Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Sep 1.
Published in final edited form as: Virology. 2010 May 31;404(2):139–147. doi: 10.1016/j.virol.2010.04.012

La Crosse Virus (LACV) Gc Fusion Peptide Mutants have Impaired Growth and Fusion Phenotypes, but Remain Neurotoxic

Samantha S Soldan a,*, Bradley S Hollidge a,c, Valentina Wagner d, Friedemann Weber d, Francisco González-Scarano a,b
PMCID: PMC2919166  NIHMSID: NIHMS199110  PMID: 20553924

Abstract

La Crosse virus is a leading cause of pediatric encephalitis in the Midwestern United States and an emerging pathogen in the American South. The LACV glycoprotein Gc plays a critical role in entry as the virus attachment protein. A 22 amino acid hydrophobic region within Gc (1066-1087) was recently identified as the LACV fusion peptide. To further define the role of Gc (1066-1087) in virus entry, fusion, and neuropathogenesis, a panel of recombinant LACV (rLACV) fusion peptide mutant viruses was generated. Replication of mutant rLACVs was significantly reduced. In addition, the fusion peptide mutants demonstrated decreased fusion phenotypes relative to LACV-WT. Interestingly, these viruses maintained their ability to cause neuronal loss in culture, suggesting that the fusion peptide of LACV Gc is a determinant of properties associated with neuroinvasion (growth to high titer in muscle cells and a robust fusion phenotype), but not necessarily of neurovirulence.

INTRODUCTION

La Crosse virus (LACV) is a member of the Bunyaviridae (genus Orthobunyavirus) and a common cause of pediatric encephalitis and aseptic meningitis in the Midwestern United States, the principal location of its major mosquito vector, Ochlerotatus (formerly Aedes) triseriatus (Thompson, Kalfayan, and Anslow, 1965). About 300,000 systemic LACV infections are estimated to occur annually, though fewer than 1.5% of these are clinically apparent (Kalfayan, 1983). Approximately half of the children with LACV encephalitis have seizures during the acute illness, and about 10% will develop epilepsy. A subset of those with encephalitis (approximately 2%) will have persistent paresis, learning disabilities, or cognitive defects. It is possible that the incidence of LACV encephalitis may be under-reported because of clinical similarities with herpes simplex virus encephalitis (McJunkin, Khan, and Tsai, 1998). Importantly, because there is a well developed mouse model for LACV encephalitis that mimics many features of the human disease, LACV infection has been used to study the neuroinvasion and neurovirulence of arthropod-borne viruses (Janssen et al., 1986).

The LACV genome consists of three single-stranded RNA segments of negative polarity, designated by size as large (L), medium (M), and small (S). These segments encode the viral polymerase, two viral glycoproteins (Gc and Gn) and the nucleocapsid protein, respectively. In addition, the M and S segments each encode nonstructural proteins, NSm and NSs. The three negative-stranded RNA segments of the LACV genome (L, M, and S) have defined roles in virus pathogenesis. Previous studies have mapped the neuroinvasive phenotype of LACV to its M RNA segment (Gonzalez-Scarano et al., 1982); (Janssen, Gonzalez-Scarano, and Nathanson, 1984; Janssen et al., 1986); (Griot et al., 1994; Griot et al., 1993), whereas at least in one instance, neurovirulence mapped to the L segment (Endres et al., 1989).

The two M-segment glycoproteins, Gn and Gc, are cleaved co-translationally from a precursor polyprotein and associate as a heteromultimer (Bupp, Stillmock, and Gonzalez-Scarano, 1996); (Eshita et al., 1985); (Fazakerley et al., 1988). This Gc/Gn heteromultimer is targeted to the Golgi apparatus, the site of viral assembly and budding, by a Golgi localization signal that has been mapped to the carboxy terminus of Gn (Lappin et al., 1994); (Ruusala et al., 1992). Gc is the exclusive target of neutralizing antibodies against LACV and has been demonstrated to play a critical role in entry as the virus attachment protein (Pekosz et al., 1995b); it is the larger of the two glycoproteins in the genus Orthobunyavirus. In addition, Gc and recombinant soluble forms that lack a transmembrane domain form homomultimers and undergo a pH dependent conformational change that is associated with virus-cell or cell-to-cell fusion (Pekosz et al., 1995b); (Gonzalez-Scarano, 1985) (Gonzalez-Scarano, Pobjecky, and Nathanson, 1984).

Previous studies from our group have demonstrated that the region corresponding to the membrane proximal two-thirds of Gc, amino acids 60-1442, is critical in LACV fusion and entry (Plassmeyer et al., 2005); (Pekosz et al., 1995a) (Pekosz et al., 1995b); this is consistent with the previously defined role of Gc as the exclusive target of neutralizing antibodies generated against LACV. Moreover, computational analysis identified structural similarities between the LACV Gc amino acid region 970-1350 and the E1 fusion protein of two alphaviruses: Sindbis virus and Semilki Forrest virus (SFV). Collectively, these studies suggested that the LACV Gc functions as a class II fusion protein, much like the alphavirus E1 glycoprotein and the flavivirus E glycoprotein, and led to the identification of a 22 amino acid hydrophobic segment (1066-1087) within Gc that was predicted to correlate structurally with a hydrophobic domain of SFV and Sindbis virus (Garry and Garry, 2004) (Plassmeyer et al., 2005). This hydrophobic domain is highly conserved among the Bunyaviridae and closely resembles the fusion domains of SFV E1 and avian sarcoma-leukosis virus (ASLV) E in the placement of its cysteine residues, which form an internal loop (Levy-Mintz and Kielian, 1991).

In a subsequent study, we used site-directed mutagenesis of conserved residues within LACV Gc 1066-1087 to demonstrate that LACV Gc 1066-1087 functions as the LACV fusion peptide (Plassmeyer et al., 2007). The effects of mutations within LACV Gc 1066-1087 were assessed using conformational and non-conformational antibodies specific to Gc, a luciferase based cell-to-cell fusion assay, and pseudotype transduction assays (Ma et al., 1999) (Plassmeyer et al., 2005; Plassmeyer et al., 2007). Several mutations within this hydrophobic domain affected glycoprotein expression to some extent, but all mutations either shifted the pH threshold of fusion below that of the wild type protein, reduced fusion efficiency, or abrogated cell-to-cell fusion and pseudotype entry altogether. Notably, a mutation at position 1066 (W1066A) was particularly informative, because it did not affect glycoprotein expression, yet abrogated fusion and entry. Collectively, these results support a role for the region Gc 1066-1087, as the LACV Gc fusion peptide.

Blakqori and Weber developed a three-cDNA plasmid reverse genetics system for LACV based on a protocol previously established for Bunyamwera virus (Blakqori and Weber, 2005). Using this technology and the existing panel of LACV Gc 1066-1087 fusion peptide mutant constructs, we have generated several recombinant viruses; we used these recombinant viruses to examine the effect of mutations in the fusion peptide region on in vitro growth characteristics, fusion phenotype, and neurotoxicity in the context of a virus infection in physiologically relevant cells and cell lines. We found that recombinant viruses with mutations in LACV Gc (1066-1087) were growth-impaired in muscle cells, insect cells, and primary rat neuronal cultures and had a diminished ability to mediate fusion from within (FFWI). As in wild-type virus, FFWI was pH and temperature dependent in these fusion peptide mutants. Nevertheless, all fusion peptide mutant rLACVs tested were as neurotoxic as wild-type virus in primary neuronal cultures in spite of their decreased growth kinetics and fusion phenotypes; these data are similar to results obtained with a monoclonal selected variant virus (V22) with reduced fusion function but unaltered neurovirulence (González-Scarano, et al., 1985). Overall, this study suggests that the fusion peptide of LACV Gc is a determinant of properties associated with neuroinvasion (growth to high titer in muscle cells, robust fusion phenotype), but not of neurovirulence.

RESULTS

Generation of recombinant La Crosse Virus (rLACV) with specific mutations in Gc

We previously generated several mutant LACV Gc constructs (Plassmeyer et al., 2007). These constructs had either (a) a single amino acid substitution within LACV Gc (1066-1087), (b) a deletion of the entire fusion peptide region, LACV (Δ1066-1087), (c) a substitution containing the entire fusion peptide of SFV instead of the LACV fusion peptide, LACV (SFV-FP), or (d) a mutation outside the fusion peptide region, in the tryptic site located at Gc 761 (Table 1). Although there are many potential proteolytic sites in the Gc glycoprotein (Gentsch and Bishop, 1979), the tryptic site at position 761 is uniquely accessible in the whole virion prior to acidification, and its accessibility is affected by the conformational changes that are induced by low pH (Gonzalez-Scarano, 1985). All of the mutations were engineered into construct pBluescript II KS(+)–LAC(M), then subcloned into the expression vector pCAGGS (Niwa, Yamamura, and Miyazaki, 1991) and sequenced for verification, as previously described (Plassmeyer et al., 2005).

Table 1. Rescue, growth, and fusion phenotypes LACV Gc (1066-1087) amino acid deletion and substitution mutant constructs and LACV Gc 761 tryptic site mutant constructs.

(Plassmeyer et al., 2007)

Construct Transfer Capable VLPs Generated rLACV rescue Growth in BHK-21 FFWI in BHK-21
rLACV-WT YES YES ++++ ++++
W1066A NO NO
G1067A YES YES +++ +
L1074A YES NO
V1076A YES YES +++ +
S1077N YES NO
D1078A YES YES +++ ++
G1083L NO NO
Δ1066-1087 NO NO
SFV FP NO NO
R761A YES NO
R761H YES YES ++ ++++
Open in a new tab

WT (wild type LACV M segment construct used to generate rLACV); VLP (virus-like particles); FFWI (fusion from within)

++++

WT growth or fusion phenotype where ≥80% of the monolayer is in a heterokaryon;

+++

slight decrease in growth or fusion phenotype (70–80% of monolayer in a heterokaryon), decrease is not significant;

++

impaired growth or fusion phenotype (50–70% of monolayer is in a heterokaryon, decrease is significant (p<0.001);

+

severely impaired growth or fusion phenotype, significant (<50% of monolayer is in a heterokaryon).

Subsequently, these mutant glycoproteins constructs were subcloned into pT7ribo-LACV-cM and recombinant LACVs (rLACVs) with mutations in the fusion peptide region were generated. Four recombinant viruses with targeted mutations in Gc were rescued from BSR-T7/5 cells using a three cDNA plasmid system consisting of the L, S, and mutant M segments in antigenomic orientation, transcribed by the T7 RNA polymerase (Table 1) (Blakqori and Weber, 2005). Based on our previous data using constructs containing these respective mutations in a luciferase based cell-cell fusion assay (Plassmeyer et al., 2007), these four rLACV Gc mutants represent a wide rage of predicted fusion phenotypes. Namely, one mutant construct has a normal fusion phenotype (rLACV-R761H), one has a moderately impaired fusion phenotype (rLACV-G1067A), and two have severely impaired fusion phenotypes (rLACV-V1076A, and rLACV-D1078A) (Plassmeyer et al., 2007). We also generated a control recombinant virus with a wild-type glycoprotein sequence (rLACV). After repeated attempts, we were not able to rescue viruses with other specific mutations in LACV Gc (Δ1066-1087, W1066A, G1083L, L1074A, S1077N, and R761A), even though some of these mutant Gc proteins (L1074A, S1077N, and R761A) were capable of mediating cell-to-cell fusion in a plasmid based fusion assay (Plassmeyer et al., 2007) and we were able to generate transfer-capable virus like particles (VLPs) with M segment constructs containing these mutations (Figure 1).

Figure 1. Generation of transfer-capable VLPs using LACV M-segment mutant constructs.

Figure 1

Open in a new tab

VLP transfer of Renilla luciferase with wild-type M-segment constructs was highly successful. The Fisher's Exact Test was used to compare the number of successful attempts to produce transfer-capable VLPs from mutant constructs to the number of successful attempts with the wild-type M-segment construct. VLP production was achieved for the mutant constructs G1067A, :1074A, V1076A, S1077N, and D1078A; there were no significant differences in the number of attempts that sucessfully produced transfer-capable VLPs for G1067A, 1074A, V1076A, S1077N, or D1078A compared to the number of attempts that generated VLPs using the wild-type M-segment construct. VLP transfer was unsuccessful for the Δ1066-1087, W1066A, and G1083L mutant constructs after at least five attempts (*p<0.01). The inability to generate VLPs with these constructs is consistent with data demonstrating that these mutant glycoproteins do not mediate fusion or pseudotype transduction (Plassmeyer et al., 2007).

Expression studies of Gc and Gn in cells infected with recombinant virus

To examine cell surface expression of rLACVs in infected cells, BHK-21 cells (a hamster fibroblastoid cell line) were infected with LACV or one of the five recombinant viruses, at a multiplicity of infection (M.O.I.) of 1.0. Forty-eight hours after infection, the cells were stained with either conformational monoclonal antibodies (MAbs) 807-31 and 807-22 against LACV Gc (50:50, vol/vol) or a commercial MAb against LACV Gn (Abcam, Cambridge, MA). Expression was then determined by flow cytometry (Figure 2). These studies indicated that cells infected with all of the recombinant viruses expressed both LACV glycoproteins and that Gc is expressed with apparent conformational integrity. Notably, expression of Gc was significantly decreased in rLACV-R761H, rLACV-G10761 and rLACV-D1078A infected cells compared to those infected wild-type LACV (Figure 2A, 2B, and 2E). In addition, Gn expression of the rLACV-R761H mutant virus was significantly decreased compared to wild-type LACV Figure (2C, 2D, and 2F). The decreased expression of Gc and Gn in rLACV-R761H infected cells could reflect the reduced replication of this mutant virus in BHK-21 cells or decreased expression efficiency of the glycoproteins. Similar results were obtained in Hela (human epithelial) and C6/36 (mosquito epithelial-like) cells (data not shown).

Figure 2. Surface expression of wild type LACV and recombinant LACV (rLACV) glycoproteins (Gc and Gn) in BHK-21 cells.

Figure 2

Open in a new tab

LACV Gc (A, B, and E) and Gn (C, D, and F) were detected in BHK-21 cells infected for 48 hours at a M.O.I. of 1.0. Expression of Gc was significantly decreased in the rLACV-R761H, rLACV-G10761 and rLACV-D1078A infected cells compared to those infected with wild-type LACV. In addition, Gn expression of the rLACV-R761H mutant virus was significantly decreased compared to that of wild-type LACV (*p<0.05, **P<0.01).

Growth Kinetics of recombinant virus

The in vitro growth characteristics of the Gc mutant recombinant viruses were first evaluated by a growth curve in BHK-21 cells and then in cell lines more reflective of tissues important in LACV pathogenesis: muscle cells (G8 cells differentiated in 1% FCS and 1% horse serum), a mosquito cell line (C6/36), and primary rat neuronal cultures (prepared from embryonic day 17 Sprague Dawley rat pups as described previously (Brewer, 1995)). Replication of rLACV fusion peptide mutants rLACV-V1076A, rLACV-G1067A, rLACV-D1078A was significantly attenuated in differentiated G8 muscle cells (Figure 3B), primary rat neuronal cultures (Figure 3C), and in the mosquito cell line C6/36 (Figure 3D) (Student’s T test; p<0.001). The tryptic site mutant, rLACV-R761H grew as well as wild type LACV and rLACV in these physiologically relevant cells, even though it replicated less efficiently than wild type in BHK-21 (Figure 3A) and in HeLa and 293 T cells (data not shown). Overall, differences in virus replication were less pronounced in the reference cell line BHK-21 (Figure 3A), which suggests that this in vitro approach will be an appropriate screen prior to future in vivo experimentation. Importantly, the growth kinetic of rLACV was not statistically different from that of LACV-WT in the cell lines and cell cultures tested, demonstrating that the recombinant virus is functionally similar to the original isolate.

Figure 3. Growth Kinetics of rLACVs in reference (BHK-21) and physiologically relevant cell lines.

Figure 3

Open in a new tab

BHK-21 (A), differentiated G8 muscle cells (B), primary rat neuronal cultures (C), and C6/36 cells (D) were infected with wild-type LACV and recombinant viruses at an MOI = 0.001. (Infection of primary rat neuronal cultures was initiated at Day in Vitro (DIV) 7). Supernatants from infected cells were harvested at 24, 48, and 72 hours post-infection and a standard plaque assay was performed on VERO cells to determine virus replication. Each curve was performed in triplicate. (*p<0.001 compared to rLACV)

In vitro characterization of recombinant LACV fusion phenotypes

To determine whether the fusion phenotypes of recombinant viruses with mutations in the fusion peptide region recapitulate the findings in the plasmid-based cell-to-cell fusion assay, we then performed fusion from within (FFWI) studies in BHK-21, differentiated G8, and C6/36 cells (Plassmeyer et al., 2007). BHK-21, differentiated G8 cells, and C6/36 cells were infected for 18 hours with LACV or rLACV at an M.O.I. of 1.0. Infected cells and uninfected control cells were exposed to pH-adjusted medium for 30-60 seconds, fixed and then stained for analysis using a Fusion Index (calculated as 1-(C/N) where C= number of cells and N = number of nuclei - a higher fusion index represents more fusion).

In BHK-21, differentiated G8 muscle cells, and C6/36 mosquito cells, fusion indices were similar for wild-type LACV, rLACV and the tryptic mutant rLACV-R761H (Figure 4). However, decreased fusion indices were observed for the three fusion peptide mutants (rLACV-G1067A, rLACV-V1076A, and rLACV-D1078A), once again supporting a role for this hydrophobic domain in LACV fusion and entry. Importantly, the FFWI data obtained with the recombinant LACV is consistent with previous data from our laboratory testing the fusion efficiency of these mutants in the context of a plasmid-based, cell-to-cell fusion assay (Plassmeyer et al., 2007). We also treated cells infected with LACV or the rLACVs with neutralizing antibodies 807.31 and 807.22 after infection and 18 hours prior to exposure to acidified medium (pH 5.5). There was a significant, dose-dependent decrease in the FFWI induced by LACV, rLACV and rLACV-R761H when virus spread was inhibited by anti-LACV neutralizing antibodies (data not shown); the antibodies also led to a slight decrease in the FFWI induced by the fusion peptide mutant viruses (rLACV-G1067A, rLACV-V1076A, and rLACV-D1078A), but the total level of fusion was lower, making a diminution harder to detect (data not shown).

Figure 4. FFWI is significantly decreased in recombinant virus with changes in the fusion peptide region.

Figure 4

Open in a new tab

(A) BHK-21 cells, (B) differentiated G8 cells, and (C) C6/36 cells were infected for 18 hours with LACV or rLACVs at an M.O.I. of 1. Infected cells and uninfected control cells were exposed to pH adjusted medium for 30 to 60 seconds, fixed, and then Giemsa-stained. The Fusion Index was calculated as 1-(C/N), where C is the number of cells and N= number of nuclei. FFWI was significantly decreased in all recombinant viruses with changes in the fusion peptide region compared to wild type LACV and rLACV (Student’s T test; p<0.001). There were n=6 replicates per condition and error bars represent Standard Error.

Temperature dependence of FFWI for recombinant LACV

Previous data from our laboratory has demonstrated a temperature dependence of FFWI with an optimum temperature of 37°C. To determine whether changes in the fusion peptide domain alter the optimal temperature of FFWI, we performed a FFWI assay in BHK cells at 4°C, 23°C, 35°C, 37°C, and 42°C. As demonstrated in Figure 5, the optimal temperature for FFWI for LACV and rLACV, rLACV-R761H, and rLACV-D1078A mutant recombinant viruses was 37°C, diminishing at increased temperatures (42°C), which is consistent with previous findings (Pobjecky, Smith, and Gonzalez-Scarano, 1986). A clear optimal temperature for FFWI for the rLACV-V1076A and rLACV-G1067A mutants was not apparent, but that may be because the magnitude of FFWI is comparatively low for these mutant viruses.

Figure 5. FFWI in BHK-21 cells is temperature dependent.

Figure 5

Open in a new tab

BHK-21 cells were infected with LACV and recombinant viruses at an M.O.I. of 1. FFWI was performed over a range of temperatures at pH 5.5. There were n=6 replicates per condition and error bars represent Standard Error of the Mean.

LACV induced neurotoxicity

A MAP2 (microtubule associated protein 2) cell-based ELISA assay was used to quantify neuronal death as previously described (Wang et al., 2007). Infection with wild type LACV, rLACV, the tryptic mutant (rLACV-R761H), and the fusion peptide mutant viruses (rLACV-G1067A, rLACV-V1076A, and rLACV-D1078A) all induced significant neuronal death as measured by decreased MAP2 staining compared to MOCK infected cells (Newman-Keuls; p<.05) (Figures 6A and 6B). Interestingly, we did not observe a significant increase in neurotoxicity in LACV or rLACV infected primary rat neuronal cultures compared to those infected by the rLACV fusion peptide mutant viruses even though these viruses grew to a higher titer in these cells (Figure 3B).

Figure 6. Neuronal loss in primary rat neuronal cultures infected with LACV and rLACVs.

Figure 6

Open in a new tab

(A) A MAP2 cell-based ELISA assay was used to quantify neuronal death as previously described (Wang et al., 2007). Infection with wild type LACV, rLACV, the tryptic mutant (rLACV-R761H), and the fusion peptide mutant viruses (rLACV-G1067A, V1076A, and D1078A) all induced significant neuronal death as measured by decreased MAP2 staining compared to MOCK infected cells (Newman-Keuls; p<.05). Data were collected as arbitrary fluorescence units and expressed as percent MAP2 positive relative to MOCK infected control. There were n=10 replicates per condition and error bars represent Standard Error of the Mean. (B) Visualization of LACV infected primary rat neuronal cultures by IFA. FITC (green)=LACV Gc; TRITC(red)=MAP2; Hoecht’s (blue). Neuronal damage and loss, as measured by loss of MAP2 staining, is visible in primary rat neuronal cultures infected with WT LACV, rLAV, the tryptic mutant virus (rLACV-R761H) and the fusion peptide mutant viruses (rLACV-G1067A, rLACV-V1076A, and rLACV-D1078A).

DISCUSSION

The availability of a reverse genetics system to generate LACV with targeted mutations represents an important strategic advance in the analysis of its neuropathogenesis and, potentially, in the development of therapeutics and vaccines to prevent infections by LACV and other bunyaviruses. In this report, we used this system to characterize the growth characteristics, fusion phenotypes, and neurotoxicity of each of four rescued Gc LACV mutant viruses. These represent a range of predicted fusion phenotypes (Plassmeyer et al., 2007): one with a normal fusion phenotype (rLACV-R761H), one with a moderately impaired fusion phenotype (rLACV-G1067A), and two with severely impaired fusion phenotypes (rLACV-V1076A, and rLACV-D1078A) (Plassmeyer et al., 2007).

With one exception (rLACV-R761H), the recombinant viruses replicated similarly in our reference cell lines (Figure 3A). However, the fusion peptide mutant viruses (rLACV-V1076A, rLACV-G1067A, and rLACV-D1078A) were significantly attenuated in the muscle cells, in the insect cells, and even in primary rat neuronal cultures. It is of great interest that the fusion peptide mutant viruses had impaired growth phenotypes in the insect cell line and physiologically relevant cells (G8 muscle cells and primary rat neuronal cultures), but not in BHK-21 cells. We speculate that the decreased growth kinetics observed in the G8s and primary rat neuronal cultures may be associated with terminal differentiation of these cells. Moreover, proteolytic processing of Gc in the insect vector is extensive and it has been suggested the role of LACV Gc in fusion and entry in mosquito cells may be different from that in mammalian cells (Sundin et al., 1987). Therefore, the consequences of mutations in the fusion peptide region may not be identical in an invertebrate system. Importantly, the fusion phenotypes observed for each of the recombinant viruses mirrored that of their parent Gc construct, validating the relevance of the plasmid based cell-to-cell fusion assay and our previous data (Plassmeyer et al., 2007). Notably we did not observe a discernable shift in pH activation with the fusion peptide mutant viruses, although there was an apparent shift (lower pH activation) in the plasmid based cell-to-cell fusion assay using the parent mutant constructs D1078A, G1067A, and V1076A (Plassmeyer et al., 2007). Paradoxically, the tryptic mutant R761H had a wild-type fusion phenotype even in cell lines where its replication was impaired. Although FFWI did not always correlate with virus replication (Figures 3 and 4), neutralizing antibodies specific for Gc were demonstrated to inhibit both the spread and FFWI of recombinant LACVs (data not shown). The ability to inhibit both virus replication and fusion with neutralizing antibodies targeting Gc underscores the importance of understanding the role of this region in virus pathogenesis.

To understand the effect that changes in the fusion peptide region may have on neuronal death, a cell property associated with neurovirulence, we used a MAP2 cell-based ELISA assay that quantifies neuronal death in primary rat neuronal cultures. We demonstrated LACV induced neurotoxicity when primary rat neuronal cultures were infected with LACV, rLACV and all of the Gc Mutant rLACVs (rLACV-V1076A, rLACV-G1067A, and rLACV-D1078A). The degree of neuronal loss was similar in the wild-type and fusion peptide mutant viruses and did not correlate with virus replication or the relative ability to mediate FFWI, suggesting that in vitro neurovirulence may not be directly affected by these changes in the fusion peptide region. These data may suggest that the neuronal damage observed in this system may be attributable, in part, to pro-apoptotic factors, cytokines, and/or excitatory amino acids released by infected cells and not exclusively caused by direct lysis of LACV infected neurons.

LACV monoclonal antibody escape variants V22 and V22F, generated with a monoclonal antibody against Gc (807-22), were associated with decreased replication in peripheral tissues that are critical for the generation of a high viremia such as striated muscle; they also had a decreased ability to mediate cell-to-cell fusion, though mutations in these viruses were not in the fusion region. Although these monoclonal antibody escape variants were not neuroinvasive, owing to the more limited viremia generated, they were as neurovirulent as the parent LACV when directly injected into the CNS. In the present study, we have demonstrated that single amino acid substitutions in the LACV Gc fusion peptide result in a diminished ability to mediate fusion from within (FFWI) and impaired growth in muscle cells and in primary rat neuronal cultures. However, these deficiencies in growth and fusion phenotypes do not correlate with reduced neuronal damage in an in vitro model of neurotoxicity using primary rat neuronal cultures. It remains to be seen whether in a mouse model the fusion peptide of LACV Gc will influence the ability of the LACV to grow to high titer in peripheral tissues and subsequently penetrate the CNS, but is ultimately not a determinant of neurovirulence, but these studies using surrogate cell lines point in that direction.

In future studies, we will use these recombinant mutant viruses to address the following questions in the existing mouse model: (1) Will mutations that affect fusion efficiency lead to decreased LACV replication in key tissues such as muscle have a decreased viremia and decreased neuroinvasiveness? (2) Conversely, do mutations in other regions of Gc, like the 761 tryptic site mutant have altered tissue tropism that increases viremia and neuroinvasion? (3) Will these fusion peptide mutant viruses be as neurovirulent as the parent LACV when injected directly into the CNS? (4) Can LACV mutants, particularly those demonstrated to have impaired neuroinvasive phenotypes, be used to generate a protective immune response and serve as the basis for an attenuated virus vaccine? These studies will serve to further elucidate the role of the LACV fusion peptide region in the neuropathogenesis of LACV encephalitis. Importantly, because the fusion peptide region is conserved among the Bunyaviridae (Garry and Garry, 2004) (Plassmeyer et al., 2007), the findings from this study and future in vivo experimentation may be extrapolated to other emerging bunyaviruses such as Crimean-Congo Hemorrhagic fever virus and Rift Valley Fever virus with the potential to lead to the development of therapeutic modalities and vaccines.

MATERIALS AND METHODS

LACV M segment constructs and site directed mutagenesis

Fusion peptide mutant LACV M segment constructs were generated as previously described (Plassmeyer et al., 2007). Briefly, the LACV M segment ORF was subcloned from pcDNA3.1 into the expression vector pCAGGS (Niwa, Yamamura, and Miyazaki, 1991) to employ the CMV enhancer and strong chicken β-actin promoter to express the polyprotein, as previously described (Plassmeyer et al., 2005; Plassmeyer et al., 2007). The LACV M ORF was amplified using rTth DNA Polymerase XL (PE Biosystems, Foster City, CA) with a primer set that introduced restriction cut sites ClaI and XhoI at the 5′ and 3′ ends, respectively. The PCR cycle consisted of a 93°C 2 minute denaturing step followed by 35 cycles of (93°C 3 for 0 seconds, 48°C for 1 minute, 72°C for 6 minutes) and an extension of 72°C 10 minutes. The resulting fragment was digested with both ClaI and XhoI and inserted into the linearized pCAGGS expression vector digested with ClaI and XhoI.

To engineer mutations (point, deletion, and insertion of SFV FP) in the putative fusion domain of LACV Gc and the Gc (761-772) tryptic site, the LACV M segment was subcloned into pBluescript II KS+ using ClaI and XhoI (Plassmeyer et al., 2005) (Plassmeyer et al., 2007). Point mutations were engineered into the M segment using the QuikChange II XL site directed mutagenesis kit following manufacturer’s instructions (Stratagene, La Jolla, CA). The primers used to engineer these substitutions have been published (Plassmeyer et al., 2007). The point mutant constructs were then cloned into the expression vector pCAGGS. These M segment clones were sequenced to verify that there were no additional mutations.

Generation of fusion peptide mutant rLACVs

rLACV were generated from the above Gc mutant constructs as described (Blakqori and Weber, 2005). The LACV minireplicon plasmids pTM-LACV N, pTM-LACV L and pLACV-vRen, the eukaryotic expression construct pI.18-LACV M, and the LACV rescue plasmids pT7Ribo-LACV-cL and pT7Ribo-LACV-cSNoEco were also described previously (Blakqori et al., 2003) (Blakqori and Weber, 2005). The T7-expressing plasmid pCAGGsT7 was kindly provided by Ramon Flick (UTMB Texas). The eukaryotic firefly luciferase expression construct pGL3-control was purchased from Promega. For cloning of the mutant LACV M virus expression constructs and rescue constructs, M ORF fragments containing the mutated region were cut out of the respective pCAGGS constructs using NheI/KpnI digestion, and inserted into the wild type M segment rescue construct pT7riboSM2-LACV-vM or the eukaryotic expression construct pI.18 LACV M, respectively, which had been cleaved with the same enzymes.

Generation of LACV VLPs

LACV VLPs were generated by a method similar to the one described for Rift Valley fever virus VLPs (Habjan et al., 2009). Human Huh7 cells grown in six-well dishes were transfected with 0.25 μg each of pTM-LACV N, pTM-LACV L and pLACV-vRen along with 0.5 μg each of pCAGGsT7 and pI.18-LACV M (WT and mutant constructs), and 0.1 μg pGL3-control, using Nanofectin transfection agent (PAA). Cells transfected without the M expression constructs served as VLP negative control. At 48 h post-transfection, supernatants were harvested and cells were lysed in 200 μl Passive Lysis buffer to measure Renilla (LACV minireplicon) and firefly (pGL3 control) luciferase activities using the Dual Luciferase kit (Promega). Cell supernatants were treated with 1 μl/ml Benzonase (Novagen) at 37°C for 3 h and centr ifuged at 12,000 g for 5 min to remove cell debris. An aliquot of 250 μl cell supernatant was then used to infect 1 well of a 6-well plate of BSR-T7/5 cells (Buchholz, Finke, and Conzelmann, 1999) which had been transfected overnight with 0.25 μg each of the LACV expression constructs pTM-LACV N and pTM-LACV L. VLP-infected cells were assayed with the Dual Luciferase system, where Renilla activity indicated succesful VLP infection, and firefly luciferase indicated background transfection.

Generation of fusion peptide mutant viruses

Recombinant LAC viruses (rLACV) were generated from the above Gc mutant constructs as described (Blakqori and Weber, 2005). Briefly, subconfluent layers of BSR-T7/5 cells grown in six-well plates were transfected with 0.5 μg each of plasmids pT7Ribo-LACV-cL, pT7Ribo-LACV-cSNoEco, and pT7ribo-LACV-vM (and mutant variants thereof) using Fugene transfection reagent (Roche). Supernatants were harvested 5 days later, and 200 μl were used to inoculate Vero cells for 1 h at 37°C. After replacing the inoculum with fresh medium, Vero cells were monitored for the appearance of a cytopathic effect for the next 5 days. wtLACV rescue (rLACV) was always included as positive control. Virus rescues were attempted 3 times with transfection of one complete 6-well plate before determined to be unsuccessful.

Plaque assays

Plaque assays for LACV and rLACVs were performed in VERO cells, an African green monkey kidney cell line. VERO cells were plated overnight at 3X105 cells/well in a 6 well plate and serial dilutions of cell culture supernatants (10−2 to 10−6) and uninfected control supernatants were applied for 1 hour. The diluted culture supernatants were then removed and a 1:1 mixture of 2% carboxymethyl-cellulose (Sigma-Aldrich, Saint Louis, MO) and 2X DMEM (Gibco) and 4% fetal calf serum were added. VERO cells were incubated for 72 hours at 32°C, 10% CO2 before the carboxymethyl-cellulose overlay was removed. Plaques were then visualized by staining with 0.1% crystal violet.

Expression of wild-type and mutant LACV Gc in cells infected by recombinant LACVs: detection by flow cytometry

BHK-21s, Hela and C6/36 cells were harvested and fixed for 30 minutes in a 2% paraformaldehyde solution. Non-specific binding was reduced by exposure to 10% goat serum (Sigma, St. Louis, MO) for 30 min. Cells were then washed three times in fluorescence-activated cell sorter (FACS) stain buffer (PBS with 1% FCS and 0.1% sodium azide) prior to incubation at room temperature for 60 minutes with either a 1:100 dilution of conformation dependent MABs 807-31 and 807-22 against LACV Gc (50:50, vol/vol), or a commercial MAB against LACV Gn (Abcam, Cambridge, MA). For each experiment, an aliquot of cells was also treated with isotype-matched control antibodies to establish background staining. The cells were then washed three times in FACS stain buffer and incubated at room temperature with a 1:100 dilution of a FITC conjugated anti-mouse IgG (Sigma) for 30 minutes. After incubation with the FITC conjugated secondary antibody, cells were washed three times in FACS stain buffer and resuspended in 400 μl FACS stain buffer before acquisition and analysis on a FACScalibur instrument (Becton Dickinson, Sunnyvale, CA) using FlowJo software (University of Pennsylvania Cancer Center).

Fusion from within assay (FFWI)

FFWI was performed as described previously (Gonzalez-Scarano et al., 1985) (Pobjecky, Smith, and Gonzalez-Scarano, 1986). BHK-21, C636, and differentiated G8 cells were plated at a concentration of 1X105 cells per well in poly-L-lysine (Sigma) pre-coated 24 well plates. Subconfluent monolayers of cells were then inoculated at a multiplicity of infection of 1.0 pfu/cell, incubated for one hour, washed with PBS and incubated with DMEM plus 2% FCS (infection medium) for 18 hours at 35°C. At that point, cells were washed again with PBS and then exposed to medium (serum free DMEM with 10 mM morpholinoethanesulfonic acid adjusted to the appropriate pH with NaOH). The acidic medium was replaced with maintenance medium after 30–60s and the cells were further incubated at 37°C (or at other temperatures for experiments described in Figure 5) for 30 min. The cells were then fixed and stained with Geimsa (Difco Laboratories, Detroit, MI), and counted. The fusion index (FI) was determined as FI= 1- (C/N) where C and N are the numbers of cells and nuclei, respectively.

Preparation of primary neuronal cultures

Primary rat neuronal cultures were prepared from embryonic day 17 Sprague Dawley rat pups as described previously (Brewer, 1995). Cells were plated on plates or German glass coverslips pre-coated with poly-L-lysine (Sigma) at a density of 4X104 cells per well in a 96 well plate or at a density of 2X105 cells per well in a 24 well plate and maintained in neurobasal medium supplemented with B27 (Invitrogen) at 37°C, 5% CO2. Primary rat neuronal cultures were infected with LACV at 7 days in vitro (DIV).

Quantification of LACV induced neurotoxicity

Microtubule associate protein 2 (MAP2) cell-based ELISA

A MAP2 cell-based ELISA assay was used to quantify neuronal death as previously described (Wang et al., 2007). This MAP2 cell-based ELISA has been demonstrated to produce results similar to those produced by hand quantification of MAP2 positive neurons (O’Donnell et al., 2006); (Wang et al., 2007). Briefly, primary rat cortical neuroglial cultures were plated in poly-L-lysine-coated 96 well plates at a concentration of 4X104 cells per well. Forty-eight hours after infection with LACV or rLAVCs at day in vitro (DIV) 7 at an M.O.I. of 1.0, cells were fixed for 30 min with 4% paraformaldehyde in 4% sucrose. After blocking for one hour with 5% normal goat serum in PBS, the plates were incubated with mouse anti-MAP2 antibody (1:1000 dilution) overnight at 4°C, followed by washing with PBS-Tween. The plates were then incubated for 30 min with goat anti-mouse secondary antibody conjugated to beta-lactamase TEM-1 (1:500), washed, and incubated at 23°C for 60 min with 0.5 μg Fluorocillin™ Green substrate per well. Fluorescence intensity was measured using a Fluoroskan Ascent fluorometer plate reader (Thermo Electron, Waltham, MA, USA) with excitation at 485 nm and emmision at 527 nm. Relative changes in the intensity of the fluorescent MAP2 signal were used to estimate neuronal damage; n=10 replicates per condition.

Immunofluroescence Assay (IFA) for MAP2 and LACV detection

Primary cortical neuroglial cultures were plated in poly-L-lysine-coated 96 well plates at a concentration of 4X104 cells per well. Forty-eight hours after infection with LACV or rLAVCs at day in vitro (DIV) 7 at an M.O.I. of 1.0, cells were fixed for 30 min with 4% paraformaldehyde in 4% sucrose in PBS for one hour followed by permeabilization in 100% methanol (10 min) and 0.2% Triton X-100/PBS (10 min). Cultures were then washed twice with PBS, blocked with 10% goat serum in PBS (45 min. at room temperature), and then exposed to anti-MAP2 antibody (60 μM in 10 % goat serum in PBS) and a 1:500 dilution of pre-adsorbed rabbit anti-LACV Gc antibody for 1 h at room temperature. After washing, a tetrametylrhodamine isothicyanate (TRITC) conjugated goat anti-IgG secondary antibody (66 μM; 45 minutes at room temperature) was used for detection of MAP2 antibody binding and a fluorescein isothiocyanate (FITC) conjugated anti-rabbit IgG antibody (1:100) was used for the detection of LACV antibody binding.

Acknowledgments

This work was supported by PHS grant 1R01AI074626-01. We thank Dennis L. Kolson, Marc A. Dichter, and Denise R. Cook (University of Pennsylvania) for providing embryonic rat brain cells, Jesse Medina for his work on this study, and Michael G. White (University of Pennsylvania) for developing the MAP2 cell-based ELISA assay.

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. Blakqori G, Kochs G, Haller O, Weber F. Functional L polymerase of La Crosse virus allows in vivo reconstitution of recombinant nucleocapsids. J Gen Virol. 2003;84(Pt 5):1207–14. doi: 10.1099/vir.0.18876-0. [DOI] [PubMed] [Google Scholar]
  2. Blakqori G, Weber F. Efficient cDNA-based rescue of La Crosse bunyaviruses expressing or lacking the nonstructural protein NSs. J Virol. 2005;79(16):10420–8. doi: 10.1128/JVI.79.16.10420-10428.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brewer GJ. Serum-free B27/neurobasal medium supports differentiated growth of neurons from the striatum, substantia nigra, septum, cerebral cortex, cerebellum, and dentate gyrus. J Neurosci Res. 1995;42(5):674–83. doi: 10.1002/jnr.490420510. [DOI] [PubMed] [Google Scholar]
  4. Buchholz UJ, Finke S, Conzelmann KK. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J Virol. 1999;73(1):251–9. doi: 10.1128/jvi.73.1.251-259.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bupp K, Stillmock K, Gonzalez-Scarano F. Analysis of the intracellular transport properties of recombinant La Crosse virus glycoproteins. Virology. 1996;220(2):485–90. doi: 10.1006/viro.1996.0336. [DOI] [PubMed] [Google Scholar]
  6. Endres MJ, Jacoby DR, Janssen RS, Gonzalez-Scarano F, Nathanson N. The large viral RNA segment of California serogroup bunyaviruses encodes the large viral protein. J Gen Virol. 1989;70 (Pt 1):223–8. doi: 10.1099/0022-1317-70-1-223. [DOI] [PubMed] [Google Scholar]
  7. Eshita Y, Ericson B, Romanowski V, Bishop DH. Analyses of the mRNA transcription processes of snowshoe hare bunyavirus S and M RNA species. J Virol. 1985;55(3):681–9. doi: 10.1128/jvi.55.3.681-689.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fazakerley JK, Gonzalez-Scarano F, Strickler J, Dietzschold B, Karush F, Nathanson N. Organization of the middle RNA segment of snowshoe hare Bunyavirus. Virology. 1988;167(2):422–32. [PubMed] [Google Scholar]
  9. Garry CE, Garry RF. Proteomics computational analyses suggest that the carboxyl terminal glycoproteins of Bunyaviruses are class II viral fusion protein (beta-penetrenes) Theor Biol Med Model. 2004;1:10. doi: 10.1186/1742-4682-1-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gentsch JR, Bishop DL. M viral RNA segment of bunyaviruses codes for two glycoproteins, G1 and G2. J Virol. 1979;30(3):767–70. doi: 10.1128/jvi.30.3.767-770.1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gonzalez-Scarano F. La Crosse virus G1 glycoprotein undergoes a conformational change at the pH of fusion. Virology. 1985;140(2):209–16. doi: 10.1016/0042-6822(85)90359-9. [DOI] [PubMed] [Google Scholar]
  12. Gonzalez-Scarano F, Janssen RS, Najjar JA, Pobjecky N, Nathanson N. An avirulent G1 glycoprotein variant of La Crosse bunyavirus with defective fusion function. J Virol. 1985;54(3):757–63. doi: 10.1128/jvi.54.3.757-763.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gonzalez-Scarano F, Pobjecky N, Nathanson N. La Crosse bunyavirus can mediate pH-dependent fusion from without. Virology. 1984;132(1):222–5. doi: 10.1016/0042-6822(84)90107-7. [DOI] [PubMed] [Google Scholar]
  14. Gonzalez-Scarano F, Shope RE, Calisher CE, Nathanson N. Characterization of monoclonal antibodies against the G1 and N proteins of LaCrosse and Tahyna, two California serogroup bunyaviruses. Virology. 1982;120(1):42–53. doi: 10.1016/0042-6822(82)90005-8. [DOI] [PubMed] [Google Scholar]
  15. Griot C, Pekosz A, Davidson R, Stillmock K, Hoek M, Lukac D, Schmeidler D, Cobbinah I, Gonzalez-Scarano F, Nathanson N. Replication in cultured C2C12 muscle cells correlates with the neuroinvasiveness of California serogroup bunyaviruses. Virology. 1994;201(2):399–403. doi: 10.1006/viro.1994.1308. [DOI] [PubMed] [Google Scholar]
  16. Griot C, Pekosz A, Lukac D, Scherer SS, Stillmock K, Schmeidler D, Endres MJ, Gonzalez-Scarano F, Nathanson N. Polygenic control of neuroinvasiveness in California serogroup bunyaviruses. J Virol. 1993;67(7):3861–7. doi: 10.1128/jvi.67.7.3861-3867.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Habjan M, Penski N, Wagner V, Spiegel M, Overby AK, Kochs G, Huiskonen JT, Weber F. Efficient production of Rift Valley fever virus-like particles: The antiviral protein MxA can inhibit primary transcription of bunyaviruses. Virology. 2009;385(2):400–8. doi: 10.1016/j.virol.2008.12.011. [DOI] [PubMed] [Google Scholar]
  18. Janssen R, Gonzalez-Scarano F, Nathanson N. Mechanisms of bunyavirus virulence. Comparative pathogenesis of a virulent strain of La Crosse and an avirulent strain of Tahyna virus. Lab Invest. 1984;50(4):447–55. [PubMed] [Google Scholar]
  19. Janssen RS, Nathanson N, Endres MJ, Gonzalez-Scarano F. Virulence of La Crosse virus is under polygenic control. J Virol. 1986;59(1):1–7. doi: 10.1128/jvi.59.1.1-7.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kalfayan B. Pathology of La Crosse virus infection in humans. In: Calisher CH, Thompso WH, editors. California Serogroup Viruses. Liss; New York: 1983. pp. 179–186. [PubMed] [Google Scholar]
  21. Lappin DF, Nakitare GW, Palfreyman JW, Elliott RM. Localization of Bunyamwera bunyavirus G1 glycoprotein to the Golgi requires association with G2 but not with NSm. J Gen Virol. 1994;75 (Pt 12):3441–51. doi: 10.1099/0022-1317-75-12-3441. [DOI] [PubMed] [Google Scholar]
  22. Levy-Mintz P, Kielian M. Mutagenesis of the putative fusion domain of the Semliki Forest virus spike protein. J Virol. 1991;65(8):4292–300. doi: 10.1128/jvi.65.8.4292-4300.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ma M, Kersten DB, Kamrud KI, Wool-Lewis RJ, Schmaljohn C, Gonzalez-Scarano F. Murine leukemia virus pseudotypes of La Crosse and Hantaan Bunyaviruses: a system for analysis of cell tropism. Virus Res. 1999;64(1):23–32. doi: 10.1016/s0168-1702(99)00070-2. [DOI] [PubMed] [Google Scholar]
  24. McJunkin JE, Khan RR, Tsai TF. California-La Crosse encephalitis. Infect Dis Clin North Am. 1998;12(1):83–93. doi: 10.1016/s0891-5520(05)70410-4. [DOI] [PubMed] [Google Scholar]
  25. Niwa H, Yamamura K, Miyazaki J. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene. 1991;108(2):193–9. doi: 10.1016/0378-1119(91)90434-d. [DOI] [PubMed] [Google Scholar]
  26. O’Donnell LA, Agrawal A, Jordan-Sciutto KL, Dichter MA, Lynch DR, Kolson DL. Human immunodeficiency virus (HIV)-induced neurotoxicity: roles for the NMDA receptor subtypes. J Neurosci. 2006;26(3):981–90. doi: 10.1523/JNEUROSCI.4617-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pekosz A, Griot C, Nathanson N, Gonzalez-Scarano F. Tropism of bunyaviruses: evidence for a G1 glycoprotein-mediated entry pathway common to the California serogroup. Virology. 1995a;214(2):339–48. doi: 10.1006/viro.1995.0043. [DOI] [PubMed] [Google Scholar]
  28. Pekosz A, Griot C, Stillmock K, Nathanson N, Gonzalez-Scarano F. Protection from La Crosse virus encephalitis with recombinant glycoproteins: role of neutralizing anti-G1 antibodies. J Virol. 1995b;69(6):3475–81. doi: 10.1128/jvi.69.6.3475-3481.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Plassmeyer ML, Soldan SS, Stachelek KM, Martin-Garcia J, Gonzalez-Scarano F. California serogroup Gc (G1) glycoprotein is the principal determinant of pH-dependent cell fusion and entry. Virology. 2005;338(1):121–32. doi: 10.1016/j.virol.2005.04.026. [DOI] [PubMed] [Google Scholar]
  30. Plassmeyer ML, Soldan SS, Stachelek KM, Roth SM, Martin-Garcia J, Gonzalez-Scarano F. Mutagenesis of the La Crosse Virus glycoprotein supports a role for Gc (1066–1087) as the fusion peptide. Virology. 2007;358(2):273–82. doi: 10.1016/j.virol.2006.08.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Pobjecky N, Smith J, Gonzalez-Scarano F. Biological studies of the fusion function of California serogroup Bunyaviruses. Microb Pathog. 1986;1(5):491–501. doi: 10.1016/0882-4010(86)90011-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ruusala A, Persson R, Schmaljohn CS, Pettersson RF. Coexpression of the membrane glycoproteins G1 and G2 of Hantaan virus is required for targeting to the Golgi complex. Virology. 1992;186(1):53–64. doi: 10.1016/0042-6822(92)90060-3. [DOI] [PubMed] [Google Scholar]
  33. Sundin DR, Beaty BJ, Nathanson N, Gonzalez-Scarano F. A G1 glycoprotein epitope of La Crosse virus: a determinant of infection of Aedes triseriatus. Science. 1987;235(4788):591–3. doi: 10.1126/science.3810159. [DOI] [PubMed] [Google Scholar]
  34. Thompson WH, Kalfayan B, Anslow RO. Isolation of California Encephalitis Group Virus from a Fatal Human Illness. Am J Epidemiol. 1965;81:245–53. doi: 10.1093/oxfordjournals.aje.a120512. [DOI] [PubMed] [Google Scholar]
  35. Wang Y, White MG, Akay C, Chodroff RA, Robinson J, Lindl KA, Dichter MA, Qian Y, Mao Z, Kolson DL, Jordan-Sciutto KL. Activation of cyclin-dependent kinase 5 by calpains contributes to human immunodeficiency virus-induced neurotoxicity. J Neurochem. 2007;103(2):439–55. doi: 10.1111/j.1471-4159.2007.04746.x. [DOI] [PubMed] [Google Scholar]

RESOURCES