Structure and Regulation of KGD1, the Structural Gene for Yeast α-Ketoglutarate Dehydrogenase

Abstract

Nuclear respiratory-defective mutants of Saccharomyces cerevisiae have been screened for lesions in the mitochondrial α-ketoglutarate dehydrogenase complex. Strains assigned to complementation group G70 were ascertained to be deficient in enzyme activity due to mutations in the KGD1 gene coding for the α-ketoglutarate dehydrogenase component of the complex. The KGD1 gene has been cloned by transformation of a representative kgd1 mutant, C225/U1, with a recombinant plasmid library of wild-type yeast nuclear DNA. Transformants containing the gene on a multicopy plasmid had three- to four-times-higher α-ketoglutarate dehydrogenase activity than did wild-type S. cerevisiae. Substitution of the chromosomal copy of KGD1 with a disrupted allele (kgd1::URA3) induced a deficiency in α-ketoglutarate dehydrogenase. The sequence of the cloned region of DNA which complements kgd1 mutants was found to have an open reading frame of 3,042 nucleotides capable of coding for a protein of M_w 114,470. The encoded protein had 38% identical residues with the reported sequence of α-ketoglutarate dehydrogenase from Escherichia coli. Two lines of evidence indicated that transcription of KGD1 is catabolite repressed. Higher steady-state levels of KGD1 mRNA were detected in wild-type yeast grown on the nonrepressible sugar galactose than in yeast grown on high glucose. Regulation of KGD1 was also studied by fusing different 5′-flanking regions of KGD1 to the lacZ gene of E. coli and measuring the expression of β-galactosidase in yeast. Transformants harboring a fusion of 693 nucleotides of the 5′-flanking sequence expressed 10 times more β-galactosidase activity when grown under derepressed conditions. The response to the carbon source was reduced dramatically when the same lacZ fusion was present in a hap2 or hap3 mutant. The promoter element(s) responsible for the regulated expression of KGD1 has been mapped to the —354 to —143 region. This region contained several putative activation sites with sequences matching the core element proposed to be essential for binding of the HAP2 and HAP3 regulatory proteins.