The 3‘Untranslated region is a critical determinant of Getah virus replication, pathogenesis, and vector competence

ABSTRACT

Getah virus (GETV), a mosquito-borne arbovirus of the Alphavirus genus, poses an emerging threat to livestock economies and public health, underscored by its expanding host range and association with recent outbreaks of heightened virulence. While the functional significance of the 3’ untranslated region (3’UTR) in alphavirus biology is recognized, its specific role in GETV remained undefined. Herein, we elucidate the virological functions of the GETV 3’UTR through a reverse genetics approach, generating a panel of viruses with targeted deletions. We demonstrate that the GETV 3’UTR is remarkably plastic, tolerating a consecutive deletion of up to 310 nucleotides while remaining viable. Deletion of conserved repeat sequence elements (RSEs) induced a cell-type-specific replication deficiency in vitro and significantly attenuated virulence in a murine model. A comprehensive deletion mutant (rGETV-KO310) exhibited further impaired replication kinetics in vitro and was profoundly attenuated in vivo, eliciting only transient morbidity with no mortality in both neonatal and weaned mice. Furthermore, this mutant displayed a significant defect in early colonization within mosquito vectors, indicating a role in vector competence. Comparative transcriptomic profiling of knee joints revealed that attenuation correlates with the altered modulation of critical host immune responses, notably the interferon and MAPK signaling pathways. Collectively, these findings establish the GETV 3’UTR as a pivotal regulator of viral fitness, pathogenesis, and transmission. This work provides a foundational rationale for the strategic development of live-attenuated vaccine candidates based on targeted 3’UTR attenuation.

HIGHLIGHTS

• The GETV 3’UTR exhibits significant plasticity, tolerating a large-scale deletion of 310 nucleotides while remaining replication-competent.

• Deletion of conserved elements within the 3’UTR attenuates GETV virulence in a murine model and causes cell-type-specific replication defects in vitro.

• A comprehensive 3’UTR mutant is highly attenuated, causing only transient morbidity with no mortality, and is impaired in its ability to colonize mosquito vectors.

• Viral attenuation is linked to the 3’UTR’s role in modulating host immune responses, specifically the interferon and MAPK signaling pathways.

Introduction

Getah virus (GETV), a member of the genus Alphavirus within the family Togaviridae, was first isolated from Culex mosquitoes in Malaysia in 1955 [1]. Since its initial identification, the range of mosquito species capable of transmitting GETV has expanded considerably, concurrent with a broadening geographical distribution and host diversity [2]. GETV infection in horses manifests as fever, dermal rash, and hind-limb edema [3,4]. Although multiple outbreaks have been documented globally, no equine fatalities have been reported. In contrast, infection in sows can result in reproductive impairments, including irregular estrus cycles, abortions, and stillbirths, while neonatal piglets may exhibit severe clinical signs such as diarrhea, hypothermia, ataxia, and mortality [5]. Recent years have witnessed several GETV outbreaks in Chinese swine populations, inflicting substantial economic losses on the pig industry, with the emergence of a highly virulent variant associated with increased mortality in piglets [6–10]. In 2019, GETV was isolated from febrile cattle in China, confirming its capacity to infect bovines [11]. The virus also induces fatal infections in wildlife, notably blue foxes [12], and natural infections have been identified in wild boars [13], red pandas [14], pangolins [15], and farmed raccoon dogs [16]. Serological evidence has demonstrated the presence of GETV-specific antibodies in humans, domestic animals (including pigs, horses, cattle, goats, dogs, and rabbits), as well as in kangaroos, chickens, and wild birds across numerous countries in Asia, Europe, and Oceania [17,18]. Of particular note, significantly elevated antibody titers have been observed in febrile human patients compared to healthy individuals, suggesting a potential etiological role of GETV in human disease [19].

The genomic architecture of GETV conforms to the typical organization of alphaviruses. Its single-stranded positive-sense RNA genome comprises approximately 11,690 nucleotides, featuring two open reading frames (ORFs) flanked by 5’ and 3’ untranslated regions (UTRs) [20]. The ORFs encode four non-structural proteins (nsP1–nsP4), which facilitate viral genome replication and modulate host gene expression and immune responses, and five structural proteins (C, E3, E2, 6K, and E1), which constitute the viral particle [21]. The capsid protein (C) encapsidates the viral RNA to form the nucleocapsid, while the envelope glycoproteins E1 and E2 form heterodimeric spikes on the virion surface. E2 mediates receptor binding and cell attachment, and E1 drives membrane fusion during viral entry [22]. The UTRs contain essential regulatory elements: the 5’UTR harbors core promoter sequences required for the initiation of negative-strand RNA synthesis, whereas the 3’UTR contains cis-acting RNA elements critical for positive-strand synthesis and replication [23].

Alphaviruses 3’UTR exhibits considerable plasticity, with frequent insertions, deletions, and duplications of sequences occurring both within and between species [24]. Despite this variability, most alphavirus 3’UTRs share a conserved core structure, characterized by short repeat sequence elements (RSEs) and a highly conserved sequence element (CSE) of approximately 19–24 nucleotides situated proximal to the poly(A) tail [25,26]. These elements collectively regulate viral replication, host adaptation, and pathogenicity through their specific sequences and structural conformations [27]. For instance, in Chikungunya virus (CHIKV), the RSEs – often designated as direct repeats (DRs) – are shaped through recombination and mutation events, yielding lineage-specific 3’UTR architectures. Variations in the copy number and organization of these repeats among CHIKV lineages influence viral fitness; deletion of DRs diminishes infectivity in mosquito vectors but enhances replication in mammalian hosts. Consequently, CHIKV undergoes dynamic gain and loss of DR copies during host switching to optimize fitness in both vertebrates and mosquitoes [24,28]. Similarly, removal of RSEs in Eastern equine encephalitis virus (EEEV) impairs replication in mosquito cells without affecting mammalian cell replication [29]. Sindbis virus (SINV) can tolerate extensive deletions within the 3’UTR, including large segments encompassing RSEs, indicating that these elements are dispensable for in vitro growth in certain contexts. Nevertheless, their absence often results in cell-type-specific replication deficiencies [30,31].

To date, the virological functions of the GETV 3’UTR have not been systematically investigated. In this study, we generated a series of 3’UTR deletion mutants to explore its role in viral replication, pathogenicity, and transmission. We found that deleting repeat sequences in the 3’UTR reduces viral replication in a cell-type-dependent manner and attenuates virulence in mice. A mutant with a 310-nucleotide deletion (rGETV-KO310) showed impaired replication in vitro, reduced virulence in vivo, and delayed early replication in mosquitoes. Transcriptomic analysis of infected mouse joints indicated that attenuation correlates with altered immune responses, including interferon and MAPK signaling. These results highlight the importance of the GETV 3’UTR in viral replication and pathogenesis, supporting its use as a target for developing live-attenuated vaccines.

Materials and methods

Cells, antibodies, and plasmids

This study utilized two mammalian cell lines, BHK-21 (ATCC CCL-10) and Vero (ATCC CCL-81), which were preserved in our laboratory [32,33], as well as the Aedes albopictus C6/36 mosquito cell line (ATCC CRL-1660), kindly provided by Prof. Li Yiping (Sun Yat-sen University, China). BHK-21 and Vero cells were maintained at 37°C under 5% CO₂ in Dulbecco’s Modified Eagle Medium (DMEM, Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, Gibco) and penicillin-streptomycin. C6/36 cells were cultured at 28°C, under 5% CO₂ in the same medium, further supplemented with 1× non-essential amino acids (NEAA). A monoclonal antibody specific to the GETV E1 protein, generated and preserved in a prior study, was used for detection [34]. An Alexa Fluor® 568-conjugated goat anti-mouse IgG secondary antibody (Invitrogen) was employed for immunofluorescence assays. The full-length infectious cDNA clone of GETV strain GX201808 (pGETV-GX), constructed previously, served as the backbone for mutant generation [32].

GETV 3’UTR sequence analysis

Viral sequences corresponding to GETV were retrieved from the NCBI database, and the 3’UTR regions were extracted. Multiple sequence alignment and characterization of the GETV 3’UTR were conducted using MegAlign software (DNASTAR, Madison, WI).

Secondary structure prediction

Secondary structure predictions for the RNA sequences were performed using the RNAfold WebServer (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi), implementing the ViennaRNA Package 2.0. Predictions were generated under the minimum free energy (MFE) model.

Plasmid construction

To construct a series of 3’UTR deletion mutants, specific primers were designed (Supplemental Table S1). Deletions were introduced via splicing by overlap extension PCR (SOE-PCR). Briefly, two fragments flanking the target deletion were amplified and subsequently fused into a single fragment by SOE-PCR. The resulting amplicon was cloned into the pGETV-GX infectious clone using the restriction enzymes BstBI and MluI, yielding full-length mutant clones with defined deletions in the 3’UTR.

DNA transfection and virus rescue

Mutant plasmids and the parental pGETV-GX clone (2 μg each) were individually transfected into BHK-21 cells at approximately 80% confluence in 6-well plates, using Lipofectamine 2000 (Invitrogen, Waltham, MA) according to the manufacturer’s protocol. Cells were incubated at 37°C with 5% CO₂. At 48 hours post-transfection (hpt), supernatants were harvested and designated as passage 0 (P0) virus stocks. To amplify viruses, P0 stocks were used to infect fresh BHK-21 or C6/36 cells. When cytopathic effect (CPE) reached approximately 90%, supernatants were collected as passage 1 (P1). Subsequent passages (P2–P3) were performed by infecting new cells at a 1:1000 dilution. Viral RNA was extracted from P3 stocks, and the 3’UTR region was verified by Sanger sequencing to confirm genetic stability.

Virus titration

Viral titers were determined by the 50% tissue culture infectious dose (TCID₅₀) assay. Ten-fold serial dilutions of virus were prepared in DMEM supplemented with 2% fetal bovine serum and used to infect monolayer cells in 96-well plates (100 μL/well), with six replicates per dilution. After 1 h adsorption at 37°C under 5% CO₂, the inoculum was replaced with fresh DMEM supplemented with 2% fetal bovine serum. Plates were incubated for 3–5 days, and CPE was recorded. The TCID₅₀ was calculated using the Reed – Muench method.

Immunofluorescence assay (IFA)

BHK-21 cells were grown in 12-well plates to ~90% confluence and then infected with parental or mutant viruses at a multiplicity of infection (MOI) of 0.1. Uninfected cells served as negative controls. At 18 hours post-infection (hpi), cells were fixed with 4% paraformaldehyde for 30 min, permeabilized with 0.1% Triton X-100, and blocked with 5% bovine serum albumin (BSA) in PBS for 30 min at 37°C. Cells were incubated with anti-GETV E1 monoclonal antibody (mAb) at 37°C for 2 h, followed by incubation with Alexa Fluor® 568-conjugated secondary antibody for 1 h at 37°C. Nuclei were stained with DAPI (1:1000; Solarbio). Images were acquired using an EVOS M5000 fluorescence microscope.

Plaque assay

Confluent Vero cells in 6-well plates were infected with 500 μL of diluted virus per well. After 1 h adsorption, the inoculum was aspirated, and cells were overlaid with 2 mL of a 1:1 mixture of 2× DMEM and 2% low melting agarose (Cambrex, Rockland, ME, USA). Following 72 h incubation at 37°C, cells were fixed with 4% paraformaldehyde for 6 h, stained with crystal violet, and plaques were imaged.

Viral growth kinetics

Multi-step growth curves were generated by infecting BHK-21, Vero, or C6/36 cells in 12-well plates (~5 × 10⁵ cells/well) at an MOI of 0.01 using P3 virus. After 1 h adsorption at 37°C (mammalian cells) or 28°C (C6/36 cells), cells were washed with PBS and maintained in DMEM supplemented with 2% fetal bovine serum. Supernatants were collected at indicated time points, and viral titers were determined by TCID₅₀ assay. Growth curves were plotted as log₁₀ TCID₅₀/mL versus time.

Animal experiments

All animal procedures were approved by the Animal Ethics Committee of Guangxi University (Protocol No. GXU2022-288). All procedures involving live viruses and animal experiments were performed in compliance with institutional biosafety regulations under Biosafety Level 2 (BSL-2) containment conditions. Specific pathogen-free (SPF) 8-week-old ICR mice were obtained from Nanning Yancheng Biotechnology Co., Ltd. Dams were individually housed, and pups were randomly redistributed into experimental groups post-delivery using a computer-based random order generator. Animals were maintained under BSL-2 conditions with ad libitum access to food and water.

To assess pathogenicity, 3- and 10-day-old ICR mice were subcutaneously inoculated with 10⁴ TCID₅₀ (equivalent to 14,000 PFU) of P3 virus or DMEM as control (n = 10 per group). Mice were monitored for 14 days for weight loss and clinical signs, scored as follows: 0, healthy; 1, lethargy/dyspnea; 2, limb paralysis/diarrhea; 3, moribund or euthanized. Each suckling mouse served as an independent experimental unit. The second author performed randomization and the third author conducted data analysis, both blinded to group assignment, while the animal care team (first, fourth, fifth, and sixth authors) remained unblinded. For viral load quantification, mice (n = 15) were humanely euthanized via cervical dislocation 1, 3, and 5 dpi (n = 5 per time point). Blood, lung, knee joints, and brain tissues were collected, homogenized in PBS, and titrated by TCID₅₀ assay.

Mosquito infection studies

To investigate the transmission potential of the 3’UTR in mosquitoes, Aedes albopictus mosquitoes were reared at 28°C and 80% relative humidity under a 12 h:12 h light:dark cycle, with ad libitum access to 10% sucrose. Female mosquitoes (5–7 days old) were starved for 24 h prior to infection. For each experimental group, 100 female mosquitoes were allowed to feed using sponge feeders. The infection and control groups received mixtures of virus or DMEM, respectively, with defibrinated sheep blood and 50% sucrose (1:1:1). From these, 50 fully engorged females were selected for each group and maintained under standard conditions. At 5, 10, and 15 days post-infection, mosquitoes were collected in pools of three individuals, with five biological replicates per time point. The samples were anesthetized at 4°C, homogenized, and total RNA was extracted using TRIzol (Invitrogen). Viral RNA loads were quantified by RT-qPCR. The group assignment, molecular analysis, and personnel responsibilities in the mosquito experiments followed the same blinding protocol as described for the mouse study.

RNA-seq analysis

Knee joint tissues were harvested from 10-day-old mice at 3 dpi. Total RNA was extracted using TRIzol. Ribosomal RNA was depleted, and mRNA was fragmented and reverse-transcribed into double-stranded cDNA. Sequencing libraries were constructed and sequenced by Tsingke Biotechnology Co., Ltd. High-quality reads were aligned to the mouse reference genome using Hisat2. Gene expression levels were quantified as FPKM. Differential expression analysis was performed using DESeq2 (v1.26.0), with significance defined as |fold change| ≥1.2 and p < 0.05.

Statistical analysis

Data analysis was conducted using GraphPad Prism 8.0.2. Data are presented as mean ± standard deviation (SD). Comparisons were made against the parental virus – inoculated group. Differences between groups were assessed using unpaired Student’s t-test or two-way ANOVA with multiple comparisons. A p-value < 0.05 was considered statistically significant.

Statement of adherence to ARRIVE guidelines

All experimental procedures in this study were conducted and reported in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines (https://arriveguidelines.org/arrive-guidelines) to ensure transparency, reproducibility, and ethical rigor in the design, execution, and reporting of the animal experiments. A completed ARRIVE checklist has been provided as supplementary material.

Results

GETV 3’UTR exhibits considerable plasticity

A subset of representative GETV strains was selected based on downloaded sequences for comparative alignment. As illustrated in Figure 1(A), almost all analyzed strains contain conserved sequence element (CSE, green-background region) and three repeat sequence elements (RSEs, yellow-background region). Strains belonging to three distinct groups exhibit a 9-nt deletion near the 5’ end of the 3’UTR. Two strains within Group II possess a 10-nt insertion at the 5’ end, while a recently circulating Group III strain from Guangdong Province, China (GDHYLC23), contains a 32-nt repeat insertion in the same region. Additionally, strains YN08 and YN12043 show 5’ terminal deletions of 44 nt and 36 nt, respectively. Numerous single-nucleotide polymorphisms, including substitutions, deletions, and insertions, are distributed throughout the 3’UTR across different strains. Secondary structure prediction of the GETV-GX201808 3’UTR revealed multiple stem-loop structures (Supplemental Figure S1).

Figure 1.

Schematic representation of GETV 3’UTR sequence alignment and deletion mutants. (A) Annotated features of the GETV 3’UTR. Genomic sequences of GETV strains were retrieved from the NCBI database. Representative isolates were selected, and multiple sequence alignment was performed using megalign software. Yellow rectangles denote three tandem 21-nucleotide (nt) repeats within the GETV 3’UTR. The green rectangle highlights a conserved 19-nt motif common among alphaviruses. (B) Schematic depiction of the 3’UTR architecture for engineered mutant viruses. Deleted sequences are indicated by dashed lines. The specific genomic positions corresponding to the repeat elements and conserved sequence element are annotated.

Deletion of GETV 3’UTR repeat sequences does not affect virus rescue

A series of mutants with deletions of different repeat sequences were constructed according to the schematic in Figure 1(B). Mutant plasmids were transfected into BHK-21 cells to rescue recombinant viruses. At 48 hpt, supernatants were harvested, clarified by centrifugation, and used to infect fresh BHK-21 cells. Pronounced CPE was evident within 24 hpi (Figure 2(A)). Successful virus rescue was confirmed by IFA using a mAb specific to the E1 protein. Immunostaining demonstrated robust E1 protein expression in cells infected with the rescued viruses (Figure 2(B)). The corresponding mutants were designated rGETV-KO1, rGETV-KO2, and rGETV-KO3. To further validate the genetic integrity of the rescued viruses, viral RNA was extracted from passage 3 (P3) stocks, and the 3’UTR region was amplified and sequenced. Amplification fragments derived from mutant viruses were visibly shorter than those from the parental virus, and Sanger sequencing confirmed the intended deletions (Figure 2(C)).

Figure 2.

Replication kinetics of 3’UTR repeat deletion mutants in vitro. (A) Cytopathic effects elicited by mutant viruses in infected BHK-21 cells. (B) Verification of E1 protein expression in rescued mutant strains via indirect immunofluorescence assay (IFA). BHK-21 cells were infected with parental or mutant viruses at a multiplicity of infection (MOI) of 0.1. Cells were fixed at 18 hours post-infection (hpi) and probed with a GETV E1-specific monoclonal antibody (mAb). (C) Confirmation of 3’UTR length in mutant viruses by RT-PCR. Viral RNA was extracted from third-passage (P3) viral stocks, reverse-transcribed into cDNA, and amplified using primers flanking the 3’UTR region. Amplified products were resolved by agarose gel electrophoresis. (D) Comparative in vitro growth kinetics of mutant and parental viruses. BHK-21, Vero and C6/36 cells were infected with P3 virus at an MOI of 0.01. Culture supernatants were harvested at the indicated time points, and viral titers (log₁₀ TCID₅₀/mL) were quantified to establish multi-step growth curves. Data represent mean ± SD and were analyzed by unpaired Student’s t-test (asterisks shown in the figure, comparing each mutant to the parental virus at each time point) and area under the curve (AUC) analysis (asterisks shown in the reference, right, comparing each mutant to the parental virus), asterisks indicate significant differences (*p < 0.05, **p < 0.01, ***p < 0.001). (E) Plaque morphology exhibited by mutant and parental viruses in Vero cell monolayers. (F) Quantitative analysis of plaque diameters derived from panel. Data represent mean ± SD; data were analyzed by two-way ANOVA with multiple comparisons (asterisks indicate significant differences between each mutant and the parental virus). (*p < 0.05, **p < 0.01, ***p < 0.001).

To assess whether deletion of the 3’UTR RSEs affects viral replication, multi-step growth kinetics were evaluated in three cell lines at an MOI of 0.01 using P3 virus. In BHK-21 cells, all mutants exhibited replication kinetics comparable to the parental virus, with no statistically significant differences in viral titers at any time point. In contrast, in Vero and C6/36 cells, the triple-repeat deletion mutant (rGETV-KO3) demonstrated attenuated replication, achieving peak titers significantly lower than those of the parental virus (Figure 2(D)). The relative replication capacity of the rGETV-KO3 mutant compared to the parental virus exhibited dynamic differences, particularly evident in Vero cells. To further characterize this, area-under-the-curve (AUC) analysis was performed. In BHK-21 cells, both the rGETV-KO2 and rGETV-KO3 mutants showed overall replication levels higher than those of the parental virus. In Vero cells, only rGETV-KO2 demonstrated a greater total replication capacity compared with the parental virus. In C6/36 cells, the overall replication level of rGETV-KO3 remained lower than that of the parental virus. Plaque assays in Vero cells indicated that all mutants produced plaques similar in morphology to the parental virus, albeit with a modest reduction in diameter that did not reach statistical significance (Figure 2(E–F)).

Deletion of three repeat sequences in the GETV 3’UTR attenuates viral virulence

The pathogenicity of parental and mutant viruses was evaluated in both 3-day-old and 10-day-old mice. In suckling mice (3-day-old), all viruses induced clinical disease, manifesting as lethargy, hind limb weakness, dyspnea, and progressive emaciation; however, mice infected with rGETV-KO3 exhibited consistently lower clinical scores than those infected with the parental virus (Supplemental Figure S2A). Weight changes in mice infected with rGETV-KO1 and rGETV-KO2 were comparable to the parental virus group. Although rGETV-KO3-infected mice initially gained weight more rapidly, a decline was observed after 5 days post-infection (Supplemental Figure S2B). While all infected mice ultimately succumbed to infection, those inoculated with RSE deletion mutants showed prolonged survival, with rGETV-KO3-infected mice surviving the longest (Supplemental Figure S2C). Viral loads in tissues were quantified by TCID₅₀ assay at multiple time points. At 24 hpi, rGETV-KO3 titers in knee joints (1.5-log₁₀ reduction), brain (0.9-log₁₀ reduction), and lung (1-log₁₀ reduction) tissues were significantly reduced compared to the parental virus. By 48 hpi, significantly lower viral loads were observed only in the knee joints (1.1-log₁₀ reduction) and lung (2-log₁₀ reduction) of rGETV-KO3-infected mice. At 72 hpi, rGETV-KO1 replicated to higher titers in the knee joints (1.4-log₁₀ increase) and lung (0.6-log₁₀ increase) than the parental virus, whereas rGETV-KO3 titers in the lung (1.6-log₁₀ reduction) remained significantly lower (Supplemental Figure S2D).

In 10-day-old mice, those infected with rGETV-KO2 and rGETV-KO3 only exhibited transient clinical symptoms at 5–6 dpi, including progressive hindlimb weakness, reduced activity, and lethargy, which lasted for approximately 6 days (Figure 3(A)), along with a significantly higher survival rate. However, the weight gain rate of the infected groups remained lower than that of the control group (Figure 3(B)). In contrast, mice infected with rGETV-KO1 still showed considerable clinical severity and mortality, displaying marked hindlimb weakness, knee joints swelling, decreased mobility, lethargy, and delayed eye-opening, with 90% of the mice succumbing to viral infection (Figure 3(C)).

Figure 3.

Pathogenicity assessment of mutant and parental viruses in ten-day-old mice. Ten-day-old ICR mice were subcutaneously inoculated with 10⁴ TCID₅₀ (equivalent to 14,000 PFU) of parental virus, mutant viruses, or DMEM (control). Mice were monitored for 14 days post-infection (dpi). Viral titers were determined by TCID₅₀ assay on Vero cells. Based on parallel plaque assays (n = 3 independent titrations), a conversion factor of 1 TCID₅₀ ≈ 1.4 PFU was established for the virus stocks used. PFU equivalents are provided in parentheses. (A) Clinical scores. Data were analyzed by two-way ANOVA with multiple comparisons. Asterisks indicate significant differences between each mutant virus and the parental virus at the corresponding time points. (*p < 0.05, **p < 0.01, ***p < 0.001). (B) Body weight change. Data were analyzed by two-way ANOVA with multiple comparisons (asterisks shown in the figure) and AUC analysis (asterisks shown in the reference, right). Asterisks denote significant differences compared to the parental virus. (*p < 0.05, **p < 0.01, ***p < 0.001). (C) Survival rates. The MOCK and rGETV-KO2 survival curves overlap, rendering the MOCK curve indistinguishable. Asterisks indicate significant differences between each mutant virus and the parental virus (log-rank test, *p < 0.05, **p < 0.01, ***p < 0.001). (D) Viral replication kinetics in tissues of ten-day-old mice infected subcutaneously (collection time points: 1, 3, 5 dpi). Data were analyzed by two-way ANOVA with multiple comparisons. Asterisks indicate significant differences in viral titers between each mutant virus and the parental virus at each time point. (*p < 0.05, **p < 0.01, ***p < 0.001). Mock-infected mice had no detectable virus in any tissue; these data are not shown in the figure for clarity.

At 1 dpi, only mice infected with rGETV-KO3 had significantly lower viral loads in both blood (0.2-log₁₀ reduction) and knee joints (0.6-log₁₀ reduction) compared to those infected with the parental virus (Figure 3(D)). By 3 dpi, there were no significant differences in viral tissue loads among the groups (Figure 3(D)). However, at 5 dpi, viral loads in the lungs of mice infected with any of the three mutant viruses were significantly lower than in the parental virus group (Figure 3(D)), with 1.6-log₁₀, 1.4-log₁₀, and 1.3-log₁₀ reductions for rGETV-KO1, KO2, and KO3, respectively. These experimental results indicate that the deletion of the three repeated sequences in the GETV 3’UTR attenuates viral virulence.

The GETV 3’UTR tolerates a 310-nt deletion

To determine the maximal deletable region within the 3’UTR, a series of sequential stem-loop deletions were engineered based on predicted secondary structure, in addition to the RSE deletions. Mutant constructs were transfected into BHK-21 cells for virus recovery (Figure 1(B)). A viable virus was successfully rescued even with a 310-nt deletion (rGETV-KO310). In contrast, transfection with a mutant bearing a 320-nt deletion did not yield visible CPE, and IFA confirmed the absence of viral protein expression (Figure 4(A-B)). After three serial passages of the transfection supernatant, RT-PCR analysis of P3 RNA still failed to detect viral rescue, whereas other deletion mutants were successfully recovered and verified by amplicon size (Figure 4(C)).

Figure 4.

Impact of large-scale 3’UTR deletion on viral replicative fitness. (A) Cytopathic effects induced in BHK-21 cells following infection with mutant viruses. (B) Detection of viral E1 protein expression by indirect immunofluorescence assay (IFA). BHK-21 cells infected at an MOI of 0.1 were fixed at 18 hpi and stained with an anti-GETV E1 mAb. (C) Validation of 3’UTR integrity in rescued mutant viruses by RT-PCR. Viral RNA from P3 stocks was reverse-transcribed, and the 3’UTR region amplified. PCR products were visualized by agarose gel electrophoresis. (D) Multi-step growth kinetics comparison. BHK-21, Vero and C6/36 cells infected at MOI 0.01 were sampled at indicated intervals. Viral titers in supernatants (log₁₀ TCID₅₀/mL) were determined (mean ± SD). Data were analyzed by two-way ANOVA with multiple comparisons; asterisks indicate significant differences between each mutant virus and the parental virus at the corresponding time points (*p < 0.05, **p < 0.01, ***p < 0.001). (E) Plaque morphology of mutant and parental viruses in Vero cells. (F) Analysis of plaque diameters derived from panel. Data represent mean ± SD; data were analyzed by two-way ANOVA with multiple comparisons; asterisks indicate significant differences between each mutant virus and the parental virus. (*p < 0.05, **p < 0.01, ***p < 0.001).

The replication kinetics of mutants with extended 3’UTR deletions were assessed in three cell lines at an MOI of 0.01 using P3 virus (Figure 4(D)). In BHK-21 cells, all deletion mutants showed significantly impaired replication compared to the parental virus. In Vero cells, large fragment deletions resulted in reduced early replication capacity, decreased peak titers, and significantly smaller plaque sizes (Figure 4(E-F)). In C6/36 cells, rGETV-KO275 and rGETV-KO285 displayed enhanced early replication but ultimately reached lower titers than the parental virus. rGETV-KO290 and rGETV-KO310 exhibited replication comparable to the parental virus during early infection but significantly lower titers at later time points.

Extensive nucleotide deletions in the GETV 3’UTR attenuate viral virulence

To further elucidate the effect of large-scale consecutive nucleotide deletions within the 3’UTR on viral pathogenicity, challenge studies were conducted in 3-day-old mice. Infection with rGETV-KO275 and rGETV-KO285 induced severe clinical manifestations (Figure 5(A)), significantly suppressed weight gain (Figure 5(B)), and resulted in uniform mortality (Figure 5(C)). In contrast, when the 3’UTR deletion reached 290 nt or greater, infected mice displayed only transient clinical signs – including occasional, temporary hind-limb weakness at 4–5 days post-infection – followed by complete recovery and survival in all cases (Figure 5A–C).

Figure 5.

Comparative analysis of pathogenicity in mice infected subcutaneously. (A-C) Disease progression in three-day-old ICR mice subcutaneously inoculated with 10⁴ TCID₅₀ (equivalent to 14,000 PFU) virus or DMEM (control) and monitored for 14 days: (A) Clinical symptom scores. Data were analyzed by two-way ANOVA with multiple comparisons; asterisks indicate significant differences between each mutant virus and the parental virus at the corresponding time points (*p < 0.05, **p < 0.01, ***p < 0.001). (B) Body weight change relative to baseline. Data were analyzed by two-way ANOVA with multiple comparisons (asterisks shown in the figure) and AUC analysis (asterisks shown in the reference, right). Asterisks denote significant differences compared to the parental virus (*p < 0.05, **p < 0.01, ***p < 0.001). (C) Survival rates. The MOCK, rGETV-KO290, rGETV-KO300, and rGETV-KO310 survival curves overlap, rendering the rGETV-KO290, rGETV-KO300, and rGETV-KO310 curves indistinguishable. (log-rank test; asterisks indicate significant differences between each mutant virus and the parental virus. *p < 0.05, **p < 0.01, ***p < 0.001). (D) Viral load quantification in tissues of subcutaneously infected three-day-old mice. Samples (blood, knee joints, lung, brain) collected at 1, 2, 3 dpi. Titers shown as log₁₀ TCID₅₀/mL or g (mean ± SD; data were analyzed by two-way ANOVA with multiple comparisons; asterisks indicate significant differences in viral titers between each mutant virus and the parental virus at each time point. *p < 0.05, **p < 0.01, ***p < 0.001). (E-G) Disease progression in ten-day-old ICR mice infected subcutaneously: (E) Clinical scores. Data were analyzed by two-way ANOVA with multiple comparisons (*p < 0.05, **p < 0.01, ***p < 0.001). (F) Body weight change. Data were analyzed by two-way ANOVA with multiple comparisons (asterisks shown in the figure) and AUC analysis (asterisks shown in the reference, right). Asterisks denote significant differences compared to the parental virus (*p < 0.05, **p < 0.01, ***p < 0.001). (G) Survival rates. The MOCK and rGETV-KO310 survival curves overlap, rendering the rGETV-KO310 curve indistinguishable. (log-rank test; asterisks indicate significant differences between each mutant virus and the parental virus. *p < 0.05, **p < 0.01, ***p < 0.001). (H) Viral load quantification in tissues of subcutaneously infected ten-day-old mice (collection time points: 1, 3, 5 dpi). Data were analyzed by unpaired Student’s t-test. Asterisks indicate significant differences between each mutant virus and the parental virus at each time point (*p < 0.05, **p < 0.01, ***p < 0.001). No virus was detected in any tissue of mock-infected mice; these data are not plotted in the figure for clarity.

The replication levels of rGETV-KO275 and rGETV-KO285 in mouse tissues were comparable to those of the parental virus (Figure 5(D)). Notably, between 24 and 48 hpi, viral loads of mutants with deletions ≥ 290 nt were significantly reduced relative to the parental virus, with rGETV-KO310 exhibiting the most substantial attenuation. By 72 hpi, viral titers of these mutants remained lower across most tissues, although the difference in lung tissue did not reach statistical significance.

Since deletions exceeding 290 nt in the 3’UTR significantly attenuated virulence in 3-day-old mice, we only investigated the pathogenicity of the maximal 3’UTR deletion (rGETV-KO310) in 10-day-old mice. Consistent with observations in suckling mice, infection of 10-day-old mice with rGETV-KO310 did not produce detectable clinical symptoms, and weight gain curves closely paralleled those of the negative control group, with no mortality observed. However, mice infected with the parental virus developed clinical signs at 5 dpi, followed by slowed body weight gain, with only one survivor remaining at 9 dpi (Figure 5(E–G)). Viral titers in tissues at 1 and 5 dpi fell below the limit of detection of the TCID₅₀ assay. Blood viral loads were significantly reduced compared to the parental virus group, and at 3 dpi, titers in the brain, knee joint and lung were markedly lower (Figure 5(H)). These findings demonstrate that extensive consecutive deletions within the 3’UTR substantially attenuate viral virulence.

Large deletions in the GETV 3’UTR impair early viral colonization in mosquitoes

To assess the influence of the 3’UTR on viral transmission potential, Aedes albopictus mosquitoes were fed blood meals containing equivalent titers of parental virus or rGETV-KO310. Mosquitoes were collected at designated time points, and total RNA was extracted for absolute quantification of viral RNA copies via RT-qPCR (Figure 6(A)). At 5 and 10 dpi, viral RNA copies of rGETV-KO310 were significantly lower than those of the parental virus; by 15 dpi, no significant difference was detected between the two groups (Figure 6(B)). Notably, to determine the stability of the viral genome during mosquito infection, viral RNA extracted from mosquitoes at 15 dpi was subjected to RT-PCR targeting the deletion site, followed by sequencing analysis. The results confirmed that the 310-nt fragment remained absent, indicating that the deletion was stably maintained throughout viral replication in mosquitoes (Supplemental Figure 3). These findings demonstrate that the 5’ proximal 310-nt segment of the GETV 3’UTR facilitates early viral colonization in mosquitoes.

Figure 6.

Replication kinetics of GETV in Aedes albopictus mosquitoes. (A) Schematic of the artificial blood-feeding infection protocol. Aedes albopictus mosquitoes were fed a solution containing virus, defibrinated sheep blood, and 10% sucrose. (B) Quantification of viral replication in mosquitoes. Total RNA was extracted from whole mosquitoes at 5, 10, and 15 days post-infection (dpi). Viral E2 gene copy number was determined by reverse transcription quantitative PCR (RT-qPCR). Data represent mean ± SD (log₁₀ copies per mosquito). *p < 0.05, **p < 0.01, ***p < 0.001 (unpaired Student’s t-test,asterisks indicate significant differences between each mutant virus and the parental virus at each time point).

Transcriptomic profiling of mice infected with rGETV-KO310 versus parental virus

To identify biological processes potentially modulated by the GETV 3’UTR during infection, RNA sequencing was performed on knee joint tissues from mice infected with rGETV-KO310 or the parental virus, with mock-infected mice serving as controls. Quality assessment of the nine sequenced samples yielded 77.00 Gb of high-quality raw data. After filtering and alignment to the reference genome, mapping efficiencies ranged from 96.88% to 98.33% (Supplemental Table S2). Sample variability and replicate consistency were evaluated (Supplemental Table S3), confirming high base quality suitable for differential expression analysis.

Principal component analysis (PCA) of gene expression profiles revealed clear separation between experimental groups, with high intra-group reproducibility (Supplemental Figure S4A). Unsupervised clustering of differentially expressed genes (DEGs) showed group-specific patterns (Supplemental Figure S4B), and Pearson correlation analysis confirmed strong intra-group coherence and clear inter-group distinction (Supplemental Figure S4C), supporting the reliability of the transcriptomic data. Comparative analysis identified 2,612 DEGs between infected and control groups. Volcano plots illustrated that both up- and down-regulated DEGs were fewer in rGETV-KO310-infected mice compared to parental virus-infected mice (Figure 7(A)). A Venn diagram further indicated that deletion of the 310-nt 3’UTR segment altered the identity and magnitude of DEGs during infection, with an overall reduction in transcriptional perturbations (Supplemental Figure S4D).

Figure 7.

Transcriptomic profiling of knee joints from virus-infected mice. (A) Volcano plot illustrating differentially expressed genes (DEGs) in knee joints of infected mice versus mock-infected controls. Significantly upregulated (red) and downregulated (green) genes are shown (|log₂(fold change)| >1, adjusted p-value < 0.05). (B) Gene ontology (GO) enrichment analysis of biological processes for the DEGs identified in (A). (C) KEGG pathway enrichment analysis of the DEGs. (D) Protein-protein interaction (PPI) network analysis of differentially expressed genes.

Gene Ontology (GO) enrichment analysis of DEGs in knee joint samples revealed that 541 DEGs in the rGETV-GX vs rGETV-KO310 comparison were enriched in 60 level-2 terms, including viral gene expression, biological adhesion, gene expression, extracellular matrix organization, myosin complex assembly, and structural molecular activity (Figure 7(B)). In both MOCK vs rGETV-KO310 and MOCK vs rGETV-GX comparisons, DEGs were significantly enriched in immune effector processes, response to interferon-γ and -β, defense response, and biological adhesion. Notably, immune effector processes and interferon responses were more strongly enriched in parental virus infection, whereas biological adhesion terms were less enriched in the rGETV-KO310 group, suggesting that the 3’UTR deletion modulates host immune activation and adhesion-related pathways.

KEGG pathway analysis (Figure 7(C)) indicated that DEGs from both infection groups were enriched in pathways involving viral protein interactions, cytokine-cytokine receptor interactions, cell adhesion molecules, and antigen processing and presentation. The enrichment in viral protein-cytokine pathways was more pronounced in the rGETV-KO310 group. Additionally, in the rGETV-GX vs rGETV-KO310 comparison, DEGs were enriched in metabolic and signal transduction pathways, including MAPK signaling, ECM-receptor interaction, ribosome, and focal adhesion. All DEGs mapped to the MAPK signaling pathway were upregulated, indicating that 3’UTR deletion alters immune regulation and metabolic responses in infected tissues.

Protein-protein interaction (PPI) network analysis was conducted using the STRING database and visualized with Cytoscape (Figure 7(D)). The resulting network revealed extensive interactions among differentially expressed proteins, with UBC, Fras1, Egf, Tnn, Fn1, and Mapk13 occupying central hubs, suggesting their potential roles in coordinating host responses to GETV infection.

Predicted secondary structures of the 3’UTR in different deletion mutants

Secondary structure predictions of the GETV 3’UTR reveal a complex architecture featuring multiple stem-loop structures (Supplemental Figure S1). The UTR of the wild-type virus (rGETV-GX) exhibits high thermodynamic stability, with a predicted free energy (ΔG) of −67.10 kcal/mol. Sequential deletions of its repetitive sequence elements (RSEs) induce significant structural remodeling, characterized by alterations in the number, positioning, and overall topology of the stem-loops. The minimal replicating virus (rGETV-KO310) shows a substantial increase in UTR free energy to −20.60 kcal/mol. Specifically, deletion of the sequence upstream of RSE2 leads to the formation of a novel stem-loop within the spacer between RSE2 and RSE1. Concurrently, RSE2, RSE3, and the conserved sequence element (CSE) each participate in forming new stem-loops, reorganizing the overall secondary structure around two newly emerged large loops. Deletion upstream of RSE3 similarly triggers the involvement of RSE3 and CSE in new stem-loop formation, resulting in a topology distinct from rGETV-GX, with the newly formed stem-loops organized around a loop proximal to the 3” end. Removal of all RSEs and their intervening spacers causes further UTR restructuring, reducing the total number of stem-loops and altering the position of the CSE within the structural context. Upon further truncation of the 3’UTR, the stem-loops at the 5” proximal region gradually diminish, accompanied by an increase in unpaired nucleotides within the CSE. Notably, except in the case of rGETV-KO300, three stem-loops at the extreme 3’ terminus remain relatively stable across deletion mutants, suggesting their structural autonomy. Of particular interest, an independent analysis of the nt 256–290 region indicates its intrinsic ability to form a stem-loop. Its deletion results in the expansion of the largest loop in the overall structure while leaving the terminal architecture unaffected, implying that this region may function as a relatively independent structural module.

Discussion

Genomic surveillance of GETV indicates that the virus is undergoing continuous evolutionary adaptation [5]. Emerging variants exhibit an accumulation of amino acid substitutions within viral protein-coding regions, suggestive of a trend toward enhanced virulence [35]. While numerous studies have focused on point mutations in the open reading frame (ORF), the evolutionary dynamics and adaptive significance of non-coding regions, particularly the UTRs, remain comparatively understudied. Recent epidemiological investigations have identified a 32-nucleotide tandem repeat insertion within the GETV 3’UTR, underscoring that substantial genetic variation also occurs in non-coding regions and may influence viral replication efficiency, immune evasion, and host adaptation [36]. Indeed, alphavirus 3’UTRs display considerable plasticity, characterized by frequent insertions, deletions, and sequence duplications across different viral lineages, in addition to point mutations.

Consistent with observations in other alphaviruses, GETV remains infectious despite the deletion of various RSEs within the 3’UTR [37]. In BHK-21 cells, mutant viruses replicated at levels comparable to the parental virus, aligning with findings from related alphavirus systems, such as EEEV [29] and CHIKV [37]. Notably, although these deletion mutants exhibited attenuated replication in Vero cells, their overall kinetic profiles remained similar. At specific time points, the relative replicative capacity of the knockout mutant exhibited dynamic variations – particularly in the case of rGETV-KO3. Nevertheless, AUC analysis revealed that its overall replication level in Vero cells showed no significant difference compared to the wild-type virus. This fluctuation suggests that the mutation may not simply cause a unidirectional block in replication within the Vero cell environment. Instead, it could subtly alter the timing of viral replication or its interaction with specific host factors, leading to changes in the shape of the replication kinetics curve – sometimes lagging and sometimes advancing – rather than a straightforward overall suppression or enhancement. Possible explanations include delays in genomic RNA synthesis, altered kinetics of viral particle assembly, or dysregulation of host translation/replication shutdown processes. However, simultaneous deletion of all three RSEs resulted in a significant fitness deficit in C6/36 mosquito cells, indicating a heightened dependency on the 3’UTR for interactions with mosquito-specific cellular factors. Analogous findings in chikungunya virus (CHIKV) have demonstrated that direct repeats (DRs) within the 3’UTR are subject to host-specific selective pressures: these elements enhance viral replication in vitro and in mosquito vectors but are dispensable in mammalian cells [28]. Further studies have revealed that 3’UTR truncation variants, generated via RNA recombination, gain a selective advantage in mammalian hosts, whereas revertants with intact 3’UTRs are positively selected upon reentry into mosquito populations [38]. Similarly, in Sindbis virus (SINV), RSEs may selectively regulate the translation of subgenomic mRNA (sgmRNA) in insect cells but not in mammalian systems. Moreover, insertion of SINV RSEs into the shortened 3’UTR of Sleeping Disease Virus (SDV) – an alphavirus lacking a known invertebrate vector – enhanced sgmRNA translation efficiency and viral replication in insect cells [39].

EEEV subjected to serial passage in mammalian fibroblasts acquires a 238-nt deletion in the 3’UTR, indicating that this region is non-essential for replication in mammalian systems [40]. In the present study, a deletion spanning 310 nt was introduced, and the resulting virus remained viable, underscoring the structural plasticity of this region. This tolerance may be attributable to compensatory RNA elements – such as conserved stem-loop structures – or other unidentified mechanisms that preserve viral fitness. Previous studies on CHIKV have shown that large 3’UTR deletions can be compensated by adaptive mutations within the coding region [41]. Notably, as the deletion size increased, the replicative capacity of mutant viruses was not only impaired in C6/36 cells but also significantly reduced in BHK-21 and Vero cells. The observed reduction in plaque size may reflect diminished cell-to-cell spread, potentially due to compromised RNA packaging or delayed virion release, thereby constraining plaque expansion. Collectively, these findings suggest that 3’UTR integrity influences viral growth kinetics by modulating the rate of RNA synthesis or the efficiency of progeny virion maturation. Secondary structure predictions indicate that nucleotide deletions alter the RNA folding landscape. Several conserved RNA secondary structures have been identified at the genomic termini of alphaviruses. For instance, a conserved Y-shaped structure (SLY) in the CHIKV 3’UTR exhibits lineage-specific variation: ECSA/IOL lineage viruses possess a single SLY domain, whereas Asian strains carry duplicated SLY elements. The presence of SLY is necessary for efficient replication in mosquito cells, and its copy number correlates positively with viral fitness [24]. Similarly, a recent study revealed that the CHIKV 3’UTR can adopt a unique double stem-loop conformation, and a 44-nt deletion alters this structure, thereby affecting viral fitness in both mammalian and mosquito hosts [42]. The mechanistic impact of analogous structural changes in the GETV 3’UTR warrants further investigation.

During early infection, we observed impaired replication of the 3’UTR deletion mutant in mosquitoes, although replication levels recovered to parental levels at later time points. This phenomenon may be linked to the structural integrity of the 3’UTR. Studies on CHIKV indicate that the full-length 3’UTR confers a replicative advantage in mosquitoes, with viruses bearing intact 3’UTRs dominating during mixed infections in later stages [43]. Conversely, 3’UTR deletion mutants exhibit reduced vector competence and transmission potential in Aedes mosquitoes, likely due to a diminished capacity to overcome anatomical barriers such as the midgut escape barrier and salivary gland infection barrier [44]. Furthermore, viral competition assays in mosquitoes have demonstrated that minor variants with enhanced replication kinetics can outcompete slower-replicating viruses in specific tissues. The infection dynamics of arboviruses in mosquito vectors are critical determinants of transmission efficiency and epidemiological spread [45–47]. Viral genetic variation influences growth kinetics and interactions with mosquito barriers, collectively shaping transmission outcomes [45,48]. Viruses with high replication rates are more likely to successfully infect mosquitoes, disseminate to secondary tissues, and be transmitted via saliva, resulting in a shortened extrinsic incubation period [44]. The recovery of rGETV-KO310 replication in mosquito cells at later stages suggests that, although extensive 3’UTR deletions impair viral fitness in the host, insect-specific factors may partially compensate, thereby sustaining viral persistence in natural cycles. Recent evidence indicates that Aedes aegypti miRNAs miR-2944b-5p and miR-2b can directly bind specific sequences in the CHIKV 3’UTR, enabling viral manipulation of host miRNA pathways to regulate mitochondrial homeostasis and balance viral replication with host cell survival [49]. It is noteworthy that the observation of equivalent replication levels between the parental virus and the mutant virus at later stages of infection suggests that, in addition to compensation by insect-specific factors, adaptive changes in other regions of the viral genome may also occur to compensate for the functional deficits caused by the 3’UTR deletion. Similarly, in this study, all in vitro replication kinetics experiments were performed using viruses passaged to the third generation (P3) to ensure phenotypic stability; although sequencing of the 3’UTR in P3 viruses confirmed the intended deletion, whole-genome sequencing was not conducted. Thus, we cannot rule out the possibility that adaptive (compensatory) mutations may have emerged in other regions of the genome during virus rescue or early passages. In conclusion, more comprehensive genomic monitoring is warranted in subsequent continuous passage studies to uncover potential compensatory mutations and their impacts on viral fitness and transmission potential.

In 10-day-old mice, extensive 3’UTR deletions further attenuated viral fitness, and mutants with deletions exceeding 290 nt were completely avirulent. Viral loads in multiple tissues of rGETV-KO310-infected mice fell below the detection limit of TCID₅₀ assays. Transcriptomic analysis revealed that, compared to control mice, knee joint tissues from infected animals exhibited significant enrichment of differentially expressed genes (DEGs) involved in immune effector processes, interferon-γ and -β responses, and defense mechanisms. This age-dependent phenotypic difference may stem from the immature immune system of neonatal mice [50]. Three-day-old mice mount weaker innate immune responses – including interferon signaling – permitting transient viral amplification in local tissues such as the knee joints. In contrast, 10-day-old mice possess more mature immune surveillance mechanisms, which, combined with the 3’UTR deletion, further restrict viral dissemination. Notably, the rGETV-KO1 mutant exhibited enhanced replication in the lungs of suckling mice compared to the parental virus, possibly due to tissue-specific adaptive changes or altered host-virus interactions [41]. Similar observations have been reported in SINV studies, where mutant viruses showed increased replication in murine brain tissue [51]. Partial 3’UTR deletions may disrupt viral mechanisms for suppressing innate immunity in specific tissues – for example, through the loss of miRNA binding sites or other immune-evasive RNA elements – thereby triggering stronger inflammatory responses that accelerate viral clearance while transiently enhancing early replication under immune pressure [29]. The conserved terminal region of the 3’UTR may contain essential sequences for viral RNA polymerase binding or genome cyclization, and their deletion could impair negative-strand RNA synthesis or virion assembly [27].

Transcriptomic profiling revealed altered host immune responses following infection with the 3’UTR deletion mutant. Compared to rGETV-GX, infection with rGETV-KO310 induced fewer DEGs, with a notably lower proportion of upregulated genes, indicating that 3’UTR integrity is critical for the virus to modulate host transcriptional programs effectively. However, the rGETV-KO310 group exhibited more pronounced enrichment in the KEGG pathway “viral protein interaction with cytokine and cytokine receptor” than the rGETV-GX group. During viral infection, nucleic acids trigger innate and adaptive immune responses in mammalian hosts via pattern recognition receptors (PRRs), leading to production of type I interferons (IFNs) and pro-inflammatory cytokines. Subsequent expression of interferon-stimulated genes (ISGs) inhibits viral replication and amplifies antiviral signaling [52]. In rGETV-KO310 infection, interferon gene expression remained unchanged, and ISGs such as ISG15 and ISG20 were not significantly upregulated. In contrast, expression of pro-inflammatory cytokines including IL6 and IL7 was significantly reduced, indicating an attenuated inflammatory response relative to parental virus infection – consistent with the milder clinical manifestations observed. Protein-protein interaction (PPI) analysis of GO and KEGG pathways identified UBC, Fras1, Egf, Tnn, and Mapk13 as hub genes within the differential interaction network, all of which are implicated in inflammatory signaling. Among these, Mapk13, a member of the p38 MAPK family, promotes expression of inflammatory mediators such as TNF-α, IL-6, and IL-8 through activation of transcription factors including NF-κB and AP-1 [53,54]. We observed significant downregulation of MAPK pathway genes in the knee joints of rGETV-KO310-infected mice compared to the parental virus group, suggesting that deletion of the 5’ 310-nt segment of the GETV 3’UTR alleviates virus-induced inflammatory responses.

An important consideration when interpreting these transcriptomic findings is whether the attenuation of host gene expression perturbations merely reflects the reduced replication efficiency of the mutant virus. We acknowledge that the lower viral antigen load resulting from attenuated replication of rGETV-KO310 likely contributes to an overall dampening of immune response intensity. This offers a plausible explanation for both the reduced number of differentially expressed genes (DEGs) and the weaker activation of pathways such as the interferon response. However, our data suggest that additional, more specific immunomodulatory effects – attributable directly to the 3’UTR deletion – are also involved. If the observed differences were driven exclusively by viral load, a proportional attenuation across all immune-related pathways would be expected. Instead, we detected directional changes in specific signaling cascades. For instance, genes within the MAPK signaling pathway were consistently downregulated in rGETV-KO310-infected mice relative to wild-type infection. Moreover, the expression profile of key pro-inflammatory cytokines, such as IL-6, did not show significant elevation following rGETV-KO310 infection – a pattern distinct from that induced by the wild-type virus. These specific alterations in the host transcriptional landscape imply that the 3’UTR deletion may directly modulate virus – host interactions – potentially via effects on RNA stability, translational efficiency, or engagement with host RBPs and cellular miRNAs.

Notably, deletion of regions outside the known replication signal elements (RSEs) in the 3’-UTR, particularly the 275–290 nt segment, resulted in more pronounced in vivo attenuation. This suggests that this region may harbor additional cis-regulatory elements. Recent studies indicate that certain miRNAs (e.g. hsa-miR-122b-5p) and RNA-binding proteins (RBPs; e.g. IFIT2) can modulate alphavirus replication by recognizing specific sequences or secondary structures within the 3’-UTR. We therefore hypothesize that the deleted region may contain binding sites for such host factors. Their absence could disrupt the equilibrium between viral replication and host regulatory networks, ultimately diminishing viral fitness. Future studies will focus on identifying the specific host molecules that interact with this region and delineating their functional mechanisms.

In summary, this study, through targeted deletion experiments, revealed that the GETV 3’UTR exhibits remarkable plasticity, tolerating extensive nucleotide deletions without compromising viral viability. Complete deletion of conserved repeat sequence elements (RSEs) resulted in cell-type-dependent attenuation of viral replication and significantly reduced virulence in murine models. Notably, the mutant strain rGETV-KO310, carrying a 310-nt deletion, demonstrated further attenuation in vivo and impaired early colonization in mosquito vectors. Transcriptomic analysis indicated that 3’UTR truncation modulates key host immune responses, including interferon and MAPK signaling pathways, correlating with reduced pathogenicity. These findings establish the 3’UTR as a rational target for the design of live-attenuated vaccines.

Funding Statement

This research was supported by the Guangxi Natural Science Foundation under Grant number [2026GXNSFBA00640123, 2023GXNSFAA026494 and 2018GXNSFDA281021]; Research Foundation for Advanced Talents of Guangxi University under Grant number [XGZ130959].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data underlying this article are available in the article and in its online Supplementary material. Data are deposited on ScienceDB at https://doi.org/10.57760/sciencedb.30626 [55].

Ethics approval

The animal experiments were approved by the Animal Experiment Committee of Guangxi University with the approval number (GXU2022-288).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21505594.2026.2687251