Comparison of a commercialized phenotyping system, antimicrobial susceptibility testing, and tuf gene sequence-based genotyping for species-level identification of coagulase-negative staphylococci isolated from cases of bovine mastitis

https://doi.org/10.1016/j.vetmic.2008.08.028Get rights and content

Abstract

In order to evaluate the usefulness of some phenotypic and genotypic methods for species identification of coagulase-negative staphylococci (CNS), isolates were obtained from bovine cases of clinical and sub-clinical mastitis from different geographical areas in Sweden. By using the Staph-Zym™ test, antimicrobial susceptibility testing, and sequencing of part of the CNS tuf gene and, when needed, part of the 16S rRNA gene we characterized 82 clinical isolates and 24 reference strains of 18 different species of staphylococci. The genotypic methods identified nine different species of CNS among the 82 milk isolates. A comparison with results obtained by tuf gene sequencing showed that Staph-Zym™ correctly identified CNS reference strains to species level more often than bovine milk CNS isolates (83% and 61%, respectively). In addition, tests supplementary to the Staph-Zym™ were frequently needed in both groups of isolates (50% of reference strains and 33% of milk isolates) to obtain an identification of the strain. It is notable that Staph-Zym™ judged two isolates as CNS, although they belonged to other species, could not give a species name in 11% of the bovine CNS isolates, and gave 28% of the isolates an incorrect species name. The present study indicates that the studied phenotypic methods are unreliable for identification of CNS from bovine intra-mammary infections.

Introduction

Bovine mastitis is often associated with staphylococcal intra-mammary infections. In Sweden, as in many other countries, Staphylococcus aureus (S. aureus) is a common finding in bovine mastitis, but coagulase-negative staphylococci (CNS) are also frequently found (Bengtsson et al., 2005). CNS have been considered to mainly cause mild sub-clinical mastitis with a slight raise in milk somatic cell count (SCC) (Honkanen-Buzalski et al., 1994), but have recently been associated with both clinical and sub-clinical mastitis (Taponen et al., 2006).

The understanding and control of CNS mastitis is complicated by the heterogeneity of this group of bacteria. So far, 16 CNS species have been isolated from bovine mastitis, and despite some variation between herds and countries, S. simulans, S. chromogenes, S. hyicus and S. epidermidis seem to be the most common (Birgersson et al., 1992, Luthje and Schwarz, 2006, Taponen et al., 2006). Little information is available about species differences in virulence, but S. chromogenes has been reported to cause more severe mastitis than other CNS (Zhang and Maddox, 2000). Some CNS species are also capable of persisting in the udder for months, or even throughout the lactation period (Taponen et al., 2007).

In spite of the increasing importance of CNS, these bacteria are not routinely identified to the species level. Conventional procedures are labour-intensive, and most commercial identification systems vary in accuracy when used on animal strains (Langlois et al., 1983, Thorberg and Brändström, 2000). However, Staph-Zym™ (Rosco, Taastrup, Denmark) was considered useful when evaluated on strains from bovine mastitis (Thorberg and Brändström, 2000). Lately, genotypic methods have contributed substantially in the classification of the genus Staphylococcus. Among such methods, sequencing of a part of the tuf gene indicates a high degree of intra-species stability, and excellent sensitivity and specificity among the genus Staphylococcus from human isolates (Martineau et al., 2001, Heikens et al., 2005). Differences in antimicrobial susceptibility could also vary between species, and thus be used for typing of strains (Devriese et al., 2002).

To identify better diagnostic tools for bovine CNS, the objective of the study was to compare two phenotypic and one genotypic methods for species identification of CNS isolated from clinical and sub-clinical cases of bovine mastitis, by using the Staph-Zym™ test, antimicrobial susceptibility testing, sequencing of part of the CNS tuf gene and, when needed, the 16S rRNA gene.

Section snippets

Reference strains and milk isolates

Twenty-four reference strains of 18 different species of staphylococci (Table 1), and 84 presumptive CNS isolates from clinical (n = 64) and sub-clinical (n = 20) cases of bovine mastitis were used in the study. Clinical isolates originated from a national Swedish survey on prevalence of udder pathogens in cases of acute clinical mastitis, conducted in 2002–2003 (Bengtsson et al., 2005). The sub-clinical mastitis isolates were from milk samples sent to the Mastitis laboratory (National Veterinary

Staph-Zym

Of the 24 reference strains, Staph-Zym™ identified 20 (83%) strains correctly (Table 1). Supplementary tests were needed for identification of twelve (50%) of the 24 strains. Two strains could not be identified, as the codes were not found in the Staph-Zym™ code list. The S. pulvereri reference strain was mis-identified as S. sciuri. One of the species of the reference strains (S. pseudintermedius CCUG 32915) was not listed in the Staph-Zym™ species list, and was misjudged as S. chromogenes.

Two

Discussion

One conclusion of this study is that Staph-Zym™ was better at identifying CNS reference strains to species level than at identifying bovine milk CNS isolates. Supplementary tests were, however, frequently needed in both groups of isolates. Moreover, Staph-Zym™ could only identify 61% of the milk isolates correctly, as judged by genotyping methods. The use of supplementary tests in combination with Staph-Zym™ was not a guarantee for correct identification of bovine milk isolates. Another finding

Conclusion

Our results support the notion that micromethods such as Staph-Zym™ based on phenotypic identification may be unable to reliably distinguish between different CNS species, and that genotypic methods are more reliable for diagnostic purposes in the identification of CNS from bovine IMI. It is notable that the accuracy of Staph-Zym™ varied widely between CNS species, and that the test had low specificity in identification of important udder pathogens such as S. chromogenes. There was also a

Acknowledgements

We are very grateful to Sara Frosth for her excellent laboratory support. This project has been supported by the The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Formas).

References (29)

  • A.N. De Paulis et al.

    Five-test simple scheme for species-level identification of clinically significant coagulase-negative staphylococci

    J. Clin. Microbiol.

    (2003)
  • L.A. Devriese et al.

    Coagulase and heat-resistant nuclease producing Staphylococcus epidermidis strains from animals

    J. Appl. Bacteriol.

    (1975)
  • L.A. Devriese et al.

    Identification of coagulase-negative staphylococci from farm animals

    J. Appl. Bacteriol.

    (1985)
  • M. Drancourt et al.

    Stenotrophomonas africana sp. nov., an opportunistic human pathogen in Africa

    Int. J. Syst. Bacteriol.

    (1997)
  • Cited by (0)

    View full text