Purification and Characterization of a Novel Membrane-bound ...

Vol. 269,No. 52,Issue of December 30, pp. 32985-32991, 1994

T H E JOURNAL OF BIOLOGICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Printed in U.S.A.

Purification and Characterization of a Novel Membrane-bound Arginine-specific Serine Proteinase from Porcine Intestinal Mucosa* (Received for publication, August 29, 1994)

Yuichi TsuchiyaS,Takayuki Takahashi+§,Yasuko SakuraiS, Akihiro Iwamatsull, and Kenji TakahashiSll From the $Department of Biophysics and Biochemistry, Faculty of Science, The University of Tokyo, Bunkyo-ku, Tbk.yo 113, Japan and llCentral Laboratories for Key ?kchnology, Kirin Brewery Co. Ltd., Yokohama-shi, Kanagawa 236, Japan

A novel membrane-boundserine proteinase has been nal proteinases are necessaryt o shed light on the mechanism purified from the microsomal membranesof porcine in- of protein and peptide degradation and its regulation in the the microsomal intestine. testinal mucosa. It was solubilized from membrane fraction with 1%sodium deoxycholate,then Many physiologically active peptides are presumed t o be purified by a series of column chromatographic steps produced from their precursors by specific cleavages at single on DE52,butyl-Toyopearl, Bio-Gel P-150,Mono Q, and or paired basic residues, and the putativeprocessing enzymes benzamidine-Sepharose in thepresence of 0.02% Lubrol are thought to have trypsin-like, but much more strict, speciPX Its molecular mass wasestimated to be 50 kDa both ficity. Thebeststudiedamongsuch enzymes istheyeast by SDS-polyacrylamide gel electrophoresis under non- KEx2 geneproduct, a Ca2+-dependentsubtilisin-like serine reducing conditions and by gel filtration, and to be 32 proteinase (1, 2). Recently, the cDNAs for mammalian homkDa by SDS-polyacrylamide gel electrophoresis under ologs of the Kex2 proteinasehave been isolated,including reducing conditions, suggesting that the enzyme may exist as a homodimer in which twosubunits are linked those for furin (3, 41, PC2 (5),PC1/3 (6, 71, PC4 (81, PC5 (91, by disulfide bond(& It had a pH optimum at around 9 PACE (lo), PACE4 (111, and PC6 (121, and the encoded enand did not require Caa+for activity. It cleaved several zymes have been expressedin heterologous systemsand peptide 4-methylcoumaryl-7-amidesubstrates almost proved t o correctly process some protein or peptide precursors exclusively after arginine residues, the best substrate into their mature forms by cleaving mainly after basic amino among those tested being t-butyloxycarbonyl-Gln-Ala- acid pairs such as Lys-Arg and Arg-Arg (for reviews, see Refs. Arg-4-methylcoumaryl-7-amide. Various neuropeptides 13-15). Many of the precursors of gastrointestinal peptides some of were also cleaved by this enzyme after arginine, mainly alsopossess such basic pairs. On the other hand, between paired basic amino acid residues, Arg-Arg or them are supposed to be cleaved after single arginine resiArg-Lys. Activitytoward protein substrates was scarcely dues, or between two basic residues (16-181, and Kex2 prodetected. Further, its partial amino acid sequences were teinase homologs have been shown to cleave at some of these highly homologous,but not identical, with those of tryp- sites, although not as efficiently (19). Therefore, the participasin-type serine proteinases. These results indicate that tion of enzymes other than Kex2 proteinase homologs is also the present enzyme is a novelarginine-specifictrypsin- expected in such processing events. To our knowledge, howlike endopeptidase, possibly involved as a processing ever, no such enzymes, including Kex2 homolog proteinases, proteinase in the production of certain gastrointestinal have been isolated and characterized so far from intestinal neuropeptides or peptide hormones from their precur- tissues, except that an arginine-selectiveendoprotease was sors, or their specific degradation. isolated from rat intestinal mucosa and shown to process the somatostatin precursor(20, 21). In the present study, we have attempted to find and isolate Intestine contains various kinds of proteinases, and some of such enzymes,with special attention to membrane-bound prothem are known to be involved in digestion of food proteins. teinases, and here we report the complete purification and characterization of a novel arginine-specific trypsin-like enOntheotherhand,protein digestion intheintestineis thought to be regulated by various kinds of gastrointestinal dopeptidase obtainedfrom the microsomal membrane fraction peptides. Therefore, some of the proteinases are assumed t o of porcine intestinal mucosa. play important roles in the processing of these peptides, inEXPERIMENTALPROCEDURES cluding activation of their precursors and degradation of their Materials-Porcine intestine was purchased from Shibaura Hormature forms. So far, however, little information is available about the nature of the proteinases involved in such process- mone Manufacturing Co. (Tokyo). The following materials were puring in intestinal tissues. Therefore, further studies on intesti- chased from the sources indicated: DE52,Whatman Chemical SeparationLtd.;butyl-Toyopearl,TosohCorporation;Bio-GelP-150 and a mixture of low molecular weight marker proteins for SDS-PAGE,’ Bio* This work was supported in part by grants-in-aid for scientific re- Rad; benzamidine-Sepharose, FPLCMono Q, Mono P, Superose 12, and search from the Ministry of Education, Science and Culture of Japan. PD-10, Pharmacia Biotech Inc.; fluorescamine, diisopropylfluorophosThe costs of publication of this article were defrayed in part by the payment of page charges. Thisarticle must thereforebe hereby marked The abbreviationsused are: PAGE, polyacrylamide gel electrophore“aduertisement”in accordancewith 18 U.S.C.Section1734solely to sis; MCA, 4-methylcoumaryl-7-amide;DFP, diisopropylfluorophosindicate this fact. phate; E-64, L-trans-epoxysuccinyl-leucylamide-(4-guanidino)-butane; Q Presentaddress: Department of Biology, Faculty of Science, TLCK, N“-tosyl-L-lysine chloromethyl ketone; TPCK, Ne-tosyl-L-phenylHokkaido University, Sapporo-shi, Hokkaido060, Japan alanine chloromethyl ketone; Boc-, t-butyloxycarbonyl-; BA”12P, bo(1 To whom correspondence shouldbe addressed. Tel.: 81-3-5689-5607; vine adrenal medulla dodecapeptide; BCA,bicinconinic acid; PVDF, Fax: 81-3-5802-2041. polyvinylidene difluoride. 32985

32986

Membrane-bound Arginine-specific Serine Proteinase TABLEI Purification of intestinal areinine-suecific uroteinase SkP

Microsomal membrane Solubilization DE52 Butyl-Toyopearl Bio-Gel P-150 FPLC Mono Q Benzamidine-Sepharose

Volume

Total activity

Protein

Specific activity

Purification

Yield

ml

units

mg

unitslmg

-fold

%

400 600 885 150 40 3.5 3.5

3,370 7,700 4,880 1,960 2,440 1,620 1,030

5,400 4,860 2.6 1,170 ND" 1.52 0.665 c0.005 >130,000

0.62 1.58 4.17 ND 1,610 2,410 >206,000

1.0 ND 1,020 1,530

100 63 25 32 21 13

'ND, not determined, because of the high concentration of ammonium sulfate. phate (DFP),trypsininhibitors from soybean and bovine pancreas, p-chloromercurybenzoic acid, EDTA, o-phenanthroline, Lubrol PX, bovine oxidized insulin B chain, Coomassie Brilliant Blue G, and various protein substrates, Sigma; E-64, phosphoramidon, pepstatin A, peptide MCA substrates, and various neuropeptides, Peptide Institute (Osaka); sodium deoxycholate, Nacalai Tesque(Kyoto);N"-tosyl-L-lysinechloromethyl ketone (TLCK), Wako Pure Chemical Ind.; N"-tosyl-L-phenylalanine chloromethyl ketone (TPCK),Aldrich; BCAm and microBCATM protein assay kits, Pierce; ProSpinTM,Applied Biosystems. Leupeptin and antipain were kindly supplied by Dr. Takaaki Aoyagi (Institute of Microbial Chemistry). All other reagents used were of analytical grade. Preparation of Intestinal Microsomal Fraction-All the procedures were carried at 4 "C essentially according to Sogawa et al. (22). Intestinal mucosa (1,600 g) from two adult pigs was minced and homogenized in a Waring blender with 2 volumes of 0.25 M sucrose solution. The homogenate was centrifuged at 10,000 x g for 30 min. The supernatant fraction was centrifuged a t 90,000 x g for 90 min. The pellet was washed once with 0.1 M borate-Na,CO, buffer, pH 8.0, and then once with the same buffer containing 1M KCI. The final pellet was suspended in 400 ml of 0.1 M borate-Na,CO, buffer, pH 8.0, and used as the microsomal membrane. To solubilize the microsomal membranes, 60ml of 10% sodium deoxycholate was added to the above membrane fraction, and this fraction was further diluted with the same buffer to 600 ml. The suspension was stirred overnight with a magnetic stirrer and was centrifuged at 90,000 x g for 90 min. The solubilized fraction was dialyzed against 0.1 M borate-Na,CO, buffer, pH 8.0, containing 0.02% Lubrol PX in order to exchange the detergent. Purification Procedures-Throughout the procedures, the buffers used contained 0.02% Lubrol PX. The solubilized fraction was divided into four equal parts (150 ml each) and separately applied to a DE52 column (2.6 x 28 cm) equilibrated with the same buffer. After washing, the column was eluted with a gradient of NaCl (0-0.3 M) in a total volume of 1,000 ml at a flow rate of 60 ml/h. The enzyme was eluted at about 0.2 M NaC1, just after the major protein peak. All of the active fractions from four runs were pooled and combined, and to this fraction was added ammonium sulfate to 20% saturation. This sample was directly applied to a butyl-Toyopearl column (2.6 x 28 cm) equilibrated with 0.1 M borate-Na,CO, buffer, pH 8.0, containing ammonium sulfate at 20% saturation. The column was washed with the same buffer containing ammonium sulfate and Lubrol PX, and then eluted with a decreasing gradient of ammonium sulfate (20 to 0% saturation) in a total volume of 1,000 ml at a flow rate of 60 ml/h. The active fractions were pooled and brought to 30% ammonium sulfate saturation. This solution was concentrated with a small volume of butyl-Toyopearlto 5.0 ml, then submitted to gel filtration on a Bio-Gel P-150 column (2.5 x 82 cm). Elution was performed with 0.1 M borate-Na,CO, buffer, pH 8.0, containing 0.2 M NaCl at a flow rate of 20 mVh. The pooled fraction was dialyzed against 20 mM Tris-HC1 buffer,pH 8.0, and applied to a FPLC Mono Q HR5/5column (Pharmacia Biotech Inc.) followedby linear gradient (0-0.5 M NaCl) elution at a flow rate of 0.5 mumin. Finally, the active fractions were pooled and applied to a small column (1.0 ml) of benzamidlne-Sepharose a t a flow rate of 5 mlh. The column was washed with 20 m~ Tris-HC1, pH 8.0, containing 0.1 M NaCl, and then eluted with the same buffer plus 50 m~ benzamidine. Active fractions were pooled and desalted with a PD-10 column (Pharmacia Biotech Inc.). Enzyme Assays toward MCA Substrates-The activity was assayed according to Yanagida et al. (23) and Tamanoue et al. (24). To an enzyme solution (0.5 ml) in 0.1 M glycine-NaOH buffer, pH 9.5, was added 5 pl of 10 mM Boc-Gln-Ala-Arg-MCA in dimethyl sulfoxide, and themixture was incubated at 37 "C for a suitable time. The reaction was stopped by adding 2.5 ml of 0.1 M sodium acetate buffer, pH 4.3, containing 0.1 M monochloroacetic acid, and fluorescence was measured with an excita-

tion wavelength at 370 nm and an emission wavelength a t 460 nm. One unit of the activity was defined as the amount of the enzyme that liberates 1 nmol of 7-amino-4-methyl coumaridmin. Digestion of Peptides and Analyses of the Resulting Peptide Fragments-Digestion was performed according to Takahashi et al. (25). Each peptide (about 1.0 nmol) was incubated with 0.1 unit of the enzyme for 2.5 h in 20 pl of 0.07 M Tris-HC1 buf€er, pH8.0, containing 10 mM EDTA. An aliquot of each digestion mixture was analyzed using an Applied Biosystems 130A separation system on an Aquapore OD-300 C,, column (2.1 x 30 mm). The elution was performed with a linear gradient of acetonitrile (0-70%) in 0.1% trifluoroacetic acid a t a flow rate of 0.8 mumin and monitored at 220 nm. Each peptide fraction was collected, lyophilized, and submitted to amino acid analysis using an Applied Biosystems derivatizer-analyzer (420Al130A-920A) equipped with an on-line automatic hydrolysis apparatus. Enzyme Assays toward Protein Substrates-Protein substrates were dissolved in 0.1 M borate-Na,CO, buffer, pH 8.0. The assay mixture (60 p1) containing each protein substrate at 1.0% (w/v),0.5 unit of the enzyme, and 10 mM EDTA, was incubated for 18 h at 37 "C. To stop the reaction, 180 1.11of 5% trichloroacetic acid was added to the mixture. The mixture was then centrifuged at 10,000 x g for 10 min, and an aliquot (100 pl) of the supernatant was removed to determine the extent of digestion by the fluorometric method using fluorescamine (22). Amino Acid Sequencing-To determine the NH,-terminal sequence of the enzyme, the purified enzyme fraction (500 units) partially desalted by passing through a PD-10 column was lyophilized and thendissolved in 50 p1 of 20 mM sodium acetate buffer, pH 4.5, containing 0.2 M NaCl, then applied onto a ProSpinm (Applied Biosystems) PVDF membrane cartridge according to the manufacturer's instruction. The membrane was directly submitted to amino acid sequencing using an Applied Biosystems Protein Sequencer model 477N120A. On the other hand, preparation and sequencing of peptide fragments were also performed according toIwamatsu (26). The purified enzyme (1,000 units) was electroblotted onto PVDF membranes after separation by SDS-PAGE. The enzyme on PVDF membranes was reduced and S-carboxymethylated, then incubated with Acromobacter protease I (lysylendopeptidase) and endopeptidase AspN, respectively. Released proteolytic fragments were separated by HPLC using a p-Bondasphere 5p C8-300A column (2.1 x 150 mm, Waters) and submitted toamino acid sequencing with a Shimadzu PSQ-1 gas phase sequencer. Electrophoresis-SDS-PAGE was performed in 12.5% polyacrylamide gel by the method of Laemmli (271, followed by CoomassieBrilliant Blue staining. Determination of Protein-Protein was determined by measuring the absorbance at 280 nm of the sample solution or by the method of Smith et al. (28) using the BCA reagent with bovine serum albumin as a standard. RESULTS

The resultsof purification of an arginine-specific proteinase from a microsomal membrane fraction of porcine intestinal mucosa are shown in Table I. The activity increased over 2-fold by oversolubilization with sodium deoxycholate, and an apparent all purification of over 130,000-fold was achieved starting from this solubilized fraction by a series of column chromatographic runs (Fig. 1).Through these procedures, the activities of contaminating proteinases such as aminopeptidase N, dipeptidyl peptidase N,neutral endopeptidase 24.11., and enteropeptidase were completely excluded. The purified enzyme gave a single bandon SDS-PAGE under both non-reducing and reducing conditions (Fig. 2), and a single peak on gel filtration (data

32987

Membrane-h0un.d Arginine-specific Serine Protcinnsc

i .

FRACTION NUMBER of intestinal arginine-specific serine proteinase. a, DE52 ion-exchangc chrnmatnnaphy. Onr was applird to a DE52 column ( 2 . 6 x YH cm I rquilihr;rtvd wlth 0.1 fourth of the enzyme fraction from the intestinal microsomal memhranc. fraction &I borate-Na,CO, hulfrr. pH 8.0, containing 0.02"; Luhrol PX. Thr column was eluted with a gradirnt of SaC.1 (0-0.3 \t J in a tntnl vnlrrmc nf l.00o ml a t a flow rate of 60 mlh. Fraction size, 10 ml. The fractions under thrhor were pooled. This oprration was pc2rformrd fnur timrs nnd thrpoolrd fractions were comhined. h. hutvl-Toyopearl hydrophohic chromatography. The sample ohtainrti hy DE52 chromatopaphy w:w atllustvd to 20'; saturation of ammonium sulfate and applied to a hutyl-Toyopearl column (2.6 x 28 cm) equilihrntrd with 0.1 \t hnratr-Sn,.('O, hufTrr. pfl H . 0 , containing 0.0Y;Luhrol PX and ammonium sulfate a t 20'; saturation. The column was elutrd with a decreasing kwdlrnt nf nmmrmlum sulfntr ( 2 0 to Or+ 1 in a total volume of 1,000 ml a t a flow r a t e o f 60 mlh. Fraction size, 11 ml. The fractions undrr thra bor wrrr pnnlrd. r, 1 3 I o 4 k . l P-150 gel filtration. The sample ohtainedhy butyl-Toyopearl chromatography \vas concentrated and applied toa Rio-(;rl P-1.50 rrrllrmn 12.5 x H2 rm 1 and eluted with 0.1 \I borate-Na,CO, huffer, pH 8.0, containing 0.02'; Luhrol PX and 0.2 >t SaCl a t a flow r a t e o f 20 ml'h. Thr frnctinnr undrr thrhnr were pooled. d , Mono Q ion-exchange chromatography. The sample ohtained hy Rio-Gel P-1.70 grl filtration was applird to n n FI'IX' \lonn ( 2 11R 5/5 column followed hy linear gradient (0-0.5 11 NaCI, elution in a total volume of 21.0 ml at n flow r a t r of 0.5 ml rnin Frnrtlon S I Z V . ( 1 5 ml. Thv fractions under the bar were pooled. Flr:. 1. Chromatographic purification

TAIII.PI 1 Effrrts of rnrrntls rrn,qcntv on thr nr.trr.rty

The enzyme f0.1 u n i t ) was prrincuhnted with rach rearrnt in 0 . 1 glycine-NaOHhufTrr, pH 9.5. for 311 mln a t rnom tempvrnture. Thr remaining activitv was assayed hv Incuhatmn wlth lk~c-(;ln-;lla-ArrM C A for 10 min.

97 66

~~~~

~

Re*ncrwt

45

~

31

TLCK TPCK Soyhean trypsin inhihitor Rovine pancrrns trypsin inhihitor Leupeptin Antipnin

21 14

DFP

FK:.2. SDS-PAGE of the purified enzyme. Electrophoresis

was

performed in a 12.5? gel followcd hv Coomassir Brilliant Rlue staining. U7"/', Positions ofthe molecular mass markers are shown in kilodaltons. dithiothreitol.

Renzamidine p-Chloromrrcurihrnzoic acid E-64 EDTA o-Phenanthroline Phosphoramidon Pepstatin CR" 7.n 2 .

~

I'onrrmtmtlnn

Inhrtvtrnn

1 .O m\c 1 .O m\c 0 . 2 mg'ml 0 . 2 mg'ml 0 . 1 mv 11.1 mv 0 . 1 m\t 1 .0 m\t 1 .O m\t 1 0 . 0 m\t 1 . O mv 1.0 mv 1 0 . 0 m\t 10.0 m\c (1.1 mv 0 . I m\t I .n m v 1.0 mv

9 .I

~~

12

x I)!) 1 I1 HH GfJ

$15 ,r,i

$15 I1 0 I1

2f1 not shown). The molecular mass of this proteinase was detertl mined to he 50 kDa both by SDS-PAGE under non-reducing A fl conditions and by gel filtration on Bio-Gel P-I50 and Superose 3 $3 27 12. On the other hand, by SDS-PAGE under reducing condi.~ tions the molecular masswas estimated to be 32 kDa, suggesting that the enzyme may exist as a homodimer connected by disulfide-bond(s) as will be discussed later. Its isoelectric point typical serine proteinase inhihitors. such a s DFP, hovine pnncreas and soybean trypsin inhihitors. leupeptin. and antipain. was approximately 4.5 as deduced from the chromatographic The enzyme did not require Ca" for activity. and was not inbehavioron a Mono P chromatofocusingcolumn(datanot hihited by metal chelators such as EDTAand o-phenanthroline. shown). As shown in Table 11, this enzyme was strongly inhibitedby Various divalent cations also had little effect on thp activity. ~~

~~~~

Membrane-bound Arginine-specific Proteinase Serine

32988

TABLE I11 Activity toward MCA substrates Each MCA substrate was incubatedwith 0.1 unit of the enzyme in 0.1 M glycine-NaOH buffer,pH 9.5, for 10 min at 37 “C. Theactivity toward Boc-Gln-Ala-Arg-MCA wastaken as loo%, and relative activities are shown. Substrate

Relative activity

Substrate”

Relative activity

%

Boc-Gln-Ala-Arg-MCA Boc-Gln-Gly-Arg-MCA Boc-Phe-Ser-Arg-MCA Boc-Leu-Gly-Arg-MCA Boc-Gln-Arg-Arg-MCA Boc-Leu-Thr-Arg-MCA Boc-Leu-Lys-Arg-MCA Boc-Gly-Arg-Arg-MCA

100

Z-Arg-Arg-MCA Bz-Arg-MCA Boc-Val-Leu-Lys-MCA Boc-GIu-L~s-LYs-MCA Suc-Ala-Ala-Pro-Phe-MCA Suc-Leu-Leu-Val-Tyr-MCA Ala-MCA Gly-Pro-MCA

61

57 45 30

23 6

Bz, benzoyl-; SUC-,succinyl-;Z-, benzyloxycarbonyl-. TABLEIV Activity toward peptide substrates The sites and extents of cleavage by the enzyme in each peptide are shown. A large arrowhead indicates a major cleavage site, and a small arrowhead, a minor cleavage site. The number above each arrowhead indicates the extent of cleavage in percent underthe specified conditions. Each peptide(1.0 nmol) was incubated with the enzyme (0.1 unit) as described under “Experimental Procedures.” Incubation was for 2.5 h at 37 “C except for secretin (7.5 h) and oxidized insulin B chain (24 h). Cta, cysteic acid;