Biochemical Characterization of a Cysteine Proteinase from Bauhinia Forﬁcata Leaves and Its Kininogenase Activity

Process Biochemistry 46 (2011) 572–578

Contents lists available at ScienceDirect Process Biochemistry

journal homepage: www.elsevier.com/locate/procbio

Biochemical characterization of a cysteine proteinase from Bauhinia forﬁcata leaves and its kininogenase activity

Sheila S. Andrade a, Rosemeire A. Silva-Lucca c, Lucimeire A. Santana a, Iuri E. Gouvea b, Maria A. Juliano b, Adriana K. Carmona b, Mariana S. Araújo a, Misako U. Sampaio a, Maria Luiza V. Oliva a,∗ a Department of Biochemistry, Universidade Federal de São Paulo, 04044-020 São Paulo, SP, Brazil b Department of Biophysics, Universidade Federal de São Paulo, 04044-020 São Paulo, SP, Brazil c Centro de Engenharias e Ciências Exatas, Universidade Estadual do Oeste do Paraná, 85903-000 Toledo, PR, Brazil article info abstract

Article history: In this work, the proteinase activity detect in the acetone precipitate (80%, v/v) of B. forficata leaves, Received 20 July 2010 trivially known as cow paw, and popularly used in folk medicine for treatment of diabetes mellitus, was Received in revised form 7 October 2010 purified by chromatography on Sephadex G-25, Canecystatin-Sepharose, and on Con A-Sepharose. The Accepted 20 October 2010 molecular weight 30 kDa was estimated by SDS-PAGE and zymography, and the N-terminal sequence and CD spectra indicated a relationship with the papain family of cysteine proteinases. Denominated Keywords: baupain, the enzyme was activated by dithiotreitol and inhibited by E-64 and iodoacetamide, but not Circular dichroism by benzamidine, TLCK, TPCK and EDTA. The S2 and S1 substrate specificity of baupain, assayed with two Cysteine proteinase Papain series of fluorescence resonance energy transfer (FRET) peptide substrates derived from Abz-KLRSSK-Q- Plant enzyme EDDnp, indicates a preference for Phe and Tyr at P2 position over Leu found in papain. Baupain releases Protein purification bradykinin from HMWK (human high molecular weight kininogen) though its proteolytic activity is −8 Substrate specificity blocked by the sequence motif QVVA of kininogen (Kiapp = 1.9 × 10 M). Canecystatin, from sugar cane, −9 which also lodges the QVVA sequence, inhibits baupain (Kiapp = 0.18 × 10 M). © 2010 Elsevier Ltd. Open access under the Elsevier OA license.

1. Introduction share a high degree of similarity within the cystatin family, including the sequence motif QXVXG [4,5]. Within the members of cysteine proteinases expressed in In plants, cysteine proteinases are widely distributed in various viruses, bacteria, animals and plants [1], papain is the archetypical tissues, and are involved in physiological events such as germina- member of these endopeptidyl hydrolases (EC 3.4.22). This group tion, senescence and environmental stress responses. Papain-like also comprise the mammalian lysosomal cathepsins, the cytoso- cysteine proteinases are often found in senescing organs partic- lic calpains (i.e., calcium activated cysteine proteinases) as well as ularly leaves [6–8], flowers [9], legume nodules [10] as well as several parasitic proteinases [2,3]. in germinating seeds [10–13]. Cysteine proteinases have been The action of these enzymes can be controlled by members intensively studied with various expression patterns, reported for of a family known as cystatins super family, which comprises different stages of plant development [14–16]. Equally studied is three families, on the basis of sub-cellular localization, molecu- cysteine proteinases role in processing and degradation of seed lar weight, disulfide bonds and sequence similarity including the storage proteins [17,18], in legume nodule development [19],in QXVXG motif. Specifically from plant, this group is known as plant response to stresses such as wounding, cold, and drought [20] as cystatins or phytocystatins (PhyCys) whose primary sequences well as in programmed cell death [21,22]. In the present work, the purification and the functional characterization of a new cysteine proteinase from B. forficata leaves are described. B. forficata is a Leguminosae known as cow paw, due to Abbreviations: FRET, fluorescence resonance energy transfer; Q-EDD, npglutaminyl-[N-(2,4-dinitrophenyl)-ethylenediamine]; Abz, ortho-aminobenzoic the characteristic bilobed aspect of its leaves from which the home- acid; HPLC, high performance liquid chromatography; LC/MS, liquid chromatogra- made extract is prepared and used in popular therapy for diabetes phy/mass spectrometry. mellitus. The studies reported that the beneficial of leaf extracts ∗ Corresponding author at: Universidade Federal de São Paulo, Departamento de (aqueous and alcoholic) in the prevention of diabetes complica- Bioquímica, Rua Três de Maio 100, 04044-020 São Paulo, SP, Brazil. tions is associated with oxidative stress since B. forficata and other Tel.: +55 11 55764444; fax: +55 11 55723006. E-mail address: [email protected] (M.L.V. Oliva). plant extracts have significant antioxidant activity [23,24].

Although this plant has been the subject of several studies 2.5. Effects of activators on enzyme activity [24–26], but a characterization of the enzyme activity is not yet ␮ ␤ l described. Baupain (0.23 M) was preincubated with DTT, -mercaptoethanol, and - cysteine (2 mM) in 0.1 M sodium phosphate buffer, pH 6.3 containing 0.4 M NaCl, The substrate specificity of the enzyme, designated as baupain 10 mM EDTA, at 37 ◦C for 30 min. The enzyme activity was determined as described, was assayed using peptides derived from the leading sequence Abz- using Z–Phe–Arg–MCA (0.4 mM) as substrate. KLRSSKQEDDnp [27–31] and the series Abz-KXRSSKQ-EDDnp and Abz-KLXSSKQ-EDDnp (X = different amino acids) used for mapping, 2.6. Effect of pH on the enzyme activity respectively, the S2 and S1 substrate specificity [29–31]. To fur- ␮ ther compare the hydrolytic properties of baupain with papain and Prior to the addition of the substrate Z–Phe–Arg–MCA 40 l of baupain (0.23 ␮M) was pre-incubated at 37 ◦C for 30 min with 100 ␮l of the following buffer human cathepsin L, we explored in vitro the ability of baupain to systems: 0.2 M sodium citrate, pH 4.0; 0.2 M sodium acetate, pH 5.0; 0.2 M sodium release kinin from HMWK. phosphate, pH 6.0; 0.2 M Tris/HCl, pH 7.0 and pH 8.0 and 0.2 M sodium bicarbonate, pH 9.0 and pH 10.0 in a final volume of 250 ␮l by adding 90 ␮l of distillated water. ␮ 2. Materials and methods The enzymatic activity was measured using 20 l of Z–Phe–Arg–MCA (5 mM) as substrate. 2.1. Enzyme purification 2.7. Proteinase inhibition studies Leaves (50 g) collected from wild Bauhinia forficata trees were homogenized in a blender with 0.15 M NaCl (700 ml). The proteins in the crude extract were pre- Effect of low molecular weight inhibitors: Baupain (0.23 ␮M) was incu- ◦ cipitated by 80% (v/v) acetone at 4 C. The sediment separated by centrifugation bated with 1.0 ␮M E-64 (l-trans-epoxysuccinyl-leucylamido [4-guanidino] butane), was dried at room temperature, and dissolved in 0.1 M sodium phosphate buffer, 1.0 mM benzamidine, 2.0 mM phenylmethanesulfonyl fluoride (PMSF), 2.0 mM pH 6.3. The enzyme fraction (1.5 ml) was applied on a size exclusion chromatogra- ortho-phenantroline, 5.0 mM ethylenediamine tetraacetic acid (EDTA), 1.0 mM phy (Sephadex G-25) equilibrated with 0.1 M sodium phosphate buffer, pH 6.3. The N-tosyl-l-phenylalanylchloromethyl ketone (TPCK), and 1.0 mM N-tosyl-l-lysyl enzyme activity was followed using Z–Phe–Arg–MCA as substrate. The fractions chloromethyl ketone (TLCK), for 10 min at 37 ◦C, before the addition of the substrate containing enzyme activity were pooled and subsequently chromatographed on a Z–Phe–Arg–MCA (0.4 mM). The assay concentration of each inhibitor was chosen canecystatin-Sepharose, equilibrated with the same phosphate buffer. The active based in suppliers information and inhibitor mode of action [37]. Enzyme activity material, eluted by 2.0 mM l-cysteine, was chromatographed on Con A Sepharose, was expressed in percent of residual activity on Z–Phe–Arg–MCA compared to the equilibrated with 0.1 M sodium phosphate buffer, pH 6.3, used for removal of pig- control. ◦ ments. The whole purification procedure was carried out at 4 C. The unbound Effect of high molecular weight (proteinaceus) inhibitors: baupain (0.23 ␮M) fractions containing enzyme activity were separated and subsequently purified by was pre-incubated for 10 min in assay buffer with the serine proteinase inhibitors HPLC on a ␮-Bondapak C18 reverse phase column. The separation was achieved by an SbTI (Soy beans trypsin inhibitor) [38] 1.39, 2.78, 5.56, 8.35 ␮M; EcTI (Enterolobium acetonitrile gradient (0–100%) in 0.1% TFA (v/v) during 75 min, at 1 ml/min flow rate contortisiliquum trypsin inhibitor) [39] 0.2, 0.7, 1.50, 2.0, 2.7 ␮M and with the cys- and room temperature. Proteins were estimated spectrophotometrically (A280) as teine proteinase inhibitors HMWK (High Molecular Weight Kininogen) 0.08, 0.16, well as by Bradford [32] assay using bovine serum albumin as the standard. 0.24, and 0.32 ␮M; canecystatin [5] 0.06, 0.12, 0.18, 0.24 and 0.34 nM and BbCI (Bauhinia bauhinioides cruzain inhibitor) [40] 1.2, 1.8, 2.4, 3.6, and 4.8 ␮M. Enzyme 2.2. N-terminal amino acid sequence activity was expressed in percent of residual activity on Z–Phe–Arg–MCA compared to the control. Purified baupain was denatured and reduced by addition of 200 ␮l50mM Tris/HCl buffer, pH 8.5, containing 6.0 M guadinium HCl, 1.0 mM EDTA, and 5.0 mM 2.8. Determination of baupain substrate specificity dithiothreitol for 3 h, at 37 ◦C. S-pyridylethylation of cysteines was achieved by addi- ◦ tion of (5 ␮l) 4-vinylpyrimidine for 3 h at 37 C in the dark and nitrogen atmosphere. The hydrolysis of two series of FRET peptides derived from Abz-KLRSSK-Q- The excess reagents were removed by reversed-phase HPLC on a C18 column using EDDnp (Q-EDDnp is the fluorescence acceptor and Abz is the fluorescence donor the same gradient conditions already described. N-terminal amino acid sequences that corresponds to glutamine-[N-(2,4-dinitrophenyl)-ethylenediamine] and ortho- were determined by Edman degradation using a PPSQ-23 Model Protein Sequencer aminobenzoic acid, respectively), in which the residues L and R and S were (Shimadzu, Tokyo, Japan). Phenylthiohydantoin derivates of amino acids were iden- substituted by natural amino acids were quantified in a Hitachi F-2500 spectroflu- tified. orimeter at 37 ◦C. Baupain concentration was fixed as 11 nM and the substrates

as 4 ␮M. Fluorescence changes were monitored continuously at ex = 320 nm and 2.3. Electrophoresis and gelatin zymography em = 420 nm. The enzyme concentrations were chosen so that less than 5% of the substrate was hydrolyzed over the course of the assay. The reaction rate was con- The homogeneity and the molecular weight of baupain were assessed by SDS- verted into nanomoles of substrate hydrolyzed per second based on a calibration PAGE under reducing and non-reducing conditions according to Laemmli [33], curve obtained from the complete hydrolysis of each peptide. The scissile bond of using 12% acrylamide gel. HMWK (10 ␮g) limited proteolysis cleavage by baupain hydrolyzed peptides were identified by isolation of the fragments using analyti- (0.23 ␮M) was demonstrated by SDS-polyacrylamide gel electrophoresis [34]. Bau- cal HPLC followed by determination of their molecular mass by LC/MS using an pain and HMWK were incubated in 0.1 M sodium phosphate buffer, pH 6.3 for 10 LCMS-2010 equipped with an ESI-probe (Shimadzu, Japan). and 60 min at 37 ◦C. The proteins were stained with Coomassie brilliant blue R-250. A broad range of molecular weight protein markers, from 25 to 175 kDa and 20 to 2.9. CD experiments 94 kDa New England BioLabs Inc. (Ipswich, MA, USA), was used. The zymography experiment was performance under non-reducing conditions Circular Dichroism (CD) measurements were taken on a Jasco J-810 spectropo- according to Becker et al. [35]. Baupain activity was detected using 10% (w/v) acry- larimeter (Jasco, Japan). Far UV-CD spectra were recorded at 25 ◦C in a cuvette of lamide (Gelatin-PAGE), with 0.04% (w/v) copolymerized gelatin included in the gel 1 mm pathlength with a 10 ␮M protein solution in the presence of 10 mM sodium as substrate for the proteinase. After electrophoresis, the gels were incubated in phosphate buffer, pH 6.3. The spectra were recorded in the 190–250 nm wave- renaturing buffer for 1 h at room temperature, followed with incubation in devel- length range. The CD intensities were expressed as mean residue ellipiticities [] ◦ ® 2 −1 oping buffer at 37 C overnight. We used Mini-Protean II Cell Bio-Rad (Hercules, (deg cm dmol ) using the formula []=e/10.C.l.N, where e is the experimental CA, USA). ellipticity in millidegrees, MRW is the mean residue weight, C is the concentration of the protein in molar, l is the cuvette pathlength in centimeters, and N being the 2.4. Enzyme activity average number of residues adopted as 110 residues baupain. For the analysis of baupain CD spectrum the CDPro program was used. CDPro software package con- Proteinase activity was measured on Z–Phe–Arg–MCA (Calbiochem Ltda, Darm- sists of three programs for analyzing the protein CD spectra for determining the stadt, Germany) and Bz-Arg-pNan (BAPA) (Sigma–Aldrich Company, St. Louis, USA) secondary structure fractions (SELCON3, CDSSTR and CONTIN) and a program for as substrates. Baupain was incubated at 37 ◦C in a microtiter plate in 250 ␮l final determining tertiary structure class (CLUSTER) [41,42]. The estimation of baupain volume of assay buffer [0.1 M sodium phosphate buffer, pH 6.3 containing 0.4 M secondary structure was performance using 43 proteins in the reference set. NaCl, 10 mM EDTA, and 2.0 mM DTT (dithiothreitol)]. The reaction was followed for 10–30 min and the reaction was stopped by the addition of 50 ␮l acetic acid 2.10. Fluorescence experiments 30% (v/v). The fluorescence release was measured on a FluoroCount PackardTM, spectrofluorometer set at 355 nm for excitation and 460 nm for emission. Steady-state fluorescence was recorded on Hitachi F-2500 spectrofluorimeter. In the case of BAPA (0.8 mM) as substrate, the reaction was followed by mea- Intrinsic tryptophan fluorescence spectrum was recorded by exciting the protein suring the absorbance of released p-nitroaniline at 405 nm in a spectrophotometer sample at 290 nm at pH 6.3, 10 mM sodium phosphate buffer. The emission spectrum PackardTM with a 50 mM Tris/HCl buffer, pH 8.0 [36]. was recorded in the range of 305–450 nm with excitation and emission slit widths 574 S.S. Andrade et al. / Process Biochemistry 46 (2011) 572–578

Fig. 1. Size exclusion chromatography on Sephadex G-25 column of B. forfi- cate leaves extract. Eluting buffer: 0.1 M sodium phosphate buffer, pH 6.3. Flow rate: 16 ml/h. The arrow indicates activity on Z–Phe–Arg–MCA (0.4 mM). Insert: Canecystatin-Sepharose column (2 mL). Sample: 0.4 mg of baupain from Sephadex G-25 column. (A) Unbound fractions were eluted with 0.1 M sodium phosphate buffer, pH 6.3 buffer. (B) The bound fraction (16) with activity on Z–Phe–Arg–MCA substrate was eluted with 0.1 M sodium phosphate buffer, pH 6.3 containing 0.001 M l -cysteine. Fig. 2. (A) SDS-PAGE (12%) of the cysteine proteinase baupain from Bauhinia forficata leaves. (1) Baupain (10 ␮g) from Con A-Sepharose under reducing conditions, Coomassie blue staining; (S), standard proteins, ␤-lactoglobulin A set a 5.0 and 2.5 nm, respectively. Baseline correction was carried out with buffer (25 kDa), triosephosphate isomerase (32 kDA), aldolase (47 kDa, glutamic dehydro- without protein. genase (62 kDa), MBP-paramyosin (83 kDa) and MBP-␤-galactosidase (175 kDa). (B) Gelatin-PAGE proteinase activity analysis of baupain 10 ␮g (1), and 20 ␮g (2). 2.11. Radioimmunoassay Clear band with dark background indicate sites of protein degradation in 10% acrylamide and 0.04% gelatin in non-reducing conditions and incubated in 0.1 M Baupain bradykinin release ability was assayed with HMWK [43]. The enzyme, sodium phosphate, pH 6.3 for 24 h. (S) standard proteins, ␤-lactoglobulin A (25 kDa), in concentrations of 0.019, 0.038, and 0.057 A280 in 0.1 M sodium phosphate, pH 6.3 triosephosphate isomerase (32 kDA), aldolase (47 kDa, glutamic dehydrogenase containing 10 mM EDTA, 0.4 M NaCl, and 2.0 mM DTT was incubated with HMWK (62 kDa), and MBP-paramyosin (83 kDa). (1.4 ␮M) in 40 ␮l for 2 h at 37 ◦C. Formed kinin was extracted in ethanol (1:4 v/v) for 10 min at −70 ◦C. The solution was concentrated and dissolved in 200 ␮l of 0.01 M phosphate buffer, pH 7.0, NaCl 0.14 M, NaN3, 0.03 M EDTA, 0.003 M 1,10 phenan- activity of the endopeptidase was increased more than 79-fold by throline and ovalbumin 0.1%. 50 ␮l of the sample was incubated with 100 ␮l anti-BK this isolation procedure and both SDS and non-denaturing PAGE of antibody [44] (1: 80.000) and 100 ␮l tyrosyl-bradykinin probe with 125I, for 20 h at the final preparation stained with Coomassie brilliant blue showed ◦ ␮ 4 C, 400 l the BSA (2 mg/ml) 0.01 M phosphate buffer, pH 7.0, 0.1% NaCl 0.14 M, a homogeneous preparation (data not shown). The purification NaN3 0.1% and 800 ␮l with polyethylene glycol 6000 25% solution were added to the samples and incubated by 10 min at 4 ◦C. Samples were centrifuged at 2000 × g steps are summarized in Table 1. for 20 min at 4 ◦C. The solution obtained was removed and the pellet was submitted The molecular weight of the purified proteinase was determined to the radiation counting and the bradykinin released was calculated. by SDS-PAGE as 30 kDa (Fig. 2). The same result was obtained by size exclusion chromatography on calibrated Superdex-200 col- 3. Results and discussion umn, where the proteolytic activity was found essentially in the main peak eluted with a volume that corresponds to a molecular 3.1. Proteinase extraction and purification mass of 33 ± 4 kDa (data not shown).

The protocol used for isolation and purification of the endopro- 3.2. N-terminal sequence teinase from B. forficata consists of three steps. Freshly prepared leaves saline extracts were treated with 80% acetone in order to Reverse phase chromatography in a C18 column was per- precipitate the protein content. The precipitate containing prote- formed with acetonitrile gradient, and the N-terminal sequences olytic activity was submitted to a size exclusion chromatography of reduced and pyridylethylated inhibitors determined by auto- in a Sephadex G-25 column (Fig. 1) and the fractions with enzyme mated Edman degradation allowed the identification of the first activity on Z–Phe–Arg–MCA were polled and concentrated by Ami- 10 amino acid residues (IPEYVDWRQQ). Using NCBI database, PSI- con ultrafiltration. BLAST [47] baupain shows similarity to those of the CA family of The strategy to apply affinity chromatography with thiopropyl- cysteine proteinases, being 90% and 40% identical to papain and Sepharose have been successfully used in the cysteine proteinase bromelain, respectively. purification since this resin presents strong binding 2-pyridyl disulfide [45,46]. In our case, the procedure resulted in low yield of 3.3. Effect of metal ions, selective inhibitors and sulfhydryl purification (data not shown), probably due to the pH 6.3 used reagents on proteinase activity where the interaction to the activated group is not efficient. Further purification was obtained using a canecystatin-Sepharose column The inhibition of enzyme activity was negligible by low (Fig. 1, insert) being the enzyme activity detected in the broad molecular weight serine proteinase inhibitors PMSF, TPLC, TLCK, peak eluted with 1.0 mM l-cysteine. The peak showing activity benzamidine, metallo proteinase inhibitors EDTA and 1,10- on Z–Phe–Arg–MCA was polled, concentrated and submitted to phenanthroline and aspartic proteinase inhibitor pepstatin A. a Concanavalin A-Sepharose affinity chromatography to complete However, the activity was completely inhibited by the sulfhydryl removal of leave pigments. Proteolytic activity was detected only modifying reagent iodoacetamide and by the papain family-specific in the unbound fraction, depleted of leave pigments. The specific inhibitor E-64. Since the reducing agents (2-mercaptoethanol, DTT S.S. Andrade et al. / Process Biochemistry 46 (2011) 572–578 575

Table 1 Puriﬁcation of cysteine proteinase from B. forﬁcata.

Steps Volume (mL) Protein (mg mL) UA Speciﬁc activity (UA mg) Purif. Yield (%)

Saline extraction (50 g) 700 – – – – – Acetone fractionation (80%) 30 1.0 – – – – Sephadex G-25 5 0.32 0.2 0.6 1.0 100 Canecystatin-Sepharose 2.2 0.48 0.4 0.8 2.0 88 Con A-Sepharose 1.6 0.43 0.5 1.2 2.5 79

Enzyme activity was determined on Z–Phe–Arg–MCA 0.4 mM as substrate. UA – ␮mols of Z–Phe–Arg–MCA hydrolyzed/min mL. Protein determinated by Bradford [32] assay.

(dithiothreitol) and l-cysteine) enhanced the hydrolytic activity of 3–4.5 and 9–10 the enzyme became irreversibly inactivated (data baupain (data not shown), all further assays where performed with not shown). of 2.0 mM DTT. The pH-profile of the hydrolytic activity of baupain on Z–Phe–Arg–MCA was measured in the pH range from 4 to 10 3.4. Circular dicroism spectra (Fig. 4A). The baupain maximum activity was achieved at pH 6.0 and very low activity observed at extreme acid and basic pHs sim- The tertiary structure of the enzyme was assessed by far UV ilar to that obtained with papain like cysteine proteinases [4],we circular dichroism. The CD spectrum (Fig. 3) shows two negative used the following assay buffer for further kinetic studies: 0.1 M CD bands at 222 nm and 208–210 nm and one positive band near sodium phosphate buffer, pH 6.3 containing 10 mM EDTA, 0.4 M 190 nm. The cluster analysis showed that the peptidase belongs to NaCl, and 2.0 mM DTT. the ␣ + ␤ class of proteins [48] and the percentages of secondary structure, calculated as 44% ␣-helix, 16% ␤-sheet and 12% ␤-turn. 3.6. Substrate specificity of baupain using FRET substrates These data are consistent with the reported data on papain like cysteine proteinases [2]. Seven residues intramolecularly quenched fluorogenic Abz- The intrinsic fluorescence spectra of the proteinase was peptidyl-EDDnp substrates, containing systematic substitutions at obtained at pH 7.0 and the maximum of fluorescence intensity the carboxyl-side of the scissile bond, were used to evaluate the was observed at 343 nm, a value characteristic of solvent accessible contribution of these positions in baupain substrates hydrolytic tryptophan residues. This data is in agreement with the fluores- rate. Two series of peptides were generated with variations at P2 cence emission maximum value of 345 nm obtained with papain and P1 positions to explore the specificity at the S2 and S1 subsites (data not shown) [49,50]. All together, these results suggest that that usually define substrate specificity among papain-like cysteine the enzyme belongs to the group of cysteine proteinases of the clan proteinases [51]. CA of the papain family, and the name baupain was proposed for this new endopeptidase. 3.6.1. Hydrolysis of Abz-KLXSSKQ-EDDnp series, for mapping S1 specificity 3.5. Proteolytic activity characterization Fig. 5A shows the baupain relative hydrolysis of the peptide series Abz-KLXSSKQ-EDDnp with different amino acids at X posi- The baupain hydrolytic activity remains after preincubation of tion. All substrates in this series were cleaved only at the X-Ser the enzyme at different pH values in the range of 4.5–9 for 30- bond, indicating that X occupied the P1 position. Baupain did not ◦ minat37 C. In the same experiments performed in the pH range present a defined specificity over the assayed peptides, being the peptide containing Gly at P1 hydrolyzed at the same rate as the one containing Phe and Arg. Deviations were observed only with the peptides containing negatively charged residues (Glu and Asp) 40 and the bulky Trp. This apparent lack of specificity at S1 contrasts )

-1 20 .dmol 2

0 deg.cm 3 ] (10 θ [ -20

-40

190 200 210 220 230 240 250 Wavelength (nm) Fig. 4. pH profile for baupain hydrolytic activity. Initial activity of hydrolysis of Fig. 3. CD spectrum of baupain (0.05 mg/ml), in 0.001 M sodium phosphate buffer, Z–Phe–Arg–MCA was carried out as described in Section 2. The relative values were pH 7.0. The spectrum was recorded over the range 190–250 nm, at 25 ◦C, in 1 mm obtained assessing the highest value as 100%. The curve drawn through the exper- cell pathlength. CD Pro program was used for estimation of secondary structure of imental points was obtained from fitting to the appropriated equation using the baupain. The calculated fractions were 44% ␣-helix, 16% ␤-sheet and 12% ␤-turn. Grafit 5.0 software (Erithacus Software, Orley, Surrey, U.K.). 576 S.S. Andrade et al. / Process Biochemistry 46 (2011) 572–578

2200 2200 2000 ABC 2000 1800 ) 1800 ) -1 1600 -1 1600 1400 1400 1200 1200 1000 1000 800 800 600 600 Activity (nM.s 400 Activity (nM.s 400 200 200 0 0 Z-FR-MCAZ-LR-MCA R* K D E G I F W RKEGAL*VMF YW Abz-KLXSSKQ-EDDnp Abz-KXRSSKQ-EDDnp

Fig. 5. (A) Baupain S1 substrate specificity. All substrates in this series were cleaved only at the X-Ser bonds, indicating that X occupied the P1 position in this series. Assay conditions are as described in Section 2. (B) Baupain S2 substrate specificity. All substrates of this series were exclusively cleaved by the enzyme at the R–S bond with X occupying the P2 position. (C) Baupain activity on Z–Phe–Arg–MCA and Z-Leu-Arg-MCA is included for comparison. Assay conditions are as described in Section 2. with that of papain, bromelain and human cathepsin L that show a As observed with papain [54], a strong inhibition of baupain × −8 preference for basic amino acids in P1 [29,51]. (Kiapp = 1.9 10 M) was also obtained with human high molecular weight kininogen (HMWK) (Fig. 6), a plasma multifunctional 3.6.2. Hydrolysis of Abz-KXRSSKQ-EDDnp series, for mapping S2 glycoprotein that besides their role as precursor of the vasoactive specificity peptides kinin, posses the sequence motif QXVXG in cysteine pro- The relative hydrolysis of the peptide series Abz-KXRSSKQ- teinase inhibitor domains [55] that is reported to inhibit cysteine EDDnp is shown in Fig. 5B. All peptides from this series, under the proteinases papain family [4]. Is worth to point out to the fact that assay conditions used, were exclusively cleaved by the enzyme at canecystatin, a baupain inhibitor also encompass the conservative the Arg-Ser bond with X occupying P2 position. The best hydrolyzed QXVXG motif in contrast to BbCI which does not affect the enzyme peptides were those containing Tyr or Phe at P2 position, whereas activity. those containing positively or negatively charged amino acids (Arg, Lys, His and Asp) as well as polar groups (Ser, Thr) were poorly 3.8. Baupain kininogenase activity hydrolyzed. It is interesting to observe that FRET substrates having aliphatic amino acids Val and Leu were hydrolyzed with 50% and The kininogenase activity of some plant cysteine proteinases 30% of the efficiency obtained with the larger aromatic residues. (papain, ficin and bromelain) on human plasma was long ago This preference was corroborated by higher hydrolytic efficiency reported by Prado [56], more recently this characteristic was of baupain over Z–Phe–Arg–MCA, in comparison to Z-Leu-Arg-MCA reported for cathepsin L but not for cathepsin B. The kininogenase (Fig. 5C). In this sense, baupain specificity differs from papain, that activity of cathepsin L was reinforce by the release of bradykinin prefers Leu and Val at P2 and from bromelain, that accepts Arg at from HMWK [57] and also from synthetic bradykinin-containing this position [51] but is similar to human cathepsin L and V pro- fragment of kininogen [58]. This property was also detected for teinases (as well as cruzain and L. mexicana cathepsin L) that also baupain. The baupain kininogenase activity was observed using prefers aromatic to aliphatic amino acids interacting at S2 subsite stoichiometric amounts of enzyme and kininogen. The kinin release [2,29,51]. was quantified by radioimmunoassay (Fig. 7A). This activity was abolished by the addition of the irreversible cysteine proteinase 3.7. Effect of proteinaceus peptidase inhibitors on baupain inhibitor E-64, demonstrating the specificity of assay. Baupain activity cleavage of HMWK generates one major band of 45 kDa and that

Inhibitors from the Bowman–Birk and plant Kunitz-type family have been characterized by proteinase specificity, primary structure and reactive site residues [52]. Since they present differences in 1 -8 proteinase inhibition profiles, they are valuable tolls for proteolytic Kiapp = 1.9 x 10 M enzyme characterization. In this sense, the distinctive substrate 0.8 preference observed with baupain was further studied accessing the effect of both plant and mammals proteinaceous peptidase inhibitors on its hydrolytic activity. 0.6 Plant Kunitz type inhibitors STI (Soybean trypsin inhibitor) and EcTI (Enterolobium contortisiliquum trypsin inhibitor), as well as the cysteine proteinase inhibitor isolated from Bauhinia bauhin- 0.4 ioides (BbCI) do not inhibit baupain at the assayed concentrations ␮ Residual Activity (%) (0.005–0.5 M). While the absence of inhibition observed with 0.2 both trypsin like inhibitors would be a expected feature due to baupain cysteine proteinase nature, it is interesting to observe that BbCI does inhibits cathepsin L, cruzain and cruzipain, but it 0 0 0.1 0.2 0.3 0.4 also fails to inhibit papain [40], sustaining a close relationship between both enzymes. Canecystatin, a recombinant cysteine pro- HMWK (uM) teinase inhibitor from sugar cane, efficiently inhibits baupain (Kiapp 10−8 M) as well as papain (K = 3.3 × 10−9 M) and cathepsin L Fig. 6. Inhibition effect of HMWK on bapain activity. Effect of increasing concentra- iapp tions of HMWK on proteolytic activity towards Z-Phe-Arg-MCA (0.4 mM) of baupain × −9 (Kiapp = 0.6 10 M) but no inhibitory effect was observed on ficin in 0.1 M sodium phosphate buffer, pH 6.3 containing 10 mM EDTA, 0.4 M NaCl, and or bromelain [53]. 2.0 mM DTT, and were pre-incubated for 20 min at 37 ◦C. S.S. Andrade et al. / Process Biochemistry 46 (2011) 572–578 577

Fig. 7. (A) Effect of baupain on bradykinin release. Bradykinin levels were determined by radioimmunoassay [44] in triplicate determinatives. (B) Degradation of HMWK (10 ␮g) by baupain (0.23 ␮M) demonstrated by SDS-PAGE 12%. (S) Standard range of molecular weight protein markers 20–94 kDa. (A) Control HMWK in 0.1 M sodium phosphate buffer, pH 6.3, containing l-cysteine 2 mM. (B) Control HMWK in 0.1 M sodium phosphate buffer, pH 6.3. (C) Baupain and HMWK were incubated for 10 min at 37 ◦C. (D) Baupain and HMWK incubated for 60 min at 37 ◦C. (E) DTT reduced HMWK. hydrolysis profile is not modified in the period of 10 at 60 min of proteinase also shares with cathepsin L and papain the capacity of incubation (Fig. 7B), probably by the impairment of the enzyme releasing bradykinin from HMWK. activity by the HMWK cysteine proteinase inhibitor domains. It is In conclusion, this work point out the kininogenase activity worth to emphasize that this impairment was not observed using specificity profile of baupain that is distinct from other cysteine other proteinaceous substrate such as insulin B-chain and the ser- proteinases and similar to cathepsin L [57]. We believe that this ine proteinase inhibitor isolated from the seeds of Bauhinia forficata property may be relevant to regroup cluster papain-like cysteine (BfTI) (data not shown). As baupain digests proteins effectively and proteinases. rapidly yielding numerous proteolytic fragments, suggesting that baupain is involved in endogenous protein degradation. Acknowledgments Bradykinin excised from HMWK requires the cleavage of the ...MK-R... and ...FR-S... bonds at the N- and C-terminal This work was supported in Brazil by Fundação de Amparo à bradykinin insertion sites, respectively. In this sense, while bau- Pesquisa do Estado de São Paulo (FAPESP) and Conselho Nacional pain specificity data support both cleavages (Phe and Met were de Desenvolvimento Científico e Tecnológico (CNPq). accepted at P2; Lys and Arg at P1; Ser at Pˇı1 – Fig. 5), the kininogenase activity data indicate that baupain also accepts Arg at Pˇı1, as observed with cathepsin L [59]. References These results provide structural and biochemical information on the cysteine proteinase isolated from the leaves of B. forficata [1] Grudkowska M, Zagdanska B. Multifunctional role of plant cysteine proteinases. Acta Biochim Pol 2004;51:609–24. which is utilized in popular therapy for diabetes mellitus. Bau- [2] Barrett AJ. Handbook of proteolytic enzymes. London; 2004. pain is similar to other cathepsins and closer to cathepsin L shown [3] Sajid M, McKerrow JH. Cysteine proteases of parasitic organisms. Mol Biochem to be essential for the development of type I diabetes in non- Parasitol 2002;120:1–21. [4] Grzonka Z, Jankowska E, Kasprzykowski F, Kasprzykowska R, Lankiewicz L, obese diabetic mice [60]. Indeed, it is important to consider the Wiczk W, et al. Structural studies of cysteine proteases and their inhibitors. kininogenase activity of baupain that releases bradykinin (BK) from Acta Biochim Pol 2001;48:1–20. HMWK. Bradykinin modulates the release of insulin in vivo [61,62] [5] Soares-Costa A, Beltramini LM, Thiemann OH, Henrique-Silva F. A sugarcane cystat recombinant expression purification antifungal activity. Biochem Bio- and the stimulation of insulin secretion by beta cells occurs through phys Res Commun 2002;296:1194–9. activation of B2 receptor [63,64]. These results were confirmed [6] Ueda T, Seo S, Ohashi Y, Hashimoto J. Circadian and senescence-enhanced using kininogen-deficient rats in which the release of insulin stim- expression of a tobacco cysteine protease gene. Plant Mol Biol 2000;44:649–57. [7] Gepstein S, Sabehi G, Carp MJ, Hajouj T, Nesher MF, Yariv I, et al. Large-scale ulated by glucose administration was lower than in normal rats identification of leaf senescence-associated genes. Plant J 2003;36:629–42. [65–68]. Our results suggest that the kininogenase activity of bau- [8] Prins A, van Heerden PD, Olmos E, Kunert KJ, Foyer CH. Cysteine proteinases pain may be involved in the hypoglycemic property of Bauhinia regulate chloroplast protein content and composition in tobacco leaves: forficata leaves. However, further studies are necessary to demon- a model for dynamic interactions with ribulose-1,5-bisphosphate carboxy- lase/oxygenase (Rubisco) vesicular bodies. J Exp Bot 2008;59:1935–50. strate the effectiveness of baupain in diabetes. [9] Eason JR, Ryan DJ, Watson LM, Hedderley D, Christey MC, Braun RH, et al. Sup- pression of the cysteine protease, aleurain, delays floret senescence in Brassica oleracea. Plant Mol Biol 2005;57:645–57. 4. Conclusion [10] Kardailsky IV, Brewin NJ. Expression of cysteine protease genes in pea nodule development and senescence. Mol Plant Microbe Interact 1996;9:689–95. [11] Callis J. Regulation of protein degradation. Plant Cell 1995;7:845–57. By acetone precipitation and different chromatographic steps an [12] Brzin J, Kidric M. Proteinases and their inhibitors in plants: role in normal endoproteinase named baupain detected in the B. forficata leaves growth and in response to various stress conditions. Biotechnol Genet Eng Rev 1995;13:421–67. was purified to homogeneity. While the N-terminal sequence sim- [13] Ling JQ, Kojima T, Shiraiwa M, Takahara H. Cloning of two cysteine proteinase ilarity, molecular mass, circular dicroism spectra and intrinsic genes CysP1 and CysP2, from soybean cotyledons by cDNA representational fluorescences profiles points to a close structural relationship to difference analysis. Biochim Biophys Acta 2003;1627:129–39. [14] Buchanan-Wollaston V, Ainsworth C. Leaf senescence in Brassica napus: cloning papain, its activity on FRET and MCA peptides indicates substrate of senescence related genes by subtractive hybridisation. Plant Mol Biol specificity more related to the mammalian cathepsin L enzyme. The 1997;33:821–34. 578 S.S. Andrade et al. / Process Biochemistry 46 (2011) 572–578

[15] Guerrero C, de la Calle M, Reid MS, Valpuesta V. Analysis of the expression of two [40] de Oliveira C, Santana LA, Carmona AK, Cezari MH, Sampaio MU, Sampaio CA, thiolprotease genes from daylily (Hemerocallis spp.) during flower senescence. et al. Structure of cruzipain/cruzain inhibitors isolated from Bauhinia bauhin- Plant Mol Biol 1998;36:565–71. ioides seeds. Biol Chem 2001;382:847–52. [16] Xu FX, Chye ML. Expression of cysteine proteinase during developmental events [41] Sreerama N, Woody RW. Estimation of protein secondary structure from cir- associated with programmed cell death in brinjal. Plant J 1999;17:321–7. cular dichroism spectra: comparison of CONTIN SELCON, and CDSSTR methods [17] Halangk W, Lerch MM, Brandt-Nedelev B, Roth W, Ruthenbuerger M, Rein- with an expanded reference set. Anal Biochem 2000;287:252–60. heckel T, et al. Role of cathepsin B in intracellular trypsinogen activation and [42] Sreerama N, Venyaminov SY, Woody RW. Analysis of protein circular dichro- the onset of acute pancreatitis. J Clin Invest 2000;106:773–81. ism spectra based on the tertiary structure classification. Anal Biochem [18] Toyooka K, Okamoto T, Minamikawa T. Mass transport of proform of a KDEL- 2001;299:271–4. tailed cysteine proteinase (SH-EP) to protein storage vacuoles by endoplasmic [43] Gozzo AJ, Nunes VA, Carmona AK, Nader HB, von Dietrich CP, Silveira VL, et al. reticulum-derived vesicle is involved in protein mobilization in germinating Glycosaminoglycans affect the action of human plasma kallikrein on kininogen seeds. J Cell Biol 2000;148:453–64. hydrolysis and inflammation. Int Immunopharmacol 2002;2:1861–5. [19] Naito Y, Fujie M, Usami S, Murooka Y, Yamada T. The involvement of [44] Shimamoto K, Ando T, Tanaka S, Nakahashi Y, Nishitani T, Hosoda S, et al. a cysteine proteinase in the nodule development in Chinese milk vetch An improved method for the determination of human blood kinin levels by infected with Mesorhizobium huakuii subsp. rengei. Plant Physiol 2000;124: sensitive kinin radioimmunoassay. Endocrinol Jpn 1982;29:487–94. 1087–96. [45] Popovic T, Puizdar V, Ritonja A, Brzin J. Simultaneous isolation of human kidney [20] Harrak H, Azelmat S, Baker EN, Tabaeizadeh Z. Isolation and characterization of cathepsins B, H, L and C and their characterisation. J Chromatogr B Biomed Appl a gene encoding a drought-induced cysteine protease in tomato (Lycopersicon 1996;681:251–62. esculentum). Genome 2001;44:368–74. [46] Popovic T, Kidric M, Puizdar V, Brzin J. Purification and characterization of [21] Solomon M, Belenghi B, Delledonne M, Menachem E, Levine A. The involve- two cysteine proteinases from Phaseolus vulgaris leaves. Plant Physiol Biochem ment of cysteine proteases and protease inhibitor genes in the regulation of 1998;36:637–45. programmed cell death in plants. Plant Cell 1999;11:431–44. [47] Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, et al. PSI- [22] Xu ZF, Chye ML, Li HY, Xu FX, Yao KM. G-box binding coincides with increased BLAST: a new generation of protein database search programs. Nucleic Acids Solanum melongena cysteine proteinase expression in senescent fruits and Res 1997;25:3389–402. circadian-regulated leaves. Plant Mol Biol 2003;51:9–19. [48] Venyaminov SY, Yang JT. Determination of protein secondary conformation. [23] Khalil NM, Pepato MT, Brunetti IL. Free radical scavenging profile and myeloper- In: Fasman GD, editor. Circular dichroism and the conformational analysis of oxidase inhibition of extracts from antidiabetic plants: Bauhinia forficata and biomolecules. New York/London: P Press; 1996. p. 69–107. Cissus sicyoides. Biol Res 2008;41:165–71. [49] Kamphuis IG, Kalk KH, Swarte MB, Drenth J. Structure of papain refined at 1.65 [24] da Cunha AM, Menon S, Menon R, Couto AG, Burger C, Biavatti MW. Hypo- A resolution. J Mol Biol 1984;179:233–56. glycemic activity of dried extracts of Bauhinia forficata Link. Phytomedicine [50] Levitt M, Chothia C. Structural patterns in globular proteins. Nature 2010;17:37–41. 1976;261:552–8. [25] Oliveira CZ, Maiorano VA, Marcussi S, Sant’ana CD, Januario AH, Lourenco [51] Choe Y, Leonetti F, Greenbaum DC, Lecaille F, Bogyo M, Bromme D, et al. Sub- MV, et al. Anticoagulant and antifibrinogenolytic properties of the aque- strate profiling of cysteine proteases using a combinatorial peptide library ous extract from Bauhinia forficata against snake venoms. J Ethnopharmacol identifies functionally unique specificities. J Biol Chem 2006;281:12824–32. 2005;98:213–6. [52] Oliva ML, Sampaio MU. Action of plant proteinase inhibitors on enzymes of [26] Lim H, Kim MK, Lim Y, Cho YH, Lee CH. Inhibition of cell-cycle progression in physiopathological importance. An Acad Bras Cienc 2009;81:615–21. HeLa cells by HY52, a novel cyclin-dependent kinase inhibitor isolated from [53] Oliva ML, Carmona AK, Andrade SS, Cotrin SS, Soares-Costa A, Henrique-Silva Bauhinia forficata. Cancer Lett 2006;233:89–97. F. Inhibitory selectivity of canecystatin: a recombinant cysteine peptidase [27] Nery ED, Juliano MA, Meldal M, Svendsen I, Scharfstein J, Walmsley A, et al. inhibitor from sugarcane. Biochem Biophys Res Commun 2004;320:1082–6. Characterization of the substrate specificity of the major cysteine protease [54] Sueyoshi T, Hara A, Shimada T, Kimura M, Morita T, Kato H, et al. Molecu- (cruzipain) from Trypanosoma cruzi using a portion-mixing combinatorial lar interaction of bovine kininogen and its derivatives with papain. J Biochem library and fluorogenic peptides. Biochem J 1997;323(Pt 2):427–33. 1988;104:200–6. [28] St Hilaire PM, Alves LC, Sanderson SJ, Mottram JC, Juliano MA, Juliano L, et al. The [55] Colman RW, Schmaier AH. Contact system: a vascular biology modulator with substrate specificity of a recombinant cysteine protease from Leishmania mex- anticoagulant, profibrinolytic, antiadhesive, and proinflammatory attributes. icana: application of a combinatorial peptide library approach. Chembiochem Blood 1997;90:3819–43. 2000;1:115–22. [56] Prado JL, Stella RC, Prado ES. Kinin liberation from equine plasma by acidifica- [29] Puzer L, Cotrin SS, Alves MF, Egborge T, Araujo MS, Juliano MA, et al. Com- tion and neutralization. An Acad Bras Cienc 1970;42:783–91. parative substrate specificity analysis of recombinant human cathepsin V and [57] Desmazes C, Gauthier F, Lalmanach G, Cathepsin L. but not cathepsin B, is a cathepsin L. Arch Biochem Biophys 2004;430:274–83. potential kininogenase. Biol Chem 2001;382:811–5. [30] Angelo PF, Lima AR, Alves FM, Blaber SI, Scarisbrick IA, Blaber M, et al. Substrate [58] Puzer L, Vercesi J, Alves MF, Barros NM, Araujo MS, Aparecida Juliano M, et al. A specificity of human kallikrein 6: salt and glycosaminoglycan activation effects. possible alternative mechanism of kinin generation in vivo by cathepsin L. Biol J Biol Chem 2006;281:3116–26. Chem 2005;386:699–704. [31] Okamoto DN, Kondo MY, Hiraga K, Juliano MA, Juliano L, Gouvea IE, et al. Salt [59] Puzer L, Cotrin SS, Alves MF, Egborge T, Araújo MS, Juliano MA, et al. Com- effect on substrate specificity of a subtilisin-like halophilic protease. Protein parative substrate specificity analysis of recombinant human cathepsin V and Pept Lett 2010;17:796–802. cathepsin L. Arch Biochem Biophys 2004;430:274–83. [32] Bradford MM. A rapid and sensitive method for the quantitation of micro- [60] Hsing LC, Kirk EA, McMillen TS, Hsiao SH, Caldwell M, Houston B, et al. Roles for gram quantities of protein utilizing the principle of protein–dye binding. Anal cathepsins S, L, and B in insulitis and diabetes in the NOD mouse. J Autoimmun Biochem 1976;72:248–54. 2010;34:96–104. [33] Laemmli UK. Cleavage of structural proteins during the assembly of the head [61] Yang C, Hsu WH. Stimulatory effect of bradykinin on insulin release from the of bacteriophage T4. Nature 1970;227:680–5. perfused rat pancreas. Am J Physiol 1995;268:1027–30. [34] Schiffman S, Mannhalter C, Tyner KD. Human high molecular weight kininogen. [62] Yang C, Hsu WH. Glucose-dependency of bradykinin-induced insulin secretion Effects of cleavage by kallikrein on protein structure and procoagulant activity. from the perfused rat pancreas. Regul Pept 1997;71:23–8. J Biol Chem 1980;255:6433–8. [63] Saito Y, Kato M, Kubohara Y, Kobayashi I, Tatemoto K. Bradykinin increases [35] Becker C, Senyuk VI, Shutov AD, Nong VH, Fischer J, Horstmann C, et al. Pro- intracellular free Ca2+ concentration and promotes insulin secretion in the teinase A, a storage-globulin-degrading endopeptidase of vetch (Vicia sativa clonal beta-cell line HIT-T15. Biochem Biophys Res Commun 1996;221:577–80. L.) seeds, is not involved in early steps of storage-protein mobilization. Eur J [64] Yang C, Chao J, Hsu WH. The effect of bradykinin on secretion of insulin, Biochem 1997;248:304–12. glucagon, and somatostatin from the perfused rat pancreas. Metabolism [36] Oliva ML, Sallai RC, Sampaio CA, Fritz H, Auerswald EA, Tanaka AS, et al. Bauhinia 1997;46:1113–5. serine proteinase inhibitors: effect on factor X, factor XII and plasma kallikrein. [65] Bhoola KD, Figueroa CD, Worthy K. Bioregulation of kinins: kallikreins, kinino- Immunopharmacology 1996;32:85–7. gens, and kininases. Pharmacol Rev 1992;44:1–80. [37] Parisi M, Moreno S, Fernandez G. Characterization of a novel cysteine peptidase [66] Damas J, Bourdon V, Lefebvre PJ. Insulin sensitivity, clearance and release in from tissue culture of garlic (Allium sativum L.). In Vitro Cell Dev Biol—Plant kininogen-deficient rats. Exp Physiol 1999;84:549–57. 2002;38:608–12. [67] Couture R, Girolami JP. Putative roles of kinin receptors in the therapeutic [38] Kunitz M. Crystallization of a trypsin inhibitor from soybean. Science effects of angiotensin 1-converting enzyme inhibitors in diabetes mellitus. Eur 1945;101:668–9. J Pharmacol 2004;500:467–85. Review. [39] Batista IF, Oliva ML, Araujo MS, Sampaio MU, Richardson M, Fritz H, et al. [68] Moreau ME, Garbacki N, Molinaro G, Brown NJ, Marceau F, Adam A. The Primary structure of a Kunitz-type trypsin inhibitor from Enterolobium con- kallikrein–kinin system: current and future pharmacological targets. J Phar- tortisiliquum seeds. Phytochemistry 1996;41:1017–22. macol Sci 2005;99:6–38. Review.