Targeted cleavage site mutations in the Gn precursor enable efficient generation of replication-competent rVSV-based surrogates for emerging nairoviruses.

Introduction

The continuous emergence of nairoviruses poses a growing threat to global health [1]. Nairoviruses belong to theNairoviridaefamily in the orderHareaviralesof theBunyaviricetesclass [2]. These viruses are primarily tick-borne, with ticks transmitting them to birds and mammals [3]. They are negative-sense, single-stranded RNA viruses with a tri-segmented genome. Crimean Congo Hemorrhagic Fever virus (CCHFV) is an orthonairovirus of medical importance and a priority pathogen due to its pandemic potential, because it causes severe illness with a case fatality rate of 5-30% or higher and can result in human-to-human transmission [4]. CCHF cases have been reported in Africa, the Middle East, Asia, and Europe, corresponding to the distribution of its primary vector, hard ticks of the genusHyalomma[5]. Humans commonly become infected by tick bites or direct contact with infectious body fluids from infected livestock or humans [6]. Beyond the well-recognized threat posed by CCHFV, several emerging nairoviruses have been reported in recent years, including Yezo [7], Beiji [8], Tacheng tick virus 1 [9], Wetland [10], and Songling viruses [11], all of which are associated with febrile illness in humans. Yezo virus (YEZV) was recently discovered in Japan, where human infections have been directly linked to tick bites [12,13]. Since its discovery in 2021, several cases of YEZV-associated febrile illness have been reported in China [13,14] and Japan [7,15].

The three genomic segments of orthonairoviruses, small (S), medium (M), and large (L), encode the nucleoprotein (NP), glycoprotein precursor (GPC), and RNA-dependent RNA polymerase (RdRp), respectively [16]. The GPC encoded by the CCHFV M segment is co-translationally cleaved into two glycoprotein precursors, PreGn and PreGc [17], and a non-structural M protein (NSm) [18]. Cleavage of PreGn requires the host cell endoprotease subtilisin kexin isozyme-1 (SKI-1), also known as site-1 protease (S1P) [19]. SKI-1 recognizes a consensus cleavage motif typically described as (R/K)-X-(hydrophobic)-Z, where Z is often leucine, phenylalanine, or threonine [20]. SKI-1 cleaves the glycoprotein precursor of orthonairoviruses at a conserved RXXL or RXLX motif to generate the mature structural Gn glycoprotein [19]. The GP38 domain, part of PreGn, closely interacts with the Gn region in its uncleaved form, but it is later cleaved by a furin-like protease at the RSKR247site in the lumen of the secretory pathway. [21,22]. Another structural glycoprotein, Gc, is produced through proteolytic processing of PreGc at the RKPL1040cleavage site [19]. Notably, the processing at the boundary of Gn and GP38 cleavage site occurs in the endoplasmic reticulum and/or the early Golgi compartments, while furin cleavage occurs in a late Golgi compartment, probably in the trans Golgi network (TGN) [19,23]. In the context of authentic CCHFV, SKI-1 deficiency results in a dramatic reduction in the release of infectious CCHFV [23] without affecting PreGn or PreGc/Gc trafficking, suggesting a deficiency in virus assembly that correlates with a block in PreGn cleavage at the RRLL site. Similarly, SKI-1 was shown to be critical for the maturation and infectivity of Nairobi sheep disease virus (NSDV) GPC, another orthonairovirus [24].

Due to the lack of licensed vaccines or approved therapeutic options and its high fatality rate, CCHFV is classified as a Select Agent by the United States Federal Select Agent Program (FSAP) and must be handled in biosafety level 4 (BSL-4) containment laboratories. Numerous vaccine candidates targeting CCHFV-GPC have demonstrated protective efficacy in animal models [25–27]. However, because CCHFV strains are highly genetically diverse, it is critical to evaluate whether these vaccines provide heterologous protection against a broad range of viral strains [28]. Similarly, no specific antivirals or vaccines are available for the other emerging orthonairoviruses. Recombinant vesicular stomatitis virus (rVSV) has proven to be a highly successful vaccine platform for emerging viral threats, most notably demonstrated by rVSV-ZEBOV (Ervebo), which received regulatory approval in 2019 [29]. This platform's success has led to the development of rVSV-based vaccines [30] currently in clinical trials for Marburg virus [31], Lassa fever virus [32], and other emerging pathogens [33–35]. The VSV platform offers robust immunogenicity, and the ability to induce both humoral and cellular immune responses [36].

In addition, rVSV bearing heterologous glycoproteins provides a rapid and easy-to-use platform for conducting neutralization studies, epitope mapping of monoclonal antibodies, and testing antiviral therapeutics under lower biosafety conditions [23–25]. VSV-based vaccines are also a promising vaccine platform showing protection against CCHFV in preclinical models [27] and could be used for development of vaccines for other nairoviruses of medical or veterinary importance.

The generation of pseudotyped or replication competent viruses expressing CCHFV glycoproteins is particularly challenging, possibly due to the complex and multistep processing required for proper glycoprotein synthesis. Unlike other replication competent VSV-based viruses, rVSV-CCHFV appears to require truncation of the C-terminal cytoplasmic tail of the CCHFV Gc protein to facilitate efficient surface expression, where VSV is assembled and released from the cell [27,37]. In addition, rescue of these replication competent viruses frequently results in mutations within or adjacent to the SKI-1 cleavage motif [25,27,38], suggesting that the selective pressures maintaining this motif in authentic CCHFV differ from those acting in VSV-based systems. To address these challenges associated with the generation of replication competent VSV-based CCHFV systems, we investigated the role of Gn cleavage for the rescue of VSV-based CCHFV and other relevant nairoviruses, closely related Hazara virus (HAZV) and medically relevant YEZV.

Materials and methods

Biosafety and ethics

All work with infectious CCHFV was conducted in a BSL-4 facility at the Centers for Disease Control and Prevention (CDC, USA) following established BSL-4 standard operating procedures. Experiments with recombinant VSV-based viruses were performed in a BSL-2 laboratory. All recombinant virus work was approved by the Institutional Biosafety Committee (IBC). International livestock sampling was conducted under CDC Institutional Animal Care and Use Committee Protocol for VHF surveillance in ungulates, with additional local approvals obtained as required by Ugandan national guidelines and institutional requirements. Human samples were obtained from patients infected during the 2024 CCHF season in and around Çorum, Turkey. Informed consent was obtained with IRB-approved protocols from Hitit University (Turkey) and University of Maryland Baltimore (IRB # HP-00082839).

Cell lines and antibodies

Huh7 (Apath LLC) and Vero-E6 cells (CDC core facility) were propagated in Dulbecco's modified Eagle medium (Gibco) supplemented with 1× non-essential amino acids, 5% (v/v) fetal bovine serum, 100 units/mL penicillin, and 100 µg/mL streptomycin (Gibco). All cells were incubated at 37°C in 5% CO2.

Murine monoclonal antibodies (MAbs) targeting CCHFV strain IbAr10200 GP38 (5A5 and 13G8), PreGc/Gc (11E7, 12), and NP (2B11 and 9D5) were obtained from the Joel M. Dalrymple-Clarence J. Peters USAMRIID Antibody Collection through BEI Resources. The PreGc/Gc MAb 8A1 was obtained as reported by Zivcec et al. [39].

Generation of recombinant VSV mutants

The GPC open reading frames (ORFs) of CCHFV IbAr10200 (NP_950235.1), Turkey 200406546 (ALT31694.1), HAZV isolate JC280 (YP_009507851.1), and YEZV isolate HH003-2020 (YP_010840879.1) were all human codon-optimized, truncated (53 amino acids for CCHFV, 55 amino acids for HAZV and 38 amino acids for YEZV) in the cytoplasmic tail, synthesized by Twist Bioscience, and cloned into the standard pol-II expression vector, pEEV-Puro and subcloned in pCDNA. Plasmids encoding VSV glycoprotein (G), polymerase (L), nucleoprotein (N), and phosphoprotein (P) were also cloned into pCAGGS to generate helper plasmids for virus rescue. Recombinant VSVs expressing either wild-type or mutant glycoproteins, together with the fluorescent reporter ZsG, were generated as described previously [40,41]. Briefly, the native VSV G gene was replaced with the glycoprotein precursor gene of the respective orthonairovirus species. The ZsG reporter gene was inserted upstream of the VSV-N gene and separated by a self-cleaving porcine teschovirus 2A (P2A) sequence, allowing co-expression of the reporter and viral proteins from a single transcriptional unit (Figure 1(A)). To rescue wild-type or mutant versions of recombinant VSV viruses, varying amounts of full-length pVSVΔG-CCHFV-GPC, pVSVΔG-HAZV-GPC, or pVSVΔG-YEZV-GPC plasmids (mutant or wild-type versions), together with pCDNA-T7, pCAGGS-VSV-L, pCAGGS-VSV-N, pCAGGS-VSV-P, and pCAGGS-VSV-G were transfected into Huh7 cells. The cell supernatants were blind passaged twice onto Huh7 cells. Recombinant viruses were titered using a standard 50% tissue culture infectious dose (TCID50) using the Reed and Muench method in Vero-E6 and Huh7 cells (Figure 1(B)). Viral genome sequences were confirmed by next generation sequencing (NGS) on a MiniSeq platform (Illumina), and sequences were analyzed using CLC genomics workbench v 24.0 (Qiagen).Figure 1.(A) Schematic diagram of CCHFV glycoprotein and mutant proteins. The schematics illustrate various constructs with the indicated mutations in the SKI-1 cleavage site and cytoplasmic tail. In this diagram, “WT” denotes the unmodified wild-type construct, while “Δ53” signifies a deletion of 53 amino acids in the cytoplasmic tail. (B) Western blot analysis of transfected plasmid constructs. pCDNA plasmids encoding indicated CCHFV glycoprotein constructs were transfected into Expi293 T cells, and cell suspensions were harvested in 2× Laemmli sample buffer (without reducing agent) after 24 h of transfection, denatured at 50°C for 10 min, separated by SDS-PAGE on 4%–12% bis-tris gels, and transferred via semi-dry blotting to nitrocellulose membranes. The primary antibodies used were against CCHFV Gn, CCHFV Gc, and tubulin (loading control). After probing with HRP-linked secondary antibodies, the proteins were visualized using a Fast Western Blot kit with SuperSignal West Dura HRP substrate, and blots were imaged with a ChemiDoc MP system (Bio-Rad). Uncropped western blots are presented in SupplementalFigure 4.Two visuals: a schematic diagram of CCHFV glycoprotein constructs and a western blot image comparing bands for wild-type and mutants.The figure shows two related visuals. The upper visual is a schematic diagram of Crimean Congo hemorrhagic fever virus glycoprotein precursor constructs. Horizontal bars represent segments labeled from left to right as signal peptide, mucin like domain, glycoprotein 38, glycoprotein N, nonstructural protein M, glycoprotein C, and transmembrane cytoplasmic tail. The first bar is marked GPC hyphen WT and ends with an intact tail. Four shorter bars beneath it are labeled GPC hyphen delta 53, GPC hyphen RRVL hyphen delta 53, GPC hyphen VRLL hyphen delta 53, and GPC hyphen AALL hyphen delta 53, each annotated with a deletion of 53 amino acids at the right end. A fifth mutant bar labeled GPC hyphen AAAA hyphen delta 53 also lacks 53 amino acids at the tail. Arrows over the central region mark the SKI hyphen 1 cleavage site with indicated tetrapeptide sequences RRLL, RRVL, VRLL, AALL, or AAAA. The lower visual is a western blot image. Four horizontal blot rows are labeled on the left as Crimean Congo hemorrhagic fever virus glycoprotein C, Crimean Congo hemorrhagic fever virus glycoprotein 38 plus glycoprotein N, Crimean Congo hemorrhagic fever virus glycoprotein N, and beta hyphen tubulin. Vertical sample lanes at the top are labeled mock hyphen transfected, non hyphen transfected, wildtype Crimean Congo hemorrhagic fever virus glycoprotein precursor, GPC delta 53, RRVL GPC delta 53, VRLL GPC delta 53, AALL GPC delta 53, and AAAA GPC delta 53. Protein bands of varying intensity appear in each row, while beta hyphen tubulin bands are similar across lanes as a loading control.

Neutralization assays

Neutralization assays were performed using bovine serum samples from Uganda, and human serum samples from Turkey. Polyclonal hyperimmune mouse ascitic fluids (HMAFs) directed against CCHFV strain IbAr10200 were obtained from CDC Viral Special Pathogens Branch (VSPB) stocks, as previously described [42]. The HMAFs were prepared as described previously [43]. HMAF for HAZV were kindly provided by the World Reference Centre for Emerging Viruses and Arboviruses (WRCEVA) at the University of Texas Medical Branch at Galveston.

For neutralization assays, 10,000 cells per well were seeded into 96-well plates one day prior to infection. Serial five-fold dilutions of antibodies, bovine sera, or human serum samples were mixed with 100 TCID50of rVSVs or CCHFV-ZsG. After incubation of the virus-antibody or virus-serum mixtures for 1 h at 37°C, the mixtures were added to Vero E6 cell monolayers in triplicate. Infected cells were incubated for 72 h at 37°C, after which fluorescence signals were measured using a BioTek plate reader.

All experiments were performed in triplicate. Data were analyzed using GraphPad Prism v10.0 (GraphPad Software). Fluorescence values of ZsG expression were normalized to background fluorescence in non-infected cells. The resulting normalized values were used to fit a 4-parameter equation to semi-log plots of the concentration–response data. The dilution of the antibodies/samples inhibiting 50% of ZsG expression (IC50) was calculated by interpolating the values.

Mix and read assay

Mix and Read (MR) assays for GP38 and NP were performed as described previously [44,45]. The positive controls and negative controls were used as mentioned [43]. In brief, 15 μL of NP or GP38 biosensor was combined with 15 μL of the test sample. Following a 20-minute incubation, Nanoluc luciferase substrate was added to the mixture. The reconstitution of luciferase upon antibody binding generates a luminescence signal once the substrate is introduced. This signal is measured 10 min after substrate addition using the Synergy instrument (BioTek). Results are normalized to the negative control serum. A positive result is defined by a signal-to-background ratio exceeding 3 for NP and 1.8 for GP38.

Western blot analysis

To verify the expression of CCHFV GPC proteins in different constructs, western blotting was performed as described previously [46]. Briefly, pCDNA-derived plasmids of various constructs were transfected into Expi293F cells, and after 24 h post-transfection, cell suspensions were harvested in 2× Laemmli sample buffer (with reducing agent), denatured at 50°C for 10 min, separated by SDS-PAGE on 4%–12% bis-tris gels, and transferred via semi-dry blotting to nitrocellulose membranes. The CCHFV rabbit anti-Gn (1:1000 dilution, was a kind gift from A. Mirazimi), CCHFV mouse anti-Gc mAb (1:1000 dilution, 11E7, BEI Resources), and anti-tubulin (1:3000 dilution, ThermoFisher Scientific #MA5-16308). After probing with HRP-linked secondary antibodies (ThermoFisher Scientific #32260 and #32230), the proteins were visualized using a Fast Western Blot kit with SuperSignal West Dura HRP substrate (ThermoFisher Scientific), and blots were imaged with a ChemiDoc MP system (Bio-Rad).

Growth kinetics of recombinant viruses

For growth kinetics, Huh7 cells were seeded at 50% confluency in 6-well plates (Costar) one day prior to infection. The cells were infected with different viruses at an MOI of 0.05. After 1 h of absorption at 37°C, the inoculum was removed, cells were washed with 1X PBS, and DMEM medium with 5% serum was added. The cell culture supernatants were collected at 0, 1-, 2-, and 3-days post infection (dpi). The virus was quantified by standard 50% tissue culture infectious dose (TCID50) protocol using the Reed and Muench method [47].

Results

Directed mutagenesis of the conserved SKI-1 cleavage site in preGn

Our initial efforts to rescue the wild-type replication competent rVSVΔG-CCHFV-GPC-ZsG, which contains the full-length, unmodified glycoprotein precursor (GPC) from CCHF virus strain IbAr10200, were unsuccessful using previously described methods [21,22].

Since previous studies showed that a 53-amino acid deletion improves VSV-ΔG-CCHFV GPC pseudotype yield [37], and recovery of rVSV-CCHFV often resulted in mutations in the RRLL cleavage motif, we investigated the effect of mutating amino acids critical for SKI-1 cleavage in combination with the 53-amino acid deletion. Several plasmid constructs were designed, including one with the same full-length wild-type CCHFV glycoprotein sequence used in our unsuccessful rescue attempts; another with a 53-amino acid deletion in the C-terminal tail of Gc (pCDNA-CCHF-GPC-Δ53); and others with mutations at the SKI-1 site of the GPC (RRVL, VRLL, AALL, and AAAA) (Figure 1(A)). The RRVL mutation was based on a previous study where this mutation was observed following the successful rescue of a replication competent VSV-CCHFV after several passages [27].

To analyze the effect of these mutations on PreGn cleavage, we performed western blot analysis. As shown inFigure 1(B), both the wild-type and Δ53 CCHFV constructs resulted in complete cleavage of PreGn into Gn. In contrast, the RRVL mutation led to incomplete PreGn cleavage, while the other three constructs (VRLL, AALL, and AAAA) completely blocked cleavage, which correlated with the known specificity of SKI-1 with the P4 Arg and P1 or P2 Leu being particularly important [48].

Abrogating cleavage at the RRLL site is required for generation of recovery of rVSVΔG encoding CCHFV GPC

Since our initial attempts to rescue rVSVΔG-CCHFV-GPC-ZsG wild type were unsuccessful, we attempted the rescue with the construct having a 53-amino acid deletion in the C-terminal tail of Gc (Figure 2(A)). This construct enabled successful recovery of infectious rVSV-CCHF-GPC-ZsG (Figure 2(B)); however, multiple independent rescues resulted in the production of low-titer virus stocks, and sequencing of the virus showed mutations in the RRLL cleavage motif to RRVL, RRLP, and QRLL variants (Figure 2(C)). Since these changes in the RRLL are predicted to reduce SKI-1 processing efficiency, we attempted to improve rescue efficiency and designed other mutations that we demonstrated earlier to block cleavage (VRLL) or by completely disrupting the RRLL cleavage site to AAAA. Western blots of VRLL and AAAA, also carrying a 53 amino acid deletion in the Gc C-terminal tail, showed a complete loss of mature Gn production (Figure 1(B)). This approach resulted in the successful generation of replication-competent recombinant viral stocks with high titers (rVSVΔG-CCHFV-VRLL-Δ53: 3 × 106TCID50/mL and rVSVΔG-CCHFV-AAAA-Δ53: 1 × 106TCID50/mL). The growth kinetics for both rescued viruses were similar in Huh7 cells, reaching very high titers at 2 days post infection (dpi) (Figure 2(D)).Figure 2.Construction of recombinant VSVΔG-CCHFV-GPC-ZsG. (A) Schematic diagram of the recombinant virus VSVΔG-CCHFV-GPC-ZsG. The recombinant virus expressing ZsGreen (ZsG) contains the ZsG fused to VSV-N (nucleoprotein) via a porcine teschovirus-1 2A ribosome-skipping peptide (P2A), VSV-P (phosphoprotein), VSV-M (matrix protein), CCHFV-GP (CCHFV glycoprotein), and L (polymerase). (B) Strategy to rescue the recombinant viruses. Plasmids encoding VSVΔG-CCHFV-GPC-ZsG, VSV-L, VSV-N, VSV-P, and VSV-G were co-transfected into Huh-7 cells to rescue rVSVΔG-CCHFV-GPC-ZsG. The supernatant was blind-passaged onto Huh-7 cells, and the resulting virus stock was titered using the TCID50determination method and sequenced by next-generation sequencing. (C) Attempts to rescue different recombinant viruses. The table summarizes the rescue attempts for the generation of various replication-competent viruses using different full-length clones (left column), where “neg” and “pos’ indicate unsuccessful or successful rescues, respectively (right column). (D) Growth kinetics of different recombinant viruses in Huh7 cells. Huh7 cells were infected at an MOI of 0.05, and titers determined at each indicated time point by TCID50.Four visuals: a genome diagram, a workflow diagram, a results table, and two line graphs of VSV CCHF virus titers over 3 days post infection.The figure shows four related visuals summarizing construction and characterization of recombinant vesicular stomatitis virus expressing Crimean Congo hemorrhagic fever glycoprotein and ZsGreen. The first visual is a genome diagram labeled rVSV delta G CCHF GP, with a ZsGreen reporter joined to the vesicular stomatitis virus nucleoprotein by a porcine teschovirus 1 2A element, followed by phosphoprotein, matrix protein, Crimean Congo hemorrhagic fever virus glycoprotein, and polymerase. Text beneath cites Crimean Congo hemorrhagic fever strain IbAr10200 and a Turkey strain. The second visual is a workflow diagram labeled generation of recombinant virus, showing plasmids encoding VSV delta G CCHF GP, VSV L, VSV N, VSV P, and VSV G transfected into Huh 7 cells, then amplification in fresh Huh 7 cells to produce virus stock for titration and next generation sequencing. The third visual is a table titled virus rescue, listing full length clones such as VSV delta G CCHF 10200 RRLL wild type, VRLL, AAAA, and multiple delta 53 variants, with rescue outcomes marked as neg or pos and notes on spurious mutations RRLP, QRLL, and RRVL. The fourth visual contains two line graphs of growth kinetics in Huh7 cells, with time in days post infection on the x axis from 0 to 3 and TCID50 per milliliter on the y axis from 10 to 10 to the power of 8. The upper graph compares VSV CCHF 10200 VRLL and VSV CCHF 10200 AAAA, both rising from below 10 to the power of 2 at day 0 to around 10 to the power of 7 by day 2, then plateauing. The lower graph compares VSV CCHF 10200 AAAA and VSV CCHF Turk AAAA, showing a similar increase to high titers by day 2 with a slight difference between curves. All data are approximate.

To validate the reverse genetics approach for generating rVSV-CCHFV, we sought to determine whether the same strategy could be applied to other endemic CCHFV strains from Turkey. Because the glycoprotein of the CCHFV-Turkey strain (200406546) exhibits only 84% amino acid similarity to that of the CCHFV-10200 strain (Supp. Figure 1), we tested this approach using the CCHFV-Turkey strain. Using this strategy, we successfully rescued a high-titer recombinant VSV-CCHFV-Turkey virus containing the AAAA mutation together with a 53-amino acid truncation in the C-terminal tail of Gc (rVSVΔG-CCHFV-Turk-AAAA-Δ53: 5 × 106TCID50/mL). When comparing the growth kinetics of rVSV-CCHFV-Turkey and rVSV-CCHFV-10200 strains in Huh7 cells, the Turkey strain showed one log higher titer at the peak of 2 dpi (Figure 2(D)). To further investigate whether SKI-1 cleavage site mutations affect cleavage of PreGn in the rescued viruses, we performed western blot analysis on cells infected with viruses having VRLL and AAAA mutations. As expected, the AAAA mutation completely blocked the cleavage of Gn and the VRLL mutation displayed low residual cleavage (Supp. Figure 2). Overall, these results show that mutation in the RRLL cleavage site that block cleavage and deletion in the C-terminal of glycoprotein is an efficient approach to rescue c-GPC.

Generation and characterization of replication-competent rVSV-HAZV and rVSV-YEZV

To further evaluate whether the same strategy could be used to generate infectious clones of other rVSVs using GPCs from other nairoviruses, we tested this approach with HAZV and YEZV, two orthonairoviruses whose glycoproteins share only 33% and 22% amino acid similarity, respectively, with the CCHFV-IbAr10200 glycoprotein (Supp. Figure 2). We attempted to rescue both viruses containing wild-type glycoproteins as well as viruses harbouring SKI-1 cleavage site AAAA mutations with a 55 and 38-amino acid deletion in the Gc cytoplasmic tail of HAZV and YEZV, respectively. Surprisingly, we efficiently rescued both SKI-1-WT and SKI-1-AAAA mutants for HAZV and YEZV, achieving high titers (Figure 3(B)). Furthermore, growth kinetics revealed that the AAAA mutation reduced viral growth of VSV-HAZV and VSV-YEZV relative to their wild-type counterparts (Figure 3(C)). Together, these results suggest that HAZV and YEZV do not require mutations in the SKI-1 cleavage site for efficient rescue.Figure 3.Construction and utilization of recombinant VSVΔG-HAZV-GPC-ZsG and rVSVΔG-YEZV-GPC-ZsG. (A) Schematic diagram of the recombinant virus, rVSVΔG, expressing Nairovirus GPC. The recombinant virus expressing ZsGreen (ZsG) contains the ZsG fused to VSV-N (nucleoprotein) via a porcine teschovirus-1 2A (P2A), VSV-P (phosphoprotein), VSV-M (matrix protein), HAZV-GP (HAZV glycoprotein)/YEZV-GP (YEZV glycoprotein), and L (polymerase) (B) Rescue of recombinant viruses: The table summarizes the rescue attempts for the generation of various VSV-based chimeric nairoviruses, where “pos’ indicates successful rescues.(C) Growth kinetics of different recombinant viruses in Huh7 cells. Huh7 cells were infected at an MOI of 0.05, and titers determined at each indicated time point by TCID50. (D) Neutralization assays using HMAFs. Vero-E6 cells were treated with serial 5-fold dilutions of indicated HMAFs mixed with 100 TCID50of the indicated recombinant viruses. After 72 h, fluorescence signal was measured using a Cytation plate reader (BioTek). Concentration-response curves were plotted using GraphPad Prism. Relative ZsG expressions are shown on the y-axis.Four visuals: one genome diagram, one table, one line graph set, and one line graph set of neutralization curves with decreasing trends.The figure shows four visuals summarizing recombinant vesicular stomatitis virus expressing Hazara and Yezo glycoproteins. The first visual is a genome diagram labeled rVSV delta G expressing Nairovirus glycoprotein precursor, with rectangles ordered from left to right as ZsGreen fused by porcine teschovirus 1 2A, nucleoprotein N, phosphoprotein P, matrix protein M, CCHF glycoprotein precursor, and polymerase L. Text beneath lists Hazara isolate Jc280 YP underscore 009507851 point 1 and Yezo virus isolate HH003 dash 2020 YP underscore 010840879 point 1. The second visual is a table titled Virus rescue with two columns labeled Full length clone and Rescue. Rows read VSV delta G Hazara AAAA delta 55 pos, VSV delta G Hazara VRLL delta 55 pos, VSV delta G Hazara RRLL delta 55 asterisk pos, VSV delta G Yezo AAAA delta 38 pos, VSV delta G Yezo VKLL delta 38 pos, and VSV delta G Yezo RKLL delta 38 asterisk pos. A footnote states asterisk WT SKI dash 1 site. The third visual shows two line graphs of growth curves with x axes Days post infection from 0 to 3 and y axes TCID50 per milliliter on logarithmic scales from 10 to 10 million, where wild type lines rise above AAAA mutant lines. The fourth visual contains three line graphs labeled VSV dash CCHF, VSV dash Hazara, and VSV dash Yezo, each plotting relative ZsGreen expression on the y axis from 0 to 150 against serum dilution factor on the x axis from 10 to 100000 on a logarithmic scale. Curves show loss of fluorescence for matched homologous HMAF at low dilution while heterologous HMAFs remain near baseline. All data are approximate.

Utility of rVSV-based replication competent nairoviruses for virus-specific neutralization assays

To evaluate the utility of these recombinant viruses, we performed cross-neutralization studies using HMAF against CCHFV and HAZV. HMAF against Rift Valley fever virus (HMAF-RVF) was used as a negative control. For neutralization studies, AAAA mutant was used for rVSV-CCHFV-10200 and rVSV-CCHFV-Turk; hereafter, the viruses are referred to by their virus names only, without explicitly mentioning the AAAA mutant. Wildtype cleavage site versions were used for neutralization studies with rVSV-HAZV and rVSV-YEZV. As shown inFigure 3(D), we observed highly specific neutralization, in which rVSV-CCHFV was neutralized by HMAF-CCHFV but not by HMAF-HAZV or HMAF-RVF. A similar pattern was observed for rVSV-HAZV, which was neutralized only by HMAF-HAZV. In contrast, rVSV-YEZV was not neutralized by any of the HMAFs tested. Overall, these data suggest that VSV-based nairoviruses can be effectively used in neutralization assays to distinguish virus-specific infections (Figure 3(D)).

Gc-Specific antibodies neutralize rVSV-CCHFV and infectious CCHFV

We tested rVSV-CCHFV in a neutralization assay using a selected group of monoclonal antibodies against Gc and GP38 regions of CCHFV glycoprotein. As shown inFigure 4, neutralizing antibodies targeting CCHFV-Gc efficiently neutralized rVSV-CCHFV, while no neutralization activity was detected for the antibodies targeting CCHFV GP38 or the Rift Valley Fever (RVFV) control antibody. To determine whether the neutralization activity observed with the VSV-based virus was comparable to that of fully infectious virus, we tested representative antibodies for Gc and GP38 using recombinant CCHFV-ZsG in a BSL-4 laboratory. As expected, representative antibodies that exhibited neutralization activity against rVSV-CCHFV also neutralized fully infectious CCHFV (Supp. Figure 3), having comparable IC50values. These results suggest that rVSV-CCHFV could be used as a BSL-2 surrogate for CCHFV to test neutralizing antibodies.Figure 4.Neutralization assays using antibodies. (A) Huh7 cells were treated with serial 5-fold dilutions of indicated antibodies mixed with 100 TCID50of the recombinant virus (rVSVΔG-CCHFV-10200). After 72 h, fluorescence signal was measured using a Cytation plate reader. Representative images of the neutralization activity of anti-CCHFV Gc (12A9 and 8A1) and anti-Rift valley fever virus (anti-RVFV) antibodies in cells infected with recombinant virus. (B) Concentration-response curves in Huh7 cells infected with recombinant virus after neutralization with anti-CCHFV-Gc-8A1 (red triangles), anti-CCHFV-Gc-12A9 (blue squares), and anti-RVFV (green circles). Relative ZsG expressions are shown on the y-axis. Each point on the graph represents mean values of triplicate wells. (C) 50% effective inhibition concentration (IC50) values of each indicated antibody against rVSVΔG-CCHFV-10200.Three visuals: a microscopy image grid, two line graphs, and two tables showing CCHFV antibody neutralization titrations and IC50 values.The figure shows three related visuals summarizing neutralization assays for recombinant vesicular stomatitis virus Crimean Congo hemorrhagic fever. The first visual is a grid of fluorescence microscopy images of Huh7 cells infected with recombinant virus and treated with monoclonal antibodies. Three horizontal rows correspond to monoclonal antibody Rift Valley Fever glycoprotein, monoclonal antibody Crimean Congo hemorrhagic fever 12A9, and monoclonal antibody Crimean Congo hemorrhagic fever 8A1. Each row contains six fields of view arranged by antibody concentration labeled along the bottom as 100000, 20000, 4000, 800, 160 nanograms per milliliter, and no antibody. The density of fluorescent cells visibly decreases at higher antibody concentrations for the Crimean Congo hemorrhagic fever glycoprotein C antibodies compared with the Rift Valley Fever control. The second visual consists of two line graphs with concentration response curves in Huh7 cells. For both graphs, the horizontal axis shows antibody concentration from 100 to 100000 nanograms per milliliter on a logarithmic scale, and the vertical axis shows relative ZsGreen expression from 0 to about 150 in arbitrary units. Curves for antibodies 12A9 and 8A1 decrease with increasing concentration, whereas the Rift Valley Fever control line remains near the top. All data are approximate. The third visual contains two tables listing monoclonal antibodies, their targets, and half maximal inhibitory concentration values in nanograms per milliliter for recombinant vesicular stomatitis virus Crimean Congo hemorrhagic fever 10200. The left table lists multiple Crimean Congo hemorrhagic fever glycoprotein C antibodies with half maximal inhibitory concentration values ranging from 200 to greater than 100000. The right table lists Crimean Congo hemorrhagic fever glycoprotein 38 antibodies and the Rift Valley Fever glycoprotein control, each with half maximal inhibitory concentration reported as greater than 100000 nanograms per milliliter.

Neutralizing activity of human and animal sera against rVSV-CCHFV and rVSV-HAZV

Furthermore, we sought to test a series of bovine samples having detectable antibodies against both CCHFV GP38 and NP. The cattle samples were collected in Uganda, a CCHFV-endemic country, and represent naturally exposed animals. Samples with a range of antibody levels measured by MR assay, were tested for their ability to neutralize the rVSV-CCHFV virus. As shown inFigure 5(A), all bovine samples with high MR antibody levels also exhibited neutralization activity against rVSV-CCHFV in a dose-dependent manner. As expected, no neutralizing activity, even at lower dilutions, was observed in samples negative for the CCHFV-specific MR assay, thereby demonstrating the specificity of the neutralization assay. The neutralizing capacity of these bovine sera was further assessed using rVSV-HAZV to evaluate the presence of cross-neutralizing antibodies. Cross-neutralizing antibodies were detected in 9 out of 20 bovine samples that neutralized rVSV-CCHFV; however, IC50values were comparatively low, with a median value of 19 for rVSV-HAZV samples versus 239 for rVSV-CCHFV.Figure 5.Neutralizing activity of bovine sera and human clinical samples. (A) Antibody reactivity was assessed with NP and GP38 mix-and-read assays. Signal is shown as the signal-to-background ratio (SBR), with positive results defined as SBR >1.8 for GP38 and >3 for NP. For the neutralization assay, Huh7 cells were treated with serial 5-fold dilutions of different bovine serum samples mixed with 100 TCID50of the rVSVΔG-CCHFV-10200 or rVSVΔG-HAZV. After 72 h, ZsG fluorescence signal was measured using a Cytation reader (BioTek). Virus neutralization IC50values of bovine sera for rVSVΔG-CCHFV-10200 and rVSVΔG-HAZV are shown. (B) Neutralization IC50values of human sera for rVSVΔG- CCHFV-Turk, rVSVΔG-CCHFV-10200, rVSVΔG-HAZV, CCHF Viral Replicon Particle (VRP)-strain Oman GPC or rCCHFV-ZsG strain IbAr10200 are shown. N.D. indicates that the neutralization assay was not performed on the samples due to limited sample quantity and white cell background indicates negative results.Two tables showing neutralization titers of bovine sera from Uganda and human sera from Turkey against CCHFV and HAZV viruses.The figure shows two data tables summarizing neutralization assays. The first table is titled Neutralization with bovine sera from Uganda and lists 10 serum identifiers in the left column and another 10 in a second block to the right. For each bovine serum, there are columns labeled MR signal GP38, MR signal NP, Neutralization titer VSV CCHF, and Neutralization titer VSV HAZ. MR signal values range from 0 to 1315. Neutralization titers against vesicular stomatitis virus Crimean Congo hemorrhagic fever span approximately 1 to 1252, while titers against vesicular stomatitis virus Hazara virus span less than 10 to 92, with some cells marked N.D. for not determined. The second table is titled Neutralization with human sera Turkey. It lists 7 named CCHV T serum samples followed by 2 control rows labeled 10 and 53. Columns include MR signal GP 38, MR signal NP, Neutralization titer VSV CCHF Turk, Neutralization titer VSV CCHF 10200, Neutralization titer VSV HAZ, Neutralization titer VRP CCHF Oman, and Neutralization titer CCHF ZsG. MR signal values range from 1 to 848. Neutralization titers across the different viruses extend from less than 10 to 4477, with some entries marked N.D. All data are approximate.

In addition, we also tested a series of human clinical samples collected from CCHF convalescent patients in Turkey. All samples with positive MR CCHFV GP-38 and NP titers efficiently neutralized rVSV-CCHFV-Turkey strain in a dose-dependent manner, while those negative in the MR assay did not exhibit neutralization activity, even at low serum dilutions (Figure 5(B)). To assess cross-neutralization activity, the samples were tested against rVSV-CCHFV-10200, yielding results similar to those observed with the Turkey strain. To further determine if the neutralization activity observed with SKI-1 mutated strains was comparable to that of the wild-type CCHFV glycoprotein, we tested samples using CCHF Viral Replicon Particle (VRP)-ZsG expressing GPC of strain Oman [49] and authentic rCCHFV-ZsG strain IbAr10200. Representative samples that exhibited neutralization activity against rVSV-CCHFV also neutralized CCHFV-ZsG VRPs (GPC Oman) and CCHFV-ZsG (Figure 5(B)). Human sera were tested with rVSV-HAZV to assess cross-neutralizing antibodies; 6 of 8 samples that neutralized rVSV-CCHFV-Turk also showed cross-neutralization. Median IC50was 19 for rVSV-HAZV and 119 for rVSV-CCHFV.

Taken together, these findings indicate that rVSV-CCHFV serves as an effective BSL-2 surrogate for infectious CCHFV. Furthermore, animal and CCHF human convalescent samples exhibit greater neutralization of rVSV-CCHFV compared to rVSV-HAZV.

Discussion

In this study, we successfully demonstrated that mutations blocking cleavage at the RRLL motif and truncation of the Gc tail are required for the rescue of replication competent rVSV-CCHFV-GPC from two different strains of CCHFV. Furthermore, we showed that replication competent VSV-based CCHFV surrogate systems can be used for detecting neutralizing antibodies in ecological and clinical samples. Importantly, neutralization assays using human serum samples from Turkey with two different strains of rVSV-CCHFV showed similar IC50values, indicating cross-neutralization. This suggests that multiple recombinant CCHFV constructs may not be required to detect neutralizing antibodies from diverse geographical regions. Furthermore, we generated novel replication competent rVSV-HAZV-GPC and rVSV-YEZV-GPC and showed that mutations in the conserved cleavage site are not necessary for the rescue of these nairoviruses. HAZV was selected because it has been widely used as a surrogate model for CCHFV in lower-containment laboratories [50], while YEZV is a recently discovered emerging pathogen associated with febrile illness in humans [7,12]. To the best of our knowledge, this is the first report in which replication competent surrogate systems are developed to study nairoviruses other than CCHFV. The replication competent VSV-based surrogate systems have been generated for use in a broad range of assays, including neutralization, antiviral, and viral entry studies. A key strength of this study is the demonstration that targeted cleavage site mutations that greatly block PreGn cleavage can facilitate the efficient rescue of rVSV-CCHFV. Our surrogate system is able to detect neutralizing antibodies with high specificity and sensitivity. Importantly, we observed significant differences between nairoviruses, with YEZV and HAZV showing distinct rescue requirements compared to CCHFV. Notably, YEZV and HAZV do not require targeted SKI-1 cleavage site mutations that would block their PreGn cleavage.

VSV-based surrogate viruses have proven to be valuable tools for studying highly pathogenic viruses and have been widely used for neutralization assays, vaccine development, antiviral screening, and receptor identification studies [38,40,51,52,53,54]. Replication competent VSV-based and pseudotyped virus systems for CCHFV have been previously described [25,26,27,37,38,52,54]. However, the generation of replication competent VSV-based CCHFV has been technically challenging due to the complex post-translational processing required to produce functional CCHFV glycoproteins and the differences related to the localization of CCHFV or VSV virion morphogenesis and budding. Truncation of the Gc cytoplasmic tail has been shown to facilitate the rescue of rVSV-CCHFV [27,37,52]. This region is thought to contain signals involved in intracellular retention, and its deletion may reduce retention within the endoplasmic reticulum and/or Golgi apparatus, thereby promoting transport of the glycoprotein to the plasma membrane [55]. Cleavage at this site is required for the generation of mature Gn but, unlike its Gc cytoplasmic tail, it is not required for traffic to the Golgi [56]. Substitutions in the SKI-1 cleavage site result in the generation of PreGn or immature cleavage intermediates, which might be more efficiently incorporated into rVSV particles, facilitating viral assembly of infectious viruses at the cell surface membrane.

In this study, we sought to determine the role of the SKI-1 cleavage site in facilitating the rescue of rVSV-CCHFV by introducing targeted mutations at this site to enable efficient recovery of an rVSV-CCHFV chimera expressing the reporter ZsG. Initially, we attempted to rescue rVSV-CCHFV containing the wild-type, human codon-optimized CCHFV glycoprotein without any modifications. Despite numerous attempts, these efforts were unsuccessful. We subsequently truncated the cytoplasmic tail of Gc, a modification previously shown to facilitate the rescue of rVSV-CCHFV [27,37]. However, during rescue attempts of rVSV-CCHFV-Δ53, we consistently observed the emergence of mutations in the SKI-1 cleavage site. Across multiple independent rescue experiments, distinct mutations arose exclusively at the SKI-1 cleavage site, a finding consistent with other reports [25,26,27]. To investigate this phenomenon further, we introduced targeted mutations into the SKI-1 cleavage site to either partially or completely inactivate it by substituting the native sequence with VRLL or AAAA motifs, respectively. Impairment of the SKI-1 proteolytic site enabled efficient rescue of rVSV-CCHFV, resulting in 10-fold higher-titer virus stocks. To assess whether this strategy could be broadly applied to other CCHFV strains, we employed the same targeted approach using the Turkey strain glycoprotein. This construct was also rescued efficiently, demonstrating the general applicability of this method for rVSV-CCHF generation. Overall, this targeted modification of the SKI-1 cleavage site provides a robust and efficient strategy for the rapid generation of replication competent VSV-based CCHFV and can be readily applied to diverse CCHFV strains.

To further investigate whether this approach can be generalized for the efficient generation of other rVSV-based nairoviruses, we tested it using two genetically diverse nairoviruses, HAZV and YEZV. We used both wild-type and SKI-1-mutated glycoproteins of HAZV and YEZV to rescue their corresponding rVSV-based viruses. We observed that both wild-type and SKI-1 mutated versions of the two viruses could be generated without difficulty. Furthermore, the wild-type versions of both viruses exhibited substantially higher growth rates compared to their SKI-1 mutated counterparts. These findings suggest that the glycoproteins of HAZV and YEZV behave differently from those of CCHFV, warranting further investigation. Additionally, these results highlight the need for caution when using HAZV as a surrogate model for CCHFV. Additionally, neutralization studies using HMAFs highlight the specificity and reliability of the rVSV-based assay for detecting neutralizing antibodies. However, we detected cross-neutralizing antibodies in bovine and convalescent human sera against HAZV, which suggests that caution should be used when interpreting neutralization test results if no cross-neutralization experiments with related viruses are conducted. The differential requirements for SKI-1 cleavage site mutations between nairoviruses represent a significant finding with important implications for future studies. While CCHFV requires these mutations for efficient rescue, HAZV and YEZV do not, suggesting fundamental differences in glycoprotein processing and viral assembly mechanisms. This observation underscores the importance of virus-specific optimization when developing rVSV-based surrogate systems for emerging nairoviruses.

In conclusion, we have demonstrated an efficient method to rescue rVSV-based CCHFV by introducing targeted mutations at the SKI-1 cleavage site of the CCHFV glycoprotein precursor. The ease of generating viruses with this approach will be particularly useful for future comparative studies of different CCHFV strains in lower-containment laboratories and for advancing vaccine development pipelines. Furthermore, this rVSV-based platform can be extended to study other emerging nairoviruses, as demonstrated in this study with HAZV and YEZV, providing a versatile and scalable approach for developing countermeasures against this medically important virus family.

References

1. LaseckaL, BaronMD. The molecular biology of nairoviruses, an emerging group of tick-borne arboviruses. Arch Virol. 2014; 159(6): 1249–1265. doi.org/10.1007/s00705-013-1940-z
2. KuhnJH, AbeJ, AdkinsS, et al. Annual (2023) taxonomic update of RNA-directed RNA polymerase-encoding negative-sense RNA viruses (realm Riboviria: kingdom Orthornavirae: phylum Negarnaviricota). J Gen Virol. 2023; 104(8). doi.org/10.1099/jgv.0.001864
3. GarrisonAR, AlkhovskySV, Avsic-ZupancT, et al. ICTV virus taxonomy profile: Nairoviridae. J Gen Virol. 2020; 101(8): 798–799. doi.org/10.1099/jgv.0.001485
4. HawmanDW, FeldmannH. Crimean-Congo haemorrhagic fever virus. Nat Rev Microbiol. 2023; 21(7): 463–477. doi.org/10.1038/s41579-023-00871-9
5. ErgönülÖ. Crimean-Congo haemorrhagic fever. Lancet Infect Dis. 2006; 6: 203–214. doi.org/10.1016/S1473-3099(06)70435-2
6. BenteDA, ForresterNL, WattsDM, et al. Crimean-Congo hemorrhagic fever: history, epidemiology, pathogenesis, clinical syndrome and genetic diversity. Antiviral Res. 2013; 100(1): 159–189. doi.org/10.1016/j.antiviral.2013.07.006
7. KodamaF, YamaguchiH, ParkE, et al. A novel nairovirus associated with acute febrile illness in Hokkaido, Japan. Nat Commun. 2021; 12(1): 5539. doi.org/10.1038/s41467-021-25857-0
8. WangYC, WeiZ, LvX, et al. A new nairo-like virus associated with human febrile illness in China. Emerg Microbes Infect. 2021; 10(1): 1200–1208. doi.org/10.1080/22221751.2021.1936197
9. LiuX, ZhangX, WangZ, et al. A tentative tamdy orthonairovirus related to febrile illness in Northwestern China. Clin Infect Dis. 2020; 70(10): 2155–2160. doi.org/10.1093/cid/ciz602
10. ZhangXA, MaYD, ZhangYF, et al. A new orthonairovirus associated with human febrile illness. N Engl J Med. 2024; 391(9): 821–831. doi.org/10.1056/NEJMoa2313722
11. MaJ, LvXL, ZhangX, et al. Identification of a new orthonairovirus associated with human febrile illness in China. Nat Med. 2021; 27(3): 434–439. doi.org/10.1038/s41591-020-01228-y
12. ApanasevichM, DubovitskiyN, DerkoA, et al. Genomic characterization of a novel Yezo Virus revealed in ixodes pavlovskyi tick virome in Western Siberia. Viruses. 2025; 17(10): 1–15. doi.org/10.3390/v17101362
13. LvX, LiuZ, LiL, et al. Yezo virus infection in tick-bitten patient and ticks, Northeastern China. Emerg Infect Dis. 2023; 29(4): 797–800. doi.org/10.3201/eid2904.220885
14. ZhangMZ, BianC, YeRZ, et al. A series of patients infected with the emerging tick-borne Yezo virus in China: an active surveillance and genomic analysis. Lancet Infect Dis. 2025; 25(4): 390–398. doi.org/10.1016/S1473-3099(24)00616-9
15. MarkarianNM, AbrahamyanL. The emerging tick-borne Yezo virus - current knowledge, challenges, and perspectives. BMC Microbiol. 2025; 8: 1–21. doi.org/10.1186/s12866-025-04624-x
16. ZivcecM, ScholteFE, SpiropoulouCF, et al. Molecular insights into Crimean-Congo hemorrhagic fever virus. Viruses. 2016; 8(4): 106. doi.org/10.3390/v8040106
17. SanchezAJ, VincentMJ, NicholST. Characterization of the glycoproteins of Crimean-Congo hemorrhagic fever virus. J Virol. 2002; 76(14): 7263–7275. doi.org/10.1128/JVI.76.14.7263-7275.2002
18. AltamuraLA, Bertolotti-CiarletA, TeiglerJ, et al. Identification of a novel C-terminal cleavage of Crimean-Congo hemorrhagic fever virus PreGN that leads to generation of an NSM protein. J Virol. 2007; 81(12): 6632–6642. doi.org/10.1128/JVI.02730-06
19. VincentMJ, SanchezAJ, EricksonBR, et al. Crimean-Congo hemorrhagic fever virus glycoprotein proteolytic processing by subtilase SKI-1. J Virol. 2003; 77(16): 8640–8649. doi.org/10.1128/JVI.77.16.8640-8649.2003
20. ZhouA, WebbG, ZhuX, et al. Proteolytic processing in the secretory pathway. J Biol Chem. 1999; 274(30): 20745–8. doi.org/10.1074/jbc.274.30.20745
21. BergeronE, ZivcecM, ChakrabartiAK, et al. Recovery of recombinant Crimean Congo hemorrhagic fever virus reveals a function for non-structural glycoproteins cleavage by furin. PLoS Pathog. 2015; 11(5): e1004879. doi.org/10.1371/journal.ppat.1004879
22. SanchezAJ, VincentMJ, EricksonBR, et al. Crimean-congo hemorrhagic fever virus glycoprotein precursor is cleaved by Furin-like and SKI-1 proteases to generate a novel 38-kilodalton glycoprotein. J Virol. 2006; 80(1): 514–525. doi.org/10.1128/JVI.80.1.514-525.2006
23. BergeronE, VincentMJ, STN. Crimean-Congo hemorrhagic fever virus glycoprotein processing by the endoprotease SKI-1/S1P is critical for virus infectivity. J Virol. 2007; 81(23): 13271–6. doi.org/10.1128/JVI.01647-07
24. BostC, HartlaubJ, Pinho Dos ReisV, et al. The proprotein convertase SKI-1/S1P is a critical host factor for Nairobi sheep disease virus infectivity. Virus Res. 2023; 329: 199099. doi.org/10.1016/j.virusres.2023.199099
25. ScherG, BenteDA, MearsMC, et al. GP38 as a vaccine target for Crimean-Congo hemorrhagic fever virus. NPJ Vaccines. 2023; 8(1): 73. doi.org/10.1038/s41541-023-00663-5
26. TipihT, LeventhalSS, Meade-WhiteK, et al. Single dose VSV-based vaccine protects mice against lethal heterologous Crimean-Congo hemorrhagic fever virus challenge. NPJ Vaccines. 2025; 10(1): 109. doi.org/10.1038/s41541-025-01164-3
27. RodriguezSE, CrossRW, FentonKA, et al. Vesicular stomatitis virus-based vaccine protects mice against Crimean-Congo hemorrhagic fever. Sci Rep. 2019; 9(1): 7755. doi.org/10.1038/s41598-019-44210-6
28. DowallSD, CarrollMW, HewsonR. Development of vaccines against Crimean-Congo haemorrhagic fever virus. Vaccine. 2017; 35(44): 6015–6023. doi.org/10.1016/j.vaccine.2017.05.031
29. Henao-RestrepoAM, CamachoA, LonginiIM, et al. Efficacy and effectiveness of an rVSV-vectored vaccine in preventing ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial (ebola Ca suffit!). Lancet. 2017; 389(10068): 505–518. doi.org/10.1016/S0140-6736(16)32621-6
30. MarziA, FeldmannH. Ebola virus vaccines: an overview of current approaches. Expert Rev Vaccines. 2014; 13(4): 521–531. doi.org/10.1586/14760584.2014.885841
31. MarziA, JankeelA, MenicucciAR, et al. Single dose of a VSV-based vaccine rapidly protects macaques from Marburg virus disease. Front Immunol. 2021; 12: 774026. doi.org/10.3389/fimmu.2021.774026
32. MalkinE, ZaricM, KiehM, et al. Safety and immunogenicity of an rVSV lassa fever vaccine candidate. N Engl J Med. 2025; 393(18): 1807–1818. doi.org/10.1056/NEJMoa2501073
33. FletcherP, O'DonnellKL, FeldmannF, et al. Fast-acting single-dose vesicular stomatitis virus-Sudan virus vaccine: a challenge study in macaques. Lancet Microbe. 2025; 6(12): 101244. doi.org/10.1016/j.lanmic.2025.101244
34. MarceauJ, SafronetzD, MartellaroC, et al. Bivalent VSV vectors mediate rapid and potent protection from Andes virus challenge in hamsters. Viruses. 2024; 16(2): 1–12. doi.org/10.3390/v16020279
35. BhatiaB, Meade-WhiteK, HaddockE, et al. A live-attenuated viral vector vaccine protects mice against lethal challenge with Kyasanur Forest disease virus. NPJ Vaccines. 2021; 6(1): 152. doi.org/10.1038/s41541-021-00416-2
36. ToberR, BankiZ, EgererL, et al. VSV-GP: a potent viral vaccine vector that boosts the immune response upon repeated applications. J Virol. 2014; 88(9): 4897–4907. doi.org/10.1128/JVI.03276-13
37. SudaY, FukushiS, TaniH, et al. Analysis of the entry mechanism of Crimean-Congo hemorrhagic fever virus, using a vesicular stomatitis virus pseudotyping system. Arch Virol. 2016; 161(6): 1447–1454. doi.org/10.1007/s00705-016-2803-1
38. MaX, MaY, WangZ, et al. The single amino acid change of R516 K enables efficient generation of vesicular stomatitis virus-based Crimean-Congo hemorrhagic fever reporter virus. J Med Virol. 2025; 97(6): e70457. doi.org/10.1002/jmv.70457
39. ZivcecM, GuerreroLIW, AlbarinoCG, et al. Identification of broadly neutralizing monoclonal antibodies against Crimean-Congo hemorrhagic fever virus. Antiviral Res. 2017; 146: 112–120. doi.org/10.1016/j.antiviral.2017.08.014
40. Shrivastava-RanjanP, JainS, ChatterjeeP, et al. Development of a novel minigenome and recombinant VSV expressing Seoul hantavirus glycoprotein-based assays to identify anti-hantavirus therapeutics. Antiviral Res. 2023; 214: 105619. doi.org/10.1016/j.antiviral.2023.105619
41. JainS, LoMK, KainulainenMH, et al. Development of a neutralization assay using a vesicular stomatitis virus expressing Nipah virus glycoprotein and a fluorescent protein. Virology. 2023; 587: 109858. doi.org/10.1016/j.virol.2023.109858
42. EricksonBR, DeydeV, SanchezAJ, et al. N-linked glycosylation of Gn (but not Gc) is important for Crimean Congo hemorrhagic fever virus glycoprotein localization and transport. Virology. 2007; 361(2): 348–355. doi.org/10.1016/j.virol.2006.11.023
43. KaraaslanE, ChiangCF, KurutasGO, et al. Rapid, sensitive, and species-independent detection of Crimean Congo hemorrhagic fever virus nucleoprotein and GP38 antibodies. EBioMedicine. 2025; 118: 105857. doi.org/10.1016/j.ebiom.2025.105857
44. BergeronE, ChiangCF, LoMK, et al. Streamlined detection of Nipah virus antibodies using a split NanoLuc biosensor. Emerg Microbes Infect. 2024; 13(1): 2398640. doi.org/10.1080/22221751.2024.2398640
45. KainulainenMH, BergeronE, ChatterjeeP, et al. High-throughput quantitation of SARS-CoV-2 antibodies in a single-dilution homogeneous assay. Sci Rep. 2021; 11(1): 12330. doi.org/10.1038/s41598-021-91300-5
46. WelchSR, ScholteFEM, SpenglerJR, et al. The Crimean-Congo hemorrhagic fever virus NSm protein is dispensable for growth in vitro and disease in ifnar(-/-) mice. Microorganisms. 2020; 8(5): 1–15. doi.org/10.3390/microorganisms8050775
47. ReedLM, MuenchH. A simple method for estimating fifty percent endpoints. Am J Epidemiol. 1938; 27(3): 493–497.
48. CendronL, RothenbergerS, CassariL, et al. Proprotein convertases regulate trafficking and maturation of key proteins within the secretory pathway. Adv Protein Chem Struct Biol. 2023; 133: 1–54. doi.org/10.1016/bs.apcsb.2022.10.001
49. ScholteFEM, SpenglerJR, WelchSR, et al. Single-dose replicon particle vaccine provides complete protection against Crimean-Congo hemorrhagic fever virus in mice. Emerg Microbes Infect. 2019; 8(1): 575–578. doi.org/10.1080/22221751.2019.1601030
50. OlejnikJ, MeierK, HerreraJN, et al. Harnessing hazara virus as a surrogate for crimean-Congo hemorrhagic fever virus enables inactivation studies at a low biosafety level. Pathogens. 2025; 14(7): 1–21. doi.org/10.3390/pathogens14070700
51. Shrivastava-RanjanP, KellyJA, McMullanLK, et al. Evaluating neutralizing antibodies in hantavirus-infected patients using authentic virus and recombinant vesicular stomatitis virus systems. Viruses. 2025; 17(5): 1–13. doi.org/10.3390/v17050723
52. SakuraiY, HiranoM, OkadaS, et al. Identification of claramine and anidulafungin as entry inhibitors of Crimean-Congo hemorrhagic fever virus. Antiviral Res. 2026; 246: 106343. doi.org/10.1016/j.antiviral.2026.106343
53. SudaY, ChamberlainJ, DowallSD, et al. The development of a novel diagnostic assay that utilizes a pseudotyped vesicular stomatitis virus for the detection of neutralizing activity against Crimean-Congo hemorrhagic fever virus. Jpn J Infect Dis. 2018; 71(3): 205–208. doi.org/10.7883/yoken.JJID.2017.354
54. WangQ, WangS, ShiZ, et al. Establishment of two serological methods for detecting IgG and neutralizing antibodies against Crimean-Congo hemorrhagic fever virus glycoprotein. Front Cell Infect Microbiol. 2024; 14: 1341332. doi.org/10.3389/fcimb.2024.1341332
55. GautamA, LalandeA, RitterM, et al. The PACS-2 protein and trafficking motifs in CCHFV Gn and Gc cytoplasmic domains govern CCHFV assembly. Emerg Microbes Infect. 2024; 13(1): 2348508. doi.org/10.1080/22221751.2024.2348508
56. FreitasN, EnguehardM, DenollyS, et al. The interplays between Crimean-Congo hemorrhagic fever virus (CCHFV) M segment-encoded accessory proteins and structural proteins promote virus assembly and infectivity. PLoS Pathog. 2020; 16(9): e1008850. doi.org/10.1371/journal.ppat.1008850

---

Republished from the open web under CC0. Authors: Jain S, Marot S, Karaaslan E, Sajadi MM, Baykam N, Yapar D, Shoemaker T, Montgomery JM, Spiropoulou CF, Albariño CG, Bergeron É. Read the original.