In this prospective study, we examined 40 eyes of 20 patients with HFMD and 36 eyes of 18 healthy children at the Nanjing Drum Tower Hospital from July 2023 to September 2023. HFMD pediatric patients were diagnosed with HFMD by an experienced doctor in the Department of Pediatric Infectious Diseases. Throat swabs were collected from HFMD cases to ensure the virus strain. This study was approved by the Ethics Committee of Children’s Hospital of Nanjing Medical University (Application No.202311015-1) and conformed to the tenets of the Helsinki Declaration. Informed consent was obtained from all patients’ parents or legal guardians.

Inclusion criteria: (1) both the diagnosis and clinical classification complied with the Diagnosis and Treatment Guidelines for Hand-Foot-Mouth Disease (2018 Edition) developed by the Ministry of Health of the People’s Republic of China; (2) the child must be able to follow the ophthalmologist’s instructions; (3) patients were positive for EV-A71 by throat swabs. Exclusion criteria: (1) systemic disease or usage of drugs that influenced the corneal epithelium; (2) a previous history of ocular surgery; (3) the presence of other ophthalmic diseases.

Oculus Keratograph 5M (Wetzlar, Germany) can provide a simple non-invasive method of examination for dry eye. Tear film metrics were assessed with the K5M to evaluate tear meniscus height (TMH) and non-invasive Keratograph tear breakup time (NIKBUT, including NIKBUT-first and NIKBUT-average) from both eyes of each subject. The examination was performed in a dusk room by a masked ophthalmologist and the subject was confirmed not to be interfered from external factors.

Human tears were collected from participants by washing each eye with 1 mL of phosphate buffered saline (PBS), collecting the wash with sterile tubes, and combining the wash from the right and left eyes of individual participants (tears collected in the same tube). All tear samples were stored at -80°C until detection. The protein expressions of Cxcl1 and IL-6 were detected by Elisa using Human Cxcl1 ELISA Kit and Human IL-6 ELISA Kit (YOBIBIO, China).

Vero cells (african green monkey kidney epithelial cells) were purchased from the American Type Culture Collection. Vero cells were grown in high-glucose DMEM (HyClone, Logan, UT, USA) plus 10% or 2% FBS (Gibco, Carlsbad, CA, USA). Cells were incubated at 37°C with a 5% CO₂ humidified atmosphere.

The BrCr strain of EV-A71 used in this study was obtained from Dr. Bin Wu (Wang et al., 2019). Vero cells were used for EV-A71 multiplication. The virus titers were calculated as the 50% tissue culture infectious dose (TCID50) using the Reed–Munch method as previously described.

The strain of mice used in this study was AG6 (IFN-α/β and IFN-γ receptor deficient). AG6 mice were kindly provided by Qibin Leng (Institute Pasteur of Shanghai, Chinese Academy of Sciences). All mice were bred under specific pathogen-free conditions in the Animal Resource Center at the Nanjing university. Both female and male mice were used in the experiment. Two-week-old AG6 mice were infected with EV-A71 (10⁷ PFU) via the intraperitoneal (i.p.) route. Eyes of the mice were observed by fluorescence sodium dyeing (Jingming, Tianjin, China), tear production was measured by phenol red-impregnated cotton threads (Jingming, Tianjin, China) and then tissues were obtained for the experiments at 4 days post-infection (dpi). All experimental procedures in animals were approved by Animal Care and Use Committee of Nanjing University and were carried out in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research.

Corneal fluorescein staining was conducted in right eye of the infected mice at different time points. 2 μl of 0.25% fluorescein sodium was dropped topically in the conjunctival sac. After closing the mouse eyelids a few times, the eyes were rinsed with 0.9% saline and examined by the slit-lamp biomicroscope under cobalt blue light. To quantify the extent of damage to the corneal epithelium, the corneal fluorescein score (CFS) method was performed. Briefly, the cornea was divided into 4 quadrants, and the score of each quadrant was evaluated as follows: 0 points (no staining), 1 point (spot-shaped staining), 2 points (spot-shaped staining), and 3 points (flake staining). The score of each cornea was the sum of 4 quadrants’ scores.

Tear production was measured by the phenol red thread tear test using cotton thread. The threads were placed in the inferior conjunctival sac for 3 min in each eye, and the length of the tear-soaked thread was recorded by the same person each time.

Eyeballs and lacrimal gland were fixed in 4% formaldehyde and embedded in paraffin. The eyeballs tissue was sliced into 4-μM-thick sections, deparaffinized with xylene, and then hydrated in an ethanol gradient. Then, the sections were stained with hematoxylin-eosin staining kit (Servicebio, Wuhan, China). The eyeballs and eyelids collected from mice were fixed in 4% paraformaldehyde, then embedded with paraffin and sectioned into 5 µm thickness. A PAS staining kit (Servicebio, Wuhan, China) was used to stain the sections. For immunohistochemistry analysis, the sections were incubated with anti-Enterovirus A71 antibodies (Abcam, ab36367) at 4°C overnight. Subsequently, the sections were incubated with secondary antibody (Alexa Fluro 488-labeled goat anti-mouse IgG) for 30 min at room temperature. The bound antibody was performed with a DAB substrate kit (Thermo Scientific). Photos were taken of the sections using a light microscope (Leica Microsystems, Wetzlar, Germany).

Total RNA was extracted from cells and tissues using the Trizol reagent (Vazyme, Nanjing, China). According to our previous work, the cDNA was synthesized by reverse transcription of 1 µg of total RNA using the HiScript II Q Select RT SuperMix (Vazyme, Nanjing, China) according to the manufacturer’s instructions. qRT-PCR was performed on ABI QuantStudio 6 Flex (Invitrogen, Carlsbad, CA, United States) using the ChamQ Universal SYBR qRT-PCR Kit (Vazyme, Nanjing, China). GAPDH or β-actin was served as an internal control, and relative expressions of genes were calculated by the 2^−ΔΔCT method. Primer sequences are listed in Supplementary Table 1 .

HMGB1 in mice serum was quantitated by a Mouse HMGB1 ELISA Kit (S203674, D&B Biological Science and Technology, Shanghai, China) according to the manufacturers’ instructions. MRX II microplate reader (Dynex, Chantilly, VA, United States) was used to measure OD values for each well at 450 nm.

Tissues were removed and placed in optimum cutting temperature (OCT) glue at −80 °C overnight. Frozen sections were taken and left at room temperature for about 1 h. To detect the apoptosis of cornea and lacrimal gland, frozen sections were tested by in situ TUNEL assay according to the manufacturer’s instructions (Roche Diagnostics GmbH, Mannheim, Germany). Photographs were captured using the Leica Thunder system (Leica, Wetzlar, Germany).

An online tool (Search Tools for the Retrieval of Interacting Genes, STRING4) was used to analyze protein interactions. The PPI pairs were screened by confidence score (>0.40), and the PPI network was visualized by the Cytoscape V3.9.0 software. MCC was calculated through CytoHubba to evaluate the importance of each node, and the top 10 nodes were selected. The hub genes were their common nodes.

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). Each experiment was repeated at least 3 times. The results were shown as mean ± standard error of the mean (SEM) and were compared using Student t-tests. *P < 0.05 represents the statistical significance.

We established a mouse model of EV-A71 infection by intraperitoneal EV-A71 infection ( Figure 1A ), as previously described (You et al., 2023). The fluorescence staining of the corneal epithelium was observed before injection and at 4 dpi in mice ( Figures 1B, C ), indicating the tear film instability and the desiccation of the ocular surface. Phenol red cotton threads were also used to assess tear secretion and it in EV-A71-infected mice decreased greatly than the normal control levels ( Figures 1D, E ). Then, the extraorbital lacrimal glands (hereinafter called lacrimal glands) were isolated for further investigated ( Figure 1F ) and our results showed that the lacrimal glands of infectious mice were statistically smaller compared to the control ones ( Figures 1G, H ). Decreased tear secretion or an imbalance in tear film homeostasis would lead to Corresponding damages to corneal epithelium cells and Goblet cells of conjunctiva (Li et al., 2022). We then conducted histomorphology assessments to investigate the structural and morphological changes in corneas and conjunctival goblet cells after infection. The corneal structure and cell morphology were first evaluated by H&E staining. The normal corneal cells were densely and neatly arranged without inflammatory cells infiltration. However, the corneal superficial epithelial layers turned disrupted and desquamated in EV-A71-infected group. The rough area of cornea in EV-A71-infected group was markedly increased about 20% compared to the normal group, with a statistically significant difference ( Figure 1I ). Next, goblet cells were specialized cells that secrete mucins to lubricate the ocular surface, and its loss could aggravate dry eye (Alam et al., 2020). As shown in Figure 1J , the PAS‐positive goblet cells are abundantly presented in the conjunctival fornix of the normal cornea. However, goblet cells of EV-A71-infected group were lost distinctly, and the average density of stained cells in the conjunctiva significantly reduced 20% compared with the control group. Finally, we investigated the relationship between the systemic inflammation and corneal injury. The fluorescein staining indicated the extent of cornea damage (Fonn et al., 2010) and the serum level of HMGB1 was applied as an indicator for the severity of infection (Zheng et al., 2017). Pearson correlation analysis showed a positive correlation between corneal fluorescein scores and HMGB1 levels ( Figure 1K ). Thus, these results suggest that EV-A71 infection could lead to ocular surface desiccation and lacrimal gland changes. We hope to determine the causes of the symptoms by carrying out research on corneas and lacrimal glands in the following studies.

It is known that corneal epithelium lesions can lead to dry eye symptoms such as a decrease in tear film stability and tear secretion (Feng and Zhang, 2023), thus we investigated the condition of inflammation and apoptosis in the corneas. We first explored whether EV-A71 existed in corneas by IHC and there was no evidence of virus existence in corneas of EV-A71-infected mice ( Figure 2A ). Our findings suggest that no apoptosis was observed in the corneas of infectious group using TUNEL assay ( Figures 2B, C ). Apoptosis marker bax and caspase-3 were also evaluated by qRT-PCR which further validated the former result ( Figures 2D, E ). Then, we assessed the mRNA expression levels of inflammatory cytokines in cornea. The qRT-PCR results revealed that, compared to the control group, infectious mice had equal mRNA expression levels of IL-1β and MMP9 in the corneas ( Figures 2F, G ). ZO-1, as a marker of corneal epithelial tight junction, are expressed in the superficial cell layer of the corneal epithelium (Yang et al., 2018). The decreased expression of ZO-1 indicated the corneal epithelial barrier disruption in ocular surface disease. We investigated the expression of ZO-1 in corneal epithelium using qRT-PCR, and found that the expression of ZO-1 showed no clear difference between the two groups ( Figure 2H ). These results, taken together, suggested that cornea should not assume main responsibility for dry eye symptoms.

We further carried out RNA-Seq analysis. In total, we obtained 36469 gene features from our sequencing study. There were 1085 differentially expressed genes (DEGs) in corneas from infectious mice compared to that from the control ones (fold change log2(fc) ≥ 0.5, p < 0.05). Of the 1085 DEGs, 255 were up-regulated and 830 were down-regulated. The volcano plot ( Figure 3A ) and heat map ( Figure 3B ) were displayed all the DEGs. As displayed in Figure 3C , GO analysis displayed that the differentially expressed genes were associated with extracellular region, extracellular matrix, extracellular space, extracellular region part, biological adhesion and cell adhesion. The top 16 significantly enriched KEGG pathways are described in Figure 3D . KEGG pathway analysis revealed top 4 enriched pathways among these DEGs, including ECM-receptor interaction, Cell adhesion molecules, Complement and coagulation cascades, and Focal adhesion. Although the virus infection did not cause significant pathological damage on corneas, severe tear deficiency caused changes in the corneal genomic level, especially the downregulation of the ECM-receptor interaction ( Figure 3E ).

We further explored whether the appearance of ocular surface symptoms was related to lacrimal glands. The presence of EV-A71 infection in lacrimal glands were confirmed using IHC and qRT-PCR ( Figures 4A, B ). We also observed an increased number of apoptotic cells in lacrimal glands of infectious mice compared with control ones ( Figures 4C, D ). The expression of apoptosis-associated genes, such as bax and caspase-3, were upregulated in the EV-A71-infected group ( Figures 4E, F ). Then, we found that the acini in the healthy gland exhibited uniformity in size, neatness in arrangement, and normality in morphology while more large vacuoles were observed in infected group (arrows) ( Figures 4G, H ). This may be an indication that EV-A71 infection can cause the dysfunction of the excretion of tear fluid from the LGs, subsequently decreasing tear drops. To further investigate the alterations in the lacrimal gland, we evaluated the expression level of AQP5, a water channel protein, in the lacrimal gland (Cao et al., 2023). Our data showed that AQP5 expression in the lacrimal glands of EV-A71-infected mice was decreased compared with that in the controls ( Figure 4I ). Moreover, Rab3D, a specific marker of secretory vesicle maturation (Meng et al., 2016), was downregulated in the lacrimal glands with EV-A71 infection, as shown by qRT-PCR ( Figure 4J ).

We then compared the transcriptomic profiles of the LGs in infected and control mice. As shown in Figures 5A, B , the analysis identified 261 DEGs between the two groups, with 149 genes being upregulated and 112 showing downregulation. GO analysis revealed that the DEGs were enriched in the biological processes of response to external stimulus, defense response, response to external biotic stimulus, and the cellular component of extracellular region, extracellular space, extracellular region part ( Supplementary Figure 1 ). In KEGG analysis, the results indicated that genes were mainly associated with IL-17 signaling pathway, TNF signaling pathway and NF-κB signaling pathway, which were significantly up-regulated ( Figure 5C ). The mRNA expression of IL-17a was assayed by using RT-qPCR and it was significantly increased in the EV-A71-infected group compared with that of control group ( Figure 5D ). STRING database analysis identified 63 nodes and 29 edges and Cytoscape analysis identified ten top hub genes, including Cxcl1, Mpo, Selp, Ccl3, Ptx3, Lcn2, Plaur, Cldn4, Ccl12, and S100a9 ( Figure 5E ). Using qRT-PCR to validate 10 core shared genes identified by GSEA analysis and the results showed that the expression of them could be up-regulated after EV-A71 infection ( Figure 5F ).

A total of 38 subjects (18 healthy and 20 HFMD) were identified. There was no significant difference between the two groups in age (p = 0.132) and gender (p = 0.489) ( Figure 6A ). The TMH value in eyes of the HFMD group was lower than that in the eyes of the control group ( Figure 6B ). The NIKBUT-f and NIKBUT-a of the HFMD was significantly shorter than the control group ( Figures 6C, D ). Tear fluid samples were collected to assess tear inflammatory factor levels from the study subjects. The levels of pro-inflammatory factors Cxcl1 and IL-6 were observed to be significantly increased in HFMD patients compared to healthy controls ( Figures 6E, F ). Although most of the children patients stated no obvious ocular discomfort, our examination found a significant reduction in tear stability and increased inflammation cytokines concentration in tears.