- Email: [email protected]
Behaviours of log phase cultures of eight strains of Escherichia coli incubated at temperatures of 2, 6, 8 and 10 °C
Available online at www.sciencedirect.com
International Journal of Food Microbiology 119 (2007) 200 – 206 www.elsevier.com/locate/ijfoodmicro
Behaviours of log phase cultures of eight strains of Escherichia coli incubated at temperatures of 2, 6, 8 and 10 °C C.O. Gill ⁎, M. Badoni, T.H. Jones Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C & E Trail, Lacombe, Alberta, Canada T4L 1W1 Received 27 June 2007; accepted 23 July 2007
Abstract The behaviours of cold-adapted, log-phase cultures of eight strains of Escherichia coli incubated at 2, 6, 8 and 10 °C for 10 days were examined by determining absorbance at 600 nm (A600), viable counts and cell size distribution as indicated by forward angle light scattering (FALS) values, obtained for samples collected each day from each culture. Cell lengths were determined from photomicrographs of samples for which the flow cytometry data indicated the mean cell lengths were maximal or minimal for each culture. At 2 °C, A600 values for all strains and viable counts for some changed little, while viable counts for other strains declined progressively by N 1 log unit. At 6 °C, A600 values for most strains increased at progressively declining rates and then remained constant while viable counts increased to reach maximum values before maximum A600 values were attained, and then declined. At 8 °C, the behaviours of most strains were similar to the behaviour at 6 °C. At 10 °C, seven of the strains grew exponentially, but for most of these the growth rate determined from A600 values differed from that determined from viable count data. Mean FALS values for cultures incubated at 6, 8, or 10 °C showed various patterns of increase and decrease, indicating fluctuations in cell lengths. For all strains, the minimum cell length was b3 μm, but the maximum cell lengths ranged from b20 to N 140 μm. The findings suggest that the formation of elongated cells or filaments is usual behaviour for E. coli growing at temperatures approaching or below the minimum for sustained growth. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved Keywords: Escherichia coli; Minimum temperature for growth; Filamentation; Growth rate estimation
1. Introduction The maximum temperatures regarded as safe for holding raw meats and other unpreserved foods have been decided largely by reference to the minimum temperature for growth of Escherichia coli, the broadly recognized indicator organism for mesophilic enteric pathogens such as Salmonella. The accepted minimum temperature of growth of E. coli is 7 °C (Cassin et al., 1998). However, it has been shown that log phase cultures of a strain of E. coli can grow at temperatures less than 7 °C with the formation of filaments, for limited times at constant temperatures but possibly indefinitely when temperatures periodically fluctuate above 7 °C (Jones et al., 2004); and with the formation of filaments during sustained growth at temperatures up to 12 °C (Jones et al., 2003).
⁎ Corresponding author. Tel.: +1 403 782 8113; fax: +1 403 782 6120. E-mail address: [email protected] (C.O. Gill).
If the formation of filaments by E. coli growing at temperatures approaching or below the supposed minimum for its growth is usual, then growth of the organism in chilled foods may be commonly underestimated by either direct determinations of numbers before and after periods of chiller storage, or on the basis of predictive models that take no account of such behaviour by the organism. Apart from the recent studies with a single strain of E. coli, there has been only one other report of filament formation by E. coli at chiller temperatures (Shaw, 1968). There have been two reports of filament formation by Salmonella incubated at refrigeration temperatures. In one study with six strains of Salmonella it was found that all formed filaments at 8 °C (Mattick et al., 2003), but in the other study only one of two strains formed filaments at 4 °C (Phillips et al., 1998). The information provided in the few available reports is then insufficient for deciding whether filament formation at temperatures likely to be experienced by chilled foods is usual among strains of E. coli and related organisms, or if such behaviour is peculiar to a few strains only. Therefore, for better
0168-1605/$ - see front matter. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved doi:10.1016/j.ijfoodmicro.2007.07.043
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206 Table 1 Strains of Escherichia coli used in the study
201
2. Materials and methods
were counted on filters bearing between 10 and 1000 such squares, and a most probable number of viable counts (cfu/ml) was obtained from that count. A 5 ml portion of each culture was centrifuged at 8000 ×g for 10 min. The pellet was washed by resuspension in 5 ml of 0.9% (w/v) saline solution, and after pelleting again the washed cells were resuspended in 4 ml of saline solution. Two 1 ml portions of the suspension were each mixed with 0.1 ml of 37% (w/v) formaldehyde solution (Fisher Scientific, Edmonton, Alberta, Canada). The preparations of fixed cells were stored at − 80 °C until they were examined by flow cytometry or microscopy. After sampling of a culture, when the A600 value was near to or exceeded 0.5, the culture was used to inoculate a fresh flask containing BHI equilibrated to the incubation temperature to obtain a further culture of A600 about 0.05. The A600 value was recorded and a sample was withdrawn for enumeration of viable counts in the new culture. On subsequent days samples were drawn from only the new culture.
2.1. Cultivation and incubation of E. coli strains
2.3. Flow cytometry
The strains of E. coli used in the study were the wild type (WT) strain isolated from meat that had been used in previous studies of filamentation (Jones et al., 2002, 2003, 2004), and seven American Type Culture Collection (ATCC) strains, including reference strains and strains used for testing microbiological and other products (Table 1). All strains were maintained in cooked meat medium (Difco, Becton Dickinson, Sparks, MD, USA) and were cultivated in half strength brain heart infusion (BHI; Difco). Stirred cultures grown to the stationary phase at 25 °C were used to inoculate 250 ml flasks containing 100 ml of BHI, which were incubated with stirring at 15 °C until values for optical absorbance at 600 nm (A600) of about 0.5 were attained. The 15 °C cultures were then used to inoculate further flasks containing BHI that had been equilibrated to 15 °C to obtain cultures of log phase cells with A600 values about 0.05 for cultures that were to be incubated at 6, 8 or 10 °C, or about 0.2 for cultures that were to be incubated at 2 °C. Immediately after inoculation, each log phase culture was placed in a water bath maintained within + 0.1 °C of a designated incubation temperature. Cultures were incubated with stirring for 10 days.
Thawed preparations of fixed cells were diluted with a solution containing brain heart infusion and NaCl at 1.85 and 8.8 g/l, respectively, to obtain a suspension containing about 106 cells/ml. A 0.5 ml portion of each suspension was mixed with 0.5 μl of a nucleic acid stain (SYTOBC bacterial stain; Molecular Probes, Eugene, OR, USA), and the mixture was held for 5 min at room temperature before it was loaded into a FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA). Forward angle light scattering (FALS) and fluorescence data were obtained. The FALS of 10,000 cells gated on fluorescence were analysed using Cell Quest software (Becton Dickinson). The 90th percentile for FALS measurements when incubation was initiated was the reference value for each culture.
2.2. Sampling of cultures
Table 2 Differences from initial values of the final values for log absorbance at 600 nm (log A600) and log numbers of viable counts (log cfu/ml) determined for cultures of Escherichia coli incubated at 2 °C for 10 days
Strain
Applications
Wild type (WT)
Used in previous studies of filament formation at low temperatures. Used for testing disinfectants, membrane filters, absorbent pads, paint resistance, etc. Type strain used for quality control testing of API products. K-12 Hfr lambda negative strain used for genetic research. Serotype 086a: K61 used for experimental induction of septic shock. Reference strain from US adult. Reference strain from Swedish adult. Serotype 0157:H7 used for quality control testing of Difco products.
ATCC 11229 ATCC 11775 ATCC 23739 ATCC 33985 ATCC 35327 ATCC 35328 ATCC 43895
understanding of the matter, the behaviours of eight strains of E. coli at temperatures around 7 °C were examined.
Samples were collected 1 h after the inoculation of each flask incubated at ≤10 °C and then at daily intervals. On each day, two 1 ml portions of each culture were used for determination of the A600 value by means of a spectrophotometer (Ultra Spec III, Pharmacia LITB Biotechnology, Uppsala, Sweden). Two other 1 ml portions were each diluted 10-fold, and each 10-fold dilution was used to prepare serial 100-fold dilutions. A 10 ml portion of each of the dilutions 10− 3, 10− 5 and 10− 7 was filtered through a hydrophobic grid membrane filter. Each filter was placed on a plate of lactose monensin glucuronate agar (LMG; Oxoid, Mississauga, Ontario, Canada) which was incubated at 35 °C for 24 h. Squares containing blue colonies
2.4. Microscopy The lengths of cells in samples from selected cultures were determined by microscopy. A drop of molten 2% agar was added to 10 μl of a fixed preparation of each selected culture on a microscope slide. The agar was covered with a cover slip, and the slide was stored at 2 °C for 24 h. The slides were viewed
Initial value – Final value
Strain a
WT 11229 11775 23739 33985 35327 35328 43895 a
log A600
log cfu/ml
− 0.03 − 0.14 − 0.02 +0.06 0.00 +0.04 − 0.02 − 0.03
− 0.32 − 0.22 − 0.06 +0.01 − 1.30 +0.13 − 1.48 − 1.18
Wild type (WT) or ATCC strain.
202
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
Table 3 Differences from initial values of the maximum and final values for log absorbance at 600 nm (log A600) and log numbers of viable counts (log cfu/ml), and the days on which maximum or near maximum values were approached or attained, for cultures of Escherichia coli incubated at 6 or 8 °C for 10 days
Table 5 The ratios of the minimum, maximum and final mean forward angle light scattering (FALS) values to the initial mean FALS value, and the days on which the minimum and maximum values were attained, for cultures of Escherichia coli incubated at 6, 8 or 10 °C for 10 days
Temperature Strain a Difference from initial value (°C)
Temperature Strain a FALS value/initial value Day value attained (°C) Minimum Maximum Final Minimum Maximum
Day maximum value approached b or attained
log cfu/ml
log A600
Maximum Final Maximum Final value value value value 6
WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895
8
a b
1.00 0.46 0.62 1.05 1.37 1.03 0.75 0.60 1.39 0.38 0.57 1.72 2.41 1.70 1.26 0.95
0.99 0.34 0.62 1.05 1.37 1.03 0.75 0.60 1.22 0.01 0.57 1.72 2.41 1.70 1.26 0.95
0.78 0.49 0.96 0.57 0.62 0.76 0.61 0.63 0.58 0.08 0.53 0.75 2.32 0.81 1.40 1.42
− 1.98 − 0.81 0.49 0.40 0.62 0.45 0.04 − 0.11 − 1.46 − 1.87 − 0.22 − 0.17 1.86 0.32 1.36 0.89
5 4 5b 8b 10 8b 8b 4b 7 4 6b 9b 9 10 10 6
2 3 4b 4 2b 6 6 3 2 1 4 3 4 5 9 7
under phase-contrast illumination using a microscope (Axioscope; Zeiss, Jena, Germany) connected to a video camera (DXC 930; Sony, Tokyo, Japan). The length of 100 cells on randomly selected images from each slide were measured with a sensitivity of 0.0001 μm, using Image Pro-Plus software, version 4 (Media Cybernetics, Silverspring, MD, USA). 2.5. Analysis of data For cultures that were prepared from an initial culture during the incubation period, A600 values and viable counts from Table 4 Apparent sustained growth rates (generations/day) determined from increases in absorbance at 600 nm (A600) or of viable counts for cultures of Escherichia coli incubated at 10 °C for 10 days
WT 11229 11775 23739 33985 35327 35328 43895 a b
Growth rate (generations/day)
(Rate A600 − Rate cfu) / Rate A600
From A600 values
From viable counts
(%)
1.76 –b 1.46 1.83 1.89 1.79 1.66 1.56
1.56 – 1.69 1.49 1.43 1.86 1.56 2.19
11 – − 16 19 24 −4 6 − 40
Wild type (WT) or ATCC strains. Growth was not sustained for 10 days.
WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895
log log A600 cfu/ml
Wild type (WT) or ATCC strain. Value increased byb0.06 log unit after this time.
Strain a
6
8
10
a
0.5 0.6 1.0 1.0 1.0 0.7 1.0 1.0 1.0 0.3 0.5 0.8 1.0 0.6 0.4 0.5 1.0 0.6 0.4 1.0 1.0 1.0 0.5 0.6
1.5 3.7 1.6 5.0 8.5 2.5 1.8 1.8 4.2 1.5 0.8 5.3 5.8 5.9 1.1 1.1 4.3 3.9 1.1 3.0 2.9 1.8 1.0 1.1
1.2 2.4 1.4 5.0 3.8 2.5 1.8 1.8 3.6 1.1 0.8 4.1 3.8 4.8 0.9 1.1 1.7 2.1 0.6 2.4 2.9 1.2 0.6 0.8
3 2 0 0 0 1 0 0 0 7 2 1 1 1 2 1 0 8 4 0 0 0 4 5
6 7 6 10 9 7 10 5 8 3 6 9 4 7 8 8 6 3 3 2 10 2 0 2
Wild type (WT) or ATCC strain.
samples from the new culture were converted to values compatible with those from the original culture. That was done by multiplying the values for samples from the new culture by the ratio of the appropriate values for the initial and the new culture obtained at the time the new culture was prepared. Increases in the mean values for duplicated log A600 and log viable count values, as observed or estimated for the initial cultures, were plotted against the times of incubation. Growth rates for cultures incubated at 10 °C were determined from the regression lines
Table 6 Minimum, maximum and mean lengths of cells in cultures of eight strains of Escherichia coli incubated at 6, 8, or 10 °C at times when the ratio of the forward angle light scattering (FALS) value to the initial FALS value was a minimum or a maximum for all the cultures of each strain Strain a Minimum ratio of FALS values
Maximum ratio of FALS values
Ratio of Cell length (μm) Ratio of Cell length (μm) FALS values FALS values Min Max Mean Min Max Mean WT 11229 11775 23739 33985 35327 35328 43895 a
0.5 0.3 0.4 0.8 1.0 0.6 0.4 0.5
1.6 1.0 2.0 2.1 2.5 2.0 2.1 1.6
5.3 2.1 6.6 17.4 9.4 11.9 16.6 14.8
Wild type (WT) or ATCC strain.
3.2 1.4 3.6 4.0 4.9 3.6 4.5 4.2
4.3 3.9 1.6 5.3 8.5 5.9 1.8 1.8
3.8 4.6 2.0 3.8 2.2 9.5 2.0 2.2
86.3 122.5 11.0 145.0 26.7 116.1 15.1 7.6
12.9 26.3 3.8 44.0 10.4 27.3 4.8 3.8
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
for such plots. Mean lengths and the mean lengths of the longest 10% of cells in selected cultures were calculated. Plotting and calculations were performed using Microsoft Excel 97 (Microsoft Corp., Redmond, WA, USA). 3. Results For most strains, the viable counts were about 7 log cfu/ml when cultures were prepared with A600 values about 0.05. However, for strain 11229, the viable counts obtained at such times were about 5 log cfu/ml. When cultures were incubated at 2 °C, the A600 values determined for each culture at all times were little different from the initial value (Table 2). For strains WT, 11229, 11775, 23739
203
and 35327 the viable counts determined for each culture at all times also differed little from the initial value. However, for the other strains, the viable counts declined progressively, by N 1 log unit after 10 days. When cultures were incubated at 6 °C the A600 values for most increased at progressively declining rates for 4 or more days and then were little different at subsequent times (Table 3). The viable counts for those cultures also increased initially, to reach maximum numbers before maximum A600 values were attained. The viable counts subsequently declined. The numbers after 10 days were mostly less than 1 log unit less than the maximum numbers, but for strains WT and 11229 the final numbers were more than 2 and more than 1 log unit less than the respective maximum values. Exceptionally, for strain 33985,
Fig. 1. Cells of Escherichia coli ATCC 11229 after incubation at (A) 8 °C for 7 days or (B) 10 °C for 3 days; E. coli ATCC 23739 after incubation at (C) 8 °C for 1 day or (D) 8 °C for 9 days; and E. coli ATCC 43895 after incubation at (E) 8 °C for 1 day or (F) 6 °C for 5 days. Size bars are 5 μm.
204
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
A600 values increased throughout the period of incubation and viable counts did not decline after near maximum numbers were attained on day 2 of incubation. When cultures were incubated at 8 °C, the A600 values for all initially increased, apparently logarithmically for the first 4 or more days, but then each rate of increase declined, progressively or abruptly. For most cultures the A600 values remained close to the maximum value after values ceased increasing (Table 3). However, the A600 values for cultures of strains 35327 and 35328 were still increasing after incubation for 10 days, while the values for strain 11229 decreased logarithmically after day 6. At 8 °C, as at 6 °C, the viable counts increased initially and most then declined with maximum values being attained before A600 values were maximal; and the viable count decreases of about 2 log units for strains WT and 11229 were again greater than those for other strains. When cultures were incubated at 10 °C, the A600 values for strain 11229 increased at a decreasing rate to attain a maximum increase of 1.13 log unit after 5 days, with the final value being 1.12 log units more than the initial value. The viable counts attained a near maximum increase of 0.32 log units after 4 days, with the final value being 0.39 log unit (data not shown). All other strains apparently grew exponentially (R2 N 0.89) during incubation at 10 °C. For strains 35327 and 35328, the rates of growth determined from A600 or viable count values differed by b6% (Table 4). However, for each of strains WT, 23739 and 33985, the rate of growth determined from A600 values was N10% more than the rate determined from viable counts; while for each of strains 11775 and 43895 the rate determined from A600 values was N 10% less than that determined from viable counts. For most cultures incubated at 6 °C, the mean FALS values increased from an initial minimum value (Table 5). However, for strains WT, 11229 and 35327, mean FALS values first declined and then increased. For strains 23739 and 35328, mean FALS values progressively increased for 10 days; but for the other strains maximum values were attained between 5 and 9 days and subsequently declined. When cultures were incubated at 8 °C, the mean FALS values for strains 23739, 33985, 35327 and 35328 decreased initially, increased after 1 or 2 days, and then decreased from the maximum value attained (Table 5). For strain WT, however, there was no initial decrease; and for strains 11775 and 43895 there was no decrease from the maximum value attained by each strain. For strain 11229 the mean FALS values increased initially and subsequently declined. When cultures were incubated at 10 °C, the mean FALS values for most strains increased initially and then decreased (Table 5). However, the mean FALS values for strain 33985 progressively increased up to 10 days, while the values for strain 35328 decreased initially then increased. The increases for strains WT, 11229, 23739 and 33985 were relatively large, and the final values were substantially above the minimum, while for the other strains the increases were relatively small and the final values were close to the minimum. For all strains, in samples that gave minimum values for the ratio of the sample FALS value to the initial value for a culture,
the minimum cell lengths observed were b3 μm (Table 6). In most of those samples the mean cell length was about or less than twice the minimum cell length, but the maximum cell lengths ranged from N2 to N 8 times the minimum cell length for a sample. In samples that gave the maximum values for FALS value ratios, the minimum cell lengths ranged from about 2 to about 10 μm. The ranges for most mean and maximum cell lengths were from about 2 to about 12 and from about 7 to about 38 times the minimum cell length for a sample, respectively. For all strains, differences in morphologies of cells in samples that gave maximal or minimal values for the ratios of sample to initial FALS values were apparent. Examples of such differences are shown in Fig. 1; for strain 11229, which formed unusually small cells under some and long filaments under other conditions; for strain 23739, the behaviour of which might be somewhat typical of that of strains that show extensive filamentation; and for strain 43895 which apparently formed few and short filaments only. 4. Discussion The ATCC strains were selected with a view to examining the behaviours of strains that originated from a variety of sources. Strain WT was included as control, to establish that under the growth conditions used its behaviour was consistent with the behaviour observed in previous studies with the organism. The organisms were cultivated at 15 °C before inoculation to media that would be incubated at lower temperatures to obtain cells that would be adapted to growth at low temperatures (Berry and Foegeding, 1997) but at which strain WT does not, and other strains would likely not form filaments (Jones et al., 2003). However, strain 11229 did form filaments at 15 °C, and consequently the viable counts were low in relation to the A600 values when cultures of the strains were prepared for incubation at temperatures less than 15 °C. When cultures were incubated at 2 °C, the behaviours of most of the strains were comparable with the behaviour previously observed for strain WT, with no substantial changes in A600 values but with progressive declines in the numbers of colonies recovered on a LMG, which is a medium that allows resuscitation of injured cells (Jones et al., 2002). However, two of the strains showed no reductions in viable counts while among the other strains the reductions in viable counts varied greatly. In a study of changes in the protein profiles of strain WT when cold adapted cells were incubated at temperatures of 8, 6 or 2 °C little change was found in the protein profiles of cells incubated at 2 °C for 24 h although large changes occurred in the protein profiles of cells incubated at the higher temperatures (Jones et al., 2006). Examination of the protein profiles of other strains incubated at 2 °C might then provide indication of a possible reason for the wide range of rates of inactivation among, and the absence of inactivation with some strains of E. coli when growth of log phase cells is totally inhibited by temperature alone. When cultures were incubated at 6 °C the behaviours of all strains were similar to the previously observed behaviour of strain WT (Jones et al., 2002), with A600 values increasing over
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
several days to maximum values that were then maintained while viable counts increased to maximum values at earlier times and then declined. However, the declines in viable counts were less for all strains than for strain WT, for which the decline in viable counts was far greater at 6 °C than at 2 °C. In contrast, there was no decline in viable counts for strain 33985 although at 2 °C the decline in viable counts had been substantial. Thus, there appears to be no consistent relationship between the rates at which cells of different strains were inactivated at 2 and 6 °C. When cultures were incubated at 8 °C the behaviours of all strains were similar to the behaviours at 6 °C, although some strains attained markedly greater maximum viable counts and/or A600 values than at 6° C; and some attained maximum values after different times than at 6 °C. The similar behaviours at 6 and 8 °C was unexpected as it is generally accepted that if the growth medium is maintained in a condition that favours growth, exponential growth of E. coli can continue indefinitely at temperatures above 7 °C (Cassin et al., 1998), while in previous studies with strain WT incubation at 8 °C was not extended beyond 4 days on the assumption that the apparently exponential growth observed during that time would continue in sequential subcultures. These findings suggest that defining a single minimum temperature for growth of strains of species of bacteria like E. coli may be misleading. It may instead be appropriate to identify at least two temperatures that define the limiting lower temperatures for growth. These would be the minimum temperature at which exponential growth can be maintained indefinitely with continuous culture or repeated subculture of log phase cells, and the maximum temperature at which growth of log phase cells will cease without obvious delay. The former temperature might be difficult to define precisely because of uncertainty about the time required for decision as to whether or not exponential growth could be indefinitely maintained. However, for the strains of E. coli that were examined, and possibly for E. coli generally, the minimum temperature for sustained growth is apparently N8 °C. It was N 10 °C for strain 11229 which behaved similarly at 8 and 10 °C. The latter temperature for all strains examined is apparently b 6 °C. It may be lower for at least some strains, as growth of strain WT at 4 °C and of a strain ATCC 23716 at 5 °C has been observed (Gill and Phillips, 1985; Jones et al., 2002). In addition, it might be appropriate to define the minimum temperature at which stationary phase cells can initiate growth. Values for such temperatures may emerge from studies aimed at the development of models for the duration of lag phases at growth permitting temperatures (Métris et al., 2005). When cultures of strains other than 11229 were incubated at 10 °C, their growth appeared to be exponential and sustained. However, the rates of growth estimated for each strain from A600 values or viable counts differed in all cases. That microbial growth parameters estimated by optical methods or by viable counts can differ is well established (Dalgaard et al., 1994). Usually it has been reported that growth rates estimated from optical absorbance data are lower than rates estimated from viable counts (Dalgaard and Koutsoumanis, 2001). That has been ascribed to the loss of viability of some fraction of the cells
205
in growing cultures (Francois et al., 2005). The non-viable cells would be detected by optical absorbance and would perturb the increases determined by that method, but determination of changes in viable counts would not be affected by the presence of a progressively increasing fraction of non-viable cells. In contrast, extensive filamentation would result in increases of optical absorbance without corresponding increases in viable counts. Those two mechanisms seem to account for the observed differences in growth rates determined by the two methods, as the rate determined from viable counts was N10% greater than that determined from A600 values in each of two strains that did not show extensive filamentation at 10 °C, while the reverse was the case for three strains which showed extensive filamentation at that temperature. The FALS data showed that some strains had a greater tendency to form filaments than others, with strains variously showing substantial cell elongation at 6, 8 and 10 °C, modest elongation at all three temperatures, substantial elongation at the lowest temperature but not at the higher temperatures, or substantial elongation at the higher temperature but not at the lowest temperature. Thus, no common pattern of filament formation was apparent. The minimum lengths observed for cells of all strains were similar, as would be expected, as most were within the range 1.5–2.5 μm. However, the maximum lengths observed for cells varied greatly, as might be expected from the FALS data, with maximum lengths being b 20 μm for some but N140 μm for other strains. Fluctuations in the FALS data over time indicate that substantial fractions of, if not all cells that have grown by elongation ultimately divide to give decreases in the average lengths of cell populations. These findings suggest that neither optical absorbance nor viable count data are wholly satisfactory for estimating growth of E. coli and related organisms at chiller temperatures. Absorbance data will tend to underestimate growth because of the presence of possibly variable fractions of viable cells, while viable count data will tend to underestimate the increase in bacterial mass because of filamentation. Under conditions favourable for growth, filaments of both E. coli and Salmonella can rapidly divide into numerous daughter cells (Jones et al., 2002; Mattick et al., 2003). Thus, if pathogens such as E. coli 0157:H7 or Salmonella can be present in foods as populations of cells of highly variable sizes, the health risks posed by the pathogens may be related more to their cell masses than their numbers. Evidently, the possibility of mesophilic pathogens growing as filaments in chilled foods should be investigated, to ensure that risks from such organisms in chilled foods are properly assessed. Filament formation by E. coli and other bacteria exposed to stresses from various environmental factors other than temperature have been reported (Jensen and Woolfolk, 1985; Wainwright et al., 1999; Everis and Betts, 2001; Bereksi et al., 2002; Kieboom et al., 2006). Filamentation may then be a usual response to environmental stress for at least some strains of E. coli, and other stresses at levels that are marginal for growth may give rise to ranges of growth responses among strains of E. coli similar to those that occur at low temperatures. However, the information
206
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
currently available is very limited, so further studies will be required to determine whether or not that is the case. Acknowledgements We thank Mss M.H. Ronsbo and M. Kamstrup for their assistance with the collection of optical absorbance and viable count data, and with the preparation of samples for flow cytrometry and microscopy. References Bereksi, N., Gavini, F., Bénézech, T., Faille, C., 2002. Growth, morphology and surface properties of Listeria monocytogenes Scott A and L028 under saline and acid environments. Journal of Applied Microbiology 92, 556–565. Berry, E.D., Foegeding, P.M., 1997. Cold temperature adaptations and growth of microorganisms. Journal of Food Protection 60, 1583–1594. Cassin, M.H., Lammerding, A.M., Todd, E.D.C., Ross, W., McColl, R.S., 1998. Quantitative risk assessment for Escherichia coli 0157:H7 in ground beef hamburgers. International Journal of Food Microbiology 41, 21–44. Dalgaard, P., Koutsoumanis, K., 2001. Comparison of maximum specific growth rates and lag times estimated from absorbance and viable count data by different mathematical models. Journal of Microbiological Methods 43, 183–196. Dalgaard, P., Ross, T., Kamperman, L., Neumeyer, K., Mc Meekin, T.A., 1994. Estimation of bacterial growth rates from turbimetric and viable count data. International Journal of Food Microbiology 23, 391–404. Everis, L., Betts, G., 2001. pH stress can cause cell elongation in Bacillus and Clostridium species. Food Control 12, 53–56. Francois, K., Devlieghere, F., Standaert, A.R., Geeraerd, A.H., Cools, I., Van Impe, J.F., Debevere, J., 2005. Environmental factors influencing the relationship between optical density and cell count for Listeria monocytogenes. Journal of Applied Microbiology 99, 1503–1515. Gill, C.O., Phillips, D.M., 1985. The effect of media composition on the relationship between temperature and growth rate of Escherichia coli. Food Microbiology 2, 285–290.
Jensen, H.R., Woolfolk, C.A., 1985. Formation of filaments by Pseudomonas putida. Applied and Environmental Microbiology 50, 364–372. Jones, T., Gill, C.O., McMullen, L.M., 2002. The behaviour of log phase Escherichia coli at temperature below the minimum for sustained growth. Food Microbiology 19, 83–90. Jones, T., Gill, C.O., McMullen, L.M., 2003. The behaviour of log phase Escherichia coli at temperatures near the minimum for growth. International Journal of Food Microbiology 88, 55–61. Jones, T., Gill, C.O., McMullen, L.M., 2004. The behaviour of log phase Escherichia coli at temperatures that fluctuate about the minimum for growth. Letters in Applied Microbiology 39, 296–300. Jones, T.H., Murray, A., Johns, M., Gill, C.O., McMullen, L.M., 2006. Differential expression of proteins in cold-adapted log-phase cultures of Escherichia coli incubated at 8, 6, or 2 °C. International Journal of Food Microbiology 107, 12–19. Kieboom, J., Kusumanignrum, H.D., Tempelaars, M.H., Hazeleger, W.C., Abee, T., Beumer, R.R., 2006. Survival, elongation, and elevated tolerance of Salmonella enterica serovar Enteritidis at reduced water activity. Journal of Food Protection 69, 2681–2686. Mattick, K.L., Phillips, L.E., Jǿrgensen, F., Lappin-Scott, H.M., Humphrey, T.J., 2003. Filament formation by Salmonella spp. inoculated into liquid food matrices at refrigeration temperatures, and growth patterns when warmed. Journal of Food Protection 66, 215–219. Métris, A., Le Marc, Y., Elfwing, A., Ballagi, A., Baranyi, J., 2005. Modelling the variability of lag times and the first generation times of single cells of E. coli. International Journal of Food Microbiology 100, 13–19. Phillips, L.E., Humphrey, T.J., Lappin-Scott, H.M., 1998. Chilling invokes different morphologies in two Salmonella enteritidis PT4 strains. Journal of Applied Microbiology 84, 820–826. Shaw, M., 1968. Formation of filaments and synthesis of macromolecules at temperatures below the minimum for growth of Escherichia coli. Journal of Bacteriology 95, 221–230. Wainwright, M., Canham, L.T., Al-Wajeeh, K., Reeves, C.L., 1999. Morphological changes (including filamentation) in Escherichia coli grown under starvation conditions on silicon wafers and other surfaces. Letters in Applied Microbiology 29, 224–227.
International Journal of Food Microbiology 119 (2007) 200 – 206 www.elsevier.com/locate/ijfoodmicro
Behaviours of log phase cultures of eight strains of Escherichia coli incubated at temperatures of 2, 6, 8 and 10 °C C.O. Gill ⁎, M. Badoni, T.H. Jones Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C & E Trail, Lacombe, Alberta, Canada T4L 1W1 Received 27 June 2007; accepted 23 July 2007
Abstract The behaviours of cold-adapted, log-phase cultures of eight strains of Escherichia coli incubated at 2, 6, 8 and 10 °C for 10 days were examined by determining absorbance at 600 nm (A600), viable counts and cell size distribution as indicated by forward angle light scattering (FALS) values, obtained for samples collected each day from each culture. Cell lengths were determined from photomicrographs of samples for which the flow cytometry data indicated the mean cell lengths were maximal or minimal for each culture. At 2 °C, A600 values for all strains and viable counts for some changed little, while viable counts for other strains declined progressively by N 1 log unit. At 6 °C, A600 values for most strains increased at progressively declining rates and then remained constant while viable counts increased to reach maximum values before maximum A600 values were attained, and then declined. At 8 °C, the behaviours of most strains were similar to the behaviour at 6 °C. At 10 °C, seven of the strains grew exponentially, but for most of these the growth rate determined from A600 values differed from that determined from viable count data. Mean FALS values for cultures incubated at 6, 8, or 10 °C showed various patterns of increase and decrease, indicating fluctuations in cell lengths. For all strains, the minimum cell length was b3 μm, but the maximum cell lengths ranged from b20 to N 140 μm. The findings suggest that the formation of elongated cells or filaments is usual behaviour for E. coli growing at temperatures approaching or below the minimum for sustained growth. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved Keywords: Escherichia coli; Minimum temperature for growth; Filamentation; Growth rate estimation
1. Introduction The maximum temperatures regarded as safe for holding raw meats and other unpreserved foods have been decided largely by reference to the minimum temperature for growth of Escherichia coli, the broadly recognized indicator organism for mesophilic enteric pathogens such as Salmonella. The accepted minimum temperature of growth of E. coli is 7 °C (Cassin et al., 1998). However, it has been shown that log phase cultures of a strain of E. coli can grow at temperatures less than 7 °C with the formation of filaments, for limited times at constant temperatures but possibly indefinitely when temperatures periodically fluctuate above 7 °C (Jones et al., 2004); and with the formation of filaments during sustained growth at temperatures up to 12 °C (Jones et al., 2003).
⁎ Corresponding author. Tel.: +1 403 782 8113; fax: +1 403 782 6120. E-mail address: [email protected] (C.O. Gill).
If the formation of filaments by E. coli growing at temperatures approaching or below the supposed minimum for its growth is usual, then growth of the organism in chilled foods may be commonly underestimated by either direct determinations of numbers before and after periods of chiller storage, or on the basis of predictive models that take no account of such behaviour by the organism. Apart from the recent studies with a single strain of E. coli, there has been only one other report of filament formation by E. coli at chiller temperatures (Shaw, 1968). There have been two reports of filament formation by Salmonella incubated at refrigeration temperatures. In one study with six strains of Salmonella it was found that all formed filaments at 8 °C (Mattick et al., 2003), but in the other study only one of two strains formed filaments at 4 °C (Phillips et al., 1998). The information provided in the few available reports is then insufficient for deciding whether filament formation at temperatures likely to be experienced by chilled foods is usual among strains of E. coli and related organisms, or if such behaviour is peculiar to a few strains only. Therefore, for better
0168-1605/$ - see front matter. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved doi:10.1016/j.ijfoodmicro.2007.07.043
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206 Table 1 Strains of Escherichia coli used in the study
201
2. Materials and methods
were counted on filters bearing between 10 and 1000 such squares, and a most probable number of viable counts (cfu/ml) was obtained from that count. A 5 ml portion of each culture was centrifuged at 8000 ×g for 10 min. The pellet was washed by resuspension in 5 ml of 0.9% (w/v) saline solution, and after pelleting again the washed cells were resuspended in 4 ml of saline solution. Two 1 ml portions of the suspension were each mixed with 0.1 ml of 37% (w/v) formaldehyde solution (Fisher Scientific, Edmonton, Alberta, Canada). The preparations of fixed cells were stored at − 80 °C until they were examined by flow cytometry or microscopy. After sampling of a culture, when the A600 value was near to or exceeded 0.5, the culture was used to inoculate a fresh flask containing BHI equilibrated to the incubation temperature to obtain a further culture of A600 about 0.05. The A600 value was recorded and a sample was withdrawn for enumeration of viable counts in the new culture. On subsequent days samples were drawn from only the new culture.
2.1. Cultivation and incubation of E. coli strains
2.3. Flow cytometry
The strains of E. coli used in the study were the wild type (WT) strain isolated from meat that had been used in previous studies of filamentation (Jones et al., 2002, 2003, 2004), and seven American Type Culture Collection (ATCC) strains, including reference strains and strains used for testing microbiological and other products (Table 1). All strains were maintained in cooked meat medium (Difco, Becton Dickinson, Sparks, MD, USA) and were cultivated in half strength brain heart infusion (BHI; Difco). Stirred cultures grown to the stationary phase at 25 °C were used to inoculate 250 ml flasks containing 100 ml of BHI, which were incubated with stirring at 15 °C until values for optical absorbance at 600 nm (A600) of about 0.5 were attained. The 15 °C cultures were then used to inoculate further flasks containing BHI that had been equilibrated to 15 °C to obtain cultures of log phase cells with A600 values about 0.05 for cultures that were to be incubated at 6, 8 or 10 °C, or about 0.2 for cultures that were to be incubated at 2 °C. Immediately after inoculation, each log phase culture was placed in a water bath maintained within + 0.1 °C of a designated incubation temperature. Cultures were incubated with stirring for 10 days.
Thawed preparations of fixed cells were diluted with a solution containing brain heart infusion and NaCl at 1.85 and 8.8 g/l, respectively, to obtain a suspension containing about 106 cells/ml. A 0.5 ml portion of each suspension was mixed with 0.5 μl of a nucleic acid stain (SYTOBC bacterial stain; Molecular Probes, Eugene, OR, USA), and the mixture was held for 5 min at room temperature before it was loaded into a FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA). Forward angle light scattering (FALS) and fluorescence data were obtained. The FALS of 10,000 cells gated on fluorescence were analysed using Cell Quest software (Becton Dickinson). The 90th percentile for FALS measurements when incubation was initiated was the reference value for each culture.
2.2. Sampling of cultures
Table 2 Differences from initial values of the final values for log absorbance at 600 nm (log A600) and log numbers of viable counts (log cfu/ml) determined for cultures of Escherichia coli incubated at 2 °C for 10 days
Strain
Applications
Wild type (WT)
Used in previous studies of filament formation at low temperatures. Used for testing disinfectants, membrane filters, absorbent pads, paint resistance, etc. Type strain used for quality control testing of API products. K-12 Hfr lambda negative strain used for genetic research. Serotype 086a: K61 used for experimental induction of septic shock. Reference strain from US adult. Reference strain from Swedish adult. Serotype 0157:H7 used for quality control testing of Difco products.
ATCC 11229 ATCC 11775 ATCC 23739 ATCC 33985 ATCC 35327 ATCC 35328 ATCC 43895
understanding of the matter, the behaviours of eight strains of E. coli at temperatures around 7 °C were examined.
Samples were collected 1 h after the inoculation of each flask incubated at ≤10 °C and then at daily intervals. On each day, two 1 ml portions of each culture were used for determination of the A600 value by means of a spectrophotometer (Ultra Spec III, Pharmacia LITB Biotechnology, Uppsala, Sweden). Two other 1 ml portions were each diluted 10-fold, and each 10-fold dilution was used to prepare serial 100-fold dilutions. A 10 ml portion of each of the dilutions 10− 3, 10− 5 and 10− 7 was filtered through a hydrophobic grid membrane filter. Each filter was placed on a plate of lactose monensin glucuronate agar (LMG; Oxoid, Mississauga, Ontario, Canada) which was incubated at 35 °C for 24 h. Squares containing blue colonies
2.4. Microscopy The lengths of cells in samples from selected cultures were determined by microscopy. A drop of molten 2% agar was added to 10 μl of a fixed preparation of each selected culture on a microscope slide. The agar was covered with a cover slip, and the slide was stored at 2 °C for 24 h. The slides were viewed
Initial value – Final value
Strain a
WT 11229 11775 23739 33985 35327 35328 43895 a
log A600
log cfu/ml
− 0.03 − 0.14 − 0.02 +0.06 0.00 +0.04 − 0.02 − 0.03
− 0.32 − 0.22 − 0.06 +0.01 − 1.30 +0.13 − 1.48 − 1.18
Wild type (WT) or ATCC strain.
202
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
Table 3 Differences from initial values of the maximum and final values for log absorbance at 600 nm (log A600) and log numbers of viable counts (log cfu/ml), and the days on which maximum or near maximum values were approached or attained, for cultures of Escherichia coli incubated at 6 or 8 °C for 10 days
Table 5 The ratios of the minimum, maximum and final mean forward angle light scattering (FALS) values to the initial mean FALS value, and the days on which the minimum and maximum values were attained, for cultures of Escherichia coli incubated at 6, 8 or 10 °C for 10 days
Temperature Strain a Difference from initial value (°C)
Temperature Strain a FALS value/initial value Day value attained (°C) Minimum Maximum Final Minimum Maximum
Day maximum value approached b or attained
log cfu/ml
log A600
Maximum Final Maximum Final value value value value 6
WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895
8
a b
1.00 0.46 0.62 1.05 1.37 1.03 0.75 0.60 1.39 0.38 0.57 1.72 2.41 1.70 1.26 0.95
0.99 0.34 0.62 1.05 1.37 1.03 0.75 0.60 1.22 0.01 0.57 1.72 2.41 1.70 1.26 0.95
0.78 0.49 0.96 0.57 0.62 0.76 0.61 0.63 0.58 0.08 0.53 0.75 2.32 0.81 1.40 1.42
− 1.98 − 0.81 0.49 0.40 0.62 0.45 0.04 − 0.11 − 1.46 − 1.87 − 0.22 − 0.17 1.86 0.32 1.36 0.89
5 4 5b 8b 10 8b 8b 4b 7 4 6b 9b 9 10 10 6
2 3 4b 4 2b 6 6 3 2 1 4 3 4 5 9 7
under phase-contrast illumination using a microscope (Axioscope; Zeiss, Jena, Germany) connected to a video camera (DXC 930; Sony, Tokyo, Japan). The length of 100 cells on randomly selected images from each slide were measured with a sensitivity of 0.0001 μm, using Image Pro-Plus software, version 4 (Media Cybernetics, Silverspring, MD, USA). 2.5. Analysis of data For cultures that were prepared from an initial culture during the incubation period, A600 values and viable counts from Table 4 Apparent sustained growth rates (generations/day) determined from increases in absorbance at 600 nm (A600) or of viable counts for cultures of Escherichia coli incubated at 10 °C for 10 days
WT 11229 11775 23739 33985 35327 35328 43895 a b
Growth rate (generations/day)
(Rate A600 − Rate cfu) / Rate A600
From A600 values
From viable counts
(%)
1.76 –b 1.46 1.83 1.89 1.79 1.66 1.56
1.56 – 1.69 1.49 1.43 1.86 1.56 2.19
11 – − 16 19 24 −4 6 − 40
Wild type (WT) or ATCC strains. Growth was not sustained for 10 days.
WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895 WT 11229 11775 23739 33985 35327 35328 43895
log log A600 cfu/ml
Wild type (WT) or ATCC strain. Value increased byb0.06 log unit after this time.
Strain a
6
8
10
a
0.5 0.6 1.0 1.0 1.0 0.7 1.0 1.0 1.0 0.3 0.5 0.8 1.0 0.6 0.4 0.5 1.0 0.6 0.4 1.0 1.0 1.0 0.5 0.6
1.5 3.7 1.6 5.0 8.5 2.5 1.8 1.8 4.2 1.5 0.8 5.3 5.8 5.9 1.1 1.1 4.3 3.9 1.1 3.0 2.9 1.8 1.0 1.1
1.2 2.4 1.4 5.0 3.8 2.5 1.8 1.8 3.6 1.1 0.8 4.1 3.8 4.8 0.9 1.1 1.7 2.1 0.6 2.4 2.9 1.2 0.6 0.8
3 2 0 0 0 1 0 0 0 7 2 1 1 1 2 1 0 8 4 0 0 0 4 5
6 7 6 10 9 7 10 5 8 3 6 9 4 7 8 8 6 3 3 2 10 2 0 2
Wild type (WT) or ATCC strain.
samples from the new culture were converted to values compatible with those from the original culture. That was done by multiplying the values for samples from the new culture by the ratio of the appropriate values for the initial and the new culture obtained at the time the new culture was prepared. Increases in the mean values for duplicated log A600 and log viable count values, as observed or estimated for the initial cultures, were plotted against the times of incubation. Growth rates for cultures incubated at 10 °C were determined from the regression lines
Table 6 Minimum, maximum and mean lengths of cells in cultures of eight strains of Escherichia coli incubated at 6, 8, or 10 °C at times when the ratio of the forward angle light scattering (FALS) value to the initial FALS value was a minimum or a maximum for all the cultures of each strain Strain a Minimum ratio of FALS values
Maximum ratio of FALS values
Ratio of Cell length (μm) Ratio of Cell length (μm) FALS values FALS values Min Max Mean Min Max Mean WT 11229 11775 23739 33985 35327 35328 43895 a
0.5 0.3 0.4 0.8 1.0 0.6 0.4 0.5
1.6 1.0 2.0 2.1 2.5 2.0 2.1 1.6
5.3 2.1 6.6 17.4 9.4 11.9 16.6 14.8
Wild type (WT) or ATCC strain.
3.2 1.4 3.6 4.0 4.9 3.6 4.5 4.2
4.3 3.9 1.6 5.3 8.5 5.9 1.8 1.8
3.8 4.6 2.0 3.8 2.2 9.5 2.0 2.2
86.3 122.5 11.0 145.0 26.7 116.1 15.1 7.6
12.9 26.3 3.8 44.0 10.4 27.3 4.8 3.8
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
for such plots. Mean lengths and the mean lengths of the longest 10% of cells in selected cultures were calculated. Plotting and calculations were performed using Microsoft Excel 97 (Microsoft Corp., Redmond, WA, USA). 3. Results For most strains, the viable counts were about 7 log cfu/ml when cultures were prepared with A600 values about 0.05. However, for strain 11229, the viable counts obtained at such times were about 5 log cfu/ml. When cultures were incubated at 2 °C, the A600 values determined for each culture at all times were little different from the initial value (Table 2). For strains WT, 11229, 11775, 23739
203
and 35327 the viable counts determined for each culture at all times also differed little from the initial value. However, for the other strains, the viable counts declined progressively, by N 1 log unit after 10 days. When cultures were incubated at 6 °C the A600 values for most increased at progressively declining rates for 4 or more days and then were little different at subsequent times (Table 3). The viable counts for those cultures also increased initially, to reach maximum numbers before maximum A600 values were attained. The viable counts subsequently declined. The numbers after 10 days were mostly less than 1 log unit less than the maximum numbers, but for strains WT and 11229 the final numbers were more than 2 and more than 1 log unit less than the respective maximum values. Exceptionally, for strain 33985,
Fig. 1. Cells of Escherichia coli ATCC 11229 after incubation at (A) 8 °C for 7 days or (B) 10 °C for 3 days; E. coli ATCC 23739 after incubation at (C) 8 °C for 1 day or (D) 8 °C for 9 days; and E. coli ATCC 43895 after incubation at (E) 8 °C for 1 day or (F) 6 °C for 5 days. Size bars are 5 μm.
204
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
A600 values increased throughout the period of incubation and viable counts did not decline after near maximum numbers were attained on day 2 of incubation. When cultures were incubated at 8 °C, the A600 values for all initially increased, apparently logarithmically for the first 4 or more days, but then each rate of increase declined, progressively or abruptly. For most cultures the A600 values remained close to the maximum value after values ceased increasing (Table 3). However, the A600 values for cultures of strains 35327 and 35328 were still increasing after incubation for 10 days, while the values for strain 11229 decreased logarithmically after day 6. At 8 °C, as at 6 °C, the viable counts increased initially and most then declined with maximum values being attained before A600 values were maximal; and the viable count decreases of about 2 log units for strains WT and 11229 were again greater than those for other strains. When cultures were incubated at 10 °C, the A600 values for strain 11229 increased at a decreasing rate to attain a maximum increase of 1.13 log unit after 5 days, with the final value being 1.12 log units more than the initial value. The viable counts attained a near maximum increase of 0.32 log units after 4 days, with the final value being 0.39 log unit (data not shown). All other strains apparently grew exponentially (R2 N 0.89) during incubation at 10 °C. For strains 35327 and 35328, the rates of growth determined from A600 or viable count values differed by b6% (Table 4). However, for each of strains WT, 23739 and 33985, the rate of growth determined from A600 values was N10% more than the rate determined from viable counts; while for each of strains 11775 and 43895 the rate determined from A600 values was N 10% less than that determined from viable counts. For most cultures incubated at 6 °C, the mean FALS values increased from an initial minimum value (Table 5). However, for strains WT, 11229 and 35327, mean FALS values first declined and then increased. For strains 23739 and 35328, mean FALS values progressively increased for 10 days; but for the other strains maximum values were attained between 5 and 9 days and subsequently declined. When cultures were incubated at 8 °C, the mean FALS values for strains 23739, 33985, 35327 and 35328 decreased initially, increased after 1 or 2 days, and then decreased from the maximum value attained (Table 5). For strain WT, however, there was no initial decrease; and for strains 11775 and 43895 there was no decrease from the maximum value attained by each strain. For strain 11229 the mean FALS values increased initially and subsequently declined. When cultures were incubated at 10 °C, the mean FALS values for most strains increased initially and then decreased (Table 5). However, the mean FALS values for strain 33985 progressively increased up to 10 days, while the values for strain 35328 decreased initially then increased. The increases for strains WT, 11229, 23739 and 33985 were relatively large, and the final values were substantially above the minimum, while for the other strains the increases were relatively small and the final values were close to the minimum. For all strains, in samples that gave minimum values for the ratio of the sample FALS value to the initial value for a culture,
the minimum cell lengths observed were b3 μm (Table 6). In most of those samples the mean cell length was about or less than twice the minimum cell length, but the maximum cell lengths ranged from N2 to N 8 times the minimum cell length for a sample. In samples that gave the maximum values for FALS value ratios, the minimum cell lengths ranged from about 2 to about 10 μm. The ranges for most mean and maximum cell lengths were from about 2 to about 12 and from about 7 to about 38 times the minimum cell length for a sample, respectively. For all strains, differences in morphologies of cells in samples that gave maximal or minimal values for the ratios of sample to initial FALS values were apparent. Examples of such differences are shown in Fig. 1; for strain 11229, which formed unusually small cells under some and long filaments under other conditions; for strain 23739, the behaviour of which might be somewhat typical of that of strains that show extensive filamentation; and for strain 43895 which apparently formed few and short filaments only. 4. Discussion The ATCC strains were selected with a view to examining the behaviours of strains that originated from a variety of sources. Strain WT was included as control, to establish that under the growth conditions used its behaviour was consistent with the behaviour observed in previous studies with the organism. The organisms were cultivated at 15 °C before inoculation to media that would be incubated at lower temperatures to obtain cells that would be adapted to growth at low temperatures (Berry and Foegeding, 1997) but at which strain WT does not, and other strains would likely not form filaments (Jones et al., 2003). However, strain 11229 did form filaments at 15 °C, and consequently the viable counts were low in relation to the A600 values when cultures of the strains were prepared for incubation at temperatures less than 15 °C. When cultures were incubated at 2 °C, the behaviours of most of the strains were comparable with the behaviour previously observed for strain WT, with no substantial changes in A600 values but with progressive declines in the numbers of colonies recovered on a LMG, which is a medium that allows resuscitation of injured cells (Jones et al., 2002). However, two of the strains showed no reductions in viable counts while among the other strains the reductions in viable counts varied greatly. In a study of changes in the protein profiles of strain WT when cold adapted cells were incubated at temperatures of 8, 6 or 2 °C little change was found in the protein profiles of cells incubated at 2 °C for 24 h although large changes occurred in the protein profiles of cells incubated at the higher temperatures (Jones et al., 2006). Examination of the protein profiles of other strains incubated at 2 °C might then provide indication of a possible reason for the wide range of rates of inactivation among, and the absence of inactivation with some strains of E. coli when growth of log phase cells is totally inhibited by temperature alone. When cultures were incubated at 6 °C the behaviours of all strains were similar to the previously observed behaviour of strain WT (Jones et al., 2002), with A600 values increasing over
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
several days to maximum values that were then maintained while viable counts increased to maximum values at earlier times and then declined. However, the declines in viable counts were less for all strains than for strain WT, for which the decline in viable counts was far greater at 6 °C than at 2 °C. In contrast, there was no decline in viable counts for strain 33985 although at 2 °C the decline in viable counts had been substantial. Thus, there appears to be no consistent relationship between the rates at which cells of different strains were inactivated at 2 and 6 °C. When cultures were incubated at 8 °C the behaviours of all strains were similar to the behaviours at 6 °C, although some strains attained markedly greater maximum viable counts and/or A600 values than at 6° C; and some attained maximum values after different times than at 6 °C. The similar behaviours at 6 and 8 °C was unexpected as it is generally accepted that if the growth medium is maintained in a condition that favours growth, exponential growth of E. coli can continue indefinitely at temperatures above 7 °C (Cassin et al., 1998), while in previous studies with strain WT incubation at 8 °C was not extended beyond 4 days on the assumption that the apparently exponential growth observed during that time would continue in sequential subcultures. These findings suggest that defining a single minimum temperature for growth of strains of species of bacteria like E. coli may be misleading. It may instead be appropriate to identify at least two temperatures that define the limiting lower temperatures for growth. These would be the minimum temperature at which exponential growth can be maintained indefinitely with continuous culture or repeated subculture of log phase cells, and the maximum temperature at which growth of log phase cells will cease without obvious delay. The former temperature might be difficult to define precisely because of uncertainty about the time required for decision as to whether or not exponential growth could be indefinitely maintained. However, for the strains of E. coli that were examined, and possibly for E. coli generally, the minimum temperature for sustained growth is apparently N8 °C. It was N 10 °C for strain 11229 which behaved similarly at 8 and 10 °C. The latter temperature for all strains examined is apparently b 6 °C. It may be lower for at least some strains, as growth of strain WT at 4 °C and of a strain ATCC 23716 at 5 °C has been observed (Gill and Phillips, 1985; Jones et al., 2002). In addition, it might be appropriate to define the minimum temperature at which stationary phase cells can initiate growth. Values for such temperatures may emerge from studies aimed at the development of models for the duration of lag phases at growth permitting temperatures (Métris et al., 2005). When cultures of strains other than 11229 were incubated at 10 °C, their growth appeared to be exponential and sustained. However, the rates of growth estimated for each strain from A600 values or viable counts differed in all cases. That microbial growth parameters estimated by optical methods or by viable counts can differ is well established (Dalgaard et al., 1994). Usually it has been reported that growth rates estimated from optical absorbance data are lower than rates estimated from viable counts (Dalgaard and Koutsoumanis, 2001). That has been ascribed to the loss of viability of some fraction of the cells
205
in growing cultures (Francois et al., 2005). The non-viable cells would be detected by optical absorbance and would perturb the increases determined by that method, but determination of changes in viable counts would not be affected by the presence of a progressively increasing fraction of non-viable cells. In contrast, extensive filamentation would result in increases of optical absorbance without corresponding increases in viable counts. Those two mechanisms seem to account for the observed differences in growth rates determined by the two methods, as the rate determined from viable counts was N10% greater than that determined from A600 values in each of two strains that did not show extensive filamentation at 10 °C, while the reverse was the case for three strains which showed extensive filamentation at that temperature. The FALS data showed that some strains had a greater tendency to form filaments than others, with strains variously showing substantial cell elongation at 6, 8 and 10 °C, modest elongation at all three temperatures, substantial elongation at the lowest temperature but not at the higher temperatures, or substantial elongation at the higher temperature but not at the lowest temperature. Thus, no common pattern of filament formation was apparent. The minimum lengths observed for cells of all strains were similar, as would be expected, as most were within the range 1.5–2.5 μm. However, the maximum lengths observed for cells varied greatly, as might be expected from the FALS data, with maximum lengths being b 20 μm for some but N140 μm for other strains. Fluctuations in the FALS data over time indicate that substantial fractions of, if not all cells that have grown by elongation ultimately divide to give decreases in the average lengths of cell populations. These findings suggest that neither optical absorbance nor viable count data are wholly satisfactory for estimating growth of E. coli and related organisms at chiller temperatures. Absorbance data will tend to underestimate growth because of the presence of possibly variable fractions of viable cells, while viable count data will tend to underestimate the increase in bacterial mass because of filamentation. Under conditions favourable for growth, filaments of both E. coli and Salmonella can rapidly divide into numerous daughter cells (Jones et al., 2002; Mattick et al., 2003). Thus, if pathogens such as E. coli 0157:H7 or Salmonella can be present in foods as populations of cells of highly variable sizes, the health risks posed by the pathogens may be related more to their cell masses than their numbers. Evidently, the possibility of mesophilic pathogens growing as filaments in chilled foods should be investigated, to ensure that risks from such organisms in chilled foods are properly assessed. Filament formation by E. coli and other bacteria exposed to stresses from various environmental factors other than temperature have been reported (Jensen and Woolfolk, 1985; Wainwright et al., 1999; Everis and Betts, 2001; Bereksi et al., 2002; Kieboom et al., 2006). Filamentation may then be a usual response to environmental stress for at least some strains of E. coli, and other stresses at levels that are marginal for growth may give rise to ranges of growth responses among strains of E. coli similar to those that occur at low temperatures. However, the information
206
C.O. Gill et al. / International Journal of Food Microbiology 119 (2007) 200–206
currently available is very limited, so further studies will be required to determine whether or not that is the case. Acknowledgements We thank Mss M.H. Ronsbo and M. Kamstrup for their assistance with the collection of optical absorbance and viable count data, and with the preparation of samples for flow cytrometry and microscopy. References Bereksi, N., Gavini, F., Bénézech, T., Faille, C., 2002. Growth, morphology and surface properties of Listeria monocytogenes Scott A and L028 under saline and acid environments. Journal of Applied Microbiology 92, 556–565. Berry, E.D., Foegeding, P.M., 1997. Cold temperature adaptations and growth of microorganisms. Journal of Food Protection 60, 1583–1594. Cassin, M.H., Lammerding, A.M., Todd, E.D.C., Ross, W., McColl, R.S., 1998. Quantitative risk assessment for Escherichia coli 0157:H7 in ground beef hamburgers. International Journal of Food Microbiology 41, 21–44. Dalgaard, P., Koutsoumanis, K., 2001. Comparison of maximum specific growth rates and lag times estimated from absorbance and viable count data by different mathematical models. Journal of Microbiological Methods 43, 183–196. Dalgaard, P., Ross, T., Kamperman, L., Neumeyer, K., Mc Meekin, T.A., 1994. Estimation of bacterial growth rates from turbimetric and viable count data. International Journal of Food Microbiology 23, 391–404. Everis, L., Betts, G., 2001. pH stress can cause cell elongation in Bacillus and Clostridium species. Food Control 12, 53–56. Francois, K., Devlieghere, F., Standaert, A.R., Geeraerd, A.H., Cools, I., Van Impe, J.F., Debevere, J., 2005. Environmental factors influencing the relationship between optical density and cell count for Listeria monocytogenes. Journal of Applied Microbiology 99, 1503–1515. Gill, C.O., Phillips, D.M., 1985. The effect of media composition on the relationship between temperature and growth rate of Escherichia coli. Food Microbiology 2, 285–290.
Jensen, H.R., Woolfolk, C.A., 1985. Formation of filaments by Pseudomonas putida. Applied and Environmental Microbiology 50, 364–372. Jones, T., Gill, C.O., McMullen, L.M., 2002. The behaviour of log phase Escherichia coli at temperature below the minimum for sustained growth. Food Microbiology 19, 83–90. Jones, T., Gill, C.O., McMullen, L.M., 2003. The behaviour of log phase Escherichia coli at temperatures near the minimum for growth. International Journal of Food Microbiology 88, 55–61. Jones, T., Gill, C.O., McMullen, L.M., 2004. The behaviour of log phase Escherichia coli at temperatures that fluctuate about the minimum for growth. Letters in Applied Microbiology 39, 296–300. Jones, T.H., Murray, A., Johns, M., Gill, C.O., McMullen, L.M., 2006. Differential expression of proteins in cold-adapted log-phase cultures of Escherichia coli incubated at 8, 6, or 2 °C. International Journal of Food Microbiology 107, 12–19. Kieboom, J., Kusumanignrum, H.D., Tempelaars, M.H., Hazeleger, W.C., Abee, T., Beumer, R.R., 2006. Survival, elongation, and elevated tolerance of Salmonella enterica serovar Enteritidis at reduced water activity. Journal of Food Protection 69, 2681–2686. Mattick, K.L., Phillips, L.E., Jǿrgensen, F., Lappin-Scott, H.M., Humphrey, T.J., 2003. Filament formation by Salmonella spp. inoculated into liquid food matrices at refrigeration temperatures, and growth patterns when warmed. Journal of Food Protection 66, 215–219. Métris, A., Le Marc, Y., Elfwing, A., Ballagi, A., Baranyi, J., 2005. Modelling the variability of lag times and the first generation times of single cells of E. coli. International Journal of Food Microbiology 100, 13–19. Phillips, L.E., Humphrey, T.J., Lappin-Scott, H.M., 1998. Chilling invokes different morphologies in two Salmonella enteritidis PT4 strains. Journal of Applied Microbiology 84, 820–826. Shaw, M., 1968. Formation of filaments and synthesis of macromolecules at temperatures below the minimum for growth of Escherichia coli. Journal of Bacteriology 95, 221–230. Wainwright, M., Canham, L.T., Al-Wajeeh, K., Reeves, C.L., 1999. Morphological changes (including filamentation) in Escherichia coli grown under starvation conditions on silicon wafers and other surfaces. Letters in Applied Microbiology 29, 224–227.