Chikungunya virus glycoproteins transform macrophages into productive viral dissemination vessels

Despite their role as innate sentinels, macrophages are cellular reservoirs for chikungunya virus (CHIKV), a highly pathogenic arthropod-borne alphavirus that has caused unprecedented epidemics worldwide. Here, we took interdisciplinary approaches to elucidate the CHIKV determinants that subvert macrophages into virion dissemination vessels. Through comparative infection using chimeric alphaviruses and evolutionary selection analyses, we discovered for the first time that CHIKV glycoproteins E2 and E1 coordinate efficient virion production in macrophages with the domains involved under positive selection. We performed proteomics on CHIKV-infected macrophages to identify cellular proteins interacting with the precursor and/or mature forms of viral glycoproteins. We uncovered two E1-binding proteins, signal peptidase complex subunit 3 (SPCS3) and eukaryotic translation initiation factor 3 (eIF3k), with novel inhibitory activities against CHIKV production. These results highlight how CHIKV E2 and E1 have been evolutionarily selected for viral dissemination likely through counteracting host restriction factors, making them attractive targets for therapeutic intervention.


13
Despite their role as innate sentinels, macrophages are cellular reservoirs for chikungunya virus 14 (CHIKV), a highly pathogenic arthropod-borne alphavirus that has caused unprecedented 15 epidemics worldwide. Here, we took interdisciplinary approaches to elucidate the CHIKV 16 determinants that subvert macrophages into virion dissemination vessels. Through comparative 17 infection using chimeric alphaviruses and evolutionary selection analyses, we discovered for the 18 first time that CHIKV glycoproteins E2 and E1 coordinate efficient virion production in 19 macrophages with the domains involved under positive selection. We performed proteomics on 20 CHIKV-infected macrophages to identify cellular proteins interacting with the precursor and/or 21 mature forms of viral glycoproteins. We uncovered two E1-binding proteins, signal peptidase 22 complex subunit 3 (SPCS3) and eukaryotic translation initiation factor 3 (eIF3k), with novel 23 inhibitory activities against CHIKV production. These results highlight how CHIKV E2 and E1 24 have been evolutionarily selected for viral dissemination likely through counteracting host 25 restriction factors, making them attractive targets for therapeutic intervention.  Chikungunya virus (CHIKV) is a highly pathogenic arthropod-borne alphavirus that has expanded 38 worldwide with emerging lineages in recent decades 9,10 . The unprecedented outbreaks from the 39 Indian Ocean islands to Southeast Asia were caused by the novel CHIKV Indian Ocean lineage 40 (IOL), characterized primarily by the E1-A226V mutation that adapted the virus from its principal 41 vector Aedes aegypti to Aedes albopictus [11][12][13] . Although CHIKV infection is typically cleared in a 42 few days, a significant percentage of individuals develops incapacitating arthralgia for up to 20 43 months 14-16 . Interestingly, CHIKV RNA and proteins persist in monocyte derived macrophages 44 (MDMs) in spleen or synovial tissue for months in macaques and humans suffering from chronic 45 arthralgia [17][18][19] . These studies propose a role for macrophages as a cellular reservoir for CHIKV 46 persistence and a niche for inflammation that is recurrently activated by viral components 17,20 . 47 However, it is not clear what mechanism drives CHIKV persistence and whether this pathogenic 48 role of macrophages is found in all arthritogenic alphavirus infections. 49 In contrast, o'nyong'nyong virus (ONNV), an arthritogenic alphavirus that shares the most genetic 50 identity with CHIKV, is confined to periodic outbreaks in Africa 9,21 . ONNV causes similar 51 symptoms in humans, but is less virulent in mouse models, requiring a higher dose than CHIKV 52 to reach the same level of mortality 22 . The evolutionary similarities yet epidemiological differences 53 make CHIKV-ONNV chimeras excellent molecular tools for probing viral determinants for host 54 adaptation. However, these studies so far have mostly focused on their differential uses of 55 mosquito vectors, such as transmission of ONNV by Anopheles gambiae 23 viruses to evade or antagonize these blockades, such as the arms race between myeloid-cell-64 specific SAMHD1 and HIV-2 Vpx 27-29 . This prompted us to question how and to what extent the 65 evolutionary pressure brought on by virus-host arms race has selected for increased CHIKV 66 survival in human macrophages. 67 Here, we found that human primary monocyte and THP-1 derived macrophage infection with 68 CHIKV (vaccine strain 181/clone 25) is much more efficient than that of ONNV at a step following 69 genome replication. By utilizing a repertoire of CHIKV-ONNV chimeras, we mapped the viral 70 determinant for efficient virion production in macrophages to the CHIKV E2 and E1 glycoproteins. 71 Interestingly, evolutionary analysis of 397 CHIKV structural polyprotein sequences isolated from 72 infected individuals uncovered signatures of positive selection mostly in E2 and E1 proteins. 73 Mutating two of the positively selected residues in CHIKV to the homologous ones in ONNV (E2-74 V460L, E1-V1029I) attenuates virion production in 293T and BHK-J cells while the E1-V1029I 75 mutation completely abolishes virion production in macrophages. We further performed affinity 76 purification-mass spectrometry (AP-MS) on macrophage interactors of CHIKV glycoproteins. We 77 discovered that E1 consistently interacts with signal peptidase complex subunit 3 (SPCS3) and 78 eukaryotic translation initiation factor 3 (eIF3k), which block CHIKV production in macrophages.

79
Taken together, we found that, in addition to their role in viral entry, CHIKV glycoproteins may 80 interfere with cellular restrictions to facilitate virion dissemination in macrophages. 81 82

CHIKV infects human macrophages more efficiently than other arthritogenic alphaviruses. 84
To evaluate the susceptibility of macrophages to different arthritogenic alphaviruses, we infected 85 human primary monocyte derived macrophages with EGFP-expressing Sindbis virus (SINV), 86 Ross River virus (RRV), ONNV, and CHIKV, and quantified infection levels at 24 hours post-87 infection (h.p.i.) by flow cytometry (Figure.1a). Despite generally low infection rates with these 88 alphaviruses (<1%), we observed a small percentage (0.76%) of macrophages highly infected with 89 CHIKV, according to intracellular EGFP expression that spans 3 logs. We then compared growth 90 kinetics of CHIKV and its closest relative, ONNV, in infected human monocytic cell line THP-1 91 derived macrophages by quantifying virion production in the supernatant ( Figure.1b). We found 92 that CHIKV produces 2-3 logs higher titers than ONNV throughout the infection time-course (up 93 to 1.05x10 7 pfu/ml for CHIKV compared to 3.75x10 4 pfu/ml for ONNV), with the titers of both 94 viruses peaking at 24 h.p.i. These results suggest that the small number of CHIKV-infected 95 macrophages is extremely efficient at producing viral progeny. 96 We asked whether the high level of CHIKV production is achieved by enhanced viral replication 97 in macrophages. To bypass viral entry, we directly transfected in vitro transcribed genomic viral 98 RNAs (vRNAs) of CHIKV and ONNV into THP-1-derived macrophages ( Figure.1c). We 99 measured intracellular negative-sense viral RNA ((-) vRNA), the replication intermediate, by 100 TaqMan RT-qPCR assays. To our surprise, the (-) vRNA levels of ONNV are significantly higher 101 than those of CHIKV following vRNA transfection, suggesting CHIKV infection is enhanced at a 102 step after genome replication in macrophages ( Figure.1c). Nevertheless, virion production of 103 CHIKV is dramatically more robust than that of ONNV and could be detected as early as 8 hours 104 post-transfection (h.p.t.) ( Figure.1d). Taken together, human macrophage infection with CHIKV 105 drives more superior virion production than that with ONNV. 106 107 CHIKV E2 and E1 synergize to mediate efficient virion production in THP-1 derived 108 human macrophages. 109 To identify the viral determinants for CHIKV infection of human macrophages, we constructed 110 several CHIKV-ONNV chimeras ( Figure.2a) and assessed their infection levels in THP-1 derived  111 macrophages, compared to parental CHIKV and ONNV. Alphaviruses express 4 non-structural 112 proteins (nsP1-4) for viral replication, and 5 structural proteins from a subgenomic mRNA (capsid, 113 E3, E2, 6K/TF, E1) for viral particle assembly and host cell entry 30 . These proteins are 114 proteolytically processed from the non-structural and structural polyproteins. Given the genome 115 organization, we generated Chimera I that contains ONNV nsP1 to capsid in a CHIKV backbone, 116 and Chimera III that contains CHIKV nsP1 to capsid in an ONNV backbone. To account for 117 potential discrepancies associated with mismatched subgenomic promoters located at the 3' end 118 of nsP4 and structural proteins, we also generated Chimeras II and IV, where the swapping of viral 119 genes starts with the subgenomic promoters in CHIKV and ONNV nsP4. We found comparable 120 levels of virion production of Chimeras I and II as CHIKV in the supernatant of infected 121 macrophages, while Chimeras III and IV recapitulate ONNV production ( Figure.2b). These data 122 demonstrate that the viral determinants for effective macrophage infection lie in the CHIKV E3-123 E2-6K-E1 structural polyprotein region. 124 To investigate the role of CHIKV structural proteins in virion production, we transfected vRNAs 125 of CHIKV, ONNV, and Chimeras I-IV into THP-1 derived macrophages to bypass viral entry. We Consistent with Figure.2b, transfection of viral genomes without CHIKV E3-E2-6K-E1 (ONNV, 129 Chimera III, and Chimera IV) led to significantly lower levels of virion production. 130 To further narrow down the viral determinants for CHIKV infection in macrophages, we 131 constructed three additional chimeras in the context of Chimera III to include CHIKV E3 (Chimera  132  III-I), E3-E2 (Chimera III-II), or E3-E2-6K (Chimera III-III) (Figure.2d). Upon macrophage  133  infection with CHIKV, ONNV, and the chimeras, we found that only Chimera III-II and Chimera  134  III-III, both possessing CHIKV E2, partially enhance virion production at 24 and 48 h.p.i. although  135 not significantly ( Figure.2e), suggesting that E2 alone is not sufficient. Chimera III-III with all the 136 CHIKV structural proteins except E1 fails to fully rescue virion production in macrophages. Taken  137 together, this supports the involvement of both CHIKV E2 and E1 in virion production. 138 To pinpoint the impact of CHIKV E2 and E1 on virion production, we generated three chimeras 139 in the ONNV backbone with CHIKV replacement of E2 (ONNV/CHIKV E2), E1 (ONNV/CHIKV 140 E1), or E2 and E1 (ONNV/CHIKV E2+E1) ( Figure.3a). Neither single replacement of CHIKV E2 141 nor E1 rescues ONNV infection of macrophages to comparable levels as CHIKV ( Figure.3b. 142 Surprisingly, macrophage infection with ONNV/CHIKV E1 is more attenuated than that with 143 ONNV, potentially due to incompatible heterodimer formation between ONNV E2 and CHIKV 144 E1. In contrast, simultaneous replacement of E2 and E1 with CHIKV homologs (ONNV/CHIKV 145 E2+E1) increased the supernatant titers to levels even higher than those of CHIKV. We then 146 transfected vRNAs into macrophages to evaluate viral replication and production ( Figure.3c). All 147 of the transfected vRNAs launched productive viral replication in macrophages; however, only the 148 transfection of ONNV/CHIKV E2+E1 RNA led to significantly enhanced virion production, albeit 149 at levels lower than those for transfection of CHIKV RNA ( Figure.3c). These results highlight the 150 requirement of both CHIKV E2 and E1 for efficient virion production in macrophages. 151 To map the viral determinants for virion production to specific domains, we strategically swapped 152 in Therefore, we asked whether residues in the CHIKV glycoproteins have been under positive 171 selection to overcome antiviral immunity and productively replicate in macrophages. We applied 172 the same methodology from a highly cited SARS-CoV-2 study 32 . Combining the use of the fixed 173 effects likelihood 35 (FEL, p≤0.05) and mixed effects model of evolution 36 (MEME, p≤0.05) 174 methods, we analyzed 397 CHIKV structural polyprotein sequences from NCBI Virus 37 database 175 that were isolated from CHIKV-infected individuals globally ( Figure.4a and Extended data.1a).

213
Gene ontology enrichment analysis visualized with enrichment map ( Figure.5c) showed that host 214 factors are highly enriched in grouped or singleton biological processes (BPs) relevant to the 215 secretory pathway in blue (negative regulation of endopeptidase activity, signal peptide processing, 216 ER-Golgi transportation, intracellular vesicle transport) and innate immune responses in green 217 (type I IFN pathway/complement activation, antigen presentation) ( Figure.5c). We also performed 218 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  PKR 51-53 , knockdown of the host genes SPCS3 and EIF3K significantly restores virion production 237 by about 5-fold. To confirm that the antiviral activities observed in Figure.6b are specific to a step 238 after viral entry, we knocked down the same host factors in macrophages followed by transfection 239 of CHIKV vRNA ( Figure.6c). We found that silencing of most of the genes enhances virion 240 production in vRNA-transfected macrophages. CHIKV production in macrophages with OAS3, 241 SPCS3, and EIF3K knockdown is significantly higher than that in NT-transfected cells, despite 242 similar intracellular vRNA levels. 243 To confirm the interaction of CHIKV glycoproteins with host proteins demonstrating antiviral 244 activities (OAS3, SPCS3, eIF3k, APOBEC3F, and PKR, Figure.6b-c), we transfected 293T cells 245 with plasmids expressing 3XFLAG-tagged host factors, followed by transfection with CHIKV 246 vRNA ( Figure.6d) or CHIKV polyprotein (E3-myc-E2-6K-E1) expressing plasmid (Extended data 247 4b). The host factors were pulled down to probe for glycoproteins in precursor or mature forms.

248
We consistently detected strong binding of E1 and moderate binding of E3-E2-6K-E1 to SPCS3 249 and eIF3k, which suggests that host factors interact with precursors and E2/E1 heterodimer 250 through binding to E1. Meanwhile, we did not observe any binding of E2 or p62 to these selected 251 host factors. Taken together, these results demonstrate that E1 interferes with intrinsic host 252 blockades through sequestering unconventional antiviral proteins. 253 254

255
Macrophages are important cellular reservoirs for persistent CHIKV infection; however, the 256 underlying mechanisms are largely unexplored. In this study, we interrogated the CHIKV 257 determinants that hijack macrophages as virion disseminators. We first demonstrated that both 258 CHIKV glycoproteins E2 and E1 mediate efficient virion production from infected macrophages 259 through comparative infection with CHIKV-ONNV chimeras. By performing selection analysis 260 on sequences of human CHIKV isolates from NCBI Virus 37 , we identified E2-V460 and E1-V1029 261 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 29, 2023. ; https://doi.org/10.1101/2023.05.29.542714 doi: bioRxiv preprint to be associated with elevated CHIKV production. We then uncovered new antiviral factors SPCS3 262 and eIF3k by AP-MS analysis that bind strongly to E1 and restrict CHIKV production in 263 macrophages. Our results suggest that positive selection of E2/E1 potentially driven by viral 264 glycoprotein antagonism of cellular restrictions contributes to efficient CHIKV production in 265 macrophages. 266 The six positive selection sites (460, 489, 571, 1020, 1029, 1175) in the structural proteins that are 267 divergent between CHIKV (vaccine strain 181/clone 25) and ONNV (strain SG650) are primarily 268 in the E2 b-ribbon arches and E1 domain II, away from the embedded fusion peptide. All of the 269 differential selection sites are exposed on the outer surface of p62-E1 that is accessible for 270 interactions with host factors in the secretory pathway and can influence E2/E1 heterodimer 271 maturation. Our results showed that mutating CHIKV E2-V460 and E1-V1029 into ONNV 272 residues disrupts virion production with the former causing unexpected cleavage of E2, suggesting 273 that the positively selected valine residues in these sites might counteract host restriction factors 274 targeting maturation. Consistent with that, silencing of E2/E1 interactors identified by AP-MS 275 revealed primarily antiviral activities at a late step in the viral life cycle after genome replication. 276 Components of the adaptive immune response, such as inhibitory antibodies and T cells, can also 277 select for escape mutations in viral glycoproteins 25 . The epitopes of currently characterized human 278 CHIKV neutralizing antibodies or broadly anti-alphavirus antibodies are mainly mapped to E2 279 domains A and B, responsible for receptor binding and cellular attachment, and E1 domain II, 280 proximal to or within the fusion loop 54-56 . None of these reported antibodies target the six 281 differential selection sites in CHIKV 181/clone 25 ( Figure.4d), suggesting that these residues are 282 likely selected by host restriction factors. 283 Among the candidate CHIKV glycoprotein interactors we identified, SPCS3 and eIF3k have 284 inhibitory activities against CHIKV production in human macrophages. SPCS3 is one of the core 285 components of the endoplasmic reticulum-associated signal peptidase complex (SPC) 57 , which 286 cleaves signal peptides during the translocation of protein precursors in the ER 58,59 . The signal 287 peptidases are presumably usurped by flaviviruses, bunyaviruses, and alphaviruses for poly-288 glycoprotein cleavage 45,46,60 . However, it is unknown what exact peptidase releases p62, 6K, and 289 E1 from the alphavirus poly-glycoprotein precursor 61 . A previous genome-wide CRISPR knockout 290 screen uncovered both SPCS1 and SPCS3 as proviral factors for flavivirus infection, and depletion 291 of SPCS1 led to inefficient polyprotein cleavage disrupting flavivirus production 60 . Surprisingly, 292 we found that SPCS3 exhibits anti-CHIKV activity and strongly associates with CHIKV E1.

293
SPCS3 overexpression does not affect CHIKV poly-glycoprotein cleavage ( Figure. Data 3), we also detected host factors associated with eIF3k, including eIF2a, a crucial subunit of 299 eIF2, and receptor for activated C kinase 1 (RACK1). eIF2a phosphorylation can be induced by 300 double stranded RNA formed during viral replication, leading to translation shutoff that suppresses 301 viral gene expression 64 . Therefore, it is plausible that eIF3k antiviral activity is related to eIF2a-302 mediated translation inhibition. Moreover, previous studies indicated that RACK1 acts as a kinase 303 scaffold to potentially regulate the phosphorylation status of several eIF3 subunits 65,66 . Future 304 studies are needed to determine whether eIF3k regulates viral translation in a RACK1-dependent 305 manner. 306 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 29, 2023. ; https://doi.org/10.1101/2023.05.29.542714 doi: bioRxiv preprint In summary, our study has uncovered CHIKV determinants for virion production in macrophages 307 in the light of evolutionary selection and proposes a novel role for viral glycoproteins as 308 antagonists of host antiviral immunity. Overall, this research provides promising viral and cellular 309 targets for therapeutic intervention to strengthen the antiviral status of macrophages and eliminate 310 CHIKV reservoirs. 311 312 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made two unique Ambion Silencer siRNAs respectively targeting G3BP1 and G3BP2 were pooled (25 351 nM) and transfected into THP-1 macrophage. The same amount of non-targeting siRNA (Thermo 352 Fisher Scientific) was transfected into THP-1 macrophages as negative control. siRNA 353 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made To construct CHIKV with myc-tagged E2 (CHIKV/myc-E2), the myc tag was inserted between 394 E3 and E2 through the NEBuilder HiFi DNA Assembly Kit. Fragment 6 was amplified from 395 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made All the primers and restriction sites used in the construction of host factor and CHIKV structural 403 polyprotein plasmids are listed in Table 2 in Supplementary information 1, respectively. 404 The cellular mRNA from THP-1 cells was reverse transcribed with oligo-dT primer through the 405 Protoscript II First Strand cDNA Synthesis Kit (New England Biolabs) after TRIzol (Thermo 406 Fisher Scientific) extraction. The host genes OAS3, PKR, SPCS3, EIF3K, and APOBEC3F were 407 amplified with specific primers containing regions overlapping the pcDNA3.1-3xFLAG vector. 408 The cDNAs of host factors were then incorporated into the pcDNA3. genes were used to synthesize viral cDNAs from total RNAs. The transcribed cDNAs were then 420 quantified by SYBR Green or TaqMan qPCR. 421 The SYBR Green assay was used to evaluate the copy number of (+) vRNA in the samples. cDNAs were then serially diluted ten-fold from 10 -1 to 10 -8 and run through the SYBR Green assay 432 (New England Biolabs) together with sample cDNAs. Specific forward primer targeting E1 and a 433 reverse primer targeting the nongenomic tag were used in 20 ul SYBR Green reaction with 1x 434 Luna qPCR Dye (New England Biolabs) according to manufacturer's instructions. The reactions 435 were run under the cycling conditions as previously reported 73 . 436 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 29, 2023. ; https://doi.org/10.1101/2023.05.29.542714 doi: bioRxiv preprint The TaqMan assay was performed to determine the copy numbers of (-) vRNAs in the samples. 437 The standard curve of (-) strand nsP1 from CHIKV or ONNV, tagged reverse transcription primers, 438 qPCR primers, and TaqMan probes were designed and generated as previously described 75 . Briefly, 439 after in vitro transcription of (-) nsP1 transcripts from standard curve plasmids, the cDNAs of (-) 440 nsP1 transcripts and viral RNA in the samples were synthesized with a forward nsP1 primer 441 containing a unique tag sequence 5'-GGCAGTATCGTGAATTCGATGC-3' by the Protoscript II 442 First Strand cDNA Synthesis Kit. The appropriate reverse nsP1 primer, tag-specific forward 443 primer, and FAM-labeled TaqMan probe were used in viral negative strand quantification with 444 Luna To validate our mass spectrometry hits, HEK-293T cells were seeded in 6-well plates at a 477 starting density of 1.5 x 10 5 cells/well, followed by transient transfection with plasmids 478 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made subjected to reduction using 5 mM Tris (2-carboxyethyl) phosphine for 30 min and free 500 cysteine residues were alkylated by 10 mM iodoacetamide for another 30 min. Samples were 501 diluted with 100 mM Tris-HCl at pH 8 to reach a urea concentration of less than 2 M, and 502 then digested sequentially with Lys-C and trypsin at a 1:100 protease-to-peptide ratio for 4 503 and 12 hours, respectively. The digestion reaction was terminated by the addition of formic 504 acid to 5% (vol/vol) with centrifugation. Finally, samples were desalted using C18 tips 505 (Thermo Scientific, 87784) and dried in a SpeedVac vacuum concentrator, and reconstituted 506 in 5% formic acid for LC-MS/MS processing. 507 Tryptic peptide mixtures were loaded onto a 25 cm long, 75 μm inner diameter fused-silica 508 capillary, packed in-house with bulk 1.9 μM ReproSil-Pur beads with 120 Å pores as described 509 previously 77 . Peptides were analyzed using a 140 min water-acetonitrile gradient delivered by a 510 Dionex Ultimate 3000 UHPLC (Thermo Fisher Scientific) operated initially at 400 nL/min flow 511 rate with 1% buffer B (acetonitrile solution with 3% DMSO and 0.1% formic acid) and 99% buffer 512 A (water solution with 3% DMSO and 0.1% formic acid). Buffer B was increased to 6% over 5 513 min at which time the flow rate was reduced to 200 nl/min. A linear gradient from 6-28% B was 514 applied to the column over the course of 123 min. The linear gradient of buffer B was then further 515 increased to 28-35% for 8 min followed by a rapid ramp-up to 85% for column washing. Eluted 516 peptides were ionized via a Nimbus electrospray ionization source (Phoenix S&T) by application 517 of a distal voltage of 2.2 kV. 518 All label-free mass spectrometry data were collected using data dependent acquisition on Orbitrap 519 Fusion Lumos Tribrid mass spectrometer (Thermo Fisher Scientific) with an MS1 resolution of 520 120,000 followed by sequential MS2 scans at a resolution of 15,000. Data generated by LC-521 MS/MS were searched using the Andromeda search engine integrated into the MaxQuant 2 522 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  Chimera I contains ONNV genome from nsP1 to capsid and CHIKV genome from E3 to E1. 812 Chimera II contains ONNV genome from nsP1 to the region prior to the subgenomic promoter in 813 nsP4 and CHIKV genome from the subgenomic promoter to E1. Chimera III contains CHIKV 814 genome from nsP1 to capsid and ONNV genome from E3 to E1. Chimera IV contains CHIKV 815 genome from nsP1 to the region prior to the subgenomic promoter in nsP4 and ONNV genome 816 from the subgenomic promoter to E1. transcript levels and secreted infectious particle titers through RT-qPCR and plaque assay, 827 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  representation of Chimera III-I, III-II, and III-III. Chimera III-I contains CHIKV  832 genome from nsP1 to E3 and ONNV genome from E2 to E1. Chimera III-II contains CHIKV 833 genome from nsP1 to E2 and ONNV genome from 6K to E1. Chimera III-III contains CHIKV 834 genome from nsP1 to 6K and ONNV E1. e, Titration of supernatant samples from THP-1 derived 835 macrophages infected with CHIKV 181/clone 25, ONNV SG650, or chimeras (III-I, III-II, III-III)  836 for 14, 24, and 48 h. The infection conditions and virus titer assessments were performed as 837 previously described in Figure. 2b. Data are representative of 2 independent experiments. Mean 838 values of biological duplicates measured in technical duplicates were plotted with SD. Asterisks 839 indicate statistically significant differences as compared to ONNV (Two-way ANOVA and 840 Dunnett's multiple comparisons test: **, p£0.01; ****, p£0.0001). 841 I, E2+E1-II). Virion production was determined through RT-qPCR and plaque assays as described 870 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 29, 2023. ; in Figure.3c. Data are representative of 4 independent experiments. Mean values of biological 871 duplicates measured in technical duplicates were plotted with SD. Asterisks indicate statistically 872 significant differences as compared to ONNV (One-way ANOVA and Dunnett's multiple 873 comparisons test: ***, p£0.001; ****, p£0.0001). 874 The histogram in each KEGG term is defined by the number of genes with a specific log2 fold 934 change value. 935 compared to the titers from cells treated with NT siRNA to assess the anti-or pro-viral effects of 946 specific host genes on CHIKV production. G3BP1 and G3BP2 (G3BP1+2) known to be proviral 947 for CHIKV replication were knocked down together as control. Data are representative of 2 948 independent experiments. The mean values of biological duplicates measured in technical 949 duplicates were plotted with SD (One-way ANOVA and Dunnett's multiple comparisons test: *, 950 p£0.05; **, p£0.01; ***, p£0.001, ****, p£0.0001). 951 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  and secreted infectious particles as previously described (One-way ANOVA and Dunnett's 973 multiple comparisons test: *, p£0.05; **, p£0.01; ***, p£0.001; ****, p£0.0001). 974

Extended data 3. 975
Enriched protein clusters from the host factors co-immunoprecipitated with CHIKV glycoproteins. 976 The