Several groups have reported apoptosis of dorsal root ganglion (DRG) cells as a prominent feature of diabetic polyneuropathy (DPN), although this has been controversial. Here, we examined subacute (4-month) type 1 diabetic BB/Wor rats with respect to sensory nerve functions, DRG and sural nerve morphometry, pro-and antiapoptotic proteins, and the expression of neurotrophic factors and their receptors. Sensory nerve conduction velocity was reduced by 13% and was accompanied by significant hyperalgesia. The numbers of DRG neurons including substance P-and calcitonin gene-related peptide-positive neurons were not altered, although they showed significant atrophy. Sural nerve morphometry showed decreased numbers of myelinated and unmyelinated fibers. Active caspase-3 and Bax expressions were increased, whereas antiapoptotic Bcl-xl and heat shock protein (HSP) 27 expressions in DRGs were increased. Nerve growth factor (NGF) contents in sciatic nerves and the expression of NGF receptor TrkA in DRGs were decreased. Immunohistochemistry showed increased numbers of active caspase-3-, HSP70-, and HSP27-positive neurons. Examinations of DRGs revealed no structural evidence of apoptosis but rather progressive hydropic degenerative changes. We conclude that apoptotic stress is induced in DRGs but is counterbalanced by survival elements in subacute type 1 diabetic BB/Wor rats and that distal nerve fiber loss reflects a dying-back phenomenon caused by impaired neurotrophic support. Diabetes 54: 3288 -3295, 2005 D iabetic polyneuropathy (DPN) is the most common late complication of diabetes and shows an increasing prevalence. Like other diabetes complications, DPN has been ascribed to hyperglycemia and subsequent metabolic abnormalities, particularly oxidative stress. Although other factors contribute to its development, such as insulin and C-peptide deficiencies and consequent deprivation of neurotrophic support (1), these have received less attention. Several investigators have reported neuronal apoptosis in relatively acute (2-to 3-month) streptozotocin-induced diabetes (STZ-D) in rats, sometimes exceeding 30% of dorsal root ganglion (DRG) cells as a contributing factor to DPN (2-4), while others have reported no significant neuronal loss in DRGs even after 12 months of STZ-D (5,6). The validity of major programmed cell death can be disputed, since the magnitude of reported apoptotic neuronal death does not correlate with a corresponding loss of axonal extensions in the peripheral nerve of the STZ-D rat. In contrast to the STZ-D rat, the spontaneously type 1 diabetic BB/Wor rat develops progressive sensory nerve fiber loss (7), as in humans, which therefore could reflect loss of parent DRG cells.Apoptotic phenomena occur in pancreatic -cells in type 1 diabetes (8), in diabetic retinopathy (9), and in hippocampal neurons in primary diabetic encephalopathy (10,11). Programmed cell death is believed to result from hyperglycemia-induced mitochondrial dysfunction (4,12) and depletion of antiapoptotic trophic factors like IGFs, i...