Richard K. Gast 1 and Peter S. Holt

Influence of the Level and Location of Contamination on the Multiplication of Salmonella enteritidis at Different Storage Temperatures in Experimentally Inoculated Eggs Richard K. Gast 1 and Peter S. Holt Southeast Poultry Research Laboratory, USDA-Agriculture Research Service, 934 College Station Road, Athens, Georgia 30605 ABSTRACT Prompt refrigeration to temperatures capable peratures able to prevent further microbial growth in of restricting microbial growth has been recom- mended as an approach to reducing the likelihood that contaminated eggs will transmit Salmonella enteritidis to humans. By using experimentally contaminated egg components, the present study determined the extent to which small numbers of S. enteritidis could grow to more dangerous levels at different temperatures over a period up to 3 d. This model was intended to simulate the potential opportunities for S. enteritidis multiplication following oviposition and prior to the achievement of internal temeggs. At a relatively warmer incubation temperature (25 C) and with higher inoculum doses (150 cells), rapid and substantial S. enteritidis multiplication often occurred, especially when the bacteria had an opportunity for access to yolk nutrients and when contaminated eggs were incubated for 2 or 3 d before sampling. Extensive multiplication of S. enteritidis was less frequently observed at lower inoculum doses (15 cells), shorter storage times (1 d), and lower temperatures (10 to 17.5 C) and when contaminants were introduced into the albumen. (Key words: Salmonella enteritidis, eggs, yolk, albumen, temperature) 2000 Poultry Science 79:559 563 INTRODUCTION During the past two decades, Salmonella enteritidis infections transmitted by internally contaminated eggs have emerged among the leading food-borne disease issues (Centers for Disease Control, 1996; Tauxe, 1997; Angulo and Swerdlow, 1999). Efforts to control this problem have focused both on reducing the incidence of S. enteritidis infections in egg-laying chickens (to lessen the likelihood that contaminated eggs will be produced) and on minimizing opportunities for egg contamination to cause human disease outbreaks (Hogue et al., 1998). In this latter context, the USDA recently announced plans to amend its regulations to require that shell eggs be stored and transported under refrigeration at an ambient temperature no higher than 7 C (US Department of Agriculture, 1998). Such policies are intended to prevent small numbers of S. enteritidis deposited in egg contents from multiplying to levels more likely to survive light cooking and to cause illness when consumed. Because growth-restricting internal temperatures cannot be achieved immediately in eggs, the prospective effectiveness of refrigeration will be contingent on the number of contaminants present, Received for publication June 14, 1999. Accepted for publication December 8, 1999. 1 To whom correspondence should be addressed: rkgast@ arches.uga.edu. the location of these organisms, and the length of time that the egg contents remain at temperatures warm enough to support bacterial growth. Evidence obtained from naturally infected commercial laying flocks suggests that contaminated eggs are normally produced infrequently, at incidences typically less than 0.03% (Schlosser et al., 1995; Kinde et al., 1996). However, experimental infection studies have also shown that much higher frequencies of egg contamination are possible (Gast and Beard, 1990; Bichler et al., 1996). Most naturally contaminated eggs have been reported to harbor very low levels of S. enteritidis cells, often fewer than 10 cells per egg (Humphrey et al., 1989), although occasional eggs containing much larger bacterial numbers have also been found (Humphrey et al., 1991). Even in eggs from hens experimentally infected with very large oral doses of S. enteritidis, contamination levels in freshly laid eggs were generally less than 10 cells/ml (Gast and Beard, 1992a). Both yolk and albumen have been identified as potential sites for internal S. enteritidis contamination of eggs (Gast and Beard, 1990; Humphrey et al. 1991). Although several components of albumen can inhibit bacterial multiplication (Stephenson et al., 1991; Baron et al., 1997), movement of S. enteritidis to reach the yolk can allow rapid growth to begin (Hammack et al., 1993; Humphrey and Whitehead, 1993; Braun and Fehlhaber, 1995). Multiplication of S. enteritidis in egg yolk proceeds briskly at 559

560 GAST AND HOLT temperatures above 20 C, but is very slow below 10 C, and becomes negligible below 7 C (Kim et al., 1989; Clay and Board, 1991; Saeed and Koons, 1993). At intermediate temperatures (near 15 C), steady bacterial growth occurs but at fairly long generation intervals (Bradshaw et al., 1990; Humphrey, 1990). Considerably more persistent survival of S. enteritidis in albumen has been reported at temperatures of 20 C or higher than at 4 C (Lock and Board, 1992). The present study sought to assess the potential for small numbers of S. enteritidis to reach more dangerous levels during the interval between oviposition and the achievement, via refrigeration, of growth-restricting internal temperatures in eggs. This objective was addressed by monitoring the multiplication of S. enteritidis inoculated into egg components at different locations and held at various temperatures. MATERIALS AND METHODS Preparation of S. enteritidis Cultures In each trial, frozen storage beads containing a phage type-8 strain of S. enteritidis were resuscitated and prepared for use by two cycles of incubation for 24 h at 37 C in tryptone soya broth. 2 This culture was centrifuged for 10 min at 3,000 g to concentrate cells, washed with 0.85% saline, centrifuged again, and then resuspended in saline. After the cell density of the resuspended culture was estimated by determining its optical density at 600 nm, further dilution in saline produced cultures containing final target cell densities of 1,500 cfu/ml and 150 cfu/ml (confirmed by subsequent plate counts). Preparation, Inoculation, and Incubation of Egg Contents Samples In each of six trials, freshly collected eggs from a specific pathogen-free, Single Comb White Leghorn flock of our laboratory were held at room temperature overnight before use. These eggs were aseptically broken, and their contents (sometimes after separating the yolk and albumen) were transferred into sterile plastic beakers. Four types of samples were prepared and inoculated with S. enteritidis. Each yolk sample consisted of a single yolk (inoculated internally). Each albumen sample consisted of albumen from a single egg (inoculated internally). Each whole egg (albumen edge) sample consisted of an unseparated liquid egg inoculated at the edge of the albumen furthest from the yolk. Each whole egg (yolk surface) sample was prepared by inoculating S. enteritidis onto the exterior surface of a single separated yolk, holding this inoculated yolk at room temperature for 5 min and then 2 Unipath Co., Oxoid Division, Ogdensburg, NY 13669. 3 Tekmar Co., Cincinnati, OH 45222. 4 Difco Laboratories, Detroit, MI 48232. 5 Sigma Chemical Co., St. Louis, MO 63178-9916. 6 GraphPad Software, San Diego, CA 92121. gently pouring the albumen from a single egg back into the beaker to surround the yolk. Twenty-four samples of each type were prepared in each trial. Half of the samples of each type were inoculated with 0.1-mL doses containing 15 cfu of S. enteritidis, and the other half were inoculated with similar doses containing 150 cfu. All samples were incubated at 25 C in two trials, at 17.5 C in two trials, and at 10 C in the remaining two trials. Enumeration of S. enteritidis in Egg Contents Samples After Incubation After 1, 2, and 3 d of incubation in each trial, one-third of the samples of each type was removed to enumerate S. enteritidis. Each sample (of yolk, albumen, or both together) was transferred to a sterile plastic bag and mixed for 30 s in a Stomacher Model 80 Lab Blender. 3 The number of S. enteritidis colony-forming units present was determined by making serial ten-fold dilutions of each sample in saline and spreading aliquots of each dilution onto plates of brilliant green agar 4 supplemented with 0.02 mg/ml novobiocin. 5 The agar plates were incubated for 24 h at 37 C, and S. enteritidis colonies were identified (Waltman et al., 1998) and counted. Because the detection threshold of this enumeration procedure was 100 cfu/ ml, any sample with no countable colonies was arbitrarily assigned a value of 50 cfu/ml. Statistical Analysis Significant (P < 0.05) differences were determined between the mean numbers of S. enteritidis found in incubated egg contents samples contaminated with different initial dose levels by application of an unpaired twotailed t-test. Significant (P < 0.05) differences between the mean numbers of S. enteritidis found in different types of incubated egg samples were determined by one-way analysis of variance followed by a Tukey-Kramer multiple-comparison test. All statistical analyses were conducted using Instat biostatistics software. 6 RESULTS The number of S. enteritidis cells in yolk samples increased rapidly during incubation at 25 C (Figure 1), reaching mean log 10 levels of 8.4 and 8.7/mL at 2 d in samples initially contaminated with doses of 15 and 150 cfu, respectively. The mean log 10 levels of S. enteritidis in whole egg (yolk surface) samples rose to only 4.3/mL (15 cfu dose) and 6.1/mL (150 cfu dose) at 2 d. In whole egg (albumen edge) samples, very little multiplication of S. enteritidis was seen at either dose (samples contaminated initially with 150 cfu reached a mean log 10 level of only 2.6/mL after 3 d). For albumen samples, no multiplication of S. enteritidis was detected at the 15-cfu dose and almost none at the 150-cfu dose (reaching only a mean log 10 level of 1.9/mL at 3 d). The mean level of S. enteritidis found in yolk samples after each of the 3 d of incubation at 25 C was significantly greater than from any other type

MULTIPLICATION OF SALMONELLA ENTERITIDIS IN EGGS 561 C (Figure 2), attaining final mean log 10 levels of 8.0/mL (15-cfu dose) and 8.2/mL (150-cfu dose) at 3 d. The S. enteritidis in whole egg (yolk surface) samples increased more slowly, reaching mean log 10 levels of 2.7/mL for the 15-cfu dose and 5.3/mL for the 150-cfu dose at 3 d. For whole egg (albumen edge) samples, relatively moderate growth of S. enteritidis (to mean log 10 levels of 2.0/mL for the 15-cfu dose and 2.3/mL for the 150-cfu dose) was detected only after 3 d of incubation. Little evidence of multiplication was apparent in albumen samples. The mean log 10 level of S. enteritidis detected in yolk samples after each of the 3 d of incubation at 17.5 C was significantly greater than from any other type of sample (P < FIGURE 1. Enumeration of Salmonella enteritidis (SE) in inoculated egg contents samples after incubation at 25 C. Two dose levels of SE of sample (P < 0.001). Likewise, S. enteritidis in whole egg (yolk surface) samples significantly (P < 0.001) exceeded that in either whole egg (albumen edge) samples or in albumen samples at all three sampling intervals. Few significant differences between the two initial inoculum doses of S. enteritidis (15 and 150 cfu) were observed when the various types of egg samples were incubated at 25 C. Significantly higher levels of S. enteritidis were associated with the larger initial dose for yolk samples after 1 d of incubation (P < 0.001) and for whole egg and yolk surface samples after 2 d of incubation (P = 0.0206). The number of S. enteritidis cells in yolk samples continued to increase throughout the 3 d of incubation at 17.5 FIGURE 2. Enumeration of Salmonella enteritidis (SE) in inoculated egg contents samples after incubation at 17.5 C. Two dose levels of SE

562 GAST AND HOLT 0.001). The S. enteritidis level in whole egg (yolk surface) samples was significantly greater than in either whole egg (albumen edge) samples or in albumen samples at 2 and 3 d(p < 0.001). Significant differences between the two initial doses were sometimes seen after incubation of samples at 17.5 C. For yolk samples, the higher (150 cfu) dose led to significantly more S. enteritidis after 1 d (P = 0.0008) and 2 d (P = 0.0238) of incubation. For whole egg and yolk surface samples, the higher dose led to significantly more S. enteritidis after 2 d(p = 0.0096) and 3d(P = 0.0047). The mean log 10 levels of S. enteritidis in yolk samples after incubation at 10 C increased only to 2.8/mL (15-cfu dose) and 3.1/mL (150-cfu dose) at 3 d (Figure 3). Little or no multiplication of S. enteritidis was observed in other types of samples incubated at this temperature. The mean log 10 levels of S. enteritidis detected in yolk samples was significantly greater than from any other type of sample after 2 and 3 d of incubation at 10 C (P < 0.001). No significant differences were evident between the two initial inoculum doses (15 and 150 cfu) in any type of sample incubated at 10 C. DISCUSSION Because most available evidence has suggested that eggs contaminated with S. enteritidis typically contain relatively few bacterial cells (Humphrey et al., 1989, 1991; Gast and Beard, 1992a), refrigeration to limit microbial multiplication is among the most frequently recommended approaches for lessening the likelihood that eggs will transmit disease to consumers. The value of refrigeration for achieving this objective depends on the extent to which bacterial growth occurs while internal egg temperatures are lowered from higher initial values that were determined by the body temperatures of the laying hens. As the present study demonstrated, S. enteritidis multiplication in eggs depends on the site and magnitude of contamination as well as the storage temperature and elapsed time. Very little change was apparent in S. enteritidis levels in albumen held at any temperature for 3 d. This result accords with considerable prior documentation that albumen is not a good medium for bacterial growth, although extended persistence of Salmonella in albumen has sometimes been noted (Clay and Board, 1991; Lock and Board, 1992). When the yolk was also present, inoculation of S. enteritidis into the albumen in the current experiments rarely led to any detectable increase in bacterial numbers during 3 d of incubation. Some previous investigations have similarly found that the physical and chemical changes in albumen viscosity and yolk membrane permeability that allow albumen contaminants to gain access to the yolk interior do not usually occur during the first week of incubation (Kim et al., 1989; Humphrey and Whitehead, 1993), but more rapid bacterial migration has also been reported (Hammack et al., 1993; Braun and Fehlhaber, 1995). FIGURE 3. Enumeration of Salmonella enteritidis (SE) in inoculated egg contents samples after incubation at 10 C. Two dose levels of SE When inoculated into the yolk contents, S. enteritidis cells proliferated quickly at temperatures of 17.5 and 25 C in the present study. At 10 C, such growth occurred far more slowly. Yolk has long been known to be an excellent nutrient medium for microbial growth at supportive temperatures (Bradshaw et al., 1990; Humphrey, 1990). Moreover, substantial increases in S. enteritidis levels were also evident for samples inoculated on the exterior yolk surface. Bacterial penetration through the yolk membrane seems to have occurred rapidly and often in this experimental model. Some previous work has suggested that this vitelline membrane may be a principal

MULTIPLICATION OF SALMONELLA ENTERITIDIS IN EGGS 563 site of yolk-associated S. enteritidis contamination (Gast and Beard, 1990). The present study demonstrated that S. enteritidis cells inoculated into or onto egg yolks could multiply to much more dangerous levels within 1 d at 25 C. At a more moderate temperature (17.5 C), this multiplication proceeded noticeably more slowly, especially for yolk samples contaminated on the exterior surface or with very small initial numbers of bacterial cells (15 cfu). These data support the need for growth-restricting refrigeration temperatures to be achieved quickly, because eggs in which S. enteritidis contaminants are deposited in close proximity to the yolk membrane may experience significant expansion in bacterial numbers even during the first day of storage. Unfortunately, conventional cooling systems may require 72 h or more to achieve such temperatures inside eggs (Curtis et al., 1995). Although some investigators have observed that albumen contamination may sometimes occur more often than yolk contamination (Humphrey et al., 1991), reports of yolk contamination at fairly high frequencies (Gast and Beard, 1990, 1992b) should serve as reminders that effective refrigeration standards must account for this contingency. ACKNOWLEDGMENT The authors gratefully acknowledge the excellent technical contributions of Rupinder Guraya. REFERENCES Angulo, F. J., and D. L. Swerdlow, 1999. Epidemiology of human Salmonella enterica serovar Enteritidis infections in the United States. Pages 33 41 in: Salmonella enterica Serovar Enteritidis in Humans and Animals. A. M. Saeed, R. K. Gast, M. E. Potter, and P. G. Wall, ed. Iowa State University Press, Ames, IA. Baron, F., M. Gautier, and G. Brule, 1997. Factors involved in the inhibition of growth of Salmonella enteritidis in liquid egg white. J. Food Prot. 60:1318 1323. Bichler, L. A., K. V. Nagaraja, and D. A. Halvorson, 1996. Salmonella enteritidis in eggs, cloacal swab specimens, and internal organs of experimentally infected White Leghorn chickens. Am. J. Vet. Res. 57:489 495. Bradshaw, J. G., D. B. Shah, E. Forney, and J. M. Madden, 1990. Growth of Salmonella enteritidis in yolk of shell eggs from normal and seropositive hens. J. Food Prot. 53:1033 1036. Braun, P., and K. Fehlhaber, 1995. Migration of Salmonella enteritidis from the albumen into the egg yolk. Int. J. Food Microbiol. 25:95 99. Centers for Disease Control, 1996. Outbreaks of Salmonella serotype Enteritidis infection association with consumption of raw shell eggs United States, 1994 1995. Morbid. Mortal. Weekly Rep. 45:737 742. Clay, C. E., and R. G. Board, 1991. Growth of Salmonella enteritidis in artificially contaminated hens shell eggs. Epidemiol. Infect. 106:271 281. Curtis, P. A., K. E. Anderson, and F. T. Jones, 1995. Cryogenic gas for rapid cooling of commercially processed shell eggs before packaging. J. Food Prot. 58:389 394. Gast, R. K., and C. W. Beard, 1990. Production of Salmonella enteritidis-contaminated eggs by experimentally infected hens. Avian Dis. 34:438 446. Gast, R. K., and C. W. Beard, 1992a. Detection and enumeration of Salmonella enteritidis in fresh and stored eggs laid by experimentally infected hens. J. Food Prot. 55:152 156. Gast, R. K., and C. W. Beard, 1992b. Evaluation of a chick mortality model for predicting the consequences of Salmonella enteritidis infections in laying hens. Poultry Sci. 71:281 287. Hammack, T. S., P. S. Sherrod, V. R. Bruce, G. A. June, F. B. Satchell, and W. H. Andrews, 1993. Growth of Salmonella enteritidis in Grade A eggs during prolonged storage. Poultry Sci. 72:373 377. Hogue, A., P. White, J. Guard-Petter, W. Schlosser, R. Gast, E. Ebel, J. Farrar, T. Gomez, J. Madden, M. Madison, A. M. McNamara, R. Morales, D. Parham, P. Sparling, W. Sutherlin, and D. Swerdlow, 1998. Epidemiology and control of eggassociated Salmonella enteritidis in the United States of America. Rev. Sci. Tech. Off. Int. Epiz. 16:542 553. Humphrey, T. J., 1990. Growth of salmonellas in intact shell eggs: influence of storage temperature. Vet. Rec. 126:292. Humphrey, T. J., A. Baskerville, S. Mawer, B. Rowe, and S. Hopper, 1989. Salmonella enteritidis phage type 4 from the contents of intact eggs: a study involving naturally infected hens. Epidemiol. Infect. 103:415 423. Humphrey, T. J., and A. Whitehead, 1993. Egg age and the growth of Salmonella enteritidis in egg contents. Epidemiol. Infect. 111:209 291. Humphrey, T. J., A. Whitehead, A.H.L. Gawler, A. Henley, and B. Rowe, 1991. Numbers of Salmonella enteritidis in the contents of naturally contaminated hens eggs. Epidemiol. Infect. 106:489 496. Kim, C. J., D. A. Emery, H. Rinke, K. V. Nagaraja, and D. A. Halvorson, 1989. Effect of time and temperature on growth of Salmonella enteritidis in experimentally inoculated eggs. Avian Dis. 33:735 742. Kinde, H., D. H. Read, R. P. Chin, A. A. Bickford, R. L. Walker, A. Ardans, R. E. Breitmeyer, D. Willoughby, H. E. Little, D. Kerr, and I. A. Gardner, 1996. Salmonella enteritidis, phage type 4 infection in a commercial layer flock in Southern California: bacteriological and epidemiologic findings. Avian Dis. 40:665 671. Lock, J. L., and R. G. Board, 1992. Persistence of contamination of hens egg albumen in vitro with Salmonella serotypes. Epidemiol. Infect. 108:389 396. Saeed, A. M., and C. W. Koons, 1993. Growth and heat resistance of Salmonella enteritidis in refrigerated and abused eggs. J. Food Prot. 56:927 931. Schlosser, W. D., D. J. Henzler, J. Mason, D. C. Kradel, and S. Hurd, 1995. The Salmonella enteritidis Pilot Project and the Pennsylvania Egg Quality Assurance Program. Proc. US Anim. Health Assoc. 98:425 430. Stephenson, P., F. B. Satchell, G. Allen, and W. H. Andrews, 1991. Recovery of Salmonella from eggs. J. Assoc. Off. Anal. Chem. 74:821 826. Tauxe, R. V., 1997. Emerging foodborne diseases: an evolving public health challenge. Dairy Food Environ. San. 17:788 795. US Department of Agriculture, 1998. Advanced notice of proposed rule-making: Salmonella enteritidis in eggs. Fed. Reg. 63:27502 27511. Waltman, W. D., R. K. Gast, and E. T. Mallinson, 1998. Salmonellosis. Pages 4 13 in: A Laboratory Manual for the Isolation and Identification of Avian Pathogens. 4th ed. D. E. Swayne, J. R. Glisson, M. W. Jackwood, J. E. Pearson, and W. M. Reeds, ed. American Association of Avian Pathologists, Kennett Square, PA.