Biochemical and molecular characterization of the Alcaligenes eutrophus pyruvate dehydrogenase complex and identification of a new type of dihydrolipoamide dehydrogenase

Abstract

Sequence analysis of a 6.3-kbp genomic EcoRI-fragment of Alcaligenes eutrophus, which was recently identified by using a dihydrolipoamide dehydrogenase-specific DNA probe (A. Pries, S. Hein, and A. Steinbüchel, FEMS Microbiol. Lett. 97:227-234, 1992), and of an adjacent 1.0-kbp EcoRI fragment revealed the structural genes of the A. eutrophus pyruvate dehydrogenase complex, pdhA (2,685 bp), pdhB (1,659 bp), and pdhL (1,782 bp), encoding the pyruvate dehydrogenase (E1), the dihydrolipoamide acetyltransferase (E2), and the dihydrolipoamide dehydrogenase (E3) components, respectively. Together with a 675-bp open reading frame (ORF3), the function of which remained unknown, these genes occur colinearly in one gene cluster in the order pdhA, pdhB, ORF3, and pdhL. The A. eutrophus pdhA, pdhB, and pdhL gene products exhibited significant homologies to the E1, E2, and E3 components, respectively, of the pyruvate dehydrogenase complexes of Escherichia coli and other organisms. Heterologous expression of pdhA, pdhB, and pdhL in E. coli K38(pGP1-2) and in the aceEF deletion mutant E. coli YYC202 was demonstrated by the occurrence of radiolabeled proteins in electropherograms, by spectrometric detection of enzyme activities, and by phenotypic complementation, respectively. A three-step procedure using chromatography on DEAE-Sephacel, chromatography on the triazine dye affinity medium Procion Blue H-ERD, and heat precipitation purified the E3 component of the A. eutrophus pyruvate dehydrogenase complex from the recombinant E. coli K38(pGP1-2, pT7-4SH7.3) 60-fold, recovering 41.5% of dihydrolipoamide dehydrogenase activity. Microsequencing of the purified E3 component revealed an amino acid sequence which corresponded to the N-terminal amino acid sequence deduced from the nucleotide sequence of pdhL. The N-terminal region of PdhL comprising amino acids 1 to 112 was distinguished from all other known dihydrolipoamide dehydrogenases. It resembled the N terminus of dihydrolipoamide acyltransferases, and it contained one single lipoyl domain which was separated by an adjacent hinge region from the C-terminal region of the protein that exhibited high homology to classical dihydrolipoamide dehydrogenases.