Search

Through a posttranslational processing mechanism, pro-growth hormone releasing hormone (proGHRH) gives rise to an amidated GHRH molecule, which in turn stimulates the synthesis and release of growth hormone. We have previously proposed a model for the biochemical processing of proGHRH [Nillni et al. (1999), Endocrinology 140, 5817-5827]. We demonstrated that the proGHRH peptide (10.5 kDa, 104 aa) is first processed to an 8.8 kDa intermediate form that is later cleaved to yield two products: the 5.2 kDa GHRH and the 3.6 kDa GHRH-RP. However, the proteolytic enzymes involved in this process are unknown. Therefore, in this study we determined which proconverting enzymes are involved in this process. We transfected different constructs in cell lines carrying different PC enzymes followed by analysis of the peptide products after metabolic labeling or Western blots. We found that in the absence of furin (LoVo cells) or CHO cells treated with BFA, only one moiety was observed, and that corresponds to the same electrophorectic mobility to the GHRH precursor. This finding strongly supports an initial role for furin in the processing of proGHRH. The results from transfections with preproGHRH alone or double or triple transfections with PC1 and PC2 in AtT-20, GH3, and GH4C1 cells indicated that PC1 is the primary enzyme involved in the generation of GHRH peptide from the 8.8 kDa intermediate form. We found that AtT-20 cells (high PC1, very low PC2) were able to generate GHRH. However, GH3 cells (high PC2, but not PC1) were able to process the 8.8 kDa peptide to GHRH only after the cotransfection with the PC1 enzyme. Transfections with preproGHRH-GFP and preproGHRH-V5 provided similar results in all the cell lines analyzed. These data support the hypothesis that proGHRH is initially cleave by furin at preproGHRH29-30, followed by a second cleavage at preproGHRH74 primarily by PC1 to generate GHRH and GHRH-RP peptides, respectively., (Copyright 2004 Humana Press Inc.)

A coordinated transcriptional response to DNA-damaging agents is required to maintain genome stability. We have examined the global gene expression responses of the fission yeast Schizosaccharomyces pombe to ionizing radiation (IR) by using DNA microarrays. We identified approximately 200 genes whose transcript levels were significantly altered at least twofold in response to 500 Gy of gamma IR in a temporally defined manner. The majority of induced genes were core environmental stress response genes, whereas the remaining genes define a transcriptional response to DNA damage in fission yeast. Surprisingly, few DNA repair and checkpoint genes were transcriptionally modulated in response to IR. We define a role for the stress-activated mitogen-activated protein kinase Sty1/Spc1 and the DNA damage checkpoint kinase Rad3 in regulating core environmental stress response genes and IR-specific response genes, both independently and in concert. These findings suggest a complex network of regulatory pathways coordinate gene expression responses to IR in eukaryotes.

Pro-hormone or pro-protein convertases are a conserved family of eukaryotic serine proteases found in the secretory pathway. These endoproteases mature precursors for peptides and proteins that perform a wide range of physiologically important and clinically relevant functions. The first member of this family to be identified was Kex2 in the yeast Saccharomyces cerevisiae. One mammalian member of this family - furin - is responsible for processing substrates that include insulin pro-receptor, human immunodeficiency virus gp160 glycoprotein, Ebola virus glycoprotein, and anthrax protective antigen. Recent determination of the crystal structures for the catalytic core domains of both Kex2 and furin - the first for any members of this family - provide remarkable insights and a new level of understanding of substrate specificity and catalysis by the pro-protein convertases.

Eglin c with mutants L45R and D42R at the P(1) and P(4) positions has been reported to become a stable inhibitor toward the proprotein convertases (PC), furin and kexin, with a K(i) of 2.3x10(-8) and 1.3x10(-10) M, respectively. The mutant was further engineered at the P(2)'-P(4)' positions to create a more potent and selective inhibitor for each enzyme. The residue Asp at P(1)' which is crucial for stabilizing the conformation of eglin c remained unchanged. The eglin c mutants cloned into the vector pGEX-2T and expressed in Escherichia coli (DH5alpha) were purified to homogeneity, and their inhibitory activities toward the purified recombinant furin and kexin were examined. The results showed that (1) Leu47 at P(2)' replaced with either a positively or negatively charged residue resulted in a decrease in inhibitory activities to both enzymes; (2) the replacement of Arg with Asp at P(3)' was favorable for inhibiting furin with a K(i) of 7.8 x 10(-9) M, but not for inhibiting kexin; (3) the replacement of Tyr with Glu at P(4)' increased the inhibitory activity to kexin with a K(i) of 3 x 10(-11) M, but was almost without any influence on furin inhibition. It was indicated that the inhibitory specificity of eglin c could be changed from inhibiting elastase to inhibiting PCs by site-directed mutation at the P positions, while the inhibitory selectivity to furin or kexin could be optimized by mutation at the P' positions.

A novel gene delivery system based on the enhanced release of plasmid DNA from polyplexes in response to intracellular proprotein convertase (PC), furin, that cleaves the carboxyl side of Arg-X-Arg/Lys-Arg, was successfully established. Polypeptides that have this recognition sequence were synthesized by the Fmoc solid phase method and evaluated as gene vectors. The fragmented polypeptides were found to lose the abilities to form polyplexes with plasmid DNA. When the Fur-polypeptide/plasmid DNA polyplexes were treated with furin in vitro, their disassembly and the release of plasmid DNA was observed in agarose gel electrophoresis analysis. The Fur-polypeptide showed higher transgene expression in COS-1 cells, while the control polypeptides which do not have the recognition sites did not. These results indicate that the cationic nature of the polypeptides is not sufficient for gene carrier and this intracellular signal-responsive gene transfer system is the useful strategy to enhance the transcription efficiency of the delivered plasmid DNA.

The mammalian convertase furin plays a significant role in tumorigenesis and its overexpression was observed in a number of different cancer types. To date, however, few mechanisms of action have been described. Most of the information available concerns the invasion step and designates MT1-MMP, through the activation of MMP-2, as the bona fide substrate mediating furin activity. However, recent reports indicate furin-independent pathways for MT1-MMP activation. To gain further insights into the role of furin in the invasion process, we studied the in vitro invasive capacity of LoVo cells, a furin-deficient adenocarcinoma cell line transfected with wild-type furin. Furin complementation resulted in an increased cell invasiveness that correlated with their capacity to produce MMP-2. Chemical blockage of MMPs activity with BB-3103 or MMP-2-specific antibodies revealed that the increased invasive capacity of furin-complemented cells was mediated by MMP-2. Unexpectedly, furin complementation did not change the status of MT1-MMP expression or activation, but instead resulted in the production of mature and bioactive TGFbeta1. Western blot-analysis of TGFbeta1 fragmentation species indicated that TGFbeta maturation step required furin activity, whereas results from TGFbeta-inducible reporter assays in the presence of MMP inhibitors or exogenous MMP-2 suggested that the activation step was under MMP influence. In addition, blockage with TGFbeta neutralizing antibodies revealed that furin-induced invasiveness was mediated by endogenous production of TGFbeta. Taken together, our findings established the existence of a novel alternative/complementary pathway by which furin increases tumor cell invasion through an amplification/activation loop between MMP-2 and TGFbeta.

We have characterized growth and protein processing characteristics of Aspergillus niger strains carrying a disrupted allele of the previously cloned and characterized kexB gene [Appl. Environ. Microbiol. 66 (2000) 363] encoding a furin-type endoprotease. Deletion of the single-copy gene confirms it to be non-essential but disruptant strains exhibit a morphologically distinct phenotype characterized by hyperbranching. Processing of homologous pro-proteins and fusion proteins comprised of a heterologous protein fused down-stream of glucoamylase and separated at the fusion junction by an endoproteolytic cleavage site was compared in wildtype and mutant strains of A. niger. We show that maturation of the native glucoamylase requires KexB, whereas maturation of aspergillopepsin does not. The processing of fusion proteins carrying Lys-Arg requires KexB, although alternative endoproteases are capable of cleaving protein fusions at sites adjacent to Lys-Arg.

Cell engineering or gene therapy may represent an alternative to current methods of treating diabetes mellitus. Cells could be engineered to secrete insulin ex vivo for transplantation or the insulin gene could be administered directly by injection into muscle. A problem has been that non-neuroendocrine cells lack the endoproteases (PC3/1 and PC2) that are responsible for the processing of proinsulin to insulin. This can be surmounted by engineering the paired basic amino acid processing sites within proinsulin to sites that would be recognized by the ubiquitously expressed protease, furin. However, in every study to date, the expression of the furin-cleavable construct was greatly reduced relative to that of the unmodified proinsulin construct. We investigated possible causes for this, including mRNA stability, the presence of additional CpG islands, and the amino acid substitutions within furin-cleavable proinsulin. Several furin-cleavable rat proinsulin I cDNAs were engineered and used to transfect human HEK293, rat L6 and mouse C(2)C(12) cell lines. The stability of wild-type and furin-cleavable proinsulin mRNA in transfected C(2)C(12) cells was measured by RT-PCR. Comparison of the decay rates in the presence of actinomycin D showed no significant difference between the two species of mRNA. A furin-cleavable proinsulin cDNA was created to contain the same distribution of CpG islands as wild-type proinsulin. Comparison of insulin-like immunoreactivity in all three cell lines transfected with either this construct or a widely used furin-cleavable proinsulin containing additional CpG islands showed that the presence of the extra CpG islands had no effect. Studies to examine amino acid substitutions used to create furin consensus sequences showed that the addition of basic residues at the C-peptide/A-chain junction was responsible for the reduced production of furin-cleavable proinsulin. Using this information, we engineered a cDNA for furin-cleavable rat proinsulin I that was efficiently processed to mature insulin and expressed at the same level as wild-type proinsulin.

Tachykinins, an evolutionary conserved family of peptide hormones in both invertebrates and vertebrates, are produced by neuronal cells as inactive preprotachykinins that are post-translationally processed into different neuropeptides such as substance P, neurokinin A, and neurokinin B. We show here that furin-mediated cleavage of the bovine respiratory syncytial virus fusion protein results in the release of a peptide that is converted into a biologically active tachykinin (virokinin) by additional post-translational modifications. An antibody directed to substance P cross-reacted with the C terminus of mature virokinin that contains a classical tachykinin motif. The cellular enzymes involved in the C-terminal maturation of virokinin were found to be present in many established cell lines. Virokinin is secreted by virus-infected cells and was found to act on the tachykinin receptor 1 (TACR1), leading to rapid desensitization of this G protein-coupled receptor as shown by TACR1-green fluorescent protein conjugate translocation from the cell surface to endosomes and by co-internalization of the receptor with beta-arrestin 1-green fluorescent protein conjugates. In vitro experiments with isolated circular muscle from guinea pig stomach indicated that virokinin is capable of inducing smooth muscle contraction by acting on the tachykinin receptor 3. Tachykinins and their cognate receptors are present in the mammalian respiratory tract, where they have potent effects on local inflammatory and immune processes. The viral tachykinin-like peptide represents a novel form of molecular mimicry, which may benefit the virus by affecting the host immune response.

ADAMTS13 belongs to the "a disintegrin and metalloprotease with thrombospondin repeats" family, and cleaves von Willebrand factor multimers into smaller forms. For several related proteases, normal folding and enzymatic latency depend on an NH2-terminal propeptide that is removed by proteolytic processing during biosynthesis. However, the ADAMTS13 propeptide is unusually short and poorly conserved, suggesting it may not perform these functions. ADAMTS13 was secreted from transfected HeLa cells with a half-time of 7 h and the rate-limiting step was exported from the endoplasmic reticulum. Deletion of the propeptide did not impair the secretion of active ADAMTS13, indicating that the propeptide is dispensable for folding. Furin was shown to be sufficient for ADAMTS13 propeptide processing in two ways. First, mutation of the furin consensus recognition site prevented propeptide cleavage in HeLa cells and resulted in secretion of pro-ADAMTS13. Second, furin-deficient LoVo cells secreted ADAMTS13 with the propeptide intact, and cotransfection with furin restored propeptide cleavage. In both cell lines, secreted pro-ADAMTS13 had normal proteolytic activity toward von Willebrand factor. In cells coexpressing both ADAMTS13 and von Willebrand factor, pro-ADAMTS13 cleaved pro-von Willebrand factor intracellularly. Therefore, the ADAMTS13 propeptide is not required for folding or secretion, and does not perform the common function of maintaining enzyme latency.

Proprotein convertases (PCs) have been proposed to play a role in tumor necrosis factor-alpha converting enzyme (TACE) processing/activation. Using the furin-deficient LoVo cells, as well as the furin-proficient synoviocytes and HT1080 cells expressing the furin inhibitor alpha(1)-PDX, we demonstrate that furin activity alone is not sufficient for effective maturation and activation of the TACE enzyme. Data from in vitro and in vivo cleavage assays indicate that PACE-4, PC5/PC6, PC1 and PC2 can directly cleave the TACE protein and/or peptide. PC inhibition in macrophages reduced the release of soluble TNF-alpha from transmembrane pro-TNF-alpha. We therefore conclude that furin, in addition to other candidate PCs, is involved in TACE maturation and activation.

Mutation of aspartic acid 187 to asparagine (D187N) or tyrosine (D187Y) in domain 2 of the actin-modulating protein gelsolin causes the neurodegenerative disease familial amyloidosis of Finnish type (FAF). These mutations render plasma gelsolin susceptible to aberrant proteolysis by furin in the trans-Golgi network, the initial proteolytic event in the formation of 71 and 53 residue fragments that assemble into amyloid fibrils. Ca(2+) binding stabilizes wild-type domain 2 gelsolin against denaturation and proteolysis, but the FAF variants are unable to bind and be stabilized by Ca(2+). Though the chain of events initiating FAF has been elucidated recently, uncertainty remains about the mechanistic details that allow the FAF variants to be processed. To test the hypothesis that impaired Ca(2+) binding in the D187 variants, but not other factors specific to residue 187, increases susceptibility to aberrant proteolysis and subsequent amyloidogenesis, we designed the gelsolin variant E209Q to remove a different Ca(2+) ligand from the same Ca(2+) site that is affected in the FAF variants. Here, we show that E209Q domain 2 does not bind Ca(2+) and is not stabilized against denaturation or furin proteolysis, analogous to the behavior exhibited by the FAF variants. Transfection of full-length E209Q into COS cells results in secretion of both the full-length and furin-processed fragments, as observed with D187N and D187Y. Mutation of the furin consensus sequence in D187N and E209Q gelsolin prevents cleavage during secretion, indicating that inhibition of proprotein convertases (furin) represents a viable therapeutic approach for the treatment of FAF. Mutations that diminish domain 2 Ca(2+) binding allow furin access to an otherwise protected cleavage site, initiating the proteolytic cascade that leads to gelsolin amyloidogenesis and FAF.

Flaviviruses assemble in the endoplasmic reticulum by a mechanism that appears to be driven by lateral interactions between heterodimers of the envelope glycoproteins E and prM. Immature intracellular virus particles are then transported through the secretory pathway and converted to their mature form by cleavage of the prM protein by the cellular protease furin. Earlier studies showed that when the prM and E proteins of tick-borne encephalitis virus are expressed together in mammalian cells, they assemble into membrane-containing, icosahedrally symmetrical recombinant subviral particles (RSPs), which are smaller than whole virions but retain functional properties and undergo cleavage maturation, yielding a mature form in which the E proteins are arranged in a regular T = 1 icosahedral lattice. In this study, we generated immature subviral particles by mutation of the furin recognition site in prM. The mutation resulted in the secretion of two distinct size classes of particles that could be separated by sucrose gradient centrifugation. Electron microscopy showed that the smaller particles were approximately the same size as the previously described mature RSPs, whereas the larger particles were approximately the same size as the virus. Particles of the larger size class were also detected with a wild-type construct that allowed prM cleavage, although in this case the smaller size class was far more prevalent. Subtle differences in endoglycosidase sensitivity patterns suggested that, in contrast to the small particles, the E glycoproteins in the large subviral particles and whole virions might be in nonequivalent structural environments during intracellular transport, with a portion of them inaccessible to cellular glycan processing enzymes. These proteins thus appear to have the intrinsic ability to form alternative assembly products that could provide important clues about the role of lateral envelope protein interactions in flavivirus assembly.

Fibrillin-1 is synthesized as a proprotein that undergoes proteolytic processing in the unique C-terminal domain by a member of the PACE/furin family of endoproteases. This family of endoproteases is active in the trans-Golgi network (TGN), but metabolic labeling studies have been controversial as to whether profibrillin-1 is processed intracellularly or after secretion. This report provides evidence that profibrillin-1 processing is not an intracellular event. Bafilomycin A(1) and incubation of dermal fibroblasts at 22 degrees C were used to block secretion in the TGN to confirm that profibrillin-1 processing did not occur in this compartment. Profibrillin-1 immunoprecipitation studies revealed that two endoplasmic reticulum-resident molecular chaperones, BiP and GRP94, interacted with profibrillin-1. To determine the proprotein convertase responsible for processing profibrillin-1, a specific inhibitor of furin, alpha-1-antitrypsin, Portland variant, was both expressed in the cells and added to cells exogenously. In both cases, the inhibitor blocked the processing of profibrillin-1, providing evidence that furin is the enzyme responsible for profibrillin-1 processing. These studies delineate the secretion and proteolytic processing of profibrillin-1, and identify the proteins that interact with profibrillin-1 in the secretory pathway., (Copyright 2003 Wiley-Liss, Inc.)

Familial British dementia (FBD) is an autosomal dominant condition caused by a point mutation in the stop codon of the BRI gene. This mutation extends the normal precursor protein (PP) of 266 amino acids to the next stop codon, which is at amino acid 277. Kim and colleagues demonstrated in vitro that furin can process both the normal protein BriPP and the extended protein ABriPP to produce C-terminal fragments of 23 and 34 amino acids. The abnormal C-terminal fragment, ABri, accumulates in FBD in the form of extracellular amyloid deposits. The objective of our study was to determine if furin is associated with ABri in FBD. Brain tissue of one case of FBD, four cases of Alzheimer's disease (AD) and two controls were studied by immunohistochemistry using antibodies against furin, beta-amyloid protein and ABri. In FBD, furin was found to be colocalized with ABri deposits and amyloid angiopathy in all areas examined. In contrast beta-amyloid deposits in AD were not immunostained by the furin antibody. In normal as well as pathological cases, clusters of neurons in the hippocampus and neocortex showed light to moderate furin immunostaining, while peptidergic neurons of the hypothalamus showed intense furin-immunostaining. These data suggest that furin may be involved in producing the pathological fragment of ABriPP in vivo and that inhibition of furin might be a method of treating this disorder.

Matrix metalloproteinases (MMPs) and integrins are essential for cell and extracellular matrix homeostasis. Both membrane type-1 MMP (MT1-MMP) and the integrin alphaV subunit are fully activated upon cleavage at a furin recognition site. Furin is shuttled to the cell surface through the trans-Golgi network and endosomal system, and its only known role on plasma membrane consists in activation of opportunistic pathogenic entities. Here, we report findings about the interaction of furin with MT1-MMP and the integrin alphaV at the cell surface. By using in vivo gene delivery, western blotting and immunogold electron microscopy, we provide evidence of significant pools of furin and proMT1-MMP along the surface of cells lining basement membranes. Moreover, furin and integrin alphaV are frequently found associated with the slit diaphragm of renal podocytes and around endothelial fenestrations. ProMT1-MMP, by contrast, is concentrated at the slit diaphragm. Coimmunoprecipitations and double immunogold labelings indicate that furin interacts with proMT1-MMP and alphaV at points of insertion of the slit diaphragm. Our results suggest that these focalized complexes could trigger basement membrane proteolysis either directly by activation of proMT1-MMP or indirectly by promoting activation of proMMP2.

Integrin alphav is involved in intracellular-extracellular signaling important for cytoskeleton alterations and control of cell movement. In vitro experiments indicate that the integrin alphav-subunit undergoes post-translational endoproteolytic cleavage. This type of activation requires the presence of suitable kexin/subtilisin-like proprotein convertases. In vitro experiments have demonstrated that, among several proprotein convertases, PC5A, and to a threefold lesser extent furin, can activate alphav integrin. The biological significance of these in vitro data would be further supported by a coexpression and coordinated regulation of the gene expression of alphav integrin and its activating enzyme PC5 in vivo. In the present study we investigated the regulation of alphav integrin and PC5 following balloon injury in vivo. Comparative immunocytochemistry revealed a coordinated regulation of alphav integrin and PC5 during vascular remodeling in rodents. Integrin alphav was found to be upregulated in PCNA-positive, proliferating vascular smooth muscle cells. Northern blots revealed no significant regulation of furin mRNA, whereas PC5A mRNA increased during vascular remodeling, suggesting that PC5 is the major convertase during neointima formation in vivo. Incubation of vascular smooth muscle cells with the Golgi-disturbing agent brefeldin A inhibited alphav integrin maturation, indicating that endoproteolytic cleavage occurs in the trans-Golgi network, were PC5 is localized. Thus, the present study further supports the concept that activation of alphav integrin occurs in the trans-Golgi network in vascular smooth muscle cells and involves PC5.

To examine the cleavage specificity of different members of the furin/propeptide convertase (PC) family of enzymes, we have selected proglucagon (PG) as a model substrate. PG was selected because it is subject to differential processing in vivo. PG is thought to be cleaved initially at an interdomain site to produce glicentin and the major proglucagon fragment (MPGF). These intermediates are subsequently cleaved, most likely by the convertases PC2 and PC1, respectively. To determine the exact sites within PG that are cleaved by PC1 and PC2, we attempted to produce milligram quantities of human PG, glicentin, and MPGF for use in an in vitro conversion assay. A methionine residue was added to the N-terminus of each protein to initiate translation. Purification was achieved using cation exchange and reversed-phase chromatography, and the integrity of the methionylated proteins was confirmed by both electrospray ionization-mass spectrometry and amino acid analysis. The combined expression and purification scheme is fast, efficient, and results in milligram quantities of > or =95% pure proglucagon, > or =95% pure MPGF, and > or =93% pure glicentin. These prohormones are cleaved by PC1 to produce product peptides consistent with the processing of PG observed in vivo, and should therefore be suitable for further analysis of the post-translational processing of PG.

Historically, the most effective means of modifying cell surface carbohydrates has required the intracellular overexpression of glycosyltransferases or glycosidases and is dependent on the enzymes occupying a cellular localization close to the carbohydrate structures they modify. We report on relocalizing the lysosomal resident glycosidase human alpha-galactosidase to other regions of the cell, Golgi and cell surface, where it is in closer proximity for cleaving the carbohydrate structure Galalpha(1,3)Gal. Relocalization of alpha-galactosidase was achieved by using the transmembrane and cytoplasmic domains from the human protein furin, which is known to localize in the trans-Golgi network (TGN) and cell surface. Two chimeric forms of alpha-galactosidase were generated, one directing it to the TGN of the cell and the other to the cell surface, as shown by confocal microscopy. The relocalized enzymes have the ability to cleave terminal alpha-galactose as detected by expression on the cell surface. Furthermore, when expressed as a transgene in mice, the TGN form of alpha-galactosidase was more effective at decreasing cell surface terminal alpha-galactose than was the native lysosomal form. When expressed in conjunction with the alpha1,2fucosyltransferase that also decreases Galalpha(1,3)Gal, the reduction was additive. The ability to relocalize enzymes that modify cell surface carbohydrate structures has far-reaching implications in biology and may be useful in such fields as xenotransplantation and treatment of glycosidase disorders.