Fundamental research I Molecular characterization I Cleavage specificity I Cellular localization I Gene organization and regulation I Research projects list
Fundamental researchProhormones and proenzymes are initially synthesized as precursors which are then post-translationally modified resulting in the final secretable active form of the protein or polypeptide hormone. Limited proteolysis of these precursors at specific amino acids represents one of the principal post-translational modifications that occur during the intracellular transport and packaging of the protein. Although the time scale and organelles in which such processing occurs have now been well elucidated, the enigma surrounding the nature of the processing enzymes and the mechanisms used by these convertases to fashion the final form of the protein have only recently been uncovered. Generally, following the co-translational N-glycosylation of the nascent protein in the endoplasmic reticulum (ER) and trimming of the sugar chains in the Golgi apparatus, initial cleavage of the proprotein seems to begin in the last lamellae of the Golgi apparatus, known as the Trans Golgi Network (TGN), and to proceed within the secretory granules. The protease(s) involved seem to preferentially cleave C-terminal to pairs of basic amino acids or post specific single basic residues. Recently, other cleavage sites involving small amino acids such as Thr, Ser and Ala as well as hydrophic ones such as Leu, Val and Met have been recognized in various precursors, including the Alzheimer disease b-amyloid precursor protein (bAPP) and the sterol regulatory element binding proteins (SREBPs).Our laboratory is interested in the elucidation at the molecular level of the biosynthetic machinery involved in the synthesis and processing of pro-hormones and pro-enzymes. This research involves a number of approaches using affinity protein purification, microsequencing, immunocytochemistry, molecular cloning, in situ hybridization, cellular expression and mutagenesis. Presently, we are seeking answers to the following fundamental questions:
Molecular characterization of convertasesBased on the concept of sequence conservation around the active sites of serine proteinases, PCR, using degenerate oligonucleotides coding for the sequence around the catalytically important Asn and the active site Ser of kexin/subtilisin-like enzymes, allowed the isolation of five out of the seven members of the mammalian subtilisin/kexin-like enzyme family. These include the ubiquitously expressed furin, the widely expressed PC7 and PACE4, the enzyme PC5 which is expressed in some endocrine and some non-endocrine cells, and the neural and endocrine convertases PC1 and PC2 and finally the enzyme PC4 which is expressed primarily in testicular germ cells and in ovaries. In addition, very recently we have cloned two novel subtilisin/kexin isozymes called SKI-1 and SKI-2, of which SKI-1 is the first enzyme responsible for the processing of SREBPs in the endoplasmic reticulum and hence is critical in the intracellular pathway leading to the generation of cholesterol and fatty acids.The comparative architectural features of the 7 mammalian subtilisin-like pro-protein convertases revealed that the sizes of the eukaryotic members vary from 637 to 1877 amino acids. All the members contain a signal peptide, and each convertase possesses a unique pro-segment which must be excised in order to generate an active proteinase. The rate of removal of the N-terminal pro-segment represents a mechanism by which the cell controls the rate of pro-protein processing. Recent studies with PC1 and PC2 demonstrated that whereas pro-PC1 is rapidly cleaved into PC1 within the ER, pro-PC2 slowly exits from the ER and is processed into PC2 within the TGN. The consequence of this different temporal activation of PC1 and PC2 is that, in cells which express both enzymes, PC1 will cleave precursors before PC2, leading to an ordered cleavage mechanism. This conclusion rationalizes the successive cleavages of either POMC or pro-insulin by PC1 and PC2. The autocatalytic zymogen activation of all the other convertases including SKI-1, occurs in the ER. The catalytic region of each convertase represents the segment exhibiting the highest protein sequence identity between the convertases (about 50-60%) and PC2 is the only convertase which shows the presence of an Asp* residue in place of the catalytically important oxy-anion hole Asn* found in all other subtilisin-like proteinases. This Asp* residue in PC2 is important for the specific binding of PC2 to the pan-neuronal protein 7B2, discovered in the laboratory in 1983. This complex formation results in an enhanced rate of zymogen activation of proPC2 to PC2.
Cleavage specificity of the convertasesThe cleavage specificity of each convertase was studied either by their cellular co-expression with precursor substrates and defining the processed products or in vitro with a number of peptide substrates. The precursors studied were POMC, pro-enkephalin, pro-dynorphin, pro-renin, pro-somatostatin, pro-insulin, pro-LHRH, the neurotrophin precursors including proNGF and the HIV-1 glycoprotein gp160. It was shown that each convertase selectively cleaved these precursors at either single (motif Arg-(X-X)n-Arg¯ , n=1, 2 or 3) or pairs of basic residues. These results are consistent with the proposed hypothesis concerning the physiological role of PCs as distinct pro-protein convertases acting alone or together to produce a set of tissue-specific maturation products both in the brain and in peripheral tissues. Very recently, SKI-1 was shown to cleave precursors such as SREBP1 at Arg-X-X-Leu¯ and pro-brain-derived neurotrophic factor (proBDNF) at Arg-X-X-Thr¯. Both clivages occur within the endoplasmic reticulum.
Cellular localization of the PCsEach convertase exhibits a unique C-terminal sequence, with PC1, PC2, PC5-A, PC4 and PACE4 being soluble enzymes and only furin, PC5-B and PC7 exhibiting a transmembrane anchor which localizes them to the TGN. Tissue distribution and intracellular localization studies demonstrated that PC1, PC2 and PC5-A is most remarkable in the region of high sequence homology which constitutes the N-terminal two thirds of the molecules, suggesting A are stored in dense core secretory granules. This is in agreement with their ability to process precursors into products stored within secretory granules (regulated pathway). In contrast, furin, PC7, PACE4 and PC5-B are the main convertases of the constitutive secretory pathway, processing precursors in the TGN or at the cell surface. Recent data demonstrated that SKI-1 primarily localizes either in the ER, or in the TGN and endosomal-like structures depending on the level of cholesterol in the medium.
Gene organization and regulationWithin the coding sequences, the intron/exon organization of the PC genes is very similar. This conservation that these genes either evolved from a common ancestral sequence (of which PC7 seems to be the closest) or one from another by deletions, insertions and other mutations around essential functional domains. Structural variations among these genes are clustered at the 3' end and may be associated with distinctive appendages which confer particular properties to each enzyme such as its intracellular localization and/or its catabolism. Regulation studies in the pituitary and in cell lines, demonstrated that the level of PC1 and PC2 transcripts are co-regulated with POMC, in either dopamine or glucocorticoid challenges. The relative expression and distinct enzymatic activities of PC1 and PC2 provided a rationale for the observed tissue-specific processing pattern of POMC. Indeed, changes in PC1 and PC2 gene expression may have dramatic effects, such as changing the biologically active peptides produced, or the effects may be more subtle, such as modifying the temporal kinetics of processing, resulting in the same peptide products together with a range of precursor intermediates. Comparative phylogenetic analysis if SKI-1 and SKI-2 suggested that these are much more ancestral than any of the PCs.
Research projects list