Fap1, a 200 kDa glycoprotein, is the major subunit of the long fimbriae of Streptococcus parasanguinis . A genomic island in the chromosome of S. parasanguinis FW213 is involved in Fap1 biosynthesis. This genomic island contains a seven-gene cluster located immediately downstream of fap1. This cluster includes secY2, gap1, gap2, gap3 (formerly orf1 to orf3), secA2, gtf1 and gtf2 that are differentially involved in Fap1 glycosylation and secretion [10, 11]. Our previous studies showed that the gene product Gap3 is involved in Fap1 glycosylation . In this study, we determined the critical amino acid residues that are important for its function. We first aligned Gap3 and its homologues to find the conserved regions and then took bioinformatics approaches to further analyze functional domains. A conserved peptide sequence, 62PDLPIL67 was found to be critical for Gap3 function and Fap1 glycosylation. This sequence domain was also found in a signal receiver domain of a bacterial polysaccharide biosynthesis regulator protein, GelA, of Sphingomonas elodea (NCBI, AAP46184). This may explain why this region is important for Fap1 glycosylation. Interestingly, the other predicted important peptide sequence from residue 144 to 157 of Gap3 is not required for Fap1 glycosylation despite the fact it has a putative coiled-coil structure (a potential protein-protein interaction domain). Therefore, we conclude the 62PDLPIL67 sequence is critical for Gap3 function.
Site-directed mutagenesis was used to replace every amino acid in the sequence 62PDLPIL67 by a disfavored amino acid  to determine which amino acid(s) in this region are important for Fap1 glycosylation. L64R, P65R and L67T mutants failed to produce fully-glycosylated mature Fap1, indicating the amino acids L64, P65 and L67 are critical for the complete glycosylation of Fap1. The replacement of the hydrophobic residues L64, P65 and L67 by charged and hydrophilic residues (R and T) did not change overall Gap3 structure as all these mutant variants expressed Gap3 equally (Fig. 2D). These findings highlight the importance of three residues for Gap3 function.
Site-directed mutagenesis of 7 other conserved amino acids V33, F35, N37, P38, S42, N54 and R59 did not alter Fap1 glycosylation despite these amino acids (with the only exception of R59) residing in putative conserved functional domains. Mutagenesis of an adjacent leucine residue L75 did not change Fap1 glycosylation either. These results support the notion that these three important residues identified in 62PDLPIL67 region are site-specific in Gap3 function.
The mechanism for how this region is involved in Gap3 function currently is unknown. Many hydrophobic residues are functionally important in the protein-protein interaction processes . The PDLPIL region has two proline and two leucine residues, both are hydrophobic amino acids and have been shown to play important roles in molecular- or substrate-recognition through hydrophobic interactions [27–29]. In this study, we replaced the conserved, hydrophobic amino acids with hydrophilic or charged, polar amino acids whose properties are totally different (L64R, P65R and L67T). The site-directed mutagenesis resulted in the attenuation of Gap3 activity in Fap1 complete glycosylation. Interestingly, the mutagenesis of a less conserved amino acid, I66 in the same region, had only a minor effect on mature Fap1 production, suggesting that I66 is less critical in Gap3 function and may not be crucial in mediating protein-protein interactions that are important for Gap3 mediated Fap1 glycosylation.
Fap1 is a structural subunit of bacterial long fimbriae and essential for fimbrial assembly and bacterial adhesion . In this study, we demonstrated that the partially glycosylated Fap1 precursor produced by the gap3 site-directed mutants, L64R, P65R and L67T could not be assembled into the long fimbriae, suggesting that glycosylation plays an important role in Fap1 mediated fimbrial formation. All mutants that failed to produce mature Fap1 could not adhere well to SHA in vitro, further demonstrating the importance of mature Fap1 in bacterial adhesion. Interestingly the I66N mutant that produces both mature and premature Fap1 still displays fimbriae on the cell surface and is able to adhere to SHA, supporting the concept that mature Fap1 is required for fimbrial biogenesis and bacterial adhesion. Understanding the underlying mechanism will help to define Fap1 biogenesis and fimbrial assembly pathways.
Our finding of 3 critical residues in Gap3 function does not preclude the possible roles of other amino acids that were not selected for this study. Gap3 protein is predicted to be involved in a glycosylation-related complex with other Gap proteins (Gap1 and Gap2). Our future investigations will be to determine if there are more active sites in Gap3 associated with protein-protein interaction during Fap1 biogenesis process.
In conclusion, we have identified that 3 highly conserved residues, L64, P65 and L67 of Gap3 that are critical for complete Fap1 glycosylation, long fimbrial formation and bacterial adhesion. Because the glycosylation-association protein is conserved in many other streptococci and staphylococci, and the three residues we identified are also highly conserved, this work may shed a light on further understanding the mechanism of biogenesis of Fap1-like serine-rich glycoproteins.