Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jul 25.
Published in final edited form as: Virology. 2009 Jun 18;390(2):338–347. doi: 10.1016/j.virol.2009.05.030

Type I interferon induction is correlated with attenuation of a South American eastern equine encephalitis virus strain in mice

Christina L Gardner 1, Jun Yin 1, Crystal W Burke 1, William B Klimstra 1, Kate D Ryman 1,*
PMCID: PMC3404509  NIHMSID: NIHMS134348  PMID: 19539968

Abstract

North American eastern equine encephalitis virus (NA-EEEV) strains cause high mortality in humans, whereas South American strains (SA-EEEV) are typically avirulent. To clarify mechanisms of SA-EEEV attenuation, we compared mouse-attenuated BeAr436087 SA-EEEV, considered an EEEV vaccine candidate, with mouse-virulent NA-EEEV strain, FL93-939. Although attenuated, BeAr436087 initially replicated more efficiently than FL93-939 in lymphoid and other tissues, inducing systemic IFN-α/β release, whereas FL93-939 induced little. BeAr436087 was more virulent than FL93-939 in IFN-α/β-deficient mice, confirming that type I IFN responses determined attenuation, but the viruses were similarly sensitive to IFN-α/β priming in vitro. Infection with BeAr436087 protected against FL93-939 disease/death, even when given 8 h afterward, suggesting that the environment produced by BeAr436087 infection attenuated FL93-939. We conclude that avoidance of IFN-α/β induction is factor for FL93-939. Furthermore, BeAr436087 could be used for vaccination and therapeutic treatment in the event of exposure to NA-EEEV during a bioterrorism attack.

Keywords: Alphavirus, interferon, lymphoid tissue, attenuation, EEEV, STAT1

Introduction

Eastern equine encephalitis virus (EEEV) is a positive-sense, single-stranded RNA virus belonging to the genus Alphavirus, family Togaviridae. Areas where EEEV is endemic range from the northeastern through the southeastern United States and into Latin and South America. EEEV is the sole virus species in the EEE antigenic complex, but phylogenetic trees have revealed distinct North American and South American lineages (Weaver et al., 1999). North American EEEV (NA-EEEV) isolates are highly conserved, differing by less than 2% in their nucleotide sequences over many years and a large geographic range (Weaver et al., 1999;Armstrong et al., 2008;Young et al., 2008). In contrast, South American EEEV (SA-EEEV) isolates evolve more rapidly and at least three genotypes can be delineated that differ by up to 25% in their nucleotide sequence (Kondig et al., 2007;Weaver et al., 1999).

The case to fatality rate for NA-EEEV infection of humans is 30-70%, with 35-80% of survivors suffering permanent, severe neurological impairments (Smith et al., 1997). In addition, NA-EEEV is also lethal in up to 90% of recognized equine cases (Rua-Domenech et al., 1999). The severity of human disease, coupled with the lack of licensed human vaccines or antiviral therapies, makes NA-EEEV a threat as both an emerging infectious disease and a biowarfare/terrorism (BWT) agent. SA-EEEV strains, on the other hand, are generally avirulent in humans and only occasionally cause disease in equines. Only two human cases of SA-EEEV encephalitis have been reported (Alice, 1956;Corniou et al., 1972), although seroprevalence associated with exposure can be as high as 25-66% (Causey & Theiler, 1958;Aguilar et al., 2005;Aguilar et al., 2007a). Although the differential virulence of NA versus SA strains in humans remains uncertain, correlation with sensitivity to the interferon alpha/beta (IFN-α/β) gamma (IFN-γ)-induced antiviral state has been demonstrated in Vero cells (Aguilar et al., 2005;Aguilar et al., 2008a).

In mice, the majority of EEEV strains exhibit a similar degree of virulence, with high mortality (Aguilar et al., 2005;Gardner et al., 2008;Vogel et al., 2005). However, SA-EEEV strain, BeAr436087 (Weaver et al., 1999), is relatively attenuated in mice compared with other strains, causing no mortality following subcutaneous administration to adult animals (Aguilar et al., 2005;Aguilar et al., 2008a). Chimeric viruses, in which the non-structural or structural protein encoding regions of the BeAr436087 genome were exchanged with those of virulent NA-EEEV isolate, FL93-939, implicated both regions in the attenuated phenotype (Aguilar et al., 2008a). Furthermore, BeAr436087 immunizes mice against subsequent challenge with a virulent NA-EEEV strain (Wang et al., 2007) and, thus, has potential for use as a live-attenuated vaccine against virulent NA-EEEV strains.

In an attempt to determine correlates of human- and mouse-attenuation of the SA-EEEV BeAr436087 strain that could be applied to rational design of alphavirus vaccines, as well as to shed light on the mechanisms underlying the virulence of NA-EEEV strains, we have performed detailed pathogenesis studies in mice comparing the association of replication potential and type I IFN induction/sensitivity with disease pathogenesis of BeAr436087 virus with that of the human- and mouse-virulent NA-EEEV FL93-939 strain. Surprisingly, in the absence of overt disease, BeAr436087 virus infection resulted in earlier and higher level virus replication in lymphoid and other non-neural tissues of mice than the uniformly fatal FL93-939 virus infection. Attenuation correlated with induction of high levels of IFN-α/β in the serum of BeAr436087-infected mice, while little or no IFN-α/β could be detected in sera from FL93-939-infected in animals. The difference in serum IFN-α/β was functionally significant as phosphorylation of the IFN signaling-activated transcription factor STAT1 was greater in the brains of BeAr436087-infected mice, a tissue associated with EEEV disease. Further supportive of a role for the IFN-α/β response in the attenuation of BeAr436087, this virus was significantly more virulent than FL93-939 for mice deficient in type I IFN responses. Differences in relative sensitivity of the two viruses to a pre-established, IFN-α/β-induced antiviral state in vitro were minor and did not appear to account for the virulence differences in vivo. Finally, infection of normal mice with BeAr436087 virus could protect from disease and reduce mortality caused by FL93-939 infection, even if the BeAr436087 virus was given 8 h after FL93-939, consistent with the BeAr436087-induced IFN-α/β acting limit the virulence to of FL93-939. From these data, we infer that FL93-939 virulence in the mouse model is primarily achieved by avoidance of IFN-α/β induction, while the attenuation of BeAr436087 virus is associated with a robust type I IFN induction and response.

Results

Mouse-virulence of BeAr436087, but not FL93-939, is strongly age-dependent

Most strains of EEEV are virulent for adult mice inoculated subcutaneously (Aguilar et al., 2005;Vogel et al., 2005). However, Weaver and coworkers identified a South American strain of EEEV (BeAr436087) that was attenuated for adult mice (Aguilar et al., 2005), but neurovirulent in 6-day-old mice inoculated intracerebrally (Wang et al., 2007). We have previously demonstrated that age-dependent attenuation of Sindbis virus (SINV) after subcutaneous inoculation, occurs coincident with increasingly curtailed replication in extraneural tissues between 5 and 11 d post-partum, associated with transition from 100% to 0% mortality and suggestive of the antiviral action of IFN-α/β (Ryman et al., 2007;Ryman et al., 2000). To determine the relative attenuation of BeAr436087 compared with SINV, we examined the age-dependence of BeAr436087 virulence versus the adult mouse-virulent NA-EEEV strain, FL93-939. Beginning with newborns, mice of increasing age were inoculated subcutaneously in the ventral thorax with equal doses of FL93-939 or BeAr436087 and monitored for signs of morbidity and mortality. As expected, FL93-939 caused severe disease and death in all ages of mice, although the average survival time (AST) extended slightly with increasing host development (Fig. 1A, B and C). BeAr436087 virus infection was similarly virulent in mice 1-day-old through 2-week-old, but the AST extended significantly and mortality rates decreased between 2 and 4 weeks of age, such that all 4-week-old mice survived. Therefore, while attenuated in older mice compared with FL93-939, BeAr436087 is significantly more virulent than SINV.

Figure 1. Age-dependence of BeAr436087 and FL93-939 virulence in mice.

Figure 1

Open in a new tab

Morbidity and mortality data from CD1 mice infected subcutaneously with 103 pfu of BeAr436087 or FL93-939. Age-dependence of mortality (A) and AST (B) in mice inoculated subcutaneously in the ventral thorax with BeAr436087 (hatched bar) or FL93-939 (black bar). Percentage survival (C) and average daily weight change (D) of 6 week-old mice infected in the hind footpad with BeAr436087 (open circle) or FL93-939 (closed circle). Error bars represent standard deviations.

Six-week-old CD1 mice infected subcutaneously with BeAr436087 exhibited minimal clinical sign of disease (Fig. 1C), but were nevertheless protected from challenge with FL93-939 virus three weeks later (data not shown), as previously observed (Wang et al., 2007). Interestingly, when we examined morbidity in 6-week-old mice (Fig. 1D), the avirulent BeAr436087 virus was found to cause greater weight loss than virulent FL93-939 virus in the early stages of infection, 1-2 d post-infection (p.i.; p < 0.01). For the related alphavirus, Venezuelan equine encephalitis virus (VEEV), early weight loss is typically associated with the so-called “lymphotrophic” stage of disease (Charles et al., 2001;Gardner et al., 2008) and a proinflammatory response (Grieder et al., 1997), whereas FL93-939 infection displays greatly reduced lymphotropism (Gardner et al., 2008). Thus, we speculated that the tropism of BeAr436087 and FL93-939 might differ, with a more pronounced lymphotropic phase in BeAr436087 infection.

BeAr436087 disseminates and amplifies more rapidly than FL93-939 in adult mice

To investigate the mechanism(s) underlying the early morbidity, as well as the ultimate attenuation of BeAr436087, we compared the replication and dissemination of BeAr436087 in 6-week-old outbred CD1 mice, an age at which BeAr436087 is attenuated regardless of inoculation route while FL93-939 is uniformly fatal. Data for FL93-939 infection were previously published (Gardner et al., 2008), but collected alongside the current experiments and, therefore, included in figures for clarity. We examined lymphoid tissues including draining lymph node (DLN) and spleen, mesenchymal tissues including femur epiphyses, metaphyses and bone aspirate, and brain, infection of which is associated with NA-EEEV-induced mortality (Vogel et al., 2005;Gardner et al., 2008;Aguilar et al., 2005;Wang et al., 2007). Overall, BeAr436087 replication was distinctly different from that of FL93-939 (Fig. 2). Perhaps counterintuitively, attenuated BeAr436087 initially replicated to higher levels than virulent FL93-939 in all tissues, suggesting that BeAr436087 exhibits higher fitness for early disseminated replication in vivo.

Figure 2. Comparison of BeAr436087 and FL93-939 replication and dissemination.

Figure 2

Open in a new tab

6-week-old CD1 mice were infected with 103 pfu of BeAr436087 (open circle) or FL93-939 (closed circle) in each hind footpad. Mice were perfused with PBS-1% DCS prior to tissue harvesting, homogenization and mechanical disruption. Virus titers in serum and tissue homogenates were determined by BHK-21 plaque titration: A) popliteal draining lymph node; B) serum; C) spleen; D) knee joint (femur epiphyses and metaphyses); E) bone aspirate (femur); F) brain. LD = limit of detection of the plaque assay and error bars represent standard deviations.

Previously it was observed that NA-EEEV strains including FL93-939 amplified poorly in the DLN and spleen (Vogel et al., 2005;Gardner et al., 2008), which we attributed to inefficient replication of the virus in myeloid-lineage cells such as macrophages and DCs (Gardner et al., 2008). Levels of BeAr436087 virus in the DLN surpassed input inoculum titers within 6 h p.i. (Fig. 2A) while FL93-93 barely exceeded input even at peak titer 18 h p.i. Furthermore, peak BeAr436087 titers were approximately 100-fold higher than FL93-939, suggestive that BeAr436087 was able either to replicate more efficiently in similar cells or in a different cell-type in the DLN than FL93-939.

Both viruses produced a serum viremia; however, BeAr436087 viremia was over 10-fold higher and slightly extended in duration compared to FL93-939 (Fig. 2B). When dissemination of the two viruses was compared (Fig. 2C—E), BeAr436087 virus was detected in the spleen 6 h earlier than FL93-939, and BeAr436087 titers in the spleen were dramatically higher than FL93-939 (>100-fold at 24 and 48 h p.i. and ~100-fold at 72 h p.i.), suggesting that BeAr436087 replicates with considerably greater efficiency in lymphoid and reticuloendothelial tissues. These data were consistent with the findings of Aguilar et al. (2008a). Differences between the two viruses were not as dramatic in the knee joint and bone aspirates, although BeAr436087 replicated more rapidly and to higher peak titers (>10-fold higher in both tissues; Fig. 2D and E).

Both viruses were detectable in the central nervous system (CNS) by 24 h p.i (Fig. 2F), but, once again BeAr436087 virus loads were higher than FL93-939 at early times. By 48 h p.i., however, clearance of BeAr436087 from the brain was evident and this virus was undetectable by 120 h p.i. In contrast, FL93-939 virus titers continued to increase until death, achieving peak titers >10,000-fold higher than were measured in BeAr436087-infected animals. These data imply that, reflective of higher and earlier replication of BeAr436087 in peripheral tissues, neuroinvasion occurs earlier. However, BeAr436087 replication is curtailed, while that of FL93-939 is poorly restrained, correlated with outcome.

IFN-α/β responses determine attenuation of Be Ar436087 in 6-week-old mice

To determine directly, the role of the innate immune response, specifically the type I IFN response, versus the adaptive immune response in the attenuation of BeAr436087 in comparison with FL93-939, we tested virus virulence in a panel of mice deficient in specific components of both immune response arms (Table 1). In the absence of functioning B and T cell activities, (RAG−/− mice), BeAr436087 remained avirulent, suggesting that these functions are not responsible or required for the high degree of attenuation. FL93-939 exhibited a slightly decreased AST and increased mortality in these mice versus C57BL/6 controls (at this dose mortality is less than 100% in C57BL/6) suggesting some role for adaptive responses in control of FL93-939 virulence. Shortened survival was primarily associated with higher FL93-939 titers in the spinal cords and brains of moribund RAG−/− mice (data not shown).

Table 1.

Mortality rates and average survival time of different mouse strains infected with EEEV.

Mouse Strain
Virus CD1 C57BL/6 RAG1−/− 129 Sv/Ev IFNGR−/− IFNAR−/− IFNAGR−/− STAT1−/−
FL93-939 Percent Mortality 100% 75% 100% 75% 100% 100% 100% 100%
AST (days) 5.0 ± 0.7 7.0 ± 2.5 5.9 ± 1.1 6.1 ± 0.9 4.9 ± 0.8 2.4 ± 0.5 2.4 ± 0.5 2.4 ± 0.3
BeAr436087 Percent Mortality 0% 0% 0% 0% 0% 100% 100% 100%
AST (days) N/A N/A N/A N/A N/A 1.8 ± 0.3 1.9 ± 0.5 2.1 ± 0.3
Open in a new tab

When we examined virulence in mice deficient in IFN-γ signaling, BeAr436087 infection did not result in mortality, similar to RAG−/− mice, FL93-939-infected IFNGR−/− mice exhibited an increase in mortality and decrease in AST suggestive of a minor role of the IFN-γ pathway in protection from disease. In contrast, both viruses caused 100% mortality in mice deficient in IFN-α/β signaling (IFNAR1−/−), both IFN-α/β and IFN-γ signaling (IFNAGR−/−) or the STAT1 transcription factor (STAT1−/−; Table 1). Furthermore, mortality occurred slightly more rapidly with BeAr436087 than FL93-939 in all three types of mice, although the differences were not highly significant (p = 0.06 for IFNAR1−/−; p = 0.07 for IFNAGR−/− and p = 0.20 for STAT1−/−). Thus, consistent with early titration data from CD1 mice, BeAr436087 exhibits equal and in some cases superior replication fitness to that of FL93-939 in vivo and the attenuation of BeAr436087 in adult mice is primarily attributable to the antiviral effects of type I IFN.

To determine the effects of IFN-α/β signaling in the pathogenesis of BeAr436087 in more detail, we examined the replication and dissemination of the virus in the three types of mice with deficits in the IFN-α/β response (IFNAR1−/−, IFNAGR1−/− and STAT1−/− strains). Serum and perfused tissues were collected to titer virus at 24 h p.i. (Fig. 3), a time at which no mice would have succumbed to infection, both BeAr436087 and FL93-939 have invaded the brain and their replication is distinguishable in CD1 mice. Each of the IFN-α/β signaling-defective mouse strains experienced higher replication of BeAr436087 than controls in all measured tissues and serum (Fig. 3A), indicating that the effects of disruption of IFN-α/β signaling upon mortality were correlated with a large increase in virus growth. In more detailed titrations comparing serum titers of STAT1−/− and 129 Sv/Ev control mice, the appearance of higher replication in the STAT1−/− animals was not observed until 18 h p.i. consistent with a requirement for induced IFN-α/β signaling for protection in the immunocompetent mice (data in not shown). Virus titers in STAT1−/− mice were significantly lower than those in IFNAR1−/− and IFNAGR−/− mice in spleen, liver and brain (p < 0.01), indicating a STAT1-independent effect of IFN signaling in these tissues. We also compared the replication of FL93-939 and BeAr436087 in the brains of IFNAR1−/− mice to determine if BeAr436087 exhibited a replication disadvantage versus FL93-939 in this tissue in the absence of IFN-α/β responses. At 48 h p.i., a time at which replication of FL93-939 had surpassed BeAr436087 in the brains of control mice (Figure 1F and 3B), BeAr436087 amplified to nearly 100-fold higher titer than FL93-939 in the absence of IFN-α/β signaling suggesting a replication advantage.

Figure 3. Effect of type I IFN responses on BeAr436087 replication and dissemination.

Figure 3

Open in a new tab

BHK cell plaque titration of virus replication in various tissues and serum of 6-8 week-old IFN-deficient and control mice infected with 103 pfu of BeAr436087 in the hind footpad. Mice were perfused with PBS-1% DCS prior to tissue harvesting, homogenization and mechanical disruption: A) 129 Sv/Ev (black bars), IFNAR1−/− (hatched bars), IFNAGR−/− (white bars) and STAT1−/− (crosshatched bars) mice were infected with BeAr436087 and samples were collected at 24 h p.i. LN = popliteal lymph node. Joint = knee joint. B) Mice were infected with BeAr436087 (hatched bars) or FL93-939 (black bars) and samples were collected at 48 h p.i. Error bars represent standard deviations.

BeAr436087 is not defective in replication in vitro compared with FL93-939

Attenuation of BeAr436087 in mice does not appear to reflect a generalized replication defect in comparison with FL93-939. To examine the viruses’ replication efficiency in more defined conditions, we infected primary cells representative of cell-types at initial sites of virus replication (conventional myeloid dendritic cells [cDCs]), fueling secondary viremia (osteoblasts) and associated with terminal disease (cortical neurons) with BeAr436087 or FL93-939. Progeny virions were titered by standard plaque assay. In primary osteoblasts and neurons, the two viruses replicated similarly and at some times BeAr436087 even produced higher titers than FL93-939 (Fig. 4A and B). Neither virus replicated efficiently in bone marrow-derived cDCs (Fig. 4C and D), or macrophages (data not shown). This is consistent with our previously published results in which FL93-939 replicated poorly in cultured myeloid-lineage cells even in the absence of IFN-α/β signaling (Gardner et al., 2008). However, these results were somewhat surprising for BeAr436087 since this virus grows to relatively high titers in vivo in DLN and spleen. It is possible that either the cells supporting BeAr436087 growth in lymphoid tissues are of a different myeloid lineage from those derived by culture of bone marrow or that there is a non-myeloid cell-type in these lymphoid tissues that supports replication of BeAr436087 significantly better than FL93-939.

Figure 4. Comparative replication of BeAr436087 and FL93-939 in relevant cell-types in vitro.

Figure 4

Open in a new tab

Primary cultures of CD1-derived osteoblasts (A) and cortical neurons (B) were infected with BeAr436087 (open circle) or FL93-939 (closed circle) and progeny virions were titered. Replication of BeAr436087 (C) and FL93-939 (D) in bone marrow-derived cDCs generated from control 129 Sv/Ev (closed square) and IFNAR1−/− (open square) mice. All cells were infected at an moi of 0.1. LD = limit of detection for the plaque assay and error bars represent standard deviations.

BeAr436087 and FL93-939 are equally sensitive to murine IFN-α/β priming in vitro

As described above, osteoblasts are an important cell-type for fueling of the serum viremia in mice infected with EEEV (Vogel et al., 2005;Gardner et al., 2008;Wang et al., 2007). Furthermore, to provide relative scale of sensitivity, we included in these studies VEEV (strain ZPC738) and SINV (strain TR339) that have high and low resistance, respectively, to the IFN-α/β-induced antiviral state in murine cells (Ryman & Klimstra, 2008). Cells were pretreated with a range of IFN-α/β concentrations, infected with viruses and subsequently observed for protection from cytopathic effect (CPE) at 24 h p.i. (when 100% CPE was observed in unprimed controls for all virus infections). As expected, VEEV was the least sensitive and SINV was the most sensitive to IFN-α/β pretreatment (Fig. 5). BeAr436087 and FL93-939 were restricted at similar concentrations of IFN-α/β used for priming (50% CPE between 5 and10 international units [IU]/mL IFN-α/β) and were intermediately sensitive between SINV and VEEV. Similar results were observed in Swiss 3t3 fibroblast cells (data not shown). Together, these data suggest that induced IFN-α/β in vivo would be similarly protective against BeAr436087 and FL93-939, since reduced virulence was not correlated with increased sensitivity to the antiviral effects of murine IFN-α/β.

Figure 5. Relative sensitivity of BeAr436087 and FL93-939 to IFN-α/β priming of primary osteoblasts.

Figure 5

Open in a new tab

Triplicate wells of primary osteoblasts generated from CD1 mice were exposed to a range of IFN-α/β concentrations for 24 h prior to infection with SINV (open square), FL93-939 (open circle), FL93-939 (closed circle) or VEEV (closed square) at an moi of 1 for each virus. Data are presented as a percentage of cells surviving at 24 h p.i. versus the dose (IU/mL) of IFN-α/β used to prime.

Systemic IFN-α/β is induced in BeAr436087-infected, but not FL93-939-infected mice

We previously demonstrated that the extensive replication of VEEV in DLN and spleen was associated with dramatically higher levels of IFN-α/β in the serum compared with FL93-939 (Gardner et al., 2008). To examine this parameter comparing FL93-939 and BeAr436087, biological IFN-α/β activity assays were conducted on serum samples at various times p.i. As previously reported (Gardner et al., 2008), FL93-939 virus only produced detectable of IFN-α/β in serum at 48 h p.i. and this was barely above the limit of detection of the assay. In contrast, BeAr436087 induced high levels of IFN-α/β at 24 h p.i., that peaked at 48 h p.i at approximately 103.5 IU/mL, over 150-fold higher than in FL93-939 infected mice (Fig. 6). These data reveal the possibility that systemic IFN-α/β induced in BeAr436087-infected mice is contributing to their survival and leading to clearance of the virus from CNS tissues.

Figure 6. Induction of systemic IFN-α/β in BeAr436087-infected versus FL93-939-infected mice.

Figure 6

Open in a new tab

Results of biological assay for detection of IFN-α/β in the serum of 6 week-old CD1 mice of infected with 103 pfu of FL93-939 (black bars) or BeAr436087 (white bars) in the hind footpad. LD = limit of detection of the IFN-α/β assay and error bars represent standard deviations.

We also examined systemic IFN-α/β in 2-week-old mice, an age in which both viruses cause 100% mortality with similar AST. Only 48 h p.i. was examined, because this was the time at which peak levels of IFN-α/β were induced in the 6 week-old mice. IFN-α/β levels similar the sera of the 2-week-old and 6-week-old mice infected with BeAr436087; however, the 2-week-old mice infected with FL93-939 produced much higher levels of serum IFN-α/β than 6 week-old mice which were comparable to mice infected with BeAr436087 (data not shown). These data suggest that tropism of one or both viruses might be altered in 2 week-old mice and that IFN-α/β responses are not fully potentiated for protection of these mice.

BeAr436087 infection leads to STAT1 phosphorylation in the CNS, but FL93-939 does not

To determine if the systemic IFN-α/β produced in the BeAr436087-infected mice was capable of inducing an antiviral state in the CNS, we infected CD1 mice with either BeAr436087 or FL93-939 and, at 24 and 48 h p.i., harvested the brain for western blot analysis of STAT1 phosphorylation status (Fig. 7). STAT1 is an IFN-α/β inducible transcription factor, activated through phosphorylation by JAK1 and Tyk2 kinases after IFN receptor signaling. Phosphorylated, dimerized STAT1 molecules translocate to the nucleus to directly transactivate IFN-inducible gene promoters (reviewed by van Boxel-Dezaire et al., 2006). A distinct increase in phosphorylation of STAT1 was observed at both times in the brains of BeAr436087 mice while the increase in the brains of FL93-939-infected counterparts was minimal. Furthermore, BeAr436087 infection caused an increase in total STAT1 consistent with the effects of IFN-α/β upon the abundance of this protein (van Boxel-Dezaire et al., 2006). Taken together, these data are consistent with the idea that BeAr436087 infection induces greater systemic IFN-α/β production than FL93-939 resulting in a more effective antiviral state in the CNS, thereby attenuating neurovirulence of the virus.

Figure 7. STAT1 phosphorylation in brains of BeAr436087-infected and FL93-939-infected mice.

Figure 7

Open in a new tab

Western blot detection of the abundance of STAT1 and its phosphorylation state at 24 or 48 h p.i. in lysates from whole brains of 6 week-old CD-1 mice infected subcutaneously in the hind footpad with 103 pfu of BeAr436087 (BeAr) or FL93-939 (FL93). (+) = positive control in which 100,000 IU of IFN-α/β were injected directly into the brains of CD1 mice and whole brain lysates were harvested 30 min later.

Priming of mice with BeAr436087 protects against FL93-939 disease

Weaver and colleagues found that avirulent BeAr436087 infection could elicit an adaptive immune response in mice that protected against FL93-939 challenge several weeks later (Wang et al., 2007), presumably by eliciting neutralizing antibody. Since the viruses are similarly sensitive to IFN-α/β priming, but BeAr436087 induces systemic IFN-α/β production while FL93-939 does not, we hypothesized that inoculation of mice with BeAr436087 would induce an innate immune response that could protect against FL93-939 challenge. CD1 mice were inoculated subcutaneously in the rear hind-limb footpad with BeAr436087 and challenged with FL93-939 either simultaneously, 8 or 24 h later in the ipsilateral or contralateral footpad (Fig. 8). Infection of mice with BeAr436087 at either 24 or 8 h prior to FL93-939 completely protected against disease and fatality following FL93-939 challenge regardless of whether the viruses were inoculated in the ipsilateral or contralateral footpad. Partial protection was observed when the viruses were inoculated in the same or opposite footpads simultaneously and even when BeAr436087 was administered 8 h after FL93-939. Higher protection for BeAr436087 given 8 h after FL93 versus simultaneous administration appeared to reflect experiment-to-experiment variation as the percentage of survivors was variable, although, the protective phenotype was consistently reproducible. A small amount of protection even developed when BeAr436087 was administered 24 h after FL93-939, but if given in the ipsilateral footpad. There is a possibility that viruses given in the same footpad could compete for cell targets; however, the fact that contralateral footpad administration was protective suggests that BeAr436087 infection activated the innate immune system to establish an antiviral state that was attenuating to FL93-939 disease, consistent with the relative sensitivity of this virus to IFN-α/β in vitro.

Figure 8. Rapid BeAr436087-mediated protection against FL93-939-induced mortality in CD1 mice.

Figure 8

Open in a new tab

6 week-old mice were untreated or inoculated with 103 pfu of BeAr436087 (“treatment” in the figure) either 24, 8 or 0 h prior to, or 8 or 24 h after, inoculation in the left hind footpad with 103 pfu of FL93-939. BeAr436087 inoculations were either the same (ipsilateral) or opposite (contralateral) footpad as FL93-939. Data are presented as the percentage of mice surviving the FL93-939 infection at 21 d p.i.

Discussion

It is a commonly held belief that pathogenic viruses have evolved mechanisms to subvert host antiviral defenses resulting in disease (reviewed in Brasier et al., 2009). Many alphaviruses initially target and replicate in myeloid-lineage cells at the site of inoculation and appear to usurp their migratory capabilities to reach the DLN, an early site of virus amplification, and seed the initial serum viremia (reviewed by (Ryman & Klimstra, 2008). The ability of a closely-related encephalitic alphavirus, VEEV, to replicate in lymphoid tissues is positively associated with virulence in the mouse model and reduced lymphotropism resulting from mutation is strongly associated with attenuation (Grieder et al., 1995;MacDonald & Johnston, 2000). During VEEV infection high systemic levels of cytokines including IFN-α/β are elicited and yet the virus manages to replicate efficiently and cause disease (Charles et al., 2001;Ryman & Klimstra, 2008;Gardner et al., 2008). On the other hand, FL93-939, representative of North American EEEV isolates, appears to circumvent the IFN-α/β-mediated antiviral response by avoiding infection of lymphoid tissues in general, and myeloid cells in particular, and limiting systemic IFN-α/β production (Gardner et al., 2008). As a direct extension of this observation, we now propose that the capacity of avirulent EEEV strains such as BeAr436087 to replicate efficiently within lymphoid tissues correlates with IFN-α/β induction and attenuation of disease.

We found that the avirulent SA-EEEV BeAr436087 strain initially replicated to higher titers in all infected tissues examined in 6-week-old mice and neuroinvaded more rapidly than the virulent FL93-939 strain, indicating no basic defect in replication in vivo. However, as infection progressed, BeAr436087 replication was cleared from the brain while that of FL93-939 increased to the point the mice succumbed to infection. Uniquely, BeAr436087 replicated much more efficiently than FL93-939 in lymphoid tissues throughout infection. These replication differences were associated with induction of high levels of serum IFN-α/β by BeAr436087 whereas FL93-939 rarely induced detectable IFN-α/β above the limit of detection of the assay, and systemic IFN-α/β induction after BeAr436087 infection was associated with STAT1 phosphorylation in the brain. BeAr436087 did not exhibit a defect in replication in osteoblasts or neurons in vitro and was similarly sensitive as FL93-939 after IFN-α/β priming of primary osteoblasts or a fibroblast cell line. While absence of T and B cell activities did not render BeAr436087 virulent, when the capacity to respond to IFN-α/β was removed as in the IFNAR1−/− mice, BeAr436087 caused more rapid mortality than FL93-939 and replicated to higher titers in the brain by 48 h p.i. Finally, administration of BeAr436087 shortly before, simultaneous with or shortly after FL93-939 inoculation, resulted in partial or complete protection from fatal FL93-939 disease. These data provide strong evidence that BeAr436087 attenuation is: i) determined by the effectiveness of the type I IFN response; ii) associated with enhanced control and clearance of replication from the brain versus FL93-939; iii) associated with IFN-α/β induction that results in STAT1 activation in the brain. Moreover, BeAr436087 infection can create an environment in vivo that attenuates FL93-939 disease. We propose that BeAr436087 attenuation in comparison with FL93-939 can, in large part, be attributed to the virus’ capacity to replicate efficiently in lymphoid tissue leading to systemic IFN-α/β production, activation of an antiviral state in the CNS and protection from lethal encephalitic disease.

Another contributing factor in the attenuation of BeAr436087 might be differential sensitivity versus FL93-939 to the antiviral state in IFN-α/β-primed cells. Previous work suggested that NA and SA strains of EEEV differed in their sensitivity to human IFN-α and IFN-β (Aguilar et al., 2008a;Aguilar et al., 2005). In contrast with these studies, we did not detect significant differences in sensitivity, which might indicate species- or cell-type specificity of responses to IFN-α/β. However, we have utilized primary cells derived from tissues infected by these viruses in mice and found little difference in IFN-α/β sensitivity between the two. Clearly, more detailed IFN-α/β sensitivity analyses will be required to determine reasons for the different results. Nonetheless, the fact that BeAr436087 induces high systemic levels of IFN-α/β that activates STAT1 in the CNS, while FL93-939 does not, implies an impact upon virulence. The idea that differences in sensitivity to the antiviral state between FL93-939 and BeAr436087 may not be correlated with virulence is further supported by the fact that administration to mice of the IFN-inducing double stranded RNA mimic, poly (I:C), attenuated FL93-939 disease (Aguilar et al., 2005), similar to our results with BeAr436087 co-infection. We have considered that the difference in brain STAT1 phosphorylation between mice infected with the two viruses is reflective of antagonism of this pathway in FL93-939-infected cells. However, at 48 h p.i. proteins from virus-infected cells represent a small fraction of total material in the lysates and, furthermore, in separate experiments, we determined that FL93-939 did not significantly block STAT1 phosphorylation in cultured mouse cortical neurons (C.L. Gardner, J. Yin, W.B. Klimstra and K.D. Ryman, unpublished observations).

It has been proposed that New World alphaviruses prevent production of IFN-α/β from infected cells by causing generalized arrest of host transcription and translation (Aguilar et al., 2008b;Garmashova et al., 2007). However, the induction of high levels of serum IFN-α/β after VEEV infection contradict this assertion in vivo; although at least some serum IFN-α/β in VEEV-infected mice may originate from uninfected, neighboring cells (Konopka et al., 2007). Recently it was demonstrated that a deletion mutation in the EEEV capsid that impaired transcriptional shutoff also greatly attenuated the virus in vivo (Aguilar et al., 2007b;Aguilar et al., 2008b;Aguilar et al., 2008b). Unfortunately, systemic IFN-α/β responses were not measured inthese studies so the relationship of this attenuation mechanism to IFN responses in vivo is unclear. It would be of interest to compare the host cell shutoff-inducing capacity of BeAr436087 to that of FL93-939.

The underlying mechanism by which BeAr436087 replicates efficiently in DLN and spleen is not known. Our results titering virus produced by cultured cells and mouse tissues suggest that, in the absence of induced IFN-α/β, BeAr436087 may have a slight growth advantage over FL93-939 in multiple cell/tissue types. We demonstrated previously that FL93-939, in contrast with SINV or VEEV, was unable to replicate within myeloid-lineage cells in vitro or in vivo due to the inability of the incoming genome to be efficiently translated and that this was attributable to effects of sequences/structures in the 5′ and/or 3′ non-translated regions and/or translation control elements in nsP1 (Gardner et al., 2008). We had expected that the higher titers of BeAr436087 in lymphoid tissues would reflect increased capacity of this virus to replicate in myeloid cells. However, BeAr436087, like FL93-939, was essentially replication-defective in bone marrow-derived cDCs and macrophages in vitro. It remains to be determined whether or not the efficient replication of BeAr436087 that is linked to IFN-α/β induction in vivo is due to tissue-specific interaction with specific cell subsets or higher numbers of infected cells in all tissues. The fact that STAT1 was not highly phosphorylated in the CNS of FL93-939-infected mice, even though this virus replicated to higher levels than BeAr436087, is suggestive that a subset of infected tissues are involved in IFN-α/β induction by BeAr436087. However, we must also consider that antagonism of IFN induction in neurons could explain the differential STAT1 phosphoryation. We feel that this is unlikely as neither of the two viruses elicit IFN-α/β from cultured cortical neurons due to rapid shutoff of host macromolecular synthesis (J. Yin, C.L. Gardner, C.W. Burke, K.D. Ryman and W.B. Klimstra, unpublished observations). Ultimately the answer to this question will come from detailed examination of the targets of FL93-939 and BeAr436087 infection in mouse lymphoid tissues.

It is possible that these results may be broadly applicable to the mechanisms of attenuation of SA-EEEV versus NA-EEEV strains in humans. Aguilar et al. (2008a) came to the conclusion that activities encoded by both structural and non-structural proteins were involved in the virulence of the NA-EEEV viruses and that increased sensitivity to human IFN-α/β of SA-EEEV strains may be one mechanism contributing to their attenuation. We now add lymphoid tissue replication resulting in high levels of systemic IFN-α/β induction by SA-EEEV strains as another possible mechanism. The relative contribution of IFN-α/β sensitivity versus induction to differences in virulence is currently not known. Further testing of our hypothesis can be achieved by examination of additional NA-EEEV and SA-EEEV isolates.

Finally, we believe our findings have applicability to strategies for both designing effective antiviral therapies and for design of a live-attenuated vaccine against EEEV. Our data combined with that of Aguilar et al. (2005) suggest that IFN-α/β administration or artificial induction of IFN-α/β responses offers promise as a therapeutic strategy during acute phase EEEV disease. The fact that BeAr436087 replicates very efficiently in lymphoid tissues suggests that it would also make an effective alphavirus vaccine or delivery system for heterologous antigens, supported by the findings of Weaver and coworkers (Wang et al., 2007;Wang et al., 2008). Furthermore, the IFN-α/β induction data indicate that this virus would elicit robust innate immune responses that are critical to adjuvant effects of alphavirus vectors (Thompson et al., 2006;Hidmark et al., 2006). Most compelling, however, was the observation that in addition to use as a prophylactic vaccine, BeAr436087 immunization could be used in the event of a BWT attack as a post-exposure therapeutic agent.

Materials & Methods

Cell lines

Baby hamster kidney (BHK)-21 and L929 murine fibrosarcoma cells were maintained in alpha minimum essential medium (αMEM) and supplemented with 10% donor calf serum (DCS) and tryptose phosphate broth, 0.25 mg/ml L-glutamine, 100 U/mL penicillin, and 0.05 mg/mL streptomycin. Swiss 3t3 murine fibroblast cells were maintained in αMEM and supplemented with 10% DCS, 0.25 mg/mL L-glutamine, 100 U/mL penicillin, and 0.05 mg/mL streptomycin.

Mice

Outbred CD1 (Charles River Laboratories), 129 Sv/Ev (Taconic Laboratories) and C57BL/6 (Jackson Laboratories) mice were purchased. Mice deficient in IFN-α/β receptor (IFNAR1−/−), IFN-γ (IFNGR−/−), IFN-α/β/γ (IFNAGR −/−), STAT1 (STAT1−/−) or recombinase activating gene 1 (RAG-1; RAG−/−) mice were bred in-house. All mice were housed in the Animal Resource Center at LSUHSC-S under specific pathogen-free conditions. CD1 mice were used at various ages as specified in figure legends. Other mouse strains were used at 6-8 weeks of age. All experiments were conducted in accordance with the guidelines of the LSUHSC-S Institutional Animal Care and Use Committee in the Biosafety Level (BSL)-3 facilities.

Primary cell cultures

cDC cultures were derived from bone marrow and cultured, as previously described (Gardner et al., 2008). Briefly, bone marrow was aspirated from femurs and tibias of mice and plated for 1 h (37°C; 5% CO2) to allow macrophage contaminates to adhere, after which the non-adherent cDC progenitors were removed and cultured. The cDCs were cultured in RPMI 1640 medium, supplemented with 10% fetal bovine serum (FBS), 1% nonessential amino acids, 1 mM sodium pyruvate, 5 mM HEPES buffer, 50 μM β-mercaptoethanol, 10 ng/mL granulocyte—macrophage colony stimulating factor, and 10 ng/mL IL-4 (Peprotech, Rocky Hill, NJ). After 7 d in culture, non-adherent cDCs were pelleted, counted and resuspended in supplemented medium for infection.

Primary murine osteoblasts were generated from calvaria of 3-5-day-old sucking mice as previously described (Gardner et al., 2008). Briefly, the calvaria were removed and gently scraped until white before digesting three times in αMEM containing 0.05% trypsin-EDTA and 10 mg/mL collagenase P. The cells were cultured for 5 d in αMEM with 15% FBS, 100 U/mL penicillin and 0.05 mg/mL streptomycin to allow outgrowth of osteoblasts from calvaria. Osteoblasts were then trypsinized, strained to remove any remaining bone fragments and seeded into 100 mm culture dishes. Once confluent, the osteoblasts were trypsinized, counted and seeded appropriately for experiments.

Primary neurons were generated from embryo cortices at 13-16 d gestation, essentially as described (Samuel et al., 2006). Briefly, the cortices were digested in Hanks balanced salt solution containing papain (1 mg/mL), DNase I (50 μg/mL), L-cysteine (0.2 mg/mL), calcium chloride (1.5 mM), and EDTA (0.5 mM). Poly-D-lysine (PLD)-coated plates were prepared by incubating 30 μg/mL PLD in the 24-well plate for 2-6 h at room temperature. Cells were plated in PLD-coated 24-well plates at 2 × 105 cells per well and cultured in neurobasal medium with 2% B27 additive (Invitrogen), 0.5 mM glutamine and 20 μg/mL gentamicin. Cells were used after 6-7 d of culture.

Viruses

cDNA clones for NA-EEEV strain FL93-939 (Aguilar et al., 2008a), SA-EEEV strain BeAr436087 (Aguilar et al., 2008a) VEEV strain ZPC738 (Anishchenko et al., 2004), and SINV strain TR339 (Klimstra et al., 1998) have been described previously. FL93-939, BeAr436087, and ZPC738 cDNA clones were generously provided by Dr. Scott Weaver (University of Texas Medical Branch, Galveston, TX). Briefly, capped infectious viral RNAs were generated by in vitro transcription (mMessage mMachine, Ambion) from linearized cDNA plasmid templates. The RNA was then electroporated into BHK-21 cells and virus particles were harvested from the supernatant ~24 h after electroporation when CPE was evident. Cell debris was clarified by centrifugation and single-use virus aliquots were stored at −80°C. Virus stocks were titered by standard BHK-21 cell plaque assay and titers were expressed as pfu/mL.

Mortality and pathogenesis studies

Virus inocula of FL93-939 or BeAr436087 containing 103 pfu in a 10 μL volume (1×105 pfu/mL) were administered subcutaneously in the each rear footpad using a 27-gauge needle and 100 μL gas-tight Hamilton syringe. Mock-infected mice received 10 μL PBS-1% DCS by the same route. Virus-infected and mock-infected mice were observed at 24 h intervals and weighed where appropriate. ASTs and percent mortality were calculated. Mouse mortality curves were compared using a Mantel-Cox log rank test with GraphPad Prism software. Titer and weight differences were compared with a one-tailed Student’s t test for samples of equal variance using GraphPad Prism software.

At predetermined intervals p.i., groups of three mice per treatment were sacrificed under isoflurane-anesthesia and blood was collected by cardiac puncture. Serum was separated from whole blood using microtainer tubes (Becton-Dickinson) and stored in single-use aliquots at −80°C. Mice were then perfused with PBS-1% DCS for 10 min at 7 mL/min to flush blood-borne virus. Tissues were harvested into preweighed eppendorf tubes and homogenized in 9 volumes of PBS-1% DCS. Supernatants were assayed for virus by standard plaque assay, expressed as pfu/DLN, mL or g.

Virus growth curves

Adherent cells were infected in 24-well plates at a moi of 1 pfu/cell or 0.1 pfu/cell, as indicated in the figure legends. Virus was incubated for 1 h at 37°C, then cells were washed three times with PBS-1% DCS diluent, appropriate medium was replaced and cells were incubated (37°C; 5% CO2). Non-adherent cDCs were infected in suspension in V-bottom 96-well plates at the desired moi (37°C; 1 h) and washed three times in PBS-1% DCS by pelleting and resuspension. Cells were resuspended in cDC culture media and seeded into 24-well plates. For virus growth curves, supernatants were harvested at 0 h p.i. and subsequent time-points for titration by standard plaque assay on BHK-21 cells.

IFN-α/β analyses

Serum IFN-α/β was measured by standard biological assay on L929 cells as described previously (Trgovcich et al., 1996), using recombinant IFN-α/β as a standard and encephalomyocarditis virus (EMCV) as the indicator of protection. Concentration of IFN-α/β was determined by the dilution of IFN-α/β needed to cause 50% protection from EMCV CPE. Levels of IFN-α/β were expressed as IU/mL.

IFN-α/β sensitivity assay

Primary murine osteoblasts or Swiss 3t3 cells in a 24-well plate were treated with different concentrations of recombinant IFN-α/β fo r 24 h, infected at an moi of 1 for 1 h (37°C; 5% CO2), then washed three times in PBS-1% DCS and appropriate media was added. The cells were scored for percentage of CPE by 24 h p.i..

STAT1 in tissue lysates

CD1 mice were infected with 103 pfu subcutaneously in both rear footpads with either FL93-939 or BeAr436087 and 24 and 48 h p.i. brains were harvested and homogenized in 800 μL of lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, 10% glycerol, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF, 2 μg/mL aprotinin, 5 μg/mL leupeptin, 5 μg/mL pepstatin, and phosphatase inhibitor cocktails I and II (Sigma). Homogenized tissues were clarified by centrifugation (20,000 × g, 10 min, 4°C) and protein concentration was determined using BioRad protein assay. One hundred μg of protein was separated on an 8% SDS-PAGE gel and transferred onto a nitrocellulose membrane (BioRad). Membranes were blocked for 1 h in 5% skim milk containing TBS/0.1% Tween 20 (TBST). The membranes were incubated with primary antibody to p-Tyr701 STAT1 (Cell Signaling), total STAT1 (M-22; Santa Cruz), or β-actin (Chemicon), in TBST with 3% BSA at 4°C. Membranes were washed with TBST and incubated with HRP-conjugated secondary antibody (Zymed) in TBST with 2% skimmed milk. The ECL-Plus chemiluminescence system (Amersham) was used to develop the membranes.

Acknowledgements

The authors wish to thank Michael Farmer, Danielle Gonzalez and DeAquanita McKinney for excellent technical assistance. We are particularly indebted to Dr. Scott Weaver (University of Texas Medical Branch, Galveston, TX) for providing us with cDNA clones from which viruses were generated.

This work was supported by National Institutes of Health grants R21 AI069158 (KDR) and R21AI072350 (WBK) and grants from NIAID/NIH through the Western Regional Center of Excellence for Biodefense and Emerging Infectious Diseases Research (WRCE) U54 AI057156 (Career Development award to KDR & major project subcontract to WBK).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Aguilar PV, Adams AP, Wang E, Kang W, Carrara AS, Anishchenko M, Frolov I, Weaver SC. Structural and Nonstructural Protein Genome Regions of Eastern Equine Encephalitis Virus Are Determinants of Interferon Sensitivity and Murine Virulence. J Virol. 2008a doi: 10.1128/JVI.02514-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aguilar PV, Leung LW, Wang E, Weaver SC, Basler CF. A five amino acid deletion of the Eastern equine encephalitis virus capsid protein attenuates replication in mammalian systems but not in mosquito cells. J Virol. 2008b doi: 10.1128/JVI.01283-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aguilar PV, Paessler S, Carrara AS, Baron S, Poast J, Wang E, Moncayo AC, Anishchenko M, Watts D, Tesh RB, Weaver SC. Variation in interferon sensitivity and induction among strains of eastern equine encephalitis virus. J Virol. 2005;79:11300–11310. doi: 10.1128/JVI.79.17.11300-11310.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Aguilar PV, Robich RM, Turell MJ, O’Guinn ML, Klein TA, Huaman A, Guevara C, Rios Z, Tesh RB, Watts DM, Olson J, Weaver SC. Endemic eastern equine encephalitis in the Amazon region of Peru. Am J Trop Med Hyg. 2007a;76:293–298. [PubMed] [Google Scholar]
  5. Aguilar PV, Weaver SC, Basler CF. Capsid protein of eastern equine encephalitis virus inhibits host cell gene expression. J Virol. 2007b;81:3866–3876. doi: 10.1128/JVI.02075-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Alice FJ. Infeccao humana pelo virus “leste” da encefalite equina. Bol Inst Biol Bahia (Brazil) 1956;3:3–9. [Google Scholar]
  7. Anishchenko M, Paessler S, Greene IP, Aguilar PV, Carrara AS, Weaver SC. Generation and characterization of closely related epizootic and enzootic infectious cDNA clones for studying interferon sensitivity and emergence mechanisms of Venezuelan equine encephalitis virus. J Virol. 2004;78:1–8. doi: 10.1128/JVI.78.1.1-8.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Armstrong PM, Andreadis TG, Anderson JF, Stull JW, Mores CN. Tracking eastern equine encephalitis virus perpetuation in the northeastern United States by phylogenetic analysis. Am J Trop Med Hyg. 2008;79:291–296. [PubMed] [Google Scholar]
  9. Brasier AR, García-Sastre A, Lemon SM. Cellular signaling and innate immune responses to RNA virus infections. ASM Press; Washington, D.C.: 2009. [Google Scholar]
  10. Causey OR, Theiler M. Virus antibody survey on sera of residents of the Amazon Valley in Brazil. Am J Trop Med Hyg. 1958;7:36–41. doi: 10.4269/ajtmh.1958.7.36. [DOI] [PubMed] [Google Scholar]
  11. Charles PC, Trgovcich J, Davis NL, Johnston RE. Immunopathogenesis and immune modulation of Venezuelan equine encephalitis virus-induced disease in the mouse. Virology. 2001;284:190–202. doi: 10.1006/viro.2001.0878. [DOI] [PubMed] [Google Scholar]
  12. Corniou B, Ardoin P, Bartholomew C, Ince W, Massiah V. First isolation of a South American strain of Eastern Equine virus from a case of encephalitis in Trinidad. Trop Geogr Med. 1972;24:162–167. [PubMed] [Google Scholar]
  13. Gardner CL, Burke CW, Tesfay MZ, Glass PJ, Klimstra WB, Ryman KD. Eastern and Venezuelan equine encephalitis viruses differ in their ability to infect dendritic cells and macrophages: impact of altered cell tropism on pathogenesis. J Virol. 2008;82:10634–10646. doi: 10.1128/JVI.01323-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Garmashova N, Gorchakov R, Volkova E, Paessler S, Frolova E, Frolov I. The Old World and New World alphaviruses use different virus-specific proteins for induction of transcriptional shutoff. J Virol. 2007;81:2472–2484. doi: 10.1128/JVI.02073-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grieder FB, Davis BK, Zhou XD, Chen SJ, Finkelman FD, Gause WC. Kinetics of cytokine expression and regulation of host protection following infection with molecularly cloned Venezuelan equine encephalitis virus. Virology. 1997;233:302–312. doi: 10.1006/viro.1997.8617. [DOI] [PubMed] [Google Scholar]
  16. Grieder FB, Davis NL, Aronson JF, Charles PC, Sellon DC, Suzuki K, Johnston RE. Specific restrictions in the progression of Venezuelan equine encephalitis virus-induced disease resulting from single amino acid changes in the glycoproteins. Virology. 1995;206:994–1006. doi: 10.1006/viro.1995.1022. [DOI] [PubMed] [Google Scholar]
  17. Hidmark AS, Nordstrom EK, Dosenovic P, Forsell MN, Liljestrom P, Hedestam G. B. Karlsson. Humoral responses against coimmunized protein antigen but not against alphavirus-encoded antigens require alpha/beta interferon signaling. J Virol. 2006;80:7100–7110. doi: 10.1128/JVI.02579-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Klimstra WB, Ryman KD, Johnston RE. Adaptation of Sindbis virus to BHK cells selects for use of heparan sulfate as an attachment receptor. J Virol. 1998;72:7357–7366. doi: 10.1128/jvi.72.9.7357-7366.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kondig JP, Turell MJ, Lee JS, O’Guinn ML, Wasieloski LP., Jr. Genetic analysis of South American eastern equine encephalomyelitis viruses isolated from mosquitoes collected in the Amazon Basin region of Peru. Am J Trop Med Hyg. 2007;76:408–416. [PubMed] [Google Scholar]
  20. Konopka JL, Penalva LO, Thompson JM, White LJ, Beard CW, Keene JD, Johnston RE. A two-phase innate host response to alphavirus infection identified by mRNP-tagging in vivo. PLoS Pathog. 2007;3:e199. doi: 10.1371/journal.ppat.0030199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. MacDonald GH, Johnston RE. Role of dendritic cell targeting in Venezuelan equine encephalitis virus pathogenesis. J Virol. 2000;74:914–922. doi: 10.1128/jvi.74.2.914-922.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rua-Domenech R, Reid SW, Gonzalez-Zariquiey AE, Wood JL, Gettinby G. Modelling the spread of a viral infection in equine populations managed in Thoroughbred racehorse training yards. Prev Vet Med. 1999;47:61–77. doi: 10.1016/s0167-5877(00)00161-6. [DOI] [PubMed] [Google Scholar]
  23. Ryman KD, Gardner CL, Meier KC, Biron CA, Johnston RE, Klimstra WB. Early restriction of alphavirus replication and dissemination contributes to age-dependent attenuation of systemic hyperinflammatory disease. J Gen Virol. 2007;88:518–529. doi: 10.1099/vir.0.82359-0. [DOI] [PubMed] [Google Scholar]
  24. Ryman KD, Klimstra WB. Host responses to alphavirus infection. Immunol Rev. 2008;225:27–45. doi: 10.1111/j.1600-065X.2008.00670.x. [DOI] [PubMed] [Google Scholar]
  25. Ryman KD, Klimstra WB, Nguyen KB, Biron CA, Johnston RE. Alpha/beta interferon protects adult mice from fatal Sindbis virus infection and is an important determinant of cell and tissue tropism. J Virol. 2000;74:3366–3378. doi: 10.1128/jvi.74.7.3366-3378.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Samuel MA, Whitby K, Keller BC, Marri A, Barchet W, Williams BR, Silverman RH, Gale M, Jr., Diamond MS. PKR and RNase L contribute to protection against lethal West Nile Virus infection by controlling early viral spread in the periphery and replication in neurons. J Virol. 2006;80:7009–7019. doi: 10.1128/JVI.00489-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Smith JF, Davis M, Hart MK, Ludwig GV, McClain DJ, Parker MD, Pratt WD. Viral Encephalitides. In: Sidell FR, Takafuji ET, Franz DR, editors. Medical Aspects of Chemical and Biological Warfare. Office of the Surgeaon General, Department of the Army; Washington, USA: 1997. p. 561. [Google Scholar]
  28. Thompson JM, Whitmore AC, Konopka JL, Collier ML, Richmond EM, Davis NL, Staats HF, Johnston RE. Mucosal and systemic adjuvant activity of alphavirus replicon particles. Proc Natl Acad Sci U S A. 2006;103:3722–3727. doi: 10.1073/pnas.0600287103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Trgovcich J, Aronson JF, Johnston RE. Fatal Sindbis virus infection of neonatal mice in the absence of encephalitis. Virology. 1996;224:73–83. doi: 10.1006/viro.1996.0508. [DOI] [PubMed] [Google Scholar]
  30. van Boxel-Dezaire AH, Rani MR, Stark GR. Complex modulation of cell type-specific signaling in response to type I interferons. Immunity. 2006;25:361–372. doi: 10.1016/j.immuni.2006.08.014. [DOI] [PubMed] [Google Scholar]
  31. Vogel P, Kell WM, Fritz DL, Parker MD, Schoepp RJ. Early Events in the Pathogenesis of Eastern Equine Encephalitis Virus in Mice. Am J Pathol. 2005;166:159–171. doi: 10.1016/S0002-9440(10)62241-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Wang E, Petrakova O, Adams AP, Aguilar PV, Kang W, Paessler S, Volk SM, Frolov I, Weaver SC. Chimeric Sindbis/eastern equine encephalitis vaccine candidates are highly attenuated and immunogenic in mice. Vaccine. 2007;25:7573–7581. doi: 10.1016/j.vaccine.2007.07.061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wang E, Volkova E, Adams AP, Forrester N, Xiao SY, Frolov I, Weaver SC. Chimeric alphavirus vaccine candidates for chikungunya. Vaccine. 2008;26:5030–5039. doi: 10.1016/j.vaccine.2008.07.054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Weaver SC, Powers AM, Brault AC, Barrett AD. Molecular epidemiological studies of veterinary arboviral encephalitides. Vet J. 1999;157:123–138. doi: 10.1053/tvjl.1998.0289. [DOI] [PubMed] [Google Scholar]
  35. Young DS, Kramer LD, Maffei JG, Dusek RJ, Backenson PB, Mores CN, Bernard KA, Ebel GD. Molecular Epidemiology of Eastern Equine Encephalitis Virus, New York. Emerg Infect Dis. 2008;14:454–460. doi: 10.3201/eid1403.070816. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES