‌An mRNA vaccine confers enhanced protection against herpes simplex virus through an IFN-I-dependent pathway Authors: Wenying Zhao, Lingjin Sun, Peng Wang, Qian Zhao, Ying Li, Yang Li, Hongyang Shi, Man Xing, Weiqian Dai, Dongming Zhou Published: 2026 ABSTRACT AbstractHerpes simplex virus (HSV) types 1 and 2 cause widespread oral or genital infections, but no prophylactic or therapeutic HSV vaccine has been approved to date. In this study, we developed three mRNA vaccine candidates expressing key viral glycoproteins: monovalent gD2, bivalent gD2-gC1, and bivalent gD2-gE1. We assessed their immunogenicity and protective efficacy in a murine model. All candidates elicited robust humoral and cellular immunity and provided significant protection against intravaginal HSV challenge. Notably, the gD2-gE1 vaccine induced markedly stronger immune responses. Mechanistically, its superior immunoprotective efficacy was associated with the stronger IFN‑I response, which thereby enhanced the adaptive immune response. Collectively, our findings provide a scientific rationale and valuable insights for the future development of HSV vaccines.Subject terms: Immunology, Microbiology Introduction Herpes simplex virus (HSV) is a double-stranded DNA virus belonging to the Herpesviridae family, with two main serotypes: HSV-1 and HSV-21. Traditionally, HSV-1 has been associated mainly with orolabial lesions, whereas HSV-2 has been recognized as the principal cause of genital herpes. However, over the past few decades, this epidemiological pattern has shifted, with a consistent increase in genital infections attributable to HSV-1. In several countries, particularly in high-income regions such as North America or Western Europe, HSV-1 has become the leading cause of first-episode genital herpes2. In 2020 alone, approximately 205 million people experienced symptomatic genital herpes. Given this substantial and lifelong global health burden, the World Health Organization has advocated for accelerated development of vaccines and other preventive measures3. Numerous HSV vaccine candidates based on various platforms, including live-attenuated, subunit, DNA, and mRNA vaccines, have entered clinical trials4. However, due to the virus’s sophisticated pathogenic and immune evasion mechanisms, most vaccines have failed during preclinical or early clinical development. While a limited number have progressed to Phase III trials, none have achieved their primary efficacy goals4–6. This history of failed clinical trials highlights two critical shortcomings: first, vaccines relying on a single antigen, most commonly glycoprotein D (gD), have proven insufficient to elicit robust and protective immunity7; second, formulations based on a single serotype often fail to confer cross-protection against both HSV-1 and HSV-2. Consequently, there is a clear consensus that traditional vaccine platforms are suboptimal for HSV, underscoring an urgent need for novel strategies that incorporate multivalent antigen design, enhanced immunogenicity, and broader cross-reactivity. The glycoproteins gD, gC and gE play distinct and critical roles in HSV pathogenesis and immune evasion, making them rational targets for a multivalent vaccine8,9. The gD is essential for viral entry, mediating fusion through its interaction with cellular receptors such as Herpesvirus Entry Mediator, nectin-1, and nectin-210–12. The gC acts as an immune evasion protein by binding to complement component C3b, thereby inhibiting the formation of the membrane attack complex and blunting complement-mediated viral clearance and T-cell co-stimulation13. Furthermore, the gE acts as an IgG Fc receptor, which sequesters host antibodies to block antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent neutralization14–17. Given gD’s pivotal role in viral entry and the potent immune evasion activities of gC and gE, we hypothesized that a vaccine targeting these antigens would not only block viral entry but also cripple key escape mechanisms, potentially eliciting a more robust and comprehensive immune response. Leveraging the well-established mRNA-LNP platform, which has proven highly effective in COVID-19 vaccine development18, we constructed a panel of vaccine candidates: a monovalent gD2 vaccine and two bivalent vaccines containing gD2 with either gC1 or gE1. We systematically evaluated the immunogenicity and protective efficacy of these candidates in a BALB/c mouse model, assessing both cellular and humoral immune responses and evaluating protection against lethal vaginal challenge. By integrating these findings with transcriptomic analysis of innate immune activation, we identified the most potent vaccine formulation and provided mechanistic insights for the development of an effective HSV vaccine. Results The gD2-gE1 mRNA vaccine elicits potent and durable antibody responses Codon-optimized, truncated gD2, gC1 and gE1 genes were cloned into the pUC57-mRNA-101PolyA vector (Supplementary Fig. 1a). Using these plasmids as templates, the corresponding mRNAs were obtained through in vitro transcription (IVT). After transfecting HEK 293T cells with mRNAs encoding gD2, gC1, and gE1, Western blot analysis confirmed the expression of all three target antigens at 24 h (Supplementary Fig. 1b, c). BALB/c mice were intramuscularly immunized with a prime-boost regimen on days 0 and 14. The vaccine groups were immunized with a monovalent vaccine containing 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gC1, or LNP-gD2 and LNP-gE1. The control group was administered 100 µL of PBS following the same schedule. To evaluate the long-term humoral immune response, serum samples were collected at 4, 6, 8, 10, and 14 weeks after the prime immunization (Fig. 1a). The gD2-gE1 mRNA vaccine elicits potent and durable antibody responses. a Immunization and serum collection strategy for mRNA vaccines in mice. Female BALB/c mice (6-8 weeks old, n = 6 per group) were intramuscularly immunized twice with 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gC1, or LNP-gD2 and LNP-gE1; the control group received 100 μL of PBS. Serum samples were then collected at the indicated time points. (Created with BioGDP.com49) Binding antibody titers (log₁₀) against inactivated HSV-1 (b), inactivated HSV-2 (c), and protein gD2 (d) were determined at different time points post-vaccination by ELISA. e Endpoint titer ratios of IgG2a to IgG1 were calculated at 6 weeks post-primer vaccination. Neutralizing antibody NT50 titers (log₁₀) against HSV-1 (f) and HSV-2 (g) were determined at 6 and 14 weeks post-primer vaccination. All data were shown as means ± SEM. P-values were analyzed with one-way ANOVA in (e). P-values were analyzed with two-way ANOVA in (f, g). (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). Specific binding antibodies against inactivated HSV-1, inactivated HSV-2, and gD2 were determined by ELISA. All three vaccines elicited high and comparable titers of specific IgG binding antibodies, in contrast to the control group, which showed no specific antibody response (Fig. 1b–d). Furthermore, these antibody titers remained elevated at 14 weeks. IgG subclass analysis revealed that all three vaccines elicited gD2-specific IgG1, IgG2a, and IgG2b antibodies (Supplementary Fig. 2). The IgG2a/IgG1 ratio ranged between 0.8 and 1.0 across all vaccine groups, indicating a balanced Th1/Th2 immune response (Fig. 1e). Neutralizing antibody titers against live HSV-1 and HSV-2 were evaluated at 6 and 14 weeks post-primer vaccination. While the control group showed no neutralizing ability against either virus, all three vaccine groups generated detectable neutralizing antibodies against both HSV-1 and HSV-2 (Fig. 1f, g). The presence of HSV-1 neutralization in mice immunized with LNP-gD2 indicates partial cross-neutralization, likely attributable to the high homology between HSV-1 and HSV-2 gD glycoprotein19. HSV-1-specific neutralizing antibody titers in the gD2-gE1 group were significantly higher than in both the gD2 and gD2-gC1 groups at 6 weeks (Fig. 1f). In contrast, no significant differences in HSV-2-specific neutralizing antibody titers were observed among the vaccine groups at this time point (Fig. 1g), suggesting that the neutralizing epitopes on gE1 don’t confer cross-neutralization against HSV-2. Additionally, neutralizing antibodies against HSV-1 and HSV-2 remained detectable at 14 weeks, with anti-HSV-2 titers showing no significant decline between 6 and 14 weeks. These results demonstrate that the mRNA vaccine can induce durable, protective antibody responses against HSV. The gD2-gE1 mRNA vaccine induces the most potent antigen-specific T cell responses T cell responses were assessed at 4 weeks post-primer vaccination using splenocytes from BALB/c mice vaccinated with gD2, gD2-gC1, or gD2-gE1 (Fig. 2a). The gD2-specific cellular immune responses were detected by enzyme-linked immunospot (ELISpot) assay. Upon stimulation with the gD2 peptide pool, all three vaccine groups exhibited a significantly higher number of IFN-γ-secreting cells compared to the control group, with the bivalent gD2-gE1 group eliciting the highest number of spot-forming units (SFUs) (Fig. 2b). The gD2-gE1 mRNA vaccine induces the most potent antigen-specific T cell responses. a Immunization and tissues collection strategy for mRNA vaccines in mice. Female BALB/c mice (6-8 weeks old, n = 6 per group) were intramuscularly immunized twice with 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gC1, or LNP-gD2 and LNP-gE1, while the control group received 100 μL of PBS. Splenocytes were collected at 2 weeks post-booster vaccination. (Created with BioGDP.com) (b). ELISpot assay was performed to detect IFN-γ secretion of splenocytes after the gD2 peptide pool stimulation. Percentages of CD4+ T cells expressing IFN-γ, TNF-α, IL-2 and IL-4 after the gD2 peptide pool (c), inactivated HSV-1 (d), or inactivated HSV-2 (e) stimulation. Percentages of CD8+ T cells expressing IFN-γ, TNF-α, IL-2, and IL-4 after the gD2 peptide pool (f), inactivated HSV-1 (g), or inactivated HSV-2 (h) stimulation. All data were shown as means ± SEM. P-values were analyzed with one-way ANOVA in (b). P-values were analyzed with two-way ANOVA in (c–h). (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). To further characterize polyfunctional T‑cell responses, intracellular cytokine staining was performed. After stimulation with the gD2 peptide pool, specific effector CD4+ T cells exhibited distinct Th1 and Th2 cytokine profiles (Fig. 2c). All three vaccine groups significantly enhanced IFN-γ production compared to the control. The frequency of CD4+ T cells secreting TNF-α was markedly elevated in the bivalently gD2‑gC1 and gD2‑gE1 groups relative to controls, while IL-2 levels didn’t differ between vaccine groups and the control group. Meanwhile, all three vaccine groups induced the production of the Th2-type cytokine IL-4. After stimulation with inactivated HSV-1, the frequency of CD4+ T cells from all three vaccine groups secreting IFN-γ, TNF-α, and IL-4 was significantly higher than those from the control group (Fig. 2d). Notably, the gD2-gE1 group exhibited the highest cytokine secretion levels, with statistically significant differences compared to the gD2 group. A similar pattern was observed after stimulation with inactivated HSV-2 (Fig. 2e). Analysis of antigen-specific effector CD8⁺ T-cell responses revealed that, following stimulation with either the gD2 peptide pool, inactivated HSV‑1, or inactivated HSV‑2, splenocytes from all three vaccine groups secreted significantly higher levels of IFN‑γ and IL‑4 compared to the control group (Fig. 2f–h). Moreover, the secretion levels in the gD2‑gE1 group were consistently higher than those in the other two vaccine groups. Collectively, these results demonstrate that all vaccine groups elicit robust antigen-specific CD4⁺ and CD8⁺ T cell responses, with the bivalent gD2-gE1 candidate inducing the most potent cellular immunity. HSV mRNA vaccines induce functional B cell immunity Germinal centers (GCs) are critical sites for the generation of high-affinity antibody-secreting plasma cells and memory B cells (MBCs); whose antibody output critically shapes the affinity, specificity, and durability of humoral immunity20. To systematically evaluate vaccine-induced B-cell responses, we analyzed key B-cell subsets across time points and tissues (Fig. 3a). HSV mRNA vaccines induce functional B cell immunity. a Immunization and cell collection strategy for mRNA vaccines in mice. Female BALB/c mice (6-8 weeks old, n = 6 per group) were intramuscularly immunized twice with 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gC1, or LNP-gD2 and LNP-gE1, while the control group received 100 μL of PBS. dLNs and spleens were harvested on days 7 and 14 post-booster vaccination; Bone marrows were collected at 14 weeks post-primer vaccination. (Created with BioGDP.com) b GC B cell responses in dLNs and spleens on days 7 and 14 post-booster vaccination. c MBC responses in spleens at 14 weeks post-primer vaccination. d LLPC responses in bone marrows at 14 weeks post-primer vaccination. All data were shown as means ± SEM. P-values were analyzed with one-way ANOVA in (c, d). P-values were analyzed with two-way ANOVA in (b). (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). First, germinal center B (GC B) cell responses in draining lymph nodes (dLNs) and spleens were analyzed by flow cytometry on days 7 and 14 after booster immunization (Fig. 3b). Compared with the control group, all vaccine groups exhibited robust GC B responses in both tissues on day 7, with a more pronounced reaction in the dLNs. On day 14, GC B cell frequencies remained elevated in the dLNs, whereas the splenic GC B response exhibited a clear decline, indicating a more persistent GC B response in lymphoid sites proximal to immunization. Given the critical role of MBCs in long-term immune protection, we further analyzed gD2-specific MBCs in the spleen at 14 weeks (Fig. 3c). All three vaccine groups showed a notable numerical increase in gD2-specific MBCs compared to the control group, with the gD2-gE1 group reaching statistical significance. To assess the durability of antibody-secreting cells, we finally examined long-lived plasma cells (LLPCs) in the bone marrow at 14 weeks (Fig. 3d). All three vaccine groups induced significantly higher frequencies of LLPCs compared with the control group, confirming the establishment of persistent antibody-secreting populations. Taken together, these findings indicate that all three vaccine candidates possess the capacity to elicit functional B‑cell responses from early GC activation to long-term memory B cells and durable LLPCs. HSV mRNA vaccines provided robust protection in a murine intravaginal HSV challenge model To evaluate the protective efficacy of the candidate vaccines, BALB/c mice were challenged intravaginally four weeks after the booster immunization with 2.7 × 10⁵ PFU of HSV-1 or 1.7 × 10⁵ PFU of HSV-2 (Fig. 4a). All mice were treated with medroxyprogesterone to synchronize and thin the vaginal epithelium 5 days prior to challenge21. HSV mRNA vaccines provided robust protection in a murine intravaginal HSV challenge model. a Immunization and challenge strategy for mRNA vaccines in mice. Female BALB/c mice (6–8 weeks old, n = 10 per group) were intramuscularly immunized twice with 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gC1, or LNP-gD2 and LNP-gE1, while the control group received 100 μL of PBS. Four weeks after the booster vaccination, all mice were challenged intravaginally with 2.7 × 10⁵ PFU of HSV-1 or 1.7 × 10⁵ PFU of HSV-2. Following the challenge, survival and weight were continuously monitored for 14 days. Vaginal swabs were collected on days 2 and 4 post-challenge for viral titer and DNA copy quantification. DRGs were harvested on days 4 and 28 post-challenge for DNA copy quantification. (Created with BioGDP.com) Survival rate (b), body weight changes (c), and vaginal lesion scores (d) after HSV‑1 challenge. e Vaginal viral titers were determined on days 2 and 4 post-HSV-1 challenge. f Vaginal HSV‑1 DNA copies were quantified on days 2 and 4 post-HSV-1 challenge. g HSV‑1 DNA copies in DRGs were quantified on days 4 and 28 post-HSV-1 challenge. Survival rate (h), body weight changes (i), and vaginal lesion scores (j) on day 14 post-HSV-2 challenge. k Vaginal viral titers were determined on days 2 and 4 post-HSV-2 challenge. l. Vaginal HSV‑2 DNA copies were quantified on days 2 and 4 post-HSV-2 challenge. m. HSV‑2 DNA copies in DRGs were quantified on days 4 and 28 post-HSV-2 challenge. All data were shown as means ± SEM. P-values were analyzed with one-way ANOVA in (e, f, k, l). P-values were analyzed with two-way ANOVA in (g, m). (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). Following HSV‑1 challenge, the survival rate of control mice was 40% on day 14, whereas all mice in the vaccine groups survived (Fig. 4b). Body weight remained stable in all three vaccine groups, while the control group showed a significant decrease (Fig. 4c). Furthermore, none of the vaccinated mice exhibited disease‑related signs such as ruffled fur, hunched posture, abnormal gait, or lethargy. In contrast, control mice displayed varying degrees of redness, swelling, and hair loss starting from day 5 post‑challenge (Fig. 4d). Viral titers in vaginal swabs collected on days 2 and 4 post‑challenge were measured to assess acute infection (Fig. 4e). Although vaginal viral titers were high across all groups on day 2, they were significantly lower in vaccine groups than in controls. On day 4, viral titers had declined sharply in the vaccine groups but remained high in controls. Quantification of HSV‑1 DNA copies by qPCR yielded consistent results: vaccinated groups showed approximately 6log₁₀ and 4log₁₀ copies on days 2 and 4, respectively, whereas controls maintained about 6log₁₀ copies on both days (Fig. 4f). These results indicate that vaccination significantly reduced viral load, reflecting effective neutralization or clearance of the virus. To evaluate neural infection, dorsal root ganglia (DRG) were harvested on days 4 and 28 post‑challenge for quantification of viral DNA copies, as peak HSV copy numbers in DRG typically occur around day 4–5 post‑infection. On day 4, HSV‑1 DNA was detectable all control mice (~5 log10 copies), but in only one mouse each from the gD2 and gD2‑gC1 groups, and in none from the gD2‑gE1 group (Fig. 4g). On day 28, high levels of HSV‑1 DNA persisted in the DRG of control mice, whereas no HSV‑1 DNA was detected in any vaccine group. Similar protection was observed against HSV‑2 challenge. All control mice succumbed on day 13, whereas all vaccinated mice survived (Fig. 4h). Body weight remained stable in vaccine groups but declined markedly in controls beginning on day 6 (Fig. 4i). Vaginal lesions appeared in controls on day 5 and increased in severity until death, whereas no lesions developed in vaccinated mice (Fig. 4j). Vaginal viral titers on days 2 and 4 were significantly lower in vaccine groups than in controls (Fig. 4k). Vaginal HSV‑2 DNA copy numbers in vaccinated mice were approximately 5log₁₀ and 4log₁₀ on days 2 and 4, respectively (Fig. 4l). In contrast, the control group consistently maintained levels around 6log₁₀ copies on both days. Although HSV‑2 DNA was detected in DRG of all groups on day 4, copy numbers were significantly reduced in all vaccine groups compared to the control group (Fig. 4m). On day 28, HSV‑2 DNA was undetectable in the gD2‑gE1 group, and was found in only two mice each from the gD2 and gD2‑gC1 groups. Collectively, these results demonstrate that all three vaccine candidates confer substantial protection against intravaginal challenge with both HSV‑1 and HSV‑2 in mice. Among them, the gD2-gE1 candidate appears to be more effective at preventing the establishment of latent HSV infection in DRG. The gD2-gE1 mRNA vaccine induces enhanced IFN‑I signaling Innate immunity acts as the host’s first line of defense against pathogens, and its activation strength is a key indicator of vaccine immunogenicity22. To comprehensively evaluate the innate immune response elicited by our candidates, we performed RNA sequencing (RNA-Seq) of dLNs collected at 24 h post-vaccination. RNA‑Seq analysis revealed that all three vaccine groups induced a substantial number of differentially expressed genes (DEGs) compared to the control (Fig. 5a, b). Pathway enrichment analysis of DEGs between the gD2‑gE1 and control groups using the Reactome databases and the Kyoto Encyclopedia of Genes and Genomes (KEGG) highlighted significant enrichment of pathways related to TLR signaling and IFN-I responses (Fig. 5c, d). Further analysis of immune‑related genes revealed significant upregulation of type I interferon (IFN-I) pathway components, including IFN‑α and IFN‑β, in all vaccinated mice (Fig. 5e). Specifically, the gD2-gE1 group exhibited significantly higher levels of IFN-I expression compared to the other two vaccine groups. Key pattern‑recognition receptors, including Toll‑like receptors (TLR3, TLR7, TLR9), the RIG‑I‑like receptor (IFIH1 and DHX58), and NOD-like receptors (NLPR3, NOD2), were also markedly upregulated, together with chemokines (e.g., CXCL9 and CXCL11) and cytokines (e.g., IL‑1β and IL‑6). The gD2-gE1 mRNA vaccine induces enhanced IFN‑I signaling. Mouse dLNs were isolated 24 h post-vaccination for bulk RNA-seq (n = 3 mice per group). a Hierarchical clustering analysis of DEGs in different groups. b DEGs are based on each group compared with the PBS group, |log2(fold change)| ≥ 1; adjusted P-value ≤ 0.05. Reactome (c) and KEGG (d) pathway enrichment analysis of upregulated DEGs in the gD2-gE1 group compared with the control group. e Average expression heatmap of selected DEGs classified by distinct immune functions. f RT‑qPCR quantification of type I IFN mRNA in dLNs 24 h post‑vaccination. g RT‑qPCR quantification of type I IFN mRNA in BMDCs 8 h after in vitro vaccine stimulation. All data were shown as means ± SEM. P-values were analyzed with two-way ANOVA in (f, g). (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). To validate the RNA-Seq findings, we first examined the expression of IFN-I in dLNs at 24 h post-vaccination by RT‑qPCR (Fig. 5f). All three mRNA candidates induced IFN-I expression in vivo, with the gD2-gE1 group showing significantly higher levels than the other two vaccine groups. To further corroborate these findings and directly assess innate immune cell activation, we performed an in vitro validation using bone marrow-derived dendritic cells (BMDCs) (Fig. 5g). Consistent with the in vivo trend, all vaccine groups induced IFN-1 production in BMDCs, and the response in the gD2‑gE1 group remained significantly stronger. IFN‑I signaling is required for the superior immunogenicity of the gD2‑gE1 mRNA vaccine RNA-seq analysis revealed that the mRNA vaccines, particularly the gD2-gE1 candidate, stimulated a potent innate immune response characterized by a robust IFN-I signature. Given the association between the robust IFN-I response and the superior adaptive immunity observed in the gD2-gE1 group, we hypothesized that IFN-I signaling plays a mechanistic role in enhancing immunogenicity. To test this, we evaluated the immune responses of the gD2-gE1 vaccine in IFNAR1‑deficient (IFNAR1−/−) mice. Wild-type (WT) and IFNAR1−/− C57BL/6 mice were immunized intramuscularly with the gD2 or gD2-gE1, and PBS as a control, using a prime-boost regimen (at 0 and 2 weeks). Blood and spleens were collected for analysis at 4 weeks (Fig. 6a). Both WT and IFNAR1−/− mice produced specific binding antibodies against gD2, inactivated HSV-1, and inactivated HSV-2 post-vaccination, and the binding antibody levels in WT mice were significantly higher than those in the corresponding IFNAR1−/− mice (Fig. 6b-d). Similarly, although neutralizing antibodies were induced in both strains, WT mice immunized with the gD2-gE1 exhibited markedly higher neutralizing titers against HSV-1 and HSV-2 compared to IFNAR1−/− mice (Fig. 6e, f). When stimulated with inactivated HSV-1, HSV-2, or a gD2 peptide pool, WT mice vaccinated with the gD2-gE1 vaccine showed a markedly increased frequency of IFN-γ⁺ CD4⁺ T cells compared to controls. This increase was substantially higher than that observed in IFNAR1−/− mice receiving the same vaccine. (Fig. 6g–i). Similarly, when stimulated with inactivated HSV-1 or HSV-2, WT mice vaccinated with the gD2 vaccine exhibited a significantly higher frequency of IFN-γ⁺ CD4⁺ T cells than IFNAR1−/− mice receiving the same vaccine. In contrast, there were no significant differences in IFN-γ+ CD8⁺ T cells frequency among all groups under any stimulation condition (Fig. 6j–l). IFN‑I signaling is required for the superior immunogenicity of the gD2‑gE1 mRNA vaccine. a Immunization scheme. Wild‑type (WT) or IFNAR1−/− C57BL/6 female mice (6–8 weeks old, n = 5 per group) were intramuscularly immunized twice with 10 µg of LNP-gD2, or a bivalent vaccine containing 10 µg each of LNP-gD2 and LNP-gE1, while the control group received 100 μL of PBS. Serum samples and splenocytes were collected at 2 weeks post-booster vaccination. (Created with BioGDP.com) Binding antibody titers against inactivated HSV-1 (b), inactivated HSV-2 (c), and protein gD2 (d) were determined by ELISA. Neutralizing antibody NT50 titers (log10) against HSV-1 (e) and HSV-2 (f) were determined. Percentages of IFN-γ-expressing CD4⁺ T cells following stimulation with inactivated HSV-1 (g), inactivated HSV-2 (h), and gD2 peptide pool (i). Percentages of IFN-γ-expressing CD8⁺ T cells following stimulation with inactivated HSV-1 (j), inactivated HSV-2 (k), and gD2 peptide pool (l). All data were shown as means ± SEM. P-values were analyzed with two-way ANOVA (nsP ≥ 0.05; *P < 0.05;**P < 0.01; ***P < 0.001; ****P < 0.0001). Collectively, these results demonstrated that the absence of IFN-I signaling impairs the adaptive immune response to gD2-gE1 vaccination, as reflected in lower humoral and cellular immunity in IFNAR1−/− mice. This supports the conclusion that the enhanced immunogenicity of the gD2-gE1 vaccine is mediated, at least in part, through the activation of innate immune responses in a IFN-I ‑dependent manner. Discussion HSV-1 and HSV-2 are widespread neurotropic pathogens that establish lifelong infections, posing a substantial global health and economic burden. The clinical manifestations range from recurrent oral and genital ulcers to severe complications such as encephalitis and neonatal herpes, particularly in vulnerable populations23,24. Despite decades of research, no prophylactic or therapeutic HSV vaccine has been approved. Lessons from previous clinical failures highlight several key challenges. In this study, all three mRNA vaccines elicited robust humoral and cellular immune responses in BALB/c mice and conferred substantial protection against intravaginal HSV challenge. However, compared with the single‑antigen gD2 commonly used in previous HSV vaccine studies, the bivalent candidates exhibited superior immunogenicity. These mRNA vaccines primarily consist of two components: antigen‑encoding mRNA and LNPs. In addition to its delivery function, LNPs also act as an adjuvant by inducing the production of cytokines such as IL-1, IL-6, and IFN-α, thereby activating innate immunity25. In this study, compared to the monovalent vaccine, the bivalent vaccine contained twice the amount of LNP and induced a stronger adaptive immune response, which may be partly attributable to the adjuvant effect of LNPs. Notably, the gD2‑gE1 group induced significantly higher HSV-1-specific neutralizing antibody titers and enhanced IFN‑γ and IL‑4 production from both CD4⁺ and CD8⁺ T cells than the other candidates. The adjuvant effect of LNPs cannot fully explain the differences between the two bivalent candidates. Therefore, the superior immunogenicity of the gD2-gE1 group can be attributed to the gE1 antigen, at least in part. A study on the neutralizing activity of major HSV-1 glycoproteins in sera from those patients with a history of HSV-1 infection suggested that gE-specific antibodies showed low neutralizing activity26. Additionally, the mRNA components may activate the innate immune system through TLR7 and TLR8 signaling pathways, which recognize single-stranded RNA27. Accordingly, this study further investigated the role of innate immunity in driving the differential adaptive immune responses observed among the three vaccine candidates. Transcriptomic analysis revealed that all three vaccines triggered robust innate immune responses; however, the gD2‑gE1 candidate elicited a notably stronger IFN‑I signature. Functional studies in IFNAR1−/− mice confirmed that IFN‑I signaling is required for optimal adaptive immunity induced by this vaccine, as its absence significantly reduced neutralizing antibody titers and IFN‑γ production by CD4⁺ T cells. IFN‑Is are known to bridge innate and adaptive immunity by promoting dendritic cell maturation, cytokine production (e.g., IL‑6), and T follicular helper cell responses, all of which support GC formation and antibody affinity maturation28–30. Previous studies have also confirmed that IFNAR1 deficiency markedly impairs antigen‑specific B‑cell activation, expansion, and antibody production31. Recently, a study investigating the impact of mRNA and LNP components on immune responses further validated the critical role of IFN‑Is in driving optimal GC reactions32. Our findings align with recent work highlighting the importance of IFN‑I signaling in mRNA-LNP vaccines for potentiating adaptive immune responses. This proposed mechanism provides valuable insights for rational vaccine optimization. For instance, co-delivery of gD2-gE1 mRNA with an IFN‑I agonist could further potentiate innate immune activation, which may improve vaccine immunogenicity, particularly in immunocompromised populations. However, the mechanistic basis for the enhanced IFN-I production elicited by the gD2–gE1 candidate remains to be fully elucidated. Several mechanisms may underlie this observation. First, differences in expression levels or transcript stability between gE1 and gC1 could affect the potency of innate immune activation. Second, the coexpression of gD2 and gE1 may exert a synergistic effect on IFN-I induction. Another plausible explanation is that the secondary structure of the gE1 mRNA may facilitate its recognition by pattern recognition receptors, thereby enhancing IFN-I signaling. Viral challenge studies showed that all three vaccine groups provided substantial protection in mice. This result is consistent with previous findings using the same dose of an HSV-2 mRNA vaccine in a challenge model8. A primary objective for prophylactic genital herpes vaccines is to prevent clinical disease and reduce or eliminate HSV DNA shedding33,34. A critical challenge is the establishment of a latent viral reservoir in the DRG, which serves as the source for recurrent reactivation and persistent asymptomatic shedding. In the challenge experiment, HSV was more rapidly and completely cleared from the DRG in the gD2-gE1 group compared to others, suggesting enhanced control of neuronal infection. Given that gE-gI complex plays an important role in virus spread from cell to cell35–37, antibodies targeting gE1 might block this function, potentially impeding viral trafficking to the DRG and contributing to the observed reduction in neural infection. Our study offers two main insights for HSV vaccine research. First, by directly comparing monovalent and bivalents candidates, we demonstrated that combining gD2 with gE1 confers cross‑serotype protection against both HSV‑1 and HSV‑2. This represents a novel aspect, as it differs from previous multivalent HSV vaccine research that typically employed antigens from the same serotype38,39. Second, we identify a mechanism through which the gD2‑gE1 vaccine enhances immunogenicity: activation of innate immunity in an IFN‑I-dependent manner, which in turn shapes the adaptive response. However, this study has some limitations. While mice serve as the principal mammalian model for investigating human immune mechanisms in both health and disease, a critical limitation is their inability to develop spontaneous reactivation or recurrent disease following HSV-1 or HSV-2 infection40. Therefore, to address this constraint and better predict clinical outcomes, further validation of the vaccine’s immunogenicity and protective efficacy in alternative models, such as guinea pigs, tree shrews, non-human primates (NHPs) is required21,41,42. In particular, NHP models are capable of mirroring the subclinical characteristics of HSV-2 infection as observed in women43. Furthermore, the mechanism by which gE1 induces IFN-I production, as well as the precise cellular sources of this interferon, warrants further investigation. Methods Study design and animal experiments A total of 197 female mice (6–8 weeks old) were used for animal experiments. BALB/c and C57BL/6J mice were obtained from Beijing Vital River Laboratory Animal Technology. IFNAR1−/− mice were acquired from Shanghai Model Organisms Center. Animals were randomly assigned to experimental groups. All mice were maintained under specific pathogen-free conditions at Tianjin Medical University or Fudan University. All mouse cages were placed on the same rack to ensure identical environmental conditions. Body weight was measured at the same time each day. Mice with a body weight loss exceeding 20% or reaching study endpoints will be euthanized, as previously described44,45. Briefly, anesthesia was induced via intraperitoneal injection with 1.25% tribromoethanol at a volume of 200 µl per 10 g body weight. Body temperature was monitored during the procedure, and anesthesia was confirmed by the loss of the righting reflex. Once anesthesia was confirmed, mice were humanely euthanized by cervical dislocation. All animal experiments were approved by the Institutional Animal Care and Use Committee of Tianjin Medical University and the Ethics Committee of Fudan University (Approval Number: 2024-A051-02). The studies were conducted in accordance with the institutional guidelines for the care and use of laboratory animals. The challenge experiments were conducted in a Biological Safety Level 2 laboratory of the Shanghai Public Health Clinical Center, Fudan University. Blinding was not performed for vaccine preparation or animal experimentation. Cells and viruses The HEK 293T and Vero E6 cell lines were obtained from the American Type Culture Collection (ATCC). Cells were maintained in complete Dulbecco’s Modified Eagle Medium (Gibco) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin-streptomycin at 37 °C in a humidified incubator with 5% CO₂. HSV-1 F strain was kindly provided by Dr. Ying Zhang (Peking University). HSV-2 VR1779 strain was purchased from ATCC. All viruses were amplified in Vero E6 cells and titrated via the TCID₅₀. Preparation of mRNA The genes encoding glycoprotein D2 (gD2; GenBank No. NC_001798.2), glycoprotein C1 (gC1; GenBank No. NC_001806.2), and glycoprotein E1 (gE1; GenBank No. NC_001806.2) were codon-optimized and synthesized by Tsingke (Beijing, China). Truncated gD2, gC1, and gE1 were amplified by PCR and cloned into the pUC57-mRNA-101PolyA vector. The resulting plasmid DNA was linearized with PmeI restriction endonuclease (New England Biolabs, R056L) and purified by phenol:chloroform:isoamyl alcohol (25:24:1, pH > 8.0) extraction (Tsingke). Finally, mRNA was synthesized by IVT using the Transcription and Capping Kit (Vazyme), and its expression was verified both in vitro and in vivo. Western blot The cells and supernatant were collected. Cells were lysed with RIPA buffer (Beyotime) containing protease inhibitors(Roche) for 30 min on ice, and protein samples were denatured for 10 min at 99 °C. The cell lysates and supernatant were loaded to 12% SDS‒PAGE for electrophoresis and were transferred to 0.2 μm PVDF membrane. 5% defatted milk blocked the PVDF membrane for 2 h. PVDF membranes were incubated with primary antibody overnight at 4 °C and were incubated with secondary antibody at room temperature for 1 h. The western blot images were acquired using the Tanon chemiluminescence imaging system. β-actin was set as the internal reference gene. The primary antibodies: gC1 (a custom-generated mouse polyclonal antibody against gC1, dilution ratio 1:50), gD2 (Santa Cruz, mouse, dilution ratio 1:800), His (Proteintech, dilution ratio 1:1000, used for detecting the His-tagged gE1 protein)and β-actin (Proteintech, dilution ratio 1:2000). The secondary antibody: Goat Anti-Mouse IgG H&L (HRP) (Abcam, ab6789, dilution ratio 1:4000). Preparation LNPs LNPs were formulated by rapidly mixing an ethanolic lipid solution with an aqueous mRNA solution using the INano L (Micro&Nano Biologics). The lipid solution consisted of Dlin-MC3-DMA (Medchemexpress), DSPC (Sigma), cholesterol (Sigma), and 14:0 PEG2000 PE (Sigma) at a molar ratio of 50:10:38.5:1.5, with a total lipid concentration of 12 mM. The aqueous phase was prepared by diluting the mRNA in 50 mM citrate buffer (pH 4.0). The two phases were combined at a 3:1 flow rate ratio (aqueous to lipid) and a total flow rate of 20 mL min−1, achieving a final ionizable lipid to mRNA ratio (N/P) of 6:1. The resulting LNP suspension was then dialyzed against PBS, concentrated using a 10 kDa Amicon Ultra centrifugal filter (EMD Millipore, Billerica), passed through a 0.45 µm filter, and stored at 4 °C until use. The hydrodynamic diameter, polydispersity index, and zeta potential of the nanoparticles were measured using a Zetasizer Nano ZS90 (Malvern Panalytical). The encapsulation efficiency of the LNPs was determined using the Quant-iT RiboGreen RNA Assay Kit (Invitroge). Mouse immunization and challenge Different immunization strategies have been presented in the main text, and the immunization dose of 10 µg and the sample size were selected with reference to previously published literature46,47. Prior to the challenge experiment, mice were anesthetized as described above. Subsequently, they were inoculated intravaginally with either HSV-1 or HSV-2 at doses of 2.7 × 10⁵ (275 LD50) and 1.7 × 10⁵ PFU (275 LD50) per mouse, respectively. Neutralizing antibody detection Vero cells were seeded in 96-well plates at a density of 1 × 10⁴ cells per well and incubated for 24 h. Serum samples were subjected to two-fold serial dilutions in DMEM, starting from 1:20 to a final dilution of 1:2560. A virus stock was diluted in DMEM to 100 TCID₅₀/50 μL. Then, 60 μL serum dilution was mixed with 60 μL virus dilution, incubated at 37 °C for 1 h. Following incubation, 100 μL mixture was added to the pre-seeded 96-well plate. Blank controls (100 μL DMEM) and virus controls (100 TCID50 virus) were set up in parallel. The plates were incubated at 37 °C for 96 h, and cytopathic effects were observed. The neutralization titers were defined as the reciprocal of the serum dilution required for 50% neutralization of viral infection. Enzyme-linked immunosorbent assay (ELISA) 96-well plates were coated with inactivated HSV-1, HSV-2, or gD2 protein and incubated overnight at 4 °C. The plates were then blocked with 5% skim milk at 37 °C for 1 h. Serial twofold dilutions of mouse serum (starting at 1:100) were added and incubated at 37 °C for 2 h. Subsequently, a 1:100000 dilution of horseradish peroxidase (HRP)-conjugated anti-mouse IgG or a 1:5000 dilution of HRP-conjugated anti-mouse IgG1/IgG2a/IgG2b antibody was incubated at 37 °C for 1 h. After washing, TMB substrate was added and incubated for 15 min, and the reaction was terminated by the addition of 2 M sulfuric acid. The absorbance was measured at 450 nm and 630 nm using a microplate spectrophotometer, and the difference (OD450–OD630) was calculated for analysis. Flow cytometry detection 1 × 10⁶ cells per well were seeded in 96-well plates and stimulated with either gD2 peptide pool or inactivated HSV-1/HSV-2 for 1 or 6 h. GolgiPlug (BD Bioscience) was then added, and the cells were incubated for 6 h. Subsequently, cells were harvested and washed with MACS buffer (PBS containing 0.5% BSA and 5 mM EDTA-Na₂). Single-cell suspensions were blocked with an anti-CD16/32 monoclonal antibody and stained using the LIVE/DEAD staining kit (Invitrogen) for 30 min. After washing, surface markers were stained with specific antibodies for 30 min. The cells were then fixed and permeabilized using a commercial Fixation/Permeabilization solution for 20 min. Intracellular cytokines were then stained with specific antibodies for 1 h. Finally, cells were fixed with 4% paraformaldehyde for 30 min. All staining steps were performed at 4 °C. Flow cytometry data were acquired on a BD LSRFortessa (BD Biosciences) and analyzed using FlowJo v.10. Details of the staining panel are provided in Supplementary Table 1–4. Gating strategies for different cell populations are shown in Supplementary Fig. 3. ELISpot assay Multiscreen HTS-IP filter plates (Millipore) were coated with anti-mouse IFN-γ antibodies according to the manufacturer’s instructions (Mabtec). The resuspended splenocytes were plated at a density of 1 × 106 cells/well. GD2 peptides were diluted in RPMI 1640 medium and added to the plate. Negative (RPMI 1640 medium) and positive controls (Cell Stimulation Mix; Thermo Fisher Scientific) were added to each assay. Plates were incubated for 48 h at 37 °C and 5% CO2 and then were incubated by adding biotinylated anti-mouse IFN-γ antibodies, HRP-conjugated streptavidin, and TMB substrate. Spots were counted using an Immuno-spot analyzer (Cellular Technology Ltd.). The results are expressed as the number of spot-forming cells in 1 × 106 cells. Viral titer determination of vaginal swabs Vero cells were seeded in 96-well plates at a density of 2.5 × 10⁴ cells per well and cultured overnight. Collected vaginal swabs were kept at 4 °C for 2 h and vortexed moderately. The samples were subjected to tenfold serial dilutions from 10−1 to 10−6, with 8 replicate wells for each dilution. Subsequently, 50 μL of each diluted sample was added to the pre-seeded 96-well plates. After 4 days of incubation, a cytopathic effect was observed, and the number of positive wells was recorded. The viral titers were calculated using the Spearman–Kärber method. RNA-seq analysis Draining lymph node cells were harvested from mice 24 h post-primer vaccination for RNA-seq. RNA-seq was performed by Shanghai Biotechnology Corporation. Raw reads were processed through quality filtering and subsequently aligned to the mouse reference genome (version mm10). Gene expression levels were quantified as Fragments Per Kilobase of transcript per Million mapped reads for each sample. Differential expression analysis was performed using a statistical model that included read count normalization, model-dependent P-value estimation and false discovery rate calculation based on multiple hypothesis testing. DEGs were defined as those with |log₂(fold change)| ≥ 1 and adjusted P ≤ 0.05. Pathway enrichment analysis of DEGs was conducted using Reactome and KEGG databases within R Studio (v3.0.3), with pathways showing an adjusted P ≤ 0.05 considered statistically significant. BMDC isolation, culture, and stimulation BMDCs were isolated from 6 to 8 weeks old female BALB/c mice and cultured as previously described48. Briefly, the femurs and tibias were dissected, and bone marrow was flushed out with PBS. The cell suspension was filtered through a 45 μm cell strainer, followed by red blood cell lysis using red blood cell lysis buffer. After centrifugation at 1200 rpm, the supernatant was discarded, and the cell pellet was resuspended in BMDC culture medium (RPMI 1640 supplemented with 10% FBS, 1% penicillin-streptomycin, 55 μM β-mercaptoethanol, and 20 ng/ml GM-CSF (Biolegend)). Cells were seeded at 1 × 107 cells per 90 mm bacterial culture dish and designated as day 0. On day 3, 8 ml of BMDC culture medium was added. On day 5, half of the medium was replaced. On day 7, BMDCs were harvested and seeded in 24-well plates at a density of 1 × 10⁶ cells per well and cultured overnight. The cells were treated with 250 ng LNP-gD2, 250 ng LNP-gD2 + 250 ng LNP-gC1, or 250 ng LNP-gD2 + 250 ng LNP-gE1, respectively. Cells were harvested 8 h post-treatment for subsequent assays. Quantitative real-time PCR DNA was extracted from vaginal swabs and DRG using a DNA extraction kit (Sangon). RNA was extracted from cells using an RNA isolation kit (Sparkjade). Total RNA was reverse transcribed into first-strand cDNA using the Hifair® III 1st Strand cDNA Synthesis SuperMix kit (Yeasen). Quantitative Real-time PCR was performed using the SYBR Premix kit (Yeasen) and analyzed using a Roche 480 Real-Time PCR System (Eppendorf). The primers used in this study are listed in Supplementary Table 5. Statistical analysis Statistical analyses were performed using GraphPad Prism 9.5.0. Data are presented as mean ± SEM. For multiple group comparisons, one-way analysis of variance (ANOVA) or two-way ANOVA was used, followed by Tukey’s multiple comparisons test. A P < – value 0.05 was considered statistically significant. Supplementary information Supplementary Information. Acknowledgements We thank Dr. Ying Zhang at Peking University for kindly providing the HSV-1 F strain. This research was supported by the National Natural Science Foundation of China (82241065) to D.Z. Author contributions D.Z., W.D., M.X., W.Z., and L.S. conceived the project and designed the experiments; W.Z. and L.S. performed experiments; W.Z., L.S., P.W., Q.Z., Ying Li, Yang Li, and H.S. contributed to methodology; D.Z. acquired funding; D.Z., W.D., M.X., W.Z. and L.S. analyzed the data; W.Z. drafted the manuscript; D.Z., W.D. and M.X. revised the manuscript; W.Z. and L.S. are co-first authors of this manuscript. All authors read and approved the final manuscript. Data availability The RNA sequencing data in this study have been deposited in the Gene Expression Omnibus (GEO) database under accession number: GSE322599. The data that support the findings of this study are available from the corresponding author upon request. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Wenying Zhao, Lingjin Sun. Contributor Information Man Xing, Email: xingman@tmu.edu.cn. Weiqian Dai, Email: 18962783896@163.com. Dongming Zhou, Email: zhoudongming@tmu.edu.cn. Supplementary information The online version contains supplementary material available at 10.1038/s41541-026-01450-8. References Associated Data Supplementary Materials Supplementary Information. Data Availability Statement The RNA sequencing data in this study have been deposited in the Gene Expression Omnibus (GEO) database under accession number: GSE322599. The data that support the findings of this study are available from the corresponding author upon request.