Prediction of Listeria spp. growth as affected by various levels of chemicals, pH, temperature and storage time in a model broth

The effects of concentration of NaCl (0.5 to 12.5%), methyl paraben (0.0 to 0.2%), sodium propionate (0.3%), sodium benzoate (0.1%), potassium sorbate (0.3%), pH (>5.9) temperature (4 to 30°C), storage time (up to 58 d) and inoculum (>10⁵ to >10⁻² per ml) on the log₁₀ probability percentage of one cell of Listeria spp. to initiate growth in a broth system were evaluated in a factorial design study. At pH 5.96 and temperature ranging from 4 to 30°C the concentrations of sodium propionate, potassium sorbate, and sodium benzoate examined allowed growth of L. monocytogenes with lag phases at 4°C of 18, 27 and 21 days, respectively. For 0.1 and 0.2% methyl paraben growth of all Listeria spp. was initiated at 8°C and 30°C, respectively. At pH 6, concentration of 12% NaCl supported the growth of L. monocytogenes at 8 to 30°C, whereas 12.5% inhibited all Listeria species. Four regression equations were derived relating probability of growth initiation to temperature, concentrations of NaCl and preservatives storage time, and Listeria species specific effects. From these equations, the number of cells needed for growth initiation can be calculated. The impact of this type of quantitative study and its possible application on the development of microbial standards for foods is discussed.     

Behaviour of Listeria monocytogenes in raw sausages (merguez) in presence of a bacteriocin-producing lactococcal strain as a protective culture

The effectiveness of a bacteriocin produced by Lactococcus lactis subsp. lactis M in reducing population level and growth of Listeria monocytogenes ATCC 7644 in fermented merguez sausage was examined. Two different formulas (with or without added nitrites) were assayed and predetermined numbers of Listeria (ca 10⁶ cfu g⁻¹) were added to sausage mixture. The effect of in situ production of the bacteriocin by Lactococcus lactis M on Listeria monocytogenes ATCC 7644 during fermentation and storage of merguez sausages at room (ca 22 °C) or at refrigeration (ca 7 °C) temperature was tested. Results indicated that counts of Listeria monocytogenes were decreased during fermentation of merguez samples fermented with either the bacteriocin-producing Lactococcus lactis M (Bac⁺) or a nonbacteriocin-producing Lactococcus lactis J (Bac⁻). However, reduction in Listeria cfu's was greater in samples fermented with the Bac⁺ than in those fermented with the Bac⁻ starter. In merguez sausage made without nitrites addition, the Bac⁺ starter induced further decrease in Listeria counts by 1.5 log cycles compared with that induced by the Bac⁻ starter. While in merguez samples with added nitrites (0.4%), the effect of the bacteriocin produced in situ was less important than in those made without nitrites addition.     

Physiological properties of Lactobacillus paracasei, L. danicus and L. curvatus strains isolated from Estonian semi-hard cheese

Growth and survival of Lactobacillus paracasei (six strains), L. danicus sp. nov. (four isolates, two strains) and L. curvatus (two strains) from semi-hard Estonian cheeses were comparatively studied in different environmental conditions of relevance for their growth in cheese and survival in gastric environment. Maximum specific growth rates for L. paracasei strains varied between 0.40 and 0.57 h⁻¹, and all strains were tolerant to low water activities, heating at 60 °C for 30 min and pH 3. The newly discovered genetically distinct species L. danicus was characterized by low maximum growth rates (0.26–0.38 h⁻¹) and low temperature optimum (<30 °C). It was acid and heat sensitive and inhibited at salt concentrations from 4% and water activities below 0.93. L. curvatus was characterized by the highest growth rates (0.65–0.70 h⁻¹), tolerance to high NaCl concentrations, but sensitivity to heating, bile salts and low pH. The study showed that genetically different LAB species isolated from cheese could be distinguished by simple cultivation experiments.     

Growth characteristics of motile Aeromonas spp. isolated from different environments

The growth of 80 strains of motile Aeromonas spp. derived from environments with temperatures above 25°C and below 15°C, respectively, were examined at five temperatures (5°C, 10°C, 25°C, 37°C and 44°C) and four salt levels (0.05%, 2%, 4% and 6% NaCl). Sixty-one strains were further examined at two pH levels (pH 7.3 and pH 5.3). All strains grew at 25°C and 10°C with the majority of the isolates proliferating from approx. 10² to approx. 10⁷ cfu/ml within 1 and 3 days, respectively. In contrast, there were significant differences in the proportion of isolates able to grow at 5°C and 37°C depending on the temperature of their source of isolation. The ecological background of the organisms thus influences their thermal growth range and their ability to proliferate at body temperature, a highly significant factor in infective disease. At 25°C and pH 7.3, all strains grew in 0.05% NaCl, 96% grew in 2% NaCl, 96% grew in 2% NaCl while few grew in broth containing 4% or 6% NaCl. Lowering the pH to 5.3 with lactic acid caused a marked increase in the lag phase at 25°C and prevented growth of a large number of isolates at suboptimal conditions. Thus, none of the isolates from warm environments and only 8% of the isolates from cold environments grew at this pH at 5°C. The observed differences in growth optima between strains from different environments are discussed in relation to food- or waterborne infection.     

Listeria innocua and aerobic mesophiles during chill storage of inoculated mechanically recovered poultry meat treated with high hydrostatic pressure

Mechanically recovered poultry meat (MRPM) was inoculated with Listeria innocua 910 CECT at a level of approximately 10⁸ CFU g⁻¹. Vacuum-packaged samples were treated by combinations of pressure (350, 400, 450 and 500 MPa), time (5, 10, 15 and 30 min) and temperature (2, 10 and 20°C) and later stored at 2°C for 2 months. Counts of L. innocua and aerobic mesophilic bacteria were determined 1, 4, 7, 15, 30 and 60 days after pressurisation. For mesophiles, in most treatments, pressurization at 2°C gave the significantly best results. High pressure caused a marked bactericidal effect on L. innocua: reductions higher than 7.5 log units were achieved in several cases. Some cells were just sublethally injured by pressure. Samples treated at 500 MPa for 30 min at 2°C had counts of only 2.3 log units after 60 days of chill storage. Noninoculated pressurised MRPM did not show Listeria growth throughout storage. These results suggest that high pressure processing can enhance the microbiological quality of MRPM.     

The occurrence of Listeria monocytogenes in cheese from a manufacturer associated with a case of listeriosis

A case of listeriosis was associated with the consumption of a soft cheese produced in England. Goats cheese and other products from the same food manufacturer were examined for the presence of Listeria over the following 11 months. Listeria monocytogenes was isolated from 16 of 25 cheese samples on retail sale, 12 of 24 cheese samples obtained directly from the factory, and from shelving within the plant. Phage-typing of 68 isolates of L. monocytogenes from cheese samples and the factory showed that 66 (97%) were indistinguishable from the strain isolated from the patient's cerebrospinal fluid and stool. L. monocytogenes was not isolated from seven goats milk or two yoghurt samples. Listeria innocua was isolated from 10 cheese samples, two of which contained no other species of Listeria. Levels of L. monocytogenes shortly after production were low (<10/g), but were higher (10⁵–10⁷ cfu/g) in six of the 16 cheese samples obtained from retail outlets. Multiplication of L. monocytogenes was demonstrated in cheeses contaminated at the factory and held at 4°C in the laboratory.     

Microbiological quality of ayib, a traditional Ethiopian cottage cheese

One-hundred samples of ayib, a traditional Ethiopian cottage cheese, were purchased from Awassa market and analysed for their aerobic mesophilic counts, psychrotropic counts, yeasts and molds, coliforms, spore-formers, enterococci, lactic acid bacteria, Listeria monocytogenes, staphylococci and Bacillus cereus. The majority of the samples showed counts of mesophilic aerobic bacteria, yeasts and enterococci of 10⁸, 10⁷ and 10⁷ cfu/g. About 55% of the samples were positive for coliforms and faecal coliforms. Listeria spp. were not detected in any of the samples. B. cereus and S. aureus were isolated at varying frequencies but at low numbers (10²–10³). The pH value of the samples varied between 3.3 and 4.6 with about 40% having pH lower than 3.7.     

Purification, partial characterisation and mode of action of enterococcin EFS2, an antilisterial bacteriocin produced by a strain of Enterococcus faecalis isolated from a cheese

Enterococcus faecalis strain EFS2, isolated from the surface of a traditional cheese, produced a bacteriocin active against Gram-positive bacteria including Listeria spp. and some Staphylococcus aureus strains. The bacteriocin, named enterococcin EFS2, has been purified to homogeneity by ammonium sulphate precipitation and reversed-phase high performance liquid chromatography (RP-HPLC). The molecular weight was determined by mass spectrometry to be 7149.6. The amino acid composition of enterococcin EFS2 revealed that it contained 67 amino acid residues and had a blocked amino-terminal end. Enterococcin EFS2 induced viability loss, efflux of K⁺ ions and ATP, and cell lysis. Kinetic study of bactericidal activity of enterococcin EFS2 on Listeria innocua strain LIN 11 indicated slower cell destruction than by nisin. At pH 7.0, the activity of enterococcin EFS2 was the highest at 35 °C and was lost at 15 °C. The bacteriocin was more active against L. innocua strain LIN11 in broth adjusted to pH 6.0, 7.0 and 8.0 than to pH 4.5 at 30 °C.     

Selective enumeration and survival of bifidobacteria in fresh cheese

Five species of bifidobacteria (15 strains), two strains of Lactococcus lactis ssp. lactis, two strains of L. lactis ssp. cremoris, and one strain of L. lactis ssp. lactis var diacetylactis were included in a study to develop a selective medium for enumeration of bifidobacteria from fresh cheese. Viable counts of bifidobacteria or lactococci on modified Columbia agar base (CAB with 0.05% cysteine-HCl) plus raffinose (0.5%) containing various selective agents were compared with non-selective media. The mCAB plus raffinose with lithium chloride (2 g L⁻¹) and sodium propionate (3 g L⁻¹) with pH adjusted to 5.1 was used successfully as a selective medium for the enumeration of bifidobacteria from fresh cheese. Using this medium, it was determined that bifidobacteria could survive up to 15 days at a level higher than 10⁶ cfu g⁻¹ in a fresh cheese stored at 4 or 12°C. The decrease in the viable counts of bifidobacteria was faster during storage at 4°C than at 12°C.     

Modeling the effect of inoculum size and acid adaptation on growth/no growth interface of Escherichia coli O157:H7

The objective of this study was to model with logistic regression the growth/no growth interface of different initial inoculation levels (10¹, 10³ and 10⁵ CFU/ml; study 1), or nonadapted vs acid-adapted (study 2) Escherichia coli O157:H7 as influenced by pH, NaCl concentration and incubation temperature. Study 1 was conducted with a mixture of four E. coli O157:H7 strains grown (35 °C, 24 h) in tryptic soy broth (TSB). Study 2 was conducted with the same mixture of four E. coli O157:H7 strains grown (35 °C, 24 h) in glucose-free TSB with 1% added glucose (final pH 4.83), or in diluted lactic acid meat decontamination runoff fluids (washings; final pH 4.92), or nonadapted cultures prepared in glucose-free TSB (final pH 6.45), or in water washings (final pH 6.87). Parameters included incubation temperature (10–35 °C), pH (3.52–7.32), and NaCl concentration (0–10% w/v). Growth responses were evaluated for 60 days turbidimetrically (610 nm) every 5 days in 160 (study 1) and 360 (study 2) combinations in quadruplicate samples, with a microplate reader. The lower the initial inoculum the higher were the minimum pH and a_w values permitting growth. Differences in the pH and a_w growth limits among inoculum concentrations increased at 15 and 10 °C. Acid-adapted cultures were able to grow at lower pH than nonadapted cultures, while at temperatures below 25 °C, growth initiation of nonadapted cultures stopped at higher a_w compared to acid-adapted cultures for the whole pH range of 3.52 to 7.32. A comparison with available data indicated that our model for acid-adapted E. coli O157:H7 in different environments may provide representative growth probabilities covering both nonadapted and stress-adapted contaminants.     

Microbiological quality of 18 degrees C ready-to-eat food products sold in Taiwan

A total of 164 samples of 18 °C ready-to-eat (RTE) food products, purchased in 1999–2000 from convenience stores and supermarkets in central Taiwan, were examined to determine the microbiological quality of these products. The 18 °C RTE food products, manufactured by 16 factories, were divided into groups based on the type of food and their major ingredients. Aerobic plate count, coliforms, Escherichia coli, Bacillus cereus, Staphylococcus aureus and psychrotrophic Pseudomonas spp. were evaluated. The incidence of E. coli and coliforms in these 18 °C RTE food products was 7.9% and 75.0%, respectively, while 49.8% and 17.9% of the samples were found to contain B. cereus and S. aureus, respectively. Among the samples tested, 1.3% of the food products contained more than 10⁵ CFU g⁻¹ of B. cereus and 0.7% contained more than 10⁵ CFU g⁻¹ of S. aureus. The pH values of the samples were all below 7.0, except for cold noodles, which had pH values ranging from 5.18 to 8.20. Among the five types of 18 °C food products tested, the highest incidence of E. coli (16%) and Pseudomonas spp. (64.0%) were detected in hand-rolled sushi in a cone shape. On the other hand, the highest incidence rate of coliforms, B. cereus, and S. aureus were found in sandwiches (88%), cold noodles (66.7%) and rice balls rolled in seaweed (25.0%), respectively. Food products made of ham contained the highest incidence of coliforms (88.0%) and E. coli (16.0%), while food products containing meat and ham as the major ingredients had the highest incidence rates of B. cereus (62.5%) and S. aureus (26.1%), respectively. For coliforms, E. coli, B. cereus and S. aureus, the percentage of 18 °C RTE food products exceeding the microbiological standards for RTE food accepted by Republic of China was 75.0%, 7.9%, 49.8% and 17.9%, respectively.     

Preparation and validation of a growth model for Bacillus cereus: the effects of temperature, pH, sodium chloride and carbon dioxide

The growth responses of a vegetative inoculum of Bacillus cereus as influenced by varying conditions of temperature, pH value and sodium chloride concentration (% w/v) and carbon dioxide concentration (% v/v) were determined in laboratory medium. Growth curves in concentrations of NaCl in the range 0.5–10.5% (w/v), pH values in the range 4.5–7.0, CO₂ concentrations in the range 10–80% (v/v) and storage temperatures from 10 °C to 30 °C were fitted using the regime of Baranyi et al. (1993). A response surface model was prepared and predictions of doubling time, growth rate, lag time and time to 1000-fold increase could be obtained for any set of conditions within the matrix studied. This model is included in Food MicroModel Version 1. Predicted doubling times from the model were compared to observed doubling times in the literature and the model was found to give realistic estimates of doubling time for a range of foods including milk, meat and poultry and carbohydrate-based products.     

Production of cured meat color in nitrite-free Harbin red sausage by Lactobacillus fermentum fermentation

Lactobacillus fermentum was substituted for nitrite to produce cured pink color in a Chinese-style sausage. Treatments included inoculations (10⁴, 10⁶, and 10⁸ CFU/g meat) followed by fermentation at 30 °C for 8 h and then at 4 °C for 16 h. Control sausage (with sodium nitrite, 60 mg/kg meat) was cured at 4 °C for 24 h without L. fermentum. The UV–Vis spectra of pigment extract from L. fermentum-treated sausage were identical to that of nitrosylmyoglobin (NO-Mb) formed in nitrite-treated control. The NO-Mb concentration and the colorimetric a^* value of sausage treated with 10⁸ CFU/g meat of L. fermentum essentially replicated those in nitrite-cured meat. Free amino acid content in sausage treated with L. fermentum was greater and the pH slightly lower compared with the nitrite-cured control sample. This study showed that L. fermentum has the potential to substitute for nitrite in the sausage production.     

Pork tenderness estimation by taste panel, Warner-Bratzler shear force and on-line methods

The extent to which modification of Warner–Bratzler shear force (WBSF) determinations, relating to storage and preparation of the meat, aperture of the V-shaped cutting blade and shearing velocity, improve the relationship with sensory tenderness perception of pork was studied. Additionally four on-line methods: pH1, FOP1 (light scattering), PQM1 (conductivity) and DDLT (Double Density Light Transmission), were evaluated for their ability to predict tenderness. Sensory tenderness evaluation was conducted on 120 frozen (at −18°C for several months) samples of m. longissimus thoracis et lumborum. After overnight thawing, the meat was grilled to an internal temperature of 74°C and scored on an eight-point scale, from extremely tough to extremely tender. The standard WBSF procedure (protocol A) consisted of heating fresh meat samples (stored for 48 h at 4°C post slaughter) at 75°C for 50 min, cooling in cold tap water for 40 min, taking cylindrical cores parallel to the fibre direction, and shearing at a velocity of 200 mm/min with a blade aperture of 60°. For the prediction of sensory tenderness, the WBSF standard procedure (protocol A) showed the lowest variance (R²=15%) and the highest standard error of the estimate (SEE=0.97 N) compared to the other WBSF protocols. A decrease in shearing velocity, from 200 to 100 mm/min and, a replacement of the cutting blade with an aperture of 60° by one with an aperture of 30° led to improvements of R² (respectively, 19% vs. 13% and 47% vs. 23%) and SEE (respectively, 0.93 N vs. 0.97 N and 0.80 N vs. 0.97 N) and thus were better predictors of tenderness. A blade aperture of 30° instead of 60° also led to considerably lower WBSF values (22.1 N vs. 30.0 N). Freezing, frozen storage and thawing of the meat, prior to WBSF measurement, resulted in higher shear force values (32.7 N vs. 28.7 N) and a better prediction of tenderness, R² (25% vs. 15%) and SEE (0.94 N vs. 1.00 N). Furthermore, preparing the frozen stored meat for WBSF determination in the same way as for the sensory evaluation, namely grilling instead of boiling, led to higher WBSF values (35.5 N vs. 32.7 N) and a further improvement in the prediction of tenderness (R²=31% vs. 25% and SEE=0.90 N vs. 0.94 N). From the on-line instruments: pH, FOP and PQM, pH was best in predicting tenderness. Linear regression with tenderness as dependent variable and the on-line techniques as independent variables revealed the following R²: 16, 8, 8 and 10% and SEE: 0.96, 1.01, 1.01 and 1.00 N for, respectively, pH1, FOP1, PQM1 and DDLT. Thus, the classical instruments and the DDLT technique, which is analogous to the CGM (Capteur Gras/Maigre), an officially accepted carcass grading apparatus in France and Belgium, are not good predictors of tenderness.     

Sorption of chloroanisole vapors by raisin packaging material

The behavior and concentration of 2,4,6-trichloroanisole (TCA) vapors migrating into low-density polyethylene film (PE) of 0.39 μm at various temperatures and desorption of TCA from PE were determined. After 12 h exposure, 1642 μg g⁻¹ TCA was sorbed at 30 °C compared with 675 μg g⁻¹ at 20 °C. For PE to reach equilibrium of 4200 μg g⁻¹ at 30 °C took 48 h, but 120 h at 20 °C. The transmission of TCA through PE occurred after 12 h at 30 °C (8.9 μg kg⁻¹ m⁻² h⁻¹) and after 36 h at 20 °C (5.0 μg kg⁻¹ m⁻² h⁻¹). Desorption of TCA from PE increased with temperature. At 80 °C, 99% TCA was desorbed in 1 h compared to 51% at 40 °C, 31% at 30 °C and 17% at 20 °C. The rate of sorption, desorption and transmission of TCA vapors by PE is highly temperature-dependent.     

Low Temperature Growth of Type E Clostridium botulinum Spores. 1. Effects of Sodium Chloride, Sodium Nitrite and pH

SUMMARY— The effects of pH and the addition of sodium chloride (NaCl) or sodium nitrite (NaNO₂) to trypticase-peptone-glucose (TPGI and trypticase-peptone-sucrose-yeast extract (TPSY) media upon spore outgrowth were investigated using seven Type E Clostridium botulinum strains. An inoculum of 1.0 × 10⁵ spores/ml was used and the pH was adjusted with hydrochloric acid to cover the range of 5.2 to 6.6. To define salt tolerance, NaCl was added to the media at intervals of 0.5% in the range of 2.0 to 5.0% and at 0.1% intervals in the 4.5 to 5.0% range. The effect of NaNO₂ was investigated with the addition of 100 and 200 ppm NaNO₂ to the media. Samples were incubated at 30, 15.6, 10, 7.2, 5.0 and 3.4°C. Outgrowth of all strains tested was inhibited at pH 5.2 at 15.6°C. Inhibition occurred at higher pH at lower temperatures. None of the strains showed outgrowth with 4.87% NaCl in the media at any of the incubation temperatures used. Addition of 100 and 200 ppm NaNO₂ to the media inhibited the outgrowth of the Minneapolis, 517, 26080 and A6247 strains but not the Kalamazoo and Seattle Forks strains.     

A practical trial of pyrethrins-in-oil surface sprays for the protection of bagged grain against infestation by Cadra cautella (Wlk.) (Lepidoptera, Phycitidae) in Kenya

Two formulations of synergized pyrethrins in technical white oil were tested as monthly protective sprays on stacks of fumigated bagged wheat, primarily against Cadra cautella (Wlk.) but also against Sitophilus oryzae (L) and Tribolium castaneum (Hbst.), under warm-temperate storage conditions in up-country Kenya. The formulations were: 0·4% pyrethrins with 2·0% piperonyl butoxide, applied at 50 ml/m², and 0·4% pyrethrins with 0·4% piperonyl butoxide at 20 ml/m².

Results were assessed by recording infestation in samples taken from each stack after 18 weeks storage and five spray applications.

Both treatments gave reasonably good protection against C. cautella but were not satisfactory against S. oryzae or T. castaneum. There was no evidence of any taint in bread made from the treated grain, but the higher application rate caused excessive staining of the bags.

It is concluded that satisfactory control of reinfestation by C. cautella can be expected in practice using 0·4% pyrethrins in oil with only a minimal quantity of added piperonyl butoxide, and that 20 ml/m² is a suitable rate for application to bagged produce.     

Identification of halothane genotypes by calcium accumulation and their meat quality using live pigs

The three halothane genotypes (NN, Nn, and nm) were identified by measuring the capacity for Ca²⁺ accumulation by sarcoplasmic reticulum in whole muscle homogenate preparations of M. longissimus dorsi with a Ca²⁺ specific electrode at 35°C. Significant differences (P < 0·001) in deterioration (%) of Ca²⁺ accumulation, 12% for NN, 35% for Nn, and 81% for nn pigs, were observed after ageing the whole muscle homogenate preparations for 24 h in ice.

Predictions of meat quality in live pigs (n = 34) based on the values for water-holding capacity, assessed as fluid (g/0·5 g wet wt LD), and pH (fluid) by using small biopsy LD samples (Cheah et al. 1993) were performed on all the halothane genotypes. The halothane genotype NN (n = 11) showed a fluid value of 0·37 ± 0·01 and a pH (fluid) value of 6·62 ± 0·03 as compared with 0·61 ± 0·02 and 5·84 ± 0·04, respectively, for the halothane genotype nn (n = 13). The Nn pigs (n = 10) showed fluid (0·49 ± 0·03) and pH (fluid) (6·19 ± 0·11) values between those values observed for the two homozygotes (NN and nn). Predictions of meat quality in live pigs from biopsy LD muscles were confirmed from assessments on post-mortem LD muscles based on pH₁ and fibre optic probe (FOP) measurements.

The extent of deterioration (%) in Ca²⁺ accumulation showed high correlations with fluid (r = −0·861) and pH (fluid) (r = −0·831) in the biopsy LD samples, and with pH₁ (r = 0·663), FOP (r = −0·812), and drip (%) loss (r = −0·777) in the post-mortem LD samples.     

Temperature shows greater impact on bovine Longissimus dorsi muscle glycogen debranching enzyme activity than does salt concentration

The degradation of glycogen progresses by the co-operation of two enzymes: glycogen phosphorylase (phosphorylase) and glycogen debranching enzyme (GDE). We studied the effect of temperature (4–42 °C) and salt concentration (0–3% NaCl) on bovine M. longissimus dorsi GDE activity. GDE activity (n = 4) decreased significantly with decreasing temperature from about 40–4 °C. GDE exhibited 52% activity at 25 °C and 11% at 4 °C compared to its optimum activity measured at 39 °C. In rapidly chilled meat, the reduction in GDE activity may substantially delay the rate of glycolysis. However, residual GDE activity at 4 °C seems sufficient to enable the attainment of normal ultimate pH if the available time is long enough. An increase in salt concentration from 0% to 2% and to 3% induced a significant (P < 0.001) increase in the ultimate pH of ground bovine meat (n = 6), but showed no effect on GDE activity.     

Study of the effects of temperature, pH and NaCl on the peptidase activities of non-starter lactic acid bacteria (NSLAB) by quadratic response surface methodology

The individual and interactive effects of temperature, pH and NaCl on the aminopeptidase-types N and A and proline iminopeptidase activities of several strains of non-starter lactic acid bacteria (NSLAB) were studied by quadratic response surface methodology. The effects on enzyme activities depended on the interactions among the independent variables, species, strains within species and type of activity. With few exceptions, the aminopeptidases N and A and proline iminopeptidase activities of Lactobacillus casei ssp. pseudoplantarum and Lb. curvatus were strongly inhibited by the hostile cheese-like conditions, while the peptidases of Lb. casei ssp. casei and especially Lb. plantarum were tolerant or not affected by variations in pH and showed the least sensitivity to NaCl or even a requirement for NaCl for the optimal activity. Strains which maintained relatively high activity under cheese-like conditions were selected within the species. The proline iminopeptidase of Lb. casei ssp. casei 2107 as well as Lb. plantarum 2788 and 2789 greatly tolerated the interactions among the independent variables. At 10°C in the presence of 3.750% NaCl and independent of the pH, the PepN activity of Lb. plantarum 2788 and 2789 was at least 33% of the maximum determined at pH 7.5 and 35°C. Proline iminopeptidase activity was most sensitive to the individual and interactive variations in temperature, pH and NaCl. On the other hand, aminopeptidase-type A activity seemed to be least sensitive to cheese-like conditions.