Mitochondrial import and processing of rat liver carnitine palmitoyltransferase II defines the amino terminus of the mature protein. Possibility of differential modification of the rat and human isoforms.

[35S]Methionine-labeled porcine heart citrate synthase (used here as a positive control) and rat liver carnitine palmitoyltransferase II (CPT II) were generated by in vitro transcription and translation of their cDNA constructs in appropriate Bluescript plasmids. Each product was imported into rat liver mitochondria in an energy-dependent manner to yield an immunoprecipitable protein of smaller size that comigrated with the corresponding purified enzyme. The size shift occurring with citrate synthase was consistent with the removal of the postulated 27-amino acid leader peptide. To determine the amino terminus of mature CPT II, [35S]methionine- or [3H]leucine-labeled material (after import and processing) was subjected to Edman degradation, followed by counting of the radioactivity released on each cycle. The results established that the precursor targeting peptide was cleaved between leucine 25 and serine 26 in the previously deduced amino acid sequence. Taken in conjunction with the recent report of Finocchiaro et al. (Finocchiaro, G., Taroni, F., Rocchi, M., Martin, A. L., Colombo, I., Tarelli, G. T., and DiDonato, S. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 661-665), the present results establish three key points concerning the rat and human forms of CPT II. First, it appears that in both species the initial translation product contains 658 amino acids and, upon mitochondrial import, is reduced in length by 25 residues through cleavage at an identical site. Second, the difference in electrophoretic mobility between the two mature proteins (documented earlier) presumably reflects either anomalous behavior of one of them on polyacrylamide gels or differential covalent modification. Finally, the recent suggestion by Brady et al. (Brady, P. S., Liu, J. S., Park, E. A., Hanson, R. W., and Brady, L. J. (1991) FASEB J. 5, A817) that our CPT II cDNA construct is incomplete in the 5'-coding region is refuted.