Two methods were established for the detection and estimation
of low levels of some of the spoilage fungi which are responsible
for severe crop losses during storage under tropical conditions.
The fungi examined produce the enzyme pectinesterase
(EC.188.8.131.52) and both methods described, made use of this fact.
A modification of the Most-Probable-Number technique (used
for estimating Coliform bacteria in liquid samples) was utilized and
this enabled low numbers of Aspergillus flavus to be detected and
estimated within 18 hours. This method was based on the use of a
medium supplemented with pectin in which the pectinesterase produced
by the fungus, hydrolyses the ester linkages of the pectin to produce
methanol and polygalacturonic acid. This results in pH depression
of the medium causing an indicator-dye to change colour. Counts by
this method were generally higher than traditional Surface and Pour
Plate methods but the method has the advantage of being capable of
accommodating large sample volumes.
In the course of the MPN work it was found that the amount of
methanol liberated varied with the size of the spore inoculum. A
dialysis-cell procedure was developed for the removal of the methanol
from the medium followed by its estimation by Gas Liquid Chromatography
and this formed the basis of the second method.
A relationship Y = ce -ß/x was established; where y = log 10
of initial spore number/ml OR dry weight of spore material (mg/ml),
x =amount of methanol (~g/ml) released after a fixed incubation
time (15 hours) and c, e and e are constants.
Linearization of this function resulted in a straight line
relationship between spore numbers/ml of inoculum and methanol
production (R = -0.963). When dry weight of spores was plotted
against methanol production a better correlation (R = -0.988)
Either of these plots when used as a calibration graph
enabled the detection and estimation of spores of some spoilage
fungi belonging to the Aspergillus and Penicillium groups in the
range 20 - 106 spores/ml within 20 hours. The accuracy obtained was
comparable to traditional standard plating methods.
The spoilage Yeast strains examined did not obey the above
relationship, neither could a generalized equation be produced
solely for the yeasts.
A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.