Microbial influx during early postnatal life fortifies the ocular surface and guards against allergic eye disease in mice

Liu, Yanbo; Sun, Hang; Song, Shulin; He, Xiezhou; Wu, Han; Ou, Shangkun; He, Hui; Zong, Rongrong; Chen, Yongxiong; Fu, Guo; Wang, Yiqiang; Wu, Huping; Quantock, Andrew J.; Liu, Zuguo; Li, Wei

doi:10.1038/s42003-025-09095-4

Download PDF

Article
Open access
Published: 26 November 2025

Microbial influx during early postnatal life fortifies the ocular surface and guards against allergic eye disease in mice

Yanbo Liu ORCID: orcid.org/0000-0003-2191-6731¹^na1,
Hang Sun²^na1,
Shulin Song¹^na1,
Xiezhou He¹,
Han Wu¹,
Shangkun Ou¹,
Hui He¹,
Rongrong Zong¹,
Yongxiong Chen¹,
Guo Fu ORCID: orcid.org/0000-0001-5451-5219³,
Yiqiang Wang⁴,
Huping Wu¹,
Andrew J. Quantock⁵,
Zuguo Liu^1,6 &
…
Wei Li ORCID: orcid.org/0000-0001-7458-2590^1,6

Communications Biology volume 8, Article number: 1694 (2025) Cite this article

1172 Accesses
1 Altmetric
Metrics details

Subjects

Abstract

Host-microbiome interplay during development governs the homeostasis of various bodily surfaces, however, postnatal colonization of the microbiome and its impact on the homeostasis of ocular surface is still unclear. Here, the changes of the conjunctival microbiome in C57BL/6 J mice were tracked in 1-week-old neonates through to 8-week-old adult mice. This disclosed that changes in the conjunctival microbiome correlate with age, especially at the 2-week and 3-week time points, which, respectively, are accompanied by eyelid-opening and weaning. Antigen presenting cells were also recruited to the conjunctival epithelium after eyelid-opening, whilst an inhibition of microbial colonization at 2-to-3 weeks of age led to a disruption of mucosal homeostasis and aggravated the development of allergic eye disease. This study improves our understanding of the development of the conjunctival microbiome in mice, and provides an indication that early microbial colonization is required for the establishment of mucosal ocular surface homeostasis, the perturbation of which leads to increased susceptibility to allergic eye disease.

Identification of an intraocular microbiota

Article Open access 09 March 2021

Ocular microbiota promotes pathological angiogenesis and inflammation in sterile injury-driven corneal neovascularization

Article 19 August 2022

Evolution of the ocular immune system

Article Open access 09 December 2024

Introduction

The ocular surface is continuously exposed to the outside environment, with the conjunctiva naturally harboring low numbers of live bacteria¹. Although many studies have reported correlations between the composition of microbiota in the ocular surface and various physiological indicators and disease manifestations in adults^1,2,3, only a few have investigated the compositional characteristics of the ocular microbiome in children. Studies conducted on newborns have found that Proteobacteria was the dominant phylum in the ocular surface microbiome, followed by Actinobacteria and Firmicutes as the next two most abundant phyla⁴. In children under 18 years of age, Firmicutes and Bacteroidetes were the next two most abundant phyla after Proteobacteria. In comparison, the ocular surface microbiome in adults exhibited decreased abundances of Firmicutes and Bacteroidetes^5,6. These studies suggest that the ocular surface microbiome varies across different age stages, however, the dynamic change of the microbiome of the ocular surface from early neonatal stages onwards has not been investigated. Here, we investigate the ocular surface microbiome of the mouse eye, which represents a good model system because the eyelids remain closed for 12–14 days after birth, allowing for continuous observation of the colonization of the ocular surface microbiome.

Antibiotic eye drops are commonly used in the treatment of bacterial infections of the ocular surface. However, because the clinical diagnosis of infectious, bacterial eye diseases in children is more difficult than in adults, the use of antibiotics might not always be the best course of action⁷. Even in the absence of ocular surface infection, given the susceptibility of newborns to pathogens, antibiotic eye drops have been widely used to provide ocular prophylaxis at birth for more than a century^8,9. It is well known that topical antibiotics applied to the eye can affect the ocular surface microbiome, so early antibiotic use may have an impact on ocular surface microbial colonization and needs to be more thoroughly understood.

From birth onwards, the colonizing microbiome of human mucosal surfaces induces the development and maturation of immune cells and tissues that facilitate tolerance to environmental exposures. The composition of those microbial communities, meanwhile, changes considerably during the development of the immune system^10,11,12. In both humans and mice, a paucity of bacteria in early life leads to immunological defects that can be sustained into adulthood and increase the susceptibility to, or exaggeration of, allergic disease^13,14. Such early-life events have been proposed in other mucosae^15,16. Due to ethical concerns and the availability of limited numbers of samples in humans, the analysis of the microbiome of the ocular surface and its role in the development of ocular surface immunity in early life remains elusive. Accordingly, to test our hypothesis that alterations in microbial composition, or depletion of the microbiota early in life, may lead to perturbations of ocular surface immune homeostasis, we have investigated the ocular surface microbiome of mice at six different growth stages between postnatal weeks 1-to-8 weeks. We also explore the likelihood of dysbiosis of ocular surface microbiota during early postnatal life induced by the use of antibiotics and its impact on ocular surface homeostasis.

Results

Postnatal dynamic changes of the conjunctival microbiome

To understand the relationship between the ocular surface microbiome and the ocular mucosa, we first characterized the conjunctival microbiome in mice between 1 and 8 weeks of age using 16S rRNA sequencing for bacterial identification. A total of 6682079 16S rRNA tags were obtained with an average of 105993 effective tags across samples (Table S1) and then we clustered 646 OTUs based on 97% sequence identity. OTUs associated with negative controls were removed from the analysis based on the indicator value (IndVal) index, employed for the identification of potential contaminants as described previously¹⁷. Community bar-plot analysis displays relative levels of the conjunctival microbial community in all 60 samples at the phylum level (Fig. S1a). A core microbiome consisting predominantly of Proteobacteria, Firmicutes, Actinobacteria, Chloroflexi, Acidobacteria, Patescibacteria, Cyanobacteria and Verrucomicrobia phyla was found, accounting for up to 85% of sequences on average. At the family level, the composition of the top ten microbials in mouse conjunctival tissue changed during eyelid-opening. In particular, at 2- and 3-week of life, the conjunctival microbiome underwent transient alterations as Enterococcaceae and Staphylococcaceae outgrew Moracellaceae and Bacillaceae (Fig. 1a). Meanwhile, at the genus level, composition and SIMPER analysis indicated that Staphylococcus and Enterococcus were significantly enriched in 2- and 3-week-old mice, while Acinetobacter and Anoxybacillus changed in an inverse direction (Fig. S1b and Table S2). BugBase was used for phenotypic prediction analysis, which revealed that the relative abundance of predicted potential pathogenic bacteria was highest in 3-week-old mice (Fig. S1c). We further examined the changes of conjunctival microbial diversity across different developmental stages. The Shannon index depicted a slight decrease in α-diversity in 2-week-old mice, followed by a dramatic decrease at 3 weeks of age. Thereafter, α-diversity increased until the 4-week of age and remained relatively constant until 8 weeks (Fig.1b). A PCoA was performed to visually depict the variations in taxon composition among these mice. In 4- to 8-week-old mice, β-diversity returned to levels similar to the initial microbial diversity observed in 1-week-old mice (Fig. 1c). No significant difference in α-diversity was observed between males and females across all samples. During postnatal development, both sexes exhibited a transient decrease in α-diversity at 3 weeks of age, while the Shannon index remained relatively stable at 1, 2, 4, 6, and 8 weeks, with no significant differences observed (Fig. S1d). These results indicated that male and female mice display a similar pattern of microbial diversity changes during developmental stages. In contrast, significant variations occurred based on gender as identified by the β-diversity index (Fig. S1e).

**Fig. 1: Dynamic changes in the conjunctival microbiome of mice aged between week 1 and 8 of neonatal development.**

We next explored bacterial co-occurrence patterns among mice at the six time points using network analysis as previously reported¹⁸. In these six networks, positive correlations were much higher than negative correlations (Fig. 1d and Table S3). Overall, the analysis of topological properties in these empirical networks revealed a significant difference between each age group (Table S3). Further comparison of co-occurrence patterns across six aggregate fractions showed that the bacterial community exhibited closer correlations (for example, a higher number of nodes) in the 1-, 4-, 6- and 8-week-old mice, while the 2- and 3-week-old mice showed a lower correlation (Table S3). Additionally, the data for average path length (APL) and average clustering coefficient (avgCC) indicated that the empirical network in 1-, 4-, 6-, and 8-week-old mice featured a more prominent “small world” module (Table S3). Subsequent structural analysis revealed the prevalence of deterministic co-occurrence patterns in the bacterial networks at intra-phyla level. Notably, bacterial OTUs in the predominant phyla, including Proteobacteria, Firmicutes, Bacteroidota, Actinobacteriota, and Verrucomicrobia exhibited a stronger propensity for co-occurrence compared to other phyla (Fig. 1d).

Due to the complex living environment of mice, the ocular surface microbiome may be affected by the animals’ surroundings during eyelid-opening. In order to track the potential source of the conjunctival microbiome, we examined possible origins, including feces samples in cages (Table S1). This indicated a significant difference in β-diversity between the conjunctival microbiome and environmental microbiome (Fig. S1f). The results of the microbial source tracking (FEAST) revealed that the microbes of the conjunctiva in 2- and 3-week-old mice likely mainly originate from the living environment (Fig. 1e), thus, we suspect that the mouse conjunctival microbiome is instantaneously affected by environmental factors at the time of eyelid-opening.

Antigen presenting cells (APCs) are transiently recruited in the conjunctival epithelium in early postnatal life

Given that the conjunctival microbes undergo instantaneous change during eyelid-opening, we speculated that microbial colonization during this period will have a major impact on the recruitment of immune cells in the conjunctiva. Evaluation of whole-mount conjunctival tissues by confocal microscopy, indeed, demonstrated a dramatic increase of CD45⁺ cells in the conjunctival epithelia of 3-week-old mice compared to the conjunctival epithelia of 1-, 2-, and 4-week-old mice (Fig. 2a, b). Further analyses by flow cytometry showed that the numbers of CD45⁺ cells in the conjunctiva dynamically changed, where it increased from 1 to 3-weeks, peaked at 3-weeks, and declined thereafter by 4-weeks (Fig. 2c). We suspected that some of these cells emerging in conjunctival epithelium could be APCs because many cellular dendrites were visible (Fig. 2d). Since APCs can be characterized by the expression of CD11b, CD11c, and MHC II¹⁹, we performed immunostaining on whole-mount conjunctival tissues and found a significant increase of intraepithelial CD11b⁺, CD11c⁺, and MHC II⁺ cells in the conjunctiva of 3-week-old mice (Fig. 2e), suggesting an enhanced recruitment of immune cells in the mouse conjunctival epithelium during eyelid-opening.

**Fig. 2: APCs are present in the conjunctival epithelium in 3-week-old mice.**

Maturation of the conjunctival epithelium during postnatal eyelid-opening

In mice, initial conjunctival epithelial stratification occurs around the time of eyelid-opening when the conjunctiva is first exposed to light, oxidative and environmental stress²⁰. The 1-to-2 cell layer-thick conjunctival epithelium in 1-week-old mice has become increasingly stratified and, respectively, has formed a squamous epithelium 3-to-4 cells thick and 5-to-6 cells thick in 2- and 3-week-old mice (Fig. 3a). The gradual increase of epithelial thickness was further confirmed by immunofluorescence staining using antibodies to conjunctival epithelial specific marker Krt19 and tight junction protein ZO-1 (Fig. 3b). The conjunctiva is a mucosal tissue, rich in mucin-producing goblet cells. In the mouse conjunctiva, goblet cells were absent in 1-week-old animals, emerged at 2-weeks of age and were highly populous in 3-week-old conjunctival tissue. MUC5AC is the major goblet cell secretory mucin, and our analysis showed that Muc5ac mRNA expression was highly upregulated in 3-week-old mice (Fig. 3c).

**Fig. 3: The conjunctival epithelial barrier is fully mature in mice at 3-weeks of age and cross-talk between APCs and antigens occurs via goblet cell associated passages (GAPs).**

We found a large number of CD11c⁺ and CD11b⁺ cells with dendrites, presumably APCs, juxtaposing to the goblet cells in conjunctiva of 3-week-old mice (Fig. 3d). We speculated that these APCs may capture the antigens colonized in the conjunctival epithelium. In intestine, goblet cells establish so-called “goblet cell-associated antigen passages” (GAPs) in the second and third weeks after birth, which allow for the delivery of bacterial antigens from the intestinal lumen to the lamina propria²¹. Similar to the intestine, antigens bind to goblet cell secreted mucins in the conjunctiva and could be delivered to APCs via GAPs²². To determine the formation of GAPs during development in mice, we topically applied OVA, in the conjunctival sac at different time points from 2- to 4-weeks postnatally. Compared to 2- and 4-week-old mice, 3-week-old mice exhibited an increased migration of OVA from the conjunctival epithelium to the stroma, with partial uptake by APCs in the stromal area (Fig. 3e). We further examined the correlation between the ocular surface microbiome and conjunctival status using Mantel tests (Fig. S2). This revealed that the ocular microbial communities exhibit a significant correlation with Muc5ac mRNA expression, conjunctival epithelial layers and the number of goblet cells in 2-week-old mice. The conjunctival microbial communities in 3-week-old mice, moreover, exhibit a significant correlation with the percentage of deep intraepithelial APCs (Fig. S2). These results demonstrate that the conjunctival microbiome in neonatal mice helps enhance maturation of conjunctival development during eyelid-opening and supports the recruitment of APCs in ocular surface mucosa, which can be located in close proximity to goblet cells and extend trans-epithelial dendrites toward the ocular surface at 3-weeks of age to sample microbes via GAPs.

Disruption of the conjunctival microbiome in early postnatal life affects the maturation of conjunctival tissue

To determine whether early elimination of commensal microbiome would affect the maturation of conjunctival tissue, 0.3% tobramycin solution was applied to the murine ocular surface around the time of eyelid-opening, i.e, 2-weeks postnatally, and for 7 days thereafter (Fig. 4a). Bacterial culture of the conjunctival tissue from antibiotic-treated group mice (Tob) showed lower bacterial colonies compared with control-group saline-treated mice (Ctrl) after 7-days treatment (i.e., at the 3-week timepoint) (Fig. S3a). The subsequent profiling of the composition of the conjunctival microbiome by 16S rRNA gene sequencing (Table S1) revealed that, in comparison with the Ctrl group mice, Tob group mice had a significantly altered conjunctival microbiome after 7-days treatment (Fig. 4b). Changes were apparent in the composition and potential functions associated with cellular process and carbohydrate metabolism predicted by PICRUST2 (Fig. S3b).

**Fig. 4: Topical antibiotic exposure in early postnatal life leads to changes in the composition of the conjunctival microbiome and disrupts mucosal homeostasis.**

Compared to control tissues, topical antibiotic exposure resulted in a reduction in the number of goblet cells in the murine conjunctiva (Fig. 4c) and a downregulation of Muc5ac, Ki67, and Krt19 gene expression (Figs. 4c and S3d, e). The examination of GAPs indicated a reduction of OVA uptake in Tob group mice compared with mice in the control group (Fig. 4d). There was no evidence of apoptosis nor disruption of cell junctions during the experimental period (Fig. S3c, f). Meanwhile, to examine the effect of tobramycin on conjunctival cells, we excised conjunctival tissue from eyelid-opening mice and cultured extracted cells as previously described²³. This revealed that all cells grow from the explants stained positively for Krt19, confirming their identity as conjunctival epithelium (Fig. S4a). Subsequently, the cells were cultured in media containing 0.03 mg/mL, 0.3 mg/mL, and 3 mg/mL tobramycin, respectively. Cell viability assays indicated no toxicity at any of the tested concentrations. Tobramycin, moreover, elicited no effect on the proliferation of the cultured cells (Fig. S4b). The number of CD45⁺ cells in the conjunctival epithelium, however, were less in the conjunctivas of Tob group mice (Fig. 4e, f). Compared with Ctrl group mice, fewer MHC II positive cells and a lower expression level of MHC II-related gene H2-Ab1 was noted in the conjunctiva of Tob group mice (Fig. 4g). To assess whether this effect was direct, we performed an in vitro co-culture of mouse cervical lymph node (CLN) cells with tobramycin. We found that, except at the highest concentration (3 mg/mL), tobramycin at 0.03 mg/mL and 0.3 mg/mL showed no significant effect on MHC II expression in CLN cells (Fig. S4c). Together, these data support the notion of a local microbiome at the ocular surface that helps maintain ocular mucosal homeostasis, and that the transient disruption of the microbiome via antibiotics in the week following eyelid-opening may delay conjunctival maturation and disequilibrate ocular mucosal homeostasis.

Longer-term effects of early postnatal antibiotic exposure on ocular surface mucosa

To investigate the longer-term effect on conjunctival tissue of disruption of the ocular surface microbiome by topical antibiotic exposure following eyelid-opening, the microbial composition characteristics of the conjunctiva were surveyed by 16S rRNA gene sequencing 1-week (i.e., mice at 4-weeks of age) or 2-weeks (i.e., 5-week-old mice) after the cessation of tobramycin/saline application for 7-days, which started at 2-weeks of age (Table S1). This analysis indicated that the effect of topical antibiotic exposure on the composition of the microbiome was still evident 2 weeks (i.e., in 5-week-old mice) after discontinued antibiotic application (Fig. 4b). At 4- and 5-weeks of age, there were significant differences in β-diversity between Ctrl and Tob groups (Fig. 5a). We further explored the bacterial co-occurrence patterns and found that there was an obvious ecological network difference between the Ctrl and Tob groups in both 4-week and 5-week mice (Fig. 5b and Table S4). Additionally, compared with Ctrl group mice, Tob group mice at 4 weeks had significantly higher Shannon and Simpson indices, however, there were no significant differences in α-diversity between the groups at 5-weeks of age (Fig. S5a). We also used PICRUSt2 to predict the functions of the microbiome in each group, which showed that in Tob group mice the microbiome pointed to a substantially different potential functions compared with Ctrl group mice at 4 weeks of age (Fig. S5b). This was not the case at 5 weeks, however, with Tob vs Ctrl groups showing smaller difference. Together, the data suggest that the diversity and potential functions of conjunctival microbiome are gradually recovering 2 weeks after discontinued topical antibiotic exposure that started during eyelid-opening, even though composition of conjunctival microbiome remained unrestored.

**Fig. 5: Topical antibiotic exposure in early postnatal life leads to longer-term effects on microbial colonization and ocular mucosal homeostasis.**

Conjunctival tissue from the experimental mice was examined, which disclosed that the number of conjunctival goblet cells and levels of mucin secretion and Krt19 expression had been restored at 4 weeks (Fig. S5c, d), but MHC II positive cell numbers were higher in Tob group mice at 4 and 5 weeks compared to numbers in Ctrl group mice (Fig. 5c). Additionally, conjunctival immune reactivity to antigens by topically applied OVA (45KD) 30 min before euthanasia revealed a significant increase in the frequency of intraepithelial CD11b⁺ cells in the conjunctiva of Tob group mice at 4- and 5-weeks of age compared to mice in the Ctrl group at these junctures (Fig. 5d). This indicates that early antibiotic exposure after eyelid-opening may have an extended effect on the colonization of the conjunctival microbiome and may impair tolerance to environmental exposures, eventually leading to a potential aggravation of the response of CD11b⁺ cells to antigens in the conjunctiva of mice.

Absence of the conjunctival microbiome in early postnatal life exacerbates the ocular allergic response

Postnatal exposure to bacteria and their products helps modulate the development of the immune system, and these influences may be durable¹⁴. If the communication between host and microbiome is disturbed in early postnatal life, excessive immune reactivity may develop in adulthood, which can lead to increased susceptibility to allergic diseases²⁴. Because the conjunctiva-commensal relationship was formed during eyelid-opening, we hypothesized that disruption of ocular surface microbiome in this period would lead to an aggravation of allergic eye disease (AED). To investigate that we studied unimmunized control mice (naive), AED mice by OVA administration and antibiotic-treated AED mice (Tob+OVA) (Fig. 6a). The data indicated that compared with naive mice, OVA mice developed a significant allergic response, whilst Tob+OVA mice presented with even more severe clinical manifestations (Fig. 6b), with higher clinical scores (Fig. 6c) and elevated serum IgE levels (Fig. 6d). The number of eosinophils infiltrated in the conjunctiva was prominently increased in Tob+OVA mice compared to the naive and OVA mice when determined by Picrosirius Red staining (Fig. 6e). The results of toluidine blue staining, a useful method to recognize mast cells and evaluate degranulation, revealed a significant increase in mast cell numbers in the conjunctiva of Tob+OVA mice compared to naive and OVA mice (Fig. 6f). We imaged freshly excised mouse conjunctival tissues based on a moxifloxacin-based fluorescence imaging method and discovered that, compared with naive mice, the number and density of goblet cells and Muc5ac mRNA expression were significantly increased in OVA mice, and were more prominent in antibiotic-treated AED mice (Fig. 6g). Furthermore, antibiotic treatment significantly increased the number of MHC II⁺ cells and CD4⁺ T cells in conjunctiva of AED mice. The mRNA expression levels of the transcription factor Gata3 in Th2 cells, along with the cytokines IL-4 and IL-5 secreted by these cells, exhibit a substantial increase in conjunctiva of antibiotic-treated AED mice (Fig. 6h). Given that secondary T helper cell responses generated in the lymph node are responsible for causing AED, as previously described²⁵, we further examined T cell subsets in CLN of mice from different group. Flow cytometry analysis revealed that, compared to the naive mice, OVA mice exhibited a significant increase in the frequency of Th1 (CD4⁺ IFN-γ⁺) and Th2 (CD4⁺ IL-4⁺) cells in the CLN cells, accompanied by a significant decrease in the frequency of Treg (CD4⁺ CD25⁺ Foxp3⁺) cells. Notably, in Tob+OVA mice, the Th2 response was further elevated, accompanied by a lower frequency of Treg cells compared to OVA mice (Fig. 7a–c).

**Fig. 6: Disruption of the conjunctival microbiome in early postnatal life exacerbates the development of allergic eye disease (AED).**

**Fig. 7: Th2 and Treg cell frequencies in the draining lymph node of AED mice are affected by topical antibiotic exposure in early postnatal life.**

The same approach (Fig. S6a) was used to examine if the topical antibiotic treatment for 7 days of weaning (3-weeks-old) or adult (6-weeks-old) mice would similarly aggravate AED, which indicated that both Ctrl and Tob groups showed comparable increases in allergic manifestations (Fig. S6b, c) and possessed equivalent numbers of eosinophilic and mast cells (Fig. S6d, e). Consistent with the above findings, antibiotic treatment did not alter the number of CD4⁺ T cells nor the mRNA expression levels of Gata3, IL-4, and IL-5 (Fig. S6f). These results suggest that, compared with Ctrl group mice, AED is not aggravated in either weaning or adult Tob group mice. Collectively, these findings highlight the potential impact of perturbations to the cross-talk between the host and conjunctival microbiome in early postnatal life on the aggravation of AED.

Discussion

This study characterizes the dynamic changes of the ocular surface microbiome during postnatal development in mice. We observed that the phyla Proteobacteria and Firmicutes dominate the ocular surface microbiome at different postnatal stages in murine conjunctiva, and these two dominant phyla have also been found in the conjunctiva of children^5,26. Unlike in humans, however, mouse eyelids remain closed during early postnatal life and open about 2 weeks after birth. Due to the diminutive size of mice eyes and the relatively low microbial load, sustained observations over time in individual eyes are unfeasible, thus, we chose to study tissue excised at different ages from eye opening through to maturity. We also note that knowledge of the ocular surface microbiome in humans comes from investigations of children of different age groups^4,5,27, and that there is a need for a more comprehensive scrutiny of ocular surface microorganisms in humans from the neonatal stage to adulthood. In this study, we found that the murine ocular surface microbiome changed significantly during eyelid-opening, allowing us to conclude that external factors have a critical impact on the ocular surface microbiome of mice. In humans, it is likely that the mode of delivery and external influences may have a significant impact on the ocular surface microbiome of newborns. The most common conjunctival bacteria are coagulase-negative Staphylococcus accompanied by bacteria characteristic of vaginal flora in the conjunctiva of infants delivered vaginally compared those born via C-section delivery^28,29. Additionally, the composition of the ocular surface microbiome underwent changes subsequent to antibiotic prophylaxis involving gentamicin in newborns⁴. This outcome parallels our observations in mice. Compared to other factors, the composition of the ocular surface microbiota is less influenced by gender, consistent with findings in humans³⁰, with male and female mice found to differ only in the β-diversity of the conjunctival microbiome.

In other sites in newborn infants, such as the skin, gut and oral cavity, the initial microbiota evolves over time and adapts to the physiological variation^31,32. The current data support the concept of a gradual adaption of the conjunctival microbiome in 4-week-old mice to 8-week-old mice, which may be a result of host adaptation and interaction with the environment. Specifically, factors such as the enhanced barrier function of the ocular surface epithelial cells, increased release of the antibacterial components in the tear fluid, and the maturation of the ocular mucosal immune system may influence the pattern of microbial colonization. Concurrently, the behavioral changes in mice, such as grooming, solid food intake and exploration, may affect the contact of ocular surface with environmental microbes. This expanded environmental exposure may promote the colonization and selection of diverse microbial communities, thereby contributing to the gradual stabilization and maturation of the conjunctival microbiome. Intriguingly, microbial presence was also detected in the conjunctiva of 1-week-old postnatal mice, even when the eyelids were still closed. Given the extremely low microbial biomass in conjunctival tissues at this stage, the possibility of trace contamination cannot be completely excluded. However, several measures were taken to minimize the risk of contamination, including sterile dissection procedure and collection of negative controls. Further analysis showed that the microbial relative abundance in the negative controls was significant lower, although these controls did reveal a variety of potential contaminants identified in prior study³³. Furthermore, a significant distinction in the microbial composition was evident between the conjunctival microbiome and negative controls. Therefore, the microbial signals detected in conjunctiva of 1-week-old postnatal mice may not be from contamination. Meanwhile, we conducted additional filtration based on the IndVal index to address potential contaminations in the conjunctival microbiome. Following filtration, multiple dominant genera still persist in the conjunctival tissues derived from 1-week-old mice, which may be due to the nature of the developmental process of the mouse eye in utero³⁴, rather than being indicative of potential contamination. However, the origin of the pre-eye-opening ocular microbiome in mice is currently unestablished and ocular mucosa-microbial interactions before eye opening in mice needs further exploration.

Microbial cohabitants are major participants in shaping and maintaining essential physiological processes³⁵. The conjunctiva is colonized by a unique commensal microbiome, and like the skin, these commensal communities stimulate and educate defense and immune responses³⁶, which contribute to the development of symbiosis between the host and microbiome. Such symbiotic relationships depend on early events taking place during perinatal life³⁷. Our results suggest that postnatal conjunctival barrier maturation and immune responses accompany the dynamic changes of commensal microbiome and that these include conjunctival epithelial cell stratification, goblet cell development, mucin secretion and temporary recruitment of APCs into the conjunctival epithelium. Notably, in 3-week-old mice, BugBase analysis predicted a higher relative abundance of potential pathogenic bacteria in the conjunctiva compared to other developmental stages, with significant alterations observed in genera such as Enterococcus, Staphylococcus, and Sphingobacterium. This period also coincided with a peak in immune activation. Despite the presence of multiple known opportunistic pathogens within these genera, their increased abundance at this stage did not induce any significant pathological changes. Previous studies suggest that, at appropriate levels, certain opportunistic bacteria may contribute the establishment of host–microbe symbiosis and even play beneficial roles in immune system development³⁸. The massive influx of microbes and microbial antigens greatly influences the expansion and differentiation of a variety of different cell types³⁹. In both humans and mice, colonization by commensal flora during early life can drive differentiation of Treg cell, thereby suppressing inflammation^40,41. In our study, the heightened immune activation at 3 weeks, followed by its reduction at 4 weeks, may be due to Treg-mediated immune regulation during this critical postnatal period. It is noteworthy that, although our data suggest an association between postnatal conjunctival barrier maturation, immune response and conjunctival microbial colonization, direct evidence of a causative relationship is lacking. Therefore, future studies using germ-free mouse are essential to determine the effects of microbial colonization on the development of the ocular mucosal immune system.

Conjunctival goblet cells are located at the interface between the external environment and stromal immune cells, which allows them to easily convey information between the external antigen and the APCs. Meanwhile, the goblet cell products presumably modulate APC function in maintaining immunological tolerance on the ocular surface^42,43. In support of this, we note that extensions of dendritic cells surrounding goblet cells project across the epithelium towards the ocular surface in the conjunctiva of TSP1^−/− mice, which are reported to harbor increased microbial frequency at the ocular surface⁴⁴. We also found a significant increase in the number of APCs in the conjunctival epithelium of 3-week-old mice, and extensions adjacent to goblet cells demonstrated ability of ocular mucosal APCs to extend trans-epithelial dendrites in response to sudden bacterial colonization at the ocular surface. In addition, conjunctival goblet cells responding to TLR signaling⁴⁴, epithelial cell-derived cytokines and chemokines^45,46, as well as neuropeptides signaling⁴⁷, may also contribute to this process. Such interactions may facilitate tolerance to environmental exposure during eyelid-opening, although further studies will be needed to incorporate FISH for a more comprehensive assessment of bacterial localization in conjunctiva and to establish the mechanism by which microbes promote ocular surface immune tolerance during early postnatal life. Furthermore, GAPs are considered a route for antigen delivery to APCs²². We observed an increased distribution of OVA in the conjunctival stroma of 3 weeks of age, with partial uptake by APCs. However, OVA antigen delivery to the stromal area were reduced despite the continued presence of GAPs in 4-week-old mice. This phenomenon may be influenced by multiple regulatory pathways, including microbial, immune and neurotransmitter signaling^21,48,49.

After birth, bacterial colonization of mucosa plays critical roles in the development and education of the host mammalian immune system⁵⁰. This period is termed the “window of opportunity” and if there is an absence of appropriate microbial exposure, the immune consequences may elicit irreversible and potentially deleterious implications for the host including inflammatory bowel disease¹⁶, allergy^51,52 and asthma¹⁵. The “window of opportunity” may also exist at other mucosal surfaces, and in this study, we report that when the eyes 2-week-old open and are treated with antibiotics for 1 week, the ocular surface microbiome is significantly altered, with an inhibition of goblet cell development and immune cell recruitment. Simultaneously, in vitro cell experiments revealed that 0.3% tobramycin exhibited no anti-proliferation effect on conjunctival epithelial cells following a 24-h incubation. In the gut, symbionts play an indispensable role for intestinal stem cell-mediated epithelial development, the use of lactic-acid-producing bacteria type symbionts can significantly increase the expansion of intestinal stem cells, Paneth cells and goblet cells in mice⁵³. Therefore, these findings substantiate that the delayed maturation of the conjunctiva may result from antibiotic-mediated elimination of the conjunctival microbiome rather than a direct impact on conjunctival epithelial cells, and we note that conjunctival goblet cell reduction has also been found in germ-free mice⁵⁴.

Antibiotic exposure in early life may have long-term effects on microbial colonization. For example, in murine gut, microbiome may take at least 13 weeks to recover following antibiotic disruption⁵⁵, whereas the ocular surface microbiome in rabbits often returns within a few weeks after cessation of treatment⁵⁶. The changes of microbiome induced by antibiotics, and the time required to restore to age-matched microbial composition, depend not only on antibiotic clearance but also on microbial colonization dynamics and the immune status of the host^57,58. Although the conjunctival microbiome composition in antibiotic-treated mice still differed from that of controls, we observed a trend toward recovery of the conjunctival microbiome two weeks after cessation of antibiotic exposure. The changes may be attributed to early antibiotic exposure, which can disrupt the interaction between the conjunctival microbiome and the ocular mucosal immune system, thereby altering mucosal immune homeostasis and delaying the re-establishment of environmentally driven commensal communities. Notably, despite the partial microbial recovery, the immune response to antigens in the conjunctiva remained changed, manifested as an increased response to antigens and aggravated AED. This was not the case in other period mice, however, indicating that there may be a similar “window of opportunity” with regards to the ocular surface, in which exposure to microbes appears critical to ensure proper conjunctival maturation that in-turn leads to conjunctival mucosal immunity and the prevention of susceptibility to allergic responses in adulthood.

Antibiotic treatment during early life has been linked to allergy development^59,60. Although, the exact mechanisms are not fully elucidated, sufficient microbial exposure during early life appears essential for the development of healthy immune system. In humans and mice, neonatal CD4⁺ T cells are biased towards Th2-type responses. Microbial exposure has been suggested to modulate this bias through the induction of Th1-type pathways or expansion of Treg cells, thereby promoting immune balance⁶¹. However, antibiotic exposure in early life reduces microbial stimulation, promoting a shift toward Th2 polarization and contributing to the development of allergic diseases^62,63. In this study, early postnatal disruption of microbiome by antibiotic did not affect the AED-induced Th1 response. However, it exacerbated the Th2 response and further reduced Treg cell frequency, indicating that early-life microbial disruption may aggravate Th2-skewed immunity and impact Treg development or maintenance. Additional studies are needed to elucidate the mechanisms by which early-life crosstalk between the host and the conjunctival microbiome contributes to the exacerbation of AED.

AED impacts 10–20% of the population, exerting a significant negative impact on the ocular surface health and quality of life. Notably, among children, the highest prevalence occurs between 3 and 11 years of age⁶⁴. The World Allergy Organization is raising awareness about the escalating burden of allergic diseases globally, particularly in children and developing countries⁶⁵. Microbiological disruption may serve as a contributing factor to this phenomenon⁶⁶, a proposition supported by our research findings. Furthermore, during the establishment of AED in mice at various ages, we observed variations in the severity of the disease. Specifically, the allergic response in adult mice was notably more intense, possibly attributed to the mature immune development in adult mice and their heightened responsiveness to external allergens. Therefore, further investigation is warranted to explore the correlation between age and the severity of AED.

In summary, the colonization of microbiota in early life is crucial to the timely maturation of conjunctival mucosa and the establishment of ocular surface homeostasis (Fig. 8). The use of antibiotics in newborns may lead to a delayed ocular mucosal development and maturation of conjunctival mucosal immunity. If the cross-talk between the microbiome and conjunctival mucosa during early postnatal life is disrupted, delayed conjunctival maturation processes could aggravate the subsequent development of allergic eye disease. Therefore, the application of antibiotic eye drops in newborn infants should be more carefully considered and the established neonate ocular prophylaxis policy throughout the world may need to be revisited.

Methods

Animal breeding and selection

C57 black 6 (C57BL/6J) mice comprise one of the most common inbred strains of laboratory mice. In this study, C57BL/6J mice were chosen to investigate the changes in the composition of the ocular surface microbiome and the immunology and functional events taking place in the conjunctiva from postnatal week-1 onward to adulthood at 8-weeks of age. All the animals used in this study were bred in the animal facility of Xiamen University Laboratory Animal Center (Xiamen, China). Initially, monogamous breeding pairs were purchased from Xiamen University Laboratory Animal Center and used through in-breeding. Post-birth, the pups were primarily fed on breast milk until the age of 3 weeks, then the animals were randomly separated into individual cages (5 mice per cage) at the age of 3–4 weeks. At the specified age, healthy non-littermates were randomly picked from the breeding pairs and divided into females and males. Mice were maintained in a standard pathogen-free environment at 25 ± 1 °C with a relative humidity of 60 ± 10% and a 12-h light-dark cycle (from 8:00 a.m. to 8:00 p.m.). The mice were fed ad libitum feed and autoclaved water. Experimental protocols were in accordance with the Association for Research in Vision and Ophthalmology’s Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Experimental Animal Ethics Committee of Xiamen University (XMULAC20170014). We have complied with all relevant ethical regulations for animal use.

Sample description

To explore the changes in the composition of the ocular surface microbiome during development, after the mice euthanized by cervical dislocation, conjunctival tissue samples were isolated from 40 C57BL/6J mice (80 eyes) at the age of 1 week, 2 weeks, 3 weeks, 4 weeks, 6 weeks, and 8 weeks under aseptic conditions, respectively. The dissection process was undertaken under sterilized conditions to avoid contamination. At each time point, 80 conjunctival tissue samples were collected and immediately frozen at −80 °C until further processing. Conjunctival tissue samples from 8 eyes were pooled as a single research sample, resulting in a total of 10 research samples per time point, comprising 5 samples from males and 5 from females. To address the challenges posed by contamination aligned to inefficiencies in microbiome analyses of low-biomass samples⁶⁷, at each sample collection event we collected laboratory environmental samples using a sterile swab exposed to the surrounding environment. These served as negative controls. Furthermore, feces samples in bedding were collected at the sampling time points and stored as described above to analyze the “environmental microbiome”.

Antibiotic treatment

Tobramycin (Sangon Biotech, Shanghai, China) was dissolved in saline to make 0.3% tobramycin solution under sterile conditions. 2.5 μl of 0.3% tobramycin solution was applied to each eye of 2-week-old C57BL/6J mice three times daily. The control group received saline solution. Mice received the antibiotic/saline for 7 days until they were 3-week-old, at which time antibiotic/saline use was discontinued. The mice were euthanized by cervical dislocation at 3-, 4-, and 5-weeks of age. The conjunctival tissue sample collection at each time point was consistent with the descriptions above.

Induction of allergic eye disease (AED) model

Mice were immunized with a subcutaneous injection of 100 μg ovalbumin (OVA, InvivoGen, vac-stova), 300 ng pertussis toxin (Abcam, ab124299) and 1 mg aluminum hydroxide (Acmec, A19200) in 100 μl PBS. After 2 weeks, mice were challenged once/day with 5 μl OVA (50 mg/ml) in PBS, applied as eye drops to each eye for 7 days. Naive mice did not receive the OVA challenge. Eyelid edema, tearing/discharge, chemosis and hyperemia were scored individually on a 0–3+ scale in a masked fashion 20 min after challenge each day for the 7 days of challenge.

DNA extraction and bacterial 16S rRNA gene sequencing

Detailed procedures DNA extraction, amplification, 16S rRNA gene sequencing, and data processing are provided in Supplementary materials and methods. Laboratory environmental samples along with blank extractions and reagents used at the time of amplification were also processed and analyzed as potential sources of contamination.

Microbiome data analysis

Based on the OTU information, α-diversity metrics of the Shannon index were calculated with QIIME software package. Bray–Curtis dissimilarities were produced as β-diversity measures and subjected to principal coordinate analysis (PCoA) with the vegan package and QIIME software package. Different groups were statistically compared using the similarity analysis, ANOSIM. The overlap of the microbial communities was determined by the R values from the ANOSIM according to the method of Buttigieg and Ramette⁶⁸. To address the risk of contamination during the analysis of low-biomass samples, we employed the indicspecies R package to evaluate the specificity of an OTU with respect to the conjunctiva or the negative control^33,69. This was achieved by calculating its IndVal index, which helped identify potential contaminants¹⁷. OTUs linked to negative controls were designated as contaminants and subsequently excluded from the analysis.

Microbial source tracking (FEAST)⁷⁰ was applied to predict the sources of the microbial communities detected in each of the six time points investigated across weeks 1-to-8. The co-occurrence networks were constructed to explore the internal community relationships across the samples¹⁸. In order to visualize associations in the network, we constructed a correlation matrix by calculating the possible pairwise Spearman’s rank correlations. A correlation between two nodes was considered to be statistically robust if the Spearman’s correlation coefficient was over 0.6 or less than −0.6 and the P < 0.01. The correlation networks were visualized using Gephi software. All microbiome analyses were conducted in R (version 4.2.3, R Development Core Team) unless otherwise stated.

Whole-mount immunostaining

Freshly harvested whole-mount conjunctival tissue obtained at each developmental stage were fixed with cold acetone at −20 °C and later stained as described in a previous study⁷¹ using primary anti-CD45 (1:150), anti-CD11b (1:200), anti-CD11c (1:200), anti-MHC II (1:100) or anti-Keratin7 (Krt7, 1:100) antibodies. Digital confocal images were captured using a laser scanning confocal microscope (FV1000MPE-B, Olympus, Japan) and a 1.16 μm Z-step. The images were processed using FV10-ASW 4.2 viewer (Olympus). The antibodies used are shown in Table S5.

Flow cytometry and analysis

Conjunctival tissues were excised and exposed to 1 mg/ml collagenase (Roche, 11088866001) for 1 h at 37 °C to obtain cell suspensions, which were stained with anti-CD45 antibodies at 4 °C for 30 min. A CytoFLEX S Flow Cytometer (Beckman Coulter, Indianapolis, IN, USA) was used and data were plotted and quantified with FlowJo software (FlowJo LLC, Ashland, OR, USA). The antibodies used are shown in Table S5. The gating strategy are shown in Fig. S7.

CLNs were collected from freshly euthanized mice and processed. Single cell suspensions were prepared and 10⁶ cells were stimulated for 4 h with leukocyte activation cocktail (BD Pharmingen, 550583). Cells were stained with Fixable Viability Dye and surface antigen CD45, CD3, CD4, and CD25 for 30 min on ice. For intracellular staining, cells were prepared with transcription factor buffer (BD Pharmingen, 562574) and subsequently stained with IL-4 and IFN-γ antibodies. Data were acquired on SONY ID7000™ Spectral Cell Analyzer using the ID A analysis software (v2.0; SONY). Data were analyzed on FlowJo software (FlowJo LLC, Ashland, OR, USA). The antibodies used are shown in Table S5. The gating strategy are shown in Fig. S7.

In vivo administration of fluorescent antigens

A 2.5 μl eye drop of ovalbumin Alexafluor 647 (O34784, Thermofisher) was applied to each eye. The mice were then euthanized by cervical dislocation 30 min later. Whole-mount freshly harvested conjunctival tissues were fixed with cold methanol at −20 °C for 10 min and stained using Hoechst dye.

Histology

Eyes, including eyelid and conjunctival tissues from each group, were surgically excised and embedded in optimal cutting temperature (OCT) compound or paraffin, cut into sagittal sections (5 μm thick), and stored at −80 °C (frozen sections) or room temperature (paraffin sections). Hematoxylin and Eosin staining, PAS staining, Immunohistochemical staining, Picrosirius Red staining and Toluidine Blue staining were performed on paraffin sections. Immunofluorescence staining and TUNEL assay were performed on the frozen sections.

Examination of conjunctival goblet cells

For PAS staining, paraffin sections were stained using a PAS staining Kit (395B-1 KT, Sigma) according to the manufacturer’s instructions. The superior and inferior conjunctiva were then examined and photographed with a digital light microscope (Eclipse 50i; Nikon, Tokyo, Japan) and the mucin-filled goblet cells counted. Three sections from each of the nasal, central and lateral parts of the eye from each animal were studied. For moxifloxacin-based fluorescence imaging, moxifloxacin ophthalmic solution (Vigamox, Alcon Laboratories, Fort Worth, USA) was used as a labeling agent, as previously reported⁷². The excised mouse conjunctival tissues were immersed in moxifloxacin ophthalmic solution for 3–5 min and imaged using a laser scanning confocal microscope (FV1000MPE-B, Olympus).

Immunofluorescent staining

For immunofluorescence staining, frozen sections were fixed in cold acetone at −20 °C for 10 min, followed by three washes with PBS for 5 min each and incubation in 0.2% Triton X-100 for 20 min. After washing the sections three times each with PBS for 5 min and pre-incubating with 2% bovine serum albumin (BSA) for 1 h at room temperature, sections were incubated with anti-Ki67 (1:300), anti-ZO-1 (1:200), anti-Krt19 (1:200), anti-MHC II (1:100) and anti-CD4 (1:200) antibodies for 16 h at 4 °C. The slides were then washed three times each with PBS for 10 min and incubated with secondary antibodies for 1 h in the dark at room temperature, followed by three washes each with PBS for 10 min. Sections were then counterstained with DAPI for 5 min, after which digital images of representative areas of the conjunctiva were taken with a laser scanning confocal microscope (FV1000MPE-B, Olympus). The antibodies used are shown in Table S5.

Quantitation of Sera IgE

Blood was collected from the submandibular vein of mice following topical challenge and serum was isolated. Total IgE was measured via ELISA, as per the manufacturer’s instructions (EM024-96, ExCell, China).

Quantitative real-time PCR analysis

Total RNA from conjunctiva was extracted using the RNeasy Mini Kit (Qiagen) following the manufacturer’s protocol. RNA quality was confirmed using Nanodrop (ThermoFisher) and reverse transcribed into cDNA by using a cDNA synthesis kit (RR047A; TakaRa, Japan). Quantitative PCR was performed with a Step One Plus Real-Time System (Applied Biosystems, Alameda, CA, USA). Table S6 provides the primer sequences used to amplify specific gene products. Expression data were normalized with β-actin as an endogenous reference.

Statistics and reproducibility

Statistical analyses were conducted with Prism 7 software (GraphPad, CA), without microbiome data analysis. The numbers of positive cells in whole-mount stained images of conjunctiva were measured using Image J (NIH). Unpaired two-tailed Student’s t test was used to assess the difference between the two groups. For continuous variables of three groups or more, a one-way analysis of variance (ANOVA) test with a Tukey-adjusted pairwise comparison was performed. Data are expressed as means ± SEM. No statistical method was used to predetermine sample size of conjunctival 16S rRNA sequencing. We determined sample sizes by referring to previously published paper in the field⁷³. P values of <0.05 were considered significant. Exact P values are provided in the source data (Supplementary Data).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All study data have been comprehensively incorporated within the article and/or Supplementary Information. The datasets of 16S rRNA genes generated during the current study have been deposited in the NCBI Sequence Read Archive (SRA) database (Accession Number: PRJNA923858). The source data behind the graphs in the main figures are provided in Supplementary Data. The other data are available from the corresponding author on reasonable request.

References

Kugadas, A. & Gadjeva, M. Impact of microbiome on ocular health. Ocul. Surf. 14, 342–349 (2016).
Article PubMed PubMed Central Google Scholar
Watters, G. A., Turnbull, P. R., Swift, S., Petty, A. & Craig, J. P. Ocular surface microbiome in meibomian gland dysfunction. Clin. Exp. Ophthalmol. 45, 105–111 (2017).
Article PubMed Google Scholar
Venugopal, R. et al. Conjunctival microbial flora in ocular Stevens-Johnson syndrome sequelae patients at a tertiary eye care center. Cornea 35, 1117–1121 (2016).
Article PubMed Google Scholar
Petrillo, F. et al. Characterization and comparison of ocular surface microbiome in newborns. Microorganisms 10, 1390 (2022).
Article CAS PubMed PubMed Central Google Scholar
Cavuoto, K. M., Galor, A. & Banerjee, S. Anatomic characterization of the ocular surface microbiome in children. Microorganisms 7, 259 (2019).
Article CAS PubMed PubMed Central Google Scholar
Cavuoto, K. M., Mendez, R., Miller, D., Galor, A. & Banerjee, S. Effect of clinical parameters on the ocular surface microbiome in children and adults. Clin. Ophthalmol. 12, 1189–1197 (2018).
Article CAS PubMed PubMed Central Google Scholar
Chen, F. V., Chang, T. C. & Cavuoto, K. M. Patient demographic and microbiology trends in bacterial conjunctivitis in children. J. AAPOS 22, 66–67 (2018).
Article PubMed Google Scholar
Pak, K. Y., Kim, S. I. & Lee, J. S. Neonatal bacterial conjunctivitis in Korea in the 21st century. Cornea 36, 415–418 (2017).
Article PubMed Google Scholar
Darling, E. K. & McDonald, H. A meta-analysis of the efficacy of ocular prophylactic agents used for the prevention of gonococcal and chlamydial ophthalmia neonatorum. J. Midwifery Women’s. Health 55, 319–327 (2010).
Article Google Scholar
Bomar, L., Brugger, S. D. & Lemon, K. P. Bacterial microbiota of the nasal passages across the span of human life. Curr. Opin. Microbiol. 41, 8–14 (2018).
Article PubMed Google Scholar
Capone, K. A., Dowd, S. E., Stamatas, G. N. & Nikolovski, J. Diversity of the human skin microbiome early in life. J. Investig. Dermatol. 131, 2026–2032 (2011).
Article CAS PubMed Google Scholar
Stewart, C. J. et al. Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature 562, 583–588 (2018).
Article CAS PubMed PubMed Central Google Scholar
Blaser, M. J. The theory of disappearing microbiota and the epidemics of chronic diseases. Nat. Rev. Immunol. 17, 461–463 (2017).
Article CAS PubMed Google Scholar
Gensollen, T., Iyer, S. S., Kasper, D. L. & Blumberg, R. S. How colonization by microbiota in early life shapes the immune system. Science 352, 539–544 (2016).
Article CAS PubMed PubMed Central Google Scholar
Russell, S. L. et al. Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma. EMBO Rep. 13, 440–447 (2012).
Article CAS PubMed PubMed Central Google Scholar
Al Nabhani, Z. et al. A weaning reaction to microbiota is required for resistance to immunopathologies in the adult. Immunity 50, 1276–1288.e1275 (2019).
Article CAS PubMed Google Scholar
Ozkan, J. et al. Biogeography of the human ocular microbiota. Ocul. Surf. 17, 111–118 (2019).
Article PubMed Google Scholar
Sun, H., Chen, F., Hao, H. & Wang, K. J. Multi-dimensional investigation and distribution characteristics analysis of gut microbiota of different marine fish in Fujian Province of China. Front. Microbiol. 13, 918191 (2022).
Article PubMed PubMed Central Google Scholar
Bian, F. et al. Age-associated antigen-presenting cell alterations promote dry-eye inducing Th1 cells. Mucosal Immunol. 12, 897–908 (2019).
Article CAS PubMed PubMed Central Google Scholar
Gupta, D., Harvey, S. A., Kaminski, N. & Swamynathan, S. K. Mouse conjunctival forniceal gene expression during postnatal development and its regulation by Kruppel-like factor 4. Investig. Ophthalmol. Vis. Sci. 52, 4951–4962 (2011).
Article CAS Google Scholar
Knoop, K. A. et al. Microbial antigen encounter during a preweaning interval is critical for tolerance to gut bacteria. Sci. Immunol. 2, eaao1314 (2017).
Barbosa, F. L. et al. Goblet cells contribute to ocular surface immune tolerance-implications for dry eye disease. Int. J. Mol. Sci. 18, 978 (2017).
Article PubMed PubMed Central Google Scholar
Tukler Henriksson, J., Coursey, T. G., Corry, D. B., De Paiva, C. S. & Pflugfelder, S. C. IL-13 stimulates proliferation and expression of mucin and immunomodulatory genes in cultured conjunctival goblet cells. Investig. Ophthalmol. Vis. Sci. 56, 4186–4197 (2015).
Article Google Scholar
Al Nabhani, Z. & Eberl, G. Imprinting of the immune system by the microbiota early in life. Mucosal Immunol. 13, 183–189 (2020).
Article CAS PubMed Google Scholar
Saban, D. R. et al. Resolvin D1 treatment on goblet cell mucin and immune responses in the chronic allergic eye disease (AED) model. Mucosal Immunol. 12, 145–153 (2019).
Article CAS PubMed Google Scholar
Yau, J. W. et al. Characterization of ocular and nasopharyngeal microbiome in allergic rhinoconjunctivitis. Pediatr. Allergy Immunol. 30, 624–631 (2019).
Article PubMed Google Scholar
Cavuoto, K. M., Banerjee, S., Miller, D. & Galor, A. Composition and comparison of the ocular surface microbiome in infants and older children. Transl. Vis. Sci. Technol. 7, 16 (2018).
Article PubMed PubMed Central Google Scholar
Eder, M. et al. Normal ocular flora in newborns delivered in two hospital centers in Argentina and Paraguay. Graefes Arch. Clin. Exp. Ophthalmol. 243, 1098–1107 (2005).
Article PubMed Google Scholar
Isenberg, S. J., Apt, L., Yoshimori, R., Ward McCarty, J. & Alvarez, S. R. Source of the conjunctival bacterial flora at birth and implications for ophthalmia neonatorum prophylaxis. Am. J. Ophthalmol. 106, 458–462 (1988).
Article CAS PubMed Google Scholar
Wen, X. et al. The influence of age and sex on ocular surface microbiota in healthy adults. Investig. Ophthalmol. Vis. Sci. 58, 6030–6037 (2017).
Article CAS Google Scholar
Koenig, J. E. et al. Succession of microbial consortia in the developing infant gut microbiome. Proc. Natl. Acad. Sci. USA 108, 4578–4585 (2011).
Article CAS PubMed Google Scholar
Yatsunenko, T. et al. Human gut microbiome viewed across age and geography. Nature 486, 222–227 (2012).
Article CAS PubMed PubMed Central Google Scholar
Salter, S. J. et al. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biol. 12, 87 (2014).
Article PubMed PubMed Central Google Scholar
Pei, Y. F. & Rhodin, J. A. The prenatal development of the mouse eye. Anat. Rec. 168, 105–125 (1970).
Article CAS PubMed Google Scholar
Harris-Tryon, T. A. & Grice, E. A. Microbiota and maintenance of skin barrier function. Science 376, 940–945 (2022).
Article CAS PubMed Google Scholar
St Leger, A. J. et al. An ocular commensal protects against corneal infection by driving an interleukin-17 response from mucosal γδ T cells. Immunity 47, 148–158.e145 (2017).
Article PubMed Central Google Scholar
Torow, N. & Hornef, M. W. The neonatal window of opportunity: setting the stage for life-long host-microbial interaction and immune homeostasis. J. Immunol. 198, 557–563 (2017).
Article CAS PubMed Google Scholar
Chisari, G., Chisari, E. M., Borzi, A. M. & Chisari, C. G. Aging eye microbiota in dry eye syndrome in patients treated with Enterococcus faecium and Saccharomyces boulardii. Curr. Clin. Pharm. 12, 99–105 (2017).
Article CAS Google Scholar
Ignacio, A., Czyz, S. & McCoy, K. D. Early life microbiome influences on development of the mucosal innate immune system. Semin. Immunol. 73, 101885 (2024).
Article CAS PubMed Google Scholar
Ohnmacht, C. et al. MUCOSAL IMMUNOLOGY. The microbiota regulates type 2 immunity through RORγt⁺ T cells. Science 349, 989–993 (2015).
Article CAS PubMed Google Scholar
Yang, B. H. et al. Foxp3(+) T cells expressing RORγt represent a stable regulatory T-cell effector lineage with enhanced suppressive capacity during intestinal inflammation. Mucosal Immunol. 9, 444–457 (2016).
Article CAS PubMed Google Scholar
Contreras-Ruiz, L. & Masli, S. Immunomodulatory cross-talk between conjunctival goblet cells and dendritic cells. PLoS ONE 10, e0120284 (2015).
Article PubMed PubMed Central Google Scholar
Alam, J., de Paiva, C. S. & Pflugfelder, S. C. Immune-goblet cell interaction in the conjunctiva. Ocul. Surf. 18, 326–334 (2020).
Article PubMed PubMed Central Google Scholar
Logeswaran, A., Contreras-Ruiz, L. & Masli, S. Conjunctival goblet cell responses to TLR5 engagement promote activation of local antigen-presenting cells. Front. Immunol. 12, 716939 (2021).
Article CAS PubMed PubMed Central Google Scholar
Niess, J. H. et al. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance. Science 307, 254–258 (2005).
Article CAS PubMed Google Scholar
Iwasaki, A. Mucosal dendritic cells. Annu Rev. Immunol. 25, 381–418 (2007).
Article CAS PubMed Google Scholar
Perner, C. et al. Substance P release by sensory neurons triggers dendritic cell migration and initiates the type-2 immune response to allergens. Immunity 53, 1063–1077.e1067 (2020).
Article CAS PubMed PubMed Central Google Scholar
Knoop, K. A., McDonald, K. G., McCrate, S., McDole, J. R. & Newberry, R. D. Microbial sensing by goblet cells controls immune surveillance of luminal antigens in the colon. Mucosal Immunol. 8, 198–210 (2015).
Article CAS PubMed Google Scholar
Kimura, M. et al. A nerve-goblet cell association promotes allergic conjunctivitis through rapid antigen passage. JCI Insight 8, e168596 (2023).
Article PubMed PubMed Central Google Scholar
Tamburini, S., Shen, N., Wu, H. C. & Clemente, J. C. The microbiome in early life: implications for health outcomes. Nat. Med. 22, 713–722 (2016).
Article CAS PubMed Google Scholar
Bashir, M. E., Louie, S., Shi, H. N. & Nagler-Anderson, C. Toll-like receptor 4 signaling by intestinal microbes influences susceptibility to food allergy. J. Immunol. 172, 6978–6987 (2004).
Article CAS PubMed Google Scholar
Gollwitzer, E. S. et al. Lung microbiota promotes tolerance to allergens in neonates via PD-L1. Nat. Med. 20, 642–647 (2014).
Article CAS PubMed Google Scholar
Lee, Y. S. et al. Microbiota-derived lactate accelerates intestinal stem-cell-mediated epithelial development. Cell Host Microbe 24, 833–846.e836 (2018).
Article CAS PubMed Google Scholar
Wang, C. et al. Sjögren-like lacrimal keratoconjunctivitis in germ-free mice. Int. J. Mol. Sci. 19, 565 (2018).
Lynn, M. A. et al. Early-life antibiotic-driven dysbiosis leads to dysregulated vaccine immune responses in mice. Cell Host Microbe 23, 653–660 e655 (2018).
Article CAS PubMed Google Scholar
Xue, R. et al. Temporal impacts of topical ceftazidime and tobramycin-vancomycin mixtures on the ocular surface microbiota in rabbits. Exp. Eye Res. 220, 109098 (2022).
Article CAS PubMed Google Scholar
Langdon, A., Crook, N. & Dantas, G. The effects of antibiotics on the microbiome throughout development and alternative approaches for therapeutic modulation. Genome Med. 8, 39 (2016).
Article PubMed PubMed Central Google Scholar
Littmann, E. R. et al. Host immunity modulates the efficacy of microbiota transplantation for treatment of Clostridioides difficile infection. Nat. Commun. 12, 755 (2021).
Article CAS PubMed PubMed Central Google Scholar
Azad, M. B. et al. Infant gut microbiota and food sensitization: associations in the first year of life. Clin. Exp. Allergy 45, 632–643 (2015).
Article CAS PubMed Google Scholar
Russell, S. L. et al. Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma. Gut Microbes 4, 158–164 (2013).
Article PubMed PubMed Central Google Scholar
Donald, K. & Finlay, B. B. Early-life interactions between the microbiota and immune system: impact on immune system development and atopic disease. Nat. Rev. Immunol. 23, 735–748 (2023).
Article CAS PubMed Google Scholar
Kuo, C. H. et al. Early life exposure to antibiotics and the risk of childhood allergic diseases: an update from the perspective of the hygiene hypothesis. J. Microbiol. Immunol. Infect. 46, 320–329 (2013).
Article CAS PubMed Google Scholar
Oyama, N., Sudo, N., Sogawa, H. & Kubo, C. Antibiotic use during infancy promotes a shift in the T(H)1/T(H)2 balance toward T(H)2-dominant immunity in mice. J. Allergy Clin. Immunol. 107, 153–159 (2001).
Article CAS PubMed Google Scholar
Das, A. V., Donthineni, P. R., Sai Prashanthi, G. & Basu, S. Allergic eye disease in children and adolescents seeking eye care in India: electronic medical records driven big data analytics report II. Ocul. Surf. 17, 683–689 (2019).
Article PubMed Google Scholar
Pawankar, R. Allergic diseases and asthma: a global public health concern and a call to action. World Allergy Organ J. 7, 12 (2014).
Article PubMed PubMed Central Google Scholar
Cavuoto, K. M., Stradiotto, A. C. & Galor, A. Role of the ocular surface microbiome in allergic disease. Curr. Opin. Allergy Clin. Immunol. 19, 482–487 (2019).
Article CAS PubMed Google Scholar
Minich, J. J. et al. KatharoSeq enables high-throughput microbiome analysis from low-biomass samples. mSystems 3, e00218 (2018).
Buttigieg, P. L. & Ramette, A. A guide to statistical analysis in microbial ecology: a community-focused, living review of multivariate data analyses. FEMS Microbiol. Ecol. 90, 543–550 (2014).
Article CAS PubMed Google Scholar
De Cáceres, M. & Legendre, P. Associations between species and groups of sites: indices and statistical inference. Ecology 90, 3566–3574 (2009).
Article PubMed Google Scholar
Shenhav, L. et al. FEAST: fast expectation-maximization for microbial source tracking. Nat. Methods 16, 627–632 (2019).
Article CAS PubMed PubMed Central Google Scholar
Ko, B. Y. Xiao, Y. Barbosa, F. L. de Paiva, C. S. & Pflugfelder, S. C. Goblet cell loss abrogates ocular surface immune tolerance. JCI Insight 3, e98222 (2018).
Kim, S. et al. In vivo fluorescence imaging of conjunctival goblet cells. Sci. Rep. 9, 15457 (2019).
Article PubMed PubMed Central Google Scholar
Singh, N., Vats, A., Sharma, A., Arora, A. & Kumar, A. The development of lower respiratory tract microbiome in mice. Microbiome 5, 61 (2017).
Article PubMed PubMed Central Google Scholar

Download references

Acknowledgements

The authors thank Jingru Huang, Haiping Zheng, Baoying Xie and Xiang You from the Core Facility of Biomedical Science, Xiamen University for technical support. This study is supported, in part, by the National Key R&D Program of China (2018YFA0107301 to W.L.), the National Natural Science Foundation of China (NSFC; No. 81770894 to W.L., No. 81970773 to W.L. and No. 82471047 to W.L.), the Fujian Provincial Natural Science Foundation of China (No. 2025J011398 to Y.L.) and the Xiamen Health and Wellness High-Quality Development Science and Technology Program (No. 2024GZL-QN004 to Y.L.). The funders played no role in the study design, data collection and analysis, decision on publishing or preparation of the manuscript.

Author information

These authors contributed equally: Yanbo Liu, Hang Sun, Shulin Song.

Authors and Affiliations

Xiamen Eye Center and Eye Institute of Xiamen University, Fujian Provincial Key Laboratory of Ophthalmology and Visual Science, School of Medicine, Xiamen University, Xiamen, Fujian, China
Yanbo Liu, Shulin Song, Xiezhou He, Han Wu, Shangkun Ou, Hui He, Rongrong Zong, Yongxiong Chen, Huping Wu, Zuguo Liu & Wei Li
State Key Laboratory of Marine Environmental Science, College of Ocean and Earth Sciences, Xiamen University, Xiamen, Fujian, China
Hang Sun
Cancer Research Center of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian, China
Guo Fu
Wisdom Lake Academy of Pharmacy, Xi’an Jiaotong-Liverpool University, Suzhou, Jiangsu, China
Yiqiang Wang
School of Optometry and Vision Sciences, Cardiff University, Cardiff, Wales, UK
Andrew J. Quantock
Department of Ophthalmology, Xiang’an Hospital of Xiamen University, Xiamen, Fujian, China
Zuguo Liu & Wei Li

Authors

Yanbo Liu
View author publications
Search author on:PubMed Google Scholar
Hang Sun
View author publications
Search author on:PubMed Google Scholar
Shulin Song
View author publications
Search author on:PubMed Google Scholar
Xiezhou He
View author publications
Search author on:PubMed Google Scholar
Han Wu
View author publications
Search author on:PubMed Google Scholar
Shangkun Ou
View author publications
Search author on:PubMed Google Scholar
Hui He
View author publications
Search author on:PubMed Google Scholar
Rongrong Zong
View author publications
Search author on:PubMed Google Scholar
Yongxiong Chen
View author publications
Search author on:PubMed Google Scholar
Guo Fu
View author publications
Search author on:PubMed Google Scholar
Yiqiang Wang
View author publications
Search author on:PubMed Google Scholar
Huping Wu
View author publications
Search author on:PubMed Google Scholar
Andrew J. Quantock
View author publications
Search author on:PubMed Google Scholar
Zuguo Liu
View author publications
Search author on:PubMed Google Scholar
Wei Li
View author publications
Search author on:PubMed Google Scholar

Contributions

Y.L., H.S., S.S., and W.L. designed research; Y.L., H.S., S.S., X.H. and H.W. (Han Wu) performed research and the data analysis; G.F., Y.W., Z.L., H.W. (Huping Wu) and W.L. supervised the work and analyzed the data; S.O., R.Z., H.H., and Y.C. contributed reagents, materials, and analytic tools; Y.L., A.J.Q., and W.L. wrote the paper with the contribution of all authors. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Wei Li.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Biology thanks the anonymous reviewers for their contribution to the peer review of this work. Primary handling editor: Dario Ummarino.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplement Information

Description of Additional Supplementary Materials

Supplementary Data

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Liu, Y., Sun, H., Song, S. et al. Microbial influx during early postnatal life fortifies the ocular surface and guards against allergic eye disease in mice. Commun Biol 8, 1694 (2025). https://doi.org/10.1038/s42003-025-09095-4

Download citation

Received: 28 October 2024
Accepted: 18 October 2025
Published: 26 November 2025
Version of record: 26 November 2025
DOI: https://doi.org/10.1038/s42003-025-09095-4