Difference and consensus of whole cell Salmonella enterica subsp. enterica serovars matrix-assisted laser desorption/ionization time-of-flight mass spectrometry spectra.

AIM: Application of MALDI-TOF MS for characterization of strains of Salmonella enterica subsp. enterica. METHODS AND RESULTS: Whole cells were analysed by MALDI-TOF MS. Spectra with a maximum of 500 mass peaks between (m/z) 0 and 25000 were examined for consensus peaks manually and by a computer software algorithm. Consensus peaks were observed by both methods for spectra of Salmonella enterica serovars Derby, Hadar, Virchow, Anatum, Typhimurium and Enteritidis. CONCLUSIONS: Differences in numbers of consensus peaks in spectra obtained by manual and computer comparison indicated that development of the software involving statistical analysis of peak accuracy is necessary. SIGNIFICANCE AND IMPACT OF THE STUDY: Development of an analysis system for peak profiles in whole cell MALDI-TOF MS spectra to enable intra and interlaboratory comparison.

A collaborative study of a method for enumeration of probiotic enterococci in animal feed.

AIMS: Validation of an enumeration method to be used as an official control method in the framework of Council Directive 70/524/EEC for probiotic enterococci used as feed additives. METHODS AND RESULTS: Twenty laboratories in 12 European countries carried out a collaborative study. A plate count method using bile esculin azide (BEA) agar was used. Precision data in terms of repeatability (r) and reproducibility (R) of the method using different feeding stuffs and three inoculation levels were determined. Enterococci were present in the samples as a single component or in mixtures with other probiotic feed additives. The enumeration of enterococci on BEA agar showed a relative standard deviation (RSD)r of 1.5-3.6% and an RSD(R) between 2.9 and 7.4%. BEA agar was selective for enterococci in the presence of other probiotic micro-organisms such as pediococci, lactobacilli and yeast. CONCLUSIONS: For routine analysis of viable enterococci concentrations in feeding stuffs, the use of BEA is recommended. This methodology is not applicable for mineral feeds. SIGNIFICANCE AND IMPACT OF THE STUDY: An official control method for enumeration of authorized probiotic enterococci in feeding stuffs was validated. The results are intended for consideration for adoption as CEN and ISO standards.

Standardized laboratory-scale preparation of mayonnaise containing low levels of Salmonella enterica serovar Enteritidis.

Salmonella enterica serovar Enteritidis PT4 and PT6 are associated with food poisoning outbreaks and are often found in food only in low concentrations. In this study a reproducible laboratory-scale procedure for preparation of mayonnaise is presented. The mayonnaise that simulates a naturally low-level contaminated product can be used for validation of new methods and is also suitable to study the behavior of low numbers of food pathogenic spoilage microorganisms in a food environment. During processing, liquid egg was artificially contaminated with low levels of S. enterica serovar Enteritidis that resulted in levels of 1 to 3 log10 CFU/g in the final mayonnaise. Cells of S. enterica serovar Enteritidis had increased stability in the mayonnaise when they were subjected to low pH in two stages, first to pH 5.8 and afterward to pH 4.5 before addition to the mayonnaise. The pH of the mayonnaise was between 4.2 to 4.5 and remained stable over the storage period. Low-level S. enterica serovar Enteritidis remained stable in artificially contaminated mayonnaise for 4 weeks at 4 degrees C.

Investigation of bacterial spore structure by high resolution solid-state nuclear magnetic resonance spectroscopy and transmission electron microscopy.

High resolution solid-state nuclear magnetic resonance spectroscopy (NMR) in combination with transmission electron microscopy (TEM) of spores of Bacillus cereus, an outer coatless mutant B. subtilis 322, an inner coatless mutant B. subtilis 325 and of germinated spores of B. subtilis CMCC 604 were carried out. Structural differences in the coats, mainly protein of spores were reflected by NMR spectra which indicated also differences in molecular mobility of carbohydrates which was partially attributed to the cortex. Dipicolinic acid (DPA) of spores of B. cereus displayed a high degree of solid state order and may be crystalline. Heat activation was studied on spores of B. subtilis 357 lux + and revealed a structural change when analysed by TEM but this was not associated with increases in molecular mobility since no effects were measured by NMR.

Formation of biogenic amine in mayonnaise, herring and tuna fish salad by lactobacilli.

The effect of amino acid decarboxylase-positive lactobacilli in mayonnaise, herring and tuna fish salads on formation of biogenic amines (BA) was investigated. Commercial mayonnaise was inoculated with either of five amine-forming lactobacilli which were selected as model contaminants: Lactobacillus curvatus LTH 975 and LTH 1859 (cadaverine, putrescine, tyramine and phenylethylamine producing), L. delbrueckii LTH 1260 (tyramine and phenylethylamine forming) and L. buchneri LTH 1388 and LTH 661 (histamine forming). Low concentrations of tyramine (4.5 ppm) were detected and an addition of precursor amino acids resulted in an increase of amine concentrations to 40 ppm putrescine, 16.5 ppm tyramine and 5.5 ppm cadaverine. Herring and tuna fish salads were inoculated either with L. curvatus LTH 975 or L. Buchneri LTH 1388. In tuna fish salad 1 ppm putrescine, 3 ppm cadaverine, 7 ppm histamine and 28 ppm tyramine were found after 4 days when L. curvatus was added. In the corresponding herring salad putrescine (14 ppm), cadaverine (11.5 ppm), histamine (17 ppm) and tyramine (72 ppm) were detected. Fish salads containing L. buchneri displayed histamine concentrations of 900 ppm in tuna and 670 ppm in herring salad, respectively. Eight lactic acid bacteria and five yeasts, isolates from spoiled delicatessen salads and ingredients, were not able to form putrescine, cadaverine, histamine, tyramine or phenylethylamine.

Effect of enhanced proteolysis on formation of biogenic amines by lactobacilli during Gouda cheese ripening.

The effect of enhanced proteolysis on amine formation by amino acid decarboxylase positive Lactobacillus sp. during Gouda cheese ripening was examined. A commercial proteolytic enzyme preparation was added to pasteurized milk prior to cheese preparation. The effect of this manipulation on the formation of putrescine, histamine and tyramine was investigated in the presence and absence of amino acid decarboxylase-positive strains during a 12-week ripening period. Four batches were supplemented with a proteolytic enzyme. Batch I contained the proteolytic enzyme only, whereas batches II-IV were additionally supplemented with Lactobacillus delbrueckii LTH 1260 (batch II), Lactobacillus buchneri LTH 1388 (batch III) or Lactobacillus brevis LTH 2560 (batch IV). In batch I putrescine was detected with 4 mg/kg, in batch II, 42 mg/kg putrescine, 238 mg/kg histamine and 636 mg/kg tyramine were found. Batch III contained 13 mg/kg putrescine and 418 mg/kg histamine, whereas in batch IV, 26 mg/kg putrescine and 776 mg/kg tyramine were present. Batch V was supplemented with all three lactobacilli but did not contain the proteolytic enzyme. In this experiment, 4 mg/kg putrescine, 179 mg/kg histamine and 337 mg/kg tyramine were detected. A control cheese batch (VI) without addition of amine forming lactobacilli or a proteolytic enzyme was produced and only 4 mg/kg putrescine were detected. An increase in amine concentration during cheese ripening under conditions of enhanced proteolysis in the presence of starter and spoilage lactobacilli was evident from the experiments.

Degradation of histamine and tyramine by Brevibacterium linens during surface ripening of Munster cheese.

Red smear formation during fermentation of Munster cheese was started by using three different strains of Brevibacterium linens as surface inocula. The cheeses were produced with and without supplementation of histamine and tyramine. After smearing the cheese surface for the first time with B. linens viable counts of 10(7) CFU/g were detected. At the end of the logarithmic growth phase cell numbers increased to 10(10) CFU/g and remained constant during the whole ripening period. During a 4-week ripening period strains of B. linens reduced histamine and tyramine content by 55 to 70%. B. linens LTH 456 and LTH 3686 degraded histamine and tyramine in a phosphate buffer (pH 7) containing 0.54 M histamine and 0.58 M tyramine when incubated with agitation at 30 degrees C. B. linens LTH 3813 did not reveal any amine degradation activity in a buffer system. The pH on the cheese surface increased from 5 to 7, whereas it increased in the center only to 5.3 after a 3-week ripening period.

Histamine and tyramine degradation by food fermenting microorganisms.

Microorganisms suitable for food fermentation were examined with regard to their potential to degrade histamine and tyramine. Out of 64 lactic acid bacteria evaluated in this study, 27 degraded histamine and one tyramine, respectively, with low activity. Among 32 strains of Brevibacterium linens and coryneform bacteria, 21 exhibited histamine and tyramine oxidase activity. None of 20 strains of Staphylococcus carnosus tested degraded histamine or tyramine. One strain out of nine strains of Geotrichum candidum degraded tyramine slightly. Among 44 strains of Micrococcus sp. examined, 17 degraded either one or two biogenic amines. In this study Micrococcus varians (M. varians) LTH 1540 exhibited the highest tyramine oxidase activity of all strains tested and was therefore investigated in detail. The enzyme was found to be located in the cytoplasm and was not membrane bound. The reaction end product p-hydroxyphenyl acetic acid was detected by HPLC analysis. An activity staining for the amine oxidase in a native polyacrylamide gel based on the formation of H2O2 during amine oxidation was developed. Resting cells of the strain exhibited optimal tyramine oxidase activity at a pH of 7 at 37-40 degrees C. The enzyme in the cell free extract had a pH optimum between 7-8. The enzyme activity was decreased by NaCl, glucose and hydralazine. Phenylethylamine and tryptamine were oxidized at lower concentrations than tyramine. The potential for amine degradation was not found to be associated with that of formation of biogenic amines, as 23 microorganisms with the ability to metabolise biogenic amines exhibited no decarboxylase activity toward histidine, tyrosine, phenylalanine, lysine or ornithine.

Tyramine degradation by micrococci during ripening of fermented sausage.

The ability of tyramine oxidase exhibiting strains of Micrococcus varians to degrade tyramine in vivo during sausage fermentation was investigated. Fermented sausage was produced with a tyramine forming strain of Lactobacillus curvatus which acquired a final tyramine concentration of 190ppm. The addition of either one of two strains of M. varians exhibiting a potential to oxidise tyramine decreased the amount of tyramine formed to 160 and 150ppm, respectively. No effect on growth of L. curvatus and pH development in the fermented sausage was observed when micrococci were present during a three weeks ripening period. Sausage fermentation was further carried out with a non-amine forming strain of L. sake and M. varians after addition of 100ppm tyramine to the raw material. The amount of tyramine in the end product was 60ppm, and during the first 10 days of ripening the decrease was faster on the outside of the sausage.

Method for the rapid quantitative detection of lipolytic activity among food fermenting microorganisms.

A standard method for the detection of free fatty acids (FFAs) in milk was modified and applied to the measurement of the lipolytic activity of microorganisms in a model system containing either homogenised pork or beef fat tissue. The increase in FFAs was measured colorimetrically using palmitic acid as a standard. Among the strains tested, two strains of Staphylococcus xylosus and one strain of Staphylococcus carnosus were found to display lipolytic activity. For all strains, a higher increase in FFA was observed in broth supplemented with pork fat than with beef fat. All three strains displayed lipolytic activity when tested on tributyrin agar plates.