Antagonizing Viral MicroRNAs Reduces Ocular HSV-1 Pathogenesis and Enhances Mucosal Immune Homeostasis

Abstract

Purpose

Herpes simplex virus 1 (HSV-1) is a globally prevalent pathogen that causes recurrent lesions at mucosal and cutaneous sites, including the cornea, leading to herpetic keratitis, a major cause of infectious blindness. While HSV-1–encoded microRNAs (v-miRs) are known to regulate viral latency and immune evasion, their role in acute mucosal infection remains unclear. This study investigates the function of v-miRs during acute HSV-1 infection of the cornea.

Methods

Using a murine model of corneal HSV-1 infection, we performed RNA sequencing and in situ hybridization to identify v-miRs enriched in the cornea during acute infection. Topical locked nucleic acid (LNA)–modified inhibitors targeting individual v-miRs (miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27) were administered, and effects on disease severity, viral replication, immune response, and lymphoid cell activation were assessed.

Results

Inhibition of select v-miRs significantly attenuated corneal keratitis, reduced viral titers, and suppressed Th1/Th17-mediated inflammation. v-miR inhibition also decreased immune cell infiltration in draining lymph nodes and enhanced the frequency of IL–10–producing CD4⁺ T cells. Expression of immunoregulatory genes, including Arg1 and CD25, was increased, and T-cell proliferation was reduced ex vivo, indicating the establishment of a tissue-protective immune environment.

Conclusions

HSV-1 v-miRs act as key immunoregulatory effectors during acute corneal infection. Their targeted inhibition using LNA-based therapy mitigates inflammation and promotes immune regulation, highlighting a novel therapeutic strategy for ocular HSV-1 disease.

Herpes simplex virus 1 (HSV-1) is a ubiquitous, double-stranded DNA virus (∼152 kb) that persists inside the host as a latent infection with intermittent reactivation.^1–4 Globally, an estimated 3.7 billion people (∼67%) are seropositive for HSV-1 infection.^5,6 Primary HSV-1 infection generally causes cold sores but remains a predominant cause of ocular morbidity and blindness. Despite the availability of antiviral drugs such as acyclovir and its analogues, the emergence of drug-resistant keratitis and the limited success of new HSV-1 vaccines for ocular keratitis underscore the urgent need to develop novel therapeutic targets to prevent disease progression and reduce associated morbidity.^7–9 HSV-1, similar to other human herpesviruses (HHVs), maintains lifelong latency inside the host.^1,10 The only manifestations of viral gene expression during latency are the accumulation of a noncoding transcript and a set of viral microRNAs (v-miRs). These small RNAs have been demonstrated to control expression of both viral and host transcripts, thereby regulating viral tropism, lytic switching, immune subversion, and so on.^11–14 Various studies have examined v-miR profiles in cell lines and HSV-1–infected mice; however, their expression profiles and function targeting the cornea have not been investigated.

HSV-1 encodes 25 mature miRs, and their expression varies depending on the infected cell and tissue types. These miRs regulate various aspects of the virus life cycle, such as immune evasion and suppression. For instance, HSV-1 encodes miR-H2, miR-H6 target viral ICP4, and ICP0 RNA to promote latency by preventing expression of IE or E genes. Two different HSV1 miRs (miR-H3 and miR-H4) downregulate the expression of lytic gene ICP34.^2,11,15–17 Viruses are solely or predominantly dependent on host machinery for their replication. KLHL24, a critical transcription factor that activates viral lytic genes (IE and E), is targeted by HSV-1 miR-H27.¹⁸ This will inhibit transcriptional activation of lytic phase–associated genes. HSV-1 miR-H8 binds to and downregulates host gene PIGT, leading to dysregulated expression of several different immune-related, glycosylphosphatidylinositol (GPI)-anchored proteins and hence evades immune cell–mediated clearance of infected cells.¹³ Most studies have focused on studying v-miRs’s role in cultured cells or latency in ganglia, but our emphasis is on severe clinical manifestations of HSV-1 infection in corneal tissues. In the cornea, the virus infects epithelial cells, leading to tissue scarring, which in turn causes partial or sometimes complete loss of vision. Identifying viral miRNAs expressed in ocular tissues during HSV-1 infection can yield targets to control disease progression.

HSV-1 persists as a subclinical, recurrent infection for an individual's lifetime; the successful evasion of various host immunosurveillance mechanisms is reasonably expected during infection. Several HSV-1 encoded proteins have been demonstrated to modulate host immune response by targeting specific innate and adaptive immune regulators. Two principal immune subsets—myeloid and lymphoid—sequentially participate in mounting an antiviral response and virus clearance. Upon virus infection, the first wave of inflammatory cells, primarily comprising myeloid cells (Mφ/dendritic cell [DC]), infiltrates the infected site, followed by a second wave of immune infiltrate containing T and B cells. HSV-1 interferes with the host defense mechanisms via suppression of HSV-1–specific immune effector function and polarization of immune cells. For instance, HSV-encoded proteins UL42 and ICP0 interact with p65/RelA and p50/NF-κB1, and they subsequently block the nuclear translocation of NF-κB, thereby resulting in reduced antiviral cytokine production.^19,20 Also, HSV-1 can interfere with the ability of myeloid and lymphoid cells to function.^21,22 V-miRs play a critical role in viral pathogenesis because they target host and viral transcripts and thus shape host–virus interaction.^23,24 However, the in vivo contribution of HSV-1 v-miRs in shaping corneal immunity is poorly studied.

This study investigates the immunomodulatory role of HSV-1 microRNAs (miRNAs) during acute primary infection in vivo by assessing the effects of their inhibition. V-miRs enriched in corneal tissues after HSV-1 infection were inhibited using custom-designed locked nucleic acid (LNA) inhibitors, and their impact on the progression of ocular keratitis was evaluated. Finally, we performed comprehensive immune profiling to assess how v-miR inhibition restores tissue homeostasis to promote disease resolution. These findings identify unique HSV-1 miRNAs that can be targeted to mitigate HSV-1 ocular infection.

Methods

Cells and Viruses

The HSV-1 (McKrae) strain used in this study to infect the mouse eye was provided by Prof. Homayon Ghiasi's laboratory at Cedars Sinai, Los Angeles, CA. The stock of HSV-1 (McKrae) was propagated and titered on Vero cells and stored at −80°C.

Animal Model of Infection and Topical Delivery of v-mir Inhibitors in the Mouse Eye

A total of n = 60 male C57BL6 mice (aged 6–8 weeks) were obtained from Charles River Laboratories (Wilmington, MA, USA) and housed at the animal facility of the University of Illinois Chicago. The studies were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Before HSV-1 infection, all mice were first anesthetized (using ketamine [100 mg/kg] and xylazine [5 mg/kg]), and then their right eyes were anesthetized with proparacaine hydrochloride ophthalmic solution (Alcon Laboratories, Fort Worth, TX, USA) for 30 seconds. Loss of toe pinch/pedal withdrawal was used to determine the efficacy of the anesthesia. Next, as we have reported previously, we performed the epithelial debridement of the right eye with a 30-gauge sterile needle, and 1 × 10⁵ plaque-forming units (PFU)/eye of HSV-1 (strain McKrae) viruses, unless stated otherwise, were overlaid on it in a total volume of 5 µL. To test the efficacy of HSV-1–specific miRNA inhibitors to control ocular HSV-1 infection, these vmiR inhibitors were mixed with Lipofectamine 2000, diluted in 1× PBS, and added to HSV-1–infected mouse eyes starting at 3 hours postinfection, three times a day until 7 days postinfection (dpi). Ocular washes were collected at the end of each day to elucidate the virus shedding, and stereoscopic images (Carl Zeiss Microimaging GmBH, Gottingen, Germany) of HSV-1–infected C57BL6 mice and noninfected mice were taken at days 2, 4, and 8 postinfection.

Small RNA Library Construction and Sequencing

Total RNA was isolated from samples using Qiazol reagent (Qiagen, Gaithersburg, MD, USA) according to the manufacturer's instructions, ensuring efficient total RNA extraction, including small RNAs. The integrity and concentration of the extracted RNA were then assessed using the Bioanalyzer 2100 system (Agilent Technologies, Inc., Santa Clara, CA, USA). Samples with an RNA Integrity Number (RIN) greater than 7 were selected for further analysis. For small RNA library construction, 1 µg total RNA from each sample was processed using the TruSeq Small RNA Sample Prep Kit (Illumina, San Diego, CA, USA), which is designed to enrich and prepare small RNA species for sequencing as described previously.¹⁴ Following the manufacturer's protocol, the prepared libraries were subjected to single-end 50-bp sequencing on an Illumina HiSeq 2500 platform at LC Sciences (Hangzhou, China). For viral miRNA detection, sequences were aligned to HSV-1 mature miRNA sequences procured from miRbase (https://www.mirbase.org/). We compiled a list of v-miRs from two different RNA sequencing performed on corneal and whole eye tissues.

miRNA Quantification by Quantitative RT-PCR

Mature miRNA levels were quantified using miScript primers and the miScript II RT kit, as described before.²⁵ Total RNA (200 ng) was reverse transcribed and amplified with miRNA-specific and universal primers (Qiagen) using SYBR Green master mix (Roche, Indianapolis, IN, USA) on an ABI Prism 7500 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). RNU6 B served as the internal control. Relative fold changes were calculated using the ΔΔCt method, and the normalized values were plotted as histograms with standard deviations.

miRNA In Situ Hybridization

miRNA in situ hybridization was performed on formalin-fixed, paraffin-embedded (FFPE) whole eye sections using double DIG-labeled LNA probes (Qiagen miRCURY LNA miRNA in situ hybridization [ISH]).²⁶ Tissue sections were cut at a 5-µm thickness, mounted on poly-L-lysine–coated glass slides, and dried at 60°C for 30 minutes to ensure optimal adhesion. The slides were dewaxed in xylene (2 × 10 min) and rehydrated through a graded ethanol series (100%, 95%, 70%, 50%, 5 minutes each) to distilled water, preparing the tissue for enzymatic treatment. To enhance probe accessibility, sections were incubated with Proteinase K at 37°C for 10 minutes in a hybridizer. After washing with PBS, hybridization was carried out by incubating the sections for 1 hour at 60°C with double DIG-labeled miRCURY LNA miRNA probes (20 nM) targeting miR-H1-3p and miR-H6-3p (Qiagen), using a hybridizer with Milli-Q water to maintain humidity. We used a scrambled miRNA probe as a negative control to validate the staining procedure and demonstrate staining specificity. Following hybridization, stringent washes at the hybridization temperature removed unbound probes. The slides were then placed in a humidified chamber and incubated with a blocking solution for 15 minutes at room temperature to reduce nonspecific binding. After removing the blocking solution, sections were treated with anti-DIG reagent (Roche, Mannheim, Germany) for 60 minutes at room temperature to enable probe detection. The slides were washed with PBS-T and incubated with AP substrate (Roche) for 2 hours at 30°C, protected from light to allow for color development. After a final wash with KTBT buffer, the sections were counterstained with hematoxylin, dehydrated through ethanol and xylene, and mounted using Eukitt mounting medium (EMS, Hatfield, PA, USA). Microscopy images were acquired using an Aperio ImageScope (Leica Biosystems, Deer Park, IL, USA), enabling visualization and analysis of miRNA localization within the tissue. To measure the signal intensity, raw images were captured with the same exposure and gain settings from all slides and saved as TIF files, and the signal was quantified using ImageJ (National Institutes of Health, Bethesda, MD, USA) analysis. Integrated density, the sum of the pixel values within a selected area, representing the total signal intensity, was calculated for each viral miRNA or scramble in uninfected and infected corneal tissues.

Isolation of Cornea, Extraction of Total RNA, and Quantitative RT-PCR

To investigate the presence of HSV-1 transcripts and associated transcripts of host antiviral genes within the corneal samples, whole eyes were removed from the euthanized HSV-1–infected mice at 4 dpi and 8 dpi time points. Briefly, the entire eye was placed on a sterile petri dish. Then, the cornea was removed using disinfected scissors and forceps under the microscope and kept in 700 µL QIAzol at −80 °C until further analysis. Before RNA isolation, the cornea samples were sonicated (5-second pulses at 20% amplitude for 30 seconds), then subjected to RNA isolation by the RNeasy Mini Kit (Qiagen) according to the manufacturer's protocol and quantified using NanoDrop (Thermo Fisher Scientific, Wilmington, DE, USA). After that, all RNA samples were equilibrated with Molecular Biology Grade Water (Corning, Corning, NY, USA) before they were reverse transcribed into cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). After that, real-time quantitative polymerase chain reaction was performed using Fast SYBR Green Master Mix on the QuantStudio 7 Flex system (Applied Biosystems) for transcript quantitation. The primers used in this study are as follows: ICP0 F (forward), 5′-ACAGACCCCCAACACCTACA-3′; ICP0 R (reverse), 5′-GGGCGTGTCTCTGTGTATGA-3′; gD F (forward), 5′-TACAACCTGACCATCGCTTC-3′; gD R (reverse), 5′-GCCCCCAGAGACTTGTTGTA-3′; IFNγ (forward), 5′-TCAGGCCATCAGCAACAACA-3′; IFNγ (reverse), 5′-GTGGACCACTCGGATGAGC-3′. ΔCt value, the difference between the cycle number of experimental conditions and their controls, was first normalized by their corresponding Ct values of the β-actin gene to get ΔΔCtv in three technical replicates, and then fold change was calculated using the 2^−ΔΔCt method.

Ocular Pathology Monitoring

After HSV-1 ocular infection, periocular disease progression in C57BL6 mice was monitored. The scoring was blinded according to pathological symptoms over time.^27,28 HSV-1–infected mice were scored on days 2, 4, and 8 for ocular pathological symptoms postinfection with the HSV-1 McKrae strain. The cumulative score for corneal opacity, blepharitis, and periocular lesion was calculated on a scale of 0 to 5, with 0 = no disease and 5 = severe disease.

Plaque Assays

Serially diluted eye samples were subjected to plaque assay, as described previously by our group.^14,29 Briefly, eye washes were obtained from HSV-1–infected mice at 2, 3, 4, 5, and 6 dpi in 10 µL PBS and added to 90 µL PBS to get a 100-µL eye wash sample per mouse. These samples were serially diluted and overlaid on the monolayer of Vero cells, 250 µL of each dilution in a 24-well plate. After 2 hours, the cells were washed with PBS and then overlaid with 5% methylcellulose (Sigma-Aldrich, St. Louis, MO, USA) made in Dulbecco's modified Eagle's medium solution, and the plates were incubated at 37°C (5% CO₂) for 72 hours until the appearance of the plaques. Then, 250 µL of 100% methanol was added to fix these cells for 10 minutes. Afterward, the cells were stained with 250 µL crystal violet solution for 30 minutes. The plaques were then manually counted after the aspiration of the solution and multiplied by the appropriate dilution factor to obtain the PFU/mL in each well of a 24-well plate.

Lymph Node Isolation

Our previously published protocol isolated lymph nodes.^28,30 After euthanizing mice, the superficial and deep cervical lymph nodes from v-miR inhibitor-treated and untreated mice were isolated and homogenized in complete RPMI media containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Next, the homogenate was filtered with a 40 µm cell strainer (Falcon, Fisher Scientific Company LLC, Hanover Park, IL, USA) and spun at 500g for 5 minutes at 4°C. The supernatant was discarded, and the cells were resuspended in 1 mL of cell staining buffer (BioLegend Incorporated, San Diego, CA, USA). Before staining, the total cell number was counted in a hemocytometer, and cell staining for flow cytometry analysis was performed with 10⁶ cells.

Flow Cytometry

The lymph node cells were washed twice with PBS supplemented with 1% (v/v) BSA. For surface staining, FITC anti-mouse CD4 antibody (clone: RM4-5, BioLegend), PerCP/Cyanine5.5 anti-mouse CD8a antibody (clone: 53-6.7, BioLegend), PerCP anti-mouse CD11c antibody (clone: N418, BioLegend), and Pacific Blue anti-mouse I-A/I-E antibody (clone: M5/114.15.2, BioLegend) were added to the cells in 200 µL PBS/1% BSA (w/v) and incubated on ice for 45 minutes. After that, the cells were washed two times with PBS/1% BSA, and data were acquired on a CytoFLEX S (Beckman Coulter, Inc., Brea, CA, USA) Flow Cytometer. For intracellular staining with IFNγ, IL-17, and IL-10, the cells were first fixed with 2% paraformaldehyde (on ice for 30 minutes) and then permeabilized with 0.3% saponin (at room temperature). After that, PerCP/Cyanine5.5 anti-mouse IFN-γ antibody (clone: XMG1.2, BioLegend), Pacific Blue anti-mouse IL-17A antibody (clone: TC11-18H10.1, BioLegend), and APC anti-mouse IL-10 antibody (clone: JES5-16E3, BioLegend) were added to the cells and incubated on ice for 1 hour. The final staining volume was 200 µL PBS/1% BSA (w/v). After that, the cells were washed, and data were acquired in the CytoFLEX S Flow Cytometer. FlowJo_v10.8.1 software (Tree Star, Ashland, OR, USA) was used to analyze the flow cytometry data.

Flow Sorting and T-Cell Proliferation Assay

We used a previously published protocol for the coculture of DCs and CD4⁺ T cells.^31,32 Sorted DCs (CD11c⁺HLA-DR⁺) were seeded (50,000/well) in 96-well round-bottom plates and added with sorted. Carboxyfluorescein succinimidyl ester (CFSE)–labeled autologous CD4⁺ T cells were added to DCs in a 1:2 ratio in the complete RPMI media (10% FBS and 1% penicillin-streptomycin) containing Ova peptide (10 µg/mL; grade V, Sigma-Aldrich) and IL-2 (50 ng/mL) following the previously published protocol. After 5 days, these cocultures were harvested, and the percentages of CD4⁺CFSE⁺ and CD4⁺CFSE− cells were analyzed on a flow cytometer (CytoFLEX). Data were analyzed on FlowJo_v10.9.0.

Statistical Analysis

The statistical analyses were performed using GraphPad Prism software (version 4.0; GraphPad Software, La Jolla, CA, USA). One-way ANOVA tests were used to compare differences in the noninfected, HSV-1–infected, and HSV-1–infected and treated groups. Significance was defined based on exceeding different P value thresholds: ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, and ∗∗∗∗P < 0.0001.

Results

HSV-1–Derived miRNAs Are Highly Expressed During the Active Phase of Ocular Herpes Infection

Various studies have examined HSV-1 miR profiles in cell lines and HSV-1–infected mice. However, their expression in ocular disease progression, contribution to ocular herpes pathogenesis, and therapeutic targeting to augment disease resolution remain understudied. To identify the set of corneal tissue–expressed HSV-1 miRNAs, we examined the dynamics of the HSV-1 miRNA repertoire in a murine model of ocular herpes. Mice were infected with HSV1 McKrae or mock-infected, and whole eyes were collected at 2, 4, 6, and 10 dpi (n = 3–4/group). Figure 1A presents a schematic overview of the experimental design used for mouse infection studies. Our model showed progressive inflammation and damage to ocular tissues, as observed by Stereoscope imaging (Fig. 1B). Small RNA sequencing identified multiple v-miRs, including miR-H1-5p, miR-H1-3p, miR-H2-5p, miR-H4-3p, miR-H5-3p, miR-H6-3p, and miR-H27, at one or more time points (Fig. 1C). Most of the v-miRs exhibited a time-dependent increase, reaching the peak expression at 4 or 6 dpi. The heatmap illustrates the kinetics of v-miR expression profiles at various time points in HSV-1–infected mice compared to their mock-infected controls (Fig. 1C). These results clearly show unique enrichment of v-miRs in ocular tissues and signify their role in viral replication and persistence.

Figure 1.

Multiple HSV-1–encoded miRNAs are detected in the infected ocular tissues. (A) Schematic representation of the experimental design outlines the workflow for the mouse infection study to investigate miRNA expression dynamics following HSV-1 infection. (B) Mice were infected with HSV-1 McKrae (10⁵ PFU each mouse). Infection images were captured at different time points (2, 4, and 8 dpi), using a V20 stereoscope (7.5×, Carl Zeiss). Representative HSV-1–infected eyes showing disease progression. (B) Total RNA isolated from HSV-1 or mock-infected ocular tissues and subjected to small RNA sequencing. (C) The heatmap shows differentially expressed v-miRs during HSV-1 infection compared to the mock control. (D) Validation of v-miR expression in the mouse cornea (8 dpi) by RT-qPCR. Low Ct values indicate higher expression of genes and vice versa. RNU6B was used as a housekeeping gene for this analysis. Each bar represents the mean ± SD. (E) Expression of HSV-1 miRNAs increases after reinfection of ocular tissues. Total RNA was isolated from whole eye or blood from infected animals, and RT-qPCR was used to examine the expression of different miR-H1-5p, miR-H6-3p, miR-H17, and miR-H27. Low Ct values indicate higher expression. RNU6B was used as an endogenous control.

To validate our RNA sequencing results, we examined the expression of nine v-miRs by quantitative RT-PCR (RT-qPCR). We included miR-H3-3p, miR-H7-3p, and miR-H17 based on their roles in the literature and our inflamed oral gingiva data sets. Our results show a time-dependent increase in v-miR expression, which peaked around days 4 to 6 postinfection. Because none of the v-miRs were detected in mock-infected animals, the data are presented as mean Ct values for the following v-miRs detected: miR-H1-5p (20.3 ± 1.15), miR-H2-5p (18.2 ± 0.47), miR-H3-3p (21.9 ± 0.40), miR-H4-3p (23.2 ± 0.57), miR-H5-3p (23.3 ± 0.70), miR-H6-3p (18.8 ± 1.6), miR-H7-3p (23.3 ± 0.72), miR-H17 (27.1 ± 0.90), and miR-H27 (24.6 ± 0.55), with disease progression on 4 dpi (Fig. 1D). As a control, we analyzed the expression of endogenous small noncoding RNA RNU6B and observed similar Ct values in mock and HSV-1–infected tissues. Together, these results validate our RNA sequencing data and show higher expression of v-miRs in the cornea.

Recurrent HSV-1 infection facilitates lifelong viral persistence by periodic reactivation and repeated disease episodes. This can lead to cumulative tissue damage, chronic inflammation, and increased risk of severe complications such as keratitis and encephalitis.⁴ To study the impact of recurrent infection on viral miRs, mice were infected with HSV-1 McKrae strain once (primary infection; in the right eye) or twice (in the left eye; reinfection after 3 days, same concentration). Animals displayed inflammation, as evidenced by reduced eyeball size, and were sacrificed after 3 days. Total RNA from the whole infected eyes was isolated, and the expression of four v-miRs (miR-H1-5p, miR-H6-3p, miR-H17, and miR-H27) was examined by RT-PCR (Fig. 1D). All the v-miRs were detected at high levels (Ct values ranging from 17–22) in ocular tissues. Expression of miR-H1-5p, miR-H6-3p, and miR-H17 was detected at higher levels at primary infection, and their levels were also detectable after reinfection, while endogenous control RNU6B did not exhibit significant variation (Ct value ranging from ∼21–22; Fig. 1E, lower right panel). Intriguingly, expression of v-miRs was also detected in the blood sample (low Ct values ranging from ∼12–18), implying their secretion into the blood, leading to dissemination of virulence factors. These results show that v-miRs are highly expressed in infected tissues and that viral miRNAs are systemically detected at much higher levels, suggesting considerable viral load in the blood.

HSV-1 miRNAs Exhibit Differential Expression in the Tissue, Suggesting Regulatory Roles During Infection

To further validate our RT-qPCR results and assess the corneal profiles of v-miRs, we examined the expression pattern of miR-H1-5p and miR-H6-3p by ISH. FFPE murine corneal samples were infected with HSV-1 at 10³ and 10⁷ PFU for 4 days. Mice infected with a high dose of HSV-1 (10⁷ PFU) showed an impaired epithelium histology compared to mice infected with a low dose of HSV-1 and uninfected control mice. Compared to low PFU, higher doses of HSV-1 caused thinning of the corneal epithelium, increased immune cell infiltration, and disruption of connective tissue. miR-H1-5p and miR-H6-3p were detected in the cytoplasm (as seen in magnified cells in the red boxes) of the corneal epithelium in all the mice infected with HSV-1 (Fig. 2A). We noticed higher expression levels of miR-H1-5p compared to miR-H6-3p, corroborating our RT-qPCR results. Uninfected mice showed no signal with miR-H1-5p and miR-H6-3p probes, indicating signal specificity in HSV-1–infected tissues. Scramble control showed no staining in the infected tissues, indicating signal specificity observed with v-miR probes.

Figure 2.

In situ hybridization exhibits differential expression pattern of HSV-1 miRs in infected cornea. (A) Representative images of miR-H1-5p and miR-H6-3p expression analyzed by ISH in the mouse corneal epithelium infected with HSV-1 (10³ and 10⁷ PFU) at 5 dpi. Inset shows a red boxed enlarged area of the sample. We use scrambled sequences as negative controls to assess signal specificity. Final magnification: ×40. Scale bar: 60 µm. (B) Bar graphs show quantification of probe signal intensity presented as the average integrated density for scramble, miR-H1-5p, and miR-H6-3p in uninfected and HSV-1–infected conditions (low dose: 10³ PFU or high dose: 10⁷ PFU). ImageJ was used for integrated density calculation. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001.

To quantify the tissue expression of miR-H1-5p and miR-H6-3p, we examined the intensity in the selected region using ImageJ analysis. Our results show significantly higher signal intensity for miR-H1-5p (120.3: low PFU; 132.5: high PFU) and miR-H6-3p (144.6: low PFU; 186.1: high PFU) in HSV-1–infected mice, although their expression was even higher at later doses (Fig. 2B, right panel). These results show that HSV-1 miRNAs are expressed in the corneal epithelium but are also noticeably detected in the underlying connective tissue, likely due to epithelium damage.

Topical Inhibition of Viral miRNAs Interferes With the Progression of Ocular HSV-1

Candidate HSV-1 v-miRs identified from in vivo screening were selected as potential targets by topical delivery of synthetic ribonucleotides to sequester v-miRs and assess their impact on the HSV-1 ocular keratitis murine model. We have developed and optimized the delivery of LNA-modified small synthetic RNA to deliver v-miR inhibitors efficiently and reproducibly in murine cornea. A key aspect of this delivery method is high levels of bioactive molecules due to LNA protection from RNases and high specificity toward the target sequence. Significantly, due to conservation of miRNA sequences across HSV-1 strains, the same inhibitor sequence can functionally suppress v-miRs.

We next investigated the impact of inhibiting nine candidate v-miRs (miR-H1-5p, miR-H2-5p, miR-H3-3p, miR-H4-3p, miR-H5-3p, miR-H6-3p, miR-H7-3p, miR-H17, and miR-H27) on HSV-1 pathogenesis (replication and disease manifestation) in the mouse eye. Virus-infected mice were treated separately with the scramble or v-miR inhibitors three times daily until day 7, and the disease progression was monitored daily until day 8.

The extent of disease was monitored by imaging the corneal surface using a V20 stereoscope (Carl Zeiss) on 2, 4, and 8 dpi. Compared to scramble control, topical delivery of v-miR inhibitors targeting miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 showed minor signs of inflammation at the periorbital region at both 4 and 8 dpi, suggesting delayed progression of ocular keratitis. On the contrary, mice treated with inhibitors targeting miR-H2-5p, miR-H4-3p, miR-H5-3p, or miR-H7-3p did not exhibit any alleviation in disease progression as compared to the control inhibitor-treated mice (Fig. 3A). We also assessed disease severity scoring over 8 days. Congruent with the above data, we observed delayed progression of disease in the mouse eye treated with inhibitors targeting miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 (Fig. 3B; Supplementary Fig. S1).

Figure 3.

Inhibition of HSV-1–encoded miRNAs controls the progression of ocular infection. Mice were infected with HSV-1 (strain McKrae; 10⁵ PFU), and the animals were treated with v-miR (miR-H1-5p, miR-H2-5p, miR-H3-3p, miR-H4-3p, miR-H5-3p, miR-H6-3p, miR-H7-3p, miR-H17, and miR-H27) or scramble inhibitors. (A) Representative images showing HSV-1–induced ocular disease progression in the eyes of mice captured by a V20 stereoscope (Carl Zeiss). Red squares show the inhibition of disease progression upon topical delivery of v-miR inhibitors. (B) Graphs presenting disease score (0-5) indicating progression and severity after topical delivery of HSV-encoded v-miR inhibitors. RT-qPCR showing the expression of HSV-1 transcripts ICP0 (C) and gD (D) on 8 dpi in the mouse cornea. β-Actin was used as an endogenous control. Viral miRNA inhibition reduced virion production in HSV-1–infected mice. Eyewashes were collected at 3 and 4 dpi after topical delivery of v-miR inhibitors. (E) Representative images showing scans of the plaque assays. (F, G) Graphs showing PFU/mL in the supernatant at different dpi. Plaques were manually counted and multiplied by the appropriate dilution factor to obtain the PFU/mL. Each bar represents mean ± SD. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001.

To further evaluate the impact of v-miR inhibitors on HSV-1 replication, we quantified HSV-1 transcripts in the infected mouse cornea treated with different v-miR inhibitors at 8 dpi. Interestingly, there was a significant reduction of ICP0 in the corneas of mice treated with v-miR inhibitors miR-H1-5p (0.54 ± 0.19; P < 0.5), miR-H3-3p (0.2 ± 0.09; P < 0.001), miR-H6-3p (0.57 ± 0.12; P < 0.01) and miR-H27 (0.6 ± 0.10; P < .01). Similarly, the expression of HSV-1 glycoprotein D (gD) was also decreased in the mice cornea after the delivery of the following v-miR inhibitors: miR-H1-5p (0.79 ± 0.10; P < 0.05), miR3-3p (0.68 ± 0.16; P < 0.05), miR-H6-3p (0.36 ± 0.14; P < 0.001), and miR-H27 (0.50 ± 0.11; P < 0.01) (Figs. 3C, 3D).

To examine the virion release, eyewashes were collected from mice treated with scramble or v-miR inhibitors on multiple days postinfection, and plaque assays were performed to assess virion production. Our results show progressive and significantly reduced plaque formation, reflecting the fewer HSV-1 virions in miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 inhibitor-treated mice compared to control (Figs. 3E–G). Corroborating this, mice treated with miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 inhibitors showed markedly smaller lymph nodes compared to the control inhibitor, indicating robust viral clearance (Supplementary Figs. S2A, S2B). These findings, for the first time, suggest the protective role of HSV-1–specific v-miR inhibitors in ocular infection.

HSV-1–Derived miRNA Inhibitors Attenuate Immune Cell Infiltration in the Cervical Lymph Nodes HSV-1

Specific tissues like the cornea and brain are devoid of the lymphatic vascular system and therefore considered immune-privileged regions.^33–35 During its replication in the cornea, HSV-1 induces keratitis and lymphangiogenesis. Typically, lymphatic vascular systems facilitate the delivery of antigen(s), cytokines, antigen-loaded APCs, and T cells to the draining lymph nodes from sites of inflammation.^36,37 Therefore, we collected the drainage lymph nodes and interrogated for the infiltration of CD11c⁺I-A/I-E⁺ dendritic cells, key antigen-presenting cells, and CD8⁺ and CD4⁺ T cells in HSV-1–infected mice.

Compared to control mice (3.4% ± 0.66%), we observed a significant decrease in the percentage of CD8⁺ T cells in the cervical lymph nodes obtained from mice eyes treated with the inhibitors of miR-H1-5p (1.1% ± 0.19%; P < 0.005), miR-H3-3p (1.2% ± 0.35%; P < 0.005), miR-H6-3p (1.8% ± 0.17%; P < 0.005), and miR-H27 (2.4% ± 0.76%; P < 0.005) at 4 dpi (Fig. 4B). However, treatment with inhibitors targeting miR-H2-5p (4.8% ± 0.42%), miR-H4-3p (2.0% ± 0.007%), miR-H5-3p (2.8% ± 0.42%), miR-H7-3p (3.5% ± 0.67%), and miR-H17 (4.1% ± 0.66%) did not exhibit significant differences (Supplementary Fig. S3). A similar trend of reduced CD8⁺ T cells counts was also observed at 8 dpi in miR-H1-5p (8.1% ± 1.5%; P < 0.005), miR-H3-3p (8.1% ± 0.56%; P < 0.005), miR-H6-3p (8.79% ± 0.73%; P < 0.005), and miR-H27 (7.3% ± 0.57%; P < 0.005) treated mice but not the others (Figs. 4A, 4C, Supplementary Fig. S4).

Figure 4.

Reduced immune cell infiltration in cervical lymph nodes in HSV-1–infected mice treated with viral miRNA inhibitors. Animals were infected with HSV-1 (strain McKrae; 10⁵ PFU) and subsequently treated with either viral miRNA inhibitors (miR-H1-5p, miR-H2-5p, miR-H3-3p, miR-H4-3p, miR-H5-3p, miR-H6-3p, miR-H7-3p, miR-H17, and miR-H27) or scrambled control inhibitors. (A) Scatterplots showing the percentages of CD4⁺ and CD8⁺ T cells obtained from cervical lymph nodes at 8 dpi. (B–E) Summary of CD4⁺ and CD8⁺ T cells in cervical lymph nodes at different time points (4 and 8 dpi). (F) Representative scatterplot showing antigen-presenting cells (CD11c⁺I-A/I-E⁺) in cervical lymph nodes. (G, H) Summary of CD4⁺ and CD8⁺ T cells in cervical lymph nodes at 4 and 8 dpi. n = 3 mice per v-miR or scramble inhibitor condition were used to generate the bar graph. Numbers in the scattered plots show percentages of positive cells for each marker. Each bar represents mean ± SD. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

We also quantified CD4⁺ T cells and noticed a significant reduction in the cervical lymph nodes of mice treated with miR-H1-5p (8.7% ± 1.18%; P < 0.05), miR-H3-3p (9.2% ± 0.26%; P < 0.05), miR-H6-3p (8.8% ± 0.28%; P < 0.05), and miR-H17 (10.4 ± 0.48%) inhibitors at 4 dpi compared to scramble, while miR-H2-5p, miR-H4-3p, miR-H5-3p, miR-H7-3p, and miR-H17 did not exhibit significant differences (Figs. 4A, 4D, Supplementary Fig. S3). A similar reduction in the CD4⁺ T cells was observed at 8 dpi in infected mice treated with miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 but not the others (Fig. 4E, Supplementary Fig. S4).

Furthermore, we observed a significant decrease in the percentages of CD11c⁺I-A/I-E⁺ DCs in the lymph nodes from mouse eyes treated with the inhibitors of miR-H1-5p (2.5% ± 0.4%; P < 0.01), miR-H3-3p (1.7% ± 0.16%; P < 0.001), miR-H6-3p (2.0% ± 0.11%; P < 0.001), and miR-H27 (2.3% ± 0.6%; P < 0.01) compared to control (3.5% ± 0.53%) at 4 dpi (Fig. 4G; Supplementary Fig. S5), and a similar trend of reduction was also followed at 8 dpi (Figs. 4F, 4H, Supplementary Fig. S6). Reduced infiltration of immune cells in the cornea and the draining cervical lymph nodes corroborate with our previous results showing low HSV-1 replication upon delivery of v-miR inhibitors targeting miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27.

Viral miRNA Inhibitors Attenuate Proinflammatory Th1 and Th17 Cell Ratio

HSV-1 infection in the cornea leads to a wide range of pathologies, from benign epithelial keratitis to more severe stromal keratitis. Various reports have demonstrated the expansion of CD4⁺ T cells in the draining lymph nodes after HSV-1 corneal infection.^38–40 Murine models further demonstrated the involvement of cytokines secreted from Th1 and Th17 in HSK pathology.^41,42 Because lymphatic drainage reflects cornea inflammation, we have investigated whether topical delivery of v-miR inhibitors diminished the percentage of proinflammatory Th1 and Th17 cells in the cervical lymph nodes at 4 and 8 dpi. Compared to control (8.4% ± 1.18%), the topical delivery of miR-H1-5p (5.02% ± 0.49%; P < 0.05), miR-H3-3p (3.7% ± 0.57%; P < 0.01), miR-H6-3p (3.0% ± 0.41%; P < 0.001), and miR-H27 (4.54% ± 1.33%; P < 0.01) reduced the IFNγ expression in CD4⁺ T lymphocytes in the cervical lymph nodes at 4 dpi (Fig. 5B). No significant changes were observed for miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 inhibitors (Supplementary Fig. S7). The reduction of IFNγ in CD4⁺ T cells at 8 dpi followed a similar pattern in cells treated with miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27. However, the abundance of CD4⁺IFNγ⁺ cells was higher at 8 dpi compared to 4 dpi (Figs. 5A, 5C, Supplementary Fig. S8).

Figure 5.

Viral miRNA inhibitor delivery in the eye affects the expression of proinflammatory cytokine secretion from CD4⁺ T cells. (A) Representative scatterplot showing the expression of IFNγ in CD4⁺ cells (4 dpi). (B, C) Total CD4⁺IFNγ⁺ T cells (Th1 type) in the cervical lymph nodes on 4 and 8 dpi of ocular HSV-1 infection. These cells were first gated at CD4⁺ T cells. (D) Representative scatterplot showing the CD4⁺IL-17⁺ cell (Th17 type) proportion of post-v-miR inhibitor delivery in the HSV-1–infected eye at 4 dpi. (E, F) Total number of Th17 cells after delivery of v-miRs. (G, H) Validation of IL-17 and IFNγ in cornea tissue by RT-qPCR analysis. (I, J) Expression of IFNα in cornea by RT-qPCR analysis. ND in RT-qPCR data indicates that the mean is not detected. β-Actin was used as an endogenous control. The scatterplots were generated using FlowJo 10.8.1. The numbers written in the scatterplots reflect the percentages of positive cells for that marker. Each bar represents mean ± SD. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001. n = 3 mice per experimental condition were used to generate the bar graph.

Although the functions of CD4⁺IFNγ⁺ cells in the pathogenesis of HSK have not been studied, the role of IL-17 has been well documented with the severity of human and mouse SK.^43,44 Contemplating this, we have evaluated the presence of CD4⁺IL-17A⁺ in the lymph nodes of HSV-1–infected mice topically treated with v-miR inhibitors. Interestingly, we noticed a marked reduction in IL-17A expression in CD4⁺ T cells from cervical lymph nodes obtained at both 4 and 8 dpi. Compared to control (13.7% ± 1.4%), the percentages of CD4⁺IL-17A⁺ cells for miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 were 8.4% ± 0.89%, 6.33% ± 0.014%, 5.61% ± 0.79%, and 6.6% ± 0.35%, respectively (Figs. 5D–F). Topical delivery of the rest of the v-miR inhibitors on Th17 abundance was not as impactful at 4 dpi (Supplementary Fig. S9). Furthermore, RT-qPCR data on the corneal transcript analysis validated markedly low expression of IFNγ in mice treated with miR-H1-5p (0.19 ± 0.11-fold; P < 0.05), miR-H3-3p (0.08 ± 0.02-fold; P < 0.005), miR-H6-3p (0.06 ± 0.005-fold; P < 0.0005), and miR-H27 (0.22 ± 0.04-fold; P < 0.005) at 4 dpi in the mice cornea (Fig. 5G). Levels of proinflammatory cytokine IL-17A also exhibited a significant reduction in the cornea treated with these inhibitors targeting miR-H1-5p (0.58 ± 0.1-fold; P < 0.005), miR-H3-3p (0.71 ± 0.11-fold; P < 0.005), miR-H6-3p (0.3 ± 0.3-fold; P < 0.05), and miR-H27 (0.41 ± 0.02-fold; P < 0.005) (Fig. 5H). Furthermore, it is reported that HSV-1 triggers type 1 and 2 IFNs from both infected epithelial cells and resident and infiltrating immune cells, which are demonstrated as crucial in its replication and clearance from the lesion.^45,46 Keeping this in view, we have also evaluated the expression of IFNα in the cornea, and we observed a significant reduction in mice treated with miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 inhibitors. Overall, these results showed a concurrent reduction in Th1, Th17 immune responses, and antiviral type 1 IFN responses in v-miR inhibitor-treated mice subjected to ocular keratitis (Fig. 5I).

Topical Delivery of v-miR Inhibitors Induces Proresolving Immune Subsets and Markers

IL-10 plays a pivotal role in the resolution of HSV-1 treatment by attenuating the proportion of T cells and neutrophils in the cornea.⁴⁷ Reduction in proinflammatory (e.g., miR-155) regulators in the murine model induces IL-10 and concomitantly reduces the generation of Th1 and Th17 effectors that orchestrate HSK lesions.⁴⁸ These two points make a strong rationale for evaluating IL-10 expression in CD4⁺ T cells in lymph nodes derived from mice treated with v-miR inhibitors. Topical delivery of miR-H3-3p, miR-H6-3p, and miR-H27 in mice results in a remarkable increase in IL-10–expressing CD4⁺ T cells in the cervical lymph nodes. We noticed higher levels of CD4⁺IL10⁺ cells at 8 dpi in miR-H3-3p (6.1% ± 0.14%), miR-H6-3p (9.3% ± 1.6%), and miR-H27 (14.2% ± 3.04%) inhibitor-treated mice compared to control (3.3% ± 0.54%) (Figs. 6C, 6D). A similar trend in higher CD4⁺IL-10⁺ cell percentages was markedly higher in cervical lymph nodes collected from animals treated with miR-H3-3p, miR-H6-3p, and miR-H27 at 4 dpi (Figs. 6A, 6B). These results suggest that v-miR inhibition suppresses inflammation and promotes disease resolution by enhancing the expression of proresolving mediators. Consistent with the expression of IL-10 by flow cytometric analysis, we also noted a marked increase in the expression of IL-10 transcript in miR-H1-5p (fold change: 1.53 ± 0.15), miR-H3-3p (fold change: 2.78 ± 0.42), miR-H6-3p (fold change: 2.09 ± 0.33), and miR-H27 (fold change: 1.59 ± 0.15) (Fig. 6E). Moreover, markedly higher expression of two key anti-inflammatory and proresolution markers, Arginase 1 (miR-H1-5p: fold change: 1.86 ± 0.21; miR-H3-3p: fold change: 1.52 ± 0.13; miR-H6-3p: fold change: 6.02 ± 2.1; miR-H27: fold change: 2.62 ± 0.2) and CD25 (miR-H1-5p: fold change: 1.3 ± 0.18; miR-H3-3p: fold change: 2.66 ± 0.35; miR-H6-3p: fold change: 6.41 ± 0.56; miR-H27: fold change: 1.88 ± 0.2), which promotes myeloid and lymphoid immune regulatory phenotypes critical for disease resolution (Figs. 6F, 6G).

Figure 6.

Induction of IL-10 in CD4⁺ T cells in the lymph nodes of HSV-1–infected mice treated with v-miR inhibitors. (A, C) Scatterplot showing the CD4⁺IL-10⁺ cells in the cervical lymph nodes at 4 and 8 dpi of ocular HSV-1 infection treated with control of miRNA inhibitors. (B, D) Comparison of the total number of CD4⁺IL-10⁺ cells at 4 and 8 dpi of ocular HSV-1 infection. The numbers written in the scatterplots reflect the percentages of positive cells for that marker. Quantitative PCR showing expression of (E) IL-10, (F) Arginase 1, and (G) CD25 transcripts in corneal tissues. Gapdh was used as an endogenous control. n = 3 mice per case were used to generate the bar graph. Scatterplots were generated using FlowJo 10.8.1. All the bar graphs were made in GraphPad Prism 10. Each bar represents mean ± SD. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Topical Delivery of v-miR Inhibitors in the Eye Reduces the Antigen Presentation Potential of CD11c⁺HLA-R⁺ Cells

Based on our previous study showing that v-miR inhibition induces IL-10 in CD4⁺IFNγ⁺ T cells, we hypothesized that this would limit HSV-1–induced expansion of CD4⁺ T cells. We sorted dendritic cells (CD11b⁺HLA-DR⁺) and CD4⁺ T cells from lymph node cells using flow sorting to test this (Supplementary Fig. S10). We performed the coculture experiment in the presence of soluble antigen ovalbumin (Ova). CD4⁺ T cells were labeled with CFSE before coculturing to assess proliferation. We observed a marked reduction in T-cell proliferation (CFSE^lo population), as assessed by flow cytometric analysis of CFSE⁺ CD4⁺ T cells in mice treated with the inhibitors of miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27. Compared to control (62.2% ± 3.30%), the percentages of CD4⁺CFSE^lo (the proliferating proportion) were observed as 43.1% ± 4.64% (miR-H1-5p), 27.3% ± 10.2% (miR-H3-3p), 49.0% ± 14.8% (miR-H6-3p), and 36.5% ± 7.39% (miR-H27) (Figs. 7A, 7B). These results suggest that v-miR inhibition promotes effector immune cell activity to promote viral clearance.

Figure 7.

Mice treated with viral miRNA inhibitors show reduced T-cell proliferation. CD11c⁺I-A/I-E⁺ DCs and CD4⁺ T cells sorted from lymph nodes were cocultured in the presence of ovalbumin. Proliferation of CFSE⁺CD4⁺ T cells was examined after 5 days of culture. (A) Representative histograms showing the proliferation of CFSE-labeled CD4⁺ T cells. The CFSE^lo population was considered a proliferating population. This phenomenon is due to the dilution of CFSE intensity after each consecutive cell cycle. CFSE-stained T cells, untreated or treated with phytohaemagglutinin (PHA), were used as negative and positive controls. Values show percent proliferation after gating for CFSE-stained and unstained T cells. (B) Bar graph showing the summary of CD4⁺ T-cell proliferation in n = 3 mice. Each bar represents mean ± SD. One-way ANOVA was used to calculate P values, and P < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Discussion

Depending on the viral life cycle, HSV-1 expresses distinct miRNA profiles. These miRNAs regulate various aspects of the virus life cycle, immune evasion, suppression, and so on. While most studies have focused on studying the v-miR role in cultured cells or latency in ganglia, this study emphasized severe clinical manifestations of HSV-1 infection in corneal tissues. In the cornea, the virus infects epithelial cells, leading to tissue scarring, which, in turn, causes partial or sometimes complete loss of vision. Identifying viral miRNAs expressed in ocular tissues during HSV-1 infection can yield targets to control disease progression. Here we have identified HSV-1 miRs in infected ocular tissues, indicating their pathogenic roles. This is the first report demonstrating that in vivo functional inhibition of primary infection expressed viral miRNAs in corneal tissues prevents the progression of ocular keratitis.

Our RNA sequencing and RT-qPCR analysis of HSV-1–infected ocular tissues revealed the abundance of miR-H1-5p, -H2-5p, -H3-3p, -H4-3p, -H5-3p, -H6-3p, -H7-3p, -H17, and -H27 at various time points (2, 4, 6, and 10 dpi) of disease progression. In situ viral miRNA expression data suggest that both miR-H1-3p and miR-H6-3p could be detected in the epithelial cells and the connective tissue, suggesting their detection in nonepithelial cells. Interestingly, reinfection of HSV-1 further increases the burden of v-miRs, indicating their pathogenic role. While epithelial cells are the primary target of HSV-1 during primary infection, it also exhibits tropism for immune cells, including monocytes, T cells, and NK cells. These target cells potentially serve as direct viral reservoirs expressing viral miRs. In addition, indirect sources of viral miRs can also contribute to detecting these viral biomolecules. Viral miRs, like cellular miRs, can be packaged into exosomes and delivered to various cell types.^26,49–51 This may permit viruses to modulate key host cell functions, including immunity, without infecting these cells.

Functional inhibition of viral miRs in ocular disease has not been investigated previously. Sequence-specific inhibition of viral miRNA activity suppression may relieve biological pathways and help restrain ocular HSV-1 infection. The results of corneal imaging (using a V20 stereoscope) and plaque assay from eye washes showed that only miR-H1-5p, miR-H3-3p, miR-H6-3p, and miR-H27 tended to restrict HSV-1–mediated ocular disease progression. Mitigation of ocular disease in these mice was evident from comparatively less visible destruction in the HSV-1–infected eye, significantly fewer HSV-1 virions in the plaque assay of eye washes, and low abundance of HSV-1–specific transcripts (ICP0, gD), indicating a role of v-miRs in viral replication and persistence. Overall, our data strongly suggest that targeting miR-H1-5p, -H3-3p, -H6-3p, and -H27 can be explored to mitigate ocular HSV-1 infection in mice. Our data corroborate with the earlier attempts to target HSV-1–induced v-miRs to inhibit HSV-1 replication and progression. Barrozo et al.⁵² showed that deletion of miR-H1 and miR-H6 impairs reactivation of HSV-1 infection in murine foot pads. Another report by Zou et al.⁵³ showed that in vitro expression of vaccinia virus protein VP55 (VP55)- from HSV-1 recombinant virus facilitated the degradation of HSV-1 encoded miR-H11, which eventually inhibited HSV-1 spread. Studies from our lab have previously shown that miR-H1 overexpression impairs HSV-1 replication in oral epithelial cells.⁵⁴ Compared to previous reports, our delivery approach is more robust, unique, and specific in the following ways: (1) sequence-specific inhibition of v-miRs and (2) topical v-miR delivery at the site of infection, the mouse eye, rather than systemic delivery.

To persist inside the host, it is of utmost importance for HSV-1 to interfere with antiviral immunity, specifically during productive infection. Viral proteins target immune function and antigen presentation pathways but can also trigger antigenic responses.⁵⁵ On the contrary, v-miRs can concurrently target numerous host and viral transcripts to regulate viral replication, are nonimmunogenic, and have unrestricted access to virtually all cells to modulate host biological functions. The role of viral miRNA in the progression of periocular diseases (such as blepharitis and conjunctivitis) and their role in perturbing host immunity is not well established. HSV-1 infection in the cornea results in infiltration of tissue-resident or early infiltrating corneal DCs that facilitate the clearance of HSV-1 by facilitating the migration of natural killer cells and inflammatory monocytes to the site of HSV-1 lesions.^56,57 After capturing the antigens in the cornea, dendritic cells and macrophages migrate to draining lymph nodes—namely, cervical lymph nodes—where these cells direct the expansion of HSV-specific CD4⁺ T cells.⁵⁸ Furthermore, various reports also exhibited the preponderance of CD4⁺ T cells in the HSK infiltrates around 7 dpi and also in genital disease in a mouse model during HSV-1 infection.^33,58,59 Our results showed a marked reduction in both migration of DCs (CD11c⁺I-A/I-E⁺) as well as expansion of CD4⁺ T cells in cervical lymph nodes in mice topically treated with the inhibitors of miR-H1-5p, -H3-3p, -H6-3p, and -H27. So far, no direct/indirect evidence demonstrates that blocking viral-encoded miRs results in disease mitigation and less CD4⁺ T-cell/infiltration expansion. Reduced viral replication during early time points (day 3) in v-miR inhibitor-treated mice suggests robust viral clearance and less immune activity. In line with this, Lukiw et al.⁶⁰ showed less infiltration of immune cells after treating human brain cells with acyclovir or Aβ42 peptides during HSV-1 infection.

Naive CD4⁺ T cells have the potential to differentiate into Th1, Th2, Th17, and induced regulatory T (iTreg) cells, depending upon the cytokine milieu.²⁰ Herein, we have investigated the nature of the immune response in ocular HSV-1 infection. Our results showed the coexistence of both Th1 (CD4⁺IFNγ⁺) and Th17 (CD4⁺IL-17⁺) in the cervical lymph nodes carrying the lymphatic drains from the mouse cornea at both time points: 4 and 8 dpi. These data were congruent with the earlier report published by Suryawanshi et al.,⁶¹ showing a significant increase of Th1 and Th17 cells in the local draining lymph nodes of mice ocular infection. However, Hirose et al.⁶² demonstrated the predominant role of Th17 cells in the mouse cornea with the progression of HSV-1. During the delivery of miR-H1-5p, -H3-3p, -H6-3p, and -H27 for disease mitigation, we found a significant decrease of both CD4⁺IFNγ⁺ and CD4⁺IL-17⁺ cells in the cervical lymph nodes, thereby suggesting their role in marginalizing the HSV-1 induced inflammatory reactions.

To control the severity of the lesion, antiviral therapies should inhibit viral replication and potentially induce the regulatory T cells for disease resolution.⁶³ The delivery of miR-H1-5p, -H3-3p, -H6-3p, and -H27 induces IL-10 in IFNγ⁺ T cells at later times of HSV-1 infection (8 dpi). In addition, two key regulatory immune markers, CD25 and Arg1, were also significantly increased in v-miR inhibitor-treated mice, further suggesting that v-miRs impair immune regulatory pathways. It implies that higher load shifts differentiation of CD4⁺ toward the proinflammatory CD4⁺IFNγ⁺ and CD4⁺IL-17⁺ phenotype, which is later shifted toward the generation of IL-10–producing regulatory T cells to resolve the infection. The pleotropic nature of v-miR inhibition controls the ocular HSV-1 by restricting viral replication and facilitating the induction of anti-inflammatory cytokine IL-10 in proinflammatory cells to prevent further damage to the corneal tissue by proinflammatory cells. Previous studies from our lab have shown that miR-H1-5p overexpression alters hundreds of host transcripts in oral epithelial cells, affecting pathways related to antiviral immunity, inflammation, and autophagy.^54,64,65 In particular, miR-H1 directly binds and regulates LIFR, ATG16L1, and TGFBR1, which play crucial roles in immune activation, cell proliferation, differentiation, and antiviral responses. Suppression of LIFR by miR-H1 likely diminishes LIF signaling, leading to reduced secretion of proinflammatory cytokines, while ATG16L1 silencing may interfere with intracellular pathogen clearance. Inhibition of miR-H1 in our study may alleviate these crucial biological pathways that promote viral clearance and antiviral immunity.

In summary, we identified HSV-1–encoded microRNAs enriched during primary in vivo infection and evaluated their therapeutic potential. Functional inhibition of these viral miRNAs (miR-H1, H3, H6, and H27) using custom LNA-modified inhibitors significantly reduced viral replication, promoted disease resolution, and enhanced tissue healing. These findings highlight novel miRNA targets for preventing or treating ocular and potentially nonocular HSV-1 infections.