Protein Engineering vol.5 no.5 pp.455-459, 1992

Active papain renatured and processed from insoluble recombinant propapain expressed in Escherichia coli

Mark A.J.Taylor, Kathryn A.Pratt, Dean F.ReveU, Kenneth C.Baker, Ian G.Sumner and Peter W.Goodenough AFRC, Institute of Food Research, Earley Gate, Whiteknights Road, Reading, RG6 2EF, UK

Introduction The reaction mechanism of cysteine proteinases is characterized by unique shifts in the pKa of the two amino acids, cysteine and histidine. These take part in the acylation and deacylation of the substrate. The electrostatics of the active site cleft control these pKa shifts (Pickersgill, 1988; PickersgiU et al., 1988, 1989). Charged side chains of residues close to the histidine and cysteine have been altered by site-directed mutagenesis to allow their contribution to the electrostatics of the active site to be quantified (Me"nard et al., 1990, 1991a-c). However, the most detailed analysis of electrostatic interactions awaits the availability of a 3-D structure for these mutants. To date sufficient material has not been obtained for these studies. Unlike the serine proteinases such as subtilisin (Dcemura et al., 1987) and the aspartic proteinase cathepsin D (Conner and Udey, 1990), which have both been cloned and expressed at high levels in Escherichia coli, there are only a few examples from the whole class of cysteine proteinases which have been expressed even at low levels in E.coli [rat and mouse cathepsin B (Mort et al., 1988), human cathepsin B (Chan and Fong, 1988), human cathepsin H (Fuchs et al., 1989), human cathepsin L (Smith and Gottesman, 1989) and papain (Vernet et al., 1989; Cohen et al., 1990)]. Only small quantities of these examples of expressed enzymes were detected. There are several possible reasons for this: (i) the

Materials and methods Restriction- and DNA-modifying enzymes were obtained from Gibco, Uxbridge, UK. E64, Leupeptin and Pepstatin were obtained from the Peptide Institute, Osaka, Japan. L-Pyroglutamyl-L-phenylalanyl-L-leucine-p-nitroanilide (Pyr-Phe-LeupNA) was acquired from Bachem, Bubendorf, Switzerland. Prestained protein molecular weight standards were supplied by Bethesda Research Laboratories, Uxbridge, UK. All other reagents were of the highest purity available. 455

Downloaded from http://peds.oxfordjournals.org/ at New York University on June 9, 2015

For the first time the pro-form of a recombinant cysteine proteinase has been expressed at a high level in Escherichia coli. This inactive precursor can subsequently be processed to yield active enzyme. Sufficient protein can be produced using this system for X-ray crystallographic structure studies of engineered proteinases. A cDNA clone encoding propapain, a precursor of the papaya proteinase, papain, was expressed in E.coli using a T7 polymerase expression system. Insoluble recombinant protein was solubilized in 6 M guanidine hydrochloride and 10 mM dithiothreitol, at pH 8.6. A protein—glutathione mixed disulphide was formed by dilution into oxidized glutathione and 6 M GuHCl, also at pH 8.6. Final refolding and disulphide bond formation was induced by dilution into 3 mM cysteine at pH 8.6. Renatured propapain was processed to active papain at pH 4.0 in the presence of excess cysteine. Final processing could be inhibited by the specific cysteine proteinase inhibitors E64 and leupeptin, but not by pepstatin, PMSF or EDTA. This indicates that final processing was due to a cysteine proteinase and suggests that an autocatalytic event is required for papain maturation. Key words: cDNA/£.co//7papain/precursor/recombinant

eukaryotic protein signal sequences, including those of cysteine proteases such as cathepsin B and papain, have been shown to be highly toxic to E. coli (Mort et al. ,1988; Vemet et al., 1989); (ii) the renaturation conditions used were incorrect; or (iii) it is possible that the pro-region has to be present as a third domain of the enzyme and is essential for correct folding, in a manner similar to that seen for subtilisin E (Dcemura et al., 1987; Ohta etal, 1991). Rat cathepsin B has been expressed in sufficient quantities in yeast for protein crystals to be grown (Lee et al., 1990). However, the protein was incorrectly processed by yeast maturation proteases to give enzymes with a six amino acid N-terminal extension, and the possibility of an extra six residues on the C-terminus. Human cathepsin L has been successfully expressed in E.coli (Mort et al., 1988) but only small quantities of correctly refolded, i.e. active, cathepsin L were recovered. Mort etal. (1988) reported that this protein was less stable than authentic human enzyme. Cohen et al. (1990) reported the expression of inactive papain in E. coli. Multiple forms of the enzyme were detected by Western blotting. Vernet etal. (1990), Menard etal. (1991a-c), Khouri et al. (1991) and Vernet et al. (1991) used the baculovirus cell culture expression system to express propapain and a series of mutants. The nucleic acid sequence they used coded for die entire preproprotein. During secretion, the first 26 amino acid residue pre-sequence was removed. In all cases only relatively low levels of protein were obtained. However, it must be assumed that the proenzyme was correctly folded, because cysteine proteinasedependent cleavage of the precursor yielded active papain. We have expressed propapain employing a similar system, in vitro using Spodopterafrugiperda(S/9) cells and in vivo with Heliothis virescens larvae, and have successfully recovered similar levels of activatable propapain in both cases. hi this study we report the development of an expression system for propapain in E. coli without the 26 residue signal sequence and demonstrate that insoluble recombinant protein can be unfolded and refolded in vitro. The product can be further autocatalytically processed to give active papain which is comparable both kinetically and by SDS-PAGE to naturally occurring papain. The system reported in this study produces ~ 3 mg of processed active papain per litre of E.coli culture broth. This is sufficient to allow crystallographic analysis of the structure and as such constitutes a significant step forward.

M.A J.Taylor et al.

Protein purification and analysis Cultures were grown in LB broth at 37°C with ampicillin at 50 /ig/ml. Chloramphenicol was used in addition at 25 ng/ml for cultures carrying pLysS. Cultures were incubated until an A^ of 0.25 was reached, then induced by addition of IPTG (isopropyl /3-r>thiogalactopyranoside) to 0.4 mM. Cells were harvested 3 h post-induction. Samples for SDS-PAGE were centrifuged, cell pellets resuspended in gel-loading buffer (Laemmli, 1970) and heated at 95°C for 2 min. Debris was removed by centrifugation and the supernatants analysed by SDS-PAGE onto 12% gels (Laemmli, 1970). Cell pellets obtained from large scale batch cultures (21) were resuspended in 50 mM NaH2PO4, pH 8.0, containing 50 mM NaCl, 1 mM EDTA and 0.5 mM phenymethylsulphonyl fluoride (PMSF). Then they were lysed by three passages through a French press (SLM AMINCO, Urbana, IL). Insoluble material was collected by centrifugation at 15 000 g for 15 min then washed with the same buffer and stored at 4°C. 5

-GGATCCATTAATGGATTTTTCTATTGTGGGTTATTC-3

Solubilization and renaturation Recombinant inclusion body material was solubilized at 10 mg/ml in 100 mM Tris—acetate, pH 8.6, containing 6 M guanidine hydrochloride and 10 mM dithiothreitol at 37°C for 1 h. Solubilized protein was then diluted 10-fold into 0.5 M Tris-acetate, pH 8.6, containing 6 M guanidine hydrochloride and 0.1 M oxidized glutathione, and incubated at 4°C for 24 h. It was then diluted a further 100-fold into 100 mM Tris-acetate, pH 8.6, containing 3 mM L-cysteine, and incubated for a further 24 h at 4°C. Activation of papain Refolded propapain was concentrated 100-fold by nitrogen pressure dialysis in an Amicon stirred cell over a YM10 membrane. Propapain was autocatalytically processed to mature papain in 100 mM acetate, pH 4.0, for 30 min at 40°C or 10 min at 60°C, in the presence of 20 mM cysteine. Enzyme activity was determined by the release of p-nitroaniline from Pyr-PheLeu-pNA at 37°C and pH 7.0 in flat-bottomed ELISA plates. Released p-nitroaniline was measured in an Anthos 2000 Labtec plate reader at 405 nM. Kinetic determinations were made at pH 7.0 and 30°C in a Perkin Elmer Lambda 15 spectrophotometer at 410 run. Data were analyzed for Michalis —Menten constants using non-linear regression analysis (Enzfitter, Elsevier—Biosoft, Cambridge, UK). Inhibition of processing Propapain was pre-incubated with various inhibitors at room temperature for 10 min. Cysteine was added to 20 mM and the pH adjusted to 4.0 with acetic acid. The enzyme was then incubated for a further 10 min at 60°C. Compound E64, leupeptin, pepstatin and PMSF were used at 1 /*M and EDTA at 1 mM. To prevent further proteolysis in reduced sample buffer prior to SDS-PAGE, free cysteine residues were carboxymethylated with 1 mM iodoacetic acid. Results

Expression of recombinant propapain BL21(DE3) cells co-transformed with pET3aPP and the T7 lysozyme-producing plasmid pLysS, expressed propapain of the correct molecular weight as determined by SDS—PAGE (Figure 2). Cells transformed with pET3aPP alone grew poorly and no protein was detected by SDS-PAGE. An additional expression vector was constructed containing the full prepropapain sequence (construction not shown). Clones containing this construct and pLysS were unable to express prepropapain. No improvement was observed using pLysE, which produces higher levels of T7 lysozyme (Studier, 1991). Cultures expressing propapain were grown to an Ay^ of 0.25, then induced by addition of IPTG. Culture samples taken hourly post-induction were analysed by SDS-PAGE. The optimum induction period proved to be 3 h,

'

M D F S I V G Y S Q . . . . .CGCCCGGATCCATTAATGGATTTTTCTATTGTGGGTTATTCTCAA.... .GCGGGCCTAGGTAATTACCTAAAAAGATAACACCCAATAAGAGTT... Asnl

S S F Y P V K N • AGCTCATTCTATCCTGTTAAAAACTGAGATTCGAAGGGGATCCTCT. TCGAGTAAGATAGGACAATTTTTGACTCTAAGCTTCCCCTAGGAGA BamHl 3-GATAGGACAATTTTTGACTCTAAGCTT-S

Fig. 1. Protein and DNA sequence of the N- and C-terminal regions of the propapain sequence in pPropap3. Primers used to generate propapain sequence by PCR shown in italics.

456

Downloaded from http://peds.oxfordjournals.org/ at New York University on June 9, 2015

Plasmids and strains The expression vector pET3a and plasmids pLysS and pLysE were kindly supplied by W.Studier (Brookhaven, NY), as was the expression host strain BL21(DE3). The host strain 'SURE' was obtained from Stratagene (Cambridge, UK). Construction of expression plasmids mRNA was extracted from Carica papaya leaves using the method of Beck et al. (1978) and a cDNA library generated in Lambda ZAP (Stratagene, Cambridge, UK) using the cDNA synthesis system of Amersham (Aylesbury, UK). A full length clone encoding prepropapain was amplified from the cDNA library by PCR. The primers used in the PCR (5'-GGATCCGTCAACCATGGCTATGA-3' and 5'-TTCGAATCTCAGTTTTTAACAGGATAG-3') were designed using the cDNA sequence published by Cohen et al. (1990). The coding sequence of the clone obtained (pPap8) was identical to the published sequence. pPap8 was subjected to a second PCR to create pPropap3. pPropap3 lacks the first 26 codons of the coding sequence (Figure 1). The introduction of an initiation codon (ATG) next to residue 27 (Asn) created an Asnl restriction enzyme site which enabled it to be subcloned into pET3a. pPropap3 was digested with Asnl and BamHl, and the fragment released ligated into pET3a cut with Ndel and BamHl. Ligation products were transformed firstly into SURE cells and a suitable clone selected (pET3aPP). The expression host strain BL21 (DE3), which expresses T7 polymerase on induction with IPTG (Studier et al., 1990), was transformed with pET3aPP. To prevent residual T7 polymerase activity prior to induction, a second plasmid, pLysS, which carries the gene for T7 lysozyme (Studier, 1991) was introduced into the cells. The lysozyme is constitutively expressed at a low level and inhibits T7 polymerase activity. BL21(DE3) was also transformed with pET3aPP alone.

Recomblnant propapain expressed in E.coti

and subsequent cultures were induced for this period. The cells were harvested and lysed as detailed in Materials and methods. The over-expressed protein fractionated with the insoluble E. coli proteins on centrifugation, yielding a relatively pure product (Figure 2). Typically 2 1 of culture broth yielded - 6 0 0 mg wet wt of insoluble material. No propapain was found in the soluble fraction even when cultures were grown at 20°C (results not shown).

1O7K 69.3K

Discussion In vivo, cysteine proteinases, as with the other proteinase classes, are synthesized as inactive precursors as preproenzymes. The pre-sequence directs protein to specific cell targets. For example, cathepsin L is directed to lysosomes by the mannose 6-phosphate lysosomal recognition marker (von Figura and Hasilik, 1986). After cleavage of the pre-region, the remaining proenzyme undergoes further processing to mature forms. Previous attempts (Vernet et al., 1989; Cohen et al., 1990) to express papain in E.coli have met with only limited success, low levels of protein being detected from cultures that grew poorly. Vernet et al. (1989) showed that the pre-leader sequence impairs the synthesis of papain. They also expressed propapain as a /3galactosidase-propapain fusion protein, but were unable to report correct folding and disulphide bridge formation. By deleting the first 26 residues of the signal sequence and using the T7 polymerase system developed by Studier et al. (1990) we have succeeded in over-expressing propapain in E.coli. The propapain coding sequence was inserted downstream of the T7 promoter in the expression vector pET3a. Prevention of propapain expression before induction, using the T7 lysozyme plasmid pLysS, was critical in allowing propapain to be expressed in E. coli. We have also confirmed the observation of Vernet et al. (1989) that the signal sequence of cysteine proteinases inhibits expression in E.coli. Propapain is over-expressed in an insoluble form. This can be solubilized in guanidine hydrochloride and the intermolecular and intramolecular disulphide bonds reduced with dithiothreitol. The protein can then be oxidized with glutathione and allowed to refold in the presence of cysteine to give soluble propapain which can be further processed to give active papain.

45.8K E .§ X

28.7K

1 o 0

10

20

30

40

50

60

Time of activation (min).

18.2K

Fig. 3. Time course of activation of refolded propapain to active papain. Activation conditions were 100 mM acetate, pH 4.0, and 20 mM cysteine. Open circles, 40°C; closed squares, 60°C,

15.4K Table 1. Kinetic constants for native and recombinant papain Km (mM) Fig. 2. SDS-PAGE analysis of propapain expression. Total cell extracts from BL21(DE3)pLysSpET3a (lane 1) and BL21 (DE3)pLysSpET3aPP (lane 2), insoluble extract from BL21(DE3)pLysSpET3aPP (lane 3), stained with Coomassie brilliant blue R250, and prestained protein molecular weight standards (lane 4).

*«(s - ) (s""'•mM)

Native papain Recombinant papain

0.348 0.145

28.6 964

82 .18 66 .5

457

Downloaded from http://peds.oxfordjournals.org/ at New York University on June 9, 2015

Renaturation and processing of propapain Insoluble propapain was solubilized with 6 M guanidine HC1 at 10 mg/ml and 10 mM dithiothreitol. A mixed disulphide was formed by dilution of the protein into oxidized glutathione. The protein was finally refolded by dilution into 3 mM cysteine at pH 8.6 to yield soluble but inactive 39 kDa pro-protein. Propapain was concentrated 100-fold by pressure dialysis over a YM10 membrane and autocatalytically processed to active papain with a mol. wt of 24 kDa by adjusting the pH to 4.0 with acetic acid and heating in the presence of 20 mM cysteine. The majority of the papain activity was released in the first few minutes of incubation of propapain at either 40 or 60°C. Extended incubation times did not significantly increase the amount of papain activity measured (Figure 3). Typically 20 mg of moist insoluble E.coli pellet yielded —0.2 mg of active papain. Recombinant processed papain obtained by this method had comparable kinetic constants, using Pyr-Phe-Leu-pNA at pH 7.0, to those published for commercial plant papain (Filippova et al., 1984) and our own observations (Table 1).

Inhibition of processing The processing of propapain to papain was inhibited by the cysteine proteinase inhibitors E64, leupeptin and iodoacetic acid but not by pepstatin, PMSF or EDTA (Figure 4).

M.A J.Taylor et al.

1

2

107K 69.3K 45.8K

28.7K

18.2K

15.4K

Vernet et al. (1991) claimed that propapain produced in the baculovirus system was processed to active papain firstly by intramolecular autoprocessing and then by intermolecular processing when free papain was released. However it is possible that endogenous cysteine proteinases (cathepsins) from the insect cells could have initiated this process. It can be shown (M.A.J.Taylor and K.C.Baker, unpublished data) that Heliothis virescens larvae infected with a baculoviral prepropapain virus construct produce Pyr-Phe-Leu-pNA activity which can be separated into two peaks by cation exchange chromatography. Only one of these peaks of activity is enhanced by processing with 20 mM cysteine at pH 4.0 and 60°C. We suspect that the other peak of activity is due to enhanced catheptic activity brought about by virally induced cellular damage. While we do not maintain that the insoluble protein pellet produced by E. coli is totally free from endogenous proteinases, the E.coli system does produce an inherently purer protein than the baculovirus system in cell culture or entire caterpillars. We agree with Vernet et al. (1991) that the processing of propapain to papain is a cysteine proteinase-dependent event. For the intramolecular autoproteolytic event to occur, a small proportion of the propapain molecules must exist in an active conformation. We have included sub-saturating levels of the cysteine proteinase inhibitors E64 and leupeptin which prevent the intramolecular autoprocessing of propapain and presumably therefore the subsequent intermolecular processing cascade leading to the active species. An added complication to successful refolding of cysteine proteinases, over and above the refolding of other classes of enzymes, is the fact that by definition they have at least one free cysteine. Any refolding regime that produces activatable propapain or papain must produce three disulphide bridges from the correct pairs of cysteines and leave the active site cysteine 458

in a non-oxidized form, able to take part in catalysis. Kinetically, the activation of propapain will depend on the connectivity of these disulphide links generated during refolding. The method of heterologous expression described will enable sufficient mutant papain to be expressed for crystal growth. Analysis of the structures will enable definitive answers to be obtained about the electrostatic and hydrogen bonding interactions in this class of proteins. We are also currently examining the nature and kinetics of the folding of propapain. Acknowledgements We would like to thank other members of our Protein Engineering Department for many useful discussions, in particular B.Fischer, R Pickersgill and LConnerton.

References Beck.F., Sommer,R., Auerswald.E.A., Kurz.Ch., Osterburg,G. and Schaller.H. (1978) Nucleic Acid Res., 5, 4995. Chan,M.M. and Fong.D. (1988) FEBS Lett., 239, 219-222. Cohen.L.W., Fluharty,C. and Dihel,L.C. (1990) Gene, 88, 263-267. Conner.G.E. and UdeyJ.A. (1990) DNA Cell Biol., 9, 1-9. FilippovaJ.Y., Lysogorskaya,E.N., Oksenoit.E.S., Rudenskaya.G.N. and Stepanov.V.M. (1984) Anal. Biochem., 143, 293-297. Fuchs.R., Fisch.B. and Gassen.H.G. (1989) Biol. Chem. Hoppe-Seyler, 370, 899-900. Ikemura.H., Takagi.H. and Inouye,M. (1987)/. Biol. Chem., 262, 7859-7864. Khouri.H.E., Vernet,T., M6nard,R., Parlati.F., LafIamme,P., Tessier.D.C., Gour-Salin,B., Thomas,D.Y. and Storer.A.C. (1991) Biochemistry, 30, 8929-8936. Laemmli.U.K. (1970) Nature, 227, 680-685. Lee.X., Ahmed.F.R., Hirama.T., Huber.C.P., Rose,D.R., T?,R., Hasnain.S., Tam.A. and Mort.J.S. (1990) J. Biol. Chem., 265, 5950-5951. Menard.R., Khouri,H.E., Plouffe.C, Dupras.R., Ripoll.D., Vernet.T., Tessier.D.C, Laliberte\F., Thomas.D.Y. and Stora\A.C. (1990) Biochemistry, 29, 6706-6713. M6nard,R., Khouri.H.E., Plouffe.C., Laflamme,P., Dupras.R., Vernet.T., Tessier.D.C., Thomas,D.Y. and Storer.A.C. (1991a) Biochemistry, 30, 5531-5538.

Downloaded from http://peds.oxfordjournals.org/ at New York University on June 9, 2015

Fig. 4. SDS—PAGE analysis of inhibition of proteolytic processing of refolded recombinant propapain. Proteins were stained with Coomassie brilliant blue R250. The lanes show prestained protein molecular weight standards (lane 1), propapain before activation (lane 2), activation of an —15 ^M solution of propapain in the absence of inhibitor (lane 3), and activation after pre-incubation with 1 yM compound E64 (lane 4), 1 fiM leupeptin (lane 5), 1 mM iodoacetic acid (lane 6), 1 /iM pepstatin (lane 7), 1 jiM PMSF (lane 8) and 1 mM EDTA (lane 9).

Recombinant propapain expressed in E.coli

Downloaded from http://peds.oxfordjournals.org/ at New York University on June 9, 2015

MiJnard.R., CarriereJ., Laflamme.P., Plouffe,C, Khouri.H.E., Vernet.T., Tessier.D.C, Thomas.D.Y. and Storer.A.C. (1991b) Biochemistry, 30, 8924-8928. M£nard,R., Plouffe.C, Khouri.H.E., Dupras.R., Tessier.D.C., Vemet.T., Thomas,D.Y. and Storer.A.C. (1991c) Protein Engng, 4, 307-311 MortJ.S., Tam,A., Steiner.D.F. and Chan.SJ. (1988) Biol. Chem. HoppeSeyler, Suppl 369, 163-167. Ohta.Y., Hojo.H., Aimoto.S., Kobayashi.T., Zhu,X., Jordan.F. and Inouye.M. (1991) Mol. Microbiol, 5, 1507-1510. Pickersgill.R.W. (1988) Protein Engng, 2, 247-248. Pickersgill.R.W., Goodenough.P.W., Sumner.I.G. and Collins.M.E. (1988) Biochem. J., 254, 235-238. Pickersgill.R.W., Sumner.I.G., Collins.M.E. and Goodenough.P.W. (1989) Biochem. J., 257, 309-312. Smith.S.M. and Gottesman.M.M. (1989) J. Biol. Chem., 264, 20487-20495. Studier,F.W., Rosenberg.A.H., DunnJJ. and DubendorffJ.W. (1990) Methods Enzymoi, 185, 60-89. Studier.F.W. (1991) J. Mol. Biol., 219, 37-44. Vernet.T., Tessier.D.C., Laliberte'.F., Dignard.D. and Thomas.D.Y. (1989) Gene, 77, 229-236. Vemet.T., Tessier.D.C., Rkhardson.C, Lalibert^.F., Khouri.H.E., Bell.A.W., Storer.A.C. and Thomas.D.Y. (1990)7. Biol. Chem., 265, 16661-16666. VernetJ"., Khouri.H.E., Laflamme,P., Tessier.D.C., Musil.R., Gour-Sahn.B., Storer.A.C. and Thomas.D.Y. (1991)7. Biol. Chem., 266, 21451-21457. von Figura.K. and Hasilik.A. (1986) Annu. Rev. Biochem., 55, 167-193. Received on March 3, 1992; revised and accepted on May 13, 1992

Note added in proof Ealdn a al. (1992) haverecentlyreportedthat cysteine proteinases from protozoan, mammalian and plant species have been shown to be expressible in E.coli in a form suitable for further processing to yield active enzyme.

Reference Eakin.A.E., Mills.A.A., Harth.G., McKerrowJ.H. and Craik.C.S. (1992) J. Biol. Chem., 261, 7411-7420.

459