Proceedings
of The World Avocado Congress III, 1995 pp. 433 - 437
ELISA
OPTIMISATION FOR MONITORING BACILLUS SUBTILIS IN PREHARVEST
BIOCONTROL PROGRAMMES
E. Towsen L. Korsten
Faculty of Biological and Agricultural Science
University
of Pretoria, Pretoria 0002, South Africa
Enzyme linked immunosorbent
assay (ELISA) was optimised for more effective monitoring of Bacillus
subtilis in avocado field spray programmes using natural antagonists.
Parameters of the ELISA that were optimised included, incubation temperature
and -time, fixatives, blocking buffers, washing buffers, antibody incubation
time, conjugate concentration, conjugate incubation time and enzyme- substrate
reaction time. Incubation conditions of 4ºC for 24 h gave the highest
signal/background value, the baking method was found more applicable due to its
3 months storage potential without loss of ELISA signals. Bovine serum albumin
gave the highest signals when used as a blocking solution. When comparing the
optimised ELISA protocol with the one used previously, a higher signal/
background value were obtained with the former ELISA.
I .
Introduction
Disadvantages associated with
the use of chemicals and their detrimental effects on human health and the
environment have resulted in an increased interest in alternative control
strategies such as biological control. Biological control has been evaluated
successfully at a pre- and postharvest level for control of avocado fruit
diseases using Bacillus subtilis sprays (Korsten et al., 1993).
When bacteria are applied as biocontrol agents in the field, their populations
should be monitored at several intervals starting from the time of application
(Spurr and Knudsen, 1985). This information is necessary to improve biocontrol
efficacy through better formulation, minimising wasteful applications and
optimising spray scheduling (Knudsen and Spurr, 1987).
Different indirect methods
such as leaf imprints and -washings and direct scanning electron microscopy
(SEM) have been evaluated to detect antagonist populations after field
applications in avocado orchards (unpublished data). The indirect techniques
gave lower viable counts compared to total higher counts with the SEM. These
techniques were found to be either inaccurate or not suited to large scale
monitoring of field sprays. Bacterial populations have been successfully
monitored using antibodies (Schaad et al., 1990). Subsequently,
monoclonal antibodies were produced against the antagonist (Unpublished data).
The purpose of this study was to optimise the ELISA
technique for efficient and rapid monitoring of antagonists in field biocontrol
studies.
2.
Materials and methods
2.1 Monoclonal antibodies
Isolate B246 originally isolated
form the avocado phylloplane and obtained from L. Korsten (Department of
Microbiology and Plant Pathology, University of Pretoria) was used for the
optimisation study. Monoclonal antibodies secreted by clone 6B7E5 (IgG2b) which
gave the highest optical density reading when tested against B246, was selected
for the optimisation studies.
2.2 ELISA optimisation
The ELISA technique described
by Verschoor et al. (1989) was used throughout the optimisation study,
changing only the various steps under investigation. A total of 11 replicates
were used per assay and a negative control (Dulbecco's modified Eagle's medium
(DMEM)) was included in each batch. B246 suspensions (100 μl) were
incubated in 96-well microtiter plates (Cooke Microtiter system M29A, Sterilin
products, Middlesex, England) for either 1, 2 or 4 h or overnight at
temperatures of either 4, 24 and 37ºC or were air dried for 1 h under a heating
lamp and laboratory fan. Various fixatives were compared viz. 70, 90 and 100 %
methanol; 70, 80 and 100 % ethanol; 60 and 70 % - 1- butanol; 0.25, 0.5 and
0.75 % formaldehyde; 0.25, 0.5 and 0.75 % glutaraldehyde (After I h of fixation
with glutaraldehyde plates were washed 3 times with 100 mM glycine in phosphate
buffered saline (PBS) pH 7.2); 0. 1 % sodium dodecyl sulphate (SDS) in
carbonate buffer; 60, 70 and 80 % isopropanol; 6 M urea in carbonate buffer;
50, 60 or 70 % acetic acid. Fixatives were incubated at room temperature for 1
h. The most effective blocking buffer were selected comparing 0.5, 1, 1.5, %
skim milk in PBS; 0.5, 1, 1.5 % gelatine in PBS and 1, 3 and 5 % bovine serum
albumin (BSA). The blocking solutions were incubated for 1, 2 h or overnight.
Different washing buffers were compared using 0.5 % Tween 20 in PBS, 1 % BSA in
PBS, 0.5 % casein in PBS and PBS. Optimum primary antibody incubation time (30,
45, and 60 min), optimum conjugate concentration (Peroxidase conjugated
goat-anti-mouse IgG; heavy and light chain specific, Dakopatts, Denmark) (1:
100, 1: 1000, 1:5000, and 1: 10000), conjugate incubation time (30, 45, and 60
min) and enzyme-substrate reaction time (10, 20, 30, 40, 50, and 60 min) using
urea peroxide and orthophenylene diamine in citrate buffer were determined.
Superiority of the optimised ELISA was determined by comparing it with the
ELISA technique previously used according to Verschoor et al. (1989).
Values were considered
positive if they were twice that of the background with a corresponding
coefficient of variation of less than 10 %.
3. Result
3.1 ELISA optimisation
A comparison of various
incubation temperatures showed that the highest signal/background ratio was
achieved at VC for 24 h and that the lowest standard coefficient of variation
was observed when micrititer plate was baked under a heating lamp and fan
(Table 1). Of the various fixative concentrations evaluated, the 0.5%
glutaraldehyde, 6M urea and 0.75% formaldehyde gave the highest
signal/background ratio (Table 2). The coefficient of variation for the three
fixatives were less than 10%. 60% 1 - butanol, 80% isopropanol, 70% acetic acid
and 0.25% glutaraldehyde gave a zero coefficient of variation (Table 2).
Differences were also found between the various blocking solutions used (Table
3). Both 1% and 5% BSA concentrations gave the highest signal/background ratio
and the coefficient of variation was also within acceptable limits. A 2 h
incubation time for 1 % BSA as blocking solution gave the highest
signal/background ratio (Table 3).
Tween 20/PBS gave the highest signal/background ratio coupled with the
lowest coefficient of variation when used at a concentration of 0.5 %, thus
making it more effective. Although the incubation time of the primary antibody
gave the highest value after 45 min, the coefficient of variation was more than
10%. An incubation time of 60 min was more acceptable. A conjugate
concentration of 1 : 5000 gave the highest signal/background ratio and 45 min
incubation time of conjugate dilution 1 : 5000 also gave the highest
signal/background ratio. All conjugate concentrations gave an acceptable
coefficient of variation. The highest signal/background ratio for different for
enzyme - substrate reaction times were obtained after 30 min, which also gave
the highest coefficient of variation. The coefficient of variation after 50 and
60 min was more acceptable.
Discussion
When comparing the optimised
ELISA protocol with the one used previously (Verschoor et al., 1989) a
higher signal/background ratio was obtained with the former ELISA protocol and
it is therefore more suitable to detect different concentrations of B.
subtilis in field studies.
Optimisation of ELISA for
commercial screening is of utmost importance, in order to rule out variation
between batches. It is therefore important, when establishing a new
immunoassay, to set acceptable levels of precision, and to achieve an
inter-assay variation of less than 10 % (McClaren et al., 1981).
Since a variety of ELISA
protocols have been described for various applications and antigen - antibody
systems (Korsten et al.,1990; Temeyer et al., 1986; Verschoor et al.,
1989), it was deemed necessary to optimise this particular system to suit
it's purpose, namely detection of low concentrations of antigen in the field.
Throughout this
study the optimum dilution, time and concentration chosen had a coefficient of
variation of less than 10 %. The highest signal/ background ratio was obtained
with incubation temperatures of 4ºC for 24 h. This adsorption procedure seemed
to give a more uniform coating. The same conditions were obtained by Korsten et
al. (1990) when optimising the ELISA to detect Pseudomonas syringae on
avocado. In contrast, Verschoor et al (1989) pointed out the advantage
of baking with a heating lamp and laboratory fan, which include storing the
plates at 4ºC for at least 3 months with no significant loss of ELISA - signal.
BSA (1 %) gave the highest signal/background value when used as a blocking
solution and Tween 20/PBS as a washing buffer. This is in accordance with a
report by (McClaren et al., 1981) who found that nonspecific binding of
the enzyme conjugate to the coated solid phase can be reduced by including
Tween 20 and BSA in the medium. When compared to the ELISA protocol used
previously (Verschoor et al., 1989) a higher signal/background ratio was
obtained with the optimised ELISA and is therefore more suited to detection of
different concentrations of B. subtilis in field studies.
The sensitivity, speed and
large amount of samples that can be processed (Clark, 1981) are attributes of
ELISA. This method is therefore more efficient when compared to leaf imprint,
washing and SEM techniques. The ELISA method can be used to monitor B.
subtilis throughout the season, which will give information to improve
biocontrol effectiveness through better formulation, adjustment of dosage and
spray scheduling.
References
Clark, M.F., 1981.
Immunosorbent assays in plant pathology. Annual Review of Phytopathology 19 :
83 - 106.
Korsten, L., De Villiers,
E.E., Weimer, F.C. and Kotz6, J.M., 1993. A review of biological control of postharvest
diseases of subtropical fruits. In: Champ, B.R., Highley, E. and Johnson, G.I.
(Eds). Postharvest handling of tropical fruits. Proceedings of an International
conference held at Chiang Mai, Thailand. Watson, Ferguson and Company : 172 -
185.
Korsten, L., Sanders, G.M.
and Kotz6, J.M., 1990. Practical application of monoclonal antibodies in the
detection of bacterial diseases of subtropical crops. Acta Horticulturae 275 :
745 - 751.
Knudsen, G.R. and Spurr,
H.W., 1987. Field persistence and efficacy of five bacterial preparations for
control of peanut leaf spot. Plant Disease 71 442-445.
McClaren, M.L., Lillywhite,
J.E. and Andrews, C.S., 1981. Indirect enzyme linked immunosorbent assay
(ELISA): practical aspects of standardization and quality control. Medical
Laboratory Sciences 38 : 245 - 251.
Schaad, N.W., SOle, S., Van
vuurde, J.W.L., Vruggink, H., Alvarez, A.M., Benedict, A.A., De Wael, L. and
Van Laere, 0., 1990. In: Klement, Z., Rudolph, K. and Sands, D.C. (Eds).
Methods in Phytobacteriology. Akadémiai Kiadò, Budapest : 153 - 190.
Spurr, H.W. and Knudsen,
G.R., 1985. Biological control of leaf diseases with bacteria. In: Windels,
C.E. and Lindow, S.E. (Eds). Biological control on the phylloplane. The
American Phytopathological Society St. Paul. Minnesota : 45 - 62.
Temeyer, K.B., Haufler, M.
and Pruett, J.H., 1986. Development of an improved ELISA for antibody detection
and use in production of a hybridoma secreting a monoclonal antibody specific
for crystal protein of Bacillus thutingiensis ssp. israelensis. Journal
of Economic Entomology 79 : 1287 - 1293.
Verschoor,
J.A., Coetzee, L. and Visser, L., 1989. Monoclonal antibody characterization of
two field strains of Haemophilus paragallinanan isolated from vaccinated
layer hens. Avian Diseases 33 : 219 - 225.