Mammary infections caused by Staphylococcus aureus are still one of the most serious problems in dairy farms all over the world (Sischo et al. 1993), and the epidemiology of the infection has not yet been completely elucidated (Aarestrup et al. 1995). Any effective modern approach to this disease must therefore be based on more comprehensive epidemiological studies, conducted with valid microbiological typing tools.

A technique for use in epidemiological studies should identify many types, and should be inexpensive, quick and easy to perform, but above all reproducible. Among the available methods, phage typing has up to now been widely and successfully used in differentiating strains of Staph. aureus isolated from cattle with mastitis (Mackie et al. 1987; Fox et al. 1991), but it has some limitations, being a technically demanding method subject to considerable experimental as well as biological variation (Maslow et al. 1993). Moreover, in some studies the number of strains that could not be typed with available bacteriophage panels has been high (Carroll & Francis, 1985; Farah et al. 1988).

Alternative methods have been investigated, and of these molecular techniques have been the most intensely studied (Aarestrup et al. 1995). In studies of human infections caused by Staph. aureus, analysis of the so-called X region of protein A gene polymorphism has been a useful epidemiological marker (Frénay et al. 1994). This gene is ∼2150 bp and harbours some functionally distinct regions: an FC-binding region, the so-called X region and, at the C terminus, a sequence required for cell wall attachment (Guss et al. 1984; Frénay et al. 1994). The X region polymorphism depends on the presence, within the region itself, of a varying number of 24 bp repeats, highlighted by the amplification of this highly polymorphic DNA region and its subsequent restriction fragment length polymorphism (RFLP) analysis (Frénay et al. 1994).

Human epidemic (MRSA) and non-epidemic methicillin-resistant Staph. aureus (non-MRSA) strains, which both cause infections but have completely different infection patterns, have been successfully distinguished by analysis of this polymorphism (Frénay et al. 1994). However, protein A has been identified in only 93% of Staph. aureus strains isolated from bovine intramammary infections (Poutrel & Ducelliez, 1979; Johne & Jarp, 1988).

The aim of the present study was to determine whether the gene for protein A of Staph. aureus (Spa) was present in Staph. aureus strains isolated from cases of subclinical bovine mastitis. This was carried out using the polymerase chain reaction (PCR), as suggested by Frénay et al. (1994). In addition, we have investigated the genetic polymorphism related to the X region of the gene, by means of PCR amplification and subsequent RFLP analysis. Finally we verified the stability of this region after in vitro subculture.