BenM is a transcriptional regulator found in the soil bacterium Acinetobacter baylyi
ADP1. Together with another transcriptional regulator, CatM, this regulator controls a large set of genes used by A. baylyi
for aromatic compound degradation (Collier et al.
; Clark et al.
; Brzostowicz et al.
; Ezezika et al.
; Clark, Haddad et al.
). BenM is a member of the family of LysR-type transcriptional regulators (LTTR), the largest homologous family of transcriptional regulators in bacteria (Schell, 1993
; Henikoff et al.
; Pareja et al.
). LTTRs control genes of diverse function that include the synthesis of virulence factors, CO2
fixation, antibiotic resistance, catabolism of aromatic compounds, nodule formation of N2
-fixing bacteria and amino-acid biosynthesis. LTTRs are present in abundant and diverse bacterial genera (Diaz & Prieto, 2000
). BenM specifically belongs to a subclass of this family involved in aromatic compound catabolism (Tropel & van der Meer, 2004
Mutational analyses of this group of regulators demonstrate that the N-terminal region is required for DNA binding. This region comprises residues 1 to approximately 66 and displays high sequence identity among members (Schell et al.
). Removal of the N-terminal region aids structural studies by circumventing the insolubility problems associated with the full-length versions of these proteins, as observed in BenM and CatM (Clark, Haddad et al.
) and other LysR-type regulators such as CysB, OxyR and Cbl (Verschueren et al.
; Tyrrell et al.
; Choi et al.
; Stec et al.
). In the case of CysB, the addition of sulfobetaines was necessary to obtain crystals (Verschueren et al.
). BenM and CatM crystals were obtained using high concentrations of NaCl, glycerol and imidazole at a nonphysiological pH (Clark, Haddad et al.
; Ezezika et al.
). As a result, the structure determinations of truncated versions of LTTRs have been more successful than those of the full-length proteins. CbnR is the only example of a complete full-length LTTR structure and high concentrations of NaCl were involved in its crystallization (Muraoka, Okumura, Ogawa et al.
). Although a complete tetrameric DntR has been crystallized, the DNA-binding domain, which contains a helix–turn–helix motif, was poorly defined owing to weak electron density in the corresponding region (Smirnova et al.
). Other LTTRs, such as AmpR and GltC, have been difficult to purify from overexpression systems (Bishop & Weiner, 1993
; Picossi et al.
CbnR is a homotetramer, but the CbnR monomers exist as two different conformations of the same polypeptide chain, making the tetrameric molecule a dimer of dimers. Tetramers are generally the active form of LTTRs, as noted for CbnR, CysB, NahR and DntR (Hryniewicz & Kredich, 1994
; Muraoka, Okumura, Ogawa et al.
; Schell et al.
; Smirnova et al.
). BenM also exists as a tetramer in its active full-length form, although the reported molecular weight of 180 kDa is 25% higher than predicted (Bundy et al.
). Gel-filtration studies carried out on truncated BenM lacking the N-terminal DNA-binding domain showed that the effector-binding domain (EBD) exists as a homodimer in solution (Clark, Haddad et al.
BenM responds synergistically to two effector ligands, benzoate and cis
-muconate (hereafter referred to as muconate), to activate transcription of the ben
genes (Bundy et al.
; Clark, Phillips et al.
). In contrast, CatM, a homolog of BenM, only responds to muconate. Recently, the structures of the EBDs of BenM and CatM were determined with and without their cognate effectors (Ezezika et al.
). These structures identify two distinct binding sites for the effectors and identify conformational changes associated with ligand binding that are likely to be associated with transcriptional activation. The primary effector-binding site is located at an interdomain cleft and can accommodate muconate or benzoate. The secondary effector-binding site can only accommodate benzoate in a hydrophobic pocket that communicates with the primary binding site. CatM lacks this secondary binding site. One unexplored area of the BenM structural studies is the interaction between its subunits. Here, we report two additional structures of the BenM EBD in different space groups that form tetramers and high-order oligomers in their unit cells. These structures suggest a general scheme by which this family of proteins might form oligomers. Surface-oligomerization domains were identified and analyzed in the context of other members of this family. Based on this analysis, we propose a model whereby the same interactions that yield biologically relevant LTTR oligomers can, under some conditions, contribute to the solubility problems associated with this family of transcriptional regulators.