key: cord-0005538-20ocbzp6
authors: Wang, Jizhen; Manicassamy, Balaji; Caffrey, Michael; Rong, Lijun
title: Characterization of the receptor-binding domain of Ebola glycoprotein in viral entry
date: 2011-06-12
journal: Virol Sin
DOI: 10.1007/s12250-011-3194-9
sha: 9fae9474725400c9be6f28818e571b7adb3ceba6
doc_id: 5538
cord_uid: 20ocbzp6

Ebola virus infection causes severe hemorrhagic fever in human and non-human primates with high mortality. Viral entry/infection is initiated by binding of glycoprotein GP protein on Ebola virion to host cells, followed by fusion of virus-cell membrane also mediated by GP. Using an human immunodeficiency virus (HIV)-based pseudotyping system, the roles of 41 Ebola GP1 residues in the receptor-binding domain in viral entry were studied by alanine scanning substitutions. We identified that four residues appear to be involved in protein folding/structure and four residues are important for viral entry. An improved entry interference assay was developed and used to study the role of these residues that are important for viral entry. It was found that R64 and K95 are involved in receptor binding. In contrast, some residues such as I170 are important for viral entry, but do not play a major role in receptor binding as indicated by entry interference assay and/or protein binding data, suggesting that these residues are involved in post-binding steps of viral entry. Furthermore, our results also suggested that Ebola and Marburg viruses share a common cellular molecule for entry.

Ebola viruses are enveloped viruses with an ~19 kb single-stranded RNA genome containing 7 genes. The GP gene encodes two glycoproteins, a secreted form called sGP and a transmembrane form called GP which is generated by RNA editing [34, 35, 44] . GP is responsible for viral entry, including attachment of viruses to the target cells and fusion of the virus-cell membranes [43, 48, 49] . GP is modified by N-linked glycosylation in the Endoplasmic Reticulum to form PreGP, and further undergoes O-linked glycosylation in Golgi to become mature GP [10] .

Ebola GP processing by proteases has been well documented. First, the N-terminal signal peptide (32 residues) of GP is cleaved cotranslationally. The GP precursor is then cleaved into GP1 and GP2 in trans-Golgi by a furin-like cellular protease, and GP1 and GP2 are linked by disulfide bonding [14, 36, 45] .

Although the cleavage site between GP1 and GP2 is conserved among the Ebola species, this cleavage event appears not required for Ebola entry [13, 28, 29, 49] .

Lastly, GP was shown to be cleaved at D637 by the tumor necrosis factor α-converting enzyme (TACE) to shed the truncated GP from the surface of infected cells [7] . In addition to these cleavage events during GP maturation, two intracellular proteases, Cathepsin B and L, residing in the endosomes, have been shown to process GP1 into an 18-19 kDa fragment during viral entry [5, 16, 37] . The endosomal cleavage appears to be critical on viral entry. It is interesting to note that these endosomal proteases play a role in entry of other viruses [9, 12, 30, 31, 39] Many cellular proteins have been implicated in facilitating Ebola entry, including asialoglycoprotein receptor, folate receptor α, dendritic cell-specific ICAM3 grabbing non-integrin, liver/lymph node-specific ICAM3 grabbing non-integrin, human macrophage galactose and N-acetylgalactosamine-specific C-type lectin, and Tyro3 receptor kinase family [1, 4, 8, 22, 32, 33, 38, 42] .

However, none of them seems to be absolutely required for the entry. Interestingly, expression of Ebola GP on the target cells can specifically enhance Ebola GP-mediated viral entry [23] .

The N-terminal 200 residues of GP1 are relatively conserved compared to the highly variable C-terminal mucin-like domain (MUC, amino acids 309~476). The MUC region is heavily modified by O-and N-linked glycosylation. It has been shown that MUC is responsible for the cytopathic effects induced by GP expression, but it is dispensable for GP-mediated viral entry, at least in tissue culture using pseudotyping systems [14, 24, 40, 50] . Previous studies have shown that the N-terminal region of GP1 (approximately 150 residues in length after cleavage of the signal peptide)

is critical for receptor binding, which is referred to as the receptor-binding domain, or RBD [3, 24] . Extensive mutational analyses have been performed to characterize the roles of RBD residues in protein folding and function in viral entry. In this report, we examined the roles of the RBD residues by substitutions in protein expression, virion incorporation, and viral entry [3, 14, 24, 26] .

By entry interference assay and binding assay, we identified several key residues involved in receptor binding and the post-binding steps in viral entry. In addition, the results further substantiated the notion that Ebola and Marburg viruses use the same or similar receptor.

Human embryonic kidney cells 293T were main- The mouse monoclonal antibody, 12B5-1-1, which recognizes the GP1 of Ebola Zaire GP, was kindly provided by Dr. Mary K. Hart [47] . The mouse anti-HIV p24 monoclonal antibody was obtained from NIH AIDS Research and Reference Reagent Program [41] .

Mouse anti-β actin antibody, goat anti-mouse IgG HRP, and anti-human IgG FITC were purchased from Sigma (St. Louis).

Ebola Zaire glycoprotein gene was synthesized by multiple rounds of overlapping PCR based on the Ebola Zaire genome sequence (GenBank accession number L11365 [15] ). ∆Mucin GP was constructed by PCR-directed mutagenesis. All alanine substitution mutations of GP gene were generated by site-directed mutagenesis using Stratagene's Quick-Change Mutagenesis kit according to supplier's protocols.

All mutations were confirmed by DNA sequencing of flanking regions (on average approximately 500 bp).

Pseudotyped viruses were produced by cotansfecting DNA of wt GP or mutants with Envdeficient HIV vector carrying a firefly luciferase reporter gene [6, 11] into producer cells. Briefly, 2μg DNA of wt GP or mutants and 2μg of pNL4-3-Luc-R --Ewere used to transfect 293T cells (90% 

To evaluate the incorporation of wt GP protein or mutants into the pseudotyped viruses, 2 mL of Ebola GP protein or mutants was detected by Western blotting as described above. Mouse anti-HIV p24 monoclonal antibody (1:5 000 dilution) was used as the primary antibody to detect the HIV p24 protein.

Human 293T cells (3×10 5 cells) were seeded in six-well plates one day prior to infection. These 

To construct plasmids carrying a Δmucin GP gene (with mutation), PshAI-BamHI fragment of fulllength Ebola GP gene (with mutation) in pcDNA3.1(+) was replaced by PshAI-BamH I fragment of Δmucin Ebola GP gene. Δmucin GP gene with/without mutation in these plasmids was amplified by PCR and inserted between BamH I and Xho I sites of the HIV vector, pHR'-CMV-Luc [equal to pHR' in [27] ], to replace luciferase gene. GFP or puromycin-resistant gene (puro) was used to replace the luciferase gene in the same vector to construct pHR'-CMV-GFP and pHR'-CMV-puro.

Human 293T cells in 6-well plate were transfected with 0.25 μg of VSV-G plasmid encoding VSV-G protein, 1μg of HIV-trans plamsid encoding HIV structural protein gag-pol [equal to pCMVΔR8.2 in [27] ], and 1 μg of pHR'.CMV plasmid carrying different genes (GFP, puro, or wt/mutant Δmucin GP).

The supernatants containing pseudotyped viruses were 

To study direct binding of Ebola GP to cells, the gene encoding a fusion protein (Ebola receptorbinding domain [RBD] fused to Fc of human IgG) was constructed and the fusion protein was purified.

Nhe I-BamH I fragment of S1-hIgG [17] 

Previous work on the conserved hydrophobic and charged residues in GP1 has shown that most of the residues involved in protein folding or viral entry are at the N-terminus. This suggests that the N-terminal region of approximately 150 residues (aa consists of an important domain of GP, receptorbinding domain, or RBD [24] . To further dissect RBD in viral entry, mutational analysis was carried out to study other residues in this domain. In total, 41 single alanine substitutions were generated on Ebola GP.

The wild type (wt) or mutant GP plasmid was cotransfected with an Env-deficient HIV vector (pNL4-3

Luc.R-E-) into 293T cells to produce pseudotyped viruses as described [24] . 293T cells were challenged by these viruses. Infected cells were lysed 48 h post-infection, and luciferase activities were measured as an indicator of entry efficiency of different pseudotyped viruses, as previously described [24] .

Western blot analysis was used to determine protein expression in 293T cells and GP incorporation on HIV particles.

Only four of the forty-one substitutions (V52A, V66A, F160A, and V181A) appeared to show low levels of protein expression and/or virion incorporation, which led to impaired GP phenotype in mediating viral entry (Fig. 1) . The lower amount of V52A, V66A, and V181A GP incorporated onto virions compared to that of wt GP could explain the inefficient viral entry (29%, 44%, and 15% of wt).

Low expression of F160A GP led to low virion incorporation and viral entry (2% of wt). In contrast, none of the remaining 37 mutant GPs showed a major defect in protein expression and GP incorporation.

Among them, two (T60A and V79A) were defective in mediating viral entry (~30% of wt), suggesting that T60 and V79 are involved in receptor binding or post-binding steps. In addition, two substitutions (T42A and N61A) did not have major effect on GP expression and viral incorporation, but they appeared to be somewhat defective in mediating viral entry (~40% of wt, Fig. 1 C and D) . In contrast, the remaining 33 substitutions did not greatly impair the ability of GP in mediating viral entry (more than 60% of wt). Indeed six of them (V96A, S119A, F151A, E156A, S167A, and T168A) displayed enhanced entry (over 200% of wt).

Among the 41 substitution mutants of GP characteri- Previously, we showed that four substitution mutants in this region (L43A, R54A, K56A, and E100A) displayed similar phenotypes, that is, they were expressed and incorporated onto virions like wt GP, but somewhat defective in mediating viral entry [24] . To further analyze the role of these residues in viral entry, fifteen double substitution mutants were generated.

Each of these mutants was characterized in protein expression, viral incorporation, and in mediating viral entry. Five mutants (V79A/L43A, V79A/L51A, V79A/ R54A, V79A/E100A, and T60A/L51A) displayed defects either in protein expression or incorporation onto viral particles ( Fig. 2A) , thus. leading to the reduced ability of GP in mediating viral entry (Fig. 2B) . In contrast, the remaining 10 double substitution mutants did not display detectable defects in protein expression or viral incorporation, but they were impaired in mediating viral entry, ranging from 2% to 44% of wt GP (Fig. 2 

We have previously shown that expression of Δmucin Ebola GP in 293T cells by transient transfection can inhibit Ebola GP-mediated viral entry [25] . The choice of Δmucin Ebola GP instead of expressed as % wt. As expected, expression of wt or mutant Ebola GP proteins did not greatly affect VSV-G-mediated viral entry (Fig. 4B) . In contrast, entry of Ebola GP pseudovirions was decreased dramatically in cells expressing either Δmucin or GP mutants (Fig. 4B) (Fig. 4C) . Furthermore, the overall inhibitory profile of viral entry mediated by Marburg GP was highly similar to that by Ebola GP. Two mutants (R64A and K95A), which were impaired in inhibiting viral entry mediated by Ebola GP, were also less effective in inhibiting viral entry mediated by Marburg GP. These results substantiate the notion that Ebola and Marburg viruses share a common receptor/co-receptor for viral entry. These results further strengthen the argument that R64 and K95 are involved in direct interaction with the cellular receptor.

To further characterize the role of R64 and K95 in receptor binding, we performed a binding assay of the recombinant Ebola GP proteins following a published protocol [17] . We constructed and expressed the wt and mutant Ebola GP RBD-hIgG fusion constructs (Fig.   5A ). The purified proteins were used to bind the target cells (293T). As expected, the control protein S1-hIgG displayed little binding to 293T cells, as determined by flow cytometry (Fig. 5B ). In contrast, wt Ebola RBD-hIgG bound to the target cells (Fig. 5B) , in a dose-dependent manner (data not shown), consistent with the previous finding [17] . Mutant I170A bound to the target cells at a level similar to that of wt.

However, mutants R64A and K95A bound to the target cells less efficiently compared to wt. These results provided direct evidence that R64 and K95 are involved in receptor binding. Since mutant I170A is similarly defective in mediating viral entry compared to R64A and K95A, but not defective in receptor binding, we conclude that I170 is involved in the post-binding steps during Ebola entry.

In this study, we focused on identification and analysis of critical residues of Ebola GP1 in receptor binding and post-binding steps. Previously we identified that residues 33 to 185 of the N-terminus consist of the receptor-binding domain, confirmed by others using a biochemical approach [17] . In this report, The role of Ebola GP residues in protein expression and viral entry was also studied by others. It has been shown that Y162A caused a defect in protein expression, virion incorporation, and entry [26] . Although it was generated on a ΔMucin GP (deletion of 315-504), the mutation displayed a similar phenotype to wt GP [24] . Consistent with our results, it has been shown that F88A rendered the infectious Ebola virus non-infectious [26] . Recently, another group studied Ebola GP function extensively using ΔMucin GP (deletion of aa 309-489) pseudotyped FIV viruses [3] . Ten mutations (T83A, S90A, V97A, N98A, T144A, E156A, F160A, T168A, V169A, and T174A) displayed similar phenotype to this work in virion incorporation and viral entry.

Study of Marburg GP-mediated viral entry showed similar results to our findings. K79, the conserved residue in Marburg GP corresponding to K95 of Ebola GP, has also been shown to be important for viral entry [25] . For the other receptor-binding residue (R64) and the putative receptor-binding residues of Ebola GP (L57, L63 and F88), the corresponding residues in Marburg GP are H48, L41, V47, and F72, respectively.

Alanine substitution of these residues in Marburg GP caused defects in protein expression, and/or virion incorporation [25] . Therefore it is more difficult to distinguish their roles in protein folding, receptorbinding, or post-binding steps.

Although a lot of host cell surface molecules have been implicated in Ebola entry, the receptor is not identified yet. Therefore we have used an entry interference assay to study the role of Ebola GP residues in receptor binding. Entry interference is one of the classical methods to determine whether different viruses use the same receptor. In infected cells, glycoprotein expressed by the virus can interact with its cognate receptor(s) to inhibit re-infection by the same virus or viruses using the same receptor.

Previously using entry interference, it was shown that expression of Δmucin Ebola GP blocked Ebola GP-mediated viral entry [25] . However the assay is less effective (with only 60% inhibition of viral entry) due to low efficiency and high variation of transient transfection. In this report, we improved the effectiveness (>90% inhibition of viral entry) and consistency using a retroviral vector to express GP protein in cells. It was shown that two alanine substitutions on Δmucin Ebola GP (R64A and K95A) were less efficient in inhibiting Ebola GP-mediated viral entry (31% and 18%) compared to wt (7%). This is consistent with the cell surface binding data. In were defective in inhibiting both Ebola and Marburg GP entry (Fig. 4 B and C) . Third, purified fusion proteins containing the Ebola RBD can block Marburg GP-mediated viral entry and vice versa [17] . In addition, small molecule compounds and their derivatives inhibited both Ebola and Marburg GP mediated viral entry while they displayed no effect on the VSV-G mediated entry [2, 51] . Taken together, these results strongly imply that the Marburg RBD adopts a similar structure as the Ebola RBD, and a common cellular factor is used by both Ebola and Marburg viruses for infection.

The results in this report and our previous study are summarized in Fig. 6 . Twelve residues (green) are important for viral entry. Among them, R64 and K95 are shown to be directly involved in receptor binding, while I170 plays a critical role in post-binding steps.

The roles of the remaining 9 residues in viral entry are not very clear at present, for example, they may be involved in either receptor-binding or the post-binding steps. Nevertheless, the results from the current study and the previous work by us and others, together with the x-ray structure of Ebola GP [18] , give insights to the structure and function of Ebola GP RBD.

C-Type Lectins DC-SIGN and L-SIGN Mediate Cellular Entry by Ebola Virus in cis and in trans

Identification of a small-molecule entry inhibitor for filoviruses

Ebola virus glycoprotein 1: identification of residues important for binding and postbinding events

Folate receptor-alpha is a cofactor for cellular entry by Marburg and Ebola viruses

Endosomal proteolysis of the Ebola virus glycoprotein is necessary for infection

Vpr is required for efficient replication of human immunodeficiency virus type-1 in mononuclear phagocytes

Ectodomain shedding of the glycoprotein GP of Ebola virus

Cell adhesion-dependent membrane trafficking of a binding partner for the ebolavirus glycoprotein is a determinant of viral entry

Cathepsin L and cathepsin B mediate reovirus disassembly in murine fibroblast cells

Biosynthesis and role of filoviral glycoproteins

Human immunodeficiency virus type 1 viral protein R (Vpr) arrests cells in the G2 phase of the cell cycle by inhibiting p34cdc2 activity

SARS coronavirus, but not human coronavirus NL63, utilizes cathepsin L to infect ACE2-expressing cells

Ebola virus glycoprotein: proteolytic processing, acylation, cell tropism, and detection of neutralizing antibodies

Covalent modifications of the ebola virus glycoprotein

The role of the charged residues of the GP2 helical regions in Ebola entry

Proteolysis of the Ebola virus glycoproteins enhances virus binding and infectivity

Conserved receptor-binding domains of Lake Victoria marburgvirus and Zaire ebolavirus bind a common receptor

Structure of the Ebola virus glycoprotein bound to an antibody from a human survivor

Marburg and Ebola viruses as aerosol threats

Fruit bats as reservoirs of Ebola virus

Multiple Ebola virus transmission events and rapid decline of central African wildlife

Differential N-linked glycosylation of human immunodeficiency virus and Ebola virus envelope glycoproteins modulates interactions with DC-SIGN and DC-SIGNR

Expression of Ebolavirus glycoprotein on the target cells enhances viral entry

Comprehensive Analysis of Ebola Virus GP1 in Viral Entry

Characterization of Marburg virus glycoprotein in viral entry

Identification of two amino acid residues on Ebola virus glycoprotein 1 critical for cell entry

Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector

Reverse genetics demonstrates that proteolytic processing of the Ebola virus glycoprotein is not essential for replication in cell culture

Proteolytic processing of the Ebola virus glycoprotein is not critical for Ebola virus replication in nonhuman primates

A mature and fusogenic form of the Nipah virus fusion protein requires proteolytic processing by cathepsin L

Endosomal proteolysis by cathepsins is necessary for murine coronavirus mouse hepatitis virus type 2 spike-mediated entry

Cellular entry of ebola virus involves uptake by a macropinocytosis-like mechanism and subsequent trafficking through early and late endosomes

Phosphoinositide-3 kinase-Akt pathway controls cellular entry of Ebola virus

Sequence analysis of the Ebola virus genome: organization, genetic elements, and comparison with the genome of Marburg virus

The virion glycoproteins of Ebola viruses are encoded in two reading frames and are expressed through transcriptional editing

Variation in the glycoprotein and VP35 genes of Marburg virus strains

Role of endosomal cathepsins in entry mediated by the ebola virus glycoprotein

Tyro3 family-mediated cell entry of ebola and marburg viruses

Inhibitors of cathepsin L prevent severe acute respiratory syndrome coronavirus entry

Ebola virus glycoproteins induce global surface protein down-modulation and loss of cell adherence

The Vif and Gag proteins of human immunodeficiency virus type 1 colocalize in infected human T cells

Human macrophage C-type lectin specific for galactose and N-acetylgalactosamine promotes filovirus entry

A system for functional analysis of Ebola virus glycoprotein

GP mRNA of Ebola virus is edited by the Ebola virus polymerase and by T7 and vaccinia virus polymerases

Processing of the Ebola virus glycoprotein by the proprotein convertase furin

Antiviral activity of a small-molecule inhibitor of filovirus infection

Vaccine potential of Ebola virus VP24, VP30, VP35, and VP40 proteins

Characterization of Ebola virus entry by using pseudotyped viruses: Identification of receptor deficient cell lines

Endoproteolytic processing of the ebola virus envelope glycoprotein: cleavage is not required for function

Identification of the Ebola virus glycoprotein as the main viral determinant of vascular cell cytotoxicity and injury

Discovery, synthesis, and biological evaluation of a novel group of selective inhibitors of filoviral entry

The laboratory research was supported by National Institutes of Health grant AI 059570 and AI077767. J.W was partially supported by a University of Illinois at Chicago Fellowship