Alteration of intestinal microbiota is associated with diabetic retinopathy and its severity: Samples collected from...

Current approaches for the therapy of diabetic retinopathy (DR), which was one of leading causes of visual impairment, have their limitations. Animal experiments revealed that restructuring of intestinal microbiota can prevent retinopathy.

To explore the relationship between intestinal microbiota and DR among patients in the southeast coast of China, and provide clues for novel ways to prevention and treatment methods of DR.

The fecal samples of non-diabetics (Group C, n = 15) and diabetics (Group DM, n = 30), including 15 samples with DR (Group DR) and 15 samples without DR (Group D), were analyzed by 16S rRNA sequencing. Intestinal microbiota compositions were compared between Group C and Group DM, Group DR and Group D, as well as patients with proliferative diabetic retinopathy (PDR) (Group PDR, n = 8) and patients without PDR (Group NPDR, n = 7). Spearman correlation analyses were performed to explore the associations between intestinal microbiota and clinical indicators.

The alpha and beta diversity did not differ significantly between Group DR and Group D as well as Group PDR and Group NPDR. At the family level, Fusobacteriaceae, Desulfovibrionaceae and Pseudomonadaceae were significantly increased in Group DR than in Group D (P < 0.05, respectively). At the genera level, Fusobacterium, Pseudomonas, and Adlercreutzia were increased in Group DR than Group D while Senegalimassilia was decreased (P < 0.05, respectively). Pseudomonas was negatively correlated with NK cell count (r = -0.39, P = 0.03). Further, the abundance of genera Eubacterium (P < 0.01), Peptococcus, Desulfovibrio, Acetanaerobacterium and Negativibacillus (P < 0.05, respectively) were higher in Group PDR compared to Group NPDR, while Pseudomonas, Alloprevotella and Tyzzerella (P < 0.05, respectively) were lower. Acetanaerobacterium and Desulfovibrio were positively correlated with fasting insulin (r = 0.53 and 0.61, respectively, P < 0.05), when Negativibacillus was negatively correlated with B cell count (r = -0.67, P < 0.01).

Our findings indicated that the alteration of gut microbiota was associated with DR and its severity among patients in the southeast coast of China, probably by multiple mechanisms such as producing short-chain fatty acids, influencing permeability of blood vessels, affecting levels of vascular cell adhesion molecule-1, hypoxia-inducible factor-1, B cell and insulin. Modulating gut microbiota composition might be a novel strategy for prevention of DR, particularly PDR in population above.