Journal of Applied Microbiology

Weinberg, Z. G.; Ashbell, G.; Azrieli, A.

doi: 10.1111/j.1365-2672.1988.tb02423.xpmid: N/A

The effect of applying a commercial lactic acid bacterial inoculant, at 5.6 × 104 cfu/g fresh material, to vetch, wheat, direct‐cut and wilted alfalfa silages has been studied under laboratory conditions, and on wheat also under farm conditions. Dry matter losses in the inoculated vetch and alfalfa silages were smaller than in the control silages, due to improved fermentation in the former as indicated by a faster and larger pH decrease and by a faster and larger lactic acid build‐up. Volatile fatty acid analysis also indicated more efficient fermentation patterns in the inoculated vetch and alfalfa silages with less ethanol, acetic and butyric acids compared with the respective control silages. The inoculant suppressed enterobacteria and clostridia in the inoculated direct‐cut alfalfa silage. The inoculant did not have a great effect on the wheat silages.

Berwick, P. G.

doi: 10.1111/j.1365-2672.1988.tb02424.xpmid: 3127373

Streptomyces cattleya, S. fradiae and S. griseus produced different amounts of growth when cultured sequentially through sporulation, vegetative and antibiotic production media. Only S. griseus grew well on all three types of medium. Streptomyces cattleya grew poorly on both sporulation and vegetative media. Growth was 1·6 and 8·0 mg/1/h respectively. For all three species, biomass yield in the final antibiotic production medium was dependent on amount of inoculum. Antibiotic yields were obtained only from production media. Under slow growth conditions l‐cysteine and l‐valine supplementation stimulated S. cattleyaβ‐lactam production, giving 1000 μg/ml β‐lactam equivalents compared with 45 μg/ml β‐lactam equivalents for no supplementation. For aminoglycosides the agar well diffusion bioassay was more sensitive towards the hydrochloride than the neutral salt. Paper chromatography confirmed the main antibiotic classes. RF values for replicate samples indicated aminoglycoside homogeneity and β‐lactam heterogeneity.

Blanco, J. L.; Domínguez, L.; Gómez‐Lucía, Esperanza; Garayzabal, J. F. F.; Goyache, J.; Suárez, G.

doi: 10.1111/j.1365-2672.1988.tb02425.xpmid: 3350782

Manchego‐type cheese, a typical Spanish cheese, was inoculated in various ways with an aflatoxigenic organism, Aspergillus parasiticus NRRL 2999, to study the production of aflatoxin. When the original milk was contaminated with a spore suspension, aflatoxin was not detected in paraffin‐covered cheeses although it was present in the top layer of non‐paraffin‐covered cheeses after ripening at 15°C for 60 d. When the cheese surface was inoculated, no aflatoxins were detected in paraffin‐covered cheeses after ripening for 60 d although they were found when the cheeses were ripened for 30 d. In non‐paraffin‐covered cheeses aflatoxins were detected only in the top layer and in the second 10 mm layer when cheeses were incubated after the normal ripening at 28°C for 30 d. When the centre of the cheese was inoculated, no aflatoxins were detected although Aspergillus grew slightly along the inoculation area. When cheese portions were inoculated, fungal growth was evident after incubation at 28° and 15°C for 6 d but there was no growth at 10°C after 50 d. At 28°C aflatoxins were detected at a concentration of 132 μg/g after 13 d, the highest level obtained. In cheese paste at 28° and 15°C, growth was intense, but the level of aflatoxins detected was lower than in cheese portions. At 10°C the growth was heavy, but aflatoxins were not detected.

Holmes, B.; Costas, M.; Sloss, L. L.

doi: 10.1111/j.1365-2672.1988.tb02426.xpmid: 3350783

Twenty‐five strains of Providencia alcalifaciens from various countries have been characterized by one‐dimensional SDS‐PAGE of cellular proteins. They comprised 15 from human faeces, one from duck faeces, one from a guinea‐pig eye and eight from unknown sources. Also included, for reference purposes, were the type strains of three other Providencia species. The protein patterns, which contained 45–50 discrete bands, were highly reproducible and were used as the basis for two numerical analyses. In the first, in which the principal protein bands (in the 33–40 kD range) were excluded, the 25 Prov. alcalifaciens strains formed, at the 83% S level, a single cluster whilst the three Providencia reference strains remained unclustered. In the second, which included all the protein bands, the 25 Prov. alcalifaciens strains formed 10 clusters at the 85% S level. We conclude that high resolution PAGE combined with computerized analysis of protein patterns provides the basis for typing clinical strains of Prov, alcalifaciens. Reference strains of each of the 10 PAGE types identified are available from NCTC for inclusion in future studies.

Macfarlane, G. T.; Allison, C.; Gibson, S. A. W.; Cummings, J. H.

doi: 10.1111/j.1365-2672.1988.tb02427.xpmid: 3127369

Protease activities in human ileal effluent were approximately 20‐fold greater than in normal faeces. Comparative studies with faeces from a person who did not have a pancreas suggested that a substantial proportion of the proteolytic activity in normal faeces was of bacterial origin. Thimerosal, iodoacetate, EDTA and cysteine significantly inhibited proteolysis in faeces, but not in small intestinal contents, showing that cysteine and metalloproteases were produced by bacteria in the large gut. These results, together with results from studies using p‐nitroanilide substrates, demonstrated that faecal proteolysis was both qualitatively and quantitatively different from that in the small intestine. Studies with pure cultures of proteolytic gut bacteria indicated that the cell‐bound proteases of Bacteroides fragilis‐type organisms were likely to contribute significantly towards proteolytic activity associated with the washed cell fraction and washed particulate fraction of faeces. Extracellular proteases were formed by Streptococcus faecalis ST6, Propionibacterium acnes P6, Clostridium perfringens C16, Cl. bifermentans C21 and Cl. sporogenes C25. Inhibition results suggested that these bacteria, and similar organisms, may be partly responsible for the extracellular proteolytic activity found in the cell‐free supernatant fraction of faeces.

Phillips, A. P.; Martin, K. L.

doi: 10.1111/j.1365-2672.1988.tb02428.xpmid: 3127370

Fluorescein‐conjugated rabbit antibodies to formalized spores of Bacillus anthracis were tested against strains of B. anthracis and other Bacillus species in a subjective immunofluorescence test. The lack of reaction of B. anthracis Vollum spores with conjugated antibody raised against B. anthracis Sterne spores indicated that spores of the Vollum strain lacked a major surface antigen present in most of the other anthrax strains tested, including the non‐encapsulated strains Sterne and the Soviet ST1, variants cured of the pX01 plasmid that codes for the toxin, and several virulent strains. Four other antibody preparations, raised against B, anthracis Vollum, New Hampshire, Ames and Strain 15, reacted to an approximately similar degree with spores of all four strains and of Sterne, indicating that Vollum has at least one spore antigen in common with these other strains. The anti‐Sterne and anti‐Vollum conjugates both displayed cross‐reactions with spores of strains of B. cereus, B. coagulans, B. subtilis, B. megaterium, B. polymyxa, B. pumilus and B. thuringiensis. Absorption of the anti‐anthrax conjugates with B. cereus NCTC 8035 and NCTC 10320 removed all these cross‐reactions, demonstrating the existence of spore antigens specific for anthrax.

Kaukas, Anne; Hinton, M.; Linton, A. H.

doi: 10.1111/j.1365-2672.1988.tb02429.xpmid: 3127371

Small groups of chickens were given feed containing either avoparcin, nitrovin, vir‐giniamycin or zinc bacitracin from the day of their purchase as day‐olds. Differences between the birds receiving growth promoters and the untreated controls were observed during the last third of the 23 d survey period. The enterococcal population of the ‘dosed’ birds contained a greater proportion of Enterococcus faecium than did that of the control birds while the converse was true for Ent. gallinarum. This apparent selection of Ent. faecium by the growth‐promoting antibiotics had an influence on the incidence of resistance to therapeutic antibiotics among the enterococcal population as a whole. This was because this species was generally more resistant than Ent. gallinarum to cephalothin, the MLS antibiotics (erythromycin, lincomycin and tylosin) and tetracycline.

Rodrigues, Ubaldina M.; Kroll, R. G.

doi: 10.1111/j.1365-2672.1988.tb02430.xpmid: 3280541

The growth patterns of macrocolonies of 59 different pure cultures were studied on eight selective solid media. A method of growing microcolonies on the surface of polycarbonate membrane filters, placed on the selective agar media, followed by staining and examination by epifluorescent microscopy was developed. The patterns of growth of the pure cultures as microcolonies were studied on the eight selective media. Only four media proved to be reliable for this purpose and the relationship between the microcolony count and plate count was studied on these media together with nutrient agar. Microcolony counts using three of these media (enriched lauryl sulphate aniline blue, pseudomonas selective agar (C‐F‐C) and Baird‐Parker medium) were capable of giving reliable estimates of coliforms (r = 0·89), pseudomonads (r = 0·93) and staphylococci (r = 0·92) after incubation at 30°C for 3 or 6 h (staphylococci) at contamination levels of above 103 bacteria/g in a variety of foods. The results are available within a working day and should allow the more efficient management of food supplies.

Shute, L. A.; Gutteridge, C. S.; Norris, J. R.; Berkeley, R. C. W.

doi: 10.1111/j.1365-2672.1988.tb02431.xpmid: 3127372

The discrimination of a set of 53 strains, taken from four closely related Bacillus species (Bacillus subtilis, B. pumilus, B. licheniformis and B. amyloliquefaciens), was examined using pyrolysis mass spectrometry. Strains were grown on six different media to examine the effect of media variation, especially batch‐to‐batch variation of a single medium, on the pyrolysis mass spectra and strain discrimination achieved. Long‐term reproducibility over a period of 14 months was also examined. It was shown that batch‐to‐batch media variation is insufficient to affect spectra and strain discrimination significantly, but different media types do affect this. It was shown that species groups could still be recovered from the data, however, with an appropriate data‐handling system. It was not possible to directly compare spectra produced 14 months apart, but the strain and species discrimination achieved using each data‐set were highly comparable.

Havelaar, A. H.; During, M.

doi: 10.1111/j.1365-2672.1988.tb02432.xpmid: 3280542

The direct plating (DP) method for enumerating Escherichia coli in food was adapted for water analysis by membrane filtration and a standardized protocol was described. The DP method was found to give equal or better recoveries of E. coli than a membrane filtration method using 0·1% sodium lauryl sulphate agar; the repeatability of the DP method was markedly better. The necessity to transfer membranes from the non‐selective medium tryptone soy agar (TSA) to the selective medium tryptone bile agar (TBA) after pre‐incubation for 4 h was considered disadvantageous for practical purposes. A double‐layer method, where the membrane filter is placed on a layer of TSA poured over TBA, with incubation in an incubator that automatically switches from 37° to 44°C after 4 h, was found to be an acceptable alternative. Recovery of E. coli and inhibition of competitive flora were equal or only slightly less than for the standard DP method.