Use of Baird-Parker's Medium to Enumerate Staphylococcus aureus in Meats 1

Transcription

1 583 Journal offood Protection, Vol. 44, No.8, Pages (,4ugust 1981) Copyright, International Association of Milk, Food, and Environmental Sanitarians Use of Baird-Parker's Medium to Enumerate Staphylococcus aureus in Meats 1 M. E. STILES2"' and L. -K. NG Faculty ofhome Economics and Department o,(microbiology, The University ofalberta, Edmonton, Alberta, Canada T6G 2M8 (Received for publication October 29, 1980) ABSTRACT The Staphylococcus aureus count was determined, using Baird-Parker (BP) medium and coagulase reaction, on 100 samples of ground beef, 23 frozen pork sausage samples and 140 retail meat cuts (beef and pork). counts on ground beef ranged from < 100 to 4,500 per g, and 16% had counts of 1,000 per g or greater. Frozen pork sausage samples had similar counts to ground beef. Retail meat cuts generally had counts less than 100 per cm 2 Typical colonies in this study on BP medium were dark grey to greyblack, and not shiny black as generally described. Type II colonies (without egg yolk clearing) did not contribute markedly to the totals. au reus count and egg yolk clearing type I colony count is suggested for routine estimation of the S. aureus count. The S. au reus isolates in this study generally had 3+ coagulase reactions. The coagulase-positive isolates were generally phosphatase-positive (96.8%), Voges-Proskauer testpositive (97.4%), DNase-positive (98.7%) and used glucose and mannitol oxidatively and fermentatively (93.5%). Most non-s. aureus colonies were "other" staphylococci or micrococci. These isolates had variable phosphatase, Voges-Proskauer and DNase reactions, generally with at least one of these three tests negative. Many selective media have been developed for enumeration of Staphylococcus aureus in foods (15,25). Among these, Baird-Parker's egg yolk-tellurite-glycinepyruvate agar (2,12), with its combined diagnostic and enumerative function, has received widespread acceptance. Typical growing as type I colonies on Baird-Parker medium are described as convex. entire, shiny black colonies, surrounded by a clear zone extending into the opaque medium (2). Type I colonies may also be surrounded by a halo of white precipitation in the cleared zone around the colonies (2,10), which is attributed to formation of calcium and magnesium salts of free fatty acids (26). Similar colonies without the zone of clearing are designated as type II and are tested with the type I colonies for ability to produce coagulase. Black. mucoid colonies larger than 2 mm in diameter, 1 Supported by Research Contract with Agriculture Canada. 'Honorary Pro.fessor, Department o,(microbiology. swarmers and non-black colonies are not counted. Such colonies usually belong to the genus Bacillus (2). in foods represent a potential food poisoning hazard (I 4-18), but at levels of 100 per g reported for refrigerated and frozen ground beef (20) no food poisoning hazard exists. Furthermore, at concentrations of 100 per g, Baird-Parker medium is being used at the lowest limit of colony detection. The level of contaminating bacteria, when using Baird Parker medium for meat samples (personal observation), and the selectivity of this medium (I 1) might adversely affect the reliability of Baird-Parker medium for enumeration of in meats. The increasing emphasis on microbiological regulations for food products prompted this study of levels of in ground beef and other meats. The principal objective of this study was to assess the diagnostic potential of Baird-Parker medium for naturally occurring in meats. MATERIALS AND METHODS Sampling Samples of ground beef, frozen pork sausages and beef and pork cuts were obtained at retail market outlets for bacteriological analysis. Comminuted samples were prepared for analysis by blending 11 g in 99 ml of sterile, 0.1 o/opeptone water. Meat cuts were sampled, using the spray gun equipment described by Clark (8). Bacteriology Appropriate dilutions of meat samples were inoculated (0.2 ml) onto prepoured plates of Baird-Parker (BP) medium (12). Plates were streaked, using a sterile, glass "hockey-stick.'' and incubated at 35 C for 48 h. The colony types were counted differentially, based on egg yolk clearing and morphology of the colonies on the BP plates (see results). Representative samples of each colony type were randomly picked and inoculated into Difco Brain Heart Infusion (BHI) broth and incubated at 35 C for 24 h. BHI cultures were used for the coagulase test, using Difco EDTA coagulase plasma in the tube technique (1). Coagulase reaction at 37 C was read at 1, 2, 4 and 6 h and recorded as 4+, 3+, 2+ (for the different degrees of organized clot formation) or 1+ or 0 (for unorganized precipitation or no change), as described by Sperber and Tatini (24). I den tilication tests lso:ates were subjected to biochemical tests to differentiate them as. "other" staphylococci, and micrococci, as shown in Table 1. JOURN.4L OF FOOD PROTECTION, VOL 44. AUGUST 1981

2 584 STILES AND NG and non-micrococcaceae and gram-positive rods (3,4,5.9,19,22). Gram-positive, catalase-positive cocci were streaked onto Difco DNase agar (23) and Phosphatase agar (6). The isolates were also inoculated into MRVP medium (Difco), to determine the Voges-f'roskauer reaction (7). Oxidation-fermentation (0/F) tests (13) for glucose and mannitol were carried out. Colonies showing an intermediate coagulase reaction of 2+. or atypical biochemical reactions for, were confirmed ass. au reus. using the thermostable nuclease test (21). TABLE 1. Identification criteria for gram-positive, catalasepositive cocci Baird-Parker (BP) medium. Tests Coagulase + Phosphatase usually+ VPTest + DNase usually+ 0/F glucose +I+ 0/F mannitol +!+or+!- Organisms Other staphylococcia usually +!- usually - +I+ +I- or-/- Micrococcus spp. usually +I- +I- or-/- +1-or-1- as. epidermidis and S. saprophyticus are classified as "other staphylococci." RESULTS Many different colony types grew on BP medium plates when they were surface-streaked with 1 :SO dilutions of meat samples for enumeration of. Colonies were classified as type I if they cleared, and as type II if they failed to clear the egg yolk. A total of five colony types were identified that gave type I clearing of egg yolk. The description of these colony types is as follows: (a) IA typical type I colonies, as described by the International Commission on Microbiological Specifications for Foods (ICMSF) (14), except that their color would more accurately be described as dark grey to greyblack on these plates; (b) IB - dark grey to grey-black. dull with a white margin, (c) IC - light grey to grey, smooth, shiny with a white margin. (d) ID - brownish grey to grey-black, rough, irregular edge and (e) IE- brownish black to black, shiny, wet appearance. The same colony types could be found as type II colonies. and in addition: (f) IIF - black. shiny colonies without narrow white margins, fitting the ICMSF description of typical type II colonies (14). The frequency of occurrence of the different colony types from ground beef samples is shown in Table 2. The predominating colony type was IIC. often outgrowing other colony types on the plates. Type la was the predominating type I colony. and was the principal type contributing to the count. Of the 100 ground beef samples. 16% had counts that exceeded per g. The maximum count on these ground beef samples was 4,500 per g. Similar data for 23 samples of frozen pork sausage are shown in Table 3. Again type IIC was the predominant colony type. and type ID increased in relative importance because of high counts on two of the samples. counts were also determined on retail meat cuts, including beef steaks and pork chops (see Table 4). The predominant colony types on the retail cuts were types lib and IIC. Most samples had less than 100 S. TABLE 2. Frequency distribution and mean count of' different colony types on Baird-Parker (BP) medium from 100 samples ofground beej: Number of samples with count types < < < 1,000 ~ 1,000 IA B c D E HA B c D E F 96 2 count TABLE 3. Frequency distribution and mean count of different colony types on Baird-Parker (BP) medium from 23 samples offrozen pork sausage. Number < < < 1,000 ~ 1,000 IA B c D E lia B c D E F count TABLE 4. Frequency distribution and mean count of dif.ti rent colony types on Baird-Parker (BP) medium from samples (!(retail meat cuts (beefand pork). Number of samples with count per cm 2 < < < ~ LOOO IA B c D E IIA , B I c D E F count JOURNAL OF FOOD PROTECTION. VOL. 44. AUGUST 1981

3 aureus per cm 2, and counts only exceeded 500 per cm2 on one sample. Correlation coefficients for coagulasepositive with type I colony counts on BP medium were r= for comminuted meats and r= for meat cuts. Random samples of206 type I and 227 type II colonies were isolated from the BP plates for further study. The data in Table 5 show the coagulase reactions of these isolates. Most coagulase-positive isolates had a 3+ coagulase reaction, only 7o/o had 4+ reactions, and 14o-/o had 2+ reactions. The principal coagulase (2+ to 4+) producing type was IA, of which 139 (85.3%) were coagulase-positive. The principal coagulase-producing type II colony was IIA (11 of 42 isolates were coagulasepositive). types ID and IE were gram-positive rods, as were types lid and IIE. Type IIF isolates were typical of colonies on BP medium, as described by the ICMSF, but they were not. These isolates were all coagulase-, phosphatase-, VP- and thermostable nuclease-negative. None of them could use mannitol anaerobically. TABLE 5. Coagulase reactions of 206 Baird-Parker type I and 227 type II Coagulase reaction IA IB IC ID IE Negative IIA lib IIC lid lie IIF Negative l a Type 1, egg yolk clearing; Type 2, no egg yolk clearing. See text for description of alphabetic codes. Other biochemical reactions of the gram-positive, catalase-positive cocci are shown in Tables 6 and 7. From the data in Table 6 it can be seen that 148 of 155 (95.5%) TABLE 6. (BP) ENUMERATING STAPHYLOCOCCI IN MEATS 585 coagulase-positive isolates were phosphatase-, VP- and DN ase-positive. Of 155 coagulase-positive (2+ to 4+) isolates, 96.8% gave positive phosphatase reaction, 97.4% gave positive VP test, and 98.7% were DNasepositive. Most coagulase-negative isolates had at least one of the above tests negative. Only 6 of 262 (2.3o/o) coagulase-negative isolates were positive for all three of these biochemical reactions. Coagulase-negative reactions did not correlate well with phosphatase, VP or DNase reactions, whereas only 5 phosphatase-negative, 2 VP-negative and 2 DNase-negative isolates had coagulase-positive reactions. From the data in Table 7 it can be seen that coagulase-positive isolates were generally able to oxidize and ferment both glucose and mannitol. A few coagulase-positive isolates (6.5%), however, failed to ferment mannitol. In contrast, the coagulase-negative isolates had variable glucose and mannitol reactions, including a large proportion that used glucose and mannitol both aerobically and anaerobically. Of 95 coagulase-negative isolates that were 0/F glucose- and mannitol-positive, only two isolates were positive for all three of the phosphatase, VP and DNase reactions. Based on the criteria shown in Table 1, isolates that had coagulase 3+ and 4+ reactions were classified as S. aureus. Isolates that had coagulase 2+ reaction, and other atypical, were confirmed as by the thermostable nuclease test. Type IA colonies were generally, including the isolates with a 2+ coagulase reaction. Only 24 of the 163 type IA isolates were not, and 18 of these coagulase-negative type IA isolates produced a zone of precipitation around the colony, within the zone of egg yolk clearing. Production of this zone of precipitation was a variable characteristic between colonies on subculturing in pure culture. These coagulase-negative isolates appeared to be either S. epidermidis or S. saprophyticus. types IB and IC could generally be classified as "other" staphylococci. However, the coagulase-positive 2+ type IC colony was confirmed as by thermostable nuclease test. The eight type IIA colonies with coagulase reaction 3+ had typical phosphatase, VP and DNase reactions for S. Relationship between coagulase. phosphatase, Voges Proskauer (VP) and DNase reactions a_( isolates on Baird-Parker meats. Test Phosphatase VP DNase Reaction ; ;- + ;- + Coagulase reaction Negative 1+ 2-r Number oflsolates JOURNAL OF FOOD PROTECllON. VOL. 44, AUGUST 1981

4 586 STILES AND NG TABLE 7. Relationship benveen coagulase reaction and ability of isolates to utilize glucose and mannitol aerobically and lltltue rlrjtjl,<:llll 0/F and mannitol a ++++ Coagulase reaction others Number of isolates Negative arepresentation of 0/F glucose and mannitol reactions: glucose oxidation and fermentation, mannitol oxidation and fermentation reactions, positive (+),negative (-). aureus and used glucose and mannitol oxidatively and fermentatively. Of the three coagulase-positive 2+ strains. two were typical as described above, the other was atypical, not producing acid from mannitol, and it was VP-negative. However, this strain was confirmed as by positive thermostable nuclease test. All other type IIA colonies that failed to produce coagulase were lacking at least one of the typical reactions, and a representative sample was shown not to be by negative thermostable nuclease test. Type lib and IIC colonies included five isolates that had coagulase reaction 2+ to 4+. All of these organisms, except one 2+ colony, were confirmed as S. au reus by the identification tests. including thermostable nuclease. The remaining type lib and IIC colonies were "other" staphylococci, except for 17 isolates which were identified as micrococci, based on their inability to oxidize or ferment glucose and mannitol. DISCUSSION AND CONCLUSIONS The numbers of per g of ground beef observed for 100 samples analyzed in this study were higher than those reported in an earlier Canadian survey (20). Even at these somewhat higher levels, in ground beef still did not represent a food poisoning hazard. However, it could be a source of in food handling and preparation. counts for frozen pork sausage samples, prepared and packaged at federally inspected packing plants, were similar to those observed for the ground beef samples, suggesting that no marked increases in counts occurred in meats due to handling at the retail level. Surface counts on retail meat cuts could not be related to counts for ground beef and pork sausage because of the determination of the count by surface area as opposed to weight. Yet the counts on the meat surfaces were considered to be appropriate for S. au reus counts on meats, because in an earlier study (Ng and Stiles. unpublished data) it was apparent that where Escherichia coli was generally present in comminuted meats at 100 to 1,000 per g, it occurred on meat surfaces at about 100 per cm 2 The results of this study indicated that for routine quality assurance, the coagulase-positive count on meats could be estimated from the egg yolk clearing colony count on BP medium. However, if a more complete estimation of coagulase-positive staphylococci is required, the estimation of the total population would require that any egg yolk clearing colonies, and any type II colonies ranging from a dark grey to black color should be included in the estimate. In these studies, where coagulase-positive type II colonies were found, they resulted in a marked increase in the S. aureus count, because of the way in which the count is computed (14), and the fact that type IIA occurred in some ground beef samples in relatively large numbers. Neither types lib nor IIC colonies described in this study (dark grey to grey-black, dull with white margin, or light grey to grey, smooth, shiny with a white margin) warranted checking for coagulase production. The differential colony counts on BP medium indicated that for these samples the "typical" colonies (types IA and IIA) were dark grey to grey-black, as opposed to the shiny black colonies described by other workers and texts (2,14). This might be due to the laboratory preparation of the BP medium; however. this color difference remained consistent throughout study, both for initial enumeration and on subsequent re-checking of the colony type on BP medium. In fact, typical colonies that could be described as "black, shiny" (type IIF in this study) were not. In many instances, typical colonies were outgrown by other colonies on the BP medium, notably by type II C. a light grey to grey colony, smooth and shiny with a white margin. Their identification as "other" staphylococci or micrococci, except for one colony out of 126 which was S. aureus. did not warrant checking of this colony type as part of the potential S. au reus count. In some instances, the incidence of this colony type was sufficient to cause difficulty in counting egg yolk clearing colonies on BP medium. This most common coagulase reaction observed with these isolates was 3+. This contrasts with 4+ reactions for reported by Sperber and Tatini for food, human and mastitis isolates (24). Phosphatase production and Voges-Proskauer (VP) reaction have also been cited as important reactions confirming (4,6). The classification in Table 1 was based on Baird-Parker (3.4.5) and Cowan and Steel (9). Mossel (19) cited anaerobic dissimilation of mannitol and coagulase production as the best criteria for distinguishing S. aureus from other catalase-positive cocci, and recommended that coagulase-negative or anaerobic mannitol-negative strains be classified as Staphylococcus epidermidis or Staphylococcus saprophyticus. In this study. S. epidermidis and S. saprophyticus were considered as those strains that failed to use mannitol aerobically and anaerobically. Those isolates that failed to use mannitol anaerobically were principally "other" JOURNAL OF FOOD PROTECTION. VOL. 44. AUG CST 1981

5 ENUMERATING STAPHYLOCOCCI IN MEATS 587 staphylococci, but 30o/o of these isolates produced coagulase and were classified as S. au reus. This might be attributable to slow or weak dissimilation of mannitol anaerobically by (11,19). The findings of this study indicated that concentrations on these raw meats were sufficiently low that their routine determination could not be justified. However, if counts in meats are regulated, BP type I counts would suffice for routine screening. S. aureus isolates in this study included strains with coagulase reactions from 2+ to 4+, suggesting a wider range of types than might be anticipated from direct human or animal origin. ACKNOWI,EDGMENT The authors thank Angeline Benjamin for her technical assistance with the enumeration and isolation of the cultures in this study. REFERENCES 1. Baer, E. F Isolation and enumeration of Staphylococcus aureus: Review and recommendations for revision of AOAC method. J. Assoc. Off. Anal. Chern. 54: Baird-Parker. A. C An improved diagnostic and selective medium for isolating coagulase-positive staphylococci. J. Appl. Bacteriol. 25: Baird-Parker, A. C A classification of micrococci and staphylococci based on physiological and biochemical tests. J. Gen. Microbiol..30: Baird-Parker. A. C The classification of staphylococci and micrococci from world-wide sources. J. Gen. MicrobiaL 38: Baird-Parker, A. C Micrococcaceae. pp Bergey's Manual of Determinative Bacteriology, 8th ed. Williams and Wilkins Co., Baltimore. 6. Barber, M., and S. W. A. Kuper Identilication of Staphylococcus pyogenes by the phosphatase reaction. J. Pathol. Bacteriol. 63: Barritt. M. M The intensification of the Voges-Proskauer reaction by the addition of alpha-naphthol. J. Pathol. BacterioL 42: Clark, D. S Improvement of spray gun method of estimating bacterial populations on surfaces. Can. J. MicrobioL 11: Cowan. S. T.. and K. J. Steel Characters of gram-positive bacteria. pp In Manual for the identification of medical bacteria, 2nd ed. Cambridge University Press. 10. Crisley, F. D.. J. T. Peeler. and R. Angelotti Comparative evaluation of five selective and differential media for the detection and enumeration of coagulase-positive staphylococci in foods. Appl. Microbiol. 13: De Waart. J. D. A. A. Mossel, R. ten Broeke and A. van de Moosdijk Enumeration of Staphylococcus aureus in foods with special reference to egg-yolk reaction and mannitolnegative mutants. J. Appl. Bacteriol. 31: Holbrook, R., J. M. Anderson and A. C. Baird-Parker The performance of a stable version of Baird-Parker's medium for isolating Staphylococcus aureus. J. Appl. Bacteriol..32: Hugh, R., and E. Leifson The taxonomic significance of fermentative versus oxidative metabolism of carbohydrates by various gram-negative bacteria. J. Bacteriol. 66: International Commission on Microbiological Specifications for Foods Microorganisms in foods.!. Their significance and methods of enumeration. 2nd edition. International Association of Microbiological Societies. University of Toronto Press, Canada. 15. Minor, T. E., and E. H. Marth Staphylococcus aureus and staphylococcal food intoxications. A review. L The staphylococci: characteristics, isolation, and behavior in artificial media. J. Milk Food Techno!. 34: Minor, T. E., and E. H. Marth Staphylococcus aureus and staphylococcal food intoxications. A review. II. Enterotoxins and epidemiology. J. Milk Food Techno]. 35: Minor, T. E., and E. H. Marth Staphylococcus aureus and staphylococcal food intoxications. A review. III. Staphylococci in dairy foods. J. Milk Food Techno!. 35: Minor, T. E., and E. H. Marth Staphylococcus aureus and staphylococcal food intoxications. A review. IV. Staphylococci in meat, bakery products, and other foods. J. Milk Food Techno!. 35: Mossel, D. A. A Attempt in classification of catalasepositive staphylococci and micrococci. J. Bacteriol. 84: Pivnick. H., I. E. Erdman, D. Collins-Thompson. G. Roberts, M. A. Johnston, D. R. Conley. G. Lachapelle, U. T. Purvis, R. Foster, and M. Milling Proposed microbiological standards for ground beef based on a Canadian survey. J. Milk Food Techno!. 39: Rayman, M. K., C. E. Park, J. Philpott, and E. C. D. Todd Reassessment of the coagulase and thermostable nuclease tests as means of identifying Staphylococcus aureus. Appl. Microbiol. 29: , 22. Shaw. C., J. M. Stitt, and S. T. Cowan Staphylococci and their classification. J. Gen. Microbiol. 5: Smith, P. B., G. A. Hancock, and D. L. Rhoden Improved medium for detecting deoxyribonuclease-producing bacteria. Appl. Microbiol.l8:99! Sperber. W. H., and S. R. Tatini Interpretation of the tube coagulase test for identification of Staphylococcus aureus. Appl. Microbiol. 29: Stiles, M. E., and P. C. Clark The reliability of selective media for the enumeration of unheated and heated staphylococci. Can. J. Microbiol. 20: Tirunarayanan. M. 0., and H. Lundbeck Investigations on the enzymes and toxins of staphylococci. Acta Pathol. Microbiol. Scand. 69: JOURNAL OF FOOD PROTECTION, VOL. 44, AUGUST 1981