THE ENUMERATION AND IDENTIFICATION
OF RHIZOBIAL BACTERIA IN LEGUME
INOCULANT QUALITY CONTROL
PROCEDURES
June, 1996
Perry E. Olsen
Eve S. Sande
Harold H. Keyser
NifTAL Center
1000 Holomua Road
Paia, HI 96779 USA
ACKNOWLEDGEMENTS
Much of the material in Chapter 5 dealing with the enumeration of rhizobia using the mostprobable-number (MPN) technique was originally collated and written by Paul Woomer while at
the NifTAL center. All of the figures, with the exception of Figure 7, have previously appeared in
the Handbook for Rhizobia by Padma Somasegaran and Heinz Hoben. Skillful assistance in the
preparation of this manuscript was provided by Patty Nakao and Sally Ekdahl.
i
THE ENUMERATION AND IDENTIFICATION OF RHIZOBIAL
BACTERIA IN LEGUME INOCULANT QUALITY CONTROL
PROCEDURES
PREFACE
The purpose of this handbook is to provide to small producers of legume inoculant a
reliable guide to the methods and procedures useful in the analysis both of rhizobial broth and
finished inoculant products. The objective is to enable the manufacturer to identify and enumerate
rhizobia (Rhizobium, Bradyrhizobium, Azorhizobium, etc.) and thereby to provide the
informational framework essential to the production of high quality inoculant. We have also
provided a referenced discussion which we hope will provide a current scientific perspective of
legume inoculant for those who are relatively new to the subject. The handbook is intended,
however, as a practical aid towards inoculant quality evaluation and not as a comprehensive review
of either techniques or theory.
Implementation of the techniques described in this handbook requires that a basic level of
facilities be devoted to the analytical function. The minimum requirement is the establishment of a
basic microbiology laboratory and either dedicated plant growth chambers or a clean room
equipped for the axenic growth of plants. Both functions need to be separated from the dust and
bacteria of the production area.
Effective microbiological analysis requires technically qualified personnel. Legume
inoculant analysis should be performed, or closely supervised by, a trained microbiologist. In
order to obtain consistently meaningful results, the technical work also needs to be performed by
qualified personnel. Organizations such as NifTAL, ICARDA, ICRISAT, and CIAT provide
excellent training courses for individuals working in the area of biological nitrogen fixation. Those
who have received such training will be familiar with much of the material provided in this
handbook. Nevertheless, this guide provides a useful compilation, in a single source, of
information and techniques which can be used in the implementation and conduct of inoculant
quality control programs.
ii
TABLE OF CONTENTS
A C K N O W L E D G E M E N T S ................................................................................................... i
P R E F A C E .................................................................................................................................... ii
L I S T O F T A B L E S ................................................................................................................. v
L I S T O F F I G U R E S ............................................................................................................... vi
A B B R E V I A T I O N S U S E D I N T E X T .............................................................................vii
CHAPTER 1
I N O C U L A N T Q U A L I T Y C O N T R O L ..........................................................................
Inoculant Quality in Relation to Inoculant Performance ....................................................
Importance of cell numbers....................................................................................
Sterile versus non-sterile carrier .............................................................................
A Perspective on Standards for Rhizobial Inoculants..........................................................
Inoculant Quality Control Testing ......................................................................................
Quality control testing of non-sterile carrier inoculant............................................
Quality control testing of sterile carrier inoculant...................................................
Quality in the Culture Collection........................................................................................
1
2
2
3
3
4
5
5
5
CHAPTER 2
M I C R O B I O L O G I C A L T E S T I N G O F I N O C U L A N T C A R R I E R ...................... 8
Carrier Sterility................................................................................................................... 8
Tests for Carrier Sterility.................................................................................................... 9
Carrier-Rhizobia Compatibility Test.................................................................................. 11
CHAPTER 3
M E T H O D S O F E V A L U A T I N G R H I Z O B I A L B R O T H Q U A L I T Y ............... 13
Growth Rate....................................................................................................................... 13
Microscopic Examination of Broth.................................................................................... 13
Microscopic Count of Total Cells in Broth ........................................................... 15
Gram Stain for Contamination in Broth ............................................................................ 18
The Peptone-Glucose Test for Contamination in Broth .................................................... 19
CHAPTER 4
M E T H O D S O F I N O C U L A N T A N A L Y S I S................................................................
S
21
Viable Cell Counts............................................................................................................. 21
iii
Serial Dilutions and Plating.................................................................................... 21
The Spread Plate Technique ............................................................................................. 25
The Drop Plate Technique................................................................................................ 27
CHAPTER 5
ENUMERATION OF RHIZOBIA IN INOCULANT USING THE
M O S T - P R O B A B L E - N U M B E R ( M P N ) P L A N T - I N F E C T I O N A S S A Y .......... 29
MPN Overview.................................................................................................................. 29
Basic Assumptions of the MPN Plant-Infection Test......................................................... 30
Designing an MPN Assay .................................................................................................. 30
A Typical MPN Design for Inoculant Evaluation . . . . . . . . . . . . . . . . . . . . .
33
Conducting the MPN Assay............................................................................................... 34
Results and Analysis .......................................................................................................... 45
CHAPTER 6
IMMUNOLOGICAL
TECHNIQUES
FOR
THE
ANALYSIS
OF
R H I Z O B I A L B R O T H A N D I N O C U L A N T S ............................................................ 57
General Comments ........................................................................................................... 57
The Cell Agglutination Reaction for Rhizobial Identity...................................................... 58
Immuno-Spot Blot Test for Rhizobial Identity................................................................... 61
Indirect Fluorescent Antibody Identification of Rhizobia In Broth.................................... 66
Colony-Lift Immunoblot (Membrane ELISA) for Inoculant Analysis............................... 70
Direct Fluorescent Antibody Enumeration of Rhizobia in Inoculant ................................. 76
CHAPTER 7
P O T E N T I A L T E C H N I Q U E S F O R I N O C U L A N T A N A L Y S I S........................
S
79
CHAPTER 8
M A N U F A C T U R E R P R O T O C O L S F O R A N A L Y S I S ............................................. 80
R E F E R E N C E S ......................................................................................................................... 84
A P P E N D I X I - M E D I A A N D R E A G E N T F O R M U L A T I O N S ........................... 91
A P P E N D I X I I - M A T E R I A L S S O U R C E S .............................................................. 95
iv
LIST OF TABLES
1. Details of the Petroff-Hausser counting chamber ................................................
17
2. YEMA rhizobial plate-count media ..............................................................
23
3. Concentrated stock solutions for preparation of YEMB .....................................
24
4. A simpler rhizobial plate-count media .........................................................
24
5. Factors for calculating the confidence intervals of
Most-Probable-Number estimates ...............................................................
32
6. Species of Root and Stem-Nodulating Bacteria and Their Hosts ..........................
35
7. Advantages and disadvantages of plant containers recommended
for use in MPN counts of rhizobia .............................................................
37
8. Plant nutrient solution composition (1) .........................................................
38
9. Plant nutrient solution composition (2) .........................................................
39
10. Commonly practiced methods for surface sterilization and scarification of seeds ....
41
11. Legume hosts commonly requiring seed scarification and approximate exposure
times to concentrated sulfuric acid for uniform germination...............................
42
12. Example of results of an MPN assay ..........................................................
46
13. Expected frequencies of equalling or exceeding the ROT for various
dilution ratio and replicate number combinations ............................................
47
14. MPN table for two replications .................................................................
50
15. MPN table for three replications ...............................................................
51
16. MPN table for four replications.................................................................
52
17. MPN table for five replications .................................................................
53
18. Comparisons between MPN and plate count enumerations of rhizobia .................
56
v
LIST OF FIGURES
1. Petroff-Hausser counting chamber ...............................................................
16
2. Procedure for a 10-fold dilution series ..........................................................
26
3. Enumeration of rhizobia by the drop plate method ..........................................
27
4. Wire rack for holding growth pouches .........................................................
36
5. Positive and negative agglutination reactions in microtiter plate wells ...................
59
6. Schematic diagram of the indirect immunofluorescence method ..........................
67
7. Immunoblot membranes and corresponding colonies on spread plates ...................
71
8. Schematic diagram of direct immunofluorescence membrane technique .................
77
vi
A B B R E V I A T I O N S U S E D I N TEXT
BCIP
BCP
BTB
bv
ELISA
EtOH
FA
FITC
g
Ig
mL
MPN
MPNES
N
NBT
PBS
PBST
ROT
ìL
ìM
UV
YEMB (YMB)
YEMA (YMA)
5-bromo-4-chloro-3-indolyl phosphate
Bromcresol purple
Bromthymol blue
biovar
enzyme-linked immunosorbent assay
ethanol
fluorescent antibody
fluorescein isothiocyanate
gram
immunoglobulin
milliliter (1.00 mL = 1000 uL)
most-probable-number
most-probable-number enumeration system
nitrogen
nitroblue tetrazolium
phosphate buffered saline
phosphate buffered saline plus 0.05% Tween 20
range of transition
microliter (0.001 mL)
micrometer
ultraviolet
yeast-extract mannitol broth
yeast-extract mannitol agar
vii
CHAPTER 1
INOCULANT QUALITY CONTROL
The key to high quality legume inoculant is an effective quality control system (Thompson,
1984). A keen desire to sell product, combined with the expense of internal quality control
systems, has left many manufacturers unable or unwilling to institute quality control procedures
(Burton, 1978). Because farmers cannot judge the quality of inoculant products at the time of
purchase, there is often even less incentive for an inoculant producer to institute quality control
programs (Thompson, 1991a), with the result that some producers will sell poor quality inoculants
(Beck et al. 1993; Brockwell and Bottomley, 1995; Olsen et al., 1996). When, however, farmers
see no benefit in the field from inoculation, their confidence in the technology is lost, and both the
manufacturers and the consumers are ultimate losers. For this reason, it is desirable that the
quality of legume inoculant be protected by some sort of governmental regulation (Burton, 1978;
FAO report, 1991). Such regulation may be voluntary, as in Australia, New Zealand and South
Africa, or legislative, as in France and Canada. Where such programs have been implemented
and properly funded, inoculant quality has risen, and farmers and manufacturers are the ultimate
winners. The summary of the FAO Report on the Expert Consultation on Legume Inoculant
Production and Quality Control (Rome, 1991) states that strong legislation coupled with a
government testing program in France and Canada has "insured high quality of local inoculants"
and "encouraged exporters to these countries to raise their standards to this level, thus benefiting all
recipients of such cultures." The expert group strongly recommended "setting up quality control
bodies on the basis of legislation, registration or mutual agreement between producers and the
responsible authorities" and "provision of the control body with ability to set realistic standards and
the authority to implement and enforce them."
Still, in most areas of the world, government programs for the regulation of inoculant
products either do not exist, or are inadequately funded and motivated. In such situations,
maintaining high quality inoculant products depends on the internal implementation of effective
quality control systems by manufacturers. Without a quality control program, the manufacturer is
unlikely to have any better information about the quality of his product than does the customer.
Unfortunately, much of the inoculant produced in the world today is of relatively poor quality, and
some of it is extremely poor (Somasegaran, 1991; Brockwell and Bottomley, 1995). There have
been recent examples of inoculants sold in both developed and developing countries that did not
contain any rhizobia at all (Thompson, 1991a; Olsen et al., 1995; Olsen et al., 1996).
The rhizobial cells in a superior powdered peat inoculant containing 2 x 109 rhizobia per
gram have been estimated to constitute only about 0.13% of the total volume of the inoculant.
There is clearly room for a further 10-fold increase in cell number (Brockwell and Bottomley,
1995). Whereas advanced research technology can anticipate improving even such superior
products by a factor of ten, implementation of relatively simple quality control programs is the
most direct and immediate way to prevent poor or worthless inoculant from reaching the market.
The techniques outlined in this manual are offered in the hope that more inoculant producers will
implement them, both in their own self-interest and in the interest of farming communities
generally.
1
INOCULANT
QUALITY
PERFORMANCE
IN
RELATION
TO
INOCULANT
Importance of cell numbers
A variety of factors combine to create "quality" in a legume inoculant. These factors include
incorporation of a superior rhizobial strain, proper packaging, clear labelling with instructions for
use on each package, a formulation that is effective and easy to apply, adequate shelf life of the
product, and freedom from excessive contamination. A factor of primary concern is that the
highest possible number of live rhizobia capable of nodulating and fixing nitrogen with the target
host be present in the product. Evaluation of inoculant quality by enumerating the viable rhizobia
present is an accurate index of inoculating potential (Hitbold, 1980). Numerical considerations are
of such significance in determining the effectiveness of inoculant that the need for quality control
cell enumeration systems is widely recognized (FAO report, 1991; Brockwell and Bottomley,
1995). The techniques described in this manual are concerned primarily with the enumeration
and/or confirmation of the identity of rhizobia in inoculant.
Circumstances exist in which proper inoculation with even the highest quality inoculant
does not result in a demonstrable increase in nodulation or yield. Inoculation failure is common
where large populations of infective rhizobia are present in the soil (Thies et al., 1991), or where
available soil N is not limiting (Beck et al., 1993). Nevertheless, evidence accumulated over the
years indicates that increases in viable rhizobial cells applied per legume seed results in increased
nodulation response and nitrogen fixation, especially under stress conditions. The basic aim of
legume inoculation is to provide the maximum number of suitable rhizobia in the rhizosphere at
the time of nodule initiation. The scientific literature attesting to the significance of high rates of
inoculation in optimizing nodulation and nitrogen fixation is extensive (e.g., Weaver and Frederick,
1974a, 1974b; Amarger, 1974; Smith et al. 1980; Amarger and Lobreau, 1982; Rice, 1982;
Nambiar et al., 1983; La Favre and Eaglesham, 1984; Lowther and Littlejohn, 1984; Bergersen et
al, 1985; Wedderburn, 1986; Brockwell et al., 1987; Berg et al., 1988; Bhuvaneswari et al., 1988;
Bordeleau, 1988; Somasegaran et al., 1988; De Oliveira and Graham, 1990; Thies et al., 1991;
Hume and Blair, 1992; Biederbeck and Geissler, 1993; Olsen et al., 1994; Patrick and Lowther,
1995; Brockwell and Bottomley, 1995). The comments of Burton in 1976 that "large numbers of
rhizobia on seed favor survival before planting," and that "large numbers of rhizobia on seed at
planting favor rhizobia multiplication in the rhizosphere and early nodulation," is supported by
overwhelming evidence today. It is also worth noting that there can be no numerical optimum (of
rhizobia), since excess numbers cannot adversely affect nodulation (Thompson, 1991a).
2
Sterile versus non-sterile carrier
Brockwell and Bottomley (1995) discuss the basic divergence in technologies used for
legume inoculant production based on solid carriers (such as peat). This divergence centers upon
the use of sterile or non-sterile material as the inoculant carrier. Considerable evidence exists that
sterile carriers generally produce superior products. Roughley and Vincent (1967) and Date and
Roughley (1977) reported better survival of rhizobia in sterile peat than in non-sterile peat. Crop
failures with non-sterile products and "consistently satisfactory" inoculant response with sterilecarrier products have been reported (Day 1991). Boonkerd (1991) showed that soybean inoculant
produced in sterile peat carrier had both higher rhizobial numbers and longer shelf life than did
inoculant produced in non-sterile peat.
The possibility of the dissemination of human, animal, or plant pathogens in rhizobial
inoculant made with non-sterile carrier has prompted French regulations preventing the sale there
of such inoculants (Catroux and Amarger, 1992). The existence of opportunistic human pathogens
in some commercial non-sterile carrier inoculant has since been demonstrated (Olsen et al., 1996).
Results of independent tests (Schall et al., 1975; Skipper et al., 1980; Somasegaran, 1991; Vincent
and Smith, 1982; Olsen et al., 1995) indicate that a substantial proportion of the inoculant
produced using non-sterile carrier appears to be unsatisfactory for farmer use, either because of
low populations of rhizobia or high numbers of contaminants (Brockwell and Bottomley, 1995).
A PERSPECTIVE ON STANDARDS FOR RHIZOBIAL INOCULANTS
Regulations on inoculant quality vary from country to country and no set of international
standards exists. Brazil (Thompson, 1991a), Canada (Olsen, et al., 1994), France (Wadoux
1991), and Uruguay (Brockwell and Bottomley, 1995) have regulatory authorities supported by
legislation. Australia, South Africa, and New Zealand have quality control programs in which
inoculant manufacturers participate voluntarily. In many other countries (including the United
States and the United Kingdom), product quality standards are left to the discretion of the
manufacturers (Smith, 1992). A great disparity exists throughout the world today in the price,
quality, and effectiveness of rhizobial inoculants available to farmers (Joly, 1991; Day, 1991).
Whether legislatively set or internally established by the manufacturer, standards for
inoculant products should be a compromise between theoretical possibilities and practical
limitations. However, without defined standards quality control cannot work (Thompson, 1991a).
There exists a consensus that the establishment of standards, whether voluntary or imposed, has
improved legume inoculant quality (Day, 1991; Olsen et al., 1994).
Two basic forms of numerical standards for rhizobial inoculants have evolved. Most
documented standards establish a minimum number of viable rhizobial cells of the appropriate
species or strain, found per unit weight of the inoculant product. A commonly accepted minimum
figure for finished peat inoculant leaving the manufacturer is 1 x 109 viable rhizobia per gram of
inoculant. Another type of standard ties the manufacturer's recommended rate of inoculant
application to a minimum number of viable rhizobia which must be delivered per seed if the
3
inoculant is applied at the suggested rate. The FAO sponsored Expert Consultation on Legume
Inoculant Production and Quality Control (Rome, 1991) has "very strongly pressed for numerical
standards to be based on numbers per seed (while recognizing that numbers per unit quantity of
carrier is the measured criterion in the quality control laboratory)."
Widely accepted minimum quantitative per-seed standards are:
103 rhizobia per small seed ( such as clover and alfalfa)
104 rhizobia per medium size seed (mungbean, pigeonpea, sainfoin)
105 rhizobia per large seed (soybean, pea, bean).
It should be emphasized that these are considered minimum standards because only a
portion of the applied inoculum will still be associated with the seed after planting. The retained
inoculum may be as low as 5% to 10% (Catroux, 1991). France has therefore established a
minimum standard of 106 Bradyrhizobium per seed for soybean inoculation. Similarly, Hume and
Blair (1992) suggested that the existing Canadian standard of 105 per soybean seed be increased to
106 per seed. Official analysts of the Canadian national Legume Inoculant and Pre-Inoculated Seed
Testing Program have also called for an increase in Canadian standards by a factor of 10 to 104,
105, and 106 for the small, medium, and large seed sizes respectively (Olsen et al., 1994). Weaver
and Graham (1994) called for a minimum of 5x103 rhizobia per seed, and noted the importance of
using an adhesive with peat inoculant. Adhesives help to keep the rhizobia on the seed and to
reduce the rate of decline in viability (Materon and Weaver, 1984).
France currently has the highest standards in the world relating to inoculant quality. These
standards extend to field testing and expiry date testing. France also has a strict requirement that
no contaminants be present in rhizobial inoculant products (Catroux, 1991). Australia also has
high standards requiring a plate count of 109 rhizobia per gram, freedom from contamination at the
10-6 dilution, and nodulation of the host plant at the 10-7 and 10-8 dilutions (Thompson 1991b).
Other reported minimum standards are 108 per gram in Uruguay, India, and Brazil (Beck et
al.,1993), 109 per gram with less than 0.001% contaminants in Rwanda (Scaglia, 1991), 5 x 107 per
g in Thailand (Boonkerd, 1991), and 108 per g in New Zealand and South Africa (Smith, 1992).
INOCULANT QUALITY CONTROL TESTING
Legume inoculant is most often made by incorporating rhizobial broth into a solid carrier.
The carrier provides a convenient base for packaging, and facilitates application and use of the
product (Smith, 1992). Peat has long been the worldwide standard for carriers and is still
considered to be the most dependable (Strijdom and Deschodt 1976; Burton, 1982; Smith, 1992;
Weaver and Graham, 1994). Peat may be unavailable in some areas, or may be locally unsuitable
in terms of rhizobial compatibility. Many carrier materials other than peat have been used
successfully. The material from which a carrier is made is less important to the options available
for inoculant quality control testing than whether or not the carrier is sterile.
Quality control testing of non-sterile carrier inoculant
4
Experience over time with inoculants in both Australia and North America has shown that
where rhizobial broth is added to non-sterile carrier and allowed to "cure" for a period of time, the
resultant non-rhizobial or "contaminant" load is often extremely high. Quality control analysis of
inoculants containing large numbers of contaminants is more complex and time consuming than
the analysis of inoculant containing relatively few contaminants (Van Rensburg and Strijdom, 1974;
Catroux, 1991; Catroux and Amarger, 1992; Olsen et al., 1994). With non-sterile carriers, the
analyst normally cannot rely on plate-count or immunological methods to enumerate rhizobia, and
must employ plant grow-out methods which take up to 4 weeks to complete. This is often more
time than the inoculant manufacturer can afford before shipping the product.
Analyses conducted during 1993 and 1994 of 100 samples of North American inoculants
made using non-sterile carrier showed they were all very heavily contaminated. More than 90% of
these products contained 2 to 1000 times more contaminants than rhizobia (Olsen et al., 1995;
Olsen et al., 1996). The high level of contamination found in inoculant produced using non-sterile
carrier creates practical difficulties directly related to inoculant quality control. It is possible to
plate-count and identify rhizobial strains in legume inoculants by immunoblot methods when
contaminant and rhizobial colony numbers are approximately equal (Olsen and Rice, 1989; Smith,
1992), but contamination levels are normally so high as to preclude the use of plate-count methods
generally. The result is that rhizobial enumeration in highly contaminated inoculant products is
limited to most-probable-number (MPN) plant assay methods.
Quality control testing of sterile carrier inoculant
Enumeration of rhizobia in sterile-carrier inoculant by plate count methods normally takes
2-3 days for fast growing rhizobia and 5-6 days for slow growers. It is essential, however, to realize
that important assumptions are made when inoculants are evaluated by plate counts. It is assumed,
for example, that the colonies produced and counted are rhizobia, and that they are of the proper
species or strain. Confirmation of the rhizobial species or strain requires a further, often
serological, procedure. In any case, it is essential that plant infection testing not be bypassed
altogether. The only certain way to demonstrate that an inoculant will nodulate its intended host is
to put it with the axenic host and wait for nodule formation.
QUALITY IN THE CULTURE COLLECTION
Most inoculant producers have a culture collection of rhizobial strains. The collection
normally includes preserved inoculant strains, "mother" cultures, strains for potential future use,
cultures obtained from researchers, and, after awhile, a great many cultures of questionable value.
Every culture accession should be treated with some skepticism and reexamined upon receipt,
even if obtained from a reputable source. Collections rarely get the attention, hard work, and
maintenance that they need. As the level of work involved in maintenance (particularly with the
very common agar slant collections) grows, the collection quality declines. It is far better to put full
effort into maintaining a small, but high quality, collection than it is to put that effort into keeping a
large collection improperly. At the very least, the most important cultures should be kept in a
5
small, separate, and thoroughly characterized and maintained collection (for review of culture
collections, see Keyser, 1987).
Quality in the collection means that the isolates are pure, the host known, that nodule
formation and effectiveness have been demonstrated, that they are alive (and likely to remain so),
that they are properly identified, and that the history of each is known precisely. Purification and
authentication are critical areas. A great many "pure cultures" of rhizobia are contaminated, often
with other strains of rhizobia. Authentication of the isolate by plant infection should be followed by
reisolation from an axenically grown nodule and this source used for the preparation of aliquots for
long term preservation. One of the aliquots prepared for preservation should then be reexamined
for purity and the ability to nodulate. Rhizobial authentication by plant passage must always be
accompanied by uninoculated negative controls. Plant infection tests which are unaccompanied by
strict negative controls are worse than useless, because they are often misleading.
The means of long term storage of a culture collection also affects the quality of the
collection. Cryopreservation and freeze-drying (lyophilization) are the best techniques available
today. Almost all laboratories working with rhizobia use agar slants for short term maintenance
and storage of cultures, but long term preservation of strains by successive transfer on agar slants is
fraught with problems. Such storage is inadequate and is extremely likely to result in the eventual
loss, genetic change, or contamination of the cultures. Frequent subculture provides an
opportunity for mutation, and variations in effectiveness between isolates from a single culture are
well documented (Herridge and Roughley, 1975; Weaver and Wright, 1987; Gibson et al., 1990).
A good principle to follow in the preservation of important strains that are used repeatedly is to
first carefully purify and characterize the strain and then carefully freeze or freeze-dry a large
number of aliquots of broth culture. In this way, the producer or researcher can always begin
inoculant production or a research experiment with the same strain, and will not be unknowingly
accumulating and propagating genetic changes.
Cryopreservation (freezing in ultra-cold conditions) of cultures is probably the best way to
preserve cultures because it is simple and fast, but it is also expensive and requires good technical
backup. Modern ultra-cold freezers are dependable, but will eventually need repair or technical
maintenance. It is wise to have two freezers, each only half-full and ready to accommodate the
contents of the other. Liquid nitrogen is an ideal cryopreservant, but requires continuing
replenishment which creates a continuing expense. If storing cryovials under liquid nitrogen
(rather than in the vapor), it is essential to use cryovials that are designed for the purpose. Some
cryovials will leak and allow liquid nitrogen along with potential contaminants into the vial.
Mechanical freezing of broth cultures mixed with 10% glycerol at a level below the recrystalization
point of ice (-135° C) has been found to be simple, fast, trouble-free, and very effective.
Cryopreservation at lower temperatures may reduce long term viability. Freeze-drying
(lyophilization) is the best alternative to cryopreservation, but involves considerable initial labor
and also requires expensive specialized equipment. Successful freeze-drying also requires more
technical competence than does cryopreservation.
Most inoculant producers do not have access to cryopreservation or freeze-drying
equipment. Producers can obtain tested, pure cultures from reputable organizations with culture
6
collections that use these methods to preserve rhizobial germplasm.
Culture collection records are important and should be kept in a bound notebook and not
scattered. Important information includes the date and location of isolate collection, host type and
soil type of origin, date of preservation, and as complete a history as possible for the strain. The
isolate history should be continuing and include the work done by the receiving lab to reexamine
the isolate, including information on repurification or nodule passage, etc. Original identification
labels for cultures received from other laboratories should be retained and used even though the
cultures will probably also get the receiving laboratory's identification number.
7
CHAPTER 2
MICROBIOLOGICAL TESTING OF INOCULANT CARRIER
As in other areas of inoculant quality analysis, there are major differences in the options
available for microbiological testing of carriers, depending on whether or not the carrier is sterile.
Inoculant carrier serves several functions, but a primary one is the ability of the carrier to
hold and protect rhizobial cells in a viable form for as long as possible. Inoculant prepared with
non-sterile carrier is usually made by adding maximally grown rhizobial broth to the carrier in the
hope that a maximum number of rhizobial cells will survive and compete adequately with fast
growing contaminants during the curing process. Inoculant prepared in sterile carrier is usually
made by adding a 1:10 to 1:100 dilution of rhizobial broth to the carrier, and allowing the rhizobia
to multiply within the carrier during curing. In either case, it is essential that the carrier does not
contain factors that will kill the rhizobia. In the case of the sterile carrier in particular, and to a
lesser degree with non-sterile carrier, it is necessary that the rhizobia actually thrive and maximally
reproduce in the carrier environment. In the situation of non-sterile carrier inoculant, no rapid
methods of estimating carrier quality in terms of rhizobial compatibility are available because the
presence of contaminants limits rhizobial enumeration methods to the most-probable-number
(MPN) test. Relative to sterile carrier, two questions are of primary significance and both can be
answered relatively quickly by plate count methods. The questions are: 1) is the carrier sterile?
and 2) if added in dilute broth form, will the rhizobia multiply to high numbers in the carrier?
CARRIER STERILITY
Sterility means that no organisms are alive within a given set of parameters. Carriers which
have been "sterilized" are almost never rigidly examined to verify that absolutely nothing living is
present. Carrier "sterility" normally means only that nothing obvious was found to grow aerobically
on common microbiological media (usually YEMA).
Carrier sterilization usually involves one of three methods; gamma irradiation, electron
beam sterilization, or autoclaving. Because of the bulk of carrier that requires sterilization during
commercial inoculant preparation, quality control checks on the effectiveness of the sterilization
should be on-going and routine. It is unlikely that complete sterilization will always be achieved.
Little information is available as to how little contamination is required in carrier to result in
significant contamination of the final product, but experience indicates that true sterility is not
always an absolute requirement.
In this chapter, three methods are suggested as ways to examine carrier sterility based on
the level of contamination present in the "sterile" carrier. These are 1) the "enrichment" detection
procedure, 2) the "sprinkle" method, and 3) plate count. A fourth test is described for evaluating
the compatibility of a sterile carrier with the rhizobia it is intended to carry.
The enrichment detection method provides a "yes" or "no" answer to the question "is there
8
any contamination in the carrier?" The sprinkle method measures from 1 to 1000 contaminants
per g carrier. The plate count method measures contaminants in the range of approximately 100
per g to the maximum possible (approximately 1011 cells per g).
For detailed methods of plate counting or for preparation of yeast-extract mannitol media, refer to
the viable cell counts section under Methods of Inoculant Analysis in this guide.
TESTS FOR CARRIER STERILITY
I. Enrichment detection procedure
1. Add sterile YEMB to the carrier bag in the same relative volume as if rhizobial broth
were being added to prepare inoculant.
2. Incubate at 30° C for 5 to 14 days, and sample periodically.
3. Prepare a 10-2 dilution of the sample and spread plate on YEMA. Only this one dilution
is needed.
4. Invert the plates and incubate. Examine plates each day for 5 days and record the time
required and the number of colonies of microbial growth.
Interpretation of enrichment detection test result:
The enrichment technique determines the presence or absence of aerobic
organisms capable of growth under these conditions on YEMA. There is little relation
between the number of organisms present at the start of the enrichment period and at its
completion, so there is no need to plate serial dilutions. If no microbial growth is detected
after the incubation period, then carrier contamination will probably not be an obstacle to
production of high quality inoculant. This does not necessarily mean that the carrier is well
suited to inoculant production because the carrier may be sterile and yet carry components
that are toxic or inhibitory to rhizobia. See the carrier-rhizobia compatibility test below.
II. Sprinkle method procedure
1. Aseptically weigh out five 0.2 g samples of carrier taken from different parts of the
carrier container.
2. Sprinkle the carrier onto five different petri plates with YEMA media and spread the
carrier as well as possible with separate sterile "hockey sticks" as in the spread-plate count
procedure.
3. Invert the plates and incubate. Examine all plates each day for 5 days and record time
9
and number of appearances of microbial growth.
Interpretation of sprinkle test result:
There are three possible results: 1) no growth; 2) countable growth; and 3)
overwhelming and uncountable growth.
(1) No growth is ideal and indicates successful sterilization.
(2) Countable growth provides a rough estimate of the level of contamination in the carrier
at the time of test (e.g., 100 colonies over the 5 plates = approximately 100 contaminants
per g carrier). Keep in mind that even low numbers of contaminating organisms could
easily explode to 109 - 1010 per gram following incubation in the moist, nutrient-enriched
environment provided by the addition of dilute rhizobial broth during preparation of an
inoculant. The advantage that the sprinkle test provides is an ability to count contaminants
without dilution and therefore to enumerate low contaminant concentrations.
(3) Overwhelming growth indicates that the inoculant is unsuitable for use in a sterilecarrier system.
III. Plate Count Method
1. Prepare 10-fold dilutions of the carrier as for plate count procedures.
2. Spread-plate in replicate fashion 100 ìL of each dilution on YEMA.
3. Invert the plates and incubate. Examine all plates each day for 5 days and record the
time of appearance and number of colonies of microbial growth.
Interpretation of plate count results:
The lowest number of contaminants detectable using this approach is one colony
per plate at the 10-1 dilution. Since 100 ìL of sample was plated, 1 colony per plate = 10 x
10 x 1 = 100 contaminants per g. While 100 contaminants per gram is the lower detectable
limit, this method can enumerate any higher level of contamination by plating at increased
dilution levels.
CARRIER-RHIZOBIA COMPATIBILITY TEST
The object of this test is to learn if rhizobial strains of interest will reproduce and
increase from low to high numbers in sterilized carrier under conditions as close to actual
inoculant production as possible. The ideal situation would be to use the carrier bagged as
intended for inoculant production and distribution. Many variations of the test are possible.
For example, different dilution rates. different rhizobial concentrations, or different curing
10
times and temperatures can be tested. This type of testing is fundamental to producing
quality inoculant and should be explored thoroughly before producing the first batch of
inoculant. This type of testing is also appropriate for routine quality control of inoculant so
that the analyst can follow development of each batch, detecting problems or surprises
early, and thereby saving time and money. The procedure described here applies to testing
of sterile carrier using plate count enumeration procedures, but non-sterile carrier can also
be testing using MPN plant assay enumeration procedures.
1. Grow the rhizobial strain of interest to early stationary phase in a small flask.
2. Aseptically dilute the rhizobial broth 1:100 with the same sterile diluent or nutrient
solution as will be used for inoculant production.
3. Aseptically sample the diluted rhizobial broth. Prepare serial 10-fold dilutions of the
sample of diluted broth (through 10-6) and plate the 10-4 - 10-6 dilutions on YEMA.
4. Add the same volume of the same dilution of broth to the carrier bag as would be added
in production.
5. Allow the carrier to wet, massage it, and aseptically sample 11 g of the freshly prepared
inoculant to 99 mL of plate count diluent (a 1:10 dilution). Prepare 10-fold dilutions of the
sample and plate the 10-3 - 10-6 dilutions on YEMA for enumeration.
6. Incubate or cure the remaining inoculant as would be done under production conditions.
7. Sample the inoculant bag every 2 to 3 days and plate as above, selecting dilutions to plate
on the basis of previous results. Continue the plating throughout the curing period, and
then, at more extended intervals, throughout the shelf-life of the product.
Interpretation of compatibility test results:
Calculate the number of rhizobia added per g carrier from the result of the broth
plate count, the volume and weight (use a figure of 1 g per mL) of broth added to the
carrier, and the weight of the carrier. For example, if mature broth containing 3 x 109
rhizobia per mL were diluted 1:100, and 16 mL of the diluted broth was then added to 22 g
of sterile carrier, then, 4.8 x 108 rhizobia were added in the process of making
approximately 38 g of inoculated carrier. This works out to approximately 1.3 x 107
rhizobia per g and serves as a base point for estimating the rhizobial load at time zero.
Graph the results of the inoculant samples as log of rhizobial count over time. If the
rhizobia die or decline in numbers upon addition to the carrier, a problem may exist with
the carrier (or with the diluent used to carry the rhizobia into the inoculant, or a
combination of the two). If the rhizobia increase to an acceptable level (at least 2 x 109 per
g and preferably higher) in the carrier over the curing period, then it is reasonable to
assume that the carrier and diluent are compatible with that rhizobial strain. If rhizobial
11
numbers decline upon addition to the carrier, but then recover to acceptable levels, perhaps
some factor in the production process can be identified and modified to avoid placing the
initial stress on the bacteria.
12
CHAPTER 3
METHODS OF EVALUATING RHIZOBIAL BROTH QUALITY
Inoculant producers regularly need to judge the quality of starter culture used to inoculate
the batch fermentor, and shortly thereafter, judge the quality of the grown fermentor broth prior to
producing inoculant with it. Two questions have the greatest priority: 1) are the desired rhizobia
present in high numbers? and 2) are contaminating microorganisms present? The time needed to
answer these questions requires that the broth be held unused until test results are obtained.
Several tests, however, will allow some estimation to be made of the quality of the broth and can be
performed within hours. These tests can detect abnormalities which provide conclusive evidence
that the broth should be discarded, but they cannot provide proof that the broth is satisfactory.
Definitive tests require at least days (plate counts) or weeks (MPN plant infection tests) to
complete.
GROWTH RATE
Most contaminants grow faster than do rhizobia. Unexpectedly quick growth is therefore a
first warning that a broth culture is contaminated. Any culture that grows to turbidity within 24
hours is highly unlikely to be rhizobia. A culture which seems to be growing faster than normal
should be immediately suspect and tested carefully. The same applies to any putative rhizobial
culture which appears unusual in terms of color or which is producing an unusual odor.
MICROSCOPIC EXAMINATION OF BROTH
Problems with a rhizobial broth culture can often be quickly detected by examining the
broth under an adequate microscope. An experienced microscopist can quickly determine
whether or not the broth contains a uniform suspension of cells with typical rhizobial morphology
and if the cells are at approximately the expected concentration. The microscopist can also often
detect contaminant microorganisms if they are present. Since many potential contaminants grow
much faster than do rhizobia, they are likely, if present, to be in high concentration. An
experienced individual can therefore often detect a bad broth after only a moments examination.
This test can save the cost of producing a faulty batch of inoculant and save the time and cost of
performing additional tests. Microscopic examination is largely subjective, and results are
therefore only as good as the experience of the microscopist. Experience can be obtained by
practice and work with small cultures of known purity, and then expanding this experience to larger
broth volumes. The microscopist must become familiar with actual broth cultures, grown in the
same media as will be used in commercial production, of the rhizobial species and strains which
will be used to make inoculant.
The microscope used need not be complicated or extremely expensive, but must be
equipped with phase contrast features and should have a built in light source with an adjustable
condenser and light diaphragm for proper Koehler illumination. If fluorescent antibody work is
anticipated, it is much cheaper to properly equip the microscope at the time of initial purchase than
13
it is to upgrade later. A 10-power dry objective for focusing, and a good 40-power phase contrast
dry objective, in conjunction with 10-power oculars, will be satisfactory for most work. The skill of
the microscopist should be such that maintenance of the microscope and adjustments for proper
illumination and phase contrast are routine and automatic. The experience of the microscopist
must be such that abnormalities in broth culture are readily discerned.
Procedure for microscopic examination of broth
1. Clean microscope slides by rubbing them with ethanol and tissue paper. Do not attempt
to reuse microscope slides. Even new "pre-cleaned" slides will often have a thin fungus like
growth on the viewing surface.
2. Apply a small drop (20 - 30 ìL) of a representative and fresh sample of rhizobial broth
to the cleaned slide. Apply a clean cover slip. The sample size should not be so large that
the cover slip skates around as the slide is tilted, but should just fill the area under the cover
slip.
3. Focus the microscope on the bacteria using the 10-power objective, then switch to the 40power objective, properly set the illumination, and adjust the phase contrast alignment to
provide optimal viewing.
4. Examine many areas of the slide and make an evaluation as follows:
a. culture appears typical - no contamination noted
or
b. inconclusive
or
c. culture is distinctly atypical or contaminated or both.
Interpretation of result:
Category a (typical) is the desirable situation. Further serological testing to establish that
the organisms are in fact the desired rhizobia can begin.
Category b (inconclusive) requires that the broth be held as long as necessary until further
testing positively establishes both rhizobial identity and purity.
Category c warrants immediate discard of the broth.
Microscopic Count of Total Cells in Broth
Total cells in broth can be counted using specialized bacterial counting chambers that
14
provide a calibrated volume of broth to be examined under the microscope. The counting
chamber most often used with rhizobia is the Petroff-Hausser chamber. Hemocytometer counting
chambers are cheaper and more widely available, but they are not suitable for counting bacteria.
The Petroff-Hausser grid consists of subdivided squares of known area (see Figure 1 and
Table 1). Counting will normally be done under dark-field or phase-contrast conditions using a dry
40-power objective. The useful range for counting is 107 to 108 cells per mL. Microscopic
methods are not useful for counting (or observing) cells in specimens with less than about 106 cells
per mL because the field of view is of such a small area that no organisms can be seen at low cell
concentrations,.
Using a Petroff-Hausser slide for repetitive counts is tedious, and not normally required on
a routine basis. The count obtained is often higher than a plate count of the same specimen
because the count includes non-viable cells. Care must be taken not to count non-bacterial
particulate matter (such as tiny grains of calcium carbonate). The method is not suitable for
counting bacteria in peat or other particulate matrices.
15
Figure 1. Petroff-Hausser counting chamber. (a) cross section, (b) top
view of the entire grid, and (c) magnified view of a single large square
16
containing 16 small squares.
Table 1. Details of the Petroff-Hausser counting chamber.
Corresponding
Volume (mL)
2
Area (cm )
Conversion
factor
Total grid
1 X 10-2
2 X 10-5
5 X 104
Large square
4 X 10-4
8 X 10-7
1.25 X 106
Small square
2.5 X 10-5
5 X 10-8
2 X 107
Procedure for total cell count using the Petroff-Hausser chamber
1. Carefully clean and dry the chamber slide and cover slip.
2. For counting fully grown rhizobial broth cultures (e.g., > 1 x 109 cells per mL) dilute the
specimen 1:10 or 1:50 with particulate-free diluent.
3. Deliver a drop of the specimen to the chamber so that it flows evenly under the cover
slip, but does not flood the channels at the edges of the etched platform. Allow the liquid
under the cover slip to stabilize for a minute or two before counting. If bacteria are seen
streaming from square to square, either too much or too little specimen has been added, or
the specimen is drying out (prepare a new slide).
4. Using the 40-power dry objective, count the bacteria in at least 10 of the larger squares or
20 of the smaller squares, depending on the density of cell concentration. Useful ranges are
about 3 - 8 cells per small square, and 8 - 80 per large square. If the bacteria are not evenly
distributed throughout the field, or the cell density is too high or too low, prepare another
slide adjusting concentration of the specimen as necessary. Count the cells over two of the
grid lines, but do not count cells on the other two lines of each square (count cells lying
over the top and bottom lines of a square, but not those over the left and right side lines).
5. Calculate the number of cells per mL of the original suspension as: average cell number
per square x dilution of the sample x factor for the square. The factor for the small squares
is 2 x 107; the factor for the larger squares is 1.25 x 106. For example: If the average cell
count per large square = 20, and the dilution of the sample was 1:10, then the total number
of cells in the original broth = 20 x 10 x 1.25 x 106 = 2.5 x 108 cells per mL.
A Petroff-Hausser chamber consists of a fragile glass slide and a reinforcing frame. The
glass slide portion of the Petroff-Hausser chamber is thin, grooved, and fragile, but allows dark
field viewing of cells (if the microscope is so equipped). The glass portion of the chamber slides
into a reinforcing frame which is large enough to prevent many microscope objectives from
swinging into position. Also, the calibrated grid on the slide is normally not centered on the glass,
and it is important to slide the glass into its frame in the direction that centers the grid. Doing so
17
provides more room for the objective lens. Still, it is often necessary to find focus by bringing the
objective down over the etched grid, while watching from the side, to a position very near the cover
slip. Then slowly lower the stage (increasing separation between slide and objective) with the focus
knob while watching through the microscope until focus is achieved. Adjusting the focus in the
wrong direction will put pressure on the expensive Petroff-Hausser slide and quickly break it.
GRAM STAIN FOR CONTAMINATION IN BROTH
The Gram stain is a rapid test that can be useful in confirming that an organism is not a
rhizobia. The technique cannot confirm that an organism is a rhizobia. Microorganisms generally
fall into two major divisions (Gram positive and Gram negative) on the basis of whether or not they
take up and retain crystal violet during the procedure. A Gram positive organism will appear deep
purple under bright field illumination under the microscope, whereas Gram negative organisms
retain a pink color (indicating adsorption of the safranin counterstain. Rhizobia almost always
show a Gram negative reaction. The stain does permit the detection of Gram positive
contaminants if present in a rhizobial culture, but it must be remembered that failure to detect
contamination with the Gram stain does not prove that the culture is contaminant free. Many
potential contaminants are as Gram negative as are rhizobia. While the technique is simple and
fast, consistently correct results with the Gram stain technique requires practice with known Gram
positive and Gram negative cultures. Do not try to judge the Gram stain reaction under phasecontrast conditions. Instead, set the microscope condenser to the bright-field stop and use more
light than with phase-contrast. Best results are obtained under oil immersion.
Gram stain reagents:
crystal violet solution
crystal violet dye
ammonium oxalate
95 % ethanol
purified water
1.0 g
0.4 g
10 mL
40 mL
iodine solution
iodine
potassium iodide
95 % ethanol
purified water
0.5 g
1.0 g
12.5 mL
50 mL
decolorizing solution
95 % ethanol
100 mL
counterstain solution
0.25 g safranin dye (in
10 mL of 95 % ethanol)
purified water
10 mL
100 mL
18
Gram Stain Procedure
1. Apply the cell culture material as a thin, even smear to a clean microscope slide. Allow
the smear to air dry completely and then heat fix the cells by a quick passage (cell side up)
of the slide through an open flame. Avoid overheating the slide.
2. Stain the smear with the crystal violet solution for about 60 seconds.
3. Using a water wash squeeze bottle, rinse off the crystal violet dye solution. Do not hit
the cell smear directly with the water stream.
4. Flick the slide to remove excess wash water and flood the smear with the iodine
(mordant) solution for about 60 seconds.
5. Rinse off the iodine solution with the water wash bottle.
6. Apply the decolorizing ethanol to the smear while holding the slide on a 45 degree slant
and rinse just until no further color is being washed from the smear. This step is critical do not over decolorize.
7. Quickly wash off excess decolorizing solution using the water wash bottle.
8. Flick the slide to remove excess wash water and counterstain the smear with the safranin
dye solution for about 60 seconds.
9. Wash the smear with the wash bottle to remove excess safranin, flick the slide to remove
excess water, allow the smear to air dry, and examine under the microscope using either a
dry objective (minimum of 40 power), or under oil immersion.
THE PEPTONE-GLUCOSE TEST FOR CONTAMINATION IN BROTH
The peptone-glucose test is designed for detection of contamination in broth cultures. The
test takes a minimum of 24 hours and as much as 48 hours, but will often give evidence of
contamination that would be missed by microscopic examination. It is possible to sample broth
during middle to late log phase of growth, but before fermentation is complete, streak the sample
on peptone-glucose agar, and have a result by the time that the broth is ready for addition to the
carrier. A positive test indicates contamination and warrants discard of the broth.
Almost all rhizobia grow poorly, if at all, on glucose-peptone media, whereas many
potential contaminants grow readily and produce pH changes. Bromthymol blue (BTB) and
bromcresol purple (BCP) are pH indicators. If either dye is incorporated in peptone-glucose
media it will indicate major pH shifts which are often associated with growth of contaminating
organisms, but not with rhizobia (because the rhizobia do not grow). Bromthymol blue turns
yellow at pH 6.0, blue at pH 7.6, and is green between pH 6.0 and 7.6. Bromcresol purple turns
19
yellow at pH 5.2 and purple at pH 6.8. Rapid growth (24 - 48 hours) of a loopful of rhizobial broth
streaked on this media, particularly if associated with color change (pH reaction), strongly indicates
a contaminated broth.
Peptone-glucose media with bromocresol purple:
1000 mL purified water
5 g glucose
10 g peptone
15 g agar
10 mL of 1% bromcresol purple in ethanol (0.10 g BCP in 10 mL EtOH).
To substitute bromthymol blue for bromcresol purple, add 5.0 mL of 0.5% BTB in ethanol in
place of the BCP.
Confirmation of Rhizobial Identity in Broth
Three immunological tests which can confirm the rhizobial identity of cells in a broth
culture are described in the section of immunological testing methods in this manual. These
methods are the cell agglutination reaction, the immuno-spot blot, and the indirect fluorescent
antibody test.
20
CHAPTER 4
METHODS OF INOCULANT ANALYSIS
VIABLE CELL COUNTS (PLATE COUNT METHODS)
Viable rhizobial cell numbers in broth and inoculants prepared with sterile carrier can be
measured by plate count methods. Most inoculant prepared using non-sterile carrier contains so
many fast-growing contaminants that plate count procedures are impractical. Viable cells are those
which can divide and produce colonies on an agar solidified nutrient media. Plate count methods
involve placement of a sample containing viable cells onto the nutrient media surface and counting
the resulting colonies after a period of incubation. Since rhizobial broth and inoculants are
expected to contain a very large number of viable cells, it is necessary to accurately dilute the
sample prior to application to the media so that a countable number of discreet colonies will grow.
The most widely used media for counting rhizobia is yeast-extract mannitol agar (YEMA or
YMA), often containing congo red to assist in differentiating contaminants from rhizobia.
Considerable latitude is permissible in formulating the media (Burton 1967; Vincent 1970;
Thompson, 1984), but levels of yeast-extract above 3.5 g per L have been shown to have
deleterious effects on rhizobial culture (Date, 1972; Skinner et al., 1977). Mannitol levels as low as
1 gram per liter are often sufficient to support rhizobial growth (Keyser, personal communication).
See Tables 2, 3, and 4 for lists of YEMA media components.
The first step in counting viable cells in either rhizobial broth or inoculant is to prepare a
dilution series which will cover the range of expected viable cells. The most commonly used
dilution ratio is 1:10 per dilution step. For broth, the schematic shown in Fig. 2 is appropriate.
For inoculant, a question arises as to what a 10-fold dilution of a powder into a liquid diluent is.
Most workers with rhizobia today adopt the view that 10 g (wet weight) of inoculant into 90 mL
diluent (or 11 g inoculant into 99 mL) is a 10-fold dilution. While this point can be (and has been)
argued, we recommend that this approach be used and reported when presenting results. For an
excellent general review of bacterial enumeration, including dilution and plating theory, see
Zuberer, 1994.
Serial dilutions and plating
Prepare serial 10-fold dilutions through the highest level required, and plate only those
dilutions that are needed to enumerate cells within the range of interest. For example, if 20
colonies resulted from a spread plate in which 100 uL of sample was applied from the 10-8 dilution,
this would correspond to 20 x 10 x 108 = 2 x 1010 viable cells per mL of the original sample. This is
a very high number for rhizobial broth, yet it would also have been obtained by a showing of 200
colonies at the 10-7 dilution. In other words, there is normally nothing to be gained by plating a 10-8
dilution of either broth or inoculant. With the spread plate technique, plating the 10-5, 10-6, and 10-7
dilutions will enumerate 3 x 107 to 3 x 1010 viable cells per mL or g of sample (based on countable
colonies per plate of 30 to 300). This covers the range of thin to dense rhizobial populations, but
21
will use nine petri plates if each dilution plated is replicated the usual 3 times. On the other hand,
the drop plate technique described here lends itself to the plating of 4 dilutions on only two plates.
The researcher using this technique should probably plate the 10-4 - 10-7 dilutions on two plates, thus
covering a broad range of population possibilities from very low to very high.
Keep in mind that the accuracy of the plate count result is dependent upon starting with a
representative sample and accurately dilute it. Mix each dilution well and do not allow settling to
take place before removing the chosen volume for the next dilution. Use a fresh pipette or pipette
tip for each dilution level. Do not let sample dilutions stand for any longer than necessary before
plating. Time the work so that it can be completed, from drawing of the sample to putting the
plates in the incubator, in a single work session. Until considerable experience has been gained,
work with one sample (start to finish) at a time.
Preparation of the inoculant dilution series is the same for drop or spread plate techniques.
For small broth samples, the sequence shown in Fig. 2 can be followed using a fresh 1.00 mL
sterile pipette at each level. For large broth samples, a larger volume should be drawn and used as
sample source. For inoculant samples, follow the dilution procedure described in the section of
this manual detailing most-probable-number technique.
22
Ingredients of a complete yeast-extract mannitol agar media for plate
counting rhizobia:
Table 2. Concentrated stock solutions for preparation of YEMB/YEMA.
mg/500 mL
1. K2HPO4
25.0 g/500 mL
6. H3BO3
50
2. NaCl
10.0 g/500 mL
ZnSO4
50
3. CaSO4⋅2H2O
10.0 g/500 mL
CuSO4⋅5H2O
25
4. MgSO4⋅7H2O
10.0 g/500 mL
MnCl2⋅4H2O
25
5. Fe-EDTA
(Sequestrene)
1.25 g/25 mL
Na2MoO4⋅2H2O
5
To prepare YEMB or YEMA add to 950 mL of purified water 10 mL of stock
concentrates 1, 2, 3, 4, 6, and 0.2 mL of stock concentrate 5.
Add mannitol, yeast extract, and if desired, agar, CaCO3, and Congo red .
23
Table 3. YEMA rhizobial plate-count media.
nutrients -
g per liter
micronutrients -
mg per
K2HPO4
0.5
H3BO3
1.0
NaCl
0.2
ZnSO4
1.0
CaSO4⋅2H2O
0.1
CuSO4⋅5H2O
0.5
MgSO4⋅7H2O
0.2
MnCl2⋅4H2O
0.5
Na2MoO4⋅2H2O
0.1
Fe-EDTA (Sequestrene)
10.0
Mannitol
Yeast extract
Agar
10.0
2.0
liter
18.0
Congo red - 2.5 mL of a 1:400 aqueous solution per L media.
CaCO3 may be added for tube slants - 0.15 g per L.
Monitor pH of autoclaved media and adjust to 6.9 - 7.1 if necessary.
For broth preparation, omit the agar.
Table 4. Simple YEMA rhizobial plate counting media (Fred and
Waksman, 1928).
Per liter purified water:
mannitol
10 g
K2HPO4
0.5 g
MgSO4⋅7H2O
0.2 g
NaCl
0.1 g
Yeast extract
0.5 g
Agar
15 g
Adjust pH to 7.0 with 1N HCl.
To add Congo red, add 10 mL of a 1:400 (0.1 per 40 mL H2O) aqueous solution of dye
to the media before autoclaving.
For broth preparation, omit the agar.
Three techniques of plate counting are commonly used: the spread plate, the drop plate, and the
24
pour plate.
THE SPREAD PLATE TECHNIQUE
The spread plate is the most reliable and commonly used method of plate count
enumeration of rhizobia. The method involves spreading a known volume (usually 100 ìL) of a
known dilution of sample over the agar solidified surface of a nutrient media. Following incubation
for growth, the resulting colonies are counted at a dilution yielding 30 - 300 colonies per plate.
Interrelating the number of colonies, the sample dilution, and the volume plated, the minimum
number of viable cells in the original sample can be calculated. Since plating at each dilution level
is normally replicated 3-5 times, the amount of material and time invested plating and counting can
become considerable.
Spread plate procedure
1. Prepare the required number of YEMA plates and sterile spreaders ("hockey sticks").
The plates should not be "wet" on the surface. Label all plates with a standard
nomenclature (and make it consistent throughout the laboratory).
2. Prepare 10-fold dilutions of the sample 10-1 - 10-7.
3. Transfer 100 ìL of the highest dilution to be plated to each of three replicate plates.
Have sterile, cool "hockey stick" spreaders ready. A 100 ìL piston-type pipette using
disposable tips is much better for sample delivery than attempting to hand deliver 0.1 mL
volumes from a 1 mL pipette. Use a fresh, sterile pipette tip for each dilution level.
4. Minimize the time that the samples are on the plate before spreading. Spread the sample
over the agar surface with a sterile spreader. Move the spreader back and forth in a straight
line while rotating the plate on a turntable for approximately 10 seconds. Do not allow the
tip of the spreader to deliver sample liquid to the extreme edge of the media where it
touches the petri dish. Do not let the spreader dig into the media.
5. Repeat steps 3 and 4 for the selected number of dilutions to be plated working from the
most dilute sample to the least dilute. Use a sterile spreader for each dilution level.
6. Allow the plates to stand right side up until the sample liquid is completely absorbed,
then invert the plates for incubation.
7. After an appropriate incubation period, count the colonies on the replicate plates of the
dilution showing discreet colonies in the countable range of 30 - 300. Do not count the
plates too early. If the colonies seem to have grown up too fast, they are probably not
rhizobia.
8. Calculate the number of viable cells in the original sample by accounting for all dilutions
25
and for the volume plated. For example, if the average number of colonies from plating
100 ìL per plate from the 10-5 dilution is 62, then 62 x 10 x 105 = 6.2 x 107 viable cells per
mL (for broth) or gram (for inoculant). Use the mean from the three replicate plates.
Figure 2. Procedure for 10-fold serial dilution.
26
THE DROP PLATE TECHNIQUE
The drop plate technique involves dropping known volumes (usually 20 - 30 ìL) of sample
dilutions onto YEM agar plates. The drops are not spread, but are allowed to absorb into the agar
surface. Resultant colonies are counted from the dilution which produces the largest number
(usually about 20) of discreet non-confluent colonies. The plate is usually marked into eight
sections to receive eight drops. Four replications of two different 10-fold dilutions can be
accommodated on a single plate (see Fig. 3). The drop plate technique uses agar plates
considerably more efficiently than the spread plate method. Experienced users report that more
practice is needed to obtain consistent results with the drop plate technique than with the spread
plate technique. If, however, a piston type pipette (using disposable tips and measuring accurately
within the 20 - 30 ìL range) is used, the advantages in time and material saving can be
considerable.
Figure 3. A drop plate showing results for two dilutions with 4 replications
27
each.
Drop plate procedure
1. Prepare appropriate media and plates. It is important that the surfaces of the plates are
not "wet." Plates prepared a few days in advance will have the chance to "dry." Mark the
plates (on the bottom) into 8 equal pie shaped sections and label them appropriately.
2. From the most dilute of the chosen range of dilutions, drop 25 ìL of sample about 2.5
cm from the edge of the plates and from a height of about 2 cm. Replicate each dilution 4
times (4 drops individually delivered to 4 pie shaped sections).
3. Using a fresh sterile pipette tip and the next least dilute sample suspension, repeat step 2
above, filling the remaining 4 sections of the plate. The plate now has eight drops in total (4
replicate drops from each of two successive dilutions). Do not spread the drops. Leave the
plate face up until the sample drops have been completely absorbed into the agar.
4. Repeat steps 2 and 3 above on a second plate, using the other two dilution levels selected
for plating. Two plates will accommodate 4 dilutions of four replicates each.
5. Allow all sample drops to be absorbed completely, then invert and incubate the plates.
6. Obtain counts from the dilution level drops which show the largest number of discreet
colonies free of confluent growth. Count from the "underside" of the plate, marking the
position of each colony with a marker as it is counted. Average the counts of the four
replicate drops and calculate the number of viable cells in the original sample by
accounting for the sample dilution and the volume of the drops used. For example, if the
average colony number is 20, the volume of the drops is 25 ìL, and the dilution level is 10-5
then: 20 x 40 x 105 = 8 x 107 viable cells per mL (or gram). To determine the
multiplication factor relative to the drop size, divide 1.0 mL (1000 ìL) by the drop volume
(25 ìL) = 40.
The pour plate technique
The pour plate technique involves mixing sample dilutions containing rhizobia with partly
cooled, but still molten, agar media and subjects the bacteria to at least some degree of thermal
shock. The technique requires careful monitoring of the media temperature to avoid heat killing of
the bacteria and has been shown to give estimates lower by 70 - 80% than those obtained by the
spread plate technique (Zuberer, 1994). We see no reason to recommend the pour plate method
for enumerating rhizobia over simpler techniques which give more accurate results.
28
CHAPTER 5
ENUMERATION OF RHIZOBIA IN INOCULANT USING THE
MOST - PROBABLE - N U M B E R ( M P N ) P L A N T - I N F E C T I O N A S S A Y
The material in this chapter provides detailed information relevant to the design, conduct,
and interpretation of results of MPN plant-infection assays conducted for the purpose of
enumerating rhizobia in legume inoculant.
MPN OVERVIEW
The most-probable-number (MPN) technique is a means to estimate microbial population
sizes. The technique is widely used to enumerate rhizobia based upon the ability of rhizobia to
nodulate appropriate host legume plants. The method relies upon the pattern of positive and
negative nodulation responses of host plants inoculated with a consecutive series of dilutions of
rhizobia containing sample suspension. The results are used to derive a population estimate based
upon the mathematics of Halvorson and Zeigler (1933). While MPN assays for rhizobia involve
plant-infection techniques, not all plant-infection tests involve MPN procedures. Workers with
rhizobia sometimes use the terms interchangeably but, for example, the authentication of a
rhizobial culture on a legume plant is a plant-infection test, but not an MPN assay. MPN
enumeration of rhizobia is dependent upon the ability of the researcher to grow the legume hosts in
a healthy and replicated fashion, and to keep the plants free from rhizobial contamination for a
period of up to 4 weeks. These conditions are not always easily met, particularly in a legume
inoculant production facility whose major function is to produce relatively enormous amounts or
rhizobia. Thought and effort invested in the establishment of a plant growth area and watering
system, which can be maintained with a minimum of outside contamination, will help greatly in
obtaining consistently meaningful MPN plant-infection results.
All rhizobial MPN plant-infection assays are similar in principle, but vary widely in design
and implementation. A large amount of relevant information has been included in this chapter so
that the researcher need not be limited to any particular method. Not all of the information,
however, applies to any single MPN assay. Any MPN assay of a legume inoculant is based on an
initial series of dilutions of the inoculant, which is then further serially diluted, and the second
series of dilutions inoculated in successive fashion onto a number of replicate plant growth units.
The means of obtaining the initial inoculant dilution, the degree of dilution of the serial dilutions
applied to the plants (called the base dilution ratio), the number of serial dilutions applied, and the
volume applied to each replicate plant unit can be selected by the researcher. Choices may be
based on the number of plant growth units which can be grown and maintained, by the size of the
plant growth units used, or by the equipment and time available. Another factor is the accuracy
required of the result obtained. Generally, the higher the number of plant replicates used and the
lower the base dilution ratio, the smaller will be the 95% confidence limits on either side of the
actual population estimate obtained.
29
BASIC ASSUMPTIONS OF THE MPN PLANT-INFECTION TEST
Rhizobial MPN enumeration is based on the following major assumptions:
1. A single, viable rhizobial cell in the root area of a young legume host of the appropriate
type will cause nodule formation in nitrogen free media.
2. Nodule formation is evidence of the presence of viable infective rhizobia.
3. The validity of the MPN assay is demonstrated by the absence of nodules on
uninoculated plants that are otherwise treated identically to the inoculated plants.
4. Absence of nodule formation is evidence of the absence of viable infective rhizobia.
5. Rhizobia present in the initial and all subsequent dilutions are evenly distributed.
DESIGNING AN MPN ASSAY
Selection of an experimental design
The process of designing an MPN assay for legume inoculants involves selecting:
1. the degree of initial dilution of the sample
2. the base dilution ratio of the serial dilutions applied to plants
3. the number of serial dilution steps applied to plants
4. the number of replicate plants to be inoculated at each dilution level
5. the volume of inoculant applied to each plant
Care must be taken to design MPN assays for which population estimates can be obtained
through MPN tables. Standard MPN tables do not provide estimates for all dilution ratio- replicate
combinations. The MPNES computer program, described later, provides considerable additional
flexibility.
The upper range of rhizobial populations which can be detected with an MPN are functions
of 1) the amount of dilution provided to the sample prior to the serial dilutions applied to the
plants, 2) the base dilution ratio of dilutions applied to the plants, and 3) the number of dilution
steps. If the number of rhizobia in an inoculant cannot be anticipated, a very wide range can be
accommodated by using a 10-fold dilution series. However, 5-fold or 4-fold dilution ratios result in
greater accuracy.
30
The base dilution ratio and the number of replicate plants per dilution level are used in the
calculation of a confidence factor which describes the reliability of the MPN result. The population
estimate is multiplied or divided by the confidence factor to establish the upper and lower limits,
respectively, of the confidence interval for the population estimate. Decreasing the base dilution
ratio or increasing the number of replicate plants per dilution level results in a narrowing of the
range of the confidence interval and a greater resolution of the MPN estimate. Conversely,
increasing the base dilution ratio or decreasing replication results in broader confidence intervals.
Confidence factors associated with different dilution ratios and replicate combinations are
presented in Table 5. For two population estimates to be significantly different from one another,
the lower limit of the greater population must be higher than the upper limit of the lesser
population.
Altering the volume of inoculant applied to the root system of the host legume is another
option in the design of plant-infection assays. Most published MPN tables assume a 1.0 mL
inoculant volume. An MPN population estimate can be obtained from those tables after applying
greater or lesser inoculant volumes by using the relationship:
POPULATION ESTIMATE = (1 / INOCULANT VOLUME) X TABULAR ESTIMATE
For example, if an inoculant volume of 2 mL is applied to the root system, the population
estimate is half that of the tabular MPN estimate. Similarly, if 0.5 mL is applied, the population
estimate is twice that of the tabular value. Increased inoculant volumes (greater than 1 ml) are
useful when larger plant growth containers are used, or when the researcher wants to lower the
range of population detection. Decreased inoculant volumes (less than 1 ml) are useful when host
plants are grown in small culture tubes or when the researcher wants to extend the upper range of
population detection. Adjustments in the inoculant volume do not affect the confidence factor.
31
Table 5. Factors for calculating the confidence intervals of Most-ProbableNumber estimates.
Factor for 95% confidence interval at various dilution ratios 1
replicates per dilution
2
3
4
5
10
1
4.01
5.75
7.14
8.31
14.45
2
2.67
3.45
4.01
4.47
6.61
3
2.23
2.75
3.11
3.40
4.67
4
2.00
2.40
2.67
2.88
3.80
5
1.86
2.19
2.41
2.58
3.30
6
1.76
2.04
2.23
2.37
2.98
7
1.69
1.94
2.10
2.23
2.74
8
1.63
1.86
2.00
2.11
2.57
9
1.59
1.79
1.93
2.03
2.44
10
1.55
1.74
1.86
1.95
2.33
1
Population estimates are multiplied and divided by the confidence factors to establish the
upper and lower confidence intervals at (p = 0.05), respectively.
Confidence factors were calculated using MPNES software (Woomer et al., 1990) after
Cochran (1950).
32
A TYPICAL MPN DESIGN FOR INOCULANT EVALUATION
We provide here a commonly used MPN design for the evaluation of legume inoculant
products. The purpose is to aid understanding by providing an example of MPN design in which
the various parameters have been established by experienced personnel. This design is that used
officially within the Canadian regulatory inoculant testing program (Anonymous, 1992).
1. Using growth pouches and a suitable supporting rack prepare 30 pouches (growth units)
with a suitable number of healthy host plants per pouch and sterile N-free nutrient solution
(30 mL for whole pouches, 15 mL per side for split pouches). This makes six sets of 5
growth units.
2. Taking a representative sample of the inoculant, dilute the sample by successive 10-fold
dilutions (in sterile diluent) to a suitable starting point based on the anticipated number of
rhizobia in the product. Given an anticipated rhizobial cell number of greater than 109 per
g, a suitable starting point for further dilutions would be a dilution of 10-6.
3. From the 10-6 dilution make an additional six serial 5-fold dilutions in sterile diluent (the
initial dilution is 10-6, the base dilution ratio is 5).
4. Apply 1.0 mL to each of four replicate growth units from each of the six 5-fold dilutions.
Leave the 5th growth unit in each series uninoculated. Whether one works from the most
dilute of the 5-fold dilutions to the least dilute, or vice versa, depends upon the technique
used to apply the inoculant suspension, but in any case work in order, and be consistent.
5. Place the growth-unit rack containing the MPN assay in an appropriate growth
environment and maintain suitable plant condition by watering with sterile water.
6. After 3-4 weeks (depending upon the legume host used) examine the plants and record
each growth pouch unit as "+" or "-" for nodulation. Even one nodule per growth unit means
a "+" score for that unit. Do not count "maybe" nodules. All six negative control units must
be free of nodules (if not, discard the test, repeat, and work on technique).
7. For each of the six successive 5-fold dilutions a number of pouches between 0 and 4 will
have been scored as "+" for nodulation. The number of "+" scores taken in order of
increasing dilution will yield a six digit number. A typical result could be: 4, 4, 4, 1, 1, 0.
8. Using either the MPNES computer program or an MPN probability table based on six 5fold dilutions, 4 replicates, and 1.0 mL volume applied per growth unit, determine the
population estimate and 95% confidence limits. This estimate refers to the number of
rhizobia present in 1 mL of the suspension from which the first 5-fold dilution applied to the
plants was made. Multiply the MPN estimate by the reciprocal of the dilution level of the
inoculant suspension prior to beginning the 5-fold serial dilutions (in this case 106) to get an
estimate for the number of viable rhizobia in the original inoculant. For the example given
above the estimate is 379 x 106 = 3.79 x 108 rhizobia per g.
33
CONDUCTING THE MPN ASSAY
Selection of an appropriate host legume
No legume host is nodulated by all rhizobia and no single Rhizobium can nodulate all
nitrogen-fixing legumes. Rather, the association of rhizobia with legumes is relatively specific;
population estimates from plant-infection counts therefore depend on the choice of host legume.
Rapid development of easily identifiable nodules occurs when the rhizobia and the host legume
belong to the same effective cross-inoculation group. When evaluating the number of rhizobia in
inoculant, smaller-seeded legume species can sometimes be substituted for the legume host of
interest when the two hosts have been shown to nodulate equally with the rhizobial strain involved.
Examples of acceptable substitutions include Macroptilium atropurpureum (siratro) for Vigna
unguiculata or Trifolium repens for T. pratense. Table 6 presents a list of species of root and
stem-nodulating bacteria and their hosts.
Selection of a plant growth system
The success of MPN plant-infection techniques is largely dependant upon the ability of
the researcher to produce healthy, uniform host plants. These assays are best conducted in a
growth chamber or growth room, however, some plant systems are suitable for the greenhouse.
Attempts to grow the host legumes next to a "sunny window" seldom provide the plants with
sufficient light, and should not be attempted.
Determine the type and number of growth containers needed. Prepare enough to discard
the least uniform 25 % of the plants. The units must be completely free of rhizobia or, preferably,
sterile.
Plant containers for host legumes
Size of the plant growth container is determined by the size of the host legume seed, the
growth rate of the plant, and the number of seeds planted in each container. A container of the
proper size will allow the roots to completely explore the media and will enhance the probability
that the roots will contact the rhizobia. The likelihood of outside contamination is also a
consideration when selecting the container. The various advantages and disadvantages of different
plant containers used in plant infection counts are summarized in Table 7.
34
Table 6. Species or root and stem-nodulating bacteria and their hosts*
Rhizobium species:
Host legumes:
Rhizobium meliloti
Medicago, Melilotus, Trigonella
Rhizobium fredii, R. xinjiangensis
Glycine max, G. soja, and other legumes
Rhizobium leguminosarum
bv. viciae
bv. trifolii
bv. phaseoli
Pisum, Vicia
Trifolium
Phaseolus
Rhizobium tropici
Phaseolus vulgaris, Leucaena spp.
Rhizobium etli
Phaseolus vulgaris
Rhizobium galegae
Galega officinalis, G. orientalis
Rhizobium loti
Lotus spp.
Rhizobium huakuii
Astralagus sinicus
Bradyrhizobium species:
Bradyrhizobium japonicum
Bradyrhizobium elkanii
Glycine max
Glycine max
Azorhizobium species:
Azorhizobium caulinodans
Sesbania rostrata
*From Martinez-Romero, 1994.
G r o w t h p o u c h e s - Growth pouches are appropriate for both small and large seeded legumes
such as: Medicago sativa, Trifolium pratense, Onobrychis viciaefoli, Glycine max, Arachis
hypogaea, Phaseolus vulgaris, Phaseolus lunatus, and Vigna unguiculata. The pouches consist of a
semi-rigid plastic bag containing a flat paper wick. Thirty mL of N-free plant nutrient solution is
placed inside the pouch. The seeds are planted in a trough at the top of the wick. The pouches are
35
normally initially free of rhizobia, and can usually be used in plant-infection assays without prior
sterilization. For small seeded legumes, the pouches can be divided vertically into smaller units
with a heat sealer thus providing two growth units from one pouch. It is sometimes helpful to
refold the paper trough of the growth pouch to better enclose and keep wet larger legume seeds. A
flexible plastic straw (or disposable Pasteur pipette) can be inserted into the pouches to facilitate
watering. This decreases the risk of cross-contamination that is associated with the watering
process. Insects can also be a source of contamination when using pouches. Partially closing the
pouches with tape, paper clips, or staples can help to reduce contamination. A simple wire rack
can be constructed to hold the growth pouches as illustrated in Figure 4. It is sometimes necessary
to provide side and end covers for the racks to prevent exposure of the developing roots to light.
A g a r s l a n t s - Small-seeded legumes can also be grown in foam (or cotton plug) stoppered glass
tubes containing N-free plant nutrient agar slants. Nodulation of uninoculated plants is rare, due in
large part to the protection afforded by the stopper at the mouth of the tube. Agar slants are timeconsuming to prepare and seed.
L e o n a r d j a r s - While Leonard jars are commonly used as aseptic growth containers in the
study of the legume - Rhizobium symbiosis, the value of Leonard jars and similar systems using
vermiculite is questionable for plant-infection MPN counts for several reasons (see Table 7).
Limitations include the inability to apply the inoculant directly to root surfaces, the difficulties in
observing and recovering nodules, and the possibility of the vermiculite preventing contact of the
rhizobia with the roots. These disadvantages frequently result in statistically unacceptable MPN
results and underestimates of the population size.
Figure 4. A rack constructed of heavy wire on a wooden base for holding
series of growth pouches.
36
Table 7. Adv antages and disadvantages of plant containers recommended
for use in MPN counts of rhizobia.
Growth
system
Agar
tubes
Advantages
Disadvantages
1. Least susceptible to rhizobial
contamination
1. Limited volume for plant
growth
2. Compact, space efficient
2. Limited inoculation volume
3. Require little maintenance
3. Not suited for large-seeded
legumes
4. Roots and nodules easily visible
5. Tubes reusable
Growth
pouches
1. Plant growth less confined
1. Requires frequent watering
2. Easy preparation and assembly
2. Higher risk of contamination
than tubes
3. Unlimited inoculation volume
Leonard
jars
4. Roots and nodules easily visible
3. Require more space than
tubes
5. Roots easily inoculated
4. Require construction of racks
6. Suitable for small- and large-seeded
legumes
5. Not reusable
1. Excellent plant growth
1. Roots cannot be observed
2. Require little maintenance
2. Roots not inoculated directly
3. Unlimited inoculation volume
3. Media interacts with rhizobia
4. Low risk of contamination
4. Assembly difficult and time
consuming
5. Suitable for large-seeded legumes
37
5. Often poor MPN results
Preparation of plant nutrient solution
Two different plant nutrient solutions which have been used successfully in MPN assays are
prepared as shown in Tables 8 and 9. It is convenient to prepare stock solutions for use in
preparation of nutrient solutions. When preparing the nutrient solution shown in Table 9, it is
important to dissolve the two chemicals for stock number 6 separately before combining. If
nutrient solution is to be autoclaved, the CaSO4 should be sterilized separately and added last and
after the autoclaving to prevent formation of an insoluble precipitate.
Table 8. Plant nutrient solution composition (1).
Component
Final
Concentration
mg/L PN
g/L stock
solutions
mL per L of stocks to
make plant nutrient
solutions
Micro-nutrients
CoCl2⋅6H2O
H3BO3
MnCl2⋅4H2O
ZnSO4⋅7H2O
CuSO4⋅5H2O
Na2MoO4⋅2H2O
H2MoO4⋅H2O
0.004
2.86
1.81
0.22
0.08
0.121
0.09
0.004
2.86
1.81
0.22
0.08
0.121
0.09
1.0
246.48
174.18
136.09
110.99
5.00
2.0
1.0
1.0
1.0
1.0
Macro-nutrients
MgSO4⋅7H2O
K2HPO4
KH2PO4
CaCl2
FeC 6H5O7⋅H2O
492.96
174.18
136.09
110.99
5.00
For complete nutrient solution add 1 mL of NH4NO3 (stock = 8 g/100 mL).
Add 15 mL of nutrient solution to split pouches and 30 mL to whole pouches.
38
T a b l e 9 . P l a n t n u t r i e n t s o l u t i o n c o m p o s i t i o n ( 2 ) aa .
Stock solution
Without N
1
2
3
4
5
6b
7
Plus N
8c
Quantity
g per liter
Quantity of stock
per liter H2O
K2SO4
MgSO4⋅7H2O
KH2PO4
K2HPO4
CaCl2
CaSO4
FeCl3
Na2H2EDTA
H3BO3
MnSO4⋅H2O
ZnSO4⋅7H2O
CuSO4⋅5H2O
Na2MoO4⋅2H2O
CoCl2⋅6H2O
NiCl2
93
493
23
145
56
3 mL
1 mL
1 mL
KNO3
(NH4)2SO4
10
133
Chemical
6.5
13
0.23
0.16
0.22
0.08
0.025
0.034
0.022
1 mL
1g
1 mL
1 mL
1 mL
a
Modified from Evans et al. (1972) as cited by Weaver and Graham (1994).
Dissolve the two chemicals separately before combining.
c
For nutrient solution containing N, substitute stock solution 8 for 1.
b
P r e p a r a t i o n a n d g e r m i n a t i o n o f s e e d an d e s t a b l i s h m e n t o f p l a n t s i n g r o w t h
units
Determine the amount of seed needed based on germination percentage determined by
prior tests. Before planting host legumes in plant containers, the seeds are usually surface sterilized
and pre-germinated. Many legumes have impermeable seed coats that require scarification for
uniform germination. These seeds can be treated with concentrated sulfuric acid, which
simultaneously surface sterilizes and scarifies the seed. Different methods of surface sterilization
and scarification are presented in Tables 10 and 11. Wash the seeds at least five times in sterile
water following surface sterilization or acid scarification. Seeds which have been surface sterilized
by means other than acid treatment can be aseptically blotted dry and stored for later use.
39
Alternatively, the seeds can be soaked in the final rinse water until fully imbibed (2-24 hours) and
sown into growth units.
Small seeded legumes can be sown unimbibed in prepared growth pouches and allowed to
germinate in darkness at 20° C. If the growth units chosen for small seeded legumes are agar
slants, the seed should be pregerminated on water agar (1%) in petri plates (temperature 20 - 30° C
depending on the host legume; temperate, tropical, etc.). Incubate the plates inverted so that the
radicles do not grow into the agar. When sowing small-seeded legumes in agar slant tubes, the
germinated seeds are individually transferred to the slant when the radicles are 0.2-1.0 cm in
length. If necessary, germinated seeds of some species can be stored under refrigeration for a day
or two prior to transplanting. Transfer germinated seeds with flamed and cooled forceps or with a
wire hook made from a microbiological transfer loop. The radicles should be placed on the
surface of the slant and not embedded into the media. Do not sow plants with damaged radicles.
For large seeded legumes sown in growth pouches, it is necessary to refold the wick so that
the paper trough is deeper to prevent seed dehydration and to poke holes in the trough of the pouch
and insert the radicle of each seed into a hole. Grow the plants for 3-6 days before inoculation, and
discard growth units showing significant lack of uniformity in the number of plants or in root or
shoot development.
40
Table 10. Common methods for surface sterilization and scarification of
seeds.
Treatment
Concentration
Procedure
Surface sterilization
no treatment
n.a.
Used only after planting many seeds (>500) in
Rhizobium-free growth containers without
subsequent nodulation. Used with commercial
quantities of extremely small-seeded, uniformly
germinating species that are adversely affected by
chemical treatment (e.g., Trifolium repens).
sodium hypochlorite
1-3%
Soak seeds for 1-5 min. followed by several (5 or
more) rinses with sterile water. Leave seeds in final
rinse until imbibed (1-4 hours). Useful for all
legumes, reduce concentration to treat smallerseeded species. This treatment should be
preceded by a 30 second rinse with 70% ethanol
or isopropyl alcohol.
mercuric chloride
0.2%
Soak seeds for 3 min. followed by repeated rinses.
Mercuric chloride is toxic and there are difficulties
associated with its safe disposal.
sulfuric acid
conc. (95%)
Immerse dry seeds for 5-60 min. depending on the
thickness of the seed coat. Drain sulfuric acid,
rinse repeatedly (8 or more rinses), leave seeds in
final rinse until imbibed (4-24 hours). Do not
attempt to rinse the seeds by adding water without
having first drained as much of the acid away as
possible.
mechanical
n.a.
Nick or abrade seed coat of recently surface
sterilized seeds with a sharp, Rhizobium-free
instrument. Regularly flame sterilize the instrument
between samples or groups of samples. Place
scarified seeds in sterile water until imbibed.
hot water
n.a.
Pour boiling water over seeds (1 liter/100 g seeds)
and allow to stand until cool. If imbibition does not
begin, drain and repeat procedure, allow seeds to
fully imbibe. This procedure must be preceded by
surface sterilization.
Scarification
41
At the NifTAL Center, seeds are routinely sterilized with sodium hypochlorite if there is no
requirement for scarification, or with concentrated sulfuric acid when scarification is required.
Species commonly requiring scarification and the recommended length of exposure to
concentrated sulfuric acid are presented in Table 11. If seeds do not imbibe after treatment, the
exposure time is likely to have been insufficient. The seeds can usually be blotted dry and
rescarified if imbibition does not occur. If sufficient seed are available conduct a scarification test.
Table 11. Legume hosts commonly requiring seed scarification and
approximate exposure times to concentrated sulfuric acid for uniform
germination.
Host legume species
Exposure time to
H 22 S O 44 ( m i n . )
Acacia albida
A. auriculiformis
A. koa
A. melanoxylon
Albizia falcata
A. lebbeck
A. saman
Cajanus cajan
Calliandra calothyrsus
Desmodium spp.
Enterolobium cyclocarpum
Glycine soya
Leucaena diversifolia
Macroptilium spp.
M. lathyroides
Neonotonia wightii
Pachyrhizus erosus
Prosopis pallida
Psophocarpus tetragonolobus
Robinia pseudoacacia
Sesbania rostrata
Sophora chrysophylla
Trifolium spp.
20
15
50
15
35
40
60
10
15
15
45
8
20
8
8
12
20
30
20
15
20
50
20
42
Preparing dilutions and inoculating the host plants
The dilution procedure must be a systematic and accurate sub-division of an inoculant.
Consequently, transfer volumes must be removed prior to settling and the dilutions shaken prior to
inoculation onto plants.
1. Prepare the diluent and dilution blanks using a minimal salt media, such as sterile
phosphate-peptone buffer of the following composition:
per 1000 mL distilled water:
peptone, 1.0 g
KH2PO4, 0.34 g
K2HPO4, 1.21 g
pH = 7.0
An alternative diluent to phosphate-peptone solution is to use a one-quarter strength
solution of the mineral salts present in YEMB (the yeast extract and mannitol are omitted).
The pH of diluents must be monitored and aseptically adjusted to near neutrality if
necessary.
Phosphate-peptone or mineral salts diluent is used instead of sterile deionized water in
order to eliminate a gradient in osmotic potential across the dilution series. The container
for the initial inoculant suspension should accommodate the addition of at least 11 g
inoculant to 99 mL (a 1:10 dilution) diluent and still have adequate room for good mixing
by shaking. Prepare the required number of sterile diluent blanks required for the MPN.
Screw-capped dilution bottles marked at the 99 mL level are very useful for preparing the
preliminary inoculant dilutions necessary prior to preparation of the second dilution series
which will be used for inoculation of the plants. These preliminary 10-fold dilutions can be
prepared in two ways: 11 mL to 99 mL = a 10-1 dilution or 1 mL to 99 mL = a 10-2 dilution.
Also needed will be dilution blanks for the serial dilution series which will be applied to the
plants. The quantity of diluent in these blanks will depend on the base dilution ratio
selected, for example, adding 2 mL suspensions to 8 mL blanks yields a 1:5 dilution.
2. Mix the inoculant thoroughly within its bag. Aseptically remove the inoculant (e.g., with
a flamed and cooled spatula), and weigh out 11 g if using 99 mL dilution blanks (10 g if
using 90 mL dilution blanks). Suspend the inoculant in sterile diluent. Place the bottle on a
wrist action shaker for 10 minutes of vigorous mixing. Adjust the bottle horizontally on the
shaker so that the suspension is hitting both ends of the bottle on each cycle. Carry out the
10-fold dilution series to the required level.
43
3. Serially dilute the final 10-fold dilution by pipetting 1-5 mL (transfer volume) into sterile
diluent to provide six serial 5-fold dilutions. The appropriate amount of diluent can be
calculated as follows:
diluent volume = [(dilution ratio -1) x transfer volume]
For example, for a dilution ratio of 5 and a transfer volume of 2 mL, the diluent
volume is (5-1) x 2 = 8 mL. Only a single dilution series per sample need be
prepared, as all replicate plant growth units at a given dilution level will be
inoculated from the same tube.
4. Inoculate replicate host plants by pipetting the inoculation volume (usually 1 mL) onto
the plant root systems (or into the paper wick trough of growth pouches). Work from the
highest dilution level to the lowest dilution level. Plant inoculation is facilitated through the
use of a repeating pipette. Those with removable, autoclavable cartridges that dispense
adjustable volumes are particularly useful. A single cartridge can be used for each sample
by inoculation of highest to lowest dilutions.
Another dilution and inoculation approach which can be used and is convenient when the
base dilution ratio level selected for inoculant application to the plants is 5 and the replicate
number of growth units is 4, involves the use of disposable 10 mL syringes. One mL of the final
10-fold inoculant dilution is pulled into the syringe and then a further 4 mL of the sterile buffer
diluent is pulled in. The syringe plunger is pulled back to the 10 mL mark and the contents shaken.
One mL is then added to each of the 4 replicate growth units leaving 1 mL of suspension in the
syringe. An additional 4 mL of sterile diluent is then pulled into the syringe, the unit shaken and
the second set of 4 growth units inoculated. The procedure is repeated through the six 5-fold
dilution steps. This method is very time effective and saves preparation of dilution blanks, but
involves the expense of disposable syringes and needles and is less accurate than using separate
pipettes or tips for each dilution level.
Maintenance and watering of growth units
Given an overall uniformity in growthroom environment and the pre-selection of plants for
uniformity prior to inoculation, it is not necessary to randomize plant growth units prior to
inoculation. During inoculation, replicate plants should be adjacent to one another to reduce the
likelihood of errors in inoculation. The dilution treatments within an MPN test should be
systematically arranged.
1. When different growth containers are being used simultaneously, as when different hosts
are being used, group the dilution series by the type and age of plant since larger plants can
shade smaller plants and affect growth.
2. To reduce the likelihood of contamination, higher dilutions should be placed adjacent to
one another, and lower dilutions should be placed adjacent to each other. If a positive
control is used, it should be located next to the units receiving the most concentrated
44
inoculant suspension rather than the least concentrated.
3. Uninoculated growth unit controls must be included in any experimental MPN design.
These controls demonstrate that nodule formation on the inoculated plants did not arise
from any source other than the inoculation.
4. Maintain the inoculated plants as required. Plants growing in agar slant tubes require
little additional maintenance. Growth pouches require periodic watering. Care must be
taken not to induce salt stress through evapotranspiration of the nutrient solution. In most
cases, addition of sterile water to replace the depleted nutrient solution results in adequate
plant growth and development. Watering is facilitated by using a gravity-feed system from
carboys connected to a UV sterilizer. It is absolutely essential that the plant water remain
rhizobia-free. A common source of nodulated negative controls in an MPN test is the water
used for plant maintenance. In all cases, the formation of nodules on uninoculated control
plants nullifies the entire MPN test.
RESULTS AND ANALYSIS
Data Collection
Nodulation occurs 2-4 weeks following inoculation. Nodulation is slowest at higher
dilutions. With some legumes, nodules are still appearing in the high dilutions during the fourth
week. Inspect plant root systems for the formation of root nodules 21-28 days following
inoculation and record the result for each unit at each dilution. Results are recorded as either
positive or negative for each growth unit.
Evaluation of acceptability of results
The results of a typical assay are presented in Table 12 below which shows a six step, 10fold serial dilution series with four replicate plants per dilution and an inoculation volume of 1 mL.
45
Table 12. Example of results of an MPN assay.
Dilution level
Total
Positive
Replicate
1
2
3
4
10-6
+
+
+
+
4
10-7
+
+
+
+
4
10-8
+
+
+
10-9
+
3
1
10-10
0
10-11
0
The number of dilution steps from and including the first not entirely positive to the last not
entirely negative dilution is called the Range Of Transition (ROT). In the above example, the
experimental code is 4-4-3-1-0-0, and the ROT = 2.
The ROT is a direct measure of experimental compliance with the principle assumptions
underlying the MPN procedure, namely, that a single cell is capable of producing a root nodule
and that the cells are randomly distributed. Lower values (ROT = 0 to 2) indicate acceptable
results. Higher ROT values (>4) indicate serious problems in experimental technique. ROT
values of 3 are acceptable under some circumstances. Every dilution series has a single ROT
value. The probability of meeting the primary assumptions underlying the MPN technique can be
determined for a given ROT, dilution ratio, and number of replicates (Table 13). When the
column for 10-fold dilutions with four replicates is located on the table, a ROT value of 2 yields a
probability of 0.271, and the dilution series is acceptable. An experimental code of 4-2-3-0-2-0
developed under similar experimental conditions has a ROT of 4 and a probability of 0.004. We
are certain to a probability value of 0.996 that the results of this dilution series do not comply with
the underlying assumptions and the results are therefore questionable.
Stevens (1958) suggested that this test of technique not be applied until a bulk of results has
been examined and that the technique be used to discover and remove procedural deficiencies.
Researchers might appropriately adopt a rule of rejecting results at p=0.01.
46
Table 13. Expected frequencies of equalling or exceeding the ROT (from Stevens,
1957; Scott and Porter, 1986) for various dilution ratio replicate number combinations.
________________________________________________________________________
Range
Probability of observing a given ROT value
of
Transition
Dilution Ratio
2
4
5
10
________________________________________________________________________
--------------------Two replicates per dilution level-------------1
2
3
4
5
6
0.930
0.820
0.625
0.415
0.246
0.136
0.717
0.373
0.123
0.034
0.009
0.002
0.660
0.281
0.075
0,015
0.003
0.0006
0.525
0.114
0.013
0.001
0.0001
2
0.00001
--------------------Three replicates per dilution level-----------1
2
3
4
5
6
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
0.891
0.511
0.208
0.060
0.015
0.004
0.851
0.435
0.123
0.027
0.005
0.001
1
2
3
4
5
6
--------------------Four replicates per dilution level-------------n.a.
0.955
0.931
n.a.
0.682
0.561
n.a.
0.294
0.178
n.a.
0.088
0.040
n.a.
0.023
0.008
n.a.
0.006
0.002
0.731
0.193
0.023
0.002
0.0002
0.00002
0.838
0.271
0.035
0.004
0.0004
0.00004
--------------------Five replicates per dilution level--------------1
n.a.
0.981
0.967
0.899
2
n.a.
0.777
0.661
0.340
3
n.a.
0.371
0.232
0.047
4
n.a.
0.118
0.054
0.005
5
n.a.
0.031
0.011
0.0005
6
n.a.
0.008
0.002
0.00005
________________________________________________________________________
Frequency distributions for 3-fold dilution series are not available.
n.a. = frequency distribution not available.
47
Determination of the viable rhizobial population estimate and the 95%
confidence levels for the estimate
Population estimates are assigned by locating the experimental results on the appropriate
MPN table (Tables 14-17). These tables are organized by replicate number. Likely results of the
dilution series are listed in the first column. When the correct code is located, the researcher
obtains a population estimate from the adjacent column which corresponds to the base ratio of the
serial dilution (2,3,4,5,or 10).
For example, given the six-step, 10-fold dilution series with 4 replicates which yielded the
experimental results 4-4-3-1-0-0 (Table 12) the researcher first locates the table for 4 replicates
(Table 16). The experimental code is located near the center of the experimental results column.
The population estimate obtained from the column for 10-fold dilutions is 1592 cells per mL. This
population estimate reflects the number of cells in the 10-5 dilution suspension and assumes a plant
inoculation volume of 1.0 mL. This number multiplied by the reciprocal of the level of dilution
prior to plant inoculation (105) provides the estimate for the population in the inoculant sample (1.6
x 108 viable rhizobia per g inoculant). If inoculation volumes were other than 1.0 mL, additional
calculations are required.
The lower 95% confidence limit is obtained by dividing the population estimate by the
confidence factor. The upper 95% limit is obtained by multiplying the population estimate by the
confidence factor. Confidence factors are listed at the bottom of each population estimate column
or can be obtained from Table 5. For the 10-fold dilution series with 4 replicates (Table 16), the
confidence factor is 3.80 (p=0.05). The experimental result 4-4-3-1-0-0 resulted in a population
estimate of 1592 cells, the lower confidence limit (p=0.05) equals 1592/3.80, or 419 cells per mL.
The upper confidence limit equals 1592 x 3.80 or 6050 cells. The final results may be expressed
as 1.6 x 108 cells per g inoculant (4.2 x 107 - 6.0 x 108, p=0.05).
Determining population estimates and 95% confidence limits
1. Locate the appropriate MPN table based on the base dilution ratio of plant applied
inoculum and the number of replicates.
2. On this table, find the experimental outcome (MPN code) identical to the results
obtained from nodulation data. Not all possible experimental outcomes are listed on MPN
probability tables. Subjective interpolation is often made, but should be related reasonably
to the ROT. To the right of the MPN code is the population estimate corresponding to the
result for that particular dilution ratio and number of replicates.
3. These tables assume an inoculation volume of 1.0 mL. If a different inoculation volume
was applied to the host legumes, the population estimate must be corrected by dividing the
inoculation volume into the tabular population estimate.
4. To obtain the upper and lower confidence limits (p = 0.05) multiply and divide the
population estimate by the confidence factor found at the base of each MPN table, or
48
obtained from Table 5.
5. Two population estimates are significantly different from one another when their
confidence intervals do not overlap.
Most-Probable-Number Enumeration System (MPNES) - a computer
program to calculate population estimates and confidence intervals.
An IBM compatible software program has been developed at NifTAL (Woomer, et al.,
1990). which provides the population estimate and confidence limits for results of six step dilution
series MPN tests. Users provide the serial dilution ratio and replicate number, and have the option
to input the inoculant volume (default = 1.0 mL). The program generates the population estimate
and confidence limits for the test sample. MPNES is useful in that solutions can be obtained for
experimental results that do not appear in standard MPN tables. MPNES also generates MPN
tables for dilution ratios <15 (including non-whole numbers) for 2,3,4,5, and 10 replicates.
49
Table 14. Most probable number of 2, 3, 4, 5 and 10-fold dilution series replicated
twice.
11
no. of positive results at
------- dilution level ------1-2-3-4-5-6
2
0-1-0-0-0-0
1-0-0-0-0-0
1-0-1-0-0-0
1-1-0-0-0-0
2-0-0-0-0-0
2-0-1-0-0-0
2-1-0-0-0-0
2-1-1-0-0-0
2-2-0-0-0-0
2-2-0-1-0-0
2-2-1-0-0-0
2-2-1-1-0-0
2-2-2-0-0-0
2-2-2-0-1-0
2-2-2-1-0-0
2-2-2-1-1-0
2-2-2-2-0-0
2-2-2-2-0-1
2-2-2-2-1-0
2-2-2-2-1-1
2-2-2-2-2-0
2-2-2-2-2-1
confidence factor
Population Estimate
----------- ratio of the dilution series ---------3
4
0.5 1.0
1.5
0.6 1.2
1.8
1.2 2.5
3.9
1.3 2.6
4.1
1.4 3.3
5.5
2.3 5.4
9.4
2.4 6.0
11.0
3.6 9.3
17.0
3.911.0
23.0
5.416.7
38.0
5.718.4
43.0
7.828.1
70.0
8.433.5
91.0
11.451.1 152.0
12.056.1 172.0
1686.4
285.0
18104.0
372.0
24162.0
638.0
26179.0
725.0
38285.0 1200.0
43359.0 1600.0
72703.0 2200.0
2
2.673.45
1
5
4.6
2.5
5.2
5.5
8.0
14.0
16.0
28.0
40.0
71.0
81.0
142.0
202.0
354.0
408.0
713.0
1016.0
1797.0
2109.0
3750.0
6250.0
15000.0
6.0
12.0
12.0
23.0
49.0
61.0
127.0
230.0
493.0
614.0
1270.0
2305.0
4844.0
5938.0
12500.0
25000.0
40000.0
60000.0
120000.0
120000.0
240000.0
4.47
6.61
4.01
This is the population density in the test substrate assuming a 1 mL inoculation
volume. Table generated using MPNES software (Woomer et al., 1990).
The population estimate is divided and multiplied by the confidence factor to
establish the lower and upper confidence limits (p = 0.05), respectively.
2
50
10
T a b l e 1 5 . M o s t- p r o b a b l e - n u m b e r o f 2 , 3 , 4 , 5 a n d 1 0 - f o l d d i l u t i o n s e r i e s w i t h 3
replicate units per dilution level.
11
no. of positive results at
------- dilution level ------1-2-3-4-5-6
population estimate
----------- ratio of the dilution series ---------2
1-0-0-0-0-0
1-0-1-0-0-0
1-1-0-0-0-0
2-0-0-0-0-0
2-1-0-0-0-0
3-0-0-0-0-0
3-0-1-0-0-0
3-1-0-0-0-0
3-1-1-0-0-0
3-2-0-1-0-0
3-2-1-0-0-0
3-3-0-0-0-0
3-3-0-1-0-0
3-3-1-0-0-0
3-3-1-1-0-0
3-3-2-0-0-0
3-3-2-1-0-0
3-3-3-0-0-0
3-3-3-0-1-0
3-3-3-1-0-0
3-3-3-1-1-0
3-3-3-2-0-0
3-3-3-2-1-0
3-3-3-3-0-0
3-3-3-3-0-1
3-3-3-3-1-0
3-3-3-3-1-1
3-3-3-3-2-0
3-3-3-3-2-1
3-3-3-3-2-2
3-3-3-3-3-0
3-3-3-3-3-1
3-3-3-3-3-2
confidence factor
3
0.3
0.7
0.7
0.8
1.3
1.4
1.9
2.0
2.7
2.8
3.7
3.9
4.8
5.0
6.1
6.4
8.0
8.4
10.0
11.0
13.0
14.0
17.0
18.0
22.0
23.0
29.0
31.0
40.0
53.0
44.0
60.0
93.0
2
2.23
4
0.7
1.5
1.5
1.7
2.8
3.3
4.7
5.0
6.7
7.2
9.8
11.0
15.0
15.0
20.0
22.0
30.0
33.0
44.0
47.0
63.0
68.0
91.0
104.0
140.0
149.0
200.0
220.0
306.0
422.0
359.0
548.0
956.0
5
10
1.1
2.3
2.4
2.7
4.4
5.5
8.0
8.6
12.0
13.0
19.0
23.0
32.0
34.0
48.0
53.0
75.0
91.0
129.0
139.0
194.0
216.0
305.0
372.0
531.0
575.0
800.0
900.0
1300.0
1900.0
1700.0
2600.0
5600.0
1.5
3.1
3.2
3.8
6.1
8.0
12.0
13.0
18.0
21.0
31.0
40.0
59.0
64.0
93.0
105.0
156.0
203.0
298.0
322.0
467.0
532.0
791.0
1035.0
1523.0
1680.0
2422.0
2969.0
4219.0
6250.0
6563.0
10000.0
20000.0
3.5
7.2
7.3
9.1
14.0
23.0
38.0
42.0
74.0
92.0
147.0
230.0
382.0
425.0
738.0
919.0
1466.0
2305.0
3829.0
4219.0
7188.0
9375.0
14375.0
25000.0
40000.0
40000.0
80000.0
80000.0
120000.0
160000.0
120000.0
240000.0
240000.0
2.75 3.11
3.40
4.67
1
This is the population density in the test substrate assuming a 1 mL inoculation
volume. Table generated using MPNES software (Woomer et al., 1990).
The population estimate is divided and multiplied by the confidence factor to
establish the lower and upper confidence limits (p = 0.05), respectively.
2
51
T a b l e 1 6 . M o s t p r o b a b l e n u m b e r o f 2 , 3 , 4 , 5 a n d 10-fold dilution series with 4
replicate units per dilution level.
no. of positive results at
population estimate
------- dilution level ------------------ ratio of the dilution series ----------11
1-2-3-4-5-6
1-0-0-0-0-0
1-1-0-0-0-0
2-0-0-0-0-0
2-1-0-0-0-0
3-0-0-0-0-0
3-1-0-0-0-0
3-2-0-0-0-0
4-0-0-0-0-0
4-1-0-0-0-0
4-1-1-0-0-0
4-2-0-0-0-0
4-2-1-0-0-0
4-3-0-0-0-0
4-3-1-0-0-0
4-3-2-0-0-0
4-4-0-0-0-0
4-4-1-0-0-0
4-4-1-1-0-0
4-4-2-0-0-0
4-4-2-1-0-0
4-4-3-0-0-0
4-4-3-1-0-0
4-4-3-2-0-0
4-4-4-0-0-0
4-4-4-1-0-0
4-4-4-1-1-0
4-4-4-2-0-0
4-4-4-2-1-0
4-4-4-3-0-0
4-4-4-3-1-0
4-4-4-3-2-0
4-4-4-4-0-0
4-4-4-4-1-0
4-4-4-4-1-1
4-4-4-4-2-0
4-4-4-4-2-1
4-4-4-4-3-0
4-4-4-4-3-1
4-4-4-4-3-2
4-4-4-4-4-0
4-4-4-4-4-1
4-4-4-4-4-2
4-4-4-4-4-3
2
confidence factor
2
0.2
0.5
0.5
1.1
0.5
1.2
0.9
1.9
1.0
2.0
1.3
2.9
1.7
3.9
1.4
3.3
1.8
4.5
2.3
5.7
2.4
6.0
2.9
7.5
3.0
8.0
3.7
10
4.4
12
3.8
11
4.7
14
5.5
17
5.6
18
6.6
22
6.8
24
8.0
30
9.5
38
8.4
33
10
43
11
53
12
56
14
69
14
74
17
94
20 118
18 104
21 135
25 169
26 179
31 225
33 244
40 312
50 404
43 355
55 485
73 689
107 1068
2.00 2.40
52
0.8
1.7
1.8
2.9
3.3
4.6
6.2
5.5
7.7
9.0
10
13
14
19
25
22
31
39
42
54
60
78
102
91
125
161
172
221
243
318
417
373
518
668
718
681
876
961
1255
1632
1650
2350
3600
2.67
3
4
1.1
2.2
2.5
3.9
4.5
6.4
8.7
8.0
11
15
16
21
24
32
43
40
57
75
81
107
121
164
218
202
287
379
409
544
610
830
1113
1035
1464
1992
2109
2812
3281
4375
5937
5064
7177
10228
15263
2.88
2.5
5.1
5.9
9.2
11
15
21
23
35
54
61
92
112
159
213
230
359
544
613
926
1123
1592
2129
2305
3594
5469
6137
9262
11239
15926
21297
230545
359439
546920
613730
1123930
1592630
2129690
3594390
5469200
6137300
9262000
11239300
3.80
5
10
1
This is the population density in the test substrate assuming 1 mL inoculation volume. Table generated using MPNES software
2
(Woomer et al., 1990). The population estimate is divided and multiplied by the confidence factor to establish the lower and upper
confidence limits (p = 0.05).
53
Table 17. Most probable number of 2, 3, 4, 5 and 10-fold dilution series with 5 replicate units
11
per dilution level.
no. of positive results at
------- dilution level ------1-2-3-4-5-6
1-0-0-0-0-0
1-1-0-0-0-0
2-0-0-0-0-0
2-1-0-0-0-0
3-0-0-0-0-0
3-1-0-0-0-0
3-2-0-0-0-0
4-0-0-0-0-0
4-1-0-0-0-0
4-2-0-0-0-0
4-3-0-0-0-0
5-0-0-0-0-0
5-0-1-0-0-0
5-1-0-0-0-0
5-1-1-0-0-0
5-2-0-0-0-0
5-2-1-0-0-0
5-3-0-0-0-0
5-3-1-0-0-0
5-3-2-0-0-0
5-4-0-0-0-0
5-4-1-0-0-0
5-4-2-0-0-0
5-4-3-0-0-0
5-5-0-0-0-0
5-5-0-1-0-0
5-5-1-0-0-0
5-5-1-1-0-0
5-5-2-0-0-0
5-5-2-1-0-0
5-5-3-0-0-0
5-5-3-1-0-0
5-5-3-2-0-0
5-5-4-0-0-0
5-5-4-1-0-0
5-5-4-2-0-0
5-5-4-3-0-0
5-5-5-0-0-0
5-5-5-0-1-0
5-5-5-1-0-0
5-5-5-1-1-0
population estimate
----------- ratio of the dilution series ---------2
0.2
0.4
0.4
0.7
0.7
1.0
1.3
1.0
1.3
1.7
2.0
1.4
1.7
1.7
2.1
2.2
2.6
2.6
3.1
3.6
3.2
3.7
4.3
5.0
3.8
4.4
4.5
5.1
5.2
5.9
6.1
6.9
7.8
7.1
8.1
9.2
10
8.4
9.4
9.6
10
3
4
5
0.4
0.8
0.9
1.4
1.5
2.1
2.7
2.2
3.0
3.8
4.7
3.3
4.1
4.2
5.1
5.3
6.4
6.7
8.0
9.5
8.5
10
12
14
11
13
13
15
16
19
20
24
28
25
30
37
44
33
39
40
48
0.6
1.3
1.4
2.2
2.3
3.3
4.3
3.6
4.8
6.1
7.6
5.5
6.9
7.2
8.9
9.3
11
12
14
18
16
20
24
30
22
28
29
35
37
45
48
59
72
64
80
100
123
91
113
117
144
0.8
1.7
1.9
2.9
3.2
4.5
5.8
5.1
6.7
8.5
10
8.0
10
10
13
14
17
18
23
29
26
34
43
53
40
51
53
66
70
88
94
119
149
132
170
215
268
202
256
268
335
54
10
1.9
4.0
4.4
6.8
7.7
10
13
12
16
21
27
23
31
32
45
48
69
78
106
138
127
168
215
270
230
312
327
452
488
691
779
1070
1382
1275
1690
2158
2715
2305
3125
3281
4531
5-5-5-2-0-0
5-5-5-2-1-0
5-5-5-3-0-0
5-5-5-3-1-0
5-5-5-3-2-0
5-5-5-4-0-0
5-5-5-4-1-0
5-5-5-4-2-0
5-5-5-4-3-0
5-5-5-5-0-0
5-5-5-5-0-1
5-5-5-5-1-0
5-5-5-5-1-1
5-5-5-5-2-0
5-5-5-5-2-1
5-5-5-5-3-0
5-5-5-5-3-1
5-5-5-5-3-2
5-5-5-5-4-0
5-5-5-5-4-1
5-5-5-5-4-2
5-5-5-5-4-3
5-5-5-5-4-4
5-5-5-5-5-0
5-5-5-5-5-1
5-5-5-5-5-2
5-5-5-5-5-3
5-5-5-5-5-4
confidence factor
11
50
12
60
12
62
14
75
16
89
15
79
17
95
20 115
23 140
18 104
20 124
21 128
24 153
24 160
28 192
29 202
33 244
39 295
35 262
41 321
49 395
59 492
72 618
43 356
52 453
65 597
83 815
118 1237
151
185
196
241
296
262
328
409
509
373
465
484
596
625
768
778
954
1167
1036
1290
1602
1980
2436
1463
1880
2421
3146
4202
2
1
This is the population density in the test substrate assuming a 1 mL inoculation
volume. Table generated using MPNES software (Woomer et al., 1990).
2
The population estimate divided and multiplied by the confidence factor to
establish the lower and upper confidence limits (p = 0.05), respectively.
55
352
442
476
600
757
669
859
1089
1357
1025
1308
1357
1699
1796
2304
2460
3125
3906
3593
4687
5937
7187
9375
5781
7500
10000
17500
16728
4922
6875
7812
10625
13750
13125
16875
21250
27500
22500
35000
35000
60000
60000
70000
80000
120000
160000
120000
160000
240000
240000
240000
240000
240000
240000
480000
1312535
Evaluation of the growth system.
For any combination of host legume and plant growth environment, it is useful to compare the
population estimate obtained by MPN to that obtained using another technique, usually a plate
count. Population estimates of pure cultures of rhizobia which have been obtained by various
researchers from plant-infection assays and plate counts are compared in Table 18. Researchers
should conduct similar comparisons at the start of any work involving the MPN plant-infection
technique.
56
Table 18. Comparison between plant infection and plate counts.
Rhizobia-Host Legume
Plate
count
System (Reference)
MPN
log
10
Bradyrhizobium japonicum
Glycine max
Glycine ussuriensis
Glycine ussuriensis
8.24
4.32
3.89
8.21
4.74
3.54
Growth pouch (1)
Tubes with agar slants (2)
Tubes with vermiculite (2)
Bradyrhizobium spp.
Vigna unguiculata
Vigna unguiculata
Siratro
Siratro
Siratro
Siratro
8.95
8.63
8.04
8.04
8.04
9.51
7.85
6.41
6.48
6.47
6.82
9.16
Growth pouch, mean 2 count (3)
Growth pouch, mean 4 strains (4)
Growth pouch, mean 2 strains (4)
Vermiculite, mean 2 strains (4)
Agar slants, mean 2 strains (4)
Agar slants (5)
R. leguminosarum bv trifolii
Trifolium subterraneum
Trifolium subterraneum
Trifolium incarnatum
2.76
2.10
6.17
2.94
1.89
5.41
Agar slants (6)
Tubes with vermiculite (6)
Tubes with sand (7)
R. leguminosarum bv vicia
"Vetch"
9.34
8.90
Jars with sand, mean 3 counts (8)
Rhizobium meliloti
Medicago sativa
Medicago sativa
Medicago sativa
Medicago sativa
Medicago sativa
8.60
4.60
3.63
3.63
8.84
8.49
4.64
3.61
3.19
8.45
Growth pouch (1)
Growth pouch (1)
Growth pouch, dilutions incubated (9)
Growth pouch, dilutions not incubated
(9)
Agar slants, mean 4 counts (10)
Rhizobium spp.
Cicer arietinum
Cicer arietinum
9.55
9.48
5.24
9.43
Tubes with sand, whole seed (11)
Tubes with sand, cotyledons removed
before planting (11)
Source citations, Table 18: 1) Weaver and Frederick, 1972. Plant and Soil 36:219-222. 2) Brockwell et al. 1975. Soil Biol. and
Biochem. 7:305-311. 3) Wilson and Trang, 1980. Trop. Agric (Trinidad) 57:232-238. 4) Boonkerd and Weaver, 1982. Soil Biol.
Biochem. 14:305-307. 5) Woomer, P. unpublished. 6) Brockwell, 1963. Appl. Microbiol. 11:377-383. 7) Tuzimura and Watanabe,
1961. Soil Sci. Plant Growth 7:61-65. 8) Date and Vincent, 1962. Aust. J. Expt Agric. and Anim. Husb. 2:5-7. 9) Scott and Porter, 1986.
Soil Biol. Biochem. 355-362. 10) Toomsan et al., 1984. Soil Biol. Biochem. 16:503-507.
57
CHAPTER 6
IMMUNOLOGICAL TECHNIQUES FOR THE ANALYSIS OF RHIZOBIAL
B R O T H A N D I N O C U LANTS
GENERAL COMMENTS
When bacteria (including rhizobia) are injected into a mammal, the animal produces antibodies
which will bind to molecular structures (antigens) on the surface of the bacteria. If the surface
antigens on the rhizobia are relatively unique and unshared by other microorganisms the binding of
the antibodies can be used to both detect and identify the target rhizobia. A variety of techniques
based on antibody-antigen reactions and which are useful in the analysis of rhizobial broth or
inoculants have been developed. Serological techniques are relatively simple procedures (such as
agglutination tests) which use unpurified antisera to obtain information. More involved methods
which use purified antibodies or antibodies which have been cross-linked with signal producing
molecules such as fluorochromes or enzymes are usually called immunological techniques. For
practical purposes, the distinction is not of significance and we will use the term "immunological" in
a broad sense.
The immunological techniques used in quality control procedures to obtain information about
rhizobial cultures and products are all based on possession by the researcher of a primary antibody
that will react with the rhizobia of interest. It is equally important, however, that the primary
antibody NOT react with other microorganisms which may be present. It is not uncommon that
anti-rhizobial antisera are produced in rabbits, shown to react with the rhizobia used as antigen, and
the reaction is simply assumed to be specific. Such antisera will, however, very often cross-react
with other rhizobial strains within the same species. This cross-reactivity may also extend to other
rhizobial biovars or species and sometimes even to members of other bacterial genera.
Information gained by virtue of the reactivity of an antibody with an antigen is only as good as is the
knowledge about whether or not the antibody is cross-reactive with non-target antigens in the
system. Appropriate controls should be devised and employed routinely wherever possible.
Cross-reactivity in a raw antisera can often be eliminated or greatly reduced by cross-adsorptive
removal of antibodies with the cross-reacting antigen (using cells of the cross-reacting organism).
Methods for the production of antisera and the removal of cross-reactivity from them, as well as
procedures to purify antibodies and cross-link them with detecting molecules are beyond the scope
of this manual (refer to Somasegaran and Hoben, 1994). Still, it is no longer necessary for every
laboratory wishing to monitor rhizobia using immunological techniques to have the facilities and
expertise required to purify, label with a detecting signal, and repurify primary antibodies. High
quality secondary antibodies which will react with the researcher's primary antibody and which
carry any of a variety of detecting signal labels are readily available commercially. With one
exception, the techniques described in this manual can be carried out using a primary antisera
raised against the target rhizobia, either alone (agglutination), or in conjunction with commercially
available secondary antibodies or reagents. The exception is the direct fluorescent antibody
enumeration of rhizobia in inoculant technique which calls for the primary antibodies to be
covalently cross-linked with fluorochrome in the researcher's own laboratory. Even this procedure
is readily adaptable for use with a commercial secondary antibody which would come with the
58
fluorochrome attached.
THE
CELL
IDENTITY
AGGLUTINATION
REACTION
FOR
RHIZOBIAL
P u r p o s e - The agglutination reaction is an immunological tool which can confirm rhizobial
identity of cells from starter or fermentor broth.
A d v a n t a g e s - The agglutination reaction is the simplest of all immunological procedures and
involves only primary antisera. No secondary antibodies or signal producing labels are needed.
The technique is fast and does not require specialized equipment.
L i m i t a t i o n s - cross-reactivity between rhizobia is common with non-adsorbed primary antisera.
The agglutination test is not as sensitive as other immunological tests and does not demonstrate
cell viability.
The simplest method of testing broth cells to verify that the cells are the desired rhizobia is the
cell agglutination reaction. Agglutination means the massive clumping of cells caused by crosslinking with antibodies. If the antibodies do not attach to the cells, no agglutination occurs.
Therefore, if the antisera used was raised against the rhizobial strain used to make the broth, and
the cells in the broth agglutinate when exposed to the antisera, this is good evidence that the cells in
the broth are the same or very similar to the cells used to make the antisera.
It is necessary to distinguish between two types of agglutination because one type is relatively
specific and the other is widely cross-reactive. The somatic reaction is between cells and is specific
and the flagellar reaction is between the flagella of cells and is cross-reactive. The somatic reaction
forms tight clumps of cells and the flagellar reaction forms large loose flocs. In practice, the nonspecificity of the flagellar reaction can be avoided in one of two ways. Rhizobial antisera should be
produced against steamed cells because this process denatures flagellar proteins while leaving the
lipopolysaccharide antigens of the cell surface undamaged. Antisera prepared in this way will not
show flagellar reactions and will have greater specificity than antisera made using unprocessed cells
as antigen. If the only antisera available show flagellar agglutination reactivity, it can be eliminated
by steaming the test sample cells. Steaming can be accomplished in several ways, but the most
effective is to steam the cells in an autoclave for 20 minutes with either the exhaust valve open or
the pressure sensor rheostat adjusted so that the temperature does not rise beyond 100° C. Many
modern autoclaves have an isothermal cycle option in which the maximum temperature reached
can be pre-programmed.
Agglutination testing in practice is very simple and can yield a quick result, but some
preliminary work is required. This involves making or obtaining antisera that will agglutinate the
desired rhizobia and an evaluation of the potency (titer) of the antisera to determine what level of
dilution is most appropriate for a particular agglutination test. Methods for determining the titer of
antisera by means of the agglutination reaction are well described (Vincent, 1970; Somasegaran
and Hoben, 1994). The best dilution of antisera for agglutination testing is usually between 1:50
59
and 1:1000. Preliminary work must also have shown that known pure culture broth cells of the
rhizobia will agglutinate with the antisera under the same conditions used to perform the
agglutination testing on actual test samples. Performance of the agglutination reaction test also
requires that the rhizobial cells in the broth are well and evenly dispersed (not already agglutinated
or growing in clumps).
Stock suspensions of washed, steam-killed rhizobial cells (of the strains to be used to prepare
broth) must be prepared. These cells should be at approximately the same concentration expected
in the starter or fermentor broths (e.g., 1 x 109 per mL in PBS) and protected with a preservative
such as 0.01% NaN3. The stock cell suspensions should be kept refrigerated. These cells must
routinely agglutinate when mixed with the appropriate homologous antiserum under the same
conditions to be used for the testing of broth cultures. The stock cell preparations are for use as
positive controls in the agglutination test. These same stock cell suspensions can also be used as
controls for other types of immunological testing such as fluorescent antibody staining, the
immuno-spot blot, and the colony-lift immunoblot.
60
Figure 5. Positive and negative (left and right, respectively) agglutination
reactions in microtiter plate wells.
Agglutination reactions taking place in small test tubes can be detected with the unaided eye under
suitable conditions. If many samples are to be run simultaneously, agglutination reactions can be
carried out in round-bottom 96-well microtiter plates (see Fig. 5). Agglutination of rhizobial cells
can also be seen under the microscope, sometimes within minutes of mixing the cells and antisera.
Generalized procedure for agglutination testing
1. Prepare the pre-determined dilution of the appropriate anti-rhizobial antiserum in PBS. For
example, to make a 1:200 dilution of the antisera add 25 ìL (.025 mL) to 5 mL of PBS and
mix well. Do not store diluted antisera.
2. Add equal volumes (e.g., 0.5 mL) of the broth culture to be tested to two small test tubes.
Add the same volume of the killed cell positive control suspension (of the same strain used to
initiate the broth production and known to react with the antisera being used) to a third tube.
3. Add an equal volume ( e.g., 0.5 mL) of diluted antiserum to one tube of broth (mark for
identification) and 0.5 mL of the diluted antiserum to the positive control cells (mark for
identification). Add 0.5 mL of PBS to the second tube of broth cells. Rock each tube gently to
mix contents and set aside (incubation at 37° - 50° C will speed the agglutination reaction). If
carrying out the reaction in round-bottom microtiter plates simply reduce volumes
appropriately (e.g, 100 ìL broth cells plus 100 ìL diluted antisera) and treat each well as if it
were a tube.
4. After 30 minutes examine the tubes. A strong agglutination reaction will often be noticeable
with the unaided eye at this point, with an unagglutinated tube remaining uniformly turbid and
an agglutinating tube showing tiny particulate formation and a clearing of the turbidity.
Reexamine the tubes at 1, 2, and 3 hours.
5. To check for agglutination under the microscope, mix very gently the contents of each tube,
and place a small drop of each on labeled microscope slides. Gently apply cover slips and
examine the slides under phase contrast using both low and high power dry objectives. If
examining the slides after only 10 minutes and finding no agglutination, make another set of
slides after 30 - 90 minutes and reexamine.
6. Test tube and microtiter plate agglutination reactions are clearest if examined after overnight
refrigeration.
Interpretation of agglutination results
The positive control should show agglutination, the negative control should not. Results other
than these indicate internal problems in the test procedure and the test sample result, whether
61
positive or negative, is not meaningful.
A good agglutination reaction will be unmistakable in comparison with the negative control. In
tubes, a positive reaction is seen as a distinct clearing of the turbidity relative to the negative control
and a deposit of clumped cells can usually be seen at the bottom of the tube. On microscope
slides, the positive control slide should show large agglutinated clumps of thousands of cells and the
negative control of broth cells not receiving antiserum should show unclumped, individually
dispersed cells. The broth sample test slide, if positive, should show large clumps of thousands of
cells. The broth sample test slide, if negative, should appear similar to the negative control slide
(which received PBS in place of antisera) and show individual well-dispersed cells. In 96-well
microtiter plates (following overnight refrigeration), a positive reaction is seen as a film settled over
the bottom of the well. A negative reaction appears as a compact dot of cells in the center of the
bottom of the U shaped well.
For all forms of the agglutination test, if both controls are as they should be, and the test sample
shows agglutination, then the broth cells are likely to be the desired rhizobia. If controls are as they
should be and the test sample does not show agglutination, the broth cells are not likely to be the
desired rhizobia and the broth must be discarded or await results of more definitive testing. If the
positive control slide shows no reaction, something has gone wrong with the test (wrong antisera,
cells, etc.). If the negative control broth cells have agglutinated (auto-agglutination sometimes
occurs) no result can be drawn from the test sample.
IMMUNO-SPOT BLOT TEST FOR RHIZOBIAL IDENTITY
P u r p o s e - Spot blot is an immunological technique (membrane ELISA) used to confirm the
identity of rhizobial cells from starter or fermentor broth.
A d v a n t a g e s - Spot blot is not complex and requires no specialized equipment. The test can be
performed with non-agglutinating antisera which have been cross-adsorbed to increase specificity.
L i m i t a t i o n s - Like all immunological techniques for identifying rhizobia, the results of the spot
blot test are only as good as the specificity of the anti-rhizobial antisera used. Also, the test does not
demonstrate cell viability.
Some high-quality antisera are not effective at agglutinating homologous cells. This is often the
case with antisera which have been cross-adsorbed to make them more specific. A quick
membrane ELISA test has been developed by NifTAL which allows the use of such antisera to
establish rhizobial identity in starter and fermentor broth cultures. The test is more refined and
sensitive than the agglutination reaction, and can be used with antisera which are much more strain
specific. The spot blot assay therefore provides more certainty about rhizobial identity in broth
than does agglutination using raw antisera.
Spot blot involves non-specific attachment of rhizobial broth cells to a nitrocellulose membrane,
reaction of the cells with antisera, and detection of bound anti-rhizobial antibodies with a second
62
antibody (commercially available) to which an enzyme (alkaline phosphatase) has been attached.
When exposed to a specific enzyme substrate, the bound enzyme acts to produce a purple spot on
the white membrane. This spot constitutes a positive reaction and indicates that the broth cells
reacted with the primary antisera and are therefore rhizobia. Unfortunately, the enzyme which
empirically works best for membrane assays (alkaline phosphatase) is also widely produced by
microorganisms. Therefore, careful controls are needed within the assay procedure to be sure that
positive reactions are due to the antibody and antibody-enzyme sequence of bindings and not due to
enzyme produced by the broth cells directly.
The spot blot is fundamentally similar to the immunoblot colony lift procedure described
elsewhere in this manual and the comments made about the immunoblot procedure generally
apply as well to the spot blot. The basic differences between the two procedures are 1) the origin of
the rhizobial cells attached to the membrane and 2) spot blot does not include an acid wash step to
denature alkaline phosphatase enzyme which may have been produced by the broth cells, but
instead includes a separate control membrane to determine if the broth cells are producing enough
alkaline phosphatase to interfere with the test. The spot blot can be used to examine rhizobial
broth directly and without delay whereas the immunoblot colony lift is designed to identify colonies
from a spread plate count thus requiring the 3 - 5 days necessary for the colonies to grow.
Conventional glass petri dishes make convenient incubation and wash containers. All solutions
must at least thoroughly cover membranes during all incubations. Wash volumes should be
maximized rather than minimized. Never touch membranes with the fingers. Clean forceps
between handling of membranes in different solutions. Keep the membrane antigens (test samples
and controls) face up during incubations and washes and when blotting (from beneath the
membrane) excess solution with paper towel. Do not rub antigen off the membrane. Do not allow
membranes to float to the surface of solutions during long incubations and thereby expose surfaces
to air drying. All reagent solutions should be prepared during the previous incubation step membranes should not be allowed to dry or even partly dry while a reagent is being prepared. Use
of a gyrotory shaker providing a minimal and gentle agitation (e.g. 15 - 30 rpm) during all wash and
incubation steps is desirable, but not essential. All steps can be carried out at room temperature.
Since antibody and enzyme reactivities are temperature dependant, refrigerated buffers used for
antibody dilution or enzyme substrate preparation should have appropriate volumes removed and
allowed to warm before use.
Materials needed for the immuno-spot blot test:
* Nitrocellulose membranes, 4 cm x 6 cm, 0.45 micron pore size.
* Washed, steam-killed cells of Rhizobium for positive and negative controls, in PBS, at
approximately 109 cells per mL. Cells of the rhizobial strain to which the primary antibody was
made should be used as a positive control (to verify that the assay is working). Cells of another
rhizobial strain or species to which the primary antibody does not react should be used as a
negative control (area of spotting should not turn purple). Note that unadsorbed antisera will often
cross-react between strains within a rhizobial species and sometimes even between strains of
different biovars (such as exist within Rhizobium leguminosarum).
63
* PBS (phosphate-buffered saline):
Per liter:
NaH2P04.H20
0.46 g
Na2HP04.7H20 1.76 g
NaCl
8.5g
purified water 1.0 liter
For each assay, prepare adequate PBS to also prepare PBST for wash steps.
* PBST (0.01M PBS, pH 7.2, plus 0.05% Tween 20): add 0.5 mL of Tween 20 per 1000 mL
PBS. Mix well.
* Blocking Solution (2% skim milk powder in PBS): dissolve 2 grams of skim milk powder in
100 mL of PBS (no Tween 20). Each assay requires 25 mL. Prepare fresh for each use.
* Primary Antibody Solution: Experimentally determine the titre of antibody which works best.
Select the most dilute antisera preparation which yields good color development in 30 minutes. A
dilution of antisera in the range of 1:200 - 1:2000 is expected. Dilute the antibody in PBST, e.g.,
25 ìL raw antisera in 25 mL PBST = 1:1000 dilution. Prepare dilute antibody fresh for each
assay.
* Secondary Enzyme-Labelled Antibody: commercially obtained sheep anti-rabbit Ig-alkaline
phosphatase (or goat anti-rabbit Ig-alkaline phosphatase). Experimentally determine and use the
most dilute antibody-enzyme dilution which will yield good color development in 30 minutes
(preparations vary between suppliers). A dilution of 1:200 - 1:2000 is expected. Dilute the enzymelabeled antibody in PBST.
* Substrate carrier buffer: carbonate buffer (0.1M NaHCO3, 1.0 mM MgCl2, pH 9.8). Add 8.4 g
NaHC03 and 0.20 g MgCl2 to 900 mL of distilled water. Adjust pH to 9.8 with sodium hydroxide
and bring volume to 1 liter. Store in refrigerator, but recheck pH with the buffer at room
temperature for each test. The pH of this buffer is important to this assay because alkaline
phosphatase has maximum activity at this pH.
* Enzyme substrate stock solutions - Prepare NBT and BCIP substrate components as stock
solutions and store cold. Note that concentrated dimethylformamide will attack plastic and that
only glass containers, pipettes, teflon lined screw caps, etc., can be used. Prepare and store NBT
and BCIP stocks separately. Larger volumes than described may be prepared.
NBT (Nitroblue tetrazolium) stock solution:
30.0 mg NBT
0.7 mL dimethylformamide
64
0.3 mL distilled water
BCIP (5-bromo-4-chloro-3-indolyl phosphate) stock solution:
15.0 mg BCIP
1.0 mL dimethylformamide
Add 0.25 mL each of the NBT and BCIP stock solutions to 25 mL of room temperature
carbonate buffer (recheck buffer pH) immediately prior to the color development step in the assay
procedure. The NBT and BCIP concentrated stock solutions may be prepared and stored in the
freezer or refrigerator for up to 1 month. The diluted, mixed NBT and BCIP color development
enzyme substrate solution should be prepared immediately before use and cannot be stored. Each
assay requires 25 mL of color development solution per membrane.
Procedure for the immuno-spot blot assay
1. Lightly mark two small rectangular membranes ("S" and "EC") with an indelible ink in the right
hand upper corners in order to maintain orientation throughout the procedure as to which side of
the membrane the antigens have been applied and to the positioning of the antigens on the
membrane. "S" = broth test sample membrane, and "EC" = enzyme control membrane. It may
also be useful to place small ink marks on the membrane to identify and locate where samples and
control antigens are applied.
2. Wet the two marked membranes by immersion into 25 mL PBS in a petri dish. Use forceps to
slip the membranes slowly into the PBS at a 45° angle, allowing the membrane pores to fill without
air pockets. Allow the membranes to soak for 5 minutes then remove from the PBS and drain off
excess liquid. Place the membranes on an adsorbent paper for 5 minutes to pull pore liquid from
the membrane.
3. Apply antigen. Adjust broth cultures to approximately 109 cells per mL by adding PBS. The
suspension should be distinctly turbid. Make two 10-fold dilutions of the 109 cell per mL sample to
approximately 108 and 107 cells per mL. Apply 1 ìL of each of the three dilutions as separate
spots, in order of dilution, onto each of the "S" and "EC" membranes. Apply similar concentrations
of positive and negative cells in rows positioned under the broth sample spots. Use fresh dry
pipette tips or cleaned, dry loops to make all transfers.
4. Immerse the enzyme control "EC" membrane in 25 mL of PBST containing no antisera. Leave
the "EC" membrane in this dish of PBST with no further processing until the enzyme substrate
color development step at the end of the procedure. Do not allow the membrane to float and dry
portions of its surface.
5. Block the "S" membrane by immersion in 25 mL of milk protein blocking solution. and incubate
for 45 minutes. Remove membrane and drain briefly.
6. Wash "S" membrane. Immerse membrane in PBST and provide gentle agitation for 5 minutes.
65
Remove, drain, and repeat the wash twice using fresh PBST. Remove membrane and drain
briefly.
7. Immerse the "S" membrane into 25 mL of the pre-determined dilution of the appropriate antirhizobial antisera in PBST and incubate for 90 min. Remove and briefly drain the "S" membrane.
Do not allow antibody containing liquid from the "S" membrane to in any way come in contact with
the "EC" membrane.
8. Wash the "S" membrane. Wash the membrane in 25 mL PBST, with gentle agitation for 5
minutes. Repeat this wash step, using fresh PBST, twice. Remove and briefly drain the "S"
membrane.
9. Incubate the "S" membrane with 25 mL of enzyme-labelled secondary antibody. Immerse the
membrane into the pre-determined dilution of alkaline phosphatase enzyme-labelled secondary
antibody (in PBST) and incubate for 60 minutes. Do not allow the enzyme-labelled secondary
antibody to come in contact with the "EC" membrane in any way.
10. Wash "S" membrane. Immerse the membrane in PBST and provide gentle agitation for 5
minutes. Repeat the wash twice using fresh PBST. Remove and briefly drain the "S" membrane.
11. Remove the "EC" membrane from its PBST dish and drain briefly.
12. Color development. Immerse the "EC" membrane, antigen side down, in 25 mL of freshly
prepared enzyme substrate (NBT + BCIP) solution. Immerse the "S" membrane in the same
enzyme substrate with the antigen side up. Color development should begin within minutes and be
well developed within 30 minutes. Remove and wash the membranes in water if the substrate
begins to auto-precipitate.
13. Stop color development by rinsing the membranes twice in purified water.
14. Dry the membranes and store in the dark.
Interpretation of spot blot results
Negative versus positive spots - this test is interpreted by whether or not colored spots develop
on a white membrane as determined by the human eye. Recognizing the subjectivity involved, we
suggest the following as guidelines. A very strong positive result will be an intense, almost black,
purple. A strong positive will be a distinct medium-dark purple. A positive reaction will be a
distinct light purple. A weak reaction will be a faint purple, but clearly seen. Intensities less than
these can be called negative. Do not ascribe positive results to very faintly colored spots. The test
technique should be "tweaked" by adjusting antibody concentrations or incubation times so that the
positive control cells fit, at a minimum, into the "strong positive" category. It is expected that
increasingly dilute antigen applied to the membranes will show decreasing degrees of purple color
intensity.
66
The "EC" enzyme control membrane. All antigen spots should be negative, i.e., they should
show no color. If positive or negative control spots on the "EC" show color, this indicates the
presence of significant alkaline phosphatase in the respective cell suspension and makes that
suspension worthless for this test. Their functions as controls on the "S" test sample membrane are
negated. If the test sample spots on the "EC" membrane show color, this indicates the presence of
significant alkaline phosphatase in the broth test sample suspension and negates the whole test.
Results should not then be drawn from the "S" membrane.
The "S" test sample membrane. The positive control cells should be positive. If they are not,
something has gone wrong with the test (incorrect antisera used, incorrect control cells used,
secondary antibody-enzyme omitted, faulty substrate, etc.). No results should be drawn from the
test. The negative control cells should be negative (show no color). If the negative control cells are
positive, something has gone wrong with the test and no results should be drawn from the test. If
the "S" membrane positive controls are positive and the "S" membrane negative controls are
negative, AND the "EC" membrane is negative, then interpretation of the test antigen result on the
"S" membrane is valid. Under these conditions, positive test broth sample spots indicate reactivity
with the primary antisera and indicate that the broth cells contain the correct rhizobia. A negative
result for the test broth sample spots then indicates that the broth does not contain significant levels
of the correct rhizobia.
INDIRECT
FLUORESCENT
RHIZOBIA IN BROTH
ANTIBODY
IDENTIFICATION
OF
P u r p o s e - To confirm the identity of rhizobia in broth cultures.
A d v a n t a g e s - The test is rapid and simple.
L i m i t a t i o n s - The test does not enumerate the rhizobia and cannot evaluate whether or not they
were alive at the time of sampling. The test requires fluorescent microscopy equipment.
This technique uses the researcher's primary antisera to bind antibodies to target rhizobia
which have been bound (non-specifically) to a microscope slide. The primary antibodies are then
reacted with a commercially obtained second antibody which has been covalently bound to a
fluorochrome. The fluorochrome, (which will only be retained through washing steps if the
sequence of antibody bindings, and hence the target rhizobia, are present) will glow brightly if
exposed to the correct wavelength of light and viewed through a properly equipped microscope (see
Fig.6). The fluorochrome used in this procedure is fluorescein isothiocyanate (FITC) which
glows green when exposed to near ultra-violet light. The net effect for a positive antibody reaction
is that the rhizobial cells glow with a green fluorescence when seen through the microscope.
67
Figure 6.
technique.
Schematic
diagram
of
69
the
indirect
immunofluorescence
This procedure begins with a wash of the rhizobial broth cells in water. The wash facilitates
binding of the cells to the glass slide, a step which is crucial and would otherwise be interfered with
by the formation of salt crystals as the smear dries. The procedure calls for the use of specialized
microscope slides (Cel-Line Associates). These slides are covered with a hydrophobic surface with
two rows of six circular openings (approximately 5 mm diameter) through the hydrophobic surface
to the glass. The effect of the hydrophobic layer is that a drop of aqueous solution placed on the
glass opening will "stand up" and not spread. Twelve different specimens can be individually
stained (even with different antisera) on a single slide. This feature greatly facilitates effective use of
the fluorescent antibody (FA) technique. The indirect fluorescence technique can be performed
using ordinary microscope slides, but it is well worth the extra effort to obtain the hydrophobic
surfaced slides.
As always with immunological procedures, it is necessary to predetermine appropriate dilutions
of the antibodies to be used. FA normally requires considerably more concentrated antibody
solutions (1:25 - 1:200 dilution) than does the enzyme-linked immunosorbent assay (ELISA).
Routine and appropriate controls are also, as always, essential. It is important to keep in mind that
the technique described determines nothing about whether or not the cells were alive when
removed from the broth. Antibodies react with dead cells as well as with live ones.
Procedure for FA staining
1. Place 0.5 mL of early stationary phase rhizobial broth into a 1.5 mL Eppendorf centrifuge tube,
add 0.75 mL purified water, mix, and pellet the cells in a high speed centrifuge.
2. Observing the pellet to be sure that it is not sucked up, remove the supernatant liquid, add 1.0
mL water, and vortex until the pellet is evenly suspended.
3. Take small (approximately 10 ìL) equal volumes of the washed broth cells (the amount
depending on cell concentration) and smear over two adjacent glass wells in the surface of the
hydrophobic layer on the printed slide. Label the wells.
4. Using stock positive control cells known to be reactive with the primary antisera and stock
negative control cells known not to be reactive with the antisera, similarly prepare one well of each
on the slide. Label the wells.
5. Allow the slide to evaporate to dryness and then lightly heat fix the cells to the glass by
momentary passage (cell side up) through a flame.
6. Apply a small amount of rabbit primary antisera, diluted to the predetermined level in PBST, as
a small drop to both the positive and negative stock cell wells and to one of the two broth cell wells.
7. Apply a similar drop of the diluent used for the primary antisera, but containing no antisera, to
the second broth cell containing well. Label the well.
8. Incubate the slide for 60 minutes in a small moist chamber (such as a petri dish with lid
70
containing a wet piece of filter paper or cotton).
9. Wash the slide 4 times, for 60 seconds each time, in PBST. Slide staining jars with vertical
grooved edges to receive the slide are ideal. Pass the slide from jar to jar.
10. Quickly remove excess liquid from the hydrophobic layer of the slide with cotton swabs. Do
not touch the cells in the wells (leave them moist).
11. Apply a small amount of the secondary FITC labelled anti-rabbit Ig antibody diluted to the
predetermined level in PBST (and preferably containing a blocking agent such as 0.1 % bovine
serum albumin or fetal calf sera) as a drop to ALL FOUR of the cell containing wells.
12. Incubate in the moist chamber for 45 minutes.
13. Wash the slide and remove excess liquid as after the primary antibody incubation.
14. Dip the slide briefly in distilled water, shake dry, and add a very small drop of buffered glycerol
(1 part PBS : 9 parts glycerine, ph adjusted to 9.3) to each cell containing well.
15. Gently apply a clean coverslip to the buffered glycerol drops and observe using an epifluorescence microscope using the proper filter sets for FITC staining.
Interpretation of FA staining results.
It is useful to adjust the microscope, even when using it for epi-fluorescence work, so that the
cells can be seen under phase contrast. By switching back and forth between phase contrast
illumination and fluorescence illumination it is possible to see cells that are not fluorescing and
which would be invisible with fluorescence illumination alone.
The positive control stock cells well should show cells fluorescing greenly. If not, something has
gone wrong with the test (wrong antisera or cells, etc.).
The negative control stock cells well should not show fluorescing cells. If it does, something has
gone wrong with the test (wrong antisera or cells, non-specific binding of FITC labelled secondary
antibody, etc.).
The negative control (no primary antisera) broth cells well should have no fluorescing cells. If
it does, something has gone wrong with the test. In conjunction with the results of the positive and
negative stock cell well results it may prove possible to identify exactly what has gone wrong with
the test. In any case, positively fluorescing broth cells which were unexposed to the primary
antisera mean that fluorescence observed in the test sample is not related to antibody specificity.
Therefore no result should be drawn from the broth sample test well.
The broth sample test well cells will either show positive or negative fluorescence. If all of the
three controls have given the expected results, then the test can be interpreted, correctly, as either
71
positive or negative on the basis of the result. Using phase contrast and visible light on the broth
sample test well, it should be possible to see if all or only a portion of the broth cells are
fluorescing. If only a portion are fluorescing, it is possible that the non-fluorescing cells are
contaminant organisms which did not react with the primary antisera.
A 40 power dry objective will give good results and is easy to use, but an oil objective using nonfluorescent microscope oil will provide much brighter fluorescence. If examining many samples,
however, it is easier to stay with the dry objective.
COLONY -LIFT
IMMUNOBLOT
INOCULANT ANALYSIS
(MEMBRANE
ELISA)
FOR
P u r p o s e - Colony-lift immunoblot analysis is an immunological technique (a form of enzymelinked immunosorbent assay) used to confirm the identity of rhizobial colonies formed by standard
spread plate count techniques.
A d v a n t a g e s - The colony-lift immunoblot simultaneously, but individually, determines whether
each colony from the surface of a plate count dish and imprinted onto the membrane reacts with a
primary antibody. Depending upon the specificity of the primary antibodies used, the technique
can detect and identify rhizobial strains. The technique is not difficult and does not require
specialized equipment.
L i m i t a t i o n s - The colony immunoblot itself takes only a few hours to perform, but it is
dependant on first having grown colonies from a plate count procedure which normally takes 2-3
days for fast growing rhizobia and 5-6 days for slow growing rhizobia. Because most contaminating
organisms grow much faster than do rhizobia, the immunoblot procedure is not of use where a
significant number of contaminants exist. If the contaminating organisms form small discreet
colonies and do not overwhelm the rhizobia present, the technique can still be used successfully.
This technique is similar in principle to the immuno-spot blot membrane ELISA described
earlier for the analysis of rhizobial broth quality. The method uses the researcher's primary
antisera to bind antibodies to target rhizobia which have been bound (non-specifically) to a
membrane. The primary antibodies are then reacted with a commercially obtained secondary
antibody which has an enzyme label (alkaline phosphatase) covalently attached. The bound
enzyme (which will only be retained through washing steps if the sequence of antibody bindings,
and hence the target rhizobia, are present) converts a chemical substrate from colorless to purple
and produces a purple precipitate at the site of the reaction. The result is purple spots on a white
membrane where the target rhizobial colonies touched the membrane. The purple spots can be
readily seen and counted with the naked eye.
The colony-lift immunoblot involves imprinting onto a membrane the pattern of colonies
formed on a petri dish media surface by standard plate count methods. A circular nitrocellulose
membrane is physically applied to the media surface picking up a portion of all colonies present in
72
a mirror image pattern of the colonies on the growth media. Identification of the colony imprints is
then made using the antibody and enzyme sequence described above. When related back to the
dilution from which the plate was made, the number of viable rhizobia which must have been
present in the original sample can be determined (see Fig. 7).
Figure 7. Rhizobial colonies on YEMA media (a and c) and corresponding
immunoblot membranes (b and d).
73
Materials needed for the colony lift immunoblot test:
* Round nitrocellulose membranes, 82.5 mm diameter (to fit standard-size petri plates), 0.45
micron pore size.
* Washed, steam-killed cells of Rhizobium for positive and negative controls, in PBS, at
approximately 109 cells per mL. Cells of the rhizobial strain to which the primary antibody used
was made will be used as a positive control (to verify that the assay is working). Cells of another
rhizobial strain or species to which the primary antibody will not bind are used as a negative
control (area of spotting should not turn purple). Note that unadsorbed antisera will often crossreact between strains and even biovars such as within the Rhizobium leguminosarum) group.
* Spread plates of Rhizobium prepared from a serial dilution of peat inoculants, selecting a
dilution level which results in 30 to 300 colonies per plate in which the colonies can be easily
counted and distinguished.
* PBS: (Phosphate-buffered saline, 0.01M, pH 7.2):
For 1 liter:
NaH2P04.H20
Na2HP04.7H20
NaCl
purified water
0.459 g
1.764 g
8.5g
1.0 liter
For each membrane assay, prepare 3 liters of PBS which will also be used to make PBST and
acidified PBS.
* Acidified PBS, pH 2.6: adjust pH of PBS down to pH 2.6 with 6N HCl (hydrochloric acid).
Each assay requires 25 mL per membrane.
* Blocking Solution (2% skim milk powder in PBS):
Dissolve 2 grams of skim milk powder in 100 mL of PBS (no Tween 20). Each assay requires 25
mL per membrane. Prepare fresh each time.
* PBST (0.01M PBS, pH 7.2, plus 0.05% Tween 20): add 0.5 mL of Tween 20 per 1000 mL
PBS.
* Primary Antibody Solution: experimentally determine the titre of primary antisera which works
best by selecting the most dilute antisera preparation which gives good color development in 30
minutes. A dilution of antisera in the range of 1:200 - 1:2000 is expected. Dilute the antibody in
PBST, e.g., 25 ìL raw antisera in 25 mL PBST = 1:1000 dilution. Each assay requires 25 mL of
diluted antibody per membrane. Prepare diluted antibody fresh for each assay.
* Secondary Enzyme-Labeled Antibody: sheep anti-rabbit Ig-alkaline phosphatase or goat antirabbit Ig-alkaline phosphatase conjugate (assuming that the primary antisera was raised in rabbit).
74
Experimentally determine the best antibody-enzyme dilution (preparations vary between suppliers).
A dilution of 1:200 - 1:2000 is expected. Dilute the enzyme-labeled antibody in PBST.
* Substrate carrier buffer: carbonate buffer (0.1M NaHCO3, 1.0mM MgCl2, pH 9.8) - Add 8.4 g
NaHC03 and 0.20 g MgCl2 to 900 mL of distilled water. Adjust pH to 9.8 with 1N sodium
hydroxide and bring volume up to 1 liter. Store in refrigerator, but recheck room temperature
buffer for each assay. The pH of the buffer is important to this assay because alkaline phosphatase
works best at this pH.
* Enzyme substrate stock solutions - Prepare the NBT and BCIP substrate components as stock
solutions and store cold. Note that concentrated dimethylformamide will attack plastic and that
only glass containers, pipettes, teflon lined screw caps, etc., can be used. Prepare and store NBT
and BCIP stocks separately. Larger volumes may be prepared.
NBT (Nitroblue tetrazolium) stock solution:
30.0 mg NBT
0.7 mL dimethylformamide
0.3 mL distilled water
BCIP (5-bromo-4-chloro-3-indolyl phosphate) stock solution:
15.0 mg BCIP
1.0 mL dimethylformamide
Add 0.25 mL each of the NBT and BCIP stock solutions to 25 mL of room temperature
carbonate buffer immediately prior to the color development step in the assay procedure. The
NBT and BCIP concentrated stock solutions may be prepared and stored in the freezer or
refrigerator for up to 1 month. The diluted color development solution should be prepared
immediately before use and cannot be stored. Each assay requires 25 mL of color development
solution per membrane.
Procedure for the colony lift immunoblot assay:
1. Select a plate with discreet colonies of typical rhizobial appearance from an appropriate
dilution of the spread plate series. A plate with 30 -50 colonies is better suited than plates with a
large number of colonies. The colony lift works best when the colonies are small (less than 1 mm
dia.) as the process of overlaying the membrane will squash the colonies, resulting in a larger
imprint on the membrane. Do not use plates that have been refrigerated. If possible, photograph
the plate using a Polaroid camera - this allows subsequent comparison of the colonies identified as
rhizobia on the membrane with the colonies as they appeared before being lifted by the membrane.
2. Place a small mark on a nitrocellulose membrane with indelible ink and similarly mark the
petri dish.
3. Colony lift - Do not touch membranes with fingers. Wet the membrane by immersion from
one edge to the other in PBST in a petri dish so that air bubbles do not become entrapped within
75
the membrane. Gently blot the membrane to surface dryness on a paper towel or Kimwipe. Using
forceps to hold the membrane, line up the mark on the membrane and the mark on the dish for
later orientation. The mark on the membrane should be down so that it touches the media
adjacent to the mark on the petri dish. Let one rim of the membrane touch the media surface next
to the far edge of the petri dish first by dragging the membrane across the rim of the dish until its
far edge falls and makes contact with the media. Slowly allow the rest of the membrane to come
down so that air is not trapped under the membrane. Do not try to drag or shift the membrane
once any of it has touched the surface. Press very lightly down on the membrane with a clean, flat
object so that the membrane touches all colonies on the media surface.
4. Peel off the membrane (using forceps) and rinse it under a running stream of water. Use
forceps to hold the membrane so that it is completely rinsed of attached colonies and all sign of
colony debris visible to the naked eye is washed away. A single layer of cells, invisible to the naked
eye, remains attached to the membrane.
5. Air dry the membrane on a paper towel or kimwipe, colony side up. Do not blot or rub the
colony imprint surface.
6. Draw small circles (3 to 4 mm in diameter) on the outside edges of the membrane to indicate
the location of positive and negative control cell spottings. Some membranes have an outer rim of
non-porous nitrocellulose. Put the control spots just inside of this rim. Label the circles "p" and "n."
Using a 1-ìL inoculating loop or, preferably, a micropipette, apply 1 ìL each of the positive and
negative control cell suspensions to the inside of the respective circles. Allow the membrane to air
dry, colony side up. A blower or laminar hood will speed this process. The membrane needs to
be dry all the way through so that it will soak up the acid solution in the next step. Thoroughly
dried membranes can be stored at this point for at least several days.
7. Soak the membrane for 20 minutes in acidified PBS (pH 2.6) in a petri dish. Insert the
membrane edgewise into the acid so that its pores fill evenly and air bubbles are not entrapped.
The purpose of this acid wash is to denature endogenous alkaline phosphatase which may have
been endogenously present in the colonies. Some microorganisms produce significant amounts of
alkaline phosphatase. Alkaline phosphatase bound to the membrane which comes from any
source other than the alkaline phosphatase-labeled secondary antibodies used in the procedure can
yield false positive results.
8. Remove the membrane from the acidified PBS and allow it to drain for a few seconds, then
soak it for 5 minutes in PBS, pH 7.2, exchanging the PBS at least twice.
9. Blot the membrane lightly, from the bottom, with tissue or filter paper. Do not touch the
colony imprint side of the membrane.
10. Soak the membrane for 30 minutes in 25 mL of the milk protein blocking solution.
Incubating at 37° C will improve the blocking action. Blocking is intended to tie up protein binding
sites on the nitrocellulose so that non-specific attachment of antibodies and associated alkaline
phosphatase is minimized.
76
11. Wash the membrane (with gentle agitation) in PBST three times (5 minutes each) using
fresh PBST for each wash.
12. Incubate the membrane with 25 mL of the dilute primary antibody solution in PBST in a
petri dish for 1 hour at room temperature.
13. Wash the membrane in PBST three times (5 min. each) using fresh PBST for each wash.
14. Incubate the membrane in 25 mL of the secondary antibody-enzyme label solution in a glass
petri dish for 1 hour at room temperature.
15. Wash the membrane as in step 13 above.
16. Incubate the membrane in 25 mL of freshly prepared NBT/BCIP substrate in a glass petri
dish. Color development should be complete after 30 minutes, during which time purple spots
corresponding to rhizobial colonies should have appeared. The enzyme substrate solution will often
begin to degrade and auto-precipitate between 30 minutes and 60 minutes after mixing the NBT
and BCIP. Remove the membrane and stop color development if this starts to occur.
17. Stop color development by rinsing the membrane twice in distilled water.
18. Dry the membrane and record the number of positive reactions. The number of primary
antibody reactive rhizobia in the original sample is calculated exactly as if the colored spots were
colonies on a spread plate.
19. The membrane may be stored in the dark (plastic pocket photograph albums work well) for
a permanent record.
20. It is necessary to verify that endogenous alkaline phosphatase which may be present in
colonies prior to being lifted by a membrane is not the cause of purple spots formed. The best way
to control for this is to run a control membrane identical to the test membrane, but incubating in
PBST alone until the color development step. The membrane used for this control should be
colony imprinted from a replicate plate from the same dilution as the plate chosen for the actual
assay membrane. This control membrane should show no spots of purple color development
(including the positive control spot). This membrane corresponds with the "EC" membrane in the
immuno-spot blot procedure.
Interpretation of colony lift immunoblot results
Calculation of the number of viable of rhizobia present in the original inoculant is obtained by
counting the number of purple spots and multiplying this number by the reciprocal of the sample
dilution x the factor for amount of sample applied per spread plate. For example, if 60 purple
spots resulted from 100 ìL of inoculant suspension applied and spread from the 10-6 dilution, the
calculation would be: 106 x 10 x 60 = 6 x 108 rhizobia per g inoculant. ELISA colony immunoblot
77
results should give comparable numbers with MPN tests run on the same sample. Colony
immunoblot 95% confidence limits can be determined from tables of confidence limits of a single
count, or, a very close approximation can be calculated as plus and minus 1.96 times the square
root of the colony count (Wardlaw, 1985).
DIRECT FLUORESCENT ANTIBODY ENUMERATION OF RHIZOBIA
IN INOCULANT
P u r p o s e - Immunofluorescent enumeration of rhizobia extracted from peat inoculants provides
a rapid means to identify and estimate the number of rhizobia present.
A d v a n t a g e s - The method is relatively quick and simultaneously allows enumeration and
identification of rhizobia.
D i s a d v a n t a g e s - The method does not distinguish between live rhizobia and dead rhizobia. The
method requires conjugation and purification of antibody with FITC fluorochrome. The method
requires fluorescent microscopy equipment.
Published procedures of this technique (Bohlool, 1987; Bottomley, 1994; Somasegaran and
Hoben, 1994) use the researcher's primary antibodies which are labelled in the researcher's
laboratory with the fluorochrome fluorescein isothiocyanate (FITC). The target rhizobia are
collected by filtration onto a membrane and reacted with the antibody-fluorochrome conjugate
while bound to the membrane. When excited with near ultraviolet light, the fluorochrome (which
will be retained through washing steps only if the sequence of antibody bindings and hence the
target rhizobia are present) will fluoresce with a green glow against the dark background of the
membrane (see Fig. 8). The net effect is that the rhizobia themselves can be seen glowing with the
green color when viewed through a fluorescence microscope. While this procedure calls for the
use of primary antibodies directly labelled with FITC, there should be no problem adapting the
procedure for use with commercial secondary antibodies already labelled with FITC. The
procedure is useful for counting populations in excess of about 106 cells per g of peat inoculant.
The bacterial filter used to trap rhizobia from an inoculant suspension will quickly clog with
particulate matter, so it is necessary to remove as much peat from the inoculant suspension as is
possible before beginning the filtration process. Different approaches have been effective.
Bottomley (1994) has used prefiltration of the sample with membranes of larger pore size (8 ìm).
Another approach is to place inoculant suspension at a 1:10 dilution in a small Eppendorf
centrifuge tube and give it a momentary (approx 3 seconds) spin in a small high speed centrifuge.
This will separate out and pack as sediment the vast proportion of peat, but leave almost all
bacteria behind. The bacteria containing supernatant liquid can then be used to prepare a 10-4
dilution for filtration which will contain very little particulate.
78
Figure 8. Schematic diagram of the di rect immunofluorescence technique.
Procedure for the direct fluorescent antibody enumeration of rhizobia in
inoculant
1. Prepare a 1:10 suspension of inoculant in phosphate -peptone diluent by adding 11 g of
inoculant to 99 mL diluent and mixing well for 15 minutes.
2. Remove the bulk of peat from the suspension by prefiltration or centrifugation.
Further dilute the partly purified 1:10 suspension to 10-4 dilution.
3. Pipette 5 mL of the 10-4 dilution through a vacuum filter apparatus (see Fig. 8) through a 25 mm
diameter, 0.2 ìM polycarbonate, black surface, membrane (Nucleopore; Poretics Corporation).
4. Remove the membrane, place it on a microscope slide, and treat it to suppress non-specific
staining with rhodamine gel (see Somasegaran and Hoben, 1994). Heat the slide at 60° C until the
conjugate is barely dry. Remove the slide and let it cool. Alternatively, omit step 4 and block the
membrane with 2% skim milk solution as was done for the colony-lift immunoblot procedure.
5. Stain the surface of the filter with an appropriate dilution of fluorescent antibody conjugate and
incubate in a moist chamber for 30 min.
6. Return the filter to the vacuum unit and destain it by pulling through it at least 50 mL of
particulate free PBS.
79
7. Mount the membrane on a clean microscope slide with a small amount of buffered glycerol (1
part PBS : 9 parts glycerine, pH adjusted to 9.3). Observe under near ultraviolet excitation with an
epi-fluorescence microscope using either the 40 power dry or an oil immersion objective.
8. Count at least 10 fields. An ocular containing outlines of squares of various sizes allows for
some adjustment of the size of field to be counted depending on cell density. The area of the
squares can be calculated for each objective by observation of a stage micrometer through each
objective and measuring the length of each square side when focussed on the micrometer.
Interpretation of FA- membrane method results
The number of rhizobia per g inoculant = N x A/a x D/V, where:
N
A
a
D
V
= mean number fluorescing bacteria per field
= effective filtering area (mm2)
= Area of microscope field of view (or square) (mm2)
= inoculant dilution factor
= volume (mL) of inoculant suspension filtered
The effective filtering area (A) is that portion of the membrane defined by the internal
circumference of the filter funnel. The number of rhizobia per g should be expressed on a wet
weight basis of the inoculant as originally sampled.
80
CHAPTER 7
POTENTIAL TECHNIQUES FOR INOCULANT ANALYSIS
All existing methods of inoculant analysis are less than ideal. It seems likely that this will
remain the case. The MPN plant infection technique is the best method because it directly
enumerates nodulating rhizobia, but 30 days is often too long a period to have to wait for a result.
Plate count methods do not prove that the cells counted are rhizobia, not even when the carrier
material is sterile. The direct immunofluorescence membrane technique is fast, but does not
prove that the rhizobia enumerated were alive at the time of collection. The colony-lift immunoblot
technique enumerates viable rhizobia, but is a cumbersome technique and, in any case, requires at
least 3 days for colonies to develop on plates before it can be conducted.
A variety of new techniques hold promise for the future in terms of the rapid and accurate
analysis of legume inoculant. Significant among these is the development of selective media which
would allow for plate counts to be performed of inoculants in which only the target rhizobia could
form colonies and be counted. Media have been reported which allow for selective recovery of
Rhizobium meliloti (Barber, 1979) and for Bradyrhizobium japonicum and B. elkanii (Tong and
Sadowsky, 1994).
A report (Olsen and Rice, 1996) of a new "quick test" technique for inoculant analysis reported
the enumeration of a minimum number of viable, identified rhizobia in a period of about 90
minutes. The technique demonstrates two aspects of analysis which may be generally useful in the
future. These aspects are 1) the use of immunomagnetic beads to selectively remove rhizobia from
an inoculant and 2) the use of biotin - streptavidin amplification technology in inoculant analysis.
The "quick test" report used agglutinating antibodies to agglutinate rhizobia within a peat inoculant
and the agglutinated rhizobia were then labelled with a secondary commercial antibody which was
itself labelled with biotin. Commercial immunomagnetic micro-beads labelled with streptavidin
were then added. Since streptavidin binds biotin with great strength the agglutinated rhizobia
became bound to the immunomagnetic microbeads and were then selectively pulled from the
inoculant suspension with a magnet. The rhizobia were then stained with a commercial nucleic acid
staining kit containing two fluorescent dyes which differentiate between live and dead cells. Under
an epi-fluorescent microscope, the live cells glow green, the dead cells glow red.
Biotin and streptavidin coupled to a variety of signal producing labels have been commercially
available for several years. These labels include enzymes such as alkaline phosphatase and
fluorochromes such as fluorescein isothiocyanate (FITC). Because the biotin - streptavidin binding
is so very potent and avid, the use of the technology allows exposure of labelled antigen (such as
rhizobia reacted with a primary antibody and secondarily labelled with biotin) to be exposed to the
signal-label bearing streptavidin molecule for relatively short times. The net effect is to reduce
background signal (which is equivalent to increasing sensitivity) and at the same time increase the
specific signal because one streptavidin molecule will bind four biotin molecules. The net result is
stronger positive reactions with minimal background noise.
81
CHAPTER 8
MANUFACTURER PROTOCOLS FOR ANALYSIS
The quality control process should proceed in a pre-planned and defined way for each batch of
inoculant produced. In this section of the manual we summarize some of the techniques which a
small manufacturer could implement for routine quality control during the inoculant production
process as well as with finished inoculant. All of the techniques are described in this manual.
PROTOCOLS
I. Carrier testing
Testing for carrier sterility
Options:
1. Perform the enrichment detection test and look for microbial growth by plate count on
YEMA.
2. Perform the sprinkle test and look for microbial growth on YEMA.
3. Perform the plate count method to quantify contaminant growth.
Testing for carrier -rhizobia compatibility
Perform the carrier-rhizobia compatibility test and look for rhizobial multiplication and survival
over time.
II. Testing of preserved, mother, starter, and fermentor cultures
A. "Mother" cultures are usually prepared on agar slants and are usually re-grown aliquots from
a stock culture derived from a preserved culture. The stock cultures should be obtained from
the preserved culture by mass transfer rather than through selecting a single colony. Mass
transfer reduces the chance of genetic variation between the mother cultures. The "mother"
cultures should also be derived through mass transfer from the stock culture. Starter fermentor
cultures are prepared from a mother culture. With this approach, tests for contamination and
plant infection tests to ensure that nodulability is maintained are carried out on the stock culture
and or a sampling of mother cultures.
It is also possible to bypass the stock culture step and prepare a season's "mother" cultures
directly from a single preserved culture. This procedure works particularly well with
cryopreserved cultures. With this approach, two or more of the "mother" cultures are used for
tests for contamination and effectiveness on plants.
82
A third alternative is to prepare a large number of aliquots of cryopreserved cultures for each
production strain, and to begin each starter broth for fermentation from a preserved aliquot.
Two or more samples of the preserved aliquots should have already been carefully examined
for contamination and nodulability at the time the aliquots were prepared. With this approach,
each preserved aliquot is, in a sense, both a stock and a mother culture. This approach
provides minimum chance for contamination in the preparation of the starter culture. Testing
for contamination and nodulability are conducted directly on the starter culture.
Fermentor broth cultures for the production of inoculant should be examined for
contamination and rhizobial identity before being combined with the carrier. It is normally
possible to sample the fermentor broth 24 h before it has reached peak cell numbers and to use
this time to perform "presumptive tests" for contamination and for the presence of the proper
rhizobia. By early testing, it is possible to make the decision to "pass" or "fail" the fermentor
broth without holding the broth in stationary phase while waiting for test results.
The "presumptive tests" for contamination and rhizobial identity in broth
c u l t u r e:
e
1. microscopic examination of the broth which can include a cell count of the broth in a PetroffHausser chamber.
2. the glucose-peptone test for contaminants.
3. the serological agglutination test for rhizobial identity.
4. the immuno-spot blot test for rhizobial identity.
5. the Gram stain test for Gram-positive contaminants.
Testing for contamination in preserved, stock, mother, or starter broth
cultures
1. Plate the culture on YEMA plus Congo red and look for abnormal colony growth
morphology, growth rate, or dye absorption.
2. Streak the culture on peptone-glucose bromcresol purple media and look for rapid growth or
color change.
3. Examine the broth microscopically and look for contaminants or atypical rhizobial
morphology.
4. Perform the Gram stain on broth samples and look for Gram-positive (non-rhizobia)
organisms.
Testing for rhizobial identity in broth culture
83
1. Perform the plant-infection test using a 10-6 dilution of broth and look for nodule formation.
Uninoculated negative controls showing no nodulation are essential.
2. Perform agglutination testing and look for somatic cell agglutination.
3. Perform the immuno-spot blot test and look for purple spot formation.
4. Perform the indirect immunofluorescent antibody test and look for fluorescent green cells of
rhizobial morphology using an epi-fluorescence microscope.
III. Testing of finished legume inoculant in carrier
Sterile carrier inoculants
1. Perform plate counts (spread plate or drop plate) to establish the number of viable cells
present. If the inoculant has been properly made, the viable cell count equals the viable
rhizobia count, but plate count tests do not prove the identity of the rhizobial strain.
2. Perform the colony-lift immunoblot test on a spread plate from the plate count to establish
the viable rhizobial cell count in the inoculant.
3. Perform the direct immunofluorescence membrane test to establish the rhizobial cell count
in the inoculant. This test does not prove that the rhizobia were alive at the time of sampling.
4. Perform the MPN plant-infection test to establish the number of live, infective rhizobia of
proper host homology in the inoculant. This is the best and most conclusive test, but takes 30
days to achieve a result.
Non-sterile carrier inoculants
Perform the MPN plant infection test to establish the number of live, infective rhizobia in the
inoculant. Contamination levels in non-sterile carrier inoculants are normally too high to allow for
meaningful plate count or colony-lift immunoblot results. At the present time there is no
established procedure that works consistently for the analysis of non-sterile carrier inoculants
except for the MPN assay.
Concluding remarks to small inoculant manufacturers
Implementation of a routine, continuing, and effective quality control system is essential to
production of high quality legume inoculant. Each manufacturer should have quality control staff
with the unquestioned authority to reject and prevent the distribution of faulty inoculant. The
quality control function should be separate from, though work closely with, the manufacturing
function. Management must be firmly committed to the quality control process and be scrupulous
about not overriding (and thus undermining) the process.
84
Quality control at the manufacturing level must begin with the rhizobia used to initiate the
inoculant production process. If the "mother" culture is not a rhizobial strain, or contains the
wrong rhizobia, or is contaminated, then an effective legume inoculant product cannot be
produced. If the inoculant production system is based on sterile carrier, then routine checks on the
sterility of every carrier batch are essential. The protocols and techniques outlined above will not
apply to all producers in all situations. Many manufacturers will, for example, find some of the
more complex immunological procedures difficult to implement. These techniques are not
absolutely essential to a good quality control program. The essential elements are: 1) the ability to
begin inoculant production with pure cultures of the correct rhizobia, 2) the abilities to identify and
quantify rhizobia, 3) the ability to keep cultures free of contamination, and 4) a commitment to
excellence in the quality of the inoculant product.
85
REFERENCES
Amarger, N. 1974. Competition pour la formation des nodsites sur la feverole entre souches de
Rhizobium leguminosarum et souches du sol. C. R. Acad. Sci. (Paris) 2 7 9 :527-530.
Amarger, N. and Lobreau, J. P. 1982. Quantitative study of nodulation competitiveness in
Rhizobium strains. Appl. Environ. Microbiol. 4 4 :583-588.
Anonymous. 1992. The methods of testing legume inoculant and pre-inoculated seed products.
Fertilizers Act, Section 23, Regulations, Feed and Fertilizer Division, Government of Canada,
Ottawa.
Barber, L. E. 1979. Use of selective agent for recovery of Rhizobium meliloti from soil. Soil Sci.
Soc. Am. J. 4 3 :1145-1147.
Beck, D. P., Materon, L. A., and Afandi, F. 1993. Practical Rhizobium-legume technology manual.
Technical manual No. 19. ICARDA, Aleppo, Syria.
Berg, R. K. Jr., Loynachan, T. E., Zablotowicz, R. N. and Lieberman, M. T. 1988. Nodule
occupancy by introduced Bradyrhizobium japonicum in Iowa soils. Agron. J. 8 0 :876-881.
Bergersen, F. J., Turner, G. L., Chase, D. L., Gault, R.R., and Brockwell, J. 1985. The natural
abundance of 15N in an irrigated soybean crop and its use for the calculation of nitrogen fixation.
Aust. J. Agric. Res. 3 6 :411-423.
Bhuvaneswari, T. V., Lesniak, A. P., and Bauer, W. D. 1988. Efficiency of nodule initiation in
cowpea and soybean. Plant Physiol. 8 6 :1210-1215.
Biederbeck, V. O. and Geissler, H. J. 1993. Effect of storage temperatures on Rhizobium meliloti
survival in peat and clay-based inoculants. Can. J. Plant Sci. 7 3 :101-110.
Bohlool, B. B. 1987. Fluorescence methods for study of Rhizobium in culture and in situ. In
Symbiotic Nitrogen Fixation Technology. Edited by G. H. Elkan. Marcel Dekker, Inc., New York.
pp.127-147.
Boonkerd, N. 1991. Inoculant quality control and standards in Thailand. In Report On The
Expert Consultation On Legume Inoculant Production And Quality Control. Edited by J. A.
Thompson. FAO, Rome, United Nations. pp. 121-129.
Bordeleau, L. M. 1988. Effects of inoculation rate on the establishment and performance of alfalfa
in Quebec. Proceedings of the Inoculant Product Technology and Seed Inoculation Research
Workshop, Sept. 27-29, 1988. Agriculture Canada Research Station, Swift Current, Saskatchewan.
Bottomley, P. J. 1994. Light microscope methods for studying soil microorganisms.
In Methods Of Soil Analysis. Part 2. Microbiological And Chemical Properties. Edited by R. W.
86
Weaver. Soil Science Society of America, Madison.
Brockwell, J. and Bottomley, P. J. 1995. Recent advances in inoculant technology and prospects for
the future. Soil Biol. Biochem. 2 7 :683-697.
Brockwell, J., Roughley, R. J., and Herridge, D. F. 1987. Population dynamics of Rhizobium
japonicum strains used to inoculate three successive crops of soybean. Aust. J. Agric. Res. 3 8 :6174.
Burton, J. C. 1967. Rhizobium culture and use. In Microbial Technology. Edited by H. J. Peppler.
Reinhold, New York. pp. 1-33.
Burton, J. C. 1976. Methods of inoculating seeds and their effect on survival of rhizobia. In
Symbiotic Nitrogen Fixation In Plants. Edited by P. S. Nutman. International Biological
Programme 7. Cambridge University Press. pp. 175-189.
Burton, J. C. 1978. Monitoring quality in legume inoculants and preinoculated seed. Proceedings
IX Reunion Latinoamericana sobre Rhizobium, Mexico. pp. 308-325.
Burton, J. C. 1982. Modern concepts in legume production. In Biological Nitrogen Fixation
Technology For Tropical Agriculture. Edited by P. H. Graham and S. C. Harris. CIAT, Cali,
Columbia. pp.105-114.
Catroux, G. 1991. Inoculant quality standards and controls in France. In Report On The Expert
Consultation On Legume Inoculant Production And Quality Control. Edited by J. A. Thompson.
FAO, Rome, United Nations. pp. 113-120.
Catroux, G. and Amarger, N. 1992. Rhizobia as soil inoculants in agriculture. In Release Of
Genetically Engineered And Other Micro-Organisms. Edited by J. C. Fry and M. J. Day.
Cambridge University Press, Cambridge. pp. 1-13.
Cochran, W. G. 1950. Estimation of bacterial densities by means of the most probable number.
Biometrics 6 :105-116.
Date, R. A. and Roughley, R. J. 1977. Preparation of legume seed inoculants. In A Treatise On
Dinitrogen Fixation. Section IV. Agronomy And Ecology. Edited by R. W. F. Hardy and Gibson
A. H. Wiley, New York. pp. 243-275.
Date, R. A. 1972. Sources and quantities of yeast extract for growth of rhizobia. J. Appl. Bacteriol.
3 5 :379-387.
Day, J. M. 1991. Inoculant production in the UK. In Report On The Expert Consultation On
Legume Inoculant Production And Quality Control. Edited by J. A. Thompson. FAO, Rome,
United Nations. pp. 75-85.
87
De Oliveira, L. A. and Graham, P. H. 1990. Speed of nodulation and competitive ability among
strains of Rhizobium leguminosarum bv phaseoli. Arch. Microbiol. 1 5 3 :311-315.
Evans, H. J., Koch, B., and Klucas, R. 1972. Separation of nitrogenase from nodules and
separation into components. Methods Enzymol. 2 4 :470-476.
FAO report, 1991. Report on the expert consultation on legume inoculant production and quality
control (19-21 March, 1991, Rome). Food And Agriculture Organization of the United Nations.
Edited by: J. A. Thompson. 145 pp.
Fred, E. B. and Waksman, S. A. 1928. Laboratory manual of general microbiology, with special
reference to the microorganisms of the soil. McGraw-Hill, New York.
Gibson, A. H., Demezas, D. H., Gault, R. R., Bhuvaneswari, T. V., and Brockwell J. 1990.
Genetic stability in rhizobia in the field. Plant Soil 1 2 9 :37-44.
Halvorson, H. O. and Zeigler, N. R. 1933. Application of statistics to problems in bacteriology. I.
A means of determining bacterial populations by the dilution method. J. Bacteriol. 2 5 :101-121.
Herridge, D. F. and Roughley, R. J. 1975. Variation in colony characteristics and symbiotic
effectiveness of Rhizobium. J. Appl. Bacteriol. 3 8 :19-27.
Hitbold, A. E., Thurlow, D. L., and Skipper, H. D. 1980. Evaluation of commercial soybean
inoculants by various techniques. Agron. J. 7 2 :675-681.
Hume, D. J. and Blair, D. H. 1992. Effect of numbers of Bradyrhizobium japonicum applied in
commercial inoculants on soybean yield in Ontario. Can. J. Microbiol. 3 8 :588-593.
Joly, C. 1991. Biological nitrogen fixation within FAO agricultural production programmes in the
context of sustainable development. In Report On The Expert Consultation On Legume Inoculant
Production And Quality Control. Edited by J. A. Thompson. FAO, Rome, United Nations. pp. 914.
Keyser, H. H. 1987. The role of culture collections in biological nitrogen fixation. In Symbiotic
Nitrogen Fixation Technology. Edited by G. H. Elkan. Marcel Dekker, New York. pp. 413-428.
La Favre, J. S. and Eaglesham, A. R. J. 1984. Increased nodulation of "non-nodulating" (rj1rj1)
soybeans by high dose inoculation. Plant Soil 8 0 :297-300.
Lowther, W. L. and Littlejohn, R. P. 1984. Effect of strain of rhizobia, inoculation level, and
pelleting on the establishment of oversown Lotus pedunculatus "Grasslands Maku". New Zeal. J.
Exp. Agric. 1 2 :287-294.
Martinez-Romero, E. 1994. Recent developments in Rhizobium taxonomy. In Symbiotic Nitrogen
88
Fixation. Edited by: P. H. Graham, M. J. Sadowsky, and C. P. Vance. Kluwer Academic
Publishers, Dordrecht, The Netherlands. pp.11- 20.
Materon, L. A. and Weaver, R. W. 1984. Survival of Rhizobium trifolii on toxic and non-toxic
arrowleaf clover seeds. Soil Biol. Biochem. 1 6 :533-535.
Nambiar, P. T. C., Ravishanker, H. N., and Dart, P. J. 1983. Effect of Rhizobium numbers on
nodulation and dinitrogen fixation in groundnut. Exper. Agric. 1 9 :243-250.
Olsen, P. E. and Rice, W. A. 1989. Rhizobium strain identification and quantification in
commercial inoculants by immunoblot analysis. Appl. Environ. Microbiol. 5 5 :520-522.
Olsen, P. E., Rice, W. A., Bordeleau, L. M., and V. O. Biederbeck. 1994. Analysis and regulation
of legume inoculants in Canada: the need for an increase in standards. Plant Soil 1 6 1 :127-134.
Olsen, P. E., Rice, W. A., and Collins M. M. 1995. Biological contaminants in North American
inoculants. Soil Biol. Biochem. 2 7 :699-701.
Olsen, P. E. and Rice, W. A. 1996. Rapid evaluation of peat-base legume inoculant using
immunomagnetic beads for cell retrieval and fluorescent nucleic acid probes for viability analysis.
Proceedings of the 15th North American Conference on Symbiotic Nitrogen Fixation, Raleigh,
North Carolina, Aug., 1995. In Press.
Olsen, P. E., Rice, W. A., Bordeleau, L. M., Demidoff, A. H., and Collins, M. M. 1996. Levels
and identities of non-rhizobial micro-organisms found in commercial legume inoculant made with
non-sterile peat carrier. Can. J. Microbiol. In press.
Patrick, H. N. and Lowther, W. L. 1995. Influence of the number of rhizobia on the nodulation
and establishment of Trifolium ambiguum. Soil Biol. Biochem. 2 7 :717-720.
Rice, W. A. 1982. Performance of Rhizobium meliloti strains selected for low-pH tolerance. Can.
J. Plant Sci. 6 2 :941-948.
Roughley R. J. and Vincent J. M. 1967. Growth and survival of Rhizobium spp. in peat culture. J.
Appl. Bacteriol. 3 0 :362-376.
Scaglia, J. A. 1991. Production and quality control of inoculants in Rwanda. In Report On The
Expert Consultation On Legume Inoculant Production And Quality Control. Edited by J. A.
Thompson. FAO, Rome, United Nations. pp. 61-69.
Schall, E. D., Shenberger, L. C., and Swope, A. 1975. Inspection of legume inoculants and preinoculated seeds. Inspection report No. 106. Purdue University Agricultural Experiment Station,
Lafayette.
89
Scott, J. M. and Porter, F. E. 1986. An analysis of the accuracy of a plant infection technique for
counting rhizobia. Soil Biol. Biochem. 1 8 :355-362.
Skinner, F. Roughley, R. J., and Chandler, M. R. 1977. Effect of yeast extract concentration on
viability and cell distortion in Rhizobium sp. J. Appl. Bacteriol. 4 3 :287-297.
Skipper, H. D., Palmer, J. H., Giddens, J. E., and Woodruff, J. M. 1980. Evaluation of
commercial soybean inoculants from South Carolina and Georgia. Agron. J. 7 2 :673-674.
Smith, R. S., Ellis, M. A., and Smith, R. E. 1980. Effect of Rhizobium japonicum inoculant rates on
soybean nodulation in a tropical soil. Agron. J. 7 2 :505-508.
Smith, R. S., 1992. Legume inoculant formulation and application. Can. J. Microbiol. 3 8 :485-492.
Somasegaran, P. 1991. Inoculant production with emphasis on choice of carriers, methods of
production and reliability testing/quality assurance guidelines. In Report On The Expert
Consultation On Legume Inoculant Production And Quality Control. Edited by J. A. Thompson.
FAO, Rome, United Nations. pp. 15-32.
Somasegaran, P. and Hoben, H. J. 1994. Handbook for Rhizobia. Springer-Verlag, New York.
Somasegaran, P., Hoben, H. J., and Gurgun, V. 1988. Effects of inoculation rate, rhizobial strain
competition and nitrogen fixation in chickpea. Agron. J. 8 0 :68-73.
Stevens, W. I. 1958. Dilution series: A statistical test of technique. J. R. Statis. Soc. Ser. B. 2 0 :205214.
Strijdom, B. W., and Deschodt, C. C. 1976. Carriers of rhizobia and the effect of prior treatment
on the survival of rhizobia. In Symbiotic Nitrogen Fixation In Plants. Edited by P. S. Nutman.
International Biological Programme 7. Cambridge University Press. pp. 151-168.
Thies, J. E., Singleton, P. W., and Bohlool, B. B. 1991. Influence of the size of indigenous
rhizobial populations on establishment and symbiotic performance of introduced rhizobia on fieldgrown legumes. Appl. Environ. Microbiol. 5 7 :19-28.
Thompson, J. A. 1984. Production and quality control of carrier-based legume inoculants.
Information Bulletin No. 17. Patancheru, A.P., India: International Crops Research Institute for
the Semi-Arid Tropics.
Thompson, J. A. 1991a. Legume inoculant production and quality control. In
Report On The Expert Consultation On Legume Inoculant Production And Quality Control.
Edited by J. A. Thompson. FAO, Rome, United Nations. pp. 15-32.
Thompson, J. A. 1991b. Australian quality control and standards. In Report On The Expert
90
Consultation On Legume Inoculant Production And Quality Control. Edited by J. A. Thompson.
FAO, Rome, United Nations. pp. 107-111.
Tong, Z. and Sadowsky, M. J. 1994. A selective media for the isolation and quantification of
Bradyrhizobium japonicum and Bradyrhyzobium elkanii strains from soils and inoculants. Appl.
Environ. Microbiol. 6 0 :581-586.
Van Rensburg, H. J. and Strijdom, B. W. 1974. Quality control of Rhizobium inoculants produced
from sterilized and non-sterile peat in South Africa. Phytophylactica 6 :307-310.
Vincent, J. M. 1970. A manual for the practical study of the root-nodule bacteria. International
Biological Programme Handbook No.15. Blackwell Scientific Publications, Ltd., Oxford.
Vincent, J. E. and Smith, M. S. 1982. Evaluation of inoculant viability on commercially inoculated
legume seed. Agron. J. 7 4 :921-922.
Wadoux, P. 1991. Inoculant production in industry using sterile carriers. In Report On The
Expert Consultation On Legume Inoculant Production And Quality Control. Edited by J. A.
Thompson. FAO, Rome, United Nations. pp. 33-42.
Wardlaw, A. C. 1985. Practical Statistics For Experimental Biologists. John Wiley and Sons, New
York.
Weaver R. W. and Frederick L. R. 1974a. Effect of inoculum rate on competitive nodulation of
Glycine max L. Merill. I. Greenhouse studies. Agron. J. 6 6 : 229-232.
Weaver R. W. and Frederick L. R. 1974b. Effect of inoculum rate on competitive nodulation of
Glycine max L. Merill. II. Field studies. Agron. J. 6 6 : 233-236.
Weaver, R. W. and Graham, P. H. 1994. Legume nodule symbionts. In Methods Of Soil
Analysis. Part 2. Microbiological And Chemical Properties. Edited by R. W. Weaver. Soil Science
Society of America, Madison.
Weaver, R. W. and Wright, S. F. 1987. Variability in effectiveness of rhizobia during culture and
in nodules. Appl. Environ. Microbiol. 5 3 :2972-2974.
Wedderburn, M. E. 1986. Effect of applied nitrogen, broadcast lime, and seed pelleting on
establishment of Lotus pedunculatus cv. "Grasslands Maku" in tussock grasslands. New Zeal. J.
Exp. Agric. 1 4 :31-36.
Woomer, P. Bennett, J., and Yost R. 1990. Overcoming the inflexibility of most-probable-number
procedures. Agron. J. 8 2 :349-353.
Zuberer, D. A. 1994. Recovery and enumeration of viable bacteria. In Methods Of Soil Analysis.
Part 2. Microbiological And Chemical Properties. Edited by R. W. Weaver. Soil Science Society
91
of America, Madison.
92
A P P E N D I X I - Media and reagent formulations
Y EMA rhizobial plate counting media
nutrients K HPO
g per liter
micronutrients
-
mg per liter
0.5
H BO
3
1.0
0.2
ZnSO
4
1.0
CaSO ⋅2H O
0.1
CuSO ⋅5H O
0.5
MgSO ⋅7H O
0.2
MnCl ⋅4H O
0.5
Mannitol
10.0
Na MoO ⋅2H O
Yeast extract
2.0
Fe-EDTA (Sequestrene)
Agar
18.0
2
4
NaCl
4
2
4
2
3
4
2
2
2
2
4
0.1
2
10.0
Congo red - 2.5 mL of a 1:400 aqueous solution per L media.
CaCO may be added for tube slants - 0.15 g per L.
Monitor pH of autoclaved media and adjust to 6.9 - 7.1 if necessary.
For broth preparation, omit the agar.
3
Concentrated stock solutions for preparation of YEMB
mg/500 mL
1. K HPO
25.0 g/500 mL
6. H BO
3
50
10.0 g/500 mL
ZnSO
4
50
3. CaSO ⋅2H O
10.0 g/500 mL
CuSO ⋅5H O
25
4. MgSO ⋅7H O
10.0 g/500 mL
MnCl ⋅4H O
25
5. Fe-EDTA
(Sequestrene)
1.25 g/25 mL
Na MoO ⋅2H O
2
4
2. NaCl
4
4
2
2
3
4
2
2
2
93
2
4
2
5
Plant Nutrient Solution (1)*
Final
Concentration
mg/L PN
Micro-nutrients
CoCl ⋅6H O
H BO
MnCl ⋅4H O
ZnSO ⋅7H O
CuSO ⋅5H O
Na MoO ⋅2H O
H MoO ⋅H O
2
3
2
3
2
2
4
2
4
2
2
4
2
4
2
2
g/L stock
solutions
mL per L of stocks to make
plant nutrient solutions
0.004
2.86
1.81
0.22
0.08
0.121
0.09
0.004
2.86
1.81
0.22
0.08
0.121
0.09
1.0
492.96
174.18
136.09
110.99
5.00
246.48
174.18
136.09
110.99
5.00
2.0
1.0
1.0
1.0
1.0
Macro-nutrients
MgSO ⋅7H O
K HPO
KH PO
CaCl
FeC H O ⋅H O
4
2
2
4
2
4
2
6
5
7
2
PBS: (phosphate -buffered saline, 0.01M, pH 7.2)
Per liter:
NaH PO ⋅H O
0.46 g
Na HPO ⋅7H O
1.76 g
NaCl
8.5 g
purified water
1.0 liter
2
2
4
4
2
2
P B S T : (phosphate-buffered saline, 0.01M, pH 7.2, plus 0.05% Tween 20):
Add 0.5 mL of Tween 20 per 1000 mL PBS. Mix well.
94
Plant Nutrient Solution (2) *
Stock solution
Chemicals
Quantity
g per liter
Quantity of stock
per liter H O
93
3 mL
2
Without N
1
K SO
2
MgSO ⋅7H O
493
1 mL
3
KH PO
K HPO
23
1 mL
56
1 mL
2
4
4
2
2
2
4
4
4
CaCl
5
CaSO
6
FeCl
Na H EDTA
6.5
13
1 mL
H BO
MnSO ⋅H O
ZnSO ⋅7H O
CuSO ⋅5H O
NaMoO ⋅2H O
CoCl ⋅6H O
NiCl
0.23
0.16
0.22
0.08
0.025
0.034
0.022
1 mL
KNO
(NH ) SO
10
133
1 mL
b
2
1g
4
3
2
7
2
3
3
4
2
4
2
4
2
4
2
2
2
2
Plus N
8
c
3
4 2
4
a
Modified from Evans et al. (1972), cited by Weaver and Graham (1994).
Dissolve the two chemicals separately before combining.
For nutrient solution containing N, substitute stock solution 8 for 1.
b
c
P H O S P H A T E -PEPTONE DILUENT
peptone
KH PO
K HPO
2
2
4
4
1L
g/4 L
g/8 L
1.00
0.34
1.21
4.00
1.36
4.84
8.00
2.72
9.68
Dissolve in purified water and add 102 mL to 99 mL milk dilution bottles and autoclave for 45 minutes.
95
PEPTONE-GLUCOSE -BROMCRESOL PURPLE MEDIA
1000 mL purified water
5 g glucose
10 g peptone
agar 15 g;
10 mL of 1% bromcresol purple in ethanol (e.g., 0.10 g BCP in 10 mL EtOH).
ALKALINE PHOSPHATASE SUBSTRATE
E n z y m e s u b s t r a t e s t o c k s o l u t i o n s : Prepare NBT and BCIP substrate components as
stock solutions and store cold. Note that concentrated dimethylformamide will attack plastic and
that only glass containers, pipettes, teflon lined screw caps, etc., can be used. Prepare and store
NBT and BCIP stocks separately. Larger volumes than described may be prepared.
NBT (Nitroblue tetrazolium) stock solution:
30.0 mg NBT
0.7 mL dimethylformamide
0.3 mL distilled water
BCIP (5-bromo-4-chloro-3-indolyl phosphate) stock solution:
15.0 mg BCIP
1.0 mL dimethylformamide
E n z y m e s u b s t r a t e c a r r i e r b u f f e r : Carbonate Buffer (0.1M NaHCO3, 1.0mM MgCl2,
pH 9.8). Add 8.4 g NaHC03 and 0.203 g MgCl2 to 900 mL of distilled water. Adjust pH to 9.8
with sodium hydroxide and bring volume up to 1 liter. Store in refrigerator. The pH of the buffer
is important to this assay because alkaline phosphatase has maximum activity at this pH.
B L O C K I N G S O L U T I O N : for blocking non-specific protein binding to nitrocellulose
membranes (2% skim milk powder in PBS): dissolve 2 grams of skim milk powder in 100 mL of
PBS (no Tween 20). Prepare fresh for each use.
M O U N T I N G F L U I D:
D for fluorescent antibody staining - buffered glycerol (1 part PBS : 9
parts glycerine, ph adjusted to 9.3)
96
APPENDIX II - MATERIALS SOURCES
Most reagents, antibodies, media components, membranes, instruments, etc., are available from
multiple sources. The following is only a very small listing of potential sources and is not intended
to endorse any particular product or company.
NBT and BCIP - alkaline phosphatase precipitating substrate
Pierce Chemical Co.
BioRad Laboratories
nitrocellulose membrane for immuno-spot blot and colony-lift immunoblot
BioRad Laboratories
secondary antibodies labelled with alkaline phosphatase, FITC, biotin, or streptavidin
Boehringer-Mannheim
Sigma Chemical
BioRad Laboratories
skim milk powder
local food store
yeast extract, mannitol, peptone, glucose, agar
Difco
Sequestrene (Fe-EDTA)
Geigy
inorganic and organic chemicals
Fisher Scientific
Sigma Chemical
immunomagnetic beads
Dynal, Inc.
live/dead BacLight kit - differential fluorescent nucleic acid stains
Molecular Probes, Inc.
Tween 20
Fisher Scientific
Sigma Chemical
bromthymol blue, bromcresol purple
Sigma Chemical
Congo red
97
Sigma Chemical
Gram stain kit
BioMerieux Vitek, Inc.
round bottom 96 well microtiter plates
Dynatech
MPNES computer software
NifTAL Center
hydrophobic microscope slides for fluorescent antibody work
Cell-Line Associates, Inc.
piston type micropipettes
Brinkman Instruments, Inc.
Oxford Labware
BioRad Laboratories
Pierce Chemical Co.
98