Inhibitory interactions of the 2,3-dihydro-6,7-dihydroxy-1H-isoindol-1-one scaffold with Bunyavirales cap-snatching endonucleases expose relevant drug design features.

The World Health Organization (WHO) identifies several bunyaviruses as significant threats to global public health security. Developing effective therapies against these viruses is crucial to combat future outbreaks and mitigate their impact on patient outcomes. Here, we report the synthesis of some isoindol-1-one derivatives and explore their inhibitory properties over an indispensable metal-dependent cap-snatching endonuclease (Cap-ENDO) shared among evolutionary divergent bunyaviruses. The compounds suppressed RNA hydrolysis by Cap-ENDOs, with IC50 values predominantly in the lower µM range. Molecular docking studies revealed the interactions with metal ions to be essential for the 2,3-dihydro-6,7-dihydroxy-1H-isoindol-1-one scaffold activity. Calorimetric analysis uncovered Mn2+ ions to have the highest affinity for sites within the targets, irrespective of aminoacidic variations influencing metal cofactor preferences. Interestingly, spectrophotometric findings unveiled sole dinuclear species formation between the scaffold and Mn2+. Moreover, the complexation of two Mn2+ ions within the viral enzymes appears to be favourable, as indicated by the binding of compound 11 to TOSV Cap-ENDO (Kd = 28 ± 3 µM). Additionally, the tendency of compound 11 to stabilize His+ more than His- Cap-ENDOs suggests exploitable differences in their catalytic pockets relevant to improving specificity. Collectively, our results underscore the isoindolinone scaffold's potential as a strategic starting point for the design of pan-antibunyavirus drugs.

Structural and functional characterization of the Sin Nombre virus L protein.

The Bunyavirales order is a large and diverse group of segmented negative-strand RNA viruses. Several virus families within this order contain important human pathogens, including Sin Nombre virus (SNV) of the Hantaviridae. Despite the high epidemic potential of bunyaviruses, specific medical countermeasures such as vaccines or antivirals are missing. The multifunctional ~250 kDa L protein of hantaviruses, amongst other functional domains, harbors the RNA-dependent RNA polymerase (RdRp) and an endonuclease and catalyzes transcription as well as replication of the viral RNA genome, making it a promising therapeutic target. The development of inhibitors targeting these key processes requires a profound understanding of the catalytic mechanisms. Here, we established expression and purification protocols of the full-length SNV L protein bearing the endonuclease mutation K124A. We applied different biochemical in vitro assays to provide an extensive characterization of the different enzymatic functions as well as the capacity of the hantavirus L protein to interact with the viral RNA. By using single-particle cryo-EM, we obtained a 3D model including the L protein core region containing the RdRp, in complex with the 5' promoter RNA. This first high-resolution model of a New World hantavirus L protein shows striking similarity to related bunyavirus L proteins. The interaction of the L protein with the 5' RNA observed in the structural model confirms our hypothesis of protein-RNA binding based on our biochemical data. Taken together, this study provides an excellent basis for future structural and functional studies on the hantavirus L protein and for the development of antiviral compounds.

The mechanism of genome replication and transcription in bunyaviruses.

Bunyaviruses are negative sense, single-strand RNA viruses that infect a wide range of vertebrate, invertebrate and plant hosts. WHO lists three bunyavirus diseases as priority diseases requiring urgent development of medical countermeasures highlighting their high epidemic potential. While the viral large (L) protein containing the RNA-dependent RNA polymerase is a key enzyme in the viral replication cycle and therefore a suitable drug target, our knowledge on the structure and activities of this multifunctional protein has, until recently, been very limited. However, in the last few years, facilitated by the technical advances in the field of cryogenic electron microscopy, many structures of bunyavirus L proteins have been solved. These structures significantly enhance our mechanistic understanding of bunyavirus genome replication and transcription processes and highlight differences and commonalities between the L proteins of different bunyavirus families. Here, we provide a review of our current understanding of genome replication and transcription in bunyaviruses with a focus on the viral L protein. Further, we compare within bunyaviruses and with the related influenza virus polymerase complex and highlight open questions.

A convenient in vivo cap donor delivery system to investigate the cap snatching of plant bunyaviruses.

Like their animal-infecting counterparts, plant bunyaviruses use capped RNA leaders cleaved from host cellular mRNAs to prime viral genome transcription in a process called cap-snatching, but in vivo systems to investigate the details of this process are lacking for them. Here, we report that Rice stripe tenuivirus (RSV) and Tomato spotted wilt tospovirus (TSWV) cleave capped RNA leaders from mRNAs transiently expressed by agroinfiltration, which makes it possible to artificially deliver defined cap donors to the two plant bunyaviruses with unprecedented convenience. With this system, some ideas regarding how plant bunyaviruses select and use capped RNA leaders can be tested easily. We were also able to obtain clear evidence that the capped RNA leaders selected by TSWV are generally longer than those by RSV. TSWV frequently uses the prime-and-realign mechanism in transcription primed by capped RNA leaders shorter than a certain length, like that has been demonstrated recently for RSV.

The Orthotospovirus nonstructural protein NSs suppresses plant MYC-regulated jasmonate signaling leading to enhanced vector attraction and performance.

Pandemics of vector-borne human and plant diseases often depend on the behaviors of their arthropod vectors. Arboviruses, including many bunyaviruses, manipulate vector behavior to accelerate their own transmission to vertebrates, birds, insects, and plants. However, the molecular mechanism underlying this manipulation remains elusive. Here, we report that the non-structural protein NSs of Tomato spotted wilt orthotospovirus, a prototype of the Tospoviridae family and the Orthotospovirus genus, is a key viral factor that indirectly modifies vector preference and increases vector performance. NSs suppresses the biosynthesis of plant volatile monoterpenes, which serve as repellents of the vector western flower thrips (WFT, Frankliniella occidentalis). NSs directly interacts with MYC2, the jasmonate (JA) signaling master regulator and its two close homologs MYC3 and MYC4, to disable JA-mediated activation of terpene synthase genes. The dysfunction of the MYCs subsequently attenuates host defenses, increases the attraction of thrips, and improves thrips fitness. Moreover, MYC2 associated with NSs of Tomato zonate spot orthotospovirus, another Euro/Asian-type orthotospovirus, suggesting that MYC2 is an evolutionarily conserved target of Orthotospovirus species for suppression of terpene-based resistance to promote vector performance. These findings elucidate the molecular mechanism through which an orthotospovirus indirectly manipulates vector behaviors and therefore facilitates pathogen transmission. Our results provide insights into the molecular mechanisms by which Orthotospovirus NSs counteracts plant immunity for pathogen transmission.

Bunyaviridae RdRps: structure, motifs, and RNA synthesis machinery.

Bunyaviridae family is the largest and most diverse family of RNA viruses. It has more than 350 members divided into five genera: Orthobunyavirus, Phlebovirus, Nairovirus, Hantavirus, and Tospovirus. They are present in the five continents, causing recurrent epidemics, epizootics, and considerable agricultural loss. The genome of bunyaviruses is divided into three segments of negative single-stranded RNA according to their relative size: L (Large), M (Medium) and S (Small) segment. Bunyaviridae RNA-dependent RNA polymerase (RdRp) is encoded by the L segment, and is in charge of the replication and transcription of the viral RNA in the cytoplasm of the infected cell. Viral RdRps share a characteristic right hand-like structure with three subdomains: finger, palm, and thumb subdomains that define the formation of the catalytic cavity. In addition to the N-terminal endonuclease domain, eight conserved motifs (A-H) have been identified in the RdRp of Bunyaviridae. In this review, we have summarized the recent insights from the structural and functional studies of RdRp to understand the roles of different motifs shared by RdRps, the mechanism of viral RNA replication, genome segment packaging by the nucleoprotein, cap-snatching, mRNA transcription, and other RNA mechanisms of bunyaviruses.

In vivo localization of iris yellow spot tospovirus (Bunyaviridae)-encoded proteins and identification of interacting regions of nucleocapsid and movement proteins.

BACKGROUND: Localization and interaction studies of viral proteins provide important information about their replication in their host plants. Tospoviruses (Family Bunyaviridae) are economically important viruses affecting numerous field and horticultural crops. Iris yellow spot virus (IYSV), one of the tospoviruses, has recently emerged as an important viral pathogen of Allium spp. in many parts of the world. We studied the in vivo localization and interaction patterns of the IYSV proteins in uninfected and infected Nicotiana benthamiana and identified the interacting partners. PRINCIPAL FINDINGS: Bimolecular fluorescence complementation (BiFC) analysis demonstrated homotypic and heterotypic interactions between IYSV nucleocapsid (N) and movement (NSm) proteins. These interactions were further confirmed by pull-down assays. Additionally, interacting regions of IYSV N and NSm were identified by the yeast-2-hybrid system and ß-galactosidase assay. The N protein self-association was found to be mediated through the N- and C-terminal regions making head to tail interaction. Self-interaction of IYSV NSm was shown to occur through multiple interacting regions. In yeast-2-hybrid assay, the N- and C-terminal regions of IYSV N protein interacted with an N-terminal region of IYSV NSm protein. CONCLUSION/SIGNIFICANCE: Our studies provide new insights into localization and interactions of IYSV N and NSm proteins. Molecular basis of these interactions was studied and is discussed in the context of tospovirus assembly, replication, and infection processes.

The N-terminal domain of the arenavirus L protein is an RNA endonuclease essential in mRNA transcription.

Arenaviridae synthesize viral mRNAs using short capped primers presumably acquired from cellular transcripts by a 'cap-snatching' mechanism. Here, we report the crystal structure and functional characterization of the N-terminal 196 residues (NL1) of the L protein from the prototypic arenavirus: lymphocytic choriomeningitis virus. The NL1 domain is able to bind and cleave RNA. The 2.13 Å resolution crystal structure of NL1 reveals a type II endonuclease α/ß architecture similar to the N-terminal end of the influenza virus PA protein. Superimposition of both structures, mutagenesis and reverse genetics studies reveal a unique spatial arrangement of key active site residues related to the PD…(D/E)XK type II endonuclease signature sequence. We show that this endonuclease domain is conserved and active across the virus families Arenaviridae, Bunyaviridae and Orthomyxoviridae and propose that the arenavirus NL1 domain is the Arenaviridae cap-snatching endonuclease.

RNA binding domain of Jamestown Canyon virus S segment RNAs.

Jamestown Canyon virus (JCV) is a member of the Bunyaviridae family, Orthobunyavirus genus, California serogroup. Replication and, ultimately, assembly and packaging rely on the process of encapsidation. Therefore, the ability of viral RNAs (vRNAs) (genomic and antigenomic) to interact with the nucleocapsid protein (N protein) and the location of this binding domain on the RNAs are of interest. The questions to be addressed are the following. Where is the binding domain located on both the vRNA and cRNA strands, is this RNA bound when double or single stranded, and does this identified region have the ability to transform the binding potential of nonviral RNA? Full-length viral and complementary S segment RNA, as well as 3' deletion mutants of both vRNA and cRNA, nonviral RNA, and hybrid viral/nonviral RNA, were analyzed for their ability to interact with bacterially expressed JCV N protein. RNA-nucleocapsid interactions were examined by UV cross-linking, filter binding assays, and the generation of hybrid RNA to help define the area responsible for RNA-protein binding. The assays identified the region responsible for binding to the nucleocapsid as being contained within the 5' half of both the genomic and antigenomic RNAs. This region, if placed within nonviral RNA, is capable of altering the binding potential of nonviral RNA to levels seen with wild-type vRNAs.

Interferon-induced Mx proteins in antiviral host defense.

Mx proteins are key components of the antiviral state induced by interferons in many species. They belong to the class of dynamin-like large guanosine triphosphatases (GTPases) known to be involved in intracellular vesicle trafficking and organelle homeostasis. Mx GTPases share structural and functional properties with dynamin, such as self-assembly and association with intracellular membranes. A unique property of some Mx GTPases is their antiviral activity against a wide range of RNA viruses, including influenza viruses and members of the bunyavirus family. These viruses are inhibited at an early stage in their life cycle, soon after host cell entry and before genome amplification. The mouse Mx1 GTPase accumulates in the cell nucleus where it associates with components of the PML nuclear bodies and inhibits influenza and Thogoto viruses known to replicate in the nucleus. The human MxA GTPase accumulates in the cytoplasm and is partly associated with a COP-I-positive subcompartment of the endoplasmic reticulum. This membrane compartment seems to provide an interaction platform that facilitates viral target recognition. In the case of bunyaviruses, MxA recognizes the viral nucleocapsid protein and interferes with its role in viral genome replication. In the case of Thogoto virus, MxA recognizes the viral nucleoprotein and prevents the incoming viral nucleocapsids from being transported into the nucleus, the site of viral transcription and replication. In both cases, GTP-binding and carboxy-terminal effector functions of MxA are required for target recognition. In general, Mx GTPases appear to detect viral infection by sensing nucleocapsid-like structures. As a consequence, these viral components are trapped and sorted to locations where they become unavailable for the generation of new virus particles.

Assessing the genetic component of the susceptibility of mice to viral infections.

Laboratory mice often exhibit wide differences in susceptibility when infected experimentally with viruses. Based on such observations, experiments have been designed to investigate the determinism of these differences at the molecular level, and a few genes that play a major role in the innate mechanisms of defence of the species toward viral aggressions have been characterised. For example, the extraordinary resistance of SJL mice to experimental infections with hepatitis virus strain A59 is the consequence of a structural alteration of a cell adhesion molecule which normally binds to the spikes of the virus, allowing its entry into the cells. If the virus cannot bind to the molecule, or if the molecule is absent, epithelial cells of the intestine and liver are not infected and mice are resistant. In the same way, most--not to say all--laboratory strains of mice are susceptible to infections with orthomyxoviruses or flaviviruses because essential molecules, the synthesis of which is normally triggered by interferon, are defective in these mice. Wild mice, by contrast --probably because they are constantly exposed to natural infections--are resistant. Finally, some mouse strains resist experimental infections by the mouse cytomegalovirus 1 (MCMV-1) because, once infected, these mice synthesise a molecule at the surface of infected cells which allows immediate recognition and killing by natural killer (NK) cells. With the exuberant development of mouse genetics and the constant generation of new mutant alleles, it is likely that many more genes with an impact on the phenotype of resistance or susceptibility will be identified in the forthcoming years. These genes are probably numerous, however, and many of them presumably interact with each other and/or have additive effects. This might slow down progress in our understanding of the innate mechanism of defence.

Proteomics computational analyses suggest that the carboxyl terminal glycoproteins of Bunyaviruses are class II viral fusion protein (beta-penetrenes).

The Bunyaviridae family of enveloped RNA viruses includes five genuses, orthobunyaviruses, hantaviruses, phleboviruses, nairoviruses and tospoviruses. It has not been determined which Bunyavirus protein mediates virion:cell membrane fusion. Class II viral fusion proteins (beta-penetrenes), encoded by members of the Alphaviridae and Flaviviridae, are comprised of three antiparallel beta sheet domains with an internal fusion peptide located at the end of domain II. Proteomics computational analyses indicate that the carboxyl terminal glycoprotein (Gc) encoded by Sandfly fever virus (SAN), a phlebovirus, has a significant amino acid sequence similarity with envelope protein 1 (E1), the class II fusion protein of Sindbis virus (SIN), an Alphavirus. Similar sequences and common structural/functional motifs, including domains with a high propensity to interface with bilayer membranes, are located collinearly in SAN Gc and SIN E1. Gc encoded by members of each Bunyavirus genus share several sequence and structural motifs. These results suggest that Gc of Bunyaviridae, and similar proteins of Tenuiviruses and a group of Caenorhabditis elegans retroviruses, are class II viral fusion proteins. Comparisons of divergent viral fusion proteins can reveal features essential for virion:cell fusion, and suggest drug and vaccine strategies.

Inhibition of translation and induction of apoptosis by Bunyaviral nonstructural proteins bearing sequence similarity to reaper.

Members of the California serogroup of bunyaviruses (family Bunyaviridae) are the leading cause of pediatric viral encephalitis in North America. Significant cell death is observed as part of the infection pathology. We now report that a Bunyaviral nonstructural protein termed NSs shows sequence similarity to Reaper, a proapoptotic protein from Drosophila. Although NSs proteins lack the Reaper N-terminal motif critical for IAP inhibition, they do retain other functions of Reaper that map to conserved C-terminal regions. Like Reaper, NSs proteins induce mitochondrial cytochrome c release and caspase activation in cell-free extracts and promote neuronal apoptosis and mortality in a mouse model. Independent of caspase activation, Bunyavirus NSs proteins also share with Reaper the ability to directly inhibit cellular protein translation. We have found that the shared capacity to inhibit translation and induce apoptosis resides in common sequence motifs present in both Reaper and NSs proteins. Data presented here suggest that NSs induce apoptosis through a mechanism similar to that used by Reaper, as both proteins bind to an apoptotic regulator called Scythe and can relieve Scythe inhibition of Hsp70. Thus, bunyavirus NSs proteins have multiple Reaper-like functions that likely contribute to viral pathogenesis by promoting cell death and/or inhibiting cellular translation.

Transient association of calnexin and calreticulin with newly synthesized G1 and G2 glycoproteins of uukuniemi virus (family Bunyaviridae).

The membrane glycoproteins G1 and G2 of Uukuniemi virus, a member of the Bunyaviridae family, are cotranslationally cleaved from a common precursor in the endoplasmic reticulum (ER). Here, we show that newly made G1 and G2 associate transiently with calnexin and calreticulin, two lectins involved in glycoprotein folding in the ER. Stable complexes between G1-G2 and calnexin or calreticulin could be immunoprecipitated after solubilization of virus-infected BHK21 cells with the detergents digitonin or Triton X-100. In addition, G1-G2-calnexin complexes could be recovered after solubilization with CHAPS (3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate), while G1-G2-calreticulin complexes were not readily detected by using this detergent. Only endoglycosidase H-sensitive forms of G1 were found complexed with calnexin. Pulse-chase experiments showed that G1 and G2 associated with both chaperones transiently for up to 120 min. Sequential immunoprecipitations with anticalreticulin and anticalnexin antisera indicated that about 50% of newly synthesized G1 and G2 was associated with either calnexin or calreticulin. Our previous results have shown that newly synthesized G1 and G2 transiently interact also with the ER chaperone BiP and with protein disulfide isomerase (R. Persson and R. F. Pettersson, J. Cell Biol. 112:257-266, 1991). Taking all of this into consideration, we conclude that the folding of G1 and G2 in the ER is catalyzed by at least four different folding factors.

Molecular determinants of Golgi retention in the Punta Toro virus G1 protein.

The G1 glycoprotein of Punta Toro virus, a member of the bunyavirus family, accumulates in the Golgi complex when it is expressed from cloned cDNA. We previously reported that the information necessary for Golgi retention of the G1 protein is located within the transmembrane domain and a portion of the cytoplasmic domain adjacent to the transmembrane domain (Matsuoka, Y., Chen, S.-Y., and Compans, R. W. (1994) J. Biol. Chem. 269, 22565-22573). To determine the features of the amino acid sequence motif required for Golgi retention, we have introduced mutations including truncations and point mutations in the transmembrane and the cytoplasmic domains and examined the cellular localization of the expressed mutant proteins. The results from truncation mutants indicate that the crucial information appears to be located within the first 10 amino acids of the cytoplasmic domain. Within this region, mutation of a proline residue yielded a protein that was transported to the cell surface. A protein was also expressed on the cell surface when one of the threonine residues in the transmembrane domain was changed to leucine. Thus the transmembrane domain may have a supportive role in Golgi retention, possibly by promoting protein interactions through hydroxylated side chains.

The 5' ends of Hantaan virus (Bunyaviridae) RNAs suggest a prime-and-realign mechanism for the initiation of RNA synthesis.

We examined the 5' ends of Hantaan virus (HTN) genomes and mRNAs to gain insight into the manner in which these chains were initiated. Like those of all members of the family Bunyaviridae described so far, the HTN mRNAs contained 5' terminal extensions that were heterogeneous in both length and sequence, presumably because HTN also "cap snatches" host mRNAs to initiate the viral mRNAs. Unexpectedly, however, almost all of the mRNAs contained a G residue at position -1, and a large fraction also lacked precisely one of the three UAG repeats at the termini. The genomes, on the other hand, commenced with a U residue at position +1, but only 5' monophosphates were found here, indicating that these chains may not have initiated with UTP at this position. Taken together, these unusual findings suggest a prime-and-realign mechanism of chain initiation in which mRNAs are initiated with a G-terminated host cell primer and genomes with GTP, not at the 3' end of the genome template but internally (opposite the template C at position +3), and after extension by one or a few nucleotides, the nascent chain realigns backwards by virtue of the terminal sequence repeats, before processive elongation takes place. For genome initiation, an endonuclease, perhaps that involved in cap snatching, is postulated to remove the 5' terminal extension of the genome, leaving the 5' pU at position +1.

Expression of the Bunyamwera virus M genome segment and intracellular localization of NSm.

Bunyamwera (BUN) virus is the prototype of the family Bunyaviridae and contains a trisegmented, single-stranded RNA genome of negative polarity. The medium (M) RNA segment encodes the two virion glycoproteins, G1 and G2, and a nonstructural protein, NSm, in the form of a polyprotein precursor which is cotranslationally cleaved. The gene order of the M segment is 5' G2-NSm-G1 3'. We have raised a monospecific antiserum in rabbits to a branched chain synthetic peptide to a region of the NSm protein which specifically immunoprecipitates NSm from BUN-infected cells. Indirect immunofluorescence experiments on BUN-infected cells using this antiserum gave a perinuclear staining pattern, suggesting that like the viral structural proteins, NSm localizes to the Golgi complex. An essentially full-length M segment cDNA was cloned into a recombinant vaccinia virus under control of bacteriophage T7 promoter and terminator sequences and expressed in cells co-infected with a second recombinant vaccinia virus which synthesizes T7 RNA polymerase. G1, G2, and NSm were detected in cells dually infected with the recombinant vaccina viruses, indicating that processing of the M segment-encoded precursor does not require other BUN proteins. Immunofluorescence experiments showed that the BUN glycoproteins expressed from this recombinant vaccinia virus system localized to the Golgi complex like authentic BUN proteins.

Golgi complex localization of the Punta Toro virus G2 protein requires its association with the G1 protein.

The glycoproteins of bunyaviruses accumulate in membranes of the Golgi complex, where virus maturation occurs by budding. In this study we have constructed a series of full length or truncated mutants of the G2 glycoprotein of Punta Toro virus (PTV), a member of the Phlebovirus genus of the Bunyaviridae, and investigated their transport properties. The results indicate that the hydrophobic domain preceding the G2 glycoprotein can function as a translocational signal peptide, and that the hydrophobic domain near the C-terminus serves as a membrane anchor. A G2 glycoprotein construct with an extra hydrophobic sequence derived from the N-terminal NSM region was stably retained in the ER, and was unable to be transported to the Golgi complex. The full-length G2 glycoprotein, when expressed on its own, was transported out of the ER and expressed on the cell surface, whereas the G1 and G2 proteins when expressed together are retained in the Golgi complex. A truncated anchor-minus form of the G2 glycoprotein was found to be secreted into the culture medium, but was retained in the Golgi complex when coexpressed with the G1 glycoprotein. These results indicate that the G2 membrane glycoprotein is a class I membrane protein which does not contain a signal sufficient for Golgi retention, and suggest that its Golgi localization is a result of association with the G1 glycoprotein.

Effect of human, recombinant interleukin 2 on Punta Toro virus infections in C57BL/6 mice.

The effect of human recombinant interleukin-2 (rIL-2) on Punta Toro virus (PTV) infection was investigated in C57BL/6 mice. Immunologic and viral parameters were assessed after mice were treated i.p. with rIL-2 for 5 days. Treatment of mice with 25000 and 12500 units/mouse of rIL-2 resulted in significant inhibition of the disease as indicated by increases in survival of mice as well as decreases in liver and serum virus titers. Serum glutamic oxalic acid and pyruvic acid transaminase levels were also lowered indicating reduced liver damage. Murine IL-2 production returned to normal or above-normal levels in rIL-2 treated mice. Natural killer cell activity was also moderately stimulated by rIL-2 treatment. Significant amounts of interferon were not detected in the sera of treated mice. Weight gain and survival rates were similar for both toxicity and normal controls indicating that rIL-2 treatments had no toxic effect.

Role of La Crosse virus glycoproteins in attachment of virus to host cells.

Data presented in this report demonstrate that the initial event of La Crosse virus (LACV) infection of cells is probably the interaction of viral glycoproteins with specific cellular receptor sites. We have shown that LACV glycoprotein G1 binds, in a dose-dependent manner, to continuous vertebrate and mosquito cell lines, but not to mosquito midguts isolated ex vivo. This binding can be inhibited by the pretreatment of cells with excess homologous glycoprotein but not with excess heterologous LACV glycoprotein. In contrast, we have shown that LACV glycoprotein G2 binds to the continuous mosquito cell line and vector midgut cells, but not to vertebrate cells. LACV infection of vertebrate cells can be inhibited by treatment of cells with purified G1, while infection in mosquito cells can be reduced by treatment of cells with a combination of G1 and G2. The results suggest that G1 is the viral attachment protein (VAP) for vertebrate cells, and that G2 serves the same purpose for mosquito midgut cells. We speculate that the protease-resistant G2 molecule may have evolved to serve as the VAP in the midgut under conditions in which G1 might be altered or removed from the virus envelope, and thus is essential to the evolution and maintenance of vertebrate-invertebrate transmission cycles.