Heat Resistance of Salmonella: the Uniqueness of Salmonella senftenberg 775W

APPLIED MICROBIOLOGY, Jan. 1969, p. 78-82 Copyright ( 1969 American Society for Microbiology Vol. 17, No. 1 Printed in U.S. A Heat Resistance of Salmonella: the Uniqueness of Salmonella senftenberg 775W HENRY NG, HENRY G. BAYNE, AND JOHN A. GARIBALDI Western Regional Research Laboratory, Agricultural Research Service, U.S. Department ofagriculture, Albany, California 94710 Received for publication 17 October 1968 Of approximately 300 cultures of Salmonella, representing 75 different serotypes, none was found to be as heat-resistant as S. senftenberg 775W. However, S. blockley 2004 was 5 times more heat-resistant and S. senftenberg 775W was 30 times more heat-resistant than S. typhimurium Tm-i, the reference strain in this study. All other strains of Salmonella tested, including 19 strains of S. senftenberg and 7 strains of S. blockley, had decimal reduction times at 57 C of about 1 min, equivalent to that of the reference organism, Tm-1. As observed in other bacterial species, strain 775W is more heat-sensitive in the log phase than in the stationary phase of growth. Cells from cultures grown at 44 C were more heat-resistant than those grown at either 35 or 15 C; the medium of growth, whether minimal or complex, made no appreciable difference in heat resistance. Cells from cultures limited by a carbon source were killed at a much slower rate than those limited by a nitrogen source and exhibited a 1-hr lag at 55 C before a significant rate of kill was attained. For any given set of growth conditions, strain 775W was always more heat-resistant than another strain of S. senftenberg, 197B, which has normal heat resistance. The thermal resistance of the genus Salmonella is a matter of great concern to all persons concerned with public health and to food processors and handlers. Although numerous reports, as cited by Bayne et al. (4), indicate that salmonellae are relatively sensitive to heat, one strain of S. senftenberg, 775W, represents an exception. This organism was first mentioned by Winter et al. (19) with three other H2S-negative strains of S. senftenberg in heat resistance studies. It survived almost 5 min of heating at 60 C in liquid egg. Later Solowey, Sutton, and Calesnick (15), designating the organism for the first time as 775W, showed it to have a decimal reduction time (DRT) at 61 C of 1.1 min in liquid whole egg and 1.19 min in a tryptose broth as compared to DRT values of less than 1 min at 58 C for other Salmonella cultures. Despite its unusual heat resistance, this organism, isolated from egg powder along with 149 other Salmonella, has not been adequately described. The only mention of its taxonomic position was the serological typing by E. H. Spaulding of the School of Medicine, Temple University, Philadelphia, Pa. (19). Although extensively used in several heat resistance studies (1-3, 10, 12, 16), its identity was neither questioned nor confirmed. In view of the uniqueness and potential public health significance of S. senftenberg 775W, we will 78 present a more thorough description of its biochemical properties, ascertain the frequency of occurrence of other Salmonella equally heatresistant, and determine how variations in growth conditions, such as age of culture, growth medium, and growth temperature, affect its heat resistance and that of a strain of the same serotype having normal heat resistance. MATERIALS AND METHODS Media. Generally, the medium was Trypticase Soy Broth (BBL) supplemented with 2% (w/v) yeast extract (Difco; TSB-YE) or Trypticase Soy Agar similarly supplemented (TSA-YE). It should be noted that commercial TSA differs from TSB not only in the presence of agar but in the absence of 0.25% (w/v) dextrose and 0.25% (w/v) dipotassium phosphate. The minimal medium used was 56 (11), having the following composition (per liter): 13.6 g of KH2PO4, 2.0 g of (NH4)2SO4, 0.01 g of CaCI2, 0.2 g of MgSO4i 7H20, and 0.0005 FeSO4. 7H20. The ph was adjusted to 7.4 with NaOH. Glucose was autoclaved separately and added aseptically to the sterile medium to give a final concentration of 0.2% (w/v). In the nitrogen- and and carbon-limited studies with S. senftenberg 775W, the concentration of (NH4)SO4 and of glucose was reduced to limiting levels 0.016 and 0.05%. respectively, in the same minimal medium. Bacterial strains. The Salmonella cultures were obtained from (i) the National Communicable Disease Center, Atlanta, Ga., (ii) The Microbiology Labora-

VOL. 17, 1969 HEAT RESISTANCE OF SALMONELLA 79 tory, California Department of Public Health, Berkeley, and (iii) our own collection. These were maintained as stab cultures in Cystine Trypticase Agar (BBL) and were carried on TSA-YE slants. Growth of cultures. Cultures were generally grown in 250-ml Erlenmeyer flasks which had been modified by the addition, as sidearms, of matched 16-by 125- mm test tubes. The flasks containing 25 ml of medium were incubated at 15, 35, or 44 0.5 C on a water bath shaker (Research Specialties, Richmond, Calif.). Growth was measured by insertion of the sidearms on the flask into a Klett colorimeter equipped with a red filter (no. 66). The turbidity readings were converted to dry weight by reference to a graph relating dry weight to Klett readings. Viable counts. The number of viable cells was determined by appropriately diluting samples in TSB-YE and spread-plating a sample onto TSA-YE plates which had been allowed to dry at room temperature for about 48 hr. The plates were incubated at 37 C overnight. (Longer incubation periods did not significantly increase the counts.) Heating of cells. A 1-ml amount of cultures was introduced into milk dilution bottles containing 99 ml of TSB-YE which had been equilibrated at 55-4 0.05 C for about 0.5 hr. Samples were then removed at various intervals to determine survivors by viable counts. Plotting the log of survivors against time allows the estimation of the DRT (DRT and D value are used interchangeably), which is the time, in minutes, required to reduce the population by 90%. Screening for thermal resistance. A relatively rapid procedure was devised to screen a large number of cultures for heat resistance. It consisted of diluting 1 ml of each culture grown for 2 days at 35 C in TSB- YE into 99 ml of ice-cold medium and dispensing 2 ml of the diluted cells into each of two 13- by 100-mm screw-cap test tubes. One tube of each culture was kept on ice, whereas the other was immersed up to the cap in a bath at 57 0.05 C for 10 min. After cooling the heated tubes in an ice bath, the viable cell count was determined on the contents of both the heated and the unheated tubes. DRT values were calculated from this single time period and were compared with the value obtained by similarly treating S. typhimurium Tm-1, a strain of normal heat resistance. RESULTS Characteristics of S. senftenberg 775W. S. senftenberg 775W is a gram-negative, nonsporeforming, motile rod. It did not produce indole, H2S, or urease. When grown on Triple Sugar Iron Agar, it produced an alklaine slant with acid and gas in the butt. It did not ferment lactose, sucrose, salicin, inositol, adonitol, glycerol, or raffinose, but did ferment sorbitol and glucose. It grew in Simmon's Citrate medium but not in malonate broth. It produced a lysine decarboxylase. The cells agglutinated the group E polyvalent antiserum and the single 0 (somatic) factor 19. The flagellar antisera g, s, and t were agglutinated. These results clearly confirm the identity of the culture as S. senftenberg according to the Kaufmann-White Schema (9). Thermal resistance of some Salmonella serotypes. Two hundred and ninety-six salmonellae of approximately 75 different serotypes were screened for heat resistance and were compared to S. typhimurium Tm-1. Only results obtained on those serotypes of which 10 or more strains were tested are presented in Table 1. A minimum, a maximum, and an average ratio among strains are given. No average ratio exceeded a value of two; i.e., no serotype on the average was more than twice as resistant as S. typhimurium Tm-1. Even the most heat-resistant strain of any one serotype, e.g., one strain each of S. heidelberg, S. pullorum, S. manhattan, and S. tel-viv (the latter two not shown on the table), was only slightly more than twice as resistant as S. typhimurium Tm-1. An exception was a strain of S. blockley 2004, which was consistently about five times as resistant. Although not indicated in Table 1, six other strains of S. blockley had only an average ratio of about 1. The survey included 19 strains of S. senftenberg other than 775W, among which were both H2S-positive and -negative strains; none was significantly more heat-resistant than S. typhimurium Tm-1. The decimal reduction times for the chief strains at 57 C and ph 6.8 were: S. typhimurium Tm-i, 1.2 min; S. blockley 2004, 5.8 min; and S. senftenberg 775W, 31 min. Effect of age of culture and temperature of growth on heat resistance. All prior studies on thermal resistance of S. senftenberg 775W were carried out on relatively old cells. It is well known that with other bacteria, old cells are more heatresistant than young cells (13, 17). Therefore, it is conceivable that the extreme heat tolerance here- TABLE 1. Heat resistance of several Salmonella serotypes as compared with S. typhimurium Tm-i Serotype No. of strains tested Ratio of heat resistance to heat resistance of S. typhimurium Tm-i Mifli- - Mean Maximum mum S. dublin... 10 O.6 0.9 1.3 S. gallinarum. 11 0.7 1.0 1.7 S. heidelberg. 15 0.9 1.3 2.2 S. muenchen. 10 1.0 1.2 1.6 S. newport... 11 0.9 1.1 1.6 S. oranienburg... 14 0.7 1.1 1.8 S. pullorum... 14 0.7 1.2 2.2 S. senftenberg4... 19 0.6 1.2 1.8 S. typhimurium... 17 0.8 1.2 1.7 a Does not include S. senftenberg 775W.

80 NG, BAYNE, AND GARIBALDI APPL. MICROBIOL. 0 10 20 10 20 20 40 60 80 Minutes Minutes Minutes FIG. 1. Survival curves for S. senftenberg at 55 C. Strain 775W grown at 15 C (a), at 35 C (b), and at 44 C (c). Strain 197B grown at 15 C (d), at 35 C (e), and at 44 C (f). Symbols: 0, log-phase cells; *, stationary cells. tofore reported for S. senftenberg 775W may be mainfested only when the culture is in the stationary phase, and normal heat resistance may be exhibited in the log phase of growth. Figures la, b, and c show that, at all growth temperatures tested, S. senftenberg 775W was indeed more heat-sensitive in the log phase than in the stationary phase; however, compared to another strain, 197B (Fig. id, e, f), which has normal heat resistance, it was still more heat-resistant in both log and stationary phases. Furthermore, strain 775W was more heatresistant than 197B at all growth temperatures. In addition, the heat resistance of both strains was a function of growth temperature; the higher the temperature, the more resistant were the cells. The cells which had been grown at 44 C and had survived the heating (Fig. ic) were reisolated and allowed to grow at 35 C and again were tested for heat resistance. They were found to have a DRT characteristic of cells grown at 35 C. Thus, neither did growth of the culture at 44 C bring about a heat-resistant mutant nor were the survivors of a heat treatment any more resistant than the original cells. Effect of growth medium on heat resistance. The heat resistance of S. senftenberg strains 197B and 775W, grown for 48 hr in either a minimal medium or in the TSB-YE, is shown in Fig. 2. Strain 775W was more resistant than 197B regardless of growth medium. Cells of 775WY however, appeared to be more resistant when grown in a complex medium than in the minimal medium, a difference not demonstrable in cells of 197B, which were equally resistant in both growth media. Effect of carbon and nitrogen limitation on heat resistance. Since the foregoing experiments indicated that cells in the stationary phase of growth are more heat-resistant than those in the expo- 24 0~~~~~~ vi *0 20 40 60 80 100 120 Minutes FIG. 2. Efiect of growth medium on heat resistance of S. senftenberg. Strain 775W grown in TSB-YE (a) and minimal medium (0). Strain 197B grown in TSB- YE (0) and in minimal medium (0).

VOL. 17, 1969 5 0 N-limited >4 N-limited (S+7kr.) (S+2hr) o 3 2 0 20 40 60 80 100 120 40 60 Minutes FIG. 3. Eject of substrate limitation on heat resistance of S. senftenberg 775W. Cells starved of carbon source for 2 hr (0), and cells starved ofnitrogen for 2 hr (E) and for 7 hr (U). i HEAT RESISTANCE OF SALMONELLA nential phase, it is important to ascertain whether cells from cultures which go into the stationary phase as a result of carbon limitation are different in heat resistance from those limited by nitrogen. Figure 3 shows that cells of 775W grown at 35 C in a medium limiting in ammonium sulfate as the sole nitrogen source and those grown in a medium limiting in glucose as the sole carbon and energy source exhibited survivor curves of different shapes. The cells subjected to heating at 55 C, after having entered the stationary phase for approximately 2 hr as a result of nitrogen limitation, were killed immediately at a constant exponential rate, whereas cells at the same stage of growth limited by carbon were killed at a much slower initial rate prior to attaining the same rate as the nitrogen-limited cells. Furthermore, extending the period of nitrogen limitation to 7 hr did not alter the heat resistance characteristics of the cells. DISCUSSION The results of the present study indicate that S. senftenberg 775W is indeed an unusual and rare organism. Although the method used for screening the several hundred cultures is approximate, it appears to be reliable for the purpose of comparing heat resistance among organisms and it is simple and rapid. Several of the cultures tested in this survey had been isolated from pasteurized egg products, but they proved to be no more heatresistant than the average. It appears, therefore, that Salmonella recovered from these pasteurized products resulted from recontamination or improperly operating equipment. On the other hand, the somewhat heat-resistant culture of S. blockley 2004 was isolated from a human salmonellosis case by the California Department of Public Health. A strain of Salmonella as heat-resistant as S. senftenberg 775W was reported (6) after completion of the present work. This strain of S. senftenberg (designated strain S8 in the authors' culture collection) had heat resistance identical to, within experimental error, that of 775W. It is a coincidence that the two strains are of the same serotype. Limited tests performed in this laboratory have failed to distinguish between the two strains. The possibility exists, therefore, that they may be the same organism even though 775W was isolated from dried eggs in the United States, whereas S8 was isolated from home-killed meat in England. Although both strains of heat-resistant Salmonella are of the same serotype, our work and that of Solowey et al. (15) and Davidson et al. (6) showed that not all S. senftenberg strains are heat-resistant nor are all heat-resistant salmonellae S. senftenberg. Furthermore, the ability to produce H2S is not correlated with heat resistance. Factors which affect heat resistance of bacteria during heating should be distinguished from those effective before heating, such as growth conditions of the cells. Hansen and Riemann (8) have amply studied and reviewed the former, whereas the latter have not been studied as thoroughly. The effect of age of culture on heat resistance was first reported by Sherman and Albus (13) and has since been confirmed (17). The results of the present study unequivocally demonstrate that S. senftenberg 775W is more resistant than other strains of Salmonella, not only in the stationary growth phase, as demonstrated so often, but also in the exponential growth phase. However, as was expected, the stationary-phase cells are many times more resistant than exponential-phase cells. The medium in which the cells are grown seems to have little or no influence on their heat resistance; those grown in a complex medium may be slightly more resistant than those grown in a glucose minimal medium. The temperature at which cells are grown appears to have a larger influence on their thermal resistance. Reports concerning this variable have been conflicting. Sherman and Cameron (14) found that cells grown at lower temperatures were more heat-resistant, and Claydon, cited by Hansen and Riemann (8), observed that Streptococcus lactis exhibited greater heat resistance when grown at 10 C than when grown at higher temperatures. However, Elliker and Frazier (7) found that E. coli survived heat treatment better when grown at 38.5 or 40 C than when grown at 81

82 NG, BAYNE, AND GARIBALDI APPL. MICROBIOL. 28, 30, or 30.5 C. White (18) also found higher heat resistance in cells of Streptococcus faecalis grown at 45 C than at 27 C. Our results agree with those of Elliker and Frazier and White. A most significant observation is that heat resistance of Salmonella can vary widely, depending on strain differences, age of cell, and temperature of growth, from a low DRT of 0.55 min for 197B grown in the log phase at 15 C to a high of 42 min for 775W grown to the stationary phase at 44C. The kinetics of killing observed in our experiments were complex, and interpretation of the data is difficult. The survivor curves took on a variety of shapes. There has been much discussion and speculation on this subject, e.g., reference 5, but its causes are unknown. In our experiments, the curve representing an initial slow rate followed by a more rapid killing rate was most frequently obtained when S. senftenberg 775W was grown to a stationary phase either in a complex medium or in a minimal medium where glucose is limiting. However, when cells were grown in a minimal medium limited by nitrogen source, an immediate exponential killing rate was obtained. The lag in kill can be 1 hr or more at 55 C so that the usual explanation of come-up time for this type of multihit kinetics is inadequate since, experimentally, the heating medium was allowed to reach temperatures before cells were added. Another explanation often advanced is that cells are clumped so that many units must be inactivated before a kill is detected. The magnitude of the lag, however, would require clumps of the order of hundreds of cells, an unlikely event. At present, no explanation can be offered for the lag in killing. The other kind of kinetics usually observed with log-phase cells is one in which an initial rapid kill is followed by a slower kill. This can be explained by the presence of a heterogeneous population in which a small percentage of cells is more heat-resistant, for example, those which failed to enter the exponential phase but remained in the more resistant stationary or lag phase. Work now in progress to determine why S. senftenberg is so much more heat-resistant than the other salmonellae may reveal what factors contribute to the heat resistance of bacteria and may reveal the mechanism by which heat kills bacteria. ACKNOWLEDGMENT We gratefully acknowledge the technical assistance of Catherine Powers, Califoria State Department of Public Health, Berkeley, who performed the biochemical and serological tests on S. senftenberg 775W. LITERATURE CITED 1. Anellis, A., J. Lubas, and M. M. Rayman. 1954. Heat resistance in liquid eggs of some strains of the genus Salmonella. Food Res. 19:377-395. 2. Angelotti, R., M. J. Foter, and K. H. Lewis. 1960. Timetemperature effects on salmonellae and staphylococci in foods. II. Behavior at warm holding temperatures. Thermaldeath-time studies. Techn. Rept. F60-5, Robert A. Taft Sanitary Engineering Center, Cincinnati, Ohio. 3. Angelotti, R., M. J. Foter, and K. H. Lewis. 1961. Timetemperature effects on salmonellae and staphylococci in foods. L. Behavior in refrigerated foods. Am. J. Public Health 51:76-88. 4. Bayne, H. G., J. A. Garibaldi, and H. Lineweaver. 1965. Heat resistance of Salmonella typhimurium and Salmonella senftenberg 775W in chicken meat. Poultry Sci. 44:1281-1284. 5. Buchanan, R. E., and E. I. Fulmer. 1930. Effect of environment upon microorganisms, p. 22. Physiology and biochemistry of bacteria, vol. 2. The Williams and Wilkins Co., Baltimore. 6. Davidson, C. M., M. Boothroyd, and D. L. Georgala. 1966. Thermal resistance of Salmonella senftenberg. Nature 212: 1060-1061. 7. Elliker, P. R., and W. C. Frazier. 1938. Influence of time and temperature of incubation on heat resistance of Escherichia coli. J. Bacteriol. 36:83-98. 8. Hansen, N. H., and H. Riemann. 1963. Factors affecting the heat resistance of nonsporing organisms. J. Appl. Bacteriol. 26:314-333. 9. Kauffman, F. 1966. The bacteriology of enterobacteriaceae. The Williams and Wilkins Co., Baltimore. 10. Licciardello, J. J., J. T. Nickerson, and S. A. Goldblith. 1965. Destruction of salmonellae in hard-boiled eggs. Am. J. Public Health 55:1622-1628. 11. Monod, J., G. Cohen-Bazire, and M. Cohn. 1951. Sur la biosynthese de la,-galactosidase (lactase) chez Escherichia coli. La specificite de l'induction. Biochim. Biophys. Acta 7:585-599. 12. Osborne, W. W., R. P. Straka, and H. Lineweaver. 1954. Heat resistance of strains of Salmonella in liquid whole egg, egg yolk, and egg white. Food Res. 19:451-463. 13. Sherman, J. M., and W. R. Albus. 1923. Physiological youth in bacteria. J. Bacteriol. 8:127-139. 14. Sherman, J. M., and G. M. Cameron 1934. Rate of growth and viability in Bacterium coli. J. Bacteriol. 27:23. 15. Solowey, M., R. R. Sutton, and E. J. Calesnick. 1948. Heat resistance of Salmonella organisms isolated from spraydried whole-egg powder. Food Technol. 2:9-14. 16. Thomas, C. T., J. C. White, and K. Longree. 1966. Thermal resistance of salmonellae and staphylococci in foods. Appl. Microbiol. 14:815-820. 17. White, H. R. 1953. The heat resistance of Streptococcus faecalis. J. Gen. Microbiol. 8:27-37. 18. White, H. R. 1963. The effect of variation in ph on heat resistance of cultures of Streptococcus faecalis. J. Appl. Bacteriol. 26:91-99. 19. Winter, A. R., G. F. Stewart, V. H. McFarlane, and M. Solowey. 1946. Pasteurization of liquid egg products. III. Destruction of Salmonella in liquid whole egg. Am. J. Public Health 36:451-460.