Fig. 2. Structure and function of the dimA gene. (A) Site of insertion of the disruption plasmid and structure of the dimA gene. The pBSR1 disruption vector was recovered from the dimA^– mutant by plasmid rescue. The insertion lies in the second exon of a 3846 bp gene, encoding a putative 1242 amino acid protein. The protein contains long stretches of asparagine (N) and glutamine (Q) residues as indicated. The region between amino acids 545-676 shows extensive homology to the DNA-binding and dimerization domains of bZIP and bRLZ transcription factors. (B) Sequence alignment of the putative dimA DNA-binding and dimerization domain with examples of bZIP and bRLZ proteins from human, mouse, Drosophila and yeast (gi:19745184, gi:10835484, gi:135304, gi:17647933 and gi:135867). (C) DimA binds DNA. Binding of total soluble protein extracts prepared from bacteria expressing the putative DimA DNA-binding/dimerization domain (dimA) fused to GST was compared with extract from cells expressing GST alone (pGEX). Equal amounts of total protein were assayed and loaded. A 48 bp fragment from the 3' half of the minimal ecmO/lacZ promoter was used as a probe and poly dAdT was included as a non-specific competitor. The probe is only retarded when mixed with DimA-expressing extract. The amount of binding is reduced by the addition of a 10-fold excess of unlabeled oligonucleotide (CC). (D) The effects of varying non-specific competitor species on DNA binding. Strongest binding is evident in the absence of non-specific competitor. The addition of poly dIdC strongly reduces binding, whereas poly dAdT addition results in a small reduction in binding. Fewer retarded bands are visible (compared with C), as electrophoresis was performed at 4°C to stabilize protein DNA interactions.