Previous Article | Next Article ![]()
Journal of Virology, April 2007, p. 3866-3876, Vol. 81, No. 8
0022-538X/07/$08.00+0 doi:10.1128/JVI.02075-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Department of Microbiology, Mount Sinai School of Medicine, New York, New York 10029,1 Center for Biodefense and Emerging Infectious Diseases,2 Department of Pathology, University of Texas Medical Branch, Galveston, Texas 77555-06093
Received 21 September 2006/ Accepted 18 January 2007
|
|
|---|
|
|
|---|
EEEV possesses a single-stranded, positive-sense RNA genome of approximately 11.7 kb. The four viral nonstructural proteins (nsP1-4) are produced from the full-length genomic RNA as a polyprotein and carry out functions important for viral RNA synthesis and polyprotein processing. A second polyprotein, translated from the 26S subgenomic mRNA, gives rise to the three major structural proteins, namely, the capsid and the two envelope proteins, E1 and E2. The envelope proteins are involved in receptor recognition, virus attachment, penetration, and membrane fusion during viral entry. These proteins also cooperate with the capsid during viral assembly and release. Based on observations made with the capsids of other alphaviruses, the EEEV capsid is expected to encapsulate the viral genomic RNA, to interact with the viral glycoproteins during assembly, and to function as a protease to release itself from the nascent structural polypeptide chain (54, 63, 65).
The alpha/beta interferon (IFN-
/ß) system plays a major role in early innate immune responses and constitutes the first line of defense against viral infection. Production of IFN-
/ß is mediated by the activation of transcription factors, such as NF-
B, ATF-2/c-Jun, and interferon regulatory factors (IRFs). IFN-
/ß released from cells binds to the IFN-
/ß receptor, activating a Jak-STAT signaling cascade that leads to the phosphorylation and activation of STAT-1 and STAT-2 transcription factors. Activation (phosphorylation) of STAT-1 and STAT-2 leads to their heterodimerization, translocation to the nucleus, and association with p48/IRF-9 to form a complex that binds DNA sequences containing interferon-stimulated response elements and promotes the induction of specific gene expression (5, 29). IFN-
/ß regulates the transcription of numerous genes, a number of which encode antiviral proteins, and leads to the induction of an antiviral state in cells.
Due to the importance of the IFN response as an antiviral defense, many viruses have developed mechanisms to inhibit this response by means of blocking or inhibiting IFN production, IFN-mediated signaling, and/or the activity of IFN-induced gene products (3, 12, 26). For example, the influenza A virus NS1 protein has been shown to suppress the initiation of the IFN response (23, 49, 53, 66, 67, 74). In another example, the highly virulent Ebola virus carries two IFN antagonist proteins: VP35 blocks IFN production, and VP24 blocks IFN signaling. The ability of these proteins to inhibit different steps in the IFN response likely plays a role in the extreme virulence of Ebola virus (6, 11, 32, 33, 59).
Among the positive-strand RNA viruses, some members of the Flaviviridae family have been shown to carry proteins that block the IFN response. For example, the hepatitis C virus NS3-4A protease inhibits signals that lead to IFN-
/ß expression (9, 20, 42, 43). In addition, other flaviviruses inhibit IFN-induced Jak-STAT signaling. For example, the dengue virus nonstructural protein NS4B strongly inhibits IFN signaling, whereas NS4A and NS2A partially block IFN signaling (14, 38, 50, 51). In contrast, the flaviviruses Japanese encephalitis virus and Langat virus use the NS5 protein to block IFN signaling, suggesting that viruses from the same family that cause different spectra of human disease may use different mechanisms to overcome the IFN response (7, 44, 45).
Studies with alphaviruses have demonstrated the importance of the IFN system in controlling infection and disease (1, 22, 30, 60, 71, 73). Mice with a defect in the IFN-
/ß response succumb faster to infections with several alphaviruses than do wild-type mice (30, 60, 73). However, it remains unknown whether alphaviruses, specifically EEEV, carry proteins that inhibit the interferon response.
Previous studies with Sindbis virus (SINV), an Old World alphavirus, suggest that nsP2 is important for the suppression of the antiviral response, including the production of IFN-
/ß in infected cells (22). Whether the same applies to the New World encephalitic alphaviruses (i.e., EEEV and Venezuelan equine encephalitis virus), which cause a different and more severe spectrum of clinical disease in humans than that of the Old World viruses (65), has not been determined. Because of the evolutionary, genetic, biological, and human disease differences between the New World and Old World alphaviruses, we hypothesized that the mechanisms by which Old and New World alphaviruses evade the innate immune response might also differ. We therefore sought to identify specific EEEV proteins with IFN antagonist properties, anticipating that such factors will play an important role in EEEV pathogenesis. Our results indicate that the EEEV capsid functions as a potent IFN antagonist protein by globally inhibiting host cell gene expression. Mechanistically, we demonstrate that the capsid protein inhibits the expression of RNA polymerase II-transcribed genes and also promotes phosphorylation of the eukaryotic initiation factor 2
(eIF2
) subunit, suggesting capsid-mediated effects upon translation as well.
(Portions of the results of this study were presented at the 2006 National RCE Meeting, New York, NY, March 2006, and the American Society of Virology Meeting, Madison, WI, July 2006.)
|
|
|---|
Construction of plasmids. The EEEV strain FL93-939, isolated from a mosquito pool in Florida in 1993 and passaged once in Vero cells and once in a suckling mouse brain, was used for reverse transcription-PCR (RT-PCR) and cDNA cloning. Structural and nonstructural proteins were cloned into the pCAGGS mammalian expression plasmid and tagged with hemagglutinin (HA) in frame at the N terminus. The pCAGGS plasmid has been described previously (52), and it contains a chicken ß-actin promoter for strong transient expression of the inserted gene and a ß-globin poly(A) signal. For cloning of the individual nonstructural protein genes, the proteolytic cleavage sites in the nonstructural polyprotein genes were chosen as boundaries (65), and a stop codon was introduced at the end of each gene. For cloning of the structural proteins, a cassette carrying the complete structural polyprotein gene was constructed, as well as a plasmid encoding the capsid protein and carrying the E3 through E1 genes. To map the specific region in the capsid protein able to inhibit gene expression, the coding regions for the N terminus (amino acids [aa] 1 to 126) and C terminus (aa 127 to 261) were also cloned into the pCAGGS mammalian expression plasmid. Expression of the proteins was confirmed by Western blotting, using antibodies against the HA tag (Sigma, St. Louis, MO). The pM1 plasmid (4) was used in this study to clone the firefly luciferase gene under the control of the T7 promoter. This low-copy-number, ampicillin-resistant plasmid (a modified version of the pBR322 vector) was selected because it lacks a T7 promoter, allowing us to introduce this promoter immediately upstream of the firefly luciferase gene. Also, a T7 terminator sequence was introduced at the end of the luciferase gene. Construction of the PR8 NS1 plasmid has been reported previously (66).
Construction of EEEV replicons. Schematic representations of the EEEV replicons generated in this study are shown in Fig. 7A. Maps and sequences are available upon request. To produce cDNA templates for RNA synthesis, plasmids were linearized using the NotI site located downstream of the poly(A) tail. In vitro transcription was performed by using an mMessage mMachine T7 kit (Ambion, Austin, TX) following the manufacturer's protocol. In vitro-transcribed RNA was electroporated into BHK cells as previously described (4), and 1 day after electroporation, the medium was replaced with DMEM-10% FBS containing 12.5 µg of puromycin. Single puromycin-resistant cells were isolated and expanded, and RNAs were extracted from the cells for PCR and sequencing purposes. Cells were lysed in reporter lysis buffer, and lysates were assayed for luciferase activity following the manufacturer's protocol (Promega, Madison, WI).
![]() View larger version (16K): [in a new window] |
FIG. 7. The capsid gene is mutated in stably selected EEEV replicons. (A) Schematic representation of the EEEV replicons constructed for this study. nsP1 to nsP4, genes for nonstructural proteins 1 to 4; Luc, luciferase gene; PAC, puromycin acetyltransferase gene (encoding puromycin resistance); C, capsid gene; arrows with 26S, 26S subgenomic promoters. (B) BHK cells were electroporated with equal amounts of in vitro-transcribed replicon RNA, and 12 h after electroporation, puromycin selection was applied to the cells. Surviving cells were expanded, and RNAs were extracted from individual clones for RT-PCR amplification and sequence analysis. The figure shows the deletions found in the capsid and luciferase genes in the stable cell lines containing the EEEV replicons with the capsid gene. EEE rep capsid, EEEV replicon encoding capsid; EEE rep, EEEV replicon lacking capsid. The gray areas indicate the deleted regions in the stable capsid replicon cell lines.
|
Reporter gene assays.
Reporter assays were performed as previously described (11). To determine whether the EEEV proteins block IFN induction, 293T cells (
1 x 106 cells) were transfected with 1 µg of expression plasmid DNA together with an IFN-ß promoter-chloramphenicol acetyltransferase (CAT) reporter (pIFNß-CAT) and the indicated firefly luciferase plasmids, using the calcium phosphate mammalian transfection method (Stratagene, La Jolla, CA). One day after transfection, the cells were infected with Sendai virus at an MOI of
8 and incubated for 1 h at 37°C in 5% CO2. After incubation, the virus inoculum was removed, 1 ml of DMEM-10% FBS was added to the cells, and the plates were incubated at 37°C in 5% CO2 for 24 h. The cells were then lysed with reporter lysis buffer (Promega). CAT activities were determined by standard methods (61), and luciferase activities were determined according to the manufacturer's protocol (Promega). Firefly luciferase activity was used as a transfection control and to normalize the CAT activity where indicated. The results are represented as x-fold induction relative to the activity of an uninduced, empty vector-transfected control. To determine whether the EEEV proteins affect IFN signaling, Vero cells were transfected with 1 µg of expression plasmid together with an IFN-
/ß-responsive ISG54 promoter-CAT reporter (kindly provided by David Levy, NYU) and a firefly luciferase plasmid, using Lipofectamine 2000 (Invitrogen). One day after transfection, the cells were washed once with phosphate-buffered saline and then maintained in 10% DMEM containing 1,000 IU/ml of human IFN-ß (PBL, Piscataway, NJ). Mock controls were maintained in 10% DMEM without IFN. One day after treatment, cells were lysed by using a reporter lysis buffer, and the CAT and luciferase activities were determined. Firefly luciferase activity was used as a transfection control and to normalize the CAT activity. The results are shown as x-fold induction relative to the activity of an uninduced, empty vector-transfected control.
Transfection of the EEEV capsid with reporter plasmids. 293T cells were transfected with 1 µg of a GFP or firefly luciferase pCAGGS mammalian expression plasmid (whose expression is driven by the RNA polymerase II promoter) in the presence or absence of the plasmid encoding the capsid protein. Twenty-four hours later, cells transfected with the firefly luciferase plasmid were lysed with reporter lysis buffer following the manufacturer's protocol, and the firefly luciferase activity was measured in the cell lysates. Cells transfected with a GFP plasmid were observed under a fluorescence microscope for a GFP signal. In another set of experiments, BSRT7 cells were transfected with 1 µg of a firefly luciferase pCAGGS plasmid (under the control of an RNA polymerase II promoter) or 1 µg of a firefly luciferase pM1 plasmid (under the control of the T7 promoter) in the presence or absence of the plasmid carrying the capsid gene. Twenty-four hours later, cells were lysed and luciferase activity was measured as described above.
Detection of phosphorylated eIF2
.
Cell lysates were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. After electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane, blocked for 1 h in 5% nonfat dry milk dissolved in Tris-buffered saline, and then probed with antibodies against phospho-eIF2
(serine 52) (Biosource, Camarillo, CA), eIF2
(Biosource), or
-tubulin (Sigma, St. Louis, Missouri). Secondary antibodies conjugated to horseradish peroxidase and a chemiluminescence detection system from Perkin-Elmer (Wellesley, MA) were used to visualize the proteins. To determine the levels of eIF2
phosphorylation in EEEV-infected cells, 293T cells were infected with EEEV strain FL93-939 (MOI = 3), and supernatants and cell lysates were obtained at 5 and 24 h postinfection (p.i.). Cell lysates were then assayed for eIF2
as described above, and the supernatants were assayed for virus titer by a plaque assay in Vero cells as previously described (1).
Quantitative RT-PCR. Quantitative RT-PCR was performed by using a previously published SYBR green protocol with an ABI7900 HT thermal cycler (75). Briefly, 293T cells were transfected with 1 µg of PR8 NS1 plasmid together with 0, 0.25, 0.5, or 1 µg of the EEEV capsid plasmid. One day after transfection, proteins and RNAs were extracted simultaneously from the cells by using a PARIS kit (Ambion) following the manufacturer's protocol. RNAs were treated with DNA-free (Ambion) following the manufacturer's protocol, and 2 µg of RNA was used for cDNA synthesis, using 50 µM of oligo(dT)20 primers. To determine the effect of capsid on the endogenous, IFN-inducible ISG54 and STAT1 genes, 293T cells were transfected with 1 µg of empty plasmid or plasmid encoding full-length capsid, the capsid N terminus (aa 1 to 126), or the capsid C terminus (aa 127 to 261), and 24 h after transfection, the cells were treated with 1,000 IU of IFN-ß (PBL Biomedical Laboratories). Twelve hours after treatment, RNAs were extracted from the cells by using an RNeasy mini kit (QIAGEN, Valencia, CA) following the manufacturer's protocol. RNAs were treated with DNA-free (Ambion) following the manufacturer's protocol, and 1 µg of RNA was used for cDNA synthesis using 50 µM of oligo(dT)20 primers. To quantify the data, a robust global normalization algorithm was used. Each transcript in each sample was assayed three times, and the median cycle threshold (CT) values were used to calculate the transcript copy number by using the following formula: transcript copy number = 2,500 x 1.93(HKi TGi), where 2,500 is an empirical estimate of the number of actin mRNAs/cell (39), 1.93 reflects a typical reaction amplification efficiency of 93%, HKi is the housekeeping gene sample's median CT value, and TGi is the target CT value. Three housekeeping genes were used for global normalization in each experiment (actin, Rps11, and tubulin genes). For Western blot analysis, rabbit NS1 polyclonal and mouse HA monoclonal antibodies were used.
Selection of stable cell lines containing EEEV replicons. BHK-21 cells were electroporated with 5 µg of in vitro-synthesized RNA. Twenty-four hours after electroporation, the medium was replaced with DMEM-10% FBS containing 12.5 µg/ml of puromycin, and surviving cells were individually selected and expanded under puromycin selection. RNAs were extracted from the stable cell lines, and PCR and sequence analyses were performed to identify potential changes within the replicons.
|
|
|---|
/ß response, we employed a previously described assay that measures the ability of individual viral proteins to complement the replication of NDV-GFP (51, 55). Briefly, A549 cells were transfected with an empty expression plasmid or a plasmid encoding Ebola virus VP35, Nipah virus W, or individual EEEV proteins. At 1 day posttransfection, cells were infected with NDV expressing GFP, and 1 day later, the cells were analyzed for GFP expression. As previously reported, transfection of empty plasmid induced the production of IFN, and therefore the replication of NDV-GFP was inhibited. In contrast, cells transfected with plasmids encoding the known IFN antagonist proteins Ebola virus VP35 and Nipah virus W supported NDV-GFP replication (55). Interestingly, when plasmids encoding the EEEV proteins were transfected, only the capsid protein was able to rescue NDV-GFP replication (Fig. 1). To further address the possibility that one of the nonstructural proteins might also influence the IFN-
/ß response, the IFN response was also examined in cells stably transfected with EEEV replicons. However, in these cell lines, we did not detect either inhibition of IFN-ß promoter activation or inhibition of IFN-induced gene expression (data not shown), further supporting the view that the capsid is the primary inhibitor of IFN responses in EEEV-infected cells.
![]() View larger version (61K): [in a new window] |
FIG. 1. Expression of the EEEV capsid protein counteracts the antiviral effects of the IFN response. A549 cells were transfected with empty plasmid or a plasmid encoding GFP (as a control for transfection efficiency), Ebola virus VP35, Nipah virus W, or the indicated EEEV proteins, and 1 day later, cells were infected with NDV-GFP. Only plasmids expressing Ebola virus VP35, Nipah virus W, and EEEV capsid rescued NDV-GFP replication. The results shown are representative of at least three different experiments. The NDV-GFP panel shows replication of this virus in untransfected cells.
|
/ß production and/or the cellular response to IFN-
/ß. 293T cells or Vero cells were transfected with an IFN-ß promoter-CAT reporter plasmid or an interferon-responsive ISG54 promoter-CAT reporter plasmid together with a constitutively expressed simian virus 40 promoter-luciferase reporter plasmid which served as a control for transfection efficiency. The 293T cells were then infected with Sendai virus to activate the IFN-ß promoter, and the Vero cells were treated with 1,000 units/ml IFN-ß to activate the ISG54 promoter. The results showed that transfection of the capsid plasmid inhibited the expression of both Sendai virus- and IFN-ß-induced reporter genes (Fig. 2A and C). However, when the results of the reporter assay were normalized to the luciferase controls, we observed that the capsid protein also down-regulated the activity of the transfection control protein (firefly luciferase), suggesting that capsid may generally inhibit gene expression (Fig. 2B and D). To determine whether the inhibition of the reporters was due to a global inhibition of gene expression, we transfected the capsid together with either luciferase plasmid alone or with a GFP expression plasmid. The results demonstrated that transfection of the capsid inhibited the expression of both luciferase and GFP (Fig. 2E and F, respectively). These results suggest that expression of the capsid protein broadly inhibits the expression of multiple genes expressed from a variety of promoters.
![]() View larger version (28K): [in a new window] |
FIG. 2. EEEV capsid is a general inhibitor of gene expression. (A and B) Vero cells were cotransfected with the ISG54 promoter-CAT reporter plasmid, a constitutively expressed luciferase reporter plasmid, and empty vector or a plasmid expressing EEEV capsid or Nipah virus V protein. At 24 hours posttransfection, the cells were mock treated or treated with 1,000 U human IFN-ß/ml for 24 h. Cells were harvested and then assayed for CAT and luciferase activities. (A) Nonnormalized CAT activity, with the untreated empty vector control value set to 1. (B) Relative CAT activity upon normalization to the luciferase control. (C and D) 293T cells were cotransfected with the IFN-ß promoter-CAT reporter plasmid, a constitutively expressed luciferase reporter plasmid, and empty vector or a plasmid expressing EEEV nsP2, EEEV capsid, Ebola virus VP35, or Nipah virus W. At 24 hours posttransfection, the cells were mock infected or infected with Sendai virus (SeV) for 24 h. Cells were harvested and then assayed for CAT and luciferase activities. (C) Nonnormalized CAT activity, with the untreated empty vector control value set to 1. (D) CAT activity upon normalization to the luciferase control. The results are representative examples from a set of at least three separate experiments. (E) 293T cells were cotransfected with a luciferase reporter plasmid and either empty vector or a plasmid expressing EEEV capsid. At 24 h posttransfection, the cells were harvested and assayed for luciferase activity. The data represent the means ± standard errors among samples from three separately transfected wells. The experiment was repeated at least twice, with consistent results. (F) An experiment was performed as described for panel E, except that a plasmid expressing GFP was used and cells were examined by microscopy for GFP expression. The experiment was repeated at least twice, with consistent results.
|
![]() View larger version (39K): [in a new window] |
FIG. 3. Expression of EEEV capsid inhibits RNA polymerase II-directed gene expression. BSRT7 cells were cotransfected with a firefly luciferase reporter plasmid expressed from an RNA polymerase II promoter (in the plasmid pCAGGS) (52) or from a T7 promoter and with increasing concentrations of a plasmid expressing EEEV capsid. At 24 h posttransfection, the cells were harvested and assayed for luciferase activity. Average relative luciferase activities are reported, and the no-capsid control value was set to 100%. The error bars represent standard errors among samples from three separately transfected wells. The experiment was repeated at least twice, with consistent results.
|
subunit of eIF2.
The previous results suggest that the EEEV capsid protein targets cellular mRNA synthesis. However, some alphaviruses are known to activate protein kinase R (PKR) and consequently induce the phosphorylation of the
subunit of the translation initiation factor eIF2 (16). Therefore, to determine whether the capsid might induce the phosphorylation of eIF2
, 293T cells were transfected with a luciferase reporter plasmid in the absence or presence of a plasmid encoding the capsid protein, the entire EEEV structural polyprotein, or the LaCrosse virus (LCV) NSs protein, previously described as a general inhibitor of host cell transcription (58). Twelve or 24 h after transfection, cells were lysed and assayed for the presence of phosphorylated eIF2
. Phosphorylation of eIF2
was evident at 1 day posttransfection in cells expressing the EEEV capsid, EEEV structural proteins (including capsid), and LCV NSs but was strongest in cells where the capsid alone was expressed (Fig. 4A). The ability of the capsid to strongly induce eIF2
phosphorylation may contribute to the inhibition of gene expression noted above. How the capsid stimulates eIF2
phosphorylation remains to be determined. However, it is notable that LCV NSs, which like the EEEV capsid inhibits mRNA expression (see below), also promotes eIF2
phosphorylation. To determine whether capsid-induced eIF2
phosphorylation might be relevant to EEEV infection, 293T cells were infected with EEEV (MOI = 3), and supernatants and cell lysates were collected at 5 and 24 h p.i. Phosphorylation of eIF2
was more evident at 24 h p.i. than at 5 h p.i. (Fig. 4B and data not shown) and also correlated with higher virus titers than those at the 5-h time point (8.6 ± 0.04 log10 versus 5.3 ± 0.07 log10, respectively).
![]() View larger version (19K): [in a new window] |
FIG. 4. Expression of EEEV capsid protein induces phosphorylation of the subunit of eIF2. (A) 293T cells were cotransfected with a luciferase reporter plasmid and empty vector (vector) or a plasmid expressing EEEV capsid, EEEV structural proteins, or the LCV NSs protein. Twelve or 24 h after transfection, the cells were harvested, and Western blotting was performed to detect the phosphorylation of eIF2 , total eIF2 , and -tubulin as a loading control. Reductions in luciferase activity were observed for cells transfected with the capsid, the EEEV structural proteins, and LCV NSs (data not shown). (B) 293T cells were mock infected or infected with EEEV strain FL93-939 (MOI = 3), and at 5 and 24 h p.i., cells were lysed and Western blotting was performed to detect the phospho-eIF2 , total eIF2- , and -tubulin as a loading control.
|
![]() View larger version (37K): [in a new window] |
FIG. 5. Expression of EEEV capsid inhibits host mRNA accumulation. 293T cells were cotransfected with a plasmid expressing the influenza A/PR8/34 (H1N1) virus NS1 protein and either empty vector or increasing amounts of a plasmid expressing EEEV capsid. At 24 h posttransfection, RNAs were extracted from the cells for quantitative real-time RT-PCR (A), and cell lysates were used for Western blot analyses of NS1, HA capsid, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a loading control (B). The data represent means ± standard errors among samples from two independently transfected wells. The experiment was repeated twice and gave consistent results.
|
(Fig. 6C). To determine whether the effects of the capsid extend to endogenous gene expression, empty vector or plasmids encoding full-length capsid, the N terminus, and the C terminus were transfected into 293T cells, and the relative mRNA copy numbers of the ISG54 and STAT-1 genes were measured by quantitative RT-PCR after IFN treatment. In the presence of the full-length capsid or the capsid N terminus, the relative copy numbers of ISG54 and STAT-1 decreased two- and fourfold, respectively, compared to those of the control, whereas the effect of overexpressing the C-terminal half of the capsid had less of an effect on mRNA accumulation (Fig. 6D). It should be noted that these experiments, which rely upon transient transfection and leave some percentage of cells untransfected, likely underestimate the effects of capsid on endogenous gene expression. When the expression of each of the capsid constructs was analyzed by Western blotting (for the HA tag), the carboxy-terminal half was expressed at much higher levels than either the full-length capsid or the amino-terminal half, possibly reflecting the ability of the latter two constructs to inhibit gene expression (data not shown). These results suggest that the inhibition of protein synthesis mediated by the EEEV capsid is encoded within the amino-terminal half of the protein and is independent of its protease function.
![]() View larger version (31K): [in a new window] |
FIG. 6. The N terminus of the EEEV capsid mediates the inhibition of gene expression. (A) 293T cells were cotransfected with a plasmid expressing the GFP reporter and either empty vector or a plasmid expressing the N terminus (aa 1 to 126) or the C terminus (aa 127 to 261) of the EEEV capsid. At 24 h posttransfection, cells were examined by microscopy for GFP expression. (B) The experiment was performed as described for panel A, except that a luciferase reporter plasmid was used and the cells were harvested and assayed for luciferase activity. (C) 293T cells were cotransfected with a luciferase reporter plasmid and empty vector (vector) or a plasmid expressing EEEV full-length capsid or the N terminus (aa 1 to 126) or C terminus (aa 127 to 261) of the capsid. Twenty-four hours after transfection, the cells were harvested and Western blotting was performed to detect the phosphorylation of eIF2 , total eIF2 , and -tubulin as a loading control. (D) 293T cells were transfected with empty plasmid or with a plasmid encoding the full-length capsid or the N-terminal or C-terminal half of the capsid. At 24 h posttransfection, the cells were treated with human IFN-ß, and 12 h after treatment, RNAs were extracted from the cells for quantitative real-time RT-PCR of STAT-1 and ISG54 mRNAs. The data represent the means ± standard errors among samples from three wells in two experiments.
|
|
|
|---|
Previous studies have defined roles for alphavirus capsid proteins in proteolytic cleavage of the structural polyprotein, RNA encapsidation, nucleocapsid formation, and binding to ribosomes to facilitate the uncoating process during viral replication (25, 35, 40, 46, 56, 62, 64, 68, 69, 72). In addition, it was reported that the capsid of SFV induces phosphorylation of PKR, leading to inhibition of protein synthesis (16). Therefore, we first investigated whether the EEEV capsid might induce phosphorylation of the
subunit of eIF2, which would suggest an activation of PKR and a means of globally inhibiting protein synthesis. Overexpression of the EEEV capsid gene in 293T cells induced the phosphorylation of eIF2
. However, attempts to establish persistent replication of an EEEV replicon harboring the capsid gene in PKR/ mouse embryo fibroblasts were unsuccessful (data not shown), suggesting that the absence of PKR does not relieve the toxic effects of the capsid-expressing replicon. Therefore, it seems possible that other mechanisms are involved in the inhibition of host protein synthesis. Thus, we sought to determine whether the capsid gene also inhibits the accumulation of host cell mRNAs, which would be a function analogous to that of the Old World alphavirus nsP2 proteins (24, 27). Cotransfection of the capsid inhibited the expression of reporter genes driven by RNA polymerase II but not by the T7 reporter construct, suggesting a mechanism whereby the capsid targets some aspect of RNA polymerase II synthesis or induces host cell RNA degradation. This result also suggests that capsid-induced inhibition of reporter gene expression is not simply due to toxic effects of capsid overexpression.
To further determine the effects of untagged capsid in the context of viral replication, we evaluated host protein synthesis of EEEV replicons, with and without the capsid gene. As previously reported for other EEEV replicons (57), no adaptive viral mutations were identified in our EEEV replicon-bearing cells that did not carry capsid. However, only the EEEV replicon without the capsid gene was readily able to establish persistent replication in IFN-deficient cells, whereas capsid-containing replicons that did yield foci contained a complete or partial deletion of the N-terminal half of the capsid. This observation is reminiscent of previous findings with SINV replicons, in which adaptive mutations in nsP2 that reduced the cytopathic effect caused by this protein were required to establish persistent replication (21, 22, 24). These data support the conclusion that EEEV capsid is a general inhibitor of host cell gene expression. They also suggest that the capsid may contribute to cell death during the course of EEEV replication.
Our results demonstrate that the expression of EEEV capsid in cells is sufficient to counteract the antiviral effects of IFN. They therefore suggest a novel mechanism by which EEEV inhibits the IFN response, through capsid-mediated suppression of host cell gene expression, and suggest an additional and important function for the capsid protein in viral pathogenesis. These results also identify the N-terminal region of the capsid as responsible for this unique function, providing evidence that this function is protease independent. In this regard, EEEV appears to evade IFN responses in a manner different from hepatitis C virus, where the NS3-4A protease cleaves IPS-1 (13, 34, 42, 48) and TRIF (43) to inhibit activation of IRF3. Instead, this mechanism is similar to that proposed for the Old World alphavirus SINV, where nsP2 acts as a general inhibitor of gene expression (24, 27). However, further studies are necessary to map specific residues involved in the inhibition of host gene expression. The N terminus of the EEEV capsid is positively charged (rich in lysine and arginine residues) and also rich in proline and is believed to be important for protein-protein interactions and encapsidation of viral RNA (65). Whether these positively charged residues play a role in the capsid-mediated inhibition of the cellular response, perhaps by specifically binding host proteins or host RNAs or by localizing the capsid protein in specific cell compartments, needs to be determined. Interestingly, the capsid protein of SFV (an Old World alphavirus) has been suggested to have two nuclear localization signals (17), and computer-based analysis has identified potential nuclear localization signals within the N-terminal half of the EEEV capsid. Further experiments should explore whether the EEEV capsid localizes to the nucleus and/or sites of mRNA translation and whether localization is important for capsid inhibition of gene expression. Further studies should also be carried out to evaluate the importance of eIF2
phosphorylation in capsid inhibition of protein expression, perhaps by using cells expressing a nonphosphorylated form of eIF2
. Such experiments will clarify whether or not the inhibition of host cell mRNA accumulation and eIF2
phosphorylation separately contribute to capsid-mediated inhibition of gene expression or whether phosphorylation of eIF2
occurs as a consequence of the inhibition of host cell gene expression. It will also be of interest to determine whether capsid that enters the host cell with incoming viral particles may contribute to suppression of host cell gene expression.
The inhibition of protein synthesis as a way to inhibit the IFN response by alphaviruses is a mechanism shared with several other arboviruses, including members of the Bunyaviridae family (70). LCV and Bunyamwera and Rift Valley fever viruses have previously been shown to inhibit the IFN response by mediating inhibition of host protein synthesis (8, 10, 36, 70). Mechanistically, they interfere with RNA polymerase II transcription (8, 41, 70). Interestingly, vesicular stomatitis virus, which is also transmitted by insects, also inhibits IFN responses by globally targeting host cell gene expression (2, 15, 19).
These observations contrast with those made recently for arthropod-borne flaviviruses, such as West Nile and dengue viruses, which target specific components of the IFN induction or signaling pathway (7, 31, 47, 50, 51). Further experiments are needed to define the exact mechanism by which New World alphaviruses inhibit cellular transcription and the full consequences of this inhibition for viral replication.
Our results raise interesting questions regarding Old World and New World alphavirus pathogenesis and the different means of inhibiting gene expression. It will be of interest to determine whether the capsid protein is used to inhibit host protein synthesis by all New World alphaviruses or whether this property correlates with a given virus's ability to cause encephalitis. Future studies using capsid and nsP2 constructs derived from other alphaviruses, including Venezuelan equine encephalitis virus and Western equine encephalitis virus, are necessary to address this question. This study also provides important information about the specific mechanisms used by EEEV to evade the host response, and more importantly, we present evidence of a novel mechanism for the capsid protein in viral pathogenesis. This information may provide insights that facilitate the development of effective measures to control EEEV infection.
We thank Robert Tesh (University of Texas Medical Branch) for providing reagents and Larry Leung (Mount Sinai School of Medicine) for critically reading the manuscript.
Published ahead of print on 31 January 2007. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»