Human [20] and camel [8]. Although the molecular structures of human (HPGRP-S) and camel (CPGRP-S) ?proteins are similar with rms deviations of 0.8 A for Ca-positions, their quaternary structures are completely different. The polypeptide chain of human PGRP-S consisting of residues from 9?175 was found to adopt the monomeric state [20] while the full chain (residues, 1?71) CPGRP-S was observed in a polymeric state [12]. The crystal structure determination of CPGRP-S showed that it consisted of four crystallographically independent molecules A, B, C and D in the asymmetric unit which is arranged in a linear chain with alternating A and C contacts (Figure 5). In such an arrangement, two ligand binding sites were observed. These are situated at the sites of A and C contacts (Figures 6). In case of HPGRP-S, the corresponding surfaces of the monomer may also act as binding sites. Indeed one of the monomeric surfaces has been shown to bind to PGN [10] while the opposite face to it was assumed to bind to non-PGN types of effector molecules [20]. The comparison of amino acid sequences of CPGRP-S and HPGRP-S [9] shows the presence of several residues in the sequence of CPGRP-S on the two surfaces of the monomer that have been reported to be favorable for dimerization of proteins [8]. These residues are Ala-5, Gly-7, Ser-8 and Asn126 at the A interface and Ilu-89, HDAC-IN-3 biological activity Lys-90, Ala-94, Pro-96, Tyr97, Pro-151 and Arg-158 at the C interface. Similarly, the binding cleft at the A contact has a favourable stereochemistry for the binding of fatty acids indicating a possibility for the recognition of cell wall molecules including mycolic acid of the Mycobacterium tuberculosis. The cleft at the C contact has been shown to be involved in the binding of cell wall molecules ofbacteria other than Mycobacterium tuberculosis. These molecules included LPS, LTA and PGN of Gram-negative and Grampositive bacteria [10]. The structure of the ternary complex of CPGRP-S with LPS and SA provides another strong evidence of the recognition potential of CPGRP-S for acting against bacterial infection. The observed forcep-like shape of the cleft formed by two a-helices Aa2 and Ba2 at the A contact provides features similar as that observed in the case of other fatty acid binding proteins [21,22]. On the other hand the cleft at the 23727046 C contact consists of a specific pocket for the recognition of glycan moieties 23727046 C contact consists of a specific pocket for the recognition of glycan moieties 1326631 such as GlcNAc and MurNAc [11]. In a contrast, it was shown in the structures of the complexes of PGRP-S domain of HPGRP-Ia and HPGRP-Ib, that the peptide moiety of PGN was the initial element of recognition by the protein [23,24]. Therefore, the real issue here was whether the specificity pocket at the C contact was more suitable for binding to glycan components of PGNs or it suited more to bind to the interlinking peptide Thus it important to understand as to which of the two moieties played a more significant role in the recognition of PGNs by PGRP-S. Since glycan moieties are the conserved chemical entities of bacterial cell wall molecules these might be preferred elements for the recognition. This has been shown by several structures of the complexes of CPGRP-S with various PAMPs [9?2,19]. On the other hand, the peptide sequences in PGNs vary considerably and may require a very promiscous peptide recognition site. Also, the peptide components in PGNs interconnect the glycan chains and hence they might not be fully accessible for specific recognition by the p.
