**Background:** Diabetic retinopathy (DR) is a leading cause of visual impairment, and recent research suggests a role for the gut microbiome via the 'gut-retinal axis.' However, direct evidence of microbial involvement in the eye is lacking. This pilot study aimed to compare the microbiome of the aqueous humour and gut in people with diabetes mellitus (DM) with and without DR, and healthy controls.
**Methods:** This prospective controlled study enrolled 17 individuals undergoing intraocular surgery at a tertiary eye care centre in India. Exclusion criteria included prior ocular surgery, infection, inflammation, systemic diseases other than DM and hypertension, and recent antibiotic use. Participants were divided into three groups: DR (n=7), DM without DR (n=4), and healthy controls (n=6). Aqueous humour samples were obtained during surgery (adequate volume in 12 participants: 4 per group). Stool samples were collected within 7 days of surgery. Dietary habits were assessed using a validated questionnaire. DNA was extracted from both sample types, and the V3-V4 region of the 16S rRNA gene was sequenced using Illumina NovaSeq 6000. Taxonomic classification was performed using QIIME and SILVA database. Alpha diversity (Shannon, Simpson, observed OTUs, Chao1) and differential abundance of genera were analysed using R software. Interaction networks were generated using CoNet in Cytoscape.
**Key Results:** The mean age of the 12 participants with paired samples was 61.16 ± 11.95 years. Mean DM duration was 9.5 ± 4.8 years (14.3 years in DR group, 4.8 years in DM group). From aqueous humour, 937 OTUs were identified. Alpha diversity analysis showed that observed OTUs and Chao1 indices were significantly different (p<0.05) between DM and DR cohorts compared to controls, but Shannon and Simpson indices were not significantly different. At the phylum level, Firmicutes and Bacteroidetes were most abundant; only Tenericutes showed significant difference between DM and DR groups (p=0.03). At the genus level, 122, 145, and 93 genera were identified in controls, DM, and DR groups, respectively. Compared to controls, DM group had one genus decreased (Shuttleworthia) and five increased (Succinivibrio, Phascolarctobacterium, Ornithinimicrobium, Turicibacter, Clostridioides). DR group had three genera decreased (Roseburia, Prevotellaceae group, Haemophilus) compared to controls. Between DM and DR groups, 24 genera decreased and 2 increased in DR. Notably, many decreased genera in DR were anti-inflammatory (e.g., Ruminococcaceae group, Dialister, Lachnospiraceae group, Faecalibacterium, Eubacterium group, Butyrivibrio, Ruminococcus group, Prevotellaceae group, Roseburia, Succinivibrio, Subdoligranulum, Turicibacter). Heat map analysis showed distinct clustering of the three groups based on discriminatory genera. Interaction networks revealed 18 hub taxa unique to controls, 31 unique to DM, and 11 unique to DR. Comparison of aqueous humour and gut microbiomes showed some common genera (e.g., Dialister, Megasphaera, Prevotella group, Ruminococcaceae group) but also unique genera: Lactobacillus, Bacillus, Bacteroides, Clostridium sensu stricto 12 were present in aqueous humour but absent in gut; Bacteroides group and Eubacterium group were present in gut but absent in aqueous humour.
**Clinical Implications:** This pilot study provides preliminary evidence that an intraocular microbiome exists in both health and disease, and that its composition differs between individuals with DR, DM without DR, and healthy controls. The consistent reduction in anti-inflammatory bacteria in the aqueous humour of DR patients suggests a potential role for microbial dysbiosis in DR pathogenesis. The intraocular microbiome differences were more pronounced than gut microbiome differences, indicating that the eye's microbial environment may be a more specific and less volatile target for therapeutic interventions. However, due to the small sample size and age differences between groups, these findings require validation in larger, well-controlled studies. If confirmed, targeting the intraocular microbiome could offer a novel approach for preventing or treating DR.