Proprotein Convertase Models based on the Crystal Structures of Furin and Kexin: Explanation of their Specificity

Proteases for processing proneuropeptides into peptide neurotransmitters and hormones

Peptide neurotransmitters and peptide hormones, collectively known as neuropeptides, are required for cell-cell communication in neurotransmission and for regulation of endocrine functions. Neuropeptides are synthesized from protein precursors (termed proneuropeptides or prohormones) that require proteolytic processing primarily within secretory vesicles that store and secrete the mature neuropeptides to control target cellular and organ systems. This review describes interdisciplinary strategies that have elucidated two primary protease pathways for prohormone processing consisting of the cysteine protease pathway mediated by secretory vesicle cathepsin L and the well-known subtilisin-like proprotein convertase pathway that together support neuropeptide biosynthesis. Importantly, this review discusses important areas of current and future biomedical neuropeptide research with respect to biological regulation, inhibitors, structural features of proneuropeptide and protease interactions, and peptidomics combined with proteomics for systems biological approaches. Future studies that gain in-depth understanding of protease mechanisms for generating active neuropeptides will be instrumental for translational research to develop pharmacological strategies for regulation of neuropeptide functions. Pharmacological applications for neuropeptide research may provide valuable therapeutics in health and disease.

Potential opportunity in the development of new therapeutic agents based on endogenous and exogenous inhibitors of the proprotein convertases

The proprotein convertases (PCs) are responsible for the endoproteolytic processing of various protein precursors (e.g., growth factors, receptors, adhesion molecules, and matrix metalloproteinases) implicated in several diseases such as obesity, diabetes, atherosclerosis, cancer, and Alzheimer disease. The potential clinical and pharmacological role of the PCs has fostered the development of various PC‐inhibitors. In this review we summarized the recent findings on PCs inhibitors, their mode of actions and potential use in the therapy of various diseases. © 2006 Wiley Periodicals, Inc. Med Res Rev, 27, No. 5, 631–648, 2007

The crystal structure of PCSK9: a regulator of plasma LDL-cholesterol

Proprotein convertase subtilisin kexin type 9 (PCSK9) has been shown to be involved in the regulation of extracellular levels of the low-density lipoprotien receptor (LDLR). Although PCSK9 is a subtilase, it has not been shown to degrade the LDLR, and its LDLR-lowering mechanism remains uncertain. Here we report the crystal structure of human PCSK9 at 2.3 A resolution. PCSK9 has subtilisin-like pro- and catalytic domains, and the stable interaction between these domains prevents access to PCSK9's catalytic site. The C-terminal domain of PCSK9 has a novel protein fold and may mediate protein-protein interactions. The structure of PCSK9 provides insight into its biochemical characteristics and biological function.

Two alpha subunits and one beta subunit of meprin zinc-endopeptidases are differentially expressed in the zebrafish Danio rerio

Meprins are members of the astacin family of metalloproteases expressed in epithelial tissues, intestinal leukocytes and certain cancer cells. In mammals, there are two homologous subunits, which form complex glycosylated disulfide-bonded homo- and heterooligomers. Both human meprin alpha and meprin beta cleave several basement membrane components, suggesting a role in epithelial differentiation and cell migration. There is also evidence that meprin beta is involved in immune defence owing to its capability of activating interleukin-1beta and the diminished mobility of intestinal leukocytes in meprin beta-knockout mice. Here we show for the first time by reverse transcription PCR, immunoblotting and immunofluorescence analyses that meprins are expressed not only in mammals, but also in the zebrafish Danio rerio. In contrast to the human, mouse and rat enzymes, zebrafish meprins are encoded by three genes, corresponding to two homologous alpha subunits and one beta subunit. Observations at both the mRNA and protein level indicate a broad distribution of meprins in zebrafish. However, there are strikingly different expression patterns of the three subunits, which is consistent with meprin expression in mammals. Hence, D. rerio appears to be a suitable model to gain insight into the basic physiological functions of meprin metalloproteases.

Differential P1 arginine and lysine recognition in the prototypical proprotein convertase Kex2

The high-resolution crystal structure of kexin (Kex2) in complex with a peptidyl-chloromethylketone inhibitor containing a noncognate lysine at the P(1) position provides the structural basis for the differential lysine/arginine selectivity that defines the prohormone (proprotein) convertase (PC) family. By comparison with the previous structures of Kex2 and furin, this structure of the acylated enzyme provides a basis for the observed decrease in the acylation rate with substrates containing a lysine at P(1) and the absence of an effect on the deacylation rate without involving mobility of the S(1) lid. The structure of the complex shows that a secondary subsite in the S(1) pocket is present, and that this site recognizes and binds the P(1) lysine in a more shallow fashion than arginine. This results in a displacement of the bound peptide away from the S385 nucleophile relative to substrates containing a P(1) arginine. It is concluded that this alternate binding site and resultant displacement of the scissile bond in the active site results in the observed decrease in the acylation rate. Studies of the inactivation kinetics of Kex2 by two peptidyl chloromethylketone inhibitors demonstrates that the selectivity between lysine and arginine at the P(1) position arises at the acylation step, consistent with what was observed with peptidyl substrates [Rockwell NC, Fuller RS (2001) J Biol Chem 276:38394-38399].

Structural and biophysical studies of PCSK9 and its mutants linked to familial hypercholesterolemia

Proprotein convertase subtilisin kexin type 9 (PCSK9) lowers the abundance of surface low-density lipoprotein (LDL) receptor through an undefined mechanism. The structure of human PCSK9 shows the subtilisin-like catalytic site blocked by the prodomain in a noncovalent complex and inaccessible to exogenous ligands, and that the C-terminal domain has a novel fold. Biosensor studies show that PCSK9 binds the extracellular domain of LDL receptor with K(d) = 170 nM at the neutral pH of plasma, but with a K(d) as low as 1 nM at the acidic pH of endosomes. The D374Y gain-of-function mutant, associated with hypercholesterolemia and early-onset cardiovascular disease, binds the receptor 25 times more tightly than wild-type PCSK9 at neutral pH and remains exclusively in a high-affinity complex at the acidic pH. PCSK9 may diminish LDL receptors by a mechanism that requires direct binding but not necessarily receptor proteolysis.

Cleavage and oligomerization of gliomedin, a transmembrane collagen required for node of ranvier formation

Gliomedin, which has been implicated as a major player in genesis of the nodes of Ranvier, contains two collagenous domains and an olfactomedin-like domain and belongs to the group of type II transmembrane collagens that includes collagens XIII and XVII and ectodysplasin A. One characteristic of this protein family is that constituent proteins can exist in both transmembrane and soluble forms. Recently, gliomedin expressed at the tips of Schwann cell microvilli was found to bind axonal adhesion molecules neurofascin and NrCAM in interactions essential for Na(+)-channel clustering at the nodes of Ranvier in myelinating peripheral nerves. Interestingly, exogenously added olfactomedin domain was found to have the same effect as intact gliomedin. Here we analyze the tissue form of gliomedin and demonstrate that the molecule not only exists as full-length gliomedin but also as a soluble form shed from the cell surface in a furin-dependent manner. In addition, gliomedin can be further proteolytically processed by bone morphogenetic protein 1/Tolloid-like enzymes, resulting in release of the olfactomedin domain from the collagen domains. Interestingly, the later cleavage induces formation of higher order, insoluble molecular aggregates that may play important roles in Na(+)-channel clustering.

Engineering of alpha1-antitrypsin variants selective for subtilisin-like proprotein convertases PACE4 and PC6: importance of the P2' residue in stable complex formation of the serpin with proprotein convertase

Furin and PACE4, members of the subtilisin-like proprotein convertase (SPC) family, have been implicated in the metastatic progression of certain tumors in addition to the activation of viral coat proteins and bacterial toxins, indicating that these enzymes are potential targets for therapeutic agents. Alpha1-Antitrypsin Portland is an engineered alpha1-antitrypsin designed as a furin-specific inhibitor and has been used as a tool in the functional analysis of furin. In this work, we engineered rat alpha1-antitrypsin to create a PACE4-specific inhibitor. Substituting Arg-Arg-Arg-Arg for Ala-Val-Pro-Met(352) at P4-P1 and Ala for Leu(354) at P2' created a potent PACE4- and PC6-specific inhibitor. This variant (RRRRSA) formed an SDS- and heat-stable serpin/proteinase complex with PACE4 or PC6 and inhibited both enzyme activities. The RRRRSA variant was efficiently cleaved by furin without formation of the stable complex. This is the first report of a highly selective protein-based inhibitor of PACE4 and PC6. This inhibitor will be useful in delineating the roles of PACE4 and PC6 localized in the extracellular matrix.

Processing and trafficking of a prohormone convertase 2 active site mutant

Processing of most PC zymogens is required for successful folding and/or passage through the secretory pathway; active site mutants are retained in the ER and degraded. We here report that the active site serine mutant of PC2 (PC2-S383A) was efficiently secreted as the intact zymogen in CHO-K1 cells, suggesting that its propeptide can productively insert into the mutated binding pocket without causing misfolding. In AtT-20 cells, PC2-S383A was cleaved at the secondary cleavage site within the propeptide; this cleavage event was pH-dependent and was inhibited by a proprotein convertase inhibitor. In vitro digestion of PC2-S383A with various convertases indicates that this site is accessible to in trans cleavage. Abundant immunoreactive S383A PC2 was found in secretory granules, supporting the idea that this protein is efficiently trafficked through the secretory pathway.

Short polybasic peptide sequences are potent inhibitors of PC5/6 and PC7: Use of positional scanning-synthetic peptide combinatorial libraries as a tool for the optimization of inhibitory sequences

Positional scanning-synthetic peptide combinatorial libraries (PS-SPCLs) are powerful molecular tools to identify enzyme substrate and potent inhibitory sequences and also to provide crucial information about active site determinants. PS-SPCLs have been surveyed for furin, proprotein convertase (PC)2, PC1/3, and PACE4 and proven efficient to identify potent peptidyl inhibitors in the low nanomolar range for furin and PC1/3. We report herein the screenings of nonamidated and acetylated hexapeptide PS-SPCLs for PC5/6A and PC7. The L-configuration library surveys distinctively revealed that L-Arg, L-Lys, and sometimes L-His in all six positions would generate the most potent inhibitors for both enzymes. Based on this clear polybasic preference, L-poly-Arg peptides ranging from four to nine residues were assayed. Inhibitory potency of these polybasic peptides increased with chain length, making nona-l-Arg a potent nanomolar inhibitor of PC5/6A and PC7 (Ki of 150 and 120 nM). PC5/6 and PC7 inhibition by nona-l-arginine was equivalent to that of furin (Ki of 114 nM) (J Biol Chem 275: 36741-36749, 2000). Nona-d-arginine was a more potent inhibitor of PC5/6 and PC7 than its levorotatory version (Ki of 19 and 81 nM), reminiscent of furin (Ki of 1.3 nM) (J Biol Chem 279:36788-36794, 2004). Our data indicate that certain poly-arginine peptides represent potent inhibitors targeting PCs of the constitutive secretory pathway (furin, PC5/6, and PC7). We conclude that basic residues within PC peptide inhibitors might be responsible for targeting PCs in general and for inhibitory potency, but that select amino acid changes will be necessary to acquire true specificity toward a single PC.

Template-assisted rational design of peptide inhibitors of furin using the lysine fragment of the mung bean trypsin inhibitor

Highly active, small‐molecule furin inhibitors are attractive drug candidates to fend off bacterial exotoxins and viral infection. Based on the 22‐residue, active Lys fragment of the mung bean trypsin inhibitor, a series of furin inhibitors were designed and synthesized, and their inhibitory activity towards furin and kexin was evaluated using enzyme kinetic analysis. The most potent inhibitor, containing 16 amino acid residues with a K_i value of 2.45 × 10⁻⁹  m for furin and of 5.60 × 10⁻⁷  m for kexin, was designed with three incremental approaches. First, two nonessential Cys residues in the Lys fragment were deleted via a Cys‐to‐Ser mutation to minimize peptide misfolding. Second, residues in the reactive site of the inhibitor were replaced by the consensus substrate recognition sequence of furin, namely, Arg at P₁, Lys at P₂, Arg at P₄ and Arg at P₆. In addition, the P₇ residue Asp was substituted with Ala to avoid possible electrostatic interference with furin inhibition. Finally, the extra N‐terminal and C‐terminal residues beyond the doubly conjugated disulfide loops were further truncated. However, all resultant synthetic peptides were found to be temporary inhibitors of furin and kexin during a prolonged incubation, with the scissile peptide bond between P₁ and P₁′ being cleaved to different extents by the enzymes. To enhance proteolytic resistance, the P₁′ residue Ser was mutated to d‐Ser or N‐methyl‐Ser. The N‐methyl‐Ser mutant gave rise to a K_i value of 4.70 × 10⁻⁸  m for furin, and retained over 80% inhibitory activity even after a 3 h incubation with the enzyme. By contrast, the d‐Ser mutant was resistant to cleavage, although its inhibitory activity against furin drastically decreased. Our findings identify a useful template for the design of potent, specific and stable peptide inhibitors of furin, shedding light on the molecular determinants that dictate the inhibition of furin and kexin.

The role of prohormone convertase-2 in hypothalamic neuropeptide processing: a quantitative neuropeptidomic study

Prohormone convertase (PC) 1/3 and 2 are involved in the generation of neuropeptides from their precursors. A quantitative peptidomic approach was used to explore the role PC2 plays in the processing of hypothalamic peptides. In this approach, extracts from mice lacking PC2 activity and from wild-type littermates were labeled with isotopic tags, combined, fractionated on a reverse phase HPLC column, and analyzed by electrospray ionization mass spectrometry. Altogether, 53 neuropeptides or other peptides derived from secretory pathway proteins were identified and sequenced using tandem mass spectrometry. These peptides arise from 21 distinct proteins: proenkephalin, proopiomelanocortin, prodynorphin, protachykinin A and B, procholecystokinin, promelanin-concentrating hormone, proneurotensin, proneuropeptide Y, provasopressin, pronociceptin/orphanin, prothyrotropin-releasing hormone, cocaine- and amphetamine-regulated transcript, chromogranin A and B, secretogranin II, prohormone convertase 1 and 2, propeptidyl-amidating monooxygenase, and proteins designated proSAAS and VGF. Approximately one third of the peptides found in wild-type mice were not detectable in PC2 knock-out mice, and another third were present at levels ranging from 25 to 75% of wild-type levels. Comparison of the cleavage sites suggests that sequences with a Trp, Tyr and/or Pro in the P1' or P2' position, or a basic residue in the P3 position, are preferentially cleaved by PC2 and not by other enzymes present in the secretory pathway.

Prediction of neuropeptide prohormone cleavages with application to RFamides

Genomic information is becoming available for an ever-wider range of animals with the genes for several well-characterized peptide families, such as the RFamides, detected in a surprisingly diverse set of these animals. While bioinformatic tools allow the prediction of the RFamide-related prohormones from genetic information, it is more difficult to accurately predict the final processed peptides because of the large number of processing steps required to convert a prohormone into mature bioactive peptides. Several statistical-based methods for predicting basic site cleavages in prohormones are described, and their ability to predict the basic site cleavages in a variety of RFamide-related peptides from vertebrates and invertebrates is reported. Specifically, the cleavages in the invertebrate FMRFamides, and the vertebrate NPFFa, RFRPa, and PrRPa peptide families are modeled. The three models compared here are based on known cleavage motifs, a logistic regression, and artificial neural networks. Improvements in the accuracy and precision of the cleavage estimates will lead to increased utilization of these models for predicting bioactive neuropeptides before experimental verification is available.

Cleavage targets and the D-arginine-based inhibitors of the West Nile virus NS3 processing proteinase

Mosquito-borne WNV (West Nile virus) is an emerging global threat. The NS3 proteinase, which is essential for the proteolytic processing of the viral polyprotein precursor, is a promising drug target. We have isolated and biochemically characterized the recombinant, highly active NS3 proteinase. We have determined that the NS3 proteinase functions in a manner that is distantly similar to furin in cleaving the peptide and protein substrates. We determined that aprotinin and D-arginine-based 9-12-mer peptides are potent inhibitors of WNV NS3 with K(i) values of 26 nM and 1 nM respectively. Consistent with the essential role of NS3 activity in the life cycle of WNV and with the sensitivity of NS3 activity to the D-arginine-based peptides, we showed that nona-D-Arg-NH2 reduced WNV infection in primary neurons. We have also shown that myelin basic protein, a deficiency of which is linked to neurological abnormalities of the brain, is sensitive to NS3 proteolysis in vitro and therefore this protein represents a convenient test substrate for the studies of NS3. A three-dimensional model of WNV NS3 that we created may provide a structural guidance and a rationale for the subsequent design of fine-tuned inhibitors. Overall, our findings represent a foundation for in-depth mechanistic and structural studies as well as for the design of novel and efficient inhibitors of WNV NS3.

Characterization of the molecular basis of the Drosophila mutations in carboxypeptidase D. Effect on enzyme activity and expression

Carboxypeptidase D (CPD) functions in the processing of proteins and peptides in the secretory pathway. Drosophila CPD is encoded by the silver gene (svr), which is differentially spliced to produce long transmembrane protein forms with three metallocarboxypeptidase (CP)-like domains and short soluble forms with a single CP domain. Many svr mutants have been reported, but the precise molecular defects have not been previously determined. In the present study, three mutant lines were characterized. svr (PG33) mutants do not survive past the early larval stage. These mutants have a P-element insertion within exon 1B upstream of the initiation ATG, which greatly reduces mRNA levels of all forms of CPD. Both svr (1) and svr (poi) mutants are viable, with a silvery body color and pointed wings. The wing shape is generally similar between these two mutants, although svr (poi) mutants have smaller wings. The svr (1) gene has a three-nucleotide deletion in exon 6, removing a leucine in a region of the protein predicted to function as a folding domain for the second CP-like domain. svr (poi) has a 1072-bp duplication of the gene that introduces a stop codon into the open reading frame, causing the truncation of the protein in the middle of the second CP-like domain. Both deletions eliminate enzyme activity of the second CP-like domain and appear to cause the misfolding of the protein. This greatly reduces the levels of the long forms of CPD protein but do not affect the levels of the short forms. Taken together, these findings suggest that lethal and viable svr alleles differ in which protein forms are affected. Flies that retain the short form are viable, whereas flies that are missing all forms of CPD do not survive past the early larval stages.

Cleavage specificity of the serine protease ofAeromonas sobria, a member of the kexin family of subtilases

Subtilisin-like proteases have been grouped into six families based on a sequence of the catalytic domain. One of the six is the kexin family, of which furin is a representative protease. All members of the kexin family, except one, are from eukaryotes. The one prokaryotic protease is a serine protease of Aeromonas sorbria (ASP). Here, we examined the substrate specificity of ASP based on the cleavage of short peptides. The results showed that ASP preferentially cleaves the peptide bond following two basic residues, one of which is Lys, but not the bond following a single basic residue. This indicates that the tertiary structure around the catalytic domain of ASP resembles, but is not identical to that of furin. Prekallikrein was cleaved into four fragments by ASP, indicating that the protein must be cleaved at specific sequences.

Cutting back on pro-protein convertases: the latest approaches to pharmacological inhibition

The secretory pathway in cells possesses an elaborate set of endoproteolytic enzymes that carry out a crucial step in protein precursor maturation. This step is proteolytic activation by cleavage at specific pairs of basic residues. These enzymes, named pro-protein convertases (PCs), are responsible for generating bioactive peptides and activating several enzymes and growth factors that are implicated in many important physiological events. PCs have roles in several pathologies including viral infections and cancers and, thus, are promising targets for therapeutic applications. Recent structural and homology-modeling studies demonstrate more similarity than expected at the catalytic site of the seven PCs, which makes the development of selective drugs to target individual PCs frustrating. Based on this information, we review the latest strategies to inhibit PCs, which might lead to the development of specific compounds.

Mutations of the PC2 substrate binding pocket alter enzyme specificity

By taking advantage of the recently published furin structure, whose catalytic domain shares high homology with other proprotein convertases, we designed mutations in the catalytic domain of PC2, altering residues Ser206, Thr271, Asp278, ArgGlu282, AlaSer323, Leu341, Asn365, and Ser380, which are both conserved and specific to this convertase, and substituting residues specific to PC1 and/or furin. In order to investigate the determinants of PC2 specificity, we have tested the mutated enzymes against a set of proenkephalin-derived substrates, as well as substrates representing Arg, Ala, Leu, Phe, and Glu positional scanning variants of a peptide B-derived substrate. We found that the exchange of the Ser206 residue with Arg or Lys led to a total loss of activity. Increased positive charge of the substrate generally resulted in an increased specificity constant. Most intriguingly, the RE281GR mutation, corresponding to a residue placed distantly in the S6 pocket, evoked the largest changes in the specificity pattern. The D278E and N356S mutations resulted in distinct alterations in PC2 substrate preferences. However, when other residues that distinguish PC2 from other convertases were substituted with PC1-like or furin-like equivalents, there was no significant alteration of the PC2 specificity pattern, suggesting that the overall structure of the substrate binding cleft rather than individual residues specifies substrate binding.