N-Arg dibasic convertase is a me tidase from rat brain cortex and testis that cleaves peptide substrates on the N terminu ofArg residues in dibasic s s.By using both an oinu ide and antibodies to screen a rat testis cDNA library, a fill-length cDNA (4,5). On the basis of its similarity to subtilisin and to furin (6), a human homolog of the Kex2 protein, a family of prohormone convertases (PCs) has been identified by PCR techniques. Their involvement in processing of a number of propeptides and proproteins was inferred mainly from cotransfection experiments (7-9), and for PC1, by the use of antisense mRNA (10).Characterization of putative processing endoproteases by classical biochemical techniques has led to the identification of a number of activities, selective for basic residues in precursors, that belong to the four classes of proteases (metallo-, serine, aspartyl, and thiol enzymes; for review, see ref. 11). This suggested that more than one processing endoprotease family could exist (12). To our knowledge, none of these basic-residue-specific enzymes has been cloned.Recently, a metalloendopeptidase was completely purified from rat testis and shown to cleave a number of peptide substrates on the N terminus of Arg residues in dibasic moieties (13). This enzyme was also present in rat brain cortex and its functional properties appeared undistinguishable from those of the somatostatin-28 convertase activity previously identified in this tissue (14,15). By using microsequencing of tryptic fragments of the purified enzyme to design an oligonucleotide probe and polyclonal antibodies raised against the purified protein (13) (20)].The in vitro highly restricted specificity ofNRD convertase for Arg residues in dibasic processing signals and its belonging to the M16 family, which contains other enzymes involved in maturation, suggest that this enzyme is the prototype of a distinct family of processing endoproteases.
MATERIALS AND METHODSIsolatin and Cherizatin of cDNA Cones E d NRD Convertae. Four tryptic fragments were sequenced after digestion of previously purified NRD convertase following native PAGE. One fragment, GMQLIYLPPSPLLAE, was used to design the following degenerate inosine-containing oligonucleotide: 5'-GGIGGIAG(A/G)TAIATIA(G/A)(T/ C)TGCATICC-3'. Two additional peptides were obtained by endolysine C treatment (13) autoradiography of the filters, positive phage plaques were isolated and rescreened to obtain single purified phage isolates. A similar aliquot ofthe library was plated onto NZ-amine medium plates for screening using polyclonal antibodies raised Abbreviations: NRD convertase, N-arginine dibasic convertase; IDE, insulin degrading enzyme; MPP, mitochondrial matrixprocessing peptidase.