Mutations within a furin consensus sequence block proteolytic release of ectodysplasin-A and cause X-linked hypohidrotic ectodermal dysplasia.

X-linked hypohidrotic ectodermal dysplasia (XLHED) is a heritable disorder of the ED-1 gene disrupting the morphogenesis of ectodermal structures. The ED-1 gene product, ectodysplasin-A (EDA), is a tumor necrosis factor (TNF) family member and is synthesized as a membrane-anchored precursor protein with the TNF core motif located in the C-terminal domain. The stalk region of EDA contains the sequence -Arg-Val-Arg-Arg156-Asn-Lys-Arg159-, representing overlapping consensus cleavage sites (Arg-X-Lys/Arg-Arg( downward arrow)) for the proprotein convertase furin. Missense mutations in four of the five basic residues within this sequence account for approximately 20% of all known XLHED cases, with mutations occurring most frequently at Arg156, which is shared by the two consensus furin sites. These analyses suggest that cleavage at the furin site(s) in the stalk region is required for the EDA-mediated cell-to-cell signaling that regulates the morphogenesis of ectodermal appendages. Here we show that the 50-kDa EDA parent molecule is cleaved at -Arg156Asn-Lys-Arg(159 downward arrow)- to release the soluble C-terminal fragment containing the TNF core domain. This cleavage appears to be catalyzed by furin, as release of the TNF domain was blocked either by expression of the furin inhibitor alpha1-PDX or by expression of EDA in furin-deficient LoVo cells. These results demonstrate that mutation of a functional furin cleavage site in a developmental signaling molecule is a basis for human disease (XLHED) and raise the possibility that furin cleavage may regulate the ability of EDA to act as a juxtacrine or paracrine factor.

The PC6B cytoplasmic domain contains two acidic clusters that direct sorting to distinct trans-Golgi network/endosomal compartments.

The mammalian proprotein convertases (PCs) are a family of secretory pathway enzymes that catalyze the endoproteolytic maturation of peptide hormones and many bioactive proteins. Two PCs, furin and PC6B, are broadly expressed and share very similar cleavage site specificities, suggesting that they may be functionally redundant. However, germline knockout studies show that they are not. Here we report the distinct subcellular localization of PC6B and identify the sorting information within its cytoplasmic domain (cd). We show that in neuroendocrine cells, PC6B is localized to a paranuclear, brefeldin A-dispersible, BaCl(2)-responsive post-Golgi network (TGN) compartment distinct from furin and TGN38. The 88-amino acid PC6B-cd contains sorting information sufficient to direct reporter proteins to the same compartment as full-length PC6B. Mutational analysis indicates that endocytosis is predominantly directed by a canonical tyrosine-based motif (Tyr(1802)GluLysLeu). Truncation and sufficiency studies reveal that two clusters of acidic amino acids (ACs) within the PC6B-cd contain differential sorting information. The membrane-proximal AC (AC1) directs TGN localization and interacts with the TGN sorting protein PACS-1. The membrane-distal AC (AC2) promotes a localization characteristic of the full-length PC6B-cd. Our results demonstrate that AC motifs can target proteins to distinct TGN/endosomal compartments and indicate that the AC-mediated localization of PC6B and furin contribute to their distinct roles in vivo.

A protein-based therapeutic for human cytomegalovirus infection.

Current antiviral strategies target viral gene products. Although initially successful, their severe toxicity and susceptibility to circumvention by the generation of drug-resistant variants limit their usefulness. By contrast, the central role of the host cell serine endoprotease furin in the proteolytic activation of numerous pathogens points to the endoprotease as a strategic target for therapeutics. Herein, we show that the production of infectious human cytomegalovirus is dramatically reduced by exogenous addition of a bioengineered serpin, alpha(1)-PDX. This protein is a potent and selective furin inhibitor (K(i) = 0.6 nM) and is 10-fold more effective than currently used antiherpetic agents in cell-culture models. The requirement of furin for the processing of envelope glycoproteins from many pathogenic viruses and for the activation of several bacterial toxins suggests that selective inhibitors of furin have potential as broad-based anti-pathogens.

Bi-cycling the furin pathway: from TGN localization to pathogen activation and embryogenesis.

Furin is a secretory pathway endoprotease that catalyses the maturation of a strikingly diverse group of proprotein substrates, ranging from growth factors and receptors to pathogen proteins, in multiple compartments within the trans-Golgi network (TGN)/endosomal system. This review focuses on recent developments in the biochemistry and cell biology of the endoprotease, including the mechanism of TGN localization, phosphorylation-dependent regulation of protein traffic, and novel insights into early embryogenesis, extracellular matrix formation and pathogen virulence.

Regulation of endosome sorting by a specific PP2A isoform.

The regulated sorting of proteins within the trans-Golgi network (TGN)/endosomal system is a key determinant of their biological activity in vivo. For example, the endoprotease furin activates of a wide range of proproteins in multiple compartments within the TGN/endosomal system. Phosphorylation of its cytosolic domain by casein kinase II (CKII) promotes the localization of furin to the TGN and early endosomes whereas dephosphorylation is required for efficient transport between these compartments (Jones, B.G., L. Thomas, S.S. Molloy, C.D. Thulin, M.D. Fry, K.A. Walsh, and G. Thomas. 1995. EMBO [Eur. Mol. Biol. Organ.] J. 14:5869-5883). Here we show that phosphorylated furin molecules internalized from the cell surface are retained in a local cycling loop between early endosomes and the plasma membrane. This cycling loop requires the phosphorylation state-dependent furin-sorting protein PACS-1, and mirrors the trafficking pathway described recently for the TGN localization of furin (Wan, L., S.S. Molloy, L. Thomas, G. Liu, Y. Xiang, S.L. Ryback, and G. Thomas. 1998. Cell. 94:205-216). We also demonstrate a novel role for protein phosphatase 2A (PP2A) in regulating protein localization in the TGN/endosomal system. Using baculovirus recombinants expressing individual PP2A subunits, we show that the dephosphorylation of furin in vitro requires heterotrimeric phosphatase containing B family regulatory subunits. The importance of this PP2A isoform in directing the routing of furin from early endosomes to the TGN was established using SV-40 small t antigen as a diagnostic tool in vivo. The role of both CKII and PP2A in controlling multiple sorting steps in the TGN/endosomal system indicates that the distribution of itinerant membrane proteins may be acutely regulated via signal transduction pathways.

PACS-1 defines a novel gene family of cytosolic sorting proteins required for trans-Golgi network localization.

We report the role of one member of a novel gene family, PACS-1, in the localization of trans-Golgi network (TGN) membrane proteins. PACS-1 directs the TGN localization of furin by binding to the protease's phosphorylated cytosolic domain. Antisense studies show TGN localization of furin and mannose-6-phosphate receptor, but not TGN46, is strictly dependent on PACS-1. Analyses in vitro and in vivo show PACS-1 has properties of a coat protein and connects furin to components of the clathrin-sorting machinery. Cell-free assays indicate TGN localization of furin is directed by a PACS-1-mediated retrieval step. Together, these findings explain a mechanism by which membrane proteins in mammalian cells are localized to the TGN.

Intracellular trafficking of furin is modulated by the phosphorylation state of a casein kinase II site in its cytoplasmic tail.

Human furin catalyzes the proteolytic maturation of many proproteins in the exocytic and endocytic secretory pathways by cleavage at the C-terminal side of the consensus sequence-ArgXaaLys/ArgArg decreases -. Both the trans-Golgi network (TGN) concentration and intracellular routing of furin require sequences in its 56 amino acid cytoplasmic tail. Here, we show that the furin cytoplasmic tail contains multiple trafficking signals. Localization to the TGN requires a cluster of acidic amino acids that, together with a pair of serine residues, forms a casein kinase II (CK II) phosphorylation site. We show that CK II efficiently phosphorylates these serines in vitro, and using a permeabilized cell system we provide evidence that CK II is the in vivo furin kinase. Analysis by mass spectrometry shows that, in vivo, furin exists as di-, mono- and non-phosphorylated forms. Finally, employing (i) furin constructs that mimic either non-phosphorylated or phosphorylated furin and (ii) the phosphatase inhibitor tautomycin, we show that the phosphorylation state of the furin cytoplasmic tail modulates retrieval of the endoprotease to the TGN. Thus, routing of furin is a two-tiered process combining a set of trafficking signals comprised of the primary amino acid sequence of the tail with its phosphorylation state.

Internally quenched fluorogenic substrate for furin.

A new substrate for furin, Abz-Arg-Val-Lys-Arg-Gly-Leu-Ala-Tyr(NO2)-Asp-OH, has been synthesized and characterized. The peptide is an internally quenched fluorogenic substrate. The kinetic parameters are Km = 3.8 microM, kcat = 29.3 s-1, and kcat/KM = 7,710,000 M-1 s-1. The substrate is efficiently cleaved by furin; its kcat/KM value is over 2000-fold higher than that of the commonly used substrate Boc-Arg-Val-Arg-Arg-AMC.

Sequence specificity of furin, a proprotein-processing endoprotease, for the hemagglutinin of a virulent avian influenza virus.

The virulence of avian influenza viruses correlates with the sensitivity of their hemagglutinin (HA) to cellular proteases. Furin, a proprotein-processing subtilisin-related endoprotease, is a leading candidate for the enzyme that cleaves the HA of virulent avian viruses. We therefore compared the specificity of furin with those of proteases in a variety of cultured cells and in a rat Golgi fraction, using the HA cleavage mutants of a virulent avian influenza virus, A/Turkey/Ireland/1378/85 (H5N8). The results indicated similar sequence specificities among the endoproteases when purified furin was used. In experiments with the vaccinia virus expression system, overexpressed furin cleaved mutant HAs that were not recognized by the endogenous proteases, resulting in an apparent broader specificity of furin. These findings authenticate the proposed role of furin as an HA-activating protease in vivo and caution against the use of expression vectors to study protease sequence specificity.

Intracellular trafficking and activation of the furin proprotein convertase: localization to the TGN and recycling from the cell surface.

Furin is a membrane-associated endoprotease that efficiently cleaves precursor proteins on the C-terminal side of the consensus sequence, Arg-X-Lys/Arg-Arg1, and has been proposed to catalyze these reactions in both exocytic and endocytic compartments. To study its biosynthesis and routing, a furin construct (designated fur/f) containing the FLAG epitope tag inserted on the C-terminal side of the enzyme's autoproteolytic maturation site was used. Introduction of the epitope tag had no effect on the expression, proteolytic maturation or activity of furin. Analysis of the localization of fur/f by immunofluorescence microscopy showed that its staining pattern largely overlapped with those of several Golgi-associated markers. Treatment of cells with brefeldin A caused the fur/f distribution to collapse around the microtubule organizing center, indicating that furin is concentrated in the trans-Golgi network (TGN). Immunoelectron microscopy showed unequivocally that furin resides in the TGN where it colocalized with TGN38. In agreement with its proposed activity in multiple compartments, antibody uptake studies showed that fur/f cycles between the cell surface and TGN. Furthermore, targeting to the TGN requires sequences in the cytoplasmic tail of the enzyme. Pulse-chase and immunofluorescence analyses demonstrated that proregion removal occurs in the endoplasmic reticulum and that cleavage may be required for exist from this compartment. Finally, we show that proregion removal is necessary but not sufficient for enzyme activation.

Arg15-Lys17-Arg18 turkey ovomucoid third domain inhibits human furin.

Turkey ovomucoid third domain with Leu18 in its reactive site is a potent inhibitor of many serine proteinases: subtilisins, chymotrypsins, and elastases. Previous studies showed that an L18K mutation made it a moderately strong inhibitor of trypsin, while an L18E mutation made it a strong inhibitor of Glu-specific Streptomyces griseus proteinase (GluSGP). For human furin substrates the consensus optimal sequence is RXKR decreases. Therefore the A15R, T17K, and L18R mutations were made in turkey ovomucoid third domain. The mutant inhibits human furin with a Ka of 1.1 x 10(7) M-1. As human furin catalyzes an obligatory step in human immunodeficiency virus proliferation, this inhibitor, along with the others already available, deserves further study.

Autophosphorylation of type II CaM kinase in hippocampal neurons: localization of phospho- and dephosphokinase with complementary phosphorylation site-specific antibodies.

We have visualized the distribution of autophosphorylated type II CaM kinase in neural tissue with the use of two complementary antibodies: a monoclonal antibody that binds to the alpha and beta subunits of the kinase only when they are autophosphorylated at threonine-286 (287 in beta) and affinity-purified rabbit antibodies that bind to both subunits only when they are not phosphorylated at these residues. We used these antibodies to double-label organotypic hippocampal cultures, detecting the mouse monoclonal antibody with rhodamine and the rabbit polyclonal antibodies with fluorescein. In double-exposed photographs, the ratios of intensities of the two fluorophores revealed the relative proportion of autophosphorylated and nonphosphorylated kinase in individual neurons throughout the cultures. We found that autophosphorylated and nonphosphorylated kinase are colocalized throughout most neurons rather than segregated within distinct cells or subcellular domains. However, the variations in intensity of the two fluorophores indicated that the proportion of autophosphorylated kinase is consistently higher in neuronal somas than in the neuropil. Incubation of the cultures in Ca2+ free medium dramatically reduced both the level of autophosphorylated kinase detected biochemically and the relative intensity of fluorescent staining with the phosphokinase specific monoclonal antibody. These results support the hypothesis that regulation of Ca(2+)-independent CaM kinase activity in vivo occurs by a dynamic equilibrium between autophosphorylation and dephosphorylation and that this equilibrium is maintained, at varying steady-state levels, in all parts of neurons.

Anthrax toxin protective antigen is activated by a cell surface protease with the sequence specificity and catalytic properties of furin.

Proteolytic cleavage of the protective antigen (PA) protein of anthrax toxin at residues 164-167 is necessary for toxic activity. Cleavage by a cellular protease at this sequence, Arg-Lys-Lys-Arg, normally follows binding of PA to a cell surface receptor. We attempted to identify this protease by determining its sequence specificity and catalytic properties. Semi-random cassette mutagenesis was used to generate mutants with replacements of residues 164-167 by Arg, Lys, Ser, or Asn. Analysis of 19 mutant proteins suggested that lethal factor-dependent toxicity required the sequence Arg-Xaa-Xaa-Arg. Based on these data, three additional mutants were constructed with the sequences Ala-Lys-Lys-Arg, Arg-Lys-Lys-Ala, and Arg-Ala-Ala-Arg. Of these mutant proteins, Arg-Ala-Ala-Arg was toxic, confirming that the cellular protease can recognize the sequence Arg-Xaa-Xaa-Arg. The mutant containing the sequence Ala-Lys-Lys-Arg was also toxic but required > 13 times more protein to produce equivalent toxicity. This sequence specificity is similar to that of the ubiquitous subtilisin-like protease furin, which is involved in processing of precursors of certain receptors and growth factors. Therefore we tested whether a recombinant soluble furin would cleave PA. This furin derivative efficiently cleaved native PA and the Arg-Ala-Ala-Arg mutant but not the nontoxic PA mutants. In addition, previously identified inhibitors of furin blocked cleavage of receptor-bound PA. These data imply that furin is the cellular protease that activates PA, and that nearly all cell types contain at least a small amount of furin exposed on their cell surface.

Human furin is a calcium-dependent serine endoprotease that recognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxin protective antigen.

Previous work demonstrated that human furin is a predominantly Golgi membrane-localized endoprotease that can efficiently process precursor proteins at paired basic residues (-Lys-Arg- or -Arg-Arg-) in transfected cells. Anion-exchange chromatography of culture supernatant from cells expressing a soluble truncated form of human furin resulted in a greatly enriched preparation of the endoprotease (approximately 70% pure as determined by protein staining). Enzymatic studies show that furin is a calcium-dependent (K0.5 = 200 microM) serine endoprotease which has greater than 50% of maximal activity between pH 6.0 and 8.5. The inhibitor sensitivity of furin suggests that it is similar to, yet distinct from, other calcium-dependent proteases. Evidence that furin may require a P4 Arg in fluorogenic peptide substrates suggested that this enzyme might cleave the protective antigen (PA) component of anthrax toxin at the sequence -Arg-Lys-Lys-Arg-. Indeed, PA was cleaved by purified furin at the proposed consensus site (-Arg-X-Lys/Arg-Arg decreases-) at a rate (8 mumol/min/mg total protein) 400-fold higher than that observed with synthetic peptides. In addition, the processing of mutant PA molecules with altered cleavage sites suggests that furin-catalyzed endoproteolysis minimally requires an -Arg-X-X-Arg- recognition sequence for efficient cleavage. Together, these results support the hypothesis that furin processes protein precursors containing this cleavage site motif in the exocytic pathway and in addition, raises the possibility that the enzyme also cleaves extracellular substrates, including PA.

Activation of human furin precursor processing endoprotease occurs by an intramolecular autoproteolytic cleavage.

Human furin is a calcium-dependent serine endoprotease that can efficiently cleave many precursor proteins on the carboxyl side of the consensus cleavage sequence, -Arg-X-Lys/Arg-Arg-, both in vivo and in vitro. Analysis of furin proteins in extracts of cells infected with a vaccinia recombinant expressing human furin show that the enzyme is present as two prominent forms of 90 and 96 kDa. Because the structurally related bacterial subtilisins require endoproteolytic removal of the NH2-terminal pro-region by an autocatalytic intramolecular cleavage, we speculated that the size heterogeneity in the furin doublet similarly may result from a proteolytic removal of an NH2-terminal pro-region. Here we report identification of the 90-kDa furin NH2 terminus and, based on the reported sequence of the furin cDNA, demonstrate that this furin protein is derived from a larger precursor by an endoproteolytic cleavage on the COOH-terminal side of a consensus furin cleavage site, -Arg-Thr-Lys-Arg107-. Expression of mutant furin molecules containing an altered cleavage site (Arg104----Ala or Arg107----Gly) resulted in the production of only the 96-kDa furin protein. Assays of furin-dependent cleavage of a protein substrate in vitro showed that proteolytic activity was associated with the 90-kDa and not the 96-kDa furin protein, demonstrating that removal of the NH2-terminal pro-region is required for furin activity. Expression of a third furin construct containing a mutation of the active site aspartate (Asp153----Asn) similarly resulted in the expression of only the 96-kDa protein, suggesting that furin activation occurs by an autoproteolytic cleavage. Finally, the production of 90-kDa furin from either site-directed furin mutant could not be potentiated by overexpressing active furin, suggesting that the autoproteolytic activation was an intramolecular event.

Autophosphorylation of type II Ca2+/calmodulin-dependent protein kinase in cultures of postnatal rat hippocampal slices.

Autophosphorylation of Thr286 on type II Ca2+/calmodulin-dependent protein kinase (CaM kinase) in vitro causes kinase activity to become partially independent of Ca2+. Here we report that Thr286 is the major CaM kinase autophosphorylation site occupied in situ in "organotypic" hippocampal cultures. Measurement of Ca(2+)-independent CaM kinase activity revealed that approximately one-third of the kinase is autophosphorylated in situ when the basal Ca2+ concentration is 15-43 nM. This proportion was substantially reduced 30 min after removal of extracellular Ca2+ or treatment of the cultures with protein kinase inhibitors and was increased by treatment with okadaic acid. Therefore, the high proportion of autophosphorylated kinase at basal Ca2+ concentrations appears to be maintained by Ca(2+)-dependent autophosphorylation. Homogenates of intact hippocampi also contain a high proportion of Ca(2+)-independent type II CaM kinase, 13-23% depending on developmental age. Thus, in hippocampal neurons, an important function of the autophosphorylation mechanism may be to produce a relatively high level of CaM kinase activity, even at basal Ca2+ concentrations, permitting both upward and downward local regulation by physiological agents.

Conserved and variable regions in the subunits of brain type II Ca2+/calmodulin-dependent protein kinase.

Brain type II Ca2+/calmodulin-dependent protein kinase is a holoenzyme composed of several copies each of three subunits, alpha (50 kd), beta (60 kd), and beta' (58 kd), in varying proportions. The deduced amino acid sequences of alpha (reported here) and beta are highly similar but not identical. The major difference between them is the deletion from alpha of two short segments (residues 316-339 and 354-392 in beta). cDNAs that appear to encode beta' are identical to beta except for the deletion of a segment encoding residues 378-392. Thus, the structural differences among alpha, beta, and beta' arise primarily from deletions (or insertions) in a variable region lying immediately carboxyl to the protein kinase and calmodulin-binding domains. The alpha and beta subunits are encoded by distinct genes expressed primarily, if not exclusively, in brain. Rather than being encoded by a third gene, beta' may arise by alternative splicing of the beta gene transcript.