to BioTechniques free email alert service to receive content updates.
Engineering a novel endopeptidase based on SARS 3CLpro
 
Chih-Jung Kuo1,2, Yan-Ping Shih1, Daphne Kan1, and Po-Huang Liang1,2
1Institute of Biological Chemistry, Academia Sinica and Core Facility of Recombinant Protein Production, Taipei, Taiwan
2Institute of Biochemical Sciences, National Taiwan University, Taipei, Taiwan
BioTechniques, Vol. 47, No. 6, December 2009, pp. 1029–1032
Full Text (PDF)

Introduction

Severe acute respiratory syndrome–coronavirus (SARS-CoV) caused an outbreak in 2003 that killed approximately 800 patients worldwide (1). A 3C-like protease from the virus, 3CLpro, is required to cleave 11 sites of the polyproteins pp1a (486 kDa) and pp1ab (790 kDa) for their maturation (2). 3CLpro is a chymotrypsin-like protease, but it uses Cys as a nucleophile for catalysis (3). Analogous to 3C proteases of picornaviruses, 3CLpro has substrate specificity in cleaving the amide bond between P1-Gln and a small amino acid such as Ser, Ala, or Gly at P1′ (4,5). As evident in its 3-D structure (6,7), this small P1′ residue is near Thr25, which likely determines the substrate specificity.

It was previously demonstrated that the recombinant SARS 3CLpro can undergo auto-processing (7,8), which indicates its potential as a tag-cleavage endopeptidase. However, it would need to be capable of cleaving Q↓M, since Met is the most common first residue at protein N-termini. In this study, we replaced the 3CLpro Thr25 with the smaller Gly residue to expand the S1′ site and found that the mutant protease cleaved peptides with larger amino acids such as Met at P1′ with high efficiency. The results presented here demonstrate that Thr25 is essential to determine P1′ substrate specificity and that the T25G mutant can be used as a novel endopeptidase for tag cleavage of recombinant fusion proteins in addition to the commonly used thrombin, Factor Xa (FXa), and tobacco etch virus protease (TEVpro). Moreover, we have constructed two vectors, using prokaryotic and eukaryotic hosts, which contained the nucleotides encoding the T25G recognition site AVLQ↓M between the tags and the N-terminal Met of the target proteins. In these vectors, PstI (CTGCAG) was chosen as a 5′ cloning site, since its sequence overlapped the nucleotide sequence (GCGGTGCTGCAG) encoding the protease recognition site. Identical 5′-PstI/3′-XhoI cloning sites in these vectors were used to allow sticky-end DNA fragments of the target genes generated by PCR (9) to ligate with these vectors simultaneously in a strategy called parallel cloning (10,11). These vectors, in conjunction with the T25G protease, provide new tools for convenient protein production in different hosts and tag cleavage to yield recombinant proteins with authentic sequences.

Materials and methods

Expression and purification of mutant 3CLpro

Expression and purification of wild-type and mutant SARS 3CLpro in Escherichia coli was accomplished according to reported procedures (12). T25G and T25S mutants were prepared from the wild-type by using the QuickChange site-directed mutagenesis kit (Cat. no. 200518; Stratagene, La Jolla, CA, USA). C-terminally His-tagged T25G was expressed using pET16b vector (Cat. no. 69662; Novagen, Darmstadt, Germany).

Construction of the expression vectors for producing tag-cleavable fusion proteins in E. coli and yeast

The UPPs-encoding gene (13) was employed as a template for PCR using primers containing the nucleotides encoding the T25G 3CLpro recognition site AVLQ, and the TEVpro recognition site EDLYFQ, respectively. The PCR products were purified from an agarose gel following electrophoresis and cloned into the pET32Xa/Lic vector (Novagen). To serve as a control, the UPPs fusion protein with AAAQ instead of AVLQ was also expressed.

For expressing EGFP fusion proteins in yeast, primers were used to generate a PCR product that was ligated into pHTPY7, which was modified from pPICZαA (Invitrogen) by incorporating nucleotides encoding a starch binding domain (SBD) (14) and AVLQ cleavage site.

Evaluation of tag removal by the proteases

The UPPs and EGFP fusion proteins were purified using NiNTA columns. To examine the tag cleavage reactions, the purified fusion proteins (5.4 µM each) were treated with 0.1 µM wild-type and two mutant (T25G and T25S) 3CLpro for 90 min at 37°C. For time course measurements, the fusion proteins (5.4 µM each) were treated with T25G (0.1 µM) at 37°C. The reactions were stopped by 2% trifluoroacetic acid after appropriate time periods and analyzed by SDS-PAGE. For comparing the tag cleavage efficiency of T25G and TEVpro (Invitrogen), the fusion proteins (Tags-AVLQ-UPPS and Tags-ENLYFQ-UPPS, 5.4 µM each) were treated with 0.1 µM T25G and TEVpro at 37°C, respectively and then analyzed by SDS-PAGE.

Substrate specificity and kinetic parameters of the mutant SARS 3CLpro

The peptides used as substrates for the T25G protease were synthesized via solid phase, using a 433A peptide synthesizer (Applied Biosystems, Foster City, CA, USA). Each peptide (100 µM) was incubated with 0.1 µM T25G for 1, 2, and 6 h, and the subsequent mixtures were analyzed by HPLC on a C-18 reverse-phase analytic column. Cleavage products were resolved using a 30-min, 2–90% linear gradient of acetonitrile plus 0.1% TFA. The product peak areas were integrated to calculate the reaction rates for each peptide substrate. For Km and kcat measurements, 0.1 µM T25G and 10–200 µM SAVLQ↓MGFRK substrate were used, and the plot of initial rates within 10% substrate consumption versus different substrate concentrations was fitted to the Michaeli-Menten equation using the KaleidaGraph computer program (Synergy Software, Reading, PA, USA).

  1    2