Purification and properties of a protease from the sarcocarp of bead tree fruit

Pergamon

0031-9422

(93) 0094-4

Pkymckemiwy,

Vd. 35, No. 6, pp 1395-1398, 1994 Elsewer Scma Ltd Printed in Great Britam 0031~9422,94 s6.00+000

PURIFICATION AND PROPERTIES OF A PROTEASE FROM THE SARCOCARP OF BEAD TREE FRUIT MAKOTO KANEDA, KAZUNARIARIMA,* HIROO YONEZAWA and TETSUYA UCHIKOBA Department

of Chemistry,

Faculty of Science, Kagoshima University, Korimoto, Kagoshima 890, Japan (Receiued in revised form 27 October 1993)

Key Word Index-Melia

azedarach; Meliaceae; bead tree; cysteine protease; plant protease; melain.

Abstract-A protease was purified from bead tree fruit (Melia azedurach L. var. japonica Makino) in four steps, including HPLC gel-filtration. The M, of the enzyme, named melain, was estimated to be 25 000 on SDS-PAGE and on HPLC gel filtration. Melain contained a carbohydrate moiety. Using casein as a substrate, the optimum pH of the enzyme was 7.5-8.5 at 37”. The enzyme was inhibited by iodoacetic acid, but was not inhibited by phenylmethanesulphony1 fluoride or EDTA. The enzyme had a wide specificity for peptide substrates such as oxidized insulin B-chain. All split sites (Pl and/or PI) were hydrophobic or charged amino acid residues. The enzymatic properties of this protease were similar to those of phytolacain, an enzyme from the fruit of pokeweed, Phytolacca umericana.

INTRODUCMON

The fruits (drupes) of Melia azedarach have been used for the treatment of chaps and chilblains of the skin in Japan. During the screening of plant proteolytic activity toward milk casein, we found that the fruit of the bead tree (Meliu azedmach L. var. japonica Makino) had very high activity for this substrate. Many plant proteases have been isolated [l, 21, however, the purification and properties of proteases from the Meliaceae have not been reported. In this paper, we describe the purification, general properties and substrate specificity of a protease from the fruit of the bead tree which we have named melain. RESULTS AND DBCUSSION

A protease from M. azedarach fruit (drupe) was purified by a four-step procedure. All purification steps were performed at 8”. The juice of ripe fruit gelatinized in a short time on account of a large amount of intrinsic pectic substances. In order to avoid gelatinization, solid ammonium sulphate was added rapidly to the fresh juice to 60% saturation. The resulting precipitate was dissolved in 60-fold of Na, K-Pi buffer, pH 5.5, and applied on a DEAE-Cellulose column. Acidic components in the juice were removed through the DEAE-Cellulose column. The eluent from DEAE-Cellulose was applied to a CMCellulose column. As shown in Fig. 1, an active peak was obtained. The protease fraction concentrated with phenyl-sepharose CL4B was further separated using SynChropak *Present address: Department

of Biochemical Engineering and Science, Faculty of Computer Science and Systems Bngineering, Kyushu Institute of Technology, Iizuka 820, Japan.

GPC-100 HPLC gel filtration. Two peaks were obtained, with the second peak retaining protease activity. From the sarcocarp of ripe bead tree fruits (1.5 kg), 4.2 mg of the purified enzyme was obtained. SDS-PAGE of purified protein showed a single band by silver staining, having a M, of ca 25000 as shown in Fig. 2. A value of 25000 was also obtained by using SynChropak GPC-100 HPLC gel filtration. A single band was also obtained with periodate-Schilf reagent on the SDS-PAGE gel of the purified enzyme. The band of carbohydrate was found at the same position as a protein band. Therefore, the enzyme is probably a glycoprotein. Carbohydrate content of the protein was 2.8% (w/w) by the phenol-sulphonic acid method. The effect of pH on the proteolytic activity and stability of melain was examined. The optimum pH for the

a 0.4 i

20

i

g-

s

10 3p

0.2

4

5 0 0

100 FrpdlomNa

0 200

Fig. 1. Elution profiles of protein (solid line) and protease activity (0) from a CM-cellulose column. Fractions of 15 ml were colkxted. Protease activity of 0.5 ml aliquot of each fraction was measured for 20 min at 35”. 1395

M.

1396

KANEDA m al.

kDa

67

Fig. 3. Specificity of melain towards oxidized insulin B-chain. Abbreviations of ammo acids follow the alphabetical system [lo]. C* indicates cysteic acid. t Indicates the bond split, ando indicates peptide identified. The number at the bottom of the arrow indicates recovery percentage from the substrate.

Table 1. Effect of various compounds on the proteolytic activity of melain against casein substrate

Fig. 2. SDS-PAGE

proteolytic

activity

of purilied melain.

of melain

was observed

to be ca

7.5-8.5 on casein. Half maximal activity was obtained at pH 6 and 10.5. At pH 2.0 and 12.0, the enzyme showed negligible activity. The pH stability of the enzyme was examined by incubating at various pH values for 3 hr at 37” prior to assay at pH 7.5 under experimental conditions. At least 70% of the activity remained after incubation between pH 5 and 9. The effect of temperature on the proteolytic activity and stability of melain was investigated. The optimum was 55” (incubated at pH 7.5). Melain was stable below 50” during a 1 hr incubation. Half proteolytic activity was retained at 65” after 1 hr, but at 90” the activity was negligible. The effects of various compounds on the proteolytic activity are shown in Table 1. The protease was completely inactivated by incubation with 1 mM iodoacetic acid for 1 hr at 35” and inhibited by 0.1 mM 5,5’-dithiobis (Znitrobenzoic acid) (DTNB) and p-chloromercuribenzoic acid (PCMB). Phenylmethanesulphonyl fluoride did not show any appreciable effect. The protease activity was slightly stimulated by EDTA. These results suggest that this enzyme is a cysteine protease. It was shown that the

Addition

Concentration (mM)*

Relative activity (%v

None Iodoacetic acid PCMBS DTNB PMSF TPCK TLCK Leupeptin EDTA CXI, MgCtZ MnCI, BaCI, COCI, CdCl, ZnCl, Cut&

2.0 1.0 0.1 1.0 0.1 0.1 0.1 1.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0

100 0 0 15 64 57 71 17 127 91 93 107 94 75 6 3 0

The enzyme was preincubated for 60 min at 35” in 1 ml 0.067 M Pi buffer, pH 7.2, containing various compounds. After preincubation, 1 ml of 2% casein in 67 mM Na, K-Pi buffer, pH 7.2, was added to 0.3 ml of the mixture and the activity assayed by the standard procedure. *Concentrations arc those in the preincubation mixture. tActivity of a control with no addition was taken as 100%. $p-Chloromercuribenzoic acid.

enzyme contained (a) essential sulphydryl group(s) as follows. The enzyme activity was gradually decreased by addition of DTNB. Addition of 2-mercaptoethanol to this reaction mixture reversed this effect and resulted in an activity exceeding the original. Melain was inactivated by 5 mM Cu2+ and Zn 2+ ions. Other bivalent ions did not have any influence.

Protease from bead tree fruit

1397

Protease assays. Proteolytic activity was measured by the method of ref. [6], with casein as a substrate. One ml of enzyme soln was added to 1 ml of 2% (w/v) casein in 67 mM Na, K-Pi, pH 7.2, at 37”. After incubation for 20 min, the reaction was stopped by the addition of 3 ml of 5% TCA. After standing for 30 min at room temp. the resulting ppt. was removed by filtration through Toyo filter paper No. 5C and A at 280 nm of the TCA-soluble peptides formed was determined. A unit of activity was defined as that amount which yielded 0.001 Azso unit of change per min under the above conditions. Determination of carbohydrate. The carbohydrate content of enzyme preparation was determined by the phenol-sulphuric acid method of ref. [7-j. D-Mannose was used as a standard. PuriJcation of melain. All procedures for purification of the melain were carried out at 8”. Step 1. Extraction. A sarcocarp of ripe bead tree fruits (1.5 kg) was homogenized and added in 11 of 67 mM Na, K-Pi buffer, pH 5.5. The homogenate was filtered through cotton cloth and centrifuged. The final supematant had a vol. of ca 1.5 1 with a light greenish colour. Solid (NH&SO4 was added rapidly to the supernatant to 60% satn, and the soln was kept for 24 hr. The resulting ppt. was diluted to 60-fold vol. with 0.067 M Na, K-Pi buffer, pH 5.5 (buffer A) and centrifuged to remove minor insoluble materials. Step 2. DEAE-Cellulose column chromatography. EXPERIMENTAL Supernatant from step 1 was placed on a column of DEAE-Cellulose (20 x 8 cm) equilibrated with buffer A. Bead tree fruits (Melia azedarach L. var. japonica After the supernatant was eluted, the column was washed Makino) were collected in Kagoshima city during July. with 500 ml of buffer A. The whole activity of the CM-Sepharose CL-6B was a product of Pharmacia. CMsupernatant was recovered in the eluate and washings. cellulose was purchased from Whatman. Casein was Step 3. CM-Cellulose column chromatography. The obtained from Merck. Protein calibration kit size II was eluent (1.5 1) from step 2 was applied to a column of CMpurchased from Boehringer. N-Tosyl-L-lysine chloroCellulose (45 x 4 cm) equilibrated with buffer A. The methylketone was a product of Sigma. N-Tosyl-L-phenylenzyme was eluted with a linear gradient from buffer A alanine chloromethylketone was from Seikagaku. Pepti(1.5 1)to 0.2 M Na, K-Pi buffer, pH 6.4 (1.5 1)at flow rate dyl p-nitroanilides were obtained from Peptide Inc. Diof 1.8 ml min- i. Proteolytic activity in each fr. was measisopropylfluorophosphate was a product of Fluka AG. ured against casein as the substrate. Step 4: the active Trichloroacctic acid, /J-mercaptoethanol, p-chloromerprotein fr. was collected and (NH&SO, was added to curibenzoic acid and monoiodoacetic acid were pur30% satn. This fr. was then applied to a Phenyl-sepharose chased from Wako. Other materials were of the highest CL4B column (4 x 1 cm) equilibrated with buffer A purity obtainable. containing (NH&SO, to 30% satn (buffer B). The colM, determination. The M, of the enzyme was estimated umn was washed with 60 ml of buffer B. Proteolytic by HPLC gel filtration of SynChropak GPC-100 activity fr. (12 ml) was eiuted with buffer A. (SynChrom, 300 x 7.8 mm) equilibrated with 50 mM Na, Step 5. HPLC gel filtration. The protease-containing K-Pi buffer, pH 7 containing 0.1 M Na,SO,. The gel fraction from step 4 was eluted on a SynChropak GPCfiltration was performed with the same buffer at a flow 100 HPLC gel filtration with 50 mM Na, K-Pi buffer, pH rate of 0.8 ml min- 1 (at 20 atm). The eflluents were moni7 containing 0.1 M NaaSO,. A portion of the purified tored by A at 215 nm. The column was calibrated using enzyme soln from step 5 was applied on SDS-PAGE and aldolase (158 000), BSA (67 000), ovalbumin (43 000), chythe other aliquots were kept at -20”. motrypsinogen (25 000) and cytochrome c (12 500). The Digestion of B-chain of insulin. Oxidized B-chain of subunit M, was determined by SDS-PAGE under redubovine insulin was prepared by the method of ref. [8]. The cing conditions according to the method of ref. [4] using a oxidized B-chain (70 pg: 20 nmol) was digested with 15% polyacrylamide slab gel. Proteins were stained with 0.2 nmol of melain in 33 pl of 67 mM Na,K-Pi buffer Ag staining by the method of ref. [S]. The marker pH 7.2, at 37” proteins used were phosphorylase (94 000), BSA (67 OOO), containing 1 mM 2-mercaptoethanol, ovalbumin (43 000), carbonate dehydratase (30 OOOZ for 24 hr. On completion of the reaction, 10 ~1 of 70% HCOOH was added to the mixt. The peptides were sepd soybean trypsin inhibitor (20000) and x-lactalbumin by reversed phase HPLC on a Aquapore RP-300 column (14400).

The oxidized B-chain of bovine insulin was hydrolysed by melain. From the amino acid compositions of resulting peptides, it was possible to locate each peptide in the primary structure of the B-chain of insulin, as shown in Fig. 3. It appeared that cleavages had occurred at carboxy1 terminal of CySO,H7, Ser9, HislO, Va112, Ala14, Leul5, Tyrl6, Leul7, CySO,H19, Glu21, Arg22, Gly23, Phe24 and Phe25 by the enzyme for 24 hr. Some split sites (Pl and/or P’l) were large hydrophobic amino acid residues as Phe, Val and Leu of HislO-Leull, Val12Glu13, Alal4-Leul5, LeulS-Tyr16, Tyrl6-Leul7, Leul7Va118, Gly23-Phe24, Phe24-Phe25 and Phe25Tyr26. Other split sites (Pl and/or P’l) were charged amino acid residues as CyS03H, Arg, Glu and His of CyS03H7Gly8, Se&HislO, CySO,H19-Gly20, Glu21-Arg22 and A&2-Gly23. But the following sites were not split; m2Asn3, HisS-h6 and bl l-Ml2 (hydrophobic amino acid residues were underlined). This difference may be due to the effect of other subsites as P2 or P2, or the influence of the peptide conformation of the insulin B-chain. A trypsin substrate, Bz-Arg-pNA is hydrolysed by papain, but not hydrolysed by melain. Phytolacain, cysteine protease from pokeweed fruit, cannot also hydrolyse the substrate [3]. The substrate specificity of melain seems to differ from that of papain.

PHYTO 35:6-C

M.

1398

KANEDA

(250 x 4.6 mm, Brounlee) with a linear gradient of 060% MeCN containing 0.1% TFA at 230 nm in Applied Biosystems 15OA HPLC system. The purified peptides were hydrolysed in 6 M HCl containing 0.1% phenol for 24 hr at 1lo”. The samples were derivatized with phenylisothiocyanate in a PICO . TAG Work Station (Waters) by the method of ref. [9]. Phenylthiocarbamyl derivatives of amino acid were sepd on a PICO *TAG amino acid analyser for PICO.TAGrn column (Waters, 150 x3.9 mm) and detected at 254 nm [9].

RJXFERENCES

1. Boiler, T. (1986) Plant Proteolytic Enzymes (Dalling, M. D., ed.), Vol. 1, pp. 67-96. CRC Press, Florida. 2. Kaneda, M. (1993) Tanpakushitsu-Bunkaikouso (in

et al.

Japanese, Tsuru, D. and Funatsu, M., eds), Vol. 2, pp. 145-180. Gakkai Shuppan, Tokyo. 3. Kaneda, M., Izumi, S., Fukuda, T., Uchikoba, T. and Tominaga, N. (1988) Phytochemistry 27, 3661. 4. Laemmli, U. K. (1970) Nature 227, 680. 5. Oakley, B. R., Kirsh, D. R. and Morris, N. R. (1980) Analyt. Biochem. 105, 361.

6. Kunitz, M. (1947) J. Gen. Physiol. 30, 291. 7. Dubois, M., Gilles, K. A., Hamilton, J. K., Rober, P. A. and Smith, F. (1956) Analyt. Chem. 28, 350. 8. Hirs, C. H. W. (1967) Meth. in Enzymol. 11, 197. 9. Bidlingmeyer, B. A., Cohen, S. A. and Tarvin, T. L. (1984) J. Chromatogr. 336, 93. Joint Commission on Biochemical 10. IUPAC-IUB Nomenclature (JCBN), Nomenclature and Symbolism for Amino Acids and Pepddes (1985) J. Biol. Chem. 260, 14.