Distribution of "Attached" Salmonella typhimurium Cells Between Poultry Skin and a Surface Film Following Water Immersion

Transcription

1 449 Joiknal of Food Protection, Vol. 49, No. 6, Pages (June 1986) Gopy-cight International Association of Milk, Food and Environmental Sanitarians Distribution of "Attached" Salmonella typhimurium Cells Between Poultry Skin and a Surface Film Following Water Immersion H. S. LILLARD United States Department of Agriculture, Agricultural Research Service, Richard B. Russell Agricultural Research Center, P.O. Box 5677, Athens, Georgia (Received for publication October 7, 1985) ABSTRACT Poultry skin was immersed in a saline solution (0.85%) containing 10 8 Salmonella typhimurium /ml. After 0.25 min, 95% of the water uptake was in a surface film and 5% in the skin. After immersion for 30 and 60 min, a significant increase in total water uptake occurred at each immersion time. A significant increase in the surface film occurred after 30 min of immersion, but not from 30 to 60 min. After 0.25 min of immersion, 94% of bacterial cells was entrapped in the water film and 6% was on the skin. As immersion time increased, the percentage of bacteria in the surface film decreased, whereas the percentage on the skin increased. After 60 min of immersion, about 39% of bacterial cells was in the surface film and 61% was on the skin. These data indicate a possible transfer of water and bacteria from surface film to skin during prolonged water immersion. Preventing the formation of the surface film by altering surface tension may reduce carcass contamination during immersion processes. Thomas and McMeekin (7,8) reported that time and temperature-dependent changes occur in the microtopography of poultry skin as a result of water uptake during immersion processing procedures. They noted that water absorption by breast skin causes "capillary-sized channels and spaces" to open in the surface layers. They indicated that, as a result of these alterations in tissue microtopography, more water is retained as a surface film compared to tissues that are only immersed for brief periods, and, hence, remain unaltered by water uptake. They hypothesized that, as a result of the increased liquid retained as a surface film, the number of contaminating bacteria might also be expected to increase (8). Also, exposed skin channels would allow bacteria present in processing water to penetrate deep into the skin surface and would thus be difficult to remove (7). They suggested that these bacteria may be the "attached bacteria" described by Notermans and Kampelmacher (4). In support of this theory, Thomas and McMeekin (7) presented the following argument: Notermans and Kampelmacher (4,5) reported an increase in numbers of bacteria retained with increases in immersion time and concluded that bacteria attach to poultry skin. McMeekin and Thomas (6) concluded that bacterial retention did not increase with time of immersion. Thomas and McMeekin (7) suggested that these differences may be explained by the fact that the two groups used skin from carcasses sampled at different stages of processing. Notermans and Kampelmacher (4,5) used skin from carcasses immediately following mechanical plucking, whereas Thomas and McMeekin (7) used skin from immersionchilled carcasses which had already absorbed water. They suggested that the increase in bacterial numbers with time of immersion reported by Notermans and Kampelmacher may simply reflect the time-dependent change in skin surface microtopography caused by water immersion and the increased volume of liquid retained as a surface film on the skin. Lillard (2) studied the effect on bacterial attachment of broiler skin samples taken from carcasses immediately after plucking and skin samples from immersion-chilled broilers. An increase in the number of bacteria retained on both types of skin was observed with increased immersion times. She concluded that skin from carcasses at different stages of processing had no effect on the pattern of bacterial attachment. Thomas and McMeekin (7) determined the weight of water held as a liquid film of skin of eviscerated, plucked and immersion-chilled carcasses by draining the carcasses for 30 s, then applying pre-weighed filter pad discs to the skin. They showed a highly significant increase in the amount of water held on the surfaces of both leg and breast skin following immersion-washing and chilling of the carcasses. This study was done to determine the effect of water immersion and length of immersion time on: (a) total water uptake by poultry skin; (b) the amount of water in the surface film formed on the skin; and (c) whether an increase in the amount of water in the film on the

2 450 LILLARD skin could account for the increase in bacteria adhering to the skin with increased immersion times. MATERIALS AND METHODS Preparation of cell suspension Nutrient broth (1 L), prepared according to the procedure of Duguid and Gillies (7), was inoculated with 2 ml of an 18- to 24-h culture of a flagellated, fimbriated strain of Salmonella typhimurium, and incubated at 35 C for 24 h. The cells were harvested by centrifugation (6000 rpm for 15 min at 9 C) and washed twice with 0.85% saline. The cells were then resuspended in 2 L of physiological saline (ph 6.2 to 6.5) to yield a cell concentration of 10 8 /ml. Sampling and experimental procedure Skin pieces (12 cm 2 ) were excised from the same location on breasts of fully processed (immersion-chilled) and uneviscerated, plucked broilers. Preweighed (analytical balance, Mettler AE-100) tagged skin pieces were immersed in a magnetically stirred S. typhimurium cell suspension (10 8 cells/ml) for 0.25, 30 and 60 min. Tagged skin pieces were removed at the appropriate time intervals, drained for 30 s, after which procedures shown in Figure 1 were followed for obtaining water uptake weights and for the enumeration of S. typhimurium cells. Experiments with skin from uneviscerated carcasses were repeated twelve times and four times with skin from immersion-chilled broilers. Procedure for spread inocula To eliminate the effect on bacterial attachment of microtopographical changes induced by water immersion of poultry skin, 0.1 ml of a cell suspension of S. typhimurium prepared as above, was spread on chicken skin with a hockey stick. The inoculum was spread on the whole skin of each broiler halfbreast which was then covered with foil (without skin contact) to prevent drying. Breasts were held 0.25, 30 and 60 min. The whole excised skin was then spray-rinsed at 50 psi, homogenized, and S. typhimurium cells were enumerated as described in Figure 1. Control samples were not rinsed before sampling in order to obtain an accurate recovery count of the spread inoculum. Experiments with spread inocula were repeated ten times. Calculations Calculations were made using the following equations: (a) Total water uptake = Weight of skin after immersion Initial weight of skin; (b) Weight of water in film = Combined weight of filter papers used to blot both sides of skin Combined initial weight of filter papers; (c) Counts due to total water uptake = Weight of total water uptake (g) x No. of bacterial cells/ml immersion fluid at each of 0.25, 30 and 60 min; (d) Counts in water film = Weight of water in film (g) x No. of bacterial cells/ml immersion fluid at each of 0.25, 30 and 60 min; and (e) Count in tissue due to retained water = Count due to total water uptake - Count in water film. Weigh tagged 12 cm 2 excised pieces of broiler breast skin from uneviscerated, plucked broilers Add skin pieces to cell Sample cell Stlr magnetlcally suspension at 20 C suspension at 0.25, 30 and 60 min Remove skin pieces at 0.25, 30 and 60 min Drain off excess fluid (30 sec) Blot both sides of skin (10 sec) with pre-welghed filter paper (Whatman 12, dlam. 5.5 cm, 23.8 cm ) Weigh filter paper after blotting Weigh skin after blotting Rinse both sides of skin samples by holding flat under cold tap water at 50 psl for 1 min (LI I lard, 1985) to remove loosely adhering bacteria and any part of water film left on skin (Bacteria that did not wash off were considered attached) Stanach samples for 1 min each with 10 ml 0.85 saline Transfer 0.1 ml of stomached sample to each of duplicate spread plates (BG Sulfa Agar, Dlfco) t Incubate spread plates at 35 C for 24 hr Enumerate.S. lyjahijuurj-uni colonies Figure 1. Sampling and experimental procedure for determining weight of total water uptake, weight of water in surface film, and the distribution of S. typhimurium cells between skin and surface film following immersion of broiler skin in a cell suspension for 0.25, 30 and 60 min. RESULTS AND DISCUSSION Unlike McMeekin and Thomas (6), Lillard (2) confirmed the results reported by Notermans and Kampelmacher (4,5) which indicated an increase in numbers of attached bacteria with increased immersion times. Lillard (2) reported that these increases occurred with skin from fully processed as well as from uneviscerated plucked carcasses. Thomas and McMeekin (7) suggested that more water may be retained in the surface film on skin from uneviscerated carcasses than on skin from fully processed carcasses. The same pattern of water uptake was observed in this study when skin from either source was used (Fig. 2, a and b). At 0.25 min for skin from both types of carcasses, almost all of the water taken up by the skin was retained in a surface film, whereas following 30- and 60-min immersion times, there was significantly more total water taken up by the skin than accounted for by the surface film. With both types of skin, there was an increase in total water uptake and in surface film with increased immersion times. After a 60-min immersion, there was about 0.3 g more total water uptake than water in the surface film for both types of tissue (Fig. 2, a and b). Because immersion-chilling is the final step in processing broilers, skin from uneviscerated carcasses was used throughout this study in order to determine realistically the effect of immersion processes on water uptake and the distribution of "attached" bacteria on poultry skin. There was a significant increase in total water uptake following 0.25-, 30- and 60-min immersion times, whereas, in the surface film, there was a significant in-

3 "ATTACHED" SALMONELLA 451 crease in water content from 0.25-min to 30- and 60-min immersion times but no difference in the amount of water in the surface film following 30- and 60-min immersion times (Table 1). These results suggest that water from the surface film may be absorbed by the skin following prolonged immersion, hence increasing total water uptake and stabilizing the amount of water retained as a surface film on the skin. The theory proposed by Thomas and McMeekin (7) that more water may be retained as a surface film following prolonged immersion of the skin than in the initial (a) CO 0.6 n FULLY PROCESSED 0.25 Total Water Uptake - - ' \ Water in Film W ~3o 15 s was not confirmed. In the first 15 s, 95% of water uptake was retained as a surface film, whereas less than 50% was retained as a surface film following 60 min of immersion of the skin (Table 1). Thomas and McMeekin (7) suggested that more bacteria may be retained in the surface film following prolonged immersion and may account for the "attached" bacteria reported by Notermans and Kampelmacher (4,5). The distribution shown here of 5. typhimurium on the skin and in the surface film following immersion in a cell suspension indicates the opposite (Table 2). After a (c) _ * CO M E o CM D U_ O E muriu c Q. +* > CO FULLY PROCESSED \ Skin N \ ^ \ ^S^ \.x^ / > - * Water Film 3'o 6 1 (b) O.6-1 UNEVISCERATED (d)? UNEVISCERATED Total Water Uptake Water Film Water in Film Figure 2. Comparison of water uptake by and bacterial attachment to skin from fully processed, immersion-chilled broilers and uneviscerated, plucked broilers.

4 452 LILLARD 0-25-min immersion, about 94% of the bacterial cells was accounted for by the water film and 6% was recovered from the skin. Following 30- and 60-min immersion times, a significant decrease occurred in the percentage of bacteria found in the surface film, with a concomitant significant increase in the percent of bacteria retained on the skin (Table 2). This redistribution of counts between skin and surface film with increased immersion time occurred with skin from both fully processed and uneviscerated plucked carcasses (Fig. 2, c and d). As the count in the surface-water film decreased, the count on the skin increased, again indicating a possible transfer from film to skin. If indeed this transfer does occur, at some point in time the counts on the skin would equal the counts in the surface film. For skin from fully processed carcasses, the counts were equal following about 30 min of immersion in the cell suspension; for skin from uneviscerated, plucked carcasses, the counts would have been equal following about 45 min of immersion (Fig. 2, c and d). Because data for skin from fully processed carcasses were based on four experiments whereas twelve experiments were done with uneviscerated carcasses, this difference may be an artifact of numbers rather than a difference due to tissue source. However, regardless of the cause for this difference, the trend for both types of tissue is the same, i.e., as counts decrease in the surface film, counts increase on the skin. The difference in counts attributable to total water uptake and the water found in the surface film would logically represent counts due to water absorbed by the skin (Table 3). The unconverted calculated mean count (2.5 x 10 6 ) in the water absorbed by the skin and the unconverted actual mean skin count (5.0 x 10 5 ) did not differ significantly when skin was immersed in a bacterial cell suspension for 0.25 min. However, the normalization of counts achieved by conversion to log 10 reveals a significant but small (0.73) difference between actual and calculated mean log 10 counts following 0.25 min of immersion. There was no significant difference between the calculated mean log 10 count due to absorbed water and actual mean log 10 count on skin for 30- and 60-min immersion times. Both the calculated count on skin due to retained water and the actual skin count (Table 3, columns C and D) reveal significant increases between the min immersion time and the 30- and 60-min immersion times, but no difference.between 30- and 60-min immersion times. These data (calculated vs. actual) tend to validate the methods and calculations presented in this study. The theory presented by Thomas and McMeekin (7) that there may be an increase in bacterial retention in the water film following prolonged immersion is rejected by data in Table 3. These data reveal that there was an increase in total water uptake by the skin with a concomitant increase in bacterial counts following 30- and 60-min immersion times. This increase in counts may be due to changes in microtopography of the skin, as suggested by Thomas and McMeekin (7), or may be simply due to increased water absorption or some other factors involved in bacterial attachment. Changes in the microtopography of poultry skin were shown to be time and temperature dependent, and these changes occurred more slowly at 2 C than at 20 C (7). Notermans and Kampelmacher (4) reported that 20 C was the optimum temperature at which bacteria attached to TABLE 1. Water uptake by broiler breast skin (12 cm 2 ) following immersion for 0.25, 30 and 60 min in 0.85% physiological saline containing 10 8 S. typhimurium/m/. Immersion time (min) 0.25 Total water uptake (g) a a ; 0.02) b Water in surface film on skin (g) a (± ; 0.02) Water uptake in surface film (%) c ; 0.09) b (± ; 0.05) (: d ; 0.14) b (± ; 0.06) a In columns and rows, values with the same letter are not significantly different (P<0.01). b Standard error of mean; standard deviation. Means were derived from 12 experiments TABLE 2. Distribution of Salmonella typhimurium counts between skin and surface water film following immersion of breast skin (12 cm 2 ) in 0.85% saline containing 10 s S. typhimurium/m/ ". Immersion time (min) Actual CFU on 12 cm 2 skin Calculated + ' CFU in surface film Total CFU skin + film Means of % CFU on skin b Means of % CFU in water film b X X X X XlO XlO XlO XlO XlO "Means derived from 12 experiments. b Percentages are significantly different (P<0.01).

5 "ATTACHED" SALMONELLA 453 TABLE 3. Mean log l0 Salmonella typhimurium colony-forming units (CFU) attributable to water uptake by broiler breast skin pieces (12 cm 2 ). Calculated count Calculated Calculated count on Actual count on Immersion due to total count in = skin due to breast skin time water uptake surface film retained water (12 cm 2 ) (min) A B C D 0.25 a 6.97a b 6.88a 6.22a 5.49a (± 0.10;0.36) c (±0.12; 0.43) (±0.15; 0.47) (± 0.14; 0.47) b 7.55b 7.59b 7.35b (±0.13; 0.43) (±0.12; 0.41) (± 0.13; 0.45) (± 0.21; 0.72) b 7.38b 7.50b 7.64b (± 0.11; 0.39) (± 0.11; 0.37) (± 0.12; 0.41) (± 0.17; 0.57) a In row 1, but not in rows 2 and 3, log 10 counts for C and D are significantly different (P<0.01) but are not different when actual counts are used. b Means derived from 12 experiments. In each column, values with the same letter are not significantly different (P<0.01). c Standard error of mean; standard deviation. poultry skin. Therefore, 20 C was used throughout this study. In actual practice, prechillers in poultry processing plants are maintained at 40 to 50 C F (4.4 to 10 C C) and are the vessels in which initial immersion chilling occurs. Experiments were repeated at 5.5 C with similar results, indicating that the time and temperature-dependent changes in microtopography did not necessarily account for the retention of bacteria on poultry skin following water immersion. In order to completely avoid changes in skin microtopography which may be induced by water immersion, inocula of S. typhimurium were spread on chicken breasts. Bacteria that did not rinse off were considered attached. Lower numbers attached to the skin with spread inocula than in the immersion experiments, but the same trend was observed with both methods. The number of attached bacteria increased significantly with increased holding times (Table 4). In this instance, bacteria were spread on the outer surface of the skin, whereas in the immersion studies both sides of the skin were exposed to the bacterial suspension. The surface area from the spread samples was at least equal to the surface area of both sides of the skin. It is possible that bacteria adhere more strongly to one surface than the other. However, considering that the calculated counts due to water retained in the skin so closely matched actual counts from homogenized skin (Table 3, columns C and D), it seems more likely that a greater number of bacteria were entrapped by the water absorbed by poultry skin, whereas a smaller number attached rapidly in the absence of water absorption. Water uptake and the surface film formed on poultry skin during water immersion play an important role in the adhesion of bacteria to the skin. Changes in microtopography may contribute to the overall process of water and bacterial retention, but factors other than water uptake must be involved in the attachment process as shown by experiments using a spread inoculum. Lillard (3) found that fimbriae are not the primary mechanism by which S. typhimurium attaches to poultry skin. Defimbriated cells attached as readily as the fim- TABLE 4. Salmonella typhimurium adhering to poultry skin (spread inoculum) in the absence of water immersion and concomitant changes in skin microtopography. S. typhimurium Sampling Total CFU recovered attached after time from breast skin rinsing (min) (%) X10 4 " 2.1 ( ; 0.19)a b X (4.55±0.09;0.31)a,b X (4.80±0.15;0.48)b Control (Not rinsed) 1.3X10 6 "Means derived from 10 experiments. In each column, values with the same letter are not significantly different (P<0.05). b Mean log 10 ± standard error of mean; standard deviation. briated, indicating the possibility of interaction between bacterial cell wall components and receptor sites on poultry skin. Also, nonfimbriated species attached readily to poultry skin immersed in a cell suspension (2). However, the significance of total water uptake and the formation of the surface film on the skin in which most bacteria are initially entrapped should not be minimized. Whether bacteria migrate from the surface film into skin crevices as more water is absorbed by the skin remains to be shown. The nature of the surface film also bears further investigation. It is possible that elimination of the formation of the surface film on poultry skin in the first 15 s of water immersion might greatly reduce the numbers retained on the skin following prolonged immersion, if indeed migration from the film to skin crevices does occur. Preventing the formation of the surface film by altering surface tension may reduce carcass contamination during immersion processes or may improve access of bacterial cells to skin surface and increase attachment.

6 454 LILLARD ACKNOWLEDGMENTS I thank R. L. Wilson for the statistical analyses and M. O. Carson for his capable technical assistance. REFERENCES 1. Duguid, J. P., and R. R. Gilles Fimbriae and adhesive properties in dysentery bacilli. J. Pathol. Bacteriol. 74: Lillard, H. S Bacterial cell characteristics and conditions influencing their adhesion to poultry skin. J. Food Prot. 48: Lillard, H. S The Role of fimbriae and flagella in the attachment of Salmonella typhimurium to poultry skin. J Food Sci 51:54-56, Notermans, S., and E. H. Kampelmacher Attachment of some bacterial strains to the skin of broiler chickens. Br. Poult. Sci. 15: Notermans, S., and E. H. Kampelmacher Heat destruction of some bacterial strains attached to broiler skin. Br. Poult. Sci. 16: McMeekin, T. A., and C. J. Thomas Retention of bacteria on chicken skin after immersion in bacterial suspensions. J. Appl. Bacteriol. 45: Thomas, C. J., and T. A. McMeekin Effect of water immersion on the microtopography of the skin of chicken carcasses. J. Sci. Food Agr. 33: Thomas, C. J., and T. A. McMeekin Effect of water uptake by poultry tissues on contamination by bacteria during immersion in bacterial suspensions. J. Food Prot. 47; Sanchis, con't. from p Jones, B. D Aflatoxin and related compounds, occurrence in food and feed. pp In T. D. Wyllie and L. G. Morehouse (eds.), Mycotoxic fungi, mycotoxins, mycotoxicoses: An encyclopedic handbook, vol. 1. Marcel Dekker, New York. 14. Roberts, B. A., and D. S. P. Anderson Detection of twelve mycotoxins in mixed feedstuffs, using a novel membrane cleanup procedure. J. Assoc. Off. Anal. Chem. 58: Stoloff, L., S. Nesheim, L. Yin, J. V. Rodricks, M. Stack, and A. D. Campbell A multimycotoxin detection method for aflatoxins, ochratoxins, zearalenone, sterigmatocystin and patulin. J. Assoc. Off. Anal. Chem. 54:91-97.