Cellular Localization and Antigenic Characterizati
http://www.100md.com
病菌学杂志 2005年第10期
Department of Microbiology, University of Pennsylvania, Philadelphia, Pennsylvania 19104
AlphaVax, Inc., 2 Triangle Drive, Research Triangle Park, North Carolina 27709
Virology Division, U.S. Army Medical Research Institute for Infectious Diseases, Fort Detrick, Frederick, Maryland 21702
Department of Microbiology, Mount Sinai School of Medicine, One Gustav L. Levy Place, New York, New York 10029
ABSTRACT
Crimean-Congo hemorrhagic fever virus (CCHFV), a member of the genus Nairovirus of the family Bunyaviridae, causes severe disease with high rates of mortality in humans. The CCHFV M RNA segment encodes the virus glycoproteins GN and GC. To understand the processing and intracellular localization of the CCHFV glycoproteins as well as their neutralization and protection determinants, we produced and characterized monoclonal antibodies (MAbs) specific for both GN and GC. Using these MAbs, we found that GN predominantly colocalized with a Golgi marker when expressed alone or with GC, while GC was transported to the Golgi apparatus only in the presence of GN. Both proteins remained endo-?-N-acetylglucosaminidase H sensitive, indicating that the CCHFV glycoproteins are most likely targeted to the cis Golgi apparatus. Golgi targeting information partly resides within the GN ectodomain, because a soluble version of GN lacking its transmembrane and cytoplasmic domains also localized to the Golgi apparatus. Coexpression of soluble versions of GN and GC also resulted in localization of soluble GC to the Golgi apparatus, indicating that the ectodomains of these proteins are sufficient for the interactions needed for Golgi targeting. Finally, the mucin-like and P35 domains, located at the N terminus of the GN precursor protein and removed posttranslationally by endoproteolysis, were required for Golgi targeting of GN when it was expressed alone but were dispensable when GC was coexpressed. In neutralization assays on SW-13 cells, MAbs to GC, but not to GN, prevented CCHFV infection. However, only a subset of GC MAbs protected mice in passive-immunization experiments, while some nonneutralizing GN MAbs efficiently protected animals from a lethal CCHFV challenge. Thus, neutralization of CCHFV likely depends not only on the properties of the antibody, but on host cell factors as well. In addition, nonneutralizing antibody-dependent mechanisms, such as antibody-dependent cell-mediated cytotoxicity, may be involved in the in vivo protection seen with the MAbs to GC.
INTRODUCTION
Crimean-Congo hemorrhagic fever virus (CCHFV) causes a hemorrhagic and toxic syndrome in humans with mortality rates of up to 50%. CCHFV infection was first described during an outbreak in Russia during the 1940s, when more than 200 cases of severe hemorrhagic fever were reported among agricultural workers and soldiers in the Crimean peninsula (15, 16). Since then, the virus has spread throughout many regions of the world, including sub-Saharan Africa (60, 61), Bulgaria, the Arabian Peninsula, Iraq, Pakistan, the former Yugoslavia, northern Greece, and northwest China (16, 23, 42-45).
CCHFV is a member of the genus Nairovirus within the family Bunyaviridae (52). Members of this enveloped-virus family have a tripartite, single-stranded RNA genome of negative polarity. The medium RNA segment (the M segment) encodes the viral glycoproteins GN and GC, which, like those of other Bunyaviridae, are synthesized as polyprotein precursors that undergo proteolytic cleavage events to yield mature glycoproteins (52). The CCHFV glycoproteins exhibit several unusual structural features and undergo several processing events. First, the CCHFV glycoproteins contain, on average, 78 to 80 cysteine residues, suggesting the presence of an exceptionally large number of disulfide bonds and a complex secondary structure. Second, the GN precursor protein (Pre-GN) contains a highly variable domain at its amino terminus that contains a high proportion of serine, threonine, and proline residues, and it is predicted to be heavily O glycosylated, thus resembling a mucin-like domain present in other viral glycoproteins, most notably the Ebola virus glycoprotein (56). The mucin-like region in the Ebola glycoprotein has been shown to play an important role in a cell-rounding phenotype and immunoevasion (39, 56). It is not known whether this domain plays an important role in CCHFV pathogenesis or whether it is even O glycosylated. A third unusual feature is that the GN glycoprotein can undergo two posttranslational proteolytic cleavage events at the conserved motifs RSKR and RRLL, potentially releasing the mucin-like domain as well as a second N-terminal domain of approximately 35 kDa (P35, or the connector domain) (59). It is not known if the released domains traffic to an intracellular compartment, if they are secreted, or what effect they may have on viral pathogenesis and antigenic structure. Similar processing strategies have not been observed for other Bunyaviridae outside of the Nairovirus group.
As the only virally encoded membrane proteins, GN and GC must interact with cell surface receptors, mediate the entry of virus into cells, and serve as targets for neutralizing antibodies. Passive transfer of neutralizing antibodies can protect susceptible animals from hantavirus infection (8, 53-55, 62), and there is a report that convalescent-human sera can afford some protection in acutely infected individuals (58). Thus, characterizing the structures and functions of these proteins will be important for understanding CCHFV tropism and pathogenesis as well as for vaccine development. In this study, we describe the first monoclonal antibodies (MAbs) raised against the CCHFV glycoproteins, map the subunits to which they bind, and characterize their abilities to neutralize virus and to protect mice from a lethal CCHFV challenge. In addition, using these MAbs, we investigated the localization of GN and GC when expressed alone or together and have begun to map the regions involved in glycoprotein localization and interactions.
MATERIALS AND METHODS
Cells, antibodies, and viruses. CCHFV prototype strain IbAr10200, first isolated in 1976 from Hyalomma excavatum ticks from Sokoto, Nigeria, was grown in African green monkey kidney Vero cells or the E6 variant (51). African green monkey kidney fibroblast (CV-1), Vero, Vero E6, human cervix carcinoma (HeLa), and human embryonic kidney (HEK 293T) cells were maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA). Similarly, the human tumor cell line SW-13 (adrenocortical carcinoma) was grown in DMEM supplemented with 2.5% fetal bovine serum. Work with CCHFV was performed in a biosafety level 4 laboratory at the U.S. Army Medical Research Institute for Infectious Diseases. A recombinant vaccinia virus expressing the T7 bacteriophage RNA polymerase (vTF1.1) was grown in HeLa cells, and titers were determined in CV-1 cells according to standard protocols (1).
Production of MAbs. MAbs were prepared against the GN and GC glycoproteins of the CCHFV strain IbAr10200 by fusion of SP2/0 myeloma cells with splenocytes from BALB/c inbred mice. We carried out five independent fusions in which mice were immunized with infected suckling mouse brain homogenates (fusions I and II; MAb 30F7), with affinity-purified virion glycoproteins precipitated from nonionic detergent lysates of gradient-purified virus preparations (fusions III and IV; MAbs 1H6, 5E3, 6C2, 5B5, 8C4, 9H3, 3E3, 8G7, 8A8, and 8F10), or with affinity-purified proteins from similar lysates of infected cells (fusion V; MAbs 8A1, 12A9, 11E7, 6C11, 13G8, 14B6, 10G4, 5A5, 6B12, 10E11, 3F3, 2C9, 7F5, 11F6, and 9C6). In earlier studies, we found that hybridoma fusions carried out with splenocytes from mice immunized with mouse brain homogenates resulted predominantly in MAbs that were reactive with the viral nucleocapsid protein. To produce immunogens enriched in viral glycoproteins, cultures of SW-13 cells in 34 T-150 flasks were infected with the IbAr10200 strain of CCHFV and incubated for 36 h at 37°C. Medium fractions were removed, clarified at 10,000 x g, and centrifuged in an SW28 rotor (Beckman) to pellet virus particles (25,000 rpm for 3 h). Virus pellets were lysed in immunoprecipitation buffer (50 mM Tris-HCl, pH 8.0, containing 150 mM NaCl, 2% Triton X-100, 1% sodium deoxycholate, and protease inhibitors) and subsequently centrifuged to equilibrium in CsCl gradients. This resulted in a separation of the viral nucleocapsid (density = 1.3 g/ml) from the viral glycoproteins which banded in a broad band at lower densities. Infected-cell monolayers were similarly lysed in immunoprecipitation buffer, clarified at 10,000 x g, and centrifuged in an SW41 rotor (Beckman) to sediment the viral nucleocapsid from the solubilized viral glycoproteins (40,000 rpm for 4 h). Fractions enriched for viral glycoproteins from both cell and virion samples were then collected on solid-phase immunoadsorbents prepared by saturating protein A-Sepharose with anti-CCHFV hyperimmune mouse ascitic fluid (D. Watts, U.S. Army Medical Research Institute for Infectious Diseases). Immunoadsorbents were then exhaustively washed with phosphate-buffered saline (PBS), emulsified directly in complete or incomplete Freund's adjuvant, and inoculated subcutaneously and intraperitoneally into mice as described below. With this procedure, the CCHFV glycoproteins were enriched but were not pure. Importantly, the viral nucleocapsid was not detectable under these conditions, and using these immunogens, we were able to prepare a large number of CCHFV glycoprotein-specific MAbs.
MAbs were produced essentially as described previously (28). Briefly, BALB/c mice were twice immunized intraperitoneally with 100 μl of antigen preparations emulsified in Freund's complete adjuvant for primary immunization and in Freund's incomplete adjuvant for secondary immunization. Mice were euthanized 3 to 5 days after a third immunization, and splenocytes were fused with Sp2/0-Ag14 myeloma cells. Hybridoma cultures were incubated at 37°C with several changes of hypoxanthine-aminopterin-thymidine medium, and the supernatant fluids were screened by immunofluorescence (IF) assays against CCHFV-infected Vero cells and by enzyme-linked immunosorbent assays (ELISA) with viral antigen from suckling mouse brain homogenates or gradient-purified virus preparations. Antigenic specificity was initially determined by immunoprecipitation from [35S]methionine-labeled infected-cell lysates (as described below) and subsequently by IF microscopy. Positive-antibody-producing cells were cloned by limiting dilution and then expanded. The immunoglobulin G subclasses of the resulting MAbs were determined by indirect ELISA analysis using hybridoma supernatants. The ELISA was developed using immunoglobulin G subclass-specific immunoglobulins (Miles Laboratories) by following the manufacturer's instructions.
Neutralization assays. Eighty percent plaque reduction neutralization (PRN-80) tests were carried out on SW-13 cell monolayers. Twofold serial dilutions of the MAbs were mixed with 200 PFU of the CCHFV IbAr10200 strain and incubated for 1 h at 37°C. Confluent monolayers of SW-13 cells in six-well plates were incubated with the virus-antibody mixture for 1 h at 37°C. The inocula were removed, and 1 ml of overlay consisting of 1 part double-strength DMEM with 5% fetal bovine serum and 1 part low-gelling-temperature agarose (Bio-Rad Laboratories, Richmond, CA) in distilled water was added. After incubation at 35°C in a sealed chamber for 2 to 5 days, the plaques were visualized by neutral red staining.
Protection studies. To evaluate the protective activities of MAbs directed against the CCHFV glycoproteins in an animal model, suckling mice, which are susceptible to infection with CCHFV (24, 50), were challenged with live virus before or after passive immunization with the CCHFV-specific MAbs. Two- to 3-day-old suckling mice were inoculated in groups of five to eight by intraperitoneal injection with 50 μl of undiluted ascitic fluid containing the different MAbs. The ascitic fluids were administrated 24 h before or after the inoculation of 100 50% lethal-dose units of the CCHFV strain IbAr10200. Ascitic fluid from Sp2/0 cells that did not contain virus-specific antibodies was used as a negative control.
Construction of CCHFV glycoprotein clones. The pCAGGS-M clone was created by cloning the entire M segment of IbAr10200 into the NruI and XhoI sites found in the plasmid pCAGGS-MCSII (41). The M segment was digested using the unique restriction enzyme sites SnaBI and SalI located in the untranslated regions of the gene. This clone was used as a template for the generation of a panel of constructs used to map functional regions on CCHFV glycoproteins (Fig. 1). Primers were synthesized according to the published sequence for IbAr10200 (51), and standard PCR technology was performed for cloning into the pcDNA3.1D/V5-His-TOPO vector (Invitrogen, Carlsbad, CA). The 5'-end primers included the CACC sequence at the 5' end and the start codon to allow for directional cloning. The 3'-end primers did not possess a stop codon to allow the inclusion of the V5 and His epitope tags at the C terminus of the protein. The cloning was performed as described by the manufacturer (Invitrogen), and all constructs were sequenced. All primer sequences are available upon request.
Protein analysis. To analyze protein expression, we transfected HEK 293T cells using Lipofectamine 2000 (Invitrogen). After 24 h, cell extracts were prepared in 50 mM Tris-HCl, pH 7.4, 5 mM EDTA, 1% Triton X-100, and Complete protease inhibitor cocktail (Roche Applied Sciences, Indianapolis, IN). Cell lysates were incubated at 4°C for 3 min and then centrifuged at 10,000 x g for 10 min. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed using 4 to 15% Tris-HCl gels (Bio-Rad, Hercules, CA), followed by Western blot analysis with mouse anti-V5 (Invitrogen) as the primary antibody and sheep anti-mouse horseradish peroxidase conjugate as the secondary antibody (Amersham Pharmacia, Buckinghamshire, United Kingdom), followed by visualization with ECL-Plus Western blotting detection reagents (Bioscience, Piscataway, NJ).
IF microscopy. Localization of the CCHFV glycoproteins by indirect IF was performed as described previously (38). HeLa cells grown to 50% confluence on glass coverslips were infected with the recombinant vaccinia virus vTF1.1 (1), followed 40 min later by transfection with the different pcDNA3.1D/V5-His constructs (Fig. 1). When proteins were expressed with pCAGGS constructs, vTF1.1 infection was not utilized, since protein expression was obtained from the chicken beta-actin promoter. At 24 h posttransfection, the cells were fixed with 2% (wt/vol) paraformaldehyde in PBS, permeabilized with 0.5% Triton X-100, and stained with a GN- or GC-specific MAb ascites fluid (diluted 1:250) or with mouse anti-V5 MAb (diluted 1:500) (Invitrogen) in PBS containing 0.5 mM MgCl2 and 4% fetal bovine serum. In addition, TGN46, a sheep antibody specific for a heavily glycosylated protein localized primarily in the trans-Golgi network, was included as a marker for Golgi localization (Serotec, Oxford, United Kingdom). Then, cells were washed with PBS and incubated for 1 h with the secondary antibodies conjugated to Alexa Flour 488 (goat anti-mouse) and Alexa Flour 594 (goat anti-sheep) (Molecular Probes, Eugene, OR) diluted 1:500 in PBS-4% fetal bovine serum. Finally, cells were washed in PBS, mounted in Fluoromount-G (Southern Biotechnology Associates, Birmingham, AL), and examined on a Nikon E600 microscope at a magnification of x60 utilizing UV illumination.
Endoglycosidase treatment. The method of Trimble and Maley (57) was slightly modified for digesting glycosylated CCHFV glycoproteins with endo-?-N-acetylglucosaminidase H (endo-H). Cell lysates from transfected HEK 293T cells were boiled for 5 min in a 50 mM sodium citrate buffer (pH 5.5) containing 1% SDS and 200 μg/ml of phenylmethylsulfonyl fluoride. After cooling, the samples were supplemented with an equal volume of 0.02 U of endo-H (New England BioLabs Inc., Beverly, MA) in a 50 mM sodium citrate buffer (pH 5.5) or the same buffer without the enzyme before incubation at 37°C for 20 h. After this incubation, the samples were analyzed by SDS-PAGE. In addition, the glycoproteins were treated with peptide-N-glycosidase F (PNGase F) (New England Biolabs). A similar protocol was followed for PNGase F treatment, but the digestion was performed in 100 mM sodium phosphate, pH 7.5, with 0.75% NP-40, 0.1% SDS, and 50 mM ?-mercaptoethanol.
Immunoprecipitations from infected cells. Immunoprecipitation assays were performed as previously described (38). Briefly, SW-13 cell monolayers (1 x106 cells) were infected with the CCHFV strain IbAr10200 at a multiplicity of infection of 5. Infected SW-13 cells were placed in methionine-free DMEM at 28 h postinfection for 30 min, and then 30 μCi of [35S]methionine (Trans-35S-label; ICN, Irvine, CA) was added. After 1 h of labeling, 50 μg of unlabeled methionine per ml was added. Infected cells were collected 24 h postlabeling or 20 h posttransfection and lysed with 0.5 ml of lysis buffer containing 50 mM Tris, pH 7.4, 150 mM NaCl, 1% CHAPSO {3-[(3-cholamidopropyl)-dimethylammonio]-2-hydroxy-1-propanesulfonate}, 5 mM EDTA, and a Complete protease inhibitor mixture (Roche). Precleared lysates were incubated for 1 h at 4°C with the MAbs. Glycoproteins were immunoprecipitated using 7 μg of the MAbs generated in this study. The MAbs were previously coupled to protein A/G Plus-agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoprecipitated proteins were eluted at 100°C for 5 min with 20 μl of sample buffer (0.08 M Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 5% ?-mercaptoethanol, 0.005% bromophenol blue). The immunoprecipitated proteins were detected after separation by 10% SDS-PAGE utilizing a personal densitometer (model 860; Molecular Dynamics, Inc., Sunnyvale, CA), and the data were analyzed with the ImageQuant NT software.
Immunoprecipitation from transfected cells. For the analysis of the different CCHFV glycoprotein constructs, HEK 293T cell monolayers (1 x106) were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. The immunoprecipitation protocol utilized was in general the same as that described above for infected cells. The V5 epitope tag present at the C terminus of the recombinant proteins was detected using 7 μg of a MAb directed against the V5 tag (Invitrogen). Samples from transfected cells were analyzed by Western blotting as described above, using a rabbit anti-V5 tag hyperimmune serum diluted 1:5,000 (Sigma).
RESULTS
CCHFV M segment processing. Glycoproteins of the nairovirus group of bunyaviruses are processed in a distinctive and complex manner (51, 59). The M segment of CCHFV encodes a polyprotein that undergoes proteolytic processing to yield a 140-kDa Pre-GN (termed GN-FL in Fig. 1) and an 85-kDa precursor of GC (Pre-GC; termed GC-FL in Fig. 1). Recently, it was shown that the N terminus of the mature GN protein contains both a mucin-like and a P35 connector domain (Fig. 1) that are removed by specific cleavage with protease SKI-1, resulting in the generation of mature GN (GN-RRLL in Fig. 1) (59). However, how GN and GC are cleaved from each other is not clear, though it is thought that the N terminus of the mature GC protein is generated by cleavage at a conserved RKPL site (59). Finally, cleavage between the mucin domain and the P35 region may be mediated by furin, which is consistent with the presence of a conserved RSKR sequence between these domains (Fig. 1) (59).
MAbs to CCHFV. To assist in our studies of CCHFV glycoprotein processing and localization, and to determine if antibodies can confer protection to CCHFV, we produced a panel of MAbs from mice immunized with a variety of CCHFV antigens as described in Materials and Methods. MAbs to CCHFV were identified by ELISA, after which both immunoprecipitation and immunofluorescence studies were done to identify the glycoprotein subunit to which each MAb bound. Only the 26 MAbs whose specificity could be clearly identified by both techniques are included in this study. For the immunofluorescence studies, HeLa cells were transfected with pcDNA3.1D/V5-His vectors expressing either GN (GN-FL, GN-RSKR, or GN-RRLL) or GC (GC-FL) and were infected with vTF1.1, a recombinant vaccinia virus that expresses the T7 RNA polymerase (1), in order to achieve high levels of protein expression. The cells were then fixed and processed for IF microscopy. All GN antibodies recognized GN-FL but most did not recognize GN-RSKR or GN-RRLL, with the exception of MAb 8F10, which recognized all the GN constructs. Since neither GN-RSKR nor GN-RRLL was correctly transported to the Golgi apparatus (see below), the failure of most GN MAbs to recognize these constructs could be due to protein misfolding rather than the loss of specific epitopes.
From the panel of 26 MAbs, we selected six directed against GN and six directed against GC for the detailed study of CCHFV glycoprotein processing and targeting. When analyzed by IF and when utilizing HeLa cells transfected with the entire M segment of CCHFV strain IbAr10200, all anti-GC MAbs gave similar patterns, with GC being localized to both the endoplasmic reticulum (ER) and Golgi regions as judged by colocalization with Golgi and ER markers (Fig. 2 and data not shown). The six anti-GN MAbs recognized GN only in the Golgi apparatus, suggesting that they recognize epitopes that are either formed after the protein reaches the Golgi apparatus or formed shortly before exit from the ER (Fig. 2). None of the MAbs detected GN or GC protein on the surfaces of unpermeabilized cells (data not shown). Our results indicate that the CCHFV glycoproteins are not delivered to the cell surface in appreciable quantities, which is consistent with studies that have shown that other Bunyaviridae bud into the Golgi apparatus and that their glycoproteins are targeted to this organelle and not delivered to the cell surface (13, 18, 29, 34, 37).
To determine if the MAbs could recognize GN and GC in different experimental contexts, both Western blot analysis and immunoprecipitation assay were performed. None of the MAbs recognized CCHFV proteins under fully denaturing conditions by Western blot analysis. However, in the absence of boiling, MAb 11E7 could recognize GC obtained from transfected cells or from virus-infected cells by Western blot analysis (Fig. 3). The failure of most of the MAbs to recognize their antigens by Western blot analysis indicates that they bind to conformation-dependent determinants that are lost upon protein denaturation.
The MAbs were also used to immunoprecipitate GN and GC from lysates of cells infected with the IbAr10200 strain of CCHFV. All of the MAbs were able to immunoprecipitate the protein subunits to which they bound, which is consistent with their ability to recognize their epitopes by IF. In addition, four of the six GC MAbs were able to immunoprecipitate both processed GC (around 75 kDa) and its precursor protein (around 82 kDa), while all of the GN MAbs analyzed were able to immunoprecipitate processed GN (35 kDa) and its precursor (140 kDa), which contains both the mucin and the P35 connector domains (data not shown).
Virus neutralization and protection studies. To analyze the in vitro neutralization activities of the panel of MAbs directed against the CCHFV glycoproteins, we performed plaque reduction assays. Twofold serial dilutions of each MAb ascites fluid sample were incubated with 200 PFU of the IbAr10200 strain of CCHFV for 1 h at 37°C prior to addition to confluent SW-13 cells. Plaques were counted 3 to 5 days later, and PRN-80 titers were calculated. None of the MAbs directed against GN exhibited neutralizing activity in this assay (Fig. 4B), though many of the MAbs directed against GC neutralized CCHFV in vitro (Fig. 4A).
Antibodies have been shown to be effective pre- and postexposure to prophylaxis treatments for a number of viruses, including cytomegalovirus (9, 10) and respiratory syncytial virus (17, 32). Convalescent-phase serum has been shown to be of benefit to individuals acutely infected with CCHFV (57). Therefore, we tested the CCHFV MAbs for their ability to protect suckling mice challenged with the virus. The MAbs were individually administered by passive immunization to 2- to 3-day-old suckling mice either 24 h before or 24 h after challenge with 100 50% infective-dose units of the IbAr10200 strain of CCHFV (Fig. 4). Protection was registered as the percentage of animals that survived challenge with the live virus (Fig. 4). In general, anti-GC MAbs that were capable of efficient neutralization in vitro protected mice to an appreciable degree when applied before and, to a lesser extent, after virus challenge (Fig. 4). In addition, anti-GC MAbs that did not neutralize CCHFV infection of SW-13 cells afforded partial protection to mice from CCHFV provided that the MAbs were administered 24 h before viral challenge. When these MAbs were administered 24 h after virus challenge, protection was usually not observed. This suggests that these MAbs do possess some neutralizing activity that was not detected in our in vitro assay or that other antibody-based effector mechanisms, such as antibody-dependent cell-mediated toxicity or complement-mediated cell lysis, function in this context. Likewise, many of the anti-GN MAbs conferred significant protection to CCHFV challenge, even when applied 24 h after virus challenge and even though they did not prevent virus infection of SW-13 cells in vitro (Fig. 4). These results show that there is an imperfect relationship between in vitro neutralization and in vivo protective ability, at least under the assay conditions used here, and that the ability of an antibody to neutralize CCHFV may depend in part on host factors, as has been observed for La Crosse virus (LACV), another bunyavirus (25-27).
GN contains a Golgi targeting signal. To analyze the contribution of each CCHFV glycoprotein to Golgi localization, we expressed each protein independently in HeLa cells. IF analyses showed that, while GN remained in the Golgi apparatus if expressed alone, GC localized entirely in the ER in the absence of GN (Fig. 5Aa and -b). When the proteins were expressed together from independent constructs, both GC and GN were localized in the Golgi apparatus (Fig. 5Ac). The restoration of GC Golgi localization in the presence of GN suggests that GN possesses a Golgi localization signal and that GC localizes to the Golgi apparatus through its interaction with GN (Fig. 5A).
We further analyzed the localization of CCHFV glycoproteins when expressed independently or together by investigating their N-linked glycosylation. The GN ectodomain of the IbAr10200 strain of CCHFV contains one predicted N-linked glycosylation site, while the GC ectodomain contains three sites. To determine whether the CCHFV glycoproteins are glycosylated and, if so, modified by medial Golgi enzymes, we used PNGase F, which removes all N-linked carbohydrate chains, or endo-H, which removes immature carbohydrate chains. When GC-FL was expressed alone, digestion with either endo-H or PNGase F caused the protein to migrate faster (Fig. 5B). When GN-FL was expressed alone, it too migrated faster following digestion with either endo-H or PNGase F. To determine if the shift observed following glycosidase treatment of GN-FL was the result of a loss of glycans on the mucin or P35 regions only, we treated the GN-RRLL protein that lacks both the mucin and the P35 domains with endo-H. The GN-RRLL protein also migrated faster after endo-H treatment, indicating the presence of N-linked carbohydrate chains on the ectodomain of the mature GN as well. Similar results were obtained when the proteins were coexpressed (Fig. 5B). Thus, both proteins are N glycosylated, but neither appears to be processed in the medial Golgi apparatus when expressed alone or together.
Contribution of GN N-terminal domains to protein localization. The CCHFV GN protein is unusual in that it has two N-terminal domains, a mucin-like domain and the P35 domain, that appear to be cleaved from the GN precursor in a posttranslational fashion (51). The function of these regions is unknown. A similar mucin-like domain has been found to be associated with the glycoprotein of Ebola virus and has been shown to induce cell rounding and detachment in vitro and possibly to be involved in the pathogenicity of the virus (56). To analyze the involvement of the mucin-like domain and P35 region in GN cellular localization, we deleted these regions from constructs that contained only the GN portion of the protein in order to generate a construct without the mucin-like domain (GN-RSKR) and a construct without both the mucin-like domain and the P35 region (GN-RRLL) (Fig. 1). In both cases, a signal sequence was introduced at the N terminus to ensure proper targeting to the ER. Upon expression in HeLa cells, IF analysis using a MAb to the GN ectodomain showed that deletion of just the mucin-like domain (GN-RSKR) resulted in a protein that localized to the Golgi apparatus in a manner similar to full-length GN-FL (Fig. 6A). In contrast, the GN-RRLL protein was not present in either the ER or the Golgi apparatus but was distributed in a punctuate pattern that resembled aggresomes, suggesting that it misfolds (Fig. 6B). Thus, the mucin-like domain is dispensable for Golgi targeting, while removal of the P35 connector region affected GN localization and also led to enhanced degradation. Interestingly, localization of GN-RRLL to the Golgi complex was recovered when the GN-RRLL protein was coexpressed with GC, suggesting that GN-GC interactions may promote the correct folding and transport of the proteins (Fig. 6C).
Mapping of the Golgi localization signal. In all Bunyaviridae glycoproteins that have been examined, the Golgi localization signal has been localized to GN, generally in the cytoplasmic tail (CT) or transmembrane domain (TM) of the protein (3-5, 12, 14, 33, 34, 47, 49). The C-terminal domain of GN contains a stretch of predicted TMs and cytoplasmic loops between the first amino acid of GC and the predicted C terminus of GN (59). The function of this unusual region on glycoprotein processing or any other step of the viral replication and cell cycle is unknown. To analyze the role of this region in GN localization, we deleted two (GN 2TM) or three (GN 1TM) of the predicted four TMs at the GN C terminus (Fig. 1). HeLa cells were transfected with the constructs and analyzed by IF (Fig. 6). We found that all of the C-terminally truncated GN constructs localized to the Golgi apparatus (Fig. 6D and E). These results suggest that the Golgi localization signal is not located in this region.
Next, we designed soluble constructs that lacked the transmembrane and cytoplasmic domains of GN (sGN) or GC (sGC) or that lacked only the CT of GN (GN-no CT) (Fig. 1). When these constructs were expressed using the vaccinia virus T7 polymerase system in HeLa cells, both sGN and GN-no CT, although also present in the ER, localized to the Golgi apparatus when analyzed by IF microscopy using MAbs against GN (Fig. 6F and G). In contrast, sGC was restricted to the ER, just like full-length GC (Fig. 6H). When sGC and sGN were coexpressed, sGC was then targeted to the Golgi apparatus (Fig. 6I), indicating that the interaction of GC and GN occurs through their ectodomains, that the proteins can fold correctly when in their soluble forms, and that Golgi targeting information resides within the ectodomain of GN.
DISCUSSION
Relatively little is known about the mechanisms by which bunyaviruses enter cells or how infection can be prevented by neutralizing antibodies, and MAbs that block CCHFV infection have not been described. Therefore, we developed a panel of MAbs to assist in our studies on CCHFV glycoprotein biology and to begin characterizing the antigenic structures of GN and GC. The large majority of MAbs bound to conformation-dependent epitopes in GN or GC. A number of MAbs against GC, but not against GN, were able to neutralize virus infection of SW-13 cells in vitro, suggesting that GC plays an important role in virus entry. Similarly, MAbs directed against the GC glycoproteins of LACV can inhibit virus infection, with some evidence indicating that this is due to a reduction in virus binding to the cell surface (19, 21, 22, 26). However, some antibodies against GN can neutralize LACV infection in an insect cell line, though not in a mammalian cell line, suggesting that virus neutralization can be dependent on the cell type being infected and that virus entry mechanisms may differ between invertebrates and vertebrates (30). Consistent with this, proteolytic degradation of GC with trypsin or pronase virtually eliminates the ability of LACV to bind to vertebrate, but not invertebrate, cell lines (30, 46).
Our results suggest that CCHFV neutralization mechanisms may be complex and context dependent. In general, MAbs directed against GN were more effective at protecting mice from a lethal CCHFV challenge than were MAbs to GC when administrated either 24 h before or after infection, even though GN MAbs did not neutralize infection of SW-13 cells in vitro. In addition, not all GC MAbs that neutralized CCHFV infection in vitro conferred high levels of protection in vivo, especially if administered after infection, as seen for the 8A1 and 1H6 MAbs. Thus, there was not a strict relationship between in vitro neutralization and in vivo protection. As noted above, these results resemble observations on the differential roles of LACV GN and GC glycoproteins in viral neutralization and protection. A soluble version of the GC La Crosse glycoprotein is sufficient to block virus infection in mammalian cells, while antibodies against GN neutralize infection in a mosquito cell line but not in a vertebrate cell line (31). The mechanisms that account for the differential inhibition of CCHFV infection are not currently clear. We assume that the conformational changes undergone by GN and GC to elicit membrane fusion and virus entry will likely be the same regardless of the cell type being infected and that these changes are likely to be induced by acid pH following endocytosis of the virus, as has been documented for La Crosse and Hantaan viruses (7, 20). However, it remains to be determined if CCHFV uses different cell surface receptors and attachment factors on different cell types. Without knowing the identity of any potential CCHFV receptors, it will be interesting to examine the ability of MAbs to neutralize CCHFV on cell lines derived from different species, including ticks.
One of the hallmarks of the Bunyaviridae family is that their glycoproteins are targeted to the Golgi apparatus from which they bud (48, 52). Based on this fact, a number of studies have sought to identify the signals responsible for targeting GN and GC to the Golgi apparatus for a number of bunyaviruses (2, 34, 36, 40, 47, 49). Generally, it has been found that the Golgi retention signals reside within the TM and/or CT of the glycoprotein closest to the N terminus of the glycoprotein precursor (18, 34, 35). However, no consensus Golgi localization motif appears to be shared among the glycoproteins of these viruses, and Nairovirus glycoprotein targeting signals and antigenic structures have not been analyzed so far.
We found that the CCHFV glycoproteins were targeted to the Golgi apparatus, as determined by IF microscopy. Since both GN and GC were sensitive to endo-H treatment, it is likely that the proteins are targeted to an early Golgi compartment. Similarly, hantavirus glycoproteins, although present in the Golgi apparatus, remain sensitive to endo-H treatment (49). When GN and GC were expressed together, GN was localized to the Golgi complex while GC was found in both the Golgi apparatus and the ER. It is obvious that GN must also be present in the ER, but our conformation-dependent MAbs either do not recognize GN in the ER or bind to epitopes that form just prior to exiting from this organelle. It is also possible that GN is transported more quickly from the ER than GC and, at steady state, is below our limit of detection in the ER. Indeed, some studies on the biosynthesis of Uukuniemi virus GN and GC proteins showed previously that GN is transported faster than GC from the ER to the site of virus budding at the Golgi complex (29). The apparent difference in transport kinetics is due to the fact that GN folds and is transported from the ER to the Golgi apparatus 30 to 45 min faster than GC (4).
The ability of CCHFV GN to localize to the Golgi apparatus when expressed independently of GC indicates that GN contains a Golgi targeting or retention motif. Since GC is restricted to the ER in the absence of GN, we conclude that its transport to the Golgi apparatus is dependent upon GN and likely results from GN-GC oligomerization. Whether GC fails to fold correctly in the absence of GN, contains an ER retention signal that is masked by GN association, or lacks a positive transport signal cannot be determined at present. Our results are in agreement with published data about other Bunyaviridae glycoproteins (3, 35, 36, 40, 47, 49) which localize to the Golgi apparatus in the absence of any other viral proteins (11, 14, 34, 47). Although for most of the Bunyaviridae glycoproteins analyzed to date, the Golgi targeting signal is contained in one of the glycoproteins, for the Hantaan virus, a member of the Hantavirus group, both glycoproteins are required to achieve Golgi apparatus targeting (49). Our results with CCHFV indicate that Golgi targeting information resides largely in the ectodomain of the GN subunit, since a soluble version of GN was largely restricted to the Golgi apparatus. However, small amounts of this protein were secreted from cells, indicating that the TM of GN may also play a role in Golgi retention. It is also evident that the ectodomains of GN and GC interact with each other and that targeting of GC to the Golgi apparatus is dependent upon its association with GN, with ectodomain interactions being important.
In summary, although the CCHFV glycoproteins are unique in many aspects with respect to the glycoproteins from other members of the Bunyaviridae family, there are some similarities with regard to Golgi targeting and the glycoprotein subunits to which neutralizing antibodies are directed. Our studies indicate that CCHFV neutralization is likely to be context dependent and that more in-depth studies of various cell lines and animal models will be needed to characterize neutralization mechanisms and to identify antibodies that could be used therapeutically. Identification of regions on the CCHFV glycoproteins involved in viral neutralization, protection, and processing will contribute to our understanding of the tropism and pathogenesis of this emerging viral pathogen.
ACKNOWLEDGMENTS
We thank Nicolette Pesik, Randall Bethke, and Loreen Hodgson for superb technical assistance on the production and characterization of the MAbs. We also thank Aura Garrison and Donald Pijak for expert technical assistance.
This work was supported in part by National Institutes of Health training grant NIH T32 AI055400 and grant PR033269 from the Department of Defense Peer Reviewed Medical Research Program (PRMRP) of the Office of the Congressionally Directed Medical Research Programs.
REFERENCES
Alexander, W. A., B. Moss, and T. R. Fuerst. 1992. Regulated expression of foreign genes in vaccinia virus under the control of bacteriophage T7 RNA polymerase and the Escherichia coli lac repressor. J. Virol. 66:2934-2942.
Anderson, G. W., Jr., and J. F. Smith. 1987. Immunoelectron microscopy of Rift Valley fever viral morphogenesis in primary rat hepatocytes. Virology 161:91-100.
Andersson, A. M., L. Melin, A. Bean, and R. F. Pettersson. 1997. A retention signal necessary and sufficient for Golgi localization maps to the cytoplasmic tail of a Bunyaviridae (Uukuniemi virus) membrane glycoprotein. J. Virol. 71:4717-4727.
Andersson, A. M., L. Melin, R. Persson, E. Raschperger, L. Wikstr?m, and R. F. Pettersson. 1997. Processing and membrane topology of the spike proteins G1 and G2 of Uukuniemi virus. J. Virol. 71:218-225.
Andersson, A. M., and R. F. Pettersson. 1998. Targeting of a short peptide derived from the cytoplasmic tail of the G1 membrane glycoprotein of Uukuniemi virus (Bunyaviridae) to the Golgi complex. J. Virol. 72:9585-9596.
Archer, D. B., D. J. Jeenes, D. A. MacKenzie, G. Brightwell, N. Lambert, G. Lowe, S. E. Radford, and C. M. Dobson. 1990. Hen egg white lysozyme expressed in, and secreted from, Aspergillus niger is correctly processed and folded. Biotechnology (New York) 8:741-745.
Arikawa, J., I. Takashima, and N. Hashimoto. 1985. Cell fusion by haemorrhagic fever with renal syndrome (HFRS) viruses and its application for titration of virus infectivity and neutralizing antibody. Arch. Virol. 86:303-313.
Arikawa, J., J. S. Yao, K. Yoshimatsu, I. Takashima, and N. Hashimoto. 1992. Protective role of antigenic sites on the envelope protein of Hantaan virus defined by monoclonal antibodies. Arch. Virol. 126:271-281.
Aulitzky, W. E., T. F. Schulz, H. Tilg, D. Niederwieser, K. Larcher, L. Ostberg, M. Scriba, J. Martindale, A. C. Stern, P. Grass, et al. 1991. Human monoclonal antibodies neutralizing cytomegalovirus (CMV) for prophylaxis of CMV disease: report of a phase I trial in bone marrow transplant recipients. J. Infect. Dis. 163:1344-1347.
Aulitzky, W. E., H. Tilg, D. Niederwieser, M. Hackl, B. Meister, and C. Huber. 1988. Ganciclovir and hyperimmunoglobulin for treating cytomegalovirus infection in bone marrow transplant recipients. J. Infect. Dis. 158:488-489.
Bupp, K., K. Stillmock, and F. Gonzalez-Scarano. 1996. Analysis of the intracellular transport properties of recombinant La Crosse virus glycoproteins. Virology 220:485-490.
Chen, S.-Y., and R. W. Compans. 1991. Oligomerization, transport, and Golgi retention of Punta Toro virus glycoproteins. J. Virol. 65:5902-5909.
Chen, S.-Y., Y. Matsuoka, and R. W. Compans. 1991. Assembly and polarized release of Punta Toro virus and effects of brefeldin A. J. Virol. 65:1427-1439.
Chen, S. Y., Y. Matsuoka, and R. W. Compans. 1991. Golgi complex localization of the Punta Toro virus G2 protein requires its association with the G1 protein. Virology 183:351-365.
Chumakov, M. P., A. M. Butenko, N. V. Shalunova, L. I. Mart'ianova, S. E. Smirnova, N. Bashkirtsev, T. I. Zavodova, S. G. Rubin, E. A. Tkachenko, V. Karmysheva, V. N. Reingol'd, G. V. Popov, and A. P. Savinov. 1968. New data on the viral agent of Crimean hemorrhagic fever. Vopr. Virusol. 13:377. (In Russian.)
Chumakov, M. P., S. E. Smirnova, and E. A. Tkachenko. 1970. Relationship between strains of Crimean haemorrhagic fever and Congo viruses. Acta Virol. 14:82-85.
Englund, J. A., P. A. Piedra, and E. Whimbey. 1997. Prevention and treatment of respiratory syncytial virus and parainfluenza viruses in immunocompromised patients. Am. J. Med. 102:61-70, 75-76.
Gerrard, S. R., and S. T. Nichol. 2002. Characterization of the Golgi retention motif of Rift Valley fever virus GN glycoprotein. J. Virol. 76:12200-12210.
Gonzalez-Scarano, F. 1985. La Crosse virus G1 glycoprotein undergoes a conformational change at the pH of fusion. Virology 140:209-216.
Gonzalez-Scarano, F., R. S. Janssen, J. A. Najjar, N. Pobjecky, and N. Nathanson. 1985. An avirulent G1 glycoprotein variant of La Crosse bunyavirus with defective fusion function. J. Virol. 54:757-763.
Grady, L. J., M. L. Sanders, and W. P. Campbell. 1983. Evidence for three separate antigenic sites on the G1 protein of La Crosse virus. Virology 126:395-397.
Grady, L. J., S. Srihongse, M. A. Grayson, and R. Deibel. 1983. Monoclonal antibodies against La Crosse virus. J. Gen. Virol. 64:1699-1704.
Hoogstraal, H. 1979. The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe, and Africa. J. Med. Entomol. 15:307-417.
Huggins, J. W. 1989. Prospects for treatment of viral hemorrhagic fevers with ribavirin, a broad-spectrum antiviral drug. Rev. Infect. Dis. 11 (Suppl. 4):S750-S761.
Kingsford, L. 1984. Enhanced neutralization of La Crosse virus by the binding of specific pairs of monoclonal antibodies to the G1 glycoprotein. Virology 136:265-273.
Kingsford, L., K. H. Boucquey, and T. P. Cardoso. 1991. Effects of specific monoclonal antibodies on La Crosse virus neutralization: aggregation, inactivation by Fab fragments, and inhibition of attachment to baby hamster kidney cells. Virology 180:591-601.
Kingsford, L., L. D. Ishizawa, and D. W. Hill. 1983. Biological activities of monoclonal antibodies reactive with antigenic sites mapped on the G1 glycoprotein of La Crosse virus. Virology 129:443-455.
Kohler, G., S. C. Howe, and C. Milstein. 1976. Fusion between immunoglobulin-secreting and nonsecreting myeloma cell lines. Eur. J. Immunol. 6:292-295.
Kuismanen, E., B. B?ng, M. Hurme, and R. F. Pettersson. 1984. Uukuniemi virus maturation: immunofluorescence microscopy with monoclonal glycoprotein-specific antibodies. J. Virol. 51:137-146.
Ludwig, G. V., B. A. Israel, B. M. Christensen, T. M. Yuill, and K. T. Schultz. 1991. Monoclonal antibodies directed against the envelope glycoproteins of La Crosse virus. Microb. Pathog. 11:411-421.
Ludwig, G. V., B. A. Israel, B. M. Christensen, T. M. Yuill, and K. T. Schultz. 1991. Role of La Crosse virus glycoproteins in attachment of virus to host cells. Virology 181:564-571.
Malley, R., J. DeVincenzo, O. Ramilo, P. H. Dennehy, H. C. Meissner, W. C. Gruber, P. J. Sanchez, H. Jafri, J. Balsley, D. Carlin, S. Buckingham, L. Vernacchio, and D. M. Ambrosino. 1998. Reduction of respiratory syncytial virus (RSV) in tracheal aspirates in intubated infants by use of humanized monoclonal antibody to RSV F protein. J. Infect. Dis. 178:1555-1561.
Matsuoka, Y., S. Y. Chen, and R. W. Compans. 1991. Bunyavirus protein transport and assembly. Curr. Top. Microbiol. Immunol. 169:161-179.
Matsuoka, Y., S. Y. Chen, and R. W. Compans. 1994. A signal for Golgi retention in the bunyavirus G1 glycoprotein. J. Biol. Chem. 269:22565-22573.
Matsuoka, Y., S. Y. Chen, C. E. Holland, and R. W. Compans. 1996. Molecular determinants of Golgi retention in the Punta Toro virus G1 protein. Arch. Biochem. Biophys. 336:184-189.
Matsuoka, Y., T. Ihara, D. H. Bishop, and R. W. Compans. 1988. Intracellular accumulation of Punta Toro virus glycoproteins expressed from cloned cDNA. Virology 167:251-260.
Melin, L., R. Persson, A. Andersson, A. Bergstr?m, R. R?nnholm, and R. F. Pettersson. 1995. The membrane glycoprotein G1 of Uukuniemi virus contains a signal for localization to the Golgi complex. Virus Res. 36:49-66.
Morais, V. A., A. S. Crystal, D. S. Pijak, D. Carlin, J. Costa, V. M. Lee, and R. W. Doms. 2003. The transmembrane domain region of nicastrin mediates direct interactions with APH-1 and the gamma-secretase complex. J. Biol. Chem. 278:43284-43291.
Nabel, G. J. 1999. Surviving Ebola virus infection. Nat. Med. 5:373-374.
Nakitare, G. W., and R. M. Elliott. 1993. Expression of the Bunyamwera virus M genome segment and intracellular localization of NSm. Virology 195:511-520.
Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108:193-199.
Olaleye, O. D., O. Tomori, and H. Schmitz. 1996. Rift Valley fever in Nigeria: infections in domestic animals. Rev. Sci. Tech. Off. Int. Epizoot. 15:937-946.
Onishchenko, G. G., V. I. Efremenko, N. G. Kovalev, N. A. Shibkov, M. Fedorov, N. Zhilina, G. N. Fedosova, G. V. Sysoliatina, M. Evchenko, A. P. Beier, V. A. Popov, K. V. Shenetts, L. N. Marchukova, E. N. Afanas'ev, and N. V. Vasilenko. 2001. Specific epidemiologic features of Crimean haemorrhagic fever in Stavropol' region in 1999-2000. Zh. Mikrobiol. Epidemiol. Immunobiol. Nov., Dec.(Suppl. 6):86-89. (In Russian.)
Onishchenko, G. G., M. Lomov, V. I. Markov, V. A. Merkulov, A. N. Mironov, I. V. Borisevich, A. A. Makhlai, N. T. Vasil'ev, S. O. Vodop'ianov, N. G. Tikhonov, T. P. Pashanina, V. V. Manankov, V. P. Smelianskii, S. T. Savchenko, and S. E. Smirnova. 2000. The laboratory diagnosis of an outbreak of hemorrhagic fever at Oblivskaya village, Rostov Province: proof of the etiological role of the Crimean-Congo hemorrhagic fever virus. Zh. Mikrobiol. Epidemiol. Immunobiol. Mar., Apr.32-36. (In Russian.)
Onishchenko, G. G., V. I. Markov, V. A. Merkulov, N. T. Vasil'ev, A. M. Berezhnoi, I. A. Androshchuk, and V. A. Maksimov. 2001. Isolation and identification of Crimean-Congo hemorrhagic fever virus in the Stavropol territory. Zh. Mikrobiol. Epidemiol. Immunobiol. Nov., Dec.7-11. (In Russian.)
Pekosz, A., C. Griot, K. Stillmock, N. Nathanson, and F. Gonzalez-Scarano. 1995. Protection from La Crosse virus encephalitis with recombinant glycoproteins: role of neutralizing anti-G1 antibodies. J. Virol. 69:3475-3481.
Pensiero, M. N., and J. Hay. 1992. The Hantaan virus M-segment glycoproteins G1 and G2 can be expressed independently. J. Virol. 66:1907-1914.
Pettersson, R. F., A. Andersson, and L. Melin. 1995. Mapping a retention signal for Golgi localization of a viral spike protein complex. Cold Spring Harbor Symp. Quant. Biol. 60:147-155.
Ruusala, A., R. Persson, C. S. Schmaljohn, and R. F. Pettersson. 1992. Coexpression of the membrane glycoproteins G1 and G2 of Hantaan virus is required for targeting to the Golgi complex. Virology 186:53-64.
Saluzzo, J.-F., and B. Le Guenno. 1987. Rapid diagnosis of human Crimean-Congo hemorrhagic fever and detection of the virus in naturally infected ticks. J. Clin. Microbiol. 25:922-924.
Sanchez, A. J., M. J. Vincent, and S. T. Nichol. 2002. Characterization of the glycoproteins of Crimean-Congo hemorrhagic fever virus. J. Virol. 76:7263-7275.
Schmaljohn, C. S. 1996. Bunyaviridae: the viruses and their replication, p. 1447-1471. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Virology, 3rd ed., vol. 1. Lippincott-Raven, Philadelphia, Pa.
Schmaljohn, C. S., J. Arikawa, S. E. Hasty, L. Rasmussen, H. W. Lee, P. W. Lee, and J. M. Dalrymple. 1988. Conservation of antigenic properties and sequences encoding the envelope proteins of prototype Hantaan virus and two virus isolates from Korean haemorrhagic fever patients. J. Gen. Virol. 69:1949-1955.
Schmaljohn, C. S., Y.-K. Chu, A. L. Schmaljohn, and J. M. Dalrymple. 1990. Antigenic subunits of Hantaan virus expressed by baculovirus and vaccinia virus recombinants. J. Virol. 64:3162-3170.
Schmaljohn, C. S., S. E. Hasty, and J. M. Dalrymple. 1992. Preparation of candidate vaccinia-vectored vaccines for haemorrhagic fever with renal syndrome. Vaccine 10:10-13.
Simmons, G., R. J. Wool-Lewis, F. Baribaud, R. C. Netter, and P. Bates. 2002. Ebola virus glycoproteins induce global surface protein down-modulation and loss of cell adherence. J. Virol. 76:2518-2528.
Trimble, R. B., and F. Maley. 1984. Optimizing hydrolysis of N-linked high-mannose oligosaccharides by endo-beta-N-acetylglucosaminidase H. Anal. Biochem. 141:515-522.
Vassilenko, S. M., T. L. Vassilev, L. G. Bozadjiev, I. L. Bineva, and G. Z. Kazarov. 1990. Specific intravenous immunoglobulin for Crimean-Congo haemorrhagic fever. Lancet 335:791-792.
Vincent, M. J., A. J. Sanchez, B. R. Erickson, A. Basak, M. Chretien, N. G. Seidah, and S. T. Nichol. 2003. Crimean-Congo hemorrhagic fever virus glycoprotein proteolytic processing by subtilase SKI-1. J. Virol. 77:8640-8649.
Williams, R. J., S. Al-Busaidy, F. R. Mehta, G. O. Maupin, K. D. Wagoner, S. Al-Awaidy, A. J. Suleiman, A. S. Khan, C. J. Peters, and T. G. Ksiazek. 2000. Crimean-Congo haemorrhagic fever: a seroepidemiological and tick survey in the Sultanate of Oman. Trop. Med. Int. Health 5:99-106.
Wood, O. L., V. H. Lee, J. S. Ash, and J. Casals. 1978. Crimean-Congo hemorrhagic fever, Thogoto, Dugbe, and Jos viruses isolated from ixodid ticks in Ethiopia. Am. J. Trop. Med. Hyg. 27:600-604.
Zhang, X. K., I. Takashima, and N. Hashimoto. 1989. Characteristics of passive immunity against hantavirus infection in rats. Arch. Virol. 105:235-246.(Andrea Bertolotti-Ciarlet)
AlphaVax, Inc., 2 Triangle Drive, Research Triangle Park, North Carolina 27709
Virology Division, U.S. Army Medical Research Institute for Infectious Diseases, Fort Detrick, Frederick, Maryland 21702
Department of Microbiology, Mount Sinai School of Medicine, One Gustav L. Levy Place, New York, New York 10029
ABSTRACT
Crimean-Congo hemorrhagic fever virus (CCHFV), a member of the genus Nairovirus of the family Bunyaviridae, causes severe disease with high rates of mortality in humans. The CCHFV M RNA segment encodes the virus glycoproteins GN and GC. To understand the processing and intracellular localization of the CCHFV glycoproteins as well as their neutralization and protection determinants, we produced and characterized monoclonal antibodies (MAbs) specific for both GN and GC. Using these MAbs, we found that GN predominantly colocalized with a Golgi marker when expressed alone or with GC, while GC was transported to the Golgi apparatus only in the presence of GN. Both proteins remained endo-?-N-acetylglucosaminidase H sensitive, indicating that the CCHFV glycoproteins are most likely targeted to the cis Golgi apparatus. Golgi targeting information partly resides within the GN ectodomain, because a soluble version of GN lacking its transmembrane and cytoplasmic domains also localized to the Golgi apparatus. Coexpression of soluble versions of GN and GC also resulted in localization of soluble GC to the Golgi apparatus, indicating that the ectodomains of these proteins are sufficient for the interactions needed for Golgi targeting. Finally, the mucin-like and P35 domains, located at the N terminus of the GN precursor protein and removed posttranslationally by endoproteolysis, were required for Golgi targeting of GN when it was expressed alone but were dispensable when GC was coexpressed. In neutralization assays on SW-13 cells, MAbs to GC, but not to GN, prevented CCHFV infection. However, only a subset of GC MAbs protected mice in passive-immunization experiments, while some nonneutralizing GN MAbs efficiently protected animals from a lethal CCHFV challenge. Thus, neutralization of CCHFV likely depends not only on the properties of the antibody, but on host cell factors as well. In addition, nonneutralizing antibody-dependent mechanisms, such as antibody-dependent cell-mediated cytotoxicity, may be involved in the in vivo protection seen with the MAbs to GC.
INTRODUCTION
Crimean-Congo hemorrhagic fever virus (CCHFV) causes a hemorrhagic and toxic syndrome in humans with mortality rates of up to 50%. CCHFV infection was first described during an outbreak in Russia during the 1940s, when more than 200 cases of severe hemorrhagic fever were reported among agricultural workers and soldiers in the Crimean peninsula (15, 16). Since then, the virus has spread throughout many regions of the world, including sub-Saharan Africa (60, 61), Bulgaria, the Arabian Peninsula, Iraq, Pakistan, the former Yugoslavia, northern Greece, and northwest China (16, 23, 42-45).
CCHFV is a member of the genus Nairovirus within the family Bunyaviridae (52). Members of this enveloped-virus family have a tripartite, single-stranded RNA genome of negative polarity. The medium RNA segment (the M segment) encodes the viral glycoproteins GN and GC, which, like those of other Bunyaviridae, are synthesized as polyprotein precursors that undergo proteolytic cleavage events to yield mature glycoproteins (52). The CCHFV glycoproteins exhibit several unusual structural features and undergo several processing events. First, the CCHFV glycoproteins contain, on average, 78 to 80 cysteine residues, suggesting the presence of an exceptionally large number of disulfide bonds and a complex secondary structure. Second, the GN precursor protein (Pre-GN) contains a highly variable domain at its amino terminus that contains a high proportion of serine, threonine, and proline residues, and it is predicted to be heavily O glycosylated, thus resembling a mucin-like domain present in other viral glycoproteins, most notably the Ebola virus glycoprotein (56). The mucin-like region in the Ebola glycoprotein has been shown to play an important role in a cell-rounding phenotype and immunoevasion (39, 56). It is not known whether this domain plays an important role in CCHFV pathogenesis or whether it is even O glycosylated. A third unusual feature is that the GN glycoprotein can undergo two posttranslational proteolytic cleavage events at the conserved motifs RSKR and RRLL, potentially releasing the mucin-like domain as well as a second N-terminal domain of approximately 35 kDa (P35, or the connector domain) (59). It is not known if the released domains traffic to an intracellular compartment, if they are secreted, or what effect they may have on viral pathogenesis and antigenic structure. Similar processing strategies have not been observed for other Bunyaviridae outside of the Nairovirus group.
As the only virally encoded membrane proteins, GN and GC must interact with cell surface receptors, mediate the entry of virus into cells, and serve as targets for neutralizing antibodies. Passive transfer of neutralizing antibodies can protect susceptible animals from hantavirus infection (8, 53-55, 62), and there is a report that convalescent-human sera can afford some protection in acutely infected individuals (58). Thus, characterizing the structures and functions of these proteins will be important for understanding CCHFV tropism and pathogenesis as well as for vaccine development. In this study, we describe the first monoclonal antibodies (MAbs) raised against the CCHFV glycoproteins, map the subunits to which they bind, and characterize their abilities to neutralize virus and to protect mice from a lethal CCHFV challenge. In addition, using these MAbs, we investigated the localization of GN and GC when expressed alone or together and have begun to map the regions involved in glycoprotein localization and interactions.
MATERIALS AND METHODS
Cells, antibodies, and viruses. CCHFV prototype strain IbAr10200, first isolated in 1976 from Hyalomma excavatum ticks from Sokoto, Nigeria, was grown in African green monkey kidney Vero cells or the E6 variant (51). African green monkey kidney fibroblast (CV-1), Vero, Vero E6, human cervix carcinoma (HeLa), and human embryonic kidney (HEK 293T) cells were maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA). Similarly, the human tumor cell line SW-13 (adrenocortical carcinoma) was grown in DMEM supplemented with 2.5% fetal bovine serum. Work with CCHFV was performed in a biosafety level 4 laboratory at the U.S. Army Medical Research Institute for Infectious Diseases. A recombinant vaccinia virus expressing the T7 bacteriophage RNA polymerase (vTF1.1) was grown in HeLa cells, and titers were determined in CV-1 cells according to standard protocols (1).
Production of MAbs. MAbs were prepared against the GN and GC glycoproteins of the CCHFV strain IbAr10200 by fusion of SP2/0 myeloma cells with splenocytes from BALB/c inbred mice. We carried out five independent fusions in which mice were immunized with infected suckling mouse brain homogenates (fusions I and II; MAb 30F7), with affinity-purified virion glycoproteins precipitated from nonionic detergent lysates of gradient-purified virus preparations (fusions III and IV; MAbs 1H6, 5E3, 6C2, 5B5, 8C4, 9H3, 3E3, 8G7, 8A8, and 8F10), or with affinity-purified proteins from similar lysates of infected cells (fusion V; MAbs 8A1, 12A9, 11E7, 6C11, 13G8, 14B6, 10G4, 5A5, 6B12, 10E11, 3F3, 2C9, 7F5, 11F6, and 9C6). In earlier studies, we found that hybridoma fusions carried out with splenocytes from mice immunized with mouse brain homogenates resulted predominantly in MAbs that were reactive with the viral nucleocapsid protein. To produce immunogens enriched in viral glycoproteins, cultures of SW-13 cells in 34 T-150 flasks were infected with the IbAr10200 strain of CCHFV and incubated for 36 h at 37°C. Medium fractions were removed, clarified at 10,000 x g, and centrifuged in an SW28 rotor (Beckman) to pellet virus particles (25,000 rpm for 3 h). Virus pellets were lysed in immunoprecipitation buffer (50 mM Tris-HCl, pH 8.0, containing 150 mM NaCl, 2% Triton X-100, 1% sodium deoxycholate, and protease inhibitors) and subsequently centrifuged to equilibrium in CsCl gradients. This resulted in a separation of the viral nucleocapsid (density = 1.3 g/ml) from the viral glycoproteins which banded in a broad band at lower densities. Infected-cell monolayers were similarly lysed in immunoprecipitation buffer, clarified at 10,000 x g, and centrifuged in an SW41 rotor (Beckman) to sediment the viral nucleocapsid from the solubilized viral glycoproteins (40,000 rpm for 4 h). Fractions enriched for viral glycoproteins from both cell and virion samples were then collected on solid-phase immunoadsorbents prepared by saturating protein A-Sepharose with anti-CCHFV hyperimmune mouse ascitic fluid (D. Watts, U.S. Army Medical Research Institute for Infectious Diseases). Immunoadsorbents were then exhaustively washed with phosphate-buffered saline (PBS), emulsified directly in complete or incomplete Freund's adjuvant, and inoculated subcutaneously and intraperitoneally into mice as described below. With this procedure, the CCHFV glycoproteins were enriched but were not pure. Importantly, the viral nucleocapsid was not detectable under these conditions, and using these immunogens, we were able to prepare a large number of CCHFV glycoprotein-specific MAbs.
MAbs were produced essentially as described previously (28). Briefly, BALB/c mice were twice immunized intraperitoneally with 100 μl of antigen preparations emulsified in Freund's complete adjuvant for primary immunization and in Freund's incomplete adjuvant for secondary immunization. Mice were euthanized 3 to 5 days after a third immunization, and splenocytes were fused with Sp2/0-Ag14 myeloma cells. Hybridoma cultures were incubated at 37°C with several changes of hypoxanthine-aminopterin-thymidine medium, and the supernatant fluids were screened by immunofluorescence (IF) assays against CCHFV-infected Vero cells and by enzyme-linked immunosorbent assays (ELISA) with viral antigen from suckling mouse brain homogenates or gradient-purified virus preparations. Antigenic specificity was initially determined by immunoprecipitation from [35S]methionine-labeled infected-cell lysates (as described below) and subsequently by IF microscopy. Positive-antibody-producing cells were cloned by limiting dilution and then expanded. The immunoglobulin G subclasses of the resulting MAbs were determined by indirect ELISA analysis using hybridoma supernatants. The ELISA was developed using immunoglobulin G subclass-specific immunoglobulins (Miles Laboratories) by following the manufacturer's instructions.
Neutralization assays. Eighty percent plaque reduction neutralization (PRN-80) tests were carried out on SW-13 cell monolayers. Twofold serial dilutions of the MAbs were mixed with 200 PFU of the CCHFV IbAr10200 strain and incubated for 1 h at 37°C. Confluent monolayers of SW-13 cells in six-well plates were incubated with the virus-antibody mixture for 1 h at 37°C. The inocula were removed, and 1 ml of overlay consisting of 1 part double-strength DMEM with 5% fetal bovine serum and 1 part low-gelling-temperature agarose (Bio-Rad Laboratories, Richmond, CA) in distilled water was added. After incubation at 35°C in a sealed chamber for 2 to 5 days, the plaques were visualized by neutral red staining.
Protection studies. To evaluate the protective activities of MAbs directed against the CCHFV glycoproteins in an animal model, suckling mice, which are susceptible to infection with CCHFV (24, 50), were challenged with live virus before or after passive immunization with the CCHFV-specific MAbs. Two- to 3-day-old suckling mice were inoculated in groups of five to eight by intraperitoneal injection with 50 μl of undiluted ascitic fluid containing the different MAbs. The ascitic fluids were administrated 24 h before or after the inoculation of 100 50% lethal-dose units of the CCHFV strain IbAr10200. Ascitic fluid from Sp2/0 cells that did not contain virus-specific antibodies was used as a negative control.
Construction of CCHFV glycoprotein clones. The pCAGGS-M clone was created by cloning the entire M segment of IbAr10200 into the NruI and XhoI sites found in the plasmid pCAGGS-MCSII (41). The M segment was digested using the unique restriction enzyme sites SnaBI and SalI located in the untranslated regions of the gene. This clone was used as a template for the generation of a panel of constructs used to map functional regions on CCHFV glycoproteins (Fig. 1). Primers were synthesized according to the published sequence for IbAr10200 (51), and standard PCR technology was performed for cloning into the pcDNA3.1D/V5-His-TOPO vector (Invitrogen, Carlsbad, CA). The 5'-end primers included the CACC sequence at the 5' end and the start codon to allow for directional cloning. The 3'-end primers did not possess a stop codon to allow the inclusion of the V5 and His epitope tags at the C terminus of the protein. The cloning was performed as described by the manufacturer (Invitrogen), and all constructs were sequenced. All primer sequences are available upon request.
Protein analysis. To analyze protein expression, we transfected HEK 293T cells using Lipofectamine 2000 (Invitrogen). After 24 h, cell extracts were prepared in 50 mM Tris-HCl, pH 7.4, 5 mM EDTA, 1% Triton X-100, and Complete protease inhibitor cocktail (Roche Applied Sciences, Indianapolis, IN). Cell lysates were incubated at 4°C for 3 min and then centrifuged at 10,000 x g for 10 min. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed using 4 to 15% Tris-HCl gels (Bio-Rad, Hercules, CA), followed by Western blot analysis with mouse anti-V5 (Invitrogen) as the primary antibody and sheep anti-mouse horseradish peroxidase conjugate as the secondary antibody (Amersham Pharmacia, Buckinghamshire, United Kingdom), followed by visualization with ECL-Plus Western blotting detection reagents (Bioscience, Piscataway, NJ).
IF microscopy. Localization of the CCHFV glycoproteins by indirect IF was performed as described previously (38). HeLa cells grown to 50% confluence on glass coverslips were infected with the recombinant vaccinia virus vTF1.1 (1), followed 40 min later by transfection with the different pcDNA3.1D/V5-His constructs (Fig. 1). When proteins were expressed with pCAGGS constructs, vTF1.1 infection was not utilized, since protein expression was obtained from the chicken beta-actin promoter. At 24 h posttransfection, the cells were fixed with 2% (wt/vol) paraformaldehyde in PBS, permeabilized with 0.5% Triton X-100, and stained with a GN- or GC-specific MAb ascites fluid (diluted 1:250) or with mouse anti-V5 MAb (diluted 1:500) (Invitrogen) in PBS containing 0.5 mM MgCl2 and 4% fetal bovine serum. In addition, TGN46, a sheep antibody specific for a heavily glycosylated protein localized primarily in the trans-Golgi network, was included as a marker for Golgi localization (Serotec, Oxford, United Kingdom). Then, cells were washed with PBS and incubated for 1 h with the secondary antibodies conjugated to Alexa Flour 488 (goat anti-mouse) and Alexa Flour 594 (goat anti-sheep) (Molecular Probes, Eugene, OR) diluted 1:500 in PBS-4% fetal bovine serum. Finally, cells were washed in PBS, mounted in Fluoromount-G (Southern Biotechnology Associates, Birmingham, AL), and examined on a Nikon E600 microscope at a magnification of x60 utilizing UV illumination.
Endoglycosidase treatment. The method of Trimble and Maley (57) was slightly modified for digesting glycosylated CCHFV glycoproteins with endo-?-N-acetylglucosaminidase H (endo-H). Cell lysates from transfected HEK 293T cells were boiled for 5 min in a 50 mM sodium citrate buffer (pH 5.5) containing 1% SDS and 200 μg/ml of phenylmethylsulfonyl fluoride. After cooling, the samples were supplemented with an equal volume of 0.02 U of endo-H (New England BioLabs Inc., Beverly, MA) in a 50 mM sodium citrate buffer (pH 5.5) or the same buffer without the enzyme before incubation at 37°C for 20 h. After this incubation, the samples were analyzed by SDS-PAGE. In addition, the glycoproteins were treated with peptide-N-glycosidase F (PNGase F) (New England Biolabs). A similar protocol was followed for PNGase F treatment, but the digestion was performed in 100 mM sodium phosphate, pH 7.5, with 0.75% NP-40, 0.1% SDS, and 50 mM ?-mercaptoethanol.
Immunoprecipitations from infected cells. Immunoprecipitation assays were performed as previously described (38). Briefly, SW-13 cell monolayers (1 x106 cells) were infected with the CCHFV strain IbAr10200 at a multiplicity of infection of 5. Infected SW-13 cells were placed in methionine-free DMEM at 28 h postinfection for 30 min, and then 30 μCi of [35S]methionine (Trans-35S-label; ICN, Irvine, CA) was added. After 1 h of labeling, 50 μg of unlabeled methionine per ml was added. Infected cells were collected 24 h postlabeling or 20 h posttransfection and lysed with 0.5 ml of lysis buffer containing 50 mM Tris, pH 7.4, 150 mM NaCl, 1% CHAPSO {3-[(3-cholamidopropyl)-dimethylammonio]-2-hydroxy-1-propanesulfonate}, 5 mM EDTA, and a Complete protease inhibitor mixture (Roche). Precleared lysates were incubated for 1 h at 4°C with the MAbs. Glycoproteins were immunoprecipitated using 7 μg of the MAbs generated in this study. The MAbs were previously coupled to protein A/G Plus-agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoprecipitated proteins were eluted at 100°C for 5 min with 20 μl of sample buffer (0.08 M Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 5% ?-mercaptoethanol, 0.005% bromophenol blue). The immunoprecipitated proteins were detected after separation by 10% SDS-PAGE utilizing a personal densitometer (model 860; Molecular Dynamics, Inc., Sunnyvale, CA), and the data were analyzed with the ImageQuant NT software.
Immunoprecipitation from transfected cells. For the analysis of the different CCHFV glycoprotein constructs, HEK 293T cell monolayers (1 x106) were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. The immunoprecipitation protocol utilized was in general the same as that described above for infected cells. The V5 epitope tag present at the C terminus of the recombinant proteins was detected using 7 μg of a MAb directed against the V5 tag (Invitrogen). Samples from transfected cells were analyzed by Western blotting as described above, using a rabbit anti-V5 tag hyperimmune serum diluted 1:5,000 (Sigma).
RESULTS
CCHFV M segment processing. Glycoproteins of the nairovirus group of bunyaviruses are processed in a distinctive and complex manner (51, 59). The M segment of CCHFV encodes a polyprotein that undergoes proteolytic processing to yield a 140-kDa Pre-GN (termed GN-FL in Fig. 1) and an 85-kDa precursor of GC (Pre-GC; termed GC-FL in Fig. 1). Recently, it was shown that the N terminus of the mature GN protein contains both a mucin-like and a P35 connector domain (Fig. 1) that are removed by specific cleavage with protease SKI-1, resulting in the generation of mature GN (GN-RRLL in Fig. 1) (59). However, how GN and GC are cleaved from each other is not clear, though it is thought that the N terminus of the mature GC protein is generated by cleavage at a conserved RKPL site (59). Finally, cleavage between the mucin domain and the P35 region may be mediated by furin, which is consistent with the presence of a conserved RSKR sequence between these domains (Fig. 1) (59).
MAbs to CCHFV. To assist in our studies of CCHFV glycoprotein processing and localization, and to determine if antibodies can confer protection to CCHFV, we produced a panel of MAbs from mice immunized with a variety of CCHFV antigens as described in Materials and Methods. MAbs to CCHFV were identified by ELISA, after which both immunoprecipitation and immunofluorescence studies were done to identify the glycoprotein subunit to which each MAb bound. Only the 26 MAbs whose specificity could be clearly identified by both techniques are included in this study. For the immunofluorescence studies, HeLa cells were transfected with pcDNA3.1D/V5-His vectors expressing either GN (GN-FL, GN-RSKR, or GN-RRLL) or GC (GC-FL) and were infected with vTF1.1, a recombinant vaccinia virus that expresses the T7 RNA polymerase (1), in order to achieve high levels of protein expression. The cells were then fixed and processed for IF microscopy. All GN antibodies recognized GN-FL but most did not recognize GN-RSKR or GN-RRLL, with the exception of MAb 8F10, which recognized all the GN constructs. Since neither GN-RSKR nor GN-RRLL was correctly transported to the Golgi apparatus (see below), the failure of most GN MAbs to recognize these constructs could be due to protein misfolding rather than the loss of specific epitopes.
From the panel of 26 MAbs, we selected six directed against GN and six directed against GC for the detailed study of CCHFV glycoprotein processing and targeting. When analyzed by IF and when utilizing HeLa cells transfected with the entire M segment of CCHFV strain IbAr10200, all anti-GC MAbs gave similar patterns, with GC being localized to both the endoplasmic reticulum (ER) and Golgi regions as judged by colocalization with Golgi and ER markers (Fig. 2 and data not shown). The six anti-GN MAbs recognized GN only in the Golgi apparatus, suggesting that they recognize epitopes that are either formed after the protein reaches the Golgi apparatus or formed shortly before exit from the ER (Fig. 2). None of the MAbs detected GN or GC protein on the surfaces of unpermeabilized cells (data not shown). Our results indicate that the CCHFV glycoproteins are not delivered to the cell surface in appreciable quantities, which is consistent with studies that have shown that other Bunyaviridae bud into the Golgi apparatus and that their glycoproteins are targeted to this organelle and not delivered to the cell surface (13, 18, 29, 34, 37).
To determine if the MAbs could recognize GN and GC in different experimental contexts, both Western blot analysis and immunoprecipitation assay were performed. None of the MAbs recognized CCHFV proteins under fully denaturing conditions by Western blot analysis. However, in the absence of boiling, MAb 11E7 could recognize GC obtained from transfected cells or from virus-infected cells by Western blot analysis (Fig. 3). The failure of most of the MAbs to recognize their antigens by Western blot analysis indicates that they bind to conformation-dependent determinants that are lost upon protein denaturation.
The MAbs were also used to immunoprecipitate GN and GC from lysates of cells infected with the IbAr10200 strain of CCHFV. All of the MAbs were able to immunoprecipitate the protein subunits to which they bound, which is consistent with their ability to recognize their epitopes by IF. In addition, four of the six GC MAbs were able to immunoprecipitate both processed GC (around 75 kDa) and its precursor protein (around 82 kDa), while all of the GN MAbs analyzed were able to immunoprecipitate processed GN (35 kDa) and its precursor (140 kDa), which contains both the mucin and the P35 connector domains (data not shown).
Virus neutralization and protection studies. To analyze the in vitro neutralization activities of the panel of MAbs directed against the CCHFV glycoproteins, we performed plaque reduction assays. Twofold serial dilutions of each MAb ascites fluid sample were incubated with 200 PFU of the IbAr10200 strain of CCHFV for 1 h at 37°C prior to addition to confluent SW-13 cells. Plaques were counted 3 to 5 days later, and PRN-80 titers were calculated. None of the MAbs directed against GN exhibited neutralizing activity in this assay (Fig. 4B), though many of the MAbs directed against GC neutralized CCHFV in vitro (Fig. 4A).
Antibodies have been shown to be effective pre- and postexposure to prophylaxis treatments for a number of viruses, including cytomegalovirus (9, 10) and respiratory syncytial virus (17, 32). Convalescent-phase serum has been shown to be of benefit to individuals acutely infected with CCHFV (57). Therefore, we tested the CCHFV MAbs for their ability to protect suckling mice challenged with the virus. The MAbs were individually administered by passive immunization to 2- to 3-day-old suckling mice either 24 h before or 24 h after challenge with 100 50% infective-dose units of the IbAr10200 strain of CCHFV (Fig. 4). Protection was registered as the percentage of animals that survived challenge with the live virus (Fig. 4). In general, anti-GC MAbs that were capable of efficient neutralization in vitro protected mice to an appreciable degree when applied before and, to a lesser extent, after virus challenge (Fig. 4). In addition, anti-GC MAbs that did not neutralize CCHFV infection of SW-13 cells afforded partial protection to mice from CCHFV provided that the MAbs were administered 24 h before viral challenge. When these MAbs were administered 24 h after virus challenge, protection was usually not observed. This suggests that these MAbs do possess some neutralizing activity that was not detected in our in vitro assay or that other antibody-based effector mechanisms, such as antibody-dependent cell-mediated toxicity or complement-mediated cell lysis, function in this context. Likewise, many of the anti-GN MAbs conferred significant protection to CCHFV challenge, even when applied 24 h after virus challenge and even though they did not prevent virus infection of SW-13 cells in vitro (Fig. 4). These results show that there is an imperfect relationship between in vitro neutralization and in vivo protective ability, at least under the assay conditions used here, and that the ability of an antibody to neutralize CCHFV may depend in part on host factors, as has been observed for La Crosse virus (LACV), another bunyavirus (25-27).
GN contains a Golgi targeting signal. To analyze the contribution of each CCHFV glycoprotein to Golgi localization, we expressed each protein independently in HeLa cells. IF analyses showed that, while GN remained in the Golgi apparatus if expressed alone, GC localized entirely in the ER in the absence of GN (Fig. 5Aa and -b). When the proteins were expressed together from independent constructs, both GC and GN were localized in the Golgi apparatus (Fig. 5Ac). The restoration of GC Golgi localization in the presence of GN suggests that GN possesses a Golgi localization signal and that GC localizes to the Golgi apparatus through its interaction with GN (Fig. 5A).
We further analyzed the localization of CCHFV glycoproteins when expressed independently or together by investigating their N-linked glycosylation. The GN ectodomain of the IbAr10200 strain of CCHFV contains one predicted N-linked glycosylation site, while the GC ectodomain contains three sites. To determine whether the CCHFV glycoproteins are glycosylated and, if so, modified by medial Golgi enzymes, we used PNGase F, which removes all N-linked carbohydrate chains, or endo-H, which removes immature carbohydrate chains. When GC-FL was expressed alone, digestion with either endo-H or PNGase F caused the protein to migrate faster (Fig. 5B). When GN-FL was expressed alone, it too migrated faster following digestion with either endo-H or PNGase F. To determine if the shift observed following glycosidase treatment of GN-FL was the result of a loss of glycans on the mucin or P35 regions only, we treated the GN-RRLL protein that lacks both the mucin and the P35 domains with endo-H. The GN-RRLL protein also migrated faster after endo-H treatment, indicating the presence of N-linked carbohydrate chains on the ectodomain of the mature GN as well. Similar results were obtained when the proteins were coexpressed (Fig. 5B). Thus, both proteins are N glycosylated, but neither appears to be processed in the medial Golgi apparatus when expressed alone or together.
Contribution of GN N-terminal domains to protein localization. The CCHFV GN protein is unusual in that it has two N-terminal domains, a mucin-like domain and the P35 domain, that appear to be cleaved from the GN precursor in a posttranslational fashion (51). The function of these regions is unknown. A similar mucin-like domain has been found to be associated with the glycoprotein of Ebola virus and has been shown to induce cell rounding and detachment in vitro and possibly to be involved in the pathogenicity of the virus (56). To analyze the involvement of the mucin-like domain and P35 region in GN cellular localization, we deleted these regions from constructs that contained only the GN portion of the protein in order to generate a construct without the mucin-like domain (GN-RSKR) and a construct without both the mucin-like domain and the P35 region (GN-RRLL) (Fig. 1). In both cases, a signal sequence was introduced at the N terminus to ensure proper targeting to the ER. Upon expression in HeLa cells, IF analysis using a MAb to the GN ectodomain showed that deletion of just the mucin-like domain (GN-RSKR) resulted in a protein that localized to the Golgi apparatus in a manner similar to full-length GN-FL (Fig. 6A). In contrast, the GN-RRLL protein was not present in either the ER or the Golgi apparatus but was distributed in a punctuate pattern that resembled aggresomes, suggesting that it misfolds (Fig. 6B). Thus, the mucin-like domain is dispensable for Golgi targeting, while removal of the P35 connector region affected GN localization and also led to enhanced degradation. Interestingly, localization of GN-RRLL to the Golgi complex was recovered when the GN-RRLL protein was coexpressed with GC, suggesting that GN-GC interactions may promote the correct folding and transport of the proteins (Fig. 6C).
Mapping of the Golgi localization signal. In all Bunyaviridae glycoproteins that have been examined, the Golgi localization signal has been localized to GN, generally in the cytoplasmic tail (CT) or transmembrane domain (TM) of the protein (3-5, 12, 14, 33, 34, 47, 49). The C-terminal domain of GN contains a stretch of predicted TMs and cytoplasmic loops between the first amino acid of GC and the predicted C terminus of GN (59). The function of this unusual region on glycoprotein processing or any other step of the viral replication and cell cycle is unknown. To analyze the role of this region in GN localization, we deleted two (GN 2TM) or three (GN 1TM) of the predicted four TMs at the GN C terminus (Fig. 1). HeLa cells were transfected with the constructs and analyzed by IF (Fig. 6). We found that all of the C-terminally truncated GN constructs localized to the Golgi apparatus (Fig. 6D and E). These results suggest that the Golgi localization signal is not located in this region.
Next, we designed soluble constructs that lacked the transmembrane and cytoplasmic domains of GN (sGN) or GC (sGC) or that lacked only the CT of GN (GN-no CT) (Fig. 1). When these constructs were expressed using the vaccinia virus T7 polymerase system in HeLa cells, both sGN and GN-no CT, although also present in the ER, localized to the Golgi apparatus when analyzed by IF microscopy using MAbs against GN (Fig. 6F and G). In contrast, sGC was restricted to the ER, just like full-length GC (Fig. 6H). When sGC and sGN were coexpressed, sGC was then targeted to the Golgi apparatus (Fig. 6I), indicating that the interaction of GC and GN occurs through their ectodomains, that the proteins can fold correctly when in their soluble forms, and that Golgi targeting information resides within the ectodomain of GN.
DISCUSSION
Relatively little is known about the mechanisms by which bunyaviruses enter cells or how infection can be prevented by neutralizing antibodies, and MAbs that block CCHFV infection have not been described. Therefore, we developed a panel of MAbs to assist in our studies on CCHFV glycoprotein biology and to begin characterizing the antigenic structures of GN and GC. The large majority of MAbs bound to conformation-dependent epitopes in GN or GC. A number of MAbs against GC, but not against GN, were able to neutralize virus infection of SW-13 cells in vitro, suggesting that GC plays an important role in virus entry. Similarly, MAbs directed against the GC glycoproteins of LACV can inhibit virus infection, with some evidence indicating that this is due to a reduction in virus binding to the cell surface (19, 21, 22, 26). However, some antibodies against GN can neutralize LACV infection in an insect cell line, though not in a mammalian cell line, suggesting that virus neutralization can be dependent on the cell type being infected and that virus entry mechanisms may differ between invertebrates and vertebrates (30). Consistent with this, proteolytic degradation of GC with trypsin or pronase virtually eliminates the ability of LACV to bind to vertebrate, but not invertebrate, cell lines (30, 46).
Our results suggest that CCHFV neutralization mechanisms may be complex and context dependent. In general, MAbs directed against GN were more effective at protecting mice from a lethal CCHFV challenge than were MAbs to GC when administrated either 24 h before or after infection, even though GN MAbs did not neutralize infection of SW-13 cells in vitro. In addition, not all GC MAbs that neutralized CCHFV infection in vitro conferred high levels of protection in vivo, especially if administered after infection, as seen for the 8A1 and 1H6 MAbs. Thus, there was not a strict relationship between in vitro neutralization and in vivo protection. As noted above, these results resemble observations on the differential roles of LACV GN and GC glycoproteins in viral neutralization and protection. A soluble version of the GC La Crosse glycoprotein is sufficient to block virus infection in mammalian cells, while antibodies against GN neutralize infection in a mosquito cell line but not in a vertebrate cell line (31). The mechanisms that account for the differential inhibition of CCHFV infection are not currently clear. We assume that the conformational changes undergone by GN and GC to elicit membrane fusion and virus entry will likely be the same regardless of the cell type being infected and that these changes are likely to be induced by acid pH following endocytosis of the virus, as has been documented for La Crosse and Hantaan viruses (7, 20). However, it remains to be determined if CCHFV uses different cell surface receptors and attachment factors on different cell types. Without knowing the identity of any potential CCHFV receptors, it will be interesting to examine the ability of MAbs to neutralize CCHFV on cell lines derived from different species, including ticks.
One of the hallmarks of the Bunyaviridae family is that their glycoproteins are targeted to the Golgi apparatus from which they bud (48, 52). Based on this fact, a number of studies have sought to identify the signals responsible for targeting GN and GC to the Golgi apparatus for a number of bunyaviruses (2, 34, 36, 40, 47, 49). Generally, it has been found that the Golgi retention signals reside within the TM and/or CT of the glycoprotein closest to the N terminus of the glycoprotein precursor (18, 34, 35). However, no consensus Golgi localization motif appears to be shared among the glycoproteins of these viruses, and Nairovirus glycoprotein targeting signals and antigenic structures have not been analyzed so far.
We found that the CCHFV glycoproteins were targeted to the Golgi apparatus, as determined by IF microscopy. Since both GN and GC were sensitive to endo-H treatment, it is likely that the proteins are targeted to an early Golgi compartment. Similarly, hantavirus glycoproteins, although present in the Golgi apparatus, remain sensitive to endo-H treatment (49). When GN and GC were expressed together, GN was localized to the Golgi complex while GC was found in both the Golgi apparatus and the ER. It is obvious that GN must also be present in the ER, but our conformation-dependent MAbs either do not recognize GN in the ER or bind to epitopes that form just prior to exiting from this organelle. It is also possible that GN is transported more quickly from the ER than GC and, at steady state, is below our limit of detection in the ER. Indeed, some studies on the biosynthesis of Uukuniemi virus GN and GC proteins showed previously that GN is transported faster than GC from the ER to the site of virus budding at the Golgi complex (29). The apparent difference in transport kinetics is due to the fact that GN folds and is transported from the ER to the Golgi apparatus 30 to 45 min faster than GC (4).
The ability of CCHFV GN to localize to the Golgi apparatus when expressed independently of GC indicates that GN contains a Golgi targeting or retention motif. Since GC is restricted to the ER in the absence of GN, we conclude that its transport to the Golgi apparatus is dependent upon GN and likely results from GN-GC oligomerization. Whether GC fails to fold correctly in the absence of GN, contains an ER retention signal that is masked by GN association, or lacks a positive transport signal cannot be determined at present. Our results are in agreement with published data about other Bunyaviridae glycoproteins (3, 35, 36, 40, 47, 49) which localize to the Golgi apparatus in the absence of any other viral proteins (11, 14, 34, 47). Although for most of the Bunyaviridae glycoproteins analyzed to date, the Golgi targeting signal is contained in one of the glycoproteins, for the Hantaan virus, a member of the Hantavirus group, both glycoproteins are required to achieve Golgi apparatus targeting (49). Our results with CCHFV indicate that Golgi targeting information resides largely in the ectodomain of the GN subunit, since a soluble version of GN was largely restricted to the Golgi apparatus. However, small amounts of this protein were secreted from cells, indicating that the TM of GN may also play a role in Golgi retention. It is also evident that the ectodomains of GN and GC interact with each other and that targeting of GC to the Golgi apparatus is dependent upon its association with GN, with ectodomain interactions being important.
In summary, although the CCHFV glycoproteins are unique in many aspects with respect to the glycoproteins from other members of the Bunyaviridae family, there are some similarities with regard to Golgi targeting and the glycoprotein subunits to which neutralizing antibodies are directed. Our studies indicate that CCHFV neutralization is likely to be context dependent and that more in-depth studies of various cell lines and animal models will be needed to characterize neutralization mechanisms and to identify antibodies that could be used therapeutically. Identification of regions on the CCHFV glycoproteins involved in viral neutralization, protection, and processing will contribute to our understanding of the tropism and pathogenesis of this emerging viral pathogen.
ACKNOWLEDGMENTS
We thank Nicolette Pesik, Randall Bethke, and Loreen Hodgson for superb technical assistance on the production and characterization of the MAbs. We also thank Aura Garrison and Donald Pijak for expert technical assistance.
This work was supported in part by National Institutes of Health training grant NIH T32 AI055400 and grant PR033269 from the Department of Defense Peer Reviewed Medical Research Program (PRMRP) of the Office of the Congressionally Directed Medical Research Programs.
REFERENCES
Alexander, W. A., B. Moss, and T. R. Fuerst. 1992. Regulated expression of foreign genes in vaccinia virus under the control of bacteriophage T7 RNA polymerase and the Escherichia coli lac repressor. J. Virol. 66:2934-2942.
Anderson, G. W., Jr., and J. F. Smith. 1987. Immunoelectron microscopy of Rift Valley fever viral morphogenesis in primary rat hepatocytes. Virology 161:91-100.
Andersson, A. M., L. Melin, A. Bean, and R. F. Pettersson. 1997. A retention signal necessary and sufficient for Golgi localization maps to the cytoplasmic tail of a Bunyaviridae (Uukuniemi virus) membrane glycoprotein. J. Virol. 71:4717-4727.
Andersson, A. M., L. Melin, R. Persson, E. Raschperger, L. Wikstr?m, and R. F. Pettersson. 1997. Processing and membrane topology of the spike proteins G1 and G2 of Uukuniemi virus. J. Virol. 71:218-225.
Andersson, A. M., and R. F. Pettersson. 1998. Targeting of a short peptide derived from the cytoplasmic tail of the G1 membrane glycoprotein of Uukuniemi virus (Bunyaviridae) to the Golgi complex. J. Virol. 72:9585-9596.
Archer, D. B., D. J. Jeenes, D. A. MacKenzie, G. Brightwell, N. Lambert, G. Lowe, S. E. Radford, and C. M. Dobson. 1990. Hen egg white lysozyme expressed in, and secreted from, Aspergillus niger is correctly processed and folded. Biotechnology (New York) 8:741-745.
Arikawa, J., I. Takashima, and N. Hashimoto. 1985. Cell fusion by haemorrhagic fever with renal syndrome (HFRS) viruses and its application for titration of virus infectivity and neutralizing antibody. Arch. Virol. 86:303-313.
Arikawa, J., J. S. Yao, K. Yoshimatsu, I. Takashima, and N. Hashimoto. 1992. Protective role of antigenic sites on the envelope protein of Hantaan virus defined by monoclonal antibodies. Arch. Virol. 126:271-281.
Aulitzky, W. E., T. F. Schulz, H. Tilg, D. Niederwieser, K. Larcher, L. Ostberg, M. Scriba, J. Martindale, A. C. Stern, P. Grass, et al. 1991. Human monoclonal antibodies neutralizing cytomegalovirus (CMV) for prophylaxis of CMV disease: report of a phase I trial in bone marrow transplant recipients. J. Infect. Dis. 163:1344-1347.
Aulitzky, W. E., H. Tilg, D. Niederwieser, M. Hackl, B. Meister, and C. Huber. 1988. Ganciclovir and hyperimmunoglobulin for treating cytomegalovirus infection in bone marrow transplant recipients. J. Infect. Dis. 158:488-489.
Bupp, K., K. Stillmock, and F. Gonzalez-Scarano. 1996. Analysis of the intracellular transport properties of recombinant La Crosse virus glycoproteins. Virology 220:485-490.
Chen, S.-Y., and R. W. Compans. 1991. Oligomerization, transport, and Golgi retention of Punta Toro virus glycoproteins. J. Virol. 65:5902-5909.
Chen, S.-Y., Y. Matsuoka, and R. W. Compans. 1991. Assembly and polarized release of Punta Toro virus and effects of brefeldin A. J. Virol. 65:1427-1439.
Chen, S. Y., Y. Matsuoka, and R. W. Compans. 1991. Golgi complex localization of the Punta Toro virus G2 protein requires its association with the G1 protein. Virology 183:351-365.
Chumakov, M. P., A. M. Butenko, N. V. Shalunova, L. I. Mart'ianova, S. E. Smirnova, N. Bashkirtsev, T. I. Zavodova, S. G. Rubin, E. A. Tkachenko, V. Karmysheva, V. N. Reingol'd, G. V. Popov, and A. P. Savinov. 1968. New data on the viral agent of Crimean hemorrhagic fever. Vopr. Virusol. 13:377. (In Russian.)
Chumakov, M. P., S. E. Smirnova, and E. A. Tkachenko. 1970. Relationship between strains of Crimean haemorrhagic fever and Congo viruses. Acta Virol. 14:82-85.
Englund, J. A., P. A. Piedra, and E. Whimbey. 1997. Prevention and treatment of respiratory syncytial virus and parainfluenza viruses in immunocompromised patients. Am. J. Med. 102:61-70, 75-76.
Gerrard, S. R., and S. T. Nichol. 2002. Characterization of the Golgi retention motif of Rift Valley fever virus GN glycoprotein. J. Virol. 76:12200-12210.
Gonzalez-Scarano, F. 1985. La Crosse virus G1 glycoprotein undergoes a conformational change at the pH of fusion. Virology 140:209-216.
Gonzalez-Scarano, F., R. S. Janssen, J. A. Najjar, N. Pobjecky, and N. Nathanson. 1985. An avirulent G1 glycoprotein variant of La Crosse bunyavirus with defective fusion function. J. Virol. 54:757-763.
Grady, L. J., M. L. Sanders, and W. P. Campbell. 1983. Evidence for three separate antigenic sites on the G1 protein of La Crosse virus. Virology 126:395-397.
Grady, L. J., S. Srihongse, M. A. Grayson, and R. Deibel. 1983. Monoclonal antibodies against La Crosse virus. J. Gen. Virol. 64:1699-1704.
Hoogstraal, H. 1979. The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe, and Africa. J. Med. Entomol. 15:307-417.
Huggins, J. W. 1989. Prospects for treatment of viral hemorrhagic fevers with ribavirin, a broad-spectrum antiviral drug. Rev. Infect. Dis. 11 (Suppl. 4):S750-S761.
Kingsford, L. 1984. Enhanced neutralization of La Crosse virus by the binding of specific pairs of monoclonal antibodies to the G1 glycoprotein. Virology 136:265-273.
Kingsford, L., K. H. Boucquey, and T. P. Cardoso. 1991. Effects of specific monoclonal antibodies on La Crosse virus neutralization: aggregation, inactivation by Fab fragments, and inhibition of attachment to baby hamster kidney cells. Virology 180:591-601.
Kingsford, L., L. D. Ishizawa, and D. W. Hill. 1983. Biological activities of monoclonal antibodies reactive with antigenic sites mapped on the G1 glycoprotein of La Crosse virus. Virology 129:443-455.
Kohler, G., S. C. Howe, and C. Milstein. 1976. Fusion between immunoglobulin-secreting and nonsecreting myeloma cell lines. Eur. J. Immunol. 6:292-295.
Kuismanen, E., B. B?ng, M. Hurme, and R. F. Pettersson. 1984. Uukuniemi virus maturation: immunofluorescence microscopy with monoclonal glycoprotein-specific antibodies. J. Virol. 51:137-146.
Ludwig, G. V., B. A. Israel, B. M. Christensen, T. M. Yuill, and K. T. Schultz. 1991. Monoclonal antibodies directed against the envelope glycoproteins of La Crosse virus. Microb. Pathog. 11:411-421.
Ludwig, G. V., B. A. Israel, B. M. Christensen, T. M. Yuill, and K. T. Schultz. 1991. Role of La Crosse virus glycoproteins in attachment of virus to host cells. Virology 181:564-571.
Malley, R., J. DeVincenzo, O. Ramilo, P. H. Dennehy, H. C. Meissner, W. C. Gruber, P. J. Sanchez, H. Jafri, J. Balsley, D. Carlin, S. Buckingham, L. Vernacchio, and D. M. Ambrosino. 1998. Reduction of respiratory syncytial virus (RSV) in tracheal aspirates in intubated infants by use of humanized monoclonal antibody to RSV F protein. J. Infect. Dis. 178:1555-1561.
Matsuoka, Y., S. Y. Chen, and R. W. Compans. 1991. Bunyavirus protein transport and assembly. Curr. Top. Microbiol. Immunol. 169:161-179.
Matsuoka, Y., S. Y. Chen, and R. W. Compans. 1994. A signal for Golgi retention in the bunyavirus G1 glycoprotein. J. Biol. Chem. 269:22565-22573.
Matsuoka, Y., S. Y. Chen, C. E. Holland, and R. W. Compans. 1996. Molecular determinants of Golgi retention in the Punta Toro virus G1 protein. Arch. Biochem. Biophys. 336:184-189.
Matsuoka, Y., T. Ihara, D. H. Bishop, and R. W. Compans. 1988. Intracellular accumulation of Punta Toro virus glycoproteins expressed from cloned cDNA. Virology 167:251-260.
Melin, L., R. Persson, A. Andersson, A. Bergstr?m, R. R?nnholm, and R. F. Pettersson. 1995. The membrane glycoprotein G1 of Uukuniemi virus contains a signal for localization to the Golgi complex. Virus Res. 36:49-66.
Morais, V. A., A. S. Crystal, D. S. Pijak, D. Carlin, J. Costa, V. M. Lee, and R. W. Doms. 2003. The transmembrane domain region of nicastrin mediates direct interactions with APH-1 and the gamma-secretase complex. J. Biol. Chem. 278:43284-43291.
Nabel, G. J. 1999. Surviving Ebola virus infection. Nat. Med. 5:373-374.
Nakitare, G. W., and R. M. Elliott. 1993. Expression of the Bunyamwera virus M genome segment and intracellular localization of NSm. Virology 195:511-520.
Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108:193-199.
Olaleye, O. D., O. Tomori, and H. Schmitz. 1996. Rift Valley fever in Nigeria: infections in domestic animals. Rev. Sci. Tech. Off. Int. Epizoot. 15:937-946.
Onishchenko, G. G., V. I. Efremenko, N. G. Kovalev, N. A. Shibkov, M. Fedorov, N. Zhilina, G. N. Fedosova, G. V. Sysoliatina, M. Evchenko, A. P. Beier, V. A. Popov, K. V. Shenetts, L. N. Marchukova, E. N. Afanas'ev, and N. V. Vasilenko. 2001. Specific epidemiologic features of Crimean haemorrhagic fever in Stavropol' region in 1999-2000. Zh. Mikrobiol. Epidemiol. Immunobiol. Nov., Dec.(Suppl. 6):86-89. (In Russian.)
Onishchenko, G. G., M. Lomov, V. I. Markov, V. A. Merkulov, A. N. Mironov, I. V. Borisevich, A. A. Makhlai, N. T. Vasil'ev, S. O. Vodop'ianov, N. G. Tikhonov, T. P. Pashanina, V. V. Manankov, V. P. Smelianskii, S. T. Savchenko, and S. E. Smirnova. 2000. The laboratory diagnosis of an outbreak of hemorrhagic fever at Oblivskaya village, Rostov Province: proof of the etiological role of the Crimean-Congo hemorrhagic fever virus. Zh. Mikrobiol. Epidemiol. Immunobiol. Mar., Apr.32-36. (In Russian.)
Onishchenko, G. G., V. I. Markov, V. A. Merkulov, N. T. Vasil'ev, A. M. Berezhnoi, I. A. Androshchuk, and V. A. Maksimov. 2001. Isolation and identification of Crimean-Congo hemorrhagic fever virus in the Stavropol territory. Zh. Mikrobiol. Epidemiol. Immunobiol. Nov., Dec.7-11. (In Russian.)
Pekosz, A., C. Griot, K. Stillmock, N. Nathanson, and F. Gonzalez-Scarano. 1995. Protection from La Crosse virus encephalitis with recombinant glycoproteins: role of neutralizing anti-G1 antibodies. J. Virol. 69:3475-3481.
Pensiero, M. N., and J. Hay. 1992. The Hantaan virus M-segment glycoproteins G1 and G2 can be expressed independently. J. Virol. 66:1907-1914.
Pettersson, R. F., A. Andersson, and L. Melin. 1995. Mapping a retention signal for Golgi localization of a viral spike protein complex. Cold Spring Harbor Symp. Quant. Biol. 60:147-155.
Ruusala, A., R. Persson, C. S. Schmaljohn, and R. F. Pettersson. 1992. Coexpression of the membrane glycoproteins G1 and G2 of Hantaan virus is required for targeting to the Golgi complex. Virology 186:53-64.
Saluzzo, J.-F., and B. Le Guenno. 1987. Rapid diagnosis of human Crimean-Congo hemorrhagic fever and detection of the virus in naturally infected ticks. J. Clin. Microbiol. 25:922-924.
Sanchez, A. J., M. J. Vincent, and S. T. Nichol. 2002. Characterization of the glycoproteins of Crimean-Congo hemorrhagic fever virus. J. Virol. 76:7263-7275.
Schmaljohn, C. S. 1996. Bunyaviridae: the viruses and their replication, p. 1447-1471. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Virology, 3rd ed., vol. 1. Lippincott-Raven, Philadelphia, Pa.
Schmaljohn, C. S., J. Arikawa, S. E. Hasty, L. Rasmussen, H. W. Lee, P. W. Lee, and J. M. Dalrymple. 1988. Conservation of antigenic properties and sequences encoding the envelope proteins of prototype Hantaan virus and two virus isolates from Korean haemorrhagic fever patients. J. Gen. Virol. 69:1949-1955.
Schmaljohn, C. S., Y.-K. Chu, A. L. Schmaljohn, and J. M. Dalrymple. 1990. Antigenic subunits of Hantaan virus expressed by baculovirus and vaccinia virus recombinants. J. Virol. 64:3162-3170.
Schmaljohn, C. S., S. E. Hasty, and J. M. Dalrymple. 1992. Preparation of candidate vaccinia-vectored vaccines for haemorrhagic fever with renal syndrome. Vaccine 10:10-13.
Simmons, G., R. J. Wool-Lewis, F. Baribaud, R. C. Netter, and P. Bates. 2002. Ebola virus glycoproteins induce global surface protein down-modulation and loss of cell adherence. J. Virol. 76:2518-2528.
Trimble, R. B., and F. Maley. 1984. Optimizing hydrolysis of N-linked high-mannose oligosaccharides by endo-beta-N-acetylglucosaminidase H. Anal. Biochem. 141:515-522.
Vassilenko, S. M., T. L. Vassilev, L. G. Bozadjiev, I. L. Bineva, and G. Z. Kazarov. 1990. Specific intravenous immunoglobulin for Crimean-Congo haemorrhagic fever. Lancet 335:791-792.
Vincent, M. J., A. J. Sanchez, B. R. Erickson, A. Basak, M. Chretien, N. G. Seidah, and S. T. Nichol. 2003. Crimean-Congo hemorrhagic fever virus glycoprotein proteolytic processing by subtilase SKI-1. J. Virol. 77:8640-8649.
Williams, R. J., S. Al-Busaidy, F. R. Mehta, G. O. Maupin, K. D. Wagoner, S. Al-Awaidy, A. J. Suleiman, A. S. Khan, C. J. Peters, and T. G. Ksiazek. 2000. Crimean-Congo haemorrhagic fever: a seroepidemiological and tick survey in the Sultanate of Oman. Trop. Med. Int. Health 5:99-106.
Wood, O. L., V. H. Lee, J. S. Ash, and J. Casals. 1978. Crimean-Congo hemorrhagic fever, Thogoto, Dugbe, and Jos viruses isolated from ixodid ticks in Ethiopia. Am. J. Trop. Med. Hyg. 27:600-604.
Zhang, X. K., I. Takashima, and N. Hashimoto. 1989. Characteristics of passive immunity against hantavirus infection in rats. Arch. Virol. 105:235-246.(Andrea Bertolotti-Ciarlet)