Portal de Periódicos da CAPES

Identification of the Amino Acid Residue Involved in Rabbit Hemorrhagic Disease Virus VPg Uridylylation

2001; Elsevier BV; Volume: 276; Issue: 30 Linguagem: Inglês

10.1074/jbc.m100707200

1083-351X

A. Machin, José M. Martín‐Alonso, Francisco Parra,

Hepatitis B Virus Studies

The virus genome-linked protein (VPg) coding region from rabbit hemorrhagic disease virus (RHDV) (isolate AST/89) was expressed in Escherichia coli by using a glutathione S-transferase-based vector. The recombinant polypeptide could be purified in good yields and was uridylylatedin vitro from [α-32P]UTP in a reaction catalyzed by the recombinant RNA-dependent RNA polymerase from RHDV in the absence of added template RNA. The use of deletion and point mutants allowed the identification of Tyr-21 as the residue involved in uridylylation and consequently in the linkage between VPg and the viral genome. These data constitute the first report on the identity of the amino acid residue involved in VPg uridylylation in a member of the Caliciviridae family. The virus genome-linked protein (VPg) coding region from rabbit hemorrhagic disease virus (RHDV) (isolate AST/89) was expressed in Escherichia coli by using a glutathione S-transferase-based vector. The recombinant polypeptide could be purified in good yields and was uridylylatedin vitro from [α-32P]UTP in a reaction catalyzed by the recombinant RNA-dependent RNA polymerase from RHDV in the absence of added template RNA. The use of deletion and point mutants allowed the identification of Tyr-21 as the residue involved in uridylylation and consequently in the linkage between VPg and the viral genome. These data constitute the first report on the identity of the amino acid residue involved in VPg uridylylation in a member of the Caliciviridae family. rabbit hemorrhagic disease virus RNA-dependent RNA polymerase glutathione S-transferase matrix-assisted laser desorption/ionization time-of-flight mass spectrometry magnesium acetate virus genome-linked protein polymerase chain reaction polyacrylamide gel electrophoresis Rabbit hemorrhagic disease virus(RHDV),1 a member of the Caliciviridae family, recently designated as the type species of the genus Lagovirus (1Pringle C.R. Arch. Virol. 1998; 143: 1449-1459Crossref PubMed Scopus (183) Google Scholar), has been identified as the causative agent of a lethal pathology in rabbits (2Ohlinger V.F. Haas B. Meyers G. Weiland F. Thiel H.J. J. Virol. 1990; 64: 3331-3336Crossref PubMed Google Scholar, 3Parra F. Prieto M. J. Virol. 1990; 64: 4013-4015Crossref PubMed Google Scholar). The viral genome is a positive polarity single-stranded polyadenylated RNA molecule of ∼7.4 kilobases, with a virus genome-linked protein (VPg) covalently attached to its 5′-end (4Meyers G. Wirblich C. Thiel H.J. Virology. 1991; 184: 677-686Crossref PubMed Scopus (160) Google Scholar). Viral particles also encapsidate an abundant VPg-linked polyadenylated subgenomic RNA of 2.2 kilobases (4Meyers G. Wirblich C. Thiel H.J. Virology. 1991; 184: 677-686Crossref PubMed Scopus (160) Google Scholar) that is collinear for its complete length to the 3′-terminal region of the genomic RNA. The data obtained from in vitro translation (5Wirblich C. Sibilia M. Boniotti M.B. Rossi C. Thiel H.J. Meyers G. J. Virol. 1995; 69: 7159-7168Crossref PubMed Google Scholar),Escherichia coli expression (6Martin Alonso J.M. Casais R. Boga J.A. Parra F. J. Virol. 1996; 70: 1261-1265Crossref PubMed Google Scholar, 7Wirblich C. Thiel H.J. Meyers G. J. Virol. 1996; 70: 7974-7983Crossref PubMed Google Scholar), detection of viral protein products after RHDV infection of cultured hepatocytes (8Konig M. Thiel H.J. Meyers G. J. Virol. 1998; 72: 4492-4497Crossref PubMed Google Scholar), and transient expression experiments (9Meyers G. Wirblich C. Thiel H.J. Thumfart J.O. Virology. 2000; 276: 349-363Crossref PubMed Scopus (94) Google Scholar) revealed that the viral RNA is translated into a polyprotein, which is subsequently cleaved to give rise to at least nine mature structural and nonstructural proteins. Seven of the eight necessary cleavage sites have been demonstrated experimentally. Amino acid sequencing studies carried out first in thePrimate calicivirus Pan-1 (10Dunham D.M. Jiang X. Berke T. Smith A.W. Matson D.O. Arch. Virol. 1998; 143: 2421-2430Crossref PubMed Scopus (32) Google Scholar) and more recently inFeline calicivirus (11Sosnovtsev S.V. Green K.Y. Virology. 2000; 277: 193-203Crossref PubMed Scopus (81) Google Scholar), together with point mutant studies performed in the RHDV system (9Meyers G. Wirblich C. Thiel H.J. Thumfart J.O. Virology. 2000; 276: 349-363Crossref PubMed Scopus (94) Google Scholar) have allowed the identification of the N-terminal sequence of a few calicivirus VPg proteins as well as the genome mapping of their coding regions. Nevertheless, since the first description of a VPg on a calicivirus RNA (12Burroughs J.N. Brown F. J. Gen. Virol. 1978; 41: 443-446Crossref PubMed Scopus (80) Google Scholar), functional studies on this viral protein and the identification of the amino acid residues putatively involved in RNA linkage are still lacking. In the poliovirus system, a closely related type of viruses, VPg has been found to be a 22-amino acid residue peptide that is genome-linked by a bond between the O4 of tyrosine and the 5′-P atom of the terminal uridylic acid residue (13Rothberg P.G. Harris T.J. Nomoto A. Wimmer E. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 4868-4872Crossref PubMed Scopus (160) Google Scholar). It has been recently found that poliovirus VPg uridylylation was mediated by the RNA polymerase 3Dpol of the virus in a poly(A)-dependent reaction (14Paul A.V. van Boom J.H. Filippov D. Wimmer E. Nature. 1998; 393: 280-284Crossref PubMed Scopus (299) Google Scholar, 15Paul A.V. Rieder E. Kim D.W. van Boom J.H. Wimmer E. J. Virol. 2000; 74: 10359-10370Crossref PubMed Scopus (239) Google Scholar). The VPg of RHDV is a larger protein of 115 amino acid residues showing no significant sequence homologies to that of polioviruses. Nevertheless, considering the similarities found between these two types of viruses concerning their genomic organization and catalytic properties of viral RNA-dependent RNA polymerases (16López Vázquez A. Martı́n Alonso J.M. Casais R. Boga J.A. Parra F. J. Virol. 1998; 72: 2999-3004Crossref PubMed Google Scholar, 17López Vázquez A. Martı́n Alonso J.M. Parra F. J. Virol. 2000; 74: 3888-3891Crossref PubMed Scopus (51) Google Scholar, 18López Vázquez A. Martı́n Alonso J.M. Parra F. Arch. Virol. 2001; 146: 59-69Crossref PubMed Scopus (20) Google Scholar), it is tempting to hypothesize that VPg could play a similar role in RHDV genome synthesis initiation and that a tyrosine residue could be involved in the VPg-genome linkage. The RHDV VPg coding region was cloned as a BamHI-EcoRI cassette by PCR amplification of RHDV cDNA using the sense oligonucleotide 3B5a (Table I) containing a BamHI restriction endonuclease site and the antisense oligonucleotide 3B3, which added a unique EcoRI site at the 3′-end of the amplified region. The 3B5a primer also included four nucleotide substitutions that cause a Val to Ser point mutation at residue 2 of the recombinant RHDV VPg. This was done to enable accurate thrombin cleavage of the GST-VPg fusion protein just before the authentic Gly N-terminal residue of VPg without the addition of unwanted amino acid extensions to the recombinant product. In this way a 351-base pair DNA fragment, corresponding to nucleotide residues 2988 to 3333 of the Spanish AST/98 RHDV isolate (GenBankTM/EMBL accession number Z49271), was amplified (Fig. 1A) and cloned into the pGEM-T vector (Promega). After digestion of the recombinant plasmid withBamHI and EcoRI, the resulting 351-base pair fragment was purified and inserted intoBamHI-EcoRI-digested expression vector pGEX-2T (Amersham Pharmacia Biotech) to generate the recombinant plasmid pGEX-VPg. This expression vector included the VPg coding region fused in frame to the 3′ end of the Schistosoma japonicumglutathione S-transferase (GST) gene.Table IOligonucleotides used for RHDV VPg cloning, mutagenesis and in vitro transcription of positive and negative genomic and subgenomic viral RNAOligonucleotideSequence1-aLowercase boldface characters indicate the mutations made with respect to the wild type RHDV AST/89 sequence (GenBank™/EMBL accession number Z49271). Lowercase characters represent nucleotide residues not related to RHDV. The T7 promoter sequence included in oligonucleotides RHDV 2, 3, and 4 is indicated in boldface. Restriction sites are underlined.3B5a5′-GGatccAAAGGCAAGACGAAACGTGG-3′3B35′-GAATTCACTCATAGTCATTGTCATAAAAGCC-3′Δ5VPg225′-GGATCCGACGAATGGCAAGCTGCACGCAG-3′Δ3VPg915′-GAATTCAGCGGATGACCTCGTTTCTCAC-3′3VPgYF5′-CAGCTTGCCATTCGTCAaACTCGTCATTGCCCAAG-3′3VPgYT5′-CAGCTTGCCATTCGTCAgtCTCGTCATTGCCCAAG-3′3VPgYS5′-CAGCTTGCCATTCGTCAgACTCGTCATTGCCCAAG-3′RHDV 15′-(T)25ATAGCTTACTTTAAACTATAAAC-3′RHDV 25′-ccGCGGCCGC TAATACGACTCACTATAGTGAAAATTATGGCGGCTATG-3′RHDV 35′-ccGCGGCCGC TAATACGACTCACTATAGTGAATGTTATGGAGGGCAAA-3′RHDV 45′-accGCGGCCGC TAATACGACTCACTATAggggcgg(T)23ATAGCTTACTTTAAAC-3′RHDV 55′-GTGAAAATTATGGCGGCTATGTCGCG-3′RHDV 65′-GTGAATGTTATGGAGGGCAAAGCCCG-3′1-a Lowercase boldface characters indicate the mutations made with respect to the wild type RHDV AST/89 sequence (GenBank™/EMBL accession number Z49271). Lowercase characters represent nucleotide residues not related to RHDV. The T7 promoter sequence included in oligonucleotides RHDV 2, 3, and 4 is indicated in boldface. Restriction sites are underlined. Open table in a new tab To obtain N- and C-terminal VPg deletion mutants, PCR amplifications were carried out using sense (3B5a and Δ5VPg22) and antisense (Δ3VPg91 and 3B3) oligonucleotides (Table I). The resulting DNA fragments were first cloned into pGEM-T vectors and then inserted intoBamHI-EcoRI digested pGEX-2T expression plasmids to generate the recombinant vectors pGEX-Δ5VPg and pGEX-Δ3VPg, respectively (Fig. 1 A). For point mutant construction directed to the VPg Tyr-21 residue, the BamHI-EcoRI fragment from pGEX-VPg plasmid was transferred into the BamHI-EcoRI sites of pBluescript SK(+) vector, originating recombinant plasmid pSKVPg. This construct was used to perform in vitrosite-directed mutagenesis using specific antisense mutagenic oligonucleotide primers (Table I) by the Chameleon double-stranded site-directed mutagenesis method (Stratagene) following the manufacturer's instructions. The resulting mutantBamHI-EcoRI DNA fragments were separately inserted into the BamHI-EcoRI-digested pGEX-2T expression vectors. Each of the mutated constructs was checked by nucleotide sequencing and then transformed into E. coli BL21 cells for point mutant VPg production. Positive and negative sense genomic and subgenomic RNAs were synthesized by in vitro transcription of purified PCR fragments to which a modified T7 promoter was added using the appropriate primers (Table I). Amplifications of viral RNA (RHDV isolate AST/89) were made using the Expand Long Template PCR System (Roche Molecular Biochemicals) following the manufacturer's protocol. The PCR fragments used to produce full-length positive (gRNA+) or negative sense genomic RNA (gRNA−) were made using primer pairs RHDV2/RHDV1 and RHDV4/RHDV5, respectively (Table I). The PCR amplicons used to produce positive (sRNA+) or negative sense subgenomic RNA (sRNA−) were made using RHDV3/RHDV1 and RHDV4/RHDV6 primer pairs, respectively (Table I). RNA transcripts were synthesized from the appropriate PCR amplicons using Ribomax, the large scale RNA production system fromPromega. The positive and negative transcripts of genomic or subgenomic length were analyzed by electrophoresis on agarose gels containing formaldehyde. The recombinant GST-VPg proteins were purified from E. coli BL21 bacterial lysates by affinity chromatography using a bulk GST purification module (Amersham Pharmacia Biotech) according to the manufacturer's instructions. The wild type or mutant VPg proteins were released from GST by thrombin cleavage and the purified recombinant proteins stored at −20 °C in 50 mmTris-HCl, pH 8.0, containing 150 mm NaCl and 0.25 mm EDTA. Purified VPg samples were analyzed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry and N-terminal sequencing. The wild type RHDV 3Dpol was produced in E. coli XL1-Blue from plasmid pGEX-3D as described previously (16López Vázquez A. Martı́n Alonso J.M. Casais R. Boga J.A. Parra F. J. Virol. 1998; 72: 2999-3004Crossref PubMed Google Scholar). The mutant D355N RNA polymerase was also produced in E. coli XL1-Blue as described elsewhere (17López Vázquez A. Martı́n Alonso J.M. Parra F. J. Virol. 2000; 74: 3888-3891Crossref PubMed Scopus (51) Google Scholar). SDS-PAGE analysis of the recombinant proteins was performed as described elsewhere (19Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207537) Google Scholar) using 15% polyacrylamide gels. The size markers used were the low molecular weigh markers (Amersham Pharmacia Biotech) and the kaleidoscope polypeptide standards (Bio-Rad). The protein concentration was determined by the Bio-Rad protein assay. The protocol used was based on that described previously for poliovirus VPg (14Paul A.V. van Boom J.H. Filippov D. Wimmer E. Nature. 1998; 393: 280-284Crossref PubMed Scopus (299) Google Scholar) with some modifications such as the use of MnCl2 instead of magnesium acetate (MgAcO) and the absence of an added RNA template. Briefly, the reaction mixtures (20 µl) usually contained 50 mm HEPES buffer, pH 7.5, 1 mm MnCl2, 100 µm UTP, 2 µCi of [α-32P]UTP (800 Ci mmol−1), and appropriate amounts of purified 3Dpol and VPg as indicated. Incubations were made at 30 °C, and the reactions were stopped at the indicated times (usually 60 min) by the addition of SDS-PAGE loading buffer and analyzed using polyacrylamide gels, which were finally dried and autoradiographed. [32P]UMP incorporated into the reaction products was quantified using an Instant-Imager (Packard Instrument Co.). To obtain large quantities of VPg we used the expression vector pGEX-2T, designed for inducible expression of foreign polypeptides in E. colias fusion proteins with GST. After induction with isopropyl-β-d-thiogalactopyranoside, a major protein band of about 38 kDa was observed in the induced extracts corresponding to the cells transformed with the recombinant vector pGEX-VPg (Fig.1 B, lane 1). The mass of this polypeptide was in the expected range for the fusion protein GST-VPg. The 38-kDa polypeptide was retained onto a glutathione-Sepharose column (Fig. 1 B, lane 2) further supporting the possibility that this band was indeed the fusion protein. The VPg moiety was released from the carrier protein byin situ thrombin proteolysis of the fusion protein adsorbed onto the column, showing the expected molecular mass of 13 kDa (Fig.1 B, lane 3). As mentioned under "Experimental Procedures," the recombinant polypeptide was almost identical in amino acid sequence to the authentic RHDV VPg, except that it contained a Val to Ser substitution, for accurate processing purposes, at the second residue from the N terminus of the RHDV mature VPg product. Nevertheless, the SDS-PAGE analysis of purified recombinant VPg consistently showed a double band pattern (Fig. 1 B, lane 3). This result was the consequence of a cleavage after the Lys-99 residue of recombinant VPg (data not shown) as deduced from the results of MALDI-TOF analysis performed on purified VPg samples. This cleavage was most probably caused by the presence of catalytic amounts of thrombin in the recombinant VPg preparations as a consequence of the purification procedure used. Point mutant substitutions directed to the Lys-99 residue did not completely abolished VPg processing (not shown) due to the appearance of secondary cleavage sites at alternative amino acid sequences of the recombinant products. Despite these results indicating some degree of C-terminal heterogeneity in the VPg preparations, this type of material was used in subsequent functional studies considering that VPg and its cleaved derivatives all sheared a common N-terminal sequence (not shown). This was a relevant finding considering that the residues putatively involved in VPg-genome linkage were thought to be located at the N-terminal region of VPg, based on sequence conservation among VPg from caliciviruses (20Boga J.A. Martı́n Alonso J.M. Marı́n M.S. Casais R. Vázquez A.L. Machı́n A. Parra F. Pandalay S.G. Recent Research Developments in Virology. Transworld Research Network, Trivandrum, India1999: 107-119Google Scholar). As a first step toward establishing a functional assay for RHDV VPg, we have used an uridylylation protocol described previously for poliovirus VPg (14Paul A.V. van Boom J.H. Filippov D. Wimmer E. Nature. 1998; 393: 280-284Crossref PubMed Scopus (299) Google Scholar). Major differences between the poliovirus protocol and the one used in this work were the use of MnCl2 instead of MgCl2 and the absence of an added poly(A) template in the reaction mixture. Under the conditions described, the recombinant RHDV VPg was radioactively labeled in the presence of [α-32P]UTP (Fig. 1 C, lane 4) and Mn2+ ions. The incorporation of labeled uridylyl residues into RHDV recombinant VPg in the absence of poly(A) RNA template was a major difference with respect to the results described previously in the poliovirus system in which the presence of poly(A) was an absolute requirement for VPg uridylylation (14Paul A.V. van Boom J.H. Filippov D. Wimmer E. Nature. 1998; 393: 280-284Crossref PubMed Scopus (299) Google Scholar). Our data also indicated that no radioactive products were detected in the absence of RHDV 3Dpol (Fig. 1 C, lane 5), or in the presence of the inactive 3Dpol D355N point mutant (17López Vázquez A. Martı́n Alonso J.M. Parra F. J. Virol. 2000; 74: 3888-3891Crossref PubMed Scopus (51) Google Scholar) (Fig. 1 C, lane 6), supporting the theory that the VPg uridylylation observed was catalyzed by the viral RNA-dependent RNA polymerase and was not due to some bacterial contaminant present in the purified VPg or polymerase preparations. We have also demonstrated that the radioactivity associated to VPg was due to a true incorporation of UMP and not to a phosphorylation event because of some residual γ-32P contaminant, as deduced from the lack of radioactivity linked to VPg in the samples in which [γ-32P]UTP was used (Fig.1 C, lanes 1–3) instead of [α-32P]UTP. The presence of divalent cations is a well known requirement forin vitro RNA polymerase activity. The standard RNA polymerase assay described previously (16López Vázquez A. Martı́n Alonso J.M. Casais R. Boga J.A. Parra F. J. Virol. 1998; 72: 2999-3004Crossref PubMed Google Scholar) contained 3 mmMgAcO, also showing that concentrations above 10 mmcompletely abolished 3Dpol activity. Our previous studies demonstrated that Mg2+ in the standard RHDV 3Dpol in vitro assays could be replaced by Mn2+ giving rise to comparable radioactive precursors incorporation into products, being 0.5 mm the optimal MnCl2 concentration (18López Vázquez A. Martı́n Alonso J.M. Parra F. Arch. Virol. 2001; 146: 59-69Crossref PubMed Scopus (20) Google Scholar). Considering that VPg uridylylation was dependent on the presence of RHDV 3Dpol, several divalent ion salts were used to investigate the metal requirements of the assay. Our results indicated that the uridylylation reaction was much more efficient in the presence of 1 mm MnCl2 and could also be detected in the presence of 3.5 mm MgAcO or CoCl2 (Fig.2A), although after a much longer exposure time. These results also showed that Mn2+could not be replaced by other ions such as Zn2+ (Fig.2 A, lane 4). It should be mentioned that several nucleic acid polymerases of different origins have been shown to be more efficient in the presence of Mn2+ instead of Mg2+ (21Brooks R.R. Andersen J.A. Biochem. J. 1978; 171: 725-732Crossref PubMed Scopus (4) Google Scholar, 22Huang Y. Beaudry A. McSwiggen J. Sousa R. Biochemistry. 1997; 36: 13718-13728Crossref PubMed Scopus (52) Google Scholar, 23Cirino N.M. Cameron C.E. Smith J.S. Rausch J.W. Roth M.J. Benkovic S.J. Le Grice S.F. Biochemistry. 1995; 34: 9936-9943Crossref PubMed Scopus (76) Google Scholar). In some cases this effect has been related to a relaxed specificity of the enzyme toward template (24Palmenberg A. Kaesberg P. Proc. Natl. Acad. Sci. U. S. A. 1974; 71: 1371-1375Crossref PubMed Scopus (36) Google Scholar) or nucleotide (22Huang Y. Beaudry A. McSwiggen J. Sousa R. Biochemistry. 1997; 36: 13718-13728Crossref PubMed Scopus (52) Google Scholar, 25Tabor S. Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4076-4080Crossref PubMed Scopus (258) Google Scholar) in the presence of Mn2+. More recently the preference for Mn2+over Mg2+ of the poliovirus 3Dpol has been attributed to a substantial reduction in the KMvalue of the enzyme for primer-template and to an increase in the number of the productive enzyme-primer-template complexes formed in the presence of this divalent ion (26Arnold J.J. Ghosh S.K. Cameron C.E. J. Biol. Chem. 1999; 274: 37060-37069Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar). Considering than in the poliovirus system VPg uridylylation was not observed in the absence of an RNA template and that the presence of some viral sequences, instead of a poly(A), could also additionally increase this reaction (15Paul A.V. Rieder E. Kim D.W. van Boom J.H. Wimmer E. J. Virol. 2000; 74: 10359-10370Crossref PubMed Scopus (239) Google Scholar), we investigated RHDV VPg uridylylation in the presence of poly(A) or synthetic genome RNA (gRNA) and subgenomic RNA (sRNA) of positive and negative polarity. The results obtained did not reveal a significant increase in VPg labeling (Fig.2 B), indicating that in the RHDV system, under the experimental conditions used, VPg uridylylation did not required the presence of an RNA template. The possibility that these results were due to a relaxed enzyme specificity as a consequence of the use of Mn2+ ions in the reaction mixture was also investigated. Nevertheless, a similar behavior was obtained in the presence of Mg2+ ions, although after much longer exposition times (not shown). The lack of a requirement for an RNA template under the conditions used for RHDV VPg uridylylation was also consistent with the finding that incorporation of the corresponding nucleotidyl residues into VPg could also be observed from any of the remaining three NTP precursors (not shown). Nevertheless, the results described in this work do not rule out the possibility that RHDV RNA-dependent RNA polymerase will eventually require a viral RNA template for optimal VPg uridylylation under improved reaction conditions. Considering these results, further biochemical characterization of the VPg uridylylation reaction was carried out using Mn2+ ions, in the absence of added RNA template. The amount of labeled VPg product synthesized in a 60-min reaction was a linear function of 3Dpol concentration from 0 to 10 µg of enzyme (data not shown). In the presence of 6 µg of 3Dpol and 2 µg of VPg, the rate of UMP incorporation into VPg was constant for about 20 min (Fig. 3A) and decreased at longer incubation times at 30 °C. Under the conditions used (6 µg of 3Dpol), the amount of UMP incorporation in 60 min was found to be a linear function of the VPg amount present in the reaction mixture up to 5 µg of the recombinant protein, showing saturation kinetics at higher concentrations of VPg (Fig.3 B). It should be noted that under these conditions the reaction yield was very low and only 0.3% of the input VPg was uridylylated. Considering that both recombinant VPg and 3Dpol were routinely purified using a buffer containing NaCl, the reaction mixture usually contained 6–15 mm NaCl, depending on the amount of protein used. Then, to optimize the uridylylation assay, we investigated the effects of increasing NaCl concentration in the reaction mixture. For this particular purpose both VPg and 3Dpol were purified using 50 mm Tris-HCl (pH 8.0) in the absence of added NaCl. Using this type of protein preparation, we found that uridylylation activity was negatively affected by the increasing amounts of NaCl (Fig. 3 C). The results indicated that maximum incorporation values were obtained in the presence of up to 15 mm NaCl. As a first approach to investigating the identity of the amino acid residues involved in VPg uridylylation, we constructed two VPg deletion mutants, which lacked 21 or 24 amino acid residues from the N or C-terminal regions of recombinant VPg, respectively (Fig. 1 A). The N-terminal deletion was made by PCR amplification using oligonucleotides Δ5VPg22 and 3B3 (TableI), whereas the C-terminal deleted VPg was made using oligonucleotides 3B5a and Δ3VPg91 (Table I). The resulting amplified DNA fragments were then cloned independently into pGEM-T vectors. After digestion of the recombinant pGEM-T derivatives with BamHI and EcoRI, the DNA inserts were purified and cloned into BamHI-EcoRI-digested pGEX-2T expression plasmid to generate the recombinant vectors pGEX-Δ5VPg and pGEX-Δ3VPg. Each of the expression vectors coding for the truncated forms of VPg were transformed into E. coliBL21 cells and used for the synthesis and purification of recombinant Δ5VPg and Δ3VPg proteins. To asses the overall purity and molecular size of the resulting recombinant polypeptides, equivalent amounts of each purified protein preparation were analyzed by SDS-PAGE, and the gels were stained with Coomassie Blue. The observed molecular masses of the deleted VPg proteins were 11 and 10.5 kDa, respectively (not shown). The uridylylation reactions performed using 2 µg of purified wild type VPg and the corresponding N- and C-terminal deleted forms indicated that Δ3VPg protein was fully active, whereas Δ5VPg protein, which lacked the first 21 N-terminal residues of wild type VPg, was not radioactively labeled in the presence of 3Dpol under standard reaction conditions (Fig.4A). These data indicated that the relevant residues involved in VPg uridylylation were located near or within the N-terminal deleted region of the protein. At this point it should be noted that in poliovirus the bond between VPg and the nucleotide was found to be O4-(5′-uridylyl)tyrosine (13Rothberg P.G. Harris T.J. Nomoto A. Wimmer E. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 4868-4872Crossref PubMed Scopus (160) Google Scholar, 27Ambros V. Baltimore D. J. Biol. Chem. 1978; 253: 5263-5266Abstract Full Text PDF PubMed Google Scholar). Therefore, it is tempting to predict that the tyrosine 21 located within the VPg N-terminal deletion could be the uridylylation acceptor residue of the protein. To confirm this hypothesis we made three point mutants in which Tyr-21 was substituted by Phe, a similar amino acid residue devoid of the acceptor hydroxyl group, or by Ser or Thr, to investigate the possibility that other hydroxyl amino acids could act as uridylyl acceptors. The method used to construct the mutant VPg proteins has been described under "Experimental Procedures." To assess the overall purity and molecular size of the resulting recombinant polypeptides, each purified protein preparation was analyzed by SDS-PAGE. Equivalent amounts of the wild type and point mutant VPg proteins were used in uridylylation assays under standard conditions. The results clearly indicated that none of the point mutant VPg were radioactively labeled (Fig. 4 B), thus supporting the belief that Tyr-21 was the acceptor residue, which could not be substituted by any other hydroxyl amino acid. These data constitute the first experimental evidence on the identity of the amino acid residue involved in VPg-genome linkage in a member of the Caliciviridae family. In addition, considering the high level of sequence conservation around this Tyr residue among several representative members of different calicivirus genera (Fig.5), we also suggest that this residue could play similar roles in their respective systems. It has been shown in some members of the Caliciviridae family, such asFeline calicivirus, that VPg is not essential for infectivity, as deduced from the in vitro synthesis of an infectious RNA genome (28Sosnovtsev S. Green K.Y. Virology. 1995; 210: 383-390Crossref PubMed Scopus (131) Google Scholar). Nevertheless, VPg plays a crucial role in RHDV genome expression and infectivity, and no infectious transcripts could be made so far in the laboratory. Consequently, the progress made toward the understanding of this essential reaction could be invaluable in designing new approaches to investigate this highly virulent pathogen, which cannot be propagated in tissue cultures. Further studies on the in vitro mechanism of RHDV RNA replication will most possibly require the expression and purification of VPg and 3Dpol precursors, in addition to other host cell proteins that are possibly involved, as the starting point for the development of an in vitro replication system based on the reconstitution of a replicase complex from individual purified components. This approach might help in part to overcome the lack of a permissive tissue culture for RHDV. We thank Dr. L. Li (Dept. of Chemistry) and Dr. K. Ng (Dept. of Biochemistry) from the University of Alberta for performing MALDI-TOF mass spectrometry on VPg samples. Dr. M. Carpenter (Dept. of Biochemistry, University of Alberta) is acknowledged for performing N-terminal sequencing. S. Morillo from Departamento de Bioquı́mica y Biologı́a Molecular (Universidad de Oviedo) and M. Morales and J. M. Torres from Centro de Investigación en Sanidad Animal (CISA-INIA) are also acknowledged for their assistance.