Sensory neurons with activated caspase-3 survive long-term experimental diabetes.

Long-term experimental diabetes may best model the prominent and irreversible sensory deficits of chronic human diabetic polyneuropathy. Whereas irretrievable loss of sensory neurons, if present, would be an unfortunate feature of the disease, systematic unbiased counting has indicated that sensory neurons survive long-term experimental diabetes. In this study, we examined whether incipient cell loss from apoptosis in chronic experimental diabetes might nonetheless be in process, or whether neurons somehow adapt to their chronic insults. We examined sensory neurons in L4 and L5 dorsal root ganglia of long-term experimental streptozotocin-induced diabetic rats using transferase-mediated dUTP nick-end labeling (TUNEL), 4',6-diamidino-2-phenylindole (DAPI) staining of nuclear morphology, and electron microscopic appraisal of cell morphology. None provided any evidence for ongoing apoptosis. Despite this confirmation that sensory neurons survive, neurons had elevated expression of activated caspase-3 in unique patterns that included their nuclei, cytoplasm, and proximal axonal segments. Bcl-2 expression, a marker of antiapoptosis signaling, was observed in similar numbers of diabetic and nondiabetic neurons. In contrast, diabetic sensory neurons had elevated expression of the DNA repair enzyme poly(ADP-ribose) polymerase (PARP) in their nuclei, cytoplasm, and proximal axonal segments not overlapping with caspase-3 localization. Diabetic sensory neurons also had an apparent rise in cytoplasmic labeling of nitrotyrosine, a marker of peroxynitrite toxicity reported to activate PARP.