Long Noncoding RNAs as Blood-Based Biomarkers of Diabetic Retinopathy.

PURPOSE: Prevalence of diabetic retinopathy closely depends on the duration of diabetes and severity of hyperglycemia. Experimental models have shown that high glucose initiates many metabolic, molecular and epigenetic changes in the retina before vascular histopathology is detectable. Several LncRNAs (RNAs with >200bp and no reading frame) are also aberrantly expressed in the retina in diabetes. These RNAs have high organ and cell specificity and show minimal gradual instability in plasma. Our goal was to examine the utility of aberrantly expressed retinal LncRNAs (nuclear DNA-encoded NEAT1, HOTAIR, HOTTIP, MALAT1, H19 and Meg3 and mitochondrial DNA-encoded CytB) in diabetes as possible biomarkers of diabetic retinopathy.

METHODS

Plasma LncRNAs were analyzed by qRT-PCR in diabetic patients with proliferative (PDR), or no retinopathy (No-DR), and in rats with streptozotocin-induced diabetes for two to six months. The results were confirmed in the retina from donors with diabetic retinopathy and from diabetic rats.

RESULTS

Plasma NEAT1, HOTAIR, MALAT1 and HOTTIP were upregulated and Meg3 and CytB were downregulated in No-DR and PDR groups and retina from human donors with diabetic retinopathy. Although these LncRNAs were also aberrantly expressed in plasma and retina within four to six months of diabetes in rats, NEAT1 and CytB had abnormal expression within two months of diabetes (without any retinal vascular histopathology).

CONCLUSIONS

NEAT1 and CytB have consistent expression patterns in plasma from diabetic patients and plasma and retina from rodents.

TRANSLATIONAL RELEVANCE: Plasma NEAT1 and CytB could serve as possible early noninvasive biomarkers of retinopathy in diabetic patients.