CONVERSION OF DENGUE VIRUS REPLICATIVE FORM RNA (RF) TO REPLICATIVE INTERMEDIATE (RI) BY NONSTRUCTURAL PROTEINS NS-5 AND NS-3

Transcription

1 Am. J. Trop. Med. Hyg., 58(1), 1998, pp Copyright 1998 by The American Society of Tropical Medicine and Hygiene CONVERSION OF DENGUE VIRUS REPLICATIVE FORM RNA (RF) TO REPLICATIVE INTERMEDIATE (RI) BY NONSTRUCTURAL PROTEINS NS-5 AND NS-3 KANAKATTE RAVIPRAKASH, MANOJ SINHA, CURTIS G. HAYES, AND KEVIN R. PORTER Infectious Diseases Department, Naval Medical Research Institute, Bethesda, Maryland; Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Henry M. Jackson Foundation for the Advancement of Military Medicine, Rockville, Maryland Abstract. Dengue viruses infect more than 100 million people each year and cause serious clinical manifestations. It is important to understand the replication of these viruses so that therapeutic and/or prophylactic agents may be designed. Dengue virus type 2 nonstructural proteins NS-5 and NS-3 were produced by in vitro transcription and translation of cloned genes. Both proteins possessed RNA-dependent RNA polymerase activity as measured by their ability to convert purified replicative form (RF) RNA to replicative intermediate (RI). The recombinant proteins, however, required one or more cellular protein(s) for their activity. Examination of NS-3 protein sequence revealed heretofore unnoticed sequence similarities with other polymerases. The replication of RNA viruses, with the exception of retroviruses, is mediated by RNA replicase or RNA-dependent RNA polymerase (RdRp). In spite of a large number of RNA viruses afflicting humans with a variety of diseases, the replication of these viruses is poorly understood. Dengue virus, a member of the flavivirus family, causes one of the major infectious diseases in the tropical regions of the world. The genome of dengue virus contains an approximately 10.5-kb single-strand RNA of positive polarity. 1 Upon infection, the virion RNA is translated into a single polyprotein from which structural and nonstructural proteins are derived by specific proteolytic cleavage. 1, 2 The nonstructural protein 5 (NS-5) bears similarity to other known RdRp sequences and contains the -G-D-D- motif sequence found in most other RdRps. 3 The NS-3 protein on the other hand bears similarity to other known helicases. 4 It is believed that the dengue virus NS-5 and NS-3 proteins function as the viral polymerase and helicase, respectively. The NS-5 and NS-3 proteins have been shown to form a complex and are believed to be parts of the viral replication complex. 5 Dengue virus and other flavivirus RNA polymerase activity has been demonstrated in lysates of virus-infected cells. 6 8 More recently, a direct demonstration of RdRp activity using dengue virus type 1 NS-5 gene expressed in Escherichia coli has been reported. 9 We have cloned the dengue virus type 2 NS-5 and NS-3 coding sequences into the pgem11zf vector and produced the recombinant proteins by in vitro transcription followed by translation in a rabbit reticulocyte lysate. We demonstrate that the in vitro produced NS-5 as well as NS-3 exhibit RdRp activity, and that there is an additive effect when both proteins are present in the reaction. Furthermore, we show that a careful examination of the NS-3 sequence reveals heretofore unnoticed protein sequence similarities with other polymerases. These studies stress the need for more information on replication of dengue viruses. Also, the replicative form (RF) to replicative intermediate (RI) conversion described here can be adapted to screen for anti-dengue agents that function by inhibition of replication. MATERIALS AND METHODS All the cell lines used and dengue virus type 2 (strain New Guinea C) were obtained from the American Type Culture Collection (Rockville, MD). The virus was grown to high titers in 90 C6/36 (Aedes albopictus) cells cultured in minimal essential medium containing 10% fetal bovine serum (FBS). Vero cells were grown in RPMI 1640 medium containing 5% FBS. Vector pgem11zf was obtained from Promega (Madison, WI). Enzymes used were obtained from Gibco Laboratories, (Gaithersburg, MD), Boehringer Mannheim Biochemicals, (Indianapolis, IN), or Promega. The following oligonucleotide primers for the polymerase chain reaction (PCR) were obtained from Cruachem Inc. (Dulles, VA): primer A: 5 -ATTAATGGATCCAG- GACCATGGGAACTGGCAACATAGGA- 3 ; primer B: 5 - CCCAAGGCTGCATTGCTTCTCACC-3 ; primer C: 5 -CTA- GGGAAGAAAAAGAACACC-3 ; primer D: 5 -ATTAA- TAAGCTTTTACCACAGGACTCCTGCCTC-3 ; primer E: 5 -ATTAATGGATCCAGGACCATGGCTGGAGTA- TTGTGGGATGTC-3 ; and primer F: 5 -ATTAATAAGCTTT- TACTTTCTTCCAGCTGCAAACTC-3. Cloning of the dengue virus NS-5 and NS-3 genes. Dengue virus from culture supernatant of infected Vero cells was pelleted by centrifugation at 26,000 rpm for 2 hr in an SW28 (Beckman Instruments, Fullerton, CA) rotor. Virion RNA was purified from the pelleted virus using Trizol reagent (Gibco Laboratories) according to the manufacturer s recommendations. The RNA was reverse transcribed to cdna in a standard reverse transcriptase reaction at 42 C for 1 hr using mouse Moloney murine leukemia virus reverse transcriptase and random hexamer primers, and the cdna was used in the PCRs. The NS5 gene (nucleotides 7570 to 10269) 10 was amplified in two parts. The sequence between nucleotides 7570 and 8798 was amplified using the primer pair A and B, and the sequence between nucleotides 8722 and was amplified using primer pair C and D. The two PCR products were cut at the common Apa L1 restriction site present at nucleotide 8748 and ligated to get the 2.7-kb NS5 gene. This was further digested with Bam HI and Hind III (sites provided by the primers) and ligated into Bam HI- and Hind III-digested vector DNA to obtain pgem-ns5. The NS-3 gene was amplified using primers E and F, and was cloned similarly to obtain pgem-ns3. In vitro transcription and translation. Five micrograms of pgem-ns5 or pgem-ns3 DNA were linearized by digesting with Hind III and used in a standard 100- l transcription reaction containing 40 mm Tris-HCl, ph 7.5, 6 mm MgCl 2, 10 mm NaCl, 2 mm spermidine, 10 mm DTT, 10 units of RNasin, 0.5 mm each of ATP, CTP, GTP, and UTP,

2 DENGUE VIRUS POLYMERASE 91 FIGURE 1. In vitro assay for dengue virus RNA-dependent RNA polymerase. The infected Vero cell lysate was incubated in a reaction containing 32 P-GTP and the reaction products were separated into replicative form (RF), single strand (SS), and replicative intermediate (RI) by electrophoresis through a 3% polyacrylamide-urea gel (lane 1). The RF was purified by lithium chloride fractionation and used as the template in a reaction containing unlabeled nucleoside triphosphates and the cell lysate from either uninfected (lane 2) or infected (lane 3) Vero cells. and units of T7 RNA polymerase. Following incubation at 40 C for 1 hr, the DNA was digested with RQ1 RNase-free DNase (Promega) and the RNA was purified by phenol extraction. Approximately 0.5 g of RNA was used in translation reactions (50 l) containing nuclease-treated rabbit reticulocyte lysate, 2 mm DTT, 79 mm potassium acetate, 0.5 mm magnesium acetate, 10 mm creatine phosphate, 50 g/ml of creatine phosphokinase, 0.02 mm hemin, 40 units of RNasin, 20 M of each amino acid except methionine, and 50 uci of 35 S-labeled methionine. In vitro translations were also performed without the isotope using 20 M of a complete amino acid mixture. Reactions were incubated at 30 C for1hr. Purification of 32 P-labeled dengue virus RF RNA. Cell lysates from Vero cells infected for 40 hr with dengue virus at a multiplicity of infection of 1 were prepared according to the method of Bartholomeusz and Wright. 7 Control lysates FIGURE 2. Ribonuclease sensitivity of lithium chloride soluble replicative form [RF] and insoluble (replicative intermediate [RI]) fractions. The RF (lane 1) was incubated with RNase A in presence of low salt (lane 2) or high salt (lane 3) concentrations. Similarly, RI (lane 4) was incubated with RNase A in presence of low salt (lane 5) or high salt (lane 6) concentrations. were prepared from uninfected Vero cells. The lysates (approximately 10 g of protein) were incubated at 37 C for 60 min in a 50- l reaction containing 50 mm Tris-HCl, ph 8.0, 10 mm magnesium acetate, 7.5 mm potassium acetate, 10 mm 2-mercaptoethanol, 6 g/ml of actinomycin D, 5 mm phosphoenol pyruvate; 0.2 units/ml of pyruvate kinase, 20 units of RNasin; 0.5 mm each of ATP, CTP, and UTP, 25 M GTP, and 10 Ci of - 32 P-GTP. Following incubation, the 32 P-labeled RF RNA was isolated by the lithium chloride fractionation described by Bartholomeusz and Wright with minor modifications. 7 In vitro conversion of RF to RI. The purified 32 P-labeled RF was incubated as above with 0.5 mm of all four nucleotide triphosphates and without - 32 P-GTP. Infected cell lysates or products of in vitro translation were used as the source of enzyme. The RF and the RI were resolved by electrophoresis through a 3% polyacrylamide-urea gel. 7 The conversion of RF to RI was measured by autoradiography and densitometry. RESULTS Activity of RdRp in a dengue virus infected Vero cell lysate. An infected Vero cell lysate was incubated with 32 P-

3 92 RAVIPRAKASH AND OTHERS FIGURE 3. Hybridization of 32 P-labeled dengue virus cdna to replicative form (RF) and replicative intermediate (RI) fractions. The RI (ii) and the RF (iii) obtained from incubating a dengue virus infected Vero cell lysate with all four unlabeled nucleoside triphosphates were hybridized to a 32 P-labeled dengue virus nonstructural protein 5 (NS-5) DNA probe. Dengue virus NS-5 RNA (iv) and products from a reaction with uninfected Vero cell lysate (i) were included as controls. FIGURE 4. In vitro expression of dengue virus nonstructural NS- 5 and NS-3 proteins. Dengue virus NS-5 and NS-3 RNA were translated using a rabbit reticulocyte lysate in the presence of 35 S-methionine. The translation mixtures were electrophoresed through a 7.5% polyacrylamide-sodium dodecyl sulfate gel and analyzed by fluorography. Lane 1 NS-5; lane 2 NS-3. Values on the right are in kilodaltons (KD). labeled GTP as described. When the reaction products were resolved by urea-polyacrylamide gel electrophoresis, three bands appeared corresponding to a double-stranded RF RNA, single-strand genome length RNA, and the RI, which migrated the slowest (Figure 1, lane 1). The RF and RI were purified by lithium chloride fractionation. The identity of these bands as RF and RI was demonstrated by their differential RNase sensitivity under high and low salt concentrations. At high salt concentrations, the RF was resistant to RNase digestion (Figure 2, lane 3), whereas the RI was partially sensitive as indicated by its conversion to the RF (Figure 2, lane 6). However, at low salt concentrations both the RF (lane 2) and the RI (lane 5) were sensitive to RNase digestion. The authenticity of the RI and RF as dengue virus specific RNAs was demonstrated by hybridization of a 32 P-labeled dengue virus NS-5 DNA probe to lithium chloride soluble (RF) and insoluble (RI) fractions obtained from a parallel nonradioactive reaction. Figure 3 shows that both the RF and RI fractions, as well as the dengue NS-5 DNA (positive control) hybridized to the probe, whereas the total unfractionated reaction products obtained from an uninfected cell lysate (negative control) did not hybridize to the probe. The purified RF, when incubated with lysate from dengue virus infected Vero cells, was converted to RI (Figure 1, lane 3), indicating the presence of RdRp activity in such cell lysates. Lysate from uninfected Vero cells did not have any effect on the RF (lane 2). Cell-free expression of dengue virus NS-5 and NS-3. Dengue virus NS-5 and NS-3 coding sequences were cloned into transcription vector as described in the Materials and Methods. In vitro transcription of linearized plasmids using T7 polymerase resulted in RNA products of expected size. When the RNAs were translated in a reticulocyte lysate system, protein products of predicted molecular sizes were synthesized. Figure 4 showed that in addition to full-length recombinant proteins (100 kd for NS-5 and 70 kd for NS-3, respectively), several smaller translation products were also seen. The smaller molecular weight proteins probably represent products of internal initiation of protein synthesis. Indeed, the molecular size of the smaller products roughly corresponded to those expected from products initiated from internal methionines. Conversion of RF to RI by both NS-5 and NS-3. The NS-5 and NS-3 RNA were translated using rabbit reticulocyte lysate, and the lysates were used in reactions in which purified 32 P-labeled dengue virus RF RNA was used as the template. The RdRp activity was measured as the ability to convert the RF to RI. The reaction products were analyzed by electrophoresis through a 3% polyacrylamide-urea gel followed by autoradiography and densitometry. The experiment was repeated several times and the representative results are shown in Figure 5. It is clear from this figure that the reticulocyte lysate in which an unrelated (luciferase) RNA had been translated was unable to convert RF to RI either by itself (lane 10) or in the presence of cell lysate from uninfected Vero cells (lane 9). The uninfected Vero cell lysate alone had no activity (lane 2). However, when the

4 DENGUE VIRUS POLYMERASE 93 FIGURE 5. RNA-dependent RNA polymerase assay with recombinant dengue virus nonstructural NS-5 and NS-3 proteins. Purified 32 P- labeled replicative form (RF) was incubated in a standard reaction in presence of infected/uninfected Vero cell lysate, recombinant NS-5, recombinant NS-3, luciferase, or various combinations thereof, as shown. The reaction products were resolved by urea-polyacrylamide gel electrophoresis and analyzed by autoradiography and densitometry. Per cent conversion of RF to replicative intermediate (RI) is shown at the bottom. uninfected Vero cell lysate was incubated in the presence of either NS-5 (lane 3) or NS-3 (lane 5), there was an appreciable conversion of RF to RI (32.5% and 31%, respectively). Furthermore, when both NS-5 and NS-3 were present in the reaction, there was an additive effect as evidenced by increased conversion (45.8%) of RF to RI (lane 7). Neither protein had any activity in the absence of lysate from uninfected cells (lanes 4 and 6), and together they only had marginal activity (lane 8). DISCUSSION Although the RdRp gene has been identified for many positive-stand RNA viruses based mostly on the presence of the -G-D-D- motif sequence, 3 only in a few cases has the biological activity of these gene products been demonstrated. 11, 12 Among flaviviruses, polymerase activity in infected cell lysates has been demonstrated for dengue virus, 7 Kunjin virus, 6 and West Nile virus. 8 The dengue virus type 1 NS-5 protein expressed in E. coli was recently demonstrated to exhibit RdRp activity. 9 Whereas NS-5 bears a similarity to other known RdRps, the flavivirus NS-3 shows sequence similarity to RNA helicases. 4 The NS-3 protein of flaviviruses also has important protease activity 13 and thus may be a multifunctional protein. The NS-3 of dengue virus is known to form stable complexes with other viral nonstuctural proteins; Arias and others have shown complex formation between NS-3 and NS-2B. 14 The NS-5 and NS-3 have been shown to form a complex, and are believed to be parts of the viral replication complex. 5 However, the precise roles of these proteins in dengue virus replication are not known. Bartholomeusz and Wright were able to inhibit dengue polymerase activity with anti-ns5 antibody, 7 whereas Grun and Brinton 8 reported RdRp activity in factions of West Nile virus infected cell lysates that were almost devoid of NS-5 protein. We have demonstrated that both dengue virus type 2 NS- 5 and NS-3 are capable of converting the RF RNA to RI. However, the conversion of RF to RI by NS-3 was surprising. It could be that the NS-3 is acting as a helicase unwinding double-stranded RF to partially single-stranded RI. The other possibility is that NS-3 is acting as a polymerase. Previous analyses of the NS-3 sequence have found an NTP binding motif between residues 195 and and two conserved helicase motifs around residues and By closer examination, we have found various stretches of sequences within NS-3 that are also parts of other RdRps and polymerases (Figure 6). For example, the se-

5 94 RAVIPRAKASH AND OTHERS FIGURE 6. Comparison of dengue and other flavivirus nonstructural protein NS-3 sequences with those of polymerases. The dengue 2 virus NS-3 protein sequence is represented at the top. Region A is the NTP-binding motif, and regions B and C are the helicase motifs. Shown in the middle is the sequence comparison for NS-3 protein from dengue type 2 (Den2) virus, West Nile virus (WNV), and Japanese encephalitis virus (JEV). At the bottom of the figure are the sequences of NS-3 that are contained in a variety of polymerase (pol) sequences. The number next to a sequence is the first residue where the homology begins. CGMV cucumber green mottle mosaic virus; BYDV barley yellow dwarf virus; IBDV infectious bursal disease virus; IPNV infectious pancreatic necrosis virus; VSV vesicular stomatitis virus (Chandipura); RdRp RNA-dependent RNA polymerase; DdRp DNA-dependent RNA polymerase. * represents a homolgous residue and represents a conservative change. quence SGSPIIDKK found in cucumber green mottle mosaic virus RdRp 17 is 100% conserved in dengue NS-3. Similarly, dengue NS-3 shares sequences found in spinach DNA-dependent RNA polymerase, barley yellow dwarf virus RdRp, vesicular stomatitis virus (chandipura) polymerase, and the RdRps of infectious bursal disease virus and infectious pancreatic necrosis virus These sequences are also fairly well conserved among other flavivirus NS-3 proteins (Figure 4) such as West Nile virus NS-3 and Japanese encephalitis virus NS-3. 23, 24 It thus appears that dengue virus NS-3, in addition to being a protease and a helicase, may also be a polymerase. This may also explain the reported West Nile virus RdRp activity in fractions devoid of NS-5. 8 Our results further indicate that one or more host factors are also involved in dengue virus replication because the RF to RI conversion occurs only in the presence of uninfected Vero cell lysate. A host protein necessary for replication of poliovirus RNA in vitro has been identified as a protein kinase that phosphorylates eukaryotic initiation factor More work is needed to define the exact nature of the host factor(s) involved, as well as the exact roles of NS-5 and NS-3 in dengue virus replication. A detailed understanding of the replication of dengue viruses will be helpful in the design of therapeutic agents, and the assays described in this study should be useful for screening of anti-dengue agents in vitro. Financial support: Funding for this work was provided by the Naval Medical Research Institute work unit 61152NMR and the Henry M. Jackson Foundation grant Disclaimer: The opinions and assertions contained herein are not to be considered as official or as reflecting the views of the Navy Services at large. Authors addresses: Kanakatte Raviprakash, Infectious Diseases Department, Naval Medical Research Institute, 8901 Wisconsin Avenue, Bethesda, MD and Henry M. Jackson Foundation for the Advancement of Military Medicine, Rockville, MD Manoj Sinha, Curtis G. Hayes, and Kevin R. Porter, Infectious Diseases Department, Naval Medical Research Institute, Building 17, Room 316, 8901 Wisconsin Avenue, Bethesda, MD REFERENCES 1. Chambers TJ, Hahn CS, Galler R, Rice CM, Flavivirus genome organization, expression and replication. Annu Rev Microbiol 44: Henchal EA, Putnak R, The dengue viruses. Clin Microbiol Rev 3: Bruenn JA, Relationships among the positive strand and double stand RNA viruses viewed through their RNA-dependent RNA polymerases. Nucleic Acids Res 19: Koonin EV, Similarities in RNA helicases. Nature 352: Kapoor M, Zhang L, Ramachandra M, Kusukawa J, Ebner KE, Padmanabhan R, Association between NS3 and NS5 proteins of dengue virus type 2 in the putative RNA replicase is linked to differential phosphorylation of NS5. J Biol Chem 270: Chu PWG, Westaway EG, Characterization of kunjin virus RNA-dependent RNA polymerase: reinitiation of synthesis in vitro. Virology 157: Bartholomeusz AI, Wright PJ, Synthesis of dengue virus RNA in vitro: initiation and the involvement of proteins NS3 and NS5. Arch Virol 128: Crun JB, Brinton MA, Characterization of West Nile virus RNA-dependent RNA polymerase and cellular terminal adenylyl and uridylyl transferases in cell free extracts. J Virol 60: Tan B-H, Fu J, Gugrue RJ, Yap E-H, Chan Y-C, Tan YH, Recombinant dengue virus type 1 NS5 protein expressed in E. coli exhibits RNA dependent RNA polymerase activity. Virology 216: Irie K, Mohan PM, Sasaguri Y, Putnak R, Padmanabhan R, Sequence analysis of cloned dengue virus type 2 genome (New Guinea-C stain). Gene 75: Neufeld KL, Richards OC, Ehrenfeld E, Purification, characterization and comparison of poliovirus RNA polymerase from native and recombinant sources. J Biol Chem 266: Sankar S, Porter AG, Expression, purification and prop-

6 DENGUE VIRUS POLYMERASE 95 erties of recombinant encephalomyocarditis virus RNA-dependent RNA polymerase. J Virol 65: Cahour A, Falgout B, Lai C-J, Cleavage of the dengue virus polyprotein at the NS3/NS4A and NS4B/NS5 functions is mediated by viral protease NS2B-NS3, whereas NS4A/NS4B may be processed by a cellular protease. J Virol 66: Arias CF, Preugschat F, Strauss JH, Dengue 2 virus NS2B and NS3 form a stable complex that can cleave NS3 within the helicase domain. Virology 193: Gorbalenya AE, Koonin EV, Viral proteins containing NTP-binding sequence pattern. Nucleic Acids Res 17: Gorbalenya AE, Koonin EV, Donchenko AP, Blinov VM, Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res 17: Ugaki M, Tomiyama M, Kakutani T, Hidaka S, Kiguchi T, Nagata R, Sato T, Motoyoshi F, Nishiguchi M, The complete nucleotide sequence of cucumber green mottle mosaic virus (SH strain) genomic RNA. J Gen Virol 72: Hudson GS, Holton TA, Whitfeld PR, Bottomley W, Spinach chloroplast rpobc genes encode three subunits of the chloroplast RNA polymerase. J Mol Biol 200: Miller WA, Waterhouse PM, Gerlach WL, Sequence and organization of barley yellow dwarf virus genomic RNA. Nucleic Acids Res 16: Maasters PM, Banerjee AK, Sequences of Chandipura virus N and NS genes: evidence for high mutability of the NS gene within vesiculoviruses. Virology 157: Morgan MM, Macreadie IG, Harley VR, Hudson PJ, Azad AA, Sequence of the small double-stranded RNA genomic segment of infectious bursal disease virus and its deduced 90- kda product. Virology 163: Duncan R, Mason CL, Nagy E, Leong J, Dobos P, Sequence analysis of infectious pancreatic necrosis virus genome segment B and its encoded VP1 protein: a putative RNA-dependent RNA polymerase lacking the Gly-Asp-Asp motif. Virology 181: Castle E, Leidner U, Nowak T, Wengler G, Wengler G, Primary structure of West Nile flavivirus genome region coding for all non-structural proteins. Virology 149: Nitayaphan S, Grant JA, Chang GJ, Trent DW, Nucleotide sequence of the virulent SA-14 strain of Japanese encephalitis virus and its attenuated vaccine derivative, SA Virology 177: Morrow CD, Gibbons GF, Dasgupta A, The host protein required for in vitro replication of poliovirus is a protein kinase that phosphorylates eukaryotic initiation factor-2. Cell 40: