Virus Research 73 (2001) 41 – 55 www.elsevier.com/locate/virusres
Arenavirus nucleocapsid protein displays a transcriptional antitermination activity in vivo M.A. Tortorici a, C.G. Albarin˜o a,1, D.M. Posik a, P.D. Ghiringhelli b, M.E. Lozano b, R. Rivera Pomar a,2, V. Romanowski a,b,* a
Instituto de Bioquı´mica y Biologı´a Molecular, Depto. de Ciencias Biolo´gicas, Facultad de Ciencias Exactas, Uni6ersidad Nacional de La Plata, Calles 47 y 115, 1900 La Plata, Buenos Aires, Argentina b Departamento de Ciencia y Tecnologı´a. Centro de Estudios e In6estigaciones. Uni6ersidad Nacional de Quilmes, Roque Saenz Pen˜a 180, 1876, Bernal, Argentina Received 16 May 2000; received in revised form 17 August 2000; accepted 8 September 2000
Abstract RNA polymerase pausing and transcriptional antitermination regulates gene activity in several systems. In arenavirus infected cells the switch from transcription to replication is subjected to a hairpin-dependent termination and requires protein synthesis to bypass this signal. The transcriptional antitermination control by Junı´n virus nucleocapsid protein N, has been demonstrated in vivo by infecting BHK-21 cells expressing this viral protein in the presence of translation inhibitors. This is the first demonstration in vivo of a transcriptional antitermination control in arenavirus-infected cells. © 2001 Published by Elsevier Science B.V. All rights reserved. Keywords: Arenaviridae; Transcriptional antitermination; Nucleocapsid protein
1. Introduction The arenavirus genome is composed of two single stranded RNA species. One of them, designated L RNA, encodes the RNA polymerase and * Corresponding author. Tel.: + 54-221-4250497 (ext. 32); fax: + 54-221-4259223. E-mail address:
[email protected] (V. Romanowski). 1 Present address: Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA. 2 Present address: Max Planck Institut fu¨r Biophysikalische Chemie, Abteilung Molekulare Entwicklungsbiologie, Am Fassberg, 37077- Go¨ttingen, Germany.
a zinc-finger-like protein (Z) (Salvato and Shimomaye, 1989; Iapalucci et al., 1989a,b; Salvato et al., 1992). The second species, termed S RNA, codes for both the viral nucleocapsid protein (N) and the precursor of the envelope glycoproteins (GPC). The open reading frames of the two RNA species are arranged in opposite orientations (ambisense) (Auperin et al., 1984) and separated by a non-coding intergenic region that is able to form a stable secondary structure (Franze-Fernandez et al., 1987; Salvato and Shimomaye, 1989; Iapalucci et al., 1989a,b; Ghiringhelli et al., 1991). The N protein is translated from an antigenomic (or viral complementary) sense mRNA species that is en-
0168-1702/01/$ - see front matter © 2001 Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 7 0 2 ( 0 0 ) 0 0 2 2 2 - 7
42
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
coded by the 3% half of the viral S RNA, while the GPC is translated from a genomic (or viral) sense mRNA corresponding to the 5% half of the S RNA (Romanowski, 1993, see Fig. 1). The genomic S RNA is transcribed to only two antigenomic forms: the 1.8 kb N mRNA and the 3.4 kb full length S RNA. The antigenomic S RNA serves as the replicative form of the virus and also as the template for GPC mRNA transcription (Fig. 1). When translation is inhibited, transcription of Junı´n virus RNA yields only the N mRNA, a phenomenon also observed for other arenaviruses (Franze-Fernandez et al., 1987; Rivera Pomar et al., 1992). This implies that the synthesis of a full length antigenomic copy of S RNA requires an antiterminator. A priori, this function could be supplied by the N protein, some other viral gene product or a cellular polypeptide with a high turnover rate. Transcription antitermination in prokaryotic and eukaryotic systems is a process that regulates gene activity bypassing termination steps (Krumm et al., 1993; Greenblatt et al., 1993; Yarnell and Roberts, 1999). Particularly, in human respiratory syncitial virus (RSV), the M2-1 protein enhances readthrough of intergenic junctions during viral
transcription (Hardy and Wertz, 1998; Fearns and Collins, 1999). Consistent with amino acid sequence and function similarities of the two largest subunits of prokaryotic and nuclear eukaryotic RNA polymerases, there are evidences on common mechanisms that control transcriptional elongation (Krumm et al., 1993). Polymerase pausing and protein binding to RNA and/or transcription complexes appear to prevent termination in several systems (O’Brien et al., 1994; Cullen, 1994; Krumm et al., 1995; Weisberg and Gottesman, 1999). Results presented here show that the N gene product — the major nucleocapsid protein — can interact in vitro with the intergenic region of Junı´n virus S RNA. This region contains two sets of self complementary nucleotide sequences capable of forming two very stable stem loop structures (Ghiringhelli et al., 1991). In conclusion, these results show that the switch from RNA-dependent transcription to replication of the Junı´n virus, a pathogenic member of the Arenaviridae family, is subjected to a hairpin-dependent transcription termination and to an antitermination control by the nucleocapsid N protein.
2. Materials and methods
2.1. Virus and cell culture
Fig. 1. Schematic of the gene arrangement in Junı´n virus S RNA. The ambisense S RNA of Junı´n virus is shown divided in two regions encoding the N gene in the viral complementary (vc) and the GPC gene in the viral polarity (v), respectively. The open reading frames corresponding to the N and GPC genes are shown as filled and open rectangles, respectively, with arrowheads indicating the direction of translation. The subgenomic mRNAs are also shown, with NNN indicating nontemplated nucleotide sequences at their 5% ends (Rivera Pomar, 1991; Raju et al., 1990). The thin curved arrows indicate explicitly the template–transcript relationships among the different RNA species.
Junı´n virus, MC2 and XJ44 strains, were propagated in BHK-21 C13 cells (ATCC, CRL8544) cells. These cells were cultured in growth medium (Dulbecco’s minimal essential medium — MEM — supplemented with 10% fetal calf serum and 2 mM L-glutamine) to 50% confluence and infected at a multiplicity of infection (m.o.i.) of 1 plaque forming unit (pfu) per cell. After virus adsorption for 1 h at room temperature, infected cells were washed with phosphate-buffered saline (PBS) and maintained at 37°C in MEM containing 2% fetal bovine serum. The viruses were recovered from the supernatant media of infected monolayers and purified by ultracentrifugation (Ghiringhelli et al., 1991). Virus titers were determined by plaque assay on Vero E6 (ATCC, CCL 1586, C1008) cell monolayers as described elsewhere (Rustici, 1984).
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
The cell monolayers were harvested 72 h postinfection by scraping into PBS and recovered by centrifugation.
2.2. RNA isolation Cultures of infected BHK-21 cells (ten dishes of 10 cm) were processed to isolate total RNA. The cells were lysed in TNE (10 mM Tris – HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA) containing 0.2% Nonidet P-40 (300 ml/dish), and the extract was clarified by spinning for 4 min at 4000 × g. The supernatant was loaded onto a 60% (w/w) CsCl cushion and centrifuged overnight at 41 000 rpm in a SW41 rotor (Beckman). The resulting pellet, containing naked RNA, was resuspended in 1 ×GTC solution (4 M guanidinium thiocyanate, 25 mM sodium citrate, 0.5% sarkosyl, 100 mM 2-mercaptoethanol), and the RNA was isolated using the Chomczynski and Sacchi (1987) method. The RNA was isolated from pelleted virions, infected and uninfected cells, using the same protocol.
43
induced by adding 1 mM isopropyl-1-thio-b-Dgalactopyranoside (IPTG). After 4 h, bacteria were collected by centrifugation, rinsed with PBS and resuspended in 50 mM Tris–HCl pH 8.0; 100 mM NaCl; 1 mM EDTA; 0.5% Nonidet P-40 plus 0.2 mM phenylmethysulfonyl fluoride (PMSF) by sonication. After centrifugation the recombinant protein was collected from the insoluble fraction, resuspended in sample buffer and analyzed by SDS–PAGE. The N gene including its own noncoding upstream sequences was amplified using ARS 1 primer 5%-CGCACAGTGGATCCTAGGC-3% plus N2 primer, and used for expression in a mammalian system. The amplified fragment was inserted by ligation into the BamH I site of pcDNA3 downstream from the cytomegalovirus promoter (pCMV) (Invitrogen). The recombinant plasmid, designated pcDNA3.NJUN, was amplified in E. coli DH5a and purified for transfection experiments using silica columns (Wizard Plus Midipreps, Promega). The expression of the N gene in a mammalian cell line (BHK-21) was tested by indirect immunofluorescence assay (IFA).
2.3. Cloning and expression 2.4. Immunological characterization A plasmid containing the cDNA of the S RNA (pBJUNS) was used for a number of constructions harboring the N gene. The N ORF was amplified by PCR using the primers N1 5%GAGATCTGGATCCATGGCACACTCCAAAGAG-3% (translation initiation sequence is indicated in bold and the Nco I site is underlined) and N2 5%-GAGATCTGGATCCTTACAGTGCATAGGCTGC-3% (sequence complementary to the stop codon is indicated in bold and the BamH I site is underlined). The PCR product was inserted into pGem-T (Promega). The N gene was recovered by digestion with Nco I and BamH I, filled-in using the Klenow fragment of DNA polymerase I, and inserted by ligation into the filled-in Nde I site of pET-22b( + ) (Novagen) to generate pET-N. This construction was made to express the N protein starting from its own methionine codon and with no fused sequences at its C-terminus. The E. coli BL21 (DE3) harboring the pETN, was grown in 20 ml of LB medium to an A550 of 0.5–0.6 and the production of N protein was
2.4.1. Preparation of N polyclonal antibody The recombinant N protein produced in bacteria was purified by SDS–PAGE using a wide well in a slab gel apparatus. The band was located by Coomassie Brilliant Blue staining of gel strips and the unstained N protein band was excized, homogenized and used for immunization of rabbits. Three rabbits were injected subcutaneously three times with 150 mg of N polypeptide each. Blood was taken 2 and 3 weeks after each injection of the protein and the antibody titer was estimated by ELISA. 2.4.2. Immunoblotting Samples of bacterial extracts were separated by SDS–PAGE (10% polyacrylamide gel), blotted onto nitrocellulose membranes, and probed with a 1/150 dilution of the rabbit antiserum specific for the N protein. Filters were then incubated with a goat anti-rabbit antibody conjugated to alkaline phosphatase (Sigma) and later incubated with
44
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
chromogenic substrates: nitro blue tetrazolium (NBT) and 5-bromo-1-chloro-3-indolyl phosphate (BCIP) (Promega).
2.4.3. Immunofluorescence assay (IFA) The immunofluorescence assay was done as described previously (Rivera Pomar et al., 1991). Briefly, BHK-21 were fixed onto glass cover slips with 4% formaldehyde in PBS and permeabilized with 0.5% Triton X-100 for 10 min at room temperature. The cells were then washed with PBS, preincubated for 45 – 60 min. at 37°C with 3% bovine serum albumin in PBS, and incubated for 30 min at 37°C with the first antibody (a 1:450 dilution of the rabbit polyclonal anti-N antiserum). After a PBS wash the second FITC-conjugated antibody (goat anti-rabbit from Sigma) was added and allowed to react at 37°C for 30 min. The cover-slips were washed with PBS and stained with Evans Blue for 10 min at room temperature. The samples were mounted using 2.5% DABCO, 10% PBS, 90% glycerol, viewed and photographed using a Nikon Episcopic-Fluorescence Microscope. 2.5. In 6itro transcription of Junı´n 6irus strand-specific probes The in vitro transcripts used in this study were derived from pT7 plasmids containing the bacterial T7 RNA polymerase promoter site. Plasmid pTVGPCN was constructed by subcloning the EcoR I-BamH I fragment of plasmid pBSNGPC into the pTV(d0,d0) (Pattnaik et al., 1992), previously digested with EcoR I and BamH I. This resulted in a plasmid containing the intergenic region of Junı´n virus S RNA (from the Sph I site at nt. 1436, to the Bgl II site at nt. 1915 viral sense polarity). The riboprobe generated from pTVGPCN was of viral sense polarity while the riboprobe generated from pBSNGPC was of viral complementary polarity. The run off transcription was carried out in 20 ml of reaction mix containing approximately 1 mg of digested plasmid DNA; 40 mM Tris–HCl pH 7.5; 6 mM MgCl2; 2 mM spermidine; 10 mM NaCl; 10 mM DTT; 0.5 mM each of UTP, CTP, and GTP; 0.05 mM ATP; 30 mCi of a [32P]ATP; 20 U of T7 RNA polymerase
(Promega); and 40 U of RNasin ribonuclease inhibitor (Promega). Reaction mixtures were incubated at 37°C for 1 h. DNA templates were removed by digestion with 1 U of RNase-free DNase RQ1 (Promega) for 10 min at 37°C. The riboprobes were purified by adsorption to glass milk (RNaid; Bio101). The eluted RNA (5× 107 dpm/mg) was diluted to 4 × 105 cpm/ml in 50 mM Tris-HCl pH 7.5, 100 mM NaCl, 5 mM MgCl2. All riboprobe preparations were analyzed by electrophoresis on a polyacrylamide gel to ensure that the RNAs were intact and that recovery was as expected.
2.6. Northwestern The RNA binding assay was based on the one described by Reichel et al. (1996). Briefly, proteins were separated by SDS–PAGE, electroblotted to nitrocellulose and renatured on the membrane by incubation in binding buffer [50 mM NaCl, 1 mM EDTA, 1 mM DTT, 10 mg/ml yeast tRNA (Sigma), 20 mM MES, pH 6.0] for 90 min, changing the buffer twice. After incubation for 1 h with the labeled RNA probe, filters were washed twice with fresh binding buffer, dried and analyzed by autoradiography.
2.7. RNase protection assay The radiolabeled in vitro transcribed riboprobes were incubated with 1 to 10 mg of RNA sample in 30 ml of hybridization buffer (40 mM PIPES, pH 6.4, 400 mM NaCl, 1mM EDTA pH 8.0, 80% formamide) at 85°C for 15 min and then, at 53°C for 14 h. After this incubation 300 ml of digestion solution (10 mM Tris–HCl, pH 7.5, 300 mM NaCl, 1 mM EDTA, pH 7.5, 30 U of RNase T1 and 3 mg/ml of RNase A) were added, and the mixture was incubated at 30°C for 1 h. After that, RNases were digested with 100 mg of proteinase K in the presence of 0.5% sodium dodecyl sulfate (SDS) at 37°C for 30 min. After phenol–chloroform extraction and ethanol precipitation, the resulting RNA species were electrophoresed on a 6% acrylamide sequencing gel. Nucleotide sequence reactions generated by the dideoxynucleotide chain termination method, using
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
a[32P]dATP and the fmol TM DNA Sequencing System (Promega), were used as size markers.
2.8. Antitermination assay and strand specific RT-PCR BHK-21 cells grown in 24-well plastic dishes (Nunc) were transfected with pcDNA3.NJUN using Lipofectamine™ (Life Technologies) following the protocol recommended by the manufacturers. Twenty four hours after transfection, BHK-21 cells expressing the N protein and those used as controls were infected with Junı´n virus (MC2 strain), in presence or absence of protein synthesis inhibitors. Different concentrations of pactamycin, cycloheximide and puromycin were used to inhibit protein synthesis yielding similar results. The studies presented in this paper were conducted using 100 mg/ml puromycin in the culture medium to achieve 98% inhibition of protein synthesis. The latter was determined by measuring 3H-valine incorporation into trichloroacetic acid precipitable material. Cells subjected to this treatment were harvested at 24 h post-infection; total RNA was prepared using the guanidinium thiocyanate method and resuspended in 20 ml sterile water (Chomczynski and Sacchi, 1987). Strand specific RT-PCR (Lanford et al., 1994; Tam et al., 1996) was used to monitor the presence or absence of N mRNA and full-length antigenomic (vc=viral complementary) S RNA. The viral sense primers vN 5%GGCATCCTTCAGAACATC-3% and vG 5%-ATGGGGCAATTCATCAG-3% were used in the cDNA synthesis. Primers vN plus vcN 5%-CGCACAGTGGATCCTAGGC-3% and vG plus vc G 5%-CCCCTTAATGTAAAGATGGC-3% were used in the PCR reactions. RNA samples were heated to 95°C for 3 min in the presence of vN and vG (0.4 mM, final concentration) and denatured with methyl mercury (II) hydroxide (10 mM, final concentration) for 5 min before initiating reverse transcription. After inactivation of the denaturant with 14 mM 2-mercaptoethanol, the first-strand cDNA synthesis was carried out in a total volume of 25 ml containing 10 ml of the total RNA, 60 mM KCl, 25 mM Tris – HCl pH 8.0, 10 mM MgCl2, 1 mM dithiothreitol, 0.5 mM (each)
45
deoxynucleoside triphosphates (dNTPs), 7.5 U RNasin and 4 U AMV reverse transcriptase. The reaction mixture was incubated for 1 h 30 min at 42°C. The first-strand cDNA reaction was ethanol precipitated in the presence of 100 mM sodium acetate and 5 mg linear polyacrylamide, washed with 70% ethanol and resuspended in sterile water. PCR amplifications were carried out in 10 ml final volume, containing one tenth of the cDNA reaction, 0.25 units of Taq Express DNA polymerase (GENPAK), 1 mM each primer, 200 mM each dNTP, 50 mM KCl, 10 mM Tris–HCl pH 8.3, 1.5 mM MgCl2 and 0.01% gelatin. The mixture was overlaid with mineral oil. The PCR cycle progression for the vN/vcN and vG/vcG primer pairs was as follows: 2 min at 92°C, followed by 35 cycles of 15 s at 92°C, 30 s at 58°C, 50 s at 68°C. The PCR products were resolved by electrophoresis on agarose gels in TAE buffer stained with 0.5 mg/ml ethidium bromide and photographed with Polaroid 667 film.
3. Results
3.1. Analysis of transcripts of Junı´n 6irus S RNA In order to define the role of the predicted secondary structure for the intergenic region (IGR) of the S RNA in the transcription process, we characterized the 3% termini of mRNAs (N and GPC) by RNase protection assay (Fig. 2). To this end, a 32P-labeled RNA probe, containing the genome-sense strand of the complete IGR of Junı´n virus S RNA (495–nt, positions 1436–1915 plus 15 nt from the vector), was used to map the N mRNA end (N probe). A 32P-labeled RNA probe, containing the antigenome-sense strand of same region (543-nt; positions 1486–1965 plus 63 nt from the vector), was used to map the GPC mRNA (G probe) (Fig. 2a). The N probe was partially protected by the N mRNA present in total RNA preparations of infected cell extracts or in mRNA pellets (Fig. 2b). It was totally protected against RNase T1+A digestion by the full length antigenomic S RNA from infected cell extracts (Fig. 2c). The differences in size of the full length probe and the completely protected
46
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
Fig. 2. (Continued)
probe, in this case are due to the vector sequences that do not hybridize with the viral antigenomic S RNA. To determine the size of the N probe
fragment protected with unencapsidated mRNA, the RNase resistant products were electrophoresed on a sequencing gel using sequencing
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
reactions from pGem-3Zf DNA as markers (Fig. 2b). Two bands corresponding to 342- and 347base-long RNase resistant fragments were detected, indicating that the N mRNA 3% termini mapped at positions 1827 – 1832 from the 5% end of the antigenomic S RNA (Fig. 2b, lanes T and P). The RNA from virions yields no detectable protection of the genome-sense probe as shown in lane V, Fig. 2c. The antigenome-sense probe (G probe) was used to determine the 3% termini of the GPC mRNA. Two bands corresponding to 155and 157-nucleotides long RNase resistant fragments were detected, indicating that the GPC mRNA mapped at positions 1590 – 1592 from the 5% end of the genomic S RNA (Fig. 2d). These results are in accordance with observations published by other authors, and indicate that there is terminal heterogeneity at the 3% end of both N and GPC mRNA (Iapalucci et al., 1991; Meyer and Southern, 1993). However, it is not possible to rule out artifacts in the RNase protection assays that could occur at either end of the mRNAs due to the breathing of the probe – target duplex. The portion of the sequencing gel shown in Fig. 2d comprises only the size range relevant for the definition of the mRNA 3% termini and, therefore, does not include the image of the antigenome sense probe protected with the viral genomic S RNA (lanes I and V). These data are presented in Fig. 2e, which shows that the G probe was totally protected against RNase digestion by the genome
47
S RNA, either packaged within virions or present in the infected cells. In order to analyze the conformation of the mRNA 3% termini, we calculated the stability of the possible secondary structures taking into account that the precise 3% end may vary according to the RNase protection assay data. The Fig. 3 shows the two most stable structures predicted in vacuo for each of the two most distant 3% ends determined for the N mRNA. It is possible to note that in all of the predicted structures two stems loops are formed that are essentially conserved, and the termination of transcription occurs 11–14 nucleotides after the last and largest stem loop. Interestingly, the 3% end of the GPC mRNA shares the same structural motif and, in this case, the termination of transcription occurs 8–10 nucleotides after the last and largest stem loop (not shown).
3.2. In 6i6o antitermination acti6ity of the N protein N mRNA is the primary transcription product of S RNA as shown when cells are infected after protein biosynthesis has been blocked (Rivera Pomar, 1991; Franze-Fernandez et al., 1993). Because the synthesis of a full length antigenomic copy of S RNA requires viral protein synthesis, we tested if the N protein could act as an antiterminator.
Fig. 2. Analysis of the 3% termini of the mRNAs encoded by the S segment. The drawing at the top (A) represents the genetic organization of Junı´n virus S RNA. The filled arrows indicate viral complementary coding sequences and the open arrows indicate the viral sense coding sequences of the N and the GPC genes, respectively, and the arrowheads show the orientations of the coding sequences. The numbers above are in viral complementary (vc) polarity and the numbers below are in viral polarity (v). The positions 1776 (vc polarity) and 1533 (v polarity) correspond to the last codon of N and GPC genes, respectively. The N probe is 495 nt long (480 nt of Junı´n sequences, indicated by a open rectangle, plus 15 nt of vector sequences, indicated by a thin line at the 5% end). This probe, containing the genome-sense strand of the complete intergenic region (IGR) of S RNA, was used to map the N mRNA. The antigenome-sense of the same region was used to map the GPC mRNA (G probe= 543 nt, 480 nt of Junı´n sequences — filled rectangle, plus 63 nt of vector sequences at the 5% and 3% ends — thin lines). RNA was isolated from uninfected cells (U), from infected cells (I) -total RNA (T) and CsCl cushion pellet RNA (P) (an amount equivalent to five 10 cm-dishes of infected cells) — and from virion RNA (V) was mapped using the N probe (B). The results of the RNase protection assays after annealing with the N probe are shown in panels B and C. Those obtained with the G probe are in panels D and E. Protected RNA fragments were resolved by electrophoresis on a 6% polyacrylamide gel alongside a sequencing ladder and detected by autoradiography (B and D). The regions of the gel containing the totally protected riboprobes are shown in panels C and E gel. The differences of length between the riboprobes without treatment (lanes ‘‘N probe’’ and ‘‘G probe’’ in panels C and E, respectively) and the protected riboprobes by RNAs (lane I in panel C; lanes V and I in panel E) are due to the vector sequences not protected by the viral RNA.
48
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
Fig. 3. Sequences and predicted secondary structure folds of the 3’ termini of the S-RNA-derived N mRNAs. The structures were calculated using the RNA mfold program (Michel Zuker’s server version 3.0, http://www.ibc.wustl.edu/zuker). The numbers indicate the positions relative to the 5’ end of the virus N mRNA (antigenomic sense). The protected species of the N mRNA terminate at different positions with maxima at nucleotide numbers 1827 and 1832 (antigenomic sense). The N mRNA secondary structure predictions start at nucleotide 1752 (viral complementary sense, vc) and ends at nucleotide 1832 (vc) in the longest structure, and at nucleotide 1827 (vc) in the shortest structure. The common stem-loop structures are indicated by the shadowed areas. Gray dots represent GC base pairings and black dots represent AU and GU (non-canonical) base pairings. The UAA stop codon for the N ORF is highlighted in bold. The most stable structure predicted for the longest N mRNA is represented in A (DG 0 = −44.9 kcal/mol). Likewise, B shows the most stable structure predicted for the shortest N mRNA (DG0 = −39.54 kcal/mol). The differences observed in the base pairing between the longest and shortest structures of the N mRNA are due to an equilibrium between the stabilizing energy of the last base pairs and the destabilizing energy of the bulge loop containing two As (nts 1756 – 1757).
3.2.1. N gene expression in BHK-21 cells We examined the transcription of S RNA in the absence of protein synthesis during viral infection and in the presence of endogenous N protein. To this end, we expressed the N gene in BHK-21 cells by transfection with a pcDNA3-derived plasmid containing the entire N gene ORF from the MC2 strain (pcDNA3N.JUN). The identity of the gene product expressed by the transfected plasmid, was assessed by indirect immunofluorescence using Nspecific polyclonal antibodies (Fig. 4c, d). Cells infected with Junı´n virus-MC2 strain were used as positive controls (Fig. 4a, b) and cells transfected with the pcDNA3 vector with no insert were used as negative controls (Fig. 4f). Fig. 4(b) and (d) were included to show the proportion of cells
expressing N after infection and transfection, respectively. The distribution of N expressed in transfected cells was similar to that of infected cells, i.e. diffuse cytoplasmic immunofluorescence and granules that are also visible with phase contrast optics (Sanchez et al., 1989; Rivera Pomar et al., 1991,). Due to the fact that not all of the cells on a cover slip became transfected, internal negative controls were present in every preparation (Fig. 4).
3.2.2. Detection of in 6i6o antitermination acti6ity We transfected BHK-21 cells with the N expression plasmid pcDNA3.NJUN (or the control vector pcDNA3) and examined the synthesis of S RNA-derived species after viral infection, in the
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
presence or absence of a protein synthesis inhibitor. Prior to infection, puromycin (100 mg/ml) was added to block translation and maintained at this concentration throughout the experiment (Fig. 5a). The presence or absence of N mRNA and full-length antigenomic S RNA was checked by strand-specific RT-PCR, as described in Section 2 (Fig. 5b). The experimental results are shown in Fig. 5c. The 186 bp long amplified cDNA fragment, indicative of sequences containing the N gene (N mRNA and/or vc S RNA), was detected in infected cells (lanes 1, 2, 3 and 4) and in uninfected cells transfected with pcDNA3NJUN (lane 5). The 310 bp long amplified cDNA is diagnostic of the presence of full length antigenomic S RNA
49
(vc S RNA). This fragment was detected in infected cells in absence of puromycin, used as positive controls (lanes 9 and 10) and was not detected in uninfected cells, which were used as negative controls (lanes 11 and 12). The key result is the information obtained from lanes 7 and 8: while puromycin inhibited the synthesis of full length antigenomic S RNA in BHK-21 (lane 8), this RNA species was readily detectable by RTPCR in transfected BHK-21 cells expressing N protein (lane 7). Identical results were also obtained using an unrelated Junı´n virus strain (XJ44) (data not shown). Different control experiments were performed to confirm the RT-PCR results. Control reactions were included in order to rule out the
Fig. 4. N-specific immunofluorescence analysis of infected and transfected cells. BHK-21 cells were infected with Junı´n virus (A, B) or transfected with the recombinant expression plasmid pcDNA3.NJUN (C, D) or with pcDNA3 with no insert (F) and inspected for N expression at 48 h post-transfection. The presence of the N protein was monitored in cells transfected and treated with puromycin for a 24 h period starting at 24 h after transfection (E). The cells were fixed, incubated with a rabbit antiserum specific for the N protein of Junı´n virus and subsequently labeled using FITC-conjugated goat anti-rabbit IgG (see Section 2).
50
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
antigenomic S RNA. However, N protein alone may not be the only regulatory requirement. It is possible that a host factor or a viral protein may also be involved.
3.3. The N protein of Junı´n 6irus binds 6iral RNA in a Northwestern blot assay Northwestern blot experiments were performed in order to show that the N protein can interact with the viral RNA in 6itro. For this purpose, the N gene was expressed in E. coli (Fig. 6a), recovered from the insoluble fraction and analyzed immunochemically with an N-specific polyclonal antibody (Fig. 6b). Cross-reactions to E. coli proteins present in this type of preparation were not apparent. To determine if the N protein was able to interact with the intergenic region of Junı´n virus, the insoluble fraction containing the recombinant protein was separated by SDS–PAGE, transferred to a nitrocellulose membrane, and incubated with a 32P-labeled RNA probe comprising the intergenic region (genome sense polarity).
Fig. 5. (Continued)
possibility that molecules other than oligonucleotides vN or vG were priming the cDNA synthesis, generating potential artifact products. To do this we performed the cDNA synthesis stage on RNAs extracted from infected BHK-21 cells and transfected BHK-21 cells expressing N protein without added primers. No amplification bands were detected in these control experiments (data not shown). We also checked the presence of N protein 48 h after transfection in cells that were treated with puromycin during 24 h prior to the immunofluorescence analysis. This assay confirmed the presence of N throughout the experiment, after translation was suppressed (Fig. 4e). The experimental results indicate that the preexisting N protein in the transfected BHK-21 cells is necessary to support the synthesis of full length
Fig. 5. (A) The upper schematic represents the time-course of the antitermination experiment. (B) Transcription of Junı´n virus RNA and antiterminator activity of N. The antigenomic forms, the N mRNA and the full length vc S RNA, are shown at the top of the figure. The primers used for the RT-PCR detection of antigenomic forms are indicated below as filled or open arrows. Only the viral sense primers (open arrows) vN and vG were used in the cDNA synthesis. Primers vN plus vcN were then used to PCR amplify antigenomic forms containing N sequences (N mRNA and antigenomic S RNA); primers vG plus vcG, to amplify antigenomic forms containing GPC sequences (full length antigenomic S RNA). The sizes of the predicted amplification products are indicated in base pairs (bp) underneath. (C) Requirement of N protein to readthrough the transcriptional termination signal. Twenty four hours after infection, total cellular RNA was prepared and RT-PCR amplifications were performed as described in Section 2. The amplification products were analyzed by electrophoresis on agarose gels in TAE buffer. Lanes labeled ‘‘B’’ contain the amplification products of BHK-21 cells RNA and lanes ‘‘N’’ correspond to BHK-21 cells transiently expressing the N gene after transfection with pcDNA3.N. Virus infection of ‘‘B’’ and ‘‘N’’ cells is indicated by an open rectangle above the corresponding lanes (1, 2, 3, 4 and 7, 8, 9, 10). Similarly, protein synthesis inhibition is indicated by a filled rectangle (lanes 1, 2 and 7, 8). Uninfected ‘‘B’’ and ‘‘N’’ cells were used as controls (lanes 5, 6, 11 and 12).
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
51
Fig. 6. Expression and analysis of recombinant N protein. E. coli transformed with the pET-N or with pET with no insert, as a control, were induced with IPTG for protein expression. (A) Proteins of total bacterial lysates were separated by SDS – gel electrophoresis on 10% polyacrylamide gels and stained with Coomassie Brilliant Blue. (B) The N protein recovered from the insoluble fraction was separated by SDS–PAGE, blotted onto a nitrocellulose membrane, and reacted with a N-specific antiserum or (C) probed with 32P-labeled intergenic region (IGR, genome sense polarity, 495-nt) or (D) with radioactively labeled non-Junı´n – specific RNA. (M, molecular weight marker. N, lysate with N protein. C, negative control).
The N protein expressed in E. coli binds to the intergenic region of Junı´n virus in this assay as shown in Fig. 6c. Apparently the binding was either sequence-specific or secondary structurespecific, because all of these binding studies were done in the presence of yeast tRNA as a non-specific competitor RNA; in addition, a radioactively labeled non-viral RNA did not bind to the N protein to a significant extent (Fig. 6c, d). N protein may also binds to other S RNA regions in accordance with its recognized structural role. This kind of N protein – RNA interaction would explain the presence of a faint band in the N lane in Fig. 6D.
4. Discussion The evidence presented above of in vivo antiterminator activity and in vitro interaction with the intergenic S RNA sequence supports the hypothesis that N functions as a transcriptional antiterminator. Results shown in Fig. 2 indicate that transcription stops at different nucleotides in the intergenic region of Junı´n virus S RNA, showing heterogeneity at the 3% ends of both mRNAs (N and GPC mRNA). The heterogeneity of the 3% termini
has been previously acknowledged for Tacaribe and LCM viruses (Iapalucci et al., 1991; Meyer and Southern, 1993) In all of the predicted structures deduced from the RNase protection data, two stem loops are formed in both mRNAs (Fig. 3). The 3% termini of mRNAs can take up alternative secondary structures by using different schemes for intramolecular base pairing. It is always difficult to define the termination point of an RNA molecule synthesized in a living cell, in which the 3% end may eventually be modified by cellular protein factors. This implies that the different 3% ends might correspond to mRNA species that are at different processing stages, or they could represent the repertoire of alternative termination events during the transcription of subgenomic mRNA species. One of the elements specifying transcription termination in prokaryotes and in some eukaryotic genes, is the stem loop structure that forms at the 3% region of the RNA transcript (Greenblatt et al., 1993; Yarnell and Roberts, 1999). This is indeed the type of structure that can be ascribed to the 3% end sequence of the S-RNA-derived mRNAs of Junı´n. This structure has a high free energy of formation, DG 0 = − 44.9 to −39.54 kcal/mol (see Fig. 3). These observations support the hypothesis that the transcript structure, rather than
52
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
particular sequences, might be involved in arenavirus transcription termination (Franze-Fernandez et al., 1993; Romanowski, 1993). As shown in Fig. 5c, both N mRNA and antigenomic S RNA were synthesized in infected BHK-21 cells but N mRNA was detected only when cells were incubated with puromycin. These results show that the transcription termination activity of the intergenic sequences that were previously shown only for Tacaribe arenavirus (Franze-Fernandez et al., 1993) must now be extended to Junı´n virus as well. However, when cells containing previously synthesized N protein (BHK-NJUN) were infected, full-length antigenomic S RNA was synthesized in absence of ongoing protein synthesis. This indicates that the pre-existing N protein could help the RNA polymerase to bypass the transcriptional termination signal. Qualitatively identical results were obtained using stable expression clones isolated from the N gene transfected BHK-21 cells and selected using the neomycin analog G418 (not shown). Furthermore, the effect of N protein expression obtained from the MC2 strain of Junı´n virus, was tested on both the homologous MC2 strain and the genealogically unrelated Junı´n virus strain XJ44, with identical results. Our results lead to different interpretations. The binding of N protein to double stranded RNA or, more specifically, to hairpin structures found in the intergenic region of the S RNA (demonstrated in vitro), could have a higher binding constant than that corresponding to ssRNA regions. Therefore, statistically binding of N to the intergenic region would occur first and so determine that the S RNA is fully transcribed. This by no means disregards the fact that N protein binds to the whole S RNA in the nucleocapsid assembly process. Alternatively, when N concentration increases to high levels after N mRNA translation, it could bind to the nascent vc S RNA which in this way would be destined to become a full length copy of the S RNA template. It is also possible that the interaction of N with the transcription/replication complex could lead to antitermination. The schematic in Fig. 7 presents a model which illustrates this hypothesis.
At this point, our results cannot completely rule out an alternative hypothesis suggested by Franze-Fernandez et al. (1993), i.e. ribonuclease processing of a longer transcript to generate N mRNA. If this were the case, the ribonuclease digestion should be prevented by N in order to maintain a steady-state level of full length vc S RNA. In the absence of N protein, the putative RNase should digest the S RNA-derived transcript either in a processive manner, stopping at the intergenic secondary structure, or endonucleolytically in this same structure. However, we feel that the proposed antitermination model is much simpler than these alternative ones, which require either a cellular nuclease or a new function for one of the four virally encoded proteins. In addition, if a processive exoribonuclease were involved, we would expect to see partially digested products on a Northern blot. If the transcripts were processed by an specific endonuclease one would expect to find cleavage products in the infected cell. Neither of these situations are apparent from previously published results (Iapalucci et al., 1991; Franze-Fernandez et al., 1993). We have also examined our data taking into account the fact that the 5%ends of LCMV mRNAs are capped, whereas those of genomic and antigenomic RNAs have a triphosphate (Meyer and Southern, 1993, 1994). Considering this scenario, the decision of making mRNAs or full length RNAs (v or vc) could be driven by the presence or absence of the capped end in the primer used by the RNA polymerase and/or differences in the protein composition between transcription and replication complexes. As an example the transcription complex could be formed by RNA polymerase only and the replication complex could be formed by RNA polymerase plus N protein (the eventual participation of other cellular and/or viral proteins cannot be excluded at this point). N could influence the ability of the transcription–replication complex to select or process the primers used for initiation, as well as allowing the bypass of the termination at the intergenic region. Our data clearly show that N is required for the synthesis of full length antigenomic S RNA. How initiation and antitermination are linked remains to be addressed.
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
The switch from transcription to replication during arenavirus infection might depend on the level of intracellular N protein in a manner similar to that proposed for rhabdovirus (Patton et al., 1984). However, there is a requirement for ongoing viral protein synthesis during replication of rhabdo and paramyxovirus and their genomic RNAs have no conspicuous secondary structures at the intergenic regions (Patton et al., 1984; Pattnaik et al., 1992; Horikami et al., 1992). More recently, Hardy and Wertz (1998) described a transcriptional antitermination activity for the M2-1 protein of human respiratory syncytial virus (RSV). In general, the mechanism proposed here is similar to that of DNA-dependent systems. In the case of Junı´n virus the mechanism of an RNA-dependent RNA polymerase also allows to switch primary viral transcription to replication mode. In vitro experiments on Tacaribe virus suggest that both L RNA encoded proteins (Z and L) are
53
required for viral RNA replication and, to a lower degree, for the synthesis of subgenomic mRNAs as well (Garcin et al., 1993). Our present study does not rule out the putative regulatory effect of Z. In fact, the virions include Z and L proteins and the amounts of both proteins that are incorporated in the host cell during the entry phase of infection might be sufficient to carry out their functions. In contrast, the number of N molecules entering the cell during the infection would not be high enough to act as a transcriptional antiterminator and it is the concentration of N protein in the cytoplasm of the infected cell that determines the switch. Our conclusion on the role of N is strongly supported by the recent observation of Lee et al. (2000) that the replication of artificial arenavirus genomes requires N in addition to L, but apparently no Z is necessary. At present, it is not known whether the switch is defined simply by the concentration of N — as a result of N mRNA translation and N consumption in nucle-
Fig. 7. Model proposed for the antitermination mechanism at the arenavirus transcription – replication switch. At lows levels of N protein the RNA polymerase pauses when the 3% end of the transcribed RNA folds into a hairpin, leading to the termination of transcription. When the N mRNA is translated, the newly synthesized N protein can bind to the growing RNA or the transcription complex destroying the termination signal or making the RNA polymerase insensitive to it. Furthermore, the N protein binds to full length RNA to form the nucleocapsid, lowering the amount of free N, acting as antiterminator, and allowing for some N mRNA synthesis. Both functions of N (transcriptional control and nucleocapsid structure) are probably the basis for the simultaneous synthesis of N mRNA and full-length S RNA.
54
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55
ocapsid formation — or implies a more complex scheme involving interactions with other viral or cellular proteins or functions such as e.g. phosphorylation of components of the transcriptionreplication complex.
Acknowledgements We thank Andrew Ball (UAB, Birmingham) for the pTV(d0,d0) vector and for critically reading the manuscript. This work was supported by grants from Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICET), Comisio´n de Investigaciones Cientı´ficas de la Provincia de Buenos Aires (CIC BA), Fundacio´n Antorchas, Universidad Nacional de La Plata (UNLP) and Universidad Nacional de Quilmes to V. R. (V. R. is a Research Career Member of CONICET, Argentina). M. A. T. and D. M. P. hold fellowships from CONICET and UNLP, respectively.
References Auperin, D., Romanowski, V., Galinski, M.S., Bishop, D.H.L., 1984. Sequencing studies of Pichinde´ arenavirus S RNA indicate a novel coding strategy, an ambisense viral S RNA. Journal of Virology 52, 897–904. Chomczynski, P., Sacchi, N., 1987. Single step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Analytical Biochemistry 162, 156– 159. Cullen, B.R., 1994. RNA-sequence-mediated gene regulation in HIV-1. Infectious Agents Diseases 3, 68–76. Fearns, R., Collins, P.L., 1999. Role of the M2-1 transcription antitermination protein of respiratory syncytial virus in sequential transcription. Journal of Virology 73, 5852– 5864. Franze-Fernandez, M.T., Zetina, C., Iapalucci, S., Lucero, M.A., Boissou, C., Lopez, R., Rey, O., Daheli, M., Cohen, G., Zakin, M., 1987. Molecular structure and early events in the replication of Tacaribe Arenavirus S RNA. Virus Research 7, 309 – 324. Franze-Fernandez, M.T., Iapalucci, S., Lopez, N., Rossi, C., 1993. Subgenomic RNAs of Tacaribe virus. In: Salvato, S. (Ed.), The Arenaviridae. Plenum Press, New York, pp. 113 – 132. Garcin, D., Rochat, S., Kolakofsky, D., 1993. The Tacaribe arenavirus small zinc finger protein is required for both mRNA synthesis and genome replication. Journal of Virology 67, 807 – 812.
Ghiringhelli, P.D., Rivera Pomar, R.V., Lozano, M.E., Grau, O., Romanowski, V., 1991. Molecular organization of Junı´n virus S RNA: complete nucleotide sequence, relationship with the other members of Arenaviridae and unusual secondary structures. Journal of General Virology 72, 2129 – 2141. Greenblatt, J., Nodwell, J.R., Mason, S.W., 1993. Transcriptional antitermination. Nature 364, 401 – 406. Hardy, R.W., Wertz, G.W., 1998. The product of the respiratory syncytial virus M2 gene ORF1 enhances readthrough of intergenic junctions during viral transcription. Journal of Virology 72, 520 – 526. Horikami, S.M., Curran, J., Kolakofsky, D., Moyer, S.A., 1992. Complexes of Sendai virus NP-P and P-L proteins are required for defective interfering particle genome replication in vitro. Journal of Virology 66, 4901 – 4908. Iapalucci, S., Lopez, R., Rey, O., Lopez, N., Franze-Fernandez, M.T., Cohen, G.N., Lucero, M., Ochoa, A., Zakin, M.M., 1989a. Tacaribe virus L gene encodes a protein of 2210 amino acid residues. Virology 170, 40 – 47. Iapalucci, S., Lopez, N., Rey, O., Zakin, M.M., Cohen, G.N., Franze-Fernandez, M.T., 1989b. The 5% region of Tacaribe virus L RNA encodes a protein with a potential metal binding domain. Virology 173, 357 – 361. Iapalucci, S., Lo´pez, N., Franze-Ferna´ndez, M.T., 1991. The 3% end termini of Tacaribe arenavirus subgenomic RNAs. Virology 182, 269 – 278. Krumm, A., Meulia, T., Groudine, M., 1993. Common mechanisms for the control of eukaryotic transcriptional elongation. BioEssays 15, 659 – 665. Krumm, A., Hickey, L.B., Groudine, M., 1995. Promoterproximal pausing of RNA polymerase II defines a general rate-limiting step after transcription initiation. Genes & Development 9, 559 – 572. Lanford, R.E., Sureau, C., Jacob, J.R., White, R., Fuerst, T.R., 1994. Demonstration of in vitro infection of chimpanzee hepatocytes with hepatitis C virus using strand-specific RT/PCR. Virology 202, 606 – 614. Lee, K.J., Novella, I.S., Teng, M.N., Oldstone, M.B.A., Delatorre, J.C., 2000. NP and L proteins of lymphocytic choriomeningitis virus (LCMV) are sufficient for efficient transcription and replication of LCMV genomic RNA analogs. Journal of Virology 74, 3470 – 3477. Meyer, B.J., Southern, P.J., 1993. Concurrent sequence analysis of 5% and 3% RNA termini by intramolecular circularization reveals 5% nontemplated bases and 3% terminal heterogeneity for lymphocytic choriomeningitis virus mRNAs. Journal of Virology 67, 2621 – 2627. Meyer, B.J., Southern, P.J., 1994. Sequence heterogeneity in the termini of lymphocytic choriomeningitis virus genomic and antigenomic RNAs. Journal of Virology 68, 7659 – 7664. O’Brien, T., Hardin, S., Greenleaf, A., Lis, J.T., 1994. Phosphorylation of RNA polymerase II C-terminal domain and transcriptional elongation. Nature 370, 75 – 77. Pattnaik, A.K., Ball, L.A., Legrone, A., Wertz, G.W., 1992. Infectious defective interfering particles of VSV from transcripts of a cDNA clone. Cell 69, 1011 – 1020.
M.A. Tortorici et al. / Virus Research 73 (2001) 41–55 Patton, J.T., Davis, N.L., Wertz, G.W., 1984. N protein alone satisfies the requirement for protein synthesis during RNA replication of vesicular stomatitis virus. Journal of Virology 49, 303 – 309. Raju, R., Raju, L., Hacker, D., Carcin, D., Compans, R., Kolakofsky, D., 1990. Non-templated bases at the 5% ends of Tacaribe virus mRNAs. Virology 174, 53–59. Reichel, C., Maas, C., Schulze, S., Schell, J., Steinbiss, H.-H., 1996. Cooperative binding to nucleic acids by barley yellow mosaic bymovirus coat protein and characterization of a nucleic acid-binding domain. Journal of General Virology 77, 587 – 592. Rivera Pomar, R.V., 1991. Ana´lisis molecular del gen de la proteı´na de la nucleoca´pside del virus Junı´n. PhD Dissertation. Facultad de Ciencias Exactas, Universidad Nacional de La Plata. Argentina. Rivera Pomar, R.V., Manzella, EM., Ghiringhelli, P.D., Grau, O., Romanowski, V., 1991. Patterns of transient expression of the arenavirus nucleocapsid protein gene in transfected cells. Microscopı´a Electro´nica y Biologı´a Celular 15, 41– 56. Rivera Pomar, R.V., Ghiringhelli, P.D., Lozano, M.E., Albarin˜o, C.G., Romanowski, V., 1992. Transcription of Junı´n virus S RNA. NATO-EEC Course and Workshop on Regulation of Gene Expression in Animal Viruses, Mallorca, Spain Romanowski, V., 1993. Genetic organization of Junı´n virus, the ethiological agent of argentine hemorrhagic fever. In:
.
55
Salvato, S (Ed.), The Arenaviridae. Plenum Press, New York, pp. 51 – 83. Rustici, S.M., 1984. Desarrollo in vitro del virus Junı´n, proteı´nas intracelulares. PhD Dissertation. Facultad de Ciencias Exactas, Universidad Nacional de La Plata. Argentina. Salvato, M.S., Shimomaye, E.M., 1989. The completed sequence of lymphocytic choriomeningitis virus reveals a unique RNA structure and a gene for a zinc finger protein. Virology 173, 1 – 10. Salvato, M.S., Schweighofer, K.J., Burns, J., Shimomaye, E.M., 1992. Biochemical and immunological evidence that the 11 kDa zinc-binding protein of lymphocytic choriomeningitis virus is a structural component of the virus. Virus Research 22, 185 – 198. Sanchez, A., Pifat, D.Y., Kenyon, R.H., Peters, S.J., McCormick, J.B., Kiley, M.P., 1989. Junı´n virus monoclonal antibodies: characterization and crossreactivity with other Arenaviruses. Journal of General Virology 70, 1125 – 1132. Tam, A.W., White, R., Reed, E., Short, M., Zhang, Y., Fuerst, T.R., Lanford, R.E., 1996. In vitro propagation and production of hepatitis E virus from in vivo-infected primary macaque hepatocytes. Virology 215, 1 – 9. Weisberg, R.A., Gottesman, M.E., 1999. Processive antitermination. Journal of Bacteriology 181, 359 – 367. Yarnell, W.S., Roberts, J.W., 1999. Mechanism of intrinsic transcription termination and antitermination. Science 284, 611 – 615.
