Construction and nucleotide sequence of a cDNA encoding the full-length preprotein for human branched chain acyltransferase

A cDNA (1.6 kilobases) for branched chain acyltransferase (E2b) isolated from a human liver library encoded only the amino-terminal half of the protein (Hummel, K.B., Litwer, S., Bradford, A.P., Aitken, A., Danner, D.J., and Yeaman, S.J. (1988) J. Biol. Chem. 263, 6165-6168). Here we report the isolation of other cDNAs which encode the carboxyl-terminal half of E2b and the construction of a cDNA which encodes the entire pre-E2b. cDNA from the original clone encoding the leader sequence, lipoate binding domain, and E3 binding domain was ligated to the cDNA from a clone which by restriction maps contained an additional 3' sequence. Both cDNAs used in the construct made a fusion protein in their origininal phage isolate recognized by antibodies to E2b. The nucleotide sequence of the constructed cDNA was determined, and the 1431 base pairs in the open reading frame encoded a protein of 477 amino acids. In vitro transcription and translation of this cDNA produced a 57-kDa protein recognized by E2b-specific antibodies. Mouse liver mitochondria imported and processed the 57-kDa protein to a 52-kDa antigenic protein which co-migrated with E2b isolated from tissue. Comparing the protein structure of this human pre-E2b protein with that for other acyltransferase proteins showed a similarity in structure throughout all the proteins suggesting evolutionary conservation. Branched chain acyltransferase from Pseudomonas putida showed the most similarity to human E2b.