Do all retinal ganglion cells respond similarly to injury and protection?

Abstract

AbstractObjectives: To investigate the responses of different retinal ganglion cell (RGC) populations to diverse injuries and protection with selective agonists of the tropomyosin related kinase B (TrkB) receptor, brain‐derived neurotrophic factor (BDNF) or 7,8‐dihydroxyflavone (DHF).Methods: Adult Sprague–Dawley rats were used. A first group had their left eye connected to a saline reservoir elevated 1.5 meters above the eye to produce an acute ocular hypertension (AOH) and induce transient ischemia for 75 minutes, these were analysed at 3, 7, 14 or 45 days (d) (n = 9–14 for each survival interval). A second group received an intraocular injection of 100 mM N‐methyl‐D‐aspartate (NMDA) to induce retinal excitotoxicity and was analysed at 3, 7, 14d or 15 months (n = 7–11 for each survival interval). A third group received a left intraorbital optic nerve transection (IONT) to induce axotomy of the RGC population, these were analysed at 7, 10, 14, 21, 30 or 60d (n = 8 for each survival interval). The first group of rats received right after AOH a single intravitreal injection of 5 μl vehicle (1% Albumin in PBS) or 5 μg BDNF. Rats of the third group were treated daily with an intraperitoneal injection of vehicle (0.9% NaCl containing 1% DMSO) or DHF (5 mg/kg diluted in vehicle), while rats of the second group were untreated. The retinas were prepared as wholemounts and immunolabelled with Brn3a and melanopsin (m) antibodies to identify Brn3a+RGCs and m+RGCs, respectively. Wholemount image reconstructions were photographed; labelled RGCs were quantified and their distribution evaluated with isodensity or neighbour maps.Results: AOH with vehicle treatment results in rapid progressive loss of Brn3a+RGCs and in significant but not progressive loss of m+RGCs; BDNF treatment afforded significant prevention which was long‐lasting for both RGC populations. Intravitreal NMDA resulted at 7d in the abrupt loss of 74% of the Brn3a+RGC population but it did not progress further. The m+RGC population showed a significant diminution by 3d that recovered to control values by 7d and did not vary thereafter. IONT with vehicle treatment resulted in typical losses of RGCs that by 14d amounted to 80% or 64% of the Brn3a+ or m+RGC population, respectively. Brn3a+RGCs further decreased up to 60d whereas m+RGCs did not, so that the proportion of surviving m+RGCs was ≈five times that of Brn3a+RGCs, further showing their resilience to IONT. Systemic administration of DHF prevented Brn3a+RGC losses up to 21d while for m+RGCs such prevention was permanent.Conclusions: Our studies document that: (i) AOH induces progressive loss of Brn3a+RGCs but not of m+RGCs, and both populations respond to TrkB‐mediated protection; (ii) NMDA induced excitotoxicity results in rapid loss of Brn3a+RGCs, whereas m+RGCs survived for up to 15 months later; (iii) IONT and vehicle treatment result in characteristic rapid loss of Brn3a and melanopsin RGCs, whereas DHF treatment affords protection of axotomized RGCs. This protection is maximal at 7d and significant until 21d for Brn3a+RGCs, whereas for m+RGCs the protection was significant at 14d and permanent. Thus, two different RGC populations (Brn3a+ and m+) respond distinctly to injury and TrkB‐mediated protection.