Unmyelinated fiber sensory neuropathy differs in type 1 and type 2 diabetes.

BACKGROUND: Neuropathic pain is common in diabetic patients. Degeneration of sensory C-fibers in peripheral nerve plays a prominent role in the generation of neuropathic pain. We examined degenerative changes of C-fibers in two rat models with type 1 and type 2 diabetes. METHODS: Type 1 insulinopenic BB/Wor and type 2 hyperinsulinemic diabetic BBZDR/Wor-rats of 8 months duration with equal exposure to hyperglycemia were examined. Thermal hyperalgesia was monitored using an infrared thermal probe. C-fiber size, number, frequencies of denervated Schwann cells, regenerating C-fibers, type 2 axon/Schwann cell relationship and collagen pockets in the sural nerve were examined morphometrically. Neurotrophic receptor expression was examined by Western blotting. Neurotrophins and neuropeptides were examined by ELISA. RESULTS: Type 1 rats showed increased thermal hyperalgesia followed by a decrease. Hyperalgesia in type 2 rats showed a slower progression. These findings were associated with a 50% (p < 0.001) loss of C-fibers, increased frequencies of denervated Schwann cells (p < 0.001), regenerating fibers (p < 0.001), collagen pockets (p < 0.001) and type 2 axon/Schwann cell relationship (p < 0.001) in type 1, but not in type 2 rats. Expression of insulin receptor, IGF-1R, TrkA and C was decreased in BB/Wor rats, whereas BBZDR/Wor rats showed milder or no deficits. NGF and NT-3 in sciatic nerve and substance P and calcitonin gene-related peptide in dorsal root ganglia were decreased in type 1, but not in type 2 rats. CONCLUSION: The more severe molecular, functional and morphometric abnormalities of nociceptive C-fibers in type 1 insulinopenic rats compared to type 2 hyperinsulinemic rats suggest that impaired insulin action may play a more important pathogenetic role than hyperglycemia per se.

A case of acquired deficiency of pituitary GH, PRL and TSH, associated with type 1 diabetes mellitus.

A 75-year-old male showed combined anterior pituitary hormone deficiency (CPHD). Basal and TRH-stimulated PRL levels were undetectable. Basal and GRH-stimulated GH levels were very low, and could barely be measured by means of an ultrasensitive enzyme immunoassay. In addition, basal TSH levels were under the normal limit, and TRH-stimulated TSH secretions were impaired. On the other hand, the secretions of ACTH, LH and FSH remained intact. There was no mutation of Pit-1 gene in this patient, and immunohistochemical studies using human pituitary and the patient's serum showed no positive staining. The HLA types frequently detected in lymphocytic hypophysitis were recognized, supporting the view that the CPHD in this case may be caused by lymphocytic hypophysitis, although magnetic resonance imaging of the pituitary gland showed no specific findings. Interestingly, a high titer of anti-glutamic acid decarboxylase antibody, suggested that the patient suffered from type 1 diabetes mellitus (DM). Five years ago, his thyroid function was normal and the treatment of DM with oral hypoglycemic agent was effective, indicating that the onset of both diseases at least occurred within the last half decade. We report here a rare case of SPIDDM with CPHD which might be caused by lymphocytic hypophysitis.

Molecular alterations underlie nodal and paranodal degeneration in type 1 diabetic neuropathy and are prevented by C-peptide.

To explore the molecular abnormalities underlying the degeneration of the node of Ranvier, a characteristic aberration of type 1 diabetic neuropathy, we examined in type 1 BB/Wor and type 2 BBZDR/Wor rats changes in expression of key molecules that make up the nodal and paranodal apparatus of peripheral nerve. Their posttranslational modifications were examined in vitro. Their responsiveness to restored insulin action was examined in type 1 animals replenished with proinsulin C-peptide. In sciatic nerve, the expression of contactin, receptor protein tyrosine phosphatase beta, and the Na(+)-channel beta(1) subunit, paranodal caspr and nodal ankyrin(G) was unaltered in 2-month type 1 diabetic BB/Wor rats but significantly decreased after 8 months of diabetes. These abnormalities were prevented by C-peptide administered to type 1 BB/Wor rats and did not occur in duration- and hyperglycemia-matched type 2 BBZDR/Wor rats. The expression of the alpha-Na(+)-channel subunit was unaltered. In SH-SY5Y cells, only the combination of insulin and C-peptide normalized posttranslational O-linked N-acetylglucosamine modifications and maximized serine phosphorylation of ankyrin(G) and p85 binding to caspr. The beneficial effects of C-peptide resulted in significant normalization of the nerve conduction deficits. These data describe for the first time the progressive molecular aberrations underlying nodal and paranodal degenerative changes in type 1 diabetic neuropathy and demonstrate that they are preventable by insulinomimetic C-peptide.

Insulin deficiency rather than hyperglycemia accounts for impaired neurotrophic responses and nerve fiber regeneration in type 1 diabetic neuropathy.

Diabetic polyneuropathy (DPN) shows more severe functional and structural changes in type 1 than in type 2 human and experimental diabetes. We have previously suggested that these differences may be due to insulin and/or C-peptide deficiencies in type 1 diabetes. To further explore these differences between type I and type 2 DPN, we examined factors underlying nerve fiber regeneration in the hyperinsulinemic type 2 BB/Z-rat and compared these with previous data obtained from the iso-hyperglycemic, insulin and C-peptide-deficient type 1 diabetic BB/Wor-rat. The expression of neurotrophic factors and cytoskeletal proteins were studied in L4 and L5 dorsal root ganglia (DRG) at various time points after sciatic nerve crush. The data were compared to those of nondiabetes-prone BB-rats. Insulin-like growth factor 1 (IGF-1) and TrkA levels were lower in DRG from type 1 than from those of type 2 and control BB-rats. On the other hand, IGF-1 receptor expression was increased at baseline in type 1 BB/Wor-rats and decreased after crush injury, whereas its expression increased after crush injury in both control and type 2 BB/Z-rats. Following crush injury, betaII- and betaIII-tubulins were upregulated in type 2 BB/Z and control rats, which did not occur in type 1 BB/Wor-rats. Furthermore, type 2 BB/Z-rats showed the normal downregulation of low and medium molecular neurofilament (NF-L and NF-M, respectively), which did not occur in type 1 BB/Wor-rats. These findings were associated with significantly milder abnormalities in axonal elongation and caliber growth of regenerating fibers in type 2 compared to type 1 diabetic rats. These data suggest that impaired insulin signaling in type 1 diabetic nerve may be of greater significance in the regulation of neurotrophic and neurocytoskeletal protein synthesis than hyperglycemia in explaining the differences in nerve fiber regeneration between type 2 and type 1 diabetes.

Impaired glucose tolerance and insulinopenia in the GK-rat causes peripheral neuropathy.

BACKGROUND: Recent studies indicate that impaired glucose tolerance (IGT) in man is a causative factor in idiopathic sensory neuropathy, and that insulinopenia may contribute substantially to the severity of diabetic peripheral neuropathy. The effect of sustained IGT and progressive insulinopenia in the absence of overt hyperglycemia on peripheral nerve abnormalities was examined in the Goto-Kakizaki (GK)-rat. METHODS: Two and eighteen-month-old GK rats with decreased glucose tolerance and overt insulinopenia, respectively, were examined with respect to nerve function, structure, morphometry and molecular integrity, and were compared to age-matched control rats. RESULTS: Both 2-(p < 0.001) and 18-month-old (p < 0.001) GK rats showed reduced body weight. Blood glucose levels following glucose tolerance tests were elevated in both the 2-month and the 18-month-old GK rats. Fasting plasma insulin levels in the 2-month GK rats were increased threefold (p < 0.05) but decreased by 71% (p < 0.001) in the 18-month GK rats. The two-month GK rats showed a normal nerve conduction velocity, whereas in the 18-month GK rats it was reduced to 76% (p < 0.001) of control values. No morphometric abnormalities were found in the 2-month GK rats, whereas the 18-month GK rats showed loss of small myelinated fibers (p < 0.001), atrophy and loss of unmyelinated axons (p < 0.05) and an increased (p < 0.01) frequency of regenerating fibers. In the older GK rats, both mRNA and protein expression of nerve growth factor (NGF) in the sciatic nerve were significantly reduced (p < 0.001 and p < 0.05), and NGFR TrkA (high affinity NGF receptor) and NGFRp75 (low affinity NGF-receptor) protein expression was reduced in dorsal root ganglia (DRG) (both p < 0.05). These changes were accompanied by significantly reduced protein expressions of substance P (SP) and calcitonin gene-related protein (CGRP) in DRG's (both p < 0.001) as well as a 40% (p < 0.001) decrease in SP and a 62% (p < 0.001) decrease in CGRP-positive DRG neurons. In the sciatic nerve, SP and CGRP protein expression was decreased by 71% (p < 0.01) and 79% (p < 0.01), respectively. CONCLUSION: IGT combined with hyperinsulinemia for 2 months have no detectable effect on peripheral nerve function or structure. In contrast, IGT and subsequent insulinopenia result in a functional and structural neuropathy associated with impaired NGF support and neuropeptide synthesis. We suggest that these abnormalities are mainly due to insulinopenia rather than hyperglycemia.

Early gene responses of trophic factors in nerve regeneration differ in experimental type 1 and type 2 diabetic polyneuropathies.

We have previously suggested that alterations in sequential early gene responses of trophic factors (IGF-1 -->c-fos-->NGF) contribute to impaired peripheral nerve regeneration in type 1 diabetic BB/W-rats. To study the role these responses may play in type 2 diabetic nerve regeneration, BB/Z-rats were subjected to sciatic nerve crush injury. The expression of IGF-1, c-fos, NGF and the receptors p75 and IGF-1R were determined at the protein and mRNA levels in sciatic nerve distal to the crush site by immunoblotting and semi-quantitative RT-PCR. In situ hybridization was performed to assess the cellular localization of IGF-1, NGF, p75, and IGF-1R mRNA and immunohistochemistry served to localize the source of p75 and IGF-1R protein expression. The data were compared to those of type 1 diabetic BB/Wor-rats and non-diabetic controls. Increased expression of IGF-1 in Schwann cells is the first growth factor response to injury and peaked at 0.5 hours (h) in control, 2 h in type 2 rats, and 24 h in type 1 rats. IGF-1R was expressed in Schwann cells and its expression was asynchronous to IGF-1 expression in type 1 rats but remained synchronous with IGF-1 in control and type 2 animals. The expression of the immediate early proto-oncogene c-fos exhibited an initial peak at 6 h in control animals, 24 h in type 2, and 2 days (d) in type 1 animals. The initial peak of NGF expression occurred at 6 h in non-diabetic rats, 24 h in type 2, and 2 d in type 1 diabetic rats. The expression of p75 was delayed and attenuated in type 1 diabetic rats; however, in type 2 diabetic rats it was similar to that of non-diabetic rats. These data indicate that early gene responses following nerve damage are significantly less perturbed in type 2 compared to type 1 diabetes. These differences may account for the more efficient nerve regeneration seen in type 2 diabetic polyneuropathy.

Expression and localization of insulin receptor in rat dorsal root ganglion and spinal cord.

The expression and localization of the insulin receptor (IR) was examined in rat dorsal root ganglia (DRG) and spinal cord using Western blotting, in situ hybridization and immunocytochemistry. Western blotting showed that the molecular weight of the IR beta subunit was higher in PNS than that found in CNS. Both IR mRNA and protein expressions were highest in small-sized sensory DRG neurons and myelinated sensory root fibers expressed higher levels of IR protein than myelinated anterior root fibers. In the spinal cord, IR immunoreactive neurons were present in lateral lamina V and in lamina X, suggesting the presence of IR in nociceptive pathways. Electronmicroscopy of DRGs revealed a polarized localization of the IR in abaxonal Schwann cell membranes, outer mesaxons in close vicinity to tight junctions of both myelinating and non-myelinating Schwann cells and to plasma membranes of sensory neurons. From these findings, we speculate that insulin may play a role in sensory fibers involved in nociceptive function often perturbed in diabetic neuropathy. The high expression of IR localizing to tight junctions of dorsal root mesaxons of DRGs may suggest a regulatory role on barrier functions compensating for the lack of a blood-nerve barrier in dorsal root ganglia. This is consistent with the colocalization of IR with tight junctions of the paranodal barrier and endoneurial endothelial cells in peripheral nerve.

Altered tubulin and neurofilament expression and impaired axonal growth in diabetic nerve regeneration.

Cytoskeletal protein expression in sensory neurons and sciatic nerve axonal growth were examined in type 1 diabetic BB/Wor rats after sciatic nerve crush injury. Diabetic male rats were subjected to sciatic nerve crush at 6 wk of diabetes. L4 and L5 dorsal root ganglia (DRG) mRNA expression of low and medium molecular weight neurofilaments (NF-L, NF-M), betaII- and betaIII-tubulin as well as protein expression of NF-L, NF-M, and beta-tubulin were examined at various time points following crush injury and compared with age- and sex-matched non-diabetic BB/Wor rats. Steady state mRNA expression of NF-L, NF-M, betaII- and betaIII-tubulin were decreased in diabetic DRG. NF-L and NF-M proteins were also decreased in DRG of uncrushed diabetic animals. After crush injury, betaII- and betaIII-tubulin mRNA were upregulated in control animals at day 2 and day 6, respectively, and beta-tubulin protein showed similarly increased expression after crush injury, while such upregulations did not occur in diabetic animals. Conversely, mRNA and protein expressions of NF-L, NF-M were downregulated to a lesser extent in diabetic animals compared to control rats. These changes were associated with impaired axonal elongation and caliber growth of regenerating fibers in diabetic rats. We propose that upregulation of tubulin has a negative feedback on NF expression in response to nerve injury, as seen in control rats. The absence of this upregulation in diabetic animals may impair its regulatory effect on NF expression and contribute to perturbed nerve regeneration seen in diabetic nerve.