Dengue virus 3’ untranslated region regulates RNA genome translation

Dengue virus 3’ untranslated region regulates RNA genome translation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Dengue virus 3’ untranslated region regulates RNA genome translation Eng Eong Ooi, Kiven Kumar, Justin Ooi, Hwee Cheng Tan, Xudong Lyu, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6513699/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Viral antigenic burden drives the inflammation-driven pathophysiology of dengue in humans. Nonetheless, the control of dengue virus (DENV) antigen expression for pathogenicity in humans remain uncertain. Herein, we examined a clinical DENV-3 isolate (Sleman/78), along with its partially and fully attenuated derivatives through 30- (Δ30) as well as 30- and 31-nucleotide deletions (Δ30/31), respectively, in the 3’ untranslated region (3′UTR) of the RNA genome; the partially and fully attenuated phenotypes of these derivatives were demonstrated in clinical trials. We found, using infectious clone technology, protein and RNA pulldown approaches, that the wild-type 3’UTR bound host translation proteins, including non-canonical eukaryotic initiation factor-3D (EIF3D) to support viral protein expression. Both Δ30/31 mutation and EIF3D silencing attenuated viral protein expression and hence replication of Sleman/78. As DENV genome is cyclized through 5’ and 3’UTR interactions, our findings the role of 3′UTR in regulating translation for infection and pathogenesis in humans. Biological sciences/Microbiology/Virology/Dengue virus Biological sciences/Microbiology/Vaccines/Live attenuated vaccines Dengue virus (DENV) attenuation 3’ untranslated region (3’UTR) translation eukaryotic initiation factors Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Dengue is an acute febrile illness caused by infection with any one of four different dengue viruses (DENV-1 to -4), all of which are transmitted by Aedes mosquitoes. Increased volume of trans-continental travel, urbanisation and climate change have expanded the geographic footprints of both DENVs and Aedes mosquitoes from the tropics to North America and Mediterranean Europe; more than half of the world’s population thus live at risk of dengue each year 1 . Clinical observations have indicated that those who develop high viraemia levels are at greater risk of severe dengue 2 , 3 ; high infection burden exacerbates pro-inflammatory responses that lead to life-threatening shock from vascular leakage, internal haemorrhage and organ dysfunction 4 . Despite the link between viraemia and severe dengue, the genetic elements that control of DENV replication in dengue patients remains poorly understood. An impediment to understanding how DENV genetics affect infection outcome is the lack of infection models that accurately recapitulate the clinical features of dengue. Live attenuated (non-chimeric) dengue vaccine strains thus provide unique opportunities to overcome such limitations. These vaccine strains were all derived from DENVs isolated from dengue patients and attenuated through empirical approaches but without knowledge on the mechanisms of attenuation 5 . The live attenuated vaccine strains that have shown low levels of viraemic infection and mild systemic symptoms in thousands of clinical trial volunteers thus provide a clinically validated context to dissect the viral genetic determinants critical causing dengue. TV003 is a tetravalent live attenuated dengue vaccine candidate that has completed phase III clinical trial 6 , 7 . TV003 contains 3 live attenuated DENVs; the fourth component of TV003 is a DENV-4/-2 chimeric virus. The attenuated DENV-1, -3 and − 4 strains of TV003 were derived through deletions in the 3′UTR. Both DENV-1 and − 4 contained a 30 nucleotide deletion (Δ30) 8 , 9 in the same position of the 3’UTR. However, the same deletion in DENV-3 produced an inadequately attenuated mutant 10 ; this strain is now used for controlled human infection studies 11 . An additional 31 nucleotide proximal to the Δ30 mutation was then deleted to produce a sufficiently attenuated mutant (DENV-3 Δ30/31) 12 . The parent wild-type DENV-3 as well as its semi- (Δ30) and fully-attenuated (Δ30/31) derivatives thus provide a unique opportunity to glean insights into the virus-host interactions necessary for pathogenesis. Host-DENV interactions critical for symptomatic infection should thus show progressive loss of function with increasing number of deletions in the 3’UTR. We found, unexpectedly, that the 3’UTR plays a critical role in initiating viral genome translation. Results Wild-type DENV-3 (Sleman/78), Δ30 and Δ30/31 mutants show different growth characteristics DENV 3’UTR contains nucleotide sequence that complement each other to form highly conserved secondary RNA structures 13 . We thus first determined the location of Δ30 and Δ31 relative to the secondary RNA structures previously defined using RNA phylogeny 14 . Figure 1 A shows the RNA structures of wild-type DENV-1 (Western Pacific), DENV-3 (Sleman/78), and DENV-4 (Dominica 814669) along with their respective attenuated derivatives. We have also included DENV-2 (Tonga/74) and the Δ30 mutant that only partially attenuated Tonga/74 virus; this mutant, like DENV-3 Δ30, is now used for controlled human infection studies 15 . To examine the effects these 3’UTR deletions have on DENV-3 fitness, we constructed infectious clones of the wild-type Sleman/78, Δ30, and Δ30/31 viruses using Gibson assembly ( Supplementary Fig. 1 ). The rescued viruses were amplified in C6/36 cell line and quantified by plaque assay in BHK-21 cells. Plaque diameter of Δ30/31 mutant was smaller compared to those of either wild-type DENV-3 or the Δ30 mutant (Figs. 1 B and C ). To determine if the differences in plaque diameter could be accounted for by virus replication dynamics, Sleman/78 and its Δ30, and Δ30/31 derivatives were inoculated onto MoDCs at a multiplicity of infection (MOI) of 5. Both viral genomic and subgenomic flavivirus RNA (gRNA and sfRNA, respectively), were measured; sfRNA was analysed as we have previously shown that changes in 3’UTR sequence can alter its abundance 16 . Levels of infection of the different viruses were comparable, as indicated by the lack of difference in both gRNA and sfRNA levels at both 2- and 4-hours post-infection (hpi) (Figs. 2 A and B ). At 6 hpi, however, levels of gRNA and sfRNA of both Sleman/78 and ∆30 virus continued on an upward trajectory whereas those for ∆30/31 virus were reduced (Figs. 2 A and B ). The ratio of gRNA:sfRNA also showed the same trend as gRNA and sfRNA levels ( Supplementary Fig. 2 ), suggesting that the nucleotide deletions did not affect the formation of sfRNA. The expression of IFN-β showed no difference relative to baseline levels for Sleman/78 and ∆30 virus infection. In striking contrast, ∆30/31 virus infection induced IFN-β expression that peaked at 6hpi, at a time when both gRNA and sfRNA levels of ∆30/31 virus were reduced (Fig. 2 C). Viral protein expression in Huh-7 cells also showed similar differences in kinetics as viral RNA replication. Unlike Sleman/78 and ∆30 viruses where viral NS2B, NS3 and NS5 proteins could be detected as early at 12 hpi, those of ∆30/31 mutant could only be detected at 24 hpi and later (Figs. 2 D-G). Taken collectively, these findings suggest that Δ30/31 delayed viral protein expression, RNA replication and compromised innate immune evasion properties. DBI and DBII binds ribosomal and translation-related proteins To understand how deletions in DBI and DBII of Sleman/78 affected DENV genome translation, virus replication and innate immune evasion, we first sought to define the functions of DBI and DBII. We reasoned that these structures engage in specific viral RNA-host protein interactions. Using stable isotope labelling by amino acids in Huh-7 cell culture coupled with quantitative mass spectrometry (SILAC-qMS) ( Supplementary Fig. 3 ), we identified proteins that preferentially bound ( p < 0.05) Sleman/78 3′UTR compared to the ∆30 and ∆30/31 mutants and further filtered for only cytoplasmic proteins based on the subcellular localization information from the UniProt database Unsupervised clustering analysis of the MS data revealed 48 proteins that were differentially enriched. A total of 34 proteins were bound to Sleman/78 more than both Δ30 and Δ30/31 3′UTR (Fig. 3 A). Most of these proteins were ribosomal proteins, eukaryotic initiation factors (EIFs), and tRNA-associated proteins, which all function to support protein translation. In contrast, only 14 proteins with mixed functions were enriched with ∆30 and ∆30/31 compared to Sleman/78 3’UTR (Fig. 3 A). To show the differential enrichment of proteins that bound Sleman/78 compared to ∆30 and ∆30/31 3’UTRs, we constructed a scatter-plot of the statistically significant hits from our SILAC-qMS data. We found 38 proteins that showed 2-fold or greater enrichment with Sleman/78 3’UTR compared to both ∆30 and ∆30/31 3’UTR (Fig. 3 B). Five proteins were enriched when Sleman/78 was compared to ∆30 UTR only, whereas 9 proteins were enriched with Sleman/78 more than ∆30/31 3’UTR only (Fig. 3 B). Similar to data shown in Fig. 3 A, most of these proteins function to support protein translation. Collectively, the ∆30/31 3’UTR resulted in a greater loss of binding of translation-related proteins to the DENV-3 positive-strand RNA. To validate our SILAC-qMS findings that specific proteins involved in translation bound Sleman/78 3’UTR more than the ∆30/31 mutant, we turned to RNA immunoprecipitation. We conducted the immunoprecipitation study using monoclonal antibodies (mAbs) to the eukaryotic initiation factors 3D and 3K (EIF3D and EIF3K), as well as ribosomal protein RL17 and RL28, given the reduced binding of these proteins to the various mutant as compared to Sleman/78 (Fig. 3 B). Cells transfected with the different 3’UTRs with comparable efficiencies (Supplementary Figs. 4A-D ) were subjected to immunoprecipitation using protein-specific mAbs. The 3’UTR RNA was then quantified using RT-qPCR. Consistent with the SILAC-qMS findings (Fig. 3 B) the amount of 3’UTR RNA in the EIF3D immunoprecipitate (Fig. 4 A), relative to the amount transfected 3’UTR were significantly higher with Sleman/78 and ∆30 compared to ∆30/31 3’UTR (Fig. 4 B). Likewise, higher amounts of Sleman/78 and ∆30 3’UTR compared to ∆30/31 3’UTR were found in the EIF3K (Figs. 4 C-D) and RL28 immunoprecipitates (Figs. 4 G-H). Immunoprecipitation of RL17 also showed consistent findings as those of SILAC-qMS, where both the ∆30 and ∆30/31 3’UTR were markedly lower than Sleman/78 (Figs. 4 E-F). These findings collectively support the notion that DBI and DBII in the 3’UTR bind proteins involved in viral RNA translation. DENV-3 DBI and DBII support genome translation Binding of proteins involved in translation suggest that Sleman/78 3’UTR may, at least in part, be involved in supporting viral protein expression. In particular, EIF3D has previously been shown to initiate EIF4E independent translation 17 , and this non-canonical pathway could be employed by DENV to translate its genome. Such an explanation is plausible as the DENV genome is cyclized through complementary sequence interactions between the 5’ and 3’UTRs 18 , 19 , 20 ; the 5′ upstream AUG region (5′UAR) in the 5′UTR base-pairs with its complementary 3′UAR counterpart at the base of the 3′ stem-loop (3′SL) in the 3′UTR to form the panhandle of the cyclized positive-strand RNA genome. DBI and DBII are thus in proximity to the translation start site on the DENV capsid gene ( Supplementary Figs. 5A and B ). To test the notion binding of EIF3D with 3’UTR is critical for DENV genome translation, we infected EIF3D-silenced Huh-7 cells (Fig. 5 A) with Sleman/78, ∆30, and ∆30/31. If EIF3D recruitment by wild-type 3’UTR is necessary for initiating translation, then silencing EIF3D would delay Sleman/78 protein expression. Compared to control cells transfected with scrambled siRNA, the levels of DENV NS2B, NS3 and NS5 proteins (Fig. 5 B), as well as viral RNA levels (Fig. 5 C) were universally lower in EIF3D silenced cells at 6 hpi. Similar differences in viral protein expression (Fig. 5 D) and abundance of viral RNA (Fig. 5 E) were also observed at 12 hpi. At 24 hpi, the levels of NS2B but not NS3 and NS5 appeared higher in EIF3D siRNA treated cells compared to controls, possibly due to depletion of EIF3D siRNA and hence initiation of translation (Fig. 5 F). However, viral RNA levels remain significantly lower in EIF3D siRNA treated cells than controls (Fig. 5 G). Reduced translation of DENV-1 and − 4 but not DENV-2 ∆30 mutants The findings with Sleman/78 and its two mutants also raised the possibility that a single 30 nucleotide stretch of deletion was sufficient to attenuate translation of DENV-1 and DENV-4, but not DENV-2 (Fig. 1 A), as evidenced by clinical trial findings 21 . To test this possibility, we used an eGFP reporter system to measure translation. We first determined if such a eGFP reporter system could recapitulate the differences in Sleman/78, ∆30, and ∆30/31 genome translation ( Supplementary Figs. 6A and 6B) ; plasmid traces were minimal in the RNA extract ( Supplementary Fig. 6C ). Consistent with our findings with viral protein expression, both the proportion of eGFP positive cells (Figs. 6 A and B ) and the mean fluorescence intensity (MFI) ( Fig. 6 C) were lower when the eGFP gene was flanked with 5’ and 3’UTR of the ∆30/31 mutant compared to either Sleman/78 or ∆30 mutant. We then constructed plasmids with eGFP reporter gene flanked with 5’UTR of DENV-1 (Western Pacific), DENV-4 (Dominican 814669) and DENV-2 (Tonga/74) on one side, and their respective 3’UTR, as well as their ∆30 and ∆30/31 mutants. Transfection of these plasmids into HEK293 cells produced comparable levels of eGFP transcripts ( Supplementary Figs. 6D and E) . As expected, unlike Sleman/78, the ∆30 mutants of both DENV-1 and − 4 showed significantly lower percentage of eGFP positive cells as well as fluorescence intensity compared to eGFP gene linked with wild-type DENV-1 and − 4 3’UTRs (Figs. 6 D-I). In contrast, no difference was found in either the percentage of eGFP positive cells or fluorescence intensity when the eGFP gene was constructed using DENV-2 (Tonga/74) ∆30 compared to wild type 3’UTR. This finding is consistent with the clinical phenotype of DENV-2 Tonga/74 ∆30 mutant (Figs. 6 J-L), which was insufficiently attenuated to be used as a vaccine strain 22 , 23 . An additional ∆31 mutation in the Tonga/74 3’UTR significantly reduced both the percentage of eGFP positive cells as well as fluorescence intensity (Figs. 6 J-L). Discussion The large body of safety and immunogenicity data on the live attenuated tetravalent TV003 dengue vaccine candidate provides a unique and clinically validated context to examine the function of DBI and DBII in the 3’UTR of DENVs. Our findings revealed that DBI and DBII bind proteins involved in viral genome translation. The 3’UTR of DENVs contain regions of nucleotide sequences that are highly conserved. Where there is variability in the nucleotide sequence, the secondary RNA structures, including DBI and DBII remain conserved through compensatory mutations 14 . Various studies have shown that these secondary and more complex RNA structures in the 3’UTR support DENV infection 24 , 25 , 26 , 27 , 28 . Deletions that either disrupted or reshaped the conserved secondary structure elements in the 3’UTR reduced DENV replication 26 . Furthermore, a study on the related West Nile virus (WNV) showed that viral genome translation was regulated through specific interactions between the 5’ and 3’UTRs to recruit and bind the 40S ribosome subunit 29 . However, this study only identified a single ribosomal 40S protein being regulated in that manner but did not adopt an unbiased approach to identify the interacting partners of WNV 3’UTR. Our findings thus add granularity to this fledgling body of knowledge and suggest that DBI and DBII function as regulators of viral genome translation. Amongst the hits on our SILAC-qMS experiment, the finding of EIF3D as a binding partner of Sleman/78 but not the ∆30/31 mutant is intriguing. EIF3D is a subunit of the EIF3 multi-protein complex 30 . It binds the mRNA cap to initiate translation when the canonical EIF4E is inactive 17 , notably during endoplasmic reticulum (ER) stress to regulate translation of specific sets of mRNA, such as those involved in cell proliferation 31 . DENV infection is known to induce ER stress 32 . The requirement for EIF3 complex for translation was also found for the closely related Zika virus (ZIKV) 33 . EIF3D-dependent translation of DENV genome could thus be strategic for the virus to ensure sustained replication even in stressed cells 34 . Besides initiating translation, EIF3D could have other non-canonical functions, including increasing RNA stability 30 , 35 . With the loss of binding of EIF3D and possibly other ribosomal proteins to the 3’UTR, the viral genome could be rendered more susceptible to RNA processing and metabolic degradation 36 . Processing of viral RNA could, in turn lead to increased activation of cytoplasmic viral RNA sensors that activate the innate immune response, especially the type-I IFN response that limit virus dissemination to attenuate systemic infection. Indeed, our viral RNA replication kinetics findings in MoDCs support this notion. The ∆30/31 mutation could thus be a “double jeopardy” on Sleman/78, reducing both viral protein expression and the cytoplasmic viral RNA half-life. Considering that DENV non-structural proteins also contribute to innate immune evasion 37 , reduced translation of ∆30/31genome could favour prompt type I IFN response that restrict systemic infection dissemination, further contributing to its attenuated phenotype. That the two DB RNA structures of Sleman/78 were found to interact with ribosomal proteins and EIF3 subunits also provide evolutionary perspectives on DENVs. Besides supporting replication of its genome, cyclization could also be a strategy for DENVs, and possibly other orthoflaviviruses, to preserve a short 5’ UTR. The 5’UTR is already highly structured to support interactions with both host and viral proteins 13 necessary for viral replication. Incorporating additional RNA elements to bind ribosomal proteins and EIF3 complex subunits would lengthen the 5’UTR and compromise cap-dependent translation 38 . Evolving RNA functional elements that recruit translation factors in the 3’ instead of 5’UTR could thus confer strategic advantage to efficient translation of cyclized DENV RNA genome. Our findings also refine the understanding of how DENV genome replicates. It has been thought that the cyclized genome is important for viral RNA replication whereas translation of the viral polyprotein occurs on the linear RNA conformation 39 . The rationale for different RNA conformation to explain replication and translation is that the polymerase and the ribosome travel in opposite direction on the positive-sense viral RNA genome. An in vitro reconstitution approach of ZIKV genome supported such an explanation although the study used a mini genome with a 3’UTR that lacked DBI and DBII 33 . As RNA conformation is highly dynamic, it is possible that once the translation machinery is assembled, in part on the 3’UTR, the cyclized RNA genome linearizes. An alternative explanation is that, once the first few negative-sense RNA strands are transcribed, the polymerase is more active on the cyclized negative-sense RNA intermediate to produce new positive-sense DENV genome, whereas translation remains exclusively on the positive-sense template. Indeed, this alternative explanation is supported by the large difference in the quantity of positive-sense viral RNA compared to the negative-sense intermediate in infected cells 40 . Further research will be needed to tease apart which of these explanations is correct. Besides defining the function of DBI and DBII of DENV-3, our findings show that DBI and DBII of the other three DENVs also function to support viral genome translation, albeit with important type-specific nuances. That ∆30 in DBII reduced translation of eGFP reporter gene when flanked with DENV-1 and DENV-4 UTRs but not those of DENV-2 are entirely consistent with the history of TV003 development 21 . DENV-2 Tonga/74∆30 strain was developed as a vaccine candidate. However, pre-clinical data showed that Tonga/74∆30 was insufficiently attenuated for further clinical development as a component of TV003 22, 23 ; the DENV-2 component of TV003 is thus composed of a chimeric DENV2/4∆30(ME) construct, with the prM and envelope (E) genes from DENV-2 New Guinea C (NGC) strain spliced into the DENV-4 ∆30 mutant backbone 22 , 41 . Whether a DENV-2 Tonga/74 ∆30/31 mutant could prove sufficiently attenuated whilst retaining immunogenicity, however, is uncertain as such a study has not been reported. Nonetheless, given the highly conserved DBI and DBII structures, it would be reasonable to expect that the interactome of these secondary RNA structures would be consistent across the different lineages and sub-lineages of each of the four DENVs. It may thus be possible to update the genomic backbones of components of TV003, which were developed using DENVs isolated in the 1970s and 1980s, with more contemporaneous strains to reduce mismatch with circulating DENVs. This study has limitations. Whilst we have shown that several proteins involved with viral RNA translation bound 3’UTR, we have not shown how they interacted with other elements of the translation machinery, including those that bind to the 5’UTR as well as the ribosomes to regulate viral protein expression. In conclusion, DBI and DBII of the 3' UTR bind proteins involved in translation, including EIF3D, to express viral proteins and promote replication. Deletions of nucleotides within these DB structures reduce viral genome translation and render viral RNA more vulnerable to processing and sensing to attenuate infection. Methods Cells HuH-7, C6/36, Vero and BHK-21 cells were purchased from the American Type Culture Collection (ATCC) and cultured according to ATCC recommendations. Primary monocytes were isolated from a healthy seronegative volunteer with informed consent (NUS-IRB-2022-501), and cultured as described 42 . Generation of infectious clone Wild-type DENV-3 Sleman/78 (GenBank: AY656169.1), ∆30 (GenBank: AY656170.1), and ∆30/3 10 were rescued using infectious clone technology. The complete genomes of Sleman/78, ∆30, and ∆30/31 were divided into seven overlapping fragments ( Supplementary Fig. 1 ). Each fragment, approximately 1800–2000 bases in length, was chemically synthesized (Company’s name). Briefly, NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs) was used to amplify each fragment with primers designed for the respective regions ( Supplementary Table 1 ). The amplified fragments were then purified using the MinElute Gel Extraction Kit (Qiagen) according to the manufacturer’s instructions. Additionally, the pUC19 vector, which contains the human cytomegalovirus (CMV) promoter and simian virus 40 polyadenylation signal (SV40-pA), was amplified using primers listed in Supplementary Table 1 . The purified genome fragments and the pUC19 vector were assembled in equimolar amounts (0.1 pmol of each fragment) using the NEBuilder HiFi DNA Assembly Kit (New England BioLabs) at 50°C for 1 hour. Five microliters of the assembled infectious clones were transfected into HEK293T cells in a 24-well plate using 3 µL of Lipofectamine 2000 (Thermo Fisher Scientific), following the manufacturer’s protocol. The supernatant containing infectious clone-derived viruses was collected 72 hours post-transfection and propagated in C6/36 cells supplemented with 3% FBS. Infected cells were observed daily for syncytia formation for 6 to 7 days, after which they were harvested and stored at − 80°C until further use. Plaque assay Plaque assay was performed on BHK-21 cells as previously described 42 . Briefly, serial dilutions (10-fold) of virus were added to BHK-21 cells in 24-well plates and incubated for 1 h at 37°C. Media was aspirated and replaced with 0.9% methyl-cellulose in maintenance media. Six days later, cells were fixed with 20% formalin and stained with1% crystal violet. The number of plaques was counted visually. Viral replication kinetics on primary monocyte derived dendritic cells (MoDCs) Peripheral blood mononuclear cells (PBMCs) were isolated from a healthy donor who had tested negative for anti-DENV antibodies using plaque reduction neutralization test 42 , 43 . CD14 + monocytes were isolated from PBMCs using CD14 microbeads (Miltenyi Biotec) according to manufacturer’s protocol. Differentiation of CD14 cells into MoDCs were done in six-well plates using RPMI 1640 supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 100 ng/ml IL-4 (eBioScience) and 50 ng/ml granulocyte macrophage-colony stimulating factor (GM-CSF, eBioScience) for 6 days with media change on the third day. MoDCs were seeded at 2 x 10 4 cells per well in 96-well tissue culture plate and infected with 5 MOI of Sleman/78, ∆30 and ∆30/31 viruses. At pre-defined timepoints post-infection, cells were washed once in PBS before RNA extraction using RNeasy Kit (Qiagen) according to manufacturer’s protocols. Total RNA was converted into cDNA using LunaScript RT SuperMix Kit (Company’s name), according to the manufacture’s protocol. Viral replication was measured using two sets of the primers ( Supplementary Table 2 ), one which targets gRNA and another which measure sfRNA were used in the qPCR. Viral RNA was semi-quantified by dividing it against RT-qPCR results on the housekeeping GAPDH gene. Protein expression Western blot was performed to assess DENV protein expression in infected cells. Briefly, Huh-7 cells were infected with Sleman/78, ∆30, or ∆30/31 virus at a multiplicity of infection (MOI) of 2. At pre-defined timepoints post-infection, cells were washed once with PBS and lysed in RIPA buffer containing protease inhibitors (1:100 dilution), followed by incubation on ice for 2 hours. The crude lysates were centrifuged at 20,000 x g, and the supernatant was collected and mixed with 4X Laemmli buffer. The samples were then heated at 95°C for 10 minutes before separation by SDS-PAGE. Proteins were transferred onto a PVDF membrane and incubated with the following primary antibodies (1:5000 dilution): mouse anti-DENV NS5 IgG monoclonal antibody (mAb) (GeneTex, GTX629447), mouse anti-DENV NS3 IgG mAb (GeneTex, GTX629477), and mouse anti-DENV NS2B mAb (GeneTex, GTX638874). The membrane was then probed with HRP-conjugated anti-mouse or anti-rabbit IgG secondary antibodies (1:5000 dilution). Detection was performed using Amersham ECL reagents (Cytiva). Blotting for β-actin with an anti-β-actin IgG mAb (1:5000; Cell Signaling Technology, 8H10D10) served as a loading control. Construct of 3’UTR and in-vitro transcription Plasmids containing the 3’UTR of Sleman/78, ∆30, ∆30/31, or NS2A were chemically synthesized with a tobramycin tag and a T7 promoter. A size-matched control, NS2A from Sleman/78, was used as a control in this experiment 44 . To generate a 560 bp fragment containing the T7 promoter, 3’UTR, and tobramycin tag, the plasmid was mixed with primers ( Supplementary Table 3 ) and NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs) for amplification. The resulting fragments were gel-purified using the MinElute Gel Extraction Kit (Qiagen) and subsequently used for in vitro transcription. The 3’UTR fragments were transcribed in vitro using a reaction mixture containing 10X TB buffer (300 mM Tris-HCl, pH 7.7, 20 mM spermidine, 30 mM MgCl₂, 50 mM NaCl, 100 mM DTT, 20% PEG 8600, and 0.05% Triton X-1000), 6 mM NTP mix, 0.4 M urea, T7 polymerase, and 10 mM DTT. The reaction was incubated at 37°C for 2 hours. The transcribed RNA was purified using the RNeasy Kit (Qiagen) according to the manufacturer’s protocol. RNA concentration was measured using a NanoDrop spectrophotometer. Adaptation of Huh-7 cells to SILAC medium Stable isotope labelling by amino acids in cell culture (SILAC) was performed to obtain labeled Huh-7 cells with light (K 0 R 0 ), medium (K 4 R 6 ), and heavy (K 8 R 10 ) isotopes ( Supplementary Fig. 3 ). The labelled Huh-7 cells were washed with PBS and then trypsinized. The detached cells were transferred to a 50 mL conical tube and centrifuged at 500 x g for 5 minutes. The resulting cell pellet was resuspended in 10–20 mL of PBS. Cell lysis and protein concentration measurement were carried out as previously described 45 . RNA-protein interactions Tobramycin RNA affinity chromatography, mass spectrometry, and data analysis were performed as previously described 16 , 44 , 45 . In brief, RNA was immobilized on a tobramycin matrix and incubated with SILAC-labelled Huh-7 cell lysates. After 2h incubation at 4°C, the matrix was washed, and bound proteins were eluted using excess tobramycin. The eluates were analysed by quantitative mass spectrometry (qMS), enabling a direct comparison of proteins bound to the Sleman/78 3’UTR, ∆30 3’UTR, ∆30/31 3’UTR, or control RNA, as illustrated in Supplementary Fig. 3 . Analysis of bound proteins The protein concentration after the elution from the column was determine by using BCA kit according manufacture protocol. Approximately 10 µg of elution proteins were mixed according the. The mixture of the proteins were analysed using quantitative mass spectrometry (qMS), allowing a direct comparison of proteins bound to the Sleman/78 3’UTR, ∆30 3’UTR, ∆30/31 3’UTR and NS2a (control RNA). Protein abundance was log 2 . transformed and normalised. Differentially bound proteins were determined using a two-tailed t-test. The data was further filtered for only cytoplasmic proteins by excluding those localized to the nuclear bodies, nucleoplasm, nuclear speckles, nucleoli, nucleolar fibrillar centre, mitotic chromosome, and nucleus based on the list of proteins and their subcellular localization obtained from the UniProt database. Hierarchical clustering was conducted using the ComplexHeatmap R package 46 . RNA immunoprecipitation (IP) Plasmids encoding the 3' UTR from Sleman/78, ∆30, and ∆30/31 was transfected into Huh-7 cells. At 4 hours post-infection (hpi), RNA IP was performed using the MAGNA-RIP kit (Merck Millipore, USA), following the manufacturer's instructions. The levels of RNA in immunoprecipitates (IP) obtained using either anti-RPL11 (Santa Cruz Biotechnology, sc-293224), anti-RL17 (Santa Cruz Biotechnology, sc-515904), anti-RL28 (MyBiosource, MBS4508582), anti-EIF3D (Santa Cruz Biotechnology, sc-271515), or anti-EIF3K (Santa Cruz Biotechnology, sc-81262) antibodies were quantified by RT-qPCR and normalized to GAPDH mRNA levels. Immunoprecipitated proteins and RNA were analysed by Western blotting and RT-qPCR, respectively. Fold enrichment (FE) was calculated using the formula: FE = 2 (-ΔΔCt) , where -ΔΔCt = -[(Ct_ RNA,IP - Ct_ GAPDH,IP ) - (Ct_ RNA,control IP - Ct_ GAPDH,control IP )]. EIF3D silencing through siRNA Huh-7 cells cultured in 24-well plates were transfected with either control small-interfering RNA (siCtrl) or Silencer Pre-designed siRNA targeting EIF3D (siRNA EIF3D) (ID:13732; Thermo Fisher Scientific) using the Lipofectamine RNAiMax reagent (Invitrogen), following the manufacturer’s protocol. Forty-eight hours post-transfection, the cells were infected with Sleman/78, ∆30, and ∆30/31 at a MOI of 5. The cells were harvested at 2, 4, 6, 12 and 24 h post-infection. Transfection efficiency was determined by Western blot analysis using anti-EIF3D (1:5000; Santa Cruz Biotechnology, sc-271515) with anti-GAPDH (1:5000; Cell Signaling Technology, 14C10) as a housekeeping protein for detection. eGFP reporter constructs The 5′ and 3′ UTRs of vaccine strains from TV003 were used, with the exception that DENV-2 was replaced by Tonga/74 (GenBank: AY744148) and Tonga/74∆30 (GenBank: AY744149). The constructs were designed as shown in Supplementary Fig. 6A and chemically synthesized in plasmids. Fragments from the synthesized plasmids were amplified using primers listed in Supplementary Table 4 with NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs). PCR products were gel-purified using the QIAquick Gel Extraction Kit (Qiagen) according to the manufacturer’s instructions. Each gel-purified fragment (0.1 pmol) was assembled with the pUC19 vector containing the CMV promoter and SV40-pA using the NEBuilder HiFi Assembly Kit (New England BioLabs) at 50°C for 1 hour. The assembled plasmids were transformed into DH5α E. coli (New England BioLabs), and the presence of the insert was confirmed by Sanger sequencing. Approximately 50 ng of plasmid was transfected into HEK293T cells in a 24-well plate using Lipofectamine 2000 (Invitrogen) and Opti-MEM (Gibco). After 24 hours, HEK293T cells were harvested, washed with 500 µL of cold PBS, and fixed in 2% paraformaldehyde (PFA) on ice for 10 minutes. Following fixation, cells were washed with PBS, resuspended in FACS buffer, and sorted using a FACSAria cell sorter (BD Biosciences). Statistical methods The graphs were plotted using GraphPad Prism software. As indicated, one-way ANOVA was used to determine significant differences ( p < 0.05). Error bars in the graphs represent the standard deviation, with * indicating p < 0.05, ** indicating p < 0.01, *** indicating p < 0.001, and # indicating p < 0.0001. Declarations Competing Interests E.E.O has served as chair and member of scientific advisory boards on dengue for Takeda, as well as in various advisory capacities for Sanofi Pasteur, Merck, Johnson & Johnson, and Novartis on dengue vaccines and therapeutics. Author Contributions E.E.O., K.K. and E.F. conceptualized the study. K.K., H.C.T., X.Y. and W.C.N. conducted the experiments. K.K. and E.F. designed and carried out the SILAC-qMS. K. K., J.SG.O., W.C.N., K.R.C., E.F. and E.E.O analysed the data. Acknowledgements We thank Tanamas Siriphanitchakorn and Clara Koh for helpful discussions and technical support. This study was supported by the Singapore Translational Research Award from the National Medical Research Council to E.E.O (MOH-001271-00), and the National Research Foundation (NRF-CRP27-2021RS-0001). References Messina JP et al (2019) The current and future global distribution and population at risk of dengue. Nat Microbiol 4:1508–1515 Vuong NL et al (2021) Higher Plasma Viremia in the Febrile Phase Is Associated With Adverse Dengue Outcomes Irrespective of Infecting Serotype or Host Immune Status: An Analysis of 5642 Vietnamese Cases. Clin Infect Dis 72:e1074–e1083 Vuong NL et al (2024) Dengue viremia kinetics and effects on platelet count and clinical outcomes: An analysis of 2340 patients from Vietnam. Elife 13 Wilder-Smith A, Ooi EE, Horstick O, Wills B (2019) Dengue Lancet 393:350–363 Ooi EE, Kalimuddin S (2023) Insights into dengue immunity from vaccine trials. Sci Transl Med 15:eadh3067 Kallás EG et al (2024) Live, Attenuated, Tetravalent Butantan-Dengue Vaccine in Children and Adults. N Engl J Med 390:397–408 Nogueira ML et al (2024) Efficacy and safety of Butantan-DV in participants aged 2–59 years through an extended follow-up: results from a double-blind, randomised, placebo-controlled, phase 3, multicentre trial in Brazil. Lancet Infect Dis Men R, Bray M, Clark D, Chanock RM, Lai CJ (1996) Dengue type 4 virus mutants containing deletions in the 3' noncoding region of the RNA genome: analysis of growth restriction in cell culture and altered viremia pattern and immunogenicity in rhesus monkeys. J Virol 70:3930–3937 Whitehead SS, Falgout B, Hanley KA, Blaney JE Jr., Markoff L, Murphy BR (2003) A live, attenuated dengue virus type 1 vaccine candidate with a 30-nucleotide deletion in the 3' untranslated region is highly attenuated and immunogenic in monkeys. J Virol 77:1653–1657 Blaney JE Jr., Hanson CT, Firestone CY, Hanley KA, Murphy BR, Whitehead SS (2004) Genetically modified, live attenuated dengue virus type 3 vaccine candidates. Am J Trop Med Hyg 71:811–821 Pierce KK et al (2024) Evaluation of a new dengue 3 controlled human infection model for use in the evaluation of candidate dengue vaccines. medRxiv Blaney JE Jr. et al (2008) Dengue virus type 3 vaccine candidates generated by introduction of deletions in the 3' untranslated region (3'-UTR) or by exchange of the DENV-3 3'-UTR with that of DENV-4. Vaccine 26:817–828 Ng WC, Soto-Acosta R, Bradrick SS, Garcia-Blanco MA, Ooi EE (2017) The 5' and 3' Untranslated Regions of the Flaviviral Genome. Viruses 9 Finol E, Ooi EE (2019) Evolution of Subgenomic RNA Shapes Dengue Virus Adaptation and Epidemiological Fitness. iScience 16:94–105 Kirkpatrick BD et al (2016) The live attenuated dengue vaccine TV003 elicits complete protection against dengue in a human challenge model. Sci Transl Med 8:330ra336 Manokaran G et al (2015) Dengue subgenomic RNA binds TRIM25 to inhibit interferon expression for epidemiological fitness. Science 350:217–221 Lee AS, Kranzusch PJ, Doudna JA, Cate JH (2016) eIF3d is an mRNA cap-binding protein that is required for specialized translation initiation. Nature 536:96–99 Alvarez DE, Filomatori CV, Gamarnik AV (2008) Functional analysis of dengue virus cyclization sequences located at the 5' and 3'UTRs. Virology 375:223–235 Filomatori CV, Lodeiro MF, Alvarez DE, Samsa MM, Pietrasanta L, Gamarnik AV (2006) A 5' RNA element promotes dengue virus RNA synthesis on a circular genome. Genes Dev 20:2238–2249 Villordo SM, Gamarnik AV (2009) Genome cyclization as strategy for flavivirus RNA replication. Virus Res 139:230–239 Durbin AP (2020) Historical discourse on the development of the live attenuated tetravalent dengue vaccine candidate TV003/TV005. Curr Opin Virol 43:79–87 Grifoni A et al (2017) Patterns of Cellular Immunity Associated with Experimental Infection with rDEN2Delta30 (Tonga/74) Support Its Suitability as a Human Dengue Virus Challenge Strain. J Virol 91 Blaney JE Jr., Hanson CT, Hanley KA, Murphy BR, Whitehead SS (2004) Vaccine candidates derived from a novel infectious cDNA clone of an American genotype dengue virus type 2. BMC Infect Dis 4:39 Mandl CW, Holzmann H, Kunz C, Heinz FX (1993) Complete genomic sequence of Powassan virus: evaluation of genetic elements in tick-borne versus mosquito-borne flaviviruses. Virology 194:173–184 Shi PY, Brinton MA, Veal JM, Zhong YY, Wilson WD (1996) Evidence for the existence of a pseudoknot structure at the 3' terminus of the flavivirus genomic RNA. Biochemistry 35:4222–4230 Proutski V, Gritsun TS, Gould EA, Holmes EC (1999) Biological consequences of deletions within the 3'-untranslated region of flaviviruses may be due to rearrangements of RNA secondary structure. Virus Res 64:107–123 Paranjape SM, Harris E (2007) Y box-binding protein-1 binds to the dengue virus 3'-untranslated region and mediates antiviral effects. J Biol Chem 282:30497–30508 Chiu WW, Kinney RM, Dreher TW (2005) Control of translation by the 5'- and 3'-terminal regions of the dengue virus genome. J Virol 79:8303–8315 Ramos-Lorente SE, Berzal-Herranz B, Romero-Lopez C, Berzal-Herranz A (2024) Recruitment of the 40S ribosomal subunit by the West Nile virus 3' UTR promotes the cross-talk between the viral genomic ends for translation regulation. Virus Res 343:199340 Ma S, Liu JY, Zhang JT (2023) eIF3d: A driver of noncanonical cap-dependent translation of specific mRNAs and a trigger of biological/pathological processes. J Biol Chem 299:104658 Lee AS, Kranzusch PJ, Cate JH (2015) eIF3 targets cell-proliferation messenger RNAs for translational activation or repression. Nature 522:111–114 Yu CY, Hsu YW, Liao CL, Lin YL (2006) Flavivirus infection activates the XBP1 pathway of the unfolded protein response to cope with endoplasmic reticulum stress. J Virol 80:11868–11880 Sanford TJ, Mears HV, Fajardo T, Locker N, Sweeney TR (2019) Circularization of flavivirus genomic RNA inhibits de novo translation initiation. Nucleic Acids Res 47:9789–9802 Rozman B, Fisher T, Stern-Ginossar N (2023) Translation-A tug of war during viral infection. Mol Cell 83:481–495 Zhang F et al (2017) EIF3D promotes gallbladder cancer development by stabilizing GRK2 kinase and activating PI3K-AKT signaling pathway. Cell Death Dis 8:e2868 Marchingo JM, Cantrell DA (2022) Protein synthesis, degradation, and energy metabolism in T cell immunity. Cell Mol Immunol 19:303–315 Zhu T, Fernandez-Sesma A (2020) Innate Immune DNA Sensing of Flaviviruses. Viruses 12 Hinnebusch AG, Ivanov IP, Sonenberg N (2016) Translational control by 5'-untranslated regions of eukaryotic mRNAs. Science 352:1413–1416 Villordo SM, Alvarez DE, Gamarnik AV (2010) A balance between circular and linear forms of the dengue virus genome is crucial for viral replication. RNA 16:2325–2335 Sanborn MA et al (2020) Analysis of cell-associated DENV RNA by oligo(dT) primed 5' capture scRNAseq. Sci Rep 10:9047 Whitehead SS, Hanley KA, Blaney JE Jr., Gilmore LE, Elkins WR, Murphy BR (2003) Substitution of the structural genes of dengue virus type 4 with those of type 2 results in chimeric vaccine candidates which are attenuated for mosquitoes, mice, and rhesus monkeys. Vaccine 21:4307–4316 Chan KR et al (2011) Ligation of Fc gamma receptor IIB inhibits antibody-dependent enhancement of dengue virus infection. Proc Natl Acad Sci U S A 108:12479–12484 Wu RS, Chan KR, Tan HC, Chow A, Allen JC Jr., Ooi EE (2012) Neutralization of dengue virus in the presence of Fc receptor-mediated phagocytosis distinguishes serotype-specific from cross-neutralizing antibodies. Antiviral Res 96:340–343 Ward AM et al (2011) Quantitative mass spectrometry of DENV-2 RNA-interacting proteins reveals that the DEAD-box RNA helicase DDX6 binds the DB1 and DB2 3' UTR structures. RNA Biol 8:1173–1186 Ward AM, Gunaratne J, Garcia-Blanco MA (2014) Identification of dengue RNA binding proteins using RNA chromatography and quantitative mass spectrometry. Methods Mol Biol 1138:253–270 Gu Z (2022) Complex heatmap visualization. Imeta 1:e43 Additional Declarations There is NO Competing Interest. Supplementary Files Supplmaterials.docx Dengue virus 3’ untranslated region regulates RNA genome translation Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6513699","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":452790469,"identity":"079c19a0-6a28-47ca-8fe7-eb613ca5ade5","order_by":0,"name":"Eng Eong Ooi","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-0520-1544","institution":"Duke NUS Medical School","correspondingAuthor":true,"prefix":"","firstName":"Eng","middleName":"Eong","lastName":"Ooi","suffix":""},{"id":452790470,"identity":"dfc9b0a8-51d0-40e3-b21a-adf8072e26bf","order_by":1,"name":"Kiven Kumar","email":"","orcid":"","institution":"Duke-NUS Medical School","correspondingAuthor":false,"prefix":"","firstName":"Kiven","middleName":"","lastName":"Kumar","suffix":""},{"id":452790471,"identity":"b94b3c81-cdfc-4a0e-b750-03620f86e6ab","order_by":2,"name":"Justin Ooi","email":"","orcid":"https://orcid.org/0000-0003-4899-8255","institution":"Duke NUS Medical School","correspondingAuthor":false,"prefix":"","firstName":"Justin","middleName":"","lastName":"Ooi","suffix":""},{"id":452790472,"identity":"5894ba63-721d-4651-843c-b278d4c0b446","order_by":3,"name":"Hwee Cheng Tan","email":"","orcid":"","institution":"Duke-NUS Medical School","correspondingAuthor":false,"prefix":"","firstName":"Hwee","middleName":"Cheng","lastName":"Tan","suffix":""},{"id":452790473,"identity":"f1c07a2d-7c53-4972-abea-1c08c7d66930","order_by":4,"name":"Xudong Lyu","email":"","orcid":"","institution":"Duke-NUS Medical School","correspondingAuthor":false,"prefix":"","firstName":"Xudong","middleName":"","lastName":"Lyu","suffix":""},{"id":452790474,"identity":"ec10907c-842b-4aea-9369-2bfd06741ca9","order_by":5,"name":"Wy Ching Ng","email":"","orcid":"https://orcid.org/0000-0001-5622-7691","institution":"Duke-NUS Graduate Medical School","correspondingAuthor":false,"prefix":"","firstName":"Wy","middleName":"Ching","lastName":"Ng","suffix":""},{"id":452790475,"identity":"4bbb8a8c-e5e0-4dc3-9432-9777c11ae48c","order_by":6,"name":"Kuan Rong Chan","email":"","orcid":"https://orcid.org/0000-0002-2427-671X","institution":"Duke-NUS Medical School","correspondingAuthor":false,"prefix":"","firstName":"Kuan","middleName":"Rong","lastName":"Chan","suffix":""},{"id":452790476,"identity":"5d6af492-8460-4ffb-a6c9-4a2738b4a9e4","order_by":7,"name":"Esteban Finol","email":"","orcid":"https://orcid.org/0000-0002-9830-439X","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Esteban","middleName":"","lastName":"Finol","suffix":""}],"badges":[],"createdAt":"2025-04-23 14:44:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6513699/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6513699/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82179426,"identity":"0c277db2-c6e0-431c-ba5a-5d9d71ade487","added_by":"auto","created_at":"2025-05-07 11:36:44","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":530719,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic representation of wild-type DENVs and their respective Δ30 and Δ31 3’UTR as well as the plaque phenotype Sleman/78 and its mutants. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) 3′UTR RNA secondary structures of the components of tetravalent TV003. The 3’UTRs of DENV-1 to -4 contain a 30-nucleotide deletion in Dumbbell-2 (DB-2), shown in blue, whereas DENV-3 contains an additional 31 nucleotide deletion in DB-1, shown in red.\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eB\u003c/strong\u003e) Representative images of the plaque forming assay for Sleman/78, ∆30 and ∆30/31 in BHK-21 cells. (\u003cstrong\u003eC\u003c/strong\u003e) The mean plaque diameter (mm) of Sleman/78, ∆30 and ∆30/31. Plaque diameters were compared using one-way ANOVA with Tukey’s in GraphPad Prism 9.5.1 software. ***\u003cem\u003ep\u003c/em\u003e £ 0.001, \u0026nbsp;\u003cem\u003ep****\u003c/em\u003e £ 0.0001.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/1cdf5a8b51ac1ca5feb00008.png"},{"id":82180340,"identity":"f56a7488-8184-402b-9fdb-f7a9192f9e6a","added_by":"auto","created_at":"2025-05-07 11:44:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":823172,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eViral RNA replication and protein expression were both reduced with Δ30/31 virus compared to Sleman/78 and Δ30 virus infection. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eLevels of Sleman/78, ∆30, and ∆30/31 virus gRNA relative to GAPDH mRNA at 2, 4, 6, 12, and 24 hpi in MoDCs. (\u003cstrong\u003eB\u003c/strong\u003e) Levels of Sleman/78, ∆30, and ∆30/31 viral sfRNA relative to GAPDH mRNA at 2, 4, 6, 12, and 24 hpi in MoDCs. (\u003cstrong\u003eC\u003c/strong\u003e) Levels of IFN-β mRNA relative to GAPDH mRNA with Sleman/78, ∆30, and ∆30/31 virus infection of MoDCs at 2, 4, 6, 12, and 24 hpi. (\u003cstrong\u003eD\u003c/strong\u003e) Western blot analysis for NS2B, NS3 and NS5 proteins with Sleman/78, Δ30, Δ30/31, virus infections (MOI = 2) or mock infection in Huh-7 cells at 12, 24 and 48 hpi. The expression of viral proteins was detected using monoclonal antibodies against NS2B, NS3 and NS5. Anti-β-actin was used as an internal loading control. Bar graphs illustrate NS5, NS3 and NS2B proteins expression relative to wildtype. (\u003cstrong\u003eE\u003c/strong\u003e) Proteins expression relative to wildtype at 12 h PI. (\u003cstrong\u003eF\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eProteins expression relative to wildtype at 24 h PI. (\u003cstrong\u003eG\u003c/strong\u003e) Proteins expression relative to wildtype at 48 h PI. Representative Western blot assays from three independent experiments are shown. Band intensities were compared using one-way ANOVA with Tukey’s. *\u003cem\u003ep\u003c/em\u003e £ 0.05, **\u003cem\u003ep\u003c/em\u003e £ 0.01, ***\u003cem\u003ep\u003c/em\u003e £ 0.001, \u0026nbsp;****\u003cem\u003ep\u003c/em\u003e £ 0.0001. Data are presented as mean ± SD.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/17c3c9c2d096c1c146119ffd.png"},{"id":82179431,"identity":"7a8b3ec4-1f7d-4546-be3c-fb5ec9d7ea3c","added_by":"auto","created_at":"2025-05-07 11:36:44","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":283042,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHost proteins involved in translation preferentially bound Sleman/78 compared to ∆30 and ∆30/31 3’UTR\u003c/strong\u003e. (\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eUnsupervised clustering revealed the abundance of cytoplasmic proteins that were differentially bound to Sleman/78 compared to ∆30/31 (\u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05) for the Sleman/78, ∆30 and ∆30/31 samples. The enriched proteins are color-coded based on their biological function. More translation-related proteins were bound to Sleman/78 compared to ∆30 and ∆30/31. The abundance of proteins is represented by Z-score\u003cstrong\u003e. \u003c/strong\u003e(\u003cstrong\u003eB\u003c/strong\u003e) Scatter plot highlights the proteins that bound Sleman/78 (WT) at 2-fold or greater rate than ∆30 and/or ∆30/31. Dotted line indicates 2-fold difference in enrichment of proteins with Sleman/78 compared to mutant 3’UTR. Red dots represent proteins that bound Sleman/78 at 2-fold or greater rate compared to ∆30/31; purple dots represent proteins that bound Sleman/78 at 2-fold or greater rate than ∆30. Black dots represent proteins that bound Sleman/78 at 2-fold or greater rate than both WT/∆30 and WT/∆30/31.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/54fb688c764280b1718e0c63.png"},{"id":82179429,"identity":"2f062b36-da5a-4bdc-8157-4946599bbae3","added_by":"auto","created_at":"2025-05-07 11:36:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":401697,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRNA Immunoprecipitation validated differential binding of translation-related proteins with Sleman/78, ∆30 and ∆30/31 3’UTR. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e, \u003cstrong\u003eC\u003c/strong\u003e, \u003cstrong\u003eE\u003c/strong\u003e and \u003cstrong\u003eG\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRNA immunoprecipitation was performed on Huh-7 cells at 4 h transfected with 3’UTR from wildtype, ∆30, and ∆30/31. The 4 h the transfected cells were subjected to pull-down using EIF3D, EIF3K, RL17, RL28 or IgG (negative control), followed by immunoblotting with EIF3D , EIF3K, RL17 and RL28 to confirm the efficacy of the pull-down. 'Input' refers to the cell lysate before RNA immunoprecipitation; 'Sup' refers to the supernatant after incubation of the antibody-coated beads with the cell lysate; 'Pellet' refers to the final centrifuged product after washing. (\u003cstrong\u003eB\u003c/strong\u003e, \u003cstrong\u003eD\u003c/strong\u003e, \u003cstrong\u003eF\u003c/strong\u003eand \u003cstrong\u003eH\u003c/strong\u003e) Bar graphs show the fold enrichment of 3’UTR after pull-down with EIF3D, EIF3K, RL17 and RL28 as measured by RT-qPCR. Mean ± SEM values for the fold enrichment were compared using one-way ANOVA with Tukey’s post-hoc test. *\u003cem\u003ep\u003c/em\u003e≤ 0.05, **\u003cem\u003ep\u003c/em\u003e ≤ 0.01, ***\u003cem\u003ep\u003c/em\u003e ≤ 0.001, ****\u003cem\u003ep\u003c/em\u003e ≤ 0.0001\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/2b3268b8975eb1331440e9e9.png"},{"id":82180342,"identity":"26268e6c-7504-45a0-a6e6-9ca19882c94a","added_by":"auto","created_at":"2025-05-07 11:44:44","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":757609,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEIF3D silencing reduced viral RNA translation and replication. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eHuh-7 cells were transfected with scrambled siRNA (siCtrl) or EIF3D-targeting siRNA for 24 hours before infection. Cell lysates were analysed by western blot to assess EIF3D levels, with GAPDH serving as a loading control. (\u003cstrong\u003eB\u003c/strong\u003e, \u003cstrong\u003eD\u003c/strong\u003e and \u003cstrong\u003eF\u003c/strong\u003e) Huh-7 cells were either mock-infected or infected with Sleman/78, ∆30, or ∆30/31 at an MOI of 5 in triplicate. Cells were collected at 2, 4, 6, 12, and 24 hours post-infection (hpi), and lysates were analysed by western blot for DENV NS2B, NS3, and NS5 proteins, with β-actin serving as loading control. Western blot analysis showed that EIF3D knockdown (siEIF3D) reduced viral protein translation for Sleman/78, ∆30, and ∆30/31 at 6 h PI\u003cstrong\u003e, \u003c/strong\u003e12h PI and 24h PI compared to siCtrl. (\u003cstrong\u003eC\u003c/strong\u003e, \u003cstrong\u003eE\u003c/strong\u003e and \u003cstrong\u003eG\u003c/strong\u003e) RT-qPCR was used to quantify viral gRNA levels at 6h PI,\u003cstrong\u003e \u003c/strong\u003e12h PI\u003cstrong\u003e \u003c/strong\u003eand 24h PI. The results showed that gRNA levels were lower in siEIF3D-transfected cells compared to the siCtrl group. Mock-infected cells served as a negative control. Values of gRNA/GAPDH were compared using one-way ANOVA with Tukey’s post-hoc test. *\u003cem\u003ep\u003c/em\u003e ≤ 0.05, **\u003cem\u003ep\u003c/em\u003e ≤ 0.01, ***\u003cem\u003ep\u003c/em\u003e ≤ 0.001, ****\u003cem\u003ep\u003c/em\u003e ≤ 0.0001\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/d2da814056667c68666925bd.png"},{"id":82180341,"identity":"aa93891b-e6c0-4c5d-be85-56a6facc0fa7","added_by":"auto","created_at":"2025-05-07 11:44:44","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1283582,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe eGFP expression level in HEK-293T cells transfected with 5’ and 3 UTR of DENV incorporated with eGFP. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e, \u003cstrong\u003eD\u003c/strong\u003e, \u003cstrong\u003eG\u003c/strong\u003e, and \u003cstrong\u003eJ\u003c/strong\u003e) Representative fluorescence microscopy images are shown DENV-3, DENV-1, DENV-4 and DENV-2. (B, E, H and K) Percentage of eGFP-positive cells for the 5' and 3' UTRs from DENV-3, DENV-1, DENV-4, and DENV-2. (\u003cstrong\u003eC\u003c/strong\u003e, \u003cstrong\u003eF\u003c/strong\u003e, \u003cstrong\u003eI\u003c/strong\u003e and \u003cstrong\u003eL\u003c/strong\u003e) Mean fluorescence intensity (MFI) of eGFP was calculated as the average fluorescence intensity of all eGFP-positive cells divided by the number of eGFP-positive cells. Percentages of eGFP-positive cells and eGFP MFI were compared using one-way ANOVA with Tukey’s post hoc test. *\u003cem\u003ep\u003c/em\u003e ≤ 0.05, **\u003cem\u003ep\u003c/em\u003e ≤ 0.01, ***\u003cem\u003ep\u003c/em\u003e ≤ 0.001, ****\u003cem\u003ep\u003c/em\u003e ≤ 0.0001, ns: nonsignificant.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/ba109d08b70cca67e28c9647.png"},{"id":84711484,"identity":"5143ebbc-16c6-4ec2-bcf2-c2211a67d175","added_by":"auto","created_at":"2025-06-16 13:24:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5247000,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/8b914e6a-72f2-4797-9d5f-59c7675b0bab.pdf"},{"id":82179430,"identity":"8c8580aa-7dbb-4187-8aaf-ce352e7beec6","added_by":"auto","created_at":"2025-05-07 11:36:44","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1263802,"visible":true,"origin":"","legend":"Dengue virus 3\u0026#x2019; untranslated region regulates RNA genome translation","description":"","filename":"Supplmaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-6513699/v1/80312c55838b10172542043a.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Dengue virus 3’ untranslated region regulates RNA genome translation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDengue is an acute febrile illness caused by infection with any one of four different dengue viruses (DENV-1 to -4), all of which are transmitted by \u003cem\u003eAedes\u003c/em\u003e mosquitoes. Increased volume of trans-continental travel, urbanisation and climate change have expanded the geographic footprints of both DENVs and \u003cem\u003eAedes\u003c/em\u003e mosquitoes from the tropics to North America and Mediterranean Europe; more than half of the world\u0026rsquo;s population thus live at risk of dengue each year\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eClinical observations have indicated that those who develop high viraemia levels are at greater risk of severe dengue\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e; high infection burden exacerbates pro-inflammatory responses that lead to life-threatening shock from vascular leakage, internal haemorrhage and organ dysfunction\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Despite the link between viraemia and severe dengue, the genetic elements that control of DENV replication in dengue patients remains poorly understood. An impediment to understanding how DENV genetics affect infection outcome is the lack of infection models that accurately recapitulate the clinical features of dengue. Live attenuated (non-chimeric) dengue vaccine strains thus provide unique opportunities to overcome such limitations. These vaccine strains were all derived from DENVs isolated from dengue patients and attenuated through empirical approaches but without knowledge on the mechanisms of attenuation\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. The live attenuated vaccine strains that have shown low levels of viraemic infection and mild systemic symptoms in thousands of clinical trial volunteers thus provide a clinically validated context to dissect the viral genetic determinants critical causing dengue.\u003c/p\u003e \u003cp\u003eTV003 is a tetravalent live attenuated dengue vaccine candidate that has completed phase III clinical trial\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. TV003 contains 3 live attenuated DENVs; the fourth component of TV003 is a DENV-4/-2 chimeric virus. The attenuated DENV-1, -3 and \u0026minus;\u0026thinsp;4 strains of TV003 were derived through deletions in the 3\u0026prime;UTR. Both DENV-1 and \u0026minus;\u0026thinsp;4 contained a 30 nucleotide deletion (Δ30)\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e in the same position of the 3\u0026rsquo;UTR. However, the same deletion in DENV-3 produced an inadequately attenuated mutant\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e; this strain is now used for controlled human infection studies\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. An additional 31 nucleotide proximal to the Δ30 mutation was then deleted to produce a sufficiently attenuated mutant (DENV-3 Δ30/31)\u003csup\u003e12\u003c/sup\u003e. The parent wild-type DENV-3 as well as its semi- (Δ30) and fully-attenuated (Δ30/31) derivatives thus provide a unique opportunity to glean insights into the virus-host interactions necessary for pathogenesis. Host-DENV interactions critical for symptomatic infection should thus show progressive loss of function with increasing number of deletions in the 3\u0026rsquo;UTR. We found, unexpectedly, that the 3\u0026rsquo;UTR plays a critical role in initiating viral genome translation.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eWild-type DENV-3 (Sleman/78), Δ30 and Δ30/31 mutants show different growth characteristics\u003c/h2\u003e \u003cp\u003eDENV 3\u0026rsquo;UTR contains nucleotide sequence that complement each other to form highly conserved secondary RNA structures\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. We thus first determined the location of Δ30 and Δ31 relative to the secondary RNA structures previously defined using RNA phylogeny\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA shows the RNA structures of wild-type DENV-1 (Western Pacific), DENV-3 (Sleman/78), and DENV-4 (Dominica 814669) along with their respective attenuated derivatives. We have also included DENV-2 (Tonga/74) and the Δ30 mutant that only partially attenuated Tonga/74 virus; this mutant, like DENV-3 Δ30, is now used for controlled human infection studies\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo examine the effects these 3\u0026rsquo;UTR deletions have on DENV-3 fitness, we constructed infectious clones of the wild-type Sleman/78, Δ30, and Δ30/31 viruses using Gibson assembly (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). The rescued viruses were amplified in C6/36 cell line and quantified by plaque assay in BHK-21 cells. Plaque diameter of Δ30/31 mutant was smaller compared to those of either wild-type DENV-3 or the Δ30 mutant (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and \u003cb\u003eC\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo determine if the differences in plaque diameter could be accounted for by virus replication dynamics, Sleman/78 and its Δ30, and Δ30/31 derivatives were inoculated onto MoDCs at a multiplicity of infection (MOI) of 5. Both viral genomic and subgenomic flavivirus RNA (gRNA and sfRNA, respectively), were measured; sfRNA was analysed as we have previously shown that changes in 3\u0026rsquo;UTR sequence can alter its abundance\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Levels of infection of the different viruses were comparable, as indicated by the lack of difference in both gRNA and sfRNA levels at both 2- and 4-hours post-infection (hpi) (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cb\u003eB\u003c/b\u003e). At 6 hpi, however, levels of gRNA and sfRNA of both Sleman/78 and ∆30 virus continued on an upward trajectory whereas those for ∆30/31 virus were reduced (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cb\u003eB\u003c/b\u003e). The ratio of gRNA:sfRNA also showed the same trend as gRNA and sfRNA levels (\u003cb\u003eSupplementary Fig.\u0026nbsp;2\u003c/b\u003e), suggesting that the nucleotide deletions did not affect the formation of sfRNA.\u003c/p\u003e \u003cp\u003eThe expression of IFN-β showed no difference relative to baseline levels for Sleman/78 and ∆30 virus infection. In striking contrast, ∆30/31 virus infection induced IFN-β expression that peaked at 6hpi, at a time when both gRNA and sfRNA levels of ∆30/31 virus were reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eViral protein expression in Huh-7 cells also showed similar differences in kinetics as viral RNA replication. Unlike Sleman/78 and ∆30 viruses where viral NS2B, NS3 and NS5 proteins could be detected as early at 12 hpi, those of ∆30/31 mutant could only be detected at 24 hpi and later (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD-G). Taken collectively, these findings suggest that Δ30/31 delayed viral protein expression, RNA replication and compromised innate immune evasion properties.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDBI and DBII binds ribosomal and translation-related proteins\u003c/h3\u003e\n\u003cp\u003eTo understand how deletions in DBI and DBII of Sleman/78 affected DENV genome translation, virus replication and innate immune evasion, we first sought to define the functions of DBI and DBII. We reasoned that these structures engage in specific viral RNA-host protein interactions. Using stable isotope labelling by amino acids in Huh-7 cell culture coupled with quantitative mass spectrometry (SILAC-qMS) (\u003cb\u003eSupplementary Fig.\u0026nbsp;3\u003c/b\u003e), we identified proteins that preferentially bound (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) Sleman/78 3\u0026prime;UTR compared to the ∆30 and ∆30/31 mutants and further filtered for only cytoplasmic proteins based on the subcellular localization information from the UniProt database\u003c/p\u003e \u003cp\u003eUnsupervised clustering analysis of the MS data revealed 48 proteins that were differentially enriched. A total of 34 proteins were bound to Sleman/78 more than both Δ30 and Δ30/31 3\u0026prime;UTR (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Most of these proteins were ribosomal proteins, eukaryotic initiation factors (EIFs), and tRNA-associated proteins, which all function to support protein translation. In contrast, only 14 proteins with mixed functions were enriched with ∆30 and ∆30/31 compared to Sleman/78 3\u0026rsquo;UTR (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003eTo show the differential enrichment of proteins that bound Sleman/78 compared to ∆30 and ∆30/31 3\u0026rsquo;UTRs, we constructed a scatter-plot of the statistically significant hits from our SILAC-qMS data. We found 38 proteins that showed 2-fold or greater enrichment with Sleman/78 3\u0026rsquo;UTR compared to both ∆30 and ∆30/31 3\u0026rsquo;UTR (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Five proteins were enriched when Sleman/78 was compared to ∆30 UTR only, whereas 9 proteins were enriched with Sleman/78 more than ∆30/31 3\u0026rsquo;UTR only (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Similar to data shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, most of these proteins function to support protein translation. Collectively, the ∆30/31 3\u0026rsquo;UTR resulted in a greater loss of binding of translation-related proteins to the DENV-3 positive-strand RNA.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo validate our SILAC-qMS findings that specific proteins involved in translation bound Sleman/78 3\u0026rsquo;UTR more than the ∆30/31 mutant, we turned to RNA immunoprecipitation. We conducted the immunoprecipitation study using monoclonal antibodies (mAbs) to the eukaryotic initiation factors 3D and 3K (EIF3D and EIF3K), as well as ribosomal protein RL17 and RL28, given the reduced binding of these proteins to the various mutant as compared to Sleman/78 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Cells transfected with the different 3\u0026rsquo;UTRs with comparable efficiencies \u003cb\u003e(Supplementary Figs.\u0026nbsp;4A-D\u003c/b\u003e) were subjected to immunoprecipitation using protein-specific mAbs. The 3\u0026rsquo;UTR RNA was then quantified using RT-qPCR. Consistent with the SILAC-qMS findings (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB) the amount of 3\u0026rsquo;UTR RNA in the EIF3D immunoprecipitate (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), relative to the amount transfected 3\u0026rsquo;UTR were significantly higher with Sleman/78 and ∆30 compared to ∆30/31 3\u0026rsquo;UTR (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Likewise, higher amounts of Sleman/78 and ∆30 3\u0026rsquo;UTR compared to ∆30/31 3\u0026rsquo;UTR were found in the EIF3K (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC-D) and RL28 immunoprecipitates (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG-H). Immunoprecipitation of RL17 also showed consistent findings as those of SILAC-qMS, where both the ∆30 and ∆30/31 3\u0026rsquo;UTR were markedly lower than Sleman/78 (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE-F). These findings collectively support the notion that DBI and DBII in the 3\u0026rsquo;UTR bind proteins involved in viral RNA translation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eDENV-3 DBI and DBII support genome translation\u003c/h3\u003e\n\u003cp\u003eBinding of proteins involved in translation suggest that Sleman/78 3\u0026rsquo;UTR may, at least in part, be involved in supporting viral protein expression. In particular, EIF3D has previously been shown to initiate EIF4E independent translation\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, and this non-canonical pathway could be employed by DENV to translate its genome. Such an explanation is plausible as the DENV genome is cyclized through complementary sequence interactions between the 5\u0026rsquo; and 3\u0026rsquo;UTRs\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e; the 5\u0026prime; upstream AUG region (5\u0026prime;UAR) in the 5\u0026prime;UTR base-pairs with its complementary 3\u0026prime;UAR counterpart at the base of the 3\u0026prime; stem-loop (3\u0026prime;SL) in the 3\u0026prime;UTR to form the panhandle of the cyclized positive-strand RNA genome. DBI and DBII are thus in proximity to the translation start site on the DENV capsid gene (\u003cb\u003eSupplementary Figs.\u0026nbsp;5A\u003c/b\u003e and \u003cb\u003eB\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo test the notion binding of EIF3D with 3\u0026rsquo;UTR is critical for DENV genome translation, we infected EIF3D-silenced Huh-7 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA) with Sleman/78, ∆30, and ∆30/31. If EIF3D recruitment by wild-type 3\u0026rsquo;UTR is necessary for initiating translation, then silencing EIF3D would delay Sleman/78 protein expression. Compared to control cells transfected with scrambled siRNA, the levels of DENV NS2B, NS3 and NS5 proteins (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB), as well as viral RNA levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC) were universally lower in EIF3D silenced cells at 6 hpi. Similar differences in viral protein expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD) and abundance of viral RNA (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE) were also observed at 12 hpi. At 24 hpi, the levels of NS2B but not NS3 and NS5 appeared higher in EIF3D siRNA treated cells compared to controls, possibly due to depletion of EIF3D siRNA and hence initiation of translation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). However, viral RNA levels remain significantly lower in EIF3D siRNA treated cells than controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eReduced translation of DENV-1 and − 4 but not DENV-2 ∆30 mutants\u003c/h3\u003e\n\u003cp\u003eThe findings with Sleman/78 and its two mutants also raised the possibility that a single 30 nucleotide stretch of deletion was sufficient to attenuate translation of DENV-1 and DENV-4, but not DENV-2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA), as evidenced by clinical trial findings\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. To test this possibility, we used an eGFP reporter system to measure translation. We first determined if such a eGFP reporter system could recapitulate the differences in Sleman/78, ∆30, and ∆30/31 genome translation (\u003cb\u003eSupplementary Figs.\u0026nbsp;6A and 6B)\u003c/b\u003e; plasmid traces were minimal in the RNA extract (\u003cb\u003eSupplementary Fig.\u0026nbsp;6C\u003c/b\u003e). Consistent with our findings with viral protein expression, both the proportion of eGFP positive cells (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cb\u003eB\u003c/b\u003e) and the mean fluorescence intensity (MFI) ( Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC) were lower when the eGFP gene was flanked with 5\u0026rsquo; and 3\u0026rsquo;UTR of the ∆30/31 mutant compared to either Sleman/78 or ∆30 mutant.\u003c/p\u003e \u003cp\u003eWe then constructed plasmids with eGFP reporter gene flanked with 5\u0026rsquo;UTR of DENV-1 (Western Pacific), DENV-4 (Dominican 814669) and DENV-2 (Tonga/74) on one side, and their respective 3\u0026rsquo;UTR, as well as their ∆30 and ∆30/31 mutants. Transfection of these plasmids into HEK293 cells produced comparable levels of eGFP transcripts (\u003cb\u003eSupplementary Figs.\u0026nbsp;6D\u003c/b\u003e and \u003cb\u003eE)\u003c/b\u003e. As expected, unlike Sleman/78, the ∆30 mutants of both DENV-1 and \u0026minus;\u0026thinsp;4 showed significantly lower percentage of eGFP positive cells as well as fluorescence intensity compared to eGFP gene linked with wild-type DENV-1 and \u0026minus;\u0026thinsp;4 3\u0026rsquo;UTRs (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD-I). In contrast, no difference was found in either the percentage of eGFP positive cells or fluorescence intensity when the eGFP gene was constructed using DENV-2 (Tonga/74) ∆30 compared to wild type 3\u0026rsquo;UTR. This finding is consistent with the clinical phenotype of DENV-2 Tonga/74 ∆30 mutant (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ-L), which was insufficiently attenuated to be used as a vaccine strain\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. An additional ∆31 mutation in the Tonga/74 3\u0026rsquo;UTR significantly reduced both the percentage of eGFP positive cells as well as fluorescence intensity (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ-L).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe large body of safety and immunogenicity data on the live attenuated tetravalent TV003 dengue vaccine candidate provides a unique and clinically validated context to examine the function of DBI and DBII in the 3’UTR of DENVs.\u003c/p\u003e \u003cp\u003eOur findings revealed that DBI and DBII bind proteins involved in viral genome translation. The 3’UTR of DENVs contain regions of nucleotide sequences that are highly conserved. Where there is variability in the nucleotide sequence, the secondary RNA structures, including DBI and DBII remain conserved through compensatory mutations\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Various studies have shown that these secondary and more complex RNA structures in the 3’UTR support DENV infection\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Deletions that either disrupted or reshaped the conserved secondary structure elements in the 3’UTR reduced DENV replication\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Furthermore, a study on the related West Nile virus (WNV) showed that viral genome translation was regulated through specific interactions between the 5’ and 3’UTRs to recruit and bind the 40S ribosome subunit\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. However, this study only identified a single ribosomal 40S protein being regulated in that manner but did not adopt an unbiased approach to identify the interacting partners of WNV 3’UTR. Our findings thus add granularity to this fledgling body of knowledge and suggest that DBI and DBII function as regulators of viral genome translation.\u003c/p\u003e \u003cp\u003eAmongst the hits on our SILAC-qMS experiment, the finding of EIF3D as a binding partner of Sleman/78 but not the ∆30/31 mutant is intriguing. EIF3D is a subunit of the EIF3 multi-protein complex\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. It binds the mRNA cap to initiate translation when the canonical EIF4E is inactive\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, notably during endoplasmic reticulum (ER) stress to regulate translation of specific sets of mRNA, such as those involved in cell proliferation\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. DENV infection is known to induce ER stress\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. The requirement for EIF3 complex for translation was also found for the closely related Zika virus (ZIKV) \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. EIF3D-dependent translation of DENV genome could thus be strategic for the virus to ensure sustained replication even in stressed cells\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBesides initiating translation, EIF3D could have other non-canonical functions, including increasing RNA stability\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. With the loss of binding of EIF3D and possibly other ribosomal proteins to the 3’UTR, the viral genome could be rendered more susceptible to RNA processing and metabolic degradation\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Processing of viral RNA could, in turn lead to increased activation of cytoplasmic viral RNA sensors that activate the innate immune response, especially the type-I IFN response that limit virus dissemination to attenuate systemic infection. Indeed, our viral RNA replication kinetics findings in MoDCs support this notion. The ∆30/31 mutation could thus be a “double jeopardy” on Sleman/78, reducing both viral protein expression and the cytoplasmic viral RNA half-life. Considering that DENV non-structural proteins also contribute to innate immune evasion\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, reduced translation of ∆30/31genome could favour prompt type I IFN response that restrict systemic infection dissemination, further contributing to its attenuated phenotype.\u003c/p\u003e \u003cp\u003eThat the two DB RNA structures of Sleman/78 were found to interact with ribosomal proteins and EIF3 subunits also provide evolutionary perspectives on DENVs. Besides supporting replication of its genome, cyclization could also be a strategy for DENVs, and possibly other orthoflaviviruses, to preserve a short 5’ UTR. The 5’UTR is already highly structured to support interactions with both host and viral proteins\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e necessary for viral replication. Incorporating additional RNA elements to bind ribosomal proteins and EIF3 complex subunits would lengthen the 5’UTR and compromise cap-dependent translation\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Evolving RNA functional elements that recruit translation factors in the 3’ instead of 5’UTR could thus confer strategic advantage to efficient translation of cyclized DENV RNA genome.\u003c/p\u003e \u003cp\u003eOur findings also refine the understanding of how DENV genome replicates. It has been thought that the cyclized genome is important for viral RNA replication whereas translation of the viral polyprotein occurs on the linear RNA conformation\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. The rationale for different RNA conformation to explain replication and translation is that the polymerase and the ribosome travel in opposite direction on the positive-sense viral RNA genome. An in vitro reconstitution approach of ZIKV genome supported such an explanation although the study used a mini genome with a 3’UTR that lacked DBI and DBII\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. As RNA conformation is highly dynamic, it is possible that once the translation machinery is assembled, in part on the 3’UTR, the cyclized RNA genome linearizes. An alternative explanation is that, once the first few negative-sense RNA strands are transcribed, the polymerase is more active on the cyclized negative-sense RNA intermediate to produce new positive-sense DENV genome, whereas translation remains exclusively on the positive-sense template. Indeed, this alternative explanation is supported by the large difference in the quantity of positive-sense viral RNA compared to the negative-sense intermediate in infected cells\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Further research will be needed to tease apart which of these explanations is correct.\u003c/p\u003e \u003cp\u003eBesides defining the function of DBI and DBII of DENV-3, our findings show that DBI and DBII of the other three DENVs also function to support viral genome translation, albeit with important type-specific nuances. That ∆30 in DBII reduced translation of eGFP reporter gene when flanked with DENV-1 and DENV-4 UTRs but not those of DENV-2 are entirely consistent with the history of TV003 development\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. DENV-2 Tonga/74∆30 strain was developed as a vaccine candidate. However, pre-clinical data showed that Tonga/74∆30 was insufficiently attenuated for further clinical development as a component of TV003\u003csup\u003e22, 23\u003c/sup\u003e; the DENV-2 component of TV003 is thus composed of a chimeric DENV2/4∆30(ME) construct, with the prM and envelope (E) genes from DENV-2 New Guinea C (NGC) strain spliced into the DENV-4 ∆30 mutant backbone\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Whether a DENV-2 Tonga/74 ∆30/31 mutant could prove sufficiently attenuated whilst retaining immunogenicity, however, is uncertain as such a study has not been reported. Nonetheless, given the highly conserved DBI and DBII structures, it would be reasonable to expect that the interactome of these secondary RNA structures would be consistent across the different lineages and sub-lineages of each of the four DENVs. It may thus be possible to update the genomic backbones of components of TV003, which were developed using DENVs isolated in the 1970s and 1980s, with more contemporaneous strains to reduce mismatch with circulating DENVs.\u003c/p\u003e \u003cp\u003eThis study has limitations. Whilst we have shown that several proteins involved with viral RNA translation bound 3’UTR, we have not shown how they interacted with other elements of the translation machinery, including those that bind to the 5’UTR as well as the ribosomes to regulate viral protein expression.\u003c/p\u003e \u003cp\u003eIn conclusion, DBI and DBII of the 3' UTR bind proteins involved in translation, including EIF3D, to express viral proteins and promote replication. Deletions of nucleotides within these DB structures reduce viral genome translation and render viral RNA more vulnerable to processing and sensing to attenuate infection.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e\n\n "},{"header":"Methods","content":"\u003ch2\u003eCells\u003c/h2\u003e\u003cp\u003eHuH-7, C6/36, Vero and BHK-21 cells were purchased from the American Type Culture Collection (ATCC) and cultured according to ATCC recommendations. Primary monocytes were isolated from a healthy seronegative volunteer with informed consent (NUS-IRB-2022-501), and cultured as described\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch3\u003eGeneration of infectious clone\u003c/h3\u003e\u003cp\u003eWild-type DENV-3 Sleman/78 (GenBank: AY656169.1), ∆30 (GenBank: AY656170.1), and ∆30/3\u003csup\u003e10\u003c/sup\u003e were rescued using infectious clone technology. The complete genomes of Sleman/78, ∆30, and ∆30/31 were divided into seven overlapping fragments (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). Each fragment, approximately 1800–2000 bases in length, was chemically synthesized (Company’s name). Briefly, NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs) was used to amplify each fragment with primers designed for the respective regions (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e). The amplified fragments were then purified using the MinElute Gel Extraction Kit (Qiagen) according to the manufacturer’s instructions. Additionally, the pUC19 vector, which contains the human cytomegalovirus (CMV) promoter and simian virus 40 polyadenylation signal (SV40-pA), was amplified using primers listed in \u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e. The purified genome fragments and the pUC19 vector were assembled in equimolar amounts (0.1 pmol of each fragment) using the NEBuilder HiFi DNA Assembly Kit (New England BioLabs) at 50°C for 1 hour. Five microliters of the assembled infectious clones were transfected into HEK293T cells in a 24-well plate using 3 µL of Lipofectamine 2000 (Thermo Fisher Scientific), following the manufacturer’s protocol. The supernatant containing infectious clone-derived viruses was collected 72 hours post-transfection and propagated in C6/36 cells supplemented with 3% FBS. Infected cells were observed daily for syncytia formation for 6 to 7 days, after which they were harvested and stored at − 80°C until further use.\u003c/p\u003e\u003ch2\u003ePlaque assay\u003c/h2\u003e\u003cp\u003ePlaque assay was performed on BHK-21 cells as previously described\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Briefly, serial dilutions (10-fold) of virus were added to BHK-21 cells in 24-well plates and incubated for 1 h at 37°C. Media was aspirated and replaced with 0.9% methyl-cellulose in maintenance media. Six days later, cells were fixed with 20% formalin and stained with1% crystal violet. The number of plaques was counted visually.\u003c/p\u003e\u003ch2\u003eViral replication kinetics on primary monocyte derived dendritic cells (MoDCs)\u003c/h2\u003e\u003cp\u003ePeripheral blood mononuclear cells (PBMCs) were isolated from a healthy donor who had tested negative for anti-DENV antibodies using plaque reduction neutralization test\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. CD14 + monocytes were isolated from PBMCs using CD14 microbeads (Miltenyi Biotec) according to manufacturer’s protocol. Differentiation of CD14 cells into MoDCs were done in six-well plates using RPMI 1640 supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 100 ng/ml IL-4 (eBioScience) and 50 ng/ml granulocyte macrophage-colony stimulating factor (GM-CSF, eBioScience) for 6 days with media change on the third day. MoDCs were seeded at 2 x 10\u003csup\u003e4\u003c/sup\u003e cells per well in 96-well tissue culture plate and infected with 5 MOI of Sleman/78, ∆30 and ∆30/31 viruses. At pre-defined timepoints post-infection, cells were washed once in PBS before RNA extraction using RNeasy Kit (Qiagen) according to manufacturer’s protocols. Total RNA was converted into cDNA using LunaScript RT SuperMix Kit (Company’s name), according to the manufacture’s protocol. Viral replication was measured using two sets of the primers (\u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e), one which targets gRNA and another which measure sfRNA were used in the qPCR. Viral RNA was semi-quantified by dividing it against RT-qPCR results on the housekeeping GAPDH gene.\u003c/p\u003e\u003ch2\u003eProtein expression\u003c/h2\u003e\u003cp\u003eWestern blot was performed to assess DENV protein expression in infected cells. Briefly, Huh-7 cells were infected with Sleman/78, ∆30, or ∆30/31 virus at a multiplicity of infection (MOI) of 2. At pre-defined timepoints post-infection, cells were washed once with PBS and lysed in RIPA buffer containing protease inhibitors (1:100 dilution), followed by incubation on ice for 2 hours. The crude lysates were centrifuged at 20,000 x g, and the supernatant was collected and mixed with 4X Laemmli buffer. The samples were then heated at 95°C for 10 minutes before separation by SDS-PAGE. Proteins were transferred onto a PVDF membrane and incubated with the following primary antibodies (1:5000 dilution): mouse anti-DENV NS5 IgG monoclonal antibody (mAb) (GeneTex, GTX629447), mouse anti-DENV NS3 IgG mAb (GeneTex, GTX629477), and mouse anti-DENV NS2B mAb (GeneTex, GTX638874). The membrane was then probed with HRP-conjugated anti-mouse or anti-rabbit IgG secondary antibodies (1:5000 dilution). Detection was performed using Amersham ECL reagents (Cytiva). Blotting for β-actin with an anti-β-actin IgG mAb (1:5000; Cell Signaling Technology, 8H10D10) served as a loading control.\u003c/p\u003e\u003cp\u003e \u003cb\u003eConstruct of 3’UTR and\u003c/b\u003e \u003cb\u003ein-vitro\u003c/b\u003e \u003cb\u003etranscription\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePlasmids containing the 3’UTR of Sleman/78, ∆30, ∆30/31, or NS2A were chemically synthesized with a tobramycin tag and a T7 promoter. A size-matched control, NS2A from Sleman/78, was used as a control in this experiment\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. To generate a 560 bp fragment containing the T7 promoter, 3’UTR, and tobramycin tag, the plasmid was mixed with primers (\u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e)\u003c/p\u003e\u003cp\u003eand NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs) for amplification. The resulting fragments were gel-purified using the MinElute Gel Extraction Kit (Qiagen) and subsequently used for \u003cem\u003ein vitro\u003c/em\u003e transcription. The 3’UTR fragments were transcribed \u003cem\u003ein vitro\u003c/em\u003e using a reaction mixture containing 10X TB buffer (300 mM Tris-HCl, pH 7.7, 20 mM spermidine, 30 mM MgCl₂, 50 mM NaCl, 100 mM DTT, 20% PEG 8600, and 0.05% Triton X-1000), 6 mM NTP mix, 0.4 M urea, T7 polymerase, and 10 mM DTT. The reaction was incubated at 37°C for 2 hours. The transcribed RNA was purified using the RNeasy Kit (Qiagen) according to the manufacturer’s protocol. RNA concentration was measured using a NanoDrop spectrophotometer.\u003c/p\u003e\u003ch2\u003eAdaptation of Huh-7 cells to SILAC medium\u003c/h2\u003e\u003cp\u003eStable isotope labelling by amino acids in cell culture (SILAC) was performed to obtain labeled Huh-7 cells with light (K\u003csub\u003e0\u003c/sub\u003eR\u003csub\u003e0\u003c/sub\u003e), medium (K\u003csub\u003e4\u003c/sub\u003eR\u003csub\u003e6\u003c/sub\u003e), and heavy (K\u003csub\u003e8\u003c/sub\u003eR\u003csub\u003e10\u003c/sub\u003e) isotopes (\u003cb\u003eSupplementary Fig.\u0026nbsp;3\u003c/b\u003e). The labelled Huh-7 cells were washed with PBS and then trypsinized. The detached cells were transferred to a 50 mL conical tube and centrifuged at 500 x g for 5 minutes. The resulting cell pellet was resuspended in 10–20 mL of PBS. Cell lysis and protein concentration measurement were carried out as previously described\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch2\u003eRNA-protein interactions\u003c/h2\u003e\u003cp\u003eTobramycin RNA affinity chromatography, mass spectrometry, and data analysis were performed as previously described\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. In brief, RNA was immobilized on a tobramycin matrix and incubated with SILAC-labelled Huh-7 cell lysates. After 2h incubation at 4°C, the matrix was washed, and bound proteins were eluted using excess tobramycin. The eluates were analysed by quantitative mass spectrometry (qMS), enabling a direct comparison of proteins bound to the Sleman/78 3’UTR, ∆30 3’UTR, ∆30/31 3’UTR, or control RNA, as illustrated in \u003cb\u003eSupplementary Fig.\u0026nbsp;3\u003c/b\u003e.\u003c/p\u003e\u003ch2\u003eAnalysis of bound proteins\u003c/h2\u003e\u003cp\u003eThe protein concentration after the elution from the column was determine by using BCA kit according manufacture protocol. Approximately 10 µg of elution proteins were mixed according the. The mixture of the proteins were analysed using quantitative mass spectrometry (qMS), allowing a direct comparison of proteins bound to the Sleman/78 3’UTR, ∆30 3’UTR, ∆30/31 3’UTR and NS2a (control RNA). Protein abundance was log\u003csub\u003e2\u003c/sub\u003e. transformed and normalised. Differentially bound proteins were determined using a two-tailed t-test. The data was further filtered for only cytoplasmic proteins by excluding those localized to the nuclear bodies, nucleoplasm, nuclear speckles, nucleoli, nucleolar fibrillar centre, mitotic chromosome, and nucleus based on the list of proteins and their subcellular localization obtained from the UniProt database. Hierarchical clustering was conducted using the ComplexHeatmap R package\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch2\u003eRNA immunoprecipitation (IP)\u003c/h2\u003e\u003cp\u003ePlasmids encoding the 3' UTR from Sleman/78, ∆30, and ∆30/31 was transfected into Huh-7 cells. At 4 hours post-infection (hpi), RNA IP was performed using the MAGNA-RIP kit (Merck Millipore, USA), following the manufacturer's instructions. The levels of RNA in immunoprecipitates (IP) obtained using either anti-RPL11 (Santa Cruz Biotechnology, sc-293224), anti-RL17 (Santa Cruz Biotechnology, sc-515904), anti-RL28 (MyBiosource, MBS4508582), anti-EIF3D (Santa Cruz Biotechnology, sc-271515), or anti-EIF3K (Santa Cruz Biotechnology, sc-81262) antibodies were quantified by RT-qPCR and normalized to GAPDH mRNA levels. Immunoprecipitated proteins and RNA were analysed by Western blotting and RT-qPCR, respectively. Fold enrichment (FE) was calculated using the formula: FE = 2\u003csup\u003e(-ΔΔCt)\u003c/sup\u003e, where -ΔΔCt = -[(Ct_\u003csub\u003eRNA,IP\u003c/sub\u003e - Ct_\u003csub\u003eGAPDH,IP\u003c/sub\u003e) - (Ct_\u003csub\u003eRNA,control IP\u003c/sub\u003e - Ct_\u003csub\u003eGAPDH,control IP\u003c/sub\u003e)].\u003c/p\u003e\u003ch2\u003eEIF3D silencing through siRNA\u003c/h2\u003e\u003cp\u003eHuh-7 cells cultured in 24-well plates were transfected with either control small-interfering RNA (siCtrl) or Silencer Pre-designed siRNA targeting EIF3D (siRNA EIF3D) (ID:13732; Thermo Fisher Scientific) using the Lipofectamine RNAiMax reagent (Invitrogen), following the manufacturer’s protocol. Forty-eight hours post-transfection, the cells were infected with Sleman/78, ∆30, and ∆30/31 at a MOI of 5. The cells were harvested at 2, 4, 6, 12 and 24 h post-infection. Transfection efficiency was determined by Western blot analysis using anti-EIF3D (1:5000; Santa Cruz Biotechnology, sc-271515) with anti-GAPDH (1:5000; Cell Signaling Technology, 14C10) as a housekeeping protein for detection.\u003c/p\u003e\u003ch2\u003eeGFP reporter constructs\u003c/h2\u003e\u003cp\u003eThe 5′ and 3′ UTRs of vaccine strains from TV003 were used, with the exception that DENV-2 was replaced by Tonga/74 (GenBank: AY744148) and Tonga/74∆30 (GenBank: AY744149). The constructs were designed as shown in \u003cb\u003eSupplementary Fig.\u0026nbsp;6A\u003c/b\u003e and chemically synthesized in plasmids. Fragments from the synthesized plasmids were amplified using primers listed in \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e with NEB Q5 Hot-Start High-Fidelity Master Mix (New England BioLabs). PCR products were gel-purified using the QIAquick Gel Extraction Kit (Qiagen) according to the manufacturer’s instructions. Each gel-purified fragment (0.1 pmol) was assembled with the pUC19 vector containing the CMV promoter and SV40-pA using the NEBuilder HiFi Assembly Kit (New England BioLabs) at 50°C for 1 hour. The assembled plasmids were transformed into DH5α \u003cem\u003eE. coli\u003c/em\u003e (New England BioLabs), and the presence of the insert was confirmed by Sanger sequencing. Approximately 50 ng of plasmid was transfected into HEK293T cells in a 24-well plate using Lipofectamine 2000 (Invitrogen) and Opti-MEM (Gibco). After 24 hours, HEK293T cells were harvested, washed with 500 µL of cold PBS, and fixed in 2% paraformaldehyde (PFA) on ice for 10 minutes. Following fixation, cells were washed with PBS, resuspended in FACS buffer, and sorted using a FACSAria cell sorter (BD Biosciences).\u003c/p\u003e\u003ch2\u003eStatistical methods\u003c/h2\u003e\u003cp\u003eThe graphs were plotted using GraphPad Prism software. As indicated, one-way ANOVA was used to determine significant differences (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). Error bars in the graphs represent the standard deviation, with * indicating \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, ** indicating \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, *** indicating \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, and # indicating \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eE.E.O has served as chair and member of scientific advisory boards on dengue for Takeda, as well as in various advisory capacities for Sanofi Pasteur, Merck, Johnson \u0026amp; Johnson, and Novartis on dengue vaccines and therapeutics.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e \u003cp\u003eE.E.O., K.K. and E.F. conceptualized the study. K.K., H.C.T., X.Y. and W.C.N. conducted the experiments. K.K. and E.F. designed and carried out the SILAC-qMS. K. K., J.SG.O., W.C.N., K.R.C., E.F. and E.E.O analysed the data.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eWe thank Tanamas Siriphanitchakorn and Clara Koh for helpful discussions and technical support. This study was supported by the Singapore Translational Research Award from the National Medical Research Council to E.E.O (MOH-001271-00), and the National Research Foundation (NRF-CRP27-2021RS-0001).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMessina JP et al (2019) The current and future global distribution and population at risk of dengue. Nat Microbiol 4:1508\u0026ndash;1515\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVuong NL et al (2021) Higher Plasma Viremia in the Febrile Phase Is Associated With Adverse Dengue Outcomes Irrespective of Infecting Serotype or Host Immune Status: An Analysis of 5642 Vietnamese Cases. Clin Infect Dis 72:e1074\u0026ndash;e1083\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVuong NL et al (2024) Dengue viremia kinetics and effects on platelet count and clinical outcomes: An analysis of 2340 patients from Vietnam. Elife 13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilder-Smith A, Ooi EE, Horstick O, Wills B (2019) Dengue Lancet 393:350\u0026ndash;363\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOoi EE, Kalimuddin S (2023) Insights into dengue immunity from vaccine trials. Sci Transl Med 15:eadh3067\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKall\u0026aacute;s EG et al (2024) Live, Attenuated, Tetravalent Butantan-Dengue Vaccine in Children and Adults. N Engl J Med 390:397\u0026ndash;408\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNogueira ML et al (2024) Efficacy and safety of Butantan-DV in participants aged 2\u0026ndash;59 years through an extended follow-up: results from a double-blind, randomised, placebo-controlled, phase 3, multicentre trial in Brazil. Lancet Infect Dis\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMen R, Bray M, Clark D, Chanock RM, Lai CJ (1996) Dengue type 4 virus mutants containing deletions in the 3' noncoding region of the RNA genome: analysis of growth restriction in cell culture and altered viremia pattern and immunogenicity in rhesus monkeys. J Virol 70:3930\u0026ndash;3937\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitehead SS, Falgout B, Hanley KA, Blaney JE Jr., Markoff L, Murphy BR (2003) A live, attenuated dengue virus type 1 vaccine candidate with a 30-nucleotide deletion in the 3' untranslated region is highly attenuated and immunogenic in monkeys. J Virol 77:1653\u0026ndash;1657\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlaney JE Jr., Hanson CT, Firestone CY, Hanley KA, Murphy BR, Whitehead SS (2004) Genetically modified, live attenuated dengue virus type 3 vaccine candidates. Am J Trop Med Hyg 71:811\u0026ndash;821\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePierce KK et al (2024) Evaluation of a new dengue 3 controlled human infection model for use in the evaluation of candidate dengue vaccines. medRxiv\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlaney JE Jr. et al (2008) Dengue virus type 3 vaccine candidates generated by introduction of deletions in the 3' untranslated region (3'-UTR) or by exchange of the DENV-3 3'-UTR with that of DENV-4. Vaccine 26:817\u0026ndash;828\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNg WC, Soto-Acosta R, Bradrick SS, Garcia-Blanco MA, Ooi EE (2017) The 5' and 3' Untranslated Regions of the Flaviviral Genome. \u003cem\u003eViruses\u003c/em\u003e 9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinol E, Ooi EE (2019) Evolution of Subgenomic RNA Shapes Dengue Virus Adaptation and Epidemiological Fitness. iScience 16:94\u0026ndash;105\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirkpatrick BD et al (2016) The live attenuated dengue vaccine TV003 elicits complete protection against dengue in a human challenge model. Sci Transl Med 8:330ra336\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManokaran G et al (2015) Dengue subgenomic RNA binds TRIM25 to inhibit interferon expression for epidemiological fitness. Science 350:217\u0026ndash;221\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee AS, Kranzusch PJ, Doudna JA, Cate JH (2016) eIF3d is an mRNA cap-binding protein that is required for specialized translation initiation. Nature 536:96\u0026ndash;99\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlvarez DE, Filomatori CV, Gamarnik AV (2008) Functional analysis of dengue virus cyclization sequences located at the 5' and 3'UTRs. Virology 375:223\u0026ndash;235\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFilomatori CV, Lodeiro MF, Alvarez DE, Samsa MM, Pietrasanta L, Gamarnik AV (2006) A 5' RNA element promotes dengue virus RNA synthesis on a circular genome. Genes Dev 20:2238\u0026ndash;2249\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillordo SM, Gamarnik AV (2009) Genome cyclization as strategy for flavivirus RNA replication. Virus Res 139:230\u0026ndash;239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurbin AP (2020) Historical discourse on the development of the live attenuated tetravalent dengue vaccine candidate TV003/TV005. Curr Opin Virol 43:79\u0026ndash;87\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrifoni A et al (2017) Patterns of Cellular Immunity Associated with Experimental Infection with rDEN2Delta30 (Tonga/74) Support Its Suitability as a Human Dengue Virus Challenge Strain. J Virol 91\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlaney JE Jr., Hanson CT, Hanley KA, Murphy BR, Whitehead SS (2004) Vaccine candidates derived from a novel infectious cDNA clone of an American genotype dengue virus type 2. BMC Infect Dis 4:39\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMandl CW, Holzmann H, Kunz C, Heinz FX (1993) Complete genomic sequence of Powassan virus: evaluation of genetic elements in tick-borne versus mosquito-borne flaviviruses. Virology 194:173\u0026ndash;184\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi PY, Brinton MA, Veal JM, Zhong YY, Wilson WD (1996) Evidence for the existence of a pseudoknot structure at the 3' terminus of the flavivirus genomic RNA. Biochemistry 35:4222\u0026ndash;4230\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProutski V, Gritsun TS, Gould EA, Holmes EC (1999) Biological consequences of deletions within the 3'-untranslated region of flaviviruses may be due to rearrangements of RNA secondary structure. Virus Res 64:107\u0026ndash;123\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParanjape SM, Harris E (2007) Y box-binding protein-1 binds to the dengue virus 3'-untranslated region and mediates antiviral effects. J Biol Chem 282:30497\u0026ndash;30508\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiu WW, Kinney RM, Dreher TW (2005) Control of translation by the 5'- and 3'-terminal regions of the dengue virus genome. J Virol 79:8303\u0026ndash;8315\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamos-Lorente SE, Berzal-Herranz B, Romero-Lopez C, Berzal-Herranz A (2024) Recruitment of the 40S ribosomal subunit by the West Nile virus 3' UTR promotes the cross-talk between the viral genomic ends for translation regulation. Virus Res 343:199340\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa S, Liu JY, Zhang JT (2023) eIF3d: A driver of noncanonical cap-dependent translation of specific mRNAs and a trigger of biological/pathological processes. J Biol Chem 299:104658\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee AS, Kranzusch PJ, Cate JH (2015) eIF3 targets cell-proliferation messenger RNAs for translational activation or repression. Nature 522:111\u0026ndash;114\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu CY, Hsu YW, Liao CL, Lin YL (2006) Flavivirus infection activates the XBP1 pathway of the unfolded protein response to cope with endoplasmic reticulum stress. J Virol 80:11868\u0026ndash;11880\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanford TJ, Mears HV, Fajardo T, Locker N, Sweeney TR (2019) Circularization of flavivirus genomic RNA inhibits de novo translation initiation. Nucleic Acids Res 47:9789\u0026ndash;9802\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozman B, Fisher T, Stern-Ginossar N (2023) Translation-A tug of war during viral infection. Mol Cell 83:481\u0026ndash;495\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang F et al (2017) EIF3D promotes gallbladder cancer development by stabilizing GRK2 kinase and activating PI3K-AKT signaling pathway. Cell Death Dis 8:e2868\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarchingo JM, Cantrell DA (2022) Protein synthesis, degradation, and energy metabolism in T cell immunity. Cell Mol Immunol 19:303\u0026ndash;315\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu T, Fernandez-Sesma A (2020) Innate Immune DNA Sensing of Flaviviruses. \u003cem\u003eViruses\u003c/em\u003e 12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHinnebusch AG, Ivanov IP, Sonenberg N (2016) Translational control by 5'-untranslated regions of eukaryotic mRNAs. Science 352:1413\u0026ndash;1416\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillordo SM, Alvarez DE, Gamarnik AV (2010) A balance between circular and linear forms of the dengue virus genome is crucial for viral replication. RNA 16:2325\u0026ndash;2335\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanborn MA et al (2020) Analysis of cell-associated DENV RNA by oligo(dT) primed 5' capture scRNAseq. Sci Rep 10:9047\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitehead SS, Hanley KA, Blaney JE Jr., Gilmore LE, Elkins WR, Murphy BR (2003) Substitution of the structural genes of dengue virus type 4 with those of type 2 results in chimeric vaccine candidates which are attenuated for mosquitoes, mice, and rhesus monkeys. Vaccine 21:4307\u0026ndash;4316\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan KR et al (2011) Ligation of Fc gamma receptor IIB inhibits antibody-dependent enhancement of dengue virus infection. Proc Natl Acad Sci U S A 108:12479\u0026ndash;12484\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu RS, Chan KR, Tan HC, Chow A, Allen JC Jr., Ooi EE (2012) Neutralization of dengue virus in the presence of Fc receptor-mediated phagocytosis distinguishes serotype-specific from cross-neutralizing antibodies. Antiviral Res 96:340\u0026ndash;343\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWard AM et al (2011) Quantitative mass spectrometry of DENV-2 RNA-interacting proteins reveals that the DEAD-box RNA helicase DDX6 binds the DB1 and DB2 3' UTR structures. RNA Biol 8:1173\u0026ndash;1186\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWard AM, Gunaratne J, Garcia-Blanco MA (2014) Identification of dengue RNA binding proteins using RNA chromatography and quantitative mass spectrometry. Methods Mol Biol 1138:253\u0026ndash;270\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu Z (2022) Complex heatmap visualization. Imeta 1:e43\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Dengue virus (DENV), attenuation, 3’ untranslated region (3’UTR), translation, eukaryotic initiation factors","lastPublishedDoi":"10.21203/rs.3.rs-6513699/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6513699/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eViral antigenic burden drives the inflammation-driven pathophysiology of dengue in humans. Nonetheless, the control of dengue virus (DENV) antigen expression for pathogenicity in humans remain uncertain. Herein, we examined a clinical DENV-3 isolate (Sleman/78), along with its partially and fully attenuated derivatives through 30- (Δ30) as well as 30- and 31-nucleotide deletions (Δ30/31), respectively, in the 3\u0026rsquo; untranslated region (3\u0026prime;UTR) of the RNA genome; the partially and fully attenuated phenotypes of these derivatives were demonstrated in clinical trials. We found, using infectious clone technology, protein and RNA pulldown approaches, that the wild-type 3\u0026rsquo;UTR bound host translation proteins, including non-canonical eukaryotic initiation factor-3D (EIF3D) to support viral protein expression. Both Δ30/31 mutation and EIF3D silencing attenuated viral protein expression and hence replication of Sleman/78. As DENV genome is cyclized through 5\u0026rsquo; and 3\u0026rsquo;UTR interactions, our findings the role of 3\u0026prime;UTR in regulating translation for infection and pathogenesis in humans.\u003c/p\u003e","manuscriptTitle":"Dengue virus 3’ untranslated region regulates RNA genome translation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 11:36:40","doi":"10.21203/rs.3.rs-6513699/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fa43a98a-177b-440f-8571-30fa6b456682","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48154001,"name":"Biological sciences/Microbiology/Virology/Dengue virus"},{"id":48154002,"name":"Biological sciences/Microbiology/Vaccines/Live attenuated vaccines"}],"tags":[],"updatedAt":"2025-06-16T13:16:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 11:36:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6513699","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6513699","identity":"rs-6513699","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}