars and even wild relatives. An important group of AAB cultivars - the “Maia maoli/Popoulu/Iholena group”- exists in Polynesia, and is being investigated with modest support of ACIAR. The Projects BIP 03 and BIP 04 had the objective to fill the identified gaps for India and Vietnam respectively. Breeding potential Characterisation (morphological as well as molecular) and (agronomic) evaluation of the collected and conserved Musa germplasm had not made great progress by the year 1990. The problem does not so much concern the popular triploids, such as the 51 so-called dessert-AAA, AAB and ABB. These have almost automatically been th studied in various research programs during the 20 century. Most of the thus accumulated knowledge had been published, especially in N. W. Simmonds’ handbook, “Bananas”, the three editions of which remain standard works for consultation world-wide. With a few exceptions (e.g. Pisang Lilin, Pisang Jari Buaya) the numerous collected edible diploids had not been subject to systematic characterisation, let alone agronomic evaluation. The situation was equally unsatisfactory for the wild diploids. During the eighties, French scientists (CIRAD) embarked on a breeding program with the aim to construct new triploids starting from selected diploids which were inter-crossed, with colchicine treatment for polyploidisation as a convenient instrument at appropriate moments during that operation (cfr “Conventional breeding”). The first products actually have been evaluated in some of the BIP projects. Although the diploids involved had been studied for female/male fertility, their potential for the construction of the desired triploids (in morphological and agronomical terms) was a matter of speculation in most cases. Project BIP 01 was an attempt at meeting this requirement. This project had a second purpose, which is explained in the subsection on "Resistance sources". Resistance sources Since the breeding programs are mainly focussed on resistance to diseases and pests, the access to resistant diploid sources and precise knowledge on the genetics of resistance/susceptibility genes during hybridisation became a paramount need. The objectives of Project BIP 01 included the screening of the numerous diploids for resistance to Fusarium races 1 and 4, as well as to yellow and black Sigatoka (the latter disease also called “Black Leaf Streak”). Screening of breeding material for resistance to Fusarium Wilt also formed part of the Project BIP 15, although not exclusively focussed on diploids. Resistance screening against nematodes had been conducted in some instances in the past (e.g. at FHIA), but, again, not systematically on diploid sources. Besides, there was the need to develop standardised early screening techniques that would be easily manageable and would reflect performances in the field. The Projects BIP 18, 19 ,20 and 21 form the first attempt in this direction, with regular contact among the Principal Investigators within a “Musa Nematology Consortium”. 52 Pest and Disease Control Black Sigatoka The expensive chemical control of this fungal disease is increasingly facing the difficulty that the pathogen is acquiring resistance to the chemicals used. The genetic background of this resistance-building, and its implications for control, form the objectives of Project BIP 17. Study of the Mycosphaerella species within the area of origin (SE Asia and Pacific), a topic of fundamental importance for understanding the generation of new and illdefined strains in various regions (e.g. India), was unfortunately not the subject of a BIP project. Fusarium Wilt The development of race 4 of Fusarium oxysporum f. sp. cubense during the recent decades became a serious menace: the race attacked the ‘Cavendish’ bananas (resistant to race 1), and, although it was not active in equatorial regions, a “move” into the whole of the tropics could not be excluded. Hence, the urgent need for much more data on the pathogen within its area of origin, as well as for a better understanding of the genetic background of its evolution. Project BIP 15 was mainly shaped for that purpose. Banana Streak Badnavirus (BSV) The Banana Bunchy Top Virus (BBTV) had been causing much damage in Asia and the Pacific for a long time, but was sparsely present in Africa and not at all in America. The more recent outbreak of the Banana Bract Mosaic Virus (BBMV) disease seems at present to be confined to some regions in Asia. Control of both these viruses is feasible and screening techniques for exchange of plant material (in vitro) are available. In contrast, the BSV appeared as a menace only two decades ago, is present in America and Africa (and perhaps in Asia as well), and seemed to escape the standard control measures. Specific screening techniques were still not available by the end of the eighties. Consequently, the BSV was seriously compromising any exchange of breeding material and promising final products, especially when generated in America and Africa, where the major breeding programs are established. A proposal from the IITA to tackle this problem was thus most welcome, with Project BIP 14 as the consequence. 53 3. Review of the Projects BIP 01 Field and Laboratory Evaluation of Diploid Bananas for their Use in Breeding Schemes Description Co-ordinates: Organisation: CIRAD-FLHOR Guadeloupe Principal Investigator: Dr. Ch. Jenny Dr. F. Carreel (associate) Partners: Dr. V. Kagi, CIRAD-FLHOR New Caledonia Dr. H. J. Fagan, WIBDECO St Lucia Dr. K. Pegg, QDPI Australia Duration: 3 years, starting April 1995 Objectives The stated objectives of the project initially were: 1. The objective of the project was to obtain preliminary evaluation data on land race diploid cultivars and on bred hybrid diploids to assist in the selection of potentially useful material in breeding programs and/or production. Preliminary evaluation data in Guadeloupe will include information on basic agronomic characters such as cycle duration, plant height, ratooning, bunch characteristics (shape, weight, number of hands and fruits), and fruit characteristics (shape, length, taste). The diploids will also be evaluated for resistance to Black Leaf Streak in New Caledonia and Fusarium oxysporum f.sp. cubense race 4 in Australia. 2. The evaluations for agronomic characteristics and disease resistance will be supplemented by laboratory studies in Guadeloupe, to determine ploidy levels and molecular characteristics. It should be noted that while the above objectives appear interim/annual reports, they have not appeared since the 1997 However, a shortened objective “to obtain preliminary evaluation accessions” figured as sub-item (4) of an Executive Summary. objectives has been reported. 54 in the earlier Annual Report. data on diploid No revision of Activities Activities have not been listed systematically, but are reported under items in the Executive Summary or under Research Activities as “achievements’. The planned means/methods for the implementation of activities have not been explained, except partly for “molecular characterisation”. However, activities are expressed in the Milestone/timechart as follows: 1) 2) 3) 4) 5) 6) 7) 8) 9) Recording of agronomic characteristics over 3 cycles ( CIRAD-FHLOR Guadeloupe) Determination of the ploidy level (CIRAD-FHLOR Guadeloupe) Molecular characterisation (cytoplasmic and nuclear data) (CIRAD-FHLOR Guadeloupe) In vitro duplication and mailing of the clones not available at the ITC (CIRAD-FHLOR Guadeloupe) Indexation of the clones not available at ITC (QDPI) Evaluation for their resistance to Fusarium race 1 and 4 (QDPI Evaluation for their resistance to yellow Sigatoka (QDPI) Evaluation for their resistance to Black Leaf Streak (CIRAD-FHLOR New Caledonia) Evaluation for their resistance to yellow Sigatoka (WIBDECO) On the other hand, the same activities are listed in a different way within the “Executive Summary” of later reports, as follows: Different sets of activities must be distinguished in this project: a. Multilocality evaluation - Multiplication, shipment and reception of the plants - Disease resistance evaluation in remote sites b. Guadeloupe evaluation of already existing accessions - Agronomic and morphotaxonomic evaluation - Laboratory evaluation (molecular markers) c. Guadeloupe evaluation of new accessions - Field planting - Agronomic and morphotaxonomic evaluation - Laboratory evaluation (molecular markers) It should be noted that some activities might not be viewed as “research” in a strict sense: for example, the activities (4) and (5) in the first list; the first activity (a) of the second list. They are routine procedures, although their importance was crucial for the running of the project. Achievements The present review of achievements is based on the progress reports as well as on the Final Completion Report. Because of the confused reporting of the planned activities (as shown above), the items under which achievements were reported changed from one report to another (e.g. what is in the 1995-96 Annual Report under “Implications/ Results”, as a sub- 55 item of “Research Activities”, re-appears in the Interim Completion Report under “Relations with other Research Programs”, as a sub-item of “Implications/Results”). Nevertheless, the Final Completion Report is well structured and provides the necessary information with profuse details. For a clear reflection of the actual achievements, it was found more useful to refer to the purpose of the project: field and laboratory evaluation of diploids (for optimal use in breeding programs). The vast majority of the accessions are from Papua New Guinea (PNG). The following articulation of activities and achievements is used for practical reasons when the actual results are considered: 1. Field evaluation: a. b. Morphotaxonomical description; preliminary performance evaluation (at Guadeloupe). Disease resistance evaluation: black and yellow Sigatoka, respectively, at NewCaledonia -FHLOR- and St Lucia – Winban Islands, WIBDECO; Fusarium Races 1 and 4 at QDPI (Australia) 2. Laboratory evaluation: a. b. Ploidy assessment with flow-cytometry Molecular markers: isozymes; mitochondrial DNA (mtDNA), chloroplast DNA (cpDNA) and nuclear DNA markers, by using the RFLP technique. The achievements can then be reported as follows: 1. Field evaluation a. Morphotaxonomic description. The INIBAP descriptor system was used and the results compiled in a CIRAD designed software system. The Completion Report does not provide details but states that the data are available through the MGIS program of the INIBAP. On a request for such data, it appeared that this was not the case at the time the Review was finalised. b. Preliminary performance evaluation (i.e. Agronomic Evaluation). This exercise is not completed. The delay is partly due to the late start of this exercise (late ‘96) for about 60 of the accessions introduced from the ITC. While the report provides an extensive table with data such as bunch weight, finger dimensions, cycle length, these are not discussed and no performance evaluation is provided. The few comments in the reports deal with morpological characters rather than with agronomical ones and are part of the INIBAP descriptor list. They would thus be accessible via the above-mentioned MGIS system. 56 A glance at the table shows very interesting features, which would have been worth some discussion; For example, the contrast among PNG diploids in bunch and fruit characters is spectacular and deserves a wide diffusion of information for breeder’s use. One case suffices here: in the ratoon stage (second cycle after planting) and following about the same growth rhythm, cultivar ‘Gulum’ produces a 4-hand bunch of 6kg, while ‘Guyod’ bunches reach 16kg with 8 hands. b. Disease resistance evaluation Regarding in vitro propagation and shipment, considerable effort has been spent in the successful propagation and dispatching of about 250 accessions to the selected evaluation sites over the world. Although this is not a research activity per se, it has been a heavy burden on the project. The actual disease resistance evaluation is still in progress, due to the late establishment of the trials. Yellow Sigatoka-resistance evaluation had already been carried out at QDPI in field collection conditions. 2. Laboratory evaluation a. Ploidy assessment was an important achievement of the project. The carefully conducted flow-cytometric measurements and interpretations led to the general confirmation of what was found by previous morphotaxonomic work, with a few remarkable exceptions. Analysis of genome sizes points to significative differences between the Sections Australimusa and Eumusa, and between M. balbisiana and M. acuminata within the latter section. The results are the subject of a paper submitted for publication (of which the text was provided in the report). For the first time, a non-Eumusa triploid was detected: ‘Asupina’ was proved to be a M. textilis triploid. The deep significance of the result in the context of the phylogenesis of cultivars should be underlined. A few presumed diploid PNG accessions (see the INIBAP Musalogue for the PNG accessions) turned out to be triploids, with important breeding implications. b. Molecular markers The first reports deal with isozyme analysis, which was in fact effected during the pre-BIP period. The impressive cytoplasm and nuclear RFLP work was an extension of what had been performed before the onset of the project (Carreel, PhD 1994). More than 50 homologous probe-enzyme combinations were used for nuclear genome analysis, 43 among them presenting a simple pattern that may be assimilated to a locus. Eleven heterologue combinations were used for cytoplasmic analysis. Resulting dendrological structures were established, and the combined results on 94 edible diploid clones allowed for the detection of identical/proximal cultivars, largely confirming the presumptions based on morhological observation. 57 STMS polymorphism study led to the choice of eight markers with high discrimination potential and clearly localised on the different linkage groups. Most clones revealed different patterns this way too. No evaluation of the combined performance of the two methods (RFLP versus STMS) is reported, and the amount of plus-value of STMS is thus not evident from the report. Evaluation Methodology A desk study was carried out on the basis of the available interim reports. The Final Completion Report was due in June 1999. The project had been visited by one of the present evaluators during the CFC MidTerm Review exercise at the end of 1995. The molecular geneticist (F. Carreel) had just arrived at that time and the facilities for her research were nearly completely established. Results Evident achievements from BIP 01, are: - - - A massive flow over long distances of diploid material with potential for breeding. It certainly was a sound performance to transfer 140 accessions in good condition to New Caledonia, 77 to St Lucia and 67 to Australia. The establishment of a network for multilocational disease resistance assessment. A critical mass of accessions has been tested in local trials at four locations throughout the world. Although the results will only be available by the end of calendar 1999, this can be considered as a major achievement of the project. The considerable substantiation of the INIBAP MGIS system for cultivar identification at diploid level. Unequivocal ploidy assessment by means of flow cytometry has been realised for all the accessions. A powerful basis has been established for any further molecular marker research on the genus Musa, with deep implications for genetic improvement. Surprising findings, such as the above-mentioned triploid textilis, open entirely new prospects in the study of the phylogenesis of bananas, with, again, far-reaching implications for genetic improvement. One other rather exciting result was that the diploid accessions from the Comores and Madagascar do form a special group with perhaps no comparable genotype anywhere else in the world. Their potential in relation with the equally unique EastAfrican Highland bananas deserves special investigation. 58 However, and despite the abundant positive results, some shortcomings should be noted. Work on bred diploids – as specified in the original project goals - is not reported. On the molecular side, the PI admits that the RFLP method has but a limited efficiency in detecting differences between diploids, which can be similar in plant behaviour but widely different in inflorescence structure. This “cries” for the AFLP method (of which the discriminating performance among the Plantains has been proved), and one wonders why the method was not adopted in the course of the project. Conclusions The objectives have been partly met. It also appears from the reports that some of the activities involved were redundant: (a) Objective 1 was over-ambitious and partly redundant. The considerable effort of propagating and dispatching 310 accessions was a real (but non-scientific) achievement, resulting in multi-location disease resistance testing which, however, could not have been achieved within the active BIP 01 period. The intended agronomic performance evaluation over 3 cycles was an unrealistic goal. The expected results for all of the PNG accessions are still to be compared with the similar work achieved at the QDPI in Australia, for the establishment of a firm agronomical assessment. It is regrettable that the morphotaxonomic data of all these accessions are not accessible through the MGIS system of the INIBAP. In addition, the intention to evaluate “bred hybrid diploids”, formulated in this objective, does not appear to have been performed, judging from the reports. (b) Objective 2 was met with great satisfaction. Ploidy assessment and genome size evaluation were major achievements and even led to spectacular findings. The not-planned AFLP method would certainly have been of great help to better discriminate the bewildering range of PNG diploids, and it is recommended that this method be essayed in the immediate future. Notwithstanding the shortcomings, the project outputs have great significance. When the results on disease resistance are available, genetic improvement programs will be fully informed on all major aspects of the germplasm in order to select the most suitable diploids for particular objectives. 59 BIP 02 Field Crosses for Understanding the Inheritance of Black Leaf Streak Resistance to Bananas Description Co-ordinates: Principal Investigator: Dr. F. BAKRY CIRAD Montpellier Collaborators: Dr. F. CARREEL Dr. C. ABADIE Dr. K. TOMEKPE Dr. J. CARLIER CIRAD Guadeloupe CIRAD Cameroon CIRAD Cameroon CIRAD Montpellier Administrative Contact: H. TEZENAS du MONTCEL, CIRAD Montpellier Duration: 3.5 years, starting April 12, 1995 Objectives 1. 2. 3. Understanding of the genetic determinism of High Resistance (HR) form and Partial Resistance (PR) form against Black Leaf Streak Disease in bananas and relations among both resistance forms. The establishment of linkages between field observations, segregation and molecular studies to map the banana genome and locate the genes and quantitative trait loci (QTLs) coding for resistance versus susceptibility (S) to the disease. Develop the use of molecular marker-assisted selection in the banana breeding programmes. Modification to objectives Research on the PR form was dropped at a late stage of the project, because of the (previously underestimated) significant influence of environment on the PR phenotype, leading to variable assessment of a supposed same degree in resistance. Activities The Milestone/time chart (as produced by the Principal Investigator in the first 6month report) distinguishes the following activities (grouped and titled by the present evaluator): 60 A. Study of the resistance inheritance with cross (HRxS) 1. Cross (HRxS) and field observation of the F1 population 2. Selfing a selected F1 issued from (HRxS) cross 3. In vitro growth and duplication of the F2 population 4. Planting and field evaluation (resistance + agronomic data) Of the F2 population in two locations B. Study of the resistance inheritance with cross (PRxS) 5. Cross (PRxS) and field observation of the F1 population 6. Selfing a selected F1 issued from (PRxS) cross 7. In vitro growth and duplication of the F2 population 8. Planting and field evaluation (resistance + agronomic data) Of the F2 population in two locations C. Study of the resistance inheritance in cross (PRxHR) 9. Cross (PRxHR) and field observation of the F1 population D. Molecular genetics 10. Search for polymorphic markers in parents and F1 11. Molecular analysis of the F2 (HRxS) population 12. Molecular analysis of the F2 (PRxS) population 13. Genetic mapping and QTL analysis Activity Groups A, B, C were to be carried out at CRBP, Cameroon, and Activity Group D at CIRAD-FHLOR, Guadeloupe. The work in CRBP was carried out at Nyombe station (altitude 60m) with duplication at Ekona (altitude 400m). All activities were originally planned for the period 1/04/95 to 31/12/98. A 6-month delay was arranged for Activity Group (D) because of the molecular geneticist (F. Carreel) arriving in the project only in late 1995. Achievements Background information included: The parents for the crosses were 3 wild accessions: - for HR: clone ‘Calcutta 4’ (M. acuminata subspecies burmannicoïdes); for PR: clone ‘Zebrina’ (M. acuminata subspecies zebrina); for S: clone ‘Madang’ (M. acuminata subspecies banksii). The three parent clones are considered homozygous in a morphological sense, but molecular markers indicate some heterozygosity in ‘Calcutta 4’ (26%) and in ‘Madang’ (9%), which could point to a certain degree of heterozygosity in physiological traits of potential influence on the resistance mechanism. The distinction between HR (High Resistance) and PR (Partial Resistance) emanates from the CIRAD concept. In HR plants, the disease symptoms are hardly developing and the attacked tissue seems to “die off”, leaving the leaf intact as a whole. In PR plants, the symptoms develop up to a variable extent but necrosis is always late (if appearing at all), allowing the plant to develop a rather normal bunch. 61 Other institutes (such as the IITA) consider both HR and PR as expression of the same (but complex) resistance genes set. Hence the Project BIP 02 could be of basic significance in that its results could settle this issue. Review of the achievements The logical and transparent Activity Scheme enabled a good appreciation of the achievements, which were extensively reported. The achievements can be briefly reviewed in parallel with the Activity Groups. A. Cross (HRxS) All the 35 F1 plants obtained initially showed HR, but 6 plants began to display a PR phenotype at later stages. The “molecular marker heterozygosity” of the ‘Calcutta 4’ parent could have played a role in this variation. The F2 population (153 plants) showed a wide variation in resistance (but no S), with only 21 plants looking firmly HR, and the rest either HR (uncertain) or PR in various degrees. The uncertain HR evaluation is due to the less vigorous growth of plants, whereby the faster-than-normal leaf decay may not have allowed for the disease symptoms to reach final stages. B. Cross (PRxS) The 205 F1 plants differed widely in resistance but no HR was detected (perhaps a point in favour of the CIRAD theory in its clear distinction between HR and PR). However, the considerable spectrum in PR was much larger than the slight (9%) molecular heterozygosity of the ‘Zebrina’ parent would have led one to expect. Further observation revealed that the classical symptomatological field method led to highly varying degrees in PR assessment for the same F1 plant at different periods of the year. It was thus not possible to distinguish environmental from genetic factors in the PR assessment. CIRAD (Montpellier) also found (via molecular markers) a genetic diversity in Mycosphaerella fijiensis for a sample of one field and collected from the one plant. CIRAD is presently developing a method under controlled environmental (e.g. glasshouse) conditions for a more accurate resistance analysis in the case of segregating populations (in the understanding that the F1 population was already segregating, due to some heterozygosity of one parent). Consequently no F2 population was created and the B group of activities was halted. 62 C. Cross (HRxPR) Observation of the 156 F1s revealed two important facts: - No S (susceptible) plant was identified; The PR phenotype was dominant over the HR one. D. Molecular Genetics Because of the above-explained difficulty with the B group activities, the study was concentrated on the (HRxS) cross products. Considerable efforts produced 40 RFLP and 40 STMS markers (of which a majority appeared to characterise the F2 progeny), and more than 200 AFLP markers, of which 149 have been analysed to construct the genetic map. Results showed significant allelic distorted segregation in the F2 progeny for 58 of these markers, which proves how a molecular genetics study can complement the field observation results. Indeed, cytogenetic studies on meiosis of the F1 selfed parent showed its heterozygosity for at least three reciprocal translocations inferring four sets of chromosomes. In a mapping effort, the 91 remaining markers were joined in 11 linkage groups, with a high correlation constraint (a Lod score of 6). A Kruskal and Wallis analysis, made with field observations ranked in 6 classes of infection severity, led to the identification of one RFLP and one AFLP marker strongly associated to the resistance, and a second RFLP one, mapped on a different linkage group, with lower significance. Evaluation Methodology A desk study was carried out on the basis of the reports, including the Final Completion Report. The project had been visited by one of the present evaluators during the CFC MidTerm Review exercise at the end of 1995. The molecular geneticist (F. Carreel) had just arrived at that time and the facilities for her research were nearly completely established. Results This project is of a high quality in several aspects. The objectives were clear-cut and realistic and the layout of the activities was very clear. The achievements are impressive, considering the pioneering character of the research. The reports systematically and carefully reflect the progress made for each activity. A major breakthrough has been made in the understanding of the genetics of resistance to Black Leaf Streak (also called black Sigatoka) Disease. 63 The three groups of field activities A, B and C produced results of basic importance: - - - The CIRAD approach of distinguishing HR (High Resistance) from PR (Partial Resistance) remains a valid instrument for investigation. No HR was found among the 205 F1s from (PRxS) which would mean that the HR-genes were absent in the (slightly heterozygotic) PR parent. However, the Group A results point to the existence of PR-genetic components in the HR parent, since F2 segregation shows a wide range of PR besides the relatively few HR plants. But this distorted segregation is explainable since PR appeared to be dominant over HR, as the Group C results showed. The absence of S (susceptible) plants in the F1 population from (PRxHR) seems to prove that the resistance mechanism is linked to the existence of specific resistance genes. The discovery that environmental factors strongly interfere with genetic ones in the resistance assessment (Group B results) has far reaching implications for the future of field observations. As for the Group D results, molecular marker analysis proved to be a powerful tool for unravelling the HR-PR interaction. The ultimate results of the project will almost certainly produce the set of markers for early screening of HR and PR, a tool which is badly needed in breeding programs. Conclusion Objective (1) has been met in a very satisfactory way. Great progress was made in the understanding of the genetics of resistance to black Sigatoka. Objective (2) was only partially attained but the identified RFLP markers for resistance can now be tested for performance. Objective (3) is not yet met, but rapid progress can be expected on the basis of the present results. The delay is a direct consequence of the very complex nature of the resistance mechanism and the genes involved. The prospects for the future of this research are evident. The BIP 02 project has laid a powerful basis for further fundamental and applied research on resistance to black Sigatoka or Black Leaf Streak Disease. 64 BIP 03 2 Collection, Characterisation and Evaluation of Nendran Banana in India Description Co-ordinates Organisation: Kerala Agricultural University, Vellanikkara Principal Investigators: Dr. Rema Menon, Associate Professor (Hort.) Dr. K. Aravindakshan, Associate Professor (Hort.) Duration: 18 months, starting July 1, 1997 Objectives 1. 2. To collect, catalogue and characterise clones of Nendran banana and wild relatives, in close collaboration with the program at the National Research Centre for Banana (NRCB) with the purpose of establishing reference data for Nendran for use by future researchers. To evaluate clones for dwarfness, short cycling and disease resistance, particularly for yellow Sigatoka and to a lesser extent to weevils and nematodes. Activities 1. 2. 3. 4. 5. Exploration (July ‘97 to March ‘98): Collection of Nendran germplasm planting material in and around Kerala State, and at the Banana Research Station; Field Collection, Kanarra: Establishment of the Field Collection at Banana Research Station, Kanarra. Starting August ’97, through first half 1998; Morphological characterisation, using the INIBAP Descriptor List; Evaluation: Initial agronomic evaluation of phenotype, i.e. pseudostem height/girth, vegetative cycle, number of functional leaves at flowering stage, resistance to yellow Sigatoka disease (using the Gauhl scoring method for degree of attack) and to nematodes and weevils (no particular techniques explained); Molecular characterisation: Isozyme analysis/zymograms with following enzymes: peroxidase, malate dehydogenase, catalase, esterase, superoxide dismutase, shikimate dehydrogenase and acid phosphatase. Achievements 2 The project originally had another title and purpose: “Developing improved banana varieties with pest and disease resistance, post-harvest superiority and maximum export potential: Kerala, Krishnan Nair.” It was realized that the objectives were too ambitious for a BIP-type undertaking and had to be reformulated and concentrated on a feasible program. Hence the rather late start of the actual project. 65 1. Exploration was effected in 8 different locations which roughly cover the following regions: Southern Kerala and Southern Tamil Nadu (Location 1), Cardamom Hills (L2), South-Central Kerala (L3), North-Central Kerala (L4), Nilgiri hills (L5), Coimbatore and Thiruchirapilly districts, i.e. Central Tamil Nadu (L6), Northern Kerala (L7), and the forest areas of Western Ghats (L8). The 14 already existing accessions at the Kanarra Station (L9) were added to the collection. Thus, the entire envisaged area has been explored. A total of 128 Nendran cultivars were collected, with 34 from Central Kerala, and relatively few (4) from Central Tamil Nadu. Three wild relatives were found in Location 8. Only one Horn Plantain was found (Northern Kerala) and no False Horn could be collected (one “Horn” and one “False Horn” existed in the Kanarra collection). 2. Field Collection By June 1998, 125 accessions had been planted. They were given the accession numbers 001 to 125. 3, 4. Morphological Characterisation/Evaluation About half of the French Plantain accessions had achieved the first vegetative cycle by mid 1998 and were tentatively grouped in 12 categories, based on the following discriminative characters: - Persistent versus deciduous male bracts; Bunch orientation (pendant versus oblique); Compact versus loosely packed bunches; Plant stature: tall, medium, small (“short”), and one dwarf; pseudostem colour. By the end of 1998, most but not all of the accessions will have fructified and a firmer classification will be possible. Yellow Sigatoka-resistance appears to be low for all observed accessions with the possible exception of Category x (the “shorter” French Plantains). Nematode and weevil resistance is under investigation. 5. Molecular Characterisation Polyphenol inhibition of enzyme activity could be overcome by adding PVP to the extraction buffer. A standardised technique for peroxidase provided zymograms where 3 of the 30 cultivars analysed so far showed a distinct band pattern. These appear to belong to the “short French Plantain” category (accession numbers 017,045 and 048) but other “shorts” (043,044) could not be distinguished. It is likely that a comparative study of zymograms with other enzymes will enable further discrimination up to a certain extent. Work on esterase and superoxide dismutase was initiated. 66 Evaluation Methodology A desk study was performed on the basis of the three reports on the BIP project to the World Bank. These reports were examined and commented upon by the SAP. The project was visited from 31 December, 1998 - 5 January, 1999, and included: - Review of the project; review of the banana research at the University; Visit to the Department of Pomology and Floriculture (with the biotech- and isozyme analysis labs); Several visits to the field collection with discussion of morphological characterisation; Visits to farmers’ fields at Chalakudy and adjoining areas; Address by evaluator to academic and scientific staff and students on Musa genetic improvement; Discussion of the BIP 03 achievements, the pests and diseases resistance/control work, and the future prospects. Results Exploration and field planting were achieved in a short time. Further exploration of plantains looks unproductive; discussions and visits to farmers proved that the PI and his associates have a profound knowledge of plantain diversity and that they would not have overlooked other cultivars. Morphological characterisation The INIBAP Descriptor Lists have been established by a scientist (Dr. Rema Menon) with “expert eyes”. The Nendran (Plantain) germplasm in the Kerala region looks quite different in composition from that in humid Central Africa, not only because it consists exclusively of French Plantains, but also because the spectrum of the latter subgroup is a narrow one. The identification of “short” French Plantains is of interest because these may turn out to be semi-dwarfs (their vegetative cycle is not different from that of the “medium” categories). Such cultivars have been found in Africa only in the central rainforest zone. Initial evaluation tends to show them as more Yellow Sigatokaresistant than the main group. The one dwarf accession turned out to be similar (if not identical to) the ‘Njock kon’ of West Central Africa, a dwarf mutant of a Giant Plantain. Agronomic evaluation A potentially very important achievement was the collection and characterisation of a few French Plantain cultivars with some resistance to drought or to flooding conditions. Such traits have not been reported for the African French Plantains, and 67 study of their physiological background could lead to the identification of structural components (such as root anatomy) with high potential for genetic improvement. Molecular Characterisation It should be stressed that DNA markers for differentiation at inter-cultivar level are not yet satisfactory, and that better knowledge of isozyme patterns is not only an alternative, but could help in the construction of more adapted DNA markers (via RNA). The Kerala attempt at using the combined result of different isozyme analyses as a marker at inter-cultivar level looks promising and this research should be encouraged further. “Yellow Sigatoka (YS)”-resistance The study of Nendran at KAU shows that the majority of the cultivars are susceptible 3 to this ill-defined Leaf Spot disease , but that the “shorter” cultivars of category X are more resistant. It is suggested to change the name of the disease into “pseudoyellow Sigatoka” (because M. musicola is not the agent) for as long as the exact pathogen has not clearly been identified. It is recommended that the KAU seeks close operational contact with the CIRAD laboratories (via INIBAP), which are actively studying the different Mycosphaerella strains. Resistance to nematodes and weevil No striking results were reported for this item and its discussion was limited to generalities. On the basis of what is known for plantains in Africa, the large majority of the Nendran cultivars should be susceptible to Radopholus and Pratylenchis coffea because they mostly grow at low altitude. But since the Nendran complex has been cultivated in Kerala for perhaps two millennia or more, and considering the results of the agronomic evaluation, there is a chance for identifying resistant accessions (see Simmonds’ “Bananas”). Conclusions The KAU team has achieved important results of international significance in a very short time. There exist enough indications that the Nendran complex in Kerala State contains several cultivars of economic importance of which some have not been reported elsewhere in the world to date. 3 Plantains are known to be resistant to YS, at least in Africa at sea level, and to be susceptible to black Sigatoka (BS). Thus the Nendran (Plantain) suffering from a Leaf Spot disease in India calls for special attention. According to David Jones (Infomusa Vol. 4, No 2) the same symptoms of this disease are seen in other parts of India, as well as in Sri Lanka, Thailand and Malaysia. Examination of the samples (including from Kerala) could reveal neither Mycosphaerella musicola nor M. fijiensis but there were indications that “a Mycosphaerella sp. with Septoria as the imperfect state” could be the agent. Because the symptoms in Malaysia are now becoming combined with BS ones, and since in West-Africa something similar may have been the case at early stages of the BS invasion, there exists the great danger of black Sigatoka appearing sooner or later in the Kerala zone. 68 Further taxonomic work should concentrate on standardisation of nomenclature, in 4 view of the desired exchange of information and material with the rest of the world . The promising effort in finding isozyme markers should proceed further and morphologic counterparts should be sought for each cultivar, in view of a solid combined identification, which could be of much help for the screening of appropriate DNA markers. Agronomic evaluation should move towards physiological research on the key factors of “drought” and “flooding” tolerance discovered in some cultivars. Among the far-reaching implications of such research is the subsequent possibility of finding corresponding DNA-sequences (the expertise and facilities of the visited biotechnology laboratory clearly offers such opportunity). Regarding Disease Studies, and apart from the above recommended collaboration with CIRAD on the “pseudo-yellow Sigatoka”, the need for intensive research on resistance to Banana Bract Mosaic Virus (BBMV) disease (a vastly growing menace in Kerala) is becoming urgent. Forging operational contacts with the QUT (Jim Dale’s lab) would mean a very productive spin-off from the BIP (with the BIP 09 project on genetic transformation in view of BBMV resistance). 4 In his book “Bananas”, N. W .Simmonds very briefly describes the previously collected French Plantains in South India, but discussion at Kerala of his proposed names revealed that they can lead to confusion because of use by people speaking different local languages. Both ‘ethan’ and ’nendran’ are rather generic names, even within the same location. Do farmers distinguish sub-subgroups within French plantains? 69 BIP 04 Collection, Evaluation and Characterisation of Genetic Resources and Improved Banana Crop Plants in Vietnam Description Co-ordinates Organisation: Institute of Agricultural Genetics (AGI), Tuliem-Hanoi Collaborators: Phu Ho Fruit Research Centre Centre of Biocontrol (at the National Institute for Plant Protection) Principal Investigator: D. Do Nang Vinh Duration: 3 years, plus 9 month extension (by Contract Variation 1, signed by the PEA on March 16, 1998; no budget implications), starting March 1, 1995; ending 31 December, 1998 Objectives A. Original Objectives: 1. 2. 3. Collection of banana germplasm and evaluation for quality, productivity and disease resistance; Development of techniques for micropropagation of high quality exports varieties of banana, and techniques for elimination of virus diseases; Improvement of banana cultivars by selection and crossing of somaclonal variants. It can be seen in the subsection “Activities” that the actual objectives covered a wider research field than the above objectives (e.g. study of diseases, germplasm conservation). The wider objectives were already being applied at the start of the project and thus before the October 1995 Workshop. B. New Objectives (following Workshop in October 1995): 1. 2. To establish a molecular diagnostic capability for the major virus diseases of banana (initially for BBTV) in material used for micropropagation for growers, and for testing in vitro and in situ collections of banana germplasm; To test and evaluate methods for the eradication of BBTV from meristem and callus tissues; 70 3. 4. 5. 6. 7. To establish a diagnostic capability for Fusarium Wilt and Sigatoka; To establish Fusarium Wilt and Sigatoka screening plots in which to select resistant variants of major bananas grown in Vietnam, and screen diploid, triploid, and tetraploid cultivars available from collections elsewhere; To establish DNA fingerprinting methods and apply these to banana germplasm used in breeding programs for increasing disease resistance, field condition tolerance, and fruit quality parameters; To analyse germplasm from Vietnamese banana collections for ploidy and genotype; To double the chromosome complement of useful diploids to use as parents capable of contributing diploid gametes in breeding experiments. The reports show that the “new” objectives were added to the previous ones, and had no influence on the original list of activities. The original list was reproduced as such in subsequent reports in the form of a milestone/timetable, with a footnote: “The workplan for the project is currently undergoing change following recommendations” -i.e. by the Workshop for “new” objectives. Activities The reports do not explain the activities in a systematic way. The “Milestone/time chart” shows the following list of activities: (1) VARIETY IMPROVEMENT 3. Improving micrpropagation technologies; 4. Developing industrial propagation; 5. Callus, suspension and shoot culture for somaclonal variation induction; 6. Developing protoplast culture procedure for genetic manipulation; 7. Tetraploid-diploid test-crossing and establishing selection procedures; 8. Field competition trials of varieties; 9. Selection of desirable traits; 10. Application of RFLP for molecular mapping of important traits (10) GERMPLASM STUDY 11. Germplasm collection in North & South; 12. Establishing field collection, Haihung Province; 13. Morphological and cytological study on genotypes; 14 Classification and characterisation of germplasm; 15. Establishing in vitro collection; 16. Study on in vitro conservation by slow growth; 17. International exchange of germplasm (18) DISEASE STUDY 19. Identification and characterisation of major diseases (fungus and virus); 20. Identification and selection of resistant varieties; 21. Study of methods for virus identification and virus elimination. 71 Most of the objectives B1-B7 can be considered as activities implied in the above list, sub 19 to 21. Achievements Introduction The final report is to be received by the PEA in the near future. Meanwhile, the review of the achievements is based on the interim/annual reports. Achievements are not systematically reported, and it is impossible to review them in parallel with the planned activities; the latter also being reported in a rather random way. However, an attempt at extracting the essentials of the reported achievements proved that they have been very important, and the following attempt is made to provide a general picture of the reported facts. Germplasm Study 1. Collection: The field collection at Phu Ho presently comprises about 125 cultivars. Most of the accessions were apparently realised within the BIP project. 2. Characterisation: All the accessions have been characterised by using the standard INIBAP Descriptor List. Ploidy of all these accessions was assessed by chromosome counting. Quite a number of newly discovered species and local cultivars have the potential of becoming important at an international level for breeding purposes. 3. Evaluation: Agronomic evaluation of many accessions has been performed. Because evaluation took place in two different environments (Phu Ho for midland; Chaugiang for Red River Delta), it was possible to qualify cultivar-specific habitats. 4. In vitro collection: Duplicates of all the Phu Ho accessions are kept in vitro under slow-growth conditions. 5.Germplasm Exchange: Several accessions were introduced from the INIBAP International Transit Centre and are growing at the Phu Ho collection. Disease Study 1. Virus diseases Banana Bunchy Top Virus (BBTV) was clearly identified as the agent of the disease attacking the commercial bananas (Cavendish group). Cucumber Mosaic Virus (CMV) and Banana Streak Virus (BSV) were found in a few cases. 72 2. Fungal diseases Fusarium Wilt: This study was made possible with a 6-month training of three Vietnamese scientists in Australia (at the laboratory of Dr. Ken Pegg, PI of another BIP project). The situation has been clarified by VGC (vegetative compatibility group) analysis, and it was found among others that the isolate VGC 0123 was infecting the bananas in South Vietnam, while VGC 0124 was dominant in North Vietnam, together with VGC 0124/5 and 0125. A list of FW sensitive accessions could be compiled and it turned out that many ABBs were very susceptible. On the other hand, Race 4 does not appear to be present in Thailand, and protective measures in germplasm introduction were advised. Sigatoka: Both black and yellow Sigatoka diseases have been identified in some areas, but pressure is apparently not yet high. The diseases were more frequently observed in South Vietnam. Other Fungi: A number of other (leaf, fruit) fungal diseases (such as Freckle, leaf speckle, fruit speckle diseases and anthracnose) were identified, with various intensities depending on the regions. 3. Bacterial Wilt. The symptoms of this disease (important in some Cavendish gardens) could be clearly differentiated from those of Fusarium Wilt. The isolated pathogen was identified as a bacterium which needs further classification. 4. Disease Control Virus elimination. The ELISA test has been successfully used for BBTV, BSV and BMV indexing, and a number of accessions were found positive, but no virusinfected samples were found in the in vitro collection. Fungus control. Antibiotics were tried with apparent success on in vitro plants for controlling black Sigatoka and Fusarium Wilt, but field tests have not yet been performed. Variety Improvement 1. Micropropagation of selected material: The technique poses no problem anymore and material has been sent to local stations for further propagation. 2. Screening for resistant cultivars: Fifteen commercial cultivars (belonging to the different genomic groups AA, AAA, AAB, ABB) were tested for resistance to Fusarium oxysporum f.sp. cubense on an experimental site incubated with Race 1 pathogen. After 24 months, all of the 8 73 ABBs died or were heavily attacked, while the other groups showed no symptoms, except the AAB, which was lightly infected. 3. Pre-Breeding Work: - Cytological analysis of pollen viability was effected on a large number of accessions; High pollen sterility was observed on many diploids; A list of seedy cultivars was compiled; Based on experimental results, cultivars of potential use in crossbreeding have been identified; Test-crosses (attempted for the first time in Vietnam) confirmed the sterility of the Cavendish group, but revealed that the ‘Chuoi tay’ (Pisang Awak) did not produce any seed with 6 pollen donors (either AAs or BBs), which is intriguing since ‘Pisang awak’ is known for its incidental seed production elsewhere. Evaluation Methodology A desk study was effected on the basis of the interim and annual reports of the PI, before visiting the project from 26 to 28 December 1998. The visit was well organised by the PI. One day was spent on a 3-hour exposé of the activities and results, followed by a 4-hour discussion on the progress made and the outlook for further research. All the collaborators of the AGI participated in the meeting. Laboratories were also visited. A visit to the Phu Ho collection took another day. Results The achievements of this project should be seen in a special context. No banana research program existed previously and the PI had to rely on staff and facilities of several AGI laboratories (which had been equipped with support from the IAEA and from the Rockefeller Foundation). Therefore, a key aspect of the results is that an actual banana research program 5 has been initiated by this project, which thus had a stimulatory effect. Germplasm study The taxonomy of the Vietnamese species and cultivars is firmly established. The Descriptor Lists of all the accessions have been compiled with great care and at least two staff members are actual specialists in the matter. The number of synonyms was reduced to a minimum. The in vitro slow-growth conservation of the accessions is under full control. The product (information and material) needs now to be made available to the world in a practical way (and fruitful discussions took place on this issue). 5 Germplasm collection and characterisation benefited from supplementary support from INIBAPUNDP. 74 Disease Several staff members proved to be particularly versed in virology and/or research on fungal diseases. The achievements with respect to Fusarium Wilt are impressive. Variety Improvement A good basis for breeding work has been established, but much needs to be realised for the improvement to be productive. The lack of operational contacts with other programs in the world is evident. Conclusions The lack of a clear strategy at the outset of this project finds its expression in the multitude of objectives and activities, which actually would be more consistent with a fully-fledged Banana Research Program. Instead, the project had only a rather modest budget of $150,000. It is therefore remarkable that only Objectives A3 and B5, 7 have not been met in a desirable way. The project has consequently been carried out in a satisfactory manner, due to the high motivation and skill of the staff involved, and despite the modest support upon which it could rely. Guidance from experts abroad was only real during the 1995 Workshop, and via the training of 3 scientists in Australia. For the rest, the PI and his staff had to rely on their own capacity, which impressed the present evaluator. The real significance of the BIP 04 project is that it has created a momentum for the development of banana research in Vietnam. The institutes involved appear to be in real need of further support, and it would be deplorable if the momentum already gained were lost. 75 BIP 05 Banana Breeding in Brazil Description Co-ordinates Organisation: Brazilian Organisation for Agricultural Research (EMBRAPA-CNPMF) Cruz das Almas-BA Principal Investigator: Dr. S. O. E Silva Dr. Aristóteles Pires de Matos Duration: Three years, starting January 1996 Objectives a) Breeding of more productive varieties resistant or tolerant to the main pest disease problems in Brazil; b) Reduction of plant stature and duration of the production cycle; c) Increase in the number of fruits per bunch and in fruit size; d) Production and evaluation of 10,000 diploid hybrids and 1,000 tetraploid hybrids, resulting in the availability of 10 superior diploid and five tetraploid cultivars resistant to pests and diseases, with fruit quality and yield better than test cultivars belonging to the Prata-type; identification of five new varieties showing superior characteristics; development of techniques for in vitro fertilisation and for propagation with reduced somaclonal variation (lower than 5%). Activities 1. 2. 3. 4. 5. Production and evaluation of AA diploid hybrids; Production and evaluation of tetraploid hybrids; Evaluation of genotype reactions to Panama Disease; Evaluation of resistance to yellow/black Sigatoka; Advanced tests of promising varieties and tetraploid hybrids in different ecosystems; 6. Evaluation of banana genotypes with respect to Radopholus similis; 7. Evaluation of banana genotypes with respect to the rhizome weevil borer; 8. Chemical analysis of fruits of banana genotypes; 9. Behaviour of promising banana genotypes in widely different levels of soil fertility; 10. Use of molecular markers for monitoring somaclonal variation; 11. Chromosomal aberrations in relation to somaclonal variation; 12. In vitro fertilisation; 13. Determination of ploidy levels in banana pollen by diameter measurements. 76 Notes: • • • • • • Evaluation of triploid hybrids was added to Activity 2, after the beginning of the project. Activities 3-7 are evaluations to ensure that hybrids are screened readily before possible release. Activity 9 was not conducted as planned. Rather, it was substituted by an experiment oriented to study the effects of different levels of minor elements as a means to avoid fruit detachment in the Pionera variety. Pionera is being promoted by EMBRAPA, but it suffers the limitation of fruit detachment, which is preventing the adoption of such a variety by farmers. Activity 11 was eliminated since it was suggested that the same result could be obtained by using RAPDs. Activity 12 was cancelled. The cancellation of the work was due to the following: a) No success in the ovule development under in vitro conditions; b) Success in obtaining viable seeds from natural hybridisation (pollination) of the cultivar Maçã; c) Some success in the process of somatic hybridisation as a tool to perform crosses among genotypes that do not produce seeds through pollination. Activity 13 was eliminated at the beginning of the project since CIRAD had already found out that there was not much practical use for this type of study. Achievements 1. Production and evaluation of AA diploid hybrids During the life of the project, 55,612 hybrid seeds from different genetic combinations were produced and more than 15,000 were evaluated under field conditions. Less than 2% were selected for clonal evaluation. In addition, one hundred bunches were pollinated during 1998, but are not yet ready for harvest. It is expected that 900 hybrid seeds will be generated from these pollinations. Currently, 368 diploid hybrids are undergoing single plant evaluation and 97 are under clonal evaluation. 2. Production and evaluation of tetraploid and triploid hybrids Tetraploids. During the life of the project, 72,217 female flowers of triploid (AAB) cultivars were pollinated, generating 979 tetraploid seeds. Under laboratory conditions, embryos of 804 seeds were rescued and cultured in vitro. Under field conditions and in both phases of single plant and clonal evaluation, one Maçã-type hybrid (YB 42-21), as well as several Prata-type hybrids, were selected. Currently, several other tetraploid hybrids are undergoing single and clonal evaluation for further selection. Fifty new seedlings are expected in the near future, from the 160 pollinations conducted during 1998. These seedlings have not yet been harvested. In addition, 77 forty tetraploid hybrids are currently under the phases of either embryo extraction/culture or hardening. The production of viable seeds and seedlings (AAAB), resulting from crossings between the cultivar Maçã (AAB) and the AA diploid Pa Musore 2 (Musa acuminata ssp. malaccensis), is one of the most relevant results of this project, since no Maçãtype tetraploid hybrid had been generated before. Around twenty such hybrids will be evaluated in coming years. The synthesis of Maçã-type tetraploid hybrids will continue. It is also worth mentioning that 300,000 plants of the tetraploid PV 0344 (Pacovan x Calcutta IV) have been multiplied for distribution. Triploids. On 14,429 female flowers of AAAB Prata-type hybrids crossed with diploid parents, 156 triploid seeds were obtained. Currently, seven triploid hybrids are under single plant evaluation, and several seedlings are under hardening to be brought to the field. The production of triploid hybrids is an activity that was born with this project. No attempt had been made before. Seven triploid hybrids are under single plant evaluation, and more than thirty additional ones are under hardening awaiting transplanting to the field. As a result of the strengthening of the diploid crossing block and the planting of more tetraploids, it is expected that the number of triploid hybrids generated will increase in the near future. 3. Evaluation of genotype reactions to Panama Disease Sixteen diploid hybrids, 15 of them generated at EMBRAPA-CNPMF and one from Honduras (SH 3263), already evaluated for reaction to black Sigatoka/yellow Sigatoka and showing superior agronomic characteristics, are undergoing evaluation for Fusarium Wilt. No external symptoms of Panama disease have been detected up to the date of this report. Internal symptoms of this disease, however, have been detected in the rhizome of one plant of SH-3263. Very severe symptoms have been observed for the reference cultivar Maçã. Considering that most of the diploids behaved as resistant to the pathogen, the experiment was installed once more in an area showing high inoculum potential. Very severe symptoms were observed on plants of the susceptible cultivar Maçã six months after planting. The following tetraploid hybrids and cultivars are undergoing evaluation: Maçã-type, YB 42-21; Prata-type, PV 42-53, PV 42-68, PV 42-81, PV 4285, PV 42-129, PV 42-143, ST 12-31, ST 42-08, SH 3640; and the cultivars FHIA-03 and Figue Pomme Naine. 4. Evaluation of resistance to yellow/black Sigatoka The 368 diploid and 31 tetraploid hybrids that are undergoing single plant evaluation are also under evaluation for resistance to yellow Sigatoka. All hybrids and/or 78 varieties selected during the conduction of this project expressed resistance to yellow Sigatoka. At the beginning of the project black Sigatoka had not been detected in Brazil. The project contemplated a study of pathogenic variability of Mycosphaerella musicola (yellow Sigatoka), in which thirty-four isolates were collected from different varieties and ecosystems. A highly significant differential interaction for the evaluated parameters was observed. These data, which suggest a high pathogenic variability within the species, were used for the Ph.D. thesis of Dr. Zilton Cordeiro, who is now back on the CNPMF staff. The project also contemplated the testing of all improved hybrids at CORBANA´s fields in Costa Rica. Now that black Sigatoka is present in Brazil, only diploid hybrids (17) are being tested at CORBANA. Ten tetraploid hybrids were sent to a black Sigatoka-affected area in Manaos to be evaluated in field conditions in 1999. 5. Advanced tests of promising varieties and tetraploid hybrids in different ecosystems A trial was planted at Cruz das Almas to evaluate the reaction to Fusarium Wilt attack of three cultivars, one tetraploid hybrid and five chemically derived Maca mutants. The results of the study of the reaction to Fusarium Wilt of nine Maca-type genotypes showed a high incidence of the fungus in all genotypes. Only the hybrid YB42-21 showed acceptable scores of Panama disease. Regarding the testing of nine cultivars in the fifteen different sites, it was found out that in most of the evaluation sites Grande Naine, FHIA-01 and FHIA-18 expressed the highest yields. It is worth noting that in the evaluation sites, black Sigatoka is still not present. In addition, the majority of the fields have irrigation systems. Regarding FHIA-01 and FHIA-18, despite their high yield, they expressed two characteristics that, according to the researchers in charge of the evaluation, would limit their commercial use: fruit detachment, and fruit taste different from the Prata-type. Other experiments: Eleven tetraploid hybrids are being tested in six farmers’ fields, but have not yet been harvested. 6. Evaluation of banana genotypes with respect to Radopholus similis Eighteen diploid hybrids, 11 tetraploids, and eight cultivars were evaluated to Meloidogyne sp. under greenhouse conditions. Another five triploid and four diploid cultivars are under evaluation in Petrolandia, State of Pernambuco. The diploids 1318-01, 1319-01 and 4223-06; the tetraploid PV42-68; and cultivars Pioneira, Prata Comum and Prata Anã did not allow a large multiplication of R. similis. The data provided for this test allowed for the selection of genotypes to conform the diploid-crossing block. The cultivars selected at Cruz das Almas were taken for further testing against Helicotylenchus sp and M. incognita to Petrolandia, State of Pernambuco. The test 79 confirmed that the cultivars were resistant to Meloidogyne sp., but had high populations of Helicotylenchus sp. However, there was no economic damage. 7. Evaluation of banana genotypes with respect to the rhizome weevil borer Four tetraploids and four diploids were evaluated under laboratory conditions. The data collected showed that the shortest cycles of the insect were observed when individuals were kept on Terra, Caipira, Lidi, Nanicão, Grande Naine and Nanica, thus suggesting that these genotypes are the most favourable to the pest’s development. Another trial related to the pest preference was installed using the diploids Tuugia, Lidi and M53, the triploid (AAB) Pacovan, triploids (AAA) Nam, Nanica, Nanicão, Caipira and Grande Naine, and the tetraploid (AAAB) Pioneira, with the Plantain Terra as the control treatment. Preliminary results showed that adults of C. sordidus expressed higher preference for the cultivar Grande Naine, while Pioneira was the least preferred cultivar. Most of the data collected indicate that there are different C. sordidus larvae multiplication rates among cultivars and also different preference levels. Up to now, these data cannot be used as a tool in selection schemes to assess resistance to the weevil borer. Therefore, the CNPMF personnel decided to start research on a biocontrol strategy using Beauveria bassiana. So far, six strains of Beauveria bassiana out of 31 have been selected as biocontrol agents of C. sordidus, which kill over 50% of the pest population in 6 to 9 days. These promising strains are under evaluation in commercial banana plantations. 8. Chemical analysis of fruits of banana genotypes Sensorial analyses were performed on two Prata Ana and two Pacovan tetraploids, and twelve more tetraploids (Prata-types) are being submitted. The sensorial analyses showed that the hybrids PA 12-03, PA 03-22, PV 03-44, and PV 03-76 had a high consumer’s preference. The results on ripening showed that cultivars Nanicão and Caipira expressed the highest firmness values and longer post-harvest longevity, while Prata Anã showed the longer shelf-life period. Fruits from Prata Anã, Nanicão and Thap Maeo showed the highest contents of soluble solids, while the lowest values were detected in Caipira and FHIA-18. Vitamin C analysis performed on 14 cultivar showed that Prata Anã, PV 03-44, FHIA-01, FHIA-18, Pioneira and the cultivar Mysore had the highest vitamin C contents, all of them higher than Gran Nain. Analysis of eleven mineral nutrients were conducted in fruit pulps of seven banana cultivars and six tetraploids. Results showed that potassium and nitrogen were the nutrients found in higher amounts in the pulp of the banana fruits, and that there were differences between genotypes, regardless of the genomic group. High amounts of potassium are found in cultivars belonging to the AAA group. 80 All studies related to fruit quality traits have been published, three in ACORBAT 1998, and the others in “Brazilian Magazine of Fruticulture”. 9. Behaviour of promising banana genotypes in widely different levels of soil fertility (Study of Fruit Detachment) Activity 9 was changed from what was originally planned. It was changed to a study of fruit detachment on the Pioneira cultivar. The experiment consisted of evaluating minor element effects on fruit detachment during two years. The soil fertility study to overcome fruit detachment by supplying different levels of minor nutrients with two different levels of N (300 and 400 kg/ha/yr), P (100 and 120 kg/ha/yr), and K (300 and 500 kg/ha/yr) did not show significant differences among treatments. Thus, it can be inferred that fruit detachment on Pioneira is genetically determined, with a low environmental effect. 10. Use of molecular markers for monitoring somaclonal variation A study to detect somaclonal variation through RAPDs in several subcultures of the diploid hybrids is currently in place. Preliminary results in samples of 10 subcultures show the presence of somaclonal variation, especially in the last subcultures. For this purpose, a set of primers has been selected because they show good potential to detect somaclonal variation, starting from the fifth subculture. Differences in fingerprinters have also been observed. It should be mentioned that detecting somaclonal variation in tetraploid and triploid hybrids exposed to tissue culture was the main goal of the study mentioned above. However, it was not possible to detect such variation by using RAPDs, as has occurred with the diploid hybrids, since it has been impossible to get primers related with somaclonal variation in tetraploid and triploid genotypes. The use of RAPDs to detect somaclonal variation in diploid genotypes, however, has allowed for the application of such techniques in other related areas: to detect variability in Sigatoka, perform molecular characterisation of germplasm, and moko disease characterisation. 11. Chromosomal aberrations in relation to somaclonal variation (dropped, see Notes sub “Activities”) Evaluation Methodology A desk study was performed based on the progress reports for June - December 6, January – June 1997, and January - June 1998, and on the Final Report. Other studies consulted were: Martins, V. et al. 1998. Evaluación de las características de la maduración poscosecha de genótipos de banano. ACORBAT Meeting, Ecuador 1998. 81 Borges, A.L. 1998. Adubacao quimic do solo sobre a producao e resistencia ao despencamento de frutos de banana “pioneira”. EMBRAPA: 65: 1-4. The project was visited on March 23 and 24, 1999, and included: • • • • Visit to EMBRAPA´s Director of CNPMF; Review of the project with Principal Investigators; Visit to the fields to observe breeding and progeny blocks and demonstration plots; Wrap-up session with project’s leaders. Results The goal of obtaining and evaluating 10,000 diploid hybrids was surpassed, since more than 55,612 hybrid seeds were obtained and more than 15,000 hybrids were evaluated under field conditions, during the lifespan of the project. The main product of this large effort is the strengthening of CNPMF´s breeding blocks, consisting now of thirty improved diploids, ten of which were in existence before the beginning of the project. The project was aimed at the production and evaluation of 1,000 tetraploid hybrids that could result in the selection of five tetraploids. This latter goal has been surpassed since the first 13 tetraploid hybrids are under field evaluation, in seventeen locations. The main objective of the project of obtaining high-yield varieties resistant or tolerant to the main pests and diseases in Brazil by conventional breeding has definitely been achieved. The tetraploid (AAAB) hybrid PV 03-44 and the Caipira cultivar, that express resistance to Fusarium Wilt, yellow and black Sigatokas, besides showing excellent agronomic characteristics, have been recommended for commercial planting in the Amazon Region. Besides the above releases, several other genotypes (diploid, tetraploid, and triploid hybrids) are in the pipeline and will be ready for recommendations in the near future. This is rather remarkable. In addition, the execution of the project allowed the breeding program at EMBRAPA/CNPMF to considerably strengthen its structure, allowing for further substantial progress in the near future. The project provided ample opportunity for scientists to obtain experience in genotype evaluation, an expertise that was lacking at the beginning of the project. Other important outputs are related to the development of screening techniques to be used in genotype evaluation. Screening techniques have been developed/adapted/adjusted for Sigatoka, weevil borer, nematode, moko disease, Fusarium Wilt, and fruit quality and post-harvest assessment. It is worth mentioning that seeds from the Maca cultivar were obtained for the first time. This is remarkable since it opens the doors for the genetic improvement of the most preferred banana cultivar in Brazil. 82 It should be pointed out that the above results, especially the massive multiplication and field evaluation of improved genotypes, are very important since black Sigatoka is now present in Brazil. Some of the results of the project have already been published or presented in scientific journals and meetings, respectively. Thirteen articles have already been published. In general, the expertise and technical preparation of the scientists involved proved to be adequate, and the excellent performance of the project is to be commended. Conclusions The project´s original expected outputs were: (a) (b) (c) Identification of 10 superior diploid hybrids; Identification of 5 new tetraploid hybrids resistant to the principal diseases and pests and with better productivity and fruit quality than their traditional equivalents; and Official launching of at least two new varieties resistant to Panama disease and to one or both Sigatoka leaf spots. These outputs have been accomplished beyond the original expectations as can be observed in the sections “Achievements” and “Evaluation: Results” of the present review. It is important to point out that with the financial support and results obtained from this project, the Banana Breeding Program at EMBRAPA/CNPMF now has a strong basis to systematically release more banana hybrids in the near future. It was observed during the evaluation of the project that the breeding program is being managed scientifically, including all the ideal components for improving a vegetatively propagated crop: single plant evaluation, clonal evaluation, early and advanced generation testing, and crossing blocks based on a set of improved diploid hybrids. The EMBRAPA Banana Breeding Program is presently associated with a private tissue culture enterprise (“CAMPO”), which has already established laboratory facilities close to the Programs offices and fields. Three hundred thousand seedlings of each Caipira and PV03-44 have been produced by this company already. Prospects. In order to better achieve or to continue to achieve the main objective, there is a need to dedicate more time and funds to obtaining larger segregating populations in order to increase the possibilities of obtaining more banana hybrids with desirable commercial traits, as was planned in this project. Indeed, since black Sigatoka is beginning to spread in Brazil, and could destroy the export banana trade, the activities undertaken in this project should be greatly enlarged. 83 The following activities need to be or will be considered in order to increase or broaden the impact of a project of this kind: a) Continue to evaluate existing genetic material (2x and 4x) as well as material emerging from the breeding activities; b) Characterise the diploid gene pool located in the crossing block as well as new acquisitions for the collection, to assess their potential use in a breeding program as sources of resistance to major pests and diseases; c) The EMBRAPA/CNPMF has about fifty non-improved natural genotypes that are being intercrossed to determine their fertility potential for breeding purposes. Their potential use as female or male parents needs to be determined; d) The project needs additional personnel for the breeding work, since this is being carried out by only one person at the moment. 84 BIP 08 Banana Biotechnology Consortium: Novel Genes for Fungal Resistance and Post-Harvest Quality Description Co-ordinates Leading Institute: Boyce Thompson Institute for Plant Research, Plants and Human Health Program, Cornell University Principal Investigator(s): Dr. Charles J. Arntzen Dr. Gregory D. May Duration: March 1, 1995 - February 28, 1997 Objectives 1. To create and make available to co-operating investigators a “molecular toolbox”6 comprised of a reliable and efficient banana genetic transformation system, including vectors for different gene introduction strategies, and genetic libraries for isolation and manipulation of genetic elements from Musa spp. to create economically useful phenotypes. 2. To identify the genes in banana which encode enzymes participating in ethylene biosynthesis, characterise the members of this gene family, which are expressed in response to fungal pathogen stress and to the onset of ripening, and create prototype transgenic bananas which have increased fungal resistance traits or enhanced fruit ripening traits as a result of modified expression of the genes or members of the gene family. Notes 1. 6 Objective (1) also appears in the BIP 10 project. For this objective, both projects formed (together with BIP 9) the so-called “Biotechnology Consortium”. Consequently, the relevant achievements in BIP 8 are to be considered in combination with those reviewed sub BIP 10 for this item. The Molecular Toolbox simply represents a collection of protocols that describe optimized procedures for the production of transgenic banana plants in detail. These protocols encompass: 1. Optimized procedure(s) for the regeneration of complete plants from embryogenic cell cultures of different banana cultivars; 2. Optimized procedure(s) for the permanent transformation (i.e. transfer and integration of agronomically relevant genes) of such embryogenic cell cultures; 3. Ready-to-use and easy-to-handle constructs for the transformation process, including vectors for the transfer and integration of the foreign genes, optimal selectable marker genes and promoters for their strong expression in the transgenic environment. 85 2. Tissue culture research became an important activity of BIP 08, when it was realised that the existing transformation method produced a number of chimeric transformed plants. The meristem tips were abandoned as transformation tissue, and replaced with success by the embryogenic cell suspension technique, which was previously established in the BIP 10 (for biolistic transformation). The “Biotechnology Consortium” again proved to be very instrumental. Some achievements in this sector do belong to the postBIP period, but are nevertheless reviewed, partly because of their key importance, but also because they prove that the BIP effort had a most stimulating effect on this research. Activities Two major areas dominated research at BTI: tissue culture and molecular biology. Tissue Culture 1. 2. 3. Development of techniques to isolate protoplasts (for transient expression assays), to transform and regenerate banana cells, and to produce somaclonal mutants; Establishment of embryogenic cell suspensions for the regeneration of nonchimeric transformants; Screening of Agrobacterium tumefaciens strains for efficient transformation of banana cells. Molecular Biology 1. 2. 3. 4. Study of gene expression during fruit ripening, specifically at different stages of pulp and peel development; Establishment of cDNA libraries from pulp, peel, leaf, and diseased leaf; Construction of selectable marker and reporter gene fusions to heterologous promoters and tests for function in transient expression assays; Besides these activities, but not forming part of the objectives, an intense research effort on the use of transgenic banana plants for the production of antigenic proteins (“edible vaccines”) involved several scientists (among them Drs. Hugh S. Mason and Tsafir S. Mor). The productivity of this group is documented by a series of publications in top journals (PNAS, Nature Medicine) and promises future breakthroughs, but is beyond the scope of this review on activities. Achievements Tissue Culture 1. A transient expression assay to test for promoter function and strength has been developed from banana pulp protoplasts, isolated from ripening fruits at 86 stage six and seven. Reporter gene constructs (e. g. GUS) with heterologous promoters (e. g. 35S promoter of Cauliflower mosaic virus). 2. Embryogenic cell suspensions from male flower tissues of Grand Naine were established, that served as substrates for Agrobacterium-mediated transformation(s) and regenerations (Dr. Joyce M. Van Eck, Nicole S. Higgs). 3. In a type of technology transfer, Dr. T. R. Ganapathi (Trombay, India) regenerated plants (AAA Grand Naine, AAB, Rasthali) from embryogenic tissues of flower buds. He employed co-cultivation of micro-cross sections of these tissues (ideally containing a single layer of cells) from Grand Naine and several lines of Rasthali with Agrobacterium tumefaciens to produce transgenic plantlets expressing a ß-glucuronidase-intron-rice and ßglucuronidase-ubiquitin-promoter fusion. Insertion of these constructs into the target genomes was verified with PCR. The time from co-cultivation to regeneration of plantlets is only four months. A chloroplast transformation protocol is presently being developed. 4. A series of Agrobacterium tumefaciens strains were tested for their potential to transform banana cells. EHA 105 proved to be the best transformer in 5 cultivars, especially Grand Naine (Dr. Greg D. May). Molecular Biology 1. cDNA libraries from fruits of different ripening stages (separately from pulp and peel), roots, leaves, and leaves challenged with Mycosphaerella fijiensis were established. Two strategies were followed: cDNAs differentially expressed during the ripening process of the banana fruit were isolated and eleven of them characterised, and cDNAs induced in infected leaves were identified. Among the eleven ripening-associated cDNAs are two metallothionein transcripts, a ß-1,3-glucanase, thaumatin-like protein and class III acidic chitinase transcripts. Northern analysis revealed a tissuespecific steady-state level of the different mRNAs, e. g. thaumatin-like protein, endochitinase and ß-1,3-glucanase transcripts are only present in pulp and peel tissue, but not in roots, corms, or leaves (Drs. S. K. Clendennen, G. D. May, J. Narvaez-Vasquez, Colombia). 2. A previously established Musa acuminata genomic library was screened with ripening-specific cDNAs, and genes encoding ACC-synthase and -oxidase as well as a polygalacturonase isolated, restriction-mapped and sequenced. Moreover, genomic clones of metallothionein and ß-1,3-glucanase genes have been characterised, and in both cases the respective promoters (up to 1500) isolated and sequenced (Dr. G. D. May). Evaluation Methodology 87 An intense desk study was performed on 9 scientific papers, the Termination Report, and two Annual Reports of the Boyce Thompson Institute for Plant Research. A total of two days was spent to become familiar with the various laboratories of the Boyce Thompson Institute and the scientists who were and are involved in BIP research, as well as some scientists who will continue research on banana improvement. Among the former category were Drs. Greg D. May, Hugh S. Mason, Tsafir S. Mor, T. R. Ganapathi, Joyce M. Van Eck, and Nicole S. Higgs; among the latter were Drs. Peter Balint-Kurti, Alice Churchill, and Peter Beetham. Basically, Dr. Greg May organised seminars and interviews for the reviewer. Results To begin with, it should be stressed that the research at Boyce Thompson Institute for Plant Research: (1) Had a good start on the basis of previous research (partly done at Texas A & M University); and (2) Was fuelled from several resources; but (3) Had only 2 years to achieve the results reported in this review. For the type of experiments necessary for the BIP, spanning from tissue culture work over molecular biology to gene technology, two years was an extremely short time. Nevertheless, major progress was made, though some of the topics of the project (e. g. promoter isolation) have only been touched on to date, or were achieved post-BIP. Molecular toolbox This toolbox, in the view of the reviewer, the centre of the whole project, comprises a collection of A. tumefaciens strains for Musa transformation (the best of which is EHA 105), a genomic library (from Grand Naine), and cDNA libraries from a variety of tissues at different developmental stages or different times after pathogen attack (here: Mycosphaerella fijiensis). Originally, the toolbox also included promoters or promoter elements with or without intron insert-ion between them and the gene, driving either a constitutive or tissuespecific gene expression in transgenic banana plants, combined with visual transformation markers, selectable markers, and reporter genes. The toolbox is now available (means: the different ingredients have been worked out), except that the different promoters are just being characterised. In any case, all the work for the production of the toolbox was very successful, despite the relatively short period of time. The toolbox itself, if fully available to other laboratories especially in developing countries, will certainly catalyse banana research in this area, since the time-consuming and costly development of the toolbox ingredients have already been done. Genes for banana improvement A series of coding sequences (as cDNAs or genomic clones) for scientifically or agronomically interesting proteins (or traits) is now available. Especially genes involved in defence reactions (encoding e. g. ß-1,3-glucanase, chitinase, thaumatin88 like protein, metallothionein, and others) can be used: (1) to characterise their structure; (2) to study their expression and its regulation; and (3) to transform susceptible banana plants to complement the deficiency. The last aspect will certainly dominate continuing work, and also interest many other laboratories, especially in banana growing (i.e. developing) countries. Since genes of ethylene biosynthesis are also available, a down-regulation approach via ACC-S-antisense technology could be used to manipulate ripening. The molecular toolbox and improvement gene aspects of the project were satisfactorily accomplished, yet there are deficiencies regarding the various promoters. Again, the short project time prevented completion of this important part, though promising work is in progress at BTI. Tissue culture work (including transformation and transient expression assays) For all the above work, reliable transformation and regeneration protocols are prerequisites. Therefore, first embryogenic cell suspension cultures were developed that reproduce rapidly. Second, a reliable regeneration protocol is at hand that should be applicable for various banana (and plantain) cultivars. Third, protoplasts can be isolated and maintained for a short period sufficient for transient expression. Fourth, a highly effective Agrobacterium tumefaciens strain has been selected for transient and permanent transformation. Fifth, a transformation protocol has been worked out, and sixth, a transient expression assay has been successfully established, using GUS reporter fused to heterologous promoters (e. g. CaMV35S, actin). The results of all this work are at least part of already published papers (e. g. BioTechnology 13: 486-492, 1995) or are being prepared for publication. Given the problems with regeneration of monocotyledonous plants generally, with regenerative potential differences between different banana cultivars, and the limited host range of A. tumefaciens (not normally including monocots), the above research now allows banana plants to be managed in different areas of tissue culture. This conclusion is confirmed by work in at least two other laboratories. The available protocols, once fully published, will certainly trigger extensive research on other banana cultivars not used in the present project, but nevertheless economically as important as the ones employed here. Conclusions Project goals Most of the project goals as listed in the CONTRACT FOR SERVICES of March 1, 1995, were reached. However, the promoter sequences to be used for various purposes and the proposed Internet “bulletin board” for information exchange could not be finished, because of limited project time. Altogether, the achievements are impressive. All these BTI activities witness the competence of the BTI researchers in banana 89 research. In conclusion, the BIP funding of research at BTI definitely catalysed progress in this somewhat neglected area. In view of the documented success, it is strongly recommended to grant funds for a longer period of time (e. g. 3-4 years), and to check all proposals in a future BIP, which I most strongly suggest, for overlaps, so that identical research at two supported institutions is a priori excluded, which was not the case in the present BIP. General performance The technical and scientific environment at the BTI was, and still is, excellent. A number of renowned plant pathologists, plant breeders and plant and fungal experts are working at Cornell University. The BTI itself is in top condition, extremely well equipped, and owns ample laboratory and greenhouse space. More importantly though, the scientists involved in the BIP, as far as I could ascertain, are exceptionally motivated and experts in their fields. As judged from the short visit, BTI is a good place for good research. Further, the efficiency is very high, as gauged by the time available and papers published (or in preparation). Also the fact, that the second PI was recently hired by NOBEL Foundation, adds to this very positive impression of the personnel at BTI. Training and technology transfer was not explicitly a requirement fixed in the Contract with the PEA. However, a fairly strong training component existed and should be appreciated, with Dr. P. Rowe (Honduras), but more so with Colombian, Mexican, Malaysian and Indian scientists. Prospects The high quality of the BIP research at BTI led to a series of spin-off activities, which are partly extending research of the BIP, partly focussing on new, but nevertheless related areas: • • • • • • Isolation, characterisation and use of banana promoter sequences for transient expression and stable transformation(s), driving tissue- or development-specific expression or elicitation of target genes in transgenic banana plants; Differentially expressed cDNA clones from ripening fruit and Mycosphaerella-elicited leaves will be further characterised and mapped on segregating populations (in co-operation with CIRAD, France, or INIBAP, whose populations will possibly become available later in 1999; Transformation and regeneration protocols for a series of other banana cultivars will be developed (e.g. for Grand Naine); Establishment of Musa BIBAC libraries (in the view of the reviewer, a central aspect of all labs working with Musa species); Non-destructive reporter gene tests will be developed (e.g. with GFP or its mutated derivates); More stress will be on Mycosphaerella fijiensis and M. musicola research per se. First, the role of fungal “secondary” metabolites in black Sigatoka disease development and the regulation of synthesis of these compounds (e. g. induction by starvation) as well as toxin identification will be dealt with more intensively. Secondly, GFP-transformed M. fijiensis and M. musicola 90 • • • • will allow study of the mechanism of infection of the host in colourful detail(s); cDNA array technology will be used to identify genes of M. fijiensis induced by plant compounds (Dr. P. Balint-Kurti, Dr. A. Churchill); Another spin-off was the detection of retrotransposons of the gypsy and copia types in the banana genome (~ 1000 copies/haploid genome), that are either specific for the A- or the B-genome. These can be used for various aspects (e. g. mapping purposes); A banana research meeting was organised by BTI for March 22-25, 1999, where all groups working on various aspects of banana/ plantain or Mycosphaerella research came together; One of the components of the Cornell University Comparative Genomics Program is a Banana Genome Initiative proposed by BTI centering around establishing an expressed sequence tag (EST) project and the physical mapping of the banana genome. 91 BIP 09 Development of Transgenic Bananas with Resistance to Banana Bunchy Top and Banana Bract Mosaic Viruses Description Co-ordinates Organisation: Queensland University of Technology (QUT) Brisbane, Australia Principal Investigator(s): Prof. Dr. J. Dale; Dr. R. Harding Duration: July 1, 1995 - June 30, 1998 Objectives 1. Development of transformation cassettes for banana bunchy top virus resistance based on current and developing knowledge for stable resistance to all strains; Characterisation of the banana bract mosaic virus genome and development of transformation cassettes for resistance; Transformation of bananas with a single resistance transgene and ultimately multiple resistance transgenes, resistance screening of transformants, and preparation of promising transgenic lines for field trials. 2. 3. Activities The activities can be grouped into three different categories: - A molecular biology part (virus characterisation, transformation cassettes, isolation of viral promoters); A transformation and regeneration part; Testing of transgenics in the greenhouse. construction of The PI additionally tried to involve the Indian Agricultural Research Institute (New Delhi) to conduct glasshouse trials with transgenic Bluggoe and Cavendish and BBMV in India, but Biosafety implications are still hampering the operation. 1. Characterisation of viral sequences Through part of the molecular toolbox consortium, QUT performed all the activities listed below with only little contact with the other partners: 92 1. 2. 3. 4. The ssDNA genomes of banana bunchy top virus (BBTV) were further characterised functionally. The genome of banana bract mosaic virus (BBMV) was further characterised. Promoter regions of the six ssDNA components of BBTV were isolated and functionally tested. The so-called BBTV6.1 promoter was more intensely characterised. Four transformation cassettes were produced using BBTV gene sequences and binary vectors, and tested in transient expression assays. 2. Transformation and regeneration of various banana cultivars 1. 2. 3. Three different transformation protocols were developed at QUT: Agrobacterium-mediated gene transfer, biolistic transfer in embryogenic cell suspensions (with the help of the KULeuven (see Project BIP 10), and transformation of embryogenic callus derived from immature male flower buds. Both Agrobacterium-mediated and biolistic gene transfers were used to transform: Bluggoe plants (reporters: GUS, sGFP); Bluggoe embryos (four different BBTV and one BBMV gene cassette); Grand Nain embryos (reporters: GUS, sGFP, two different resistance BBTV gene cassettes, and one BBMV resistance gene cassette). Embryogenic cell suspensions from male flower explants of Cavendish (Williams and Grand Nain) and Goldfinger were developed, and plants regenerated from these embryos (Douglas K. Becker). 3. Greenhouse testing of transgenic banana for tolerance towards BBTV Characterisation, multiplication and glasshouse challenging of potentially virusresistant transgenic Bluggoe and Grand Nain plants. Achievements Characterisation of viral sequences 1. The viral genomes: a) BBTV: The BBTV genome consists of six (or more) circular ssDNA components each about 1 kb in size and each containing at least one ORF (BBTV DNA-1: two ORFs). The major DNA-1 gene encodes a replication initiation protein (Rep), DNA-3 the viral coat protein (cp), and DNA-4 a movement protein. The function(s) of the other ORFs are still unknown. The intergenic (non-coding) regions of DNA 1-6 share three regions of homology: 1. the 66 - 92 nt common region, CR-M, the binding site for an endogenous ssDNA primer for complementary strand synthesis; 2. the stem-loop common region, CR-SL, a target for the replicase; and 3. a TATA-box. These results were published in a total of 6 publications. 93 b) BBMV: The BBMV genome was further characterised, and the coat protein gene isolated and sequenced. 2. The viral promoters: a) Intergenic regions of BBTV DNA-1 to DNA-6 were structurally and functionally characterised. Structure: three regions of homology (see above), and TATA-like sequences present. b) Function: The potential promoter sequences of all intergenic regions were fused to β-glucuronidase (GUS) and green fluorescent protein (sGFP) genes, and their functions tested in transient expression assays using tobacco and banana cells (Bluggoe, Grand Nain). All the promoters are active, but the different activities vary considerably in one and the same system (e. g. BBTV 4, 5 and 6 are more active than the others are), and between tobacco and banana. The promoter of BBTV DNA-6 was analysed by deletions generated by restriction or PCR, and contains Gand I-boxes in a 272 bp minimal promoter region. This promoter (BBTV 6.1) was also used for stable transformation of banana. 3. The transformation cassettes A series of transient expression and transformation cassettes was constructed. The different BBTV resistance cassettes used either: (1) the full-length replication initiation protein (rep) gene; (2) a truncated rep gene with an introduced stop codon (“mutated rep gene”); (3) a DNA-1 internal ORF; (4) a DNA-5 ORF; or (5) a DNA-5 UT ORF. The BBMV resistance cassette was based on the coat protein gene of this virus. All the viral genes were driven by the 35S promoter. The NPT II selectable marker gene was controlled by the Nos promoter. Transformation and regeneration of various cultivars a) A series of transgenic Bluggoe lines was established expressing GUS and sGFP under the control of the 35S promoter and the npt II gene driven by either the 35S promoter or the BBTV 6.1 promoter. b) A series of transgenic Bluggoe lines was established containing the various virus resistance genes (PCR positive): Construct Number of lines BBTV DNA-1 rep BBTV DNA-1 internal ORF BBTV DNA-5 ORF BBTV DNA-5 UT ORF BBMV cp 94 33 40 4 37 41 c) A series of transgenic Williams and Grand Nain lines was established expressing GUS, sGFP, BBTV DNA-1 rep, BBTV DNA-1 internal ORF, and BBMV coat protein gene under the control of the maize polyubiquitin promoter. The selective marker gene for NPT II was driven by the BBTV 6.1 promoter. PCR-positive Grand Nain lines Construct Number of lines BBTV DNA-1 internal ORF BBTV DNA-5 UT ORF BBMV cp 16 55 9 Molecular analysis of transgenic plants A fraction of the potentially transgenic bananas were analysed for transgene presence and copy number by Southern blotting hybridisation. a) b) Bluggoe: A total of 28 independent lines contained at least one, maximally seven copies of the transgene. Grand Nain: A total of 12 independent lines contained at least one, maximally 14 copies of the transgene. Performance of the transgenic plants Three transgenic plants were challenged with BBTV-carrying aphids in the greenhouse (amongst them a Cavendish line expressing BBTV-5 UT ORF, and a Bluggoe line expressing BBTV DNA-1 internal ORF). Evaluation Methodology A desk study was performed on 6 scientific publications, the two interim reports, the Termination Report and a manuscript (accepted for Plant Cell Reports). Two days were spent at the Centre for Molecular Biotechnology, School of Life Sciences, QUT. This relatively short time was a consequence of a severe influenza brought from Hong Kong, and a cold that prevented Dr. Dale from coming to work. During Day 1, Dr. Robert M. Harding and Dr. Douglas K. Becker were interviewed. They also showed me the Carseldine Campus and the transgenic bananas. Day 2 was used to speak to Dr. James L. Dale, see the laboratories and growth chambers with the transgenics, and discuss some aspects of the project with Dr. Becker’s work. Results It should be stressed, that Dr. J. Dale and his group have more than ten years of experience with BBTV, two years with BBMV, well-equipped laboratories, and several funding sources for their research. Moreover, his research in the frame of BIP has been supported for three full years. 95 1. Molecular toolbox The contribution of QUT to the molecular toolbox comprises: a) b) (c) Several transformation cassettes with genes, or gene variants, thought to confer partial or total resistance to BBTV or BBMV; A total of six different viral promoters, of which BBTV 6.1 is used in stable transformation of bananas; Integration of three different transformation protocols , developed in other laboratories of the Biotech Consortium (see BIP 8 and 10), of which Agrobacterium tumefaciens (strain LBA 4404) and micro-particle bombardment transformation was used. The components of the toolbox have been worked out in parallel with University of Hawaii, but no direct contacts were established between the two institutions. In any case, all the work for the development of the above toolbox had already began before BIP, but was continued and accelerated by BIP, and altogether is very successful. Since no construct has yet proven to confer BBTV or BBMV resistance, the perspectives remain vague. However, once a viral gene or gene variant has proven to be effective, the toolbox, if available to researchers world-wide, will trigger activities to generate various cultivars with resistance to BBTV, BBMV, and other viruses as well. 2. Transformation and regeneration of cultivars For the application of the toolbox, reliable transformation and effective regeneration protocols are needed. Therefore, various cultivars were regenerated first (cv. Bluggoe, cv. Williams and cv. Grand Nain). Second, the function of potential promoters, derived from the intergenic region of the virus, could be successfully tested in transient expression assays using GUS and sGFP reporter genes. Third, a series of transformants could be generated with various constructs and resistance cassettes, which were partially characterised in molecular terms (Southern analysis, copy number). All these achievements prove that QUT has the capacity to produce the constructs, to transform the target cells (either with particle gun or Agrobacterium), and to regenerate the transformed cells. Moreover, a regeneration scheme has been developed for Williams and Goldfinger (male flower explants, apparently an adaptation of a similar CIRAD technique). Also here, the results have already been published in part (e. g. J. Gen. Virol, 79: 2301 - 2311, 1998), or are prepared for publication. So, QUT has been adding to the success of the BIP considerably. 3. Performance of transformants in the greenhouse: The extremely small number of transgenic bananas challenged with BBTV in the glasshouse and the negative results (no resistance) represent a deficiency of the project. Certainly more transgenics could have been tested from the beginning in collaboration with the KULeuven, as a part of the Consortium activities, and it is to be expected that resistance will be found with increasing numbers of plants, but only if the general concept is correct (resistance conferred by viral proteins). 96 Conclusions Project goals Most of the project goals were reached. However, it was a priori not proven that the general strategy of expressing a viral gene in the host plant and thereby conferring virus resistance, holds also for gemini viruses, as e.g. BBTV. Therefore, BIP-funded research at QUT aimed at testing this strategy with different BBTV genes and one BBMV gene was important. So far, no resistance to BBTV (or BBMV) could be observed in transgenic plants. However, since only three plants were challenged, the results cannot be conclusive, especially since only three of a total of 65 transgenic plants were resistant to BBTV in Dr. Hu’s experiments (University of Hawaii). Nevertheless, the achievements at QUT are altogether impressive, though final proof for the resistance concept is missing. It is for this reason, that multiple resistance transgenes could not be tested either. General performance There is absolutely no doubt, that much competence is accumulated at QUT. The scientific environment is therefore excellent. The Department of the PI is extremely well equipped and organised, although the greenhouse with the transgenic plants is remote from the laboratory. The scientists at QUT, of whom I met only few, are experts in their areas and very motivated. The efficiency of the scientific work is extremely high (published papers per time unit). All the QUT achievements are a consequence of the high competence and motivation of the researchers, and it is to be expected that QUT will play a major role in developing transgenic bananas resistant to BBTV and BBMV in future. However, it is advisable in the future that more attention is given to checking all future proposals for overlaps, so that identical research at two supported institutions will be excluded (specifically, QUT and UH). Prospects Research at QUT on virus resistance in bananas started before BIP, continued with the help of BIP, and will focus on related research topics in future: - Further characterisation of the BBMV genome and isolating more genes with potential for resistance to viruses; The cloning, sequencing and elucidation of functions of the BBTV genes, especially DNA-1 internal ORF, DNA-2, -4, -5, and -6 ORFs; The “erratic rate of multiplication of transgenic lines” (habituation: no response to hormones) has to be overcome; Many more transgenic lines have to be regenerated from different bombardments, and challenged with BBTV or BBMV, respectively, in the glasshouse; 97 - Organisation of field trials in India, using transgenic plants whose resistance has been proven in glasshouse trials. 98 BIP 10 Genetic Transformation of Prototype Bananas for Black Sigatoka-Resistance Description Co-ordinates Organisation: Katholieke Universiteit Leuven (KUL) Leuven, Belgium Principal Investigator(s): Prof. Dr. R. Swennen Dr. L. Sagi Duration: November 1, 1994 - October 31, 1996 Objectives 1. To create and make available to co-operating investigators a “molecular toolbox” comprised of a reliable and efficient banana genetic transformation system, including vectors for different gene introduction strategies, with the aim to create economically useful phenotypes. 2. To identify heterologous regulatory DNA elements to drive constitutive or tissue-specific gene expression and construct efficient expression cassettes for stable transformation of bananas. Activities The activities at KUL can be broadly categorised into: 1. 2. 3. 4. Optimisation of Agrobacterium tumefaciens-mediated and biolistic gene transfer techniques; The construction of banana transformation vectors with (twin) selectable marker genes and strong promoters; Study of the efficiency of heterologous promoters in transient expression assays; The production of many independent transgenic lines of several different cultivars using the optimised techniques. 99 Achievements Gene transfer techniques 1. A systematic study on chemotaxis and attachment of A. tumefaciens, strain LBA 4404, to wounded banana tissues was performed, and the transformation efficiency tested with a GUS reporter system. Results were not conclusive during the BIP project period, but the system was improved afterwards. 2. Optimum conditions for biolistic transformations of banana embryogenic cell suspensions (ECSs) of cv. Bluggoe, cv. Williams and cv. Three Hand Planty were established. The optimised particle bombardment of ECS was superior to A. tumefaciens, when the latter was performed on meristems and allowed to produce a large number of independent transgenic lines. Banana transformation vectors For consecutive transformation steps, plant expression vectors containing (twin) selectable marker genes (geneticin and hygromycin resistances) driven by a rice actin promoter and flanked by a poly(A) signal from A. tumefaciens gene 7 were constructed. Heterologous and homologous promoters 1. Several heterologous promoters for high gene expression in bananas have been identified by fusing them with the gusA gene and introducing these fusions into target cells by the improved particle bombardment technique. Among these promoters were the maize polyubiquitin gene promoter containing the first intron, the rice Act 1 gene promoter, the recombinant Emu promoter, the duplicated CaMV 35S promoter together with the untranslated leader sequence of AMV, and a duplicated CaMV 35S promoter. The promoter performance decreased in the same order, if tested in transient expression assays with the GUS reporter. 2. For the isolation of homologous promoters, a genomic Grand Nain library was screened with polyubiquitin and actin cDNA clones, and a vectorette approach used to isolate the corresponding promoters with vague results only for the actin promoter. Probably to the same end, a leaf cDNA library was established, but not further used. Transgenic banana lines The improved particle bombardment procedure allowed the transformation of ECSs from cvs. Williams, Three Hand Planty, Bluggoe, Cardaba, Monthan, and Musa balbisiana to produce many independent transgenic lines ( as proven by transient GUS expression). Stable transformations were achieved with cv. Williams, Bluggoe and cv. Three Hand Planty. 100 Evaluation Methodology On two occasions the laboratory of the PI at KUL was visited. Additionally, an intensive desk study was performed on the Termination Report and 7 scientific papers. Also, several contacts by phone and e-mail helped the understanding of specific details. Results From the above study and personal experiences with the PI’s laboratory, it can be stated that research at KUL on bananas: (1) had already proceeded for more than a decade; (2) has accumulated a lot of expertise; (3) produced a series of pacemaking results in banana improvement; (4) was fuelled from several sources; and (5) was accomplished by a series of excellent scientists. It should be stressed, that KUL had only two years of BIP funding, but nevertheless made major progress. 1. Gene transfer techniques Monocotyledons are not natural hosts for Agrobacterium tumefaciens, so that their transformation has to be triggered (by e. g. acetosyringon or other virA effectors), or to be optimised. This has successfully been achieved with only relatively few monocots. Consequently, the trial to establish an effective Agrobacterium-based transformation protocol with banana host cells originally remained inconclusive, though the bacteria were obviously attracted by wounded banana tissues and attached. The team has improved the technique in the post-BIP period, with satisfactory results. In contrast, particle bombardment of banana tissues proved to be effective, has been optimised considerably, and been used successfully to transform a series of Musa cultivars. Though in neither case the integration of the introduced DNA, or its physical integrity, or its copy number has been shown, particle bombardment seems nevertheless to be the method of choice for banana transformation. 2. Banana transformation vectors The idea of twin selectable marker genes for consecutive transformation steps is not new, but has first been shown to be applicable to banana cells. Since such cells are sensitive to the corresponding antibiotics (geneticin, hygromycin), such dual selection vectors will certainly be very useful in future banana transformation experiments (if available to institutions not funded by BIP). 3. Heterologous and homologous promoters Since banana promoters were not available at the start of the BIP, a logical objective was to search for effective promoter sequences from other plants. From the outset of a research program it is also acceptable to test only constitutive promoters (as the ones selected here), but clearly inducible (or regulated) promoters are needed for 101 realistic banana improvement. Nevertheless, the promoters identified to work in banana cells served and will serve to optimise constructs, cassettes and expression conditions. Although it might have been advisable to give more attention to the isolation of banana promoters that are, for example, induced by pathogens, this aspect was not part of the contract with the PEA. Notwithstanding this critical remark, there is no doubt that the promoter work at KUL was very successful. 4. Transgenic banana lines The successful transformation of several cultivars with the optimised biolistic transfer technique demonstrates its potential. The scientists at KUL have already successfully transferred this technique to eight students from other laboratories, and it is to be expected that it will be used more frequently in banana research. The various aspects of the KUL project have been competently and successfully accomplished, in spite of the fact that work at KUL has only been funded for two years by BIP. Conclusions Project goals If the objectives as stated in this report are taken as guidelines, then most of the project goals were reached. However, tissue-specific gene expression was only marginally addressed, though it is highly important, for example in resistance breeding. Nevertheless, the achievements at KUL are impressive. General performance The institution of the PI has long-standing experience in banana research, and the PI himself is one of the leading experts in this field. The scientific environment at KUL is also stimulating, and links to industry are established (via Dr. B. Cammue). The scientists involved in BIP-funded research, as far as is known to the reviewer, are altogether highly motivated and active. Again, the efficiency of KUL research (publications per unit time) is very high, and compares well with BTI and QUT. There was also a strong training, technology transfer and co-operation aspect in the BIP-funded research at KUL. All these activities will certainly continue and are proof of the competence of the KUL scientists in banana research. Prospects The future research at KUL will continue in the following areas: - Improvement of the Agrobacterium-mediated gene transfer system for bananas; 102 - Control of somaclonal variation in the generative phase of newly established cell suspensions; Development of large insert genomic libraries for the isolation of agronomically interesting genes;. Isolation of tissue-specific and pathogen-inducible promoters via differential display techniques; Characterisation of the isolated polyubiquitin and actin genes to isolate their promoters. 103 BIP 11 Genetic Engineering of Ethylene Biosynthesis in Bananas Description Co-ordinates Organisation: Hongkong University of Science and Technology (HKUST) Biology Department Principal Investigator(s): Prof. Dr. S. F. Yang Dr. N. Li Dr. Yi-Min Chen Duration: August 1, 1995 - August 31, 1997. Objectives 1. 2. 3. 4. 5. Isolation of full length ACC-synthase and ACC-oxidase encoding sequences from banana genomic libraries; Northern blot analysis of expression of both genes during fruit ripening and after attack of leaves by Mycosphaerella (fijiensis, musicola), and roots by Fusarium oxysporum f.sp. cubense; Isolation and sequencing of ACC-synthase and ACC-oxidase promoters for transformation experiments; Engineering constructs of sense and antisense ACC-synthase and ACCoxidase genes driven by the above promoters; Transformation of banana plants with these constructs, and molecular (Northern analysis), biochemical (ACC-synthase/ACC-oxidase activities, levels of ACC and/or ethylene), and agronomic characterisation of the transformants (shelf life of fruits, post-harvest behaviour, and response to exogenous ethylene, response to Mycosphaerella and Fusarium attack). This project aimed at engineering bananas with reduced ethylene production and reduced disease susceptibility, but enhanced fruit quality. Activities The activities of HKUST were restricted to three areas: the isolation of ACC synthase and ACC oxidase genes from bananas (Musa acuminata cv Grand Nain); determination of their steady-state transcript levels in pulp and peel tissues, and construction of sense and anti-sense overexpression vectors for banana transformation. 104 1. 2. 3. ACC synthase/ACC oxidase genes: Genomic phage λ libraries were established from green M. acuminata cv. Grand Nain leaves, screened with RT-PCR generated cDNA clones of both genes from ripening fruits, positive clones restriction-mapped and some partially sequenced. Both ACC synthase and ACC oxidase gene expression were tested by conventional Northern blotting analysis in pulp and peel tissues, using ACC synthase and ACC oxidase gene probes. Construction of an ACC synthase / ACC oxidase sense and anti-sense gene fusion driven by a doublet 35S promoter and 3‘ flanked by a nos terminator. The construct was sent to KUL for transformation of embryogenic plantain cell suspensions. Achievements 1. Isolation and sequencing of ACC synthase and ACC oxidase genes Genomic DNA from cv. Grand Nain was partially digested with Sau 3 A, the fragments size-selected for 12 - 23 kb, ligated into an EMBL 3/Bam HI vector and packaged. The titers were medium, and insert sizes low. The primary libraries were screened with RT-PCR-generated cDNA clones encoding ACC synthase and ACC oxidase, a total of 5 positive clones detected, cloned, and restriction mapped. A total of 3 distinct ACC synthase, and 2 distinct ACC oxidase genes were finally available, one representative of each family sequenced, and parts of the flanking regions determined by primer walking. 2. Expression analysis Northern blot hybridisation was used to determine the regulation of ACS and ACO gene expression in pulp and peel tissues at various developmental stages. Both ACS 1 and ACO 1 genes were expressed in pulp at the preclimacteric mature green stage, and this basic level of both transcripts was made responsible for the basal level of ethylene production (S. F. Yang’s system 1 ethylene biosynthesis). Both genes are ethylene-inducible, and the stimulation is drastic. MA-ACO 1 transcript levels in peel and pulp tissues peaked prior to the climacteric ethylene production (S. F. Yang’s system 2 ethylene production), and were linked to the rise of ethylene synthesis during banana fruit ripening. The increase in MA-ACO 1 mRNA levels in pulp tissue of green mature fruits preceded the increase in peel tissue upon ethylene induction, which conflicts with published results, but is logical in a ripening process from inside out. 3. Gene fusion A gene fusion was constructed from MA-ACS 1 and MA-ACO 1 genes in both a sense and an anti-sense orientation driven by the 35S promoter. These constructs were sent to KUL as part of a direct co-operation in early 1998, but so far no transgenic plants (here: plantain “Three Hand Planty”) have been received by HKUST for molecular characterisation and greenhouse evaluation. 105 Evaluation Methodology A desk study was performed on 6 scientific papers, the Termination Report, and a manuscript to be submitted for publication in a scientific journal. A total of 2 days were devoted to familiarisation with the Hong Kong University of Science and Technology Campus, the Department of Biology, and the various laboratories of Dr. Ning Li. Dr. Li gave an exhaustive overview on his part in the BIP, whereas Dr. Wai M. Chung (Development Manager, Technology Transfer Centre) provided information on budgetary issues. Results It should be stressed that Dr. Li, a former student of Professor S. F. Yang (Taiwan) and still linked to him, has long-standing experience in the field of fruit ripening and ethylene. Moreover, Dr. Li had experimented with ACC synthase genes before the BIP program. Last, but not least, he has laboratories packed with sophisticated and new instruments at his disposal, but almost no human resources. His research in the BIP has only been supported for one single year, which is close to nothing in view of the objectives. As a consequence, only part of the proposed work could be completed. 1. ACC synthase and ACC oxidase genes Gene techniques nowadays are elaborate and relatively easy to adapt. Therefore the straightforward establishment of an EMBL 3 genomic library from M. acuminata cv. “Grand Nain” DNA, the screening of this bank with 32 P-labelled cDNAs from ACC-S and ACC-O genes, the restriction mapping of the isolated clones, and their sequence analysis did not meet any major difficulties. This task then was performed satisfactorily, and adds three banana ACC-S and two ACC-O genes and their sequences and internal structure to the list of corresponding genes from other plants. 2. Expression of ACC synthase and ACC oxidase genes Again, techniques for expression analysis are so advanced that Northern experiments with ACC-S and ACC-O gene probes were successfully performed. The results support the ethylene autocatalysis model (in the author’s view). At the least, ethylene induces both ACS and ACO genes in pulp, and in peel with a delay. The results of the isolation and characterisation of both genes and their expression analysis were compiled in a manuscript (submitted). 3. Gene constructs The sense and anti-sense ACS 1 and ACO 1 gene fusions, though not originally mentioned, could not be tested in a transgenic environment, since a serious mite problem at KUL delayed the regeneration of the transgenic plants. As a result, nothing can be said of the fidelity of the constructs or their function(s) in vivo. 106 Conclusions Project goals The number of project goals was forcefully reduced by limiting the financial support to only one single year. While this peculiar decision is surprising, it is nevertheless good to see that HKUST at least performed one task: the isolation of two classes of genes catalysing ethylene biosynthesis in bananas. It is the view of the reviewer that this would not have been achieved without additional support (e.g. HKUST granted about US$ 10,000 for a three-month fellowship) and some previous work. Therefore the major goals of this particular project could not be reached: transgenic banana plants with reduced ethylene production capacity and their characterisation as well as evaluation in the field (reduced disease development, and enhanced fruit quality). General performance The technical environment at HKUST is excellent. All the most modern instruments are available, and, as far as was ascertained, well maintained. There are also some basic funds available for Dr. Ling Ni’s research. However, human resources are limited (e.g. only one Ph. D. student is currently working in Dr. Ni’s lab), partly due to the high salaries at HKUST. Dr. Li has proven competence in the field of gene technology, but the time within the BIP was simply not enough to achieve more. Also, there could have been more contact with, for example, QUT (Australia). For example, the QUT group was able to regenerate not only cv. Bluggoe, but also Cavendish-type bananas, and maintains experience in transformation. Also, they would have been prepared to insert the ACS-ACO fusion into their lines. In any case, substantially more could have been achieved, if the funds had not been limited to one single year. Prospects - - - The performance of the sense and anti-sense ACS and ACO fusions in target plants and the effect(s) of their expression on fruit ripening and host resistance to pathogens await further experimentation. A major goal remains the transformation of several (regenerable) Cavendish cultivars of economical importance for Southeast Asia with genes controlling ethylene biosynthesis with the aim of increasing shelf life of fruits. The role of pathogen and wound-induced ethylene synthesis in bananas will be evaluated further. The ultimate aim of all this work will be transgenic banana plants with reduced ethylene production in fruits (and also leaves). The over-expression of CTR (constitutive triple response) gene(s) in ripening fruits and the inhibition of ethylene response with concomitant increase in post-harvest fruit quality will remain a future challenge. 107 BIP 12 Use of Biotechnology to Produce Transgenic Bananas Resistant to Banana Bunchy Top Virus Infection Description Co-ordinates Organisation: University of Hawaii, Department of Plant Pathology Principal Investigator(s): Dr. J. S. Hu Dr. R. M. Manshardt Duration: March 1, 1995 – June 30, 1998 Objectives 1. 2. 3. Transformation of Cavendish bananas, cv. Williams with BBTV gene constructs, using a modified Agrobacterium tumefaciens and a biolistic transfer protocol; Molecular characterisation and challenge of transgenic plants with BBTV in the greenhouse; Field trials with identified resistant transgenic banana plants. Activities The activities fell into two broad categories: the production of BBTV-resistant (tolerant) transgenic banana plants, and their evaluation in the greenhouse. 1. Production of transgenic banana plants This activity encompassed several approaches listed below, and was based on the molecular toolbox developed by BTI and KUL: 1. 2. 3. 4. 5. Isolation and cloning of the rep and coat protein (cp) genes of BBTV isolates from Hawaii; Isolation and cloning of cp, movement protein, and rep genes from Chinese and Philippine BBTV isolates; Construction of transformation and expression vectors for the rep and cp genes; Optimisation of biolistic and Agrobacterium-mediated transformation procedures developed at Leuven (KUL) and Cornell (BTI), respectively; Optimisation of the antibiotic selection procedure for putative transformants; 108 6. 7. Transformation of Cavendish ‘Williams’ with an improved Agrobacteriummediated system developed at BTI; Screening of the transgenics in the greenhouse. 2. Evaluation of transgenic bananas for tolerance towards BBTV Achievements Transgenic banana plants 1. 2. 3. A series of constructs containing the BBTV rep gene, truncated versions, antisense-rep, and a gene encoding a 19 kDa protein, as well as a coat protein gene from banana mosaic virus (cucumber mosaic virus: CMV) have been produced. It was not disclosed what promoter(s) was (were) used to drive transcription of the transgenes. Particle gun and Agrobacterium-mediated transformation protocols were compared for their relative transformation efficiency, modified and refined. The original biolistic technique developed at KUL, though improved and optimised, was not used, because contamination occurred in the suspension cultured cells, which in turn took too long to grow. The changes in the agrotransformation technique (transformation of adventitious meristems growing on corm tissue, and induction of multiple shoots from the above meristems) made it suitable for Cavendish transformation. Optimal antibiotic concentration during recovery from transformation and subsequent regeneration was determined (10 mg/l kanamycin). Evaluation of transgenic banana plants A total of 300 putative transgenic plants were generated, of which 65 were screened for BBTV tolerance in the greenhouse. Of these, 21 plants exhibited a 2 - 3 weeks delay of symptoms, and only 3 plants showed no BBTV-specific symptoms at all. ELISA testing for BBTV was negative on these plants. Evaluation Methodology The evaluation was based exclusively on the Termination Report, direct contacts with Dr. Hu, and cross-references in Reports of other BIP partners. Results The group at the University of Hawaii was an informal part of the Biotechnology Consortium, and therefore links existed to BTI at Cornell University (transformation technology) and KULeuven (constructs, transformation technology). This cupertino certainly favoured progress, though the major goal of testing BBTV resistant transgenic banana plants in the field could not be reached in the three years of the project. 109 Transgenic banana plants: 1. Gene technology is nowadays standard, so that the production of various constructs for engineering BBTV resistance obviously was not difficult. The work also included cloning cp, mp and rep genes from Chinese and Philippine BBTV isolates, which means that one had to engineer resistance to both BBTV groups in one step: double transformation. This was regrettably not done. Therefore, the transgenic banana plants obtained in this project may only be of use in the Pacific area. Another problem, the instability of monogenic traits, has not been addressed in the relatively short time. It was not mentioned what type of promoter was used; though rapidly inducible promoters certainly would have advantages. 2. A dozen gene transfer techniques are now available, from which the scientists at UH have selected particle gun and Agrobacterium-mediated transfer. Both techniques had already been used for banana transformation at BTI and KUL, but had to be adapted for the specific needs at UH (Grand Nain, meristems). Therefore, much time was devoted to improving the original protocols. In the end, a modified version of Agrotransfer was employed. The use of cell suspension cultures from cv. Williams and cv. Grand Nain was obviously avoided, since endophytic bacteria were present. There was, however, no effort made to tackle this problem. 3. Optimisation of selection procedures as part of transformation/regeneration protocol was achieved rapidly. the whole 4. The main outcome of the project are 300 potentially transgenic banana plants (cv. Williams), of which 65 were tested under greenhouse conditions, and found partly (32%) or fully resistant (5%) to BBTV infection (vectors: aphids). It is to be expected that a number of the non-tested transgenics are false positives, especially since it was shown that not all of the tested genomes contain the BBTV gene (Southern hybridisation). The three fully resistant plants have probably been multiplied by now, and evaluation of the remaining putative transgenics for BBTV resistance may be complete. The researchers regrettably do not comment on the partly resistant plants (symptom delay by several weeks), because partial resistance may be the preferred and durable type of virus tolerance. As a whole, the construction of various gene fusions, their transfer by Agrobacterium vectors into target plants, their regeneration on selective media, and testing a fraction of the obtained transgenic plants in the greenhouse were accomplished. Though the resistant positives are few, the general approach of engineering resistance to ssDNA viruses such as BBTV in bananas is now promising. Nevertheless, questions remain: Did the experiments with an anti-sense construct fail? Is a single gene-based resistance durable enough, and if not, why was there no effort made to combine several genes? It is also disappointing that no real field tests were possible in the relatively short time of three years. 110 The availability of virus resistance gene constructs, conferring partial or complete resistance to BBTV, and the whole technology of transformation and regeneration of Cavendish “Williams” bananas has certainly opened this area to many banana researchers world-wide. Conclusions Project goals If one reduces the goals of this project to produce transgenic bananas with tolerance to BBTV by inserting viral sense or anti-sense genes into the host’s genome, then the project goals have been reached. Given the difficulties encountered with establishment of cell suspension cultures, the time spent on improvement or adaptation of already existing transformation and regeneration techniques, and the problems with the monocot banana in particular, the results are satisfying, and certainly will stimulate further research. On the other hand, there remain deficiencies, especially the missing double or triple transformations with different viral genes, the transformation with a combination of genes from the two BBTV groups, and the lack of field-testing. General performance The group at University of Hawaii co-operated intensely with BTI and KUL, but apparently by long-distance cupertino only (except for the project meetings). A laboratory training session, e.g. in cell suspension culture at KUL, or in Agrobacterium-mediated transformation at BTI would certainly have been catalytic. Nevertheless, the work at UH demonstrates competence in different fields ranging from gene technology over tissue culture to transformation and regeneration methodology. Prospects Basically, three spin-off activities are planned: • • • First, the remaining putative transgenic plants have to be tested for BBTV resistance in the greenhouse, the promising candidates to be propagated and field tests (in different environments) to be conducted. The resistant plants (including the partially resistant ones) have to be molecularly characterised (e. g. Southern and Western analysis, stability of the transgenes) and the heritability of the trait to be confirmed. Second, resistance to other viruses (e. g. banana streak badna virus, BSV, and cucumber mosaic virus, CMV) can and will be engineered in several co-operative projects. Third, continuing collaborations with the BIP partners BTI and KUL, as well as a Philippine group and Chinese groups, will focus on producing BBTV-resistant bananas for world-wide use (and possibly distribution). 111 BIP 14 Elimination of Banana Streak Badnavirus (BSV) from Improved Musa Germplasm and Related Studies on Transmission and Host Plant/Virus/Vector Interactions Description Co-ordinates Leading Institute: IITA Principal Investigators: G. Dahal J. d’A. Hughes G. Thottappilly J. H. Crouch Project Virologist7 Virologist Biotechnologist/Virologist Biotechnologist IITA IITA IITA IITA Collaborating Institutions8: Several IITA Scientists John Innes Centre, UK; QDPI, Australia; University of Minnesota, USA; KULeuven, Belgium; Cocoa Research Institute, Ghana; Museum National d’Histoire Naturelle, France. Duration: 3 years (with 3 months no-cost extension), starting Nov 1, 1994 Objectives 1. 2. 3. To develop reliable therapeutic methods to eliminate BSV from Musa germplasm; To determine if BSV is transmitted through pollen, embryo, endosperm; To study host plant/virus/vector relationships. Modification: In the course of the project, Objective 1 was adapted to the discovery of the episomal versus genomic nature of the virus: 1. 7 8 Identify Musa plant tissues, which are free of both genomic and episomal BSV and consequently produce ‘virus-free’ Musa germplasm. Recruited for this project No budget implications. Linking operations funded by other instances. 112 Activities Initial formulation: 1. Develop virus-free materials; 2. Develop and validate indexing method; 3. Transmission studies; 4. Vector identity studies; 5. Symptom expression and systemicity studies; 6. Screen for resistance. Definitive formulation (without budget implications): 1. Development of protocols for sensitive detection of BSV in Musa tissue (Objective 1). 2. Development of a protocol for virus elimination through meristem dissection, chemotherapy, thermotherapy, or a combination of all three (Objective 1). 3. Transmission of BSV through in vivo seed germination and in vitro embryo rescue (Objective 2). 4. Laboratory and field-tests to determine the identity of BSV vector(s) (Objective 2). 5. Field and laboratory experiments to study vector transmission in relation to natural dissemination through vegetative material (Objective 2). 6. Field and laboratory experiments to determine the effect of environmental factors 7. Identification of genotypes with tolerance or resistance to BSV, especially temperature and host genotype on virus systemicity, persistence, and expression of symptoms (Objective 3). The reformulation reflects the progress made during the implementation of this project, as well as the care taken by the PIs to optimise it. Achievements The PIs always took care to report extensively on each activity. This facilitated the understanding of the achievements to a great degree. 1. BSV detection Considerable efforts were made to perform this activity in a successful way. Collaboration of the University of Minnesota (Prof. Ben Lockhart) played a crucial role in this process. The basic reason for this painstaking work is the elusive character of this Badnavirus, which at one time can be integrated in the genomes of a Musa plant (and be inoffensive), and become episomal (and thus harmful) at another time, on the same plant. The said elusive character of the virus misled the researchers for some time, in that apparently well functioning detection techniques afterwards turned out to be unreliable. 113 Eventually, the following (briefly sketched) battery of techniques, applied in a cascade way for precise detection, was found to be quite efficient: - A first ELISA (enzyme-linked immunoabsorbent assay) using sap samples; Negatively tested plantlets to undergo a second ELISA test (using both sap and partially purified virus preparations); Negatively tested plants to be tested by ISEM (immunoabsorbent electron microscopy) (same type of material); Negatively tested plantlets to be tested by IC-PCR (immuno-capture polymerase chain reaction) (same type of material). The entire procedure takes at least 6 months (visual inspections included), and has been internationally accepted as a reliable Indexing Technique for BSV. 2. Virus elimination The elusive nature of BSV discovered in the course of the project explains why not one of the techniques tried out under this activity could eliminate the virus. Even the plants generated from embryogenic cell suspensions (provided by the KULeuven laboratory) tested positive. 3. Transmission through in vivo seed germination and in vitro embryo rescue The studies (600 plants from 12,500 seeds in vivo; 1750 plantlets from 18,000 embryos in vitro) suggest that BSV is transmitted through both in vivo and in vitro germination. Genomic DNA extracts from flower stigmas with pollen from BSV symptomatic plants also indicated (PCR amplification) the presence of BSV. The PIs also checked what happens in the case of propagation through suckers: the majority of suckers tested positive for BSV by ISEM within a few months. 4. and 5. Vector transmission studies Among the different mealybug populations found active in bananas, even the two most aggressive ones (Dysmococcus brevipes and Ferrisia virgata) could not be proved to be vectors of BSV. In the field, there was no evidence of radial or jump spread that is usually associated with mealybug vectored viruses. Attempts at mechanical transmission were unsuccessful. Tissue culture material exhibited more severe symptoms when compared with suckers (in vitro stress). The PIs conclude: “Circumstantial evidence suggests that there is very little, if any, vector transmission or radial spread and that the outbreaks are often associated with plants that are, or were, stressed (e.g. drought stress).” 114 6. Effect of environmental factors A scoring system on a 0-3 scale for the symptom severity was devised and different types and combinations of symptoms grouped into five major categories. The system is likely to be adopted internationally. By using these systems, the complex dynamics in relation to the environment were revealed (e.g. different symptom-combinations in the rainy versus dry season). Symptom expression varied within the same plant and even on the same leaf. This suggests that the virus may be incompletely or partially systemic, and the different concentrations of BSV found in different plant parts (TAS-ELISA) seem to confirm this. Yield loss attributable to BSV depended on time of infection and crop cycle. A significant cultivar x environment interaction was found. 7. Resistance/tolerance to BSV A relatively cool environment with low rainfall and potential evapotranspiration appeared to be favourable for BSV and enabled good screening of cultivars for resistance or tolerance. The important field collection at IITA was thus screened. A differentiation at genomic level is tempting but not clear: - The observed M.acuminata group (AA and AAA ) did not show symptoms; The ABBs and the AAB French Plantains seldom showed BSV-like symptoms; But the AAB False Horn Plantain showed the symptoms in most of the environments; All plantain hybrids showed the symptoms, although significant differences among the hybrids were observed. It was tentatively concluded that: - Transgressive segregation (->epistasis) may be the main non-additive gene interactive system controlling the response to virus(es) in plantain hybrids; The overall variation in a field would be determined by a combination of additive clonal and environmental effects. Evaluation Methodology A desk study was performed on the basis of the Termination Report. Visits to the IITA laboratories involved at Ibadan and Onne, discussions with the virologists Dr. G. Dahal and Dr. J. d’A.Hughes, who prepared the Terminal Report in place of Dr. J. Crouch (who left the Institute in the course of the project), as well field visits (August ‘98) allowed familiarisation with details of the project. 115 Results All of the seven activities were pertinent to the concept of the project and led to major breakthroughs in knowledge on this virus which is presently menacing banana cultivation in at least two continents, Africa and America. As a consequence of the project, whereby the PIs worked in collaboration with several specialists and laboratories over the world, it became clear that: - The “nuclear phase” of the virus is responsible for its elusive character; A reliable BSV detection method was nonetheless developed; The virus cannot be eliminated by any of the available techniques. Vector transmission is unlikely; Rather precise environmental factors (weather) have been found to be triggering factors for BSV outbreaks in the field; Conventional breeding for BSV-resistance is possible, although considerably more knowledge on genetic aspects are required; Genetic manipulation (in combination with breeding) may again be the more useful method for obtaining resistance/tolerance. Conclusions The three objectives have been met satisfactorily. The entire initial domain for investigation of the disease (which did not become important before the eighties) has been covered, and the results are of basic importance for the future. The international scientific community is now prepared for well-focussed further research. While the project formed an integral part of IITA’s program on BSV research, it undoubtedly played the motor-role in the progress made so far. No less than 22 papers in international journals were one of its product outputs. One spin-off of the project is the continued close collaboration of several Institutes and Universities (John Innes Centre, University of Minnesota, KULeuven) which have now attracted support from donors, such as the Gatsby Foundation, in an attempt to achieve resistance/tolerance through genetic manipulation. 116 BIP 15 Variability and Relationships within Populations of Fusarium Oxysporum F. SP. Cubense from its Centre or Origin Description Co-ordinates Leading Institute: Plant Protection Unit, Dept. of Primary Industries, Indooroopilly, Australia Principal Investigator: Ken G. Pegg Collaborating Institutes: Queensland Department of Primary Industries (QDPI) with Dr. N. Y. Moore, Dr, M. K. Smith, Mr P. W. Langdon, Mr R. D. Davis CRC for Tropical Plant Pathology, University of Queensland Duration: 3 years (1 March, 1995 to 28 February, 1998) Objectives Original Objectives: The project was to be implemented in very close collaboration with the University of Florida (Dr. Randy Ploetz). Both institutes were in fact to carry out two similar projects (BIP15 and BIP16) with the same objectives: 1. To collect isolates of Fusarium oxysporum f.sp. cubense (Foc) in the Asia/Pacific Region to assess diversity in the centres of origin for Musa acuminata and Musa balbisiana. 2. To assess relationships within and among these populations of Foc (and populations from elsewhere) using phenetic and genetic procedures. 3. To develop early screening methods whereby host:pathogen responses could be quickly and reliably ascertained in a quarantine environment. 4. To study the influence of host genotype on VCG (vegetative compatibility groups) composition in the field. 5. To evaluate diploid bananas being used in breeding programs, as well as their hybrids, for resistance to Fusarium Wilt. 117 Final Objectives: Because the Project Executing Agency decided to drop proposal BIP16, the following objectives were agreed upon: 1. To collect isolates of Foc in the Asia/Pacific region in order to assess phenetic and genetic variability in the pathogen. Isolates will be assessed using production of volatile aldehydes, vegetative compatibility and arbitrarily primed PCR (RAPD and DAF techniques) to generate DNA fingerprints. A PCR-based DNA detection and diagnostic system will be developed using the SCARs (Sequence Characterised Amplified Regions) approach. 2. To assess the reliability of using plants derived from tissue culture as opposed to conventional planting material (bits/suckers) when evaluating banana genotypes in the field for resistance or susceptibility to Foc. 3. To evaluate breeding diploids and hybrids, from international breeding programs, in the field against Australian and Asian populations of Foc. This change meant: - - - The original Objectives (3) and (4) were dropped (which caused concern for the PI because (3) would have optimised a handy technique of general importance); Objectives (1) and (2) were merged, with the essential difference that the RFLP method (in which Dr. Ploetz’s group was specialised) was no longer included. The decision had important implications for the classification of the isolates at the end of the project (as will be shown in “Results”). The “new” Objective (1) is spelled out in a number of activities; A genuinely new Objective (2) was inserted, dealing with the interesting problem of the “non-hardy” tissue culture plants (which frequently show susceptibility to diseases and pests for plants that become resistant in later stages of their development). Surprisingly, a fourth objective appears in the Completion Report only: 4. To increase the disease resistance and hardiness of plantlets derived from tissue culture using mycorrhizae, fungal endophytes and rhizosphere bacteria. Activities It was apparently not found opportune by the PI to spell out the 3 objectives in a number of relevant activities. The present reviewers admit that such would be very difficult with objectives that are, in fact, activities themselves. The effort would lead to the description of the very techniques, repeatedly used in close sequences within each objective for the “waves” of collected Foc isolates and of assayed Musa genotypes. Activities for Objective (1) are mentioned within the paragraph on Objectives (see above). 118 Discussion of the timetable produced by the PI is unproductive in the present context because all of the objectives have been met. The PI preferred to highlight in each of the interim reports the progress made for each of the objectives, which makes reading quite captivating. Achievements Achievements are reviewed here by objective for the reason explained in “Activities” above. 1. Understanding the variability in Foc within the primary centre of Musa diversity Considerable progress was achieved in this domain, as the interim reports and the Completion Report clearly show. This is not surprising since the PI is an authority of world-wide reputation in research on Foc. The most salient achievements were: - - - - - - The host Institute of the PI is now depositor of the most important Foc collection in the world, with more than 2000 isolates from the centre of origin as well as from various parts of the world (leaving quite some cumbersome work to screen this huge collection for the duplicates); 77 Foc specimens were collected during an expedition to Indonesia (1996); About 400 isolates were subject to a DNA fingerprint study of which the results confirmed the validity of the VCG classification; Nine clonal lineages were established on the basis of DNA fingerprinting (and referred to since as DNA fingerprint groups (DFGs). Five of these DFGs contained numerous closely related VCGs and genotypes, while the remaining four DFGs (or lineages) each contained only a single VCG or genotype; The 9 lineages were then divided into two major groups, by means of phenetic analysis. These two groups seem to have a phylogenetic/geographical significance (i.e. two areas of Foc diversification within the primary centre of Musa diversity). Tropical race 4 populations which are Cavendish competent in tropical regions were restricted to south-east Asia and far northern Australia; The hypothesis of co-evolution (Foc-Musa) within the primary centre of origin of Musa was confirmed and can be accepted as a reliable theory. The absence of Foc in PNG (at least in the traditional context), where the susceptible species Musa schizocarpa and M.acuminata ssp banksii are endemic, is of great practical importance in this respect (quarantine rationale); Strain-specific PCR primers for development of PCR-based detection systems for Foc were identified and a DNA library (banding patterns) established; In the study on rDNA, the internal transcribed spacer (ITS) region and the intergenic spacer region (IGS) of the rDNA genes were characterised and 119 - sequenced. Genetic variation could be identified among different races and VCGs based on the IGS haplotype; A phylogeny was eventually constructed for the genotypes of Foc based on DNA fingerprinting and IGS sequence information. 2. The value of tissue culture plantlets By contrast with Objective (1), achievements for this objective within the BIP project have not been very impressive, beyond the confirmation that such plantlets are more susceptible to Fusarium Wilt than conventional material such as suckers and “bits”. Attached to the Completion Report is a copy of a paper reporting a strong indication that this greater susceptibility cannot be the consequence of any lower photosynthetic activity nor sink-source dynamics in photosynthates. It thus becomes clear (to the present reviewers at least) that the cause of the susceptibility should be sought either in the structural (histological) elements of these plants (lack of silicate compounds?) or in a lack of beneficial associations with endophytes. But the paper is the product of work carried out with non-BIP support (Queensland Banana Industry Protection Board; and the Horticultural Research and Development Corporation). This support is acknowledged in the paper. 3. Screening of breeding material for resistance to Foc A large number of edible diploids and improved diploids and triploids were introduced mostly from CIRAD (Guadeloupe) and FHIA, and a couple of hybrids from IITA. The operation of testing all that material for resistance to Foc proved to be a very time-consuming one. Strict quarantine regulations, virus indexing, propagation, careful selection of fields infected with the critical VCGs, eventually led to the bulk of the material being planted not before 1997. Since reliable testing of resistance versus susceptibility to Foc cannot be achieved within the first two years after planting (two cycles are needed), the majority of the results will be obtained this year (1999),and thus no longer can be reported within the BIP framework. The operation extends to countries other than Australia and the first “waves” of the material were already planted in Indonesia and Malaysia. A finding of fundamental importance to breeding was the discovery in Indonesia of a population of Musa acuminata ssp malaccensis, which segregated for resistance to the subtropical race 4. While several acuminata subspecies are resistant, this is the first opportunity to study the genetics of resistance in a precise genetic context and perhaps to understand the molecular basis of it. 120 Evaluation Methodology A desk study was performed on the basis of the Completion Report as well as of all the interim reports. Results and Conclusions Because the activities coincide with the objectives, the present review embraces both aspects. Objective 1 has been amply met, as the above long list of achievements demonstrates. The project enabled the establishment of a solid basis for all further research on the origin, the diversification and the significance of the pathogen. While the results of the massive DNA fingerprinting work do provide a firm rationale for further molecular research, it is unfortunate that this endeavour could not be backed by the RFLP analysis carried out separately at Florida University (due to the BIP16 proposal being dropped): the conjunction would by now have produced a great theory of international authority in this domain. One element of the objective has not yet been achieved: the detection kit for fast resistance-to-Foc screening. But a convenient PCR primer was found and the kit is likely to be ready in the near future. Objective 2 was but partly met, and some reported results appear to be the product of non-BIP support. Much work still lies ‘down the road’ and the impromptu addition by the end of this project of a fourth objective dealing exactly with that research is significant in the context. Objective 3 will be met in a satisfactory way during the post-BIP period. Even the reported first results are useful. The relatively short time-frame of 3 years may have been insufficient to accomplish such a task, given its time-consuming nature. With respect to Objective 1, the prospects for further progress are manifest. Many operational links with institutes over the world were established and are likely to remain active. The research involved now has a rather routine character (further collection of isolates, continued DNA fingerprinting, interpretation and completion/modification of the established classification). The BIP project attracted ample co-financing in Australia, and part of this has a good chance of success in the light of very promising achievements to date. Still, international interest should be maintained because the Asian mainland (beyond Malaysia, Thailand and Vietnam) remains a vast area of investigation for the study of Foc. 121 BIP 17 Origin and Distribution of Fungicide-Resistant Strains of Mycosphaerella Fijiensis in Banana Plantations in Costa Rica Description Co-ordinates Organisation: Corporación Bananera Nacional de Costa Rica (CORBANA) Principal Investigator: Ronald Romero, Project Leader Mauricio Guzmán-Quesada Alexander Jiménez-Loria Randall Vargas Note: Eng. M. Guzmán-Quesada became PI after the departure of Dr. R. Romero and managed the project in a very satisfactory way. Duration: Three years, starting March 1, 1995 Objectives 1. To study changes in the dynamics of isolates of Mycosphaerella fijiensis resistant to benomyl and propiconazole in commercial banana fields subjected to different fungicide regimes; To determine the production of conidia of benomyl and propiconazoleresistant or sensitive isolates in culture. Similarly, to study conidia production of isolates that are resistant to both benomyl and propiconazole; To study the parasitic fitness (aggressiveness) of benomyl and propiconazoleresistant isolates of Mycosphaerella fijiensis on young banana plants in the greenhouse. In the same way, study the aggressiveness of isolates that are both benomyl and propiconazole-resistant; To study the ability of fungicide-resistant isolates of Mycosphaerella fijiensis to produce conidia and ascospores, and to study the inheritance of the character. 2. 3. 4. Note: - Objectives 2 and 3 are hereafter combined due to their similarity and convenience in reporting. They are reviewed under Activity/Achievement (2), while Objective (4) is dealt with under Activity/Achievement (3). 122 Activities 1.1 Testing the effect of a propiconazole-free period on the changes in sensitivity of Mycosphaerella fijiensis to this fungicide The original plan which contemplated the use of four commercial farms was accomplished. Three treatments were considered: DMI-free period of six months, DMI-free period of three months, and no DMI-free period. The sensitivity to propiconazole was determined by calculating the effective dose that caused 50% inhibition of colony growth (EC50), by regressing growth against the logarithm of the fungicide concentration as well as by using the germ tube inhibition method. 1.2 Determining the effect of a mixture of propiconazole with mancozeb on the control of the disease, and on the changes of Mycosphaerella fijiensis sensitivity to propiconazole. This experiment was conducted on commercial farms as well. 2. Evaluation of isolates of Mycosphaerella fijiensis that are resistant to both benomyl and propiconazole for their ability to produce conidia in culture Three runs of this experiment were conducted, each using a different set of isolates of M. fijiensis, with a range of sensitivities to the fungicide. The number of conidia produced for each isolate was determined by transferring the individual colonies to a test tube and counting the number of conidia in a hemocytometer. A related activity was to determine whether the acquisition of resistance to fungicides affects the ability of these strains to sporulate, which is an important component of fitness. 3. Inter-crossing isolates with different degrees of sensitivity and testing their progenies (monoascosporic isolates) for their sensitivity to propiconazole. Achievements Effect of fungicide regimes 1. No significant differences in the EC50 of populations (due to a DMI-free period) were observed by either method of sensitivity test used. The population of less sensitive isolates was not affected by any duration of a triazole-free period carried out for two consecutive years. The treatments did not affect the level of practical control of the disease, since the Infection Index and all other variables related to disease did not show any differences among treatments. 123 Consequently, according to the results of this activity, the common practice of using a triazole-free period as a way of controlling black Sigatoka seems to have no important implications on the dynamics of the fungicide-resistant populations, since a DMI or a triazole-free period of six months did not show any effect. However, this practice still has importance in regions where M. fijiensis has appeared just recently or in regions with no shift or where only a small shift has been observed. In fact, CORBANA has been promoting it as an important factor to consider in any black Sigatoka control strategy. 2. The most interesting result is that with doses of fungicide concentration of 0.05 and 0.1 µgml of each fungicide tested, the mixtures of propiconazole and tridemorph (each of them with the same concentration as the fungicide alone) consistently gave better control than each of its components separately. This identified synergism would allow safe use of important fungicides, which should be of great value for any integrated pest management program, as well as cost-effective. The use of fungicide mixtures at low concentrations should be studied further to determine the potential in any integrated pest management program. Parasitic fitness and ability to produce conidia The results indicate that there is no loss of parasitic fitness associated with the acquisition of resistance genes, since no relationship was found between the level of resistance to propiconazole and the ability of the isolate to produce conidia in culture. Significant variation in conidia production among isolates was observed, and is possibly due to the existing natural genetic variation in the population. In general, isolates that produced the highest number of conidia were from the fungicide treated areas, and they possessed intermediate levels of resistance to propiconazole, but isolates from the wild type population did not have any isolate producing high numbers of conidia. These preliminary results may help explain the high frequency of individuals with resistance to propiconazole found in the fungus population in several commercial banana plantations. Although the DMI-fungicides are facing a widespread resistance in M. fijiensis, they are still used because their performance has not completely failed. Possibilities to continue using DMI-fungicides against black Sigatoka in bananas would strongly depend on the effectiveness of strategies aimed at reducing the frequency of the resistant strains. Inter-crossing ability of M. fijiensis isolates It was proven that crosses between different M. fijiensis isolates are possible. Crosses of triazole-sensitive versus resistant isolates generated progenies with low and intermediate levels of sensitivity, and only a few isolates that appeared to accumulate higher levels of resistance than the original parents. 124 Crosses of resistant x resistant, sensitive x sensitive, and intermediate x intermediate isolates gave progenies with resistant, sensitive, and intermediate levels of sensitivity, respectively. Conidia were obtained from any of the crosses involving different isolates of M. fijiensis. Monoascosporic isolates were also obtained from all possible cross combinations. In general, most crosses in which a resistant isolate was used against an intermediate or a sensitive one, the progeny rendered intermediate levels of resistance, some with high resistance. Similar results were observed when isolates with intermediate levels of sensitivity were crossed between them and with sensitive isolates. Evaluation Methodology Desk Study: A desk study was performed based on the Project´s Original Contract, the Annual Reports for March 1995 to March 1996, March 1996 to December 1996, January 1997 to December 1997, a Six-Month Progress Report of January 1998 to June 1998, and on the Final Report. Field Visit: The project was visited on February 12, 1999 and included: - Official visit to Corbana´s Research Director at Guapiles, Costa Rica; Formal presentation of results by the project´s leader; Discussion with personnel in charge of project; Visit to Pest Control Laboratory facilities. Results All the project objectives were met and their activities to achieve them were conducted effectively. Crosses of isolates with different degrees of sensitivity to propiconazole were performed on banana leaves. Isolates were obtained from a work collection existing at CORBANA, with known EC50 to this fungicide. Isolates with high, intermediate and low EC50 values were crossed in all possible combinations, except selfing. Then, monoscosporic isolates were obtained from these crosses (progenies) and their sensitivity to propiconazole tested. Regarding fitness, a great deal of information has resulted from this project and its use should be promoted to design environmentally sound strategies to control black Sigatoka. In consequence, this will result in the reduction and safer use of fungicides as well as less costly applications. It should be taken into consideration that this can only be possible if less sensitive propiconazole populations can be obtained. It is worth mentioning that fungicide-free periods are no longer the only methods to reduce resistance. 125 Conclusions CORBANA’s principal investigators merit special recognition for the excellent work achieved in studying M. fijiensis under strictly controlled growth conditions and for having been able to cross different strains of the fungus and obtain monioscosporic isolates from the various combinations (RxR; SxS; IxI and SxR). The technical and scientific environment at CORBANA as well as the expertise of the principal investigators was adequate to conduct the project. The co-operation provided by the multinational companies Dole, Del Monte and Geest Caribbean (now called CARIBANA) for the conduct of the experiment is to be commended. During the conduct of this project, inter-institutional co-operation was present. For example, CORBANA provided isolates of M. fijiensis to CIRAD to study genetic variability and to KUL for testing transgenic plants. Prospects According to the desk study and the field visit, the following studies should be considered to broaden the results of this project: 1. 2. 3. 4. 5. Research on the resistance of M. fijiensis to triazoles needs to be undertaken further. From our point of view, we believe that the problem of increased triazole resistance still exists and studies to decipher its origin should be strengthened and funded. Synergism between tridemorph and propiconazole, which may be important to improve the control of the disease, should be studied further. This BIP project did not have the necessary time to devote to this aspect of the research, however, it represents an avenue for further research to improve the control of the pathogen in populations with a shift in sensitivity to triazoles. Evaluate all possible alternatives for black Sigatoka control within an Integrated Pest Management strategy, including: mixtures of fungicides; different levels of fungicides; interactions of fungicide by environment; DMIfree periods, fungicide rotations. Related to number 3, it will be important to study the economic implications of the use of a more environmentally-based control strategy in contrast to the chemically-oriented one. Parameters studied could be the possible economical losses at the beginning of the strategy implementation, the time needed to recover control of the disease once the strategy has been in place, and the actual cost of implementing such a strategy. “Sensibility Gene Pools”—this refers to the relative value of having pools of sensitive genes to fungicides within commercial plantations to allow sexual recombination with resistant strains in order to bring down the sensibility value of the population to fungicides. The effective percentage of “non-treated areas” nearby or within commercial plantations should be determined in order to allow an efficient recombination of susceptible and resistant isolates of M. fijiensis. 126 6. Studies to measure the movement/migration of isolates and their recombination percentage with neighbouring populations. Genetic and inheritance studies of M. fijiensis, such as those already initiated by CORBANA, should be encouraged in order to shed more light on the wellknown complicated case of black Sigatoka. 127 BIP 18 Study of Resistance of Bananas and Plantains to Nematodes Description Co-ordinates Organisation: Centre Régional de Recherches sur Bananiers et Plantains (CRBP) NYOMBE,CAMEROON Principal Investigator: Dr. Roger Fogain Duration: March 1995 - December 1998 Objectives 1. To identify sources of resistance to R. similis and P. goodeyi in the CRBP Musa germplasm in Nyombe (Cameroon). 2. To identify cultivars with lower levels of susceptibility to these nematodes. 3. To evaluate improved materials from breeding programs for resistance to nematodes. Activities 1. Collection of roots, isolation of nematodes for axenic culture; 2. Pot screening for resistance to Radopholus similis, using diploid Musa, for rapid selection of interesting materials for field studies; 3. Field screening of diploids for resistance to R. similis and P. goodeyi; 4. Pot screening of triploids for tolerance to R. similis; 5. Pot evaluation of hybrids from breeding programs (IITA, FHIA, CRBP, CIRADFHLOR). Achievements The three objectives implied the same methods/techniques which were not subject to research per se beyond the check of ‘pot versus field’ results for consistency. Screening for ‘tolerance’, i.e. non-resistant genotypes allowing for nematode colonisation without much suffering, was left out early in the course of the project when it appeared that root damage, using the root lesion index (RLI), was not a good indicator (see BIP 20 for the same difficulty ). The pot technique is rather labour and space-consuming, since each pot contains one sucker and allows the growth of normal roots for more than two months. but this cannot be considered as a handicap in tropical countries. the technique is not time-consuming, the nematode development being already measured two months after inoculation. 128 Because the mass of the inoculum (1000 individuals) is the same as in the BIP projects 19 and 20, results can be compared for the same Musa genotypes tested. Consequently, the Achievements are reviewed ‘across-objectives’ and they only deal with resistance/susceptibility. 1. Pot Experiments Radopholus similis (dominant in the lowlands, i.e. altitude less than 900m in Cameroon) Few of the 61 tested diploids (AA, AB, BB) showed good resistance, comparable to that of the control ‘Yangambi Km5’. The resistance of ‘Pisang Jari Buaya’(PJB) and the lower susceptibility of the tested BBs were confirmed. Of the diploids used as male parents in current breeding programmes, only the wild AA ‘Calcutta 4’ had lower susceptibility. The 5 accessions from Papua New Guinea and the wild M.acuminata ssp banksii (resistant to BS) were susceptible as well. Of a total of 62 triploids (inclusive of the 34 plantain cultivars previously tested and several East-African AAAs), only the ‘Ibota’-group (presumably synonyms of, or at least akin to, ‘Yangambi Km5’) proved to be resistant. No significant variation could be detected within the plantains or the East-African AAAs which are all susceptible. Equally susceptible were the 13 plantain hybrids tested (from CRBP) and the 4 CIRAD hybrids IRFA 909, 910, 911, 912. Pratylenchus goodeyi (dominant on highlands, i.e. altitude more than 900m in Cameroon) The unusually low reproduction level of nematodes (even on clones known for their susceptibility such as the plantains), did not allow for differentiation in resistance/susceptibility of the 10 accessions tested. The PI offers as explanations: (a) that the inoculum (1000 nematodes/plant) is too low in the case of P. goodeyi or/and (b) that the counting should be effected at a later time (which may be a consequence of (a). 2. Field Experiments and Pot/Field Consistency Radopholus similis The pot-evaluation indicated that 23 of about 142 screened accessions could be moderately susceptible or resistant. These were planted, together with three FHIA hybrids at two sites (Njombè and Mbalmayo; randomised blocks, five replications), together with the susceptible standards ‘Grande Naine’ (Cavendish) and ‘French Sombre’ (Plantain) and the resistant one ‘Yangambi Km5’. Nematodes were counted at 3, 6, 9 and 12 months. The 6-month data proved to be reliable at both sites as they were confirmed by the later observations (classical statistics with LSD and Confidence Interval 95%). A positive correlation (R2= 0.650) was found between the data from the pot trials and those from the field experiments. The moderately susceptible/resistant status of 129 these accessions was thus confirmed in general. The PI also concludes that “the results of pot studies are valid and consequently the early screening method used in this study can be recommended”, but see subsection on ‘Evaluation’. Pratylenchus goodeyi The same accessions were planted on an infested field at Bova (highland). The nematode counting at 3 months appeared to have been destructive as most of the plants died because of the excavation of more than 50% of their total root mass, and the experiment was stopped. First indications pointed to the BB and ‘Calcutta 4’ having the same moderate susceptibility/resistance as to R. similis, but PJB being susceptible. Evaluation Method A desk study was based mainly on the very comprehensive Completion Report (with 17 tables). During a visit to the CRBP (August ’98) interim results were briefly discussed with the Principal Investigator. Results This project was focussed on screening of Musa material for resistance to the nematodes R. similis and P. goodeyi, and not on screening methodology/techniques (as with Projects BIP19 and 20). The experiments were carefully carried out and the data duly analysed, so that their results are of great importance. Remarkable progress was thus achieved in the knowledge on resistance/susceptibility to P. similis. It becomes clear that convenient sources for resistance breeding are scarce. The resistant triploid Yangambi Km5 can hardly be considered as a convenient source (female sterile and erratic pollen production) and the wild ‘Calcutta 4’ turns out (again, see BS work) to be the most practical source, despite its poor bunch characteristics (the other resistant wild AA ‘truncata’ and ‘malaccensis’ bear even less attractive bunches). The PI’s conclusion that pot-evaluation is a reliable fast technique for screening cannot be shared. Indeed, the positive correlation found between pot and field data does not coincide with a strict regression. The tables show for example that the hybrid ‘IRFA 903’, found to be resistant in the pot-trial (log-value of nematode number per 100g roots = 2.680 < 3 ) was quite susceptible in the field (log-value = 4.5 > 4). Less pronounced, but still significant discrepancies can be noticed for ‘Pisang Tongat’, ‘P.trimulin’, ‘Palimberry’ and even for Musa balbisiana. Discussions at early stages of the project and within the ‘Nematode Consortium’, on methodology and techniques could have been helpful, but there are no indications that these took place. 130 It is unfortunate that almost no progress could be made in the investigation on resistance/susceptibility to P. goodeyi. This lesion nematode dominates on the highlands and is a particular menace for the bananas in East-Africa. Breeding of the East-African AAA for nematode-resistance remains handicapped since no certain sources are available beyond the ‘Calcutta 4’. Conclusions Although the objectives have only partly been met (i.e. the gap in knowledge about resistance to P. goodeyi), the considerable work carried out at CRBP within the BIP project has produced results of international importance. Research on techniques for reliable fast screening (of resistant/susceptible Musa genotypes) should be pursued and preferably be conceived in collaboration with other nematode specialists within the now active Nematode Consortium. Much research on the Musa-P. goodeyi interaction remains to be achieved to the benefit of banana/plantain growers in Africa. 131 BIP 19 Identification of Durable Nematode Resistance Sources in Banana and Plantain Description Co-ordinates Organisation : Laboratory of Tropical Crop Improvement, Catholic University of Leuven (KULeuven), Belgium Principal Investigator : Dr. D. De Waele Collaborating : Associated : Dr. R. Swennen Ir. R. Stoffelen J. Reynders Duration : 1 March, 1995 - 31 December, 1998 Objectives 1. To provide the existing classical breeding programs for Musa improvement of FHIA in Honduras and of IITA in Africa (Nigeria and Uganda) with a nematological component. 2. To adapt or develop early, rapid and reliable methods for nematode resistance screening. 3. To screen banana and plantain genotypes for nematode resistance. Activities 1. 2. 3. 4. 5. Establishment and maintenance of stock cultures of Musa nematodes; Characterisation of nematode populations by reproductive fitness on carrot discs; In vitro screening experiments for nematode resistance; In vitro culture of Radopholus similis on carrot callus (inserted afterwards); Greenhouse screening experiments for nematode resistance. Achievements Stock culture Populations from various parts of the world were established on carrot discs for Radopholus similis (6), Pratylenchus coffeae (3) and Pratylenchus goodeyi (3). Populations of the root-knot Meloidogyne spp. (6) were established on transformed tomato roots. The P. goodeyi populations were not used in the project. 132 Characterisation The reproductive fitness of R. similis populations was higher than that of the P. coffeae populations, with the sample from Cuba showing the fastest multiplication. In vitro screening The method has a great potential: it is very precise, has to be carried out in fully controlled conditions, and requires a minimum of space within a single room. The techniques involved have been established within this project (probably for the first time in this research area). Many components of the screening had to be adapted. An example is the inoculum mass where good performance was found with 25 mature female nematodes (per tested plantlet), individually collected with a micropipette and placed in a drop of sterile water on the surface of the growth medium. The results are promising but the method still needs further investigation. A serious problem was the microbial contamination of the nematodes collected from the carrot discs and ‘boosted’ by the rich in vitro growth media. It prompted the PI to search for an alternative (next activity). Musa genotypes of known resistance/susceptibility against R. similis were tried out : ‘Pisang Jari Buaya’ and Yangambi Km5 (resistant); ‘Gros Michel’ and ‘Grande Naine’ (respectively moderately and highly susceptible). Tentative results are in agreement, but the resistance of Yangambi Km5 was only partly expressed, and since that resistance is clear in the classical testing conditions (pots, fields), the in vitro technique needs to be adjusted. In vitro culture of R. similis on carrot callus Callus tissue developed from carrot discs enabled nematode proliferation (200-fold increase in 10 weeks which is, however, 10 times lower than with carrot discs) and no further microbial contamination was reported. But the nematodes did not proliferate further on banana plants in the greenhouse pots, while those from carrot discs multiplied 7 times. An alternative culturing method for rearing sterile nematodes is thus still needed. Screening for resistance in greenhouse conditions In vitro plantlets (roots developed) were placed in plastic pots (diameter 12cm) filled with autoclaved soil, and progressively exposed (4 weeks, flexible plastic cover) to the greenhouse atmosphere, whereupon inoculation took place and observation was effected 56 days later. Data were expressed in number of nematodes per gram of fresh roots. The method proved to be efficient (see also BIP 20). About 80 Musa genotypes were screened, including more than 30 edible AA from Papua New Guinea, plantains, one Highland (East-African AAA), accessions from Vietnam and Malaysia, 14 cultivars considered to belong to the ‘Pisang Jari Buaya’ (PJB)-subgroup, various AAB and ABB, a couple of new hybrids and even ‘Fe’I’ (Australimusa) clones. 133 The “PJB-subgroup” is probably an artefact since some cultivars were quite susceptible. The standard PJB as well as the ‘Yangambi Km5’, both resistant to R. similis, were susceptible to P. coffeae and Meloidogyne spp. Results of resistance screening against one nematode cannot be extrapolated. New sources of resistance to R. similis were found in the ‘Fe-I’ group. Evaluation Method A desk study was performed on the basis of the very comprehensive reports. The project was visited regularly and the PI consulted for technical details. Results The way the objectives have been met needs some explanation in order to evaluate the achievements. Objective (2) is the key one. Two early-screening methods were established of which one, in vitro screening, is still under investigation and has consequently not been applicable as yet. The other method, small-pot screening in greenhouse, proved to be efficient and was applied for Objective (3) as well as Objective (1) of which the FHIA component was in fact a major element of Project BIP 20, and of which the IITA component is not part of any BIP project. Consequently, the achievements within this project are concerned with Objectives (2) and (3) only. Objective (2) was only partly met but the first results are still promising: the small-pot method ‘works’ and the in vitro method has a good chance to become applicable in the near future. Finding an alternative nematode rearing method (to the carrot disc method) remains a feasible goal. Objective (3) has been met with satisfaction for the R. similis case. The results also made clear that the Musa germplasm has to be screened for probably each of the nematode species concerned, a most important factor in the future organisation of the screenings of hybrids. However, and beyond the stated objectives, the BIP 19 project played a motor role in the establishment and the running of a ‘Musa Nematologists Consortium’. The partners (FHIA, IITA, KULeuven) are presently forging operational links with specialists in South America, Caribbean, Asia and Australia, and are thus preparing a precious instrument for reliable nematological evaluation under different agroecological conditions of cultivars and new hybrids. 134 Conclusions Due to Project BIP 19, a mechanism with specialised laboratories and proper methods and techniques for nematode resistance screening of Musa genotypes has successfully been initiated. It can be expected that the mechanism will be developed further and become productive in the coming years, mainly under the impulse of the newly created ‘Musa Nematologists Consortium’. 135 BIP 20 Identification of Durable Nematode Resistance Sources in Banana and Plantain Description Co-ordinates Organisation: Fundación Hondureña de Investigación Agrícola (FHIA) Principal Investigators: Dr. Phillip Rowe Dr. Nicole Viaene Dr. Mauricio Rivera Duration: Three years, starting March 1, 1995 Objectives 1. To provide the existing conventional breeding program for Musa improvement of FHIA with a nematological component. 2. To adapt or develop early, rapid and reliable methods for nematode resistance and/or tolerance screening in banana and plantain. 3. To screen banana and plantain genotypes for nematode resistance and tolerance. Activities 1. Development and maintenance of nematode cultures in the laboratory to provide a continuous supply of fresh inocula used to infest soil of pots and bags in the shadehouse experiments. 2. Shadehouse testing of 35 genotypes of interest, following INIBAP´s guidelines for recording data, to evaluate resistance and tolerance to R. similis and P. coffeae in comparison to several reference genotypes of known reaction to the nematode R. similis. 3. Field testing of fifteen genotypes for resistance and tolerance to nematodes was conducted in a field located at Guaruma, La Lima, Cortés, well known for the predominant occurrence of P. coffeae, although Meloidogyne sp., R. similis and H. multicinctus are also found occurring in very low numbers. The field was planted on August 1996 with corms of banana cv. Grand Nain as a blank crop to increase the natural populations of P. coffeae in the field. 4. A second field trial was planted on October 1997 in lot C3N, Finca La Paz, a commercial banana plantation at Isletas, Colón, for testing the resistance and tolerance to nematodes of 20 genotypes (including the 15 genotypes in the test mentioned in item 3). 136 Achievements Nematode Cultures: The technique used for in vitro development of the nematode populations proved to be satisfactory in the case of R. similis. P. coffeae showed to be less amenable for in vitro multiplication under local conditions. Culturing of Meloidogyne sp. on bananas (FHIA-03 and Williams) gave excellent results: high quantities of eggs were obtained and maintenance of the plants was very easy. Rearing Meloidogyne on eggplants also gave good results. Establishment of a culture of Meloidogyne sp. in Petri dishes on excised tomato roots did not succeed due to contamination with bacteria and fungi. No tests were conducted with M. incognita as a result of not being able to develop in vitro cultures and besides this nematode is of no economic significance in the region. Shadehouse Tests Significant differences in resistance to R. similis and P. coffeae were found among genotypes. Materials which consistently showed resistance to either one, or both of the nematodes were identified, some of them promising for current of future use in FHIA’s breeding program. Complete resistance to R. similis was shown by the reference genotype Pisang Jari Buaya (source FHIA), the natural diploids Pisang Tunjuk II-28, Mundan AVP-36, and the improved parental diploids SH-2095 and SH-3142. The reference genotype Yangambi Km5 (source FHIA), together with the improved parental diploids, SH3723, SH-3362 and SH-3624, and the wild diploids Pisang Jari Buaya (source ITC), Calcutta 4, Pisang Batuau II-95, Biu Batu II-92, Siang Hil II-82 and Tangung II-77 all showed to be highly resistant to R. similis. Moderately resistant to R. similis were SH-3648, reference genotype Yangambi Km5 (source ITC), FHIA-01, FHIA-18, SH3386, SH-3437, FHIA-23 and Pelipita. As for P. coffeae, the lack of reference genotypes of known reaction to this nematode forced the investigators to consider the resistance reaction grading with some reserve. Nevertheless, a consistent, continuous variation in reaction was evident in the genotypes. Complete resistance to P. coffeae was found only in Calcutta 4. The genotypes SH-3437, SH-3217, Yangambi Km5 (both sources), Lowgate, Biu Batu II-92, Siang Hil II-82, and Tangung II-77 were highly resistant to P. coffeae. Moderate resistance to P. coffeae was found in Highgate, SH-3624, FHIA-17, Prata Enana, Mundan AVP-36 and Pisang Batuau II-195. Distinct high to complete resistance to both R. similis and P. coffeae in the same genotypes did occur in the natural diploids Calcutta 4, Yangambi Km5 (FHIA source), Biu Batu II-92, Siang Hil II-82 and Tangung II-77. The natural diploids Mundan AVP-36 and Pisang Batuau II-195, the cv. Prata Enana, and also the improved diploids SH-3624 and SH-3437 appeared to carry resistance to both nematodes, although in variable levels. Field Tests In the field, in Guarumas specifically, collection of samples three months after planting, at shooting and at harvest was completed during the first cycle. During the second cycle, sampling at shooting and harvest covered only 35% and 3%, 137 respectively, due to the destruction of the trial by floods brought about by Hurricane Mitch. The first sampling date was the only one performed at Finca La Paz, due also to the destruction of the trial by the floods. Sampling at first shooting and first harvest was only 74% and 30%, respectively. At Guarumas 1, P. coffeae was the predominant banana nematode present. At all three sampling dates at this site, there were differences in nematode populations and damage between main lots (much lower in nematicide-treated than in nonnematicide treated lots) and among genotypes within main lots. Like in the shadehouse tests, lack of reference genotypes of known reaction to P. coffeae forced us to consider the resistance reaction grading with some reserve. Based on the results of all three sampling dates, regardless of applying or not applying nematicide, Yangambi Km5 and Highgate appeared as completely resistant and highly resistant, respectively; FHIA-01, FHIA-02, FHIA-03, FHIA-18 and SH-3624 fell in a dubious category that included what would be considered moderately resistant and moderately susceptible. FHIA-20, FHIA-21, Grand Nain, SH-3723, SH-3697, SH-3142 and SH-3362 were moderate to highly susceptible. These data are still preliminary considering that at least two cropping seasons in the field are required to obtain conclusive data. Nevertheless, the data clearly showed the tendency of most of the materials to have smaller nematode populations and lower damage ratings, in close agreement in most cases with the rankings or the resistance grading resulting from the shadehouse tests. At La Paz, it was confirmed that R. similis was the predominant nematode species, although Helicotylenchus sp. and Meloidogyne sp also occurred in very low numbers. At this site the results were not as clear as at Guarumas 1, most likely as a result of the natural variation of nematode populations in the soil as well as the problem of frequent failure of plant germination at the initial planting date and in replants done later. As a result, no major attempt was made to try to interpret the data. In spite of their short duration, the results of the field trial at Guarumas 1 provided strong evidence that there is a close correlation and agreement between field results and results obtained from shadehouse tests. Although more information is necessary in order to fully document this correlation, these results partially validate the results obtained in the shadehouse for most genotypes. In fact, shadehouse tests should provide a better estimate of the reaction of the genotypes since their environment is free of the problems of aggregation of nematode populations that naturally occur in cultivated soils and which undoubtedly provoke unexpected results in materials of known reaction, as occurred in some cases with the field trials. Evaluation Methodology A desk study was performed based on two interim reports for activities in 1996, a 1997 annual report, and an interim report for the first semester of 1998, as well as the final report. 138 The project was visited on February 16, 1999 and included: - Review of the project; Visit to the shadehouse where testing was conducted; Visit to the Guaruma field where testing was conducted (heavily affected by Hurricane Mitch). Results 1. Nematode Cultures: In vitro development of nematode cultures of R. similis was accomplished readily. To the contrary, in vitro nematode cultures of P. coffeae were not easily accomplished. Having enough quantities of these cultures is very important for the pathologist and breeder testing for resistance to these nematodes. Therefore, it will be important to continue to develop the best protocol to develop cultures of P. coffeae more efficiently. 2. Shadehouse and Field Genotype Testing for Resistance to Nematodes: The genotypes tested showed significant differences for resistance to R. similis and P. coffeae. Similar rankings for genotype resistance were observed for the tests in the shadehouse and in the field. This is a rather important finding for the plant breeder who is searching for genotype resistance in segregating populations. Regarding the planting material used, essentially the same degree of resistance to nematodes occurred whether the genotypes were tested as tissue culture (TC) or as corm (CO) plants. Besides obtaining information on resistance of the Musa genotypes, the results of the screening tests have also provided insight into the influence of experimental conditions (such as initial nematode infestation level, type and age of planting material, influence of environment) and on nematode reproduction and nematode damage caused to the plants. This information will be useful in future interpretation of results and in improving screening methods. Because field testing was partly affected by the flooding caused by Hurricane Mitch, it is advisable to perform more field tests in order to corroborate the high correlation observed between the resistance in the field and in the shadehouse. Conclusions By implementing this project, FHIA has now developed the expertise and resource capability to perform on a routine basis all the laboratory, shadehouse and field activities required for determination of Musa resistance to Radopholus similis and Pratylenchus coffeae. The techniques and methods used in determination of the reaction of the genotypes to both nematodes proved to be quick and also reliable since repetitive results were obtained. In addition, improved and natural genotypes were identified which consistently showed resistance to either nematode. This 139 knowledge will enable the breeder at FHIA to design the crossing schemes in a way that gives maximum chance to the resulting progenies to express resistance to both nematodes. Therefore, all the three project objectives were met. This is all the more remarkable since the field tests were hampered for some time because of the flooding caused by Hurricane Mitch. The field data that it was possible to obtain were judiciously analysed and results could thus be included in the final report, with the necessary care in interpretation. The expertise and technical preparation of the scientists involved was very adequate. Prospects The following studies will be undertaken or considered in the near future to strengthen the results of this project: 1. Determination of the possible differences in pathogenicity that may occur between inocula of P. coffeae obtained from continued subculturing in laboratory and inocula obtained from roots of infected banana plants. 2. Determination of the relationship between nematode populations, magnitude of damage on corm and roots and their effect on total plant fresh weight as a means of measuring tolerance to nematodes in potted or bagged plants in shadehouse conditions. 3. Determination at the end of the exposure time of the relationship that might exist between nematode populations in the roots, nematode populations in the soil, and the magnitude of damage expressed in roots and corms. This research is aimed at trying to explain the occurrence of high nematode damage in some genotypes and yet contrastingly very low nematode populations in the roots. Error could be occurring in the quantification of the true nematode population as a result of migration taking place to the surrounding soil once they have exhausted their source of food. 4. Research is planned in order to define a set of reference genotypes with known reaction to P. coffeae. 5. Formal determination of the response of Musa genotypes to P. coffeae when using TC and CO plants. 6. Comparison of the reaction expressed by the reference genotypes of the FHIA and ITC sources. 7. Retesting (for conclusiveness) those genotypes in which the available evidence of their reaction to the nematodes was based on one, or at the most two, tests. 8. Provided financial resources are available, it has been considered to establish a field study with select nematode-resistant genotypes to determine the durability of their resistance to those nematodes after continued exposure for several production cycles. 140 BIP 21 Identification of Durable Nematode Resistance Sources in Banana and Plantain Description Co-ordinates Organisation: Instituto de Recerca I Tecnologia Agroalimentares (IRTA) Collaborating Institute: Insituto Canario de Investigaciones Agrarias (ICIA) Principal Investigator: Dr. Jorge Pinochet Collaborating: Dr. Montserrat Jaumot (IRTA) and Mariá de Carmen Jaizme (ICIA) (abroad) Dr. D. De Waele (KULeuven), Dr. P. Rowe (FHIA), Dr. P. Speyer (IITA) Duration: 1 March 1995 - August 1996 Objectives 1. 2. 3. To provide the existing classical breeding programs for Musa improvement of FHIA in Honduras and of IITA in Africa (Nigeria and Uganda) with a nematological component. To adapt or develop early, rapid and reliable methods for nematode resistance screening. To screen banana and plantain genotypes for nematode resistance. Note: Although the three objectives are stated in the Completion Report, they are in fact only reflecting what was originally proposed by the PI, and conform to the intention to work in close collaboration with the ‘Nematological Consortium’ for three years (see same objectives for BIP 19). The project did not, however, receive the required funds for the total operation and had to be terminated after 1.5 years. Only the activities related to Objectives (2) and (3) could be undertaken (in part). The practical purpose of the project was to screen relevant Musa genotypes at the ICIA (Canary Islands), and this could eventually be achieved with other funds (personal communication: Dr.De Waele). Activities 1. Nematode culture and inocula preparation; 141 2. Resistance screening in pots, under greenhouse conditions. Achievements Nematode culture A total of 21 isolates were established and maintained at IRTA. They belong to the species R. similis, P. coffeae, P. goodeyi and Meloidogyne spp. The rearing of the nematodes on carrot discs was difficult (see same problem in BIP 20). Around 10 to 15 cultures per isolate were made readily available for immediate use (20,000 to 100,000 nematodes per culture). Several cultures were sent to members of the ‘Nematological Consortium’. Resistance screening in Greenhouse By mid-1996, screening trials for lesion and root-knot nematodes were running. Evaluation Method A desk study was carried out with the Completion Report as the only source. Results Considering the time constraint and the fact that funds were not received before September 1995, leaving hardly one year for implementation, the work achieved can be regarded as very satisfactory. In this short time: - Many abundant and pure nematode cultures were established, maintained, and even distributed to partners; Greenhouse tests were initiated, thus inaugurating a program of great importance for banana research in the Canary Islands. Had the project been supported as was proposed, it would have been very beneficial for the ‘Musa Nematological Consortium’. The PI was already one of the only Musa nematode specialists in the world at the onset of the project, and had previously been active in Musa resistance screening (at FHIA). Conclusions The first results of the project were very promising and it is unfortunate for the ‘Nematology Consortium’ that the project could not evolve beyond the scope of the BIP. 142 BIP 22 Breeding Hybrid Musaceae with Resistance to Multiple Diseases, especially Black Sigatoka and Panama Disease Description Co-ordinates Organisation: Fundación Hondureña de Investigación Agrícola (FHIA) Principal Investigator: Dr. Phillip Rowe Duration: Two years, starting January 1, 1997 Objectives To develop hybrid bananas with resistance to multiple diseases, especially black Sigatoka (Mycosphaerella fijiensis) and Panama Disease (Fusarium oxysporum f.sp. cubense) by conventional breeding Activities Pollinations were made between two diploid parents (SH-3142 and SH-3362) and two triploid banana cultivars, the seed-fertile Lowgate and Dwarf Prata, for the production of tetraploid hybrids. The SH-3444 tetraploid was pollinated in 4x x 2x crosses to produce triploid hybrids. Achievements Pollinations between diploid and triploid parents From the 2,200 pollinated bunches of Lowgate, 16 hybrid seeds were obtained. Four embryos were viable, and three hybrid plants were evaluated in the field (the other is still in the nursery). Three tetraploids were derived from the Lowgate x SH-3362 crosses and were evaluated in the field. All these hybrids had the Gran Nain-like short plant height, but were susceptible to black Sigatoka and had mediocre bunch features. A total of 685 seeds were produced on 450 pollinated bunches of Dwarf Prata. A total of 250 hybrid seeds had embryos. From these 250 cultured embryos, 123 new hybrid plants have been transplanted to the field. At the end of the project, the 123 plants derived from crosses onto Dwarf Prata had not yet flowered in the field. 143 Pollinations between SH-3444 and diploid parents: A total of 84 seeds were obtained from 600 pollinated bunches of SH-3444. A total of 20 embryos were cultured. From these 20 cultured embryos, 17 hybrid plants were obtained. Two of these 17 have the gene for dwarfness and only these two dwarf plants will be transplanted to the field for further evaluation (these two plants are presently in the nursery). Evaluation Methodology A desk study was performed based on the annual report for 1997 and a Progress Report for the period January 1, 1998 to June 30, 1998, as well as the Final Report. The project was visited on February 15, 1999 and included: - Review of the project with Principal Investigator; Visit to the fields where the hybrid evaluation is being conducted (heavily affected by Hurricane Mitch); Visit to the crossing blocks in the field (heavily affected by Hurricane Mitch). Results The results in this project have shown that very few seeds are produced from pollinating Lowgate: one hybrid per 500 bunches. This indicates that several thousand bunches of Lowgate should be pollinated annually in order to obtain a large segregating population to increase the possibilities for selection of commercial hybrids with the desired combinations of disease resistance, productivity, and fruit qualities. Although low numbers of viable embryos were obtained, the 3x x 2x and 4x x 2x crossing schemes used in this project have proven to be effective in the development of disease-resistant hybrids. There is no question about whether these crossing schemes work. It is just a matter of time and larger segregating populations from which to select the best hybrids. Conclusions The objective of the project to develop hybrid bananas with resistance to multiple diseases especially black Sigatoka (Mycosphaerella fijiensis) and Panama Disease (Fusarium oxysporum f.sp. cubense) by conventional breeding was not achieved, as was expected, in the time frame defined for the project. The eventual infertility of some of the genetic material used handicapped progress, especially in the development of large segregating populations. However, it was proven that the crossing schemes (3x x 2x and 4x x 2x) are valid. 144 It is critically important that all the activities in this project be continued, the FHIA scheme being the only one with the proper plant material and expertise to create resistant hybrids that would reproduce the organoleptic qualifications of the ‘Lowgate’. This cultivar belongs to the ‘Gros Michel’ subgroup of dessert bananas, of which the fruit quality is commonly recognised as superior to that of the ‘Cavendish’ subgroup. The time frame defined to achieve the main objective of developing excellent hybrid dessert bananas with resistance to multiple diseases especially black Sigatoka (Mycosphaerella fijiensis) and Panama Disease (Fusarium oxysporum f.sp. cubense) by conventional breeding was not sufficient for a project of this kind. The expertise and scientific knowledge of the Principal Investigator is well known and of excellent quality. Prospects In order to better achieve the objective, there is a need to dedicate more time and funds to obtaining larger segregating populations in order to increase the possibilities of obtaining hybrid bananas with desirable commercial traits, as was planned in this project. Indeed, since black Sigatoka could destroy the export banana trade, the activities undertaken in this project should be greatly enlarged. FHIA will continue limited work on genetic improvement of bananas with its own funds. However, these reduced activities will be only a fraction of what needs to be undertaken. Therefore, a next phase for this project should be planned and funded. 145 CONCLUSIONS AND RECOMMENDATIONS The Concept As was explained in the Introduction, the Terms of Reference of this Project evolved in the course of its preparation, with the main result that it eventually covered a much larger research field than was initially conceived. One is consequently dealing with a large research program, rather than with a project. The program consisted of eighteen projects which together did not form a coherent and fully-fledged entity since several relevant research domains were missing (e.g. the multi-faceted work on important pests and diseases, such as the Borer Weevil, the Banana Bunchy Top Virus, the Banana Bract Mosaic Virus; research on post-harvest technology; and research on various agronomic issues). The only project directly addressing the black Sigatoka disease (BIP 17) was restricted to one aspect of the resistance of M. fijiensis to fungicides. The entire domain of the host-pathogen relation and of its co-evolution was left untouched. A key question, therefore, is whether the actual implementation of the BIP addressed the following two original objectives, entirely and correctly: 1. To develop and evaluate, over a five-year period, a range of improved banana varieties with export potential, incorporating increased productivity with durable disease resistance, using conventional and non-conventional breeding techniques. 2. To develop more efficient and integrated disease management practices for black Sigatoka disease, based on an improved understanding of disease epidemiology and pathogen variability in different areas and over different seasons. The answer appears to be: “Partially yes, but several positive results concern subjects that were not implied by the above objectives, while other relevant activities are missing”. This “skewed” product is the result of the change in orientation of the BIP during the initial implementation phase as explained in the Introduction. The present Review Team does not consider it as its task to evaluate the judiciousness of such reorientation, which apparently was effected by the PEA with the implicit agreement of the CFC and the Supervisory Body. The Team had to evaluate the implementation and the products of the BIP as it was eventually established. But it should be borne in mind that the re-orientation, while addressing very important research subjects that were not in the original Terms of Reference, had as a consequence the abandonment (or at least the considerable “shrinking”) of the following three major activities: - Testing and post-harvest evaluation and propagation of new hybrids; 146 - Research on the Sigatoka diseases; 9 Information, documentation and training. IMPLEMENTATION AND RESULTS Genetic Improvement It can be stated that the actual BIP, although no longer exclusively focussed on the development and evaluation of “improved banana varieties with export potential”, has nevertheless produced important results to that end. Project BIP 22 directly addressed this theme, while the results of most other projects, especially the ‘biotechnology’ ones, will be of great (future) help for such improvement. The BIP projects dealing with Conventional Breeding can be considered as an interim support to long-term programs. They nevertheless had a direct (BIP 05, 22) or indirect (BIP 02) boosting effect on the breeding effort. The EMBRAPA project (BIP 05) produced the most striking results, since black Sigatoka-resistant hybrids are presently propagated right in the zone of the recent attack in Brazil. Project BIP 22 at FHIA had a duration of 2 years only, and adequate further support for the effort to develop improved ‘Highgate’ commercial export bananas is an absolute necessity for the breeding program to become fully productive. While progress was made in the understanding of the genetics of resistance to Black Leaf Streak (black Sigatoka) with the help of project BIP 02, the results reveal the complexity of that research theme as well as the need for still more improved observation techniques in highly controlled environments in order to make the promising results applicable for breeders. Indeed, it can be stated that the results of these three projects risk becoming obsolete unless the programs involved receive substantial further support. This leads to the question of whether such long-term efforts can survive on projects with limited duration, such as the BIP. The Review Team advances the opinion that in this particular research area a support system based on projects of limited duration is less productive in the end, because of the production of fragmented and often inapplicable results. The Team hopes that sponsors and supporters will reconsider the short duration approach to funding. As documented in the different reviews on the biotechnological part of the BIP, i.e. the Non-Conventional Breeding (encompassing BIP 08, 09, 10, 11 and 12), the results were a mixture of deficiencies and remarkable achievements. Even if viewed very critically, the biotechnology projects of the BIP were together productive, and some of the outcomes will doubtlessly be of future importance for the nonconventional genetic improvement of banana varieties. A few examples should illustrate this point. First, successful culture of tissues and cells, and reproducible regeneration of complete plants is no more restricted to one (or at best a few) cultivars. On the basis of the established regeneration protocols, it is fair to expect, that any cultivar will be regenerable in future. Second, the 9 Conferatur Annex 1, Objective (1), sub Outputs 1 and 3; Objective (2). 147 transformation of banana has left the purely experimental stage, and the BIPsupported development of transformation technologies will make stable banana transformation less of a trial-and-error type of experiment. Third, the scientific community now knows which promoters perform best in transgenic bananas, and which genes are available that confer meaningful agronomic traits. Fourth, cDNA libraries are available from different tissues that await exploitation beyond the isolation of fruit ripening-associated messages. Genomic libraries allow the isolation of full-length genes and relevant promoters. Fifth, some, not always encouraging, experiences have been made with transgenic banana plants, from which lessons for the future can be learned. For example, the number of proven transgenics should definitely be higher (> 50 plants) than available at the end of BIP. There are other achievements of the BIP, that will last and positively influence banana research in this area (see the individual project reviews). The ground for all of these achievements was prepared before the BIP started, and was supported by various other sources. In some cases, BIP funding was taken as an additional, yet welcome increase in financial potential; in other cases it was catalytic. In all BIP groups working on the biotechnology aspects, research on the above topics will continue with other, as yet limited and short-duration funds only. So, was BIP only an episode in an otherwise already established banana improvement program? Certainly yes! As far as volume and supported activities are concerned, the BIP was no more than any other research program: too short, too spartanic. However, it shares with many other programs, the provision of an incentive to the banana research community to focus and to collaborate, albeit with limited success, as might be expected in view of the limited time and resources applied. In the area of Genetic Resources and Resistance Sources, the BIP produced results of unequal value, for several reasons. Positive is the newly acquired and multifaceted knowledge on diploids as breeding sources (BIP 01), the much better understanding on plantains elsewhere than in Africa (BIP 03), the establishment of a good-performing banana research capacity in Vietnam (BIP 04), and the considerable progress made in the screening of breeding material for nematoderesistance (BIP 18 to 21). Shortcomings in BIP 01 were mainly due to over-ambitious objectives (e.g. agronomic evaluation over three cycles is an unrealistic goal within three years of implementation; the link between morphological and molecular markers remains to be established). Unrealistic objective setting was a weakness also in BIP 04. The present review also shows that this project evolved in a rather isolated scientific environment, despite the BIP Workshop organised there in its initial stages. Lack of scientific advice from abroad was a handicap for project BIP 03 as well, on top of its extremely short duration (18 months). Still, both these projects, implemented by modest but skilful local scientists, paved the way for solid further research of great importance. Further support is, however, again the problem. The BIP has been instrumental in the initiation of an international network for mematode-resistance screening methods. In particular, the implementers of BIP 20 148 can be commended for the remarkable progress made in the methodology for assessing resistance to nematodes. But much further work is necessary for the improvement of the relevant techniques, particularly at laboratory level, as project BIP 19 reveals. Yet, the momentum created by the BIP risks stalling by lack of further support. The Musa Nematology Consortium, which proved to be an instrumental innovation, cannot rely in the future solely on the goodwill of its members and calls for sufficient international support. Pest and Disease Control With the exception of project BIP 17, no progress can be registered for the BIP in research on the Sigatoka diseases. The reason for this shortcoming was explained earlier (“The Concept”). On the other hand, the BIP can be considered as “the architect” of the considerable progress made in research on Fusarium Wilt (BIP 15) and on Banana Streak Virus (BIP 14). The results of both projects do have far-reaching and long-lasting positive implications for banana breeding. The projects were also instrumental in some informal network building among specialists and collaborators at international level. IMPACT As is the case for many research projects, the impact of the BIP on the future of banana/plantain production can hardly be formulated in economic terms, because the integration of its results into the production and the marketing sectors needs quite some time. The Review Team is of the opinion that the impact on further research is what interested parties (including donors) would like to see in the first place. In this respect, the BIP’s significance is undeniable. Entire panels of research have been catalysed through the BIP in each of the domains where it was active. On the basis of the above conclusions regarding “Implementation and Results”, one can predict with confidence that many results will positively and deeply influence further research. A “spin-off” is already discernible in the following cases: - - - For the first time in Brazil, black Sigatoka-resistant hybrids, suited to the local market, have been propagated for further testing and many newly produced resistant diploid hybrids have been introgressed for the development of still better performing end-products, which will play an important role in banana production for local consumption in Latin-America and elsewhere. Research on the genetics of the resistance to black Sigatoka as well as the characterisation/ evaluation of numerous diploid cultivars are likely to culminate during the coming years in the production of appropriate “building stones” (improved diploids) of all sorts of resistant hybrids. Genetic transformation of Musa tissues and subsequent plant regeneration is presently implemented in a wide cultivar spectrum and the catalytic role of the 149 - - BIP in this evolution is undeniable. The cDNA- and genomic libraries established via the BIP now offer the opportunity to extend the production of “transgenics” for different purposes. Systematic screening of breeding material for resistance to nematodes has been achieved for the first time on an international scale and the established basic methodology has been improved. The considerable progress realised in the study for Fusarium Wilt has already had an impact on resistance-building adapted to local host-pathogen conditions. The better understanding of the Banana Streak Virus in all its aspects is of great assistance to breeders. The BIP “came in time”. However, interested parties should be aware of the fragile character of this potential impact, in a world where financial support from the public sector for such research is apparently much less certain than it was only a decade ago. For nearly all of the eighteen projects, the cessation or even reduction of financial support would seriously compromise the impact of further research. The BIP has brought banana research near the point of multi-purpose application, and a decrease in public sector interest in this research could run the risk of being viewed in the future as a historical error. RECOMMENDATIONS A major lesson for the future is that a 5-year project such as the BIP should not be compromised in its implementation by a fundamental reconsideration of its agreed activities leading to a two-year delay in “action on the field”. Many subprojects dealing with genetic improvement would have achieved the set objectives with a four-to-five year duration. The argument that the competitive bidding system was a guarantee for an equilibrated performance of the whole BIP is only partly valid. A major part of the funds was allocated anyway to institutes with existing expertise, and these institutes would have achieved a lot more within the agreed five-year framework if more direct support had been given. Competitive bidding (for a substantial remaining part of the total funding) could then still have had a meaning for institutes where research on Musa was/is being developed. That part would probably have attracted necessary co-financing, which apparently and most unfortunately has been lacking in the present BIP. Instead, the BIP tried to involve too many different aspects. While this was desirable in a sense, it led to a dilution of funds with the result that that no single activity could be sufficiently supported. For example, the inclusion of HKUST in the BIP for only one single year would not have had any effect on banana research, if HKUST had not invested considerable funds to save the project. Another example is the excellent but very incomplete work on plantains in India (BIP 03) for which only 18 months were planned, due to previous insufficient investigation in local research capacities and needs. 150 Regarding biotechnology research, genomics and transcriptomics (i. e. the huge repertoire of techniques for genome analysis and gene expression analysis) should be involved to isolate more agronomically interesting genes (e. g. resistance genes, genes for fruit quality, or genes for yield increase), and promoters to drive these genes under various environmental regimes. These genes and promoters should then be used to improve commercially important local cultivars for a particular trait, and transgenics should be studied for trait inheritance and stability. Cornell University (among others) has already started a similar program very much along the same lines. It is certain, that industry would be involved in such a Banana Genome Initiative (BGI). So, less (projects) can be more (progress) in research. It is recommended that projects of the importance of the BIP should not only 10 restrict the objectives to a minimum , but should assure sufficient time (5 years) and funds for the actual implementation of the (previously) carefully selected11 activities involved. An important aspect of the BIP was its involvement in long-term research (conventional breeding and fundamental aspects in several subprojects, as, for example, the genetics of black Sigatoka-resistance, BIP 02). It should be made clear to interested parties that this type of research, so vital for the future, should not have to rely on projects with limited duration, but rather form part of long-term core support. While the BIP was very beneficial for that research, the question now is whether the effort will be continued (if funds still have to come via short-duration projects). To paraphrase this: Everybody mentions “durable resistance/productivity/etc”; where is the “durable support”? It is recommended that conventional breeding and connected genetics research (including the search for proper molecular markers) benefit from appropriate long-term core support in an institutional context. Pathogens and their host plant evolved and still evolve simultaneously in a dynamic and rather closed system (co-evolution). Therefore, it is obligatory to study both in order to predict the consequences of any intervention in the system. While research on virus resistance within the BIP met this criterion in the case of the BSV, analogous work on the important black Sigatoka (and yellow Sigatoka) diseases, for example, was clearly under-represented. The productivity of the BIP suffered from a serious discrepancy in efforts between “resistance-building” and “understanding what this resistance actually means”12. As a result, genetic improvement is still short of precise and adequate physiological criteria (and of involved genetic markers) which are needed for the construction of a reliable resistance to the disease: failures and shortages appear on hybrids grown in different environments, and these can as yet not be explained for the breeder to adapt methods and techniques. 10 As was actually the case for the BIP. That is: envisaged and then selected during the preparatory phase. 12 In physiological and ecological terms. 11 151 It is recommended that genetic improvement for resistance to a particular disease or pest be flanked by sufficient research on the host-pathogen/pest interaction. The INIBAP-promoted PROMUSA system, with its working groups articulated on genetic improvement as well as on the relevant diseases and pests, could offer a convenient forum. In the experience of the Review Team, contact between the different participants in the BIP existed, but in no case was intense co-operation established. Such an interaction would have prevented some duplication, which generally is not at all rewarding. And such duplication existed in several cases in the biotechnology part of the BIP. It is recommended that projects, during their preparatory phase, encourage the desired co-operation between groups with similar interests (e.g. BBTV resistance research) to submit joint and interlinked research proposals. The INIBAP should be instrumental in such endeavours. Though transgenic bananas were considered a product from the beginning of the BIP, obviously issues of deliberate release, availability of national guidelines, and public perception were not considered sufficiently at the onset of the program. As a consequence, even after termination of the BIP funding, no clear decision was possible as to the release of transgenics in, for example, India. It is recommended that sensitive issues in relation to the production of transgenic bananas be positively considered prior to, or early in, a project/program. 152 ANNEX 1 : ORIGINAL TERMS OF REFERENCE OF THE BIP Objectives: 1. To develop and evaluate, over a five-year period, a range of improved banana varieties with export potential, incorporating increased productivity with durable disease resistance, using conventional and non-conventional breeding techniques;. 2. To develop more efficient and integrated disease management practices for black Sigatoka disease, based on an improved understanding of disease epidemiology and pathogen variability in different areas and over different seasons. Outputs and Activities Objective 1: Output 1. New banana varieties with export potential Activity 1: Conventional breeding of improved Musa germplasm Activity 2: Multi-locational testing of new varieties Activity 3: Rapid propagation of new varieties Activity 4: Post-harvest evaluation Activity 5: Disease screening methodologies Activity 6: Use of modern biotechnologies in banana breeding. Output 2. Musa germplasm collection and conservation Activity 1: Germplasm collecting, conservation and utilisation Activity 2: Collection of pest, pathogens and biocontrol agents Output 3. Documentation, information and training Activity 1: Information and documentation Activity 2: Workshop and training Objective 2: Output 1. Sigatoka Disease Forecasting System Activity 1: Epidemiological studies Output 2. Sigatoka variability description Activity 1: Genetic and ecological studies on the Sigatoka disease complex in different regions and over different seasons. 153 ANNEX 2: LIST OF ACRONYMS ACIAR: AGI: Australian Centre of International Agricultural Research Institute of Agricultural Genetics (Vietnam) BBMV: BBTV: BIP: BMV: BS: BSV: BTI: Banana Bract Mosaic Virus Banana Bunchy Top Virus Banana Improvement Project Banana (bract) Mosaic Virus Black Sigatoka Banana Streak Virus Boyce Thompson Institute (Cornell University, USA) CFC: CIRAD-FLHOR: Common Fund of Commodities Centre de coopération Internationale en Recherche Agronomique pour le Développement – Département productions fruitières et horticoles Cucumber Mosaic Virus Corporación Bananera Nacional de Costa Rica Centre de Recherche sur les Bananiers et les Bananiers Plantains CMV: CORBANA: CRBP: EMBRAPA: Brazilian Organisation for Agricultural Research FAO: FHIA: Food and Agriculture Organisation Fundación Hondureña de Investigaciõn Agricola GFP: GUS: Green Fluorescence Protein β- Glucuronidase HKUST: Hongkong University of Science and Technology IAEA: ICIA: IGG: IITA: INIBAP: IRTA: ITC: International Atomic Energy Agency Instituto Canario de Investigaciones Agrarias (Spain) Intergovernmental Group on Bananas International Institute of Tropical Agriculture International Network for the Improvement of Banana and Plantain Instituto de Recerca I Tecnologia Agroalimentares (Spain) INIBAP Transit Center KAU: KUL: Kerala Agricultural University (India) Katholieke Universiteit Leuven (Belgium) NRCB: National Research Center for Banana (India) PEA: PI: Project Executing Agency Principal Investigator 154 PNG: QDPI: QUT: Papua New Guinea Queensland Department of Primary Industries (Australia) Queensland University of Technology (Australia) UH: University of Hawaii WIBDECO: Winban Development Corporation YS: Yellow Sigatoka 155 156 PART III Impact Assessment for Banana Improvement Project 157 158 EXECUTIVE SUMMARY This report provides a brief assessment of the impact of the Banana Improvement Project (BIP) co-sponsored by the Common Fund for Commodities, the InterGovernmental Group on Bananas and the World Bank. The BIP consisted of 18 individual projects funded between 1994 and 1998. The original target audience for these projects was the world export banana industry with the focus altering during the life of the program to take in the needs of small holder farmers. The principal objective was to increase the productivity of banana production through the development and use of higher yielding disease-resistant varieties and by reducing the cost of production, especially the cost of pesticide applications. The BIP was managed by the World Bank with the assistance of a Program Manager and a Scientific Advisory Panel. The Panel was involved in industry consultation, the definition of research priorities and project selection. The Panel also met regularly to review the progress of the portfolio after projects had commenced. Before the projects were called for and selected, the World Bank commissioned a report to investigate the then current status of banana research and priority areas for the BIP. The report (Buddenhagen, 1996) highlighted the need for combining innovative breeding approaches with advances in plant biotechnology. This influenced the composition of the program portfolio. Most of the projects fell into two categories, breeding and biotechnology and many of the projects were long term and strategic in nature, where knowledge was the primary output. The knowledge produced has been in the following areas: • • • The collection and characterisation of a wide range of banana germplasm that will allow increased effectiveness of the banana-breeding programs around the world; Increased efficiency of breeding programs through improved testing, screening and typing methods and development of protocols; use of additional information assembled through use of genetic markers, particularly for disease resistance to Sigatoka, Panama Disease and nematodes; Improved and refined technology for genetic manipulation including transformation processes for banana and improved knowledge of disease resistance mechanisms in banana. The use of this knowledge will provide positive outcomes for plantains and all types of banana production whether for cooking, domestic consumption as dessert bananas or for production of export bananas. However, the application of this knowledge in the future to provide tangible outcomes for producers and exporters will require further R&D investment. While the program did not specifically target applied outputs, there were a number of areas where the program has had or will have direct industry impacts in the short term. 159 These include some black Sigatoka-resistant clones produced in the BIP currently being used to assist control of the recent black Sigatoka incursion into the Amazonas region of Brazil. Also, the ability to now type strains of Fusarium in a given location when considering new banana developments will be beneficial to new banana enterprise developments. The BIP enhanced co-ordination of effort and increased collaboration in international banana plant improvement. This was particularly evident in relation to what was termed the ‘biotechnology consortium’ (a joint effort between KUL, BTI and QUT). This alliance fostered staff exchanges between the three institutions involved and assisted the transfer and refinement of both particle bombardment and Agrobacterium transformation and regeneration technology. Another area in which increased collaboration was fostered was that of nematode screening for resistance through what was termed the ‘nematode consortium’. This consortium will continue after the BIP is completed and will be important in the screening and understanding of nematode resistance in the future (de Waele, 1995). A number of other collaborative efforts have been stimulated by the BIP. The BIP was successful, through the competitive grants approach, in attracting new research providers into banana improvement. Moreover, due to the necessity for many research institutions to attract external funding, such as through the BIP, the attraction of more resources into banana plant improvement was achieved. While some of the resources attracted into BIP projects would have probably been committed to banana improvement anyway, it is likely that, due to the BIP, there were additional resources attracted into the overall effort on banana improvement. In summary, the BIP has provided a unique set of outputs in terms of both content and process. The BIP has produced knowledge of a strategic nature as well as some technologies that have been, and can be, applied in the short term. In terms of process, the competitive bid system for identifying the best projects to fund within a strategic vision has proved successful, as has the leverage of resources into the banana R&D world, as well as increased collaboration and co-ordination among research teams. Outcomes of the BIP are defined as the activities or events that have occurred, or are likely to occur, as a result of the outputs from the BIP. Outcomes are directly relevant to producers of bananas and to others in the banana value chain including consumers. Outcomes can be positive or negative and are usually associated with benefits and costs. These benefits and costs may fall differentially on the various participants in the banana value chain. Two important short-term outcomes stand out. The first was the ability to type Fusarium strains before locating new banana developments. The world-wide collection of the various strains of Fusarium will serve as a reference collection in the future enabling the strains causing an outbreak of Panama disease to be quickly identified. New banana developments can be more safely implemented through determining the strains present in any given location before planting begins. For example, recent outbreaks of Fusarium 160 Wilt in Cavendish plantations in the eastern tropics (peninsular Malaysia, Sumatra and Halmahera) caused enormous damage. These outbreaks have ominous implications for the durability of western trades that are based on Cavendish clones. The avoidance of future losses will be one of the short-term benefits from the BIP. The second was the Brazilian genetic material used in Amazonas to combat the spread of black Sigatoka as it enters Brazil. Some banana clones resistant to black Sigatoka have been produced in Brazil with assistance from the BIP and are being used to slow the rate of advance of the disease in northern Brazil. This would have a major reduction in the costs of those who can potentially afford to spray with fungicides as well as reduce yield losses of those who cannot afford chemical control technologies. While the short-term outcomes and associated benefits are likely to have been significant the most important outcomes and contributions from the BIP are likely to be derived from the strategic knowledge contributions of BIP. Outcomes associated with knowledge generated from the BIP are inherently uncertain. However, in terms of assessing the impact of the BIP, it is possible to make some assumptions about the future and the role that the BIP will have played in the future development of the export banana industry. The key events shaping the future use of knowledge from the BIP with respect to export banana industries would include the development (in full or in part) of a Cavendish or alternative export banana type that is resistant to disease. Benefits that might be attributed at least in part to the BIP can be grouped into economic, environmental, and social and health benefits. The principal economic benefit most likely associated with the development of a disease-resistant banana is the reduced cost of production of export bananas through chemical replacement. The principal environmental benefit associated with the BIP is the likelihood of lesser amounts of pesticides being used if resistant or more tolerant varieties of export bananas are developed. The lower the pesticide applications, the lower the potential damage to the environment. As with the environment, a range of statements has been made that fungicides and nematicides used in banana production are associated with negative health impacts on banana plantation workers, and those living in banana-producing communities. Some of the perceived benefits related to health that would arise from developing disease-resistant bananas would be improved health of banana industry workers and higher labour productivity, as well as improved health and quality of life of communities in banana-producing regions. In order to evaluate the past investment in the BIP in quantitative terms, it is necessary to construct future scenarios that assume that the particular knowledge generated from the BIP is used in further R&D efforts that are eventually successful with the resulting technologies or products being adopted by industry. This requires a number of assumptions to be made about the type of future investment and associated costs, the time period before disease-resistant bananas will actually be produced, the likelihood 161 that the technologies will be produced within the timeframe assumed, and the proportion of total benefits that might be attributed to the BIP. The two key future scenarios assumed were: 1. The production of a transgenic Cavendish that is reasonably resistant to black Sigatoka and nematodes; 2. The development of export banana types that are resistant to black Sigatoka and nematodes. This is assumed to be achieved through a combination of transformation and regeneration technologies together with breeding components and has resulted in a robust form of resistance. Each of these scenarios was analysed in terms of potential benefits and costs that might be expected ‘with’ and ‘without’ the R&D program. The analysis was effected through a discounted cash flow analysis that ran for a period of thirty-five years, commencing in the 1994 calendar year. All costs and benefits were discounted back to 1994 using a discount rate of 7%. All values were expressed in 1998 US dollars. Investment criteria estimated included the Net Present Value (NPV), the Benefit-Cost Ratio (B/C Ratio), and the Internal Rate of Return (IRR). Criteria were estimated for each of the two scenarios separately. The investment criteria estimated using the base assumptions for Scenario 1 are summarised below: Expected Investment Criteria for Scenario 1 (Transgenic Cavendish) (Discount rate 7%) Total Investment BIP Investment only NPV ($ million) 578 27.1 B/C Ratio 9:1 10:1 IRR (%) 21.4 21.9 Expected Investment Criteria for Scenario 2 (Alternative Banana Type) (Discount rate 7%) Total Investment BIP Investment only NPV ($ million) 484 22.8 B/C Ratio 8:1 8:1 IRR (%) 20.0 20.5 The estimation of benefits is conservative in two ways. First, the actual parameters used in the quantitative analysis could be considered to be conservative. Second, there are a number of other benefits, outlined in earlier sections, that have not been included in the quantitative analysis. Environmental and human health benefits are two such benefits 162 that are usually difficult to value but are likely to be high and significant. Such benefits could also provide justification for public investment in banana improvement since benefits could be captured by communities, as opposed to the private sector, in banana-exporting countries. Given the assumptions made in the analysis, the results suggest that the BIP could be expected to provide a rate of return on funds invested of at least 20%. Conservative assumptions have been used through the analyses, and the sensitivity analyses and break-even parameters estimated show that the results are robust across a credible range of assumptions. The results of the investment analysis demonstrate that the payoffs being sought in terms of disease-resistant bananas can justify a large investment in banana improvement research. The BIP has provided the methods for the transformation and regeneration of bananas and has moved the world banana industry closer to the reality of a transgenic banana. Transformation and regeneration have now been accomplished. There has also been significant progress demonstrated in terms of new germplasm for use in breeding programs and methods for making breeding programs more effective in the future. The most important output from the BIP was strategic knowledge that can be used in further R&D projects in the future. 163 1. Introduction 1.1 Background and Methods Employed This report provides an impact assessment of the Banana Improvement Project (BIP) managed by the World Bank over the five-year period 1994 to 1998. The BIP is hereafter sometimes referred to as the ‘program’ rather than the ‘project’ so as not to confuse the ‘total portfolio of projects’ with the 18 individual research projects. The terms of reference for the impact assessment are provided in Annex 1. Most of the information assembled in this report was gained from progress and termination reports for individual projects. Most of this material was made available by the Program Co-ordinator before and during the consultant’s visit to the World Bank in Washington, D.C. In addition, personal visits were made to six individual project sites (two in Australia, two in Belgium and two in France), where discussions on specific projects were held with Principal Investigators. Discussions with each of the three members of the BIP Advisory Panel and further exchanges with Advisory Panel members also took place. These discussions and a visit to the World Bank in Washington, D.C. were held over the period 5 to 17 October, 1998. In addition, various industry personnel associated with the world export banana industry were also contacted to assist with various information requirements for the assessment. 1.2 Structure of Report Section 2 of the report provides a brief background to the BIP, including its aims and objectives, and a description and classification of the constituent projects that were funded. Section 3 provides a summary of the principal outputs from the program, while Section 4 describes the actual and likely outcomes of the research, as well as the benefits that have been derived to date and those that are likely to ensue in the future. The assumptions and results of an investment analysis for the funds invested in the BIP are reported in Section 5, and a brief conclusion of the report is contained in Section 6. 2. Description of Banana Improvement Program 2.1 Introduction and Background The BIP was sponsored by the Common Fund for Commodities (CFC), the FAO InterGovernmental Group on Bananas (FAO/IGB), and the World Bank. Resources of approximately US$ 3.5 million were provided by the CFC to fund the research portfolio. The BIP consisted of 18 projects located in a range of R&D organisations around the 164 world. The projects were selected through a competitive bidding system whereby over 200 national and international research organisations were invited to prepare and submit project proposals to the World Bank. The World Bank managed the BIP with the assistance of a Program Manager and a Scientific Advisory Panel. The Panel was involved in industry consultation, the definition of research priorities and project selection. The Panel also met regularly to review the progress of the portfolio after projects had commenced. All BIP projects will be completed by the end of December 1998. 2.2 Aims, Objectives and Priorities The goal addressed by the BIP is to make banana production more profitable and to reduce the use of pesticides. The aim of the program is to increase the productivity of the crop through the development and use of higher yielding, disease-resistant varieties and by reducing the cost of production, especially the cost of pesticide applications. The two key objectives of the BIP are to support research which will: 1. Develop and evaluate a range of improved banana varieties with export potential, incorporating increased productivity and durable disease resistance through conventional and non-conventional breeding techniques; 2. Develop more efficient and integrated disease management practices, especially for black Sigatoka disease. The objectives of the BIP were consistent with the strategies of the FAO/IGB, which included: • • • • • 2.3 Making bananas a more profitable crop for smallholder producers by increasing productivity and reducing the costs of production, especially by improved disease control based on durable disease resistance; Reducing pesticide use, thus both reducing costs of production and limiting the environmental damage and human health risks caused by excessive use of fungicides and nematicides; Improving the reliability of supply of high–quality products; Expanding the export market, including opening up niche markets for speciality bananas; Improving access to markets. Projects Funded Before the projects were called for and selected, the World Bank commissioned a report during 1994/95 to investigate the then current status of banana research and priority areas for the BIP. The report (Buddenhagen, 1996) highlighted the need for combining 165 innovative breeding approaches with advances in plant biotechnology. This influenced the composition of the program portfolio. The projects funded concentrated on banana improvement rather than agronomy or integrated pest management. Both breeding and biotechnical projects were funded. A list of projects funded within the program is provided in Table 2.1. Further details are provided in Annex 2. Table 2.1: Constituent Projects of BIP Project ID PN1 PN2 PN3 PN4 PN5 PN8 PN9 PN10 PN11 PN12 PN14 PN15 Title Field and laboratory evaluation of diploid bananas for their use in breeding schemes Field crosses for understanding the inheritance of black leaf streak resistance in bananas Collection, characterisation and evaluation of Nendran bananas in India Collection, evaluation and characterisation of genetic resources and improvement of banana in Vietnam Banana-breeding in Brazil Novel genes for fungal resistance and post-harvest quality The development of transgenic bananas with resistance to banana bunchy top and banana mosaic viruses Molecular tool box Genetic engineering of ethylene biosynthesis in bananas Use of biotechnology to produce transgenic bananas that are resistant to BBTV infection Elimination of banana streak badnavirus (BSV) from improved Musa germplasm and related studies on transmission and host plant/virus/vector interactions Variability and relationships within populations of Fusarium oxysporum f sp cubense from its centre of origin 166 Research Provider CIRAD-FLHOR, France CIRAD-FLHOR, France Kerala Agricultural University, India Institute of Agricultural Genetics, Vietnam EMBRAPA, Brazil Boyce Thompson Institute for Plant Research, USA QUT, Australia KUL, Belgium Hong Kong University of Science and Technology, Hong Kong University of Hawaii, USA IITA, Nigeria QDPI, Australia PN17 PN18 PN19 PN20 PN21 PN22 2.4 Origin and distribution of fungicideresistant strains of M. Fijiensis in banana plantations in Costa Rica Study of tolerance and resistance of banana in nematodes Identification of durable resistance sources in banana and plantain Identification of durable resistance sources in banana and plantain Identification of durable resistance sources in banana and plantain Breeding of hybrid Musaceae with resistance to multiple diseases, especially black Sigatoka and Panama Disease CORBANA, Costa Rica CRBP, Cameroon KUL, Belgium FHIA, Honduras IRTA, Spain FHIA, Honduras Type of Research Table 2.2 shows the projects presented by project number, title, whether strategic or applied and whether the project was biotechnology or breeding-orientated in nature. A project was defined as strategic when the intended output from the project could not be used directly by industry at the present time, and where the output was intended as an input to further research. On the other hand, a project was considered applied if the outputs could be used directly by industry. Some projects had a mix of both applied and strategic components. Strategic projects usually produce knowledge as their primary outputs. Such information or knowledge can be used in other projects to eventually produce applied outputs which are implementable, by industry. Strategic projects are usually long term in nature. As can be seen from Table 2.2, the portfolio of projects funded by the program is predominantly strategic in nature. 167 Table 2.2: Classification of Constituent Projects of BIP Project ID PN1 PN2 PN3 PN4 PN5 PN8 PN9 PN10 PN11 PN12 PN14 PN15 PN17 PN18 Title Field and laboratory evaluation of diploid bananas for their use in breeding schemes Field crosses for understanding the inheritance of black leaf streak resistance in bananas Collection, characterisation and evaluation of Nendran bananas in India Collection, evaluation and characterisation of genetic resources and improvement of banana in Vietnam Banana-breeding in Brazil Novel genes for fungal resistance and post-harvest quality The development of transgenic bananas with resistance to banana bunchy top and banana mosaic viruses Molecular tool box Genetic engineering of ethylene biosynthesis in bananas Use of biotechnology to produce transgenic bananas that are resistant to BBTV infection Elimination of banana streak badnavirus (BSV) from improved Musa germplasm and related studies on transmission and host plant/virus/vector interactions Variability and relationships within populations of Fusarium oxysporum f sp cubense from its centre of origin Origin and distribution of fungicideresistant strains of M. Fijiensis in banana plantations in Costa Rica Study of tolerance and resistance of banana in nematodes 168 Type Strategic Target Activity Breeding with biotechnology component Breeding with biotechnology component Strategic Strategic with an applied component Breeding Strategic with an applied component Applied with a strategic component Breeding Breeding with biotechnology component Strategic Biotechnology Strategic Strategic Biotechnology Biotechnology Strategic Biotechnology Strategic Biotechnology Strategic Biotechnology and breeding Strategic Breeding and biotechnology Applied with a strategic component Pest and disease control methods Strategic Breeding PN19 PN20 PN21 PN22 Identification of durable resistance sources in banana and plantain Identification of durable resistance sources in banana and plantain Identification of durable resistance sources in banana and plantain Breeding of hybrid Musaceae with resistance to multiple diseases, especially black Sigatoka and Panama Disease Strategic Breeding with biotechnology component Strategic Breeding Strategic Breeding Applied Breeding Another observation was that projects focused more strongly on the first objective of the program (developing the improved banana types) compared with the second objective (more efficient and integrated disease management practices for black Sigatoka). 3. Summary of Principal Outputs of the Program 3.1 Introduction The outputs of the program have been: • • • Increased knowledge that will input into further projects which are targeted towards producing disease-resistant export bananas; Some applied and more immediate outputs that can already be used by industry; Leverage of additional resources into banana R&D, and increased collaboration and international co-ordination in plant improvement R&D on bananas. A summary of the achievements of the BIP in each of these areas follows. 3.2 Strategic Knowledge Strategic knowledge has been the principal output from the program. This knowledge has been in the following areas: • • The collection and characterisation of a wide range of banana germplasm that will allow increased effectiveness of the banana-breeding programs around the world (PN3 and PN4); Increased efficiency of breeding programs through improved testing, screening and typing methods, development of protocols; and use of additional information assembled through use of genetic markers, particularly for disease resistance to 169 • Sigatoka, Panama Disease and nematodes (PN1, PN2, PN14, PN15, PN18, PN19, PN20, and PN21); Improved and refined technology for genetic manipulation, including transformation processes for banana and improved knowledge of disease resistance mechanisms in banana (PN5, PN8, PN9, PN10, PN11,PN12, PN14 and PN18). The use of this knowledge will provide positive outcomes for plantains and all types of banana production whether for cooking, domestic consumption as dessert bananas or for production of export bananas. The application of this knowledge in the future to provide tangible outcomes for producers and exporters is addressed in Section 5. 3.3 Applied Outputs While the program did not specifically target applied outputs, there were a number of areas where the program may have direct industry impacts, albeit not necessarily in the export banana industries. These include projects PN3, PN5, PN15, PN17 and PN22. PN5 and PN22 were projects embedded in the breeding programs of EMBRAPA (Brazil) and FHIA (Honduras) respectively. Black Sigatoka clones produced in the BIPsupported part of the EMBRAPA breeding program are currently being used to assist control of the recent black Sigatoka incursion into the Amazonas region of Brazil. Also, the FHIA attempt to develop disease-resistant bananas with a short stature could result in such types being commercially accepted. Project PN3 evaluated Indian germplasm and the results may have implications for the use of more short-statured bananas and a cost reduction in propping in South Indian production systems. Project PN15 will have direct application by industry in being able to type strains of Fusarium in a given location when considering new banana developments. The finding in Project PN17 concerning the continuation of resistance to chemicals once fungicides are no longer applied is of direct industry relevance. These applied outputs should be viewed as direct benefits from the BIP investment and are considered again later in a description of outcomes. 3.4 Leverage Other investments in banana research have been leveraged by the BIP into the banana realm. This has not only been achieved within many of the 18 projects themselves, but also outside the projects. For example, the establishment of the BIP has leveraged the Flemish Association for Development Co-operation and Technical Assistance (VVOB) 170 and the Belgium Development Agency (BDA) to provide significant funding to the Nematode Consortium through positioning scientific personnel in various breeding programs around the world. This will be ongoing after the BIP is completed. A conservative estimate of the funds contributed to the BIP projects by the host institutions (co-financing) was $2 million. Other instances of leverage and attraction of other resources into banana research include NZ support in Vietnam for post-harvest production studies and additional funds outside the original project contributed by the University of Hong Kong and the University of Hawaii. 3.5 Co-ordination and Collaboration Because of the nature of the program that was established, there was enhanced coordination of effort and increased collaboration in international banana plant improvement. This was particularly evident in relation to what was termed the ‘biotechnology consortium’ (a joint effort between KUL, BTI and QUT). This alliance fostered staff exchanges between the three institutions involved and assisted the transfer and refinement of both particle bombardment and Agrobacterium transformation and regeneration technology. However, in the latter stage of the BIP, this consortium faded in activity due to the pursuit of more specific goals by each research provider involved in the biotechnology initiative within the BIP. The other area in which increased collaboration was fostered was that of nematode screening for resistance through what was termed the ‘Nematode Consortium’. This consortium will continue after the BIP is completed and will be important in the screening and understanding of nematode resistance in the future (de Waele, 1995). A number of other collaborative efforts have been stimulated by the BIP. These include continuing interaction between Australian and Vietnamese scientists and Australian involvement in testing genetic material for a range of breeding programs. Unfortunately, attempts to integrate the classical breeding approaches in Honduras with molecular aids to breeding have not been successful. Integration between the FHIA approach to breeding and that of the French breeding program would have been beneficial to both areas of endeavour. The BIP was successful, through the competitive grant approach, in attracting new research providers into banana improvement. Moreover, due to the necessity for many research institutions to attract external funding, such as through the BIP, the attraction of more resources into banana plant improvement was achieved. While some of the resources attracted into BIP projects would have probably been committed to banana improvement anyway, it is likely that significant additional resources were encouraged into banana improvement due to the BIP. 171 Two examples suffice: • • 3.6 CIRAD have now secured an additional permanent position for banana improvement due to the program; Generic resources provided by VVOB in Belgium have been committed to banana improvement through the Nematode Consortium Conclusion The BIP has provided a unique set of outputs in terms of both content and process. The BIP has produced knowledge of a strategic nature as well as some technologies that have been, and can be, applied in the short term. In terms of process, the competitive bid system for identifying the best projects to fund within a strategic vision has proved successful, as has the leverage of resources into the banana R&D world, as well as increased collaboration and co-ordination among research teams. 4. Summary of Actual and Potential Outcomes and Benefits 4.1 Introduction Section 3 has provided a brief summary of the reported outputs of the program up to November 1998. There are still a number of projects to be completed due to a range of reasons and a number of research providers will be continuing with research instigated under the BIP. It will be important for such future outputs to be reported even after the official end of the BIP in 1999. How this will be achieved is uncertain, but this issue highlights the unsatisfactory nature of short-term funding such as provided by the BIP. Outcomes of the BIP are defined as the activities or events that have occurred, or are likely to occur, as a result of the outputs from the BIP. Outcomes are directly relevant to producers of bananas and to others in the banana value chain including consumers. Outcomes can be positive or negative and are usually associated with benefits and costs. These benefits and costs may fall differentially on the various participants in the banana value chain. Outcomes can be short term or long term. Short-term outcomes are often preferable, but not always easy to achieve in agricultural research. Outcomes can also be classified into those of content (products, technologies, and knowledge), and those of a structural nature (e.g. changed priorities and resource allocation, attraction of additional resources). The following describes the outcomes and associated benefits and costs that have arisen due wholly or partially to the BIP. 172 4.2 Short-Term Outcomes and Benefits Short-term outcomes and benefits have arisen from the BIP. Two important outcomes were: • • The ability to type Fusarium strains before locating new banana developments; The Brazilian genetic material used in Amazonas to combat the spread of black Sigatoka as it enters Brazil . Typing Fusarium strains The world-wide collection of the various strains of Fusarium will serve as a reference collection in the future, enabling the strains causing an outbreak of Panama disease to be quickly identified. New banana developments can be more safely implemented through determining the strains present in any given location before planting begins. For example, recent outbreaks of Fusarium Wilt in Cavendish plantations in the eastern tropics (peninsular Malaysia, Sumatra and Halmahera) caused enormous damage. These outbreaks have ominous implications for the durability of western trades that are based on Cavendish clones. The avoidance of future losses will be one of the shortterm benefits from the BIP. Brazilian material Some banana clones resistant to black Sigatoka have been produced in Brazil with assistance from the BIP and are being used to slow the rate of advance of the disease in northern Brazil. This would have a major reduction in the costs of those who can potentially afford to spray with fungicides as well as reduce yield losses of those who cannot afford spraying control technologies. 4.3 Long-Term Outcomes and Benefits While the short-term outcomes and associated benefits are likely to be significant, the most important outcomes and contributions from the BIP are likely to be derived from the strategic knowledge contributions. Outcomes associated with knowledge generated from the BIP are inherently uncertain. However, in terms of assessing the impact of the BIP, it is necessary to make some assumptions about the future and the role that the BIP may play in the future development of the export banana industry. The key events shaping the future use of knowledge from the BIP with respect to export banana industries would include the development (in full or in part) of a Cavendish or alternative export banana type that is resistant to the two principal fungal diseases of black Sigatoka and Panama Disease, as well as to nematodes and viral diseases. 173 These events may utilise both conventional breeding technologies and biotechnologies such as genetic markers or genetic engineering. The key uncertainties associated with events of this type would include the following: • • • • • • • • • The number of years until they will occur; The annual investment required for them to occur within a given period with a specific probability of research success; The applicability of the resistant material (across pathogens and strains within a pathogen); The durability of resistance inherent in the new material; The level of use of chemicals still required; Any interaction or enhancement with cultural techniques including integrated pest management; The level of maintenance of resistance through breeding/biotechnology required on an ongoing basis after the events are realised; The cost and level of availability to industry of any new material produced; The levels of adoption of new export banana types. There are still serious constraints to be able to effectively introduce disease-resistant genes and have them expressed in banana plants. A major constraint is still the degree of somaclonal variation in the recipient banana material. The cell target tissue needs to be able to be produced in vitro fast and effectively to reduce the degree of variation. Promoters need to be more specific so that they are not expressed in all tissue types including the fruit. Specific promoters that are only expressed in certain tissues (e.g. roots) or when an infection is present are required. However, it is not known how precisely specific promoters need to be. Target genes need to be identified and evaluated so that they can be combined in a manner that confers robust protection. Finally, it is not known how strongly expression of the gene in the laboratory will be related to resistance in the field. These technical issues will influence the time, resources and probability of success of the events actually occurring. For example, time and resources will vary depending on whether a short-term solution is assumed or one of a more durable nature incorporating multiple sources of resistance. There are also considerable commercial constraints to the development of transgenic or disease-resistant bananas. One of these is the intellectual property issue. Viral transfer constraints (banana streak virus) (Frison and Sharrock, 1998), and the need to test against multiple strains of black Sigatoka will also reduce the rate of adoption of any transgenic banana across the whole export industry. 174 In addition, there are legal issues that might retard progress in introducing a transgenic banana into the field. These may be related to consumer concerns about genetically engineered food crops. However, the area of genetically engineered food crops is increasing every year and it is likely that concerns will dissipate with education and clear labelling. The reduction in chemical use associated with the production of genetically engineered bananas, if promoted, is likely to outweigh any negative perceptions. If such a transgenic banana is not a Cavendish type, then there may be additional difficulties of market development and a slower rate of adoption throughout the industry. The greater the investment in biotechnology R&D, the broader and more robust may be the resulting form of resistance. Pathogens can become resistant to gene products, especially if the transgenic banana plants are used intensively and the genes bestowing the resistance properties are not broad based in their action. Combining different forms of resistance in one or more transgenic plants, combined with the use of some chemical spraying, is likely to provide a more enduring strategy. In any case, the extent of spraying is likely to be considerably reduced. On the other hand resistance against Fusarium is likely to be more robust as the fungal organism in this case has less chance of recombination due to its asexual form of propagation. While Sigatoka-resistant genetic material should result in significant chemical savings and, in the case of production for domestic markets, significant quality improvements, the technology is not likely to be made available outside the private interests that are most likely to develop the clones in the first instance. If the development is through the public sector there will still be the royalties that need to be paid to the owners of the embedded technologies. This may result in significant costs for the resistant planting material. Any increase in the cost of the planting material is still likely to be less than the chemical costs saved, thus conferring a net benefit to the industry. Additional benefits to society should result from the reduced impact of chemicals on the environment as well as the reduced impact on human health. The following provides a qualitative description of the type of long-term benefits that might be attributed in part to the program. The benefits are classified as economic, environmental, and social and health benefits. 4.3.1 Economic Benefits The principal benefit most likely to be associated with such key events as specified earlier is the reduced cost of production of export bananas. This might be achieved through: • • Reduced chemical spraying costs; Reduced importation costs of chemicals in banana-exporting countries; 175 • • • • Higher yields (which can be translated into a cost reduction); Higher quality fruit resulting in less wastage along the value chain (which can be translated into a cost reduction); Reduced other control costs, such as saved rouging costs in the case of bunchy, top or lowered replanting and relocation costs in the case of Panama Disease; The ability to reclaim land lost for banana production due to disease. The economic benefits should accrue to both large plantations producing for export as well as smallholder banana producers producing for local markets. Smallholders often cannot afford chemical treatments so losses due to disease such as black Sigatoka can be significant. With higher productivity, less land may be required to produce a given level of banana production or more bananas can be produced, both of which will result in improved profitability to individual producers, regions and countries. The replacement of chemicals is required due to the ever-increasing costs of chemical control. Some chemicals have been, or are being, banned from use. Resistance to chemicals is increasing and new chemicals are needing to be developed which involves more costs. Other benefits can flow on from a more profitable banana export industry, as the sector is very important in a number of countries in terms of its contributions to export income. Improved consumer confidence and a higher level of food security in banana consuming countries from less use of chemicals can also provide benefits to banana-producing and exporting countries. 4.3.2 Environmental Benefits The principal environmental benefits associated with the BIP are those resulting from the use of lesser amounts of pesticides applied if resistant or more tolerant varieties of export bananas are used or if a more efficient chemical regime is identified. The lower the pesticide applications, the lower the potential damage to the environment. The damage to the environment has not been verified or precisely quantified for the whole of the banana export industry around the world, but there have been various statements made in relation to specific countries that suggest that at least some negative impact is occurring (Andreatta, 1997; Colburn 1997). In addition to ecosystem damage, and the possible reduction in biodiversity, there is one instance of a potential causal negative impact on another export industry through fungicide runoff into prawn farming locations in Ecuador (Colburn, 1997). Short-term impacts on the environment are serious but can be dealt with more easily than potential long-term impacts which are uncertain and less focused and disparate. In both cases, there is difficulty in dealing with these issues in terms of government policies in many banana-exporting countries due to the conflict between economic development goals and environmental sustainability. 176 The objectives of the BIP pertaining to developing disease-resistant types requiring less chemicals takes on even more importance if the political and government policy vehicles are incapable of resolving such issues. Hence, benefits from even partial success in developing resistant banana varieties should have a positive environmental impact through: • • • • • 4.3.3 Reduced impact of chemicals on industries relying on water quality in rivers and estuaries (e.g. fishing, prawn farming); Reduced impact of chemicals on other agricultural industries; Improved conservation of ecological systems in banana-producing countries with associated amenity and aesthetic benefits; Improved environment for tourism activities; Lesser need for increased land areas for bananas so potentially conserving native forests. Social and Health Benefits As with the environment, a range of statements have been made that fungicides and nematicides used in banana production are associated with negative health impacts on banana plantation workers and those living in banana-producing communities. Again this is claimed by many to be due to the use of various chemicals, and particularly those applied through spraying where overuse and misuse is reported as occurring. For example, Swennen (1997) claims that people associated with banana production have a 10% higher chance of damage to their vital organs. Both Andreatta (1997) and Murray (1994) suggest that chemical use in banana production is associated with human health problems. Some of the perceived benefits related to health that would arise from developing disease-resistant bananas would be improved health on the part of banana industry workers and higher labour productivity, as well as improved health and quality of life of communities in banana-producing regions. Other social improvements from increasing productivity in the banana industry are the maintenance of, or increase in employment in banana-producing and exporting countries. 4.4 Process and Structural Outcomes and Benefits A number of process benefits have arisen from the BIP investment. As identified in Sections 3.4 and 3.5, there has been leverage of additional resources into banana research as well as improved co-ordination and collaboration between research providers due to the BIP. Both these outputs should contribute to more effective 177 investment in banana improvement in the future and should result in lessening the time that it might take to produce the disease-resistant types in the future. 4.5 A Framework for Analysis The following framework has been provided as a guide to analysing the value for money obtained by the Common Fund for Commodities (CFC) in its investment in the Banana Improvement Project. Effecting an investment analysis for a program of what is really strategic research is not a simple or easy undertaking. However, it is possible to effect a quantitative analysis using some of the benefits that were described in Section 4.3. Clearly there is a wide range of scenarios for the future. Assessing the impact of the past BIP investment will require assumptions about the future and needs to take into account the uncertainty concerning the success and timing of any scenarios postulated. Potential gross benefits are those benefits that can be derived from the outcomes such as the development of disease-resistant export bananas. Additional costs involved in realisation of defined potential gross benefits need to be subtracted from the potential gross benefits. For example, additional costs of planting material that might arise due to intellectual property rights for transgenic material, or due to the additional value chain costs of exporters changing to alternative export banana types, would need to be subtracted from the production cost reductions that might occur due to resistant banana types being introduced. Once additional costs incurred are subtracted, the remaining benefits can be termed potential net benefits. Potential net benefits need to be assessed relative to what would have been likely to have happened to industry costs in the event of these positive outcomes not having taken place at some stage in the future. It is reasonable to assume a ‘no intervention’ strategy where there is no R&D investment. In this case, assumptions would have to be made about escalating costs of chemical control, banning of nematicides, or increasing environmental impacts over time. The term ‘potential net benefit’ is used to signify that the benefits may not necessarily be realised. When multiplied by some estimate of the probability of the event actually occurring then the resulting parameter may be termed ‘expected net benefits’. Expected net benefits are those that might be realised by the industry and consumers of export bananas. The distribution of benefits to producers, exporters, retailers and consumers is not part of the framework specified here. Such distributional aspects require a more sophisticated framework and more detailed assumptions than can be made in the current impact assessment. Expected net benefits may not be fully realised if the technologies produced are not applicable to all geographic regions or all export banana industries, all pathogen strains 178 etc. The actual benefits that might be ascribed in part to the BIP will depend on how the technologies are actually adopted by industry. In other words, even though expected net benefits to multinational companies might be positive, there may be other constraining factors that might retard or negate the technologies being adopted by industry. Hence the rate of adoption of any ensuing technologies needs to be taken into account. A quantitative analysis using this framework is provided in Section 5. 5. Investment analysis 5.1 Approach In order to evaluate the past investment in the BIP in quantitative terms, it is necessary to construct future scenarios that assume that the particular knowledge generated from the BIP is used in further R&D efforts that are eventually successful with the resulting technologies or products being adopted by industry. This requires a number of assumptions to be made about the type of future investment and associated costs, the time period before disease-resistant bananas will actually be produced, and the likelihood that the technologies will be produced within the timeframe assumed. A part of the benefits that might arise in the future can then be attributed to the BIP and the remaining benefits to the non-BIP investment. Assessing the relative contributions is difficult. One way is to ascertain the proportion of the BIP investment to the total investment that will be required and attribute benefits to the BIP according to that proportion. This attribution method was the principal method used in the investment analysis described in this section. Another way in which the likely returns of the BIP investment can be estimated is through an assumption that the BIP ‘sped up’ the delivery of the final products that are achieved. For example, it could be assumed that the development and commercialisation of a disease-resistant export banana would have occurred in the year 2017, and that the additional investment and impetus provided by the BIP would allow it to occur in 2012. An exercise using this method has also been used in the present analysis and is hereafter known as the ‘earlier impact’ method. In this case it is assumed that investment in R&D would have occurred without the BIP but that the BIP contributed to the effectiveness of the mainstream R&D investment through speeding up of the realisation of scenarios envisaged, that is, the benefits from the introduction of the transgenic and alternative banana would have been received earlier. The two key future scenarios specified earlier were: 1. The production of a transgenic Cavendish that is reasonably resistant to black Sigatoka and nematodes; 179 2. The development of export banana types that are resistant to black Sigatoka and nematodes. This is assumed to be achieved through a combination of transformation and regeneration technologies together with breeding components and has resulted in a robust form of resistance. For the application of the attribution method, each of these scenarios is analysed in terms of potential benefits and costs that might be expected ‘with’ and ‘without’ the R&D program. In the case of benefits for the ‘with’ R&D situation, these are principally the reduction (not necessarily elimination) in chemical costs and any reduction in yield losses from improved disease control. These benefits may differ in each of the two scenarios analysed. In the case of costs for the ‘with’ R&D situation, these will include the costs of R&D to develop the disease-resistant bananas. In the case of the necessity to move away from Cavendish (Scenario 2), it will also be necessary to include the cost of producing and exporting systems for changing to another banana type. These benefits and costs for the ‘with’ R&D situation are then compared with the ‘without’ R&D situation where, while there are no R&D costs, there are assumed to be future increases in chemical costs and yield losses. In the case of the ‘earlier impact’ method, the costs of the BIP are placed against the difference in benefits expected in two situations. The benefits in the ‘with BIP’ situation were assumed to commence in 2012, but in the ‘without BIP’ scenario, the benefits were assumed to commence later. It is the difference in the benefits between these two situations that is assumed to be the benefit due to the BIP investment. 5.2 Assumptions Assumptions used in the analysis using the attribution method are summarised in Table 5.1. Table 5.1: Assumptions Used in Investment Analysis for BIP Item Without and R&D Program Annual banana exports (million t) 12.6 (a) Banana yield (t/ha) 40 (b) Export banana production costs ($/t) 169 (c) Cost of fungicides ($/ha/annum) 1150 (d) Cost of nematicides ($/ha/annum) 225 (d) Yield loss due to black Sigatoka with efficient chemical 10.0 (d) control (%) Yield loss due to nematodes with efficient chemical control 17.5 (d) (%) 180 Annual increase in cost of fungicide (% per annum) Annual increase in cost of nematicide (% per annum) Resource cost of BIP ($ million per annum) (1994-1998) Other resources devoted to banana improvement worldwide (scientist years per annum) (1994-1998) Other resources devoted to banana improvement worldwide (scientist years per annum) (1999-2012) Resource cost of a scientist year ($ per annum) Year in which transgenic disease-resistant Cavendish commercialised and adopted by growers Probability of development and commercialisation of transgenic Cavendish (%) Reduction in cost of fungicides used per ha after transgenic Cavendish introduced (%) Reduction in yield loss due to black Sigatoka after transgenic Cavendish introduced (%) Reduction in cost of nematicides after transgenic Cavendish introduced (%) Reduction in yield loss due to nematodes after transgenic Cavendish introduced (%) Maximum adoption level of transgenic Cavendish (%) Year in which new disease-resistant banana type commercialised and adopted by growers Probability of development and commercialisation of new disease-resistant banana type (%) Reduction in cost of fungicides used per ha after new banana type introduced (%) Reduction in yield loss due to black Sigatoka after new banana type introduced (%) Reduction in cost of nematicides after new banana type introduced (%) Reduction in yield loss due to nematodes after new banana type introduced (%) Maximum adoption level of new banana type (%) Cost to export banana industry of changing banana type ($ million) 181 1 (e) 1 (e) With and R&D Program 0.7 (f) 30 (g) 35 (e) 180 000 (h) Outcome: Scenario One (Transgenic Cavendish) 2012 (I) 50 (I) 60 (I) 30 (I) 60 (I) 30 (I) 80 (I) Outcome : Scenario Two (Alternative Banana Type) 2012 (i) 50 (i) 80 (i) 50 (i) 80 (i) 50 (i) 50 100 (i) (j) (a) CIRAD (1998) (b) Estimated from CIRAD statistics for major exporting countries (c) Estimated from figures supplied by Corbana, Costa Rica (d) Mid-point of range supplied by Corbana, Costs Rica (e) Estimate by Agtrans Research (f) Estimate of annual investment by the BIP (g) Estimate by Agtrans after discussions with CIRAD (h) Estimate by Agtrans after discussions with Advisory Panel (i) Estimate by Agtrans and Advisory Panel (j) Estimate by Advisory Panel; this cost will depend on what level of existing banana production will need to be ploughed out due to the changeover, the degree to which the new banana type resembles Cavendish for post-production activities, and whether, and to what degree, market research and promotion is required. The cost of applying for and holding patents was assumed to be only 1-2 million dollars and was not included in the analysis due to its relative insignificance in comparison with other costs. However, the cost of royalty payments was considered potentially significant. Such royalty payments could be considered part of the total benefits estimated in the analysis. Assuming such royalty payments do not inhibit the adoption of the technologies, the total benefits will remain the same, but may not accrue solely to banana producers, exporters or banana consumers. No estimate has been made of the size of any royalty payments or the distribution benefits between the banana export industry, and research organisations or biotechnology companies. Further, there has been no attempt to distribute the total benefits that might accrue between banana producers and consumers. The analysis was effected through a discounted cash flow analysis that ran for a period of thirty-five years, commencing in the 1994 calendar year. All costs and benefits were discounted back to 1994 using a discount rate of 7%. All values were expressed in 1998 US dollars. Investment criteria estimated included the Net Present Value (NPV), the Benefit-Cost Ratio (B/C Ratio), and the Internal Rate of Return (IRR). Criteria were estimated for each of the two scenarios separately. Sensitivity analyses were carried out to assess how investment criteria changed, with changes in the value of a range of assumptions including adoption level, probability of development and commercialisation, and the reduction in the fungicide and chemical costs after the disease-resistant type is developed. In the case of the ‘earlier impact’ approach, the basic assumption was that the BIP caused the impact to occur two years earlier than without the BIP. However, a sensitivity analysis was carried out on the length of this time period. 182 5.3 Attribution of Benefits to BIP Attribution of benefits to the BIP are made on the basis of the present value of costs of the BIP in relation to the present value of other funding available from 1994 to 1998 as well as that required from 1998 onwards to achieve the outcomes specified. It is assumed that 30 scientist years per annum were contributed outside of the BIP over the period 1994 to 1998. In addition, it is assumed that a further 35 scientist years per annum will need to be committed between 1998 and 2012 for the two scenarios specified in the assumptions table to be achieved. Valued at $180,000 per scientist year (full cost), this assumption means that the investment in the period 1998 to 2012 will be of the order of $6.3 million per annum and will include both private and public sector investments. 5.4 Results Using the Attribution Method 5.4.1 Results for Scenario 1 – The Transgenic Cavendish The investment criteria estimated using the base assumptions for Scenario 1 are shown in Table 5.2. Table 5.2: Expected Investment Criteria for Scenario 1 (Transgenic Cavendish) (Discount rate 7%) Total Investment BIP Investment only NPV ($ million) 578 27.1 B/C Ratio 9:1 10:1 IRR (%) 21.4 21.9 The results show that, even with the conservative assumptions used, there is an expected positive NPV from the BIP investment of $27 million, with a B/C Ratio of 10:1. This is despite the time lag of 18 years assumed between the commencement of the BIP and the introduction of a disease-resistant Cavendish banana. The $27.1 million NPV for the BIP investment could be larger if a higher level of benefits from the transgenic Cavendish is attributed to the BIP. The leverage of additional resources into banana improvement as a result of the BIP has not been included in the current attribution; in addition, the influence of the BIP on the degree of interest from the private sector has been significant, but the implications of this influence have not been included in the attribution of benefits made to the BIP. The estimation of benefits is conservative in two ways. First, the actual parameters used in the quantitative analysis could be considered to be conservative. Second, there are a number of other benefits outlined in earlier sections that have not been included in the quantitative analysis. Environmental and human health benefits are two such benefits, which are usually difficult to value but are likely to be high and significant. Such benefits could also provide justification for public investment in banana improvement since 183 benefits could be captured by communities as opposed to the private sector in bananaexporting countries. Table 5.3 shows the sensitivity of investment criteria to small chemical cost increases that are likely due to an increasing amount of chemicals applied to control banana diseases as well as the increase in per unit costs of chemicals due to the need to continually develop new products to overcome chemical resistance. While the influence of the rate of increase of this cost is significant, it would have been even greater if not for the lag between the BIP investment and the likely introduction of a disease-resistant variety. Up until the year 2012, the cost increases are assumed to apply in both the ‘with’ and ‘without’ R&D situations so there is no benefit to the R&D until the year 2012. It is only after the year 2012 that a difference is apparent. Table 5.3: Sensitivity of Investment Criteria to Rate of Chemical Cost Increase Rate of increase (% pa) 1 (base) 2 3 5 10 NPV ($ million) B/C Ratio 27.1 31.3 36.4 50.5 125.5 10:1 11:1 13:1 17:1 42:1 Table 5.4 shows that the investment criteria are highly sensitive to the probability of success of the R&D. The base level probability assumed was 0.5. The break-even probability (where the BIP investment would break even at a 7% discount rate) was 0.05, indicating that the chances of development and commercialisation do not have to be very high in order for the investment to provide a worthwhile expected payoff. Table 5.4: Sensitivity of Investment Criteria to Probability of Success Probability of success by 2012 0.10 0.25 0.50 (base) 0.75 1.00 NPV ($ million) B/C Ratio 3.0 12.0 27.1 42.2 57.3 2:1 5:1 10:1 15:1 20:1 Table 5.5 shows that the maximum level of adoption of the transgenic Cavendish needs to be above about 8% before the investment will break even. Since the changeover costs of adopting a disease-resistant Cavendish are likely to be minimal, as the changeover could be phased in over a period, it is likely that the adoption level among banana-exporting producers will be quite high. This is reflected in the base assumption used, which was 80% adoption. 184 Table 5.5: Sensitivity of Investment Criteria to Level of Adoption Level of adoption (%) 10 25 50 80 (base) 100 NPV ($ million) B/C Ratio 0.7 6.4 15.8 27.1 34.7 1.2:1 3:1 6:1 10:1 12:1 The level of reduction of chemicals due to the use of a transgenic Cavendish is difficult to assess without further assumptions about the source of resistance and the mechanisms involved in conferring such resistance. It is assumed that at least some chemicals will still need to be used. The base level reduction was assumed to be 60% of what the use would have been without the transgenic Cavendish. As illustrated in Table 5.6, the investment criteria were very sensitive to changes in this assumption. The NPV is still positive even if there is no reduction in chemical usage, due to the assumption that yield losses will still be reduced. Table 5.6: Sensitivity of Investment Criteria to Level of Reduced Chemical Usage Reduction usage 0 10 30 60 (base) 100 in chemical NPV ($ million) 8.7 11.8 17.9 27.1 39.4 B/C Ratio 4:1 5:1 7:1 10:1 14:1 Investment criteria changed significantly when the assumption concerning the reduction in yield loss due to the transgenic Cavendish was varied (Table 5.7). Table 5.7: Sensitivity of Investment Criteria to Reduced Yield Losses Reduction in level of yield loss (%) 0 10 30 (base) 60 100 NPV ($ million) B/C Ratio 15.3 19.4 27.1 37.8 50.5 6:1 7:1 10:1 13:1 17:1 185 5.4.2 Results for Scenario 2 – The Alternative Banana Type The investment criteria estimated using the base assumptions for Scenario 2 are shown in Table 5.8. Table 5.8: Expected Investment Criteria for Scenario 2 (Alternative Banana Type) (Discount rate 7%) NPV ($ million) Total Investment 484 BIP Investment only 22.8 B/C Ratio 8:1 8:1 IRR (%) 20.0 20.5 The results show that, again with the conservative assumptions used, there is a highly positive expected rate of return from the BIP, with a B/C Ratio of 8:1. The main difference between this future possibility and the transgenic Cavendish outcome is that Scenario 2 assumes that there is a cost to the industry of changing the banana type away from Cavendish. The magnitude of this cost is difficult to estimate as it will depend, in part, on the similarity of the new export banana type to Cavendish. Depending on the market development strategy and how quickly change is accommodated, there may be some replanting costs (additional replanting costs to what would have been the case with continuing with Cavendish). But handling, storage and shipping facilities may also need to be altered and the associated costs could be significant. While the base assumption is a cost of $100 million, this variable is subject to a sensitivity analysis. Results are shown in Table 5.9. The break-even changeover cost that could be sustained and still provides a 7% return on R&D investment was estimated at $1.75 billion. Table 5.9: Sensitivity of Investment Criteria to Changeover Cost Changeover cost ($ million) 100 (base) 500 1000 2000 NPV ($ million) B/C Ratio 22.8 17.2 10.3 -3.5 8:1 7:1 4:1 -0.14:1 The other two major differences between Scenarios 1 and 2 were: • • The maximum adoption level was assumed to be lower for Scenario 2; The reductions in chemical costs and yield losses were assumed to be higher in Scenario 2. Most other sensitivity analyses carried out for the new banana type scenario showed similar trends to those for Scenario 1. 186 The analyses to date have assumed that either one or the other of Scenarios 1 and 2 could occur. Although it is possible that both Scenarios 1 and 2 could occur, to assess the combined effect would require recasting the assumptions regarding adoption rates. 5.5 Results Using the Earlier Impact Method 5.5.1 Results for Scenario 1 – The Transgenic Cavendish Results for the investment analysis using the ‘earlier impact’ method and assuming an earlier impact of the BIP of two years (Scenario 1 occurring in 2012 instead of 2014) are shown in Table 5.10. Table 5.10: Expected Investment Criteria for Scenario 1 (Transgenic Cavendish) (Discount rate 7%) BIP Investment NPV ($ million) 120 B/C Ratio 40:1 IRR (%) 33 A sensitivity analysis of investment criteria to the time period for the earlier impact is shown in Table 5.11. Table 5.11: Sensitivity Analysis of Investment Criteria to Time Impact of BIP Difference in impact timing (years) 1 2 (base) 3 5 5.5.1 NPV ($ million) B/C Ratio 61 120 175 275 21:1 40:1 58:1 90:1 Results for Scenario 2 – The Alternative Banana Type Results for the investment analysis for the ‘earlier impact” method for the alternative banana type scenario are shown in Table 5.12. This result is based on an assumption of a two-year speeding up of the realisation of benefits. Table 5.12: Expected Investment Criteria for Scenario 2 (Alternative Banana Type) (Discount rate 7%) BIP Investment NPV ($ million) 104 187 B/C Ratio 35:1 IRR (%) 31% 5.6 Conclusion Given the assumptions made in the analysis, the results reported in this section suggest that the BIP could be expected to provide a rate of return on funds invested of over 20%. Conservative assumptions have been used throughout the analyses, and the sensitivity analyses and break-even parameters estimated show that the results are robust across a credible range of assumptions. The benefits estimated from this analysis are only a part of the total benefits that would accrue to the investment in banana improvement and, in this regard, the results are likely to be an underestimate of the performance of the investment. The attribution method of analysis gave a lower expected rate of return than the ‘earlier impact’ approach, even when the impact of the transgenic Cavendish or the development of the alternative banana type was assumed only one year earlier (2012 instead of 2013) due to the BIP. This is not surprising since under the attribution method, the BIP’s share of the benefits from the total investment was less than 5%, whereas with the earlier impact method the full benefits of the total investment for two years were credited to the BIP investment. 6. Conclusion The BIP has provided the methods for transformation and regeneration of bananas and has moved the world banana industry closer to the reality of a transgenic banana. Transformation and regeneration have now been accomplished. There has also been significant progress demonstrated in terms of new germplasm for use in breeding programs and methods for making breeding programs more effective in the future. The most important output from the BIP was strategic knowledge, which can be used in further R&D projects in the future. Short-term and long-term benefits, and both content and process benefits, are likely to be derived from the BIP. The long-term benefits include those related to reduced costs of production, the environment and health. Some of the economic benefits were quantified and used as input to a benefit-cost analysis of the BIP investment. The results of the investment analyses demonstrate that the expected payoffs from development of disease-resistant bananas can justify a large investment in banana improvement research. Assumptions were generally conservative and only limited sets of benefits were included in the analysis. Based on the various assumptions used, the expected rate of return was estimated at 20% to 33%, the NPV at $ 22 to $ 120 million and the B/C Ratio at 8:1 to 40:1. 188 References Andreatta, S. L. (1997) ‘Bananas, Are They the Quintessential Health Food? A Global/Local Perspective’, Human Organisation Vol. 56, No 4, pp. 437-449. Buddenhagen (1996) ‘Banana Research Needs and Opportunities’. In Persley G. J. and George P. (eds) ‘Banana Improvement Research Challenges and Opportunities’, World Bank, USA, pp. 1-20. CIRAD (1998) ‘Banana World Production and Trade Estimates’, FruiTrop, No 51, Oct, CIRAD-FLHOR, France. Colburn, F. D. (1997) ‘Shrimp or Bananas’, Journal of Business Research, 38, 97-103. De Waele, D. (1995) ‘Musa Hematologists’ Consortium: A Step Towards Global Collaboration’, INIBAP Annual Report, Montpellier, France. Frison, E. A. and Sharrock, S. L. (eds) (1998) ‘Banana Streak Virus: A Unique Virus Musa Interaction’, Proceedings of a Workshop of the ProMusa Virology Working Group, held in Montpellier, France, INIBAP. Murray, D. L. (1994) ‘Cultivating Crisis: The Human Cost of Pesticides in Latin America’, University of Texas Press, Austin, USA. Speijer and De Waele, D. (1997) ‘Screening of Musa Germplasm for Resistance and Tolerance to Nematodes’, INIBAP Technical, IPGRI. Swennen, R. (1997) ‘Breeding our Future’, Second National Banana Industry Congress, Gold Coast, Australia. 189 ANNEX 1: TERMS OF REFERENCE To prepare an impact assessment statement on the results of the contracted research undertaken by the BIP portfolio of projects on the future production techniques and costs of the export banana industry. In undertaking this objective, the consultant will: • • • • • Analyse and summarise program and project documentation; Develop a framework for benefit-cost analysis for the program; Visit project locations as appropriate for consultation with Principal Investigators of individual projects; Visit the World Bank Office in Washington DC to meet with the Project Coordinator to discuss the report and access further documentation and write the initial report; Interact with members of the BIP Scientific Advisory Panel. 190 ANNEX 2: SUMMARY AND IMPACTS OF INDIVIDUAL PROJECTS PN1: Field and laboratory evaluation of diploid bananas for their use in breeding schemes CIRAD-FLHOR, France Introduction This project focused on producing basic information on diploids in order to increase the efficiency of breeding programs. The project characterised new genetic material as well as material that was already used in breeding programs, but was not well characterised. The additional information provided by this project can be used to select parents for crossing that are more likely to provide improved clones with desirable characteristics. The information assembled included information from both the field and laboratory. Field data assembled included agronomic and fruit characteristics, morphological data and disease resistance. With respect to disease resistance, material was tested for black Sigatoka in New Caledonia, for Panama Disease in Australia, and for yellow Sigatoka in St Lucia. Laboratory data included information on ploidy level and molecular markers. Results Information for about 150 clones has been assembled from the project. Assembly of agronomic and morphological data and the information on ploidy levels and molecular markers will be completed by the end of calendar 1998. However, the resistance information will not be completed by that time. This is because there were delays in obtaining and multiplying the material for the field trials. In addition, not all clones arrived at the same time in some locations, and problems in growing plants from tubes occurred in one location. These constraints led to only 150 clones being tested (original intention was 200) and were the main reasons for the delayed production of results. Overall these problems, as well as the necessity to index for viruses, demonstrated the difficulties in establishing and implementing international network activities. It is likely that all disease resistance information will be available by about June 1999. The information will be provided to all banana breeders through the MGIS (Musa Germplasm Information System) sponsored by INIBAP. Impact While the information is provided on diploids only, it can be useful for all breeding programs, as diploid material has to be used in all programs including the classical breeding approach. Nevertheless, the information is likely to be most useful in the CIRAD and CRBP programs and will most certainly be used in such programs. There is also some chance that the information will be used in the EMBRAPA, IITA and FHIA programs. While additional information produced will refer in some instances to dessert bananas, the evaluation showed that the new material was more related to cooking 191 bananas. Where the information is used in the selection of diploid parent material for crossing, it is likely that such use will increase the probability of producing improved clones from any particular crossing. This is likely to increase the total amount of improved material produced over a given time period, or alternatively, decrease the time to produce a hybrid of a particular type that has been targeted. It has been conservatively estimated by CIRAD that the use of the information could result in ten times the number of improved dessert type bananas, compared with breeding strategies that did not utilise the additional information. This may mean that for any given set of resources applied to breeding it may take only one tenth of the time to reach a specific target compared to non-use of the information. However, it should be recognised that there are different philosophies in breeding programs and that the information, while relevant to all breeding programs, is particularly targeted at the CIRAD program. A breeding cycle for bananas is approximately five years. What may be achieved in five years with the information, therefore, may have taken up to 50 years without information. There will also be a cumulative effect over time if further information is assembled as it is produced from crosses made. The value of the information produced from this BIP project will depend to a large extent on the success of the CIRAD breeding program itself. While this success is not guaranteed, there are signs that the program will be able to produce dessert bananas that may be suitable for export. Again the issue is one of time and how significant constraints on market development and adjustment along the value chain for alternative Cavendish types can be overcome. PN2 : Field crosses for understanding the inheritance of black leaf streak resistance in bananas CIRAD-FLHOR, France Introduction This project focused on understanding the basis of the resistance of banana types to black Sigatoka. This entailed the testing of resistance in the field of crosses of specific wild types of bananas for which resistance characteristics were already known, and correlating resulting observations with molecular studies in the laboratory. The idea was to use the resulting information to use molecular markers to quickly determine the likely resistance levels of germplasm that might be used in future breeding strategies (marker assisted selection). 192 Results Due to environmental interactions in the field, it was difficult to accurately assess the degree of resistance of crossed material to black Sigatoka. This required the development of a leaf test (SLP test) that could be carried out under controlled conditions. This test allowed the determination of the level of resistance for each cross to be made. The necessity to develop this test within the project delayed the progress of the project. Molecular markers for highly resistant material will be in place by the end of calendar 1998. Highly resistant material appears to be controlled by a recessive gene so making it difficult to introduce to new plant material through breeding programs. The work on partial resistance will not be completed by the end of calendar 1998. The completion of this work may take another 12 months, but should be completed outside of the project by the end of calendar 1999. This is because finding molecular markers for the set of genes involved will be quite difficult, but this information will be critical. The major output from the project will be increased knowledge about the mechanisms involved in resistance in bananas to black Sigatoka. This may allow a higher rate of progress in producing black Sigatoka clones in breeding programs for both cooking and dessert type bananas. Further, the knowledge may allow the identification of natural resistance genes or gene segments that may be used in transformation processes. Impact Molecular markers are not yet available to screen for resistance, but should be available by the end of the 1999 calendar year. There will probably be a low probability of being able to confer total resistance to black Sigatoka through breeding programs; this may not be desirable in the long term anyway due to the ease of the black Sigatoka agent in overcoming resistance to a single gene. However, prospects for conferring partial resistance appear bright if and when molecular markers that correlate with resistance are identified. This outcome will therefore be time related. Banana hybrids with resistance to black Sigatoka have already been bred outside the BIP and are now in the validation phase. More hybrids will be selected soon and there are currently 10 hybrids being tested. The BIP has had some part in the development of these later bred hybrids. Even with the development of genetically resistant bananas, spraying will still most likely be necessary. However, one assumption is that spraying frequency for black Sigatoka could be reduced by at least 50% and possibly more. It is likely also that partial resistance will need to be retained through a continuous breeding program. In this respect the molecular marker technology could be of significant value in future breeding programs. 193 It is likely that a wide range of resistant material will be required over space and time, due to the diversity of the fungus and its ability to recombine. Hence, the understanding and identification of genetic resistance mechanisms will be a critical component of future strategies to reduce the negative economic and environmental impact of existing chemical control methods. The BIP will not be able to claim any specific direct commercial impact with respect to the development of resistant material. The contribution through the BIP supported project has been in the main one of additional knowledge. However, it may be appropriate to attribute up to 5% of the benefits that might eventuate from black Sigatoka-resistant clones that could be produced and adopted by industry under the CIRAD program. The appropriate level of attribution could be roughly equated to the relative financial contribution of the BIP project to that of the total CIRAD breeding program aimed at producing resistant types. The benefits from the breeding program as a whole will depend on the program producing export type bananas, the timing and cost of such activities, and the rate and level of adoption of such types by industry. Note: The figure for two BIP projects is US$ 450,000 for three years, compared with CIRAD’s contribution over 16 years of 6 million francs per year or 96 million francs; this is about US$ 20 million; relative contribution is about 2.25%. PN3: Collection, characterisation and evaluation of Nendran bananas in India Kerala Agricultural University, India Introduction The objective of this project was to collect and categorise genetic material on Nendran bananas and their wild relatives in southern India. The idea was to establish reference data for further research and to make the germplasm available world-wide for input into breeding programs. Results A total of 128 accessions were collected and evaluated for dwarfism, short-cycling and disease resistance, particularly for yellow Sigatoka and to lesser extent weevils and nematodes. Impact There may be some local impact on Indian banana production system in terms of using the more suitable genotypes for different farming systems. The evaluation may result in the use of more short-stature bananas, which in turn may reduce the cost of propping as well as increasing suitability to fit into annual rotation systems. Moreover, the 194 germplasm may be very useful in terms of expanding the range of germplasm available to breeding programs around the world. However, it is not clear as to whether the germplasm has yet been made available to the international germplasm collection. PN4: Collection, evaluation and characterisation of genetic resources and improvement of banana in Vietnam Institute of Agricultural Genetics, Vietnam Introduction This project was aimed at the collection and evaluation of new germplasm for improved productivity and disease resistance. Other components were also included in the original objectives of the project, but the main emphasis was on germplasm collection. Results The collection of the germplasm was effected and much of the material has already been evaluated and classified. Characterisation of morphological and agrobiological traits of local cultivars has been completed, with a number of valuable clones selected and recommended for production. The collections have been established in tissue culture, but some accessions have exhibited difficulty in tissue culture multiplication, which is delaying their availability to contribute to the genetic improvement of Musa. Impact The new material will eventually be available for use in breeding programs. As this region, together with south China is near to the origin of the banana, it is likely that the material does contain new genetic types. The material could provide an important resource for future banana improvement. PN5: Banana-breeding in Brazil EMBRAPA, Brazil Introduction This project was associated with the Brazilian banana-breeding program and was aimed at obtaining high yielding varieties resistant to the main pests and diseases in Brazil. The project included the production and evaluation of 10,000 diploid hybrids and 1000 tetraploid hybrids. Although yellow Sigatoka is present in Brazil, black Sigatoka has only just been detected in the Amazonas region; hence it is likely that this project will assist the efforts to control the spread of the disease. 195 Results Diploid hybrids are being produced and evaluated. Breeding techniques to utilise sterile material are currently being developed. Molecular markers for germplasm characterisation are being identified and used. Impact The project should allow improved breeding for resistance to black Sigatoka and other pests and diseases in banana. Of more immediate impact is that two black Sigatokaresistant clones, to whose development and testing the BIP project contributed, are being used to assist control of the black Sigatoka outbreak in the Amazonas. These clones may slow down the rate of spread of the disease. This would have a major reduction in the cost of those who can potentially afford to spray with fungicides as well as reduce yield losses of those who cannot afford spraying control technologies. PN8: Novel genes for fungal resistance and post-harvest quality Boyce Thompson Institute for Plant Research (BTI), USA Introduction This project was a component of the molecular tool box project and was aimed at identifying and characterising genes controlling ethylene metabolism. Results The project allowed an improved understanding of the molecular events that occur during banana fruit development and ripening and those events that are associated with pathogen challenge in fungal resistant and non-resistant banana varieties. While some genes were successfully identified, there were still some problems in successfully applying the Agrobacterium method in the transformation process. It is understood that these issues have now been partly overcome as the group at the University of Hawaii is currently using a modified version of the technique for gene introduction after collaborating with the group at BTI. Impact The major impact of this project will be through the additional knowledge generated on ethylene control and the further development of the Agrobacterium method of transformation. 196 PN9: The development of transgenic bananas with resistance to Banana Bunchy Top (BBTV) and Banana Mosaic (BBMV) viruses Queensland University of Technology (QUT), Australia Introduction The aim of the project was to control BBTV and BBMV viruses through transformation with resistance genes. These viral diseases are costly to control, largely through pulling out infected plants. While this is effective in Australia, such control needs a strong legal backing and responsive producers. If bunchy top is not controlled, it can reduce yields by up to 100%. Losses may average about 30%. Cavendish is one of the most susceptible clones to bunchy top and shows serious symptoms. Some other banana clones are difficult to infect. Banana Bract Mosaic virus can create losses in yield of up to 20-40%. Results The particle bombardment transformation process was used to successfully transform both Bluggoe and Cavendish banana types with anti-viral genes. This has been achieved for four different BBTV-potentially resistant genes and one gene for BBMV. Regeneration was achieved but propagation of some of the regenerated cells proved troublesome. It is expected that about 50% of the transgenic lines produced will be resistant. This proportion is expected to be increased to 90% within a year as the process is further refined. Impact The technology to generate potentially virus-resistant Cavendish plants, including transformation, regeneration, and expression of the new genes that should confer resistance, has now been demonstrated. Transgenic plants generated so far need to be tested in the field to ensure that the expression of the gene does confer resistance. It was estimated by the Principal Investigator that the chance of the resistance breaking down will be only 5-10% within a period of 10 years from release. Field trials may take some two years to complete and will face the potential constraint of biosafety regulations. The technologies developed will not only be relevant to viral resistant transgenics but will also be relevant to the generation of transgenic resistance to other banana diseases. Without the BIP project, efforts would have continued in this area at QUT, but a different approach would have been taken and the result would not have been achieved. Hence, there is a direct relationship between the BIP and the achievement of these results. 197 PN10: Molecular tool box KUL, Belgium Introduction PN10 consisted of two phases: a generic phase involving three different institutions working in collaboration with one another; and a second phase where the project was split up into two specific components, one associated with ethylene and one associated with viruses. A project component targeting anti-fungal biotechnologies did not proceed due to intellectual property issues. The idea of the molecular toolbox was to improve the collaboration of a range of efforts to produce transgenic bananas with commercial application. The idea was to produce synergy so that efforts were focused and the rate of gain was increased. Results The first phase of the project, the development of a molecular toolbox, was successfully completed and demonstrated effective co-operation among the three research providers making up the consortium. The project resulted in the refinement and optimisation of the transformation process using the particle bombardment process. However, there is still some doubt about the repeatability of the Agrobacterium process. Isolation of promoters has not been completed, although some genes eliciting high expression have been defined. The project could not work with the antifungal gene as it had already been patented by Zeneca. The DNA library has been made available to other interested laboratories. After the completion of the generic part of the project, the synergy exhibited earlier in the project weakened. Impact Transformation was possible before the molecular toolbox project, but the project allowed systems to be refined and additional parameters to be specified. In this regard, the BIP accelerated the ability to transform banana. The BIP also helped to publish some of the previous work from KUL and Texas A&M. This has resulted in others using transformation technologies such as those at QUT. BIP partners at QUT, Hawaii and Hong Kong supplied constructs to KUL, where they were inserted into banana according to the refined protocol. Due in part to the BIP, there has been increased confidence within the private sector in the ability to capture the potential gains that biotechnology has to offer. It is now understood that a number of the multinational banana-producing and exporting companies are pursuing biotechnology solutions, mostly in conjunction with private 198 technology companies. It is likely that the BIP project has contributed to this increasing interest on the part of the private sector. Most private sector groups are concentrating on Cavendish transformation. If these initiatives are successful, the BIP could claim maybe 5-10% of the total benefits through the indirect triggering of such private investment. Alternatively, the BIP project could be attributed to bringing forward the benefits of the transgenic bananas by some two to three years. Any benefit that can be attributed to the BIP will rely on the actual production and use of transgenic bananas in the future, whether produced by the private sector or through other investment taking place via public research institutions. PN11: Genetic engineering of ethylene biosynthesis in bananas Hong Kong University of Science and Technology, Hong Kong Introduction This project set out to clone genes involved in ethylene biosynthesis in banana as well as to genetically engineer bananas with reduced ethylene production capacity and evaluate transformed plants in relation to reduced disease development and enhanced fruit quality. Results The project achieved only the first objective of identifying and cloning two genes involved in ethylene biosynthesis. The project constructed a banana genomic library and identified several genes. However, the results were somewhat inconclusive and much more work is needed to link the gene structure to their functions in disease resistance and fruit ripening. Impact The project results produced knowledge of the genetic information associated with ethylene biosynthesis. This may be useful in future research aimed at post-harvest qualities of fruit as well as in disease control. PN12: Use of biotechnology to produce transgenic bananas that are resistant to BBTV infection University of Hawaii, USA Introduction The project was aimed at developing further transformation technology using a modified version of the Agrobacterium-mediated banana transformation system developed by BTI at Cornell University. The project focused on BBTV gene constructs and their 199 incorporation into ‘Williams’ bananas. Results Much of the project was associated with the transformation technologies and, in that regard, it appears that the Agrobacterium process was refined, although some problems still remain. Three hundred transgenic plants were produced using pathogen-derived genes and 65 of these were evaluated. This resulted in three of the 65 showing no BBTV-specific symptoms and this was confirmed with ELISA testing. A further 21 transgenic plants showed two to three week delays in symptom development. Impact Field experiments still remain to be conducted to test the field performance of the plants. In addition, the isolate used was that found in Hawaii and it is uncertain whether the transgenic plants would also be resistant to the other BBTV virus group endemic in other parts of Asia. However, the impact of these results could be highly significant in that it has now been demonstrated that it is possible to produce viral resistant plants. Further refinement of the process could result in considerable savings in losses from bunchy top virus disease, which is present in Asia, the Pacific and in African countries. PN14: Elimination of banana streak badnavirus (BSV) from improved Musa germplasm and related studies on transmission and host plant/virus/vector interactions IITA, Nigeria Introduction BSV (banana streak virus) is present in many banana cultivars and in germplasm used in breeding programs. The presence of the virus has serious implications for the safe movement of germplasm around the world. This project was aimed at identifying Musa tissue that was free of both genomic and episomal BSV to test the hypothesis that some cells could be free of both BSV types. The project was also aimed at testing whether BSV could be eliminated from germplasm, and how it is transmitted. Results Diagnostic techniques were developed and transmission mechanisms defined. Transmission through seeds and suckers was confirmed as was propagation through tissue culture. No evidence of spread via mealybugs or field spread of BSV was confirmed. Elimination of BSV does not appear possible using conventional virus elimination methods. 200 Impact The project produced increased knowledge and understanding of BSV. This included its etiology, epidemiology and possibility of control. Information relevant to the safe movement of germplasm and appropriate quarantine activities was produced. The virus remains a serious constraint to the movement of germplasm due to its incorporation into the banana genome. BSV is now also increasingly recognised as a (field) production constraint in certain banana growing areas. PN15: Variability and relationships within populations of Fusarium oxysporum f sp. cubense from its centre of origin Queensland Department of Primary Industries (QDPI), Australia Introduction The essence of this project was an assessment of variability in the pathogen causing Fusarium Wilt (Panama Disease) in bananas. Other aims of the project were to assess the reliability of using plants derived from tissue culture as opposed to conventional planting materials when assessing resistance of banana germplasm to the pathogen. A third component was the assessment of breeding material against the Australian and Asian populations of the pathogen. There is no control mechanism for Panama disease. Infection means the planted area has to be abandoned. The pathogen will remain in the soil for long periods. Results The QDPI now has a world-wide collection of over 2000 isolates of the pathogen, as well as a DNA database for more than 500 isolates representing the 20 recorded VCGs and the 33 known genotypes, and would be able to provide a service in the future for typing specific pathogen types. Protocols have been drawn up for field evaluation of host genotypes against pathogen strains. One banana type is resistant to tropical race 4 of the pathogen (strain 4 is the Asian pathogen to which Cavendish is extremely susceptible). However, strain 4 is not yet in Central and Latin America or the Caribbean areas. The pathogen displays significant genetic variability and is the product of co-evolution in Asia and has moved with the planting material around the world. Some data provides evidence for the pathogen’s origin being outside the banana’s centre of origin. Plants derived from tissue culture are more susceptible when compared to conventional planting material. The idea of using this to advantage in attempting to increase resistance with beneficial and bio-control organisms is being pursued. The project has been screening material from the CIRAD program and from the 201 Brazilian and FHIA breeding programs for resistance against the Australian and Asian strains of the pathogen. Impact The world-wide collection of the various strains of the pathogen will serve as a reference collection in the future, enabling the strains causing an outbreak of Panama disease to be quickly identified. The information on the genetic diversity of the pathogen will also be useful in breeding programs to screen against a more complete profile of the pathogen. New banana developments can be more safely implemented through determining the strains present in any given location before planting begins. For example, recent outbreaks of Fusarium Wilt in Cavendish plantations in the eastern tropics (peninsular Malaysia, Sumatra and Halmahera) caused enormous damage. These outbreaks have ominous implications for the durability of western trades that are based on Cavendish clones. Protocols for the field evaluation of hybrids and diploid material for resistance against Panama Disease have been developed and these should be useful in future breeding programs. Professional links with CIRAD, FHIA,EMBRAPA and other breeding programs around the world have been developed and strengthened due to the BIP. The Australian group at QDPI is collaborating with these other initiatives as a result of the program. PN17: Origin and distribution of fungicide-resistant strains of M. fijiensis in banana plantations in Costa Rica CORBANA, Costa Rica Introduction This project was aimed at providing information that would be useful in developing management strategies for chemical control of black Sigatoka disease. Part of the project was associated with identifying inheritance of resistance in the progenies of the black Sigatoka pathogen. The population dynamics of the pathogens that are resistant to two major fungicides (benomyl and propiconazole) were studied and attempts made to determine how many genes were involved in the resistance to fungicides. Results It was found that the acquisition of genes for resistance to chemicals did not have an effect on the fitness of pathogen populations. It was found that resistance patterns persisted even after fungicide use was stopped. 202 The project also provided a greater understanding of the parasitic fitness of resistant strains (conidial sporulation, aggressiveness, and ability to cross) and on the ability of the pathogen to carry double resistance (to benzimidazoles and triazones). Impact The absence of an effect of a chemical-free period is useful information in the development of strategies to more effectively use chemicals and lessen the impact of populations of pathogens resistant to fungicides. While this was a negative finding in some respects, it has confirmed that strategies for improved use of chemicals for control of the black Sigatoka pathogen are not easy to develop and that the continuing development of new chemicals, together with its associated costs, is the principal avenue by which some retention of chemical control can be managed. PN18: Study of tolerance and resistance of banana to nematodes CRBP, Cameroon Introduction The primary aim of the project was to identify sources of resistance of bananas from the CRBP Musa germplasm collection at CRBP to major nematode species. Screening and testing were carried out with trials in greenhouses, with young suckers, as well as in the field. The greenhouse inoculation was artificial, but that in the field was natural. A total of 142 accessions from different genotypes were screened including diploids, triploids and hybrids from CIRAD, CRBP, and FHIA. The aims were: (i) (ii) To identify types to use in the breeding program in CRBP; To look for less susceptible plantains that can be recommended to farmers to replace the susceptible ones Results It was found that many genotypes displayed resistance but not real immunity. As the resistance was in different genotypes, it was suspected that many genes were involved. It was likely therefore that a range of barriers could provide sources of resistance e.g. penetration, biochemical, physiological etc. The suspected source of resistance is polygenic and therefore, if built further into banana types, it would be less susceptible to being overtaken by the organism. However, polygenic resistance would be less simple to incorporate into future types through breeding. It was also shown that the plantain clones are all susceptible to the major nematode species, indicating that there is need to incorporate nematode resistance into fertile plantains. 203 Impact The most important impact will be through the selection of parents in the breeding program in CRBP. It is likely that an increased rate of genetic gain with respect of resistance could therefore be forthcoming. Further, increased knowledge on the resistance of Musa to nematodes has been generated. Some of the resistant clones also have resistance to other banana pests and pathogens. This will have implications for the effectiveness of strategies to incorporate multiple resistance into banana types. An early screening method using suckers was developed and validated. This method will help to rapidly discard susceptible breeding lines in the selection process. PN19: Identification of durable nematode-resistance sources in banana and plantain KUL, Belgium Introduction The project PN19 examined Musa germplasm for nematode resistance. The original project proposal was submitted by FHIA, IITA, IRTA (Spain - Canary Islands) and KUL. The project (PN19) was part of this original project concept with other projects also being separately funded. A broad network of collaboration developed between several BIP projects on nematodes as well as other projects outside BIP. Results PN19 confirmed that there are durable resistance types for nematodes (PJB types), and gave validity of screening resistance at the pot level and across locations. An important finding was that if sufficient time (eight weeks) was allowed for roots to have developed sufficiently, then the test in the pots was reliable. The publication of the screening protocol was a particular output from the BIP. The project found that it was possible to use in vitro screening and thus save a few months in time, as opposed to growing material out in plots. However, it was found that contamination with bacteria was high and there was a need to improve the culture preparations to minimise contamination. Hence the project is not quite completed. Screening that might have taken eight to nine months can now be effected in three to four months and would take less space so that more material can be evaluated for a given set of resources in breeding programs. In this regard the project output will increase the efficiency of future breeding programs. The project also found new sources of resistance, mainly in genotypes from PNG. These materials may not be very useful in breeding programs as they generally displayed low productivity, were rather infertile, and were not of the Cavendish type. However, the materials may be useful in understanding the mechanisms of resistance 204 to nematodes. A major accomplishment of the project was the establishment of the Nematode Consortium, which is continuing today. Impact The efficiency of future breeding programs will be increased through the findings and protocol developed within this project and through PN20. The IRTA initiative would not have started without the BIP, but some of the other initiatives might have commenced in spite of BIP. Further, FHIA screening would not have started without BIP. BIP also triggered a whole network of research through the consortium, including leveraging further resources from VVOB (Flemish government) into bananas. The impact of the consortium will continue under the IMPT3. (The International Musa Trial Program supported by INIBAP and UNDP). IMPT3 will benefit from the similar testing of genotypes against nematodes on a world-wide basis. The original BIP nucleus will be the mainstay of this testing program; without the BIP this network would have had to start from scratch. Before this project, in vitro methods were not available for testing for nematode resistance. Both PN19 and PN20 have improved existing methods and resulted in a better understanding of the reliability of the use of different methods of testing for resistance (both in vitro and in vivo) and also improvement of nematode inoculum. PN19 provided a high level of leverage for the introduction of other resources into nematode-related research. It has been estimated that for every dollar put into the nematode projects PN19 and PN20 by BIP a further $ 6.20 was invested by others, including the Flemish and Belgian governments. PN20: Identification of durable nematode-resistance sources in banana and plantain FHIA, Honduras Introduction PN20 was carried out in Honduras by FHIA. Screening was against two of the most important nematodes, and both corms and tissue culture were used. Both parents and hybrids were screened to enable breeders to follow what they were doing in their crossing program. The FHIA evaluation initiative would not have happened without the BIP. KUL personnel experienced with nematodes provided supervision for PN20. It has been estimated that losses due to nematodes may average around 20%. 205 Results The project produced systematic information on resistance and brought forward the possibility of screening for resistance. Also some sources of resistance were found that were not known to exist beforehand. Impact Before PN20, there had been no validation of resistance materialised in breeding programs. The information from this project provided knowledge of resistance when selecting parents for crossing within the breeding program. This would have the potential of increasing the rate of genetic gain with respect to nematode-resistant clones. Also new sources of resistance to nematodes were found, and this could lead to more information on the mechanism for resistance, which could have implications for biotechnological approaches. However, the nematode-resistant varieties of other crops which are reported in the literature were developed by conventional breeding. The PN20 findings are being used in the breeding program and the new material from PNG is currently being prepared. However, there has been no information provided on genetic mechanisms for resistance. It will probably be necessary to keep breeding for resistance. There is not only a range of nematode types, but some of the less aggressive types may become more aggressive when the more dominant ones are controlled. Such 'new' dominants may not be resistant to the banana type produced for resistance to the previous dominant strain. This possibility indicates a need to breed for multiple resistance, that is, against a range of nematode types. Little is known about how robust any resistance mechanisms to the existing principal nematodes might be in this regard. Nematicides may not have to be used once resistance is built into the genome. There is also a possibility that this research will lead to the identification of resistant genes, although due to the mechanism of feeding of the nematode, any target toxin will have to be tested via an engineered living plant cell. The work effected under PN19 and PN20 may have shortened the time until a nematode-resistant banana type is available. This would apply to export bananas that would still need to be developed and commercialised. Cavendish is highly susceptible to the major nematodes and a biotechnology solution may need to be sought in the case of providing a nematode-resistant Cavendish. However, if a nematode-resistant alternative to Cavendish is targeted instead, then the FHIA breeding program which has developed agronomically-advanced, nematode-resistant diploid parental lines which are readily usable in cross pollinations for breeding resistant, commercial type hybrids for export, may well be successful. The FHIA-23 ‘Gros Michel’ type hybrid is currently being cultivated on about 2000 hectares in Cuba. It is replacing Cavendish as rapidly as planting material of the hybrid can be multiplied. 206 The evidence that conventional breeding works in breeding nematode-resistant commercial hybrids has been confirmed in the FHIA-19 apple-flavoured hybrid. FHIA-18 was derived from crossing the burrowing nematode-resistant SH-3142 diploid onto the Dwarf Prata triploid, and this resultant nematode-resistant tetraploid is now being cultivated on about 900 hectares in Cuba. PN21: Identification of durable nematode-resistance sources in banana and plantain IRTA, Spain Introduction This project was a one-year project aimed at evaluating Musa genotypes for durable resistance against the main nematode pests that affect banana production. This was effected by establishing a live collection of migratory nematodes and initiating screening trials. Results The live collection was established and the material prepared for resistance evaluation. Screening trials were initiated and were continuing at the time of the termination report. Impact The germplasm screening for resistance will enable more effective breeding programs to be initiated through selection of resistant material. The development of more resistant material should reduce costs of nematode control and lower the use of nematicides. PN22: Breeding hybrid Musaceae with resistance to multiple diseases, especially black Sigatoka and Panama disease FHIA, Honduras Introduction The objective of the project is to produce commercial-type dessert bananas with a short stature and which are resistant to both black Sigatoka and Panama Disease. Already some success has been obtained with resistant types but the plant has been too tall for ready acceptance by commercial producers, except in Cuba where it is reported that the area of the resistant type is increasing. 207 Results Cross pollinations have been made with improved diploids crossed onto seed fertile triploids. The first tetraploid hybrids produced under this strategy have not yet been evaluated. Impact If varieties with appropriate stature and disease resistance are produced, this could have a major impact on the banana export industry. This positive impact will only occur if the resulting hybrids are seedless, productive, exhibit desirable bunch features, and maintain disease resistance over time. The black Sigatoka-tolerant tetraploid (FHIA-23) being cultivated in Cuba illustrates that conventional breeding can work for development of disease-resistant alternatives to Cavendish. However, as the sources of resistance have not been identified through molecular studies, the nature of resistance conferred will not be understood. This would be relevant if biotechnical solutions are to be pursed as well as conventional breeding solutions. 208