DELAYED INITIATION OF REGENERATION IN DIABETIC NERVE IS PRECEDED BY ATTENUATION OF IMMEDIATE EARLY GENE RESPONSES
Shy M.E., Jiang H., Xu G., Sima A.A.F. Wayne State University, Detroit, MI., U.S.A.
Nerve fiber regeneration is impaired in diabetic nerve and contributes to the relentless nerve fiber loss characterizing this disorder. Immediate early gene responses constitute the initial response to nerve injury and include upregulation of NGF and IGF-1 primarily by Schwann cells. These responses are believed to initiate macrophage recruitment necessary for initiation of axonal regeneration. We examined NGF, IGF-l and CNTF mRNA in sciatic nerve at 10 timepoints (0.5hr to 24d) following sciatic nerve crush in diabetic BB/W-rats. The peak of the immediate upregulation of IGF-l and NGF occurred at 0.5 and 6 hrs respectively in control nerves and was delayed to 24 hrs for both IGF-1 and NGF in diabetic nerves. Also the expression of the NGF p75 receptor was significantly attenuated in diabetic nerve. CNTF mRNA showed an immediate downregulation following nerve crush with no significant differences between control and diabetic rats. These findings suggest that attenuations of the immediate gene responses of para- and autocrine IGF-1 and NGF in diabetic nerve may be responsible for the earlier reported defect in macrophage recruitment and delayed initiation of nerve fiber regeneration. Supported by JDFI and Thomas Foundation
LOCALIZATION OF INSULIN RECEPTOR ISOFORMS IN RAT PERIPHERAL NERVE
Sugimoto K., Zhang W., Xu G., Sima. A.A.F. Wayne State University, Detroit, MI., U.S.A.
Insulin deficiency and hyperinsulinemia are independent risk factors for the development and progression of peripheral neuropathy in diabetes. Studies in the insulinoma rat model show that hyperinsulinemia correlates with an increase in myelinated fiber and microvessel densities and a reduction in myelinated axon size in peripheral nerves. It is, however, unclear how and via which cellular components insulin exerts these effects. In the present studies, we examined localization of the insulin receptor (IR) and expression of two alternately spliced IR mRNAs in rat peripheral nerve. Eight to eleven-month old male non-diabetic BB/W-rats were used (n=7). Single teased nerve fibers and paraffin sections demonstrated immunoreactive IR:s localized to Schwann cell cytoplasm and the nodes of Ranvier of myelinated fibers and to microvessels in the sciatic nerve. RT-PCR using a primer pair flanking the alternately spliced exon 11 revealed that the major PCR product in the sciatic nerve was derived from IR mRNA without exon 11 (197 base pair), which encodes the high affinity IR isoform. This product was identical to that observed in brain, but different from that in liver (233 base pair). We conclude that peripheral nerve, like brain, expresses the high affinity IR. It is therefore possible that alteration in the availability of insulin may modulate vascular tone and nodal and axonal function and structure.
SCHWANN CELL PRODUCTION OF GLIAL DERIVED GROWTH FACTOR I IN NORMAL AND DIABETIC RATS
Sullivan K.A., Peacock M.L., Dixon M.K. University of Michigan, Departments of Int. Medicine and Neurology, Ann Arbor, MI
Diabetic neuropathy is a common complication of diabetes and
its cause remains unknown. Neurotrophic factors support and influence
the growth and regenerative capacity of axons and are produced
by target tissues and Schwann cells (SC). Decreased neurotrophism
has been implicated in diabetic neuropathy in animal models (streptozotocin
(STZ) induced) of diabetes. SC pathology includes, thinning of
the myelin sheath, demyelination, and eventual breakdown of the
SC. We hypothesize that high glucose interferes with the production
of neurotrophic factors, and that this lack of local trophic support
contributes to axonal injury. Glial derived neurotrophic factor
(GDNF) is an important regulator of axonal growth and repair expressed
within nerves by SC. The current study examines the appearance
of GDNF in the STZ treated rat. Male Sprague-Dawley rats (200-300g)
were assigned to three groups, controls(C) food restricted (WC,
weight maintained within 10% of diabetic animals) and STZ treated
(D). SCs from normal rats express GDNF and GDNFRa
protein. GDNF immunoreactivity (GDNF-IR) is localized within axons
and SC. Axonal staining of GDNF is similar in shape and size to
neurofilament-IR (NF-IR). GDNF-IR is variable, not every axon
is positive and staining intensity varies from strong to moderate.
SC are densely stained in both longitudinal and cross section.
No differences in GDNF-IR were detected among the groups. The
pattern of GDNFRa-IR is similar to
that observed GDNF-IR, however, less axonal staining is evident
overall. A similar proportion of SC contain GDNFRa-IR.
In addition the endoneurium/basement membranes are also positive.
GDNFRa-IR was similar between diabetic
and weight control animals, however both of these groups demonstrated
a slight decrease from control staining. Changes in GDNF-IR were
not detected by immunohistochemistry, however, a slight decrease
in GDNFRa-IR was noted between diabetic
and healthy rats. Further quantification of GDNF protein and mRNA
by ELISA and RT-PCR is underway to fully investigate changes in
GDNF expression in diabetic animals. Supported by the Michigan
Diabetes Research & Training Center Pilot Feasibility Program.
IGF-I PREVENTS HYPERGLYCEMIC INDUCTION OF OXIDATIVE STRESS LOSS OF MITOCHONDRIAL MEMBRANE INTEGRITY AND CASPASE CLEAVAGE IN MODELS OF DIABETIC NEURONAL PROGRAMMED CELL DEATH
Russell J.W., Rout P., van Golen C.M., Feldman E.L. University of Michigan, Department of Neurology Ann Arbor, MI
Recent evidence indicates that oxidative stress is associated
with mitochondrial (Mt) membrane depolarization (MMD), downregulation
of Bcl programmed cell death (PCD) regulators, and procaspase
cleavage. To study the role of oxidative stress and MMD in the
pathogenesis of diabetic sensory neuropathy, we examined changes
in vivo and in vitro in dorsal root ganglia (DRG)
sensory neurons and in human neuro-blastoma cells. Hyperglycemia
induces apoptotic changes, swelling and disruption of the Mt cristae
in diabetic DRG neurons but only rarely in control neurons. Similar
Mt changes are observed in cultured E15 DRG and human SH-SY5Y
neurons exposed to 20 mM excess glucose, but not when treated
with IGF-I. In timed experiments (0-24 h), 20 mM excess glucose
induces initial Mt swelling. Mt swelling is maximal at 3-6 h post
glucose exposure and is inhibited by 10nM IGF-I (p < 0.01).
In live neurons, after 12 h of excess glucose, MMD is significantly
increased (p < 0.001), and there is downregulation of Bcl-2
and Bcl-x(L), prevented by IGF-I. Interestingly, mean H2DCFDA
levels in live neurons (a measure of reactive oxygen species [ROS])
peak between 3 and 6 h, corresponding to Mt swelling, and decrease
as neurons undergo PCD. In contrast, IGF-I inhibits both oxidative
stress and PCD. We find in vitro and in vivo that
elevated glucose levels, commonly seen in uncontrolled diabetic
patients, result in initial Mt swelling, followed by MMD and disruption.
These results indicate that an early glucose induced rise in ROS
is associated with Mt swelling in vitro, and in turn MMD
precedes caspase cleavage and PCD. IGF-I inhibits production of
ROS and MMD, upstream of Bcl downregulation and caspase activation.
This supports the growing evidence of the importance of oxidative
stress and hyperglycemic induced apoptosis in diabetic neuropathy
and suggest that IGF-I therapy may preclude this PCD. Supported
by the Veterans Administration, NIH NS01938 (JWR); NIH R01 NS32843,
R01 NS38849, JDFI and ADA (ELF)
SORBITOL AND MYO-INOSITOL LEVELS AND MORPHOLOGY OF SURAL NERVE IN RELATION TO PERIPHERAL NERVE FUNCTION AND CLINICAL NEUROPATHY IN MEN WITH DIABETIC, IMPAIRED, AND NORMAL GLUCOSE TOLERANCE
Sundkvist G., Dahlin L-B, Nilsson H., Eriksson K-F, Lindgärde F., Rosén I., Lattimer S.A., Sima A.A.F., Sullivan K., Greene D.A. Malmö University Hospital, Malmö, Sweden, Wayne State University, Detroit MI, and University of Michigan, Ann Arbor, MI
Accumulation of sorbitol and depletion of myo-inositol (MI) are thought to be important for the development of diabetic neuropathy. In this study, sorbitol and MI levels and morphology of sural nerve were compared with nerve function and clinical neuropathy in men with diabetic, impaired (IGT), and normal glucose tolerance. Sural nerve conduction velocities (SNVC) (median [interquartile range]) (41.0 [6.0] m/s vs 47.0 [3.0] m/s; p = 0.014) and evoked action potential (SNAP) amplitudes (3.7 [3.5] mV vs 11.3 [10.6]) mV; p = 0.04) were significantly lower in diabetic than in IGT subjects whereas there were no differences in nerve function between IGT and normal subjects. Sorbitol levels were significantly higher in diabetic subjects (643 [412] nmol/mg prot.) compared with IGT (286 [83] nmol/mg prot.) and normal subjects (296 [250] mnol/mg prot.) and correlated with SNAP (rs = -0.693 ; p = 0.0376) in those with IGT. There were no differences in sural nerve morphology between the three groups. In diabetic and IGT subjects, however, myelinated nerve fiber density (MNFD) was significantly (p = 0.0021) lower in 9 subjects with clinical neuropathy (4076 [1091] nr fibers/mm2) than in 10 without (5219 [668] nr fibers/mm2). MI levels correlated positively with myelinated nerve fiber regeneration (density of myelinated fibers in clusters). MI levels were significantly lower (p = 0.0283) in diabetic patients with clinical neuropathy (25049 [3985] nmol/mg prot) than in those without (32253 [7503] nmol/mg prot). In conclusion, reduced MNFD correlated with neuropathy status but not glucose tolerance status, increased sorbitol levels were associated with disturbed nerve function even in subjects with IGT, and low MI levels were associated with clinical neuropathy and lower nerve fiber regeneration.
DISTRIBUTION OF EPIDERMAL NERVES IN CONTROL AND DIABETIC SUBJECTS.
Wendelschafer-Crabb G., Kennedy W.R., Hazen E., Wabner K. Dept. Neurology, Univ. of Minnesota, Minneapolis, MN, USA
The distal degenerative characteristic of diabetic neuropathy
and the early onset of sensory symptoms suggest that the first
morphological signs of peripheral nerve involvement occur in distal
segments of sensory epidermal nerve fibers (ENFs). We previously
quantified a reduction of ENFs in skin biopsies from diabetic
patients. In some patients ENF loss was accompanied by only mild
thinning of the subepidermal plexus and near normal innervation
of sweat glands, arterioles and arrector pilorum muscles. This
suggested that quantitation of ENFs in diabetic patients can be
useful for early diagnosis of neuropathy and that regeneration
of ENFs might be an objective indicator of improvement during
clinical therapeutic trials. The purpose of this study was to
determine ENF content at several body locations in normal and
diabetic subjects to study the distribution of neuropathy in these
superficial sensory nerves. Skin biopsy and skin blister samples
were removed from five sites: foot, calf, thigh forearm, and hand.
Thick sections were cut, immuno-stained, and imaged to quantify
ENFs. ENF content varied with body location, maximal in the forearm
and least in the calf. ENF loss in diabetic neuropathy was greatest
in the foot and calf, less in hand and thigh and rare in forearm.
ENF clustering accompanied the reduction in number. Severe ENF
loss was accompanied by loss of nerve in the subepidermal neural
plexus (SNP). Proximal hyperinnervation was often present. ENF
counts in skin biopsies were closely correlated to those in skin
blisters. The innocuous nature of the biopsy/blister method allows
sampling from multiple locations. The distribution of ENF loss
is greater than suspected clinically. This allows a greater choice
of biopsy sites and gives an appreciation of the distribution
of neuropathy. In instances where ENF regeneration might be expected,
as during a clinical trial, it is critical to make a pre-trial
selection of the site(s) to be sampled because if ENF and SNP
loss are too severe there will not be adequate neural source for
ENF regeneration.
Back to Program | PNS
Home