Skip to main content
NCBI home page
BMC Microbiology logoLink to BMC Microbiology
. 2026 Mar 5;26:335. doi: 10.1186/s12866-026-04878-z

Uncovering the relationship between the human ocular surface microbiome and gut microbiome

Nina Eldridge 1, Livia Spörri 1, Marco Kreuzer 2, Gianna Haldimann 1, Martin S Zinkernagel 1,3, Denise C Zysset-Burri 1,3,✉
PMCID: PMC13069832  PMID: 41781852

Abstract

Background

The human body harbors diverse microbial communities that play essential roles in health and disease. While the gut microbiome (GM) has been extensively studied and linked to numerous systemic conditions, the ocular surface microbiome (OSM) remains less well understood due to its low microbial biomass and technical challenges in sequencing. Previous work has suggested possible connections between gut dysbiosis and ocular disease, but whether the OSM and GM are directly related remains unclear.

Results

This study investigated the relationship between the ocular surface microbiome (OSM) and the gut microbiome (GM) by analysing matched samples from both niches in the same individuals. Using high-throughput sequencing, we characterised microbial composition and diversity at each niche, followed by statistical analysis to show common taxa. A Mantel test was applied to assess potential correlations between the two microbial communities. Our results revealed no significant correlation between the composition of the OSM and GM, neither in structure nor in diversity.

Conclusion

This study suggests that the two niches are shaped independently, likely due to their distinct environmental conditions and microbial densities. These findings highlight the uniqueness of each microbiome and underscore the need for niche specific approaches when investigating host microbiome interactions.

Trial registration

ClinicalTrials.gov, TRN: NCT02438111, Registration date: 28 April 2015, and TRN: NCT 04658238, Registration date: 01 December 2020.

Keywords: Ocular surface microbiome, Gut microbiome, Whole metagenome shotgun sequencing

Background

The ocular surface microbiome (OSM) refers to the community of microorganisms and their genomes that inhabit the surface structures of the eye, including the lid, cornea and conjunctiva [1–6]. Although initially considered a sterile environment due to its constant exposure to antimicrobial components in tears and blinking, recent advances in high-throughput sequencing technologies have revealed a resident microbiome even in this low biomass site [7–11]. The OSM contributes to ocular surface health and disease, playing roles in local immune regulation, competitive inhibition of pathogens and maintenance of epithelial barrier integrity [12–16]. Studies suggest that the composition of the OSM is influenced by both host and environmental factors [17–20]. Disruptions in this microbial balance (dysbiosis) have been associated with ocular diseases [17–19]. However, OSM research remains in its early stages in terms of stability, interindividual variability and potential for therapeutic treatment [1, 11, 21–23].

In contrast, the gut microbiome (GM) is one of the most densely populated and metabolically active microbial ecosystems in the human body, consisting of up to 100 trillion microorganisms [16, 24, 25]. This complex and dynamic community plays a central role in host physiology, including digestion, energy metabolism, immune system development and the maintenance of intestinal homeostasis [16, 26]. Characterised primarily through 16S rRNA gene sequencing and whole metagenome shotgun sequencing, the GM shows interindividual variation and is shaped by factors such as diet, age, genetics, medication and environment [24, 27, 28]. Alterations in gut microbial composition have been linked to a wide range of diseases, including metabolic disorders, inflammatory bowel disease, neurodegenerative conditions and cardiovascular disease [20, 29, 30]. More recently, the gut eye axis has gained attention, as studies have suggested that dysbiosis in the gut may influence autoimmune response in the eye [31–38].

To date, studies of the OSM and the GM have been conducted independently. In this study, samples from both the ocular surface and the gut were collected from each participant, providing a unique opportunity to investigate potential correlations between these two distinct microbial niches. Although this approach allows to better understand how different microbiome niches interact with each other, the analysis poses significant challenges. Since the gut is a high-biomass, anaerobic environment with a densely populated and metabolically rich microbial community, current sequencing technologies may be applied to characterise its microbiome [11, 16, 39–41]. In contrast, the ocular surface represents a low biomass niche with constant exposure to air, light and antimicrobial tear components, which results in sparse microbial colonisation and makes sequencing more technically demanding due to the risk of contamination and low DNA yields [11, 42–44]. These fundamental differences in microbial density and environmental conditions must be considered in both, the interpretation of results and the optimisation of analytical methods.

The aim of this retrospective study was to investigate whether there is a correlation between the microbial communities of the ocular surface and the gut on a community level and on an individual basis. By comparing the composition and diversity of microbiomes from both niches, we determined whether patterns or shared features exist despite the differences in their environments. Identifying such correlations could reveal common factors shaping microbiome composition across body sites. Ultimately, these findings may contribute to a better understanding of microbiome interactions and open new possibilities for diagnostics or therapeutic strategies targeting microbiome related diseases.

Methods

Study design and recruitment

This research involved a retrospective analysis of microbiome samples from a cohort of 51 participants whose samples were originally collected as part of Spörri et al. [45] and another unpublished study of our Department. The studies were approved by the Ethics Committee of the Canton of Bern (ClinicalTrials.gov: NCT02438111 and NCT04658238). Secondary use of the samples for this analysis was also approved under the same ethical protocol. Participants were recruited from the Department of Ophthalmology of the University Hospital Bern (Inselspital), Switzerland. Study procedures adhered to the CONSORT guidelines as well as to the principles outlined in the Declaration of Helsinki and the International Ethical Guidelines for Biomedical Research Involving Human Subjects, as established by the Council for International Organizations of Medical Sciences. Written informed consent was obtained from all participants after they were provided with both oral and written explanations of the study. Individuals were excluded if they had undergone ocular surgery, taken regular medication, had a systemic disease or had received systemic or topical antibiotics within the preceding three months. To be included in the study, participants had to meet the following criteria: willingness and ability to provide written informed consent, age of 18 years or older, being non-smoker, not wearing contact lenses and no history of ocular surgery in the preceding three months.

Sample collection

After written informed consent was given and no exclusion criteria were met, a total of 51 participants were included in the study (M: 22 (43.1%), F: 29 (56.9%); Mean age 73.04, s.d. ±9.45). To avoid temporal variation in samples, sampling was conducted at the same time of the day and during the same season. After applying a local anaesthetic (Tetracaine 1%, Théa, Schaffhausen, Switzerland), conjunctival swabs using flocked nylon swabs (FLOQSwabs #518CS01, Copan, Brescia, Italy) were collected.

Stool samples were collected by the study participants, immediately cooled, and transported to the research facility within 16 h after defecation. Upon arrival, samples were promptly frozen at − 20 °C until further analysis.

DNA isolation and sequencing

The ocular surface swabs were processed, including host-DNA depletion, and DNA was isolated on the same day using the QIAamp DNA Microbiome Kit (51704) from QIAGEN (Hilden, Germany). Until further processed, DNA samples were stored at -20 °C. Library preparation was carried out using the NEBNext Ultra II DNA Library Prep Kit (SKU E7645L). Dual-indexed libraries were subsequently sequenced on an Illumina NovaSeq 6000 system using 150 bp paired-end reads. Whole metagenome shotgun sequencing was conducted at the Next-Generation Sequencing Platform, University of Bern, Switzerland as described in Spörri et al. [45].

Stool samples were processed and DNA isolated using the PSP® Spin Stool DNA Plus kit (Stratec Biomedical AG, Beringen, Switzerland). Until further processed, DNA samples were stored at -20 °C. The TruSeq DNA PCR-Free Library Preparation kit was used for library preparation and whole metagenome shotgun sequencing was done on an Illumina NovaSeq 6000 system by the Next-Generation Sequencing Platform of the University of Bern, Switzerland, as described in Morandi et al. [32].

Taxonomic annotation

DNA sequences were generated containing Unique Molecular Identifiers (UMIs; [46]) for OSM samples and not containing UMIs for GM samples. UMIs are specifically applied to mitigate PCR amplification bias and improve quantification accuracy in low-biomass microbiome samples. These UMIs were initially extracted using UMI-tools (v.1.1.4.). Following extraction, read quality filtering was performed with fastp (v.0.20.1; [47]). To remove host-derived sequences, the filtered reads were aligned to the human genome reference GRCh38. Reads that did not map to the host genome were isolated using SAMtools (v.1.10). The non-host reads were subsequently aligned to the ChocoPhlAn reference database (version mpa_vOct22_CHOCOPhlAnSGB_202212) using bowtie2 (v. 2.3.4.1; [48]). The resulting SAM alignment files were converted into BAM format, sorted, and indexed using SAMtools (v. 1.10; [49]). These sorted BAM files of OSM samples were then deduplicated based on both UMI and alignment position using UMI-tools (v. 1.1.4). The GM BAM files were not deduplicated. The deduplicated BAM file of the OSM and the sorted BAM file of the GM served as input for the MetaPhlAn4 pipeline (v. 4.0.4; [50, 51]) for taxonomic profiling.

Statistical analysis

The bioinformatic pipeline used for this project is summarised in Fig. 1. All statistical analyses were performed using R Statistical Software (v. 4.2.2). All analyses were done on the phylum and genus level. To visualise shared microbial taxa between the two sampled niches, a Venn diagram was generated using the ggplot2 package (v. 3.5.0; Fig. 2A-B). The distribution of shared taxa between the two niches and across participants was further illustrated using an UpSet plot (Fig. 2C-D). In addition, a barplot was created to quantify the number of shared taxa for each individual (Fig. 3A). To assess within sample microbial diversity, the Shannon diversity index was calculated using the vegan R package (v. 2.6.4) and visualised as boxplots for the two niches with ggplot2 (Fig. 3B).

Fig. 1.

Fig. 1

Open in a new tab

Overview of the pipeline applied for descriptive, correlation and type analysis between the ocular surface microbiome (OSM) and the gut microbiome (GM)

Fig. 2.

Fig. 2

Open in a new tab

Taxonomic overlap of the ocular surface and gut microbiomes. Number of shared and unique phyla (A) and genera (B) between the ocular surface microbiome (OSM) and the gut microbiome (GM). UpSet plots illustrating the distribution of phyla (C) and most prevalent genera (D) per individual of shared taxa. A total of 5 phyla and 43 genera were found to be in common in both the ocular surface and gut microbiomes. Set size indicating the prevalence of the taxa per sample and intersection size describing the co-occurrence of taxa and their prevalence

Fig. 3.

Fig. 3

Open in a new tab

Shared taxa and microbial diversity between the ocular surface and gut microbiomes. (A) Counts of the number of shared microbial taxa per individual are shown at the phylum (orange) and genus level (blue; n = 102). (B) The Shannon diversity showed significant differences between the gut microbiome (GM; red) and the ocular surface microbiome (OSM; turquoise) at both, phylum (p-value < 0.001) and genus level (p-value < 0.00001, paired t-test, n = 102)

To explore overall community structure a non-metric multidimensional scaling (NMDS) analysis was performed based on the Bray-Curtis dissimilarity. The Bray-Curtis dissimilarity was calculated on all the microbial taxa found in any of the two niches, using the vegan R package (v. 2.6). The analysis yields a two-dimensional solution with a stress value, which were visualised using ggplot2 (Fig. 4). To assess the similarity of microbial community structures within niches, Bray–Curtis dissimilarity matrices were calculated separately for each niche (Fig. 5A-B). A Mantel test was then applied to correlate the two distance matrices, yielding a correlation coefficient r and associated p-value to evaluate the strength and significance of the relationship.

Fig. 4.

Fig. 4

Open in a new tab

Clustering of microbial genera of the ocular surface microbiome and the gut microbiome. Non-metric multidimensional scaling (NMDS) of genera in the ocular surface microbiome (OSM; turquoise) and the gut microbiome (GM; red) samples based on Bray-Curtis dissimilarity. The two distinct clusters correspond to the two niches, highlighting their compositional differences with a stress value of 0.09, indicating a good representation of the data in two dimensions (n = 102)

Fig. 5.

Fig. 5

Open in a new tab

Dissimilarity matrices of the ocular surface microbiome and the gut microbiome. Bray-Curtis dissimilarity matrices illustrate pairwise compositional differences between samples within the gut microbiome (GM; n = 51; (A)) and the ocular surface microbiome (OMS; n = 51; (B)) at genus level. The matrices show that OSM samples can be very similar, while GM samples display high inter-individual variability with fewer similarities across individuals. There was no correlation between the microbial community structures of the two niches, correlation coefficient r = − 0.12, p-value = 0.97 (Mantel test)

To identify distinct community structures, the Jensen-Shannon distance was calculated per niche, followed by Partitioning Around Medoids (PAM) clustering. To determine the optimal number of clusters, the Calinski–Harabasz (CH) index was calculated for values of k ranging from 3 to 10 using the clusterCrit package (v. 1.3.0). The number of clusters corresponding to the highest CH index was selected as the optimal solution. Final clustering was performed using the selected k for the PAM clustering, and cluster robustness was evaluated using silhouette width. The clusters were visualised by performing a Principal Component Analysis (PCA) using the ade4 package (v. 1.7–23) and subsequently applying BCA based on the clustering assignments (Fig. 6) [52].

Fig. 6.

Fig. 6

Open in a new tab

Entero- and oculartypes. Clustering of the gut microbiome (GM; n = 51) and ocular surface microbiome (OSM; n = 51) samples based on Jensen-Shannon divergence using Partitioning Around Medoids (PAM). Calinski-Harabasz index plots for the GM (A1) and the OSM (A2), respectively, indicate k = 3 as the optimal number of clusters. PAM clustering for the GM (B1) and OSM (B2) samples. Driving genera for the enterotypes were Alistipes (cluster 1, C1), Prevotella (cluster 2, C2), and Bacteroides (cluster 3, C3). Driving genera for the oculartypes were Cutibacterium (cluster D1), Limosilactobacillus (cluster 2, D2), and Corynebacterium (cluster 3, D3)

Results

Read statistics

Sequencing libraries from negative controls did not pass quality and concentration requirements, suggesting that contaminating DNA, if present, was below detection limit. All samples from both microbiomes passed standard quality control checks using FastQC.

In total, 6.38 billion 150 bp paired-end reads were generated from the ocular surface samples with a median read depth of 132.8 million per sample. Following adapter trimming and host read removal, approximately 653 million high-quality (10.24%), non-human reads remained for the ocular surface samples, averaging 12.79 ± 7.92 (s.d.) million reads per sample. As reported in previous studies as well as expected due to low microbial biomass, the majority of reads originated from the host [42]. Subsequent deduplication using (UMIs) resulted in a total of 1.23 million unique reads available for downstream analysis, with a mean of 24’000 ± 34’000 (s.d.) reads per sample.

A total of 3.69 billion paired-end reads were generated from the gut samples with a median read depth of 66.1 million per sample. After adapter trimming and host removal, 3.56 billion high-quality (96.48%), non-human reads remained for the gut samples, with an average of 67.23 ± 24.86 (s.d.) million reads per sample, for downstream analysis.

Common taxa

A total of 19 phyla and 939 genera were identified across all samples. Among them, 5 phyla and 43 genera were shared between the two niches, 4 phyla and 54 genera were unique to the ocular surface and 10 phyla and 842 genera to the gut (Fig. 2A). The UpSet plot of the phyla across both niches (Fig. 2C) showed that Actinobacteria was the most prevalent phylum, detected in 49 out of 51 individuals. Firmicutes and Proteobacteria were present in 40 and 39 individuals, respectively. The UpSet plot reveals that the combination of the three most prevalent phyla was found in 25 individuals, almost half of the cohort.

The UpSet plot with the most prevalent genera (Fig. 2D), showed that Escherichia was present in 12 and Pseudomonas in 11 out of 51 individuals. Furthermore, the combination of the two was found in 6 individuals. The results showed that there was no single shared taxa found in all individuals.

A bar plot illustrating the number of taxa found in both niches per individual (Fig. 3A) showed that on phylum level, most participants had three taxa in common between the two niches, while only one shared none. Notably, no individual had all five shared phyla, showing a limited overlap between the two niches. A strong shift in shared taxa was seen on genus level, with 31 individuals having no common genera between the two niches. Only 3 individuals had 3 common genera and only one had 4 common genera.

To assess differences in microbial diversity between the two niches, the Shannon diversity index was calculated for each sample. Boxplots (Fig. 3B) showed that the GM exhibited a higher median Shannon index compared to the OSM on both, phylum and genus level. A paired t-test confirmed this difference to be statistically significant for both, phylum (p < 0.001) and genus (p < 0.00001).

Non-metric multidimensional scaling

The NMDS of the microbial communities from both niches revealed distinct clustering patterns for the two niches (Fig. 4). Samples from the gut clustered more homogenously than samples from the ocular surface. The analysis yielded a stress value of 0.09.

Bray-curtis dissimilarity matrix and mantel test

The Bray–Curtis dissimilarity matrix generated for the gut samples showed a homogenous distribution of values (Fig. 5A). The ocular surface matrix revealed a broad range of dissimilarity values, indicating that while some samples were highly similar, others were very different (Fig. 5B). The high dissimilarity in most of the samples reflect a high inter-individual variability. The ocular surface matrix displayed a rather scattered pattern with limited clustering. When comparing the dissimilarity patterns between the two niches, no consistent overlap in clustering was observed. The Mantel test resulted in a correlation coefficient r of − 0.12 and a p-value of 0.97.

Ocular- and enterotypes

PAM clustering based on the Jensen–Shannon distance revealed that the optimal number of clusters was k = 3 for both OSM and GM samples (Fig. 6A1, A2), as determined by the Calinski–Harabasz index. The BCA plots illustrated separated clusters in the gut dataset (enterotypes), with cluster 1 comprising 31 individuals, cluster 2 seven individuals, and cluster 3 thirteen individuals (Fig. 6B1). In contrast, the ocular surface clusters appeared less distinct, with cluster 1 containing thirteen individuals, cluster 2 containing 32 and cluster 3 six individuals (Fig. 6B2). Silhouette scores were calculated for each point and averaged per cluster. Scores for the gut clusters were 0.23 for cluster 1, 0.44 for cluster 2 and 0.11 for cluster 3. Slightly higher scores were obtained for the ocular surface clusters with 0.54 for cluster 1, 0.26 for cluster 2 and 0.49 for cluster 3.

Taxonomic profiling revealed dominant genus within each cluster: Alistipes in gut cluster 1, Prevotella in cluster 2 and Bacteroides in cluster 3 (Fig. 6C1-3). In the ocular surface, cluster 1 showed a higher abundance of Cutibacterium, cluster 2 Limosilactobacillus and cluster 3 Corynebacterium (Fig. 6D1-3), which we refer to here as oculartypes, by analogy to gut enterotypes. However, no consistent pattern of sample clustering between the two niches was observed, indicating that clustering of individuals in one niche did not correspond to clustering in the other.

Discussion

This study highlights a big difference in microbial abundance between the OSM and GM. As expected, the gut exhibited a far greater microbial richness and diversity [10], while only a limited number of taxa were found to be shared between the two niches. This discrepancy became even more pronounced at genus level. This was further emphasised by individual level comparisons with bar plots (Fig. 3A), which revealed that over half of the participants had no genera in common between their gut and ocular samples. The UpSet plot (Fig. 2C) indicated that the phylum Actinobacteria was present in nearly all individuals, which is in accordance with previous studies, which have found Actinobacteria to be pivotal in the maintenance of gut homeostasis [6, 53, 54]. However, this apparent overlap largely diminished if assessed at higher taxonomic resolution. At genus level, the most prevalent shared genus Escherichia was found in only a quarter of the participants (Fig. 2D), underscoring the limited microbial overlap across these anatomically and functionally distinct niches.

The limited overlap between the ocular surface and gut microbiomes can be further understood in the context of ocular surface biology. The tear film contains antimicrobial peptides, lysozyme, and immunoglobulins that actively suppress microbial colonisation. Physical factors, such as blinking, tear flow, and the small surface area of the eye, further limit bacterial colonisation and persistence [5]. In addition, the ocular surface represents a largely aerobic environment, whereas the gut is predominantly anaerobic, which constrains the types of microbes that can survive in each niche and further contributes to their limited overlap. Together, these biological features create a selective and constrained environment, which likely contributes to the distinct composition and low diversity of the OSM relative to the gut.

The comparison of alpha diversity between the OSM and GM revealed significant differences at both, phylum and genus levels, with more pronounced differences at genus level (Fig. 3B). This highlights not only the distinct microbial ecosystems that characterise each niche but also the considerable variability in microbial richness and evenness. Given the limited number of taxa shared between the two niches (Fig. 2A), especially at higher taxonomic resolution, a direct comparison or attempt to correlate the two microbiomes is challenging. Previous studies have tried to define and determine a stable microbial core, but neither for the OSM nor the GM has one been consistently identified [55]. The lack thereof further complicates efforts to explore inter niche microbial relationships.

The NMDS plot (Fig. 4) revealed a clustering of samples according to their niche, emphasising the substantial differences in microbial community composition between the two niches. With a stress value of 0.09, the NMDS provided a reliable two-dimensional representation of the high-dimensional data, reinforcing the assumption that the two microbial environments have minimal overlap.

The Bray-Curtis dissimilarity matrices provided further insight into intra niche variability (Fig. 5). For the ocular surface, we observed a broad range of dissimilarity values, suggesting that while some samples are relatively similar, others are highly divergent. A few samples showed slight clustering, indicating high similarity within these samples. In contrast, the gut samples exhibited consistently high dissimilarity values, highlighting the pronounced inter-individual variability of the GM [6]. No apparent clustering was observed among gut samples. Given the absence of clustering in the gut, it was not feasible to directly compare cluster structures between the two niches. This was further assessed by the Mantel test, which yielded a correlation coefficient (r) of -0.12 and a p-value of 0.9, indicating no detectable correlation between the dissimilarity matrices of the OSM and GM with the current methodology. Previous studies on the GM have identified distinct clusters of individuals characterised by driving genera, commonly referred to as enterotypes [52]. To the best of our knowledge, this classification has not yet been applied to the OSM so far. In this study, we conducted a clustering analysis separately for each niche (Fig. 6), as the minimal overlap between their microbial communities made a combined approach unfeasible. Using the Calinski-Harabasz index, we determined that three clusters represented the optimal grouping for both the OSM and GM. The driving genera of the gut clusters—Alistipes, Prevotella and Bacteroides—are consistent with previously described enterotypes. In contrast, for the OSM, this is the first indication that three distinct oculartypes may exist, characterised by Cutibacterium, Limosilactobacillus and Corynebacterium. While the silhouette scores were moderate, they support the presence of meaningful clustering in both niches. It should be noted that in high-dimensional microbiome data, silhouette scores are generally lower due to the inherent sparsity and complexity of community composition. No association in metadata such as sex or age was found between the cluster distributions of matched samples across the two niches, further supporting their independent microbial profiles. However, the entero- and oculartyping should be considered exploratory, as they are sensitive to sequencing depth, distance metrics, and clustering algorithms.

Given the low microbial biomass of the ocular surface, the absence of negative controls represents a methodological limitation of this study. In addition, different library preparation strategies were applied to the two niches, including the use of Unique Molecular Identifiers (UMIs) for ocular surface samples but not for gut samples. This reflects the substantially lower biomass and higher susceptibility to amplification bias in the ocular surface microbiome. However, as our primary objective was to investigate relative community patterns and distance relationships rather than absolute taxonomic composition, the consistent niche-specific separation observed across multiple analytical approaches suggests that the main findings are unlikely to be driven by technical artefacts. Taken together, these methodological considerations should be taken into account when interpreting the results but do not invalidate the central conclusion of niche-specific and largely independent microbial profiles.

Conclusions

This study presents a comprehensive comparison of the OSM and GM within the same individuals, offering a novel view into cross-niche microbial relationships. Our analyses showed that the two niches are compositionally distinct, with minimal taxonomic overlap and no detectable correlation in microbial structure. While three potential microbial clusters were identified in each niche, their distributions were independent, underscoring the uniqueness of each microbial environment. The low microbial biomass of the ocular surface poses significant technical challenges for sequencing, potentially limiting the detection of rare taxa and complicating direct comparisons with the highly abundant and diverse GM. These methodological constraints may partly account for the missing correlation. Overall, our findings emphasise that host microbiome interactions are likely niche specific and highlight the need for optimised approaches tailored to low biomass environments like the ocular surface.

Acknowledgements

Not applicable.

Authors’ contributions

Conceptualisation, D.C.Z.-B. and M.S.Z; methodology, N.E., M.K. and D.C.Z.-B.; formal analysis N.E and M.K.; investigation, L.S. and G.H.; writing-original draft preparation, N.E.; writing-review and editing, D.C.Z.-B.; visualisation, N.E.; supervision, D.C.Z.-B; project administration, D.C.Z.-B.; funding acquisition, D.C.Z.-B. and M.S.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Foundation Bertarelli Catalyst Fund, EPFL (Ecole Polytechnique Fédérale de Lausanne), Lausanne, Switzerland (CF10000044-EPFL SCR0237812).

Data availability

The Ocular and Gut Microbiome Sequencing Dataset is available in the European Nucleotide Archive. The dataset can be accessed using the following DOI or accession IDs: https://identifiers.org/ena.embl/PRJEB74248, https://identifiers.org/ena.embl/PRJEB74773, https://identifiers.org/ena.embl/PRJEB74775, https://identifiers.org/ena.embl/PRJEB74776. Original R scripts are available in GitHub (https://github.com/ninaeld/OSM to GM correlation).

Declarations

Ethics approval and consent to participate

The studies were approved by the Ethics Committee of the Canton of Bern and can be accessed using the following links: https://clinicaltrials.gov/study/NCT02438111?term=NCT02438111&rank=1 and https://clinicaltrials.gov/study/NCT04658238?term=NCT04658238&rank=1. Study procedures adhered to the CONSORT principles as well as to the principles outlined in the Declaration of Helsinki and the International Ethical Guidelines for Biomedical Research Involving Human Subjects, as established by the Council for International Organizations of Medical Sciences. Consent to participate: Informed consent was obtained from all subjects involved in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s Note

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

References

  • 1.Delbeke H, Younas S, Casteels I, Joossens M. Current knowledge on the human eye microbiome: a systematic review of available amplicon and metagenomic sequencing data. Acta Ophthalmol (Copenh). 2021;99(1):16–25. [DOI] [PubMed] [Google Scholar]
  • 2.Kalyana Chakravarthy S, Jayasudha R, Ranjith K, Dutta A, Pinna NK, Mande SS, et al. Alterations in the gut bacterial microbiome in fungal Keratitis patients. PLoS ONE. 2018;13(6):e0199640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Petrillo F, Pignataro D, Lavano MA, Santella B, Folliero V, Zannella C, et al. Current Evidence on the Ocular Surface Microbiota and Related Diseases. Microorganisms. 2020;8(7):1033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Huang Y, Yang B, Li W. Defining the normal core microbiome of conjunctival microbial communities. Clin Microbiol Infect Off Publ Eur Soc Clin Microbiol Infect Dis. 2016;22(7):e6437–64312. [DOI] [PubMed] [Google Scholar]
  • 5.Peter VG, Morandi SC, Herzog EL, Zinkernagel MS, Zysset-Burri DC. Investigating the Ocular Surface Microbiome: What Can It Tell Us? Clin Ophthalmol Auckl NZ. 2023;17:259–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Zysset-Burri DC, Schlegel I, Lincke JB, Jaggi D, Keller I, Heller M, et al. Understanding the Interactions Between the Ocular Surface Microbiome and the Tear Proteome. Invest Ophthalmol Vis Sci. 2021;62(10):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Keilty RA. The Bacterial Flora of the Normal Conjunctiva with Comparative Nasal Culture Study. Am J Ophthalmol. 1930;13(10):876–9. [Google Scholar]
  • 8.Perkins RE, Kundsin RB, Pratt MV, Abrahamsen I, Leibowitz HM. Bacteriology of normal and infected conjunctiva. J Clin Microbiol. 1975;1(2):147–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.McNatt J, Allen SD, Wilson LA, Dowell VR. Anaerobic flora of the normal human conjunctival sac. Arch Ophthalmol Chic Ill 1960. 1978;96(8):1448–50. [DOI] [PubMed] [Google Scholar]
  • 10.Doan T, Akileswaran L, Andersen D, Johnson B, Ko N, Shrestha A, et al. Paucibacterial Microbiome and Resident DNA Virome of the Healthy Conjunctiva. Invest Ophthalmol Vis Sci. 2016;57(13):5116–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Turnbaugh PJ, Ley RE, Hamady M, Fraser-Liggett CM, Knight R, Gordon JI. The human microbiome project. Nature. 2007;449(7164):804–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Schlegel I, De Goüyon M, Lincke JB, Keller I, Zinkernagel MS, Zysset-Burri DC. The Human Ocular Surface Microbiome and Its Associations with the Tear Proteome in Dry Eye Disease. Int J Mol Sci. 2023;24(18):14091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.St Leger AJ, Desai JV, Drummond RA, Kugadas A, Almaghrabi F, Silver P, et al. An Ocular Commensal Protects against Corneal Infection by Driving an Interleukin-17 Response from Mucosal γδ T Cells. Immunity. 2017;47(1):148–e1585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kugadas A, Gadjeva M. Impact of Microbiome on Ocular Health. Ocul Surf. 2016;14(3):342–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Miller D, Iovieno A. The role of microbial flora on the ocular surface. Curr Opin Allergy Clin Immunol. 2009;9(5):466–70. [DOI] [PubMed] [Google Scholar]
  • 16.Heintz-Buschart A, Wilmes P. Human Gut Microbiome: Function Matters. Trends Microbiol. 2018;26(7):563–74. [DOI] [PubMed] [Google Scholar]
  • 17.Aragona P, Baudouin C, Benitez Del Castillo JM, Messmer E, Barabino S, Merayo-Lloves J, et al. The ocular microbiome and microbiota and their effects on ocular surface pathophysiology and disorders. Surv Ophthalmol. 2021;66(6):907–25. [DOI] [PubMed] [Google Scholar]
  • 18.Hotta F, Eguchi H, Kuwahara T, Nakayama-Imaohji H, Shimomura Y, Kusaka S. Disturbances in the ocular surface microbiome by perioperative antimicrobial eye drops. Front Cell Infect Microbiol. 2023;13:1172345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Zhong Y, Fang X, Wang X, Lin YA, Wu H, Li C. Effects of Sodium Hyaluronate Eye Drops With or Without Preservatives on Ocular Surface Bacterial Microbiota. Front Med. 2022;9:793565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hall AB, Tolonen AC, Xavier RJ. Human genetic variation and the gut microbiome in disease. Nat Rev Genet. 2017;18(11):690–9. [DOI] [PubMed] [Google Scholar]
  • 21.Satokari R, Fuentes S, Mattila E, Jalanka J, de Vos WM, Arkkila P. Fecal transplantation treatment of antibiotic-induced, noninfectious colitis and long-term microbiota follow-up. Case Rep Med. 2014;2014:913867. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Morgan XC, Segata N, Huttenhower C. Biodiversity and functional genomics in the human microbiome. Trends Genet TIG. 2013;29(1):51–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Joyce SA, Gahan CGM. The gut microbiota and the metabolic health of the host. Curr Opin Gastroenterol. 2014;30(2):120–7. [DOI] [PubMed] [Google Scholar]
  • 24.Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, et al. Diversity of the human intestinal microbial flora. Science. 2005;308(5728):1635–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Pheiffer C, Riedel S, Dias S, Adam S. Gestational Diabetes and the Gut Microbiota: Fibre and Polyphenol Supplementation as a Therapeutic Strategy. Microorganisms. 2024;12(4):633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kirschner R, Hsu T, Tuan NN, Chen CL, Huang SL. Characterization of fungal and bacterial components in gut/fecal microbiome. Curr Drug Metab. 2015;16(4):272–83. [DOI] [PubMed] [Google Scholar]
  • 27.Abraham P, Pratap N. Dysbiosis in Irritable Bowel Syndrome. J Assoc Physicians India. 2023;71(9):75–81. [DOI] [PubMed] [Google Scholar]
  • 28.Li J, Wei J, Wang J, Xu T, Wu B, Yang S, et al. Association between gut microbiota and spinal stenosis: a two-sample mendelian randomization study. Front Immunol. 2024;15:1360132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A, Ley RE, et al. A core gut microbiome in obese and lean twins. Nature. 2009;457(7228):480–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Karlsson FH, Fåk F, Nookaew I, Tremaroli V, Fagerberg B, Petranovic D, et al. Symptomatic atherosclerosis is associated with an altered gut metagenome. Nat Commun. 2012;3:1245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Horai R, Zárate-Bladés CR, Dillenburg-Pilla P, Chen J, Kielczewski JL, Silver PB, et al. Microbiota-Dependent Activation of an Autoreactive T Cell Receptor Provokes Autoimmunity in an Immunologically Privileged Site. Immunity. 2015;43(2):343–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Morandi SC, Herzog EL, Munk M, Kreuzer M, Largiadèr CR, Wolf S, et al. The gut microbiome and HLA-B27-associated anterior uveitis: a case-control study. J Neuroinflammation. 2024;21(1):120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zysset-Burri DC, Morandi S, Herzog EL, Berger LE, Zinkernagel MS. The role of the gut microbiome in eye diseases. Prog Retin Eye Res. 2023;92:101117. [DOI] [PubMed] [Google Scholar]
  • 34.Lincke JB, Christe L, Unterlauft JD, Zinkernagel MS, Zysset-Burri DC. Microbiome and Retinal Vascular Diseases. Am J Pathol. 2023;193(11):1675–82. [DOI] [PubMed] [Google Scholar]
  • 35.Zinkernagel MS, Zysset-Burri DC, Keller I, Berger LE, Leichtle AB, Largiadèr CR, et al. Association of the Intestinal Microbiome with the Development of Neovascular Age-Related Macular Degeneration. Sci Rep. 2017;7:40826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Zysset-Burri DC, Keller I, Berger LE, Neyer PJ, Steuer C, Wolf S, et al. Retinal artery occlusion is associated with compositional and functional shifts in the gut microbiome and altered trimethylamine-N-oxide levels. Sci Rep. 2019;9(1):15303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Huang L, Hong Y, Fu X, Tan H, Chen Y, Wang Y, et al. The role of the microbiota in glaucoma. Mol Aspects Med. 2023;94:101221. [DOI] [PubMed] [Google Scholar]
  • 38.Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Microbial ecology: human gut microbes associated with obesity. Nature. 2006;444(7122):1022–3. [DOI] [PubMed] [Google Scholar]
  • 39.Lichtenegger AS, Posadas-Cantera S, Badr MT, Häcker G. Comparison of the diversity of anaerobic-cultured gut bacterial communities on different culture media using 16S rDNA sequencing. J Microbiol Methods. 2024;224:106988. [DOI] [PubMed] [Google Scholar]
  • 40.Wu J, Singleton SS, Bhuiyan U, Krammer L, Mazumder R. Multi-omics approaches to studying gastrointestinal microbiome in the context of precision medicine and machine learning. Front Mol Biosci. 2023;10:1337373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Riedl RA, Burnett CML, Pearson NA, Reho JJ, Mokadem M, Edwards RA, et al. Gut Microbiota Represent a Major Thermogenic Biomass. Funct Oxf Engl. 2021;2(3):zqab019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Herzog EL, Kreuzer M, Zinkernagel MS, Zysset-Burri DC. Challenges and insights in the exploration of the low abundance human ocular surface microbiome. Front Cell Infect Microbiol. 2023;13:1232147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Delbeke H, Casteels I, Joossens M. DNA extraction protocol impacts ocular surface microbiome profile. Front Microbiol. 2023;14:1128917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Chowdhary A, Van Gelder RN, Sundararajan M. Methodologic Considerations for Studying the Ocular Surface Microbiome. Ophthalmol Sci. 2023;3(4):100408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Spörri L, Uldry AC, Kreuzer M, Herzog EL, Zinkernagel MS, Unterlauft JD, et al. Exploring the Ocular Surface Microbiome and Tear Proteome in Glaucoma. Int J Mol Sci. 2024;25(11):6257. 10.1038/s41525-026-00550-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Smith T, Heger A, Sudbery I. UMI-tools: modeling sequencing errors in Unique Molecular Identifiers to improve quantification accuracy. Genome Res. 2017;27(3):491–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9(4):357–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Blanco-Míguez A, Beghini F, Cumbo F, McIver LJ, Thompson KN, Zolfo M, et al. Extending and improving metagenomic taxonomic profiling with uncharacterized species using MetaPhlAn 4. Nat Biotechnol. 2023;41(11):1633–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Truong DT, Tett A, Pasolli E, Huttenhower C, Segata N. Microbial strain-level population structure and genetic diversity from metagenomes. Genome Res. 2017;27(4):626–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Arumugam M, Raes J, Pelletier E, Le Paslier D, Yamada T, Mende DR, et al. Enterotypes of the human gut microbiome. Nature. 2011;473(7346):174–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Dong Q, Brulc JM, Iovieno A, Bates B, Garoutte A, Miller D, et al. Diversity of bacteria at healthy human conjunctiva. Invest Ophthalmol Vis Sci. 2011;52(8):5408–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Binda C, Lopetuso LR, Rizzatti G, Gibiino G, Cennamo V, Gasbarrini A. Actinobacteria: A relevant minority for the maintenance of gut homeostasis. Dig Liver Dis Off J Ital Soc Gastroenterol Ital Assoc Study Liver. 2018;50(5):421–8. [DOI] [PubMed] [Google Scholar]
  • 55.La-ongkham O, Nakphaichit M, Nakayama J, Keawsompong S, Nitisinprasert S. Age-related changes in the gut microbiota and the core gut microbiome of healthy Thai humans. 3 Biotech. 2020;10(6):276. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The Ocular and Gut Microbiome Sequencing Dataset is available in the European Nucleotide Archive. The dataset can be accessed using the following DOI or accession IDs: https://identifiers.org/ena.embl/PRJEB74248, https://identifiers.org/ena.embl/PRJEB74773, https://identifiers.org/ena.embl/PRJEB74775, https://identifiers.org/ena.embl/PRJEB74776. Original R scripts are available in GitHub (https://github.com/ninaeld/OSM to GM correlation).

Articles from BMC Microbiology are provided here courtesy of BMC

RESOURCES