Edinburgh Napier University

  The nucleotide sequence of the sucB gene, which encodes the dihydrolipoamide succinyltransferase component (E2o) of the 2‐oxoglutarate dehydrogenase complex of Escherichia coli K12, has been determined by the dideoxy chain‐termination method. The results extend by 1440 base pairs the previously reported sequence of 3180 base pairs, containing the sucA gene. The sucB structural gene comprises 1209 base pairs (403 codons excluding the initiating AUG), and it is preceded by a 14‐base‐pair intercistronic region containing a good ribosomal binding site. The absence of a typical terminator sequence and the presence of an IS‐like sequence downstream of sucB suggest that there may be further gene(s) in the suc operon. The IS‐like sequence is homologous with other intercistronic sequences including that between the sdhB and suc A genes, the overall gene organisation being: sdhB‐IS‐sucAsucB‐IS‐. The patterns of codon usage indicate that sucB may be more strongly expressed than suc A, consistent with the disproportionate contents of their products in the oxoglutarate dehydrogenase complex.
The predicted amino acid composition and Mr (43 607) of the succinyltransferase component agree with previous studies on the purified protein. Comparison with the corresponding acetyltransferase component of the pyruvate dehydrogenase complex (E2p, aceFgene product) indicates that each contains two analogous domains, an amino‐terminal lipoyl domain linked to a carboxy‐terminal catalytic and subunit binding domain. The lipoyl domain of the acetyltransferase (E2p) comprises three tandemly repeated ∼ 100‐residue lipoyl binding regions containing two short (∼19 residues) internal repeats, whereas the lipoyl domain of the succinyltransferase (E2o) contains just one ∼ 100‐residue lipoyl binding region, with ∼27% homology to each of the three comparable regions in E2p, and no detectable internal repeats. The catalytic and subunit binding domains, each ∼300 residues, have an overall homology of 34 % and, consistent with their combination of analogous and specific functions, some regions are more homologous than others. Both sequences feature segments rich in proline and alanine. In E2p these occur at the carboxy‐terminal ends of each of the three lipoyl binding regions, there being a particularly extended sequence at the end of the third repeat, whereas in E2o the main proline‐alanine segment is found ∼50 residues into the subunit binding domain. It is suggested that these segments could form flexible hinge‐like regions which may account for the observed mobility in these subunits and be important in the mode of action of the bound lipoyl coenzyme. The results provide a basis for defining the structure‐function relationships of the two acyltransferases and they suggest that these components are closely related in evolution.