Résumé
Transcriptional antiterminators of the BglG/SacY family are bacterial regulatory proteins able to prevent the premature arrest of transcription through specific binding to a ribonucleic antiterminator (RAT) sequence. The RNA recognition module of these regulators is made of the 55-amino acid long N-terminal domain which can by itself promote efficient antitermination activity
in vivo and RNA binding
in vitro. The structure of this domain, which was called CAT for co-antiterminator, has first been determined for SacY from
Bacillus subtilis and the putative surface contacting RNA has been defined by NMR footprinting. Here we have performed a genetic mapping of the SacY-CAT RNA binding site by substituting 24 amino acid residues including those previously identified by NMR, the highly conserved residues in the 55 homologous antiterminators recognised in the databases and all the positively charged residues. A total of 57 SacY-CAT variants have been constructed and tested
in vivo for their antitermination efficiency. A few of these variants were then purified in order to analyse their RNA binding properties by surface plasmon resonance and to check their structural integrity by NMR. The present study validates and clarifies the RNA interacting surface previously mapped by NMR. The residues that are the most intolerant to substitutions, Asn8, His9, Asn10, Gly25, Gly27, and Phe30, are aligned across the CAT dimer interface and form the core of the RNA binding site. Three highly conserved residues stand outside the interaction surface but are essential for maintaining the CAT dimeric structure (Phe47) or may play an important functional role in the full length protein (Glu20 and Lys32). Interestingly, none of the twelve positively charged residues of SacY-CAT are crucial for the antitermination activity. By replacing three Lys residues and combining the Ala26→Arg mutation that significantly enhanced the affinity for RNA, we engineered a SacY-CAT variant that should be suitable for NMR study of the complex.