Quantitative analysis of pupillary light reflex by real-time autofluorescent imaging in a diabetic mouse model

Here we (i) introduce a novel laser-based quantitative method of measuring pupillary light reflex (PLR) and applied it for the in vivo PLR monitoring of early diabetic retinopathy (DR) in a mouse model, (ii) investigate if melanopsin-expressing retinal ganglion cells (mRGCs) are implicated in the early progression of DR and, if so, is there an impact on PLR and (iii) determine if changes in PLR precede the onset of retinal hypoperfusion. A base-line PLR measurement is captured from C57BL/6J wild type mice followed by a single intraperitoneal administration of 200 mg/kg streptozotocin (STZ) and citrate buffer (vehicle) for the "diabetic" (n = 5) and "control" (n = 5) mice respectively, the very next day. PLR measurements are repeated once a week for four weeks. The PLR data comprises retinal autofluorescence intensity (AFI) values sampled over a 5-min period under confocal excitation with 488 nm high intensity blue laser light. AFI is used here as an indirect measure of pupil size since the amount of excitation light entering and emission light leaving the eye is proportional to the pupil area. Immunohistochemistry (IHC) staining of mRGCs and RT-PCR of melanopsin mRNA are performed at the end-point. The vascular calibre of both control and STZ-treated diabetic mice is assessed via in vivo fluorescein angiography (FA) on day 0 (base-line), 1/2, 1 and 4 months post-STZ treatment. The PLR profile shows a more rapid pupil constriction and slower dilation in diabetic mice compared to the control. Changes in PLR coincide or even precede the onset of retinal hypoperfusion. Extensive dendritic network of the mRGCs in retinal whole-mounts and increased melanopsin mRNA from the whole eye are also observed in diabetic mice. These pathological changes to mRGCs during early DR may in turn contribute towards changes in PLR. We present here a quantitative method of measuring PLR which enables an early detection of DR with potential application in the clinical setting. In contrast to conventional measurements of PLR, we are able to calibrate the amount of light reaching the retina which is a crucial parameter in longitudinal studies.