TEXT

Top Abstract Text References

Actinobacillus seminis is a common cause of ovine epididymitis and ram infertility throughout the world (2, 6, 10, 21). It was first isolated in the United Kingdom in 1991 (7) and in one survey was found to be present in the semen of 19% of infertile rams (11). Clinical infections are usually unresponsive to treatment, and because infected animals are of considerable financial value, economic losses can be considerable. A. seminis is a fastidious, slow-growing, pleomorphic, weakly fermentative bacterium. Primary isolation and presumptive identification take several days; there are relatively few distinguishing tests for this and phenotypically similar organisms such as Histophilus ovis, which is also associated with epididymitis; and there have been cases of misidentification (20). It has been proposed that gram-negative pleomorphic bacteria which are catalase, oxidase, nitrate, and ornithine decarboxylase positive and indole, urease, phosphatase, and -galactosidase negative should be considered A. seminis (6), and the API-ZYM system has been shown to be useful for provisional identification (4). However, because identification is laborious and because there is no published phenotypic or genotypic method for intraspecies discrimination of A. seminis isolates, knowledge of the mechanisms of transmission and persistence in rams and ewes and of the pathogenic potential of different isolates is uncertain. In a previous study, a combination of PCR ribotyping, repetitive extragenic palindromic element (REP)-based PCR, and enterobacterial repetitive intergenic consensus (ERIC)-based PCR techniques was used for identification and fingerprinting of Haemophilus somnus isolates of ovine and bovine origin (1). These methods were also found to be applicable to the characterization of A. seminis strains.

Bacterial isolates. Twenty-four A. seminis isolates were included in this study (Table 1). The type strain of A. seminis, NCTC 10851, was obtained from the National Collection of Type Cultures, Colindale, United Kingdom. All other field strains were isolated in Scottish Agricultural College Veterinary Services Laboratories, except for SA32 and SA33, which were kindly provided by P. J. Heath (7), and strain X16, which was isolated by one of the authors (S. Appuhamy) from the reproductive tract of a cow from slaughterhouse materials. H. ovis strains were isolated by Scottish Agricultural College Veterinary Services. All isolates were stored at 80°C in brain heart infusion broth (Oxoid) supplemented with 10% glycerol, 1% Tris (BDH), 1% soluble starch (BDH), 0.5% sodium-L-aspartate (Sigma) and 0.001% thiamine monophosphate (Sigma), pH 7.8 (1). They were propagated on brain heart infusion agar (Oxoid) containing 5% sheep blood and 0.5% yeast extract (Oxoid), and plates were incubated at 37°C for 48 h in a candle jar. The identity of A. seminis isolates was confirmed by a panel of cultural and biochemical tests (17) and by the API-ZYM system (BioMerieux, Marcy l'Etoile, France) (4, 11). All A. seminis isolates showed strong positive reactions in tests for leucine arylamidase, acid phosphatase, and -glucuronidase. Variable intensities for the reaction of alkaline phosphatase were observed, and the bovine isolate X16 was negative for this reaction. Only one isolate (SA35) showed a weak positive reaction for lipase esterase. All other tests were negative.

PCR amplification. Bacteria were suspended in 1-ml volumes of sterile distilled water to a turbidity equivalent to McFarland no. 5 standard (BioMerieux), heated to 100°C for 20 min, and centrifuged at 15,000 × g for 10 min; the supernatant was used as the source of template DNA for PCR. The primers were GIRRN (GAAGTCGTAACAAGG) and LIRRN (CAAGGCATCCACCGT) for PCR ribotyping (8), REP-IRDT (IIINCGNCGNCATCNGGC) and REP2-DT (NCGNCTTATCNGGCCTAC) for REP-PCR (18), and ERIC-IR (ATGTAAGCTCCTGGGGATTCAC) and ERIC-2 (AAGTAAGTGACTGGGGTGAGCG) for ERIC-PCR (18). They were obtained from Life Technologies Ltd. (Paisley, United Kingdom). PCR methods were optimized for template, deoxynucleoside triphosphate, primer, and magnesium ion concentrations by a modified Taguchi method based on the use of orthogonal arrays as described by Cobb and Clarkson (3). Annealing temperature and extension time were also optimized. High intensity, resolution, and sharpness of bands with a low background in an agarose gel were used as the criteria for optimization. The optimized reaction mixture (25 µl) contained 10 mM Tris-HCl (pH 8.4), 50 mM KCl, 3 mM MgCl₂, 0.2 mM each deoxynucleoside triphosphate (Boehringer Mannheim, Lewes, United Kingdom), 100 pM each primer, 0.625 U of Taq DNA polymerase (Life Technologies Ltd.), and 2.5 µl of template DNA preparation. Twenty-five microliters of liquid paraffin was used to overlay each reaction mixture. Amplification was performed in a thermocycler (Techne Ltd., Cambridge, United Kingdom) for 35 cycles, consisting of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 6 min, with a final extension at 72°C for 6 min. The amplified products were electrophoresed in 2.0% agarose type II-A (Sigma) in Tris-borate-EDTA buffer containing 0.5 µg of ethidium bromide/ml (15) in a horizontal submarine electrophoresis apparatus (E-C Apparatus Corporation, St. Petersburg, Fla.). The amplimers were visualized and photographed under UV light. Whenever a distinct PCR profile, in terms of the number and positions of the clearly visible bands, was observed, the corresponding strain was given a unique number or letter designation.

Typing of strains. REP-PCR, ERIC-PCR, and PCR ribotyping have been used to identify species and also to differentiate strains within species of both gram-positive and gram-negative bacteria (1, 8, 9, 18). Boiled cell extracts as the source of template DNA produced the same results as extracted chromosomal DNA for all three PCR methods (1, 9). PCR-based fingerprinting is therefore simple and rapid and can be performed with very small quantities of bacterial cultures. Reproducibility was good for all three methods not only with the same template, but also with template samples derived from different cultures, although some day-to-day variation in the intensity of amplimers, particularly of the minor bands, was observed.

View larger version (146K): [in this window] [in a new window]   FIG. 1.   Fingerprints obtained by REP-PCR for A. seminis isolates. Lanes M, 1-kb DNA ladder; lanes 1 to 24, A. seminis isolates NCTC 10851 (type strain), SA25, SA31, SA32, SA36, SA39, SA67, SA71, SA65, SA66, SA33, SA37, SA43, SA60, SA63, SA70, SA30, SA34, SA35, SA38, SA61, SA62, SA64, and X16, respectively. The profiles have been arranged so that isolates of a similar type are grouped together: type 1 in lanes 1 to 8, type 2 in lanes 9 to 10, type 3 in lanes 11 to 16, type 4 in lanes 17 to 23, and type 5 in lane 24 (see Table 1).

View larger version (141K): [in this window] [in a new window]   FIG. 2.   Fingerprints obtained by ERIC-PCR for A. seminis isolates. Lanes M, 1-kb DNA ladder; lanes 1 to 24, A. seminis isolates SA39, NCTC 10851 (type strain), SA37, SA60, SA65, SA66, SA31, SA36, SA67, SA71, SA70, SA30, SA33, SA34, SA35, SA38, SA43, SA61, SA62, SA64, SA25, SA32, SA63, and X16, respectively. The profiles have been arranged so that isolates of a similar type are grouped together: type A in lane 1, type B in lanes 2 to 6, type C in lanes 7 to 10, type D in lane 11, type E in lanes 12 to 20, type F in lane 21, type G in lane 22, type H in lane 23, and type I in lane 24 (see Table 1).

View larger version (134K): [in this window] [in a new window]   FIG. 3.   Differentiation of A. seminis from H. ovis by PCR methods. Lanes M, 1-kb DNA ladder; lanes 1 to 4, different isolates of H. ovis; lanes 5 to 8, A. seminis strains SA32, SA33, SA37, and SA39, respectively.

Comparison with H. ovis. H. ovis is an ovine pathogen that can cause epididymitis and orchitis. It may well be present in the same clinical specimens as A. seminis, and it may be difficult to distinguish biochemically between the two species. Stephens et al. (17) distinguished two isolates of A. seminis from the Haemophilus-Histophilus group by their lack of yellow pigment, production of catalase, and differences in cell wall envelope protein profiles. Recent genetic studies have shown a clearer picture of species differentiation, for example, by restriction endonuclease profiles for BamHI (12, 13) and DNA-DNA hybridization techniques (14, 19). In 1990, Sneath and Stevens (16) defined the properties of A. seminis and proposed it as a new species on the basis of cultural, biochemical, and DNA-DNA hybridization methods. When a number of representative H. ovis isolates were compared with A. seminis strains by the three PCR methods, the profiles were completely different (Fig. 3). The profiles were also distinct from those of H. somnus isolates examined previously (1) (data not shown). Thus, the PCR methods can readily be used to differentiate between A. seminis and the Haemophilus-Histophilus group.