Mapping the daily rhythmic transcriptome in the diabetic retina

Retinal function changes dramatically from day to night. Yet, clinical diagnosis, treatments, and experimental sampling occur during the day, leaving a significant gap in our understanding of the pathobiology occurring at night. While there is evidence that diabetes disrupts the circadian system that optimizes our physiology to the environmental light/dark cycle, the impact of such disruption is not well understood. This study investigates whether diabetes affects the retina's daily rhythm of gene expression to understand the pathobiology of diabetic retinopathy, a common complication of diabetes. Control and hyperglycemic littermate mice (Ins2Akita/J) were kept under a standard 12h:12h light/dark cycle until four months of age. Bulk mRNA sequencing was conducted in retinas collected every 4 hours throughout the 24 hr light/dark cycle. Computational approaches were used to detect rhythmicity, predict acrophase, identify differential rhythmic patterns, analyze phase set enrichment, and predict upstream regulators. The retinal transcriptome exhibited a tightly regulated rhythmic expression with a clear 12-hr axis of transcriptional rush, peaking at midday and midnight. The functions of day-peaking genes were enriched for DNA repair, RNA splicing, and ribosomal protein synthesis, whereas night-peaking genes were enriched for metabolic processes and growth factor signaling. Although the 12-hr transcriptional axis is retained in the diabetic retina, it was phase advanced by approximately 1-3 hours with a wider distribution. Upstream regulator analysis for the genes that showed phase shifts identified oxygen sensing mechanisms and HIF1alpha as regulators, but not the circadian clock, which remained in phase to the light/dark cycle. We propose a model in which early in diabetes, the diabetic retina experiences a ‘jet lag’ caused by the entrained circadian clock and its output being in one phase and metabolic pathways related to neuronal dysfunction and hypoxia driving advancement of gene expression to a different phase. Further studies are now required to evaluate the chronic implications of such internal jet lag for development of diabetic retinopathy. Control and hyperglycemic C57BL/6-Ins2Akita/J (Ins2Akita) mice were selected randomly at four months of age, and euthanized at one of six time-points over a 24hour light/dark cycle (ZT: 1, 5, 9, 13, 17, 21 where ZT0 is indicative of lights on and ZT12 lights off), with 4-5 biological replicates for each time-point