Design and evaluation of neutralizing and fusion inhibitory peptides to Crimean-Congo hemorrhagic fever virus

, Crimean-Congo hemorrhagic fever virus (CCHFV). CCHFV is endemic to Asia, the Middle East, South-eastern Europe, and Africa and is transmitted in enzootic cycles among ticks, mammals, and birds. Human infections are mostly subclinical or limited to mild febrile illness. Severe disease may develop, resulting in multi-organ failure, hemorrhagic manifestations, and case-fatality rates up to 30%. Despite the widespread distribution and life-threatening potential, no treatments have been approved for CCHF. Antiviral inhibitory peptides, which antagonize viral entry, are licensed for clinical use in certain viral infections and have been experimentally designed against human pathogenic bunyaviruses, with in vitro and in vivo efficacies. We designed inhibitory peptides against CCHFV with and without conjugation to various polyethylene glycol and sterol groups. These additions have been shown to enhance both cellular uptake and antiviral activity. Peptides were evaluated against pseudotyped and wild-type CCHFV via neutralization tests, Nairovirus fusion assays, and cytotoxicity profiling. Four peptides neutralized CCHFV with two of these peptides shown to inhibit viral fusion. This work represents the development of experimental countermeasures for CCHF, describes a nairovirus immunofluorescence fusion assay, and illustrates the utility of pseudotyped CCHFV for the screening of entry antagonists at low containment settings for CCHF.


Introduction
Crimean-Congo hemorrhagic fever (CCHF) is one of the world's most medically relevant tick-borne viral diseases, and it is caused by the bunyavirus, Crimean-Congo hemorrhagic fever virus (CCHFV).The virus contains a tri-segmented, negative sense, and single-stranded RNA genome, classified within the family Nairoviridae, genus Orthonairovirus (Adams et al., 2017).CCHFV is widely distributed across 30 countries within the Eastern hemisphere and is maintained in epizootic cycles between animals and Hyalomma ticks (Bente et al., 2013;Hoogstraal, 1979;Spengler et al., 2016).It can cause life-threatening human disease with associated case fatality rates of up to 30% (Ergönül, 2006;Whitehouse, 2004).The World Health Organization has classified it as a priority pathogen for research and development and a Biodefense Category A pathogen by the United States National Institutes of Health (Keshtkar-Jahromi et al., 2011).In many non-endemic countries worldwide, work with infectious CCHFV is restricted to biosafety level 4 (BSL4/P4/CL4) laboratories, which can hamper the timeline for research and treatment developments (Bente et al., 2013;Keshtkar--Jahromi et al., 2011;Weidmann et al., 2016).There are no broadly licensed vaccines to prevent CCHFV.The only vaccine available for human use is prepared from suckling mouse brain and is only used in Bulgaria (Mousavi-Jazi et al., 2012).Given the widespread distribution and sporadic outbreaks associated with CCHFV, product development and deployment of vaccine(s) to individuals who could stand some benefit are likely quite challenging (Dowall et al., 2017;Keshtkar-Jahromi et al., 2011).
Ideally, an antiviral would be an effective means to treat suspected and/or clinical cases of CCHF.However, there are no licensed antiviral therapeutics, despite ribavirin, a guanosine analogue, compassionately used to treat CCHF cases with contentious efficacies (Espy et al., 2018;Soares-Weiser et al., 2010).Other experimental therapeutics have shown promise with pathogenic members from the viral order Bunyavirales in either in vitro or in vivo testing [reviewed in (Gowen and Hickerson, 2017)].The most promising have been small molecule inhibitors such as Favipiravir, a pyrazinecarboxamide-based antiviral (Furuta et al., 2013), immunoglobulins (Ahmed et al., 2005;Bertolotti-Ciarlet et al., 2005;Nicacio et al., 2000), and synthetic peptides that antagonize viral entry by targeting virion glycoproteins (Barriga et al., 2016;Koehler et al., 2013).
Nairoviruses, like CCHFV, produce their glycoproteins from the genomic M-segment, which encodes for an approximately 1700 residue polyprotein precursor, termed the GPC (Kuhn et al., 2016;Lasecka and Baron, 2014).Co-translation and host cell processing via the endoplasmic reticulum and then the Golgi apparatus can yield multiple non-structural proteins along with two structural glycoproteins termed G N and G C , which are presumably studded as heterodimers on mature nairovirus particles (Bergeron et al., 2007;Clerx et al., 1981;Sanchez et al., 2006).These dimers contain functionally relevant domains pertinent to the initial events of viral infection.CCHFV entry studies have revealed microtubule, clathrin, early endosome, cholesterol, and pH-dependencies (Garrison et al., 2013;M. Melinda Simon et al., 2009;Melinda M. Simon et al., 2009).Studies have also shown that the nairovirus G C contains a pH-dependent class II viral fusion protein (Fels et al., 2021;Garry and Garry, 2004;Hua et al., 2020;Mishra et al., 2022), which has been empirically demonstrated for related bunyaviruses such as La Crosse, Tahyna, Andes, Hanta, and Rift Valley fever viruses (Cifuentes-Munoz et al., 2011;de Boer et al., 2012;Plassmeyer et al., 2005;Tischler et al., 2005).The G C has also been shown for CCHFV to possess neutralizing and B-cell epitopes, along with a putative surface-receptor binding domain (Bertolotti-Ciarlet et al., 2005;Fritzen et al., 2018;Goedhals et al., 2015;Xiao et al., 2011;Zivcec et al., 2017).Taken together, we hypothesized that the CCHFV-G C was an ideal target and that inhibitory peptides could be designed to this glycoprotein to antagonize CCHFV infections.
Inhibitory peptides have been successfully developed as a viralmembrane fusion inhibitor approved by the United States Food and Drug Administration (known as Fuzeon™; DP178; T20; Enfuvirtide) used to treat human immunodeficiency virus 1 (HIV-1) infections (Hardy and Skolnik, 2004).Inhibitory peptides have proven successful in pre-clinical studies for multiple human pathogenic viruses such as the flaviviruses (Hrobowski et al., 2005;Aaron G Schmidt et al., 2010;Aaron G. Schmidt et al., 2010;Yu et al., 2017), coronaviruses (de Vries et al., 2021;Outlaw et al., 2020), orthomyxoviruses (Lee et al., 2011;Saito et al., 2021), paramyxoviruses (Mathieu et al., 2015;Porotto et al., 2006Porotto et al., , 2010a;;Welsch et al., 2013), filoviruses (Miller et al., 2011;Pessi et al., 2019) and bunyaviruses (Barriga et al., 2016;Koehler et al., 2013).Other studies have supported the conjugation of various sterol lipid moieties onto inhibitory peptides to facilitate plasma membrane incorporation and/or subsequent endosomal uptake to enhance antiviral efficacies (Lee et al., 2011).Incorporation of these sterol-conjugated peptides within plasma membranes or endosomes results in binding of the peptides to the viral fusion protein to inhibit infection by one of two mechanisms: 1) inhibition of virus binding to cellular receptors or 2) inhibition of virus and endosome membrane fusion through prevention of fusion protein conformational changes (de Vries et al., 2021).Therefore, we sought to design inhibitory peptides to conserved regions within the CCHFV-G C and to evaluate these peptides in vitro on multiple cell types, while taking these two potential mechanisms into consideration.We adapted a pre-screening approach with a pseudotyped vesicular stomatitis virus (VSV) at low containment (BSL-2) to augment and streamline inhibitory peptide approaches for CCHFV (Rodriguez et al., 2019).We investigated fusion inhibition as a potential mechanism of the screened peptides' inhibitory abilities and developed a fluorescence-based nairovirus fusion assay to test this hypothesis.

Identification of regions of interest
Residues from the open reading frames (ORF) of all GenBank available nairovirus genomic M-segments were aligned using a Blosum62 score matrix via the global alignment function with free end gaps and an open gap penalty of 12 and extension penalty of 3, using Geneious software (version 11.0).Aligned sites were placed in columns with a score equal to or exceeding a specified threshold for either identities or similarities.Similarities were based on a scoring matrix for residue sizes, hydrophobicities, polarities, and isoelectric points.Regions of 15 or more amino acids with a minimum of 80% similarities were then assessed for Wimley-Whit interfacing hydrophobicity scoring using in silico Membrane Protein eXplorer.Regions with a positive Wimley-Whit score were selected for lipopeptide synthesis.

Cell culture, plasmids, viruses, and antibodies
Baby hamster kidney cells (BHK), human hepatocarcinoma cells (HuH-7) (Thermo Scientific), and a clone from the SW-13 human adrenocortical carcinoma cell line (SW-13-CDC), were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA), 2 mM L-glutamine (Invitrogen), and 1% penicillin-streptomycin (Invitrogen), cumulatively called D10.A codon-optimized version of the open reading frame of the CCHFV M-segment, pCAGGS-CCHFV-GPC, was kindly provided by J. Kortekaas.Hazara virus (HAZV) and Dugbe virus (DUGV) were obtained from the World Reference Center for Emerging Viruses and Arboviruses (WRCEVA, University of Texas Medical Branch, Galveston, TX).Three different viruses of CCHFV strain IbAr10200 were used: (1) wild-type virus for PRNT screenings, (2) a ZsGreen reporter version expressing the fluorescence tag on the N-terminus of the genomic S-segment ORF, designated rCCHFV-ZsG for determining the peptide IC 50 , and (3) an mCherry reporter version expressing the fluorescent reporter in the Nterminus of Pre-G N , designated rCCHFV-M/Cherry for fusion assays.All in vitro work with CCHFV was performed in a biosafety level 4 facility at the Galveston National Laboratory, University of Texas Medical Branch, Galveston, TX, in accordance with the approved Institutional Biosafety Committee protocols.A recombinant vesicular stomatitis virus (rVSV) strain: Indiana, with a green fluorescence protein tag, designated throughout this manuscript as VSV.A replication-competent pseudotyped CCHFV, designated as ΔGrVSV-CCHFV-GPC, was developed and described in a previous publication (Rodriguez et al., 2019).Hyperimmune mouse ascitic fluid (HMAF) was kindly provided by T. Ksiazek, University of Texas Medical Branch, Galveston, TX.The CCHFV monoclonal antibodies 12A9, 11E7, and 9D5 were obtained from B.E.I.Resources (ATCC) and 8A1 was kindly provided by the United States Armed Forces Research Institute for Infectious Diseases, Frederick, MD, from the Peters-Dalrymple Collection.

Virus titrations
Confluent monolayers of BHK, Huh7, or SW-13 cells were infected with VSV, ΔGrVSV-CCHFV-GPC, CCHFV-IbAr10200, or rCCHFV-ZsG, at 50-100 plaque-forming units (PFU) for 1 h at 37 • C with 5% CO 2 with rocking at 15-min intervals.Plaque assays for viral titrations were carried out analogously, with an overlay media final concentration of 1.25% Avicel (Sigma-Aldrich, St. Louis, MO) in 1X Eagle's Minimum Essential Medium (MEM) with 5% FBS and 1% P/S.Infected BHK cells were incubated for 24-48 h post-infection, and Huh-7 cells were incubated until four days post-infection.After incubations, overlays were aspirated, and monolayers were fixed and stained with 10% buffered formalin supplemented with 1X crystal violet for 1 h.Plaques were enumerated, and PFU were determined by averaging technical replicates per sample.

Peptide screening by plaque reduction neutralization test (PRNT)
To assess the use of ΔGrVSV-CCHFV-GPC as a neutralizing screening tool, CCHFV-GPC neutralizing antibodies (11E7, 12A9, 8A1, and HMAF) and one CCHFV-NP antibody (9D5, negative control) were tested by PRNT against the pseudotyped virus.ΔGrVSV-CCHFV-GPC was incubated with serial dilutions of each antibody for 1 h before inoculation on BHK cell monolayers for plaque assay as described above.The pseudotyped virus was then used to screen peptides for virus inhibition.BHK cell monolayers and the pseudotyped virus were treated separately with peptides or controls before virus inoculation.Each peptide was suspended in cell culture media with a final concentration of 10 μM peptide and incubated onto cells for 1.5 h.A known concentration of ΔGrVSV-CCHFV-GPC was incubated with each peptide at a final concentration of 10 μM peptide for 1 h before inoculation.The peptide inoculum was removed from the monolayers before infection with the virus/peptide mixture.HMAF at a final dilution of 1:160, with an additional control of chloroquine treatment at 10 μM, was also used.Plates were fixed and stained at two days post-infection, and the number of plaques were counted and assessed for percent plaque reduction compared to infected cells without any peptide treatment.To assess specificity, selected peptides were assayed against wild-type VSV using the same PRNT methods.Peptides were also screened against wild-type recombinant CCHFV, strain IbAr10200, using SW-13 cells in the BSL-4 under the same peptide treatment conditions with fixing and staining at three days post-infection.

Determination of 50% cytotoxicity concentrations (CC 50 )
Peptides were assessed for their effect on cellular viability as a measurement of relative toxicity in vitro using the MTT Cell Proliferation Assay (ATCC).Manufacturers' instructions were followed, but briefly, monolayers of Huh-7 cells were treated with peptides ranging from 0.001 μM to 100 μM for two and a half hours, then peptides were removed, and cells were supplemented with D10 and incubated for two days.Treatment time was chosen to replicate PRNT and IC 50 treatment and incubation times.The percent viability was normalized to cells without peptide treatment.
The rCCHFV-Ibar10200-ZsGreen virus was co-incubated with each peptide dilution for 1 h on ice.The peptide treatment was removed from the cell monolayers, and the virus/peptide mixture of the corresponding concentration was inoculated and incubated for 1 h.Cells were then supplemented with D10 and incubated until 48 h post-infection.The relative fluorescence of each well was measured using a GFP plate reader.Values for 100% and 0% fluorescence were obtained by triplicate wells inoculated with virus only without peptide treatment and SF-DMEM only without virus, respectively.

Nairovirus fusion assay and lipopeptide fusion inhibition assay
An in vitro acid-induced syncytium formation assay, referred to herein as a fusion assay, was developed for nairoviruses.BioCoat™ 8well Culture Slides (Corning) were pre-treated with 2.5 μg/well of mouse laminin I and seeded with HuH-7 or BHK cells at 80-90% confluency.Cells were transfected with 750 ng of codon-optimized pCAGGS-CCHFV-GPC (strain: IbAr10200) per well or infected with a multiplicity of infection (MOI) of three with ΔGrVSV-CCHFV-GPC, HAZV, DUGV, or rCCHFV-M/Cherry was exposed to either sterilefiltered DMEM with variable pH (pH 5.2-7.4, in 0.2 increments) or normal media (infection only control) for 30 min.Media was then replaced with 1 μM CellTracker™ Green CMFDA Dye (ThermoFisher) in serum-free DMEM post-infection for 2 h.At 18 h post-infection, cells were fixed in 4% paraformaldehyde, mounted with ProLong™ Diamond Antifade Mountant with DAPI (ThermoFisher), and imaged.Three representative fields of view were imaged and coded.Two blinded analyzers counted nuclei and individual cells or polykaryocytes to determine the fusion index.Fusion index was calculated according to fusion index = 1 -(number of cells/number of nuclei) from three fields of view.Fusion indices were plotted for each pH 0.2 incremental step of the assay to determine the optimal pH for fusion induction.To assess the impact of peptide treatment on fusion inhibition, the fusion assay was completed using rCCHFV-M/Cherry.Huh-7 cells were treated with lipopeptides at 10 μM for 1.5 h, while rCCHFV-M/Cherry (MOI = 3) was co-incubated with lipopeptides at a final concentration of 10 μM for 1 h.The peptide treatment was removed from the cells, and the virus/lipopeptide mixture was inoculated.Cells were exposed to acidified media, fixed, stained, imaged, and counted as above.

Statistical analysis
Statistically significant differences in the mean of plaque reduction and fusion indices between peptides were assessed using a one-way ANOVA with Tukey's Post-Hoc Test to correct for multiple comparisons.The CC 50 and IC 50 values were determined using a four-parameter, variable slope, nonlinear regression model with a 90% confidence interval for errors.All statistical analyses were completed using GraphPad Prism software version 8 (La Jolla, CA, U.S.A.).

Identification of regions of interest and peptide synthesis
A consensus mapping of all GenBank available prototype nairovirus M-segment, GPC open reading frames (ORF) was created.Residues were mapped based on identities (Supplemental Fig. 1), and nine regions of interest were identified along the C-termini of nairovirus GPCs, modeled off of empirical and putative nairovirus G C data (Bergeron et al., 2007;Clerx et al., 1981;Kuhn et al., 2016;Marriott et al., 1992;Punch et al., 2018;Sanchez et al., 2002).These nine regions possessed either multiple contiguous residues, more than 80% identity at the family level, or were implicated by previous studies to possess functional relevance (Bertolotti-Ciarlet et al., 2005;Garry and Garry, 2004;Sanchez et al., 2002).Peptides were designed against these nine regions of interest with cysteine residues replaced with serine residues to maintain secondary structure integrity of CCHFV-G C upon peptide bindings.
From the nine regions of interest, seven peptide sequences were selected for synthesis, and four of the seven peptides were synthesized in yields sufficient for PEG/sterol conjugations and subsequent purifications; these peptides were designated as Sequences 1 through 4 and are outlined in Fig. 1.Sequence 1 (abbreviated as Seq1 throughout) was generated from residues 147-169 within the CCHFV-G C mature protein and had the highest total nairovirus identity, with 40.9% of residues being identical at the family level and contains the class II nairovirus fusion loops (Garry and Garry, 2004;Mishra et al., 2022).Seq2 and Seq3, each 15 and 21 amino acids long, respectively, were designed against elements within the CCFV-G C domain II (Garry and Garry, 2004;Mishra et al., 2022).Seq4 was designed against 16 residues in a region upstream of the transmembrane domain, which possessed six highly conserved amino acids among the consensus nairovirus GPC sequence data (Supplemental Fig. 1).
Each of these four sequences was then linked using a -GSG-linker to a sterol (either cholesterol or tocopherol) and variable-sized PEGs on either the N-terminus or C-terminus of the synthesized peptide.Either cholesterol or tocopherol was conjugated onto the linker, or no sterol group was added.A PEG linker of 4 or 24 polyethylene glycol groups was also incorporated with the various sterol group conjugations.Ultimately, Seq1 and Seq4 had both N-and C-terminus peptides linked with cholesterol groups PEGylated with either a PEG4 or PEG24 (termed 'Chol4' and 'Chol24', respectively), a tocopherol PEGylated with a PEG24 (termed 'Toc24'), and groups with no sterol conjugation (termed 'No Sterol').Seq2 was created similarly; however, an additional lipopeptide with a PEG4 tocopherol (termed 'Toc4') was also synthesized.Seq3 was synthesized in quantities to yield only N-terminus modified peptides with Chol4, Chol24, and Toc24 conjugated molecules.In total, 29 different peptide-conjugate combinations were generated in sufficient quantity for screening and further assays.Fig. 1 describes these peptides, their binding regions, and their conjugations.

Peptide screening
The previously developed pseudotyped recombinant vesicular stomatitis virus (rVSV) expressing CCHFV surface glycoprotein, ΔGrVSV-CCHFV-GPCΔ, was used as a neutralizing screening tool in lowcontainment settings (BSL-2) (Rodriguez et al., 2019).To validate the pseudotype as a screening tool for antiviral peptides targeting CCHFV-G C , CCHFV-G C neutralizing antibodies, 11E7, 12A9, and 8A1, demonstrated neutralization of the pseudotype, with the antibody 11E7 demonstrating the highest neutralization with a PRNT50 > 1:6400 (Supplemental Fig. 2).Given that CCHFV-GPC antibodies could successfully bind to and neutralize the pseudotype, all peptide-conjugate combinations were screened by PRNT using this virus.
Screening results are shown in Fig. 2A and Supplementary Fig. 3, where all treatment and control groups were normalized to mock, i.e., if the number of starting plaque-forming units were not changed by the treatment, zero percent reduction to mock was the reported value.Chloroquine and HMAF were selected as controls to establish a baseline PRNT value for the pseudotype/peptide-inhibitor PRNT assay and are shown in black in Fig. 2A.All twenty-nine lipopeptides were screened against the pseudotyped virus at 10 μM (Fig. 2A Residues from the open reading frames (ORF) of all GenBank available nairovirus genomic M-segments were aligned to identify regions of interest.Regions of interest possessed either multiple contiguous residues, more than 80% identity at the family level, or implicated by previous studies to possess functional relevance.Four sequences (Seq1-Seq4) were synthesized in yields sufficient for PEG/sterol conjugations and subsequent purifications.Each of these four sequences was then linked using a -GSG-linker to a sterol (either cholesterol or tocopherol) and variable sized PEG molecule on either the N-terminus or C-terminus of the synthesized peptide.Peptides not conjugated to a sterol group "No Sterol" included the -GSG-linker.Each peptide sequence is shown in the table with its respective G C domain, residue location, and percent identity across all available nairovirus ORFs.The schematic shows the relative location of each peptide sequence in relation to the CCHFV G C .Boxes are colored to indicate N-terminal or C-terminal sterol conjugations, which are maintained throughout the manuscript.
To assess the interactions of peptides and the VSV backbone (i.e., non-CCHFV-GPC related proteins) in our screening approach, we ran a PRNT assay with four peptides that demonstrated high levels of inhibition of the pseudotyped virus (Seq2-N-Chol24, Seq2-C-Chol24, Seq4-N-Chol4, and Seq4-C-Chol4).Supplemental Fig. 4 shows that under similar conditions, a wild-type VSV has decreased levels of percent inhibition when using the highest performing peptides from the pseudotyped virus assay.This data reinforced that the identified peptides were inhibiting the CCHFV glycoproteins on the pseudotyped surface, and therefore, ran this assay on wild-type CCHFV.
Four of the seven lipopeptides demonstrating greater than or equal to 70% pseudotyped virus neutralization were further analyzed by PRNT on wild-type CCHFV, strain IbAr10200, chosen based on lipopeptide availability (Seq2-N-Chol24, Seq2-C-Toc4, Seq4-N-Chol4, and Seq4-C-Chol4).The percent of plaque reduction by each lipopeptide is shown in Fig. 2B.Three of the four lipopeptides demonstrated greater inhibition of wildtype CCHFV than the pseudotyped virus: Seq2-N-Chol24, 89% vs 72%; Seq2-C-Toc4, 95% vs 70%, and Seq4-C-Chol4, 98% vs 97%.The lipopeptide Seq4-N-Chol4 was the only lipopeptide to show a reduction in inhibition against the wildtype virus compared to the pseudotyped virus, with a 53% reduction against wildtype CCHFV versus 79% reduction against the pseudotyped virus.Overall, these data demonstrate that the pseudotyped virus can serve as an effective screening tool for virus inhibition studies before confirmation of the results in high containment settings using wild-type CCHFV.

Determination of 50% cell cytotoxic concentrations (CC 50 )
The MTT Cell Proliferation Assay (ATCC) was used to assess the cytotoxicity of the peptides from 0.001 μM to 100 μM in Huh-7 cells.Fig. 3A shows the CC 50 curves for the four top peptides and scramble control.Two peptides, Seq2-N-Chol24 and Seq2-C-Toc-4, showed minimal impact on cell viability at 100 μM, and the CC 50 could not be determined.The CC 50 concentrations were determined for three of the five peptides assessed and were approximately 23.2 μM, 14.6 μM, and 1.6 μM for Scramble-Chol4, Seq4-N-Chol4, and Seq4-C-Chol4, respectively.These results suggest that cytotoxicity is not likely to be a confounding factor on viral inhibition when screening at 10 μM, for Seq2-N-Chol24, Seq2-C-Toc-4, and Seq4-N-Chol4.

Nairovirus fusion assay
CCHFV relies on early endosomal trafficking and multivesicular bodies for cell entry (Garrison et al., 2013;Shtanko et al., 2014), and the pH during these steps can range down to approximately 6.0 (Fels et al., 2021;Garrison et al., 2013;Hua et al., 2020;Mishra et al., 2022).To develop a nairovirus fusion assay, pH was altered in cell culture systems with medium buffered at 0.2 increments descending from 7.4 (approximate basal pH of the body) down to 5.2, the late endosomal pH, at which nairoviruses are not expected to induce fusion (Garrison et al., 2013;Shtanko et al., 2014).The fusion indices for Fig. 2. Screening and selection of peptides on pseudotyped virus (A) and wildtype CCHFV (B).BHK or SW-13 cell monolayers and either the pseudotyped virus or wildtype CCHFV were treated separately with peptides or controls prior to virus inoculation.Each peptide was suspended in cell culture media with a final concentration of 10 μM peptide and incubated onto cells for 1.5 h.A known concentration of virus was incubated with each peptide at a final concentration of 10 μM peptide for 1 h prior to inoculation.HMAF at a final dilution of 1:160, and chloroquine treatment at 10 μM were used as controls.Panel A shows 23 of the 29 total peptides screened against the pseudotyped virus that demonstrated reduction in virus infection compared to mock.Panel B shows four peptides that had greater than or equal to 70% pseudotype reduction selected for assessment of plaque reduction against wildtype CCHFV, strain IbAr10200.Percent reduction is calculated by comparison to mock infected (~50 pfu) control wells on BHK cells, and ****p = <0.0001.Individual data points are shown, with bars indicating means (±SD).For screening on the pseudotyped virus (A) data points represent the average of two technical replicates from two independent experiments.For screening on wildtype CCHFV, data points are two technical replicates from three independent experiments.Fig. 3. Determination of 50% cytotoxicity (A) and 50% virus inhibition (B) with selected peptides.(A) Huh-7 cells were incubated with increasing concentrations of the lipopeptides for two and a half hours before the media was replaced.Cells were incubated for 48 h after treatment to mimic the cell treatment during the PRNT screening experiment.Cellular metabolism was then measured using the MTT assay as a surrogate for cellular cytotoxicity.Dose response curves were then generated based on the relative absorbance at 570 nm.The CC 50 concentrations were approximately 23.2 μM, 14.6 μM, and 1.6 μM for Scramble-Chol4, Seq4-N-Chol4, and Seq4-C-Chol4, respectively.Only Seq4-N-Chol4 and Seq4-C-Chol4 had reported CC 50 errors of ±847.0 and ± 1.1, respectively (90% confidence interval).(B) Huh-7 cells were mock infected (Mock control), infected with 2.5e3 PFU rCCHFV-Ibar10200-ZsGreen without inhibitors ('Virus Only' control), or infected with 2.5e3 PFU rCCHFV-Ibar10200-ZsGreen in the presence of increasing concentrations of fusion inhibitory peptides (colors).Relative fluorescence was measured 48 h post infection, and relative fluorescent units are shown as a function of the (log-scale) concentration of peptide.The IC 50 concentrations for Seq4-C-Chol4, Seq2-C-Toc4, Seq2-N-Chol24, and Seq4-N-Chol4 are approximately 6.4, 10.9, 10.7, and 15.7 μM, respectively.Only Seq4-C-Chol4 and Seq2-N-Chol24 had reported IC 50 errors of ±60.2 and ± 307.0, respectively (90% confidence interval).Data points are means (±SD) for three biological replicates.The CC 50 and IC 50 values were determined using a four parameter, variable slope, nonlinear regression model in GraphPad Prism.were infected with a MOI of 2 in 8-well chamber slides or transfected at 1 μg per cover slip in a 6-well plate.Cells shown in panels are example images taken from HAZV infected cells.Cell culture medium was pH-calibrated and sterile filtered and incubated for 30 min to induce fusion.Cells were then stained and fixed between 18 and 24 h post infection/treatment.Nairovirus-induced fusion is induced between pH 6.6-6.2.Cells were fixed briefly with paraformaldehyde and left unpermeabilized.Stained with a CellMask Green Plasma Membrane Stain (ThermoScientific).CCHFV-GPC transfected cells, ΔGrVSV-CCHFV-GPCΔ, HAZV, DUGV, and CCHFV fusion index was calculated as stated in methods from blinded counts of nuclei and cell bodies, with three fields of view per technical replicate with experiments were done in triplicate.Dotted baseline indicates the averaged fusion index in infected but not pH-treated control cells.pCAGGS-CCHFV-GPC, ΔGrVSV-ΔCCHFV-GPC, HAZV, DUGV, and CCHFV, from pH 7.4 to 5.2 are shown in Fig. 4. Transfection of Huh-7 cells with pCAGGS-CCHFV-GPC demonstrated few polykaryocytes at pH 5.8 and increased down to pH 5.2.These findings did not align with the current published data supporting that fusion occurs during the early endosomal pathway, though transfection efficiencies and surface expressions could impact syncytial formations (Garrison et al., 2017;Hua et al., 2020).The CCHFV pseudotype, ΔGrVSV-ΔCCHFV-GPC, was assessed for fusion induction in BHK cells, and polykaryocyte formation occurred below pH 6.4.Nairovirus (HAZV, DUGV, and CCHFV) infected Huh-7 cells demonstrated polykaryocyte formation at pH < 6.6, with widespread polykaryocyte formation and a fusion index near 1.0 for DUGV and CCHFV at pH ≤ 6.4.The pseudotyped ΔGrVSV-ΔCCHFV-GPC demonstrated utility in both examining fusion using the CCHFV-G C in a different cell type and had a similar pH range as wild-type CCHFV.In contrast, the plasmid base system (pCAGGS-CCHFV-GPC) did not perform similarly.Cumulatively, this data range supports the findings of other studies that nairovirus fusion occurs between pH 6.6-6.2 (Fels et al., 2021;Garrison et al., 2013;Hua et al., 2020;Mishra et al., 2022).

Lipopeptide fusion inhibition assay
To assess fusion inhibition properties of the selected peptides Seq2-N-Chol24, Seq2-C-Toc4, Seq4-N-Chol4, and Seq4-C-Chol4 against CCHFV-IbAr10200, immunofluorescence microscopy was used to observe the effects of cell-to-cell fusion by polykaryocyte formations (Fig. 5).Treatment with all four peptides significantly reduced the fusion index compared to the 'No Peptide' and Scramble-Chol4 controls.The peptides Seq4-C-Chol4 and Seq2-C-Toc4 demonstrated the greatest fusion inhibition, with average fusion indices of 0.25 and 0.33, respectively.The peptides Seq2-N-Chol24 and Seq4-N-Chol4 were less effective at inhibiting fusion and had average fusion indices of 0.58 and 0.72, respectively.

Discussion
Inhibitory peptides are long being explored for their potential to inhibit viral infection.Thus far, only one inhibitory peptide is used in the clinic, but pre-clinical studies for all major virus families of human concern are being performed (Barriga et al., 2016;de Vries et al., 2021;Hrobowski et al., 2005;Koehler et al., 2013;Lee et al., 2011;Mathieu et al., 2015;Miller et al., 2011;Outlaw et al., 2020;Pessi et al., 2019;Porotto et al., 2010aPorotto et al., , 2006;;Saito et al., 2021;Aaron G Schmidt et al., 2010;Aaron G. Schmidt et al., 2010;Welsch et al., 2013;Yu et al., 2017).Inhibitory peptides have multiple advantages over other therapies and may also be used preventatively.They have a long shelf-life, do not need cold chains, are highly specific with low immunogenicity, and, dependent on the application, are suggested to be administered via aerosol or inhalation, making treatment of hard-to-reach-populations possible.Challenges in inhibitory peptide design include their rapid clearance and proteolytic susceptibility leading to a relatively short half-life.This can be partly overcome by lipid-anchoring (Chong et al., 2017;Porotto et al., 2010b;Xia et al., 2020) as lipopeptides are targeted to cellular membranes and dimerization of peptides (de Vries et al., 2021;Pessi et al., 2012), as shown by their retention in the respiratory tract after nasal administration.In contrast, monomeric peptides are washed out into the bloodstream (de Vries et al., 2021;Figueira et al., 2017).
We sought to design inhibitory peptides to CCHFV and explore potential mechanisms of their inhibition.Using conserved regions, we identified key areas on the CCHFV-G C , synthesized peptides to these regions, conjugated them to sterol-PEG groups, and screened peptides in a neutralization assay against pseudotyped CCHFV.Peptides that neutralized the pseudotype comparably, or better than, CCHFV hyperimmune mouse ascitic fluid, were further evaluated for cytotoxicities and neutralizations on wild-type CCHFV.A nairovirus immunofluorescence fusion assay was also developed for HAZV, DUGV, CCHFV, pseudotyped CCHFV, and in trans expression of CCHFV-GPC, and selected peptides were evaluated on their abilities to impact viral fusion.
Peptides were synthesized using sequences from four conserved regions of the CCHFV Gc (Fig. 1; Supplemental Fig. 1).Surprisingly, peptides synthesized using Seq1, which is part of the Gc loops, demonstrated low levels of inhibition of the pseudotyped virus.Treatment with peptides synthesized using Seq3 resulted in increased plaque-forming units compared to the mock infection controls (Supplementary Fig. 3).This region overlaps with a peptide region assessed by Koehler et al. (2013), where the RVFV peptide did not inhibit RVFV infection (Koehler et al., 2013).These data suggest that targeting the regions of Seq1 and Seq3 may not be an efficient strategy for countermeasures developments against CCHFV.It is unknown why peptides targeting Seq3 enhanced plaque-forming units in the assay; however, it could be an artifact of the pseudotyped CCHFV used to screen or could be a conformational change to the G C leading to increased binding of the pseudotype.Peptides synthesized using Seq2, downstream of the fusion domain, were highly effective for virus inhibition.This region was not targeted by other related bunyavirus inhibitory peptides studies, though this selected region has a high degree of similarity among nairoviruses.This region is directly downstream of an N-linked glycan epitope, the linear binding region of the highly neutralizing monoclonal antibody 11E7, and the region that putatively binds cell surface nucleolin (Ahmed et al., 2005;Erickson et al., 2007;Xiao et al., 2011).Our results suggest that neutralization of CCHFV by these peptides likely involves fusion reduction.This neutralization mechanism is supported by other studies.Mishra et al., show that a probable mechanism for neutralizing antibodies is inhibiting CCHFV entry with a mechanism that prevents membrane fusion (Mishra et al., 2022).However, given that our methodology employed pre-incubation and incubation during in vitro infections with either pseudotyped or wild-type CCHFV, the mechanism remains unclear.Peptides synthesized using Seq4 were also effective for virus inhibition.This region is directly upstream of the original targeted 'stem' region, which in ANDV and RVFV inhibitory peptide studies had the greatest reduction in the infectivity of both viruses (Koehler et al., 2013).For these studies, peptides from Seq5-7, corresponding to the stem and transmembrane domain region, were not synthesized in sufficient quantities for downstream applications such as screening, though it would be interesting to test this region as well.
Previous nairovirus fusion assays have relied on conical-filtration apparatus concentrations and urografin gradient density purification to induce viral-fusion induced polykaryocyte formations at approximately pH 6.0 (Green et al., 1992) or have used viral supernatant incubations with pH gradient media (Garrison et al., 2013).Similar transfection methods have been employed with CCHFV-GPC and a GFP reporter in a cell-to-cell immunofluorescence fusion assay (Hua et al., 2020).Other studies have utilized monomeric and oligomeric forms of modified CCHFV-G C in acidified conditions (Mishra et al., 2022).Transfection of Huh-7 cells with CCHFV-GPC demonstrated a few polykaryoctes beginning to form at pH 5.8 and increasing down to pH 5.2.These findings did not align with the current published data supporting that fusion occurs during the early endosomal pathway.A rationale for this, is that the expression system does not adequately allow for fusion events at the cell surface using the CCHFV-G C or that transfection expression efficiencies limit the quantity of CCHFV-G C that can act at the cell surface in a biologically functional manner.While, overall, transfection may achieve the goal of nairovirus fusion, it may not be the most straightforward approach to be able to assay fusion inhibitors with.
Our study limitations involve differentiating the type of neutralizations employed by the peptides.Since we incubated pre-infection and during infection, we cannot discern whether the peptides bind to infecting viral particles or if the peptides are binding to G C residues within an acidified infected endosome.We chose a broader approach to detect as many potent inhibitory peptides as possible within our screen.Future studies should be employed to further dissect mechanisms of neutralization as employed by other functional studies.Additionally, our study is limited in testing single peptides and did not seek to evaluate synergistic effects.It would be ideal to test combinations of these peptides in future studies.The IC50 values of these peptides are in the micromolar range.These concentrations are in agreement with the IC50 concentrations of ribavirin and favipiravir (T-705) in vitro (<10 μM), where favipiravir has demonstrated in vivo efficacy (Hawman et al., 2018;Oestereich et al., 2014).Our screening results (Fig. 2), and other studies, have shown that conjugate type and linker influence inhibition capacity (Lee et al., 2011).It is possible that sterol conjugate modifications to Seq2 or Seq4 peptides could yield a lipopeptide demonstrating virus inhibition within the nanomolar range, as has been seen with other fusion inhibitory peptides.
Overall, our study agrees with other studies suggesting countermeasure designs that target the fusion or stem regions of CCHFV-G C could be effective protective epitope regions.Vaccines eliciting responses that target these regions may offer protection, and therapies that bind or sterically hinder these areas could be useful prophylactically, as peptides targeting these regions appeared most neutralizing and prevented fusion induction in our nairovirus fusion assay.This study further demonstrated the utility of pseudotyped CCHFV, as much of the inhibition (either PRNT or fusion indices) trends were similar between pseudotyped and wild-type viruses.Lastly, this study provides two new peptide inhibitors for CCHFV that warrant further studies to examine inhibitory concentrations and in vivo studies.
M.C.Mears et al.
M.C.Mears et al.

Fig. 1 .
Fig. 1.Selection of peptide sequences on CCHFV-G C .Residues from the open reading frames (ORF) of all GenBank available nairovirus genomic M-segments were aligned to identify regions of interest.Regions of interest possessed either multiple contiguous residues, more than 80% identity at the family level, or implicated by previous studies to possess functional relevance.Four sequences (Seq1-Seq4) were synthesized in yields sufficient for PEG/sterol conjugations and subsequent purifications.Each of these four sequences was then linked using a -GSG-linker to a sterol (either cholesterol or tocopherol) and variable sized PEG molecule on either the N-terminus or C-terminus of the synthesized peptide.Peptides not conjugated to a sterol group "No Sterol" included the -GSG-linker.Each peptide sequence is shown in the table with its respective G C domain, residue location, and percent identity across all available nairovirus ORFs.The schematic shows the relative location of each peptide sequence in relation to the CCHFV G C .Boxes are colored to indicate N-terminal or C-terminal sterol conjugations, which are maintained throughout the manuscript.

Fig. 5 .
Fig. 5. Influences of selected inhibitory peptides on CCHFV fusion.Huh-7 cells were treated with lipopeptides at 10 μM for 1.5 h, while rCCHFV-M/Cherry (MOI = 3) was co-incubated with lipopeptides at a final concentration of 10 μM for 1 h.The peptide treatment was removed from the cells, and the virus/lipopeptide mixture was inoculated.Cells were exposed to acidified media, fixed, stained, imaged, and counted as described in Methods.Representative images of each treatment condition are shown on the left.Quantification of fusion using the fusion index is shown on the right.Individual data points are shown with bars for means (±SD) for technical triplicates across two independent experiments or more, ***p = 0.0007, ****p = <0.0001.