(Ostrolenk and Hunter, 1946) revealed that approximately 7 per cent of the spec- 197

Transcription

1 COMPARATIVE STUDIES OF ENTEROCOCCI AND ESCHERICHIA COLI AS INDICES OF POLLUTION MORRIS OSTROLENK, NORMAN KRAMER, AND ROBERT C. CLEVERDON U. S. Food and Drug Administration, Washington, D. C. Received for publication October 3, 1946 The presence of coliform bacteria in water and in or on other foods depends on many environmental factors, each of which must be determined separately for each item of food in question. When a correlation is found to exist between insanitary methods of food production and the presence of Escherichia coti in the finished product, the presence of confirmed fecal strains of coliform organisms can be interpreted as having sanitary significance. Under these circumstances, however, if environmental factors tend to permit or promote multiplication of the normal flora of organisms, together with the coliform contaminants, an interpretation of the sanitary significance of E. coli becomes involved. Despite these limitations, the coliform bacteria have long been considered "the only bacterial group which can be used with even a reasonable degree of accuracy as a measure of pollution" (Hunter, 1939). It is not improbable that intestinal microorganisms, other than E. coli, have not been investigated because of technical difficulties. With the development of "SF" medium (Hajna and Perry, 1943) and the application of a few taxonomic requirements (Sherman, 1937), the enterococci represent a group of intestinal microorganisms which can now be readily isolated and identified. The applicability of the enterococci as an index of pollution thus merits consideration. It has been demonstrated that enterococci are normally present in the intestinal tract of man and many warm-blooded animals (Sherman, 1937; Ostrolenk and Hunter, 1946). The work of the latter demonstrated that, in. 37 per cent of 51 fecal specimens examined, enterococci occurred in equal or in greater numbers than E. coli. In the remaining -63 per cent of the specimens, E. coli exceeded enterococci numerically by from one to five decimal dilutions. The lower number of enterococci, as compared with E. coli, in human and animal feces does not in itself necessarily minimize the potential sanitary significance of fecal streptococci. It does, in fact, support the group as a more reliable index of pollution, provided there is a significant relationship between the presence of the organism and recognized pollution sources. Other factors that must be considered include distribution and reproductivity in nature, survival of temperature and chemical treatments, and longevity. METHODS Comparative studies were made of 53 specimens of commercial black walnut and pecan meats and fresh animal feces to determine the most efficient method and medium for the recovery of enterococci. Previous studies with "SF" broth (Ostrolenk and Hunter, 1946) revealed that approximately 7 per cent of the spec- 197 Downloaded from on October 11, 218 by guest

2 198 M. OSTROLENK, N. KRAMER, AND R. C. CLEVERDON [vol. 53 imens examined gave false positive reactions. It was therefore anticipated that similiar.results would be obtained on the examination of soils and washings from commercial foods. As a consequence, Hajna and Perry's "SF" broth was modified to contain 6.5 per cent sodium chloride,.4 g K2HPO4, and.15 g KH2PO4 per liter, and bromthymol blue indicator was substituted for bromcresol purple. This modified medium was evaluated against the original "SF" medium with incubation at 45.5 C, both in a constant temperature water bath and in a water-jacketed incubator. Preincubation at 37.5 C was also tried. Daily examinations of all inoculated tubes for 72 hours revealed that the method most productive of fecal streptococci consisted of the use of Hajna and Perry's unmodified "SF" broth for primary enrichment, incubated at 45.5 C in a waterjacketed incubator for 24 and 48 hours. Broth tubes showing acid production were streaked on "SF" agar ("SF" broth plus 1.5 per cent agar) and the plates were then incubated at 45.5 C for 24 or 48 hours. Five hundred and thirty-one colonies, which produced acid on "SF" agar plates and consisted of gram-positive streptococci, were isolated for study. These subcultures conformed to the taxonomic classification for enterococci (Sherman, 1937), i.e., possessing the ability to grow at 37.5 C in glucose broth containing 6.5 per cent sodium chloride, in alkaline glucose broth with a ph of 9.6, and in milk containing.1 per cent methylene blue, and to acidify and coagulate litmus milk. No culture produced typical hemolysis on Casman's blood agar (Casman, 1946) incubated in a candle jax at 37.5 C. Following the development of the methods described above, data were accumulated on the longevity of enterococci and E. coli in soil, feces, and on foods. Studies were also made to determine the occurrence of these organisms on various commercial foods and their distribution in food-producing establishments. Parallel coliform determinations were made in the conventional manner using standard lactose broth followed by streaking on Levine's eosin methylene blue agar. Typical, well-isolated colonies were fished for confirmation in"imvic" media. EXPERIMENTAL Survival studies. To determine the longevity of enterococci as compared with E. coli under varying conditions of storage and from various sources, the following survival studies were conducted: Fresh rat and mouse feces were stored in 14-mm open petri dishes at room temperature and at 7.22 C (45 F). Virgin soil and soil fertilized with chicken manure were each thoroughly mixed and then stored in open 5-ml glass beakers at room temperature and at 7.2 C. Five pounds each of pecan meats and soil were artificially contaminated with 18-hour washed suspensions of E. coli and enterococci isolated from rat feces. The pecan meats and soil were then stored in large, flat, shallow, open pans at room temperature. Fifty-gram amounts of each of the soil and pecan specimens and 5-gram amounts of each of the fecal specimens were examined periodically for the test Downloaded from on October 11, 218 by guest

3 1947] STUDIES OF ENTEROCOCCI AND ESCHERICHIA COLI 199 organisms. The soil fertilized with chicken manure ("B" in table 1) contained initially more than 1 enterococci but less than 1, per gram, and E. coli was demonstrated in a maximum of 1/1,-gram portion of the soil washings. Eleven examinations of this specimen, covering a period of 16 days, revealed no fecal streptococci after 21 days in the portion stored at room temperature ("B"), TABLE 1 Longevity of enterococci and Escherichia coli in soil, in animal feces, and on food NUMBER OF DAYS IN STORAGE ig 19 1 g ' 1-2 log 13 o O log log 1 g 1 g 1 g g 1_ ' 1-' l 1 g 1 g 1-' ' g ' ' ' ' 1-' 1-1-' 1-' 1' Figures (1, 1, 1-1, etc.) represent dilution (per g) at which organism was found; means not found in 1 g. but E. coli was present in 1-gram amounts of soil on the 66th day. In the second portion of the same soil ("B-1"), stored at 7.2 C, both test organisms were recovered on the 19th day but not on the 123rd day. In the unfertilized soil, stored at room temperature ("A"), no E. coli could be demonstrated after the initial examination prior to storage whereas enterococci were present in 1-gram amounts on the 19th day. In the refrigerated specimen ("A-1"), E. coli Downloaded from on October 11, 218 by guest

4 2 M. OSTROLENK, N. KRAMER, AND R. C. CLEVERDON survived for 66 days and enterococci for 123 days. Soil ("C"), which was artificially contaminated, showed a gradual reduction in numbers of both organisms from an initial count of approximately 1,, per gram to 1 per gram following room storage for 16 days. The pecan meats ("P"), simultaneously contaminated with E. coli and enterococci, showed no apparent reduction in coliform count following 16 days' storage at room temperature. Under the same conditions of storage, the fecal streptococci practically disappeared during this period. The results of the examination of rat and mouse feces stored at room temperature and at 7.2 C ("R", "R-1", "M", and "M-1") indicate that enterococci survive longer than E. coli under such conditions of storage, and that refrigeration has a preserving action on both test organisms (table 1). Commercial foods. Three hundred and eighteen specimens, representing four classes of fresh and frozen foods, were examined for enteroccocci and E. coli: TABLE 2 Incidence of enterococci and Escherichia coli on commercial food8 TYPES OF FRESH AND FROZEN FOODS SUBDMSIONS SHOWING Nut meats Fruits and Frozen Frah cmbmeat berries vegetables Fehcama (No.) (%) (No.) (%) (No.) (%) (No.) (%) No strep.; no E. coli Strep. equal to E. coli Strep.; no E. coli E. coli; no strep Strep. greater than E. coli E. coli greater than strep Totals [VOL domestically shelled nut meat samples (three varieties); 27 specimens of frozen fruits and berries representing seven varieties; 46 frozen vegetable samples (eight varieties); and 96 specimens of fresh-picked crabmeat representing claw meat, white meat, backfin lump, and fancy pack. The nut meat and fresh crabmeat specimens were examined shortly after commercial packaging, whereas the specimens of frozen fruits, berries, and vegetables were examined following storage of these products of from several months to more than one year. Neither coliform bacteria nor enterococci were isolated from 16 per cent ofthenutmeat specimens and 25 per cent of the crabmeat specimens; 7.3 per cent of the former specimens and 9.5 per cent of the latter specimens contained approximately the same number of E. coli and enterococci, whereas 53.3 per cent of the nut meat and 58.3 per cent of the crabmeat specimens contained enterococci without any evidence of E. coli. Both types of food showed 2 per cent of the specimens to contain E. coli but no demonstrable enterococci. In 21.3 per cent of the nut Downloaded from on October 11, 218 by guest

5 1947] STUDIES OF ENTEROCOCCI AND ESCHERICHIA COLI 21 meat specimens and 5.2 per cent of the crabmeat specimens both fecal streptococci and E. coli were found, the enterococci occurring in larger numbers. Total bacterial counts were also obtained on all the specimens examined. The esti- TABLE 3 Incidence of Escherichia coli and enterococci on surfaces in crabmeat plants CEBMEAT ESTABLISM ENT NUMBER CULTURETFRO E.c. FS EB. FS E.c. FS E.c. FS EBc. FS E.c. FS E.c. FS Pickers' hands Picking knife Picking table Crabmeat. + + _ Pickers' hands Picking knife Picking table Crabmeat. + + _ + Pickers' hands Picking knife Picking table _- Crabmeat._.... Pickers' hands _ Picking knife _ _ Picking table _ Crabmeat._... Pickers' hands + + Picking knife Picking table + + Crabmeat _ + Packers' hands _ Packing table Crab shovel Cooling floor Crab bucket Misc. surfaces Total samples Designations: E.c. (E. coli); FS (enterococci); + (positive); - (negative). mates of microorganisms ranged from less than 1, to greater than 2 million per gram. Fifty-six of the nut meat samples that contained enterococci and coliform bacteria but no E. coli had total bacterial counts of 3, or more per gram of specimen. Downloaded from on October 11, 218 by guest

6 22 M. OSTROLENK, N. KRAMRR,. C. CLEVERDON Of the 26 specimens of frozen fruits and berries examined, 3 (11.5 per cent) contained enterococci but no E. coli; 1 (3.9 per cent) contained E. coli but no fecal streptococci. The remaining 22 specimens (84.6 per cent) contained neither enterococci nor coliform bacteria. Thirty-eight (82.6 per cent) of the frozen vegetables contained enterococci alone; 2 (4.4 per cent) contained fecal streptococci and E. coli; and the remaining 6 (13 per cent) contained neither organism (table 2). Food establishment specimens. Seven crabmeat-producing establishments were studied for the presence of E. coli and enterococci. The following specimens were collected for examination: washings from the hands of the operators, swabbings from representative surfaces of the equipment used in picking and packing crabmeat, wash water, hand and crabmeat can chlorine dip waters, the cooling ice employed in the manufacturing process, and picked crabmeat. Of 117 specimens examined by the methods described, 45 (38.4 per cent) contained neither E. coli nor enterococci; 12 (1.3 per cent) contained no E. coli but did contain enterococci; 4 (3.5 per cent) contained E. coli but no fecal strepto- TABLE 4 Relative incidence of Escherichia coli and enterococci on surfaces in crabmeat plants SPECIENS SIHOWNG NUMBR SPEcJIS Escherichia coli negative; enterococci negative (38.4%) Escherichia coli negative; enterococci positive (1.3%) Escherichia coli positive; enterococci negative... 4 ( 3.5%) Escherichia coli positive; enterococci positive (47.8%) [VOL. 53 Total number of specimens examined Number of specimens polluted (index enterococci) (58.1%) Number of specimens polluted (index E. coli)... 6 (51.3%) cocci; and 56 (47.8 per cent) contained both types of microorganisms (tables 3 and 4). It should be noted (table 3) that in the food establishments numbered "3" and "7" an excellent correlation between sanitary conditions and the presence of enterococci and E. coli was obtained. Although both establishments possessed essentially all the equipment necessary to produce a wholesome, unpolluted product, rodent infestation, poor supervision, and lack of personal hygiene in plant number "3" resulted in the, pollution of all surface items examined. In plant "7", good control over rodent contamination, proper supervision of operations, and the control of personal hygiene among the employees resulted in the elimination of both E. coli and enterococci in the specimens examined. The correlation between plant sanitation and the presence of intestinal types of bacteria on contact surfaces is further reflected in the data shown in table 4. Sixty specimens (51.3 per cent) were found to be polluted as judged by the preseiice of E. coli, whereas 68 specimens (58.1 per cent) were polluted when enterococci were used as the index of pollution. Downloaded from on October 11, 218 by guest

7 1947] STUDIES OF ENTEROCOCCI AND ESCHERICHIA COLI 23 SUMMARY A method is described for the isolation of fecal streptococci, employing incubation at 45.5 C in.5 per cent sodium azide broth ("SF" broth) as a primary enrichment, followed by streaking on the same medium solidified by 1.5 per cent agar ("SF" agar), and incubating the plates at 45.5 C. By this method, 531 cultures were studied (from nut meats, frozen fruits and berries, frozen vegetables, and freshly packed crabmeat) and found to be predominantly Streptococcus faecalis or Streptococcus liquefaciens. Enterococci in artifically contaminated soils and pecan meats and in normal feces appear to survive longer than Escherichia coli under identical conditions of storage. The adverse effect of high total bacterial counts on the isolation of Escherichia coli did not obtain with the use of "SF" media to isolate enterococci. It is suggested that this observation enhances the value of fecal streptococci as an index of pollution. Data obtained from food-producing establishments show excellent correlation to exist between sanitation and recovery of both Echerichia coli and enterococci. REFERENCES CASMAN, E. P Studies on blood agar. 1. An improved blood agar base. J. Bact., 51, 126. HAJNA, A. A., AND PERRY, C. A Comparative study of presumptive and confirmative media for bacteria of the coliform group and for streptococci. Am. J. Pub. Health, 33, HUNTER, A. C Uses and limitations of the coliform group in sanitary control of food production. Food Research, 4, OSTROLENK, MORRIS, AND HUNTER, A. C The distribution of enteric streptococci. J. Bact., 51, SEIRMAN, J. M The streptococci. Bact. Revs., 1, Downloaded from on October 11, 218 by guest