Bioengineered Peptides Based on α1-PDX Structure as Inhibitors of Furin: Design, Synthesis and Comparative Efficacy

Document Type: Research Paper


1 Faculty of Pharmacy, Tabriz University of Medical Sciences, Tabriz, Iran, Diseases of Aging Program, Regional Protein Chemistry Center, Ottawa Health Research Institute, University of Ottawa, Ottawa, Ontario, Canada,

2 Faculty of Pharmacy, Tabriz University of Medical Sciences, Tabriz, Iran,

3 Research Center for Pharmaceutical Nanotechnology, Tabriz University of Medical Sciences, Tabriz, Iran

4 Diseases of Aging Program, Regional Protein Chemistry Center, Ottawa Health Research Institute, University of Ottawa, Ottawa, Ontario, Canada,


     Furin is a Ca 2+ -dependent serine protease which cleaves proprotein substrates at the Arg-Xaa-(Lys/Arg)-Arg    site to generate biologically active proteins. Furin’s critical role in many cellular events associated with health disorders such as HIV, SARS, anthrax, and influenza as well as cancer has made inhibitors of this enzyme as therapeutic targets. To this date, the most potent inhibitor of furin is the bioengineered serpin (serine protease inhibitors) protein namely α1-PDX. It was already demonstrated that the reactive site loop (RSL) of all serpins are prime interactive domains responsible for their protease inhibitory function. Therefore, the objective of the present study was to develop small peptides with the RSL structure of serpin α1-PDX, as inhibitors of furin activity. Fifteen peptides were designed from reactive site loop structure of α1-PDX (sequences 367-394) with different mutations in this site, and were synthesized using a solid-phase peptide synthesizer, and  characterized by MALDI-tof mass spectrometry and amino acid analysis. The inhibitory effects of the designed peptides against furin activity were evaluated by spectroflourometry using QVEGF-C [Abz-QVHSIIRRßSLP-Y(NO2)-A-CONH2, Abz = 2-amino benzoic acid and Y(NO2) = 3-nitro tyrosine] as substrate. The results showed that all of the designed peptides inhibit furin activity with different efficacies in a time and concentration dependent pattern. Peptides containing His  or straight alkyl side chain amino acids in positions P2, P3, P6 and P8 have higher efficacy for blocking furin activity in comparison with peptides containing Arg or Lys in that position. The study further revealed that the peptides inactivate furin in a slow tight binding pattern. Our study provides an alternate strategy for development of efficient peptide-based inhibitors of proprotein convertases such as furin.


1. Introduction

    The discovery of mammalian subtilases,the proprotein convertases (PCs), in 1990 came as a culmination of a prolonged and sustained effort to hunt for enzyme/s responsible for proteolytic maturation of inactive precursor proteins [1-3]. Since then seven PCs namely PC1/3, PC2, PC4, PC5/6, PC7/8/LPC, furin/PACE (paired-basic-amino acid-converting enzyme) and PACE4 have been discovered that cleave at the carboxyl

terminal of the consensus Arg/Lys-X-X/Lys- Arg  ↓ site [4]. Proteolytic processing of larger inactive precursor proteins into smaller functionally bioactive forms is an important cellular event implicated in both normal and abnormal situations. Because of this role, proteolytic enzymes remain in the forefront as important therapeutic targets for intervention of many diseases and metabolic disorders, even though several of these proteases are also linked to normal development and important physiological functions [1-4].

     Furin is the first and so far the best characterized enzyme of the mammalian subtilisin-like family of Ca+2-dependent cellular endoproteases. As well as activating pathogenic agents, furin has an essential role in embryogenesis, and catalyses the maturation of a diverse collection of proprotein substrates. These range from growth factors and receptors to extracellularmatrix proteins and even other protease systems that control disease. Furin’s crucial role in so many different cellular events, and in diseases ranging from anthrax and bird flu to cancer, dementia and Ebola fever, has caused researchers to re-evaluate it [5-8].

Table 1. List of various peptides derived from the amino acid sequence of reactive site loop of Alpha 1-PDX containing the furin cleavage motif.

Peptide I


Peptide II


Peptide III


Peptide IV


Peptide V


Peptide VI


Peptide VII


Peptide VIII


Peptide IX


Peptide  X


Peptide XI


Peptide XII


Peptide XIII


Peptide XIV


Peptide XV



     The development of potent and specific protease inhibitors became an important research topic in drug design and therapeutic applications. Already a number of inhibitors directed against furin have been described in the literature for treatment of conditions such as viral infections like AIDS, inflammatory diseases such as chronic obstructive pulmonary disorder, prostate cancer and others [9-12]. Despite several reports, not much significant success has been achieved in designing potent small compound inhibitors of furin, that are more attractive as therapeutic agents  because  of  their  stability  and accessibility. In the present study, we report the  development  of  peptide  molecule inhibitors of furin consisting of the RSL structure of a1-PDX with different mutations in P1 to P8 positions that block furin activity with different potencies. These findings were based on in vitro experiments using small peptide and physiological protein substrates.

 Figure 1. RP-HPLC chromatograms of crude peptide (VIII) (A) and its corresponding MALDI mass spectra (B).

2. Materials and methods

2.1. Materials

     All Fmoc (fluoren-9-yl methoxycarbonyl) amino acids (L-configuration), the coupling agents and organic solvents were purchased from PE Applied Biosystems (Framingham, MA, USA), Calbiochem–Novabiochem (San Diego, CA, USA), Chem-Impex International (Wood Dale, IL, USA) and Aldrich Chemical Company    (Milwaukee,    WI,    USA). Recombinant furin enzyme and its substrate Boc-RVRR-MCA   (MCA   =   4-methyl coumaryl-7-amide) used for kinetic study were purchased from New England BioLabs Inc. (MA, USA) and Peptides International (Louisville, KY, USA), respectively. The intramolecularly quenched fluorogenic substrate,QVEGF-C[Abz-QVHSIIRR227 ßSLP-Y(NO2)-A-CONH2, MW 1701, Abz = 2-amino benzoic acid and Y(NO2) = 3-nitro tyrosine] used in the present study as a furin substrate [13]. This peptide was derived from the physiological processing site RR227ßSL of VEGF-C. Acetonitrile (ACN), 4-  hydrox- ycinnamic acid (CHCA), HCl, triflouro ethanol (TFE), and all other reagents were purchased from Sigma-Aldrich Company (Milwaukee, MI, USA).

Figure 2. RP-HPLC chromatograms of pure peptide (VIII) (A) and its corresponding MALDI mass spectra (B).

2.2. Peptide synthesis

     All peptides were synthesized as C- terminal amides on an automated solid-phase peptide synthesizer (Pioneer, PE-Perseptive Biosytems Inc., Framingham, MA, USA), following O-hexafluoro-phospho-[7-azaben- zotriazol-1-yl]-N,N,N’,N’-tetramethyluronim (HATU)/di-isopropylethylamine (DIEA)- mediated Fmoc chemistry using unloaded PAL-PEG (polyamino linker polyethylene glycol) resin. The following amino acid side chain protecting groups were used: t-butyloxy- carbonyl  (Boc)  for  Lys;  2,  2,  4,  6,  7- pentamethyl dihydrobenzo furan-5-sulphonyl (Pbf) for Arg; t-butyl for Ser, Thr, Asp and Tyr and trityl for His, Asn and Gln. A deprotection cocktail (Reagent B) containing TFA, phenol, water and TIPS (triisopropylsilane) was used and added to the peptide-bound resin (5 ml) [13].  After 3  h  of treatment at the room temperature, the peptide was recovered as described [13].

2.3.   RP-HPLC   protocol   for   peptide purification

     All peptides were purified by reversed- phase      high      performance      liquid chromatography    (RP-HPLC,    Rainin Dynamax) using a semi-preparative column (CSC    Exsil,    C18,    25    cm×1.0    cm, Chromatography Specialty Co., St-Laurent, Quebec, Canada). The buffer system consisted of an aqueous 0.1% (v/v) trifluoroacetic acid solution and an organic phase of acetonitrile also containing 0.1% (v/v) trifluoroacetic acid. The elution was carried out using a linear gradient from 15-60% organic phase in 105 min. following a 10 min. isocratic step at 100% aqueous phase (gradient A, semiprepar- ative run) or from 0-60% organic phase in 60 min.  (gradient A, for analytical run). The flow rate was adjusted to 2.0 or 1.0 ml/min., respectively, and the separation was monitored by UV absorbance at 230 nm.

Table 2. Amino acid analysis of peptide VIII (n = 3).





















Mean # of residue

Known observed



times (min.)





per molecule

± SD









2.22 ± 0.13









2.95 ± 0.19


















3.13 ± 0.09









1.12 ± 0.11









1.95 ± 0.18









1.10 ± 0.12









1.25 ± 0.05









4.34 ± 0.25









0.93 ± 0.07









1.94 ± 0.13









1.00 ± 0.06


















3.06 ± 0.08









3.08 ± 0.11









1.25 ± 0.09



















2.4. Amino acid analysis

     Quantitative amino acid analyses were performed following 24 h hydrolysis in 6 N HCl in a sealed tube at 110 °C in vacuo using the ion chromatography system (ICS-2500, Dionex,  Oakville, Canada) equipped with conductance detector [14]. Norluocin was used as internal standard.

     The quantity of each residue in the peptide molecules was calculated using the below procedure. First of all the molecular weight of each peptide was divided by 112 (approximate average residue weight) to estimate total residue per molecule. After that the calculated µmol of all amino acids (according to the obtained relative area from ion exchange chromatograms  as  well  as  calibration equation) were sum totaled and divided by the calculated total residues per molecule to estimate µmol of peptide analyzed. µmol of each amino acid was then divided by µmol of peptide analyzed to obtain residues per peptide molecule [14].

2.5. Mass spectral analysis

     The identity of each peptide was fully confirmed by matrix-assisted laser-desorption ionization-time of-flight (MALDI-TOF) mass spectrometry  (Voyageur-DE  Pro,  PE- Biosystems Inc., Framingham, MA, USA) or   surface  enhanced  laser  desorption ionization time of flight (SELDI-TOF) (Ciphergen,  CA,  USA)  on  gold  chips ( using 4-hydrox- ycinnamic acid or 1, 2 dihydroxy benzoic acid as energy absorbing matrix [15].

2.6. Enzyme assay and kinetic evaluation

    All in vitro enzyme studies were performed using initial rate and/or end-time assays at room temperature in a final volume of 100 µl in 96-well flat-bottom black plates (Micro- fluor, Dynatec, USA). The buffer consisted of 25 mM Mes, 25 mM Tris and 2.5 mM CaCl2 (pH 7.4). Assays were performed using Q- VEGF-C in the concentration of 10 µM. The release of highly fluorescent 2-amino benzoic acid (Abz) containing N-terminal fragment from QVEGF-C were monitored by a spec- trofluorometer instrument (Perkin Elmer model LS50B) at excitation and emission wavelengths of 370 nm/460 nm. Recombinant furin was pre-incubated for 30 min at the room 25 °C with various concentrations of the mini PDX peptide. The fluorogenic substrate QVEGF-C was then added, and the released Abz-containing peptide was measured [16]. For measurement of IC50 (the concentration necessary to achieve 50% inhibition of the enzymic  activity) values, the inhibitor concentrations were varied over a range wide enough to yield residual activities of 25-75% of the control value [16, 17].

Figure 3. Ion exchange chromatogram for peptide VIII.

Figure 4. Bar graphs based on amino acid analysis of hydrolyzed and derivative peptide VIII.

3. Results and discussion

3.1. Design of inhibitor peptides

     Furin’s critical role in many cellular events associated with infections of HIV, SARS, Ebola, anthrax, and influenza as well as cancer,    neurological    dementia    like Alzheimer’s disease has made this enzyme a therapeutic target [18-20]. So far, several furin inhibitors have been reported that are either macromolecular proteins or small molecule peptides to non peptides [9]. Among them  the most potent is the bioengineered serpin protein namely a1-PDX with IC50 value of 0.6 nM [21]. It was already demonstrated that RSL (reactive site loop) of α1-PDX and also almost all of other members of serpin (serine proteinase inhibitors) family are prime interactive domains responsible for their protease inhibitory function [22, 23]. To find an optimum region of α1-PDX to design small inhibitors against furin, alignment of amino acid sequence of the most important members  of  serpin family, namely α1- antithrypsin, Antithrombin, C1-inhibitor, PAI (plasminogen  activator inhibitor), d-SPN4 (Drosophila),  ovalbumin  and  antichy- motrypsin, within  their reactive site loop (RSL) and ranking regions were performed using SIM alignment  tool available at -- The results of alignment indicate a resemblance of types of amino acids present at various P and P’ positions as there are near conserved alkyl side chain containing hydrophobic amino acids at P6, P5, P4  and P’2  positions. Also they all contain negatively charged Glu and/or Pro residues at positions between P’3-P’6. Thus based on these resemblances and the hypotheses of RSL (located between sites 350-394) significant role in furin inhibition, a 30-mer peptide from residue (367-394) was designed and to study the effect of different residue, several mutations in the designed peptide was made to obtain the best choice for interacting with furin. Table 1 lists the sequences of the designed peptides.


Figure 5. Bar graphs based on stop time assay to evaluate the effect of the designed peptides against furin activity after 10 minutes incubation of the digestion mixture, as measured with Q-VEGF-C (10 mcM) as a substrate (n = 3).

3.2. Synthesis and characterization of the designed peptides

     The designed peptides were synthesized using Fmoc protected amino acid residues by an automated peptide synthesizer machine. Following synthesis and deprotection of the peptides from PEG-PS (Poly Ethylene Glycol- Poly   Styrene)   resins.   Peptides   were lyophilized and then purified using RP-HPLC.

     Peaks eluting at Rt  = 51.721, 51.222, 49.631, 51.044, 51.531, 50.101, 49.361,50.035, 49.165, 49.996, 46.339, 45.671, 46.256, 45.931, 44.259 min., respectively for peptides I, II, III, IV, V, VI, VII, VIII, IX, X, XI, XII, XIII, XIV, XV, were collected and further analyzed by analytical RP-HPLC to confirm their homogeneities.

MALDI-tof mass spectra of the pure compounds showed single peaks at m/z 3364.0740, 3355.8001, 3447.1228, 3384.0178 (M+H) +, respectively, consistent with the calculated molecular weight of designed peptides compared to that of crude material which exhibited multiple peaks along with the correct mass. RP-HPLC chromatogram as well as MALDI-tof mass spectrum of peptide VIII instantly was shown in Figures 1 and 2, respectively.

































Figure 6. Bar graphs based on stop time assay to evaluate the effect of the designed peptides against furin activity after 30 min. incubation of the digestion mixture, as measured with Q-VEGF-C (10 μM) as a substrate (n = 3).

3.3. Amino acid analysis

     The  purified  peptides  were  further characterized by amino acid analysis. As described in materials and methods section, the peptides after hydrolysis by 6 N HCl, were  analyzed  using  cation  exchange chromatography    with    post    column derivatization. The results of ion exchange chromatography for hydrolyzed and derivative peptide VIII for instance was shown in Tables 2 and Figures 3 and 4. The areas under the curves for each peak were divided by those of standard (norleucine) and the amount of the residues in each sample was calculated according to the calibration equations of each residue. Finally the quantity of each residue in the  peptide molecules was calculated. According to the observed residue quantity of the peptide and comparison of them with the calculated quantity shown in bar graphs of Figure 4, the structures of the peptides were in consistent with the expected structure.

3.4. Effect of the designed peptides against furin activity

     In order to examine in vitro furin inhibition by designed peptides, a substrate namely IQVEGF-C (intramolecularly quenched vascular endothelial growth factor), [Abz-QVHSIIRR227¯SLP-Y (NO2)-A-CONH2] were selected.

Figure 7. Bar graphs based on stop time assay to evaluate the effect of the designed peptides against furin activity after 60 min. incubation of the digestion mixture, as measured with Q-VEGF-C (10 ∞ M) as a substrate (n = 3).

     Human VEGF-C, a potential marker of tumorigenesis, is a protein of 419aa residues with a predicted molecular mass of 59 kDa. It is intracellularly processed from a precursor named  pro-VEGF-C, that is a proprotein consisting of an N-terminal signal sequence (aa  1-12)  followed  by  an  N-terminal propeptide (aa 13–102), the VEGF homology domain (aa 103–227), and a cysteine-rich C- terminal segment (aa 228–419). ProVEGF-C is secreted as a disulfide-bonded homodimer that is proteolytically processed to release active form of the peptide. Upon examination of the amino acid sequence of the VEGF-C precursor, a dibasic motif, 220Q-VHSIIRR SLP230, resembling those recognized by the PCs is found, suggesting the involvement of these convertases in the maturation of VEGF- C. In fact furin was later found  to  be the processing enzyme [16].

     In order to examine in vitro the processing effects of furin on the maturation of pro-VEGF-C, an intramolecularly quenched fluorogenic (IQF) peptide, IQ-h-VEGF-C [Abz-220QVHSIIRR  SLP230-Y(NO2)-A- CONH2] containing the processing site of proVEGF-C     was     synthesized     and characterized in vitro previously by Khatib and  coworkers using an automated solid- phase    peptide    synthesizer.   Abz    is 2-(O)-aminobenzoic acid (fluorogenic group) and Y(NO2) is 3-nitrotyrosine (fluorescence quench group) [16]. According to the results of this study IQ-VEGF-C is a good substrate for furin in the case of in vitro studies and can be used to evaluate the effect of various parameters, like inhibitors, in furin activity [16].

By the cleavage after IIRR¯ site in this substrate, the highly fluorescent compound of 2-amino benzoic acid was released and monitored    using    spectrofluorometer instrument. The intramolecularly quenched fluorogenic (IQF) substrate contains the physiological furin activation site of human proVEGF-C [16].

     Three different concentrations of the designed peptides (2, 5 and 10 µM) were prepared in dH2O, and furin activity to cleave the above mentioned substrate was monitored in the absence and presence of the designed peptides, by both on line and stop time fluorescence spectroscopy.

     The results of stop time assays using 2, 5 and 10 µM of peptides I to XV at 10, 30 and 60 min using IQ-VEGF-C as substrate, were shown in Figures 5 to 7, respectively. It is evident that the pattern of inhibition of furin by the peptides is time and concentration dependent as it can be observed from the comparison of the Figures 5 to 7, and the comparison of bar graphs related to different concentrations in each figure. In the case of peptides I, II, III, VI, VII, VIII and XIII in all of incubation times and concentrations, an inhibition is monitored and intensified by increasing the concentrations and incubation times. In spite, in the peptides IV, V, IX, X, XI, XII, XIV and XV almost there is no inhibition at the first 10 min. in any of concentrations; in fact slight activation is noted in some cases. However, after 30 and 60 min. of incubation, relative inhibition is considered and heightened by increasing the concentrations. According to a study of Jean et al., the inhibition of furin by α1-PDX obeyed slow- binding inhibition kinetics whereby rapid formation of a loose α1-PDX-furin complex is followed by a slow isomerization to the tightly bound product [21]. Therefore, in the case of peptides IV, XI, XII, XIV and XV, the presence of more Arg or Lys residues at the positions P2, P6 and P8 may convert the peptides to an attractive substrate for furin, as it was shown by Klimpel et al. [24] as well as Molloy et al. [25] that furin cleaves an Arg- Xaa-(Lys/Arg)-Arg site with an E10-fold higher efficiency than that of Arg-Xaa-Xaa- Arg [24, 25]. Also three research groups have used a series of fluorogenic peptides to screen their effectiveness as substrates for furin. These studies have revealed that the Arg residues at the P4  and P1  positions are absolutely required for cleavage by furin. A basic residue (Lys or Arg) at the P2 position enhances the cleavage efficiency [21, 26, 27]. These basic residue requirements are in good agreement with those determined by in vitro studies using protein substrates [24, 27, 28].

     Thus, before the loose complex of furin- peptide could invert to a tightly bound product, the designed peptides were cleaved and degraded by furin.  On the other hand, replacement of Arg or Lys by His at the positions P2, P4 and P6, as well as mutating IL to Lue or Val at the position P3 intensified the inhibitory activity of the designed peptides. It can be concluded that the presence of straight rather than branch alkyl chain containing amino acid at P3 position can be stabilized the peptide-furin complex and lead to a more potent inhibition.


     The authors gratefully acknowledge the financial support from Ministry of Health and Medical Education, government of Iran for the award of scholarship to Farzaneh Lotfipour. The funding from CFI, government. of Canada and OIT, Govt. of Ontario and CIHR is also gratefully acknowledged.


[1]   Thomas G. Furin at the cutting edge: From protein traffic to embryogenesis and disease. Nature Rev Mole Cell Biol 2002; 3: 753-66.

[2]   Seidah  NG,  Chrétien  M.  Proprotein  and prohormone convertases: A family of subtilases generating diverse bioactive polypeptides. Brain Research 1999; 848: 45-62.

[3]   Denault JB, Leduc R. Furin/PACE/SPC 1: A convertase involved in exocytic and endocytic processing of precursor proteins. FEBS Lett 1996; 379: 113-6.

[4]   Rockwell NC, Krysan DJ, Komiyama T, Fuller RS. Precursor processing by Kex2/furin proteases. Chem Rev 2002; 102: 4525-48.

 [5]   Steiner DF. The proprotein convertases. CurrOpin Chem Biol 1998; 2: 31-9.

[6]   Basak A, Zhong M, Munzer JS, Chrétien M, Seidah  NG.  Implication  of  the  proprotein convertases furin, PC5 and PC7 in the cleavage of surface glycoproteins of Hong Kong, Ebola and respiratory syncytial viruses: A comparative analysis using fluorogenic peptides. Biochem J 2001; 353: 537-45.

[7]   Chrétien M, Mbikay M, Gaspar L, Seidah NG. Proprotein convertases and the pathophysiology of human diseases: Prospective considerations. Proc Assoc Am Physicians 1995; 107: 47-66.

[8]   Khatib AM, Siegfried G, Prat A, Luis J, Chretien M,  Metrakos  P,  Seidah  NG.  Inhibition  of proprotein convertases is associated with loss of growth and tumorigenicity of HT-29 human colon carcinoma cells: Importance of insulin-like growth factor-1 (IGF-1) receptor processing in IGF-1- mediated functions. J  Biol Chem 2001; 276:30686-93.

[9]   Villemure M, Fournier A, Gauthier D, Rabah N, Wilkes BC, Lazure C. Barley serine proteinase inhibitor-2 derived cyclic peptides as potent and selective inhibitors of convertases PC1/3 and furin. Biochemistry 2003; 42: 9659-68.

[10] Zhong M, Munzer JS, Basak A, Benjannet S, Mowla SJ, Decroly E, Chretien M, Seidah NG. The  prosegments of furin and PC7 as potent inhibitors of proprotein convertases: In vitro and ex vivo assessment of their efficacy and selectivity. J Biol Chem 1999; 274: 33913-20.

[11] Kang UB, Baek JH, Kim SHRJ, Yu MH, Lee C. Kinetic mechanism of protease inhibition by α1- antitrypsin. Biochem Biophys Res Comm 2004; 323: 409–15.

[12] Molloy SS, Bresnahan PA, Leppla SH, Klimpel KR, Thomas G. Human furin is a calcium- dependent serine endoprotease that recognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxin  protective antigen. J Biol Chem 1992; 267: 16396-402.

[13] Siegfried G, Basak A, Cromlish JA, Benjannet S, Marcinkiewicz J, Chrétien M, Seidah NG, Khatib AB. The secretory proprotein convertases furin, PC5, and PC7 activate VEGF-C to induce tumorigenesis. J Clinical Invest 2003; 111: 1723-32.

[14] Amino acid analysis, Perkin Elmer Applied Biosystems; 2004 [on line], Available from a.html.

[15] Brown  RS,  Lennon  JJ.  Mass  resolution improvement by incorporation of pulsed ion extraction    in    a    matrix-assisted    laser desorption/ionization linear time-of-flight mass spectrometer. Anal Chem 1995; 67: 1998-2003.

[16] Khatib AM, Siegfried G, Chretien M, Metrakos P, Seidah NG. Proprotein convertases in tumor progression and malignancy: Novel targets in cancer therapy. Am J Pathol 2002; 160: 1921-35.

[17] Jean F, Basak A, Dugas H, Lazure C. Design of ctapeptide loop structures by modeling the active site region of barley serine proteinase inhibitor. In: Hodges  RS, Smith JA, (editors). Peptides, chemistry, structure and biology ESCOM. Leiden, Netherlands, Peptides,  Proceedings of 13th American Peptide Symposium 1994; pp. 613-5.

[18] Decroly E, Vandenbranden M, Ruysschaert JM, Jacob GS, Howard SC, Marshall G, Kompelli A, Basak A, Jean F, Lazure C, Benjannet S, Chretien M, Day R, Seidah NG. The convertases furin and PC1 can both cleave the human immunodefi- ciency virus (HIV)-1 envelope glycoprotein gp160 into gp120 (HIV-1 SU) and gp41 (HIV-I TM). J Biol Chem 1994; 269: 12240-7.

[19] Bergeron E, Vincent MJ, Wickham L, Hamelin J, Basak A, Nichol ST, Chretien M,  Seidah NG. Implication of proprotein convertases in the processing and spread of severe acute respiratory syndrome  coronavirus. Biochem Biophys Res Comm 2005; 326: 554-63.

[20] Volchkov VE, Feldmann H, Volchkova VA, Klenk HD. Processing of the Ebola virus glycoprotein by the  proprotein convertase furin. Proc Natl Acad Sci  1998; 95: 5762-7.

[21] Jean FK, Stella L, Thomas G, Liu Y, Xiang AJ, Reason A, Thomas G. α1-Antitrypsin Portland, a bioengineered serpin highly selective for furin: Application as an antipathogenic agent. Proc Natl Acad Sci 1998; 95: 7293-8.

[22] Gettins PGW. Serpin structure, mechanism and function. Chem Rev 2002; 102: 4751-803.

[23] Irving JA, Pike RM, Lesk AM, Whisstock JC.

Phylogeny of the serpin superfamily: Implications of patterns of amino acid conservation for structure and function. Genome Res. 2000; 10:1845-64.

[24] Klimpel KR, Molloy SS, Thomas G, Leppla SH.Anthrax toxin protective antigen is activated by a cell surface protease with the sequence specificity and catalytic properties of furin. Proc Natl Acad Sci 1992; 89: 10277-81.

[25] Molloy SS, Bresnahan PA, Leppla SH, Klimpel KR, Thomas G. Human furin is a calcium- dependent serine endoprotease that recognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxin  protective antigen. J Biol Chem 1992; 267: 16396-402.

[26] Molloy SS, Thomas L, van Slyke JK, Stenberg PE, Thomas G. Intracellular trafficking and activation of the furin proprotein convertase: Localization to the TGN and recycling from the cell surface. EMBO J 1994; 13: 18-33.

[27] Hatsuzawa K, Nagahama M, Takahashi S, Takada K, Murakami K, Nakayama K. Purification and characterization of furin, a Kex2-like processing endoprotease, produced in chinese hamster ovary cells. J Biol Chem 1992; 267: 16094-9.

[28] Takahashi  S,  Hatsuzawa  K,  Watanabe  T, Murakami K, Nakayama K. A Kex2-related endopeptidase activity present in rat liver specifically processes the insulin proreceptor. J Biochem (Tokyo) 1994; 116: 47-52.