Introduction

Staphylococcus aureus can colonize and infect a variety of members of the animal kingdom, including mammals, reptiles, and birds (van Leeuwen et al. 2005). This organism is the most common etiologic agent of contagious bovine mastitis, with relevant losses in the dairy industry (Zecconi et al. 2003). Intramammary infections, caused by S. aureus and also by other bacterial species, are due to the entry of bacteria through the teat canal into the mammary gland. The induction of intramammary infection is thought to be due in part to its expression of a large number of secreted and cell surface-associated virulence factors enabling adherence, colonization, and invasion of the mammary cells of the bovine host by the S. aureus cells, evasion from the immune defense mechanism and survival in the host environment. The first step in the colonization of the mammary gland by S. aureus seems to be adhesion to epithelial cells, which prevent the bacteria from flowing out of the gland during milking (Brouillette et al. 2003) and thus, a risk factor for invasive disease (Foster and Hook 1998). Several studies have been demonstrated in vitro and in vivo that S. aureus is able to adhere to and penetrate inside bovine mammary epithelial cells (Gudding et al. 1984; Sutra and Poutrel 1994; Hebert et al. 2000; Hensen et al. 2000). However, the mechanism of S. aureus adhesion to mammary epithelial cells is not well known, but is likely to be multifactorial. It is now recognized that S. aureus express various types of wall-associated proteins promoting adherence to host cells and/or tissue components. S. aureus adhesins are grouped into a single family named microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) (Foster and Hook 1998). These surface proteins play a major role in host–bacterium interaction. Two types of interactions could be involved: (1) non-specific physicochemical interactions, and (2) specific interactions between bacterial cell wall-associated receptors and host components (Sutra and Poutrel 1994). Among the surface proteins produced by S. aureus are a diverse group of adhesins capable of binding host proteins present in the extracellular matrix. The gene encoding cna was reported as an important virulence factor associated to bacterial adhesion (Montanaro et al. 1999; Nashev et al. 2004; Xu et al. 2004). The cna (collagen adhesin) gene (Fig. 1) encodes a 135-kDa protein with structural features common to surface proteins from other Gram-positive bacteria. An amino-terminal signal sequence is followed by a large non-repetitive region (55 kDa) which is referred to as ligand-binding A domain. Immediately following the A domain is a series of 187 amino acid repeated motifs present in one to four copies. On the basis of the number (one, two, three, or four) of B domains, the S. aureus collagen adhesion occurs in at least four forms. Finally, the repetitive region is followed by domains that anchor collagen binding protein (CBP) to the cell envelope. These domains included a 64-amino acid proline- and lysine-rich cell wall-associated region, a hydrophobic membrane-traversing domain, and finally, a short, positively charged carboxy-terminus located in the bacterial cytoplasm. The collagen-binding domain (CBD) of the collagen adhesin has been localized to a fragment containing amino acids 151–318 of the intact protein (within the non-repetitive region) (Patti et al. 1995). The crystal structure of the binding domain has been determined at 2 Å resolution (Symersky et al. 1997), and has a ‘jelly-roll’ topological pattern that is composed of two anti-parallel R-sheets and two short helices. A groove on one of the R-sheets exhibits the best complementarity to triple-helical collagen model probes and is the most likely collagen-binding region.

Fig. 1
figure 1

Model of the collagen MSCRAMM gene (cna) from S. aureus FDA strain 574 (Symersky et al. 1997). S, signal peptide-encoding region; A, non-repetitive domain; Bl, B2, B3, repeated regions; W, cell wall domain; M, membrane-spanning domain; C, carboxy-terminal domain. The 19-kDa CBD (collagen binding domain) in the A region is also indicated

Full size image

Cna, encoded by the cna gene, is the primary identified adhesion protein responsible for the ability of S. aureus to bind collagen substrates and collagenous tissues (Switalski et al. 1989; Patti et al. 1992; Gillaspy et al. 1998). This adhesin has been shown to significantly contribute to tissue colonization in various pathological conditions such as eye keratitis (Rhem et al. 2000; Jett and Gilmore 2002), osteomyelitis and septic arthritis (Hudson et al. 1999; Smeltzer and Gillaspy 2000; Elasri et al. 2002), and indwelling medical devices (Hudson et al. 1999; Arciola et al. 2005).

Investigating distributions and the virulent factors of S. aureus provides important information for establishing infection control strategies. Hence, we suggest that a study for adhesion molecules may help in clarifying the relevance of the different adhesion mechanisms in the pathogenesis of mammary-associated infections. The role of Cna in the virulence of S. aureus during intramammary infections (IMI) has not been definitively demonstrated. Molecular biology could provide the tools to clarify this point. We, therefore, investigated the prevalence of collagen adhesion gene in a collection of 58 S. aureus strains isolated from bovine mastitis, and analyzed for any correlation between the presence of cna gene and the different genotypes of S. aureus.

Materials and methods

Bacterial isolates and preparation of bacterial DNA

The 58 S. aureus strains used in this study were the same as those used in previous studies (Saei et al. 2009, 2010). Briefly, the tested isolates were obtained from clinical and subclinical bovine mastitis cases from nine dairy herds. All isolates were characterized by means of classic microbiological methods.

For chromosomal DNA purification, bacterial cells grown in 10 ml of nutrient broth (Merck, Germany) were collected by centrifugation (Sigma, Germany) at 13,000 rpm for 5 min and resuspended in 200 μl of TE buffer in a 1.5 microtube. Bacterial DNA was then extracted using Genomic DNA purification Kit (Fermentas, Germany) according to the manufacturer’s protocol.

PCR mediated identification and genotype analysis

The determination of species specific parts of the genes encoding thermonuclease (nuc) was performed with oligonucleotide primers as described previously (Brakstad et al. 1992) and PCR conditions reported elsewhere (Saei 2010).

The results from previous study (Saei et al. 2009), in which amplification of the variable region of the coa gene and restriction enzyme analysis of the resultant coa amplicons were performed by a modification of the procedures described by Hookey et al. (1999), were used in the present study.

Detection of the cna gene

The amplification was carried out by a modification of the procedure described by Montanaro et al. (1999). The primers to detect the presence of the cna gene were as follows: 5′-AAAGCGTTGCCTAGTGGAGA (forward primer) and 5′-AGTGCCTTCCCAAACCTTTT (reverse primer), including a region of 192 bp (corresponding to nucleotides 1,291–1,482). The thermal cycler programs were: 94°C for 5 min, 1 cycle; 94°C for 45 s, 55°C for 45 s, 72°C for 45 s, 40 cycles; 72°C for 10 min, 1 cycle. PCR products (8 μl) were analyzed on 1.2% (wt/v) agarose gel stained with ethidium bromide (0.5 μg/μl), and visualized under ultraviolet transillumination and photographed using gel doc apparatus. The S. aureus ATCC 25923 was used as a positive control for cna gene detection.

Results

According to culture and biochemical properties, all 58 isolates used in the present study could be identified as S. aureus. The identification of the isolates was also confirmed by PCR amplification of thermonuclease gene, nuc. The amplicons of this gene showed a uniform size of approximately 270 bp (data not shown).

Results from coa gene PCR-RFLP analysis of the studied isolates have been described previously in details (Saei et al. 2009). Briefly, amplification of the variable region of the coa gene from these isolates produced five different PCR products ranging in size from approximately 490 to 850 bp. To obtain RFLP patterns of the PCR products, they were subjected to digestion with restriction endonuclease HaeIII. Nine coa gene RFLP patterns, numbered I–IX, were observed, with 23 isolates (39.66%) assigned to RFLP pattern I and 14 isolates (24.14%) assigned to RFLP pattern III. The genetic analysis of S. aureus isolates showed that all of them were positive for coa gene as expected, and thus, this gene was further analyzed by enzyme restriction pattern using HaeIII.

The PCR technique allowed the identification of cna + strains by the appearance of an amplified DNA fragment of 192 bp, as described earlier (Montanaro et al. 1998, 1999) (Fig. 2). The amplicon indicating the presence of cna gene was detected in 84.5% of all the 58 isolates. As a result, majority of the bovine mastitis-associated S. aureus isolates had gene encoding collagen adhesin.

Fig. 2
figure 2

Agarose gel electrophoresis of polymerase chain reaction (PCR) amplification of cna gene. Lane M, 100 bp DNA molecular size marker; lane 1, Positive control (S. aureus ATCC 25923); lanes 2–4, PCR amplicons obtained with DNA amplification of S. aureus; lane 5, negative control (reaction mixture without DNA)

Full size image

The different prevalence of cna among the nine categories of coa types is reported in Table 1. As shown, the cna gene was detected in both predominant and rare coa types but mostly in predominant types than rare types. On the other hand, different genotypes exhibit differences in prevalence of the cna gene. As shown, all of the 37 predominant types were positive for cna gene, but nine out of the 21 rare genotypes were negative for the mentioned gene.

Table 1 Presence of cna gene among the nine categories of coa types
Full size table

Discussion

In the present study, the frequency of the isolates classified in the different restriction pattern clusters confirmed that there is a prevalent clone of S. aureus isolated from bovine mastitis. This is in agreement with the finding of other researchers (Gilot et al. 2002; Schlegelova et al. 2003).

It is now well established that to initiate infection at a particular site, bacteria must adhere to host cells or to layers covering these cells. There is mounting evidence to suggest that proteins that belong to a family defined as MSCRAMMs are involved in the adhesion process. These include proteins covalently bound to bacterial cell wall peptidoglycan (Foster and Hook 1998), such as collagen-binding protein (Cna) (Hudson et al. 1999; Smeltzer and Gillaspy 2000). The role played by the Cna adhesin as a virulence determinant in the pathogenesis of septic arthritis is well documented (Patti et al. 1994). The presence of collagen adhesin (Cna) is also correlated with pulmonary manifestations (Gonzalez et al. 2005). Similarly, inactivation of the collagen adhesin gene, cna, has been correlated to reduce virulence in animal models of endocarditis (Patti et al. 1994; Hienz et al. 1996). In this study, the presence of gene encoding for the CBP in the majority of the 58 isolates of S. aureus, suggests that adherence to epithelial cells may play an important role in the pathogenesis of S. aureus mastitis and the presence of cna gene and expression of collagen adhesin may be very relevant, almost essential, trait for the virulence action in bovine mammary tissue. Receptors for collagen have been demonstrated on strains isolated from bovine IMI (Mamo et al. 1988, 1992). However, our finding is controversial when compared with earlier observations where most strains of S. aureus do not encode cna (Smeltzer et al. 1997). Results from previous studies have shown that the frequency of isolates harboring cna gene both in human (Montanaro et al. 1999; Nashev et al. 2004) and in animal isolates (van Leeuwen et al. 2005; Zecconi et al. 2005; Reinoso et al. 2008) was lower. In a study carried out by Reinoso et al. (2008), the cna gene of 15 isolates from bovine subclinical mastitis yielded no amplicon. Further expanded studies using large number of isolates from clinical and subclinical mastitis is necessary to investigate the extent of cna presence in mastitis-associated S. aureus strains.

In the current study, distribution of the cna gene among different categories of the coa genotypes indicate that certain subtypes of S. aureus may have special properties increasing their potential for adherence to and colonization of bovine udder tissue in comparison with other subtypes less adapted to bovine udder. This result is in agreement with the findings of other researchers, who reported that the adherence capacity of S. aureus isolated from bovine mastitis to mammary cells to be strain dependent (Frost et al. 1977; Opdebeeck et al. 1988; Iturralde et al. 1993; Kuzma et al. 2006). Furthermore, a study by Otsuka et al. (2006) demonstrated that the pandemic type (ST30) and continent-specific type (ST1, ST8, ST59 or ST80) of community-acquired methicillin-resistant S. aureus (CA-MRSA), each had a unique adhesin gene. These findings are in consistent with the results from Poland, where the prevalence of cna positive strains was significantly higher in spa ≤ 7 type than in spa > 7 type (Kuzma et al. 2006). Results from study by Hensen et al. (2000) also indicate that strain differences of adherence and invasion exist for S. aureus. The study by van Leeuwen et al. (2005) showed that the cna gene was not equally distributed among the different lineages of S. aureus. These results may explain differences in the ability of the strains to spread. However, distribution of other adhesins such as fibronectin (FnbA and FnbB), fibrinogen (ClfA, ClfB and Efb) and elastin (EbpS) binding proteins must be investigated between different subtypes of S. aureus isolates. Overall, we suggest that different types of S. aureus seemed to use different adhesion mechanisms, some mediated by collagen binding, others independent of collagen binding. This issue may influence the target cell specificities of the major subtypes in comparison with rare types. This idea is supported by the observation that the presence of (combinations of) virulence factors plays an important role in host or even tissue specificity in S. aureus infections (van Leeuwen et al. 2005).

It would also be interesting, if a vaccine containing purified surface proteins involved in S. aureus adhesion (e.g., CBP) belonging to predominant types could be used to immunize cows. An advantage of vaccines directed at surface proteins is that they will block adherence of bacteria as well as promoting phagocytosis of the non-adherent bacteria. However, results presented by Mamo et al. (1994a), mice vaccinated with the glutathione S-transferase-CBP fusion protein were not protected against challenge infection with S. aureus. A possible explanation for this finding might be the fusion protein used in that study contained only the A domain responsible for collagen binding; it is still possible that CBP presented in another way could induce protection. Other speculations could be that the synthesis of the proteins present on the bacterial surface is influenced by growth conditions, such as the composition of the growth medium (Ellwood and Tempest 1972). On the other hand, it must be noted that S. aureus adheres to several different host proteins, so it might be necessary to include epitopes from several different binding proteins to be fully effective. A preliminary study has shown that vaccination of cows with a fusion protein containing the fibronectin-binding domain of Fnbp lead to significant protection against experimental S. aureus IMI (Nelson et al. 1991). Immunizations with fusion proteins encompassing the fibronectin binding D-domains of a FnBP from S. aureus have been shown to induce the production of antibodies with adherence blocking activity (Luk et al. 1989; Ciborowski et al. 1992). It was also reported that immunization with these fusion proteins induced protection against S. aureus rat endocarditis (Schennings et al. 1993) and mouse mastitis (Mamo et al. 1994b).

In conclusion, although S. aureus can bind to a verity of host proteins, isolates associated with bovine mastitis seem to have an enhanced propensity to bind to collagen and predominantly express receptors for this matrix protein. Hence, it can be supposed that CBP encoded by cna gene could mediate S. aureus adhesion, especially predominant types, to epithelial cells and to micro-lesions of the mammary epithelium where basal lamina and inflammatory conjunctive tissue, which are rich in collagen, are exposed. Identification and characterization of the S. aureus adhesins that contribute to collagen binding is important in light of the fact that S. aureus strains that cause mammary gland infections almost invariably bind collagen. This correlation suggests that therapeutic strategies aimed at the inhibition of collagen binding might be useful for prevention and treatment of mammary gland infections. However, the development of such strategies will require a clear understanding of the bacterial factors that contribute to collagen binding by performing both in vitro and in vivo models of adherence and infection of mammary cells.