Microbial Contamination in Inoculated Shell Eggs: I. Effects of Layer Strain and Hen Age 1

Transcription

1 Microbial Contamination in Inoculated Shell Eggs: I. Effects of Layer Strain and Hen Age 1 D. R. Jones,*,2,3 K. E. Anderson,*,4 P. A. Curtis,,5 and F. T. Jones *Department of Poultry Science and Department of Food Science, North Carolina State University, Raleigh, North Carolina 27695; and Department of Poultry Science, University of Arkansas, Fayetteville, Arkansas ABSTRACT Three Ottawa control strains and a current These data suggest that PF is a primary invader of eggs commercial laying stock were reared and housed under identical environmental and management conditions. that is more capable of contaminating egg contents through the shell membranes than SE. The PF and SE Eggs were collected from each strain when hens were 32, data suggest that bacterial contamination of air cells, 45, 58, 71, and 84 wk of age. The eggs were inoculated with shells, and egg contents is more easily achieved in eggs Salmonella enteritidis (SE), Pseudomonas fluorescens (PF), or from older hens than from younger hens. There were also differences between the strains. Control Strain 10 a combination of the two. After storage at 26 C, bacterial consistently maintained a lower level of contamination counts were obtained from the exterior shell surfaces for both organisms in each sampling location. The overall (rinse), air cell, egg contents, and shell structure. SE and results of this study suggest that genetic selection has PF survived at different rates on the shell surface with altered the ability of eggs to resist microbial contamination and that screening for microbial integrity should as much as a 1 log difference during a given collection period. Egg content counts tended to be higher than eggshell counts in PF, whereas the opposite was true for SE. be considered in the selection process among the laying egg breeders. (Key words: egg, Salmonella enteritidis, Pseudomonas fluorescens, pathogen, spoilage) 2002 Poultry Science 81: INTRODUCTION Numerous production and processing changes have occurred within the commercial egg industry over the past 20 yr that have affected the quality of egg being delivered to consumers in the United States (Bell, 1998). However, recognition of Salmonella enteritidis (SE) as an emerging pathogen in eggs (St. Louis et al., 1988; Gast and Beard, 1990) led to perhaps the most dramatic shift in the consumer definition of egg quality in recent years (Thorton, 1991). Prior to SE, consumers defined egg quality in physical and visual terms (i.e., size of the air cell, color of the yolk, height of the albumen) and few consumers expressed concern about the microbial load contained 2002 Poultry Science Association, Inc. Received for publication August 2, Accepted for publication November 27, Mention of trademark, proprietary product, or specific equipment does not constitute a warranty by North Carolina State University and does not imply its approval to the exclusion of other products that may be suitable. 2 Current address: Russell Research Center, Poultry Processing and Meat Quality Research Unit, USDA-ARS, Athens, GA To whom correspondence should be addressed: drjones@saa. ars.usda.gov. 4 Reprint requests: Ken_Anderson@ncsu.edu. 5 Current address: Department of Poultry Science, Auburn University, Auburn, AL on or within commercially processed eggs. Confirmation of Grade A shell eggs as a major source of SE infections (Gast, 1994; Humphrey, 1994) led to a shift in consumer attention and research efforts. Although a majority of these research efforts focused on understanding and controlling in ovo transmission of SE, certain research groups concentrated on bacterial contamination via more traditional routes. Egg washing procedures were re-examined (Lucore et al., 1994) and when the bacterial microflora from eggs collected from research houses was characterized (Lucore, 1994) results were similar to those obtained by earlier researchers (Board et al., 1964). Gram-positive organisms accounted for 50 to 60% of isolates, whereas Pseudomonas isolates comprised 10 to 15% (Lucore, 1994; Board et al., 1964). Board et al. (1964) also noted that fluorescent Pseudomonas isolates accounted for 1.2% of the isolates obtained from eggshells. Although earlier experiments documented that Pseudomonas was among the genera of bacteria commonly found in rotten eggs (Board and Tranter, 1995), rotten eggs occurred in earlier eras because eggs were produced during the spring and summer months and had to be subjected to prolonged periods of cold storage so that eggs were available in the fall and winter. Modern flock Abbreviation Key: BPW = buffered peptone water; CCS = current commercial stock; CS5 = control strain 5; CS7 = control strain 7; CS10 = control strain 10; PF = Pseudomonas fluorescens;se = Salmonella enteritidis. 715

2 716 JONESETAL. management methods have virtually eliminated seasonal differences in egg production reducing the need for prolonged cold storage. In fact, most eggs reach the retail outlet within 36 h of processing (American Egg Board, 1981). Attempts (Anderson, K. E., P. A. Curtis, and F. T. Jones, 1993, unpublished data) have failed to replicate experimental results obtained by earlier researchers (Brant et al., 1965) who inoculated eggs. Because the bacterial microflora of the eggs used in these experiments was similar to that documented by earlier researchers, it was postulated that the genetic improvements in layers could have influenced the susceptibility of eggs to microbial contamination. Roberts and Brackpool (1994) noted that in Single Comb White Leghorn hens, eggshell thickness differences related to genetic strain and hen age exist. Tharrington et al. (1999) noted that genetic improvements in commercial layer strains have had a dramatic impact on egg size and shape. Therefore, this research project was conducted to determine if eggs from three historical strains and one current strain of commercial laying hen differed in bacterial counts after inoculation with a pathogen (SE), a spoilage organism [Pseudomonas fluorescens (PF)], or both (SE and PF). In addition, the effects of hen age on the microbial integrity of eggs were investigated. MATERIALS AND METHODS Source of Shell Eggs Eggs were collected from three Ottawa Control Strains, provided by Agriculture Canada, and a current commercial laying strain. Each strain was hatched and reared under identical environmental and management conditions (Anderson, 1996) at the North Carolina Department of Agriculture Piedmont Research Station, Salisbury, North Carolina. The random-bred control strains utilized were strains 5 (CS5), 7 (CS7), and 10 (CS10) as described by Gowe et al. (1993) and Fairfull et al. (1983). Selection was closed in these strains in 1950, 1959, and 1972, respectively. The current commercial stock (CCS) was a 1993 laying stock with a common ancestral linkage to the control strains, the H&N Nick Chick. Molt was induced in all strains at 62 wk of age. The first production cycle peak occurred from 28 to 32 wk of age. The second production cycle peak in egg production occurred at 68 wk of age. Egg samples were collected from all strains when hens were 32, 45, 58, 71, and 84 wk of age. Experimental Design Pseudomonas fluorescens and a nalidixic acid resistant SE (Brant et al., 1965) were used to inoculate eggs. Pseudomonas fluorescens was chosen because it comprised a very small portion of the initial bacterial population pres- 6 Spectronic Unicam, Rochester, NY. ent on shell eggs. Eggs from each of the four hen strains were inoculated with SE, PF, or a combination of SE and PF. Treatments were arranged in a 3 4 factorial design, and bacterial counts were obtained from four locations: external shell rinse, internal air cell, egg contents, and shell contents. Inoculation Technique The standard inoculation technique (Brant et al., 1965) was used for PF and SE inoculation procedures. Fresh, nest run, clean, large eggs, not older than 72 h, were obtained. Before transportation to the laboratory on sanitized plastic flats, eggs were washed under commercial conditions and candled with a mercury vapor light to remove eggs with cracks, blood, or other defects. Flats containing 30 eggs, one from each test strain, were allowed to equilibrate for 18 to 24 h at 26 C. Control eggs were also tested each collection period. The control egg sample comprised an equal number of eggs from each of the hen strains that were immersed in sterile buffered peptone water (BPW) according to the procedure described below. The SE inoculum was grown in 100 ml of trypticase soy broth for 24 h at 35 C. The PF inoculum was grown in 100 ml of trypticase soy broth for 48 h at 26 C. Turbidity of the inoculum was measured with a Spectronic 20 6 spectrophotometer set at 590 nm, and calibration curves showing the relationship between inoculum concentration and turbidity were established. Approximately 8 L of BPW was placed in a 10 L container and autoclaved at 121 C for 15 min. The BPW was cooled to 7 C lower than the egg temperature. The appropriate amount of stock culture was added to provide a population density of 10 6 cfu/ml for each organism. A sample of the medium was removed aseptically and a plate count was performed to confirm inoculum concentration. Two stainless steel baskets, each containing 24 eggs, were immersed simultaneously in the inoculum and removed after 10 s. Eggs were drained in the basket for approximately 1.5 min and then placed on sanitized plastic egg flats to air dry for 10 min. Inoculated and control eggs were stored in a 26 C incubator at 90% RH ± 5% until tested. After inoculation, eggs were tested weekly for 5 wk. Data on the effect of egg age on bacterial contamination will be presented in a subsequent manuscript. Microbial Sampling Techniques Control eggs were always assayed prior to inoculated eggs. External shell wash samples were obtained with the procedure described by Gentry and Quarles (1972) with 100 µl of saline egg wash spiral plated onto the appropriate media. All other bacterial counts required eggs being aseptically cracked, which was achieved by first dipping the intact egg into 95% ethanol, allowing it to air dry, and cracking the egg on the lip of a sanitized beaker. Egg contents were then placed in a sterile sample bag. Internal air cell counts were obtained by rinsing the

3 HEN STRAIN AND AGE EFFECTS ON BACTERIAL CONTAMINATION OF EGGS 717 albumen from the membrane with sterile distilled water and then rinsing with 70% ethanol. The shell membrane was broken with a sterile calcium alginate swab, which was used to swab the air cell shell surface, and transferred to a tube containing 3 ml of citrate buffer. After equilibration for 15 min, 500 µl of inoculated buffer was introduced onto a spread plate of appropriate media. Counts of egg contents were determined by pooling the contents of six eggs in a sterile sample bag, stomaching for 60 s in a laboratory stomacher, and spreading 500 µl onto appropriate media with a sterile bent glass rod. Eggshell content samples were obtained by first rinsing the inside of the shell with sterile saline to remove any adhering albumen, air drying the shell, spraying the shell inside and out with ca 5 ml of 70% ethanol, placing six egg shells in a sterilized glass blender jar 7 with 200 ml of sterile saline, and blending the pool of shells on high speed for 30 s. The blended shell mixture was allowed to stand for 2 min before 500 µl was spread on appropriate media. Duplicate plates were made for each sample of the egg content and shell content. SE (nalidixic acid resistant) was cultured on MacConkey agar with 200 µl/ml nalidixic acid added incubated at 35 C for 48 h prior to counting on Quebec-type colony counters. PF was enumerated on Pseudomonas isolation agar incubated at 26 C for 48 h before enumeration. Statistical Analyses All data were subjected to a log transformation before analysis was conducted. Due to the prominence of threeway interactions, data were sorted for hen age. Data were analyzed using the general linear model of SAS software (SAS, 1989) with egg age as a block and using control egg counts as a covariant in the analysis of treatment groups. Means were separated by the least-squares method. RESULTS Count data obtained from eggs that were inoculated with both SE and PF were virtually identical to data obtained from eggs inoculated with only one organism, and few significant interactions were noted. Thus, for brevity, these data are not presented. Because counts obtained from control eggs were always lower than those obtained from inoculated eggs, these data were also omitted. The lowest external rinse counts for SE were found in eggs collected during the first cycle egg production peak at 32 wk (Figure 1). Subsequent collection periods resulted in similar average SE levels present on the shell surfaces. These average values ranged from 1.5 to 2.1 log cfu/ ml. Whenever significant differences occurred among the strains, CS10 maintained the lowest levels of external SE contamination. At 58 wk of age (end of first cycle 7 Waring, Torrington, CT. FIGURE 1. Effect of hen age and hen genetic strain on Salmonella enteritidis counts obtained from exterior rinse samples of inoculated shell eggs. ****Hen age and genetic strain interaction, P < a,b Genetic strains with the same letter are similar within a hen age, P < A,B Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); production), the CCS had the greatest level of SE external contamination (2.98 log cfu/ml). Exterior rinse samples from PF-inoculated shell eggs produced similar results to those observed for the SEinoculated eggs, although absolute levels were lower for PF (Figure 2). Once again, when significant differences occurred among strains, CS10 had the lowest PF concentration. At 58 wk of age, the CCS had a sharp increase in SE and PF present on the shell surface after inoculation. This increase occurred before the molt (62 wk of age) when shell quality is low. Shell populations of SE and PF remained elevated after the molt. These results suggest that SE and PF survive at a greater level on eggs with decreased shell quality. Because identical concentrations of SE and PF were used to inoculate the shell surface, the FIGURE 2. Effect of hen age and hen genetic strain on Pseudomonas fluorescens counts obtained from exterior rinse samples of inoculated shell eggs. a,b Genetic strains with the same letter are similar within a hen age, P < Y,Z Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); Control Strain 10 (white); current commercial strain (horizontal stripe);

4 718 JONESETAL. FIGURE 3. Effect of hen age and hen genetic strain on Salmonella enteritidis counts obtained from the interior (air cell) of inoculated shell eggs. **Hen age and genetic strain interaction, P < a,b Genetic strains with the same letter are similar within a hen age, P < Y,Z Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); Control Strain 10 (white); current commercial strain (horizontal stripe); average for all strains (black). current data suggests that SE was better able to survive on the shell surface under the conditions utilized in the current study. Figure 3 illustrates the effects of hen genetic strain on interior (air cell) SE counts obtained from inoculated eggs. In general, internal SE counts tended to increase with hen age. At two periods, 45 wk (P < 0.05) and 84 wk (P < ), significant differences were found among the strains for air cell SE contamination levels. Each of these times, CS10 had one of the lowest contamination levels. As was observed with exterior rinse values, at 58 wk the CCS had a large increase in SE contamination. The greatest average SE air cell contamination was observed at 84 wk of age (1.25 log cfu/ml). The contamination of the interior (air cell) of the inoculated eggs with PF is shown in Figure 4. When significant FIGURE 5. Effect of hen age and hen genetic strain on Salmonella enteritidis counts obtained from egg contents of inoculated shell eggs. **Hen age and genetic strain interaction, P < ****Hen age and genetic strain interaction, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); Control Strain 10 (white); current commercial strain (horizontal stripe); differences occurred among the strains, CS10 once again had one of the lowest contamination levels of PF in the air cell. At 58 wk, the CCS had a significant (P < ) increase in PF air cell contamination (1.51 log cfu/ml). Unlike exterior contamination levels, the concentrations of SE and PF isolated from the air cell were similar. Therefore, both organisms appear to traverse the shell matrix at comparable rates. SE contamination levels in the egg contents of inoculated eggs generally increased with hen age (Figure 5). As was the case in previous measurements, there was an increase in the number of SE organisms present in the egg contents of the CCS at 58 wk of age (2.52 log cfu/ ml). Furthermore, CS10 had one of the lowest SE contamination levels at each collection period. The levels of PF present in the egg contents increased with hen age in each of the two production cycles (Figure 6). The lowest levels of PF contamination occurred at 71 FIGURE 4. Effect of hen age and hen genetic strain on Pseudomonas fluorescens counts obtained from the interior (air cell) of inoculated shell eggs. *Hen age and genetic strain interaction, P < a,b Genetic strains with the same letter are similar within a hen age, P < A,B Genetic strains with the same letter are similar within a hen age, P < Y,Z Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); FIGURE 6. Effect of hen age and hen genetic strain on Pseudomonas fluorescens counts obtained from the egg contents of inoculated shell eggs. ***Hen age and genetic strain interaction, P < A,B Genetic strains with the same letter are similar within a hen age, P < Y,Z Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe);

5 HEN STRAIN AND AGE EFFECTS ON BACTERIAL CONTAMINATION OF EGGS 719 DISCUSSION FIGURE 7. Effect of hen age and hen genetic strain on Salmonella enteritidis counts obtained from eggshells of inoculated shell eggs. A,B Genetic strains with the same letter are similar within a hen age, P < y,z Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); wk (average of 0.58 log cfu/ml), just after the second cycle production peak. The CCS had a significantly (P < ) greater PF contamination in the egg contents at 58 wk of age (4.10 log cfu/ml). The CS10 also maintained one of the lowest levels of PF contamination at each collection period. For each of the organisms, results were similar for egg contents and shell contamination (Figures 7 and 8). CS10 maintained one of the lowest contamination levels among the strains. For SE, shell contamination levels were greater than those observed for egg contents. The opposite was true for PF, in which contents contamination levels were greater than those observed for the shell during the same period. FIGURE 8. Effect of hen age and hen genetic strain on Pseudomonas fluorescens counts obtained from the eggshells of inoculated shell eggs. ****Hen age and genetic strain interaction, P < a,b Genetic strains with the same letter are similar within a hen age, P < Key: Control Strain 5 (gray); Control Strain 7 (diagonal stripe); Control Strain 10 (white); current commercial strain (horizontal stripe); average for all strains (black). Although generally all counts tended to increase with increased hen age, the slope of this increase varied with the sampling location and organism. External rinse SE counts rose drastically during the first production cycle and then remained fairly constant on subsequent samplings. In contrast, external rinse PF counts tended to rise gradually with each subsequent sampling. Because exterior rinse counts should reflect the populations of organisms existing on or near the shell surface, these data suggest that the conditions present in this trial allowed SE to survive at a greater level than PF. The storage conditions, after inoculation, used in this study do not represent common industry practice. The relative humidity and temperature in this study better represent a worst-case scenario. Although interior (air cell), pooled egg content, and pooled eggshell counts for SE and PF appeared to respond similarly with increased hen age, differences existed with respect to the organisms examined. Interior (air cell) SE counts were lower at 32 wk than were interior (air cell) PF counts. The SE levels gradually increased with hen age during the life of the flock. There was a period of decreased PF contamination in the air cell, which coincided with the beginning of the second egg production cycle. These results show that PF appeared to be more dependent on shell and egg quality. Egg content PF counts tended to be higher than eggshell PF counts, whereas the opposite was true for SE. As previously noted, PF can invade eggs during prolonged cold storage and is easily capable of contaminating egg contents through the shell membranes (Florian and Trussell, 1957), whereas SE is less able to move through shell membranes at the expense of competing organisms (Humphrey, 1994). The data from this study also illustrate these points. Nevertheless, data from PF and SE suggest that bacterial contamination of air cells, shells, and egg contents is more common in eggs from older hens than from younger hens. The egg layer strains used in the current study had different rates of contamination for SE and PF at the various sampling locations. CS5 was initiated during the time of the Florian and Trussell study. This strain had one of the greater levels of PF contamination in the air cell, contents, and shell. Therefore, it can be understood how Florian and Trussell obtained those results. Throughout the collection periods of this study, CS10 generally maintained the greatest level of microbial integrity for SE and PF contaminations. The CCS and CS5 consistently had the highest microbial contamination for the parameters tested. Previous research conducted with this flock found that CCS had the greatest level of egg production and the highest percentage of large and extralarge eggs. This finding would suggest that microbial integrity was affected during the selection for more efficient egg production. In conclusion, Salmonella enteritidis and Pseudomonas fluorescens each exhibited ability to survive at different

6 720 JONESETAL. rates in the various portions of the egg. The pathogenic SE was more able to survive on the exterior surface, whereas the spoilage organism PF was better able to traverse the shell membranes and infect the contents of the egg. Hen age also affected the microbial integrity of the shell egg. As a hen ages and shell and egg quality decreases, organisms are better able to infect the egg. It was more difficult for PF to contaminate the egg early in each laying cycle. SE contamination levels did not decrease after molting, as was the case with PF. Instead, SE contamination levels remained consistent with those before the molt. As the hen aged, increases were seen in SE contamination throughout the egg. The overall results of this study suggest that screening for microbial integrity should be included in the selection process among the laying hen breeders. Throughout history, this selection has not been a major economic concern for the industry, but with the current emphasis on food safety, microbial integrity should be evaluated in breeder stocks. This study also illustrates the need to maintain random-bred control stocks to serve as a point of reference for changes that have taken place through selection. ACKNOWLEDGMENTS The authors acknowledge Agriculture Canada for providing the hatching eggs for the control strains. The authors also appreciate the laboratory assistance provided by George Gartrell, Mike Mann, and Joanna Tharrington. They also thank Pam Jenkins for performing the statistical analysis. The authors are also grateful to the staff of the North Carolina Department of Agriculture, Piedmont Research Station for maintaining the research flock. REFERENCES American Egg Board Eggcyclopedia. American Egg Board, Park Ridge, IL. Anderson, K. E Final report of the thirty-first North Carolina layer performance and management test. NCLP&MT Report 33. North Carolina State University, North Carolina Cooperative Extension Service, Raleigh, NC. Bell, D Economic aspects in layer production. Pages in Proceedings of the International Seminar in Avian Pathology and Production. University of Georgia, Athens, GA. Board, R. G., and H. S. Tranter The microbiology of eggs. Pages in Egg Science and Technology. W. J. Stadelman and O. J. Cotterill, ed. Food Products Press, New York. Board, R. G., J. C. Ayres, A. A. Kraft, and R. G. Forsythe The microbiolgical contamination of eggshells and egg packing materials. Poult. Sci. 43: Brant, A. W., P. B. Starr, and J. A. Hamann The bacteriological, chemical, and physical requirements for commercial egg cleaning. Marketing Research Report No USDA, ARS, Washington, DC. Fairfull, R. W., V. A. Garwood, J. L. Spencer, R. S. Gowe, and P. C. Lowe The effects of geographical area, rearing method, caging density, and lymphoid leukosis infection on adult performance of egg stocks of chickens. Poult. Sci. 62: Florian, M. L. E., and P. C. Trussell Bacterial spoilage of shell eggs. IV. Identification of spoilage organisms. Food Technol. 11: Gast, R. K Understanding Salmonella enteritidis in laying chickens: The contributions of experimental infections. Int. J. Food Microbiol. 1: Gast, R. K., and C. W. Beard Production of Salmonella enteritidis contaminated eggs by experimentally infected hens. Avian Dis. 34: Gentry, R. F., and C. L. Quarles The measurement of bacterial contamination on eggshells. Poult. Sci. 51: Gowe, R. S., R. W. Fairfull, I. McMillan, and G. S. Schmidt Strategy for maintaining high fertility and hatchability in multiple-trait egg stock selection program. Poult. Sci. 72: Humphrey, T. J Contamination of egg shell and contents with Salmonella enteritidis: A review. Int. J. Food Microbiol. 21: Lucore, L. A Effect of egg temperature and wash water temperature on internal and external bacterial counts, cooling rate, bacterial analysis of contents, and bacterial populations on shell eggs processed in a modern commercial-type processor. M.S. Thesis. North Carolina State University, Raleigh, NC. Lucore, L. A., F. T. Jones, K. E. Anderson, and P. A. Curtis Internal and external bacterial counts from shells of eggs washed in a commercial-type processor at various wash-water tempeature. J. Food Prot. 60: Roberts, J. R., and C. E. Brackpool The ultrastructure of avian egg shells. Poult. Sci. Rev. 5: SAS Institute A User s Guide to SAS. Sparks Press, Inc., Cary, NC. St. Louis, M. E., D. L. Morse, M. E. Potter, T. M. DeMilfi, J. J. Guzeqich, R. V. Tauxe, and P. A. Blake The emergence of Grade A eggs as a major source of Salmonella enteritidis infections. JAMA 259: Tharrington, J. B., P. A. Curtis, F. T. Jones, and K. E. Anderson Comparison of physical quality and composition of eggs from historic strains of Single Comb White Leghorn chickens. Poult. Sci. 78: Thorton, G Salmonella enteritidis: The undefined threat. Egg Ind. 97:14 22.