24:75-87. FhuA (14), TbpA is usually proposed to adopt an antiparallel -barrel structure (28). TbpB (65 to 90 kDa) is considered to be an outer membrane protein that is anchored to the membrane via the lipidated N-terminal part of the protein (11). There is evidence which indicates that TbpB consists of two globular domains which correspond to the N- and C-terminal halves of the protein (23). It has been shown in vitro that htf binds primarily to the TbpA part of the receptor and with a lower affinity to TbpB (13). The presence of TbpB, however, increases the affinity of TbpA for the iron-loaded form of htf (13), which thus allows the bacterial receptor to compete successfully with the htf receptor. TbpB has been shown to be a potential candidate for an antimeningococcal vaccine (17). Meningococcal infections represent a major worldwide health problem (2), and there is no vaccine available against serogroup B isotype I meningococcal strain, B16B6, using a novel approach based on the prediction of ligand-binding sites from a protein sequence. A recently described prediction method was shown to identify ligand-binding sites from a diverse range of protein database entries with a satisfactory score (10). We have used this method to identify the transferrin-binding site on TbpB. As mentioned above, TbpB binds to two different ligands (TbpA and htf), and the output of the prediction method corresponds to an overall prediction of binding sites for both ligands without being able to differentiate between ligand types. Site-directed mutants were thus prepared in order to verify the implication of the predicted regions in transferrin binding. Immunization experiments were carried out with TbpB mutants devoid of transferrin-binding activity to assess the immunogenicities of these proteins. MATERIALS AND METHODS Site-directed mutagenesis, protein expression, and purification of MBP-TbpB fusion proteins. The construction of a recombinant plasmid corresponding to the expression vector pMAL-c2 (Biolabs) made up of a DNA fragment coding for amino acids 2 to 579 of TbpB from B16B6 (the full-length mature protein devoid of the N-terminal cysteine) in fusion with the maltose-binding protein (MBP) has been described previously (22). Site-directed mutations were introduced into this plasmid using the Mouse monoclonal to EGF Quick Change site-directed mutagenesis kit (Stratagene) following the manufacturer’s instructions. Lysine (K) or PCI-34051 arginine (R) residues were replaced by asparagine (N) or glutamate (E) using pairs of mutagenic oligonucleotides, which are listed in Table ?Table1.1. XL-2 Blue (Stratagene) was transformed with the resulting plasmids, and bacterial cultures were produced on Luria broth medium supplemented with 100 g of ampicillin ml?1. In all cases, the introduction of the desired point mutations was verified by DNA sequencing of the entire insert. TABLE 1. Primers used for site-directed mutagenesis of MBP-TbpB of strain B16B6 strains B16B6 (B:2a:P1.2), 2713 (B:NT:P1.2), 2717 (B:NT:P1.2), N106/NK (B:NT:P1.5), and 64/92 (B:2b:NST) were kindly provided by D. Caugant (NIPH, Oslo, Norway) via B. Rokbi (Aventis Pasteur). Each strain was grown in Mueller-Hinton broth medium with the chelator ethylene diamine di-B16B6. The prediction of ligand-binding sites in the absence of any three-dimensional structural information remains a challenge. However, a recently described method (10) has been shown to identify binding domains with satisfactory reliability. The method is based on the determination of the mean hydrophobicity ( strain B16B6. (A) Residues 53PRRNA57 were found to lie within the RBD zone. (B to D) Evaluation PCI-34051 of effects of single-amino-acid replacements around the positions of the residues in the RDB zone, indicating changes in binding potential. K51N has no significant effect on the RBD zone (B); R54N (C) and R55N (D) abolish the binding potential (the RBD zone is almost empty). The sequence of TbpB B16B6 has been submitted to this ligand-binding prediction method (10), and amino acids with some degree of binding potential are highlighted in Fig. ?Fig.2.2. Based on this initial analysis, three clusters of amino acids with detectable binding PCI-34051 potential could be distinguished and were called RBD1, -2, and -3. The identification of RBD1 is illustrated in Fig. ?Fig.1,1, which shows the plot of strain B16B6. Amino acids predicted to have a strong (red), moderate (blue), or weak (green) binding potential are highlighted. The three major receptor-binding domains are boxed. Identification of amino acids within the three RBDs with a major.