Isolation and Characterization of a Cone Snail Protease with Homology to CRISP Proteins of the Pathogenesis-related Protein Superfamily

Abstract

The pathogenesis-related (PR) protein superfamily is widely distributed in the animal, plant, and fungal kingdoms and is implicated in human brain tumor growth and plant pathogenesis. The precise biological activity of PR proteins, however, has remained elusive. Here we report the characterization, cloning and structural homology modeling of Tex31 from the venom duct of Conus textile. Tex31 was isolated to >95% purity by activity-guided fractionation using a para-nitroanilide substrate based on the putative cleavage site residues found in the propeptide precursor of conotoxin TxVIA. Tex31 requires four residues including a leucine N-terminal of the cleavage site for efficient substrate processing. The sequence of Tex31 was determined using two degenerate PCR primers designed from N-terminal and tryptic digest Edman sequences. A BLAST search revealed that Tex31 was a member of the PR protein superfamily and most closely related to the CRISP family of mammalian proteins that have a cysteine-rich C-terminal tail. A homology model constructed from two PR proteins revealed that the likely catalytic residues in Tex31 fall within a structurally conserved domain found in PR proteins. Thus, it is possible that other PR proteins may also be substrate-specific proteases. The pathogenesis-related (PR) protein superfamily is widely distributed in the animal, plant, and fungal kingdoms and is implicated in human brain tumor growth and plant pathogenesis. The precise biological activity of PR proteins, however, has remained elusive. Here we report the characterization, cloning and structural homology modeling of Tex31 from the venom duct of Conus textile. Tex31 was isolated to >95% purity by activity-guided fractionation using a para-nitroanilide substrate based on the putative cleavage site residues found in the propeptide precursor of conotoxin TxVIA. Tex31 requires four residues including a leucine N-terminal of the cleavage site for efficient substrate processing. The sequence of Tex31 was determined using two degenerate PCR primers designed from N-terminal and tryptic digest Edman sequences. A BLAST search revealed that Tex31 was a member of the PR protein superfamily and most closely related to the CRISP family of mammalian proteins that have a cysteine-rich C-terminal tail. A homology model constructed from two PR proteins revealed that the likely catalytic residues in Tex31 fall within a structurally conserved domain found in PR proteins. Thus, it is possible that other PR proteins may also be substrate-specific proteases. Of the genomes that have been completely sequenced, as many as 2% of the gene products encode known proteases, many of which regulate physiological processes such as blood coagulation, fibrinolysis, the complement system, and the processing of protein hormone precursors by specific convertases. Prohormones and neuropeptides (3–40 amino acids), important coordinators of cellular function in the endocrine and nervous systems, are often synthesized as propeptides. The propeptide is subsequently processed by substrate-specific proteases to yield the mature bioactive form (1Docherty K. Steiner D. Annu. Rev. Physiol. 1982; 44: 625-638Crossref PubMed Scopus (397) Google Scholar). As many of the substrate-specific proteases remain unidentified, predicting new bioactive peptides from cDNA sequences is presently difficult, if not impossible (2Bogyo M. Ploegh H. Nature. 1998; 396: 625-627Crossref PubMed Scopus (17) Google Scholar, 3Nakai K. J. Struct. Biol. 2001; 134: 103-116Crossref PubMed Scopus (41) Google Scholar). Linderström-Lang coined the term “limited proteolysis” to describe enzymes with restricted specificity (4Linderström-Lang K. Otteseen M. Nature. 1947; 159: 807-808Crossref PubMed Scopus (27) Google Scholar). Enzyme activity can be restricted by substrate structure, enzyme processing state, endogenous protease inhibitor levels or a combination of these factors (5Neurath H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 10962-10963Crossref PubMed Scopus (66) Google Scholar). Peptides derived from prohormones are typically flanked by a pair of dibasic residues (Lys-Arg, Arg-Arg, Lys-Lys, or Arg-Lys) and are cleaved by proteases found in secretory vesicles (1Docherty K. Steiner D. Annu. Rev. Physiol. 1982; 44: 625-638Crossref PubMed Scopus (397) Google Scholar). However, many precursor peptides contain multiple sets of basic residues, suggesting that highly substrate specific or differentially expressed proteases can determine processing outcomes. The venom of cone snails (predatory marine molluscs of the genus Conus) has yielded a rich source of novel neuroactive peptides (“conotoxins”) (6Olivera B.M. Cruz L.J. Yoshikami D. Curr. Opin. Neurobiol. 1999; 9: 772-777Crossref PubMed Scopus (45) Google Scholar). However, the enzymes responsible for cleaving the conotoxins from their propeptide precursors have not been isolated. Here we report the purification, cloning, and initial characterization of Tex31, a protease from venom Conus textile. Interestingly, Tex31 sequence was found to be homologous to members of the PR protein superfamily. Substrate Design and Activity Assay—Based on the cDNA sequence of the propeptide and the mature amino acid sequence of the TxVIA peptide, a δ-conotoxin contained in the venom of C. textile, propeptide cleavage site substrate analogues were designed. The peptidic para-nitroanilide (pNA) 1The abbreviations used are: pNA, para-nitroanilide; Bistris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; Abu, 2-aminobutyric acid; HPLC, high performance liquid chromatography; AEBSF, 4-(2-aminoethyl)benzenesulfonyl fluoride; PR, pathogenesis-related. substrate analogues were synthesized by solid phase synthesis (7Abbenante G. Leung D. Bond T. Fairlie D. Lett. Pept. Sci. 2001; 7: 347-351Google Scholar). A typical reaction mixture consisted of 100 mm Bistris propane, pH 8.5, 100 μm pNA substrate (acetyl-KLNKR-pNA or Ac-LNKR-pNA) and protease from an appropriate source. The reaction mixture was incubated at 37 °C for 1 h. All assays were carried out in 96-well plates and their absorbance measured at 405 nm on a Spectromax spectrometer. To confirm Tex31 could cleave amide bonds, an 11-residue peptidic substrate (acetyl-KLNKRWAbuKQSG-NH2), which is made up of five residues N-terminal and six residues C-terminal to the propeptide cleavage site, was assembled on Rink amide MBHA resin (Novabiochem) using standard Fmoc (N-(9-fluorenyl)methoxycarbonyl) SPPS chemistry (8Alewood P.F. Croft M. Schnolzer M. Kent S.B.H. Giralt E. Andreu D. Peptides 1990. ESCOM, Leiden, The Netherlands1990: 174-175Google Scholar). To prevent any possible interaction arising from the presence of a free cysteine residue, which in the native conotoxin propeptide may be involved in a disulphide bond, it was replaced with the cysteine isostere 2-aminobutyric acid (Abu). The peptide was deprotected and cleaved from the solid support with 95% trifluoroacetic acid, 2.5% triisopropylsilane, and 2.5% water (25 ml/g of resin) for 1 h at room temperature. The crude peptide was purified by reversed-phase HPLC. A typical peptide substrate reaction mixture contained 100 mm Bistris propane, pH 8, 10 mm CaCl2, 80 μm peptide substrate, and SEC I (150 pmol) purified protease. The reaction mixture was incubated at 37 °C and the rate of digestion monitored by reversed-phase HPLC. The masses of resulting peptide fragments were confirmed by mass spectrometry. Isolation and Purification of Tex31—Venom ducts from three C. textile specimens were removed and homogenized with a mortar and pestle in chilled extraction buffer (100 mm Bistris propane, pH 7, containing 0.1% Triton X-100). Extracted material was collected and centrifuged at 10,000 × g for 30 min (4 °C). The supernatant was concentrated with an Ultrafree 15 Biomax 5K cutoff centrifugal filter (Millipore) in preparation for the purification of Tex31. The Tex31 isolated for sequencing was first purified on a reversed-phase C4 semi-preparative column eluted with a 1% gradient of 0–80% acetronitrile containing 0.1% (v/v) trifluoroacetic acid. Active fractions were pooled, concentrated, and applied to a Superdex 75 column (Amersham Biosciences) eluted with 50 mm Tris-HCl, pH 8, containing 150 mm NaCl. Active fractions contained a major protein with an apparent molecular weight of ∼28,000 and of >95% purity as assessed by SDS-PAGE (Fig. 1A, lane 4). These fractions were pooled, concentrated, and applied to an analytical reversed-phase C4 HPLC column eluted with a 0.5% gradient of 40–70% acetronitrile containing 0.1% (v/v) trifluoroacetic acid. A single major peak (C4) eluting at 52% acetronitrile, with an apparent molecular weight of ∼28,000 and purity >99% as assessed by SDS-PAGE, was collected and sequenced (Fig. 1A, lane 5). To confirm that activity was associated with the ∼28,000-Da band and to purify sufficient material for biochemical analysis, additional Tex31 was purified using a modified method. First, an ion-exchange DEAE-Fast Flow Hi-Trap column (Amersham Biosciences) was equilibrated, and crude venom duct extract was eluted with 20 mm Bistris propane, pH 7. Tex31 was eluted with a linear 0–1 m NaCl gradient in 20 mm Bistris propane, pH 7, and fractions assayed in 50 mm Tris-HCl, pH 8.5. Pooled active fractions eluting at ∼150 mm NaCl were concentrated by ultrafiltration (as described above) and applied to a Biosep-SEC-S 2000 column (Phenomenex) equilbrated in 20 mm Bistris propane, pH 7, containing 150 mm NaCl. The eluted active fractions of > 95% purity as assessed by SDS-PAGE were pooled (SEC I) (Fig. 1B, lane 3) and used to characterize Tex31 proteolytic activity. This pooled fraction was reapplied to the SEC column eluted at 0.3 ml/min to obtain an active fraction that gave a single band on SDS-PAGE (SEC II) ∼28,000-Da protein (Fig. 1B, lane 4), confirming that proteolytic activity corresponded to the sequenced material. Mass Spectrometry and Sequencing—Electrospray quadrupole-time of flight mass spectra of Tex31, SEC II, and C4 were acquired on an API QSTAR pulsar mass spectrometer (Applied Biosystems). Purified C4 Tex31 was digested with trypsin (l-1-tosylamido-2-phenylethyl chloromethyl ketone-treated, Sigma) for 6 h at 37 °C in 200 μm N-ethylmorpholine, pH 8.5. Digested peptides were purified on a reversed-phase C18 column (Zorbax C18–300SB). Amino acid sequencing was performed on both the intact Tex31 and tryptic peptides by Edman degradation on an Applied Biosystems 477A protein sequencer. Cloning—TRIzol (Invitrogen) was used to extract total RNA from a single C. textile venom duct, previously stored in RNAlater (Ambion) at –20 °C, according to the manufacturer's protocol. Single-stranded cDNA was 3′ end-synthesized from total RNA templates using an oligo(dT)16 primer (PE Biosystems) extended with Superscript II reverse transcriptase (Invitrogen). Degenerate PCR oligonucleotide primers were designed based on the N-terminal protein sequence (5′-TAYTATTgYCTNACNCCNgCNCAYACNATg-3′) and the internal tryptic peptide (5′-CATRTTNgCNgCRTCNgTNACRTCNgC-3′), designated Pr1S and Pr2R, respectively. Touchdown PCR (64 °C to 50 °C) was carried out on C. textile cDNA using Red Hot DNA polymerase (Advanced Biotechnologies), 3 mm MgCl2, 200 μm dideoxynucleotides, primers Pr1S and Pr2R, producing a 140-bp cDNA fragment. The PCR fragment was ligated into a TA cloning vector (Invitrogen) and sequenced using Big Dye terminator chemistry (Applied Biosystems). The SMART rapid amplication of cDNA ends (Clontech) kit was then used to clone the 5′- and 3′-untranslated regions using the gene specific primers, Pr3S (5′-CgCgCACACAATgTCCTgA-3′) and Pr4R (5′-CATgTTggCAgCATCTgTTACgTCggCgCg-3′). The 5′ and 3′ cDNA PCR fragments were ligated into cloning vectors and sequenced as described above. The primers Pr5S (5′-gggAgTCACACAgACACAgC-3′) and Pr6R (5′-CgACgTCTTCAggTTggATT-3′) were used to amplify full-length Tex31 cDNA. Biochemical Characterization—The optimal pH for protease activity was determined over a pH range of 7.2–9.4 in 100 mm Bistris propane buffer. pNA substrate analogues based on the TxVIA conotoxin propeptide cleavage site along with three control substrates were assayed at 100 μm in 100 mm Bistris propane, pH 8.5 at 37 °C. Protease inhibitors were preincubated with Tex31 for 30 min at 37 °C before the addition of 100 μm Ac-LNKR-pNA and the reaction incubated for a further 1 h. RVKR-CMK (decanoyl-Arg-Val-Arg-Lys-chloromethylketone) was assayed at 100 μm, Kunitz-soy bean trypsin inhibitor, 100 μm; leupeptin, 100 μm; EGTA, 1 mm; EDTA, 1 mm; AEBSF (4-(2-aminoethyl)benzenesulfonyl fluoride) 100 μm, N α-tosyl-Lys-chloromethylketone, 100 μm; pepstatin A, 2 μm; E-64, 20 μm; and benzamidine, at 1 mm. Tex31 was preincubated for 30 min at 37 °C in buffers containing 2 mm metal ion, 1 mm EDTA, 100 mm Bistris propane, pH 8.5, before the addition of 100 μm Ac-LNKR-pNA. Tex31 in 100 mm Bistris propane, pH 8.5, buffer was included as a control. Initial velocities (v) were determined with 0–150 μm Ac-KLNKR-pNA, Tex31 SEC I pool (200 pmol) in 100 mm Bistris propane, 1 mm CaCl2, pH 8.5, at 37 °C. Kinetic parameters were calculated from nonlinear regression of initial velocities as a function of seven substrate concentrations [S] using GraphPad Prism software. The k cat/K m values were calculated assuming Michaelis-Menten kinetics. Tex31 SEC I pool (200 pmol) and 100 μm Ac-LNKR-pNA was used unless stated. Tex31 protein concentration was determined using the BCA protein assay kit (Pierce). Homology Modeling—The PR protein, Ves v 5, and P14a structures were overlaid within the InsightII (9Accelrys Inc.InsightII Modelling Enviroment. Accelrys Inc., San Diego, CA2001Google Scholar) environment to ascertain structural homology. The structurally conserved regions were then formally superimposed using InsightII. The structural alignment and homology modeling of Tex31 was carried out using the homology module with subsequent refinement using Discover 3. The validity of the model was tested using WHATCHECK and PROCHECK (10Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-291Crossref Google Scholar). Electrostatic potentials were calculated using GRASP (11Nicholls A. Sharp K. Honig B. Proteins. 1991; 11: 281-296Crossref PubMed Scopus (5318) Google Scholar). Using a Ac-LNKR-pNA substrate based on the propeptide cleavage site of conotoxin TxVIA (12Woodward S.R. Cruz L.J. Olivera B.M. Hillyard D.R. EMBO J. 1990; 9: 1015-1020Crossref PubMed Scopus (172) Google Scholar) in a simple colormetric assay, we isolated and purified to homogeneity a major protease Tex31 from the C. textile venom duct (Fig. 1). The venom duct contents contained the majority of Ac-LNKR-pNA-specific cleavage activity, suggesting that Tex31 is secreted from the venom duct tissue. Tex31 appeared not to be specifically associated with the insoluble granules found in the crude venom (data not shown). Under reducing conditions, Tex31 migrated with an apparent molecular weight of 28,000 on 12% SDS-PAGE gel (Fig. 1A, lanes 3–5, and B, lanes 2–4). Mass spectral analysis of Tex31 showed a mass of 30,825 Da, as well as a number of slightly heavier masses (Fig. 1C) that may represent modified forms of the enzyme. Similar profiles of related masses were seen in batches of Tex31 prepared by two separate purification methods (SEC II and C4; data not shown). Edman degradation of C4 purified native Tex31 yielded a single N-terminal sequence, HHXDSKYYELTPAHTM. Based on this protein sequence, and a tryptic digest fragment ADVTDAANMLK, we designed two degenerate primers to PCR-amplify a 140-bp fragment of cDNA encoding Tex31. Utilizing this sequence, we designed specific primers to isolate the 5′- and 3′-untranslated regions using SMART rapid amplication of cDNA ends. The final assembled cDNA sequence encoded a 300-amino acid protein (GenBank™/EBI data base: accession number AJ4913118) (Fig. 2). The predicted post-translationally processed secreted protein of 276 residues contains both the N-terminal and tryptic digest sequences and has a calculated molecular weight of 30,875. This result is consistent with SDS-PAGE (∼28,000 Da) and mass spectrometry (30,853 Da allowing for 11 disulfide bonds) data obtained from the native protease. A BLAST analysis (14Altschul S.F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Nucleic Acids Res. 1997; 25: 3389-3402Crossref PubMed Scopus (60233) Google Scholar) of the SwissProt and PIR databases indicated that Tex31 is a novel protein, with closest sequence identity to members of the pathogenesis-related (PR) protein family. The term pathogenesis-related protein was initially given to plant proteins that are induced in pathological or related situations (15Antoniw J. White R.F. Phytopathol. Z. 1980; 98: Scopus Google Scholar). are plant PR The of PR proteins in situations of or a in cellular Physiol. 1999; Scopus Google Scholar). proteins with high sequence homology to the plant family have been in the and including and fungal the human brain tumor PR proteins Zhang W. J. 159: PubMed Scopus Google Scholar) and T. PubMed Scopus Google mammalian protein M. J. K. PubMed Scopus Google from venom B. 1990; PubMed Scopus Google from venom G. M. D.R. J. 1993; Google and proteins from J.M. J. 1993; PubMed Scopus Google Scholar). alignment of PR proteins a highly conserved of amino (Fig. suggesting that may a This has been as that PR proteins from a and into a protein T. C. C. K. Proc. Natl. Acad. Sci. U. S. A. 1998; PubMed Scopus Google Scholar). the PR proteins, Tex31 the sequence identity to a of PR proteins found in and known as the cysteine-rich secretory proteins the CRISP U. D. J. B. A. C. T. PubMed Scopus Google which M. J. K. PubMed Scopus Google Scholar) and B. 1990; PubMed Scopus Google Tex31 a cysteine-rich C-terminal with of cysteine residues found in the Biochemical characterization that Tex31 is Tex31 cleaved the conotoxin propeptide substrate at the predicted propeptide cleavage site (Fig. The of the two resulting peptide and was confirmed by mass spectrometry. A number of pNA substrate analogues were used to the substrate specificity of Tex31 (Fig. Tex31 cleaved single basic amino acid pNA substrates However, the addition of a and cleavage activity by and respectively. A for the leucine at in a in activity, that the substrate specificity of Tex31 a leucine at activity was with the addition of a found in the native propeptide the of activity. The on cleavage activity of the and residues that the specificity of Tex31 is by residues removed from the cleavage To the other conotoxin family of Tex31, the substrate based on the cleavage site of the family was Tex31 also cleaved this substrate for substrates with a leucine at and suggesting that a related protease could be used by cone snails to cleave The of Tex31 to cleave the native substrate was for control substrates and which are the substrates for the 2 protease D. K. White H. G. Fairlie J. Biol. 2001; PubMed Scopus Google and for the substrate, which is cleaved by trypsin S. J. J. PubMed Scopus (27) Google Scholar). Tex31 is active over a pH range with activity at pH 8.5 (Fig. the pH of C. textile crude venom Tex31 the substrate with an apparent of 80 μm and a proteolytic cat/K of × (Fig. To further characterize Tex31 activity, the of metal and and inhibitors were assessed (Fig. 5, and The inhibitors and Kunitz-soy bean trypsin inhibitor Tex31. a inhibitor of and cysteine proteases, of the activity of Tex31, the inhibitor the inhibitor AEBSF which in a to AEBSF, was not tested to and The of catalytic residues modified by inhibitors using mass spectral peptide fragment analysis was made impossible the protease of Tex31. The addition of the metal and cleavage activity by and the addition of 1 mm a in substrate cleavage (Fig. This result that may to the structural 1991; PubMed Scopus Google Scholar) of Tex31. were found to for (Fig. to the of many protease the of to which protease Tex31 may be by To the of Tex31, we constructed a homology model for two PR protein The highly conserved amino from two PR protein the of from venom v A. K. H. M. Proteins. 2001; PubMed Scopus Google Scholar) and the of the plant protein P14a C. T. T. E. K. J. Biol. 1997; PubMed Scopus Google were with Tex31 and a model constructed in InsightII (Fig. enzyme active are in the of an enzyme N M. Thornton J. Sci. PubMed Google Scholar). the homology we in to (Fig. within a well (Fig. their and we that and are two members of a catalytic that basic A conserved typically the of the catalytic for proteases, which is predicted from the AEBSF (Fig. is from the homology However, the conserved (Fig. 3) is a possible it is on a that could it in to and Tex31 is the first member of the PR protein superfamily with a at four residues N-terminal of the cleavage site, and including a leucine at characterize by Tex31. A possible basic substrate site, which the likely catalytic residues and was in a Tex31 homology that the catalytic residues in Tex31 fall within structurally conserved regions of the PR proteins, we that PR proteins may presently substrate-specific proteases. of of the specific substrate for PR proteins may have to their proteolytic activity. Tex31 may the first into the by PR proteins in the in to and A. for mass D. Leung for control pNA D. for peptide for with and T. of for support this