Function of the Pro Pep Tide Region in ant Expression of Active sin L in Escherichia Coli J Biochem-1999-Ogino-78-83

8/4/2019 Function of the Pro Pep Tide Region in ant Expression of Active sin L in Escherichia Coli J Biochem-1999-Ogino-78-83

1/6

J. Biochem. 126, 78-83 (1999)

Fun ction of the Propeptide R egion in Recombinant Exp ression ofAc tive Procathepsin L in Escherichia coli1Tetsuya Ogino,* Toshio Kaji,* Mitsuhiko Kawabata,* Kazumasa Satoh,* Koji Tomoo,*Toshim asa Ishida,*-2 Hiroshi Y amazaki,' Kazumi Ishidoh,1 and Eiki Kominami*'Department of Physical Chemistry, Osaka U niversity of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki,Osaka 569-1094; and ^Department of Parasitology an d ^Department of Biochemistry, Juntendo University Schoolof M edicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8421Received March 12, 1999; accepted April 15, 1999

In order to determine the functional role of the procathepsin L propeptide region for thepreparation of active recombinant rat cathepsin L (CL), cDNAs encoding two short-lengthpropeptides (C-terminal 2 and 27 residues) and the full-length (96 residues) one plus theentire CLwere expressed as two so luble fusion proteins with a fragment of ma ltose-bindingprotein and an insoluble fusion protein with glutathione-S-transferase in Escherichia coli,resp ectively . A fter refolding of the insoluble fusion protein, each gene productwas purifiedto homogeneity by amylose or glutathione-Sepharose-4B affinity column, and digestionwi th factor Xa and a-thrombin under alkaline conditions (pH ~8 .0 ) led to the elution of twopure short-length procathepsin Ls (PCLs) and a full-length one, respectively. The en-zymatic activity, estimated by hydrolytic assaying of benzoxycarbonyl-Phe-Arg-7-(4-methyl)coumarylamide under acidic conditions (pH5.5), indicated that the two short-length PCLs exhib ited in a great loss of the activity, as compared with the full-length PCL.The CD spectra of the short-len gth PCLs were different from that of the full-leng th one. Thepresent res ults clearly show that the full-length propeptide is essential for construction ofthe active tertiary structure of CL at the stage of recombinant protein ex pression , althoughthe expression of CL itself in E. coli does not require the propeptide. Based on the tertiarystructure of PCL, the propeptide region necessary for the construction of the CL activestructure has been discussed.Key words: activity, autoprocessing, CD spectra, fusion protein, propeptide, rat procathep-sin L, recombinant protein expression.

Cathepsin L [EC 3.4.22.15] (hereafter ab brevia ted as CL) either a single polypeptide or two-chains of an heavy chainis the mo st active of the lysosomal cysteine proteases with involving an active cysteine residue and a light chain, whichregard to the ability to hydrolyze azocasein (1) , elastin (2) , are linked by a disulfide bond (1) . In order to obtain a readycollagen (3), and class Il-associatedlipeptide (4), and plays source of easily purifiable cathepsin L for studies on itsa major role in intracellular protein proteolysis because it is structural /biological function, some system s for genemost commonly found in lysosomes (5). Since evidence has expression have been constructed {9-11). For the efficientbeen accumulating of its action in tumor invasion and preparation of active recombinant CL, generally, a cDNAmetastasis (6) , bone resorption (7) , and rheumatoid arthri- clone has been expressed as procathepsin L (PCL) and thentis (8), the structu re -activity relationship of CL has been converted to the matu re form through autoprocessing.receiving increasing atten tion for the characterization of its Although the propeptide region of PCL has been proposedbiological function and the development of a specific in- to be importan t for gene expression, folding, and /orhib itor. processing (9, 12), it remains to be determined for whichCL is synthesized as a preproenzym e in the cell, which is biological process the propeptide is absolutely necessary. Inprocessed to a proenzyme and then to a mature form, Le . this study, th us, the function of the propeptide, to prep are. . _^ , ^-. r r . lnt ..-n,n.. an active recombinant CL, was investigated by means of1 This study was suppo rted in par t by a research grant (06453194) , . . . , . . . , .from the Ministry of Education, Science, Sports and Culture of Japan. rec om bin an t p ro te m ex pre ss ion , folding, an d pr oc es sin g/1 To whom correspondence should be addressed. acti vity ex am ina tio n of th re e kin ds of ra t P C Ls , i.e., tw oAbbreviations: /J-ME, ^-mercaptoethanol; CD, circular dichroism; sh ort -len gth p ro pe pt id es (C -te rm ina l 2 an d 27 re sid ueCL, cathepsin L; DTT, dithiothreitol; GST, glutathione-S-trans- pr op ep tid es ), an d a full- leng th one (96 res id ue s) plus th eferase; IPTG, isopropyl-/J-D-thiogalactopyranoside; PCL, procathep- en tir e C L.sin L; P2CL, C-terminal 2 residues of propeptide plus entire CL;P27CL, C-terminal 27 residues of propeptide plus entire CL; P96CL,full-length (96 residues) prope ptide plus entire CL; Z-Phe-Arg-MCA, MATERIALS AND METHODSben2yloxycarbonyl-Phe-Arg-7-(4-methyl)coumarylamide. MaterialsEnzymes (T4 DNA ligase, T4 polynucleotide 1999 by The Japanese Biochemical Society. kin ase, res tric tio n end on ucle ases , RN ase A, a--thrombin,

78 J. Biochem.
http://jb.oxfordjournals.org/


2/6

79Xa)were purchased from Takara Shuzo, Toyobo,a Research L aborato ries, P-L Biochemicals, Nova-for theEscherichia coli BL21 (Novagen) was

usion p lasmid] were purchased fromThe affinity res ins for purification [glutathion e-Seph a-

Toyobo. Gene mutation was carried outShuzo). The m arke rs for molecular weight estima-

on of an expression vector were created by meansConstruction of an Expression VectorTo insert rat

Sail and Notlat both the N- and C-terminal sites of theith ATP (10 nM) using T4 polynucleotide kinase (5mM T ris-HCl buffer (pH 8.0) containing 10

2, 10 mM /S-ME, and 1 mM spermine (totalml) at 37'C for 60 -90 min. The gene was thenoom tem per atur e over a period of 4 h in the p resen ce of

2mM A TP. On the oth er hand, the pGEX -4T-2Sail (20 U)d Notl (20 U); the cleavage was performed in 10 mMCl buffer (pH 7.5) containing 150 mM NaC l, 7 mM7 mM /?-ME, and 0.01% Triton (for Notl) at 37Cr 5 min. T he phos phorylated cDNA of PCL (0.5 pM) wased to 66 mM T ris-H Cl buffer (pH 7.6) containing6 mM MgCl2) 0.5 mM EDT A, and 10 mM y9-ME, andby means of T4 DNA ligase

For the co nstruction of expression vectors for the s hort-and 27 residues) PC Ls, the pM AL-p2 plasmid wasEcoBl-HindHl or Pstl-HindUl restriction sites

(13).Recombinant Protein ExpressionCalcium chloride-E. coli cells (100-200 //I), which had been pre pared(14), were transformed

% seeding in 3 ml medium ) a t 37'C in a jar ferm enter.

Induction w as continued for 12 h, after which the cells wereharv ested . The cells were disrupted by sonication at 20 kHzand 100 W (30 s, 10 time s; Model UD-200, T omy ). Thepellet and supernatant were separated by centrifugation(15,000 rpm a t 4'C for 30 min). On the oth er hand, the E.coli cells transformed with the pMAL-p2-P2CL or -P27CLvector were plated on ampicillin (20-50 m g/m l) containingM9 casa mino acid me dium (300 ml) a t 37"C in a jarfermen tor. After ferm entation for 3 h, the gene was in-duced by th e ad dition of 0.4 mM EPTG. Induction w ascontinued for 8 h, and then the cells were h arves ted anddisrupted.The whole cell proteins w ere analyzed by 15% SDS-PAGE and Western blotting using antibodies to CL.Purification of a Fusion Protein and Isolation of PCL.Full-Length PCLThe pellets were resuspended in thedissolution buffer [8 M ure a, 50 mM Tris-HC l (pH 8.0), 5mM ED TA, 50 mM NaCl, 10 mM D TT] and then stood for20 h at 4'C. After centrifugation at 15,000 rpm for 1 h, thecombined supernatants were subjected to renaturationaccording to either of the following two different methods.One was the dropwise method, where the supernatantswere dropwise (2 ml/ 1 h) added to 300-400 v olumes of therena turation buffer [50 mM potassium ph osphate (pH10.7), 5 mM EDTA, 1 mM reduced g lutathione, 0.1 mMoxidized glutathio ne], stirred overnight at 4C, adjusted topH 8. 0 w ith HC1, and then subjected to DEAE anionexchange chromatography with a buffer comprising 50 mMpotassium phosphate (pH 8.0), 5 mM EDTA, and 2 mMcysteine. The other one was the dialysis method, w here thesolubilized den atured protein w as dialyzed in a tube againsta buffer comprising 100 mM T ris-HCl (pH 8.6 ), 0.015%Triton X-100 , 3m M cysteine, and 0.1 mM cystine, fol-lowed by stirring overnight at 4C. The renatured proteinsolution was then applied at the flow rate of 10 ml/h to a 2ml glutathione-Sepharose 4B affinity column equilibratedwith buffer A [140 mM NaC l, 2.7 mM KC1, 10 mMN a2HPO 4, 1.8 mM KH 2PO 4] and washed with the samebuffer until the optical density returned to the base line.The fusion protein, GST-P96CL, was then eluted withbuffer B (buffer A+lOOmM glutathione). On the otherhand, the GST-bound column was reacted with 0.5-2.0uni ts/m g of ff-thrombin in 20 mM Tris-HC l (pH 8.4)buffer containing 150 mM NaCl and 2.5 mM CaCl2 at 20'Cfor 20 h, and then P96CL was eluted with 50 mM Tris-HC lbuffer (pH 8.0) and purified by ion-exchange chrom atog-raphy on a Mono-Q column.

Short-Length PCLsBacteriolysis of the cultured cellswas carried out for 1 h at 37'C by adding lysozyme (60 n%lml), followed by sonication. After 30 min centrifugation at8,000 rpm, the supernatant was obtained by filtration andthen subjected to affinity-chromatography (amylose resin );a buffer consisting of 20 mM Tris-HCl (pH 7.5) and 0.5 MNaCl was used for the evaluation, and the same buffercontaining a 10 mM maltose w as used for the elution. Thedigestion of the fusion protein with factor Xa was perform-ed for 24 h at 25'C in the elution buffer, w hich contained thesame volume of water, 2 mM CaCl J( and factor Xa at theweight ratio of 1/400 with respect to the fusion protein.The purification of the isolated P2CL or P27CL wasperformed in the same way as for the full-length PCL.

Enzyme Assaying of CLSince the recombinant PCL isknown to cause autohydrolysis under acidic conditions, its


3/6

80 T. Ogino et al.enzyme activity w as determined by assaying of the isolatedrecombinant PCL with Z-Phe-Arg-MCA (1). The assay wasperformed at 37C using a system comprising 20 mMacetate-EDTA buffer (pH 5.5) containing 8 mM cysteine,50 mg/m l fusion protein, and 40 mM Z-Phe-Arg-MCA;although th e reaction proceeded faster at pH 4.0, th eexperim ent was performed at pH 5.5 to obtain reliabledata.

Circular Dichroism MeasurementsCircular dichroism(CD) spectra were recorded with a Jasco J-20C spectro-polarometer with a DP-500N data processor using 2-mmpath-length cells. The sample tem perature in the cuvettewas regulated with a circulating w ater bath and was kept at25'C. The recombinant PCLs, dissolved in 20 mM phos-phate buffer (pH 7.0), were preincubated for several hoursbefore recording the spectra. The concentrations of PCLs

were adjusted to about 20 //M. The results were expressedin terms of molar ellipticity [0] .RESULTS

Construction of a PCL Expression VectorThe cDNAused for recombinant rat PCL consists of a pre-peptide(N-terminal 17 residues), a pro-peptide (96 residues), andthe full sequence of mature CL (221 residues). The expres-sion system for the full-length PCL gene was constructed asa fusion with the GST protein using the pGEX-4T-2plasmid and host bacterial strain BL21. To construct theexpression vector (pGEX-4T-P96CL), the plasmid wasdigested with Sall-Notl restriction enzymes, and thenrecovered by 1% (w/v) agarose gel electrophoresis. Theplasmid and a cDNA fragment of PCL were then religatedtac Promoter

GSTIMCS

f

Saila^rr^rfnjnf^^nnSF^^^^H^^HMH

Not IPro (9 6 ) Mature Catbeprin L fern

T4 DNA UgaSall-Notl

Pro (2 7 ) Mature Caibrprin L eene

tacGST (a)P r o ( 9 6 )ature Cathepsin L gene

tac Promoter

Pro ( 2 ) Mature Catbeprin L gene

mdSE

Pro (27 ) (b)Mature Cathepsin L genetecZamalE (0

PH VHind n iP r o ( 2 )

Mature Cathepsin L genetecZa

Fig. 1. Exp ression plasm id construction of the full-length PCL as an insoluble fusion protein w ith GST, pGEX-4T-P97CL (a), andshort-le ngth PCLs as soluble fusions with the maltose-b inding protein, pMAL-p2-P27CL (b), and pMA L-p2-P2CL (c).

(kDa)97.466.2 J45.0

(a)M 1 2 3 4

(b )(kDa) 3 4

MAL-P27CL

P27CL

M(kDa)97.4 J66.2 - ^45.0 a^j31.0 ^21.5 ^1 4.4 ^ .

(c)1 2 3 4

isI s - ^h- MAL-P2CL^t - P2CL

14.4

Fig. 2. SD S- PA GE an a ly se s of th e P96CL (a), P27 CL (b ), and protein with


4/6

Function of the Procathepsin L Proregion 81using T4 DNA ligase (Fig. la ). The E. coliBL21 strain cellswere transformed with the vector on ampicillm-containingLB medium, the rat PCL cDNA being finally expresse d asan insoluble fusion with the GST protein (Fig. 2a, lane 1).On the other hand, two short-length PCLs were bothexpressed as soluble fusions with the maltose bindingprotein (Fig. 1, b and c) . To construct a expression vector,the pMAL-p2 plasmid and cDNA were digested withEcoBl-HindHi. or Pstl-HindUI restriction enzymes, andthe respective fractions were recovered by 1% (w/v)agarose gel electrophoresis. The plasmid and the cDNAfragments were then religated using T4 DNA ligase. Theresult ing vectors, named pMAL-p2-P27CL or pMAL-p2-P2CL, encode a 27 and 2 propeptide sequence plus theentire CL, respectively. The E. coli BL21 strain was thentransformed with these vectors on ampiciUin-containingLB mediu m. The induction of the tac prom oter by IPTG ledto the emergence of a 72 kDa band for pMAL-p2-P27CLan d a 68 kDa one for pMAL-p2-P2CL on 10% SDS-PAGE(Fig. 2, b and c, lane 1) of the supernatants; these bandscorrespond to the molecular weights of the matu re CL plusth e 27 and 2 propeptide residues, respectively.

Purification and Isolation of a Recombinant PCLThefull-length PCL expressed as an insoluble fusion proteinwas solubilized by the addition of urea and renatured byeither of the dropwise or dialysis method; no notabledifference was observed betwe en the methods, although the

TABLE I. Catalytic activity of recombinant PCL antoprocess-ed under pH 5.5.

Enzyme P96CL P27CL P2CL//mol/min/mg"Deviation of significance" 18.38 0.02+ 0.01"The value corresponds to ji mol of MCA produced by the enzyme from

1 mg of Z-Phe-Arg-MCA in 1 min. The value was determ ined fromthe fluorescence intensity of MCA. bF, and Fc represent the fluores-cence intensities of the sample and standard, respectively. The limitof detention was F, F c < 0.1% for the fluorometer used in this work,and + implies F.-Fc>0.1%.

PHig. 3. pH activity profiles of short-length PCLs for the hydro-

ysi s of Z-Phe-Arg-MCA. P27CL () and P2CL ( ). Thectivity of P27CL at pH 4.0 is taken as 100%, and the other values are

former method was a little superior to the la t ter one.Purification of the renatured fusion protein was performedon a glutathione-Sepharose 4B affinity column. After thecolumn had been thoroughly washed with buffer A, thepurified fusion protein was obtained by elution with bufferB (Fig. 2a, lane 2). On the other hand, the bound fusionprotein, GST -P96CL, was digested with tf-thrombin on the

(a) P96CL7 4.0

zo0 0

200 210 220 230 240 150 260

Wavelength (nra)

(b) P27CL- ^ 20

0 0

-2.0

-4.0

200 210 220 230 240 230 260

Wavelength (ran)

(c) P2CL- 20

0.0

-1 0

-4.0

20 0 210 220 230 240 230 260Wavelength (nm)

Fig. 4. CD spectra of the purified recombinant fall-length PCL(a) and two short length PCLs, Le. 27 residue (b) and 2 residue(c) ones. Samples were dissolved in 20 mM phosphate buffer (pH7.0).

ol. 126, No. 1, 1999


5/6

82 T. Ogino et al.equilibrated affinity column and then eluted with buffer A(Fig. 2a, lane 3), where the scission of GST-P96CL wasmo st effective in the case of 1.5 u nits /m g (ratio of a -throm-bin per fusion protein) . About 0.4 mg of pure recom binantPCL was obtained from 8.5 g E. coli cells. The 40-kDaform, which corresponds to the molecular mass of theful l-length PCL , was obtained almost ent irely as a 29-kDaband on ion-exchange chrom atography at pH 5.5, which isof the molecular mass expected for the autoprocessed CL(Fig. 2a, lane 4).On the other hand, the purification procedure for theshort-length PCLs w as carried out according to the schemeshown in Fig. 2, b and c (lanes 2-4). The soluble fractionobtained on centrifugation was purified by affinity chro-matography on amylose resin (lane 2). After proteolysiswith factor Xa (lane 3), th e short-len gth PCLs were purifiedby ion-exchange chromatography on a Mono-Q column atpH 7.5 (lane 4). About 8 mg (P2CL) and 18 mg (P27CL) ofthe pu re sho rt-length PC Ls w ere obtained from 5 g of E.coli cells.Catalytic Activity of the Recom binant Rat CLThecatalytic activities of the two short-length PCLs and thefull-length one autoprocessed under acidic conditions (pH5.5) were measured against Z-Phe-Arg-MCA, an optimalsynthet ic substrate for CL, according to the method ofBarre t t and Kirschke (1). The results are given in Table I.In contrast w ith the full-length PCL , the short-length PCLs(both P2CL and P27CL) exhibited a great loss of activity.Since no notab le difference w as observed in the PAG E bandposi t ion b etween th e sho rt-length PC Ls isolated at pH 7.7and processed at pH 5.5, this m ay be due to that auto-processing had not taken place, as judged from the resultsof N-terminal amino acid analysis, although the pH-profileof activity (Fig. 3) showed signs of active CL (9).CD SpectraThe CD spectra of the three kinds of PCLsare shown in Fig. 4; nearly the same spectra were obtainedfor the respective autoprocessed PCLs. It is obvious thatthe tert iary structures of the two short-length PCLs,al though they have similar structures, are quite differentfrom that of the ful l-length PCL. The lat ter spectrumwould reflect the active stru ctu re of CL, because almost thesame spectrum has been reported for the act ive form ofrecombinant human ful l-length PCL (C25S mutant) {15).

DISCUSSIONIt has been reported that the propeptide region of PCLfunctions as a regulator of the catalytic activity and stabil-i ty of CL in nonlysosomal compartments (16, 17), and isessential for the proper folding of CL in mammalian cells(12) . On the other h and, i t has also been reported tha t thepropeptide is required for the renaturat ion of insolublehuman CL expressed in E. coli (9). However, i t remainsunclear at which step the pro peptide functions impo rtantlyin the processing of active CL, i.e., the recombinant proteinexpression, the renatu rat ion, or the autoprocessing of PCL.Therefore, we have at tempted the preparat ion of pro-region-al tered re com binant PC Ls and their autoprocessingto CL. With reference to the crystal structure of humanrecombinant ful l-length PCL (18), several proregion sizesof PCL were considered to correspond to the functionalstructural fragments, and three proregion-al tered cDNAswere finally used in this study, based on the intended

expression conditions. The full-length P9 6CL was exp ress-ed as an insoluble fusion with GST, which necessitatedsolubilization and consequent careful refolding to preparethe active enzyme. On the other hand, two short-lengthPCLs, P2CL and P27CL, were both secreted as solubleforms, which could be much more advantageous for prepa-ration of the active enzyme than the insoluble form.The present results showed that the full-length PCL,prepared by the usual renaturation operation, could beautoprocessed to the active CL under acidic conditions,while neithe r of the two short-leng th PCLs showed rem ark-able CL activity under the same conditions, in spite of theadvantage of their soluble expression. On the other hand,the CD spectra showed the considerable difference betweenthe tert iary structures of the two short-length PCLs and thefull-length one. Judging from these results, it is obviousthat the full-length propeptide region of PCL is necessaryfor the proper protein folding (including the formation ofthe proper secondary structures and their correct mutualorientation) at stage of the recombin ant protein expressionin E. coli, which leads to autoprocessing to active matureCL, although the importance of the propeptide at the stepof renatu ration of this insoluble full-length PCL could n otbe ruled out. On the other hand, it can be said from thepresent results that the propeptide moiety does not affectthe expression of the recom binant protein itself apar t fromthe active/inactive form.

The crystal structure of human recombinant full-lengthPC L (18) showed that the propeptide consists of twostructural components: a long, extended portion (1-20sequence), and a globular domain comprising sequence 25-96 (three a-hel ices, 21-28, 45-71 , and 77-90, and anextended connecting loop, 37-40). The 17-25 region islocated in the active-site region of CL, where the extende dsequence lines the cleft along the S and S ' subsites and thecontacts are extensive . Judging from the pres ent resu lts forP27CL, it can be said that the length of the propeptide,which covers th e whole active site cleft of CL, is not enoughfor the tertiary structure of CL to be orientated correctly,and the globular domain consisting of propep tide sequence25-96 is indispensable for construction of the CL activestructure at the recombinant protein expression of itscDNA in E. coli, meaning that this globular domain func-tions as a stabilizer and /or pro mo ter for the prop er foldingof CL. Interestingly, this discussion appears to agree wellwith results of CL inhibition by the propeptide fragment(19), i.e., the propeptide segment (sequence 16-69), whichlacks residues 1-15 of the PCL propeptide, exhibits goodinhibitory activity, while the peptide sequence 1-44 con-tributes little to the inhibition.We wish to thank Dr. Hiroshi Miyoshi and Prof. Kin-ichiro Miura,Institute for BLmolecular Science, Faculty of Science, GakushuinUniversity, for the valuable discussions.

REFERENCES1. Barrett, A. and Kirschke, H. (1981) Cathepsin B, cathepsin H,and cathepsin L. Methods Enzymol. 80, 535-5612. Mason, R.W., Johnson, D.A., Barrett, A.J., and Chapman, H.A.(1986) Elastinolytic activity of human cathepsin L. Biochem. J.233, 925-9273. Kirschke, H., Kembhavi, A.A., Bohley, P., and Barrett, A.J.(1982) Action of rat liver cathepsin L on collagen and other

J. Biochem.


6/6

83substrates. Biochem. J. 201, 367-3724. Nakagawa, T., Roth, W., Wong, P., Nelson, A., Farr, A.,Deussing, J., Villadangos, J.A., P loegh, H., Pe ters, C , andRudensky, A.Y. (1998) Cathepsin L : critical role in Ii degradationand CD4 T cell selection in the thy mus. Science 280, 450-4535. Katunuma , N . and Kominami, E. (1987) Distribution and locali-zations of lysosomal cysteine proteinases and cystatins. Rev.PhysioL Biochem. Pharmacol 108, 1-206. Yagel, S., Warner, A.H., Nellans, H.N., Lala, P.K., Waghorne,C , and Denhardt, D.T. (1989) Suppression by cathepsin Linhibitors of the invasion of amnion mem branes by m urine cancercells. Cancer Res. 49, 3553-35577. Kakegawa, H., Nikawa, T., Tagami, K., Kamioka, H., Sumitani,K., K awata, T., Drobnic-Kosorok, M ., Lenarcic, B., Turk, V., andKatunuma, N. (1993) Participation of cathepsin L on boneresorption. FEBS Lett. 321, 247-2508. Traba ndt, A., Aicher, W.K., Gay, R.E., Sukhatme, V .P., Nilson-Hamilton, M., Himilton, R.T., McGhee, J.R., Fassbender, H.G.,and Gay, S. (1990) Expression of the collagenolytic and Ras-induced cysteine proteinase cathepsin L and proliferation-asso-ciated oncogenes in synovial cells of MRL/I mice and patientswith rheumatoid. Matrix 10, 349-3619. Smith, S.M. and Gottesman, M.M. (1989) Activity and deletionanalysis of recombinant human cathepsin L expressed in Escheri-chia coli. J. Biol. Chem. 264, 20487-20495

. Dolinar, M., Maganja, D.B., and Turk, V. (1995) Expression offull-length human procathepsin L cDNA in Escherichia coli andrefolding of the expression product. Biol. Chem. Hoppe-Seyler376, 385-388. Nomura, T., Fujishima, A., and Fujisawa, Y. (1996) Characteri-zation and crystallization of recombinant human cathepsin L.

Biochem. Biophys. Res. Commun. 22S, 792-79612. Tao, K., Steams, N.A., Dong, J., Wu, Q., and Sahagian, G.G.(1994) The proregion of cathepsin L is required for properfolding, sta bility, and ER exit. Arch. Biochem. Biophys. 3 1 1 ,1 9-2713. Ishidoh, K., Towatari, T., Imajoh, S., Kawasaki, H., Kominami,E., Katunum a, N., and S uzuki, K. (1987) M olecular cloning andsequencing of cDNA for rat cathepsin L. FEBS Lett. 223, 69-7314. Morrison, D.A. (1979) Transformation and preservation ofcompetent bacterial cells by freezing. Methods Enzymol. 6 8 ,326-33115. Menard, R., Carmona, E., Takebe, S., Dufour, E., Plouffe, C ,Mason, P., and Mort, J.S. (1998) Autocatalytic processing ofrecombinant human procathepsin L. J. Biol Chem. 273, 4478-448416. Mason, R.W., Gal, S., and Gottesman, M.M. (1987) The iden-tification of the major excreted protein (MEP) from a transform-ed mouse fibroblast cell line as a catalytically active precursorform of cathepsin L. Biochem. J. 248, 449-45417. Mason, R.W., W ilcox, D., Wilkstrom, P., and Shaw, E.N . (1989)The identification of active forms of cysteine proteinases inKirsten-virus-transform ed mouse fibroblasts by use of a specificradiolabelled inhibitor. Biochem. J. 257, 125-12918. Coulombe, R., Grochulski, P., Sivaraman, J., Menard, R., Mort,J.S., and Cygler, M. (1996) Structure of human procathepsin Lreveals the molecular basis of initiation by the prosegment.EMBO J. 15 , 5492-550319. Carmona, ., Dufour, E., Plouffe, C, Takebe, S., Mason, P.,Mort, J.S ., and M enard, R. (1996) Potency and selectivity of thecathepsin L propeptide as an inhibitor of cysteine proteases.Biochemistry 35 , 8149-8157

Documents

Function of the Pro Pep Tide Region in ant Expression of Active sin L in Escherichia Coli J Biochem-1999-Ogino-78-83