VOLUME 54 (1), 1999

 TITLE: PHENOTYPIC CHARACTERISTICS OF STAPHYLOCOCUS AUREUS STRAINS ISOLATED FROM MILK AND DAIRYMEN ON DAIRY FARMS IN TRINIDAD  

AUTHOR:   A.A. Adesiyun1, L.A. Webb1 and H. Romain2  
    1. School of Veterinary Medicine, Faculty of Medical Sciences, The     University of the West Indies, St. Augustine, Trinidad
    2. Veterinary Public Health Unit, Ministry of Health, Port of Spain, Trinidad 

Read it all      Introduction      Materials and Methods      Results     Discussion

Summary
    The phenotypic characteristics of Staphylococcus aureus strains isolated from bulk and composite milk and from nasal and hand swabs of dairymen were compared. Of 105 strains of S. aureus isolated from bulk milk, 70 (66.7%) and 41 (39.0%) were pigmented and hemolytic respectively, compared to 121 (82.9%) and 33 (22.6%) atrains recovered from 333 composite milk isolates. The differences were statistically significant (P­0.01; X2). Amongst bulk milk isolates, 56 (53.3%), 79 (75.2%) and 72 (68.6%) produced a halo on Baird-Parker agar (BPA), and were mannitol fermeters and slide coagulase-test positive, respectively. Halo production on BPA, fermentation of mannitol and a positive slide coagulase test were found in 84 (57.5%), 104 (71.2%) and 108 (74.0%) strains, respectively from composite milk. Of 83 strains of S. aureus isolated from hand swabs of dairymen, 24 (28.9%) and 67 (78.3%) produced haemolysin (s) and were mannitol fermenters, respectively, while out of 66 strains isolated from human nasal swabs, 32 (48.5%) and 42 (63.6%) were haemolytic and fermented mannitol, respectively. The differences were statistically significant (P­0.05; X2). Strains of S. aureus isolated from the hands of dairymen were not statistically significantly (P­0.05; X2) different from the composite milk isolates for any of the phenotypic characteristics investigated. However, S. aureus strains isolated from dairymen were significantly (P­0.05; X2) more pigmented (81.9%) than strains from bulk milk (66.7%).

It was concluded that based on the phenotypic characteristics, there was some interchange of S. aureus strains between dairymen and their cows at the farm level in Trinidad.

Introduction

Staphylococci are frequently recovered from the raw milk of animals (1,2). Although Staphylococcus aureus has been reported to be an important aetiological agent of subclinical and clinical mastitis (3, 4), recent avidence suggests that coagulase-negative staphylococci (CNS) may also have clinical significance in mastitis (5, 6).

S. aureus constitutes the normal floea of the anterior nares and skin of man and animals (7, 8) and has also been demonstrated that interchange of S. aureus occurs between humans and animals due to close contact (9, 10), therefore posing zoonotic and food hygiene threats.

In laboratories where the technology is available, procedures such as typing (11), ribotyping (12), plasmid typing (13) and pulse-field gel electrophoresis (14) are used to characterize S. aureus strains in epidemological investigations (15). However, common phenotypic characteristics such as pigmentation of S. aureus strains, haemolysin production, production of a halo in Baird-Parker agar (BPA) due to lecithinase production and the ability to ferment mannitol (10, 16, 17, 18) may also contribute in characterization of strains. The ability of different strains of S. aureus to coagulate plasma from several animal sources and produce bound or free coagulase have also helped in determining the origin of strains (19, 20, 21).

  The study employed phenotypic characteristics to determine the relatedness of S. aureus strains isolated from bulk and composite milk and from nasal and hand swabs of dairymen on farms in Trinidad.

Back To Top    Introduction      Materials and Methods      Results     Discussion

Materials and Methods

Source of Samples

Dairy farms supplying milk to eight collection centres earlier studied (22) served as sources of samples.

Type of Samples

Bulk milk produced by a total of 175 dairy farms across the eight centres was sampled. After the morning milking, approximately 25 ml of bulk milk representative of the churns of milk from the farm was collected. One bulk milk was collected per farm during the study resulting in a total of 175 bulk milk samples. Approximately 20% of all lactating cows on each farm were selected for the study and were randomly sampled by collecting milk from the four quarters and pooling into a sterile universal bottle, as a composite milk sample. Overall, one- fifth of all lactating cows on the 175 dairy farms visited were therefore sampled for the study.

The hand and nasal swabs of the dairymen on each dairy farm visited were collected prior to the milking. Swab samples were

dipped into transport medium (Becton-Dickinson Co., U.S.A), All samples were taken to the laboratory ice-cooled within 2 h of collection.

Isolation Identification and Characterization of Staphylococci

All samples collected were streaked for isolation on Baird-Parker agar (BPA) plates and incubated at 370C for 48 hr. Typical black or greyish-black colonies were Gram-stained and their catalase activity determined. Presence or absence of a clearing or halo on BPA plates was noted. Further identification of Gram-positive, catalase-positive cocci was carried out using standard methods (18,23).

To determine the ability of strains to produce haemolysin(s) and to detect pigment production each isolate of Staphylococcus classified as S. aureus was streaked for isolation on 5% defibrinated sheep blood agar and incubated at 370C for 24 hrs. For this investigation, non-white colonies were considered as pigmented while evidence of any complete clearing around colonies on blood agar (BA) was classified as haemolysin production.

Mannitol fermentation by S. aureus strains was determined with mannitol salt agar (MSA) as recommended by the Sub-committee on Taxonomy on Staphylococci and Micrococci (24). Production of yellowish discoloration around colonies on MSA was considered as evidence of acid production.

To determine the production of bound and free coagulase by S. aureus strains, rabbit plasma was used in tests described by Baer et al. (18) with the test interpretation protocol of Sperber and Tatini (25). All strains were initially subjected to the slide coagulase test for detection of bound coagulase. All slide coagulase-negative strains were subjected to the tube coagulase test but slide coagulase test-positive strains were not further tested. 

Back To Top    Introduction      Materials and Methods      Results     Discussion

Results
    Table 1 shows the sources of the strains of S. aureus tested. The 175 samples of bulk milk yielded 105 (60.0%) strains of S. aureus compared to 146 (43.8%) of 333 composite milk, positive for the microorganism. The difference was statistically significant (P­ 0.001; X2). Of 183 samples of hand and nasal swabs cultured, 83 (45.4%) and 63 (34.4%) yielded S. aureus strains, respectively. The difference was statistically significant (P­0.05; X2).

Table 1: Distribution of S. aureus strains tested according to centre of origin.

    The characteristics of S. aureus strains isolated from bulk milk are shown in Table 2. Of 172 staphylococcal isolates tested, 105 (61.0%) were S. aureus with the highest frequency (75.0%) amongst staphylococcal isolates from centres 5H and 2G while the lowest frequency (33.3%) was found in centre 2C. Seventy (66.7%), 41 (39.0%) and 56 (53.3%) of 105 strains of S. aureus were pigmented, haemolytic and produced a halo on BPA, respectively. Seventy-nine (75.2%) strains were mannitol fermenters while 33 (31.4%) strains were negative by the slide coagulase test but coagulated rabbit plasma by the tube test.

Table 2: Characteristics of s. aureus strains isolated from bulk milk of dairy cows.

a - Baird Parker agar.       b - Negative by slide test but positive by tube test using rabbit plasma.

    Table 3 shows the characteristics of S. aureus strains isolated from composite milk. Of 284 staphylococcal isolates tested, 146 (51.4%) were S. aureus. Centre 2G had the highest (65.2%)   proportion of S. aureus strains compared to centre 6H with the lowest proportion (30.0%). Overall, of the 146 S. aureus strains, 121 (82.9%), 33 (22.6%) and 84 (57.5%) were pigmented, haemolytic and produced a halo on BPA, respectively. One hundred and four (71.2%) were mannitol fermenters but only 38 (26.0%) were classified as coagulase-positive by the tube test only.

Table 3: Characteristics of s. aureus strains isolated from composite milk of dairy cows.

a - Baird Parker agar.       b - Negative by slide test but positive by tube test using rabbit plasma.

    Phenotypic characteristics, whose overall frequency of occurrence statistically significantly (P­ 0.05; X2 ) different between strains of S. aureus isolated from bulk and composite milk, were production of pigmented colonies and haemolytic colonies. A comparison of these two phenotypic characteristics between isolates from bulk milk and composite milk on the same farms revealed a statistically significant difference (P­ 0.05; X2) for pigmented colonies only for milk from 3 (37.5%) of 8 centres (5C, 5H and 26). On the other hand the difference between haemolytic strains among S. aureus isolated from bulk and composite milk from the same centres differed significantly (P< 0.05; X2) in 2 (25.%) of 8 centres (5C and 2B).

A comparison of the strains of S. aureus isolated from bulk and composite milk showed that strains isolated from all bulk milk samples were statistically significantly (P­0.01; X2) less pigmented than strains recovered from composite milk. Strains isolated from bulk milk in 6 of the 8 centres were significantly (P­ 0.01; X2) more common amongst bulk milk strains than those isolated from composite milk. Again, in 6 of 8 centres, a higher frequency of haemolysin(s) production was detected amongst bulk milk than composite milk isolates of S. aureus.

Eighty-three (48.8%) of 170 staphylococcal isolates from dairymen were S. aureus (Table 4). Pigmented, haemolytic and halo-producers in BPA were 68 (81.9%), 24 (28.9%) and 47 (56.6%) strains, respectively. Mannitol fermenters were 67 (78.3%) of 83 strains of S. aureus while 20 (24.1%) were classified as coagulase-positive by the tube test only.

Table 4: Characteristics of s. aureus strains isolated fromthe hand swabs of handlers of dairy cows

a - Baird Parker agar.       b - Negative by slide test but positive by tube test using rabbit plasma.

Table 5 shows the characteristics of S. aureus strains isolated from the nasal swabs of dairymen. Sixty-six (40.0%) of 165 staphylococcal isolates were S. aureus, with 53 (80.3%), 32 (48.5%) and 43 (65.2%) being pigmented, haemolytic and produced a halo on BPA, respectively. Forty-two (63.6%) strains fermented mannitol while 20 (30.3%) strains, although classified as coagulase-negative by the slide test, were positive by the tube test.

Strains of S. aureus isolated from the nasal and hand swabs were similar but haemolysin production was statistically significantly (P­0.05; X2) less common amongst isolates from hands (28.9%) compared to isolates from nasal swabs (48.5%). Frequency of haemolysin production was also higher amongst S. aureus isolated from nasal than hand swabs in 6 of 8 milking centres. However, strains of S. aureus from hand swabs (78.3%) were significantly (P­0.05; X2) more mannitol-fermenters than strains isolated from nasal swabs (63,6%). Amongst strains of S. aureus isolated across the centres, mannitol fermenters were more frequent in hand swabs than in nasal swabs of dairymen in 4 out of 8 centres.

Table 5: Characteristics of s. aureus strains isolated from the nasal swabs of handlers of dairy cows.

a - Baird Parker agar.       b - Negative by slide test but positive by tube test using rabbit plasma.

Overall, for a total of 251 strains of S. aureus isolated from both bulk and composite milk, 74 (29.5%), 191 (76.1%), 183 (72.9%), 140 (55.8%) and 71 (28.3%) were haemolysin producers, pigmented, mannitol fermenters produced halo in BPA and were coagulase-positive by the tube test alone, respectively. Similarly, for 149 S. aureus strains isolated from both the nasal and hand swabs of dairymen, 56 (37.6%), 121 (81.2%), 109 (73.2%), 90 (60.4%) and 40 (26.8%) were haemolytic, pigmented, fermenters of mannitol, halo-positive on BPA and were coagulase-positive only by the tube test, respectively. The differences in the incidence of all the characteristics between milk and human strains of S. aureus were not statistically significant (P• 0.05; X2).

Back To Top    Introduction      Materials and Methods      Results     Discussion

Discussion

The frequencies of detection of S. aureus strains from bulk milk (61%) and composite milk (51.4%) which originated from dairy cows free of clinical mastitis are much higher than the 13% found by Langlois et al. (26) from bovine milk. However, the finding that 40-50% of staphylococcal isolates from human handlers (nasal and hand swabs) are S. aureus is significantly lower than the 61% found elsewhere (26). S. aureus are known flora of the anterior nares and skin of human beings (15,17,18). The relatively low prevalence of S. aureus strains amongst staphylococci in milk cannot be readily explained but it has been demonstrated that staphylococci strains from various sources have different abilities to coagulate plasmas of several species of animals (19,21). It is also pertinent to mention that although all coagulase-positive staphylococci, using rabbit plasma, were classified as S. aureus in this study, some subspecies of S. hyicus were shown to be coagulase positive (26,27,28). It may also be assumed that a majority of the coagulase-negative staphylococci (CNS) isolated from both human and milk in this study are S. epidermidis since the species is known to constitute approximately 75% of CNS (6,7).

A comparison of the strains of S. aureus isolated from bulk and composite milk samples from dairy farms across Trinidad clearly showed that they were significantly different in their ability to produce pigments and haemolysins. Others studies have demonstrated that human strains of S. aureus were more pigmented and haemolytic than bovine strains (16,17,18). The findings in this study therefore suggests that some of the S. aureus strains isolated from bulk and composite milk may have been human biotypes.

Hand-milking of lactating cows, a practice found at over 95% of the dairy farm studied (Adesiyun and others, unpublished) may have contributed to an interchange of S. aureus between milkers and dairy cows, particularly during milking. Published reports (9,10,29) have established that an interchange of S. aureus occurs between man and animals while human biotypes are responsible for bovine mastitis (29).

It was of interest to detect that strains S. aureus isolated from the nasal and hand swabs of milkers differed significantly in their ability to ferment mannitol and to produce haemolysins. These findings again suggest some differences in the biotypes of strains   between sites on milkers. This was not totally unexpected since hand milking may have resulted in a change of the flora of S. aureus on the milkers hands with a high possibility of acquisition of bovine strains of S. aureas. Therefore, the significantly higher frequency of haemolysin producers amongst nasal (48.5%) isolates compared with hand isolates (28.9%) agrees with published reports that human strains of S. aureus are more haemolytic than bovine strains (16,17). Poor sanitary practices such as nose picking prior to or during milking may further lead to interchange of S. aureus strains between the hands and anterior nares of daitymen as earlier in another study (30).

Futher evidence for interchange of S. aureus strains between two sites (hand and anterior nares) of hummi handlers and between humans and lactating cows (bulk and composite milk) is that analysis of the data based on all strains from milk and humans failed to show any stntistically significant difference. Although only 20% of all lactating cows on each farm was sampled for composite milk, the abilities of strains of S. aureus isolated from both milk and humans to produce haemolysin, pigments, produce halo on BPA, ferment mannitol and coagulate rabbit plasma by the tube test only were not significantly different. Adesiyun (31) showed by phage typing that only one-third of S. aureus strains isolated from mastitic milk were actually bovine strains. Furthermore, phage typing of the S. aureus from bulk milk, composite milk and dairymen in the current study has demonstrated their relatedness (33) as suggested by phenotypic characteristics documented here. The importance of the present finding is that in laboratories where the technology for sophisticated techniques like zymotying (32) ribotying (12), plasmid typing (13) or pulse-field gel electrophoresis (14) are not available, may therefore use a combination of these phenotypic characteristics in epidemiological investigations.

It was concluded that the determination of phenotypic characteristics of S. aureus strains isolated from milk and dairymen suggests that an interchange of strains occurred, which could have both milk hygiene and zoonotic significance.

Acknowledgment
    The support of Nestle Trinidad and Tobago Limited is appreciated. We thank the staff of the Veterinary Public Health Unit of the Ministry of Health for their assistance in sample collection. The technical input of Mr. Gerard Ramirez is acknowledged. We thank Ms. Liza Garraway for typing the manuscript.

  Back To Top    Introduction      Materials and Methods      Results     Discussion

References
1. Sanchez, I.C., Auban, E.G., Lopez, B-G,     and Fernandez, F.U.:         Relationship between the California mastitis test and chemical and bacteriological composition of cow’s raw milk. An. Bromatol. 34: 191-198, 1993.

2. Freedman, B,: Milk quality. In: Sanitarian’s Handbook Theory and Administrative Practice for Environmental Health. Peerless Publishing, USA. pp. 564-589, 1977.

3. Sudharma, D., Nair, G.K., Pillair, R.M. and Sulochana, S.: Bacterial organisms associated with mastitis of cattle and goats in Kerala and their antibiograms. Kerala J. Vet. Sci. 16:99-108, 1985.

4. Flinois, J.: Bacteriology of subacute bovine mammary infections. Discussion and prospects. Bull. Mens. Soc. Vet. Prat. Fr. 69:623-640, 1985.

5. Devriese, L.A. and De Keyser. H.: Prevalence of different species of coagulase-negative staphylococci on teats and in milk samples from dairy cows. J. Dairy Res. 47: 155-158, 1980.

6. Sewell, C.M., Clairidge, J.E., Young, E.J., and Guthrie, R.K.: Clinical significance of coagulase-negative staphylococci. J. Clin. Microbiol. 16: 236-239, 1982.

7. Rosdahl, V.T., Gahrn-Hansen,B., Moller, J.K. and Kjaeldgard, P.: Phage-typing of coagulase-negative staphylococci. APMIS 98: 299-304, 1990.

8. Kloos, W.E., Zimmerman, R.J. and Smith, R.F. : Preliminary studies on the characteristics and distribution of Staphylococcus and Micrococcus species on animal skin. Appl. Environ. Microbiol 31: 53-59, 1976.

9. Swartz, R., Jooste, P.J. and Novello, J.C.: Bacteriophage typing of Staphylococcus aureus strains isolated from Blomenforntein dairy herds. J.S. Afr. Vet. Assoc. 56:69-73, 1985.

10. Adesiyun, A.A. and Usman, B.: Isolation of enterotoxigenic strains of staphylococci from dogs. Vet. Microbiol 8:459-468, 1983.

11. Blair, J.E. And Williams, R.E.O.: Phage typing of staphylococci. Bull. World Hlth. Org. 24: 771-784, 1961.

12. Prevost, G., Jaulhac, B. and Piemont, Y.: DNA fingerprinting by pulse-field gel electrophoresis is more effective than ribotyping in distinguishing among methicillin-resistant Staphylococcus aureus isolated. J. Clin. Microbiol. 30: 967- 973, 1992.

13. Cardoso, M. and Schwarz, S.: Chloramphenicol resistance plasmids in Staphylococcus aureus isolated from bovine subclinical mastitis. Vet. Microbiol. 30: 223-232, 1992.

14. Schlichting, C., Branger, C., Fournier, J., Witte, W., Boutonnier, A., Wolz, C., Goullet, P. and Doring, G.: Typing of Staphylococcus aureus by pulse-field gel electrophoresis, zymotyping, capsular typing, and phage typing: Resolution of clonal relationship. J. Clin. Microbiol. 31: 227-232, 1993.

15. Parker, M.T.: The significance of phage typing pattens of Staphylococcus aureus. In: Staphylococcal Infections, Vol. I. Clinical and Epidemiological Aspects. (C.S.F. Easmon and C. Adlan, eds.). Academic Press, pp. 34-61, 1993.

16. Elek, S.O. and Levy, E.: Distribution of hemolysins in pathogenic and non-pathogenic staphylococci. J. Pathol. Bacteriol. 62:541-554, 1950.

17. Hajek, V. and Marsalek, E.: A study of staphylococci of bovine origin Staphylococcus aureus var. bovis. Zentralbl. Bakteriol. 209;541-554, 1969.

18. Baer, E.F., Gray, R.J. and Orth, D.S.: Methods for isolation and enumeration of  Staphylococcus aureus. In: Compendium of Methods for Microbiological Examination of Food (M.L. Speck ed.). American Public Health Association, Washington DC, pp. 374-385, 1976.

19. Live, T.: Differentiation of Staphylococcus aureus of human and of canine origins: coagulation of human and of canine plasma, fibronolysin activity, and serologic reaction. Am. J. Vet. Res. 33:385-391, 1972.

20. Rayman, M.K., Park, C.E., Philpott, J. and Todd, E.C.: Reassessment of the coagulase and thermostable nuclease tests as means of identifying S. aureus. Appl. Microbiol. 29:451-454, 1975.

21. Adesiyun, A.A. and Shehu, L.M.: Detection of staphylo-coagulase using plasmas from various animals. Vet. Microbiol. 10:387-392, 1995.

22. Adesiyun, A.A. Webb, L.A. and Rahaman, S.: Microbiological quality of cow’s raw milk at collection centres in Trinidad. J. Food Prot. 58: 139-146, 1995.

23. Macfaddin, J.F.: Biochemical Tests for Identification of Medical Bacteria. Williams and Wilkins, New York, N.Y, 1977.

24. Subcommittee on Taxonomy of Staphylococci and Micrococci, Minutes of the First Meeting. Int. Bull. Bacteriol. Nomencl. Taxon. 15: 109-110, 1965.

25. Sperber, W.H. and Tatini, S.R.: Interpretation of the tube coagulase test for identification of S. aureus. Appl. Microbiol. 29:502-505, 1975.

26. Langlois, B.E., Parlindungan, A.K., Harmon, R.J. and Akers, K.: Biochemical characteristics of Staphylococcus species of human and bovine origin. J. Food Prot. 53: 119-126, 1990.

27. Devries, L.A., Hajek, V.M., Oeding, P., Meyer, S.A. and Schleifer, K.H.: Staphylococcus hyicus (Sompolinsky 1953) comb. nov. and Staphylococcus hyicus subsp. chromogenes subsp. nov. Int. J. Syst. Bacteriol. 28:482-490, 1978.

28. Adesiyun, A.A., Tatini, S.R. and Hoover, D.G.: Production of enterotoxin(s) by Staphylococcus hyicus. Vet. Microbiol. 9:487-495, 1984.

29. Garcia, M.L., Garcia, M.C., Garcia, M.R., Preton, M., Otero, A. and Fernandez, M.F.: Biovars of coagulase-positive staphylococci isolated from bovine mastitic milk. Microbiologica 6:31-36, 1990.

30. Adesiyun, A.A., Singh, D., Gunness, R.I .: Toxic shock syndrome toxin-1 (TSST-1) production and phage susceptibility of Staphylococcus aureus strains from human vagina and anterior nares in Trinidad. Int. J. Med. Microbiol. 280:371-381, 1994.

31. Adesiyun, A.A.: Characteristics of Staphylococcus aureus strains isolated from bovine mastitic milk: Bacteriophage and antimicrobial susceptibility and enterotoxigenicity. J. Vet. Med. B. 42:129-139, 1994.

32. Adesiyun, A.A., Romain, H., and Webb, L.A.: Relatedness of Staphylococcus aureus strains isolated from milk and ISRAEL JOURNAL OF VETERINARY MEDICI

Back To Top    Introduction      Materials and Methods      Results     Discussion