Abstract: Random amplified polymorphic DNA (RAPD) analysis was employed on twenty-four Streptococcus uberis isolates from cases of clinical and sub-clinical mastitis from five smallholder dairy herds in Malaysia. Three ten-mer primers namely OPA-01, OPA-05 and OPA-07 were selected out of a 20 primers set. A high degree of genetic polymorphism was revealed among the isolates of S. uberis by RAPD-PCR. Fourteen distinct RAPD fingerprint patterns were generated with primer OPA-05, while primers OPA-01 and OPA-07 produced 12 and 11 RAPD patterns respectively. The discriminatory power of this technique could be further improved by the parallel use of several primers. Most of the isolates were genetically different strains, however, identical patterns were noted among different cows within the same farm or from different cows from different farming regions. These findings indicate that RAPD analysis is a rapid and reproducible method for molecular sub-typing of bovine S. uberis isolates and therefore, represents a powerful tool for epidemiological studies.
Introduction
Bovine mastitis, an inflammation of the udder, usually arises as a result of intramammary infection by bacteria and is, economically, the most important disease to the dairy industry (Aarestrup and Jensen, 1996). Financial losses in mastitis occur from reduced milk quality and production, culling of the affected cow, and treatment costs. Recently estimated cost of a case of mastitis averages £180 (Kossaibati and Esslemont, 1997) that equates an annual loss of around £170 million in UK, and in USA, the cost is US$ 130 – 320 per mastitis case (Wilson et al., 1996). Three species, Streptococcus uberis, S. agalactiae and S. dysgalactiae are known as the principal causes of bovine streptococcal mastitis (Lammler, 1991; Gillespie et al., 1997). Widespread application of mastitis control measures the reduced prevalence of Streptococcus agalactiae and Staphylococcus aureus intramammary infections. However, the proportion of intramammary infections caused by environmental pathogens such as S. uberis have increased markedly in recent years (Leigh, 1999). Recent surveys conducted in last ten years in Ontario (Canada), Ohio (USA), the Netherlands and the UK indicated that S. uberis was responsible for between 14 and 26% of all clinical mastitis (Hogan and Smith, 1997). Another study estimated that the level in UK was as high as 33% (Hillerton et al., 1993). Streptococcus uberis differs from other mastitis causing streptococci in that it can also be isolated from udder surface and from other sites of the body of cows, also the cows environment can serve as the reservoir of infections (Bramely, 1982; King, 1981).
Epidemiological studies of S. uberis have been limited due to lack of convenient, reliable and reproducible method for comparing the strains. Conventional typing methods based on biochemical profiles, serotyping and antibiotic resistance patterns are inadequate for identifying the closely related strains (Jayarao et al., 1992). The development and application of nucleic acid based diagnostic techniques for the identification and differentiation of streptococci and other bacteria have been described with increasing frequency (Bentley and Leigh, 1995). Random Amplified Polymorphic DNA fingerprinting analysis has become an accepted and reliable tool for epidemiological subtyping of a wide variety of bacteria within a genus or species (Jayarao et al., 1996). Studies on other organisms have previously shown RAPD analysis to discern different strains better than REA, Ribotyping and RFLP but its potential use has been limited to a few bacterial species belonging to the genera Streptococcus and Enterococcus (Jayarao et al., 1996). The aim of present study was to analyze S. uberis strains from bovine mastitis in Malaysia by means of RAPD fingerprinting and to evaluate the value of this technique molecular epidemiological studies.
Materials and Methods
Streptococcal isolates: The study was carried out during the period from January to December 1999 at the Faculty of Veterinary Medicine, University Putra Malaysia. Twenty-four Streptococcus uberis isolates were recovered from cases of clinical and subclinical bovine mastitis from five different dairy herds in the state of Selangor, Malaysia. The S. uberis reference strains used in this study: 0140J, EF20, 2022 and 2038 (kindly provided by Dr. James Leigh, Compton Lab, UK). All strains of S. uberis were identified based on the following criteria- non-hemolytic on blood agar, CAMP negative, hydrolyzed esculin and hippurate, fermented inulin, mannitol and salicin according to Watts (1988). Furthermore, all the isolates were identified using API 20 strep systems according to the manufacturer’s guidelines (bioMerieux sa, France).
Genomic DNA preparation: Bacterial genomic DNA was extracted using commercial Genomic DNA purification kit (Wizard, Promega, Madison, WI, USA) according to the manufacturer’s instructions. In addition to the reagents supplied with the kit, Mutanolysin (5U/ul)(Sigma, UK) and freshly prepared Lysozyme (30 mg/ml) (Sigma, UK) were added to augment the lysis of bacterial cell wall. The concentrations and purity of DNA was determined spectrophotometrically.
Selection of primers: A set of twenty 10-mer primers (OPA-01 to OPA-20) (Operon technologies, Almeda, CA) were tested with the above mentioned reference strains. Three primers namely OPA-01, OPA-05, and OPA-07 with nucleotide sequences of 5’-CAGGCCCTTC-3’, 5’- AGGGGTCTTG-3’, and 5’- GAAACGGGTG-3’ respectively with 50-70% G+C content were selected, as they gave reproducible patterns comprising fragments with a large size range and a small number of low-intensity bands. Primer OPA-01, OPA-05 and OPA-07 generated distinct bands for the four reference strains of S. uberis. The reproducibility of the RAPD patterns obtained with these three primers was verified by repeating the experiments under same conditions.
RAPD-PCR conditions: RAPD fingerprinting was performed in a total volume of 25ul mixture containing 10x Mg-free buffer (10mM Tris-HCl, 50mM KCl), 2.5mM MgCl2; 100μM each of the four dNTPs (Boehringer Mannheim, Germany), 0.2μM primer, 0.5μ of Taq DNA polymerase and 25mg of DNA template. Each sample was subjected to the first cycle of amplification at 94 °C for 4 min, at 36 °C for 1 min and at 72°C for 2 min in a DNA thermal cycler (Techne, Cyclogene, Germany) followed by 44 subsequent cycles consisting of denaturation at 94 °C for 1 min, annealing at 36 °C for 1 min and extension at 72 °C for 2 min and for the last cycle extension at 72 °C for 10 min.
Amplified products were separated by electrophoresis on 1% agarose gel (Gibco BRL) and visualized by UV trans-illumination following ethidiumbromide staining and photographed with 665 Polaroid film. A 1-kb DNA ladder (Life technologies) was used in each gel as molecular size marker. The comparison of different fingerprint profiles generated by the primers was made with unaided eye and the degree of dissimilarity between the strains was determined by cluster analysis program.
Results and Discussion
Amplification of DNA from Streptococcus uberis isolates with primers OPA-01, OPA-05, and OPA-07 resulted in characteristic banding patterns. The reproducibility of the RAPD was examined by repeating the RAPD analysis on three separate occasions. No change in DNA fingerprints was observed in any of the replicate experiments (data not shown). For 24 S. uberis strains, 12 fingerprint patterns composed of two to ten bands of 0.3 to 1.8 Kbp with primer OPA-01 (Fig. 1), 11 fingerprints composed of 3 to 9 bands of 0.3 to 2.5 Kbp with primer OPA-07 (Fig. 2), and 14 fingerprints with one to nine bands of 0.4 to 2.8 Kbp were achieved with primer OPA-05 (Fig. 3). One isolate (#116, lane 6 in Figs. 1, 2 & 3) of S. uberis was not amplified with primer OPA-05 but produced different RAPD fingerprints with primer OPA-01 and OPA-07. The dendrogram obtained with the combined RAPD patterns of three primers generated two main clusters (A & B) and most of the isolates were grouped in cluster B with varying degree of genetic dissimilarities (Fig. 4). Each of the clusters A and B contained several subclusters indicating that the bacterial strains are genetically heterogeneous. The genetic difference among the strains could be estimated by comparing the clusters or subclusters (Fig. 4) with the scale (0 - 100%). Primer OPA-05 had the most discriminatory power for S. uberis isolates. The distribution of RAPD fingerprints generated with different primers among the isolates of S. uberis recovered from different herds and cows are depicted in Table 1.
| Fig. 1: | RAPD fingerprint patterns of S. uberis isolates generated with primer OPA-01 separated on 1% agarose gel. Lane M: 1 Kb DNA molecular size marker, lanes 1 to 24: S. uberis isolate numbers. |
| Fig. 2: | RAPD fingerprint patterns of S. uberis isolates generated with primer OPA-07 and electrophoresed in 1% agarose gel. Lane M: 1 Kb DNA molecular size marker; lanes 1 to 24: S. uberis isolate numbers. |
| Fig. 3: | RAPD fingerprint patterns of S. uberis isolates generated with primer OPA-05 and separated on 1% agarose gel. Lane M: 1 Kb DNA molecular size marker; lanes 1 to 24: S. uberis isolate numbers. |
| Fig. 4: | Dendrogram of the cluster analysis based on the combined RAPD fingerprints of Streptococcus uberis strains with primers OPA-01, OPA-05 and OPA-07. |
Identical fingerprint profiles were observed among the isolates recovered from different and the same herds. Furthermore, isolates from different cows within the same and different herds also generated similar RAPD fingerprint profiles. Detection and analysis of polymorphisms in a DNA sequence for a given species has been used widely in the field of epidemiology. RAPD has been used for typing and identification of a number of closely related species of bacteria and assessment of their genetic relationships. Its results are usually consistent with those of DNA-DNA homology studies and can be used to estimate the genetic distances (Bert et al., 1996). Recently RAPD analysis was evaluated as better than REA for typing of group A streptococci (Seppala et al., 1994). RAPD analysis is faster, technically less demanding and more economical than the older genomic typing methods like REA, RFLP and ribotyping. The amount of DNA needed for RAPD analysis is much smaller and unlike conventional PCR, data on DNA sequence of the organisms are not a prerequisite for RAPD analysis (Seppala et al., 1994).
| Table 1: | RAPD fingerprint profiles of Streptococcus uberis isolates (n=24) generated with primers OPA-01, OPA-05 and OPA-07. |
| * na = not amplified | |
However, optimization of PCR conditions and the selection of appropriate primers are the two prerequisites to obtain better results from RAPD-PCR.
In present study, the RAPD-PCR assay was applied to characterize 24 S. uberis isolates from cases of clinical and subclinical mastitis in cows. All strains of S. uberis were distinguishable using primer OPA-01 and OPA-07 respectively. But primer OPA-05 did not amplify one S. uberis isolate (#116), this however, generated reproducible fingerprint profiles when primers OPA-01 and OPA-07 were used. This may be due to small DNA sequence changes in course of mutations, deletions or insertions. These changes may, however, produce a new RAPD pattern by creating or destroying a target site of the primer (Williams et al., 1990). This is in agreement with the notions that discriminatory power of RAPD could be increased further using two or more independent primers (Bert et al., 1996, Zucher et al., 1996). Precise molecular analysis of S. uberis strains using RAPD-PCR has not been reported before. In present study, 14 RAPD patterns were observed among 24 S. uberis isolates and to some extent, the isolates from cows of the same and different farms shared RAPD profiles. Diversity among S. uberis isolates was also reported in earlier epidemiological studies using different molecular techniques such as DNA restriction fragment length polymorphism (Jayarao et al., 1991, Jayarao et al., 1993) and PCR-fingerprinting (Jayarao et al., 1992). More recently, several studies using pulsed field gel electrophoresis (PFGE) have also demonstrated great diversities among S. uberis strains (Baseggio et al., 1997; Wang et al., 1999; Douglas et al., 2000). The high level of heterogeneity among S. uberis isolates in this study supports the classical epidemiological studies, suggesting an environmental reservoir with limited transmission, if any, from cow to cow during milking process. Discriminatory power of PFGE is similar to RAPD, but the specialized equipments required, the length of time and expertise for the protocol make it more difficult and complex method than RAPD (Bert et al., 1996). The high level of polymorphism among the strains of this environmental Streptococcus might be due to genetic exchange within the species, as frequent interactions in the environment and within the animal hosts create opportunities for genetic exchange. Genetic exchange within the species has been demonstrated for environmental organisms (Whittam et al., 1993). All the data established that RAPD-PCR is one of the most efficient, rapid and reproducible DNA based fingerprinting techniques for precise epidemiological studies of Streptococcus uberis strains of bovine origin.
Acknowledgment
The study was financially supported by the Malaysian Government IRPA research grant No. 51380.