Scientific Publications - Work Done by Microbiology Reader Bioscreen C

Food Microbiology, October 1 1997, Vol. 14, No. 5, pp. 403-412

Variability of the response of 66 Listeria monocytogenes and Listeria innocua strains to different growth conditions

C. Begot, I. Lebert and A. Lebert*

ABSTRACT

The growth of 58 strains of Listeria monocytogenes and eight strains of Listeria innocua iso­lated from meat products (68%) and industrial sites (23%), were compared in four con­ditions of temperature, water activity (aw) and pH. Temperatures ranged from 10-37 °C, pH from 5·6-7·0 and aw from 0·96-1. Growths were performed in a meat broth with an auto­mated turbidimeter (Bioscreen C, Labsystem). Growth curves were fitted using the Gom­pertz function, and growth parameters were calculated. The differences between strains in lag phase duration were much greater than in growth rate. The greatest differences occurred at 10 C, pH 7 and aw 0·96 : lag time values ranged from 4 h to 4 days. the Listeria population was separated into five groups, according to the lag time and maximal growth rate values using clustering analysis. The majority of the strains isolated from industrial sites were grouped together and showed faster growth than the others in the four con­ditions studied. The serotype or the nature of the meat from which the strains were isolated did not influence growth. The variability observed among strains raises questions about the consequences in quantitative risk assessment and about the construction of models in pre­dictive modeling.

Introduction

Listeria bacteria are widespread in the environment. Among the different species, Listeria monocytogenes and Listeria innocua are the two most commonly isolated from food processing (Cox et al. 1989). L. monocy­togenes can cause the death of both the very young and immunocompromised individuals. Most cases are traced to the contamination of raw or processed foods with L. monocyto­genes. Listeriosis is therefore a major threat to human health.

Listeria monocytogenes can grow at low temperatures (Walker et al. 1990), low pH (George et al. 1988) and low water activity (aw) (Farber et al. 1992, Nolan et al. 1992); they are therefore able to survive and multi­ply in a wide range of food products.

Work on predictive microbiology has been carried out in an attempt to improve the shelf life and safety of food (Gould 1989). Predic­tive models of microbial growths are set out with respect to the main controlling factors of the environment such as temperature, pH and water activity. Despite variations in growth among strains (Junttila et al. 1988, Walker et al. 1990) numerous models have been constructed using only one or a small pool of strains. In this work, we compared the growth of 66 strains of L. monocytogenes and L. innocua, isolated from meat, meat prod­ucts and industrials sites. Strains were clus­tered according to their growth in a broth medium in four conditions of temperature, pH and aw. Consequences on the construction of models are discussed.

Materials and Methods

Strains

Fifty-eight strains of L. monocytogenes and eight strains of L. innocua were studied: 45 strains were isolated from meat and meat products, 15 from meat and dairy plants (materials, floors, walls) four were involved in outbreaks (Table 1). Five of 13 serotypes of L. monocytogenes were represented: 1/2a, 1/2b, 1/2c, 4b, 4d. Listeria cultures were stored on tryptic soy agar (TSA, Difco, Detroit, MI, USA) slopes at 4 C.

Media

Culture inocula were prepared in a meat medium (MM): meat peptone (Merck) 10 g l 1_, yeast extract (Difco) 5 g l ^1, glucose (Merck) 5 g l ^1. The pH was adjusted to 7·0 with NaOH (Prolabo) 1 mol l ^1. The medium was autoclaved at 120 C for 20 min.

Growth experiments were carried out in a tryptic meat broth (TMB) which was steril­ized by filtration derived from tryptone meat agar (GTV5p, Fournaud et al. 1973): meat extract (Merck) 10 g l ^1, proteose peptone (Merck) 10 g l ^1, tryptone (Difco) 5 g l ^1, glu­cose 5 g l ^1. The medium was buffered with a K2HOP4-KH2PO4 (Merck), 0·1 mol l ¹ solution in proportion 1:1 (v/v) and adjusted to pH 7·0 with NaOH 1 mol l ^1. aw was set by adding NaCl (Merck) in accordance with Chirife and Resnik (1984).

Tryptone sodium chloride medium (TSC) (tryptone 1 g l ^1, NaCl 8·5 g l ^1, pH 7·0) was used for dilutions, and TSA and APT agar (Difco) were used for plate counts and agar slope cultures.

Growth

Growths were measured by optical density in an automated turbidimeter, Bioscreen C (Labsystem, Finland). All strains were inocu­lated in MM and incubated for the same period of time (17 h) at 30 C. All cultures were therefore in stationary phase, thus avo­iding disparities in the subsequent growth kinetics. Nine millilitres of TMB were inocu­lated with the subculture. The inoculum size was confirmed by plate counts. The inocu­lated TMB was dispensed aseptically in 300 µl volumes into honeycomb microplates (10 10) of the Bioscreen C. For each strain, eight successive wells of the same column were filled. The last two wells received the same volume of non-inoculated medium in order to determine the growth medium optical density (OD) and controlled any poss­ible contamination. Ws methodology is simi­lar to that used by Be got et al. (1996). Growth conditions are summarized in Table 2.

Experimental design

A Plackett Burman design (Plackett and Bur­man 1943) was used to test the effects of three factors: temperature, pH and water activity. Each factor was studied twice at two levels: high and low (Table 3). While a fac­torial design would have required eight experiments, four were sufficient with a Plackett Burman design.

Curve fitting

OD data were transferred from the Bioscreen C to Excel software (Microsoft Windows) and transformed for each measurement. Four quantities were calculated at time t: (ODi)_t, the average of the OD of eight replicates; (ODni)t, the average of the OD of the non­inoculated medium

(ΔOD)t=(ODi)t (ODni)t

Log10[(ΔOD)_t/ΔODmin]

where ΔODmin was the lowest ΔOD value above the detection threshold.

In the linear range of the Bioscreen C ( ΔOD<1'2) (Begot et al. 1996), growth curves were fitted using the modified Gompertz equation (Zwietering et al. 1990) (1):

where A is the logarithmic increase of bac­terial population, L, the lag time, µ, the maxi­mal growth rate and t, the time.

Growth parameters were determined by non-linear regression with Statitcf software (Technical Institute of Cereals and Fodders, France). Generation time (T) was derived from the maximal growth rate (2):

T = [Log10(2)]/μ

Generation times were called T001, T010, T111 and T100 according to Table 3. The same denominations were given to L, µ and A.

Statistical analysis

Two techniques were used to compare the growth of strains. Lag times (L) and maxi­mum growth rates ( ) were both considered. Average and standard deviation were calcu­lated for both variables. Each of the 66 values were then normalized: the average was sub­tracted and the result divided by the stan­dard deviation.

Each strain was depicted in a space of 2x4=8 dimensions. Principal component analy­sis (PCA) was used to reduce the dimension of the sample space. The method looks for the unique subspace of a given dimension which explains the major part of the variance in the original data. This analysis was performed using Statitcf software. A cluster analysis (Everitt 1980) was used

and calculated with Statitcf software. Classi­fication is achieved through successive aggre­gations of individuals and then groups of individuals. The principle is to regroup indi­viduals according to their distance from the centre of gravity for the set of experimental points. Two individuals or two classes were aggregated when their fusion involved a minimal increase of the intra-class disper­sion. The average value of each parameter was then calculated for each group and an overall average was calculated for the 66 strains.

Results

Table 4 shows statistics of L, µ and A for each of the four conditions. Generation time and lag time increased when temperature increased and were higher in condition 001 than in condition 100 (conditions described in Table 3). The logarithmic increase of popu­lation, A, was maximal in condition 111. Par­ameter A was higher in both conditions of pH 7·0 than those of pH 5·6.

The mean, below and above which there is an equal number of strains, was below the average value for the parameters considered in all conditions (Table 4). A proportion of 50-60% of the strains was under the average for each parameter.

Large variations in lag time and gener­ation time were noted among the strains in

all conditions. The largest differences in lag time among strains were recorded in con­dition 001. In this condition, the values were multiplied by a factor 25 from the minimum to the maximum value, taking the value from 4 h to 4 days (Table 4). Larger variations among lag times were found successively in condition 001, 100, 010 and 111. The more severe the growth conditions, the larger the number of disparities found in lag time. Fewer differences were found in generation time. In condition 100, generation times tripled from the minimal to the maximal value. In the other conditions, generation times doubled when comparing the slowest and fastest strain for each condition. Slight variations were found in parameter A which corresponds to the logarithm increase of the population density. We observed that strains exhibiting the longest generation time of the entire population did not show the maximal lag time. For example L. monocytogenes ATCC 19115 which has the longest genera­tion time in condition 001, showed a lag time of 50·8 h whereas the longest lag time was 97·9 h. Similarly, the strain exhibiting the shortest generation time did not show the shortest lag time. Results were identical in all conditions tested.

PCA

Strains were separated according to their ability to grow in each condition by using the

Figure 1. Representation in Plane 1-2 (a) and in Plane 1-3 (b) of the principal component analysis, of the five groups obtained after a cluster analysis on 66 strains of Listeria. Groups were separated according to the growth ability of the strains in conditions of temperature, aw and pH. Group 1 (X) was characterized by fast growth, group 2 (n) by slow growth, group 3 (j) by average growth, group 4 (u ) and group 5 (s) by slow growth except in condition 100 and 001 respectively.

variables L and µ. We chose three axes which explained 68% of the initial information. Axis 1 separated strains according to the values of parameters L and µ. Lag time and maximal growth rate were negatively and highly corre­lated in all conditions. Strains exhibiting fast growth (low lag time and high maximal growth rate) were isolated from strains which grew slowly (high lag time and low maximal growth rate). The second axis scaled the con­ditions from worst to best: 001-100-010-111 (see Table 3). Strains separated again accord­ing to growth conditions. Axis 3 segregated condition 001 from the others. Strains that grew well in this condition, where tempera­ture and water activity were low, were clus­tered. Fig. 1(a) and (b) show the distribution of the strains according to the three axes.

Cluster analysis

The cluster analysis divided the population into five groups as shown by Fig. 2. These groups are shown in Figs. 1(a) and (b). Each group was characterized as follows:

Group 1 (10 strains) was clearly separated from other groups. As shown in Figs. 1(a) and (b), it was comprised of strains which grew faster than the overall average in all the con­ditions studied (Table 5). The majority (66%) of the strains belonging to this group orig­inated from industrials sites (Fig. 3).

Group 2 (15 strains) was characterized by strains which grew more slowly than the overall average (Table 5), which is in oppo­sition to group 1 in Figs. 1(a) and (b). L. inno­cua was not present (Fig. 3).

Figure 2. Distribution of 66 strains of Listeria in five groups. Group 1 (fast growth) is immediately separated from the others; groups 2 and 3 (respectively low growth and average growth) are separated from groups 4 and 5 (low growth except in one condition).

Figure 3. Distribution of the 66 strains of Listeria according to their source in the five groups defined by the cluster analysis (█ Industrial strains, ▒ L. innocua, □ Epidemic strains, ▓ Others).

Group 3 (28 strains) was intermediate. Strains exhibited lag times and maximal growth rates around the overall average (Table 5). It is located at the centre of the PCA figures (Figs. 1(a) and (b)). All origins were represented (Fig. 3).

Group 4 (eight strains) clustered which grew more slowly than the overall average except for condition 100 where they grew fas­ter (Table 5). No epidemiological strains of L. monocytogenes were present in this group (Fig. 3).

Group 5 (five strains) was composed of strains which grew more slowly than the average of the strains, except for condition 001 where they grew faster (Table 5). 80% of group 5 strains were L. innocua (Fig. 3). L. monocytogenes Scott A, which has been widely used to set up models, belonged to this group.

The nature of the meat they were isolated from had no effect on the cluster analysis:

each group was composed of strains of diff­erent serotypes (Table 6).

Discussion

This study highlighted differences in growth among L. monocytogenes strains. Larger vari­ations between the 66 strains were shown in lag times rather than in generation times. Particularly wide variations were observed in condition 001 (10 C, aw 0·96, pH 7·0) where lag times extended from 4 h to 4 days. These results tally with those of Barbosa et al. (1994) who reported that among 39 strains of L. monocytogenes, lag times ranged from 1·5-2·9 days at 10 C, pH 7·2. Lag time is known to be the time required for the organ­isms to adapt their cellular components to the new environment. As a consequence, it is affected by the difference between subculture and culture condition (Buchanan and Klaw­itter 1991), the physiological state of the inoculum (McMeekin et al. 1993) and by the strain. Indeed, in our study, as we were care­ful to homogenize the conditions of the sub­cultures so as to avoid disparities in the sub­sequent growth kinetics, the differences we obtained in the growth parameters were attributed to the nature of the strain.

Such variability in lag times added to the variability in generation times characterizes strains with slow or fast growth. As the Inter­national Commission on Microbial Specifi­cation for Foods (ICMSF 1994) set the toler­ance limit in food at <100 L. monocytogenes per gram at the point of consumption, it was possible to calculate the time needed to reach this limit in the four conditions studied for both 'slow' and 'fast' strains as the inoculum varied (Table 7). The maximal difference between both strains is seven days and is reached in condition 001, when the inoculum is 1 cfu ml ^1. Such results can have great consequences in quantitative risk assess­ment. Listeriosis causes illness in certain categories of the population (immuno­compromised, old or pregnant people) but the conditions of foodborne outbreaks are still not well known. The infectious dose is supposed to be high (Van Schothorst 1996) but the minimal dose causing illness depends on the vulnerability of the person who ingested the food, the type and quality of the food, the level of the pathogen in the food and the viru­lence of the strain. As a consequence, such unknown data linked to the variability of the growth response show the difficulties than can be faced in estimating the probability of occurrence of listeriosis.

In our study, the strains were clustered according to their growth ability in the four conditions. They were not grouped round their serotype nor the nature of the meat from which they were isolated. However it is interesting to note that group 1 was in majority composed of strains isolated from industrial sites. These strains grew faster than the others in all the conditions tested. This suggests that they are accustomed to growing in a wide range of harsh conditions (e.g. high osmolarities, low temperatures). Studies have recently shown the ability of Listeria to adapt their cellular physiology to harsh environments (Beumer et al. 1994, Ko et al. 1994, Deneer et al. 1995). Most of the L. innocua studied were found in group 5, and grew more slowly than the average popu-

lation, except in condition 001 (10 C, aw 0·96, pH 7·0). These results concord with those of Barbosa et al. (1994) but not with studies where modified or selective media were used (Petran and Swanson 1993, Duh and Shaffner 1994, Macdonald and Sutherland 1994). Listeria monocytogenes Scott A was also clustered in group 5; in condition 001, it has the fastest lag time (3·9 h) and the fastest generation time (9·4 h) of all the 66 strains. These results do not tally with those of Bar­bosa et al. (1994) who studied this strain among 39 other strains and showed that it had the longest lag time at 4 and 10 C. Other strains that were the causes of epidemic out-breaks were found in different groups: L. monocytogenes OH Wisconsin was clustered in group 1 (characterized by fast growth), L. monocytogenes CDC Atlanta in group 2 (slow growth) and L. monocytogenes CA Wisconsin, in group 3 (average growth).

Such results point out the difficulties in choosing a strain to build a predictive model. Indeed numerous studies have involved the testing of one or a cocktail of three or five strains. L. monocytogenes Scott A was fre­quently used. The results of Barbosa et al. (1994) along with those presented here raise some questions as to predictive models: is it wise to construct models with only one strain? If so, which strain? That with an average, fast or slow growth? If not, how can we integrate the growth variability in mod­els? If the fastest strain were considered, the model would always predict a shorter lag time and lower growth rate than those found for the majority of strains, and would thus always provide safer growth estimates. But in this case, the industry could not support the costs involved in reducing the shelf life of a product. Furthermore, there is nothing to guarantee that a faster strain will not be found in the future. For public health safety, the slowest strains should be avoided. In order to take the variability of the strains into account, two ways are proposed: the first approach is the production of two predictive models using two strains that are charac­terized by a fast and a slow growth; this would result in giving a growth response interval. The second approach would be a model integrating first the study of a strainhaving an average behavior in a large range of experimental conditions, and second the results of this study.

We have shown the importance of studying a large number of strains of various origins before constructing models to assess strain variability. However, other factors should also be considered to produce effective mod­els, for example inoculum level, interactions between bacteria, interactive effects between strain and product as shown by Rosenow and Marth (1987) and comparison of growth in liquid media to growth on solid media.

Acknowledgements

This work was supported by four food com­panies, collaborating in a research Program UNIR (Ultra-propre Nutrition Industrie Recherche), the French Research Depart­ment and the French Agricultural Department. We are sincerely grateful to the labora­tories which provided bacterial strains.

References

Barbosa, W. B., Cabedo, L., Wederquist, H. J., Sofos, J. N. and Schmidt, G. R. (1994) Growth variation among species and strains of Listeria in culture broth. J. Food Protect. 57, 9, 765 ´-769.

Be got, C., Desnier, I., Daudin, J. D., Labadie, J. C. and Lebert, A. (1996) Recommendations for calculating growth parameters by optical density measurements. J. Microbiol. Methods 25, 225-232.

Beumer, R. R., Te Giffel, M. C., Cox, L. J., Rom-bouts, F. M. and Abee, T. (1994) Effect of exogenous proline, betaine, and carnitine on growth of Listeria monocytogenes in a minimal medium. Appl. Environ. Microbiol. 60, 4, 1359 -1363.

Buchanan, R. L. and Klawitter, A. (1991) Effect of temperature history on the growth of Listeria monocytogenes Scott A at refrigeration temperatures. Int. J. Food Microbiol. 12, 235 -246.

Chirife, J. and Resnik, S. L. (1984) Unsaturated solutions of sodium chloride as reference sources of water activity at various temperatures. J. Food Sci. 49, 1486-1488.

Cox, L. J., Kleiss, T., Cordier, J. L., Cordellana, C., Kondel, P., Perazzini, C., Beumer, R. and Siebenga, A. (1989) Listeria spp. in food pro-

cessing, non-food and domestic environments. Food Microbiol. 6, 49-61.

Deneer, H. G., Healey, V. and Boychuk, I. (1995) Reduction of exogenous ferric iron by a surface-associated ferric reductase of Listeria spp. Microbiology 141, 1985-1992.

Duh, Y. H. and Schaffner, D. W. (1993) Modelling the effect of temperature on the growth rate and lag time of Listeria innocua and Listeria monocytogenes. J. Food Protect. 56, 3, 205 -210.

Everitt, B. S. (1980) Cluster Analysis 2nd edition, London, Heineman Educational Book Ltd.

Farber, J. M., Coates, F. and Daley, E. (1992) Minimum water requirements for the growth of Listeria monocytogenes. Lett. Appl. Microbiol. 15, 103-105.

Fournyud, J., Sale P. and Valin, C. (1973) XIX Re union Europe enne des Chercheurs en viande. Paris 2-7 Septembre 1973.

George, S. M., Lund, B. M. and Brocklehurst, T. F. (1988) The effect of pH and temperature on initiation of growth of Listeria monocytogenes. Lett. Appl. Microbiol. 6, 153-156.

Gould, G. (1989) Predictive mathematical modelling of microbial growth and survival in foods. Food Sci. Technol. Today 3, 2, 89-92.

International Commission on Microbiological Specifications for Foods (ICMSF). (1994) Choice of sampling plan and criteria for Listeria monocytogenes. Int. J. Food Microbiol. 22, 89 -96.

Junttila, J. R., Niemela S. I. and Hirn, J. (1988) Minimum growth temperatures of Listeria monocytogenes and non-haemolytic Listeria. J. Appl. Bacteriol. 65, 321-327.

Ko, R., Smith. T., L. and Smith, G. M. (1994) Glycine betaine confers enhanced osmotolerance and cryotolerance on Listeria monocytogenes. J. Bacteriol. 176, 2, 426-431.

MacDonald, F. and Sutherland, A. D. (1994) Important differences between the generation times of Listeria monocytogenes and Listeria innocua in two Listeria enrichment broths. J. Dairy Res. 61, 433-436.

McMeekin, T. A., Olley, J., Ross, T. and Ratkowsky, D. A. (1993) Predictive Microbiology: Theory and Application. pp. 340. Taunton; Research Studies Press.

Nolan, D. A., Chamlin, D. C. and Troller, J. A. (1992) Minimal water activity levels for growth and survival of Listeria monocytogene and Listeria innocua. Int. J. Food Microbiol. 16, 323-335.

Petran, R. L. and Swanson, K. M. J. (1993) Simultaneous growth of Listeria monocytogenes and Listeria innocua. J. Food. Protect. 56, 7, 616 -618.

Plackett, R. L. and Burman, J. P. (1943) Design of optimum multifactorial experiments. Biometrika 33, 305-325.

Rosenow, E. M. and Marth, E. H. (1987) Growth of Listeria monocytogenes in skim, whole and chocolate milk, and in whipping cream during incubation at 4, 8, 13, 21 and 35 C. J. Food Protect. 50, 6, 452-459.

Van Schothorst, M. (1996) Setting of Criteria for Listeria monocytogenes based on risk assessment. Food Safety 96, ASEPT Editor, Laval, pp. 157-168.

Walker, S. J., Archer, P. and Banks, J. G. (1990) Growth of Listeria monocytogenes at refrigeration temperature. J. Appl. Bacteriol. 68, 157 -162.

Zwietering, M. H., Jongenburger, I., Rombouts, F. M. and Van't Riet, K. (1990) Modelling of the bacterial growth curve. Appl. Environ. Microbiol. 56, 1875-1881.

  (order Full Text from publisher)

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005