COMMUNAUTE FRANCAISE DE BELGIQUE
ACADEMIE UNIVERSITAIRE WALLONIE-EUROPE
UNIVERSITE DE LIEGE – GEMBLOUX AGRO-BIO TECH
Contribution to the study of semiochemical slow-release
formulations as biological control devices
Stéphanie HEUSKIN
Essai présenté en vue de l’obtention du grade de docteur en sciences
agronomiques et ingénierie biologique
Promoteur : Prof. Georges Lognay
2011
COMMUNAUTE FRANCAISE DE BELGIQUE
ACADEMIE UNIVERSITAIRE WALLONIE-EUROPE
UNIVERSITE DE LIEGE – GEMBLOUX AGRO-BIO TECH
Contribution to the study of semiochemical slow-release
formulations as biological control devices
Stéphanie HEUSKIN
Essai présenté en vue de l’obtention du grade de docteur en sciences
agronomiques et ingénierie biologique
Promoteur : Prof. Georges Lognay
2011
Copyright. Aux termes de la loi belge du 30 juin 1994, sur le droit d'auteur et les droits voisins, seul l'auteur a
le droit de reproduire partiellement ou complètement cet ouvrage de quelque façon et forme que ce soit ou d'en
autoriser la reproduction partielle ou complète de quelque manière et sous quelque forme que ce soit. Toute
photocopie ou reproduction sous autre forme est donc faite en violation de la dite loi et des modifications
ultérieures.
Heuskin Stéphanie (2011). Contribution to the study of semiochemical slow-release formulations as
biological control devices (PhD thesis). Gembloux, Belgium, University of Liege – Gembloux Agro-Bio
Tech, 224 p., 21 tabl., 31 fig.
Abstract: Semiochemicals – informative molecules used in insect-insect or plant-insect interactions – have
been widely considered within various integrated pest management (IPM) strategies. In the present thesis, two
sesquiterpenoids, E-β-farnesene and E-β-caryophyllene, were formulated for their related properties as aphid
enemy attractants. E-β-farnesene, the alarm pheromone of many aphid species, was also identified as a
kairomone by attracting and inducing oviposition of aphid predators (Episyrphus balteatus De Geer (Diptera:
Syrphidae)) and by attracting aphid parasitoids (Aphidius ervi Haliday (Hymenoptera: Braconidae)). E-βcaryophyllene was identified as a potential component of the aggregation pheromone of the Asian ladybird,
Harmonia axyridis Pallas, another aphid predator. The two products were purified from essential oils of
Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L. (Lamiaceae) for E-β-farnesene and E-βcaryophyllene, respectively. Natural and biodegradable slow-release formulations were then investigated in order
to deliver these molecules on crop fields for a long period of time as biological control devices. Due to their
sensitivity to oxidation, both sesquiterpenes needed to be protected from oxygen degradation. For this purpose,
alginate – hydrophilic matrix with low oxygen permeability – was used as polymer for the formulations: the
main objective was to deliver semiochemical substances in the air in a controlled way. Consequently, a careful
selection of alginates was realised. Formulated beads showed different structural and encapsulation properties
depending on various formulation factors. Alginate formulations were characterized by texturometry and by
confocal microscopy in order to observe the distribution of semiochemicals in alginate network. The last step of
alginate bead characterisation consisted in studying release rate of semiochemicals in laboratory-controlled
conditions by optimised trapping and validated Fast-GC procedures. Finally, the efficiency of formulations as
aphid predator (Syrphidae) and parasitoids (A. ervi) attractants was demonstrated by field trapping and
olfactometry experiments.
Heuskin Stéphanie (2011). Contribution à l’étude de formulations à liberation progressive de
sémiochimiques en tant qu’outils de contrôle biologique (Thèse de doctorat). Gembloux, Belgique,
Université de Liège – Gembloux Agro-Bio Tech, 224p., 21 tabl., 31 fig.
Résumé: Les sémiochimiques – molécules véhiculant des informations au sein des interactions insecte-insecte
ou plante-insecte – ont été largement utilisés dans de nombreuses stratégies de lutte intégrée. La présente thèse
de doctorat décrit la formulation de deux sesquiterpènes, le E-β-farnésène et le E-β-caryophyllène, pour leurs
propriétés relatées dans la littérature en tant qu’attractants des ennemis des pucerons. En effet, le E-βfarnésène, la phéromone d’alarme de nombreuses espèces de pucerons, a également été identifié comme
kairomone en attirant des parasitoïdes (Aphidius ervi Haliday (Hymenoptera: Braconidae)) et des prédateurs de
pucerons (Episyrphus balteatus De Geer (Diptera: Syrphidae)) et en induisant un comportement d’oviposition
chez ces derniers. Le E-β-caryophyllène a récemment été identifié comme un composé probable de la
phéromone d’agrégation des coccinelles asiatiques, Harmonia axyridis Pallas, également prédatrices de
pucerons. Les deux produits ont été purifiés au départ des huiles essentielles de Matricaria chamomilla L.
(Asteraceae) et de Nepeta cataria L. (Lamiaceae) respectivement dans le cas du E-β-farnésène et du E-βcaryophyllène. Des formulations biodégradables et d’origine naturelle ont ensuite été développées comme outil
de contrôle biologique afin de libérer les deux composés de façon progressive. En raison de leur sensibilité à
l’oxydation, ces deux sesquiterpènes devaient être protégés de l’oxygène. A cette fin, de l’alginate – matrice
hydrophile offrant une faible perméabilité à l’oxygène – a été utilisé comme polymère de formulation : le
principal objectif était de délivrer les substances sémiochimiques dans l’air d’une façon contrôlée. En
conséquence, l’alginate a été sélectionné de façon minutieuse. Les billes formulées ont montrés des propriétés
structurelles et d’encapsulation différentes en fonction de nombreux facteurs intervenant lors de la formulation.
Les formulations ont été caractérisées par texturométrie et par microscopie confocale de façon à observer la
distribution des sémiochimiques au sein du réseau d’alginate. La dernière étape de caractérisation des billes a
consisté en l’étude du taux de libération des sémiochimiques dans des conditions contrôlées de laboratoire après
avoir optimisé la technique de piégeage des volatils et validé la procédure d’analyse par GC rapide. Finalement,
l’efficacité d’attraction des formulations envers les prédateurs et les parasitoïdes de pucerons a été démontrée
par des expériences de piégeage sur cultures et des essais d’olfactométrie.
I dedicate this work to my parents
for their love, their support and for
all the things they sacrificed for helping me
Acknowledgments
Many people contributed to the achievement of this PhD thesis. All these actors must be
acknowledged hereafter:
-
My promoter, Prof. Georges Lognay, who gave me the opportunity to realise this
fascinating and challenging work. He always was there for me when I had some
doubts. Since the beginning of my thesis he trusted in me and gave me opportunities to
follow my ideas. Through the many discussions we had together, he helped me to
become more self-confident in my scientific decisions.
-
The other members of my PhD committee: Prof. Jean-Paul Wathelet for leading the
committee, Prof. Jean-Paul Barthélémy for his role as secretary of the committee,
Prof. Eric Haubruge, Prof. Thierry Hance, Prof. Philippe Thonart and Prof.
Michèle Mestdagh. They have to be acknowledged for the precious advices they gave
me during four years.
-
Prof. François Béra for his precious help in the slow-release study approach. He
should be acknowledged for the many discussions we had and for the time he spent to
solve the material and scientific problems we encountered all along this work.
-
My colleagues from the SOLAPHID project: Stéphanie Lorge, Isabelle Frère and
Frédéric Muratori and all the technicians from their team, Ahmed Sabri, Pascal
Leroy and Julien Farmakidis. They must be acknowledged for the many ideas we
shared together during almost five years. A special thank to Stéphanie and Pascal with
which I collaborated more closely and who contributed actively to the achievement of
publications.
-
My colleagues of the Department of Analytical Chemistry (past and present): Quentin
Denis,
Hélène
Flamant,
Fabienne
Piscart,
Liliane
Hannot,
Sophie
Vancraenenbroeck, Christelle Marlet, Christophe Fischer, Maryse Vanderplanck
and Michaël Dermience for the nice moments we had together. I also have a special
thought to Jean Taziaux.
-
Dany Trisman and Vincent Hôte from the Department of General and Organic
Chemistry for their precious help.
-
The bachelor and master students I supervised during my thesis and who contributed
directly to this research: Bruno Godin, Julien Farmakidis, Aise Kilinc and
Christophe Fischer.
-
My friend Sabrine Attia from the Earth and Life Institute of the Catholic University
of Louvain with which I had nice and constructive discussions about essential oils.
-
The financial support of the Walloon Region Ministry for the SOLAPHID project
which allowed me to conduct my PhD thesis in the best conditions of equipment and
gave me the financial opportunities to take part in many interesting international
symposia.
-
My parents and my family who always trusted in me. They must be acknowledged
for all the love they gave me in my life. I also would like to thank my two angels who
protect me since thirty years.
-
My friends “Les Hystériques” for their support and their interest in my research and
particularly for their frame of mind.
-
My love Daniel who was always by my side and who helped me to find motivation
and energy to overtake the problems.
Abbreviations
α
Selectivity factor
ANCOVA
Analysis of covariance
Aw
Water activity
CLSM
Confocal laser scanning microscopy
D
Diffusion coefficient (m²/s)
EBF
E-β-farnesene
EI
Electronic impact
E.O.
Essential oil
FID
Flame ionisation detector
GC
Gas chromatography
GC-MS
Gas chromatograph – mass selective detector
GLP
Good laboratory practices
Hmin
Minimum height of theoretical plate
HETP
Height equivalent of theoretical plate
HPLC
High performance liquid chromatography
(I)
Rentention index
I.D.
Internal diameter
IS
Internal standard
IPM
Integrated pest management
LLOQ
Low limit of quantification
LOD
Limit of detection
LOQ
Limit of quantification
M/G
Mannuronate/Guluronate ratio
Mt
Cumulative mass of semiochemical released at time t (µg)
M∞
Cumulative mass of semiochemical released at time ∞ (µg)
Mw
Molecular weight
(N)
Number of theoretical plates
NMR
Nuclear magnetic resonnance
R²
Determination coefficient
RH
Relative humidity
Rs
Chromatographic resolution
RSD
Relative standard variation (%)
RT
Retention time
RT’
Reduced retention time
SD
Standard deviation
SS
Sum square
UFGC
Ultra fast gas chromatograph
UFM
Ultra fast module
W
Peak width of chromatographic peak
Index
Chapter I General introduction…………………………………………..
19
General introduction………………………………………………………………..
21
References…………………………………………………………………………..
25
Chapter II Bibliography…………………………………………………..
29
Publication 1: The use of semiochemical slow-release devices in integrated pest management
strategies ………………………………………………………………………………..
31
Abstract……………………………………………………………………………...
32
Résumé………………………………………………………………………………
33
1. Introduction……………………………………………………………………….
34
2. Semiochemicals…………………………………………………………………...
35
2.1. Definitions…………………………………………………………………..
35
2.2. Chemistry and properties of semiochemicals………………………………
36
3. IPM strategies using semiochemicals…………………………………………….
37
3.1. IPM strategies………………………………………………………………
37
3.1.1. Monitoring………………………………………………………..
38
3.1.2. Trapping………………………………………………………….
38
3.1.3. Mating disruption…………………………………………………
38
3.1.4. Push-pull strategy………………………………………………....
39
3.1.5. Biological control…………………………………………………
39
4. Slow-release of semiochemicals………………………………………………….
40
4.1. Slow-release dispensers…………………………………………………….
40
4.2. Slow-release rate studies…………………………………………………...
42
4.2.1. Techniques to estimate release rate……………………………….
42
4.2.2. Release rate studies……………………………………………….
43
5. Conclusions……………………………………………………………………….
48
References……………………………………………………………………………
49
Chapter III Methodology of the research…………………………………. 55
Chapter
IV
Characterisation
of
essential
oils
and
semiochemical
purification………………………………………………………………..
Objectives…………………………………………………………………………
59
61
Publication 2: Fast gas chromatography characterisation of purified semiochemicals from
essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L.
(Lamiaceae) ……………………………………………………………………………..
63
Abstract………………………………………………………………………………
64
1. Introduction……………………………………………………………………….
65
2. Experimental………………………………………………………………………
67
2.1. Chemicals and materials……………………………………………………
67
2.2. GC-MS analyses…………………………………………………………….
67
2.3. Fast GC analyses……………………………………………………………
68
2.4. NMR spectra………………………………………………………………… 68
2.5. Essential oil fractionation and purification…………………………………
68
2.6. Method validation…………………………………………………………..
69
2.7. Calculation of direct resistively heated column (UFM) efficiency…………
70
3. Results and discussion……………………………………………………………
71
3.1. Fast GC analytical method validation……………………………………..
71
3.2. Analytical performances of UFM column………………………………….
74
3.3. Analysis of essential oils……………………………………………………
76
3.4. Essential oil fractionation………………………………………………….
80
Acknowledgments……………………………………………………………………
81
References……………………………………………………………………………
83
Chapter V Quantification of semiochemical in formulations……………
Objectives…………………………………………………………………………
85
87
Publication 3: Validation of a fast gas chromatographic method for the study of
semiochemical slow release formulations………………………………………………
91
Abstract………………………………………………………………………………
92
1. Introduction……………………………………………………………………….
93
2. Experimental………………………………………………………………………
95
2.1. Chemicals and Reagents……………………………………………………
95
2.2. Fast GC analyses……………………………………………………………
95
2.3. Flash Chromatography……………………………………………………..
97
2.4. Formulation of alginate gel beads………………………………………….
97
2.5. Determination of the sesquiterpene protection efficiency of formulations…
98
2.6. Method validation…………………………………………………………..
100
2.6.1. Solutions used for calibration…………………………………….
100
2.6.2. Solutions used for validation……………………………………..
101
3. Results and discussion……………………………………………………………
102
3.1. Resolution and selectivity of the method…………………………………...
102
3.2. Validation by use of accuracy profile approach……………………………
102
3.2.1. Analysis of the response functions and determination of the best
regression models………………………………………………………..
103
3.2.2. Trueness of the method…………………………………………..
107
3.2.3. Precision of the method…………………………………………..
107
3.2.4. Accuracy of the method…………………………………………..
108
3.2.5. Limits of detection, quantification and range…………………….
108
3.2.6. Linearity…………………………………………………………..
108
3.3. Measurement uncertainty…………………………………………………..
111
3.4. Protection efficiency of the formulations…………………………………..
114
4. Conclusion………………………………………………………………………..
116
Acknowledgments…………………………………………………………………..
116
References…………………………………………………………………………..
117
Chapter VI Optimisation, efficiency and slow-release study of semiochemical
formulations ………………………………………………………………
123
Chapter VI.1………………………………………………………………
125
Objectives…………………………………………………………………………
127
VI.1. Publication 4: Optimisation of a semiochemical slow-release alginate formulation
attractive towards Aphidius ervi Haliday parasitoids………………………………….
131
Abstract……………………………………………………………………………...
132
1. Introduction……………………………………………………………………….
133
2. Materials and methods……………………………………………………………
136
2.1. Chemicals and reagents…………………………………………………….
136
2.2. 1H-NMR characterisation of sodium alginate M/G ratio…………………..
137
2.3. Alginate gel bead formulation………………………………………………
137
2.4. Volatile collection system…………………………………………………...
139
2.4.1. Description of the design………………………………………….
139
2.4.2. Efficiency evaluation of the design……………………………….
140
2.5. Quantification of semiochemicals and fast GC analyses…………………...
140
2.6. Texturometry of alginate beads……………………………………………..
141
2.7. Confocal laser scanning microscopy………………………………………..
142
2.8. Olfactometry bioassay………………………………………………………
142
3. Results and discussion……………………………………………………………
144
3.1. Preliminary experiments for formulation optimisation…………………….
144
3.1.1. Determination of alginate type and cross-linker ion……………...
144
3.1.2. Determination of the optimal ionic strength…………………..…
145
3.2. Optimisation of the alginate formulations…………………………………
146
3.3. Characterisation of alginate beads………………………………………..
148
3.3.1. Bead texture analyses…………………………………………….
148
3.3.2. Polymeric network characterisation……………………………...
149
3.4. Semiochemical slow-release measurement…………………………………
150
3.4.1. Evaluation of the volatile collection system performances………
150
3.4.2. Semiochemical release rate measurements……………………….
151
3.5. Olfactometry bioassay………………………………………………………
153
Acknowledgments……………………………………………………………………
155
References……………………………………………………………………………
156
Chapter VI.2………………………………………………………………… 163
Objectives..………………………………………………………………………….
165
VI.2 Publication 5: A semiochemical slow-release formulation in a biological control
approach to attract hoverflies……………………………………………………………
167
Abstract………………………………………………………………………………
168
1. Introduction……………………………………………………………………….
169
2. Material and methods……………………………………………………………..
170
2.1. Chemicals and reagents…………………………………………………….
170
2.2. Semiochemical purification…………………………………………………
170
2.3. Alginate bead formulation…………………………………………………..
171
2.4. Slow-release measurement………………………………………………….
171
2.5. Fast GC analyses…………………………………………………………...
172
2.6. Field experiments……………………………………………………………
173
2.7. Data analysis………………………………………………………………..
174
3. Results…………………………………………………………………………….
175
3.1. Semiochemical release rate measurements…………………………………
175
3.2. Field-trapping experiments…………………………………………………
178
4. Discussion and conclusion………………………………………………………..
180
Acknowledgments…………………………………………………………………...
181
References……………………………………………………………………………
182
Chapter VII General Discussion, Conclusions and Perspectives………… 187
VII.1 General discussion………………………………………………………………..
189
VII.2 Conclusions and perspectives…………………………………………………….. 201
References....................................................................................................................
Chapter VIII List of scientific productions………………………………
204
209
1. Publications……………………………………………………………………….
211
2. Oral presentations…………………………………………………………………
211
3. Posters…………………………………………………………………………….
212
List of figures……………………………………………………………...
215
List of tables……………………………………………………………….
221
Chapter I
General introduction
Chapter I General Introduction
___________________________________________________________________________
General introduction
Since the seventies and the eighties, integrated pest management (IPM) strategies are more
and more developed with the intention of reducing pesticide use. Indeed, pesticides were
largely reported to induce human diseases and to be harmful for the environment (Witzgall,
2001). In this context, biological control, an IPM approach which consists in modifying the
behaviour of pest natural enemies, was recently related (Stoner, 2004). To date, this tactic is
still very few developed and the some experiments conducted to reduce pests by this manner
were generally managed by directly releasing beneficial insects (predators or parasitoids of
the pests) on infested crops. Nevertheless, sometimes invasive problems occurred when exotic
insects were employed like it was the case with the Asian ladybeetle, Harmonia axyridis
Pallas a few years ago (Huelsman et al., 2002; Roy et al., 2006; Brown et al., 2008) in order
to reduce aphid populations. For this reason, attraction of local beneficial insects by
exploiting insect communication signals is preferred.
Aphids (Homoptera: Aphididae) constitute a major economical and agricultural problem in
many temperate crops by transmitting viruses and creating direct damages by feeding phloem
of plants (Dedryver et al., 2010). Wellings et al. (1989) reported that in Europe, aphids were
directly responsible for mean annual losses of 700000 tonnes of wheat, 850000 tonnes of
potatoes and 2000000 tonnes of sugar beet. The success of their control by means of
pesticides is limited due to a series of reasons: (a) pesticides are very expensive compared to
the crop losses avoided; (b) generally pesticides are non species-specific and can cause
damages towards beneficial insects; (c) many pests develop resistances towards chemical
treatments; (d) pesticides are recognised to be unsafe for environment and human health
(Dedryver et al., 2010). The main problem to control aphid populations is due to their ability
to reproduce rapidly by parthenogenesis alternating with sexual reproduction. During sexual
reproduction, female aphids release a sex pheromone produced by glandular cells of the tibiae
to attract males. Studies led on various aphid species identified two components as aphid
sexual pheromones: (+)-(4aS, 7S, 7aR)-nepetalactone (or Z,E-nepetalactone) and (-)-(1R, 4aS,
7S, 7aR)-nepetalactol. The ratio between both compounds differed from one species to
another. These sexual pheromones also proved to act as kairomones – allelochemical
substance which benefits to the receptor of the chemical signal detrimently to the emmitor –
by attracting aphid parasitoids, Aphidius ervi Haliday (Hymenoptera: Braconidae) (Glinwood
et al., 1999). Aphids are also known to produce alarm pheromones when they are attacked by
21
Chapter I General Introduction
___________________________________________________________________________
enemies (predators or parasitoids). It results, in the colony, dispersal or defensive behaviour
after the chemical signal is perceived by conspecifics. The main alarm pheromone released by
the cornicles of most aphid species is a sesquiterpene hydrocarbon: E-β-farnesene (Bowers at
al., 1972; Edwards et al., 1973; Wientjens et al., 1973). Other molecules were also identified
as aphid alarm pheromones in various aphid species: germacrene A in Therioaphis maculate
(Bowers et al., 1977; Nishino at al., 1977) and α-pinene in Megoura viciae (Pickett et al.,
1980). Furthermore, Francis et al. (2005a) reported that in some aphid species other volatile
compounds were released in the same time than E-β-farnesene. In a recent study, Verheggen
et al. (2009) shown that E-β-farnesene, released as alarm pheromone, is also continuously
emitted by aphids in colony at lower concentration levels (on the order of 1 ng per aphid) than
as in response to predation. The authors demonstrated that the continuous emission of alarm
pheromone depends on the social environment of aphids, and the presence and abundance of
conspecifics in the colony. Similarly to nepetalactone/nepetalactol sexual pheromones, E-βfarnesene, being released continuously by Aphididae, has kairomonal activity by attracting
aphid predators like ladybeetles (Nakamuta, 1991; Zhu et al., 1999; Acar et al., 2001; Francis
et al., 2004), hoverflies (Francis et al., 2005b, Almohamad et al., 2007; Verheggen et al.,
2008, 2010) and lacewings (Boo et al., 1998; Zhu et al., 1999), and aphid parasitoids (Du et
al., 1998; Powell et al., 2003).
The use of insect communication signals (semiochemicals) to modify the behaviour of aphid
natural enemies is an interesting way very few exploited presently. The main developments of
semiochemical use in integrated pest management were achieved in mating disruption
approach. Nevertheless, some authors related biological control of aphid populations by
means of aphid sexual pheromone to attract aphid parasitoids (Powell et al., 1993; Glinwood
et al., 1998; Agelopoulos et al., 1999). In this context of aphid natural enemy attraction, E-βcaryophyllene could play a significant role. Indeed, this compound was identified as a
component of the aggregation pheromone (Brown et al., 2006; Verheggen et al., 2007) of
Harmonia axyridis Pallas ladybeetles, important aphid predators.
One major problem in the biological control strategy conducted with chemical compounds is
the unstability of such volatile molecules due to degradation by oxidation or UV light (Bruce
et al., 2005). Ideally, these volatile compounds must be formulated in slow-release devices to
be protected over time. Many semiochemical slow-release formulations and devices exist in
IPM practices, principally for the mating disruption of moths and butterflies which devastate
22
Chapter I General Introduction
___________________________________________________________________________
orchards (Kehat et al., 1994; Bradley et al.,1995; Atterholt et al., 1999; Stipanovic et al., 2004;
Cork et al., 2008) but, at present, nothing is proposed in biological control tactic against
aphids. The main characteristics of such volatile delivery systems must be: to be safe for the
environment and human health, to ensure a sufficient rate of active compounds to be
perceived by targeted insects; to release semiochemicals during a period of time long enough
to avoid highly manpower costs for the numerous replacements of releasers on fields; to
protect the semiochemical compounds against degradation due to UV light or oxygen
(Witzgall, 2001). In general, synthetic pheromones are used in IPM slow-release devices but
syntheses are commonly long and expensive for producing low quantities of compounds.
Essential oils could be a good natural alternative to multi-steps chemical syntheses in order to
obtain semiochemicals. Indeed, plants are known to produce many volatile components
similar to insect communication signals.
The present PhD thesis was realised in a goal to develop semiochemical slow-release
formulations from natural origins in order to attract predators and parasitoids of aphids to
control their populations on crops. This challenging work was achieved considering various
questions as presented hereafter:
-
How to isolate semiochemicals from plant matrixes with a high purity?
-
How to analyse and quantify semiochemicals on a fast and accurate
manner?
-
How to formulate semiochemicals in slow-release devices not harmful for
the environment and human health?
-
Is the formulation efficient in terms of semiochemical diffusion and in terms
of biological control to attract aphid predators and parasitoids?
Before presenting the results and the answers to the previous questions, a bibliographic
research was conducted in order to have an overview of the use of semiochemicals in
integrated pest management, and more particularly about diffusion studies of various slowrelease devices. The methodology of the research is presented in chapter III.
In chapter IV, we attempted to answer the first question of this thesis. The characterisation of
essential oils by gas chromatography was followed by a fractionation of the oils in order to
purify semiochemicals.
23
Chapter I General Introduction
___________________________________________________________________________
Chapter V deals with the development of a fast and accurate analytical procedure in order to
quantify semiochemicals in formulations. This chapter responds to the second question of this
work.
Chapter VI firstly describes the otpimisation and the characterisation of semiochemical
formulation, and responds to the fourth question. In a second part, this chapter gives
information on the efficiency of the formulations in terms of aphid natural enemy attractants
and in terms of semiochemical release.
General discussion, conclusion and perspectives are presented in chapter VII. The last chapter
presents all the scientific productions realised during this PhD thesis.
24
Chapter I General Introduction
___________________________________________________________________________
References
Acar E.B., Medina J.C., Lee M.L., Booth G.M. (2001). Olfactory behaviour of convergent
lady beetles (Coleoptera: Coccinellidae) to alarm pheromone of green peach aphid
(Hemiptera: Aphididae). Can. Entomol. 133, 389-397.
Agelopoulos N., Birkett M.A., Hick A.J., Hooper A.M., Pickett J.A., Pow E.M., Smart L.E.,
Smiley D.W.M., Wadhams L.J., Woodcock C.M. (1999). Exploiting semiochemicals in
insect control. Pest. Sci. 55, 225-235
Almohamad R., Verheggen F.J., Francis F., Haubruge E. (2007). Predatory hoverflies select
their oviposition site according to aphid host plant and aphid species. Entomol. Exp.
Appl. 125, 13-21.
Atterholt C.A., Delwiche M.J., Rice R.E., Krochta J.M. (1999). Controlled release of insect
sex pheromones from paraffin wax and emulsions. J. Controlled Release, 57, 233-247.
Boo K.S., Chung I.B., Han K.S., Pickett J.A., Wadhams L.J. (1998). Response of the
lacewings Chrysopa cognate to pheromones of its aphid prey. J. Chem. Ecol. 24, 631643.
Bowers W.S., Nault L.R., Webb R.E., Dutky S.R. (1972). Aphid alarm pheromones: isolation,
identification, synthesis. Science 177, 1121-1122.
Bowers W.S., Nishino C., Montgomery M.EM, Nault L.R., Nielson M.W. (1977).
Sesquiterpene progenitor, germacrene A: an alarm pheromone in aphids. Science 196,
680-681.
Bradley S.J., Suckling D.M., McNaughton K.G., Wearing C.H., Karg G. (1995). A
temperature-dependent model for predicting release rates of pheromone from a
polyethylene tubing dispenser. J. Chem. Ecol., 21(6), 745-760.
Brown A.E., Riddick E.W., Aldrich J.R., Holmes W.E. (2006). Identification of (−)-βCaryophyllene as a Gender-Specific Terpene Produced by the Multicolored Asian Lady
Beetle. J. Chem. Ecol. 32, 2489-2499.
Brown P. M. J., Adriaens T., Bathon H., Cuppen J. , Goldarazena A., Hägg T., Kenis M.,
Klausnitzer B. E. M., Kovář I., Loomans A. J. M., Majerus M. E. N., Nedved
O., Pedersen J., Rabitsch W., Roy H. E., Ternois V., Zakharov I. A., Roy D. B. (2008).
25
Chapter I General Introduction
___________________________________________________________________________
Harmonia axyridis in Europe: spread and distribution of a non-native coccinellid.
BioControl, 53(1), 5-21.
Bruce T.J.A., Birkett M.A., Blande J., Hooper A.M., Martin J.L., Khambay B., Prosser I.,
Smart L.E., Wadhams L.J. (2005). Response of economically important aphids to
components of Hemizygia petiolata essential oil. Pest Manag. Sci. 61, 1115-1121.
Cork A., De Souza K., Hall D.R., Jones O.T., Casagrande E., Krishnaiah K., Syed Z. (2008).
Development of a PVC-resin-controlled release formulation for pheromones and use in
mating disruption of yellow rice stem borer, Scirpophaga incertulas. Crop Prot., 27,
248-255.
Dedryver C.-A., Le Ralec A., Fabre F. (2010). The conflicting relationships between aphids
and men : A review of aphid damage and control strategies. C.R. Biologies 333, 539553.
Du Y., Poppy G.M., Powell W., Pickett J.A., Wadhams L.J., Woodcock C.M. (1998).
Identification of Semiochemicals Released During Aphid Feeding That Attract
Parasitoid Aphidius ervi. J. Chem. Ecol. 24, 1355-1368.
Edwards L.J., Siddall J.B., Dunham L.L., Uden P., Kislow C.J. (1973). Trans- ß-farnesene,
alarm pheromone of the green peach aphid, Myzus persicae (Sulzer). Nature 241, 126127.
Francis F., Lognay G., Haubruge E. (2004). Olfactory responses to aphid and host plant
volatile releases: (E)-ß-farnesene an effective kairomone for the predator Adalia
bipunctata. J. Chem. Ecol. 30, 741-755.
Francis F., Vandermoten S., Verheggen F., Lognay G., Haubruge E. (2005a). Is the (E)-ßfarnesene only volatile terpenoids in aphids?, J. Applied Ento. 129, 6-11.
Francis F., Martin T., Lognay G., Haubruge E. (2005b) Role of (E)-β-farnesene in systematic
aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae), Eur. J.
Entomol. 102, 431-436.
Glinwood R., Smiley D.W.M., Hardie J. Pickett J.A., Powell W., Wadhams L.J., Woodcock
C.M. (1998). Aphid sex pheromones: manipulation of beneficial insects for aphid
population control. 9th IUPAC Congress of Pesticide Chemistry, 1, Abstract 3D-0002.
26
Chapter I General Introduction
___________________________________________________________________________
Glinwood R.T., Du Y.-J., Powell W. (1999). Responses to aphid sex pheromones by the pea
aphid parasitoids Aphidius ervi and Aphidius eadyi. Entomol. Exp. Appl. 92, 227-232.
Huelsman M.F., Kovach J., Jasinski J., Young C., Eisley B. (2002). Multicolored Asian lady
beetle (Harmonia axyridis) as a nuisance pest in household in Ohio. Ohio State
University, Integrated Pest Management Program, Proceedings of the 4th International
Conference on Urban Pests, www.icup.org.uk/reports%5CICUP226.pdf. (06/07/10).
Kehat M., Anshelevich L., Dunkelblum E., Fraishtat P., Greenberg S. (1994). Sex pheromone
traps for monitoring the codling moth: effect of dispenser type, field aging of dispenser,
pheromone dose and type of trap on male captures. Entomol. Exp. Appl., 70, 55-62.
Nakamuta K. (1991). Aphid alarm pheromone component, (E)-ß-farnesene, and local search
by a predatory lady beetle, Coccinella septempunctata Bruckii mulsant (Coleoptera:
Coccinellidae). Appl. Entomol. Zool. 26, 1-7.
Nishino C., Bowers W.S., Montgomery M.E., Nault L.R., Nielson M.W. (1977). Alarm
pheromone of the spotted alfalfa aphid, Therioaphis maculata Buckton (Homoptera:
Aphididae). J. Chem. Ecol. 3, 349-357.
Pickett J.A., Griffiths D.C. (1980). Composition of alarm aphid pheromones. J. Chem. Ecol.
6, 349-360.
Powell W., Hardie J., Hick A.J., Mann J., Merritt L.A., Wadhams L.J., Wright A.,
Nottingham S.F., Holler C., Wittenrich J. (1993). Responses of the parasitoid Praon
volucre (Hymenoptera: Braconidae) to aphid sex pheromone lures in cereal fields in the
autumn: implications for parasitoid manipulation. Eur. J. Entomol. 90, 435-438.
Powell W., Pickett J.A. (2003). Manipulation of parasitoids for aphid pest management:
progress and prospects. Pest Manag. Sci. 59, 149-155.
Roy H., Brown P., Majerus M. (2006). Harmonia axyridis: a successful biocontrol agent or an
invasive threat? In: Eilenberg J. and Hokkanen H.M.T., eds. An Ecological and Societal
Approach to Biological Control. Springer Netherlands, 295-309.
Stipanovic A.J., Hennessy P.J., Webster F.X., Takahashi Y. (2004). Microparticle dispensers
for the controlled release of insect pheromones. J. Agric. Food Chem., 52, 2301-2308.
Stoner K., 2004. Approaches to the biological control of insects, University of Maine,
Cooperative Extension Bulletin 7144,
www.umext.maine.edu/onlinepubs/PDFpubs/7144.pdf (06/07/10).
27
Chapter I General Introduction
___________________________________________________________________________
Stoner, K. (2004). Approaches to the biological control of insects. University of Maine,
Cooperative Extension Bull. 7144 ,
www.umext.maine.edu/onlinepubs/PDFpubs/7144.pdf (06/07/10).
Verheggen F.J., Fagel Q., Heuskin S., Lognay G., Francis F., Haubruge E. (2007).
Electrophysiological and behavioral responses of the multicolored asian lady beetle,
Harmonia axyridis Pallas, to sesquiterpene semiochemicals. J. Chem. Ecol. 33, 21482155.
Verheggen F.J., Arnaud L., Bartram S., Gohy M., Haubruge E. (2008). Aphid and plant
volatiles induce oviposition in an aphidophagous Hoverfly. J. Chem. Ecol. 34, 301-307.
Verheggen F.J., Haubruge E., De Moraes C.M., Mescher M.C. (2009). Social environment
influences aphid production of alarm pheromone. Behav. Ecol. 20, 283-288.
Verheggen F.J., Haubruge E., Mescher M.C. (2010). Alarm pheromones. In:Litwack G., ed.,
Pheromones. Elsevier.
Wellings P.W., Ward S.A., Dixon A.F.G., Rabbinge R. (1989). Crop loss assessment, in:
Minks A.K., Harrewijn P. (Eds.), Aphids, their biology, natural enemies and control,
2C, Elsevier, NL, pp. 49-69.
Wientjens W.H.J.M., Lakwijk A.C., van der Marel T. (1973). Alarm pheromone of grain
aphids. Experientia 29, 658-660.
Witzgall P. (2001). Pheromones – future techniques for insect control? Pheromones for Insect
Control in Orchards and Vineyards IOBC wprs Bulletin, 24(2), 114-122.
Zhu J.W., Cosse A.A., Obrycki J.J., Boo K.S., Baker T.C. (1999). Olfactory reactions of the
twelve-spotted lady beetle, Coleomegilla maculate and the green lacewing, Chrysoperla
carnea to semiochemicals released from their prey and host plant: electroantennogram
and behavioral responses. J. Chem. Ecol. 25, 1163-1177.
28
Chapter II
Bibliography
Chapter II Bibliography
___________________________________________________________________________
The use of semiochemical slow-release devices in integrated pest
management strategies
Stéphanie Heuskin1,2, François J. Verheggen3, Eric Haubruge3, Jean-Paul Wathelet2, Georges
Lognay1
1
Department of Analytical Chemistry, Gembloux Agro-Bio Tech, University of Liege, Passage des
Déportés 2, B-5030 Gembloux (Belgium).
2
Department of General and Organic Chemistry, Gembloux Agro-Bio Tech, University of Liege,
Passage des Déportés 2, B-5030 Gembloux (Belgium).
3
Department of Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of
Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
Reference: Accepted for publication in Biotechnologie Agronomie Société et Environnement
(in press).
31
Chapter II Bibliography
___________________________________________________________________________
Abstract
The development of integrated pest management (IPM) strategies is increasing since many
problems appeared with the use of synthetic pesticides. Semiochemicals – informative
molecules used in insect-insect or plant-insect interaction – are more and more considered
within IPM strategies as alternative or complementary approach to insecticide treatments.
Indeed, these species-specific compounds do not present any related adversely affectation of
beneficial organisms and do not generate any risk of pest insect resistance as observed with
insecticides. Because of their complex biological activity, their dispersion in the environment
to be protected or monitored needs the elaboration of slow-release devices ensuring a
controlled release of the biologically active volatile compounds. These sensitive molecules
also need to be protected from degradation by UV light and oxygen.
Many studies were conducted on estimation of release-rate from commercialized or
experimental slow-release devices. The influence of climatic parameters and dispenser type
were estimated by previous authors in order to provide indications about the on-field
longevity of lures. The present review outlines a list of slow-release studies conducted by
many authors followed by a critical analysis of these studies.
Key words
Semiochemical, pheromone, insects, integrated pest management, biological control, slowrelease devices, predators, release rate.
32
Chapter II Bibliography
___________________________________________________________________________
Résumé
Le développement des stratégies de lutte intégrée est en croissance depuis que de nombreux
problèmes sont apparus suite à l’utilisation abusive et non raisonnée des pesticides de
synthèse. Les sémiochimiques – molécules informatives utilisées dans les interactions insecteinsecte ou plante-insecte – sont de plus en plus considérés, au sein des stratégies de lutte
intégrée, comme des approches alternatives ou complémentaires aux traitements insecticides.
En effet, ces composés, spécifiques à chaque espèce, ne présentent pas d’effets négatifs
relatés dans la littérature envers les organismes bénéfiques et n’engendrent aucun risque de
résistance chez les insectes ravageurs comme observés avec les insecticides. En raison de leur
activité biologique complexe mais de leur risque de dégradation par les rayons ultraviolets ou
à l’oxygène de l’air, leur dispersion dans l’environnement nécessite l’élaboration de systèmes
garantissant une libération lente et contrôlée des composés volatils actifs.
Plusieurs études ont été menées afin d’estimer le taux de libération de systèmes
commercialisés ou mis au point en laboratoire. L’influence des paramètres climatiques et du
type de diffuseur a été estimée par plusieurs autres auteurs afin de fournir des indications sur
la longévité des diffuseurs sur terrain. La présente revue analyse et critique une liste d’études
de systèmes à libération lente.
Mots-clés
Sémiochimique, phéromone, insectes, gestion intégrée des nuisibles, lutte biologique,
systèmes à libération lente, prédateur, taux de libération.
33
Chapter II Bibliography
___________________________________________________________________________
1. Introduction
During the seventies and the eighties, environmental and social side effects of synthetic
pesticides led to the development of integrated pest management (IPM) programs in the USA
and Asia. Since then, many IPM strategies have been successful worldwide. Indeed, the
overuse of insecticides presents many drawbacks like the appearance of insect resistances,
environmental concerns, and risks for human health. Moreover, the action of pesticides is
generally non species-specific with the risk of disturbing the natural ecological equilibrium
(Witzgall, 2001).
IPM implies various strategies, which ideally have to be combined at different levels. In 1998,
Kogan defined IPM as “a decision support system for the selection and the use of pest control
tactics, singly or harmoniously coordinated into a management strategy, based on
cost/benefit analyses that take into account the interests of and the impacts on producers,
society and the environment”.
The efficiency of these approaches needs an interdisciplinary collaboration between
agronomists, entomologists, chemists having an experience in pest behaviours, technologists,
and finally the crop producers. It is particularly true when the IPM tactic implies the use of
insect semiochemical slow-release devices as tools to modify the behaviour of insect pests.
Indeed, the release systems must be ideally economical, effective, environmentally safe
without harmful side effects, and field-tested to prove the efficiency towards targeted insects
before legal authorization and commercialization. The validation of all these manufacturing
steps is not possible without the interaction of multi-disciplinary fields of knowledge.
On an historical point of view, the role of sexual pheromones in insect mating was
demonstrated in the late 19th century. The characterization of the first insect sex pheromone
was established by Butenandt and Hecker in 1959, and was isolated from female Bombyx
mori (Lepidoptera). This technological overhang led, in the mid-seventies, to an increase of
commercial activities in synthesis of semiochemicals previously identified as potential agents
for controlling pests. This was the first step to replace synthetic insecticides with pheromone
products (Cork, 2004). In the same time, the researches on insect chemical communication
grew up and led to the emergence of a new scientific discipline: the chemical ecology. In
1971, Edward Wilson gave a definition of the chemical communication: “this is the emission
of a stimulus by one individual and which induces a reaction in another one, the reaction
34
Chapter II Bibliography
___________________________________________________________________________
being beneficial to the emitter, to the receptor or both”. In parallel, the gas chromatography
appeared in chemistry and brought simplicity in identification of volatile molecules. Rapidly,
the economical interest for using pheromone compounds in pest controls was updated and
included in integrated pest management programs (Brossut, 1997).
The present review deals with the development of the major approaches to control pests by
using semiochemical (chemical communication signal) slow-release devices. Furthermore, the
authors will focus on the techniques implemented in the study of release rates and on the
estimation of on-field longevity of semiochemical devices.
2. Semiochemicals
2.1. Definitions
Semiochemicals, from semeion (in Greek) or signal, can be defined as chemicals emitted by
living organisms (plants, insects…) that induce a behavioural or a physiological response in
other individuals. These compounds can be classified in two groups considering whether they
act
as
intraspecific
(pheromones)
or
interspecific
(allelochemicals)
mediators.
Allelochemicals include allomones (emitting species benefits), kairomones (receptor species
benefits) and synomones (both species benefit) (Figure 1). However, a single chemical signal
may act as both as pheromone and allelochemical.
Semiochemicals
fic
eci
p
s
a
Intr
s
tion
c
a
ter
in
Inte
rsp
eci
fic
inte
rac
Pheromones
tion
s
Allelochemicals
- Sex pheromones
- Aggregation pheromones
- Allomones (+ emitter)
- Alarm pheromones
- Kairomones (+ receptor)
- Trail pheromones
- Synomones (+ emitter AND receptor)
- Host marking pheromones
Figure 1: Semiochemicals in classes
35
Chapter II Bibliography
___________________________________________________________________________
There are different types of pheromones according to the response they induce on the
perceiving individuals. The most common are presented hereafter (Brossut, 1997; Cork,
2004):
Sex pheromones are generally produced by females of a species in order to attract males of
the same species for mating. Some exceptions exist where male butterflies (Bicyclus anynana
for e.g.) produce sex pheromones to seduce females during the courtship (Nieberding et al.,
2008). Sex pheromones consist in individual molecules or specific blend of compounds in a
given ratio. The most studied, and used in IPM, sex pheromones are that emitted by
Lepidoptera.
Aggregation pheromones are released by one gender of a species to attract individuals (both
sexes) of the same species in order to exploit a specific resource (food, appropriate mating
site...). They are mainly emitted by Coleopterous species.
Alarm pheromones alert conspecifics in case of threats. Generally the response behaviour
results in dispersion of congeners. These pheromones, characteristic of social or gregarious
insects, occur in some important insect pests including Aphididae and Thripidae. This class of
pheromones has potential in IPM (Verheggen et al., 2010).
Trail pheromones are present in social colonies to indicate the trail to be followed when
some scout insects locate food resource. Walking insects, like ants, typically produce these
pheromones.
Host marking pheromones reduce the competition between members of the same species,
like it is observed in parasitoids that mark a host in which they have laid an egg.
2.2. Chemistry and properties of semiochemicals
Pheromones and semiochemicals in general, consist in a wide range of organic molecules
which could be volatile or non-volatile. Non-volatile semiochemicals include cuticular
hydrocarbons, acting in mate recognition or in cannibalism regulation of several insect
species. Wilson et al. (1963) suggested that the volatile pheromones naturally exploited in
insect communication have between 5 and 20 atoms of carbon with molecular weights
ranging from 80 to 300. Those having a molecular weight above 300 are not sufficiently
36
Chapter II Bibliography
___________________________________________________________________________
volatile to allow a communication at long distance. Cork (2004), in his Pheromone manual,
cites the major pheromones identified in moths and butterflies according to their chemical
classes.
The biosynthesis of such semiochemical molecules is supposed to come from the food. They
are generally synthesized de novo by excreting cells. The biosynthesis of sexual pheromones
is well known in Lepidoptera and Diptera. In both cases, the pheromones consist in long
carbon chains (alcohols, aldehydes and acetates for Lepidoptera; hydrocarbons having high
molecular weight for Diptera) derived from the metabolism of fatty acids (Brossut, 1997).
The efficiency of semiochemical substances in chemical communication depends on various
physical properties including chemical nature, volatility, solubility and lifetime of the
molecules in the environment. An important abiotic factor controlling the effectiveness of the
pheromones is the temperature which increases the diffusion of the molecules in the air. The
stability of these volatile compounds also affects the efficiency in IPM.
3. IPM strategies using semiochemicals
There are many benefits to formulate semiochemical substances in integrated pest
management outline. These molecules are naturally occurring and are generally
environmentally friendly. Additionally, in IPM strategies the compounds are generally used at
concentrations close to those found in nature and, due to their high volatility, they can act at
long distances and dissipate rapidly. The risk to human health and environment is also
reduced compared to pesticides. For all these reasons, semiochemicals are compounds of
potentially high interest in IPM.
3.1. IPM strategies
Various strategies exist depending on the goals and scopes to achieve. Some of them are
described hereafter:
37
Chapter II Bibliography
___________________________________________________________________________
3.1.1. Monitoring
Monitoring of insect populations has generally three purposes: to detect the presence of
invasive pests; to estimate the relative density of a pest population at a specific site; to
indicate the first emergence or peak flight activity of a pest species in a given area. The
appropriate control actions (local insecticide treatment for e.g.) can then be carried out
(Weinzierl et al., 2005).
3.1.2. Trapping
Trapping with pheromone lures is a mechanical control action that consists in removing large
number of pests in an area after monitoring step. The traps can be used simultaneously with a
killing substance (“lure and kill” strategy) which has the benefit of not being in direct contact
with the crop. This technique is also useful in stored-product pest control (Phillips, 1997).
3.1.3. Mating disruption
The technique of mating disruption by using species-specific sex pheromones in large
quantity is principally applied to control moth populations in orchards. In moth, females
generally release sex pheromones to attract males, at relatively long distances (several
kilometres), for reproduction. The females lay their eggs on orchard trees and larvae develop
inside fruits which are then no more eatable.
Mating disruption consists in affecting the behaviour of males in their search of a female for
mating by releasing high quantities of synthetic female pheromones in the atmosphere. The
disruption of males can be achieved by affecting different biological mechanisms which were
originally defined by Bartell (1982). These mechanisms have been recently revised by Miller
et al. (2006a, 2006b) and were synthesized in a review by Stelinski (2007). To be an efficient
technique to control pests, surrounding orchards or fields must ideally also be part of IPM
programs. When the population of moth is too large, mating disruption can be associated with
targeted pesticides at local and punctual applications.
38
Chapter II Bibliography
___________________________________________________________________________
3.1.4. Push-pull strategy
Also called stimulo-deterrent diversion, push-pull strategy is a more recent approach than the
other described IPM practices. It consists in a combination of repellent and attractive stimuli
modifying the behaviour of insect pests and/or of their natural enemies. The insects are
deterred or repelled away from the crops (push strategy). They are simultaneously attracted by
lures (pull strategy) and concentrated in other areas where they are trapped or killed in a
controlled manner. This strategy requires a clear understanding of the pest biology, chemical
ecology, and of the interactions with hosts, conspecifics and natural enemies (Cook et al.,
2007).
3.1.5. Biological control
Biological control of the insect pests is defined (Stoner, 2004) as “the use of living organisms
(insects or pathogens) to suppress pest populations, making them less damaging than they
would otherwise be”. Insect natural enemies, also called beneficial insects, can be classified
in two classes: predators and parasitoids. Beneficial insects, sometimes exotic, can be
artificially introduced in infested fields. This practice must be cautiously managed in order to
verify that no-indigenous species will not have an adverse environmental and economic
impact, like it was the case with the introduction of the Asian ladybeetle Harmonia axyridis
Pallas (Coleoptera: Coccinellidae) (Huelsman et al., 2002; Roy et al., 2006; Brown et al.,
2008).
A new concept consists in attracting local beneficial insects on crops by means of kairomonal
substances as explained in Heuskin et al. (2009) for the biological control of aphids with their
parasitoid wasps (Aphidius ervi Haliday (Hymenoptera: Braconidae) (Du et al., 1998; Powell
et al., 2003) and their hoverflies predators (Episyrphus balteatus De Geer (Diptera:
Syrphidae) (Francis et al., 2005; Verheggen et al., 2008; Verheggen et al., 2009).
39
Chapter II Bibliography
___________________________________________________________________________
4. Slow release of semiochemicals
4.1. Slow release dispensers
Major volatile semiochemicals being extremely unstable due to their chemical structure, it is
necessary to formulate them so that they are protected from degradation by UV light and
oxygen. Moreover, the formulation must ensure a controlled release of semiochemicals. To be
efficient in IPM strategies, semiochemical slow-release devices must have particular
specifications: the aerial concentration after release must be sufficiently high to be detected by
insects; the release of semiochemical must be effective during all the period of insect
occurrence; the production of dispenser must be reproducible. The application of dispensers
must be realized early in the season when the pest density is not too high, given that their
release rates, for the majority of devices, decrease with time (Witzgall, 2001).
Several formulations and dispensers have been developed and commercialized with various
slow-release capacities. Some examples of dispensers are described hereafter. The majority of
them involve mating disruption of moth. Three groups can be distinguished: solid matrix
dispensers, liquid formulations to spray and reservoirs of formulations. On an historical point
of view, the first related and the most commonly used pheromone dispenser is the natural
rubber septum (Roelofs et al., 1972).
Solid matrix dispensers are hand-applied on crops or in orchards. The semiochemicals are
incorporated in a solid matrix. Because of the various materials that can be used to constitute
a matrix, the release rates for a single molecule, can differ significantly from one device to
another, as demonstrated by Golub et al. (1983) for the measurement of release rate of
gossyplure ((Z,Z)- and (E,Z)-7,11-hexadecadien-1-yl acetate), the sex pheromone blend of the
pink bollworm (Pectinophora gossypiella Saunders, Lepidoptera: Gelechiidae) from different
formulations.
The most common solid matrix used in dispensers are polyethylene tubes (twist tie dispensers
like Isomate®), polyethylene sachets (Torr et al., 1997), polyethylene vials (Johansson et al.,
2001; Zhang et al., 2008), membrane dispensers (CheckMate CM-XL®), spiral polymer
dispensers (NoMate CM®) (Tomaszewska et al., 2005), polymer films, rubber septa
(McDonough, 1991; Möttus et al., 1997), rubber wicks, polyvinyl chloride (PVC), hollow
40
Chapter II Bibliography
___________________________________________________________________________
fibers (Golub et al., 1983), impregnated ropes, wax formulations, gel-like dispensers matrices
(Atterholt et al., 1999).
Drawbacks encountered with solid matrix dispensers include the difficulty to maintain a zeroorder release kinetic (constant release rate) during a long period of time, and the decreasing of
aerial semiochemical concentration with the distance from the dispenser. Consequently, these
dispensers are only efficient to attract and trap insects at short distance. A way to by-pass this
problem is to apply devices in sufficient sites in the crop or in the orchard. The resulting
disadvantage is the high manpower needed for application of dispensers in the fields. Another
shortcoming is the non biodegradability of the formulated polymers (Stipanovic et al., 2004).
The effective lifetime of the biggest solid matrix dispensers can range from 60 to 140 days.
Sprayable slow-release formulations are generally composed of a biodegradable liquid
matrix compound in which the semiochemical is dissolved. Regularly, other components can
be added to protect the semiochemicals, like UV-stabilizers, antioxidants and surfactants.
Frequently, the sprayable formulation consists in a micro emulsion, resulting in polymeric
micro beads containing the semiochemicals (microencapsulated pheromones) dispersed in a
liquid matrix (de Vlieger, 2001). In 1999, Atterholt et al. studied the release rates of oriental
fruit moth sexual pheromones formulated in aqueous paraffin emulsions as carrier material.
The time of efficiency of such formulations ranges from days to weeks depending on
environmental factors, micro bead size, release capacities, and the pheromones chemical
properties (Welter et al., 2005).
The major advantage of sprayable formulations compared to solid matrix dispensers is that
the entire crop can be treated.
Reservoir dispensers generally consist in two parts, a reservoir and a diffusion area. Hofmeyr
et al. (1995) described a dispenser consisting in glass tube acting as a pheromoneimpermeable reservoir attached to a short polyethylene tube through which the pheromone
can diffuse. Another reservoir was tested by Shem et al. (2009) as repellent allomone device
against tsetse flies. The upper part (reservoir) was made of aluminum and the diffusion area
was made from Tygon® silicon tubing.
41
Chapter II Bibliography
___________________________________________________________________________
Aerosol emitters (Suttera® puffer for e.g.), consisting in electronically programmed reservoirs
of formulation, release large amounts of pheromone by means of a pressurized aerosol. Puffs
can be emitted at fixed time intervals. The advantage of this system is the use of fewer
dispensers per surface to treat.
Reservoir systems are the most suitable to approach zero-order release kinetic of
semiochemicals (Atterholt et al., 1999).
4.2. Slow release rate studies
Release rate study does not specify the biological efficiency of a semiochemical delivery
dispenser, but gives an idea of the release kinetic over time according to climatic conditions.
Many dispensers do not guarantee a release at a steady rate, inducing a decrease of release
rate during the season. However, the most important is to know at which moment the quantity
of released semiochemical is no more sufficient to influence insect behaviour, and to change
the dispenser.
4.2.1. Techniques to estimate release rates
Given that it is not easy and reliable to measure release rates directly in the field, estimations
of semiochemical release rates from formulations were performed in laboratory or semicontrolled conditions. Three different techniques were improved over time: the gravimetric
method, the total organic solvent extraction, and the dynamic collection of volatiles. The first
procedure, less and less used, consists in weighing dispensers at daily intervals over the
season and to determine the percentage of mass loss with time. The major weakness of this
technique is the lack of precision and accuracy to set up release rates. Sometimes, the mass
increases instead of decreasing due to the presence of humidity and dust deposited on the
dispensers.
The second technique implies the total organic solvent extraction of semiochemicals from
dispensers to determine the residual concentration of compound in field-aged devices. The
condition to have an optimal pheromone extraction implies the complete dissolution of
compound contained in the dispenser (Lopez et al., 1991; Möttus et al., 1997). This technique
has the benefit to permit to qualify and quantify the pheromone and its potential volatile
42
Chapter II Bibliography
___________________________________________________________________________
degradation products by gas chromatography (GC) analysis. However, it presents a risk of not
permitting detection of non volatile degradation products by GC (Tomaszewska et al., 2005).
The third method to determine release rate consists in a dynamic sampling and an adsorbent
trapping of volatile compounds from field-aged dispensers. The evolution of release rate is
estimated according to field-age of devices. It is essential to measure the rate every time in the
same conditions of atmospheric pressure, temperature, relative humidity and airflow to obtain
analogous analyses over time. The volatile collection system is generally composed of a
chamber in which air flows through the dispenser. The carried volatile semiochemicals are
trapped on an adsorbent cartridge, followed by solvent extraction or thermal desorption, and
GC analysis. Various adsorbents have been tested like Super Q (Mayer et al., 1998; Atterholt
et al., 1999; Meagher, 2002), silica gel (McDonough et al., 1992; Pop et al., 1993), Tenax
(Cross, 1980), Carbograph, Porapak Q (Cross et al., 1980), activated charcoal, polyurethane
foam (PUF) (Van der Kraan et al., 1990; Tomaszewska et al., 2005). The choice of the
adsorbent depends on the semiochemical properties, and on the maximum airflow to apply on
the cartridge without breakthrough of the compounds.
Considering the advantages and shortcomings of the three techniques, the last one is the most
appropriate and accurate in order to estimate release rate of semiochemicals from dispensers.
4.2.2. Release rate studies
The release of volatile semiochemicals in the atmosphere is reliant on two major factors: the
diffusion speed of the compound through the dispenser matrix and the evaporation speed of
the molecule in the air (Krüger et al., 2002). The first factor depends on the characteristics of
the dispenser (type of matrix (Golub et al., 1983), size (Hofmeyr at al., 1995), shape,
thickness, distribution of the semiochemical in the matrix (Stipanovic et al., 2004)) while the
second factor (speed of evaporation) mainly relies on environmental parameters like air
temperature, wind speed, relative humidity, and the physical properties of the compound
itself. (Alfaro-Cid et al., 2009; http://www.cbceurope.it/images/stories/file/biocontrol/Guida
BioENG.pdf). In the case where the evaporation process of pheromone from the surface of
dispenser is slower than the diffusion step, the speed of evaporation is the limiting factor, and
the first-order release kinetic equation is considered:
C0 = Ct e-kt,
43
Chapter II Bibliography
___________________________________________________________________________
where C0 is the amount of compound in the dispenser at the beginning of evaporation, Ct is
the amount of compound at time t, and k is the evaporation rate constant. In case of a firstorder kinetic, a half of the amount of the pheromone compound will be evaporated after a
time t1/2, called half-life of the compound (McDonough et al., 1989; Möttus et al., 2001).
Many studies were conducted to give an estimation of the release rate of pheromone over time
from dispensers in definite experimental conditions. However, very few studies dealt with the
conception of rate kinetic predictive models according to abiotic parameters (temperature,
relative humidity, wind speed…). Moreover, these experiments checked parameters one by
one rather than considering their combination regarding an experimental design to finally
obtain a realistic rate modeling, close to the kinetic expected on the field.
Table 1 summarizes studies considering the type of dispenser, the semiochemicals and insects
of the research, the targeted crop and the main conclusions of the release rate evaluation. Most
studies concluded to first-order release kinetics, semiochemical rates decreasing with time and
release being dependent on the amount of compound present in the dispenser. Already in
1979 and 1981, Butler et al. showed that alcohol and acetate molecules (sex pheromones of
many moth species) were released from rubber septa following a first-order kinetic. Indeed,
they concluded that pheromone molecular sizes, double bond positions and isomers
conditioned the evaporation rates and the half-life times of the molecules. McDounough et al.
(1992) described a modeling of pheromone (codling moth sex pheromones) release rate by
determining the half-life times of compounds delivered from field-aged hollow plastic tube
dispensers. In 1994, Kehat et al. also found that these codling moth sex pheromones were
desorbed from field-aged rubber septa dispensers following a first-order kinetic. Zhang et al.
(2008) measured release rate of female sex pheromones of cocoa pod borer, Conopomorpha
cramerella, from polyethylene vials placed in a fume hood (20-25°C; 129 ft/min face
velocity). They obtained the same kinetic of pheromone delivery. PVC-resin controlled
release formulations developed by Cork et al. (2008) for the delivery of yellow rice stem
borer sex pheromones were tested at various temperatures (from 22°C to 34°C). Releases
followed a first-order kinetic. Moreover, the temperature highly influenced pheromone rates,
half-lives decreasing with an increase of the temperature. Considering several other studies,
temperature is one of the most important climatic parameter that affects volatile release rates.
In 1990, Van der Kraan et Ebbers determined the influence of temperature and air velocity on
a variety of dispensers delivering moth sex pheromones (tetradecen-1-ol acetate). The authors
44
Chapter II Bibliography
___________________________________________________________________________
concluded that the impact of temperature was more important than wind speed on the kinetic
of release. Bradley et al. (1995) proposed a linear rate-temperature relationship model to
predict release of light brown apple moth pheromones (E11-14:OAc; E9,E11-14: OAc; Z1114: OAc) from polyethylene tubing dispensers. Two years later, Torr et al. (1997) studied the
release of tsetse flies kairomonal substances from polyethylene sachets. Even though release
rates were independent of the semiochemical amount present in the dispenser, they increased
exponentially with temperature. Atterholt et al. (1999) investigated the release of oriental fruit
moth pheromone from paraffin emulsions at three temperatures from 27°C to 49°C. At the
lowest temperature, the release rate was constant over time (during 100 days). The release rate
was higher at 38°C and 49°C. However, the rate decreased with time at these highest
temperatures due to pheromone oxidation and degradation phenomena. Once again, in 2001,
Johansson et al. illustrated the increase of sawflies sex pheromone release rate with
temperature from polyethylene vial dispensers. More recently, Shem et al. (2009) studied the
influence of temperature on the release rate of a blend of allomones derived from waterbuck
odor (carboxylic acids, ketones, 2-methoxyphenol, δ-octalactone), in a reservoir type
dispenser, to control tsetse flies. As expected, the release rate increased according to the
temperature.
It is not easy to develop and formulate semiochemical delivery systems, which guarantee the
diffusion of effective amount of compound along the season. With first-order release kinetics,
semiochemical rates decrease quickly and, as a consequence dispenser field-life is often too
short to cover the period of pest occurrence.
45
Table 1: Development of semiochemical dispensers and formulations, and release rate studies.
A.
Type of dispenser or
formulation
Solid dispensers
Author
(Year)
Semiochemicals and target insect
Protected crop
Torr et al. (1997)
Tsetse flies (Diptera: Glossinidae: Glossina
sp.) kairomones: 1-Octen-3-ol, 4-methylphenol
and 3-n-propylphenol.
No crop (trapping).
Gravimetric and volatile collection methods.
Release rates are independent of the amount present in dispenser
(zero-order release kinetic), are related directly to surface area,
inversely related to wall thickness and increase exponentially with
temperature.
Rochat et al. (2002)
Male aggregation pheromone of Dynast beetle,
Scapanes australis Bsdv. (Coleoptera:
Scarabaeidae): 2-butanol, 3-hydroxy-2butanone, 2,3-butanediol.
Coconut.
No rate study.
Van der Kraan et al. (1990)
Lepidopteran sex pheromones for mating
disruption: tetradecen-1-ol acetates (Z9-14: Ac
and Z11-14 : Ac).
Orchards
Comparison of polyethylene tubes with other dispenser materials.
Volatile collection on Polyurethane foam (PUF) cartridges +
solvent elution.
Release rate depends on: type of dispenser, temperature, wind
velocity.
McDonough et al. (1992)
Codling moth, Cydia pomonella L.
(Lepidoptera: Olethreutidae), mating disruptant
blend: (E,E)-8,10-dodecadien-1-ol / dodecan1-ol / tetradecan-1-ol.
Orchards
Volatile collection on silica gel cartridges + solvent elution.
Release rate is function of the change of pheromone content with
time (first-order release kinetic).
Bradley et al. (1995)
Sex pheromones of light brown apple moth,
Epiphyas postvittana (Walker) (Lepidoptera:
Tortricidae): E11-14: OAc / E9,E11-14:OAc /
Z11-14:OAc.
Orchards
Volatile collection + measure of liquid pheromone length over
time.
Comparison of measured and modeled predicted release rate
considering real on-field measured temperature (linear release rate
– temperature relationship).
Pine trees.
Gravimetric method.
Release rates increase with temperature.
Cacao, Theobroma
cacao L.
Total solvent extraction method .
First-order release rate.
Polyethylene sachets
Polyethylene vials and tubes
(for e.g. Shin-Etsu®)
Release rate studies (method of measurement and
observations)
Johansson et al. (2001)
Zhang et al. (2008)
Sex pheromones of sawflies: Neodiprion
sertifer Geoffr. and Diprion pini L.
(Hymenoptera: Diprionidae).
Acetates of pentadecanol / (2S, 3S, 7S)-3,7dimethyl-2-tridecanol / (2S, 3R, 7R)-3,7dimethyl-2-tridecanol.
Female sex pheromones of cocoa pod borer,
Conopomorpha cramerella (Snellen)
(Lepidoptera: Gracillariidae): (E,Z,Z)- and
(E,E,Z)-4,6,10-hexadecatrienyl acetates and
corresponding alcohols.
Butler et al. (1979, 1981)
Rubber septa
Kehat et al. (1994)
Hollow fibers
Golub et al. (1983)
Lopez et al. (1991)
Plastic dispensers (PVC, PVC-resin…)
Cork et al. (2008)
B.
Sex pheromones of codling moth, Cydia
pomonella L. (Lepidoptera: Olethreutidae):
(E,E)-8,10-dodecadien-1-ol
Sex pheromone blend of the pink bollworm
(Pectinophora gossypiella Saunders
(Lepidoptera: Gelechiidae): (Z,Z)- and (E,Z)7,11-hexadecadien-1-yl acetate.
Sex pheromones of Helicoverpa zea (Boddie)
(Lepidoptera: Noctuidae): (Z)-11-hexadecenal
/ (Z)-9-hexadecenal / (Z)-7-hexadecenal.
Sex pheromones of yellow rice stem borer
Scirpophaga incertulas (Walker) (Lepidoptera:
Pyralidae): (Z)-9-hexadecenal / (Z)-11hexadecenal.
Orchards.
Apple and pear
orchards.
Total solvent extraction method.
Pheromone molecular size is one of the major features determining
evaporation rates in rubber septa.
Double bond positions and isomers condition the half-lives.
Volatile collection on Porapak Q cartridges + solvent elution.
Release rates decrease with field aging of dispensers.
Orchards.
Total solvent extraction.
Comparison of release rate for various devices: hollow fibers, red
rubber septa, red rubber wick. Rate is different according to the
type of matrix.
Corn and cotton
fields.
Volatile collection on Tenax cartridges + solvent elution.
Linear decrease of release rate with time.
Rice crops.
Total solvent extraction.
Half lives of pheromone decrease with an increase of temperature.
First-order release rate kinetic.
Orchards.
Volatile collection on Super Q cartridges + solvent elution.
Release rate:
- is dependent of the formulation and the evaporative surface area.
- increases with temperature.
Orchards.
Gravimetric method.
Release rate depends on coating of the microcapsule, surface area,
micropore volume.
Orchards.
Gravimetric method.
Release rate is function of the size of polyethylene tubing.
No crop
Gravimetric method.
Release rate is dependent of the temperature. The compounds in
the blend interact with each other. The rate kinetic is different for
one compound (zero-order) and for the blend (first-order).
Sprayable formulations
Paraffin emulsions
Atterholt et al. (1999)
Microcapsules: pheromone
immobilized on a porous substrate
coated with a polymer film membrane.
Stipanovic et al. (2004)
C.
Alcohol and acetate molecules found as sex
pheromones of various moth species.
Oriental fruit moth Grapholita molesta (Busck)
(Lepidoptera: Tortricidae) mating disruptant
blend : (Z)-8-dodecen-1-yl-acetate / (E)-8dodecen-1-yl-acetate / (Z)-8-dodecen-1-ol.
Sex pheromones of codling moth, Cydia
pomonella L. (Lepidoptera; Olethreutidae)
(codlemone: (E,E)-8,10-dodecadien-1-ol) and
gypsy moth, Lymantria dispar L. (Lepidoptera:
Lymantriidae) (disparlure: (Z)-7,8-epoxy-2methyloctadecane).
Reservoirs
Home-made reservoir dispensers:
glass and polyethylene tubing
Reservoir with silicon diffusion area
Hofmeyr et al. (1995)
Shem et al. (2009)
Pheromone trap blend against false codling
moth, Cryptophlebia leucotreta (Meyr.)
(Lepidoptera: Tortricidae): (E)-7-dodecenyl
acetate / (E)-8-dodecenyl acetate / (Z)-8dodecenyl acetate.
Blend of allomones (waterbuck odour)
(carboxylic acids, ketones, δ-octalactone, 2methoxy-4-methyl phenol) against tsetse fly
(Diptera: Glossinidae: Glossina sp.).
Chapter II Bibliography
___________________________________________________________________________
5. Conclusions
At the end of this review, two questions remain: what kind of dispenser is the best in IPM
programs? What is the lifetime of dispenser in terms of semiochemical diffusion efficiency?
To answer the first question, the choice of dispenser (solid matrix, formulation, reservoir,
puffer) will mainly depend on the needs of the crop farmers, taking into account the labour
and the manpower costs to implement IPM strategies. Other important decisional criteria are
the targeted pest, the season of occurrence of the insects (with the knowledge of the mean
climatic conditions) and the IPM tactic itself. Moreover, environment protection can also be
determinative in the dispenser selection. Biodegradable matrix, environmentally safe, could
be preferred as slow-release device material for semiochemical delivery. Alfaro-Cid et al.
(2009) recently attempted to develop an eco-friendly biodegradable dispenser for codling
moth mating disruption. Additionally this experimental system seemed to have small
sensitivity to climatic conditions.
The second question implies the knowledge of the semiochemical release rate kinetic. As
demonstrated all along the review, this kinetic relies on the type of molecule, the dispenser,
and the climatic conditions. The perspective to develop case by case (semiochemicaldispenser) predictive slow-release models taking into account the climatic parameters is an
ideal but difficult approach. Experiments conducted to reproduce the environmental
conditions faced the constraint that the fluctuations observed in field are too unpredictable
and random to be duplicated in laboratory. The laboratory studies can only predict limitations
of use in fixed conditions and give theoretical information on dispenser lifetime. Furthermore,
such studies are generally time and money consuming. For these reasons, the best way to
estimate diffusion efficiency consists in regularly measuring the residual semiochemical
quantity and/or determining release rate from field-aged dispensers. This approach, generally
less time consuming, gives a direct indication of the dispenser release effectiveness and the
moment to replace pheromone delivery system on field.
In conclusion, the perspectives of semiochemical use in IPM programs seem to be promising
with the increasing worldwide biological agriculture. Slow-release dispenser and formulation
improvement will continue with the contribution of multiple scientific fields of research
(entomology, chemistry, ecology…) and the crop farmer skills.
48
Chapter II Bibliography
___________________________________________________________________________
References
Alfaro-Cid E., Esparcia-Alcazar A.I., Moya P., Femenia-Ferrer B., Sharman K., Merelo J.J.,
2009. Modeling pheromone dispensers using genetic programming. In: M. Giacobini et
al., eds. Evo Workshops 2009. LNCS 5484, Springer-Verlag Berlin Heidelberg, 635644.
Atterholt C.A., Delwiche M.J., Rice R.E., Krochta J.M., 1999. Controlled release of insect sex
pheromones from paraffin wax and emulsions. J. Controlled Release, 57, 233-247.
Bartell R.J., 1982. Mechanisms of communication disruption by pheromone in the control of
Lepidoptera: a review. Physiol. Entomo., 7, 353-364.
Bradley S.J., Suckling D.M., McNaughton K.G., Wearing C.H., Karg G., 1995. A
temperature-dependent model for predicting release rates of pheromone from a
polyethylene tubing dispenser. J. Chem. Ecol., 21(6), 745-760.
Brossut R., 1997. Phéromones: La communication chimique chez les animaux. Belin : Croisée
des Sciences.
Brown P. M. J., Adriaens T., Bathon H., Cuppen J. , Goldarazena A., Hägg T., Kenis M.,
Klausnitzer B. E. M., Kovář I., Loomans A. J. M., Majerus M. E. N., Nedved
O., Pedersen J., Rabitsch W., Roy H. E., Ternois V., Zakharov I. A., Roy D. B., 2008.
Harmonia axyridis in Europe: spread and distribution of a non-native coccinellid.
BioControl, 53(1), 5-21.
Butler L.I., McDonough L.M., 1979. Insect sex pheromones: Evaporation rates of acetates
from natural rubber septa. J. Chem. Ecol., 5(5), 825-837.
Butler L.I., McDonough L.M., 1981. Insect sex pheromones: evaporation rates of alcohols
and acetates from natural rubber septa. J. Chem. Ecol., 7(3), 627-633.
Cook S.M., Khan Z.R., Pickett J.A., 2007. The use of push-pull strategies in integrated pest
management. Annu. Rev. Entomol., 52, 375-400.
Cork A., 2004. Pheromone Manual. Natural Resources Institute, Chatham Maritime ME4
4TB, UK.
Cork A., De Souza K., Hall D.R., Jones O.T., Casagrande E., Krishnaiah K., Syed Z., 2008.
Development of a PVC-resin-controlled release formulation for pheromones and use in
49
Chapter II Bibliography
___________________________________________________________________________
mating disruption of yellow rice stem borer, Scirpophaga incertulas. Crop Prot., 27,
248-255.
Cross J.H., 1980. A vapor collection and thermal desorption method to measure
semiochemical release rates from controlled release formulations. J. Chem. Ecol., 6(4),
781-787.
Cross J.H., Tumlinson J.H., Heath R.E., Burnett D.E., 1980. Apparatus and procedure for
measuring release rates from formulations of lepidopteran semiochemicals. J. Chem.
Ecol., 6(4), 759-770.
De Vlieger J.J., 2001. Development of a sprayable slow-release formulation for the sex
pheromone of the Mediterranean corn borer, Sesamia nonagroides. IOBC wprs Bulletin,
24(2), 101-106.
Du Y., Poppy G.M., Powell W., Pickett J.A., Wadhams L.J., Woodcock C.M., 1998.
Identification of Semiochemicals Released During Aphid Feeding That Attract
Parasitoid Aphidius ervi. J. Chem. Ecol., 24(8), 1355-1368.
Francis F., Martin T., Lognay G., Haubruge E., 2005. Role of (E)-β-farnesene in systematic
aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae), Eur. J.
Entomol. 102, 431-436.
Golub M., Weatherston J., Benn M.H., 1983. Measurement of release rates of gossyplure
from controlled release formulations by mini-airflow method. J. Chem. Ecol., 9(3), 323333.
Heuskin S., Godin B., Leroy P., Capella Q., Wathelet J.-P., Verheggen F., Haubruge E.,
Lognay G., 2009. Fast gas chromatography characterisation of purified semiochemicals
from essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L.
(Lamiaceae). J. Chromatogr. A, 1216, 2768-2775.
Hofmeyr. H., Burger B.V., 1995. Controlled-release pheromone dispenser for use in traps to
monitor flight activity of false codling moth. J. Chem. Ecol., 21(3), 355-363.
Huelsman M.F., Kovach J., Jasinski J., Young C., Eisley B., 2002. Multicolored Asian lady
beetle (Harmonia axyridis) as a nuisance pest in household in Ohio. Ohio State
University, Integrated Pest Management Program, Proceedings of the 4th International
Conference on Urban Pests, www.icup.org.uk/reports%5CICUP226.pdf. (06/07/10).
50
Chapter II Bibliography
___________________________________________________________________________
Johansson B.G., Anderbrant O., Simandl J., Avtzis N.D., Salvadori C., Hedenström E.,
Edlund H., Högberg H.-E., 2001. Release rates for pine sawly pheromones from two
types of dispensers and phenology of Neodiprion sertifer. J. Chem. Ecol., 27(4), 733745.
Kehat M., Anshelevich L., Dunkelblum E., Fraishtat P., Greenberg S., 1994. Sex pheromone
traps for monitoring the codling moth: effect of dispenser type, field aging of dispenser,
pheromone dose and type of trap on male captures. Entomol. Exp. Appl., 70, 55-62.
Kogan M., 1998. Integrated pest management: historical perspectives and contemporary
developments. Annu. Rev. Entomol., 43, 243-270.
Krüger A.J., Tolmay A.T., 2002. Prediction of the release characteristics of alcohols from
EVA using a model based on Fick’s second law of diffusion. J. Appl. Poly. Sci., 84,
806-813.
Lopez J.D., Leonhardt B.A., Shaver T.N., 1991. Performance criteria and specifications for
laminated plastic sex pheromone dispenser for Helicoverpa zea (Lepidoptera:
Noctuidae). J. Chem. Ecol., 17(11), 2293-2305.
Mayer M.S., Mitchell E.R., 1998. Rapid measure of sex pheromone emission from plastic
rope dispensers: exemple of utility in sexual communication disruption of the
diamondback moth, Plutella xylostella. Phytoparasitica, 26(2), 1-9.
McDonough L.M., 1991. Controlled release of insect sex pheromones from a natural rubber
substrate. In: Hedin P.A., ed. Naturally Occurring Pest Bioregulators. American
Chemical Society, 106-124.
McDonough L.M., Brown D.F., Aller W.C., 1989. Insect sex pheromones. Effect of
temperature on evaporation rates of acetates from rubber septa. J. Chem. Ecol., 15(3),
779-790.
McDonough L.M., Aller W.C., Knight A.L., 1992. Performance characteristics of a
commercial controlled-release dispenser of sex pheromones for control of codling moth
(Cydia pomonella) by mating disruption. J. Chem. Ecol., 18(12), 2177-2189.
Meagher R.L. Jr, 2002. Trapping noctuid moths with synthetic floral volatile lures. Entomo.
Exp. Appli., 103, 219-226.
51
Chapter II Bibliography
___________________________________________________________________________
Miller J.R., Gut L.J., de Lame F.M., Stelinski L.L., 2006a. Differentiation of competitive vs.
non-competitive mechanisms mediating disruption of moth sexual communication by
point sources of sex pheromone (Part I): Theory. J. Chem. Ecol., 32 (10), 2089-2114.
Miller J.R., Gut L.J., de Lame F.M., Stelinski L.L., 2006b. Differentiation of competitive vs.
non-competitive mechanisms mediating disruption of moth sexual communication by
point sources of sex pheromone (Part II): Case Studies. J. Chem. Ecol., 32 (10), 21152143.
Möttus E., Nômm V., Williams I.H., Liblikas I., 1997. Optimization of pheromone dispensers
for diamondback moth Plutella xylostella. J. Chem. Ecol., 23(9), 2145-2159.
Möttus E., Liblikas I., Ojarand A., Kuusik S., Nikolaeyva Z., Ovsjannikova E., Borg-Karlson
A.-K., 2001. Calculation and using of pheromone communication channel parameters
for optimization of pheromone dispensers. In: Metspalu L., Mitt S., eds. Practice
oriented results on the use of plant extracts and pheromones in pest control.
Proceedings of the international workshop, Estonia, Tartu, 24-25 January 2001, 101125.
Nieberding C, Schneider MV, de Vos H, Lassance JM, Estramil N, Andersson J, Bång J,
Hedenström E, Lofstedt C, Brakefield P, 2008. Male sex pheromones in the butterfly
Bicyclus anynana: towards an evolutionary analysis. PLoS ONE 3(7):e2751.
Phillips T.W., 1997. Semiochemicals of stored-product insects: research and applications. J.
stored Prod. Res., 33(1), 17-30.
Pop L., Arn, H., Buser H.-R., 1993. Determination of release rates of pheromone dispensers
by air sampling with C-18 bonded silica. J. Chem. Ecol., 19(11), 2513-2519.
Powell W., Pickett J.A., 2003. Manipulation of parasitoids for aphid pest management:
progress and prospects. Pest Manag. Sci., 59(2), 149-155.
Rochat D., Morin J.-P., Kakul T., Beaudoin-Olivier L., Prior R., Renou M., Malosse I.,
Stathers T., Embupa S., Laup S., 2002. Activity of male pheromone of Melanesian
rhinoceros beetle Scapanes australis. J. Chem. Ecol., 28(3), 479-500.
Roelofs W.L., Bartell R.J., Hill A.S., Cardé R.T., Waters L.A., 1972. Codling moth sex
attractant-field trials with geometrical isomers. J. Econ. Ento., 65, 1276-1277.
52
Chapter II Bibliography
___________________________________________________________________________
Roy H., Brown P., Majerus M., 2006. Harmonia axyridis: a successful biocontrol agent or an
invasive threat? In: Eilenberg J. and Hokkanen H.M.T., eds. An Ecological and Societal
Approach to Biological Control. Springer Netherlands, 295-309.
Shem P.M., Shiundu P.M., Gikonyo N.K., Ali A.H., Saini R.K., 2009. Release kinetics of a
synthetic tsetse allomone derived from waterbuck odour from a Tygon silicon dispenser
under laboratory and semi field conditions. Amer.-Euras. J. Agric. & Environ. Sci., 6(6),
625-636.
Stelinski L.L., 2007. On the physiological and behavioral mechanisms of pheromone-based
mating disruption. Pestycydy, 3(4), 27-32.
Stipanovic A.J., Hennessy P.J., Webster F.X., Takahashi Y., 2004. Microparticle dispensers
for the controlled release of insect pheromones. J. Agric. Food Chem., 52, 2301-2308.
Stoner K., 2004. Approaches to the biological control of insects, University of Maine,
Cooperative Extension Bulletin 7144,
www.umext.maine.edu/onlinepubs/PDFpubs/7144.pdf (06/07/10).
Tomaszewska E., Hebert V.R., Brunner J.F., Jones V.P., Doerr M., Hilton R., 2005.
Evaluation of pheromone release from commercial mating disruption dispensers. J.
Agric. Food Chem., 53, 2399-2405.
Torr S.J., Hall D.R., Phelps R.J., Vale G.A., 1997. Methods for dispensing odour attractants
for tsetse flies (Diptera: Glossinidae). Bull. Entomol. Res., 87, 299-311.
Van der Kraan C., Ebbers A., 1990. Release rates of tetradecen-1-ol acetates from polymeric
formulations in relation to temperature and air velocity. J. Chem. Ecol., 16(4), 10411058.
Verheggen F.J., Arnaud L., Bartram S., Gohy M., Haubruge E., 2008. Aphid and plant
volatiles induce oviposition in an aphidophagous Hoverfly. J. Chem. Ecol., 34(3), 301307.
Verheggen F.J., Haubruge E., De Moraes C.M., Mescher M.C., 2009. Social environment
influences aphid production of alarm pheromone. Behav. Ecol., 20, 283-288.
Verheggen F.J., Haubruge E., Mescher M.C., 2010. Alarm pheromones. In:Litwack G., ed.,
Pheromones. Elsevier.
53
Chapter II Bibliography
___________________________________________________________________________
Weinzierl R., Henn T., Koehler P.G., Tucker C.L., 2005. Insect attractants and traps.
University of Florida, IFAS Extension, http://edis.ifas.ufl.edu/in080, (05/05/10).
Welter S.C., Pickel C., Millar J., Cave F., Van Steenwyk R.A., Dunley J., 2005. Pheromone
mating disruption offers selective management options for key pests. California Agric.,
59(1),
16-22,
http://fruitsandnuts.ucdavis.edu/uops/pheromone_welter_03_05.pdf,
(16/05/10).
Wilson E.O., Bossert W.H., 1963. Chemical communication among animals. Recent Progr. In
Hormone Res., 19, 673-716.
Witzgall P., 2001. Pheromones – future techniques for insect control? Pheromones for Insect
Control in Orchards and Vineyards IOBC wprs Bulletin, 24(2), 114-122.
Zhang A., Kuang L.F., Maisin N., Karumuru B., Hall D.R., Virdiana I., Lambert S., Purung
H.B., Wang S., Hebbar P., 2008. Activity Evaluation of cocoa pod borer sex pheromone
in cacao fields. Environ. Entomol., 37(3), 719-724.
http://www.cbceurope.it/images/stories/file/biocontrol/GuidaBioENG.pdf
54
Chapter III
Methodology of the research
Chapter III Methodology of the research
___________________________________________________________________________
Methodology of the research
The global objective of the present thesis consisted in developing semiochemical slow-release
formulations as biological control devices attractive towards aphid natural enemies. On the
basis of the literature, two sesquiterpene compounds were considered in this research for their
respective properties as insect communication signals and more precisely as aphid predator
and parasitoid attractants: E-β-farnesene and E-β-caryophyllene. A third molecule, Z,Enepetalactone, one component of the sex pheromone of some aphid species, was also
investigated for its role as semiochemical. Nevertheless, after preliminary olfactory bioassays
led on parasitoids (Aphidius ervi Haliday) and predators (Episyrphus balteatus De Geer), this
molecule did not induce any attractive behaviour and future researches on it were given up.
Two essential oils, Matricaria chamomilla L. and Nepeta cataria L., were considered as
natural matrixes to obtain the previously cited compounds.
The methodology of the research is presented in Figure 1. The first step of the work consisted
in a minutious characterisation by GC-MS and ultra fast GC of both essential oils in order to
verify their global composition and particularly their percentage in compounds of interest. On
the other hand, the fast analytical methods were validated to garantee accurate quantification
of semiochemicals in formulations.
Once characterised, essential oils were submitted to fractionation process in order to purify
semiochemicals. All the steps of this methodology were optimised and validated. Purified
semiochemicals were then encapsulated in order to be released over time. The formulations
were optimised and characterised. Their biological efficiency was tested by bioassays on
aphid predators and parasitoids. Slow-release study was also conducted in order to estimate
semiochemical diffusion kinetic over time and according to abiotic parameters.
57
Essential oils
E.O. characterisation
GC-MS
(identification)
E.O. fractionation
Ultra fast GC
(quantification)
Small scale
Column chromatography
Flash
chromatography
UFM performance
evaluation
Analytical validation
Solvent evaporation
Classical validation
Accuracy profile
Ultra fast GC analysis
(purity)
Purified semiochemicals
Formulation
Alginate gel beads
Bioassays
Slow-release study
Characterisation
Figure 1: Methodology of the research
Chapter IV
Characterisation of essential oils and
semiochemical purification
Chapter IV
___________________________________________________________________________
Objectives
The objective of this first experimental part was to respond to the first question of the thesis:
“How to purify semiochemicals from plant matrix with a high purity?”. To reach this goal,
essential oils of Matricaria chamomilla (originated from Nepal) and Nepeta cataria
(originated from France) were first characterised by GC-MS and by ultra fast GC in order to
know their percentage in compound of interest: E-β-farnesene, Z,E-nepetalactone and E-βcaryophyllene. The fast GC analytical method was validated in terms of linearity of the
calibration curve, LOD, LOQ, accuracy and trueness. The analytical performances of the
Ultra Fast Module column were also investigated.
Essential oils were then subjected to fractionation by liquid column chromatography on
silicagel to purify semiochemicals. E-β-farnesene, E-β-caryophyllene and Z,E-nepetalactone
were obtained at high purity after evaporation of the solvent of elution.
The methodology developed in this chapter proved to be efficient to obtain purified
semiochemicals. The different steps of the process are described in Figure 1’ with the main
experimental conditions.
61
Figure 1’: Essential oil characterisation and fractionation
Chapter IV
___________________________________________________________________________
Fast gas chromatography characterisation of purified semiochemicals from
essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta
cataria L. (Lamiaceae)
Stéphanie Heuskin1,2, Bruno Godin1, Pascal Leroy3, Quentin Capella3, Jean-Paul Wathelet2,
François J. Verheggen3, Eric Haubruge3, Georges Lognay1
1
Department of Analytical Chemistry, Gembloux Agro-Bio Tech, University of Liege, Passage des
Déportés 2, B-5030 Gembloux (Belgium).
2
Department of General and Organic Chemistry, Gembloux Agro-Bio Tech, University of Liege,
Passage des Déportés 2, B-5030 Gembloux (Belgium).
3
Department of Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of
Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
Reference: Journal of Chromatography A, 1216 (2009), 2768-2775.
63
Chapter IV
___________________________________________________________________________
Abstract
The chemical composition of Matricaria chamomilla L. and Nepeta cataria L. essential oils
was determined by GC-MS on an apolar stationary phase by comparison of the characteristic
fragmentation patterns with those of the Wiley 275L database. The GC-MS chromatograms
were compared with those obtained by fast GC equipped with a direct resistively heated
column (Ultra Fast Module 5% phenyl, 5m x 0.1 mm, 0.1µm film thickness). Analytical
conditions were optimised to reach a good peak resolution (split ratio = 1:100), with analysis
time lower than 5 min versus 35-45 min required by conventional GC-MS. The fast
chromatographic method was completely validated for the analysis of mono- and
sesquiterpene compounds. Essential oils were then fractionated by column chromatography
packed with silicagel. Three main fractions with high degree of purity in E-β-farnesene were
isolated from the oil of Matricaria chamomilla. One fraction enriched in (Z,E)-nepetalactone
and one enriched in β-caryophyllene were obtained from the oil of Nepeta cataria. These
semiochemical compounds could act as attractants of aphid’s predators and parasitoids.
Keywords
Aphids; Matricaria chamomilla; Nepeta cataria; fast GC; method validation; E-β-farnesene;
β-caryophyllene; nepetalactone; fractionation
64
Chapter IV
___________________________________________________________________________
1. Introduction
Since a few years, essential oils and their constituents with semiochemical properties are
more and more used for insect control in integrated pest management programs to encounter
or drastically reduce the pesticides treatments [1-4]. There are many advantages for isolating
semiochemicals from plant matrixes, like the essential oil fractionation technique, rather than
by a chemical synthesis: the compounds of interest are natural molecules, the fractionation
process is fast and simple to implement and the production costs are low. For supporting this
technique, it is necessary to work with state-of-the-art analytical instrument for the
determination and the quantification of products, like fast gas chromatography. Indeed, it is
particularly suitable when a large number of fractions have to be checked.
In the present study, the main goals consist in isolating aphid pheromones molecules from a
plant source and formulating them to attract aphid predators and/or parasitoids on the
infested fields. E-β-farnesene and (Z,E)-nepetalactone are respectively the alarm and the
sexual pheromones of many aphids species [5-7]. Moreover, β-caryophyllene is a molecule of
interest having biological activity against aphid reproduction [2] and was identified as the
aggregation pheromone of the Asian lady beetle Harmonia axyridis [8]. One of the main
interest of these compounds is that they could act as attractants and oviposition inductors of
some aphid predators (Episyrphus balteatus De Geer (Diptera: Syrphidae)) and parasitoids
(Aphidius ervi Haliday (Hymenoptera: Braconidae)) [9-13]. The essential oil of Matricaria
chamomilla L. (Asteraceae), popularly known as German chamomile (other synonyms:
Matricaria recutita L. and Chamomilla recutita L.), was reported to contain a high
proportion of E-β-farnesene (EBF). The percentage of this compound can vary in function of
the cultivar, the chemotype and the manufacturing process [14], and the part of the plant [15,
16]. (Z,E)-nepetalactone and β-caryophyllene are present as the major constituents in the
essential oil of Nepeta cataria L. (catnip oil) (Lamiaceae) [17, 18]. Another isomer of
nepetalactone, (E,Z)-form, is present in small proportions in the catnip essential oil and is
reported to be repellent to cockroaches [18].
The fractionation of these essential oils by liquid column chromatography, with pentane as
elution solvent, is a fast and simple separation method to isolate groups of components
(monoterpenes, sesquiterpenes, oxygenated compounds…). The solvent, with a low bowling
point, could be evaporated rapidly without significant loss of compounds of interest. The
65
Chapter IV
___________________________________________________________________________
isolation of E-β-farnesene from essential oil of Matricaria chamomilla by this technique was
reported by Bungert et al. [19]. The method proposed by these authors combined adsorption
chromatography and argentation HPLC and is quite laborious. The procedure we describe
here is faster and leads to adequate EBF purification for performing biological tests.
As for most volatile terpenoids of essential oils, EBF, nepetalactones and β-caryophyllene are
generally analysed by conventional GC and GC-MS, but GC analytical methods are still time
consuming, principally for the analysis of a great number of essential oil fractions. The
necessity for fast GC methods is growing for routine analyses with repeatable and
reproducible results. The efficiency of the fast GC technique with a direct resistively-heated
column (ultra fast module-GC) was demonstrated for the analyses of various types of
samples: essential oils, pesticides, lipids… [20-22].
The present research describes a completely validated fast GC method for the analysis and
the quantification in less than five minutes of different mono- and sesquiterpenes. The
method proposed herein could be easily transposed to other components of essential oils. The
fast method was validated in term of repeatability, reproducibility, linearity, accuracy,
selectivity and LOD/LOQ. The sample capacity and the column efficiency were also
evaluated respectively, with the evolution of the number of theoretical plates in function of
the amount of sample injected, and with the Van Deemter plots. The gain of analytical time is
about of a factor ten compared with conventional GC, with an optimal peak resolution. The
original GLC method described in the present paper allows very high throughput and is of
particular interest for the study of slow release formulations (ongoing investigations).
66
Chapter IV
___________________________________________________________________________
2. Experimental
2.1. Chemicals and materials
Essential oil of Matricaria chamomilla was purchased from Vossen & Co. (Brussels,
Belgium) and was originated from Nepal (Lot no.: CHA06MI0406). Essential oil of Nepeta
cataria was purchased from APT-Aromatiques (Saint-Saturnin les Apt, France) and was
originated from France (Lot no.: 18007).
E-β-farnesene from chemical synthesis was kindly supplied by Dr. S. Bartram and Prof. W.
Boland (Max Planck Institute for Chemical Ecology, Jena, Germany). β-caryophyllene, nbutyl-benzene, α-pinene and longifolene as reference compounds were purchased from
Sigma-Aldrich (Bornem, Belgium). The purity of the references was determined by fast GC.
Solution of each compound was prepared in n-hexane at an approximate concentration of
1µg/µl. Three replicates were analysed. List of reference compounds mean purity, with
standard deviation (SD) and coefficient of variation (% RSD) is given in Table 1.
Table 1: Purity of reference compounds
Compound
Mean purity (%)
SD
RSD (%)
E-β-farnesene
98.17
0.0009
0.10
β-caryophyllene
94.67
0.0071
0.75
Longifolene
98.01
0.0003
0.03
n-butyl-benzene
100.00
0.0000
0.00
Limonene
100.00
0.0000
0.00
α-pinene
100.00
0.0000
0.00
2.2. GC-MS analyses
Conventional GC-MS analyses were carried out on a Thermo Trace GC Ultra coupled with a
Thermo Trace MS Finnigan mass selective detector (Thermo Electron Corp., Interscience,
Louvain-la-Neuve, Belgium) and equipped with an Optima 5 MS (Macherey-Nagel)
capillary column (30m x 0.32mm I.D., 0.25µm film thickness). The oven temperature
program was initiated at 40°C, held for 5 min then raised first at 5°C/min to 230°C, raised in
a second ramp at 30°C/min to 280°C with a final hold at this temperature for 5 min. Carrier
gas: He, constant flow rate of 1.5 ml/min. Injection volume: 1 µl. Split ratio = 1:20. Injection
temperature: 240°C. Interface temperature: 280°C. MS detection was performed with
67
Chapter IV
___________________________________________________________________________
electron impact (EI) mode at 70 eV by operating in the full-scan acquisition mode in the 35
to 350 amu range. The identification of the volatile compounds was performed by comparing
the obtained mass spectra with those from the Wiley 275L spectral library.
Retention indices (I) were determined relative to the retention times of a series of n-alkane
standards (C9 to C30, Sigma, 0.025 µg/µl in n-hexane), measured under the chromatographic
conditions described above, and compared with literature values [23].
2.3. Fast GC analyses
Fast GC analyses were carried out on a Thermo Ultra Fast Trace GC gas chromatograph
operated with a split/splitless injector and a Thermo AS 3000 autosampler (Thermo Electron
Corp., Interscience, Louvain-la-Neuve, Belgium). The GC is equipped with an Ultra fast
module (UFM) incorporating a direct resistively heated column (Thermo Electron Corp.,
Interscience, Louvain-la-Neuve, Belgium): UFC-5, 5% phenyl, 5m x 0.1mm I.D., 0.1µm film
thickness. The following chromatographic conditions are those of the fast GC validation
method. Temperature program of UFM: initial temperature at 40°C, held for 0.1 min, ramp 1
at 30°C/min to 95°C, ramp 2 at 35°C/min to 155°C, ramp 3 at 200°C/min to 280°C, held for
0.5 min. Injection temperature: 240°C. Injection volume: 1µl. Carrier gas: He, at constant
flow rate of 0.5 ml/min. Split ratio = 1:100. Detection: the GC unit has a high frequency fast
FID detector (300 Hz), at 250°C. H2 flow: 35 ml/min; air flow: 350 ml/min; makeup gas flow
(N2): 30 ml/min. Data processing was by Chrom-card software (Version 2.3.3).
2.4. NMR spectra
All NMR spectra were recorded on Varian VNMR system (100, 400 and 600 MHz)
spectrometers operating at 14.1 Tesla or 9.4 Tesla for 20 µl of sample diluted in 700 µl of
CDCl3. The signal of solvent was used as internal reference of chemical displacement (1H:
7.26 ppm, 13C: 77.16 ppm).
2.5. Essential oil fractionation and purification
Liquid column chromatographic separation of essential oils was used to obtain fractions
enriched in compounds of interest. For that purpose, 1 ml (0.9306 g for M. chamomilla and
68
Chapter IV
___________________________________________________________________________
0.9525 g for N. cataria) of essential oil was fractionated over 11 g of silica gel G60 (70-230
mesh: ref. no.815330.1, from Macherey-Nagel) previously dried during 16 hours at 120°C
and packed in a glass column (15 mm I.D.) with glass wool plug at the bottom. Essential oil
of Matricaria chamomilla was eluted with 125 ml n-pentane to yield 5 fractions respectively
of 25, 10, 45, 25 and 20 ml. Essential oil of Nepeta cataria was first eluted with 125 ml npentane to yield 4 fractions respectively of 20, 40, 50 and 15 ml, followed by a second
elution step with 70 ml n-pentane:diethylether (80:20) leading to 2 fractions of 35 ml each.
Fifty µl of different fractions were diluted 30 times in n-hexane prior to GC-MS and fast GC
analyses.
Solvent-free purified compounds were obtained after evaporation of solvents from fractions
at atmospheric pressure and at 40°C with a Büchi rotatory evaporator without vacuum.
Solvent-free fractions were diluted in n-hexane and analysed at fast GC.
2.6. Method validation
The validation of the method for the quantification of volatile compounds (mono- and
sesquiterpenes) was done for 2 ranges of concentrations (range 1: 0.008 to 0.100 µg/µl, range
2: 0.080 to 1.000 µg/µl). Longifolene was used as internal standard for the sesquiterpene
components (E-β-farnesene and β-caryophyllene). n-Butyl-benzene was used as internal
standard for the monoterpenes (limonene and α-pinene). This molecule has close molecular
weight and is chromatographically well resolved in the prementioned conditions. Calibration
curves were obtained by plotting the ratio of analysed peak area / IS peak area, versus the
concentration ratio (analysed component / IS). For the first range, the concentration of
longifolene IS was at 0.0497 µg/µl and the concentration of n-butyl-benzene IS was at
0.0534 µg/µl. For the second range, the longifolene and the n-butyl-benzene concentrations
were at 0.4966 µg/µl and 0.5340 µg/µl respectively. For the calibration curves of each
component, 6 standard solutions from 0 to 0.100 µg/µl in n-hexane (range 1) and 6 standard
solutions from 0 to 1.000 µg/µl in n-hexane (range 2) were used as data points. Each of the
12 concentration levels were analysed in triplicate. The calibration curves were calculated
using the method of least squares fit analysis. The linearity was considered satisfactory when
correlation coefficient (r2) was higher than 0.996 [24].
69
Chapter IV
___________________________________________________________________________
The accuracy of the method was expressed as the bias (%) between the assigned value and
the measured value. The accuracy was judged satisfactory when comprised between 90% and
110% [24].
The precision was evaluated by the determination of the repeatability and the reproducibility.
To measure the repeatability, ten replicates of a sample were analysed at 0.05 µg/µl and 0.5
µg/µl, for the first and the second range of concentrations respectively, on the same day by
one analyst (n=10). To define the reproducibility, ten replicates at 0.05 µg/µl and ten at 0.5
µg/µl, were analysed 5 times on 5 days (n=50). Maximum allowed values (%) for
repeatability and reproducibility were depending on the concentration (AOAC norm, 2006)
(Range 1: repeatability of 8%, reproducibility of 16% ; Range 2: repeatability of 6%,
reproducibility of 12%).
The limit of detection (LOD) is the lowest quantity of a substance that can be distinguished
from the blank within a stated confidence limit (LOD=3*SDblank) for 8 replicates of blank.
The limit of quantification (LOQ) was arbitrarily set at 2*LOD.
The selectivity of the method was defined with the selectivity factor (α) between the two
nearest peaks (longifolene and β-caryophyllene):
α = (RT’β-caryophyllene / RT’longifolene),
where RT’ are the reduced retention times.
2.7. Calculation of direct resistively heated column (UFM) efficiency
The analytical performances of the UFM column were determined with the calculation of
theoretical plates in function of the quantity of component injected in the chromatographic
column. For each compound, 12 quantities were injected in triplicate from 0.08 ng to 50 ng
range. The mean value of the number of theoretical plates for the 3 replicates was plotted in
function of the quantity injected on the column. The height of theoretical plates was
calculated for each component in function of the carrier gas velocity. 0.5 ng of all
components were injected on the column for 11 velocities ranging from 6.70 cm/s to 66.49
cm/s. Three replicates were realised. The mean value of the height of theoretical plates was
plotted in function of the carrier gas velocity.
70
Chapter IV
___________________________________________________________________________
3. Results and discussion
The first part of this study concerns the validation of the fast GC analytical method for the
quantification of monoterpene and sesquiterpene hydrocarbons with the calculation and study
of theoretical plates, while the latter part deals with the purification of semiochemical
compounds from essential oils of Matricaria chamomilla and Nepeta cataria.
3.1. Fast GC analytical method validation
Figure 1 shows the chromatogram obtained with the fast GC analytical method for the
reference compounds (α-pinene, limonene, E-β-farnesene and β-caryophyllene) and the
internal standards, n-butyl-benzene and longifolene for monoterpenes and sesquiterpenes
respectively. The concentrations of these compounds were at 0.500 µg/µl in n-hexane.
Another solution of reference compounds was analysed at 0.050 µg/µl in n-hexane. In each
case, good separation of analytes was achieved with acceptable peak resolution and
symmetry within a total runtime of 5 minutes, with a split ratio of 1:100.
Figure 1: Chromatogram of reference compounds and internal standards analysed with
optimised UFGC method. For analysis conditions, see text.
The linearity of the method is summarised in Table 2 which shows calibration data and
detection limits for reference compounds in the two working ranges (6 points from 0.008 to
0.100 µg/µl and 6 points from 0.080 to 1.000 µg/µl; n=3), as well as the accuracy of the
calibration curves. The linearity of the calibration curves was validated with the r2
coefficients largely upper than 0.996 and with the Grubbs’s test where reduced residual are
lower than 2.754 in absolute value [25]. The results show that within the indicated
concentration ranges, there was a good correlation between peak area and concentration of
compounds. The accuracy of calibration curves was dependent of compounds and ranges of
concentration. The observed values, close to 100%, are comprised in the theoretical
71
Chapter IV
___________________________________________________________________________
acceptable limits (90 – 110 %), and give a very strong accuracy for each compound in the
two ranges of concentrations. The LOD and LOQ values are expressed in pg. They were
calculated accounting for the dispersibility of 8 blank replicates. As shown in Table 2, the
values are dependent of the ranges of concentration and the compounds. The lowest values of
LOD and LOQ are for β-caryophyllene with 0.74 pg and 1.48 pg respectively.
Table 3 shows the precision of the method with repeatability and reproducibility for each
compound at 2 concentrations (mean concentration of each range). The relative standard
deviations (RSD%) for repeatability were lower than the values of the AOAC norm which
requires 8% and 6 %, for the first and the second range of concentrations respectively. For
the reproducibility, the RSD % were lower than 16% and 12% for the ranges 1 and 2. These
values of RSD are very good and show strong repeatability and reproducibility of the method
for each reference compounds, the worst being α-pinene with RSD % repeatability at 2.07 %
and 1.78 %, for ranges 1 and 2 respectively, and with RSD % reproducibility at 3.48 %
and7.51 % for ranges 1 and 2. Considering these results, the precision of the fast GC method
was widely satisfactory at both high and low concentrations.
72
Table 2: Linearity data for the fast GC validation method
β-caryophyllene
E-β-farnesene
α-pinene
Limonene
0.008 – 0.100
0.080 – 1.000
0.008 – 0.100
0.080 – 1.000
0.008 – 0.100
0.080 – 1.000
0.008 – 0.100
0.080 – 1.000
y = 0,9592x - 0,0028
y=0,9558 x + 0,0053
y=0,8381 x + 0,0030
y=0,8408 x + 0,0056
y=0,9767 x + 0,0024
y=0,9690 x + 0,0019
y=0,8004 x + 0,0012
y=0,8097 x - 0,0013
r2
0,9998
0,9999
0,9998
0,9999
0,9998
0,9999
0,9988
0,9993
Reduced residual (Grubb’s test)
2.668
1.866
2.147
1.880
2.394
1.826
2.134
2.276
Accuracy of calibration curves
99.19
99.86
99.77
99.90
100.85
100.67
103.65
102.36
Longifolene
Longifolene
Longifolene
Longifolene
n-butyl-benzene
n-butyl-benzene
n-butyl-benzene
n-butyl-benzene
LOD (pg)
2.38
2.40
1.79
0.74
2.43
1.37
2.11
2.05
LOQ (pg)
4.76
4.80
3.58
1.48
4.86
2.74
4.22
4.10
Range (µg/µl)
Equation of the calibration curve
(%)*
Internal standard
Table 3: Precision of the fast GC method expressed as repeatability and reproducibility
β-caryophyllene
E-β-farnesene
α-pinene
Limonene
Concentration (µg/µl)
0.050
0.500
0.050
0.500
0.050
0.500
0.050
0.500
Repeatability (RSD %)
1.16
0.70
0.43
0.12
0.70
0.42
2.07
1.78
Reproducibility
3.00
2.82
0.89
0.81
1.98
1.89
3.48
7.51
%)
(RSD
Chapter IV
___________________________________________________________________________
The selectivity of the method is expressed in Table 4 as the selectivity factor () between the
two nearest peaks, longifolene and -caryophyllene, presented with S.D. and R.S.D. (%). The
selectivity was good with  at 1.016 for both ranges of concentrations.
Table 4: Selectivity of the fast GC method.
Concentration range: 0.008 – 0.100
Concentration range: 0.080 – 1.000
µg/µl
µg/µl
1.016
1.016
S.D.
0.001
0.001
R.S.D. (%)
0.07
0.07
Selectivity
factor
( )
between
longifolene
and
-
caryophyllene
3.2. Analytical performances of UFM column
Figure 2 shows the evolution of the number of theoretical plates in function of the quantity
of compounds injected on the UFM column. The number of plates is higher for
sesquiterpenes than for monoterpenes, the retention times of sesquiterpenes being longer. As
it can be seen, the maximum theoretical plate number is constant until a threshold amount of
sample. It was determined that for the monoterpenes, quantities of up to 10 ng can be
accomodated without affecting chromatographic resolution, while for the sesquiterpenes the
same is true for values below 1 ng. The injection of higher quantities led to fronting peak
distorsions.
The efficiency of analyses in function of velocity is shown in Figure 3 where Van Deemter
plots (derived from experimental data) were drawn, with the height of theoretical plates in
function of carrier gas velocity (cm/s) for monoterpene and sesquiterpene reference
compounds. The optimal height of a theoretical plate corresponded at the minimum value of
H in the curve: Hmin. For the two groups of components (monoterpenes and sesquiterpenes),
the Hmin value is obtained at a velocity of 35.86 cm/s. Hmin is at 0.036 mm, 0.017 mm and
0.014 mm, for α-pinene, limonene and n-butyl-benzene repectively. The validation of the
method was realised at a velocity of 43.94 cm/s, near optimal velocity, which allows faster
analyses without affecting the efficiency of the separation.
74
Chapter IV
___________________________________________________________________________
Figure 2: Number of theoretical plates (N) in function of the quantity of compounds injected on the Ultra Fast
Module column. (A: E-β-farnesene, B: β-caryophyllene, C: longifolene, D: n-butyl-benzene, E: limonene, F: αpinene).
Figure 3: Van Deemter plots for Ultra Fast GC (A: α-pinene, B: limonene, C: n-butyl-benzene, D: longifolene,
E: β-caryophyllene, F: E-β-farnesene).
75
Chapter IV
___________________________________________________________________________
3.3. Analysis of essential oils
Essential oils of Matricaria chamomilla and Nepeta cataria were analysed by GC-MS
(5µg/µl in n-hexane) to determine their compositions in compounds of interest: E-βfarnesene, β-caryophyllene and (Z,E)-nepetalactone.
The essential oil composition of Matricaria chamomilla was dominated by monoterpene and
sesquiterpene hydrocarbons: 16 hydrocarbon sesquiterpenes (59.00%), 9 oxygenated
sesquiterpenes (31.63%), 7 hydrocarbon monoterpenes (1.25%) and 3 oxygenated
monoterpenes (0.41%).Through comparison of the retention index with those of the literature
[23] and EI-mass spectra of each peak with the library, it was possible to identify 41
individual components, representing 99.57 % of the total amount (Table 5). The major peaks
observed in the chromatogram were attributed to sesquiterpenes with E-β-farnesene
(42.59%), (E,E)-α-farnesene (8.32%), germacrene D (2.93%), bicyclogermacrene (1.99%),
chamazulene (1.18%) and oxygenated sesquiterpenes with α-bisabololoxide A (21.2%), αbisabolone oxide A (4.53%) and α-bisabolol oxide B (4.43%). Two sesquiterpenes
representing 0.33 % and 0.10 % respectively, remained unknown.
Table 6 shows the composition of the Nepeta cataria essential oil. Twenty components were
identified (EI-mass spectra and retention index comparison). Among them, 6 were
oxygenated sesquiterpenes (77.17%), 2 hydrocarbon sesquiterpenes (10.53%), 9 hydrocarbon
monoterpenes (2.84%) and 2 oxygenated monoterpenes (0.09%). The main identified
constituents of this essential oil are 4aα,7α,7aα-nepetalactone ((Z,E)-nepetalactone or
4aS,7S,7aR-nepetalactone) (73.27%), β-caryophyllene (9.72%), β-caryophyllene oxide
(1.81%), cis-β-ocimene (1.64%) and 4aα,7β,7aα-nepetalactone (1.10%). Three isomers of
nepetalactone were present in the essential oil of Nepeta cataria, but could not be identified
with absolute configuration certainty by GC-MS.
The stereoisomery of (Z,E)-nepetalactone was confirmed by 1H and
(data not shown) and by comparison of the chemical
13
13
C NMR spectrometry
C displacements with those of the
literature [26]. Ten minor compounds of the essential oil could not be identified with the
spectral library and retention index.
76
Chapter IV
___________________________________________________________________________
Table 5: Constituents of the essential oil of Matricaria chamomilla identified by GC-MS.
No.
Components
Retention time (min)
Retention index (measured)
%
1
α-pinene
9.92
922
0.03
2
sabinene
11.38
965
0.04
3
6-methyl-5-hepten-2-one
11.83
977
0.03
4
2-pentyl-furan
12.08
988
0.05
5
p-cymene
13.14
1018
0.11
6
limonene
13.34
1024
0.10
7
trans-β-ocimene
13.74
1036
0.11
8
cis-β-ocimene
14.08
1046
0.69
9
γ-terpinene
14.36
1054
0.17
10
artemesia ketone
14.44
1057
0.32
11
artemesia alcohol
15.20
1079
0.06
12
isoborneol
17.68
1153
0.03
13
4,8-dimethyl-nona-3,8-dien-2-one
20.87
1262
0.04
14
α-copaene
23.78
1365
0.04
15
β-maaliene
23.88
1369
0.07
16
α-isocomene
24.06
1376
0.26
17
β-elemene
24.19
1382
0.07
18
sativene
24.55
1397
0.04
19
α-gurjunene
24.69
1403
0.04
20
β-caryophyllene
24.89
1411
0.17
21
aromadendrene
25.40
1433
0.07
22
E-β-farnesene
25.95
1456
42.59
23
Not identified sesquiterpene (MW: 204)
26.15
1465
0.10
24
Germacrène D
26.47
1478
2.93
25
β-selinene
26.59
1483
0.22
26
(Z,E)-α-farnesene
26.77
1491
0.83
27
bicyclogermacrene
26.85
1494
1.99
28
(E,E)-α-farnesene
27.13
1506
8.32
29
δ-cadinene
27.48
1521
0.18
30
sesquirosefuran
28.15
1549
0.18
31
Not identified sesquiterpene (MW: 204)
28.33
1157
0.33
32
trans-nerolidol
28.39
1559
0.17
33
dehydronerolidol
28.46
1562
0.09
34
dendrolasin
28.63
1569
0.21
35
spathulenol
28.71
1573
0.63
36
globulol
28.85
1578
0.23
37
α-bisabololoxide B
30.52
1649
4.43
38
α-bisabolone oxide A
31.10
1673
4.53
39
chamazulene
32.08
1715
1.18
40
α-bisabololoxide A
32.55
1735
21.16
41
Cis-ene-yne-dicycloether
35.15
1802
5.94
42
Trans-ene-yne-dicycloether
35.30
1807
0.99
43
(E)-phytol
39.79
2107
0.23
77
Chapter IV
___________________________________________________________________________
Table 6: Constituents of the essential oil of Nepeta cataria identified by GC-MS.
No.
Components
Retention time (min)
Retention index (measured)
%
1
α-thujene
9.72
916
0.01
2
α-pinene
9.92
922
0.05
3
sabinene
11.37
965
0.10
4
β-pinene
11.44
967
0.24
5
β-myrcene
12.12
987
0.02
6
limonene
13.34
1024
0.28
7
trans-β-ocimene
13.74
1036
0.47
8
cis-β-ocimene
14.08
1046
1.64
9
linalool
15.71
1095
0.05
10
α-terpineol
18.47
1177
0.04
11
Not identified monoterpene (MW: 136)
19.09
1196
0.54
12
Not identified monoterpene (MW: 136)
19.78
1216
0.03
13
Not identified monoterpene (MW: 136)
19.88
1219
0.12
14
Not identified monoterpene (MW: 136)
20.74
1245
0.10
15
(4aα,7α,7aα)-nepetalactone
23.51
1353
73.27
16
(4aα,7α,7aβ)-nepetalactone
23.94
1371
0.44
17
(4aα,7β,7aα)-nepetalactone
24.07
1377
1.10
18
dihydronepetalactone
24.56
1397
0.46
19
β-caryophyllene
24.89
1411
9.72
20
α-humulene
25.77
1449
0.81
21
Not identified
27.82
22
Benzoate (Z)-3-hexen-1-ol
28.48
1551
0.07
23
β-caryophyllene oxide
28.87
1579
1.81
24
humulene epoxide II
29.45
1604
0.09
25
Not identified
32.46
0.06
26
Not identified
33.69
0.05
27
Not identified
33.76
0.03
28
Not identified
33.91
0.38
29
Not identified
34.00
0.13
30
6,10,14-trimethyl-2-pentadecanone
34.59
7.82
1840
0.07
The essential oils were then analysed by fast GC on an UFM column of the same polarity as
in GC-MS (apolar stationary phase). The retention times of the components of interest from
the essential oils were compared to those of the reference compounds. Figure 4 and Figure 5
report the patterns, respectively of the Matricaria chamomilla and the Nepeta cataria
essential oils, under study analysed by GC-MS and fast GC, together with a list of their
characteristic components. With conventional GC-MS, analysis time was about 40 min,
while for fast GC it was less than 5 min, with the same chromatographic profile and similar
resolution.
78
Chapter IV
___________________________________________________________________________
Figure 4: Profiles of a Matricaria chamomilla essential oil analysed by GC-MS (a) and Ultra Fast GC (b). For
analysis conditions see text. List of the main components: (1) E-β-farnesene; (2) germacrene D; (3)
bicyclogermacrene; (4) (E,E)-α-farnesene; (5) α-bisabolol oxide B; (6) α-bisabolone oxide A; (7) chamazulene;
(8) α-bisabolol oxide A; (9) cis-ene-yne-dicycloether.
Figure 5: Profiles of a Nepeta cataria essential oil analysed by GC-MS (a) and Ultra Fast GC (b). For analysis
conditions see text. List of the main components: (1) (Z,E)-nepetalactone; (2) (E,Z)-nepetalactone; (3) βcaryophyllene; (4) unknown compound; (5) β-caryophyllene oxide.
79
Chapter IV
___________________________________________________________________________
3.4. Essential oil fractionation
Tables 7 and 8 present the mean composition (%) ± SD (n=3) of Matricaria chamomilla and
Nepeta cataria fractions recovered of elution on adsorption chromatography and solvent
evaporation. The fractions were analysed on fast GC and the identification of peaks was
realised by comparison of the different retention times. As it can be seen, each fraction
differs from the original oil in its chromatographic profile and in the percentage of the
constituents.
Five fractions were collected for Matricaria chamomilla based on the evolution of E-βfarnesene percentage in function of n-pentane solvent elution volume. EBF is present in each
fraction and was always associated with (E,E)-α-farnesene. Germacrene D was also
associated in small percentage to the two previous cited compounds in the three first
fractions. F3 has the highest relative percentage of EBF (79.61%) and, for this reason, is the
most interesting in this study. The percentage of (E,E)-α-farnesene grows from F1 (3.98 %)
to F5 (25.22 %), with a mean value of 14.34 % in F3. Chamazulene was detected in fraction
F4 (0.69 %), but clearly appeared in F5 (4.52 %) with a blue coloration of the fraction. This
compound also served as coloured indicator during the fractionation process to stop the
elution; the percentage of EBF decreasing with the apparition of chamazulene.
The fractionation of Nepeta cataria essential oil was reached in two elution steps. The first
elution with n-pentane allowed the collection (4 fractions from F1 to F4) of non-polar
compounds such as β-caryophyllene in very high proportion as regards the percentage of this
compound in the pure essential oil (9.72 %). Therefore, β-caryophyllene is mainly present in
fraction F2 at 79.54 % purity. The same fractionation process on another catnip oil from
USA led to totally pure β-caryophyllene (> 99.9 % by fast GC) (data not shown). The second
fractionation step (2 fractions, F5 and F6) was realised with a more polar elution solvent mix
(n-pentane 80% - diethylether 20%) to collect nepetalactone isomers. Fraction F5 was
enriched to 96.97 % in (Z,E)-nepetalactone, the aphid sexual pheromone, with caryophyllene
oxide (2.19 %) as associated compound.
Compared with the pure essential oils, fractions obtained by column chromatography are
more concentrated in compounds of interest. This purification technique led to the isolation
and the collection of various polarity compounds (monoterpenes, sesquiterpenes, oxygenated
80
Chapter IV
___________________________________________________________________________
compounds…) in a very fast way. Flash Chromatography is currently developed to obtain
purified compounds at higher scale and shows identical results.
It is noteworthy that such fractionation and fast GC analyses are powerful and simple
methods to produce and analyse essential oils constituents.
Acknowledgments
The authors are grateful to Prof. M. Luhmer (Département de chimie, Laboratoire de RMN
haute resolution, ULB, Belgium) for NMR analyses and to Dr. S. Bartram and Prof. W.
Boland from the Max Planck Institute for Chemical Ecology (Germany) for providing EBF
from chemical synthesis. This research was funded by the Walloon Region Ministry grant
(WALEO2: SOLAPHID-RW/FUSAGX 061/6287) and by the FRFC grant (n° 2.4586.04F).
81
Table 7: Major components of Matricaria chamomilla fractions (mean ± SD of triplicate)
Matricaria chamomilla
Sum of monoterpenes
E-β-farnesene
(E,E)-α-farnesene
germacrene D
chamazulene
F1
composition
4.38%
51.79%
3.98%
7.52%
0%
SD
1.32%
0.68%
0.40%
0.18%
0%
F2
composition
2.73%
74.90%
9.20%
5.79%
0%
SD
0.58%
2.28%
1.91%
0.41%
0%
F3
composition
0%
79.61%
14.34%
1.25%
0%
SD
0%
0.06%
0.81%
0.51%
0%
F4
composition
0%
76.78%
20.86%
0%
0.69%
SD
0%
1.50%
0.87%
0%
0.66%
F5
composition
0%
68.99%
25.22%
0%
4.52%
SD
0%
2.16%
1.63%
0%
2.44%
Table 8: Major components of Nepeta cataria fractions (mean ± SD of triplicate)
(Z,E)-nepetalactone
(E,Z)-nepetalactone
β-caryophyllene
caryophyllene oxide
F1
Composition
0.00%
0.00%
0.00%
0.00%
SD
0.00%
0.00%
0.00%
0.00%
F2
Composition
0.00%
0.00%
79.54%
0.00%
SD
0.00%
0.00%
4.68%
0.00%
Nepeta cataria
F3
Composition
SD
0.00%
0.00%
0.00%
0.00%
26.88%
25.86%
0.00%
0.00%
F4
Composition
0.00%
0.00%
0.00%
0.00%
SD
0.00%
0.00%
0.00%
0.00%
F5
Composition
96.97%
0.19%
0.00%
2.19%
SD
1.70%
0.32%
0.00%
0.39%
F6
Composition
84.63%
14.96%
0.00%
0.10%
SD
12.59%
13.15%
0.00%
0.18%
Chapter IV
___________________________________________________________________________
References
[1] A.L. Tapondjou, C. Adler, D.A. Fontem, H. Bouda, C. Reichmuth, J. Stored Prod. Res. 41
(2005).
[2] B.S. Tomova, J.S. Waterhouse, J. Dobersky, Entomol. Exp. Appl. 115 (2005).
[3] J. Wang, F. Zhu, X.M. Zhou, C.Y. Niu, C.L. Lei, J. Stored Prod. Res. 42 (2006).
[4] S. Rajandran, V. Sriranjini, J. Stored Prod. Res. 44 (2008).
[5] G.W. Dawson, D.C. Griffiths, N.F. Janes, A. Mudd, J.A. Pickett, L.J. Wadhams, C.M.
Woodcock, J. Chem. Ecol. 16 (1990).
[6] M.A. Birkett, J.A. Pickett, Phytochem. 62 (2003).
[7] F. Francis, S. Vandermoten, F. Verheggen, G. Lognay, E. Haubruge, Entomol. Exp. Appl.
129 (2005).
[8] F.J. Verheggen, Q. Fagel, S. Heuskin, G. Lognay, F. Francis, E. Haubruge, J. Chem. Ecol.
33 (2007).
[9] Y. Du, G.M. Poppy, W. Powell, J.A. Pickett, J. Wadhams, C.M. Woodcock, J. Chem.
Ecol. 24 (1998) 8.
[10] R.T. Glinwood, Y.-J. Du, W. Powell, Entomol. Exp. Appl. 92 (1999).
[11] R. Almohamad, F. Verheggen, F. Francis, E. Haubruge, Comm. Appl. Biol. Sci. Ghent
University 71 (2006) 2b.
[12] N. Harmel, R. Almohamad, M.-L. Fauconnier, P. Du Jardin, F. Verheggen, M. Marlier,
E. Haubruge, F. Francis, Insect Sci. 14 (2007).
[13] F.J. Verheggen, L. Arnaud, S. Bartram, M. Gohy, E. Haubruge, J. Chem. Ecol. 34 (2008)
3.
[14] H. Schulz, M. Baranska, H.-H. Belz, P. Rösch, M.A. Strehle, J. Popp, Vibr. Spectr. 35
(2004).
[15] M. Das, G. Ram, A. Singh, G.R. Mallavarapu, S. Ramesh, M. Ram, S. Kumar, Flav.
Fragr. J. 17 (2002).
[16] E. Szöke, E. Maday, G. Marczal, E. Lemberkovics, Acta Hort. 597 (2003).
83
Chapter IV
___________________________________________________________________________
[17] C. Bourrel, F. Perineau, J. Essent. Oil Res. 5 (1993).
[18] G. Schultz, E. Simbro, J. Belden, J. Zhu, J. Coats, Environ. Entomol. 33 (2004) 6.
[19] M. Bungert, R. Thiel, P. Goedings, H. Becker, Z. Naturforsch. 57c (2002).
[20] C. Bicchi, C. Brunelli, C. Cordero, P. Rubiolo, M. Galli, A. Sironi, J. Chrom. A 1024
(2004).
[21] L. Mondello, A. Casilli, P.Q. Tranchida, R. Costa, B. Chiofalo, P. Dugo, G. Dugo, J.
Chrom. A 1035 (2004).
[22] C. Bicchi, C. Brunelli, C. Cordero, P. Rubiolo, M. Galli, A. Sironi, J. Chrom. A 1071
(2005).
[23] R.P.Adams, in C. Stream (Editor), Identification of essential oil components by gas
chromatography/quadrupole mass spectroscopy. Allured Publishing Corporation,
Illinois, 2001, 456 p.
[24] L. Roland. Etapes de validation d’une méthode de dosage de résidus par CPG ou HPLC.
Application pratique de la procédure VALIDANA.P01. BEAGx, Gembloux, 2002.
[25] P. Dagnelie, in De Boeck & Larcier (Editor), Statistique théorique et appliqué: 2.
Inférence statistique à une et à deux dimensions. Broché, Bruxelles, 2006.
[26] M. Wang, K.-W. Cheng, Q. Wu, J.E. Simon, Phytochem. Anal. 18 (2007).
84
Chapter V
Quantification of semiochemicals in
formulations
Chapter V
__________________________________________________________________________
Objectives
After the development of a method to purify semiochemicals from essential oils in the first
part of the work, this second experimental chapter answers the question: “How to analyse
and quantify semiochemicals on a fast and accurate manner?”. The quantification of
semiochemicals in formulations was validated according to a methodology never used in
IPM field of research: the accuracy profile concept. This principle of validation differed from
classical validation methods in that it considered the total error concept, being a combination
of the systematic and the random errors. The fast GC analytical procedure was accurate along
the range validated. The different steps of this validation method are presented in Figure 1’.
This chapter also presents the purification of semiochemicals from essential oils at larger
scale, by flash chromatography. This technique was optimised and allowed fractionation of
10 fold higher quantities than presented in chapter IV of the PhD thesis.
Various formulations were also evaluated in terms of protection efficiency of sesquiterpene
semiochemicals towards oxidation. Alginate bead formulations proved to be more efficient
than sunflower oil to protect compounds from degradation. The various steps of the
procedure are summarized in Figure 2’ with some experimental conditions.
87
UFGC accuracy profile validation
Calibration standards
Validation standards
Experimental conditions
Experimental conditions
3 concentration levels
3 replicates per level
3 series
5 concentration levels
3 replicates per level
3 series
3 calibration curves
Back-calculation of
predicted concentrations
Resolution and selectivity
+
Validation criteria
Linearity of the method
Trueness: bias = systematic error
Precision: repeatability – intermediate precision
= random error
Accuracy: Trueness + Precision
= Total error
Accuracy profile
LOQ
Range
Figure 1’: Accuracy profile validation procedure
+
Uncertainty measurements
Figure 2’: Steps of the procedure to measure the protection efficiency of formulations towards sesquiterpenes
Chapter V
__________________________________________________________________________
Validation of a fast gas chromatographic method for the study of
semiochemical slow-release formulations
Stéphanie Heuskin1,2,*, Eric Rozet3,*, Stéphanie Lorge4, Julien Farmakidis1, Philippe Hubert3,
François J.Verheggen5, Eric Haubruge5, Jean-Paul Wathelet2, Georges Lognay1
1
Department of Analytical Chemistry, Gembloux Agro-Bio Tech, University of Liege, Passage des
Déportés 2, B-5030 Gembloux (Belgium).
2
Department of General and Organic Chemistry, Gembloux Agro-Bio Tech, University of Liege,
Passage des Déportés 2, B-5030 Gembloux (Belgium).
3
Laboratory of Analytical Chemistry, Bioanalytical Chemistry Research Unit, CIRM, University of
Liege, B36, B-4000 Liege (Belgium).
4
Institute of Condensed Matter and Nanosciences, Catholic University of Louvain, Croix du Sud 2,
B-1348 Louvain-la-Neuve (Belgium).
5
Department of Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of
Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
* These authors contributed equally to this work
Reference: Journal of Pharmaceutical and Biomedical Analysis 53 (2010), 962-972.
91
Chapter V
__________________________________________________________________________
Abstract
The validation of a fast GC-FID analytical method for the quantitative determination of
semiochemical sesquiterpenes (E-β-farnesene and β-caryophyllene) to be used in an
integrated pest management approach is described. Accuracy profiles using total error as
decision criteria for validation were used to verify the overall accuracy of the method results
within a well defined range of concentrations and to determine the lowest limit of
quantification for each analyte. Furthermore it allowed to select a very simple and reliable
regression model for calibration curve for the quantification of both analytes as well as to
provide measurement uncertainty without any additional experiments.
Finally, this validated method was used for the quantification of semiochemicals in slow
release formulations. The goal was to verify the protection efficiency of alginate gel beads
formulations against oxidation and degradation of sesquiterpenes. The results showed that the
alginate beads are adequate slow release devices which protect the bio-active molecules
during at least twenty days.
Keywords
Validation; Accuracy profile; Sesquiterpenes; Alginate beads; Semiochemicals
92
Chapter V
__________________________________________________________________________
1. Introduction
Semiochemicals, which can be defined as chemical communication signals between living
organisms, are more and more used in integrated pest management programs, acting as insect
control or monitoring devices [1]. This increasing interest is linked to the need for reducing
the pesticides treatments on the infested fields. However, the compounds used in such
systems are generally obtained by chemical synthesis [2, 3] instead of being extracted from
natural sources, like plant matrixes.
Indeed, essential oils of many plant species contain a lot of molecules which are also reported
in insect communication. E--Farnesene, the alarm pheromone of many aphid species [4],
can be isolated, with a high purity degree, from Matricaria chamomilla L. (Asteraceae)
essential oil [5] by means of a fast and simple process [6]. On a biological point of view, this
sesquiterpene is also considered as attractant and oviposition inductor of predators
(Episyrphus balteatus De Geer (Diptera: Syrphidae)) [7-9] and aphid parasitoids (Aphidius
ervi Haliday (Hymenoptera: Braconidae)) [10]. -Caryophyllene, identified recently as a
potential component of the aggregation pheromone of the Asian ladybeetle Harmonia
axyridis Pallas [11], is present as a major compound of Nepeta cataria L. (Lamiaceae)
essential oil [6, 12]. This molecule can also have a biological activity against aphid
reproduction [13]. These two sesquiterpene compounds are therefore considered as
allelochemicals (kairomones: receptor species benefits), being produced by members of one
species and influencing the behaviour of individuals of another species.
An interesting way to promote the allelochemical properties of these molecules consists in
the development of natural and biodegradable semiochemical slow release formulations for
attracting and/or maintaining populations of predators and/or parasitoids on aphid infested
fields in a biological control approach. Alginate gel beads were largely described as efficient
releasers for aroma and flavour volatile compounds in the food industry [14] or for essential
oils acting as antimicrobial agents [15, 16]. The beads are rather simple to produce on a labscale, easy to manipulate and have low impact on the environment [17]. Furthermore,
alginate, a polysaccharide derived from marine brown algae (Phaeophyceae), is a hydrophilic
matrix with low oxygen permeability properties which can protect the volatile molecules
from oxidation [18]. Indeed, the sesquiterpenes present double bonds which are preferential
sites for oxidation reactions (hydroxylation, epoxidation, oxidative clivage of double bonds
93
Chapter V
__________________________________________________________________________
like ozonolyse) [19]. Some papers relate the oxidation of cyclic sesquiterpenes like βcaryophyllene [20-22], but on our knowledge, only one experiment has been conducted on
the oxidation of E--farnesene, a linear molecule [23].
The purpose of the present research consists in verifying this protection efficiency of alginate
gel beads towards incorporated semiochemicals. The procedure developed involves the
quantification of compounds in the formulations over time, when exposed to air and light, by
means of fast gas chromatography (< 5 min) coupled with a high frequency (300 Hz) flame
ionisation detector (fast GC-FID). The paper describes also the validation of a fast analytical
GC method for the sesquiterpenes analysis with accurate results. The necessity for fast GC
methods is growing for routine analyses. As a matter of fact, conventional GC methods are
still time consuming, principally for the analysis of a great number of essential oil fractions,
but also for validation steps. This validation was conducted by means of the accuracy profile
concept based on the guidelines of the Société Française des Sciences et Techniques
Pharmaceutiques (SFSTP) [24-26]. The present procedure was largely described for
pharmaceutical [27-32] and food [33, 34] analytical methods. The method described herein is
the first application of accuracy profile validation within the field of integrated pest
management combined with fast GC – incorporating a direct resistively heated column (Ultra
Fast Module) – analytical tool. Besides, most applications of this validation methodology
used constant acceptance limits all over the concentration range investigated. At least since
the work of Horwitz et al. [35], it is well known that the relative standard deviation of any
assay increases with decreasing concentrations, thus leading to higher random error. Here,
acknowledging that making more total error at small concentrations is acceptable, the
acceptance limits were therefore settled larger at the expected lower limit of quantification to
take into account this natural behaviour of analytical methods. Such larger acceptance limits
for decreasing concentration is common place for the random errors as well as for the
systematic errors in other fields of applications [36-38]. However, to our knowledge, it is the
first time two levels acceptance limits are used with the accuracy profile validation approach.
Nonetheless in order to interpret and compare adequately results obtained by laboratories
with their analytical methods, it is essential to estimate measurement uncertainty [39]. In this
respect, measurement uncertainties of this fully validated method were also computed,
without any additional experiments, thus increasing the reliability evaluation of the analytical
results obtained and thus the adequacy of the developed method.
94
Chapter V
__________________________________________________________________________
2. Experimental
2.1. Chemicals and reagents
Essential oil of Matricaria chamomilla was purchased from Vossen & Co. (Brussels,
Belgium) and was originated from Nepal (lot no. CHA06MI0406). Essential oil of Nepeta
cataria was purchased from Essential7.com (Roswell, NM, USA) and was originated from
Canada (lot no. EO0020f).
E-β-Farnesene from chemical synthesis was kindly supplied by Dr. S. Bartram and Prof. W.
Boland (Max Planck Institute for Chemical Ecology, Jena, Germany). β-Caryophyllene, used
as reference compound for the method validation, was extracted by flash chromatography
from Nepeta cataria L. essential oil. (+)-Longifolene as internal standard was purchased
from ABCR (Karlsruhe, Germany). The mean purities of the terpenes, shown in Table 1 with
standard deviations (SDs) and relative standard deviations (RSDs), were determined by fast
GC. A solution of each compound was prepared in n-hexane at a concentration of 1 µg µL-1.
Ten replicates were performed.
n-Hexane of GC grade was purchased from VWR (Leuven, Belgium). n-Pentane extra pure
was purchased from Acros Organics (Geel, Belgium).
Table 1: Purity of compounds analysed by fast GC.
Compound
E--Farnesene (from synthesis)
E--Farnesene (from F3 Matricaria chamomilla)
-Caryophyllene (from F2 Nepeta cataria)
(+)-Longifolene (from synthesis)
Retention time (min.)
3.52
3.52
3.41
3.39
Mean purity (%)
99.4
83.8
97.7
100.0
SD
0.2
0.3
0.5
0.0
RSD (%)
0.2
0.4
0.5
0.0
2.2. Fast GC analyses
Fast GC analyses were conducted on a Thermo Ultra Fast Trace GC gas chromatograph
operated with a split/splitless injector and a Thermo AS 3000 autosampler (Thermo Electron
Corp., Interscience, Louvain-la-Neuve, Belgium). The GC system was equipped with an
Ultra fast module (UFM) incorporating a direct resistively heated column (Thermo Electron
Corp.): UFC-5, 5% phenyl, 5 m x 0.1 mm I.D., 0.1 µm film thickness. The following
chromatographic conditions were determined for good resolution of terpenes (mono- and
sesquiterpenes) analyses in a previous paper [6]. Temperature programme for UFM was the
95
Chapter V
__________________________________________________________________________
following: initial temperature at 40 °C, held for 0.1 min, ramp 1 at 30 °C min-1 to 95 °C,
ramp 2 at 35 °C min-1 to 155 °C, ramp 3 at 200 °C min-1 to 280 °C, final hold of 0.5 min at
280°C. Injection temperature: 240 °C. Injection volume: 1 µL. Carrier gas: He, at constant
flow rate of 0.5 mL min-1. Split ratio = 1:100. The GC unit had a high-frequency fast flame
ionization detector (300 Hz FID), at 250 °C. H2 flow: 35 mL min-1; air flow: 350 mL min-1;
makeup gas flow (N2): 30 mL min-1. Data processing was realised by Chromcard software
(Version 2.3.3).
Figure 1a and Figure 1b show the chromatograms obtained with this fast GC analytical
method for the reference compounds (E-β-farnesene (81.6 ng µL-1) and β-caryophyllene
(80.5 ng µL-1) with the internal standard longifolene (102.6 ng µL-1)) and for a blank sample
(matrix without sesquiterpenes), respectively.
Figure 1: Chromatograms of analytes (E-β-farnesene at 81.6 ng µL-1 and β-caryophyllene at 80.5 ng µL-1) and
internal standard (longifolene at 102.6 ng µL-1) (a) and of a blank alginate beads matrix sample (b) analysed
with optimised fast GC method. For analysis conditions, see text.
96
Chapter V
__________________________________________________________________________
2.3. Flash chromatography
In a previous paper [6], the fractionation process was realised by a classic liquid column
chromatographic separation of essential oils. The present research relates the use of flash
chromatography (Flash-chromatography assembly with threaded joints, Sigma-Aldrich,
Bornem, Belgium) to obtain purified extracts of relatively large scale quantities at higher
speed (in less than 10 minutes). Ten mL of essential oil (9.306 g for Matricaria chamomilla
and 9.525 g for Nepeta cataria) were fractionated under pressure (N2 at 0.5 bar) over 110 g
of silica gel G60 (70-230 mesh: ref. no. 815330.1, from Macherey-Nagel) previously dried
16 hours at 120 °C and packed in a glass column (35 mm I.D.) with glass wool plug at the
bottom. The silicagel bed was 34 cm high. Essential oil of Matricaria chamomilla was eluted
with 1200 mL n-pentane to yield five fractions of 250 (F1), 200 (F2), 400 (F3), 200 (F4) and
150 (F5) mL, respectively. Essential oil of Nepeta cataria was eluted with 1050 mL npentane leading to three fractions of 250 (F1), 350 (F2) and 450 (F3) mL, respectively. Fifty
µL of each fraction were diluted 30 times in n-hexane prior to fast GC analyses. Solvent-free
purified compounds were obtained after solvent evaporation from fractions at atmospheric
pressure and at 40 °C with a Büchi rotatory evaporator (rotation: 1.6 tour s-1). The recovery
of this evaporation mode was measured in five replicates (96.3% ± 0.94%) and judged
satisfactory according to the AOAC norm (2006) which requires recoveries comprised
between 90% and 108%. Solvent-free fractions were diluted in n-hexane and analysed by fast
GC. The large amounts (approximately 5 ml and 8 ml for E-β-farnesene and β-caryophyllene,
respectively) of purified semiochemicals obtained by that way were stored at 4°C until use.
2.4. Formulation of alginate gel beads
A solution of sodium alginate (Molar mass: 235.5 kDa; 1.56 Mannuronate/Guluronate ratio)
(Sigma Low viscosity, Sigma-Aldrich, Bornem, Belgium) was prepared in distilled water at
1.5 % w/v. In the same time, a 0.2 M calcium chloride (Acros Organics, Geel, Belgium)
solution was prepared in distilled water. The ionic strength of this solution was fixed at 0.5
M.
Eight mL of sodium alginate solution added to 1.8 mL of sunflower oil and 0.2 g of E-βfarnesene (solvent-free fraction (F3) from fractionation of Matricaria chamomilla essential
oil) or β-caryophyllene (solvent-free fraction (F2) from fractionation of Nepeta cataria
97
Chapter V
__________________________________________________________________________
essential oil) were mixed with an ultraturax system (IKA T18 Basic, QLab, Vilvoorde,
Belgium) at 24000 rpm during 20 s to obtain a thin and homogeneous emulsion. For the αtocopherol (Sigma-Aldrich, Bornem, Belgium) alginate beads type, 150 mg of this
component were added in the mix before the ultraturax emulsion process.
The emulsion was extruded by needle (0.4 mm I.D.) and the drops fell into agitated
(magnetic stir bar at 600 rpm) CaCl2 solution to form the alginate gel beads containing
semiochemical compounds. The distance between needle and CaCl2 solution was fixed at 20
cm to obtain spherical beads. The beads stayed 48 h in the CaCl2 solution to stabilise the
syneresis phenomenon. The beads were dried before use to eliminate surface water. First,
they were drained off on a filter paper during a few seconds. Then they were dried under air
pressure at 2 bars during 30 minutes.
2.5. Determination of the sesquiterpene protection efficiency of
formulations
For each tested sesquiterpene (E-β-farnesene and β-caryophyllene) obtained by flash
chromatography, four different formulations were compared in terms of protection efficiency
during twenty days. For each day of analysis, three replicates of each formulation were
prepared. The flasks were put under sunlight at room temperature until analysis, except for
the day 0 where the analysis took place directly after the preparation of the samples. A Hobo
data logger (Miravox, Hoevenen, Belgium) was installed near the flasks all along the 20 days
of experiment to measure the lab temperature.
The first formulation consisted in compounds, isolated by flash chromatography, and not
formulated in solvent or encapsulated in matrix. The purities of E--farnesene and caryophyllene from essential oils are presented in Table 1. Eighty mg of extract were
introduced in a 10 mL flask and let in the previous described conditions until analysis. For
the analysis, the flasks were filled up to the mark with n-hexane. One and a half mL of this
solution was transferred into another 10 mL flask with 250 µL of longifolene (internal
standard) at 10 µg µL-1. The flask was filled up to the mark with n-hexane. The solutions
were then analysed by fast GC for internal quantification.
98
Chapter V
__________________________________________________________________________
For the second tested formulation, the compounds were mixed with sunflower oil in a 1:10
(w/w) ratio. Seven hundreds mg of the mix were introduced in a 10 mL flask. For the
experiments, the flask was filled up with n-hexane. Before the fast GC analysis, sunflower oil
containing triglycerides had to be discarded from the solution. For this purpose, 1 mL of the
solution (sesquiterpene - sunflower oil - n-hexane) was fractionated over 1.5 g of silica gel
G60 (70-230 mesh: ref. no. 815330.1, from Macherey-Nagel) previously dried 16 hours at
120 °C and packed in a glass column (10 mm I.D.) with glass wool plug at the bottom. The
silicagel bed was 3 cm high. The deposited sample was eluted with 50 mL n-pentane. Nine
mL of this eluted extract were introduced with 250 µL of longifolene (internal standard) at 10
µg µL-1 in a 10 mL flask and filled up to the mark with n-hexane. The solution was then
analysed by fast GC for internal quantification.
The elution volume (50 mL) necessary to elute the semiochemicals from the column was
determined by measuring the elution recovery of a known quantity of E-β-farnesene
deposited on the silica gel. The recovery of 91.2% ± 0.4% was judged satisfactory according
to the AOAC norm (2006) which requires recoveries comprised between 85% and 110%
considering the concentrations tested.
The two other formulations to test consisted in sesquiterpene alginate beads with or without
added α-tocopherol, respectively. The α-tocopherol was added in the beads formulations as
an antioxidant. The protocol of protector effect determination was the same for the two types
of formulations. Ninety mg of alginate beads were introduced in a SOVIREL tube. Two mL
of pentasodium tripolyphosphate (Na5P3O10), at 25 µg µL-1 in water, were added to
destabilize the alginate beads cohesion and liberate the semiochemicals and the sunflower oil
contained in the beads. In the same time, 250 µL of internal standard (longifolene) at 10 µg
µL-1 were added in the tube for further quantification. The tube was let at rest during 30
minutes. Three successive extractions of sesquiterpene compounds were conducted with npentane as extraction solvent. For each extraction, 5 mL n-pentane were added to the tube,
the solution was homogenised for 10 min and centrifuged at 5000 rpm at 20 °C ± 2 °C. The
pentane phases coming from the 3 extractions were transferred cautiously in a flask. The
solvent was then evaporated until 1 mL at atmospheric pressure and 40 °C with a Büchi
rotatory evaporator. The same fractionation process than the one previously described was
elaborated to quantify semiochemical compounds without injecting sunflower oil on the GC
column. For this purpose, the 1 mL extract was deposited on 1.5 g of silica gel G60
99
Chapter V
__________________________________________________________________________
previously dried and packed in a glass column (10 mm I.D.) with glass wool plug at the
bottom. n-Pentane was used as elution solvent. Fifty mL were collected. This extract was
analysed by fast GC for internal quantification of E-β-farnesene or β-caryophyllene.
The protection efficiency of each formulation was expressed by the mean residual percentage
of compounds at each day of analysis compared to the mean values of day 0 (day 0 = 100%).
2.6. Method validation
The validation step was performed using the accuracy profile concept [24-26]. The range of
concentration levels for E-β-farnesene and β-caryophyllene was from 80 ng µL-1 to 1000 ng
µL-1 in n-hexane for the calibration standards and in a reconstituted matrix for the validation
standards. Longifolene was used as internal standard (I.S.) for these sesquiterpene
components at 102.6 ng µL-1 in each level of concentration for the calibration and for the
validation standards.
The choice of longifolene as internal standard was made for different reasons: this compound
belongs to the same family than the analytes (sesquiterpenes), the retention time of
longifolene was close to the retention time of the analytes without coelution (see 3.1.) and
longifolene was absent of the samples to analyse in routine.
2.6.1. Solutions used for calibration
For each component, three standard solutions were prepared in three replicates for three
series (three separated days) of analyses. The concentration levels are shown in Table 2.
Each solution was analysed by fast GC. The calibration curves were obtained for each series
of analyses by plotting the ratio of analysed peak area / I.S. peak area, versus the
concentration of analyte.
Table 2: Levels of calibration standards for E-β-farnesene and β-caryophyllene
Level
1
2
3
Total
100
Concentration of E-βfarnesene (ng µL-1)
25.5
509.8
1019.7
Concentration of βcaryophyllene (ng µL-1)
24.9
499.6
999.2
9 samples / series (day)/ compound
Concentration of I.S.
(ng µL-1)
102.6
102.6
102.6
Chapter V
__________________________________________________________________________
2.6.2. Solutions used for validation
Five independent standard solutions were prepared in three replicates for three series (days)
of analyses. The solutions consisted in matrix of alginate gel beads without sesquiterpene
treated as explained in 2.5., and spiked with fixed amounts of reference sesquiterpenes. The
concentration levels of the validation standards are shown in Table 3. These standards were
treated like real samples on fast GC.
Table 3: Levels of validation standards for E-β-farnesene and β-caryophyllene
Level
1
2
3
4
5
Total
Concentration of E-βfarnesene (ng µL-1)
81.6
163.2
367.1
734.2
1019.7
Concentration of βcaryophyllene (ng µL-1)
80.5
160.9
362.1
724.2
1005.8
Concentration of I.S.
(ng µL-1)
102.6
102.6
102.6
102.6
102.6
15 samples / series (day)/ compound
101
Chapter V
__________________________________________________________________________
3. Results and discussion
3.1. Resolution and selectivity of the method
The selectivity of the chromatographic method depends on the resolution of the targeted
compounds and on the absence of interference. In the present paper, the total resolution was
defined between the two nearest compounds, longifolene and -caryophyllene as:
Rs = 2(tR β-caryophyllene – tR longifolene) / (Wlongifolene + W β-caryophyllene),
where tR are the retention times and W are the peak widths of the two nearest compounds.
The resolution was good with Rs (1.65) higher than 1.5.
Moreover, by comparing 4 replicates of a blank injection (alginate beads with α-tocopherol
and sunflower oil, but without sesquiterpene, treated as explained in 2.5.) to a diluted mixture
(6 replicates) of E--farnesene (81.6 ng µL-1), -caryophyllene (80.5 ng µL-1) and
longifolene (102.6 ng µL-1), no peak or interference was observed at the retention times
corresponding of the analytes and the internal standard. Figure 1b shows the chromatogram
of a blank injection compared to the chromatogram of analytes (Figure 1a). The retention
times of the reference compounds are presented in Table 1.
3.2. Validation by use of accuracy profile approach
This concept of validation largely described in practice and theoretically explained by Hubert
et al. [24-26] and by Rozet et al. [40-42], can be summarized as follows:
-
Step 1: analyse the three series of calibration standards and draw calibration curves as
explained in 2.6.1. Test different regression models for the calibration curves.
-
Step 2: analyse the validation standards of each series. Back-calculate the predicted
concentrations by means of the peak area ratio obtained and introduced in the
corresponding regression equation.
-
Step 3: determine the mean bias (estimating the trueness) for each concentration
level, which corresponds to the systematic error.
102
Chapter V
__________________________________________________________________________
-
Step 4: calculate the precision parameters: repeatability and intermediate precision for
each concentration level, which correspond to the random error.
-
Step 5: determine the relative tolerance limits (β-expectation tolerance interval) for
each validation standard concentration level with a prespecified probability level β.
-
Step 6: plot the accuracy profile as the mean bias, the relative tolerance limits and the
acceptance limits in function of the concentrations, in relative values.
-
Step 7: determine the linearity of the method by plotting the back-calculated
concentrations of all the series (N=45) in function of the introduced concentrations
(concentration levels of the validation standards). This step is necessary to verify that
the analytical method gives results (in terms of predicted concentrations) strictly
proportional to the tested concentrations.
3.2.1. Analysis of the response functions and determination of the best
regression models
For each analyte and for each series of calibration standards, using the three calibration levels
ranging from 80 ng µL-1 to 1000 ng µL-1, different regression models were tested for E-βfarnesene and β-caryophyllene in order to define the most adequate one. The regression
models tested were the simple linear regression and the linear regression model through zero
fitted with the maximum concentration level of the calibration standards.
The calculation of the different validation parameters (trueness, precision, accuracy, linearity,
limits of detection (LOD) and quantification (LOQ) and range) was realized from each
regression model. Moreover, in each case, an accuracy profile, with a maximum risk limit of
5%, was constructed as it can be seen in Figure 2 and Figure 3 for E-β-farnesene and βcaryophyllene, respectively. Furthermore, in order to take decision about the validity of the
method, acceptance limits in terms of maximum total error have to be set. For this
application, two levels of acceptance limits were defined. First, for the concentration levels
ranging from 160 ng µL to 1000 ng µL-1, the acceptance limits was settled at +/- 15%,
meaning that a symmetric maximum total error of 15% around the true concentration of
analyte present in the sample could be accepted. Second, for the targeted lower limit of
quantification, a slightly larger acceptance limits was defined: +/- 25%. Indeed, it is
103
Chapter V
__________________________________________________________________________
reasonable to authorize increasing error for small concentrations as on one hand the absolute
value of this error (in concentration for e.g.) will still be acceptable and on the other hand this
concentration dependant behaviour of error has been reported since long time ago [43] and is
integrated in various validation guidelines [44, 45].
For the E-β-farnesene, two models were tested: a simple linear regression model (Figure 2a)
and a linear regression model through zero fitted with the maximum level of concentration
(Figure 2b). In both cases, the relative β-expectation tolerance limits were inside the
acceptance limits fixed at ± 15% and ± 25%. Nevertheless, considering on one hand the risk
profiles illustrated in Figure 4, and on the second hand its practical advantage the second
model seemed to be the most appropriate with a maximum risk of 2.8% (Figure 4a) instead
of 4.4% in the first model. Indeed, these risk profiles express that the probability to obtain
future results outside the specified acceptance limits for the concentration range tested are of
maximum 4.4% for the simple linear regression model and at most 2.8% for the highly
simple and economic one level calibration scheme.
Figure 2: Accuracy profiles of E-β-farnesene obtained by considering a simple linear regression model (a) and
by considering a linear regression model through zero fitted with the maximum level of concentration (b); plain
line: relative bias, dashed lines: β-expectation tolerance limits (β=95%), dotted curves: acceptance limit (+/25% and +/-15%) and dots: relative back-calculated concentrations of the validation standards.
104
Chapter V
__________________________________________________________________________
Figure 3: Accuracy profiles of β-caryophyllene obtained by considering a simple linear regression model (a)
and considering a linear regression model through zero fitted with the maximum level of concentration (b)
without correction of the bias. Accuracy profiles of β-caryophyllene obtained by considering a simple linear
regression model (c) and considering a linear regression model through zero fitted with the maximum level of
concentration (d) with a correcting factor of the bias. Plain line: relative bias, dashed lines: β-expectation
tolerance limits (β=95%), dotted curves: acceptance limit (+/-25% and +/-15%) and dots: relative backcalculated concentrations of the validation standards.
105
Chapter V
__________________________________________________________________________
Figure 4: Risk profiles of E-β-farnesene (a) and β-caryophyllene (b) for the chosen regression models. Dotted
line: maximum risk of 5%; dashed line: effective risk of having results falling outside the specified acceptance
limits.
In the case of β-caryophyllene, the study of accuracy profiles was more complex due to the
presence of a systematic bias at each concentration level for the two tested regression
models. Figure 3a and Figure 3b show the accuracy profiles obtained for, respectively, the
simple linear regression model and the linear regression through zero fitted with the
maximum concentration level. In both profiles, the β-expectation tolerance limits were
completely outside the acceptance limits fixed at ± 15%. Indeed, a strong proportional
systematic error was observed as given by the following equations of the linearity between
true concentration (X) and back-calculated concentration (Y) for the simple linear and the
forced through zero regressions, respectively:
Y  2.218  0.7361X
Y  4.629  0.7348 X
In order to correct this systematic error a correction factor was determined as the inverse of
the slopes of the linearity equations (1.3585 and 1.3609 for the simple linear and the forced
through zero regressions, respectively) [46]. The correction factors were then applied to the
results obtained using their respective regression model, as explained in Hubert et al. [47].
Figure 3c and Figure 3d show the corrected accuracy profiles with acceptance limits fixed at
± 15%, except at the lowest concentration level (± 25%). In the simple linear regression
model (Figure 3c), the upper tolerance limit stepped outside the 15% acceptance limit at the
second concentration level (160 ng µL-1). For the second model (Figure 3d), the tolerance
106
Chapter V
__________________________________________________________________________
limits were over the whole concentration range tested inside the acceptance limits except for
the first concentration level.
Nonetheless, after examination of the risk profile obtained using the forced through zero
calibration curve (Figure 4b), the true risk associated to the first concentration level was
very close (5.15 %) to the maximum risk limit fixed initially at 5%. Indeed, this slight
increased risk to obtained future results outside the ± 25% is perfectly acceptable with
respect to the final use of the method. This regression model was thus finally chosen as the
most appropriate for the quantification of β-caryophyllene.
3.2.2. Trueness of the method
The trueness expresses the closeness of agreement between the mean value obtained from a
series of measurements and the value which is accepted as the true value [25]. Trueness is
expressed in terms of bias (in relative (%) and absolute (ng µl-1) values) which corresponds
to the systematic error. As presented in Table 4, the relative biases are not too high ranging
from -0.67 % and from 0.38 % for E-β-farnesene and β-caryophyllene, respectively,
illustrating the good trueness of the method developed for each analyte when using a linear
regression curved forced through zero and using only the maximum concentration level of
the calibration standard.
3.2.3. Precision of the method
The precision expresses the closeness of agreement between a series of measurements
obtained from multiple sampling of the same sample [25]. The precision is evaluated in terms
of repeatability (same analytical procedure, same operator, and same day) and intermediate
precision (same analytical procedure but different operators and different days) expressed by
relative standard deviations (RSDs %).
As shown in Table 4, the relative standard deviations of repeatability and intermediate
precision are lower than 5% for both analytes, except at the lowest concentration levels
where, nevertheless, they never exceeded 10%.
107
Chapter V
__________________________________________________________________________
3.2.4. Accuracy of the method
The accuracy of an analytical procedure expresses the closeness of agreement between the
value found and the value accepted as the conventional true value. The closeness of
agreement observed is the resultant (total error) of the sum of the systematic and random
errors, also the sum of the trueness and the precision [25].
Figure 2b and Figure 3d show the accuracy profiles of E-β-farnesene and β-caryophyllene
for the chosen regression models. The plain central lines represent the relative biases. The
dashed lines are the 95% β-expectation tolerance limits (the probability that each future
results will fall inside these β-expectation tolerance limits is of 95%) and the dotted lines
represent the acceptance limits. The method is considered giving accurate results as long as
the β-expectation limits do not cross the acceptance limits. Table 4 gives the β-expectation
intervals for each analyte at all concentration level tested of the validation standard.
3.2.5. Limits of detection, quantification and range
The lower LOQ (LLOQ) is the lowest amount of the targeted analyte in the sample which
can be quantitatively determined under the experimental conditions prescribed with a well
defined accuracy [25]. Therefore, the LLOQ of both analytes are the smallest concentrations
tested for which the β-expectation tolerance intervals are included inside the acceptance
limits previously settled. As discussed earlier, based on a risk analysis the LLOQ was defined
as 80 ng µL-1 for β-caryophyllene. The LOD (limit of detection) was arbitrarily defined as
1/2 LOQ. The limit of detection is generally defined as the lowest quantity of a substance
that can be distinguished from the blank but which can not be quantified. LODs and LOQs
(in ng µL-1) are presented in Table 4.
3.2.6. Linearity
The linearity of the results generated by an analytical procedure is the ability within a given
range to obtain test results that are directly proportional to the concentrations (amounts) of an
analyte in the sample [25,48].
In practice, the linearity was determined, for the two compounds, by drawing a regression
line of the back-calculated concentrations (for all the series, N=45) in function of the
108
Chapter V
__________________________________________________________________________
introduced concentrations (validation standards, 5 levels ranging from 80 ng µL-1 to 1000 ng
µL-1).
For each compound, the linearity of the results obtained by the analytical method was
demonstrated using β-expectation tolerance limits (β=95%) fully included with the
acceptance limits expressed in concentration units as shown in Figure 5a and 5b for E-βfarnesene and β-caryophyllene (after correction of the results), respectively. The
determination coefficients (r2) as well as the linearity equations are given on Table 4.
Figure 5: Linearity profiles of (a) E-β-farnesene and (b) β-caryophyllene (after correction of the results). Plain
line: identity line (Y=X), dashed lines: β-expectation tolerance limits (β=95%), dotted curves: acceptance limit
expressed in ng µL-1 and dots: back-calculated concentrations of the validation standards.
109
Table 4: Validation results for E-β-farnesene and β-caryophyllene for the chosen regression models
β-caryophyllene
80.5 – 1005.8
E-β-farnesene
81.6 – 1019.7
Range (ng µL-1)
Response function
(m=3, n=3)
Slope
Trueness (n=3, p=3)
Concentration levels
1
2
3
4
5
Precision (n=3, p=3)
Concentration levels
1
2
3
4
5
Accuracy (n=3, p=3, β=0.95)
Concentration levels
1
2
3
4
5
Linearity (n=3, m=5, p=3), N=45
Range (ng µL-1)
Slope
Intercept
r2
Lower LOQ (ng µL-1)
Lower LOD (ng µL-1)
Series 1
0.0089
Absolute bias (ng µL-1)
-0.1
4.8
0.8
2.9
-6.8
Repeatability (RSD %)
8.8
3.4
2.9
0.8
1.0
β-expectation tolerance
limits in ng µL-1
[63.9 – 99.1]
[154.2 – 181.7]
[333.9 – 401.9]
[654.3 – 820.0]
[970.5 – 1055.0]
Series 2
0.0091
Series 3
0.0089
Relative bias (%)
-0.1
2.9
0.2
0.4
-0.7
Intermediate
Precision (RSD %)
8.8
3.4
3.4
2.6
1.4
β-expectation tolerance
limits in %
[-21.7 – 21.4]
[-5.5- 11.4]
[-9.0 – 9.5]
[-10.9 – 11.7]
[-4.8 – 3.5]
81.6 – 1019.7
0.9928
3.7050
0.9989
81.6
40.8
Series 1
0.0071
Absolute bias (ng µL-1)
3.5
9.3
4.9
9.7
3.8
Repeatability (RSD %)
9.8
3.4
2.3
0.7
1.1
β-expectation tolerance
limits in ng µL-1
[64.7 – 103.2]
[156.8 – 183.6]
[346.6 – 387.4]
[659.1 – 808.6]
[970.7 – 1049.0]
Series 2
0.0072
Series 3
0.0070
Relative bias (%)
4.3
5.7
1.3
1.3
0.4
Intermediate
Precision (RSD %)
9.8
3.4
2.3
2.3
1.4
β-expectation tolerance
limits in %
[-19.6 – 28.3]
[-2.5 – 14.1]
[-4.3 – 6.7]
[-9.0 – 11.6]
[-3.5 – 4.3]
80.5 – 1005.8
0.9998
6.3450
0.9991
80.5
49.5
Chapter V
__________________________________________________________________________
3.3. Measurement uncertainty
To allow a correct interpretation of results obtained by an analytical procedure, their
reliability must be demonstrated. Validation ensures that the method is fit for its future
purpose, however it is not sufficient if one aims at interpreting and comparing results
correctly. Uncertainty of measurements should therefore be evaluated to ensure this. One
major advantage of the applied validation methodology is that it can, without any additional
experiments, give estimation of uncertainty of measurements. Indeed Feinberg et al. [49]
demonstrated the mathematical link between the variance used to compute the β-expectation
tolerance interval and the uncertainty of the measurements as defined in the ISO/DTS 21748
[50]. Therefore, as long as the experimental design used for the validation is representative of
the sources of variability that will be encountered during routine analysis, this uncertainty
estimate is relevant for the results obtained in the laboratory having validated the analytical
procedure. Several estimations of uncertainty were thus computed without any additional
experiments and are presented in Table 5. The expanded uncertainty was computed using a
coverage factor of k=2 [39,51,52], representing an interval around the results where the
unknown true value can be observed with a confidence level of 95%. For the particular case
of β-caryophylene, a correction factor was introduced in order to alleviate the strong
systematic error observed. The introduction of this factor has two effects. First, it expands the
values of the uncertainty of the uncorrected results by a factor corresponding to its value, as
expected in theory. Second, the uncertainty of this factor should be taken into account. Since
this correction factor is the inverse of the slope, its uncertainty is the standard error of the
slope obtained from the least square linear regression. Its value is given in Table 5. As can
be seen in this table, the uncertainty of this correction factor is far from being the most
important source of uncertainty. Nonetheless, the combined standard uncertainty, expanded
uncertainty and relative expanded uncertainty were computed by incorporating this
supplementary uncertainty and are given in Table 5. It is also for this reason that the minor
additional uncertainty conveyed by the correction factor was not introduced in the accuracy
profiles of Figures 3c and 3d.
As shown in Table 5, the relative expanded uncertainty of each semiochemical irrespective
of the concentration levels did not exceed 10%, except for the smallest concentration levels
for which it is around 20%. In other words, this means that with a confidence level of 95%
the unknown true value is situated at maximum ±10% around the measured result for
111
Chapter V
__________________________________________________________________________
samples ranging from 160 to 1000 ng µL-1 and at maximum ±20% around the measured
result for samples at 80 ng µL-1.
112
Table 5: Estimates of the measurement uncertainties related to E-β-farnesene and β-caryophyllene, at each concentration level investigated in validation using the selected
regression models.
Analyte
E-β-farnesene
β-caryophyllene
Concentration
(ng µL-1)
Uncertainty of the bias
(ng µL-1)
81.6
163.2
367.1
734.2
1019.7
2.400
1.880
5.290
10.610
6.630
80.5
160.9
362.1
724.2
1005.8
2.623
1.822
2.784
9.410
6.147
Uncertainty of the correction
factor
(ng µL-1)
0.003286
0.003286
0.003286
0.003286
0.003286
Combined
Uncertainty
(ng µL-1)
7.58
5.94
13.71
21.80
15.78
8.30
5.81
8.95
19.51
15.76
Expanded
uncertainty
(ng µL-1)
15.16
11.88
27.41
43.59
31.56
16.60
11.61
17.90
39.02
31.52
Relative expanded
uncertainty
(%)
18.6
7.3
7.5
5.9
3.1
20.8
7.3
5.0
5.4
3.2
Chapter V
__________________________________________________________________________
3.4. Protection efficiency of the formulations
During the twenty days of analysis, the temperature was measured in the lab where the
experiments were conducted. The mean observed temperature was of 23.06 °C ± 1.90 °C.
The evolution of the protection capacity of the different formulations, expressed in terms of
residual percentage of compound, is presented in Figure 6a and Figure 6b for E-β-farnesene
and β-caryophyllene, respectively.
Figure 6: Residual percentage of E-β-farnesene (a) and β-caryophyllene (b) in formulations during twenty days.
For E-β-farnesene, the most stable formulation is the alginate beads without α-tocopherol.
The residual percentage of semiochemical in this formulation decreases slowly until day 10,
and then stays relatively stable until the end of the experiment at a value close to 85%. The
residual percentage in the beads with α-tocopherol becomes stable after 5 days at a lower
value close to 60 %. The quantity of E-β-farnesene formulated in sunflower oil and in the
non formulated pure E-β-farnesene decreases rapidly with a ½-life period of 2 and 1.2 days,
respectively.
114
Chapter V
__________________________________________________________________________
The case of β-caryophyllene is slightly different from that of E-β-farnesene in terms of
protection efficiency of formulations. There is no important difference in the residuals
percentage evolution for the three formulations (sunflower oil, beads with α-tocopherol and
beads without α-tocopherol), the stability being comprised between 60% and 70%. The non
formulated compound is rapidly degraded with a ½-life period of 1.6 days.
The alginate beads are more protective for the components formulated and are easier to
manipulate as slow release devices to put on fields in integrated pest management programs.
Slow release studies of semiochemicals are presently conducted on these alginate beads to
determine a mathematical kinetic model of release considering the impact of physicochemical parameters like temperature, relative humidity, and wind. The results of this
experiment will be presented in a following paper.
115
Chapter V
__________________________________________________________________________
4. Conclusion
The method for the quantification of semiochemical sesquiterpenes by fast GC-FID was
completely validated by applying the concept of total error using the accuracy profile as
decision tool. The accuracy profiles were constructed for the two analytes (E-β-farnesene and
β-caryophyllene) by considering a probability of 95% and a linear regression through zero
fitted with the maximum level of concentration as calibration curve. The different validation
criteria were evaluated and the lowest limits of quantification were determined. The method
developed is providing accurate results along the concentration range evaluated for the two
analytes, i.e. from 80 ng µl-1 to 1000 ng µl-1. In addition the measurements uncertainties were
estimated without any additional experiments thanks to the validation methodology, allowing
correct interpretation and comparison of the results in a cost effective manner.
Moreover, the alginate gel beads formulations were estimated in terms of protection
efficiency of sesquiterpenes. The results showed that the beads protect the compounds at
relatively high levels (between 60% and 85% of residual percentages with a stabilisation of
the degradation) during minimum twenty days. Some experiments presently in study (lab
controlled conditions) show that the release of such formulations can be conducted during at
least 80 days. The biological effects of the slow release devices (tested on Episyrphus
balteatus De Geer and on aphid parasitoids) have also been demonstrated in lab experiments
and in naturally conditions (field experiments) (unpublished results).
Acknowledgements
The authors are grateful to Dr. S. Bartram and Prof. W. Boland from the Max Planck Institute
for Chemical Ecology (Jena, Germany) for providing E-β-farnesene from chemical synthesis.
This research was funded by a Walloon Region Ministry grant (WALEO2: SOLAPHIDRW/FUSAGX 061/6287) and by an FRFC grant (no. 2.4586.04F).
A research grant from the Belgium National Fund for Scientific Research (FRS-FNRS) to E.
Rozet is also gratefully acknowledged.
116
Chapter V
__________________________________________________________________________
References
[1] P.D. Cox, Potential for using semiochemicals to protect stored products from insect
infestation, J. Stored Prod. Res. 40 (2004) 1-25.
[2]
A.D. Graves, E.H. Holsten, M.E. Ascerno, K.P. Zogas, J.S. Hard, D.P.W. Huber, R.A.
Blanchette, S.J. Seybold, Protection of spruce from colonization by the bark beetle, Ips
perturbatus, in Alaska, Forest Ecol. Manag. 256 (2008) 1825-1839.
[3]
E.M. Hegazi, M.A. Konstantopoulou, A. Herz, B.E. Mazomenos, W.E. Khafagi, E.
Agamy, A. Zaitun, G.M. Abd El-Aziz, S. Showiel, S.M. Abdel-Rahman, Is mating
disruption effective in controlling the olive moth, Prays oleae?, Crop Protec. 28 (2009)
181-189.
[4]
F. Francis, S. Vandermoten, F. Verheggen, G. Lognay, E. Haubruge, Is the (E)-ßfarnesene only volatile terpenoids in aphids?, J. Applied Ento. 129 (2005) 6-11.
[5]
E. Szöke, E. Maday, G. Marczal, E. Lemberkovics, Analysis of biologically active
essential oil components of chamomiles in Hungary (In vivo – In Vitro), Acta Hortic.
597 (2003) 275-284.
[6]
S. Heuskin, B. Godin, P. Leroy, Q. Capella, J.-P. Wathelet, F. Verheggen, E. Haubruge,
G. Lognay, Fast gas chromatography characterisation of purified semiochemicals from
essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L.
(Lamiaceae), J. Chromatogr. A 1216 (2009) 2768-2775.
[7]
F.J. Verheggen, L. Arnaud, S. Bartram, M. Gohy, E. Haubruge, Aphid an plant
volatiles induce oviposition in an aphidophagous Hoverfly, J. Chem. Ecol. 34 (2008)
301-307.
[8]
F. Francis, T. Martin, G. Lognay, E. Haubruge, Role of (E)-β-farnesene in systematic
aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae), Eur. J.
Entomol. 102 (2005) 431-436.
[9]
F.J. Verheggen, E. Haubruge, C.M. De Moraes, M.C. Mescher, Social environment
influences aphid production of alarm pheromone, Behav. Ecol. 20 (2009) 283-288.
[10] Y. Du, G.M. Poppy, W. Powell, J.A. Pickett, L.J. Wadhams, C.M. Woodcock,
Identification of semiochemicals released during aphid feeding that attract parasitoid
Aphidius ervi, J. Chem. Ecol. 24 (1998) 1355-1368.
117
Chapter V
__________________________________________________________________________
[11] F.J. Verheggen, Q. Fagel, S. Heuskin, G. Lognay, F. Francis, E. Haubruge,
Electrophysiological and behavioral responses of the multicolored asian lady beetle,
Harmonia axyridis Pallas, to sesquiterpene semiochemicals, J. Chem. Ecol. 33 (2007)
2148-2155.
[12] A.H. Gilani, A.J. Shah, A. Zubair, S. Khalid, J. Kiani, A. Ahmed, M. Rasheed, V.U.
Ahmad, Chemical composition and mechanisms underlying the spasmolytic and
bronchodilatory properties of the essential oil of Nepeta cataria L., J. Ethnopharmaco.
121 (2009) 405-411.
[13] B.S. Tomova, J.S. Waterhouse, J. Boberski, The effect of fractionated Tagetes oil
volatiles on aphid reproduction, Entomol. Exp. Appl. 115 (2005) 153-159.
[14] M.E. Malone, I.A.M. Appelqvist, Gelled emulsion particles for the controlled release of
lipophilic volatiles during eating, J. Contr. Release 90 (2003) 227-241.
[15] F. Lai, G. Loy, M. Manconi, M.L. Manca, A.M. Fadda, Artemesia arborescens L.
essential oil loaded beads: preparation and characterization, AAPS Pharm. Sci. Tech. 8
(3) (2007) E1-E7.
[16] C.P. Chang, T. Dobashi, Preparation of alginate complex capsules containing eucalyptus
essential oil and its controlled release, Colloids and Surf. B 32 (2003) 257-262.
[17] J.J.G. van Soest, Encapsulation of fragrances and flavours: a way to control odour and
aroma in consumer products, in: R.G. Berger (Ed.), Flavours and Fragrances, Springer,
2007, pp.439-455.
[18] M.A. Rojas-Graü, M.S. Tapia, F.J. Rodriguez, A.J. Carmona, O. Martin-Belloso,
Alginate and gellan-based edible coatings as carriers of antibrowning agents applied on
fresh-cut Fuji apples, Food Hydrocoll. 21 (2007) 118-127.
[19] R.A. Sheldon, I. Arends, U. Hanefeld, Green Chemistry and Catalysis, Wiley-VCH,
2007, 448 p.
[20] A. V. Tkachev, The chemistry of caryophyllene and related compounds, Chem. Nat.
Comp. 23 (4) (1987) 393-412.
[21] M. Sköld, A.-T. Karlberg, M. Matura and A. Börje, The fragrance chemical βcaryophyllene—air oxidation and skin sensitization, Food Chem. Toxic. 44 (4) (2006)
538-545.
118
Chapter V
__________________________________________________________________________
[22] J. Parshintsev, J. Nurmi, I. Kilpeläinen, K. Hartonen, M. Kulmala, M.L. Riekkola,
Preparation of beta-caryophyllene oxidation products and their determination in
ambient aerosol samples, Anal. Bioanal. Chem. 390 (3) (2008) 913-919.
[23] I. Kourtchev, I. Bejan, J.R. Sodeau and J.C. Wenger, Gas-Phase Products and Secondary
Aerosol Yields from the Ozonolysis of (E)-β-farnesene, European Aerosol Conference
2009, Karlsruhe, Abstract T021A01.
[24] P. Hubert, P. Chiap, J. Crommen, B. Boulanger, E. Chapuzet, N. Mercier, S. BervoasMartin, P. Chevalier, D. Grandjean, P. Lagorce, M. Lallier, M.C. Laparra, M.
Laurentie, J.C. Nivet, The SFSTP guide on the validation of chromatographic methods
for drug bioanalysis: from the Washington Conference to the laboratory, Anal. Chim.
Acta 391 (1999) 135-148.
[25] P. Hubert, J.J. Ngyen-Huu, B. Boulanger, E. Chapuzet, P. Chiap, N. Cohen, P.A.
Compagnon, W. Dewe, M. Feinberg, M. Lallier, M. Laurentie, N. Mercier, G. Muzard,
C. Nivet, L. Valat, Validation des procédures analytiques quantitatives. Harmonisation
des démarches, STP Pharma Pratiques 13 (3) (2003) 101-138.
[26] P. Hubert, J.J. Ngyen-Huu, B. Boulanger, E. Chapuzet, N. Cohen, P.A. Compagnon, W.
Dewe, M. Feinberg, M. Laurentie, N. Mercier, G. Muzard, L. Valat, Validation des
procédures analytiques quantitatives: Harmonisation des démarches. Partie II –
Statistiques, STP Pharma Pratiques 16 (1) (2006) 28-58.
[27] P. Chiap, Ph. Hubert, B. Boulanger, J. Crommen, Validation of an automated mehod for
the liquid chromatographic determination of atenolol in plasma: application of a new
validation protocol, Anal. Chim. Acta 391 (1999) 227-238.
[28] S. Block, D. Brkic, Ph. Hubert, J. Quetin-Leclercq, A validated method for the
quantification of pimarane and trachylobane diterpens in the leaves of Croton
zambesicus by capillary gas chromatography, Phytochem. Anal. 16 (2005) 342-348.
[29] R.D. Marini, P. Chiap, B. Boulanger, S. Rudaz, E. Rozet, J. Crommen, Ph. Hubert, LC
method for the determination of R-timolol in S-timolol maleate: Validation of its ability
to quantify and uncertainty assessment, Tanlanta 68 (2006) 1166-1175.
[30] E. Strahm, S. Rudaz, J.-L. Veuthey, M. Saugy, C. Saudan, Profiling of 19-norsteroid
sulfoconjugates in human urine by liquid chromatography mass spectrometry, Anal.
Chim. Acta 613 (2008) 228-237.
119
Chapter V
__________________________________________________________________________
[31] D.S.S. Kpoviéssi, F. Gbaguidi, J. Gbénou, G. Accrombessi, M. Moudachirou, E. Rozet,
Ph. Hubert, J. Quetin-Leclercq, Validation of a method for the determination of sterols
and triterpenes in the aerial part of Justicia anselliana (Nees) T. Anders by capillary
gas chromatography, J. Pharm. Biom. Anal. 48 (2008) 1127-1135.
[32] J. Mantanus, E. Ziémons, P. Lebrun, E. Rozet, R. Klinkenberg, B. Streel, B. Evrard, Ph.
Hubert, Moisture content determination of pharmaceutical pellets by near infrared
spectroscopy: Method development and validation, Anal. Chim. Acta 642 (2009) 186192.
[33] M. Feinberg, Approche globale et harmonisée de la validation, Spectra Analyse 249
(2006) 16-23.
[34] C. Brasseur, F. Brose, A. Pirlot, C. Douny, G. Eppe, G. Maghuin-Rogister, M.-L.
Scippo, Validation of the analytical procedure for the determination of polyaromatic
hydrocarbons in smoke flavourings using high performance liquid chromatography
coupled to an ultraviolet, diode array or fluorescence detector, Accred. Qual. Assur. 12
(2007) 535-542.
[35] W. Horwitz, L.A., Kamps, K.W. Boyer, Quality assurance in the analysis of foods for
trace constituents, J. Assoc. Off. Anal. Chem. 63 (1980) 1344-1354.
[36] Guidance for industry: Bioanalytical Method Validation, US Department of Health and
Human Services, Food and Drug Administration, Center for Drug Evaluation and
Research (CDER), Center for Biologics Evaluation and Research (CBER), Rockville,
May 2001.
[37] C.T. Viswanathan, S. Bansal, B. Booth, A.J. DeStefano, M.J. Rose, J. Sailstad, V.P.
Shah, J.P. Skelly, P.G. Swann, R.Weiner, Workshop/Conference Report —
Quantitative Bioanalytical Methods Validation and Implementation: Best Practices for
Chromatographic and Ligand Binding Assays, AAPS J. 9(1) (2007) E30-E42.
[38] 2002/657/EC: Commission Decision of 12 August 2002 implementing Council
Directive 96/23/EC Concerning the Performance of Analytical Methods and the
Interpretation of Results. Off. J. Eur. Commun. L221 (2002) 8-36.
[39] S. L. R. Ellison, M. Rosslein, A. Williams (Eds.), Eurachem/Citac guide, Quantifying
the uncertainty in analytical measurement, 2nd Edition, 2000, 120 p.
120
Chapter V
__________________________________________________________________________
[40] E. Rozet, R. Morello, F. Lecomte, G.B. Martin, P. Chiap, J. Crommen, K.S. Boos, Ph.
Hubert, Performances of a multidimensional on-line SPE-LC-ECD method for the
determination of three major catecholamines in native human urine: Validation, risk
and uncertainty assessments, J. Chromatogr. B 844 (2006) 251-260.
[41] E. Rozet, A. Ceccato, C. Hubert, E. Ziemons, R. Oprean, S. Rudaz, B. Boulanger, Ph.
Hubert, Analysis of recent pharmaceutical regulatory documents on analytical method
validation, J. Chromatogr. A 1158 (2007) 111-125.
[42] E. Rozet, V. Wascotte, N. Lecouturier, V. Préat, W. Dewé, B. Boulanger, Ph. Hubert,
Improvement of the decision efficiency of the accuracy profile by means of a
desirability function for analytical methods validation. Application to a diacetylmonoxime colorimetric assay used for the determination of urea in transdermal
iontophoretic extracts, Anal. Chim. Acta 591 (2007) 239-247.
[43] W Horwitz, L R Kamps, K W Boyer, Quality assurance in the analysis of foods and
trace constituents, J Assoc Off Anal Chem 63 (6) (1980)1344-54.
[44] Guidance for industry: Bioanalytical Method Validation, US Department of Health and
Human Services, Food and Drug Administration, Center for Drug Evaluation and
Research (CDER), Center for Biologics Evaluation and Research (CBER), Rockville,
May 2001.
[45] European Commission Decision 2002/657/EC implementing Council Directive
96/23/EC concerning the performance of analytical methods and the interpretation of
results.
[46] M. Feinberg, M. Laurentie, A global approach to method validation and measurement
uncertainty, Accred. Qual. Assur. 11 (2006) 3–9.
[47] Ph. Hubert, J.-J. Nguyen-Huu, B. Boulanger, E. Chapuzet, N. Cohen, P.-A. Compagnon,
W. Dewé, M. Feinberg, M. Laurentie, N. Mercier, G. Muzard, L. Valat, E. Rozet,
Harmonization of strategies for the validation of quantitative analytical procedures: A
SFSTP proposal. Part IV. Examples of application, J. Pharm. Biomed. Anal. 48 (2008)
760-771.
[48] International Conference on Harmonization (ICH) of Technical Requirements for
registration of Pharmaceuticals for Human Use, Topic Q2 (R1): Validation of
Analytical Procedures: Text and Methodology, Geneva, 2005.
121
Chapter V
__________________________________________________________________________
[49] M. Feinberg, B. Boulanger, W. Dewé, Ph. Hubert, New advances in chemical data
quality: method validation and measurement uncertainty, Anal. Bioanal. Chem. 380
(2004) 502-514.
[50] ISO/DTS 21748 (2003) Guide to the use of repeatability, reproducibility and trueness
estimates in measurement uncertainty estimation. ISO, Geneva.
[51] Analytical Methods Committee, Uncertainty of measurement - implications for its use in
analytical science, Analyst 120 (1995) 2303-2308.
[52] EA-4/16, EA Guidelines on the Expression of Uncertainty in Quantitative Testing, 2004,
http://www.european-accreditation.org.
122
Chapter VI
Optimisation, efficiency and slowrelease study of semiochemical
formulations
Chapter VI.1
Chapter VI
_________________________________________________________________________
Objectives
Preliminary experiments conducted in the previous chapter on semiochemical
formulations shown that alginate beads were efficient to protect sesquiterpenes
from oxidative degradation.
The first part of the present chapter deals with the optimisation of these alginate
gel bead formulations in terms of semiochemical encapsulation capacity. The
alginate beads were also characterised in order to determine the influence of
formulation parameters on their texture and to observe the dispersion of
semiochemicals in the polymeric network. The alginate formulations were also
tested in terms of semiochemical release in fixed laboratory conditions in order to
verify their efficiency as slow-release devices. The capacity of semiochemical
beads to attract aphid parasitoids, Aphidius ervi, was also evaluated by
olfactometry bioassays.
Figure 1’ describes the different experimental steps of the part 1 of this chapter.
The principle of alginate bead formulation is presented in Figure 2’. Figure 3’
shows the volatile collection system developed to study semiochemical release
from formulations.
127
Figure 1’: Steps of the alginate bead formulation optimisation
Figure 2’: Alginate bead formulation process
Figure 3’: Volatile collection system
Chapter VI.1
__________________________________________________________________________
VI. 1
Optimisation of a semiochemical slow-release alginate formulation
attractive towards Aphidius ervi Haliday parasitoids
Stéphanie Heuskin1,2,*, Stéphanie Lorge3,*, Bruno Godin1, Pascal Leroy4, Isabelle Frère5,
François J. Verheggen4, Eric Haubruge4, Jean-Paul Wathelet2, Michèle Mestdagh3, Thierry
Hance5, Georges Lognay1
1
Department of Analytical Chemistry, Gembloux Agro-Bio Tech, University of Liege, Passage des
Déportés 2, B-5030 Gembloux (Belgium).
2
Department of General and Organic Chemistry, Gembloux Agro-Bio Tech, University of Liege,
Passage des Déportés 2, B-5030 Gembloux (Belgium).
3
Institute of Condensed Matter and Nanosciences, Catholic University of Louvain, Croix du Sud 2,
B-1348 Louvain-la-Neuve (Belgium).
4
Department of Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of
Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
5
Earth and Life Institute, Catholic University of Louvain, Croix du Sud 4-5, B-1348 Louvain-laNeuve (Belgium).
* These authors contributed equally to this work.
Reference: Accepted for publication in Pest Management Science journal.
131
Chapter VI.1
__________________________________________________________________________
Abstract
BACKGROUND
Optimisation of alginate formulations is described in order to develop semiochemical (E-βfarnesene and E-β-caryophyllene) slow-release devices in biological control approaches by
attracting predators and parasitoids of aphids. Various formulation criteria were optimised
considering semiochemical encapsulation capacity. Moreover, optimised formulation was
characterised by texturometry and confocal microscopy. Slow-release rates of semiochemical
were calculated in laboratory controlled conditions.
Attractiveness of semiochemical formulations towards Aphidius ervi was demonstrated by
olfactometry.
RESULTS
Two major parameters were highlighted in encapsulation optimisation: type of alginate
(Sigma Low Viscosity) and type of cross-linker ion (Ca2+). Other formulation parameters
were optimised: ionic strength (0.5 M), Ca2+ (0.2M) and alginate (1.5%) concentrations, and
maturation time of beads in CaCl2 solution (48 h). After physical characterisation of beads,
semiochemical slow-release measurements showed that alginate formulations were efficient
sesquiterpene releasers with 503 µg E-β-farnesene and 1791 µg E-β-caryophyllene totally
released in 35 days.
The efficiency of semiochemical alginate beads as attractants for female parasitoids was
demonstrated with high percentages of attraction for semiochemical odours (88% and 90%
for E-β-farnesene and E-β-caryophyllene, respectively) and significative statistical results.
CONCLUSION
Semiochemical alginate beads can be considered as efficient slow-release systems in
biological control. These formulations could be very useful to attract aphid parasitoids on
crop fields.
Keywords
Semiochemical; biological control; alginate beads; slow-release; aphids; Aphidius ervi
132
Chapter VI.1
__________________________________________________________________________
1. Introduction
Aphids constitute a major problem in many crops by causing direct injuries in plants and by
transmitting viral diseases 1. To reduce their population, pesticides have been used for a long
time. Pesticide utilisation is, nevertheless, not harmless for human health and environment.
Adversely side effects have often been related like the development of pest insect resistance
towards synthetic pesticides 2. In order to decrease pesticide use, integrated pest management
(IPM) strategies have been conducted in many crop protection programs 3. Semiochemicals –
informative molecules used in insect-insect or plant-insect interactions – have been
considered within various IPM strategies
4-10
. Biological control is one of these IPM tactics
which implies the utilisation of natural enemies (predators and parasitoids) to reduce
problems due to phytophagous pest insects 11.
In the present research, two sesquiterpenoids: E-β-farnesene and E-β-caryophyllene have
been formulated for their potential biological properties as aphid predator and parasitoid
attractants. E-β-farnesene, the alarm pheromone of many aphid species
12
, has also been
identified as a kairomone (allelochemical substance produced by members of one species and
which benefits to individuals of another species) by attracting and inducing oviposition of the
aphid predators Episyrphus balteatus De Geer (Diptera: Syrphidae)
13-16
and by attracting
aphid parasitoids: Aphidius ervi Haliday (Hymenoptera: Braconidae) 17,18. E-β-caryophyllene
was identified as a potential component of the aggregation pheromonal blend of the Asian
ladybird beetle: Harmonia axyridis Pallas
(2005)
21
19,20
. On a biological point of view, Tomova et al.
reported that this molecule has an activity against aphid (Acyrthosiphon pisum)
reproduction. Sasso et al. (2009)
22
studied the behavioural response of A. ervi to tomato
plant volatiles among which E-β-caryophyllene sesquiterpene was identified. The results
obtained by electroantennography and by wind tunnel assays showed attractiveness of
parasitoids to this compound.
Aphidius ervi (Haliday) (Hymenoptera: Braconidae: Aphidiinae) is an effective biological
control agent
18
of various aphid species among which the pea aphid, Acyrtosiphon pisum
(Harris) is its main host
23-27
. Semiochemical cues coming from host-damaged plants28-32 or
from the hosts themselves, are of crucial importance in the foraging behaviour (host location
process) of female parasitoids 33, 34.
133
Chapter VI.1
__________________________________________________________________________
These two compounds of interest have been purified from essential oils of Matricaria
chamomilla L. (Asteraceae) and Nepeta cataria L. (Lamiaceae) for E-β-farnesene and E-βcaryophyllene, respectively 35,36. Slow-release devices have then been investigated in order to
deliver semiochemicals as biological control systems. Due to their sensitivity to air and
oxygen 37-40, the molecules need to be protected from degradation. For this purpose, alginate,
being a hydrophilic matrix with low oxygen permeability
41
, was considered as polymer for
semiochemical formulations. Alginate, a polysaccharide present in brown algae
(Phaeophyceae), chemically consists of linear polymers of (1→ 4) linked α-L-guluronate (G)
and β-D-mannuronate (M) residues organised in a non-regular, blockwise pattern along a
linear chain. These two sugars are present in three kinds of polymer segments: one consisting
essentially of D-mannuronic acid units (MMMMM), the second of L-guluronic acid units
(GGGGG) and the third of alternating D-mannuronic acid and L-guluronic acid residues
(MGMGMG). The type of arrangement of these blocks and the proportion of each one
control the gelling properties and the affinity of the alginate for divalent metal cations like
Ca2+ or Cu2+. Guluronic acid-rich alginates are most suited for gelling, even if all alginates
will gel in the presence of divalent metal ions. Gels made with mannuronic acid-rich
alginates are more deformable. The mode of alginate gel formation is generally explained by
the “egg-box” model. The rigid shape of the poly-G sections contains cavities favourable to
divalent cations considering that they are linked with carboxylate and other electronegative
oxygen atoms (Figure 1) 42-45.
Figure 1: Mannuronnate (Poly M), guluronate (Poly G) and Poly MG blocks constitutive of alginate polymer.
(Source: http://www.kjemi.uio.no/Polymerkjemi/Research/Alginate.htm)
134
Chapter VI.1
__________________________________________________________________________
Because of their gelling properties and their inherent low toxicity, alginates play a significant
role in the field of encapsulation and immobilisation of various molecules: food aromas 46-48,
essential oils 49,50 and pharmaceuticals 51,52.
The main objective of the present work was to hold and deliver semiochemical substances in
a controlled way and to test their attractiveness efficiency towards A. ervi. Consequently, a
careful selection of alginates was realised to use the most suitable type for encapsulation.
Formulated beads showed different structural and encapsulation properties depending on
various parameters of formulation. The gel structure influenced its mechanical strength and
diffusion properties and therefore the efficacy of the immobilisation processes. The
mechanical properties of the gel were characterised by submitting gels to uniaxial
compression force. The rigidity of the gel depended on the content of guluronic acid
on alginate and ion concentrations
54
53
and
. Alginate formulations were also characterized by
confocal microscopy in order to observe the distribution of semiochemical in alginate
network.
Indeed, several studies utilised the confocal microscopy to measure the
encapsulation efficiency, distribution and release mechanisms of biomolecules incorporated
in microcapsules and microspheres
55-57
. Semiochemical release rate measurements were
realised in laboratory controlled conditions followed by olfactometry bioassays on female A.
ervi parasitoids.
135
Chapter VI.1
__________________________________________________________________________
2. Materials and methods
2.1. Chemicals and reagents
Essential oil of Matricaria chamomilla was purchased from Vossen & Co. (Brussels,
Belgium) and was originated from Nepal (lot no. CHA06MI0406). Essential oil of Nepeta
cataria was purchased from Essential7.com (Roswell, NM, USA) and was originated from
Canada (lot no. EO0020f).
E-β-Farnesene from chemical synthesis (99.4 % ± 0.2 %) was kindly supplied by Dr. S.
Bartram and Prof. W. Boland (Max Planck Institute for Chemical Ecology, Jena, Germany).
E-β-Farnesene (83.8 % ± 0.3), E-β-caryophyllene (97.7 % ± 0.5 %) from natural origin were
extracted by flash chromatography from essential oils of M. chamomilla L. and N. cataria L.,
respectively as reported by Heuskin et al. (2010) 36. (+)-Longifolene (GLC purity > 99.9%),
as internal standard, was purchased from ABCR (Karlsruhe, Germany). The mean purities
with standard deviation (ten replicates) of these sesquiterpenes were determined by fast gas
chromatography.
n-Hexane of GC grade was purchased from VWR (Leuven, Belgium). n-Pentane extra pure
was purchased from Acros Organics (Geel, Belgium).
Three different sodium alginates were tested for formulation and are presented in Table 1
with their mannuronate / guluronate (M/G) ratios (this ratio was determined by polarimetry
on alginate of the same origin in a previous study
58
; M/G ratio of present alginates was
confirmed by nuclear magnetic resonance (NMR)) and their molar mass (obtained by
viscosimetry using the Mark-Houwink equation 58).
Table 1: Sodium alginates with their characteristics (Mannuronate/Guluronate ratio and molar mass)
Compound
M/G1 M/G (NMR)
Molar mass2 (kDa)
Satialgine SG 500 (SKW Biosystems, Paris, France)
0.6
0.86
271
Alginic acid, sodium salt. Medium viscosity (Sigma n°A-2033, 1.22
/
429.5
Bornem, Belgium)
Alginic acid, sodium salt. Low viscosity (Sigma n°A-2158, Bornem, 1.56
1.56
235.5
Belgium)
1
M/G ratio determined by polarimetric method 58
2
Molar mass obtained by viscosimetry using the Mark-Houwink equation 58:  = KMa, where : intrinsic
viscosity; M: molar mass; K and a: parameters dependent of the polymer-solvent system.
136
Chapter VI.1
__________________________________________________________________________
Calcium chloride was purchased from Acros Organics (Geel, Belgium) and copper chloride
was purchased from Sigma-Aldrich (Bornem, Belgium).
The fluorescent marker, FM4-64, used for confocal microscopic analyses was purchased
from Invitrogen (Merelbeke, Belgium).
2.2.
1
H-NMR characterisation of sodium alginate M/G ratio
NMR experiments were run on a Bruker DRX spectrometer operating at 500 MHz for 1H
spectra. The spectrometer was equipped with a 5 mm z-gradient broadband inverse probe.
Samples at a concentration of 20 mg mL-1 of alginate were prepared in D2O
59
. One-
dimensional 1H spectra were measured at 90°C to decrease sample viscosity and therefore
optimise resonance baseline. An aliquot of d4-TSP (sodium 3-trimethylsilyl propionate
2,2,3,3-d4) was added as internal reference. The 90° pulse width was calibrated in order to
maximise the signal and varied between 10 and 12 µs. The 1H spectral width was set at 10
ppm. Each spectrum acquisition was composed of 1024 scans with delay time of 15 s
between scans to allow spin relaxation.
2.3. Alginate gel bead formulation
Sodium alginate solutions were prepared in distilled water from 1 % to 3% w/v. Calcium
chloride and copper chloride solutions were prepared in distilled water from 0.05 M to 0.5 M.
The ionic strength of these Ca-/ Cu- chloride solutions was adjusted to 0.5 M or 1.0 M with
pure sodium chloride (Fluka, Bornem, Belgium). The ionic strength was only determined by
the quantity of salt added in the solution and did not take into account the ionic strength
coming from alginate polymer. Eight mL of sodium alginate solution added to 1.8 mL of
sunflower oil, 150 mg of α-tocopherol (Sigma-Aldrich, Bornem, Belgium) and 0.2 g (2 %
w/v) to
2.0 g (20 % w/v) E-β-farnesene
or
E-β-caryophyllene (from essential oil
fractionations) were mixed with an ultraturax system (IKA T18 Basic, QLab, Vilvoorde,
Belgium) at 24000 rpm for 20 seconds to obtain a thin and homogeneous emulsion. The
emulsion was extruded by needle (from 0.4 to 1.6 mm I.D.) and droplets fell into agitated
(magnetic stir bar at 600 rpm) calcium or copper chloride solution to form the alginate gel
beads containing semiochemical components. The distance between needle and chloride
solution was fixed at 20 cm to obtain spherical beads. The beads stayed between 20 minutes
and 48 hours in the ionic solution to stabilise the syneresis phenomenon. To eliminate surface
137
Chapter VI.1
__________________________________________________________________________
water the beads were dried before use as slow-release devices. They were first drained off on
a filter paper for a few seconds. Then they were dried under air pressure at 2 bars for 30
minutes at 21°C ± 2°C.
The various experimental conditions were applied following two experimental designs as
presented in Tables 2 and 3.
Table 2: Experimental design n°1
Run
order
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
138
Type of
alginate
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Satialgine
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma L
Sigma M
Sigma M
Sigma M
Sigma M
Sigma M
Sigma M
Sigma M
Sigma M
Sigma M
Alginate %
(w/v)
1
1
1
1.5
1.5
1.5
2
2
2
2.5
2.5
2.5
1
1
1.5
1.5
1.5
2
2
2
2.5
2.5
2.5
1
1
1
1.5
1.5
2
2
2.5
2.5
Type of ion
Ion (M)
Calcium
Copper
Copper
Calcium
Calcium
Copper
Calcium
Copper
Copper
Calcium
Calcium
Copper
Calcium
Copper
Calcium
Copper
Copper
Calcium
Calcium
Copper
Calcium
Copper
Copper
Calcium
Calcium
Copper
Calcium
Copper
Calcium
Copper
Calcium
Copper
0.05
0.136
0.3
0.05
0.136
0.136
0.3
0.05
0.136
0.136
0.3
0.05
0.05
0.136
0.3
0.05
0.3
0.05
0.136
0.3
0.136
0.05
0.136
0.05
0.136
0.3
0.05
0.136
0.136
0.05
0.3
0.05
Diameter of the
needle (mm)
0.9
1.6
0.4
0.4
1.6
0.9
0.4
1.6
0.9
0.4
1.6
0.9
0.9
1.6
1.6
1.6
0.4
0.9
0.4
1.6
0.9
0.4
1.6
1.6
0.4
0.9
1.6
0.4
0.9
1.6
0.9
0.4
Time of maturation
(h)
24
0.3
0.3
24
48
2
2
48
24
2
0.3
48
2
2
2
2
48
0.3
0.3
24
48
0.3
24
48
24
48
0.3
0.3
48
2
2
24
Chapter VI.1
__________________________________________________________________________
Table 3: Experimental design n°2
Run
Alginate Sigma Low viscosity %
order
(w/v)
1
3
2
2.5
3
2
4
2.5
5
3
6
1
7
2
8
1.5
9
2.5
10
1.5
11
3
12
1
13
3
14
2
15
2
16
2.5
17
1.5
18
1.1
19
1.2
1
Ca2+/COO- ratio expressed in equivalents ratio.
Calcium
(M)
0.050
0.050
0.050
0.070
0.070
0.136
0.070
0.070
0.200
0.136
0.136
0.200
0.200
0.136
0.200
0.136
0.200
0.200
0.200
(Ca2+)/(COO-)1
1.64
1.98
2.30
2.50
2.77
3.50
4.46
4.67
5.39
6.56
6.80
7.92
9.07
10.00
13.33
13.60
16.00
18.00
20.00
COO- equivalents = (polymer concentration (g L-1)* guluronate molar fraction) / guluronate molar mass.
2.4. Volatile collection system
2.4.1. Description of the design
The collection system dedicated to the measure of volatile compounds was constituted of 0.5
L Teflon containers (Isoflon, Diemoz, France) containing slow-release formulations (100 mg
of semiochemical alginate beads). Each Teflon chamber was connected to a pump (Escort Elf
Air Sampling Pump, Sigma-Aldrich, Bornem, Belgium) programmed to pump air at a flow of
0.5 L min-1. Airflow produced by the pump was checked every day with an airflow meter. To
purify air, it entered the chamber through an activated charcoal filter cartridge. The volatile
compounds released by the alginate formulations were retained on traps (collection and
security traps). The traps consisted in 60 mg of HayeSep Q 80-100 mesh (Alltech, Lokeren,
Belgium) packed in Teflon tubing (4 mm I.D. x 1 mm wall thickness) between 2 pieces of
inox (AISI 304) wire cloth of 325 mesh (Haver Belgium S.A., Battice, Belgium) and 2 glass
tubing (4 mm O.D. x 0.8 mm wall thickness). Volatile collection was running 24 hours a day.
The system was only switched off during the change of trap cartridges every day (less than 5
min). All tubing connections were constructed of Teflon. To realise sampling everyday in the
same conditions, the volatile collection system was completely included in a thermally
139
Chapter VI.1
__________________________________________________________________________
controlled room (Maxi Artic Jouan, Vel, Louvain-la-Neuve, Belgium). Temperature was
programmed at 20 °C ± 1 °C. Temperature and relative humidity were continuously recorded
during all the duration of the experiments by means of a Hobo data logger (Miravox,
Hoevenen, Belgium).
After volatile collection, each trap (collection and security traps) was eluted 4 times with 250
µL of n-hexane. Twenty microlitres of internal standard (longifolene) at 1 µg µL-1 in nhexane were added at each elution sample before analysis and quantification by fast GC.
2.4.2. Efficiency evaluation of the design
In order to verify the quantitative efficiency of trap elutions, known quantities of E-βfarnesene and E-β-caryophyllene (in n-hexane) were directly injected onto HayeSep Q
cartridges at three levels (1 µg, 5 µg and 14 µg) in five replicates. A total solvent elution
volume of 750 µL was determined to be sufficient to elute with a satisfactory recovery the
highest quantity of compound. A safety margin was respected by considering a total solvent
volume of 1 mL for each trap elution. Elutions were made as previously described and the
quantification of compound in the extract was determined by internal standard quantification
and fast GC.
The recovery of the method was expressed as:
Recovery (%) = (quantity recovered in the extracts after elution / quantity injected onto the
cartridge) * 100
The recoveries were judged satisfactory when comprised between 85 % and 110 % (AOAC
norm, 2006). Maximum allowed values (RSD %) for repeatability were dependent of the
concentration and must be lower than 6 % in respect of the AOAC norm.
2.5. Quantification of semiochemicals and fast GC analyses
Optimisation of alginate bead formulation was depending on the optimal quantity of
semiochemicals incorporated in the beads according to the experimental conditions. This
quantification of semiochemicals formulated in alginate beads was realised as described in
Heuskin et al. (2010)
140
36
: solvent (n-pentane) extraction of compounds was followed by a
Chapter VI.1
__________________________________________________________________________
fractionation process (on silica gel) necessary to discard sunflower oil from the extract before
GC analysis.
Sesquiterpene analyses of volatile collection extracts and of bead extracts by fast GC were
validated according to the accuracy profile procedure
36
with longifolene as internal standard
for quantification.
Fast GC analyses were conducted on a Thermo Ultra Fast Trace GC gas chromatograph
associated to a split/splitless injector and coupled to a Thermo AS 3000 autosampler
(Thermo Electron Corp., Interscience, Louvain-la-Neuve, Belgium). The GC system was
equipped with an Ultra Fast Module (UFM) incorporating a direct resistively heated column
(Thermo Electron Corp.): UFC-5, 5% phenyl, 5 m x 0.1 mm I.D., 0.1 µm film thickness. The
chromatographic method was optimised for good resolution of terpenes (mono- and
sesquiterpenes) analyses
35
. Temperature programme for UFM was the following: initial
temperature at 40 °C, held for 0.1 min, ramp 1 at 30 °C min-1 to 95 °C, ramp 2 at 35 °C min-1
to 155 °C, ramp 3 at 200 °C min-1 to 280 °C, final hold of 0.5 min at 280 °C. Temperature of
injection: 240 °C. Volume injected: 1 µL. Carrier gas: He, at constant flow rate of 0.5 mL
min-1. Split ratio = 1:100. GC unit had a high-frequency fast flame ionization detector (300
Hz FID), at 250 °C. Flame gases: H2 flow at 35 mL min-1, air flow at 350 mL min-1, make up
gas (N2) flow at 30 mL min-1. Data were processed with Chromcard software (Version 2.3.3).
2.6. Texturometry of alginate beads
In order to determine the influence of the different formulation parameters on the texture of
the alginate polymer network, beads were subjected to texture measurements using a texture
analyser TA-XTPlus (Texture Technologies Corp., Surrey, UK) equipped with a 5 kg load
cell. The tests were conducted by means of a cylinder probe in Plexiglas P25L (25 mm
diameter). The measurements were done at room temperature on individual beads placed on
the lower platform and centrally positioned under the probe. The beads were compressed
until 90% of their initial volume at a velocity of 1 mm s-1. Data were processed with Texture
Exponent 32 software (Version 4.0.9.0).
141
Chapter VI.1
__________________________________________________________________________
2.7. Confocal laser scanning microscopy
Confocal microscopy for the characterisation of the polymer bead network was performed
using a LSM 710 laser scanning confocal microscope (Zeiss GmbH, Jena, Germany)
equipped with an Ar-laser and a diode used at 5% of the maximum power (25 mW). One
fluorescence channel was used in the red excitation line of 514 nm to visualise the oil phase
and another channel was used in the blue excitation line of 405 nm to visualise the alginate
polymer network. All confocal fluorescence images were taken with a Zeiss 10X objective.
Before confocal observation, alginate beads were colored during 10 minutes with fluorescent
marker FM4-64 (Invitrogen) in water. Complete and cut beads were observed under confocal
microscope. Optical cross-sections were taken at various depths of the beads to determine
semiochemical distribution in the polymeric network. Imaging of the beads was computed
with Zen 2009 software.
2.8. Olfactometry bioassay
Aphidius ervi parasitoids were reared on Sitobion avenae (F.) aphids feeding on wheat plants.
After emergence, adult parasitoids were placed in Petri dishes at 10°C  1°C with a 16L:8D
photoperiod and access to dilute honey solution. Female insects used for experiments were 34 days old. A glass Y-tube olfactometer was used to investigate the attraction of female
parasitoids to semiochemical odours released by alginate beads. The two arms of the tube
were connected to small glass chambers (4 cm diam.) containing alginate beads. Chamber
caps were sealed with Parafilm in order to have an airtight system. Air was pushed through
an activated charcoal filter at a rate of 50 mL min-1 into each glass chamber. In order to
guarantee homogeneous light, experiments were realised in a darkroom at 20°C  1°C with a
lamp (11 Watt) placed centrally above the olfactometer. Parasitoids being very sensitive to
light, the attraction could be biased if the light was not homogeneous.
Alginate beads (100 mg, 500 mg or 1.5 g) with (from 0.2 % to 2.0 %) and without
semiochemicals were deposited in one chamber of the olfactometer 4 h before experiments.
The other chamber was empty. At mid-experiment (after testing 15 parasitoids), the odour
source and the empty chambers were changed (interchange of olfactometer arms) in order to
avoid any bias towards one of the arms. Between each experiment, the arms of the
142
Chapter VI.1
__________________________________________________________________________
olfactometer were cleaned with n-hexane to eliminate residual odours adsorbed on glass.
They were dried at room temperature for 2 min.
For each experiment, thirty female parasitoids were individually introduced in the third arm
at the opposite side of the two test arms. In the centre of the olfactometer, where the three
arms were connected, females had choice between empty or semiochemical odour arms. The
test ended when the female entered one of both chambers or after 15 min of observation.
A 2 goodness-of-fit test was used to compare the distribution of the number of parasitoids in
both olfactometer arms, for each experiment (Minitab 15).
143
Chapter VI.1
__________________________________________________________________________
3. Results and discussion
Optimisation of alginate beads formulations was conducted by considering E-β-farnesene
encapsulation capacity as the major decisional factor. Parameters playing a role in the
alginate gelling process were modified in respect of two experimental designs (Tables 2 and
3) in order to determine their influence on the semiochemical quantity encapsulated in the
beads. After optimisation of the bead formulation, various tests were undertaken in order to
characterise this E-β-farnesene optimised formulation in terms of texture and polymeric
network arrangement. E-β-caryophyllene, another sesquiterpene molecule important in
biological control strategy against aphids, was then also formulated according to the same
optimised formulation process. A release rate study was conducted on both semiochemical
formulations (E-β-farnesene and E-β-caryophyllene) in laboratory controlled conditions.
Finally, optimised formulations were tested by olfactometry on female parasitoids in order to
determine attractiveness efficiency of alginate beads towards these insects.
3.1. Preliminary experiments for formulation optimisation
The first experimental design (Table 2) took into account the following parameters: alginate
type (Satialgine, Sigma L, Sigma M), alginate concentration (from 1 % to 2.5 % (w/v)), type
of cross-linker cation (Ca2+ or Cu2+) of the chloride solution, cation concentration (from 0.05
M to 0.3 M), extrusion needle diameter (from 0.4 mm to 1.6 mm) and maturation time of
beads in chloride solution (from 20 min to 48 h).
3.1.1. Determination of alginate type and cross-linker ion
Measurement of total E-β-farnesene (at 2% w/v in the mix before encapsulation process)
quantity encapsulated in 100 mg of alginate beads for each experimental test of Table 2
highlighted two major impacting parameters: type of alginate and type of cross-linker ion.
As shown in Figure 2 and Figure 3 representing the mean quantity of E-β-farnesene in 100
mg of alginate beads according to the type of alginate and the type of ion, respectively, the
optimal formulation was obtained for Sigma Low Viscosity (Sigma L) (mean of 1693.6 µg of
E-β-farnesene) and for calcium cross-linker ion (mean of 1499.3 µg of E-β-farnesene).
144
Chapter VI.1
__________________________________________________________________________
Figure 2: Encapsulation rate (in g) of E--farnesene in 100 mg of beads (sticks) and resistance force (in N)
(spots) according to 3 types of alginate (Satialgine, Sigma L, Sigma M).
Figure 3: Encapsulation rate (in g) of E--farnesene in 100 mg of beads (sticks) and resistance force (in N)
(spots) according to the type of cross-linker ion (Ca2+ or Cu2+).
3.1.2. Determination of the optimal ionic strength
A comparative study was performed on the slow-release rate of beads formulated at two
different ionic strengths of salt solutions, 0.5 M and 1.0 M. As shown in Figure 4, the
release rate of the 0.5 M ionic strength formulation (Figure 4a) was higher than for the 1.0
M ionic strength beads (Figure 4b). The cumulative released quantity of E-β-farnesene after
6 days was of 134.2 µg ± 5.1 µg (3 replicates) at 0.5 M and of 89.7 µg ± 0.5 µg (3 replicates)
at 1.0 M. This factor was also considered in the optimisation of the formulations.
145
Chapter VI.1
__________________________________________________________________________
Figure 4: Cumulative quantity (in g) of E--farnesene released from 100 mg of 0.5 M ionic strength alginate
(sigma L) beads (a) and from 100 mg of 1.0 M ionic strength alginate (sigma L) beads (b).
3.2. Optimisation of the alginate formulations
Based on the previously described results, the following parameters were chosen in such a
way they induced the optimal value in terms of sesquiterpene encapsulation: chloride
solution (Ca2+), type of alginate (Sigma L) and ionic strength (0.5M).
The optimisation of beads was completed by modifying the alginate and calcium chloride
concentrations as presented in Table 3.
The results of this second experimental design were expressed according to the ratio between
the calcium ions and the carboxylate ions from alginate (Ca2+/ COO-). This ratio plays an
important role in the gelling mechanism of the beads. Fang et al. (2007) 60 demonstrated that
when the ratio was higher than 0.55, polymer network was very dense and beads were
regular and homogeneous.
146
Chapter VI.1
__________________________________________________________________________
The relationship between the E-β-farnesene encapsulation rate (in µg) according to the Ca2+/
COO- ratio (Figure 5), with a maturation time of 20 min, increased linearly until a ratio of
7.92 where a plateau was then reached. To produce beads with a constant encapsulation rate,
the optimal formulation was chosen in the middle of the plateau, at a Ca2+/ COO- ratio of
13.3 (E-β-farnesene at 1567.9 µg ± 765.4 µg, n=4). Strong variations (SD) of the
encapsulation rates were observed even on the plateau. The formulation of beads being not
instantaneous, the beads underwent a phenomenon of syneresis which only stabilized after 48
hours of maturation in CaCl2 solution. The time of maturation was thus modified from 20
minutes to 48h (Figure 5). By increasing the time of maturation to 48 h, beads were more
homogeneous within the same batch and encapsulated more semiochemicals: at Ca2+/ COOratio of 13.3 for a maturation time of 48 h, the encapsulation rate increased to 3467.7 µg ±
878.6 µg (n= 3 ). The gel network tightened and prevented the ejection of the oilsemiochemical mixture during the later manipulations of the beads (as confirmed by the
measure of the oil mass (data not reported)).
Figure 5: Encapsulation rate (in g) of E--farnesene in 100 mg of alginate beads according to the Ca2+ / COOratio, for maturation times of 20 min. () and 48 h ().
The optimal chosen formulation had the following characteristics: Sigma Low viscosity; at
concentration of 1.5 %; Ca2+ 0.2M as cross-linker ion; Ionic strength: 0.5 M; Maturation
time: 48 h.
Repeatability of the formulation process was also estimated in terms of RSD (%). It was
determined by quantifying E-β-farnesene encapsulated in 100 mg of beads for 5 replicates.
147
Chapter VI.1
__________________________________________________________________________
RSD (%) of repeatability was of 2.6 %, value accepted according to the AOAC norm (2006)
which prescribed RSD of repeatability inferior to 4 %.
3.3. Characterisation of alginate beads
3.3.1. Bead texture analyses
Texturometry analyses were conducted in order to determine the impact of the various most
influencing formulation parameters on the elasticity of the semiochemical slow-release
alginate beads. As shown in Figure 6, the bead resistance increased with the alginate type in
the way Sigma L < Sigma M < Satialgine at different alginate concentrations. This
observation can be made in connection with the M/G ratio of the various alginates as
presented in Table 1. The G units conditioned the gelling process and their increasing
percentage (decreasing M/G ratio) increased the bead resistance. Figure 2 shows the mean
resistance force (in N) measured by texturometry in relation with the mean sesquiterpene
encapsulation capacity for the three tested alginates (Sigma L at 2.37 N, Sigma M at 3.07 N
and Satialgine at 3.78 N).
Figure 6: Resistance force (in N) of beads for different alginates (Sigma L, Sigma M, Satialgine) at various
concentrations (1.0 %; 1.5 %; 2.0 %; 2.5 %).
Type of ion was also an important parameter affecting the bead texture and their properties.
Indeed, as demonstrated in Figure 3, with the mean semiochemical encapsulated quantity,
the mean resistance force (in N) was higher for copper ion (3.59 N) than for calcium (2.60
N). This can be explained by differences existing in the cross-linking reactions between
calcium-alginate and copper-alginate networks. The affinity of alginate was higher for copper
148
Chapter VI.1
__________________________________________________________________________
ion than for calcium ion. As a result, different responses to deformation were observed with a
denser gel with Cu2+.
The resistance of the beads from the second experimental design (Sigma L, ionic strength:
0.5 M, ion: Ca2+) was finally analysed in function of the Ca2+ / COO- ratio at different
alginate concentrations (Figure 7). The resistance force of the beads increased according to
the alginate concentration (also demonstrated in Figure 6) and to the Ca2+/COO- ratio; the
network becoming denser. The best formulations in terms of bead shapes and their
homogeneity are circled in Figure 7.
Figure 7: Resistance force (in N) of Sigma L alginate beads according to the Ca 2+ / COO- ratio for different
alginate concentrations. Bold circle represents the best formulation in terms of appearance and homogeneity of
beads.
3.3.2. Polymeric network characterisation
Confocal laser scanning microscopy (CLSM) was used to characterise the structure of the
polymeric bead network (optimised formulation). This technique, to be effective in the
understanding of the semiochemical distribution in the beads, required fluorescent marker
molecules (FM4-64 mix) which attached preferentially to alginate polymer or to sunflower
oil macromolecules.
Figure 8a and Figure 8b show the CLSM imaging of a dried alginate bead (Aw: 0.42) and a
wet alginate bead (Aw: 0.98), respectively. In the case of the dried bead, oil-semiochemical
phase, in red, was distributed in the entire polymer network, in blue. The CLSM picture of
the wet bead was of poor quality to observe the semiochemical-oil arrangement in alginate
network. A thin water film disturbed the confocal microscopic observation. Considering
149
Chapter VI.1
__________________________________________________________________________
semiochemical slow-release rate measurements realised on dried and wet beads, it could be
assumed that the polymeric network was modified in wet beads in such a way that the
semiochemical release was stopped.
(a)
(b)
Figure 8: CLSM imaging of a dried (Aw=0.42) E--farnesene alginate bead (a) and of a wet (Aw=0.98) E-farnesene alginate bead (b).
3.4. Semiochemical slow-release measurement
After optimization of E-β-farnesene alginate beads as a biological control formulation,
another sesquiterpene, E-β-caryophyllene was encapsulated in optimized beads for its
interesting potential biological properties as aphid predator attractant.
Slow-release rate studies were conducted on both semiochemical formulations in laboratory
controlled conditions. These preliminary tests were realized in order to further
mathematically modelization of the semiochemical release kinetic according to abiotic
parameters like temperature, relative humidity and wind speed.
3.4.1. Evaluation of the volatile collection system performances
In order to obtain the most accurate and reliable measurements of semiochemicals released
by alginate beads over time, volatile collection system must encounter different
specifications, as described hereafter.
150
Chapter VI.1
__________________________________________________________________________
Teflon material was used for chambers and for all tubing used for the volatile collection
system to avoid adsorption of semiochemicals on wall surfaces, as it was reported with glass
61
and to have quantitative collection of released volatiles on adsorbent cartridges.
Verifications were realised by washing all Teflon parts with hexane after volatile collection.
No trace of semiochemical compounds was recovered as demonstrated by fast GC analysis.
Activated charcoal filters were used on each Teflon boxes to purify the air entering the
collection system. Blank sampling, without semiochemical beads in Teflon chambers, was
analysed in order to verify the efficiency of activated charcoal filters in retaining air
impurities.
n-Hexane was chosen as elution solvent for its apolar property ideal to desorb
sesquiterpenoids from HayeSep Q adsorbent. A volume of 750 µL of n-hexane was measured
to be sufficient to elute the maximum deposited quantity of the two investigated
semiochemicals. Nevertheless a security margin of 250 µL (total volume of 1 mL) was
considered. Recoveries of elution were determined by eluting known quantities (1 µg, 5 µg
and 14 µg) of semiochemical compounds directly deposited on the adsorbent cartridge.
Average recoveries were judges satisfactory in both cases (E-β-farnesene and E-βcaryophyllene) being comprised between 87.1 % and 91.2 % for the three quantities, with
relative standard deviations of repeatability inferior to 6 % (from 3.5 % to 3.9 %).
In order to avoid loss of compounds by breakthrough of the volatile collection cartridge
during the 24 h of sampling, a security cartridge was added in series. The elution step of this
second cartridge was realised in the same manner as described above.
3.4.2. Semiochemical release rate measurements
Release rate study was conducted during 35 days in laboratory controlled conditions (20 °C,
65% of relative humidity, airflow at 0.5 L min-1) on 100 mg of E-β-farnesene and 100 mg of
E-β-caryophyllene alginate beads. Figure 9a and Figure 9b show the cumulative released
quantity of semiochemicals from beads over time. After 35 days, 503.4 µg of E-β-farnesene
were released, representing 36.9 % of the total quantity of semiochemical encapsulated in
100 mg beads. For E-β-caryophyllene, 1790.9 µg were released after 35 days, representing
46.8 % of the total quantity of encapsulated compound. The daily release rates were
151
Chapter VI.1
__________________________________________________________________________
comprised between 9.1 µg and 32.6 µg for E-β-farnesene and between 16.1 µg and 99.5 µg
for E-β-caryophyllene.
These results showed that optimised semiochemical alginate beads were efficient slowrelease devices. The speed of release was very different between both compounds. This
phenomenon could be studied more in details to determine whether the conformation of each
compound could have an influence.
Diffusion kinetics could be influenced by abiotic parameters like temperature, relative
humidity or wind speed. For this reason, further study on semiochemical release from beads
in real environmental conditions is required in order to optimize the utilisation of these
formulations in agriculture as biological control devices.
Figure 9: Cumulative released quantity (in g) of E--farnesene (a) and of E--caryophyllene (b) from 100 mg
of optimised formulation during 35 days in laboratory controlled conditions (20°C; 65% relative humidity;
airflow at 0.5 L min-1).
152
Chapter VI.1
__________________________________________________________________________
3.5. Olfactometry bioassay
The main goal of the present bioassay consisted in determining whether 3-4 day old female
parasitoids were attracted by E--farnesene or E--caryophyllene odours released from
alginate gel bead formulations. Therefore, the optimal dose of semiochemical formulation
was determined.
The first step of the research was dedicated to demonstrate that alginate beads without
semiochemical were not attractive towards A. ervi. Y-tube olfactometer experiments were
conducted with 500 mg of alginate beads without sesquiterpene as control in one arm and
with a blank (no odour) in the other arm of the device. The percentages of insects in both
arms were of 40.0 % and 60.0 % for control and blank, respectively. 2 test gave a p-value of
0.0007 (<0.001) which meant that their existed a significative difference between the 2 ways
of the olfactometer, alginate beads as such (without semiochemical) being not attractive for
parasitoids. Preliminary tests were then conducted on 100 mg of beads at 2% w/v (before
encapsulation) of semiochemical. Nether for E--farnesene, nor for E--caryophyllene,
females made a choice significantly different of the blank (no odour). Indeed, p-values
obtained by 2 tests were of 0.1938 and 0.1460 (largely higher than 0.05) for E--farnesene
and E--caryophyllene, respectively. It was therefore assumed that the quantity of
semiochemical released during the experiment (approximately 0.35 g/15 min and 1.03
g/15 min for E--farnesene and E--caryophyllene, respectively) was not sufficient to be
attractive for parasitoids.
A new experiment was then carried out by increasing the quantity of semiochemicals
released from the alginate beads. Five hundred mg of beads formulated with 20% w/v of
compounds (in the mix before encapsulation process) were tested. These quantities were
adapted in order to obtain a release rate comprised between 10 g and 100 g during 15 min
of experiment, as it was related by Du et al. (1998)
17
and by Foster et al. (2005) 2. In this
experiment, females of A. ervi made rapidly a choice between both ways of the olfactometer.
Figure 10 represents the percentage of females in each arm of the olfactometer for E-farnesene and E--caryophyllene compared to the blank. For both tested molecules, the arm
with semiochemical odour was preferred by females compared to the other arm. Indeed,
66.7% and 90.0% of females moved towards E--farnesene and E--caryophyllene odour
source, respectively. A 2 test gave p-values of 0.0001 (< 0.001) in both cases, which meant
153
Chapter VI.1
__________________________________________________________________________
that parasitoids were attracted by semiochemical alginate beads with a very highly
significative difference compared to blank. In terms of percentages of insects attracted by
semiochemicals, a difference between E--farnesene and E--caryophyllene was noticed.
This phenomenon could be explained by the difference of release rate between both
compounds which was observed in laboratory controlled slow-release measurements. Indeed,
E--caryophyllene formulation had a release rate of semiochemical almost 3 times higher
than E--farnesene alginate beads.
The olfactometry experiment was adapted by increasing the quantity of E--farnesene beads
to 1.5 g instead of 500 mg. The percentage of attraction increased to 88 % (Figure 10) for
this compound (p-value < 0.001) and became almost similar to the percentage obtained with
E--caryophyllene.
It is important to note that all tested parasitoids made a choice between arms of the
olfactometer.
Figure 10: Percentages of female A. ervi in both arms of Y-tube olfactometer, for 500 mg E--caryophyllene
alginate beads, 500 mg and 1500 mg E--farnesene alginate beads.
The results obtained by Y-tube olfactometry for semiochemical slow-release alginate beads
were in agreement with the conclusions of previous studies showing the attractiveness of E-
-farnesene
154
2,17
and of E--caryophyllene
22
towards A. ervi female parasitoids. In these
Chapter VI.1
__________________________________________________________________________
previous studies, synthetic standard components dissolved in solvent were tested by
olfactometry or wind tunnel bioassays on parasitoid wasps. In the present work, some
experimental differences must be shown. Semiochemical compounds were not from synthetic
origin but were purified from natural plant matrices: essential oils of Matricaria chamomilla
and Nepeta cataria. Secondly, alginate formulations guaranteed a slow-release of compounds
during the complete duration of the experiment at a rate comprised between 10 g and 100
g per 15 min. Finally, no solvent was used in the experiments.
The high percentages of response (88 % and 90 % for E--farnesene and E--caryophyllene,
respectively) of parasitoids to semiochemical odors in the Y-tube olfactometer indicated that
attractiveness is not only dependent of visual cues like it was often assumed in previous
works 62-64 but odours are of crucial importance.
In conclusion, the results obtained by olfactometry on female A. ervi show that alginate beads
are efficient semiochemical slow-release systems to attract aphid natural enemies in a
laboratory arena. However, in further studies, it could be interesting to test the efficiency of
such devices in greenhouses and finally in field experiments where climatic conditions could
influence semiochemical release rate from alginate beads.
Aknowledgements
This research was funded by a Walloon Region Ministry grant (WALEO2: SOLAPHIDRW/FUSAGX 061/6287).
The authors are grateful to Dr. S. Bartram and Prof. W. Boland from the Max Planck Institute
for Chemical Ecology (Jena, Germany) for providing E-β-farnesene from chemical synthesis.
We would like to thank Cécile Le Duff from the Institute of Condensed Matter and
Nanosciences, Catholic University of Louvain (Louvain-la-Neuve, Belgium) for her precious
help in NMR analyses. Thanks also to Abdelmounaim Errachid from the Institute of Life
Sciences, Catholic University of Louvain (Louvain-la-Neuve, Belgium) for confocal laser
scanning microscopy analyses.
155
Chapter VI.1
___________________________________________________________________________
References
[1] Hooks C.R.R., Fereres A., Protecting crops from non-persistently aphid-transmitted
viruses: A review on the use of barrier plants as a management tool. Virus Res. 120: 1–
16 (2006).
[2] Foster S.P., Denholm I., Thompson R., Poppy G.M., Powell W., Reduced response of
insecticide-resistant aphids and attraction of parasitoids to aphid alarm pheromone; a
potential fitness trade-off. Bull. Entomol. Res. 95: 37-46 (2005).
[3] Kogan M., Integrated pest management: historical perspectives and contemporary
developments. Annu. Rev. Entomol. 43: 243-270 (1998).
[4] Phillips T.W., Semiochemicals of stored-product insects: research and applications. J.
stored Prod. Res., 33: 17-30 (1997).
[5] Torr S.J., Hall D.R., Phelps R.J., Vale G.A., Methods for dispensing odour attractants for
tsetse flies (Diptera: Glossinidae). Bull. Entomol. Res., 87: 299-311 (1997).
[6] Rochat D., Morin J.-P., Kakul T., Beaudoin-Olivier L., Prior R., Renou M., Malosse I.,
Stathers T., Embupa S., Laup S., Activity of male pheromone of Melanesian rhinoceros
beetle Scapanes australis. J. Chem. Ecol. 28: 479-500 (2002).
[7] Stipanovic A.J., Hennessy P.J., Webster F.X., Takahashi Y., Microparticle dispensers for
the controlled release of insect pheromones. J. Agric. Food Chem. 52: 2301-2308
(2004).
[8] Weinzierl R., Henn T., Koehler P.G., Tucker C.L., Insect attractants and traps. University
of Florida, IFAS Extension, http://edis.ifas.ufl.edu/in080, (05/05/10).
[9] Zhang A., Kuang L.F., Maisin N., Karumuru B., Hall D.R., Virdiana I., Lambert S.,
Purung H.B., Wang S., Hebbar P., Activity Evaluation of cocoa pod borer sex
pheromone in cacao fields. Environ. Entomol, 37: 719-724 (2008).
[10] Shem P.M., Shiundu P.M., Gikonyo N.K., Ali A.H., Saini R.K., Release kinetics of a
synthetic tsetse allomone derived from waterbuck odour from a Tygon silicon dispenser
under laboratory and semi field conditions. Amer.-Euras. J. Agric. & Environ. Sci., 6:
625-636 (2009).
156
Chapter VI.1
___________________________________________________________________________
[11] Stoner, K., Approaches to the biological control of insects. University of Maine,
Cooperative Extension Bull. 7144 (2004),
www.umext.maine.edu/onlinepubs/PDFpubs/7144.pdf (06/07/10).
[12] Francis F., Vandermoten S., Verheggen F., Lognay G., Haubruge E., Is the (E)-ßfarnesene only volatile terpenoids in aphids? J. Applied Ento. 129: 6-11 (2005).
[13] Francis F., Martin T., Lognay G., Haubruge E., Role of (E)-β-farnesene in systematic
aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae), Eur. J.
Entomol., 102: 431-436 (2005).
[14] Verheggen F.J., Arnaud L., Bartram S., Gohy M., Haubruge E., Aphid and plant
volatiles induce oviposition in an aphidophagous Hoverfly. J. Chem. Ecol. 34: 301-307
(2008).
[15] Verheggen F.J., Haubruge E., De Moraes C.M., Mescher M.C., Social environment
influences aphid production of alarm pheromone. Behav. Ecol. 20: 283-288 (2009).
[16] Verheggen F.J., Haubruge E., Mescher M.C., Alarm pheromones. In:Litwack G., ed.,
Pheromones. Elsevier, (2010).
[17] Du Y., Poppy G.M., Powell W., Pickett J.A., Wadhams L.J., Woodcock C.M.,
Identification of Semiochemicals Released During Aphid Feeding That Attract
Parasitoid Aphidius ervi. J. Chem. Ecol. 24: 1355-1368 (1998).
[18] Powell W., Pickett J.A., Manipulation of parasitoids for aphid pest management:
progress and prospects. Pest Manag. Sci. 59: 149-155 (2003).
[19] Brown A.E., Riddick E.W., Aldrich J.R., Holmes W.E., Identification of (−)-βCaryophyllene as a Gender-Specific Terpene Produced by the Multicolored Asian Lady
Beetle. J. Chem. Ecol. 32: 2489-2499 (2006).
[20] Verheggen F.J., Fagel Q., Heuskin S., Lognay G., Francis F., Haubruge E.,
Electrophysiological and behavioral responses of the multicolored asian lady beetle,
Harmonia axyridis Pallas, to sesquiterpene semiochemicals. J. Chem. Ecol. 33: 21482155 (2007).
[21] Tomova B.S., Waterhouse J.S., Boberski J., The effect of fractionated Tagetes oil
volatiles on aphid reproduction. Entomol. Exp. Appl. 115: 153-159 (2005).
157
Chapter VI.1
___________________________________________________________________________
[22] Sasso, R., Iodice, L., Woodcock, C.M., Pickett, J.A., Guerrieri, E., Electrophysiological
and behavioural responses of Aphidius ervi (Hymenoptera: Braconidae) to tomato plant
volatiles. Chemoeco. 19: 195-201 (2009).
[23] Gonzalez, D., White, W., Hall, J., Dickson, R.C., Geographical distribution of
Aphidiidae (Hym.) imported to California for biological control of Acyrthosiphon
kondoi and Acyrthosiphon pisum (Hom.: Aphididae). Entomophaga 23: 239-248
(1978).
[24] Stary, P., The fate of released parasitoids (Hymenoptera, Braconidae, Aphidiinae) for
biological control of aphids in Chile. Bull. Entomol. Res. 83: 633-639 (1995).
[25] Michaud, J.P., The oviposition behavior of Aphidius ervi and Monoctonus paulensis
(Hymenoptera: Aphididae) encountering different host species (Homoptera: Aphididae)
in sequential patches. J. Insect Behav. 9: 683-694 (1996).
[26] Michaud, J.P., Sloderbeck, P.E., Nechols J.R.,. Biological control of insect pests on field
crops in Kansas. Kansas State University Agricultural Experiment Station and
Cooperative Extension Service (2008), http://www.oznet.ksu.edu (06/07/10).
[27] Libbrecht, R., Gwynn, D.M., Fellowes, M.D.E., Aphidius ervi preferentially attacks the
green morph of the pea aphid, Acyrthosiphon pisum. J. Insect Behav. 20: 25-32 (2007).
[28] Du Y., Poppy, G.M. Powell, W., Relative importance of semiochemicals from first and
second trophic levels in host foraging behavior of Aphidius ervi. J. Chem. Ecol. 22:
1591-1605 (1996).
[29] Du Y., Poppy G.M., Powell W., Wadhams L.J., Chemically mediated associative
learning in the host foraging behavior of the aphid parasitoid Aphidius ervi
(Hymenoptera: Braconidae). J. Insect Behav. 10: 509-522 (1997).
[30] Glinwood R.T., Du Y.-J., Smiley D.W.M., Powell W., Comparative responses of
parasitoids to synthetic and plant-extracted nepetalactone component of aphid sex
pheromones. J. Chem. Ecol. 25: 1481-1488 (1999).
[31] Guerrieri E., Poppy G.M., Powell W., Tremblay E., Pennacchio F., Induction and
systematic release of herbivore-induced plant volatiles mediating in-flight orientation of
Aphidius ervi. J. Chem. Ecol. 25: 1247-1261 (1999).
158
Chapter VI.1
___________________________________________________________________________
[32] Guerrieri E., Poppy G.M., Powell W., Rao R., Pennacchio F., Plant-to-plant
communication mediating in-flight orientation of Aphidius ervi. J. Chem. Ecol. 28:
1703-1715 (2002).
[33] Schwörer, U., Völkl, W., Foraging behavior of Aphidius ervi (Haliday) (Hymenoptera:
Braconidae: Aphidiinae) at different spatial scales: Resource utilization and suboptimal
weather conditions. Biologic. Contr. 21: 111-119 (2001).
[34] Stilmant, D., Van Bellinghen, C., Hance, T., Boivin, G., Host specialization in habitat
specialists and generalists. Oecologia 156: 905-912 (2008).
[35] Heuskin S., Godin B., Leroy P., Capella Q., Wathelet J.-P., Verheggen F., Haubruge E.,
Lognay G., Fast gas chromatography characterization of purified semiochemicals from
essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L.
(Lamiaceae). J. Chrom. A. 1216: 2768-2775 (2009).
[36] Heuskin S., Rozet E., Lorge S., Farmakidis J., Hubert Ph., Verheggen F., Haubruge E.,
Wathelet J.-P., Lognay G., Validation of a fast gas chromatographic method for the
study of semiochemical slow release formulations. J. Pharma. Biomed. Anal. 53: 962972 (2010).
[37] Tkachev A. V., The chemistry of caryophyllene and related compounds. Chem. Nat.
Comp. 23: 393-412 (1987).
[38] Sköld M., Karlberg A.-T., Matura M., Börje A., The fragrance chemical βcaryophyllene—air oxidation and skin sensitization. Food Chem. Toxic. 44: 538-545
(2006).
[39] Parshintsev J., Nurmi J., Kilpeläinen I., Hartonen K., Kulmala M., Riekkola M.L.,
Preparation of beta-caryophyllene oxidation products and their determination in
ambient aerosol samples. Anal. Bioanal. Chem. 390: 913-919 (2008).
[40] Kourtchev I., Bejan I., Sodeau J.R., Wenger J.C., Gas-Phase Products and Secondary
Aerosol Yields from the Ozonolysis of (E)-β-farnesene. European Aerosol Conference,
Karlsruhe, Abstract T021A01 (2009).
[41] Rojas-Graü M.A., Tapia M.S., Rodriguez F.J., Carmona A.J., Martin-Belloso O.,
Alginate and gellan-based edible coatings as carriers of antibrowning agents applied on
fresh-cut Fuji apples. Food Hydrocoll. 21: 118-127 (2007).
159
Chapter VI.1
___________________________________________________________________________
[42] Grant G.T., Morris E.R., Rees D.A., Smith P.J.C., Thom D., Biological interactions
between polysaccharides and divalent cations. The egg-box model. FEBS Letters., 32:
195-198 (1973).
[43] Ingar D.K., Gudmund S.-B., Olav S., Alginate based new materials. Int. J. Biol.
Macromolec. 21: 47-55 (1997).
[44] Braccini L., Grasso R.P., Perez S., Conformational and configurational features of acidic
polysaccharides and their interactions with calcium ions: molecular modeling
investigation. Carbohydrate Res. 317: 119-130 (1999).
[45] Sankalia M.G., Mashru R.C., Sankalia J.M., Sutariya V.B., Papain entrapment in
alginate beads for stability improvement and site-specific delivery: Physicochemical
characterization and factorial optimization using neural network modeling. AAPS
PharmSciTech. 6: E209-E222 (2005).
[46] DeGhorain B., Contribution à l’étude de l’encapsulation et du relargage contrôlé de
molécules aromatiques par des billes d’alginate de calcium. Master thesis, UCL,
Belgium. (1997).
[47] Malone M.E., Appelqvist I.A.M., Gelled emulsion particles for the controlled release of
lipophilic volatiles during eating. J. Contr. Release. 90: 227-241 (2003).
[48] Madene A., Jacquot M., Scher J., Desobry S., Flavour encapsulation and controlled
release – a review. Int. J. Food Sci. Tech. 41: 1-21 (2006).
[49] Chang C.P., Dobashi T., Preparation of alginate complex capsules containing eucalyptus
essential oil and its controlled release. Colloids and Surf. B. 32: 257-262 (2003).
[50] Lai F., Loy G., Manconi M., Manca M.L., Fadda A.M., Artemesia arborescens L.
essential oil loaded beads: preparation and characterization. AAPS PharmSciTech. 8:
E1-E7 (2007).
[51] Tonnesen H.H., Karlsen J., Alginate in drug delivery systems. Drug Dev. Indust. Pharm.
28: 621-630 (2002).
[52] George M., Abraham T.E., Polyonic hydrocolloids for the intestinal delivery of protein
drugs: alginate and chitosan – a review. J. Contr. Rel. 114: 1-14 (2006).
[53] Smidsrod O., Grasdalen H., Polyelectrolytes from seeweeds. Hydrobiologia 116: 19-28
(1984).
160
Chapter VI.1
___________________________________________________________________________
[54] Ouwerx C., Velings N., Mestdagh M.M., Axelos M.A.V., Physico-chemical properties
and rheology of alginate gel beads formed with various divalent cations. Polym. Gels
Net. 6: 393-408 (1998).
[55] Determan A.S., Trewyn B.G., Lin V. S-Y., Nilsen-Hamilton M., Narasimhan B.,
Encapsulation, stabilization, and release of BSA-FITC
from
polyanhydride
microspheres. J. Contr. Release. 100: 97-109 (2004).
[56] Lamprecht A., Schäfer U.F., Lehr C-M., Vizualisation and quantification of polymer
distribution in microcapsules by confocal laser scanning microscopy (CLSM). Int.
Journal Pharm. 196: 223-226 (2000).
[57] Berkland C., Kipper M. J., Narasimhan B., Kim K., Pack D. W., Microsphere size,
precipitation kinetics and drug distribution control drug release from biodegradable
polyanhydride microspheres. J. Contr. Release. 94: 129-141 (2004).
[58] Velings N., Mestdagh M.M., Physico-Chemical Properties of Alginate Gel Beads.
Polymer Gels Networks. 3: 311-330 (1995).
[59] Grasdalen H., Larsen B., Smidsrod O., A P.M.R. Study of the Composition and
Sequence of Uronate Residues in Alginates. Carbohydrate Res. 68: 23-31 (1979).
[60] Fang Y., Al-Assaf S., Phillips G.O., Nishinari K., Funami T., Williams P.A., Multiple
steps and critical behaviors of the binding of calcium to alginate. J. Physical Chem. B.
111: 2456-2462 (2007).
[61] Cross J.H., A vapour collection and thermal desorption method to measure
semiochemical release rates from controlled release formulations. J. Chem. Ecol. 6:
781-787 (1980).
[62] Michaud J.P., Mackauer M., The use of visual cues in host evaluation by aphidiid
wasps. I. Comparison between three Aphidius parasitoids of the pea aphid. Ent. Exp.
Appl. 70: 273-283 (1994).
[63] Battaglia D., Pennacchio F., Romano A., Tranfaglia A., The role of physical cues in the
regulation of host recognition and acceptance behavior of Aphidius ervi Haliday
(Hymenoptera: Braconidae). J. Insect Behav. 8: 739-750 (1995).
[64] Battaglia D., Poppy G., Powell W., Romano A., Tranfaglia A., Pennacchio F., Physical
and chemical cues influencing the oviposition behaviour of Aphidius ervi. Ent. Exp.
Appl. 94: 219-227 (2000).
161
Chapter VI.2
Chapter VI.2
___________________________________________________________________________
Objectives
The semiochemical alginate beads being optimised, the second part of chapter VI contributed
to more understand the diffusion behaviour of semiochemicals from alginate beads according
to the variation of abiotic factors such as temperature, relative humidity and air flow rate.
Diffusion coefficients were estimated for various experimental conditions as presented in
Figure 1.
The attraction efficiency of semiochemical formulations was then evaluated on field
experiments by catching aphid predators, Syrphidae species. The experimental conditions are
presented in Figure 1’ with a summary of the main results
165
Figure 1’: Different steps of the semiochemical slow-release study according to abiotic factors and of the field bioassays
Chapter VI.2
___________________________________________________________________________
VI. 2
A semiochemical slow-release formulation in a biological control
approach to attract hoverflies
Stéphanie Heuskin1,2, François Béra3, Stéphanie Lorge4, Aise Kilinc1, Pascal Leroy5, Eric
Haubruge5, Jean-Paul Wathelet2, Yves Brostaux6, Georges Lognay1
1
Department of Analytical Chemistry, Gembloux Agro-Bio Tech, University of Liege, Passage des
Déportés 2, B-5030 Gembloux (Belgium).
2
Department of General and Organic Chemistry, Gembloux Agro-Bio Tech, University of Liege,
Passage des Déportés 2, B-5030 Gembloux (Belgium).
3
Department of Food Processing Industry, Gembloux Agro-Bio Tech, University of Liege, Passage
des Déportés 2, B-5030 Gembloux (Belgium).
4
Institute of Condensed Matter and Nanosciences, Catholic University of Louvain, Croix du Sud 2, B1348 Louvain-la-Neuve (Belgium).
5
Department of Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of
Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
6
Department of Statistics, Mathematics and Applied Computer Sciences, Gembloux Agro-Bio Tech,
University of Liege, Passage des Déportés 2, B-5030 Gembloux (Belgium).
Reference: Submitted to Entomologia Experimentalis et Applicata journal.
167
Chapter VI.2
___________________________________________________________________________
Abstract
E-β-Farnesene, the alarm pheromone of many aphid species, and E-β-caryophyllene, recently
identified as one of the possible component of the aggregation pheromone of Harmonia
axyridis Pallas, are considered as two sesquiterpenes attractive for aphid predators, among
which Syrphidae species.
In the present research, both compounds were formulated in alginate gel beads as slowrelease devices in a biological control approach towards aphids. Semiochemical diffusion
from beads was studied in the laboratory according to abiotic parameters. Efficiency of
formulations as hoverfly attractant was demonstrated in field experiments from June to
August 2009.
Keywords
Pheromones, sesquiterpenes, Syrphidae, aphids, alginate, controlled-release
168
Chapter VI.2
___________________________________________________________________________
1. Introduction
Biological control is an integrated pest management (IPM) strategy which consists in using
living organisms (insect predators or parasitoids, pathogens) to suppress pest populations,
making them less damaging that they would otherwise be (Stoner, 2004). IPM programs are
increasing since the emergence of environmental side effects and risks for human health with
the use of synthetic pesticides. Furthermore, development of pest resistance towards
synthetic pesticides was related (Foster et al., 2005) as well as no species-specificity of
pesticides with lethal effects on beneficial insects (Moens et al., 2010).
Hoverfly larvae (Diptera: Syrphidae) are considered since many years as one of the most
important predator of aphids. Already in 1985 and 1986, Chambers et al. reported the
economical importance of hoverfly species as aphidophagous biological control agents.
Other studies highlighted the role of volatile compounds – coming from aphid-damaged
plants and from aphids themselves – as semiochemical cues in the foraging process of female
Syrphidae (Almohamad et al., 2006; Harmel et al., 2007). Among these molecules, E-farnesene, the alarm pheromone of many aphid species (Francis et al., 2005a), was largely
implicated in oviposition induction of Syrphidae (Francis et al., 2005b; Verheggen et al.,
2008, 2009, 2010).
In the current study, E--farnesene and E--caryophyllene, another sesquiterpene compound
having interesting biological properties – aggregation pheromone of the Asian ladybeetle,
Harmonia axyridis Pallas (Brown et al., 2006; Verheggen et al., 2007) and reducer of aphid
reproduction (Tomova et al., 2005) – were formulated in slow-release devices. The efficiency
of such devices was demonstrated in terms of semiochemical slow-release according to
abiotic factors and in terms of attraction of female Syrphidae during field-trials.
In respect with biological control approaches, environmentally safe materials were
considered. First of all, volatile organic compounds were extracted and purified from
essential oils of Matricaria chamomilla L. (Asteraceae) and Nepeta cataria L. (Lamiaceae)
for E--farnesene and E--caryophyllene, respectively (Heuskin et al., 2009, 2010).
Secondly, alginate biodegradable matrix was chosen to formulate semiochemicals. This
polysaccharide having low oxygen permeability can protect the sesquiterpenes from
oxidation. The protection efficiency of alginate beads towards both molecules was
demonstrated in Heuskin et al. (2010).
169
Chapter VI.2
___________________________________________________________________________
2. Material and methods
2.1. Chemicals and reagents
Essential oil of Matricaria chamomilla L. was purchased from Vossen & Co. (Brussels,
Belgium) and was originated from Nepal (lot no. CHA06MI0406). Essential oil of Nepeta
cataria L. was purchased from Essential7.com (Roswell, NM, USA) and was originated from
Canada (lot no. EO0020f).
E-β-Farnesene (83.8 % ± 0.3) and E-β-caryophyllene (97.7 % ± 0.5 %) from natural origin
were extracted by flash chromatography from essential oils of M. chamomilla and N. cataria,
respectively as reported by Heuskin et al. (2010). (+)-Longifolene (GLC purity > 99.9%), as
internal standard for validation and quantification steps, was purchased from ABCR
(Karlsruhe, Germany). The mean purities with standard deviation (ten replicates) of the
sesquiterpenes were determined by fast gas chromatography.
n-Hexane of GC grade was purchased from VWR (Leuven, Belgium). n-Pentane extra pure
was purchased from Acros Organics (Geel, Belgium).
Sodium alginate and α-Tocopherol used in bead formulations were purchased from SigmaAldrich (Bornem, Belgium). Calcium chloride was purchased from Acros Organics (Geel,
Belgium).
2.2. Semiochemical purification
Adsorption chromatography (Flash chromatography assembly, Sigma-Aldrich, Bornem,
Belgium) was used to obtain purified semiochemicals. Ten millilitres of essential oil (9.31 g
for Matricaria chamomilla and 9.53 g for Nepeta cataria) were fractionated under N2
pressure (0.5 bar) over 110 g silica gel G60 70-230 mesh (Macherey-Nagel) previously dried
16 h at 120 °C and packed in a glass column (35 mm I.D.) (silica gel bed was 34 cm high).
Essential oil of M. chamomilla was eluted with 850 ml n-pentane to yield 3 fractions of 250
ml (F1), 200 ml (F2) and 400 ml (F3). Essential oil of N. cataria was eluted with 600 ml npentane to obtain 2 fractions of 250 ml (F1) and 350 ml (F2). Solvent-free purified extracts
were obtained after solvent evaporation at atmospheric pressure and at 40 °C with a Büchi
rotatory evaporator. Solvent-free fractions were analysed by fast GC (dilution in n-hexane).
170
Chapter VI.2
___________________________________________________________________________
(Heuskin et al., 2009, 2010). Fractions F3 from M. chamomilla and F2 from N. cataria were
the richest in E--farnesene (83.8 %  0.3 % E--farnesene; 11.6 %  0.2 % (E,E)-farnesene; 1.8 %  0.1 % germacrene D; 0.9 %  0.1 % bicyclogermacrene) and E-caryophyllene (97.7 %  0.5 % E--caryophyllene; 2.1 %  0.3 % -humulene),
respectively.
2.3. Alginate bead formulation
Alginate bead formulation was optimised in a previous study (submitted in Pest Management
Science), in terms of semiochemical formulation efficiency. The experimental conditions are
described hereafter. Sodium alginate solution was prepared in distilled water at 1.5 % w/v.
Calcium chloride solution was prepared in distilled water at a concentration of 0.2M in such
a way the ionic strength of this CaCl2 solution was 0.5 M.
Eight mL of sodium alginate solution added to 1.8 mL sunflower oil, 150 mg α-tocopherol as
antioxidant and 0.2 g (2 % w/v) E-β-caryophyllene or 0.6 g (6 % w/v) E-β-farnesene (from
essential oil fractionations) were mixed with an ultraturax system (IKA T18 Basic, QLab,
Vilvoorde, Belgium) at 24000 rpm during 20 seconds to obtain a thin and homogeneous
emulsion. The emulsion was extruded by needle (0.4 mm I.D.) and droplets fell into agitated
(magnetic stir bar at 600 rpm) CaCl2 solution to form the alginate gel beads containing
semiochemical components. The distance between needle and chloride solution was fixed at
20 cm to obtain spherical beads. The beads stayed 48 hours in the ionic solution to stabilise
the syneresis phenomenon. In order to eliminate surface water, beads were dried before use.
They were first drained off on a filter paper during a few seconds. Then they were dried
under air pressure at 2 bars during 30 minutes at 21°C ± 2°C.
2.4. Slow-release measurement
The collection system dedicated to the measure of volatile compounds consisted in double
glass funnel-shaped devices containing slow-release formulations (200 mg semiochemical
alginate beads) deposited side by side on a sintered surface. An air generator pulled the air to
diffuse in a bottle with supersaturated saline solution. Saline solutions were used to fix
relative humidity (RH) at different values (distilled water: RH at 100%; BaCl2: RH at 90%;
KCl: RH at 85%; NaCl: RH at 75%; CH3COOK: RH at 25%) in the sampling system.
171
Chapter VI.2
___________________________________________________________________________
Airflow adjusted from 0.05 L/min to 1.00 L/min was checked every day with an airflow
meter.
The volatile compounds released by the alginate formulations were retained on two traps
(collection and security traps). The traps consisted in 60 mg of HayeSep Q 80-100 mesh
(Alltech, Lokeren, Belgium) packed in Teflon tubing (4 mm I.D. x 1 mm wall thickness)
between 2 pieces of inox (AISI 304) wire cloth of 325 mesh (Haver Belgium S.A., Battice,
Belgium) and 2 glass tubing (4 mm O.D. x 0.8 mm wall thickness). Volatile collection was
running 24 hours a day. During that period no breakthrough of the tested molecules was
observed. The system was only switched off during the change of trap cartridges every day
(less than 5 min). All tubing connections were made in Teflon. The volatile collection system
was completely included in a thermally controlled room (Maxi Artic Jouan, Vel, Louvain-laNeuve, Belgium) programmed at 20 °C or 40 °C (± 1 °C). Temperature and relative humidity
were continuously recorded during all the duration of the experiments by means of a Hobo
data logger (Miravox, Hoevenen, Belgium).
After volatile collection, each trap (collection and security traps) was eluted 4 times with 250
µL n-hexane. Twenty microlitres of internal standard (longifolene) at 1 µg µL-1 in n-hexane
were added at each elution sample before analysis and quantification by fast GC.
2.5. Fast GC analyses
Sesquiterpene analyses and quantification of volatile collection extracts by fast GC were
validated according to the accuracy profile procedure described in Heuskin et al. (2010) with
longifolene as internal standard for quantification.
Fast GC analyses were conducted on a Thermo Ultra Fast Trace GC gas chromatograph
associated to a split/splitless injector and coupled to a Thermo AS 3000 autosampler
(Thermo Electron Corp., Interscience, Louvain-la-Neuve, Belgium). The GC system was
equipped with an Ultra Fast Module (UFM) incorporating a direct resistively heated column
(Thermo Electron Corp.): UFC-5, 5% phenyl, 5 m x 0.1 mm I.D., 0.1 µm film thickness. The
chromatographic method was optimised for good resolution of terpenes (mono- and
sesquiterpenes) analyses (Heuskin et al., 2009). Temperature programme for UFM was the
following: initial temperature at 40 °C, held for 0.1 min, ramp 1 at 30 °C min -1 to 95 °C,
ramp 2 at 35 °C min-1 to 155 °C, ramp 3 at 200 °C min-1 to 280 °C, final hold of 0.5 min at
172
Chapter VI.2
___________________________________________________________________________
280 °C. Temperature of injection: 240 °C. Volume injected: 1 µL. Carrier gas: He, at
constant flow rate of 0.5 mL min-1. Split ratio = 1:100. GC unit had a high-frequency fast
flame ionization detector (300 Hz FID), at 250 °C. Flame gases: H2 flow at 35 mL min-1, air
flow at 350 mL min-1, make up gas (N2) flow at 30 mL min-1. Data were processed with
Chromcard software (Version 2.3.3).
2.6. Field experiments
Field experiments were conducted to estimate the attraction efficiency of semiochemical
formulations on hoverflies. These experiments were run from June to August 2009 on three
no pesticide-treated crop fields (beet, horse bean, winter wheat) of the University of Liege,
Gembloux Agro-Bio Tech in Gembloux (Belgium). The following experimental conditions
were investigated. One experimental plot (latin square design) was delineated in each crop at
the 20 x 60 m dimensions. Two formulations (E-β-farnesene and E-β-caryophyllene alginate
beads) and a control (alginate beads without semiochemical) were tested and replicated three
times in a plot. Two hundreds mg of formulations were deposited in sticky delta traps (Pirlot,
Waremme, Belgium). Distances between traps were of 10 m and 30 m for the lines and the
columns of the latin square design, in respect with one replicate of each formulation per line
and per column.
In order to evaluate the biological efficiency of formulations, the number of Syrphidae in
each trap was recorded. In the same time the population size of aphids nearby each trap (in an
area of 1 m around the trap) was approximated to statistically determine the effect of
formulations or aphids on the hoverfly attraction.
The weather conditions (temperature and relative humidity) were recorded during all the
duration of experiments by means of a Hobo data logger (Miravox, Hoevenen, Belgium). The
data with means, standard deviation (SD), minimal and maximal values, are presented in
Table 1.
Table 1: Weather conditions from June to August 2009 presented with means, standard deviations, minimum
and maximum values of temperature (°C) and relative humidity (%).
Temperature (°C)
Relative humidity (%)
Mean
SD
20.9
63.1
8.4
24.5
Min. value
Max. value
3.3
23.6
41.9
100.0
173
Chapter VI.2
___________________________________________________________________________
2.7. Data analysis
Data were presented as total number of female Syrphidae caught per formulation and per
crop from June to August 2009. Data were subjected to one-way or two-way analysis of
covariance (ANCOVA) followed by a Dunnett’s test (95%) (comparison with a control) to
compare the trap data observed for the two semiochemical formulations to the control. The
influence of covariates was tested: lines and columns of the latin square plot, and aphid
population size in 1 m around the traps. Aphid population size was estimated for all the
duration of experiments and was reported in classes according to the density of aphids (class
1: < 5 aphids; class 2: from 5 to 50 aphids; class 3: from 50 to 500 aphids; class 4: > 500
aphids). All statistical analyses were conducted by using Minitab v15 for Windows®.
174
Chapter VI.2
___________________________________________________________________________
3. Results
3.1. Semiochemical release rate measurements
Semiochemical release rates were measured at various experimental conditions (Table 2) in
order to evaluate the impact of abiotic factors (temperature, relative humidity, sampling
airflow) on the diffusion process of sesquiterpenes from alginate formulations.
Table 2: Laboratory experimental conditions to determine release rates and diffusion coefficients for E-βfarnesene and E- β-caryophyllene.
Experimental
test
N° 1
N° 2
N° 3
N° 4
N° 5
N° 6
N° 7
N° 8
Relative
humidity
(%)
25
25
25
75
75
85
90
100
Airflow
(L/min)
0.05
0.50
1.00
0.50
0.50
0.50
0.50
0.50
Temperature
(°C)
20
20
20
20
40
20
20
20
Diffusion coefficient
for E-β-farnesene
(m²/s)
1.98 * 10-14
3.40 * 10-14
3.71 * 10-14
1.23 * 10-14
2.12 * 10-14
1.56 * 10-15
6.15 * 10-33
1.03 * 10-32
Diffusion coefficient for
E- β-caryophyllene
(m²/s)
1.35 * 10-15
1.57 * 10-15
1.13 * 10-15
7.39 * 10-14
1.03 * 10-14
1.33 * 10-32
8.26 * 10-33
9.93 * 10-31
In each case, diffusion coefficient was estimated according to the theoretical equation for the
diffusion in a sphere (Cranck, 1975):
Mt
6  1
 1  2  2 exp(  Dn ² ²t / a ²) ,
M
 n1 n
where Mt (in µg) is the cumulative quantity of semiochemical released at time t, M∞ (in µg) is
the cumulative quantity of semiochemical released at time ∞ (supposed to be the total
quantity of semiochemical incorporated in the bead at time t=0), a (in m) is the radius of one
bead; t (in s) is the diffusion time; n is the number of terms in the sum and D (in m²/s) is the
effective diffusion coefficient. The previous equation was applied according to the following
assumptions: (i) diffusion was realised in a non-steady state, (ii) the semiochemical surface
concentration was considered constant over time, (iii) the radius of the bead was assumed to
be constant over time. In practice, diffusion coefficient was approximated at the value which
minimised the sum square (SS) between the experimental and the theoretical Mt / M∞ values,
by using the solver tool in Excel (v. 2003). In each experimental test, M∞ was noticed at 8160
µg/200 mg alginate beads and 7600 µg/200 mg alginate beads for E-β-farnesene and E-βcaryophyllene, respectively. These values were found in a previous study (results submitted
175
Chapter VI.2
___________________________________________________________________________
in Pest Management Science). Figures 1a and 1b show the cumulative release rates over
time for the eight experiments for E-β-farnesene and E-β-caryophyllene, respectively.
Calculated diffusion coefficients are presented in Table 2 for both molecules.
Considering release graphs and diffusion coefficients obtained for both molecules, the most
limiting abiotic factor on release kinetic was the relative humidity (comparison of tests n °2,
4, 6, 7 and 8). Indeed, from 85 % to 100 % relative humidity, semiochemical releases
decreased for E-β-farnesene and E-β-caryophyllene with diffusion coefficients close to zero
(10-31 – 10-33) after 10 days. It was assumed that water absorbed by alginate created a
physical barrier to semiochemical diffusion. At lower relative humidities, diffusion
behaviours were not the same for both compounds. In the case of E-β-caryophyllene, tests led
at 75 % relative humidity gave higher release quantities and higher diffusion coefficients
than at 25 %. For E-β-farnesene, release values (cumulative quantity and diffusion
coefficients) were in the same range for 25 % and 75 % relative humidity. On field, relative
humidity was subjected to high variations from day to day, alginate beads suffering from
cyclic water absorption and desorption. In order to verify the release efficiency of such
beads, a complementary laboratory study consisted in applying from 1 to 4 water absorption
– desorption cycles to beads and to measure release during 10 days after each cycle. As
presented in Table 3, diffusion coefficients were not highly influenced by multiple water
absorption and desorption phenomena.
The second abiotic parameter which seemed to impact semiochemical diffusion was the
temperature. Two temperatures were laboratory-tested: 20 °C and 40 °C (tests n° 4 and 5). As
expected with the literature (Van der Kraan et al., 1990; Torr et al., 1997; Atterholt et al.,
1999; Johansson et al., 2001; Cork et al., 2008; Shem et al., 2009), release rates increased at
higher temperature for the two sesquiterpenes. Temperature had a direct impact on the rate of
evaporation of molecules in the air (Krüger et al., 2002).
The last tested factor to estimate release was sampling airflow. Compared to the two other
parameters (temperature and relative humidity), airflow had very few influence on diffusion
kinetic. Three airflow values (0.05, 0.50 and 1.00 L/min) (tests n° 1, 2 and 3) were evaluated
in the release study. With E-β-caryophyllene formulation, cumulative released masses and
diffusion coefficients were approximately the same for the three airflow values. In the case of
E-β-farnesene, airflow at 0.05 L/min led to a lowest released quantity of semiochemical than
at 0.50 and 1.00 L/min.
176
Figure 1: Cumulative mass released (in µg) of semiochemical (E-β-farnesene (a) and E-β-caryophyllene (b)) from alginate beads (200 mg) under eight different experimental
conditions.
Chapter VI.2
___________________________________________________________________________
Table 3: Diffusion coefficients of E-β-farnesene and E- β-caryophyllene obtained after 1 to 4 water absorptiondesorption cycles.
Number of cycle
1
2
3
4
Diffusion coefficient
for E-β-farnesene
(m²/s)
1.23 * 10-14
2.74 * 10-14
3.94 * 10-15
1.43 * 10-14
Diffusion coefficient for
E- β-caryophyllene
(m²/s)
7.39 * 10-14
9.88 * 10-15
2.10 * 10-15
1.25 * 10-15
As shown in this study, temperature and relative humidity factors were of significant
importance in the efficiency of semiochemical release from alginate beads. In field
experiments, these two climatic parameters were supposed to highly condition the trapping of
Syrphidae from day to day.
3.2. Field-trapping experiments
Bioassay results with statistical values are presented in Table 4 with mean numbers (± SD)
of female Syrphidae captured per formulation for each crop and for the three crops. In each
case, influence of covariates (lines and columns of the latin square plot, and aphid population
size) was statistically evaluated. Lines and columns had no significant influence (p-values >
0.05) on the attraction of hoverflies. On the other hand, aphid population size significantly
influenced the catches with p-values of 0.0500, 0.0040 and 0.0010 for beet, horse bean and
for the global data (3 crops), respectively. No aphid influence was shown for wheat, aphid
population density being the same (class 1) for each trap of the crop.
Dunnett’s test results showed differences of hoverfly attraction between formulations and
control, and between crops after aphid influence subtraction (adjusted p-values). In beet crop,
only E-β-caryophyllene formulation was highly significantly different compared to the
control (p-value = 0.0011). On the other hand, in horse bean crop, only E-β-farnesene
differed from the control (p-value = 0.0052). In wheat field, both formulations were not
different from the control in terms of attraction. In order to reduce the differences observed
between crops, the three crops were treated in the same statistical analysis. By considering
the globality of experiments (for 3 crops), both E-β-farnesene and E-β-caryophyllene were
significantly different from the empty alginate beads (control) with adjusted p-values of
0.0416 and 0.0064, respectively.
178
Chapter VI.2
___________________________________________________________________________
Table 4: Catches of female Syrphidae (mean ± SD) per formulation and per crop with the statistical data
Mean ± SD Control
Mean ± SD E-β-farnesene
Mean ± SD E-β-caryophyllene
Aphid influence (p-value)
Dunnett’s test (Adjusted p-value)
E-β-farnesene
E-β-caryophyllene
*
**
Beet
Horse Bean
Wheat
Global data
(3 crops)
45.6 ± 7.6
76.0 ± 36.4
145.7 ± 13.5
0.0500
78.3 ± 11.6
139.7 ± 29.8
91.0 ± 31.4
0.0040
27.0 ± 12.3
40.0 ± 14.8
38.3 ± 8.7
/
50.3 ± 24.3
85.2 ± 50.2
91.7 ± 49.7
0.0010
0.0513
0.0011**
0.0052**
0.1723
0.3796
0.4621
0.0416*
0.0064**
Significant difference compared to the control
High significant difference compared to the control
These results demonstrated the effective and sufficient release of volatile compounds during
the period of experiment. Indeed, as shown in Table 1, mean relative humidity value was
63.1 %. That is in the optimal releasing conditions as demonstrated in paragraph 3.1.
179
Chapter VI.2
___________________________________________________________________________
4. Discussion and conclusion
Adult hoverflies are considered as important pollinators of flowering plants given that they
feed on nectar and pollen. On the other hand, larvae are one of the most important aphid
predators (Ghahari et al., 2008). Some studies demonstrated the attraction of adult hoverfly
species by means of selected flowering plants established close to crop fields in order to
reduce aphid population size after females have laid their eggs (Ambrosino et al., 2006;
Sadeghi, 2008). Sadeghi (2008) showed the higher Syrphidae attractiveness with Matricaria
chamomilla flowering plant.
In the present research, the bioassay results were in accordance with previous works
(Almohamad et al., 2006; Harmel et al., 2007) showing the attraction efficiency of E-βfarnesene towards hoverflies. Nevertheless, these previous studies dealt with the use of
synthetic pheromones as attractant contrarily to the current work where semiochemicals from
natural origin were used. In the case of E-β-caryophyllene, the attraction phenomenon was
higher than for E-β-farnesene. Furthermore, to our knowledge, it was the first time this
molecule was tested as a potential attractant of hoverflies. Nevertheless, E-β-caryophyllene
was already demonstrated to be attractive towards aphid parasitoids, Aphidius ervi Haliday
(Sasso et al., 2009). The attraction efficiency of this compound, found in the volatile blend of
tomato plants, was shown by electroantennography and wind tunnel bioassays.
Many semiochemical slow-release devices were reported in the literature, principally as
mating disruption systems. These releasers generally consist in solid matrix (polymer)
dispensers through which semiochemicals can release over time (McDonough, 1991; Torr et
al., 1997; Johansson et al., 2001; Tomaszewska et al., 2005; Zhang et al., 2008). Some
sprayable slow-release formulations were also developed in such a way the semiochemical
compounds are dissolved in a biodegradable liquid matrix (Atterholt et al., 1999; de Vlieger,
2001; Welter et al., 2005).
Alginate bead release devices, as presented in this paper, consist in a compromise between
solid dispensers and liquid formulations given that a liquid (semiochemical and sunflower
oil) is enclosed in a solid polymeric network. Alginate gel beads were largely described as
efficient releasers for aroma and flavour volatile compounds in the food industry (Malone et
al., 2003) or for essential oils acting as antimicrobial agents (Chang et al., 2003; Lai et al.,
180
Chapter VI.2
___________________________________________________________________________
2007). Furthermore, the beads are simple to produce on a lab-scale, easy to manipulate and
have low impact on the environment (van Soest, 2007).
In the present research, alginate beads proved their effectiveness as semiochemical slowrelease systems on field experiments despite their limitation of use at high relative humidity.
A study of semiochemical release in environmental conditions is currently in prospect as well
as the understanding of diffusion phenomenon inside alginate bead network according to
abiotic factors.
This biological control approach could be improved and be successful over time if agrienvironmental measures are developed in the same time in the farmlands surrounding the
experimental crops. In this manner, it could be assumed that the global density of pest
predators and parasitoids will increase.
Acknowledgements
The authors are grateful to Prof. Bernard Bodson from Gembloux Agro-Bio Tech, University
of Liege (Phytotechnie des Régions Tempérées) for giving the authorization to realise field
experiments on crops.
This research was funded by a Walloon Region Ministry grant (WALEO2: SOLAPHID –
RW/FUSAGX 061/6287).
181
Chapter VI.2
___________________________________________________________________________
References
Almohamad R., Verheggen F.J., Francis F., Haubruge E. (2006). Evaluation of hoverfly
Episyrphus balteatus De Geer (Diptera: Syrphidae) oviposition behaviour towards
aphid-infested plants using a leaf disc system. Communication in Applied Biological
Science 71/2b: 403-412.
Ambrosino M.D., Luna J.M., Jepson P.C., Wratten S.D. (2006). Relative frequencies of visits
to selected insectary plants by predatory hoverflies (Diptera: Syrphidae), other
beneficial insects and herbivores. Journal of Environmental Entomology 35(2): 394400.
Atterholt C.A., Delwiche M.J., Rice R.E., Krochta J.M. (1999). Controlled release of insect
sex pheromones from paraffin wax and emulsions. Journal of Controlled Release 57:
233-247.
Brown A.E., Riddick E.W., Aldrich J.R., Holmes W.E. (2006). Identification of (-)-βcaryophyllene as a gender-specific terpene produced by the multicolored Asian lady
beetle. Journal of Chemical Ecology 32: 2489-2499.
Chambers R.J., Sunderland K.D., Stacey D.L., Wyatt I.J. (1985). Control of cereal aphids in
winter wheat by natural enemies: aphid-specific predators, parasitoids and pathogenic
fungi. Annals of Applied Biology 108: 219-231.
Chambers R.J., Adams T.H.L. (1986). Quantification of the impact of hoverflies (Diptera:
Syrphidae) on cereal aphids in winter wheat: an analysis of field populations. Journal of
Applied Ecology 23: 895-904.
Chang C.P., Dobashi T. (2003). Preparation of alginate complex capsules containing
eucalyptus essential oil and its controlled release, Colloids and Surfaces B 32: 257-262.
Cork A., De Souza K., Hall D.R., Jones O.T., Casagrande E., Krishnaiah K., Syed Z. (2008).
Development of a PVC-resin-controlled release formulation for pheromones and use in
mating disruption of yellow rice stem borer, Scirpophaga incertulas. Crop Protection
27: 248-255.
Cranck J. (1975). The mathematics of diffusion. Clarendon Press, Oxford.
182
Chapter VI.2
___________________________________________________________________________
De Vlieger J.J. (2001). Development of a sprayable slow-release formulation for the sex
pheromone of the Mediterranean corn borer, Sesamia nonagroides. IOBC wprs
Bulletin, 24(2): 101-106.
Foster S.P., Denholm I., Thompson R., Poppy G.M., Powell W. (2005). Reduced response of
insecticide-resistant aphids and attraction of parasitoids to aphid alarm pheromone; a
potential fitness trade-off. Bulletin of Entomological Research 95: 37-46.
Francis F., Vandermoten S., Verheggen F., Lognay G., Haubruge E. (2005a). Is the E-βfarnesene only volatile terpenoids in aphids? Journal of Applied Entomology 129: 6-11.
Francis F., Martin M., Lognay G., Haubruge E. (2005b). Role of E-β-farnesene in systematic
aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae). European
Journal of Entomology 102: 431-436.
Ghahari H., Hayat R., Tabari M., Ostovan H. (2008). Hover flies (Diptera: Syrphidae) from
rice fields and around grasslands of Northern Iran. Munis Entomology and Zoology 3:
275-284.
Harmel N., Almohamad R., Fauconnier M.-L., Du Jardin P., Verheggen F., Marlier M.,
Haubruge E., Francis F. (2007). Role of terpenes from aphid-infested potato on
searching and oviposition behavior of Episyrphus balteatus. Insect Science 14: 57-63.
Heuskin S., Godin B., Leroy P., Capella Q., Wathelet J.-P., Verheggen F., Haubruge E.,
Lognay
G.
(2009).
Fast
gas
chromatography
characterisation
of
purified
semiochemicals from essential oils of Matricaria chamomilla L. (Asteraceae) and
Nepeta cataria L. (Lamiaceae). Journal of Chromatography A 1216: 2768-2775.
Heuskin S., Rozet E., Lorge S., Farmakidis J., Hubert Ph., Verheggen F., Haubruge E.,
Wathelet J.-P., Lognay G. (2010). Validation of a fast gas chromatographic method for
the study of semiochemical slow release formulations. Journal of Pharmaceutical and
Biomedical Analysis 53: 962-972.
Johansson B.G., Anderbrant O., Simandl J., Avtzis N.D., Salvadori C., Hedenström E.,
Edlund H., Högberg H.-E. (2001). Release rates for pine sawly pheromones from two
types of dispensers and phenology of Neodiprion sertifer. Journal of Chemical Ecology
27: 733-745.
183
Chapter VI.2
___________________________________________________________________________
Krüger A.J., Tolmay A.T. (2002). Prediction of the release characteristics of alcohols from
EVA using a model based on Fick’s second law of diffusion. Journal of Applied
Polymer Science 84: 806-813.
Lai F., Loy G., Manconi M., Manca M.L., Fadda A.M. (2007). Artemesia arborescens L.
essential oil loaded beads: preparation and characterization, AAPS PharmSciTech 8(3):
E1-E7.
Malone M.E., Appelqvist I.A.M. (2003). Gelled emulsion particles for the controlled release
of lipophilic volatiles during eating, Journal of Controlled Release 90: 227-241.
McDonough L.M. (1991). Controlled release of insect sex pheromones from a natural rubber
substrate. In: Hedin P.A., ed. Naturally Occurring Pest Bioregulators. American
Chemical Society, pp. 106-124.
Moens J., De Clercq P., Tirry L. (2010). Side effects of pesticides on the larvae of the
hoverfly Episyrphus balteatus in the laboratory. Phytoparasitica DOI 10.1007/s12600010-0127-3.
Sadeghi H. (2008). Abundance of adult hoverflies (Diptera: Syrphidae) on different
flowering plants. Caspian Journal of Environmental Sciences 6: 47-51.
Sasso R., Iodice L., Woodcock C.M., Pickett J.A., Guerrieri E. (2009) Electrophysiological
and behavioural responses of Aphidius ervi (Hymenoptera: Braconidae) to tomato plant
volatiles. Chemoecology 19: 195-201.
Shem P.M., Shiundu P.M., Gikonyo N.K., Ali A.H., Saini R.K. (2009). Release kinetics of a
synthetic tsetse allomone derived from waterbuck odour from a Tygon silicon dispenser
under laboratory and semi field conditions. American-Eurasian Journal of Agriculture
& Environmental Science 6(6): 625-636.
Stoner K. (2004). Approaches to the biological control of insects, University of Maine,
Cooperative Extension Bulletin 7144, (www.umext.maine.edu/onlinepubs/PDFpubs
/7144.pdf; 06/07/10).
Tomaszewska E., Hebert V.R., Brunner J.F., Jones V.P., Doerr M., Hilton R. (2005).
Evaluation of pheromone release from commercial mating disruption dispensers.
Journal of Agricultural and Food Chemistry 53: 2399-2405.
184
Chapter VI.2
___________________________________________________________________________
Tomova B.S., Waterhouse J.S., Boberski J. (2005). The effect of fractionated Tagetes oil
volatiles on aphid reproduction. Entomologia Experimentalis et Applicata 115: 153159.
Torr S.J., Hall D.R., Phelps R.J., Vale G.A. (1997). Methods for dispensing odour attractants
for tsetse flies (Diptera: Glossinidae). Bulletin of Entomologycal Research 87: 299-311.
Van der Kraan C., Ebbers A. (1990). Release rates of tetradecen-1-ol acetates from
polymeric formulations in relation to temperature and air velocity. Journal of Chemical
Ecology 16: 1041-1058.
Van Soest J.J.G. (2007). Encapsulation of fragrances and flavours: a way to control odour
and aroma in consumer products, in: R.G. Berger (Ed.), Flavours and Fragrances,
Springer, pp.439-455.
Verheggen F.J., Fagel Q., Heuskin S., Lognay G., Francis F., Haubruge E. (2007).
Electrophysiological and behavioral responses of the multicolored asian lady beetle,
Harmonia axyridis Pallas, to sesquiterpene semiochemicals. Journal of Chemical
Ecology 33: 2148-2155.
Verheggen F.J., Arnaud L., Bartram S., Gohy M., Haubruge E. (2008). Aphid and plant
volatiles induce oviposition in an aphidophagous Hoverfly. Journal of Chemical
Ecology 34: 301-307.
Verheggen F.J., Haubruge E., De Moraes C.M., Mescher M.C. (2009). Social environment
influences aphid reproduction of alarm pheromone. Behaviour Ecology 20: 283-288.
Verheggen F.J., Haubruge E., Mescher M.C. (2010). Alarm pheromones. In Litwack G. (ed.),
Pheromones. Elsevier.
Welter S.C., Pickel C., Millar J., Cave F., Van Steenwyk R.A., Dunley J. (2005). Pheromone
mating disruption offers selective management options for key pests. California
Agriculture 59(1): 16-22.
Zhang A., Kuang L.F., Maisin N., Karumuru B., Hall D.R., Virdiana I., Lambert S., Purung
H.B., Wang S., Hebbar P. (2008). Activity Evaluation of cocoa pod borer sex
pheromone in cacao fields. Environmental Entomology 37(3): 719-724.
185
Chapter VII
General discussion, Conclusions and
Perspectives
Chapter VII
___________________________________________________________________________
General discussion, conclusions and perspectives
VII.1. General discussion
In the current context of environment protection, the use of pesticides must be cautiously
managed and is more and more associated with integrated pest management strategies more
friendly for the environment and human health. In the present PhD thesis, a biological control
approach was investigated in order to reduce the aphid population – causing many damages
to crop plants – by means of aphid natural enemies. Biological control strategies already
proved their efficiency in past researches (Agelopoulos et al., 1999; Powell et Pickett, 2003;
Michaud et al., 2008). Generally, beneficial insects (predators or parasitoids) were directly
released on pest infested fields, with sometimes the use of exotic natural enemies (for e.g.
Harmonia axyridis Pallas) causing invasion and ecological problems (Brown et al., 2008).
The goal of the present research consisted in developing a natural semiochemical formulation
having low impact on the environment and which guaranteed a release of compounds at a
sufficient rate to be perceived by local beneficial insects and to attract them. This research
was original at different levels. Firstly, semiochemicals were quickly isolated from natural
plant origin instead of coming from synthesis as in the most of published works. Secondly,
for the first time in integrated pest management field of research, semiochemical
characterisation and quantification analyses were conducted by means of an ultra fast GC in
respect with an accuracy profile validation method. Thirdly, alginate bead formulations
revealed their originality in terms of “auto-regulation” process with relative humidity abiotic
factor.
The research consisted in answering to various questions:
-
Which semiochemically active molecules could be chosen?
-
From which natural origin the semiochemicals could be extracted?
-
How to analyse and quantify semiochemicals?
-
How to purify semiochemicals?
-
How to formulate semiochemicals?
-
Is the formulation efficient in terms of semiochemical release and in terms of
biological control device?
The following discussion will be presented as answers to the previous questions.
189
Chapter VII
___________________________________________________________________________
Which semiochemically active molecules could be chosen?
E-β-Farnesene and Z,E-nepetalactone were both semiochemicals interesting for their
previously related properties as aphid parasitoid and predator attractants. Indeed, E-βfarnesene, firstly considered as the alarm pheromone of many aphid species (Francis et al.,
2005a), could also have a role of kairomone by attracting and inducing oviposition of aphid
predator, Episyrphus balteatus De Geer (Diptera: Syrphidae) (Francis et al., 2005b;
Verheggen et al., 2008, 2009, 2010) and by attracting Aphidius ervi parasitoids (Du et al.,
1998; Powell et al., 2003). On the other hand, Z,E-nepetalactone, a compound of the sexual
pheromone of some aphid species, was also reported to be attractive towards A. ervi.
(Glinwood et al., 1999).
E-β-caryophyllene was identified as a potential component of the aggregation pheromone of
another aphid predator, Harmonia axyridis Pallas (Brown et al., 2006; Verheggen et al.,
2007).
From which natural origin the semiochemicals could be extracted?
In general, synthetic pheromones are used in IPM slow-release devices. Molecules from plant
origin were preferred to synthetic compounds for ecological and cost reasons: the natural
origin of the sesquiterpenes; multi-step chemical syntheses are generally time consuming and
high solvent consuming for producing low quantities of compounds; synthetic products are
very expensive and therefore could not be used for the purpose of the present thesis. After a
rapid literature screening on various plant composition, Matricaria chamomilla L.
(Asteraceae), the German chamomile, was found to contain up to 60 % of E-β-farnesene in
the essential oil according to the cultivar, the age of the plant and the distilled part of the
plant. For example, Szöke et al. in 2003 studied the essential oil composition of six
Matricaria chamomilla cultivars in Hungary. The results showed E-β-farnesene percentages
from 7 % to 15 % in flowers, from 52 % to 59 % in herbs (stems and leaves) and from 25 %
to 40 % in roots, according to the cultivar. On the other hand, E-β-caryophyllene was one of
the major components of the Nepeta cataria L. (Lamiaceae) essential oil, firstly chosen for
its high content in Z,E-nepetalactone.
190
Chapter VII
___________________________________________________________________________
How to analyse and quantify semiochemicals?
In order to select the richest oils, two essential oils of Matricaria chamomilla and two of
Nepeta cataria from different origins were characterised by GC-MS and by fast GC (data not
submitted for publication). The percentages of compounds in the oils were of 42.6 % and
21.3 % in E--farnesene in the Matricaria chamomilla essential oils. The composition of
both Nepeta cataria essential oils were of 9.7 % in E--caryophyllene and 73.3 % in Z,Enepetalactone in the first oil (originated from France), and 58.9 % in E--caryophyllene and
8.4 % in Z,E-nepetalactone in the second one (originated from Canada). The richest ones in
active molecules were retained for further purification.
Identifications of essential oil components were achieved by GC-MS by comparing mass
spectra with those of a spectral library and confirmed with retention index calculations. Fast
GC analytical method was optimised in order to guarantee a good resolution of
semiochemical peaks for further sample analyses and quantifications in less than five minutes
The method was developed for the analysis of monoterpenes (α-pinene and limonene) and
sesquiterpenes (E--farnesene and E--caryophyllene) in the same time. Indeed,
monoterpenes are compounds commonly found in essential oils and which can act as
semiochemicals (Miller et Lindgren, 2000; Poland et al, 2004). Quantification was realised
by means of internal standards. n-Butyl-benzene was used as internal standard for the
monoterpenes, and longifolene for sesquiterpene quantifications. Ideally, internal standard
must meet certain specifications: to belong to the same family, to have close molecular
weight and to have close retention time than the compounds to quantify. Furthermore,
internal standard must be absent from the original sample to analyse. Nevertheless, it is
difficult to find a molecule which meets all these criteria. n-Butyl-benzene being not a
monoterpene, was chosen as internal standard due to its close molecular weight (Mw = 134
for n-butyl-benzene and Mw = 136 for monoterpenes) and its good peak resolution with the
optimised analytical fast GC method.
Two validations of the fast GC optimised analytical method were then conducted in order to
quantify sesquiterpenes in formulation samples and to estimate their rate of release in an
accurate manner. The first validation method based on the classical ISO 5725, AOAC (2006)
norms and on the GLP standard operating procedures developed in the Department of
Analytical Chemistry, gave for each validation criterion a value in agreement with the
191
Chapter VII
___________________________________________________________________________
limitations of the previously cited norms. Two ranges of concentrations (range 1: 0.008 to
0.100 µg/µl; range 2: 0.080 to 1.000 µg/µl) were evaluated in terms of linearity of the
regression model (calibration curve), accuracy, precision, LOD and LOQ of the method. On
the other hand, the accuracy profile validation technique, based on the guidelines of the
Société Française des Sciences et Techniques Pharmaceutiques (SFSTP) (Hubert et al., 1999,
2003, 2006), was a more harmonized approach which took into account the notion of total
error (bias (systematic error) + standard deviation (random error)) in the acceptation or reject
decision of the validation procedure rather than dealing with acceptance criteria one by one
as realised with the classical validation strategies. Total error was expressed by the accuracy
as a combination of trueness and precision. Furthermore, the accuracy profile validation
concept introduced the notion of risk (expressed by the tolerance limits with a maximum β
risk) in order to assess the quality of the results produced by an analytical procedure. This
approach was principally developed for validation procedures in the pharmaceutical industry.
To our knowledge, in the present thesis it was the first time that the accuracy profile
validation concept was described in the field of environment and more particularly in
biological control research with a fast GC analytical method. Furthermore, to date, no other
paper describes the use of two levels acceptance limits in an accuracy profile validation
methodology. Indeed, a higher level of acceptance (25 %) was settled at the lower
quantification limits (81.6 ng/µl and 80.5 ng/µl for E-β-farnesene and E-β-caryophyllene,
respectively) considering a higher relative standard deviation (higher random error) at lower
concentrations as related by Horwitz et al. (1980). The accuracy profiles were constructed
with a maximum β risk of 5 % and according to a calibration curve corresponding to a linear
regression model through zero fitted with the maximum level of concentration. The
analytical method (with a split ratio 1:100) was demonstrated to be accurate along the tested
concentration range from 80 ng/µl to 1000 ng/µl for both sesquiterpenes.
To complete the analytical validation, the performances of the UFM column were also
demonstrated, firstly in terms of the evolution of the theoretical plates number (N) with the
quantity of molecules injected on the column. This number of theoretical plates reflects the
resolution capacity of the column. UFM performances were also evaluated with the
minimum eight of a theoretical plate (Hmin) obtained at a specific carrier gas velocity (cm/s).
This optimal carrier gas velocity was of 35.86 cm/s. Optimised analytical method was
conducted at a velocity of 43.94 cm/s, close to the optimal value without affecting peak
separation efficiency and allowing faster analyses.
192
Chapter VII
___________________________________________________________________________
Before this PhD thesis, very few studies were conducted with an ultra fast gas chromatograph
similar to the one we used (Bicchi et al., 2004, 2005). The fast chromatography is known
since many years and consists in manually modifying classical GC parameters in order to
increase the analytical speed. These modifications could be: to reduce the length (from 1 to 5
m compared to 15 to 50 m in classical GC) and the internal diameter (from 0.05 to 0.2 mm
compared to 0.25 to 0.32 mm in classical GC) of the GC column coated with a thinner film
(from 0.1 to 0.25 µm compared to 0.25 to 1 µm in classical GC), to change the carrier gas (to
replace helium with hydrogen), to increase carrier gas flow. Analysis time could be reduced
to a few minutes compared to the thirthy to sixty minutes in classical GC analysis. Reed
(1999) gave a complete historic of the fast GC evolution. Ultra fast GC is a new technology
developed by Thermo Electro Corporation (Italy) which allows analyses in less than 5
minutes with a specialized instrumentation. The GC column, from 2.5 m to 10 m in length
with an internal diameter comprised between 0.1 to 0.32 mm and a film thickness from 0.1 to
0.5 µm, is completely integrated in a box (Ultra Fast Module®) which is directly heated.
Ramps of temperature could increase until 1200°C/min. The detector, adapted to the fastness
of analysis, has an acquisition frequency of 300 Hz. In the present study, the ultra fast GC
demonstrated its usefulness during the analytical validations which could be realised in a
short time with a maximum of replicates, as well as for the analysis of many fractions during
the liquid column chromatography optimisation for semiochemical purification.
This GC tool also proved its efficiency in many studies of chemical ecology realised in
collaboration with the Department of Analytical Chemistry and which conducted to
publications (.
How to purify semiochemicals?
The efforts of purification were first conducted on E-β-farnesene by means of a frozen
trapping system. Headspace volatile compounds (from dried chamomile flowers and from
chamomile essential oils) retained on cold trap were analysed and quantified after solvent
extraction (data not shown). This technique was not appropriate in terms of purification:
percentages of E-β-farnesene in the extracts being almost the same than in the essential oils
or in the headspace volatile content of dried plants. Furthermore, the mass recovery of the
technique was very poor. This purification process was given up for the development of a
liquid column chromatography separation of essential oils.
193
Chapter VII
___________________________________________________________________________
Liquid column chromatography fractionation was firstly optimised at small scale on 11 g
dried silicagel for sesquiterpene purification. In the beginning of the research, E-βcaryophyllene was not investigated as a potential semiochemical of interest. The efforts were
thus concentrated on E-β-farnesene and Z,E-nepetalactone purification. Fractionation of
essential oils was optimised in order to select enriched fractions of compounds of interest by
plotting the evolution of component purity with the elution volume. In that way, five
fractions were obtained for Matricaria chamomilla, with fraction F3 being the richest in E-βfarnesene (79.6 %). In the case of Nepeta cataria essential oil (originated from France), the
fractionation process was realised in two steps. The first one, realised with n-pentane as
mobile phase, led to four fractions among which the fraction F2 was enriched in E-βcaryophyllene (79.5 %). The second step was dedicated to the purification of Z,Enepetalactone with a more polar mix of elution solvents (n-pentane/diethylether 80:20). Two
more fractions were obtained where fraction F5 contained 97 % of Z,E-nepetalactone.
Considering the three isomers of nepetalactone present in the essential oil of Nepeta cataria,
the solvent-free purified fraction F5 was analysed by 1H and 13C NMR spectrometry in order
to verify the stereoisomery of the Z,E form. In later experiments, Z,E-nepetalactone was no
more purified considering poor results obtained by preliminary olfactometry bioassays led on
aphid parasitoids and hoverflies.
The technique of fractionation was then rapidly developed at higher scale and realised in
routine by manual flash chromatography both for E-β-farnesene and E-β-caryophyllene
isolation from essential oils. To isolate E-β-caryophyllene with a high purity, another
essential oil of Nepeta cataria (originated from Canada) was chosen for its composition (58.9
% E-β-caryophyllene). Flash chromatography had many advantages for the success of the
present work. First of all, the rapidity and the simplicity of the technique were very useful.
Moreover, high masses of essential oils (9.3 g and 9.5 g for Matricaria chamomilla and
Nepeta cataria, respectively) were fractionated in less than 10 minutes by means of the
pressurised (N2) system, leading to the achievement of important quantities of purified
natural sesquiterpenes (from 3.5 g to 6.5 g for E-β-farnesene at 83.8 % of purity and E-βcaryophyllene at 97.7 % of purity, respectively) for each flash chromatography batch. These
quantities of semiochemicals were sufficient to formulate from 12 g (for E-β-farnesene) to 65
g (for E-β-caryophyllene) alginate beads.
194
Chapter VII
___________________________________________________________________________
The optimisation of the fractionation process took also into account the choice of the elution
solvent. A first selection was realised by comparing various solvents by thin layer
chromatography of essential oils. N-pentane was finally chosen for its boiling point (36.1 °C)
lower than the one of sesquiterpenes. The solvent evaporation step was feasible at 40 °C by
Büchi rotatory evaporator with a small loss of semiochemicals. Recoveries of the evaporation
process were measured in five replicates for E-β-farnesene (96.3 % ± 0.9 %) and were judged
satisfactory according to AOAC norm.
How to formulate semiochemicals?
Semiochemical formulation was a key step in the realisation of the present PhD thesis.
Various formulations were tested such as paraffin oil, paraffin wax, and sunflower oil before
succeeding with alginate beads containing mixed semiochemical and sunflower oil. The two
first formulations were rapidly given up due to problems encountered for obtaining a
homogenous distribution of semiochemical compounds in the matrix. Indeed, in both cases,
phase separations were observed between matrixes and sesquiterpenes. Furthermore,
sesquiterpenes from these formulations were released at a random rate from day to day (from
6 µg/day to 12 µg/day) and during a short period of time (11 days). Finally, the perspective
to use these formulations on crops turned out to be complex and many time consuming
considering the frequent replacements of formulation on fields. On the other hand, sunflower
oil formulation was evaluated in terms of protection efficiency of sesquiterpenes towards
oxygen degradation with time. This protection effect for E-β-farnesene proved to be lower
than the one observed for alginate beads with and without α-tocopherol. It was important to
notice the difference of degradation time between E-β-farnesene and (E,E)--farnesene in
alginate formulations (data not published) as demonstrated by B. Godin (2008). Indeed,
(E,E)--farnesene half-life times (corresponding to 50 % residual quantity in formulations)
were of 4.5 days and 5.2 days in alginate beads with and without α-tocopherol, respectively.
At these periods of time, the residual percentages of E-β-farnesene in formulations were at 92
% and 61 %, respectively. It was assumed that this higher degradation observed for (E,E)-farnesene was due to the auto-oxidative process of this molecule leading to a hydroperoxide
molecular form after passing through a very reactive conjugated triene intermediate radical
(Rowan et al., 1995). This mechanism is well known in fruit superficial scald disorder
occurring after storage at low temperature (Rowan et al., 1995; Whitaker et al., 1997;
Pechous et Whitaker, 2004). The consequence of this degradation process was a higher
195
Chapter VII
___________________________________________________________________________
percentage of E-β-farnesene (> 92 %) in the volatile blend released from alginate
formulation. In the case of E-β-caryophyllene, the difference of protection efficiency
between the three formulations was less significant. However, as for paraffin oil matrix,
sunflower oil formulation was difficult to implement in the field due to its liquid condition in
contrast with alginate beads being in a gel solid state easier to manipulate.
Semiochemical alginate beads were formulated in close collaboration with Stéphanie Lorge
from the Institute of Condensed Matter and Nanosciences (UCL, Belgium). Alginate, natural
polymer derived from marine brown algae (Phaeophyceae), was considered as biodegradable
formulation matrix (Dornish et al., 1996; Aggarwal et al., 1999; Chung et al., 2002). Alginate
gel beads, having no harmful impact on the environment, presented many advantages for the
achievement of the present thesis. Indeed, alginate was already known as efficient aroma and
flavour releaser in the food industry (Chang et al., 2003; Malone et al., 2003; Madene et al.,
2006; Lai et al., 2007). Furthermore, this polymeric matrix has low oxygen permeability
(Rojas-Graü et al., 2007) and thus, could protect sesquiterpenes against oxidation as
demonstrated hereunder.
The formulations were optimised by considering the sesquiterpene encapsulation capacity as
well as the compound rate of release. Various parameters implied in the gel structure and in
the encapsulation properties were studied like the type (Sigma L, Sigma M, Satialgine) and
the concentration (from 1 % to 3 % w/v in water) in alginate, the type (Ca2+ or Cu2+) and the
concentration (from 0.05 M to 0.5 M) in cross-linker ion, the ionic strength (0.5 M or 1.0 M),
and the maturation time (20 minutes or 48 hours). The optimised formulation, with an
optimal quantity of encapsulated E-β-farnesene at 3468 µg ± 878 µg in 100 mg alginate
beads, was the following: Alginate Sigma Low viscosity (Mw = 235.5, M/G = 1.56) at 1.5 %
(w/v), Ca2+ at 0.2 M as cross-linker cation, ionic strength at 0.5 M and maturation time of
48h.
The formulations were then characterised in terms of texture of the alginate beads. It was
noteworthy that the bead resistance was lower for the Sigma Low viscosity alginate type
compared to the two others. This phenomenon was due to the proportional relationship
between the percentage of guluronate units (G) in alginate, conditioning the polymer
gelification, and the bead resistance (Smidsrod et Grasdalen, 1984; Kakita et Kamishima,
2008). Indeed, Sigma L alginate type had mannuronate/guluronate ratio (M/G = 1.56) higher
(weak percentage of G units) than Sigma M (M/G = 1.22) and Satialgine (M/G = 0.66) types.
196
Chapter VII
___________________________________________________________________________
The texture of alginate beads was also influenced by the type of cross-linker ion. Differences
of deformation were observed between Ca2+ and Cu2+ alginate beads. These experiments also
supported the suppositions already related by Velings et al. (1995). In that paper, Velings et
al. studied the volume reduction during maturation of Cu- and Ca-alginate beads. The authors
observed that, at a fixed ionic strength (1.0 M) obtained by NaCl addition, the copper
concentration had no effect on the volume loss contrarily to calcium concentration which
modified the bead volume. The explanations were that Na+ ions displaced a small proportion
of Ca2+ ions from the alginate network but not Cu2+ ions considering the 10 times higher
affinity between copper and alginate than between calcium and the polymer (Haug et
Smidsrod, 1970). This phenomenon resulted in a denser gel with an increased rigidity for Cualginate beads. Consequently, in our study, we observed a higher mean resistance force for
Cu-beads (3.59 N) than for Ca-beads (2.60 N).
Beads were then submitted to confocal laser scanning microscopy (CLSM) in order to
observe the semiochemical dispersion in the alginate network. The main advantage of the
confocal microscopy is that this technique can provide information on polymeric network
close to the reality without the need to cut the beads (Liu et al., 2002). Very few studies
report the characterisation of alginate formulations by CLSM. Nevertheless, Liu et al. (2002)
related the characterisation of alginate beads by this technique in order to compare the
homogeneity of alginate network for two gelation methods based on internal and external
source of Ca2+, the last technique corresponding to ours. They concluded that by external
gelation, Ca2+ and alginate distributions within the bead were inhomogeneous due to
concentration gradients, both concentrations being higher at the surface and lower in the
core. Their results were in accordance with previous papers (Skjak-Braek et al., 1989; Quong
et al., 1998, Vandenberg and De La Noüe, 2001). In the present PhD thesis, we did not
considered this phenomenon but differences were pointed out between dried (Aw: 0.42) and
wet (Aw: 0.98) alginate beads. It was the first time such a comparison was conducted and
related. In dried polymeric formulation, semiochemical mixed with sunflower oil appeared to
be spread in the entire alginate network. On the contrary in wet bead, sunflower oil
containing the sesquiterpenic molecules was only present at the edge of the bead. We
supposed that at high water activity, the alginate network changes. This could confirm our
observations during semiochemical release rate measurement where diffusion of
sesquiterpenes stopped at relative humidity higher than 85 %. To our knowledge, this
mechanism was not previously related in the literature.
197
Chapter VII
___________________________________________________________________________
The use of (micro)capsules to release pheromones to the atmosphere was recently related by
some authors. Nevertheless, all the formulations described in the literature were different
from the ones presented in this PhD work. For example, Kong et al. (2009) reported the
preparation of capsules for releasing dodecanol in the air in a mating disruption approach
against codling moth (Cydia pomonella). Their formulations were prepared via coacervation
of gelatine and arabic gum as polymeric matrix and consisted in capsules from 1 to 340 µm
in size. A second paper (Yosha et al., 2008) studied the slow release of dodecyl acetate to the
atmosphere from gelatine-alginate beads. Their principle of formulation by extrusion of
polymer in CaCl2 is quite the same than the one we used. However, they added gelatin to
alginate as emulsifier in order to stabilise the emulsion. The authors obtained beads of the
same size than in our work (1.5 to 2 mm in diameter). They estimated pheromone release rate
by residual pheromone quantification after the beads were placed in the following conditions:
0.5 m/s air flow speed, 35° temperature and 3-5% humidity. They concluded that pheromone
release rate was dependent on the alginate concentration: as polymer concentration increased,
the pheromone release rate decreased. These observations could be of great interest to control
release rate of our semiochemicals in future prospects.
Is the formulation efficient in terms of semiochemical release and in terms of
biological control devices?
We studied semiochemical diffusion from alginate beads in laboratory controlled conditions
(temperature, relative humidity and air flow rate). In preliminary experiments we
demonstrated that the release of molecules was efficient during a longer period of time (more
than 80 days, data not shown) compared to other tested formulations like paraffin oil and wax
in the same lab conditions (20 °C, 75 % relative humidity and 0.5 L/min air flow rate). The
objective was then to estimate semiochemical diffusion (expressed by diffusion coefficient D
in m²/s) behaviour when beads were placed in different experimental conditions. In this
estimation, we supposed that the beads were spherical with a homogeneous structure. The
lab study described in the present work highlighted the important impact of relative humidity
on diffusion of both active compounds. Indeed, at a relative humidity higher than 85 % the
release of semiochemicals was stopped. It was assumed that water absorbed by alginate
polymer limited the diffusion of semiochemicals from alginate network to the air. However,
this phenomenon was reversible, beads starting to release again once at a lower relative
humidity. This “auto-regulation” process of release was very useful considering the goal of
198
Chapter VII
___________________________________________________________________________
such formulations to attract predators and parasitoids of aphids in real environmental
conditions.
When semiochemical release was not stopped due to high relative humidity, we observed
diffusion coefficients from 1.13 E-15 to 7.39 E-14 m²/s for E-β-caryophyllene, and from 1.56
E-15 to 3.71 E-14 m²/s for E-β-farnesene. These values were lower than the ones observed by
Velings et al. (1996) for the diffusion of different solutes, but the authors studied diffusion of
compounds in a solution and not in the air. These authors also demonstrated that the diffusion
coefficient decreased with increasing molecular weight of compounds. The differences of
diffusion coefficients observed in our experiments between both semiochemicals could not
be explained by molecular weight differences. Indeed, both semiochemicals are
sesquiterpenes of Mw = 204. Nevertheless, they possess different conformations, E-βfarnesene being a linear molecule and E-β-caryophyllene being cyclic. Very few papers
report calculation of diffusion coefficient for estimating the release of volatile compounds in
the air from alginate beads. A recent study (Hambleton et al., 2010) described the diffusion
behavior in the air of two aroma compounds, n-hexanal and D-limonene, through emulsified
(with lipid addition) and non-emulsified sodium alginate-based films. The authors compared
determined coefficients from sorption kinetics following the second Fick’s law following the
same procedure we adapted for spherical model. They obtained estimated diffusion
coefficients close to the one of this PhD thesis: 6.05 E-16 m²/s and 7.65 E-16 m²/s in the case
of D-limonene for alginate films without and with lipids, respectively; 1.45 E-15 m²/s and
1.35 E-15 m²/s in the case of n-hexanal for alginate films without and with lipids,
respectively.
The efficiency of semiochemical alginate formulations as biological control devices was
finally demonstrated by olfactometry on aphid parasitoids, Aphidius ervi, and by field
experiments to trap aphid predators. In both cases, alginate beads loaded with semiochemical
molecules proved to be more attractive towards beneficial insects than beads without
compound. E-β-farnesene and E-β-caryophyllene formulations were almost similarly
attractive towards aphid parasitoids. However, the attractiveness of E-β-caryophyllene
towards aphid predatory Syrphidae species proved to be slightly higher than with E-βfarnesene compound. The role of E-β-caryophyllene as parasitoid and Syrphidae attractant is
surprising. Indeed, this molecule was supposed to attract Harmonia axyridis Pallas
ladybeetles according to the papers of Brown et al. (2006) and Verheggen et al. (2007), but
199
Chapter VII
___________________________________________________________________________
no ladybeetle was found in the traps we used on field. We supposed that sticky delta traps
were not adapted to catch robust insects such as Coleoptera. Only one paper (Sasso et al.,
2009), to our knowledge, described the attractiveness of A. ervi towards E-β-caryophyllene.
On a biological point of view, this molecule, present in the volatile blend of tomato plants,
could be considered by parasitoids as a chemical cue associated with aphids feeding on those
plants. Nevertheless, in the case of Syrphidae, nothing in the literature could explain the
attractiveness towards this sesquiterpene.
As demonstrated all along this thesis, encapsulation of semiochemical molecules in alginate
gel beads is a powerful tool in order to release the volatile compounds in the air. Another
study is presently in progress in the Department of Functional and Evolutionary Entomology,
in close collaboration with the Department of Analytical Chemistry, in order to formulate
other semiochemicals in alginate beads in a new IPM program of research.
200
Chapter VII
___________________________________________________________________________
VII.2. Conclusions and perspectives
This work contributed to the development of an efficient natural formulation able to release
semiochemicals in a biological control approach in order to attract aphid predators and
parasitoids. To reach this goal, various methodologies were developed, optimised and
validated such as:
-
the purification of semiochemicals from essential oils by flash chromatography in
order to obtain compounds at high purity. This purification tool presents the
advantage to be easily transposed for the extraction of other active compounds from
various essential oils.
-
the analysis and the quantification of sesquiterpenes by ultra fast GC and the
validation of the analytical method.
-
the formulation of semiochemicals in alginate beads, and the characterisation of the
optimised formulation in terms of structure and in terms of protection of compounds
against oxidation.
-
the slow-release study of semiochemicals by means of a volatile sampling system
otpimised in terms of recovery and breakthrough.
Alginate beads presented many advantages among which the easiness of formulation on a lab
scale at low costs and the biodegradable property of the polysaccharide (Dornish et al., 1996;
Aggarwal et al., 1999; Chung et al., 2002). In future prospects, other semiochemical
compounds from other essential oils could also be formulated in alginate beads for slow
release in an integrated pest management approach.
The efficiency of formulations was demonstrated by olfactometry on aphid parasitoids. Onfield experiments from June to August 2009 also proved attractiveness of semiochemical
formulations towards Syrphidae species.
In order to formulate these semiochemical alginate beads at an industrial scale, some
experiments and developments are still needed:
-
The flash chromatography process must be adapted. Indeed, the major drawback of
the technique, as such, was the high quantity of solvent needed to elute the column.
In a prospect of industrial development at larger scale, automated flash
201
Chapter VII
___________________________________________________________________________
chromatography systems exist (Buchi Corporation®; Yamazen®; Interchim®) with a
fraction collector and minimizing solvent consumption. In an eco-friendly
perspective, a solvent recycling system (ARC Sciences® for e.g.) could be coupled
to the automated fractionation process.
-
Production of semiochemical alginate beads at larger scale is possible in an industrial
prospect. Indeed, Brandenberger et al. (1998) described an automated and modular
multinozzle encapsulation/immobilisation system (Nisco Engineering Inc.®,
Switzerland) dedicated to the production of mono-dispersed alginate beads under
reproducible conditions from 0.2 mm to 1 mm diameter. Furthermore, this tool is
equipped with a cleaning system which allows to run several formulation cycles and
to achieve productivity up to 5 L of droplets per hour. Similar industrial process
systems are proposed by Brace® GmbH (Germany) allowing bead sizes ranging
from 50 µm to 5 mm, and by GeniaLab® GmbH capable of generating up to 13000
beads per second per nozzle (up to 15 L of 1 mm diameter droplets per hour).
-
A system containing alginate beads on crops must be studied to ensure
semiochemical release over time. Ideally, the shape of this design should protect the
beads from direct humidity (rain) and from mechanical degradation. Delta trap
systems are very useful in such application. Indeed, as it was realised during the field
experiments to trap Syrphidae species in this PhD thesis, semiochemical alginate
beads were deposited in a cloth attached inside the delta trap. Beads were protected
and air was able to go through the system allowing volatiles to be released.
-
The time of degradation of alginate might be evaluated outdoors by regularly
examining field-aged semiochemical beads. In the same time, a microbiological
study of these beads could be considered in order to identify possible
microorganisms responsible of the alginate degradation over time. On the other hand,
considering the bactericide and fungicide properties of some essential oil
components, a comparison between alginate beads with and without semiochemicals
is required in terms of degradation.
-
It could be interesting to improve the knowledge about the evolution of
semiochemical release over time under environmental conditions. Such a study is
presently in progress, in close collaboration with Prof. F. Béra from the Laboratory
202
Chapter VII
___________________________________________________________________________
of Food Processing Industry (Gembloux Agro-Bio Tech, ULg), with the goal to
mathematically simulate the sesquiterpene release over time according to the abiotic
factors (temperature, relative humidity, air flow rate).
-
Field experiments could be repeated to determine the maximal distance at which
semiochemicals could be perceived by beneficial insects. To realise this study,
parasitoids and hoverflies marked with dyes could be released on fields and caught in
sticky traps at various distances from the insect release area. Secondly, the optimal
semiochemical rate at which insects are sensitive in natural environment could be
estimated by means of alginate beads formulated with various sesquiterpene
quantities.
-
Finally, the success over time of the biological control approach to reduce aphid
populations by attracting insect predators and parasitoids will depend on the
development of other agri-environmental methods in surrounding crops. Such
measures are integrated in agriculture to protect environment, fauna and flora. It is
supposed that by these practices, density of aphid predators and parasitoids will
increase.
To close this chapter, I would like to highlight the importance of E-β-farnesene in the biofuel
industry in order to replace petroleum-derived fuels, as recently related by Ritter (2011). The
author describes the process developed by the Amyris Company to produce E-β-farnesene
obtained from yeasts after sugar fermentation. The purification of the molecule is very
simple: E-β-farnesene having low solubility in water, it separates quickly from the
fermentation broth without the need of distillation. The purified sesquiterpene is then
hydrogenated to farnesane which can be used as biodiesel. In future prospects, this way of Eβ-farnesene purification could replace the essential oil fractionation process, providing that
the sesquiterpene is producted with a high purity and at lower costs.
203
Chapter VII
___________________________________________________________________________
References
Agelopoulos N., Birkett M.A., Hick A.J., Hooper A.M., Pickett J.A., Pow E.M., Smart L.E.,
Smiley D.W.M., Wadhams L.J., Woodcock C.M. (1999). Exploiting semiochemicals in
insect control. Pest. Sci. 55, 225-235.
Aggarwal N., HogenEsch H., Guo P., North A., Suckow M., Mittal S. K. (1999).
Biodegradable alginate microspheres as a delivery system for naked DNA. Can. J. Vet.
Res. 63(2), 148-152.
Bicchi C., Brunelli C., Cordero C., Rubiolo P., Galli M., Sironi A. (2004). Direct resistively
heated column gas chromatography (Ultrafast module-GC) for high-speed analysis of
essential oils of differing complexities. J. Chrom. A 1024, 195-207.
Bicchi C., Brunelli C., Cordero C., Rubiolo P., Galli M., Sironi A. (2005). High-speed gas
chromatography with direct resistively-heated column (ultra fast module-GC)separation measure (S) and other chromatographic parameters under different analysis
conditions for samples of different complexities and volatilities. J. Chrom. A 1071, 312.
Brandenberger H., Widmer F. (1998). A new mutlinozzle encapsulation/immobilisation
system to produce uniform beads of alginate. J. Biotech. 63, 73-89.
Brown A.E., Riddick E.W., Aldrich J.R., Holmes W.E. (2006). Identification of (-)-βcaryophyllene as a gender-specific terpene produced by the multicoloured Asian lady
beetle. J. Chem. Ecol. 32, 2489-2499.
Brown P.M.J., Adriens T., Bathon H., Cuppen J., Goldarazena A., Hägg T., Kenis M.,
Klausnitzer B.E.M., Loomans A.J.M., Majerus M.E.N., Nedved O., Pedersen J.,
Rabitsch W., Roy H.E., Ternois V., Zakharov I.A., Roy D.B. (2008). Harmonia
axyridis in Europe: spread and distribution of a non-native coccinellid. BioControl,
53(1), 5-21.
Chang C.P., Dobashi T. (2003). Preparation of alginate complex capsules containing
eucalyptus essential oil and its controlled release. Colloids and Surf. B. 32, 257-262.
Chung T.W., Yang J., Akaike T., Cho K.Y., Nah J.W., Kim S.I., Cho C.S. (2002).
Preparation of alginate/galactosylated chitosan scaffold for hepatocyte attachment.
Biomaterials 23, 2827-2834.
204
Chapter VII
___________________________________________________________________________
Dornish M., Aarnold M., Skaugrud. (1996). Alginate and chitosan: biodegradable
biopolymers in drug delivery systems. Eur. J. Pharm. Sci. 4, S153.
Du Y., Poppy G.M., Powell W., Pickett J.A., Wadhams L.J., Woodcock C.M. (1998).
Identification of Semiochemicals Released During Aphid Feeding That Attract
Parasitoid Aphidius ervi. J. Chem. Ecol. 24,1355-1368.
Francis F., Vandermoten S., Verheggen F., Lognay G., Haubruge E. (2005a). Is the (E)-ßfarnesene only volatile terpenoids in aphids? J. Applied Ento. 129, 6-11.
Francis F., Martin T., Lognay G., Haubruge E. (2005b). Role of (E)-β-farnesene in
systematic aphid prey location by Episyrphus balteatus larvae (Diptera: Syrphidae).
Eur. J. Entomol. 102, 431-436.
Glinwood R.T., Du Y.-J., Powell W. (1999). Responses to aphid sex pheromones by the pea
aphid parasitoids Aphidius ervi and Aphidius eadyi. Entomol. Exp. Appl. 92, 227-232.
Godin B. (2008). Contribution à la production et à l’étude du relargage de substances
volatiles sémiochimiques dans le cadre de la lutte biologique contre les pucerons.
Faculté Universitaire des Sciences Agronomiques de Gembloux, pp 75.
Hambleton A., Voilley A., Debeaufort F. (2010). Transport parameters for aroma compounds
through i-carrageenan and sodium alginate-based edible films. Food Hydrocol., in
press.
Haug A., Smidsrod O. (1970). Selectivity of some anionic polymers for divalent metal ions.
Acta Chem. Scand. 24, 843-854.
Horwitz W., Kamps L.A, Boyer K.W. (1980). Quality assurance in the analysis of foods for
trace constituents. J. Assoc. Off. Anal. Chem. 63, 1344-1354.
Hubert Ph., Chiap P., Crommen J., Boulanger B., Chapuzet E., Mercier N., Bervoas-Martin
S., Chevalier P., Grandjean D., Lagorce P., Lallier M., Laparra M.C., Laurentie M.,
Nivet J.C. (1999). The SFSTP guide on the validation of chromatographic methods for
drug bioanalysis: from the Washington Conference to the laboratory. Anal. Chim. Acta
391, 135-148.
Hubert Ph., Ngyen-Huu J.J., Boulanger B., Chapuzet E., Chiap P., Cohen N., Compagnon
P.A., Dewé W., Feinberg M., Lallier M., Laurentie M., Mercier N., Muzard G., Nivet
205
Chapter VII
___________________________________________________________________________
C., Valat L. (2003). Validation des procedures analytiques quantitatives. Harmonisation
des demarches. STP Pharma Pratiques 13, 101-138.
Hubert Ph., Ngyen-Huu J.J., Boulanger B., Chapuzet E., Cohen N., Compagnon P.A., Dewé
W., Feinberg M., Laurentie M., Mercier N., Muzard G., Valat L. (2003). Validation des
procedures analytiques quantitatives: Harmonisation des demarches. Partie II –
Statistiques. STP Pharma Pratiques 16, 28-58.
Kakita H., Kamishima H. (2008). Some properties of alginate gels derived from algal sodium
alginate. J. Applied Phycol. 20, 93-99.
Kong X.Z., Gu X., Zhu X., Zhang Z. (2009). Spreadable dispersion of insect sex pheromone
capsules, preparation via complex coacervation and release control of the encapsulated
pheromone component molecule. Biomed. Microdevices 11, 275-285.
Lai F., Loy G., Manconi M., Manca M.L., Fadda A.M. (2007). Artemesia arborescens L.
essential oil loaded beads: preparation and characterization. AAPS PharmSciTech. 8,
E1-E7.
Liu X.D., Yu W.Y., Zhang Y., Xue W.M., Yu W.T., Xiong Y., Ma X.J., Chen Y., Yuan Q.
(2002). Characterization of structure and diffusion behaviour of Ca-alginate beds
prepared with external or internal calcium sources. J. Microencapsulation 19, 775-782.
Madene A., Jacquot M., Scher J., Desobry S. (2006). Flavour encapsulation and controlled
release – a review. Int. J. Food Sci. Tech. 41, 1-21.
Malone M.E., Appelqvist I.A.M. (2003). Gelled emulsion particles for the controlled release
of lipophilic volatiles during eating. J. Contr. Release. 90, 227-241.
Michaud J.P., Slodebeck P.E., Nechols J.R. (2008). Biological control of insect pests on field
crops in Kansas. Kansas State University, Agricultural Experiment Station and
Cooperative Extension Service, 1-12.
Miller D.R., Lindgren B.S. (2000). Comparison of a-pinene and myrcene on attraction of
mountain pine beetle, Dendroctonus ponderosae (Coleoptera: Scolytidae) to
pheromones in stands of western white pine. J. Entomol. Soc. Brit. Columbia 97, 41-46.
Pechous S.W., Whitaker B.D. (2004). Cloning and functional expression of an (E,E)-afarnesene synthase cDNA from peel tissue of apple fruit. Planta 219, 84-94.
206
Chapter VII
___________________________________________________________________________
Poland T.M., de Groot P., Haack R.A., Czokajlo D. (2004). Evaluation of semiochemicals
potentially synergistic to α-pinene for trapping the larger European pine shoot beetle,
Tomicus piniperda (Col., Scolytidae). JEN 128, 639-644.
Powell W., Pickett J.A. (2003). Manipulation of parasitoids for aphid pest management:
progress and prospects. Pest. Manag. Sci. 59, 149-155.
Quong D., Neufeld R., Skjak-Braek G., Poncelet D. (1998). External versus internal source
of calcium during the gelation of alginate beads for DNA encapsulation. Biotech.
Bioengineer. 57, 438-446.
Reed G.L. (1999). Fast GC: Applications and theoretical studies. PhD thesis. Faculty of
Virginia Polytechnic Institute and State University, 150 pp.
Ritter S.K. (2011). Race to the pump. Biofuel technologies vie to provide a sustainable
supply of transportation fuels. C&EN February 2011, 11-17.
Rojas-Graü M.A., Tapia M.S., Rodriguez F.J., Carmona A.J., Martin-Belloso O. (2007).
Alginate and gellan-based edible coatings as carriers of antibrowning agents applied on
fresh-cut Fuji apples. Food Hydrocoll. 21, 118-127.
Rowan D.D., Allen J.M., Fielder S., Spicer J.A., Brimble M.A. (1995). Identification of
conjugated triene oxidation products of α-farnesene in apple skin. J. Agric. Food Chem.
43, 2040-2045.
Sasso R., Iodice L., Woodcock C.M., Pickett J.A., Guerrieri E. (2009). Electrophysiological
and behavioural responses of Aphidius ervi (Hymenoptera: Braconidae) to tomato plant
volatiles. Chemoeco. 19, 195-201.
Skjak-Braek G., Grasdalen H., Smidsrod O. (1989). Inhomogeneous polysaccharide ionic
gels. Carbohydrate Poly. 10, 31-54.
Smidsrod O., Grasdalen H. (1984). Polyelectrolytes from seaweeds. Hydrobiologia 116, 1928.
Szöke E., Maday E., Marczal G., Lemberkovics E. (2003). Analysis of biologically active
essential oil components of chamomiles in Hungary (In vivo – In vitro). Acta Hortic.
597, 275-284.
207
Chapter VII
___________________________________________________________________________
Vandenberg G.W., De La Noüe J. (2001). Evaluation of protein release from chitosanalginate microcapsules produced using external or internal gelation. J. Microencaps.
18, 433-441.
Velings N., Mestdagh M.M. (1995). Physico-Chemical Properties of Alginate Gel Beads.
Polymer Gels Networks 3, 311-330
Velings N., Ouwerx C., Mestdagh M., Axelos M. (1996). Physico-chemical properties of
alginate gel beads and their incidence on retention/release of proteins. Proc.
International Worskshop on Bioencapsulation, Talk 37, September 1996.
Verheggen F.J., Fagel Q., Heuskin S., Lognay G., Francis F., Haubruge E. (2007).
Electrophysiological and behavioral responses of the multicolored Asian lady beetle,
Harmonia axyridis Pallas, to sesquiterpenes semiochemicals. J. Chem. Ecol. 33, 21482155.
Verheggen F.J., Arnaud L., Bartram S., Gohy M., Haubruge E. (2008). Aphid and plant
volatiles induce oviposition in an aphidophagous Hoverfly. J. Chem. Ecol. 34, 301-307.
Verheggen F.J., Haubruge E., De Moraes C.M., Mescher M.C. (2009). Social environment
influences aphid production of alarm pheromone. Behav. Ecol. 20, 283-288.
Verheggen F.J., Haubruge E., Mescher M.C. (2010). Alarm pheromones. In:Litwack G., ed.,
Pheromones. Elsevier.
Whitaker B.D., Solomos T., Harrison D.J. (1997). Quantification of α-farnesene and its
conjugated trienol oxidation products from apple peel by C18-HPLC with UV detection.
J. Agric. Food Chem. 45, 760-765.
Yosha I., Shani A., Magdassi S. (2008). Slow release of pheromones to the atmosphere from
gelatin-alginate beads. J. Agric. Food Chem. 56, 8045-8049.
208
Chapter VIII
List of scientific productions
Chapter VIII
___________________________________________________________________________
List of scientific productions
The following scientific productions were realised during this PhD thesis.
1. Publications
1. Heuskin S., Godin B., Leroy P., Capella Q., Wathelet J.-P., Verheggen F.,
Haubruge E., Lognay G. (2009). Fast Gas Chromatography characterisation of
purified semiochemicals from essential oils of Matricaria chamomilla L.
(Asteraceae) and Nepeta cataria L. (Lamiaceae). Journal of Chromatography A,
1216: 2768 - 2775.
2. Heuskin S., Rozet E., Lorge S., Farmakidis J., Hubert Ph., Verheggen F., Haubruge
E., Wathelet J.-P., Lognay G. (2010). Validation of a fast gas chromatographic
method for the study of semiochemical slow release formulations. Journal of
Pharmaceutical and Biomedical Analysis 53: 962-972.
3. Heuskin S., Haubruge E., Verheggen F., Wathelet J.-P., Lognay G. (2010). The use
of semiochemical slow-release devices in integrated pest management strategies.
Accepted for publication in Biotechnologie Agronomie Société et Environnement.
4. Heuskin S., Lorge S., Godin B., Leroy P., Frère I., Verheggen F.J., Haubruge E.,
Wathelet J.-P., Mestdagh M., Hance Th., Lognay G. (2010). Optimisation of a
semiochemical slow-release alginate formulation attractive towards Aphidius ervi
Haliday parasitoids. Accepted for publication in Pest Management Science.
5. Heuskin S., Béra F., Lorge S., Kilinc A., Leroy P., Haubruge E., Wathelet J.-P.,
Brostaux Y., Lognay G. (2010). A semiochemical slow-release formulation in a
biological control approach to attract hoverflies. Submitted to Entomologia
Experimentalis et Applicata.
2. Oral presentations
1. Heuskin S., Farmakidis J., Lorge S., Debatty-Mestdagh M., Wathelet J.-P., Lognay
G. (2009). Flash presentation: Contribution to the study of semiochemical slow
211
Chapter VIII
___________________________________________________________________________
release formulations. Development of flash chromatographic methods. Meeting of
the CHIM Doctoral School, Namur, Belgium, 24th April 2009.
2. Heuskin S., Lorge S., Wathelet J.P., Verheggen F., Lognay G. (2010). Biological
control formulations incorporating essential oil components. 41st International
Symposium on Essential Oils (ISEO), Wroclaw, Poland, 5-8 September 2010.
3. Posters
1. Heuskin S., Lognay G. (2007). Lutte biologique par l’utilisation de médiateurs
chimiques : apport des techniques chromatographiques. Journée Découverte
Entreprise, FUSAGx, Gembloux, Belgium, 7th October 2007.
2. Heuskin S., Leroy P., Capella Q., Godin B., Wathelet J.-P., Verheggen F.,
Haubruge E., Lognay G. (2008). Fast purification and characterisation of
semiochemical compounds from essential oils of Matricaria chamomilla L. and
Nepeta cataria L. using column chromatography fractionation and Ultra Fast GC
analysis. 32nd International Symposium on Capillary Chromatography, Riva del
Garda, Italy, 25-30 May 2008.
3. Heuskin S., Farmakidis J., Lorge S., Debatty-Mestdagh M., Wathelet J.-P., Lognay
G. (2009). Contribution to the study of semiochemical slow release formulations.
Development of flash chromatographic methods. Meeting of the CHIM Doctoral
School, Namur, Belgium, 24th April 2009.
4. Heuskin S., Lorge S., Leroy P., Verheggen F., Wathelet J.-P., Lognay G. (2010). A
fast gas chromatographic method for the study of semiochemical slow release
formulations. 11th International Symposium on Hyphenated Techniques in
Chromatography and Hyphenated Chromatographic Analyzers (HTC-11), Bruges,
Belgium, 27-29 January 2010.
5. Heuskin S., Lorge S., Debatty-Mestdagh M., Wathelet J.-P., Lognay G. (2010).
Optimisation of slow-release formulations as biological control devices. One-Day
Symposium on Chemical Entomology (150th Anniversary of Gembloux University),
Gembloux Agro-Bio Tech – ULG, Gembloux, Belgium, 5th May 2010.
212
Chapter VIII
___________________________________________________________________________
6. Heuskin S., Lorge S., Verheggen F.J., Haubruge E., Mestdagh M., Wathelet J.-P.,
Lognay G. (2010). Development of biological control formulations incorporating
components of plant origin. Journée Scientifique Annuelle de la SRC, Gembloux,
Belgium, 14th October 2010.
7. Heuskin S., Lorge S., Leroy P., Haubruge E., Mestdagh M., Wathelet J.-P., Lognay
G. (2011). Development and optimisation of formulations incorporating essential
oil components as biological control devices against aphids. Global Conference on
Entomology, Chiang Mai, Thailand, 5-9 March 2011.
213
List of figures
List of figures
__________________________________________________________________________
List of figures
Chapter II
Figure 1: Semiochemicals in classes…………………………………………………………… 35
Chapter III
Figure 1: Methodology of the research………………………………………………………… 58
Chapter IV
Figure 1’: Essential oil characterisation and fractionation........................................................... 62
Figure 1: Chromatogram of reference compounds and internal standards analysed with optimised
UFGC method............................................................................................................................... 71
Figure 2: Number of theoretical plates (N) in function of the quantity of compounds injected on the
Ultra Fast Module column. (A: E-β-farnesene, B: β-caryophyllene, C: longifolene, D: n-butylbenzene, E: limonene, F: α-pinene)............................................................................................... 75
Figure 3: Van Deemter plots for Ultra Fast GC (A: α-pinene, B: limonene, C: n-butyl-benzene, D:
longifolene, E: β-caryophyllene, F: E-β-farnesene).....................................................................
75
Figure 4: Profiles of a Matricaria chamomilla essential oil analysed by GC-MS (a) and Ultra Fast
GC (b). For analysis conditions see text. List of the main components: (1) E-β-farnesene; (2)
germacrene D; (3) bicyclogermacrene; (4) (E,E)-α-farnesene; (5) α-bisabolol oxide B; (6) αbisabolone
oxide
A;
(7)
chamazulene;
(8)
α-bisabolol
oxide
A;
(9)
cis-ene-yne-
dicycloether................................................................................................................................... 79
Figure 5: Profiles of a Nepeta cataria essential oil analysed by GC-MS (a) and Ultra Fast GC (b).
For analysis conditions see text. List of the main components: (1) (Z,E)-nepetalactone; (2) (E,Z)nepetalactone; (3) β-caryophyllene; (4) unknown compound; (5) β-caryophyllene oxide........... 79
Chapter V
Figure 1’: Accuracy profile validation procedure……………………………………………… 88
Figure 2’: Steps of the procedure to measure the protection efficiency of formulations towards
sesquiterpenes…………………………………………………………………………………… 89
Figure 1: Chromatograms of analytes (E-β-farnesene at 81.6 ng µL-1 and β-caryophyllene at 80.5 ng
µL-1) and internal standard (longifolene at 102.6 ng µL-1) (a) and of a blank alginate beads matrix
sample (b) analysed with optimised fast GC method…………………………………………… 96
217
List of figures
__________________________________________________________________________
Figure 2: Accuracy profiles of E-β-farnesene obtained by considering a simple linear regression
model (a) and by considering a linear regression model through zero fitted with the maximum level of
concentration (b); plain line: relative bias, dashed lines: β-expectation tolerance limits (β=95%),
dotted curves: acceptance limit (+/-25% and +/-15%) and dots: relative back-calculated
concentrations of the validation standards……………………………………………………… 104
Figure 3: Accuracy profiles of β-caryophyllene obtained by considering a simple linear regression
model (a) and considering a linear regression model through zero fitted with the maximum level of
concentration (b) without correction of the bias. Accuracy profiles of β-caryophyllene obtained by
considering a simple linear regression model (c) and considering a linear regression model through
zero fitted with the maximum level of concentration (d) with a correcting factor of the bias. Plain line:
relative bias, dashed lines: β-expectation tolerance limits (β=95%), dotted curves: acceptance limit
(+/-25% and +/-15%) and dots: relative back-calculated concentrations of the validation
standards………………………………………………………………………………………… 105
Figure 4: Risk profiles of E-β-farnesene (a) and β-caryophyllene (b) for the chosen regression
models. Dotted line: maximum risk of 5%; dashed line: effective risk of having results falling outside
the specified acceptance limits………………………………………………………………….. 106
Figure 5: Linearity profiles of (a) E-β-farnesene and (b) β-caryophyllene (after correction of the
results). Plain line: identity line (Y=X), dashed lines: β-expectation tolerance limits (β=95%), dotted
curves: acceptance limit expressed in ng µL-1 and dots: back-calculated concentrations of the
validation standards……………………………………………………………………………... 109
Figure 6: Residual percentage of E-β-farnesene (a) and β-caryophyllene (b) in formulations during
twenty days……………………………………………………………………………………… 114
Chapter VI
Figure 1’: Steps of the alginate bead formulation optimisation………………………………... 128
Figure 2’: Alginate bead formulation process………………………………………………….. 129
Figure 3’: Volatile collection system…………………………………………………………... 130
Chapter VI.1
Figure 1: Mannuronnate (Poly M), guluronate (Poly G) and Poly MG blocks constitutive of alginate
polymer………………………………………………………………………………………….. 134
Figure 2: Encapsulation rate (in g) of E--farnesene in 100 mg of beads (sticks) and resistance force
(in N) (spots) according to 3 types of alginate (Satialgine, Sigma L, Sigma M)……………….. 145
218
List of figures
__________________________________________________________________________
Figure 3: Encapsulation rate (in g) of E--farnesene in 100 mg of beads (sticks) and resistance force
(in N) (spots) according to the type of cross-linker ion (Ca2+ or Cu2+)…………………………. 145
Figure 4: Cumulative quantity (in g) of E--farnesene released from 100 mg of 0.5 M ionic strength
alginate (sigma L) beads (a) and from 100 mg of 1.0 M ionic strength alginate (sigma L) beads
(b)………….................................................................................................................................. 146
Figure 5: Encapsulation rate (in g) of E--farnesene in 100 mg of alginate beads according to the
Ca2+ / COO- ratio, for maturation times of 20 min. () and 48 h ()…..……………………... 147
Figure 6: Resistance force (in N) of beads for different alginates (Sigma L, Sigma M, Satialgine) at
various concentrations (1.0 %; 1.5 %; 2.0 %; 2.5 %)…………………………………………… 148
Figure 7: Resistance force (in N) of Sigma L alginate beads according to the Ca 2+ / COO- ratio for
different alginate concentrations. Bold circle represents the best formulation in terms of appearance
and homogeneity of beads………………………………………………………………………. 149
Figure 8: CLSM imaging of a dried (Aw=0.42) E--farnesene alginate bead (a) and of a wet
(Aw=0.98) E--farnesene alginate bead (b)…………………………………………………….. 150
Figure 9: Cumulative released quantity (in g) of E--farnesene (a) and of E--caryophyllene (b)
from 100 mg of optimised formulation during 35 days in laboratory controlled conditions (20°C; 65%
relative humidity; airflow at 0.5 L min-1)……………………………………………………….. 152
Figure 10: Percentages of female A. ervi in both arms of Y-tube olfactometer, for 500 mg E-caryophyllene alginate beads, 500 mg and 1500 mg E--farnesene alginate beads……………. 154
Chapter VI.2
Figure 1’: Different steps of the semiochemical slow-release study according to abiotic factors and of
the field bioassays………………………………………………………………………………. 166
Figure 1: Cumulative mass released (in µg) of semiochemical (E-β-farnesene (a) and E-βcaryophyllene (b)) from alginate beads (200 mg) under eight different experimental conditions.. 177
219
List of tables
List of tables
__________________________________________________________________________
List of tables
Chapter II
Table 1: Development of semiochemical dispensers and formulations, and release rate studies... 46
Chapter IV
Table 1: Purity of reference compounds………………………………………………………... 67
Table 2: Linearity data for the fast GC validation method……………………………………... 73
Table 3: Precision of the fast GC method expressed as repeatability and reproducibility……... 73
Table 4: Selectivity of the fast GC method……………………………………………………... 74
Table 5: Constituents of the essential oil of Matricaria chamomilla identified by GC-MS…… 77
Table 6: Constituents of the essential oil of Nepeta cataria identified by GC-MS…………….. 78
Table 7: Major components of Matricaria chamomilla fractions (mean ± SD of triplicate)…... 82
Table 8: Major components of Nepeta cataria fractions (mean ± SD of triplicate)……………. 82
Chapter V
Table 1: Purity of compounds analysed by fast GC……………………………………………. 95
Table 2: Levels of calibration standards for E-β-farnesene and β-caryophyllene……………… 100
Table 3: Levels of validation standards for E-β-farnesene and β-caryophyllene………………. 101
Table 4: Validation results for E-β-farnesene and β-caryophyllene for the chosen regression
models…………………………………………………………………………………………... 110
Table 5: Estimates of the measurement uncertainties related to E-β-farnesene and β-caryophyllene, at
each concentration level investigated in validation using the selected regression models……... 113
Chapter VI.1
Table 1: Sodium alginates with their characteristics (Mannuronate/Guluronate ratio and molar
mass)…………………………………………………………………………………………….. 136
Table 2: Experimental design n°1……………………………………………………………… 138
Table 3: Experimental design n°2……………………………………………………………… 139
223
List of tables
__________________________________________________________________________
Chapter VI.2
Table 1: Weather conditions from June to August 2009 presented with means, standard deviations,
minimum and maximum values of temperature (°C) and relative humidity (%)……………….
173
Table 2: Laboratory experimental conditions to determine release rates and diffusion coefficients for
E-β-farnesene and E- β-caryophyllene………………………………………………………….. 175
Table 3: Diffusion coefficients of E-β-farnesene and E- β-caryophyllene obtained after 1 to 4 water
absorption-desorption cycles……………………………………………………………………. 178
Table 4: Catches of female Syrphidae (mean ± SD) per formulation and per crop with the statistical
data……………………………………………………………………………………………… 179
224