Thermal Inactivation of Campylobacter Species, Yersinia enterocolitica, and Hemorrhagic Escherichia coli 0157:H7 in Fluid Milk

Heat treatment of raw milk in an HTST pasteurizer operated at 60.0 to 72.0°C for a minimum holding time of 16.2 s rapidly inactivated mixtures of hemorrhagic Escberichia coli 0157:H7, Yersinia enterocolitica and Campylobacter spp. (C. fetus, C. coli, and C. je]uni). Each of the three genera in the mixture was inoculated at a level of approximately 1.0 × l0 s cfu/ml. At 60.0°C, hemorrhagic E. coli showed a maximum 2 log10 reduction in counts and no viability at ~>64.5°C. Yersinia enterocolitica and Campylobacter spp. showed greater heat sensitivity with a 4 log lo reduction in counts at 60.0°C and absence of viable cells at /> 63.0°C. These findings reiterate the need for stringent control of thermal processes in the manufacture of dairy products from raw or heat-treated (non-pasteurized) milk. I N T R O D U C T I O N Recent outbreaks attributed to bovine milk and dairy products contaminated with Campylobacter, Yersinia, Salmonella, and Listeria spp. are disquieting and underline the growing importance of these commodities as vehicles of human infection. Numerous episodes of salmonellosis and campylobacteriosis frequentReceived April 19, 1988. Accepted July 21, 1988. Contribution Number 766. ly involving school children on visits to farms have been associated with the consumption of raw milk in the European and North American continents (11, 20, 31, 32). Thermally processed fluid milk has also figured prominently as a source of human illness where inadequate pasteurization and posttreatrnent contamination were major contributing factors in many incidents (32). More specifically, the largest foodborne outbreak ever to be reported in the United States was linked to improperly pasteurized milk contaminated with Salmonella typhimurium (22). Although the source of contamination was never confirmed, a crosslink between raw and pasteurized milk lines was likely responsible for more than 16,000 confirmed cases of salmonellosis. Pasteurized milk was also implicated in an earlier incident of 49 cases (14 deaths) of listeriosis in the state of Massachusetts (12). In spite of the prevalence and ability of Yersinia enterocolitica to grow in refrigerated raw milk (34), comparatively fewer outbreaks have been associated with this pathogen. Consumption of raw milk by several groups of school children attending a sugar bush party resulted in 138 cases of illness (8). Other reports of milkborne outbreaks of yersiniosis have been limited to thermally processed milk (4, 32, 38) and possibly milk powder (33). Dairy products manufactured from sheep and goat milk are equally susceptible to contamination with pathogenic microorganisms, and their potential impact on public health should not be minimized (24). The propensity for human disease from contaminated cheese products is equally notable. The presence of low numbers of 1988 J Dairy Sci 71:3230-3236 3230 THERMAL INACTIVATION OF PATHOGENS IN MILK 3231 Salmonella heidelberg and Salmonella typhimurium in Cheddar cheese engendered major outbreaks in the United States and Canada, respectively (7, 13). Contamination of a Mexican-type cheese with Listeria monocytogenes was devastating and resulted in no fewer than 150 cases of illness and 84 deaths (1). The presence of Escherichia coli 0157:H7 has been reported in a variety of raw meats (9) and in dairy herds, which may act as reservoirs of hemorrhagic and other verotoxin-producing strains of E. coli (5, 24, 25, 39). Isolated foodborne episodes have been associated with this new pathogen. For example, raw milk contaminated with E. coli 0157:H7 and served to children in three kindergarten classes resulted in an extensive outbreak of gastroenteritis with several cases developing a hemolytic uremic syndrome (24). Ground beef from infected dairy cattle was also incriminated in related restaurant and nursing home outbreaks (25). Widespread use of nonpasteurized milk in cheese production (6, 32) and recent developments in the epidemiology of milk and milk products led to two earlier studies on the thermal resistance of Salmonella and Listeria at subpasteurizing temperatures (6; Farber et al., 1988, unpublished). The present study examines the effect of heat on the survival of hemorrhagic E. coli 0157:H7, 11. enterocolitica, and Campylobacter spp. suspended in raw milk. approximately 1.0 x 10 9 cfu/ml. The mixture (100 ml) containing a total of 1.0 to 2.0 x 1011 cfu was sedimented at 16,000 x g for 15 min, and the pellet was suspended in 50 ml of raw milk (herd 1). Equal portions (25 ml) of the suspension were then added to separate 1-L volumes of raw milk (herd 1) and stored overnight at 4°C. The TSA slant cultures of 15 strains of 11. enterocolitica were suspended separately in physiological saline solution to a final cell concentration of 1.0 to 1.2 x 109 cfu/ml. Portions (10 ml) from each bacterial suspension were then combined and sedimented at 16,000 x g for 15 min. The cell pellet was suspended and equilibrated overnight at 4°C in 1 L of raw milk (herd 1). To prepare Campylobacter spp. inocula, working cultures maintained on Albimi semisolid agar medium were plated on Brucella-FBP agar (Gibco Laboratories) and incubated for 48 h at 35°C under microaerobic conditions (17). Growth on the plate surface was taken up in Brucella broth (Gibco Laboratories) to a final cell concentration of 1.0 to 1.5 x 101° cfu/ml. Equal 10-ml portions from each of the 15 Campylobacter suspensions were combined, sedimented at 16,000 × g, and stored at 4°C in raw milk (herd 1) as previously described. Raw milk inoculated with E. coli, Y. enterocolitica, and Campylobacter spp. and equilibrated overnight at 4°C was combined into a single tank of milk (herd 2) to be heat processed. M A T E R I A L S A N D METHODS Bacterial Cultures and Inocula Ten strains ofE. coli 0157:H7 (Table 1)and 15 strains of IT. enterocolitica obtained from various sources were maintained on tryptic soy (TSA) agar slants stored at room temperature. Working cultures of 15 strains of Campylobacter spp. (Table 2) were stored at 35°C on Brucella agar (Albimi) semisolid medium (Gibco Laboratories, Madison, WI). Raw milk for the preparation of inocula (herd 1) and for thermal processing (herd 2) was obtained from different dairy farms. To prepare raw milk inocula, TSA slant cultures of 10 strains of E. coli were grown separately in brain-heart infusion (BHI) broth for 18 to 20 h at 35°C. Equal portions (10 ml) from each test strain were then mixed to a final cell density of Thermal Processing of Milk Inoculated raw milk was thermally processed under conditions used previously for the inactivation of Salmonella (6). Milk was processed in a regenerative plate HTST pasteurizer operated in the temperature range of 60 to 72°C and a minimum holding time of 16.2 s. At the beginning of each trial, a 1-L sample of raw milk was withdrawn from the bulk tank for determination of aerobic colony counts (ACC) and quantitation of E. coli and hemorrhagic E. coli 0157:H7 by direct plating techniques (15, 16, 36). Inocula of test organisms previously prepared in raw milk (herd 1) and stored overnight under refrigeration were then added to the bulk tank containing approximately 1200 L of raw milk (herd 2). The inoculated milk was stirred mechanically for 10 min and a sample taken for enumeration of test organisms. Journal of Dairy Science Vol. 71, No. 12, 1988