Dual Role of the Lymphocytic Choriomeningitis Viru
http://www.100md.com
病菌学杂志 2005年第6期
Department of Neuropharmacology, The Scripps Research Institute, La Jolla, California
ABSTRACT
Each genome segment of the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV), encodes two genes in ambisense orientation, separated by an intergenic region (IGR). The 3' ends of subgenomic viral mRNAs have been mapped to a stem-loop structure within the IGR, suggesting structure-dependent transcription termination. We have studied the role of the LCMV IGR by using a minigenome (MG) rescue system based on RNA analogues of the short genome segment. An ambisense MG coding for chloramphenicol acetyltransferase (CAT) and green fluorescent protein reporter genes instead of the nucleoprotein and glycoprotein open reading frames, respectively, served as a template for synthesis of full-length anti-MG (aMG) replicate and subgenomic size mRNA for reporter gene expression. An analogous MG without IGR was amplified by the virus polymerase with equal efficiency, but subgenomic mRNA was undetectable. Reporter gene expression from IGR-deficient aMG CAT-sense RNA of genomic length was approximately 5-fold less efficient than that from subgenomic CAT mRNA derived from an IGR-containing MG, but at least 100-fold more efficient than that from a T7 RNA polymerase transcript with the same sequence. Therefore, in the absence of IGR-mediated transcription termination, a fraction of full-length aMG RNA appears to behave as bona fide mRNA. Unexpectedly, MGs without IGR were dramatically impaired in their ability to passage reporter gene activity via infectious virus-like particles. These data suggest that the LCMV IGR serves individual functions in transcription termination for enhanced gene expression and in the virus assembly and/or budding, which are required for the efficient propagation of LCMV infectivity.
TEXT
Arenaviruses are enveloped viruses with a bisegmented negative-strand RNA genome. Each segment uses an ambisense coding strategy to direct the synthesis of two gene products. Besides lymphocytic choriomeningitis virus (LCMV), its prototype member, the arenavirus family comprises important human pathogens, such as Lassa fever virus (LFV) and the South American hemorrhagic fever viruses. The short (S; 3.4-kb) (31) segment of LCMV encodes the viral glycoprotein (GP) precursor (GP-C; 75 kDa) and nucleoprotein (NP; ca. 63 kDa). The large (L; 7.2-kb) (30, 32) segment carries the open reading frames (ORF) for the viral RNA-dependent RNA polymerase (RdRp) gene L (ca. 200 kDa) and for the small zinc finger protein Z (11 kDa). NP encapsidates the genome to generate the template recognized by the virus polymerase. NP and L represent the minimal viral transacting factors required for viral RNA transcription and replication (14), whereas Z is the main driving force of arenavirus budding (23). Studies using minireplicon systems for the Old World arenavirus LCMV (5, 6) as well as for the New World arenavirus Tacaribe virus (TV; 18) have revealed an inhibitory activity of Z on virus transcription and replication. Consistent with this finding, cells expressing LCMV or LFV Z proteins are refractory to infection with LCMV or LFV, respectively (7). For TV, this inhibitory activity was correlated with the ability of Z to interact with the L protein. The GP of LCMV is a type I transmembrane protein. Posttranslational processing of GP-C by the cellular protease S1P (16, 26) generates GP-1 (40 to 46 kDa) and GP-2 (35 kDa) polypeptides, which remain noncovalently associated (3). Oligomers of the GP-1/GP-2 complex are present as club-shaped projections at the surface of mature virions and mediate receptor binding and cell fusion. Each genome segment contains noncoding 5' and 3' untranslated regions (UTRs) and an intergenic region (IGR) (30-32). These sequences represent sufficient cis-acting elements for RNA replication and transcription mediated by the arenavirus polymerase (14). The replicase function of the RdRp directs synthesis of mostly encapsidated (29), uncapped (17) full-length antigenomic and genomic RNA species. These encapsidated RNA species serve as templates for the transcriptase activity of the viral RdRp to direct synthesis of subgenomic size, unencapsidated (29), capped (20, 29), and nonpolyadenylated (29) mRNA for translation of viral proteins. The 5' end of viral mRNAs contains, in addition to a cap structure, four to five random nontemplated bases (20), indicating a transcription initiation mechanism involving a cap-oligonucleotide primer of host origin, similar to cap-snatching of orthomyxoviruses. The 3' ends of the nonpolyadenylated viral mRNAs are heterogeneous and have been mapped within the IGR (21, 33). All arenavirus IGR sequences are predicted to fold into single or double stem-loop structures (10), suggesting a structure-dependent transcription termination mechanism reminiscent of rho-independent termination in prokaryotes (33, 35). For the New World arenavirus Junin, selective association of the IGR with the respective viral NP has been shown in vitro (33). No specific functions of the IGR have so far been experimentally confirmed, but a minigenome (MG) for TV containing solely the 5' and 3' UTRs as cis-acting elements was competent in genome amplification and reporter gene expression (18). This TV MG rescue system was based on T7 RNA polymerase provided from a recombinant vaccinia virus. In this case, the capping and polyadenylation activities provided by the enzymatic machinery of vaccinia virus introduce confounding factors for the assessment of the translation efficiency of RNA synthesized by the TV polymerase but lacking the IGR. Here, we present studies using a LCMV MG rescue system to address the role of the S segment IGR in the viral life cycle. This system is based on intracellular synthesis of a LCMV S segment genome analogue (MG) via transient transfection and mediated by either RNA polymerase I (Pol-I (25) or bacteriophage T7 RNA polymerase (T7), together with the use of polymerase II (Pol-II)-driven plasmids for the expression of viral proteins and T7 (15).
Effects of IGR on MG RNA expression. We have previously described a MG construct consisting of the 5' and 3' UTRs and the IGR of the LCMV Armstrong S segment that expresses the chloramphenicol acetyltransferase (CAT) reporter gene substituting for the NP ORF (MG/S-CAT; 15, 25). For the present study, we have generated an ambisense MG expressing an additional green fluorescent protein (GFP) reporter gene replacing the GP ORF (MG/S-CAT/GFP) (Fig. 1A). We also constructed MG/S-CATIGR and MG/S-CAT/GFPIGR by deleting the IGR in the two previous constructs, respectively (Fig. 1A). We used Northern blot analysis to determine levels of MG RNA and derived RNA species in transfected cells. Pol-I-mediated intracellular synthesis of MG RNA in the absence of the viral transacting factor L was unimpaired in constructs lacking an IGR (Fig. 1B, lane 4 versus lane 2 and lane 8 versus lane 6). Consistent with our previous findings (25), MG amplification was, however, strictly L dependent. This process was unimpaired in MGs lacking an IGR (Fig. 1B; lane 1 versus lane 2, lane 3 versus lane 4, lane 5 versus lane 6, and lane 7 versus lane 8). Likewise, anti-MG (aMG) RNA, the full-length replicative intermediate, was found in similar amounts in the presence or absence of an IGR (Fig. 1C, lane 3 versus lane 1 and lane 7 versus lane 5). Notably, subgenomic CAT mRNA was detected only in cells expressing either MG/S-CAT or MG/S-CAT/GFP and not in cells expressing the corresponding constructs lacking an IGR (Fig. 1C, lanes 1 and 5 versus lanes 3 and 7). Northern blot analysis has only limited resolution power for analysis of RNA species of similar sizes, which might have jeopardized the detection of MG/S-CATIGR-derived subgenomic CAT mRNA terminated at the 3'-end of the CAT ORF. Such an argument does not, however, apply to the MG/S-CAT/GFPIGR construct, thus confirming that transcription termination by the virus polymerase was critically dependent on the IGR. Besides the absence of a CAT mRNA band, we noticed a smear extending from below aMG/S-CAT/GFPIGR (Fig. 1C, lane 7 and data not shown), suggesting that a small fraction of the transcriptase complexes on IGR-deficient templates might have terminated prematurely. The lack of a distinct pattern and a very weak hybridization signal indicated, however, that low processivity rather than bona fide transcription termination was the underlying mechanism.
Effect of IGR on MG reporter gene expression. Consistent with our previous findings (25), CAT expression from MG/S-CAT (Fig. 2A, lane 1 versus lane 2) or MG/S-CAT/GFP (Fig. 2A, lane 5 versus lane 6) required coexpression of L and NP but not of GP or Z (data not shown). Likewise, GFP expression from MG/S-CAT/GFP (Fig. 2B) depended also on L and NP, despite the GFP ORF having a sense orientation with respect to the Pol-I promoter. This finding indicated that according to the arenavirus ambisense coding strategy, the primary Pol-I transcript was not efficiently translated. MG constructs lacking an IGR (MG/S-CATIGR and MG/S-CAT/GFPIGR) (Fig. 1A) also expressed CAT and GFP, respectively, but quantitative analysis of CAT activity by phosphorimager revealed reduced expression levels in the absence of an IGR that were in the fivefold range for each respective construct (Fig. 2A, lane 3 versus lane 1 and lane 7 versus lane 5). These reductions were independent of GP and Z polypeptides (data not shown).
Inefficient CAT translation from T7-derived full-length anti-MG RNA. The levels of reporter gene expression in the apparent absence of bona fide transcription seemed surprisingly high. We therefore considered the possibility that the single distinct RNA species of CAT sense (aMG) polarity derived from MG/S-CAT or MG/S-CAT/GFP (Fig. 1C, lanes 3 and 7) consisted indeed of a heterogeneous population of (i) uncapped bona fide aMG RNA and (ii) capped full-length antigenomic RNA active in translation. To investigate this possibility, we first examined the efficiency of CAT translation from RNA of full-length aMG sequence but lacking 5' cap and 3'-end polyadenylation modifications. For this experiment, we constructed pT7-aMGhrz and pT7-aMGdrz (Fig. 3A). Cotransfection of these constructs together with a plasmid pC-T7 (15) for Pol-II-driven expression of T7 generates a RNA of MG/S-CAT/GFPIGR complementary sequence with a 5' end modified by only two G residues derived from the minimal T7 promoter sequence. Upon self-processing of the hepatitis delta ribozyme (pT7-aMGdrz) (24) or of a sequence-adapted hairpin ribozyme (pT7-aMGhrz) (14), respectively, a 3' end identical to MG/S-CAT/GFP (pT7-aMGhrz) or extended by only a single C residue (pT7-aMGdrz) is generated. This T7 transcript coding for CAT in sense polarity should have yielded CAT activity if the flanking LCMV noncoding regions had, unexpectedly, allowed for efficient translation independent of 5'- and 3'-terminal modifications of the RNA. We compared aMG RNA levels and derived CAT activity obtained from pT7-aMGhrz and pT7-aMGdrz transcripts with the corresponding values for CAT sense RNA synthesized by the LCMV polymerase by using MG/S-CAT/GFPIGR as template (Fig. 3B to E; Table 1). Because self-processing of the ribozyme in the T7 transcripts was incomplete, we considered the possibility that either the processed RNA species alone, the unprocessed RNA species alone, or both RNA species served as templates for CAT translation, and we calculated the respective relative translation efficiency as arbitrary units of CAT activity per arbitrary amount of aMG RNA (Table 1). Even under the assumption that only processed aMG RNA derived from pT7-aMGdrz or pT7-aMGhrz was translated, the CAT activity per RNA molecule was approximately 100-fold below the translational activity of aMG RNA produced by the LCMV polymerase with MG/S-CAT/GFPIGR as template (Table 1). This large difference in translational efficiency was most likely an underestimation considering that (i) translation was unlikely to have discriminated between processed and unprocessed T7 transcripts and that (ii) a fraction of the aMG RNA band produced by the LCMV polymerase corresponds to encapsidated replicative RNA species that are not substrate for the translational machinery of the cell. This finding supported our hypothesis that the LCMV polymerase used the MG/S-CAT/GFPIGR template for a transcription-like process to generate CAT sense molecules of full antigenomic length that carried modifications for efficient translation, putatively a 5' cap structure. We cannot formally rule out the possibility that the small amount of heterogeneous CAT sense RNA molecules (corresponding to a weak smear of subgenomic aMG/S-CAT/GFPIGR RNA in Fig. 1C, lane 7) made a minor contribution to the CAT activity generated from MG/S-CAT/GFPIGR (Fig. 2A).
Effect of the IGR on VLP formation. We next examined a possible contribution of the IGR to virus assembly and budding by determining the ability of the different MG constructs to passage CAT or GFP reporter gene activity, or both, via infectious virus like particles (VLPs), following established procedures (23) (Fig. 4A and B). For this experiment, we coexpressed GP and Z proteins together with the MG expression plasmid and the minimal transacting factors NP and L. To assess the production of infectious VLPs, we passed supernatants from transfected cells onto fresh cell monolayers and used superinfection with LCMV as a source of transacting factors to drive reporter gene expression in the passage culture. As predicted, efficient VLP formation depended on L protein-mediated genome amplification in the transfection culture (Fig. 4A, lane 1 versus lane 2 and lane 5 versus lane 6; Fig. 4B, sample 5 versus sample 6). Notably, though, passage of CAT activity by MG/S-CATIGR and MG/S-CAT/GFPIGR was dramatically reduced compared to that by MG/S-CAT and MG/S-CAT/GFP, respectively (Fig. 4A, lane 1 versus lane 3 and lane 5 versus lane 7). We detected only very low, though specific, CAT activity in passage cultures of IGR-deficient MGs when CAT reactions were tuned for the highest sensitivity. However, detailed quantitative analysis revealed approximately 100-fold less efficient reporter gene passage by an IGR-deficient MG (data not shown). Similarly, GFP expression was widespread in passage cultures of MG/S-CAT/GFP but practically absent in passages from cells transfected with MG/S-CAT/GFPIGR (Fig. 4B, sample 5 versus sample 7). In some experiments, a small number of isolated GFP-positive cells could be detected very early (approximately 20 h) after passage of MG/S-CAT/GFPIGR culture supernatants (Fig. 4B, sample 7, single cell in passage culture). These cells disappeared, however, while GFP expression in MG/S-CAT/GFP passage cultures spread to form large foci at 48 to 72 h after transfer (data not shown). Similarly inefficient passage of reporter gene activity was observed when LCMV was added directly to the transfection culture (15) to allow passage of IGR-deficient MGs via reassortment (reference 15 and data not shown).
These results suggested that the IGR plays an important role in either assembly or budding of VLPs or in both processes. To address the relative budding efficiency of IGR-deficient MGs, we determined levels of VLPs present in supernatants (SN) of transfected cells by using established protocols (23). We transfected 293T cells with pT7-MG/S-CAT/GFP and pT7-MG/S-CAT/GFPIGR encoding the respective MG under control of a modified T7 promoter (24), followed downstream by a hepatitis delta ribozyme and the T7 terminator (Fig. 4C) (15). Autocatalytic cleavage of the ribozyme in the primary transcript generates the respective MG active in transcription and replication (14, 15). Cotransfection of pT7-MG/S-CAT/GFP or pT7-MG/S-CAT/GFPIGR together with pC-T7, pC-L, pC-NP, pC-GP, and pC-Z resulted in a pattern of CAT activity in transfection and passage cultures (data not shown) that recreated those found with the corresponding Pol-I constructs (Fig. 2A and B and 4A and B). We purified VLPs from the SN of transfected cells by ultracentrifugation, and MG RNA in VLPs was measured by semiquantitative reverse transcription (RT)-PCR (Fig. 4D). MG RNA was recovered from both MG-CAT/GFP and MG-CAT/GFPIGR. In both cases, detection of cell-free MG RNA resulted from a specific viral budding process, as demonstrated by its dependence on GP and Z (Fig. 4D, lane 1 versus lane 2 and lane 3 versus lane 4). The signal obtained from MG-CAT/GFPIGR was, however, reduced (about fivefold) compared to MG-CAT/GFP, indicating impaired budding activity of the IGR-deficient MG.
Based on the present data, we propose a mechanism for transcription termination and replication of LCMV as outlined in Fig. 5. Consistent with our previous findings (25) and as proposed for other NS RNA viruses (9, 27), the LCMV polymerase may exist in two distinct functional complexes, namely, as transcriptase or as replicase. The transcriptase initiates RNA synthesis by using a cap-oligonucleotide primer (20), while the replicase starts by a prime-and-realign mechanism (29). Only transcriptase complexes terminate within the IGR sequence, probably via a structure-dependent mechanism that operates mainly in the transcript rather than the template (35). Replicase complexes may be intrinsically resistant to structure-dependent termination, or the termination signal may be attenuated in the context of encapsidated and uncapped (17) replicate RNA but not of capped (20) transcripts. On IGR-deficient MG templates, transcriptase complexes proceed to the very end of the template due to the lack of an appropriate termination signal. These transcripts have 3' ends identical to those of the replicate but carry 5' nontemplated bases and a cap structure of host origin like bona fide transcripts. The fivefold reduction in CAT activity observed in this context could likely be due to an influence of 3' IGR sequences on translation efficiency, although alternative explanations could equally be envisaged. In any case, IGR-mediated transcription termination enhances apparently viral gene expression.
The very inefficient passage of reporter gene activity by MG/S-CATIGR and MG/S-CAT/GFPIGR was unexpected and raises the question of whether the IGR contains an encapsidation or packaging signal. Considering the fact that the LCMV polymerase recognizes only templates encapsidated by NP (14), efficient amplification of IGR-deficient MGs argued against a critical role of the IGR in encapsidation. This does not, however, rule out specific interactions of the LCMV NP with the IGR as reported for Junin virus (33). The IGR stem-loop parallels strong secondary RNA structures that serve as packaging signals in various viral families ranging from retroviruses to unenveloped icosahedral plant viruses (13, 19, 28, 34). Further, zinc binding domains found in almost all retroviral Gag proteins (2) are known to interact specifically with certain stem-loops in retroviral psi elements (1, 8) and thereby to recruit viral genomes into infectious particles (11). It therefore seems reasonable to speculate that the arenavirus Z protein may, in addition to its role in budding (23), contribute to virus assembly (23) at least in part via interaction with the IGR stem-loop, thereby recruiting LCMV genomes into particles. Our semiquantitative RT-PCR data showing reduced budding activity of IGR-deficient MGs would support this notion. Residual budding activity and reporter gene passage in the absence of an IGR indicate, however, that additional cis-acting sequence signals may also contribute to this process. At first sight, the modestly reduced budding of the IGR-deficient MG seems insufficient to fully explain the >100-fold reduction in GFP and CAT activity upon passage. But the formation of large GFP-positive foci by MG/S-CAT/GFP suggested that reporter gene activity in passage culture wells reflected the integral of multiple rounds of MG propagation. Thereby, even a modestly reduced propagation of IGR-deficient MGs may account for strongly reduced CAT activity upon multiple rounds of spreading. A modest impairment in budding could have an even more dramatic impact on the propagation of infectious LCMV in vivo. We cannot, however, rule out the possibility that reduced infectivity of IGR-deficient particles may have contributed to our experimental findings. Similar observations have been made with human immunodeficiency virus genomes exhibiting mutations in the stem-loop of the dimer initiation site that is contained within the psi element (4, 12, 22). Whether the IGR contains a putative S/L segment dimerization site and how this site may relate to arenavirus packaging and infectivity remains to be elucidated. A better understanding of these functions may provide the basis for the rational design of engineered infectious arenaviruses (26) with reduced propagation capacity as candidate live attenuated vaccines against arenaviral hemorrhagic fevers.
ACKNOWLEDGMENTS
This work was supported by a fellowship from the Gebert Rüf Stiftung, Switzerland, to Daniel D. Pinschewer and by NIH grant AI47140 to Juan Carlos de la Torre.
We thank Phi Lam for excellent technical assistance.
Present address: Institute of Experimental Immunology, Department of Pathology, University Hospital of Zurich, 8091 Zürich, Switzerland.
REFERENCES
Bacharach, E., and S. P. Goff. 1998. Binding of the human immunodeficiency virus type 1 Gag protein to the viral RNA encapsidation signal in the yeast three-hybrid system. J. Virol. 72:6944-6949.
Berg, J. M. 1986. Potential metal-binding domains in nucleic acid binding proteins. Science 232:485-487.
Buchmeier, M. J., P. J. Southern, B. S. Parekh, M. K. Wooddell, and M. B. Oldstone. 1987. Site-specific antibodies define a cleavage site conserved among arenavirus GP-C glycoproteins. J. Virol. 61:982-985.
Clever, J. L., and T. G. Parslow. 1997. Mutant human immunodeficiency virus type 1 genomes with defects in RNA dimerization or encapsidation. J. Virol. 71:3407-3414.
Cornu, T. I., and J. C. de la Torre. 2002. Characterization of the arenavirus RING finger Z protein regions required for Z-mediated inhibition of viral RNA synthesis. J. Virol. 76:6678-6688.
Cornu, T. I., and J. C. de la Torre. 2001. RING finger Z protein of lymphocytic choriomeningitis virus (LCMV) inhibits transcription and RNA replication of an LCMV S-segment minigenome. J. Virol. 75:9415-9426.
Cornu, T. I., H. Feldmann, and J. C. de la Torre. 2004. Cells expressing the RING finger Z protein are resistant to arenavirus infection. J. Virol. 78:2979-2983.
De Guzman, R. N., Z. R. Wu, C. C. Stalling, L. Pappalardo, P. N. Borer, and M. F. Summers. 1998. Structure of the HIV-1 nucleocapsid protein bound to the SL3 psi-RNA recognition element. Science 279:384-388.
Fearns, R., M. E. Peeples, and P. L. Collins. 1997. Increased expression of the N protein of respiratory syncytial virus stimulates minigenome replication but does not alter the balance between the synthesis of mRNA and antigenome. Virology 236:188-201.
Franze-Fernandez, M. T., S. Iapalucci, N. Lopez, and C. Rossi. 1993. Subgenomic RNAs of Tacaribe virus, p. 113-131. In M. S. Salvato (ed.), The Arenaviridae. Plenum Press, New York, N.Y.
Gorelick, R. J., L. E. Henderson, J. P. Hanser, and A. Rein. 1988. Point mutants of Moloney murine leukemia virus that fail to package viral RNA: evidence for specific RNA recognition by a "zinc finger-like" protein sequence. Proc. Natl. Acad. Sci. USA 85:8420-8424.
Hill, M. K., M. Shehu-Xhilaga, S. M. Campbell, P. Poumbourios, S. M. Crowe, and J. Mak. 2003. The dimer initiation sequence stem-loop of human immunodeficiency virus type 1 is dispensable for viral replication in peripheral blood mononuclear cells. J. Virol. 77:8329-8335.
Junker-Niepmann, M., R. Bartenschlager, and H. Schaller. 1990. A short cis-acting sequence is required for hepatitis B virus pregenome encapsidation and sufficient for packaging of foreign RNA. EMBO J. 9:3389-3396.
Lee, K. J., I. S. Novella, M. N. Teng, M. B. Oldstone, and J. C. de La Torre. 2000. NP and L proteins of lymphocytic choriomeningitis virus (LCMV) are sufficient for efficient transcription and replication of LCMV genomic RNA analogs. J. Virol. 74:3470-3477.
Lee, K. J., M. Perez, D. D. Pinschewer, and J. C. de la Torre. 2002. Identification of the lymphocytic choriomeningitis virus (LCMV) proteins required to rescue LCMV RNA analogs into LCMV-like particles. J. Virol. 76:6393-6397.
Lenz, O., J. ter Meulen, H. D. Klenk, N. G. Seidah, and W. Garten. 2001. The Lassa virus glycoprotein precursor GP-C is proteolytically processed by subtilase SKI-1/S1P. Proc. Natl. Acad. Sci. USA 98:12701-12705.
Leung, W. C., H. P. Ghosh, and W. E. Rawls. 1977. Strandedness of Pichinde virus RNA. J. Virol. 22:235-237.
López, N., R. Jacamo, and M. T. Franze-Fernández. 2001. Transcription and RNA replication of Tacaribe virus genome and antigenome analogs require N and L proteins: Z protein is an inhibitor of these processes. J. Virol. 75:12241-12251.
McBride, M. S., and A. T. Panganiban. 1996. The human immunodeficiency virus type 1 encapsidation site is a multipartite RNA element composed of functional hairpin structures. J. Virol. 70:2963-2973.
Meyer, B. J., and P. J. Southern. 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. J. Virol. 67:2621-2627.
Meyer, B. J., and P. J. Southern. 1994. Sequence heterogeneity in the termini of lymphocytic choriomeningitis virus genomic and antigenomic RNAs. J. Virol. 68:7659-7664.
Paillart, J. C., R. Marquet, E. Skripkin, C. Ehresmann, and B. Ehresmann. 1996. Dimerization of retroviral genomic RNAs: structural and functional implications. Biochimie 78:639-653.
Perez, M., R. C. Craven, and J. C. De La Torre. 2003. The small RING finger protein Z drives arenavirus budding: implications for antiviral strategies. Proc. Natl. Acad. Sci. USA 100:12978-12983.
Perez, M., and J. C. de la Torre. 2003. Characterization of the genomic promoter of the prototypic arenavirus lymphocytic choriomeningitis virus. J. Virol. 77:1184-1194.
Pinschewer, D. D., M. Perez, and J. C. De La Torre. 2003. Role of the virus nucleoprotein in the regulation of lymphocytic choriomeningitis virus transcription and RNA replication. J. Virol. 77:3882-3887.
Pinschewer, D. D., M. Perez, A. B. Sanchez, and J. C. de la Torre. 2003. Recombinant lymphocytic choriomeningitis virus expressing vesicular stomatitis virus glycoprotein. Proc. Natl. Acad. Sci. USA 100:7895-7900.
Qanungo, K. R., D. Shaji, M. Mathur, and A. K. Banerjee. 2004. Two RNA polymerase complexes from vesicular stomatitis virus-infected cells that carry out transcription and replication of genome RNA. Proc. Natl. Acad. Sci. USA 101:5952-5957.
Qu, F., and T. J. Morris. 1997. Encapsidation of turnip crinkle virus is defined by a specific packaging signal and RNA size. J. Virol. 71:1428-1435.
Raju, R., L. Raju, D. Hacker, D. Garcin, R. Compans, and D. Kolakofsky. 1990. Nontemplated bases at the 5' ends of Tacaribe virus mRNAs. Virology 174:53-59.
Salvato, M., E. Shimomaye, and M. B. Oldstone. 1989. The primary structure of the lymphocytic choriomeningitis virus L gene encodes a putative RNA polymerase. Virology 169:377-384.
Salvato, M., E. Shimomaye, P. Southern, and M. B. Oldstone. 1988. Virus-lymphocyte interactions. IV. Molecular characterization of LCMV Armstrong (CTL+) small genomic segment and that of its variant, Clone 13 (CTL–). Virology 164:517-522.
Salvato, M. S., and E. M. Shimomaye. 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.
Tortorici, M. A., C. G. Albarino, D. M. Posik, P. D. Ghiringhelli, M. E. Lozano, R. Rivera Pomar, and V. Romanowski. 2001. Arenavirus nucleocapsid protein displays a transcriptional antitermination activity in vivo. Virus Res. 73:41-55.
Weiss, B., U. Geigenmuller-Gnirke, and S. Schlesinger. 1994. Interactions between Sindbis virus RNAs and a 68 amino acid derivative of the viral capsid protein further defines the capsid binding site. Nucleic Acids Res. 22:780-786.
Yarnell, W. S., and J. W. Roberts. 1999. Mechanism of intrinsic transcription termination and antitermination. Science 284:611-615.(Daniel D. Pinschewer, Mar)
ABSTRACT
Each genome segment of the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV), encodes two genes in ambisense orientation, separated by an intergenic region (IGR). The 3' ends of subgenomic viral mRNAs have been mapped to a stem-loop structure within the IGR, suggesting structure-dependent transcription termination. We have studied the role of the LCMV IGR by using a minigenome (MG) rescue system based on RNA analogues of the short genome segment. An ambisense MG coding for chloramphenicol acetyltransferase (CAT) and green fluorescent protein reporter genes instead of the nucleoprotein and glycoprotein open reading frames, respectively, served as a template for synthesis of full-length anti-MG (aMG) replicate and subgenomic size mRNA for reporter gene expression. An analogous MG without IGR was amplified by the virus polymerase with equal efficiency, but subgenomic mRNA was undetectable. Reporter gene expression from IGR-deficient aMG CAT-sense RNA of genomic length was approximately 5-fold less efficient than that from subgenomic CAT mRNA derived from an IGR-containing MG, but at least 100-fold more efficient than that from a T7 RNA polymerase transcript with the same sequence. Therefore, in the absence of IGR-mediated transcription termination, a fraction of full-length aMG RNA appears to behave as bona fide mRNA. Unexpectedly, MGs without IGR were dramatically impaired in their ability to passage reporter gene activity via infectious virus-like particles. These data suggest that the LCMV IGR serves individual functions in transcription termination for enhanced gene expression and in the virus assembly and/or budding, which are required for the efficient propagation of LCMV infectivity.
TEXT
Arenaviruses are enveloped viruses with a bisegmented negative-strand RNA genome. Each segment uses an ambisense coding strategy to direct the synthesis of two gene products. Besides lymphocytic choriomeningitis virus (LCMV), its prototype member, the arenavirus family comprises important human pathogens, such as Lassa fever virus (LFV) and the South American hemorrhagic fever viruses. The short (S; 3.4-kb) (31) segment of LCMV encodes the viral glycoprotein (GP) precursor (GP-C; 75 kDa) and nucleoprotein (NP; ca. 63 kDa). The large (L; 7.2-kb) (30, 32) segment carries the open reading frames (ORF) for the viral RNA-dependent RNA polymerase (RdRp) gene L (ca. 200 kDa) and for the small zinc finger protein Z (11 kDa). NP encapsidates the genome to generate the template recognized by the virus polymerase. NP and L represent the minimal viral transacting factors required for viral RNA transcription and replication (14), whereas Z is the main driving force of arenavirus budding (23). Studies using minireplicon systems for the Old World arenavirus LCMV (5, 6) as well as for the New World arenavirus Tacaribe virus (TV; 18) have revealed an inhibitory activity of Z on virus transcription and replication. Consistent with this finding, cells expressing LCMV or LFV Z proteins are refractory to infection with LCMV or LFV, respectively (7). For TV, this inhibitory activity was correlated with the ability of Z to interact with the L protein. The GP of LCMV is a type I transmembrane protein. Posttranslational processing of GP-C by the cellular protease S1P (16, 26) generates GP-1 (40 to 46 kDa) and GP-2 (35 kDa) polypeptides, which remain noncovalently associated (3). Oligomers of the GP-1/GP-2 complex are present as club-shaped projections at the surface of mature virions and mediate receptor binding and cell fusion. Each genome segment contains noncoding 5' and 3' untranslated regions (UTRs) and an intergenic region (IGR) (30-32). These sequences represent sufficient cis-acting elements for RNA replication and transcription mediated by the arenavirus polymerase (14). The replicase function of the RdRp directs synthesis of mostly encapsidated (29), uncapped (17) full-length antigenomic and genomic RNA species. These encapsidated RNA species serve as templates for the transcriptase activity of the viral RdRp to direct synthesis of subgenomic size, unencapsidated (29), capped (20, 29), and nonpolyadenylated (29) mRNA for translation of viral proteins. The 5' end of viral mRNAs contains, in addition to a cap structure, four to five random nontemplated bases (20), indicating a transcription initiation mechanism involving a cap-oligonucleotide primer of host origin, similar to cap-snatching of orthomyxoviruses. The 3' ends of the nonpolyadenylated viral mRNAs are heterogeneous and have been mapped within the IGR (21, 33). All arenavirus IGR sequences are predicted to fold into single or double stem-loop structures (10), suggesting a structure-dependent transcription termination mechanism reminiscent of rho-independent termination in prokaryotes (33, 35). For the New World arenavirus Junin, selective association of the IGR with the respective viral NP has been shown in vitro (33). No specific functions of the IGR have so far been experimentally confirmed, but a minigenome (MG) for TV containing solely the 5' and 3' UTRs as cis-acting elements was competent in genome amplification and reporter gene expression (18). This TV MG rescue system was based on T7 RNA polymerase provided from a recombinant vaccinia virus. In this case, the capping and polyadenylation activities provided by the enzymatic machinery of vaccinia virus introduce confounding factors for the assessment of the translation efficiency of RNA synthesized by the TV polymerase but lacking the IGR. Here, we present studies using a LCMV MG rescue system to address the role of the S segment IGR in the viral life cycle. This system is based on intracellular synthesis of a LCMV S segment genome analogue (MG) via transient transfection and mediated by either RNA polymerase I (Pol-I (25) or bacteriophage T7 RNA polymerase (T7), together with the use of polymerase II (Pol-II)-driven plasmids for the expression of viral proteins and T7 (15).
Effects of IGR on MG RNA expression. We have previously described a MG construct consisting of the 5' and 3' UTRs and the IGR of the LCMV Armstrong S segment that expresses the chloramphenicol acetyltransferase (CAT) reporter gene substituting for the NP ORF (MG/S-CAT; 15, 25). For the present study, we have generated an ambisense MG expressing an additional green fluorescent protein (GFP) reporter gene replacing the GP ORF (MG/S-CAT/GFP) (Fig. 1A). We also constructed MG/S-CATIGR and MG/S-CAT/GFPIGR by deleting the IGR in the two previous constructs, respectively (Fig. 1A). We used Northern blot analysis to determine levels of MG RNA and derived RNA species in transfected cells. Pol-I-mediated intracellular synthesis of MG RNA in the absence of the viral transacting factor L was unimpaired in constructs lacking an IGR (Fig. 1B, lane 4 versus lane 2 and lane 8 versus lane 6). Consistent with our previous findings (25), MG amplification was, however, strictly L dependent. This process was unimpaired in MGs lacking an IGR (Fig. 1B; lane 1 versus lane 2, lane 3 versus lane 4, lane 5 versus lane 6, and lane 7 versus lane 8). Likewise, anti-MG (aMG) RNA, the full-length replicative intermediate, was found in similar amounts in the presence or absence of an IGR (Fig. 1C, lane 3 versus lane 1 and lane 7 versus lane 5). Notably, subgenomic CAT mRNA was detected only in cells expressing either MG/S-CAT or MG/S-CAT/GFP and not in cells expressing the corresponding constructs lacking an IGR (Fig. 1C, lanes 1 and 5 versus lanes 3 and 7). Northern blot analysis has only limited resolution power for analysis of RNA species of similar sizes, which might have jeopardized the detection of MG/S-CATIGR-derived subgenomic CAT mRNA terminated at the 3'-end of the CAT ORF. Such an argument does not, however, apply to the MG/S-CAT/GFPIGR construct, thus confirming that transcription termination by the virus polymerase was critically dependent on the IGR. Besides the absence of a CAT mRNA band, we noticed a smear extending from below aMG/S-CAT/GFPIGR (Fig. 1C, lane 7 and data not shown), suggesting that a small fraction of the transcriptase complexes on IGR-deficient templates might have terminated prematurely. The lack of a distinct pattern and a very weak hybridization signal indicated, however, that low processivity rather than bona fide transcription termination was the underlying mechanism.
Effect of IGR on MG reporter gene expression. Consistent with our previous findings (25), CAT expression from MG/S-CAT (Fig. 2A, lane 1 versus lane 2) or MG/S-CAT/GFP (Fig. 2A, lane 5 versus lane 6) required coexpression of L and NP but not of GP or Z (data not shown). Likewise, GFP expression from MG/S-CAT/GFP (Fig. 2B) depended also on L and NP, despite the GFP ORF having a sense orientation with respect to the Pol-I promoter. This finding indicated that according to the arenavirus ambisense coding strategy, the primary Pol-I transcript was not efficiently translated. MG constructs lacking an IGR (MG/S-CATIGR and MG/S-CAT/GFPIGR) (Fig. 1A) also expressed CAT and GFP, respectively, but quantitative analysis of CAT activity by phosphorimager revealed reduced expression levels in the absence of an IGR that were in the fivefold range for each respective construct (Fig. 2A, lane 3 versus lane 1 and lane 7 versus lane 5). These reductions were independent of GP and Z polypeptides (data not shown).
Inefficient CAT translation from T7-derived full-length anti-MG RNA. The levels of reporter gene expression in the apparent absence of bona fide transcription seemed surprisingly high. We therefore considered the possibility that the single distinct RNA species of CAT sense (aMG) polarity derived from MG/S-CAT or MG/S-CAT/GFP (Fig. 1C, lanes 3 and 7) consisted indeed of a heterogeneous population of (i) uncapped bona fide aMG RNA and (ii) capped full-length antigenomic RNA active in translation. To investigate this possibility, we first examined the efficiency of CAT translation from RNA of full-length aMG sequence but lacking 5' cap and 3'-end polyadenylation modifications. For this experiment, we constructed pT7-aMGhrz and pT7-aMGdrz (Fig. 3A). Cotransfection of these constructs together with a plasmid pC-T7 (15) for Pol-II-driven expression of T7 generates a RNA of MG/S-CAT/GFPIGR complementary sequence with a 5' end modified by only two G residues derived from the minimal T7 promoter sequence. Upon self-processing of the hepatitis delta ribozyme (pT7-aMGdrz) (24) or of a sequence-adapted hairpin ribozyme (pT7-aMGhrz) (14), respectively, a 3' end identical to MG/S-CAT/GFP (pT7-aMGhrz) or extended by only a single C residue (pT7-aMGdrz) is generated. This T7 transcript coding for CAT in sense polarity should have yielded CAT activity if the flanking LCMV noncoding regions had, unexpectedly, allowed for efficient translation independent of 5'- and 3'-terminal modifications of the RNA. We compared aMG RNA levels and derived CAT activity obtained from pT7-aMGhrz and pT7-aMGdrz transcripts with the corresponding values for CAT sense RNA synthesized by the LCMV polymerase by using MG/S-CAT/GFPIGR as template (Fig. 3B to E; Table 1). Because self-processing of the ribozyme in the T7 transcripts was incomplete, we considered the possibility that either the processed RNA species alone, the unprocessed RNA species alone, or both RNA species served as templates for CAT translation, and we calculated the respective relative translation efficiency as arbitrary units of CAT activity per arbitrary amount of aMG RNA (Table 1). Even under the assumption that only processed aMG RNA derived from pT7-aMGdrz or pT7-aMGhrz was translated, the CAT activity per RNA molecule was approximately 100-fold below the translational activity of aMG RNA produced by the LCMV polymerase with MG/S-CAT/GFPIGR as template (Table 1). This large difference in translational efficiency was most likely an underestimation considering that (i) translation was unlikely to have discriminated between processed and unprocessed T7 transcripts and that (ii) a fraction of the aMG RNA band produced by the LCMV polymerase corresponds to encapsidated replicative RNA species that are not substrate for the translational machinery of the cell. This finding supported our hypothesis that the LCMV polymerase used the MG/S-CAT/GFPIGR template for a transcription-like process to generate CAT sense molecules of full antigenomic length that carried modifications for efficient translation, putatively a 5' cap structure. We cannot formally rule out the possibility that the small amount of heterogeneous CAT sense RNA molecules (corresponding to a weak smear of subgenomic aMG/S-CAT/GFPIGR RNA in Fig. 1C, lane 7) made a minor contribution to the CAT activity generated from MG/S-CAT/GFPIGR (Fig. 2A).
Effect of the IGR on VLP formation. We next examined a possible contribution of the IGR to virus assembly and budding by determining the ability of the different MG constructs to passage CAT or GFP reporter gene activity, or both, via infectious virus like particles (VLPs), following established procedures (23) (Fig. 4A and B). For this experiment, we coexpressed GP and Z proteins together with the MG expression plasmid and the minimal transacting factors NP and L. To assess the production of infectious VLPs, we passed supernatants from transfected cells onto fresh cell monolayers and used superinfection with LCMV as a source of transacting factors to drive reporter gene expression in the passage culture. As predicted, efficient VLP formation depended on L protein-mediated genome amplification in the transfection culture (Fig. 4A, lane 1 versus lane 2 and lane 5 versus lane 6; Fig. 4B, sample 5 versus sample 6). Notably, though, passage of CAT activity by MG/S-CATIGR and MG/S-CAT/GFPIGR was dramatically reduced compared to that by MG/S-CAT and MG/S-CAT/GFP, respectively (Fig. 4A, lane 1 versus lane 3 and lane 5 versus lane 7). We detected only very low, though specific, CAT activity in passage cultures of IGR-deficient MGs when CAT reactions were tuned for the highest sensitivity. However, detailed quantitative analysis revealed approximately 100-fold less efficient reporter gene passage by an IGR-deficient MG (data not shown). Similarly, GFP expression was widespread in passage cultures of MG/S-CAT/GFP but practically absent in passages from cells transfected with MG/S-CAT/GFPIGR (Fig. 4B, sample 5 versus sample 7). In some experiments, a small number of isolated GFP-positive cells could be detected very early (approximately 20 h) after passage of MG/S-CAT/GFPIGR culture supernatants (Fig. 4B, sample 7, single cell in passage culture). These cells disappeared, however, while GFP expression in MG/S-CAT/GFP passage cultures spread to form large foci at 48 to 72 h after transfer (data not shown). Similarly inefficient passage of reporter gene activity was observed when LCMV was added directly to the transfection culture (15) to allow passage of IGR-deficient MGs via reassortment (reference 15 and data not shown).
These results suggested that the IGR plays an important role in either assembly or budding of VLPs or in both processes. To address the relative budding efficiency of IGR-deficient MGs, we determined levels of VLPs present in supernatants (SN) of transfected cells by using established protocols (23). We transfected 293T cells with pT7-MG/S-CAT/GFP and pT7-MG/S-CAT/GFPIGR encoding the respective MG under control of a modified T7 promoter (24), followed downstream by a hepatitis delta ribozyme and the T7 terminator (Fig. 4C) (15). Autocatalytic cleavage of the ribozyme in the primary transcript generates the respective MG active in transcription and replication (14, 15). Cotransfection of pT7-MG/S-CAT/GFP or pT7-MG/S-CAT/GFPIGR together with pC-T7, pC-L, pC-NP, pC-GP, and pC-Z resulted in a pattern of CAT activity in transfection and passage cultures (data not shown) that recreated those found with the corresponding Pol-I constructs (Fig. 2A and B and 4A and B). We purified VLPs from the SN of transfected cells by ultracentrifugation, and MG RNA in VLPs was measured by semiquantitative reverse transcription (RT)-PCR (Fig. 4D). MG RNA was recovered from both MG-CAT/GFP and MG-CAT/GFPIGR. In both cases, detection of cell-free MG RNA resulted from a specific viral budding process, as demonstrated by its dependence on GP and Z (Fig. 4D, lane 1 versus lane 2 and lane 3 versus lane 4). The signal obtained from MG-CAT/GFPIGR was, however, reduced (about fivefold) compared to MG-CAT/GFP, indicating impaired budding activity of the IGR-deficient MG.
Based on the present data, we propose a mechanism for transcription termination and replication of LCMV as outlined in Fig. 5. Consistent with our previous findings (25) and as proposed for other NS RNA viruses (9, 27), the LCMV polymerase may exist in two distinct functional complexes, namely, as transcriptase or as replicase. The transcriptase initiates RNA synthesis by using a cap-oligonucleotide primer (20), while the replicase starts by a prime-and-realign mechanism (29). Only transcriptase complexes terminate within the IGR sequence, probably via a structure-dependent mechanism that operates mainly in the transcript rather than the template (35). Replicase complexes may be intrinsically resistant to structure-dependent termination, or the termination signal may be attenuated in the context of encapsidated and uncapped (17) replicate RNA but not of capped (20) transcripts. On IGR-deficient MG templates, transcriptase complexes proceed to the very end of the template due to the lack of an appropriate termination signal. These transcripts have 3' ends identical to those of the replicate but carry 5' nontemplated bases and a cap structure of host origin like bona fide transcripts. The fivefold reduction in CAT activity observed in this context could likely be due to an influence of 3' IGR sequences on translation efficiency, although alternative explanations could equally be envisaged. In any case, IGR-mediated transcription termination enhances apparently viral gene expression.
The very inefficient passage of reporter gene activity by MG/S-CATIGR and MG/S-CAT/GFPIGR was unexpected and raises the question of whether the IGR contains an encapsidation or packaging signal. Considering the fact that the LCMV polymerase recognizes only templates encapsidated by NP (14), efficient amplification of IGR-deficient MGs argued against a critical role of the IGR in encapsidation. This does not, however, rule out specific interactions of the LCMV NP with the IGR as reported for Junin virus (33). The IGR stem-loop parallels strong secondary RNA structures that serve as packaging signals in various viral families ranging from retroviruses to unenveloped icosahedral plant viruses (13, 19, 28, 34). Further, zinc binding domains found in almost all retroviral Gag proteins (2) are known to interact specifically with certain stem-loops in retroviral psi elements (1, 8) and thereby to recruit viral genomes into infectious particles (11). It therefore seems reasonable to speculate that the arenavirus Z protein may, in addition to its role in budding (23), contribute to virus assembly (23) at least in part via interaction with the IGR stem-loop, thereby recruiting LCMV genomes into particles. Our semiquantitative RT-PCR data showing reduced budding activity of IGR-deficient MGs would support this notion. Residual budding activity and reporter gene passage in the absence of an IGR indicate, however, that additional cis-acting sequence signals may also contribute to this process. At first sight, the modestly reduced budding of the IGR-deficient MG seems insufficient to fully explain the >100-fold reduction in GFP and CAT activity upon passage. But the formation of large GFP-positive foci by MG/S-CAT/GFP suggested that reporter gene activity in passage culture wells reflected the integral of multiple rounds of MG propagation. Thereby, even a modestly reduced propagation of IGR-deficient MGs may account for strongly reduced CAT activity upon multiple rounds of spreading. A modest impairment in budding could have an even more dramatic impact on the propagation of infectious LCMV in vivo. We cannot, however, rule out the possibility that reduced infectivity of IGR-deficient particles may have contributed to our experimental findings. Similar observations have been made with human immunodeficiency virus genomes exhibiting mutations in the stem-loop of the dimer initiation site that is contained within the psi element (4, 12, 22). Whether the IGR contains a putative S/L segment dimerization site and how this site may relate to arenavirus packaging and infectivity remains to be elucidated. A better understanding of these functions may provide the basis for the rational design of engineered infectious arenaviruses (26) with reduced propagation capacity as candidate live attenuated vaccines against arenaviral hemorrhagic fevers.
ACKNOWLEDGMENTS
This work was supported by a fellowship from the Gebert Rüf Stiftung, Switzerland, to Daniel D. Pinschewer and by NIH grant AI47140 to Juan Carlos de la Torre.
We thank Phi Lam for excellent technical assistance.
Present address: Institute of Experimental Immunology, Department of Pathology, University Hospital of Zurich, 8091 Zürich, Switzerland.
REFERENCES
Bacharach, E., and S. P. Goff. 1998. Binding of the human immunodeficiency virus type 1 Gag protein to the viral RNA encapsidation signal in the yeast three-hybrid system. J. Virol. 72:6944-6949.
Berg, J. M. 1986. Potential metal-binding domains in nucleic acid binding proteins. Science 232:485-487.
Buchmeier, M. J., P. J. Southern, B. S. Parekh, M. K. Wooddell, and M. B. Oldstone. 1987. Site-specific antibodies define a cleavage site conserved among arenavirus GP-C glycoproteins. J. Virol. 61:982-985.
Clever, J. L., and T. G. Parslow. 1997. Mutant human immunodeficiency virus type 1 genomes with defects in RNA dimerization or encapsidation. J. Virol. 71:3407-3414.
Cornu, T. I., and J. C. de la Torre. 2002. Characterization of the arenavirus RING finger Z protein regions required for Z-mediated inhibition of viral RNA synthesis. J. Virol. 76:6678-6688.
Cornu, T. I., and J. C. de la Torre. 2001. RING finger Z protein of lymphocytic choriomeningitis virus (LCMV) inhibits transcription and RNA replication of an LCMV S-segment minigenome. J. Virol. 75:9415-9426.
Cornu, T. I., H. Feldmann, and J. C. de la Torre. 2004. Cells expressing the RING finger Z protein are resistant to arenavirus infection. J. Virol. 78:2979-2983.
De Guzman, R. N., Z. R. Wu, C. C. Stalling, L. Pappalardo, P. N. Borer, and M. F. Summers. 1998. Structure of the HIV-1 nucleocapsid protein bound to the SL3 psi-RNA recognition element. Science 279:384-388.
Fearns, R., M. E. Peeples, and P. L. Collins. 1997. Increased expression of the N protein of respiratory syncytial virus stimulates minigenome replication but does not alter the balance between the synthesis of mRNA and antigenome. Virology 236:188-201.
Franze-Fernandez, M. T., S. Iapalucci, N. Lopez, and C. Rossi. 1993. Subgenomic RNAs of Tacaribe virus, p. 113-131. In M. S. Salvato (ed.), The Arenaviridae. Plenum Press, New York, N.Y.
Gorelick, R. J., L. E. Henderson, J. P. Hanser, and A. Rein. 1988. Point mutants of Moloney murine leukemia virus that fail to package viral RNA: evidence for specific RNA recognition by a "zinc finger-like" protein sequence. Proc. Natl. Acad. Sci. USA 85:8420-8424.
Hill, M. K., M. Shehu-Xhilaga, S. M. Campbell, P. Poumbourios, S. M. Crowe, and J. Mak. 2003. The dimer initiation sequence stem-loop of human immunodeficiency virus type 1 is dispensable for viral replication in peripheral blood mononuclear cells. J. Virol. 77:8329-8335.
Junker-Niepmann, M., R. Bartenschlager, and H. Schaller. 1990. A short cis-acting sequence is required for hepatitis B virus pregenome encapsidation and sufficient for packaging of foreign RNA. EMBO J. 9:3389-3396.
Lee, K. J., I. S. Novella, M. N. Teng, M. B. Oldstone, and J. C. de La Torre. 2000. NP and L proteins of lymphocytic choriomeningitis virus (LCMV) are sufficient for efficient transcription and replication of LCMV genomic RNA analogs. J. Virol. 74:3470-3477.
Lee, K. J., M. Perez, D. D. Pinschewer, and J. C. de la Torre. 2002. Identification of the lymphocytic choriomeningitis virus (LCMV) proteins required to rescue LCMV RNA analogs into LCMV-like particles. J. Virol. 76:6393-6397.
Lenz, O., J. ter Meulen, H. D. Klenk, N. G. Seidah, and W. Garten. 2001. The Lassa virus glycoprotein precursor GP-C is proteolytically processed by subtilase SKI-1/S1P. Proc. Natl. Acad. Sci. USA 98:12701-12705.
Leung, W. C., H. P. Ghosh, and W. E. Rawls. 1977. Strandedness of Pichinde virus RNA. J. Virol. 22:235-237.
López, N., R. Jacamo, and M. T. Franze-Fernández. 2001. Transcription and RNA replication of Tacaribe virus genome and antigenome analogs require N and L proteins: Z protein is an inhibitor of these processes. J. Virol. 75:12241-12251.
McBride, M. S., and A. T. Panganiban. 1996. The human immunodeficiency virus type 1 encapsidation site is a multipartite RNA element composed of functional hairpin structures. J. Virol. 70:2963-2973.
Meyer, B. J., and P. J. Southern. 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. J. Virol. 67:2621-2627.
Meyer, B. J., and P. J. Southern. 1994. Sequence heterogeneity in the termini of lymphocytic choriomeningitis virus genomic and antigenomic RNAs. J. Virol. 68:7659-7664.
Paillart, J. C., R. Marquet, E. Skripkin, C. Ehresmann, and B. Ehresmann. 1996. Dimerization of retroviral genomic RNAs: structural and functional implications. Biochimie 78:639-653.
Perez, M., R. C. Craven, and J. C. De La Torre. 2003. The small RING finger protein Z drives arenavirus budding: implications for antiviral strategies. Proc. Natl. Acad. Sci. USA 100:12978-12983.
Perez, M., and J. C. de la Torre. 2003. Characterization of the genomic promoter of the prototypic arenavirus lymphocytic choriomeningitis virus. J. Virol. 77:1184-1194.
Pinschewer, D. D., M. Perez, and J. C. De La Torre. 2003. Role of the virus nucleoprotein in the regulation of lymphocytic choriomeningitis virus transcription and RNA replication. J. Virol. 77:3882-3887.
Pinschewer, D. D., M. Perez, A. B. Sanchez, and J. C. de la Torre. 2003. Recombinant lymphocytic choriomeningitis virus expressing vesicular stomatitis virus glycoprotein. Proc. Natl. Acad. Sci. USA 100:7895-7900.
Qanungo, K. R., D. Shaji, M. Mathur, and A. K. Banerjee. 2004. Two RNA polymerase complexes from vesicular stomatitis virus-infected cells that carry out transcription and replication of genome RNA. Proc. Natl. Acad. Sci. USA 101:5952-5957.
Qu, F., and T. J. Morris. 1997. Encapsidation of turnip crinkle virus is defined by a specific packaging signal and RNA size. J. Virol. 71:1428-1435.
Raju, R., L. Raju, D. Hacker, D. Garcin, R. Compans, and D. Kolakofsky. 1990. Nontemplated bases at the 5' ends of Tacaribe virus mRNAs. Virology 174:53-59.
Salvato, M., E. Shimomaye, and M. B. Oldstone. 1989. The primary structure of the lymphocytic choriomeningitis virus L gene encodes a putative RNA polymerase. Virology 169:377-384.
Salvato, M., E. Shimomaye, P. Southern, and M. B. Oldstone. 1988. Virus-lymphocyte interactions. IV. Molecular characterization of LCMV Armstrong (CTL+) small genomic segment and that of its variant, Clone 13 (CTL–). Virology 164:517-522.
Salvato, M. S., and E. M. Shimomaye. 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.
Tortorici, M. A., C. G. Albarino, D. M. Posik, P. D. Ghiringhelli, M. E. Lozano, R. Rivera Pomar, and V. Romanowski. 2001. Arenavirus nucleocapsid protein displays a transcriptional antitermination activity in vivo. Virus Res. 73:41-55.
Weiss, B., U. Geigenmuller-Gnirke, and S. Schlesinger. 1994. Interactions between Sindbis virus RNAs and a 68 amino acid derivative of the viral capsid protein further defines the capsid binding site. Nucleic Acids Res. 22:780-786.
Yarnell, W. S., and J. W. Roberts. 1999. Mechanism of intrinsic transcription termination and antitermination. Science 284:611-615.(Daniel D. Pinschewer, Mar)