IgG-binding protein from Staphylococcus and nucleotide sequence encoding this protein

#The present invention relates to a new protein and a nucleotide sequence encoding said protein. More precisely, the invention relates to a DNA molecule coding for a protein expressed by a bacterium of the genus Staphylococcus aureus, said protein and polypeptide fragments of said protein. Vectors comprising the nucleotide sequence coding for the protein, the protein and fragments thereof, and antibodies specifically binding to the protein may all be used for different vaccines against Staphylococcal infections in mammals. The invention also relates to a method of isolating and/or purifying apolipoprotein H from e.g serum with an immobilised protein or polypeptide of the invention.

#Staphylococcus aureus is a pathogen responsible for a wide variety of diseases in humans and animals, including endocarditis, osteomyelitis, wound sepsis and mastitis. The bacterium produces several potential virulence factors such as alpha-, beta-, gamma- and delta-toxins, toxic shock syndrome toxin (TSST), enterotoxins, leucocidin, proteases, coagulase and clumping factor.

#It is generally accepted that adhesion to tissues is required for bacterial colonisation to occur. For this purpose staphylococci express surface adhesins, which interact with host matrix proteins such as fibronectin, vitronectin, collagen, laminin and bone sialoprotein. In addition, staphylococci are able to bind several serum proteins, such as IgG, fibronectin, fibrinogen, and thrombospondin, possibly masking the bacteria from the immune system of the host. However, the contribution and importance of each of these binding functions in different infections is still unclear.

#FIG. 1 shows a) the cloned 3 kb PstI-XbaI fragment from S. aureus strain 8325-4 containing the sbi-gene of the invention (clone pPX1), and b) schematic drawing of the protein Sbi aligned with the peptides encoded by the phagemid clones isolated by panning against IgG. Stars indicate that the clone was isolated several times independently. Ig4 represents the clone isolated earlier (Jacobsson & Frykberg, 1995). S denotes the signal sequence and the shaded bar represents a proline-rich region.

#FIGS. 2a+2b show the complete nucleotide sequence of the sbi-gene (SEQ ID NO:1) of the invention from S. aureus 8325-4, and the deduced amino acid sequence of the encoded protein. Features of the sequence are indicated as follows: underbar, putative promotor sequences; double underbar, possible ribosomal binding site; overbar, putative transcriptional stop signal (inverted repeat); single vertical arrow indicates the cleavage site of the signal sequence and in bold the repeated proline residues. Also shown in bold is the minimal IgG- binding domain as deduced by shot gun phage display mapping.

#FIG. 3 shows the specificity of the binding domain of the protein Sbi (clone Ig4) for various IgGs and human IgG3κ, IgG3λ, IgM and IgA. Each column represents a mean value of two independent experiments with the same phage stock. The y-axis shows the number of c.f.u. in 50 μl of the eluate.

#Analysis of IgG binding activity in S. aureus by Western blots usually reveals more than one protein band, which interact with IgG, and it has been assumed that these polypeptides represent breakdown products of protein A. In view of the present results, it is not surprising that protein Sbi has escaped detection. For example, analysis of the expression of the protein Sbi in S. aureus has been hampered by the ability of protein A to interact with IgGs from most mammalian species. Furthermore, both proteins migrate similarly in SDS-PAGE and have the same specificity for all tested immunoglobulins. As disclosed in Example 4 below, the commercially available anti-protein A antibodies cross-react with full-length protein Sbi but not with MAL-SbiΔ (SEQ ID NO:4) (lacking the known IgG-binding domain). Most likely, the cross-reaction is not unique to this source of antibodies and may have contributed to the failure to identify protein Sbi. Now the antibodies directed against the protein Sbi of the invention produced in chicken allows discrimination between expression of protein A and protein Sbi.

#The present inventors have cloned and sequenced the sbi-gene from S. aureus strain 8325-4. The detection of protein Sbi, expressed from its own promoter in S. epidermidis and as a MAL-fusion in E.coli with HRP-labelled IgG, proves that this gene encodes a second IgG-binding protein. This protein consists of 436 amino acids (SEQ ID NO:2) and contains a signal sequence but lacks the cell wall sorting LPXTG (SEQ ID NO:7) motif, suggesting that the protein is not anchored in the cell-wall. However, in strain Newman 4, which produces high amounts of protein Sbi, no protein Sbi is detected in the culture growth medium (data not shown). Instead, protein Sbi is released from the cell surface by addition of sample buffer (Example 4), in contrast, release of protein A requires treatment with lysostaphin (Example 4). This indicates that protein Sbi is associated to the cell surface by a different mechanism. There are also other examples of cell surface associated proteins that lack the LPXTG motif, such as the S. aureus elastin-binding protein (Park et al 1996) and MHC class II analogue (Jönsson et al 1995).

#Mapping of protein Sbi by shot-gun phage display strongly suggests that the protein has only one IgG-binding domain with a deduced minimal binding domain of 52 amino acids. Furthermore, expression and analysis of NH2-terminally truncated Sbi-protein, i.e. the C-terminal part of the protein, consisting of aa 143-436 ( SEQ ID NO:4), shows that no IgG-binding activity is located in this part of the protein (Example 4). The IgG-binding domain shows a significant homology to the IgG-binding repeats of protein A (FIG. 3).