Axonal transport of acetylcholinesterase in the diabetic mutant mouse

Age-dependent slowing of enteric axonal transport in insulin-resistant mice

AIM: To investigate retrograde tracer transport by gastric enteric neurons in insulin resistant mice with low or high glycosylated hemoglobin (Hb).

METHODS: Under anesthesia, the retrograde tracer fluorogold was superficially injected into the fundus or antrum using a microsyringe in KK Cg-Ay/J mice prior to onset of type 2 diabetes mellitus (T2DM; 4 wk of age), at onset of T2DM (8 wk of age), and after 8, 16, or 24 wk of untreated T2DM and in age-matched KK/HIJ mice. Six days later, mice were sacrificed by CO₂ narcosis followed by pneumothorax. Stomachs were removed and fixed. Sections from fundus, corpus and antrum were excised and mounted on a glass slide. Tracer-labeled neurons were viewed using a microscope and manually counted. Data were expressed as the number of neurons in short and long descending and ascending pathways and in local fundus and antrum pathways, and the number of neurons in all regions labeled after injection of tracer into either the fundus or the antrum.

RESULTS: By 8 wk of age, body weights of KKAy mice (n = 12, 34 ± 1 g) were heavier than KK mice (n = 17, 29 ± 1 g; F (4, 120) = 4.414, P = 0.002] and glycosylated Hb was higher [KK: (n = 7), 4.97% ± 0.04%; KKAy: (n = 6), 6.57% ± 0.47%; F (1, 26) = 24.748, P < 0.001]. The number of tracer labeled enteric neurons was similar in KK and KKAy mice of all ages in the short descending pathway [F (1, 57) = 2.374, P = 0.129], long descending pathway [F (1, 57) = 0.922, P = 0.341], local fundus pathway [F (1, 53) = 2.464, P = 0.122], local antrum pathway [F (1, 57) = 0.728, P = 0.397], and short ascending pathway [F (1, 53) = 2.940, P = 0.092]. In the long ascending pathway, fewer tracer-labeled neurons were present in KKAy as compared to KK mice [KK: (n = 34), 302 ± 17; KKAy: (n = 29), 230 ± 15; F (1, 53) = 8.136, P = 0.006]. The number of tracer-labeled neurons was decreased in all mice by 16 wk as compared to 8 wk of age in the short descending pathway [8 wk: (n = 15), 305 ± 26; 16 wk: (n = 13), 210 ± 30; F (4, 57) = 9.336, P < 0.001], local antrum pathway [8 wk: (n = 15), 349 ± 20; 16 wk: (n = 13), 220 ± 33; F (4, 57) = 8.920, P < 0.001], short ascending pathway [8 wk: (n = 14), 392 ± 15; 16 wk: (n = 14), 257 ± 33; F (4, 53) = 17.188, P < 0.001], and long ascending pathway [8 wk: (n = 14), 379 ± 39; 16 wk: (n = 14), 235 ± 26; F (4, 53) = 24.936, P < 0.001]. The number of tracer-labeled neurons decreased at 24 wk of age in the local fundus pathway [8 wk: (n = 14), 33 ± 11; 24 wk: (n = 12), 3 ± 2; F (4, 53) = 5.195, P = 0.001] and 32 wk of age in the long descending pathway [8 wk: (n = 15), 16 ± 3; 32 wk: (n = 12), 3 ± 2; F (4, 57) = 2.944, P = 0.028]. The number of tracer-labeled enteric neurons was correlated to final body weight for local fundus and ascending pathways [KK: (n = 34), r = -0.746, P < 0.001; KKAy: (n = 29), r = -0.842, P < 0.001] as well as local antrum and descending pathways [KK (n = 36), r = -0.660, P < 0.001; KKAy (n = 31), r = -0.622, P < 0.001]. In contrast, glycosylated Hb was not significantly correlated to number of tracer-labeled neurons [KK (n = 17), r = -0.164, P = 0.528; KKAy (n = 16), r = -0.078, P = 0.774].

CONCLUSION: Since uncontrolled T2DM did not uniformly impair tracer transport in gastric neurons, long ascending neurons may be more susceptible to persistent hyperglycemia and low effective insulin.

Decreased glycolytic and tricarboxylic acid cycle intermediates coincide with peripheral nervous system oxidative stress in a murine model of type 2 diabetes

Diabetic neuropathy (DN) is the most common complication of diabetes and is characterized by distal-to-proximal loss of peripheral nerve axons. The idea of tissue-specific pathological alterations in energy metabolism in diabetic complications-prone tissues is emerging. Altered nerve metabolism in type 1 diabetes models is observed; however, therapeutic strategies based on these models offer limited efficacy to type 2 diabetic patients with DN. Therefore, understanding how peripheral nerves metabolically adapt to the unique type 2 diabetic environment is critical to develop disease-modifying treatments. In the current study, we utilized targeted liquid chromatography-tandem mass spectrometry (LC/MS/MS) to characterize the glycolytic and tricarboxylic acid (TCA) cycle metabolomes in sural nerve, sciatic nerve, and dorsal root ganglia (DRG) from male type 2 diabetic mice (BKS.Cg-m+/+Lepr(db); db/db) and controls (db/+). We report depletion of glycolytic intermediates in diabetic sural nerve and sciatic nerve (glucose-6-phosphate, fructose-6-phosphate, fructose-1,6-bisphosphate (sural nerve only), 3-phosphoglycerate, 2-phosphoglycerate, phosphoenolpyruvate, and lactate), with no significant changes in DRG. Citrate and isocitrate TCA cycle intermediates were decreased in sural nerve, sciatic nerve, and DRG from diabetic mice. Utilizing LC/electrospray ionization/MS/MS and HPLC methods, we also observed increased protein and lipid oxidation (nitrotyrosine; hydroxyoctadecadienoic acids) in db/db tissue, with a proximal-to-distal increase in oxidative stress, with associated decreased aconitase enzyme activity. We propose a preliminary model, whereby the greater change in metabolomic profile, increase in oxidative stress, and decrease in TCA cycle enzyme activity may cause distal peripheral nerves to rely on truncated TCA cycle metabolism in the type 2 diabetes environment.

Criteria for creating and assessing mouse models of diabetic neuropathy

Diabetic neuropathy (DN) is a serious and debilitating complication of both type 1 and type 2 diabetes. Despite intense research efforts into multiple aspects of this complication, including both vascular and neuronal metabolic derangements, the only treatment remains maintenance of euglycemia. Basic research into the mechanisms responsible for DN relies on using the most appropriate animal model. The advent of genetic manipulation has moved mouse models of human disease to the forefront. The ability to insert or delete genes affected in human patients offers unique insight into disease processes; however, mice are still not humans and difficulties remain in interpreting data derived from these animals. A number of studies have investigated and described DN in mice but it is difficult to compare these studies with each other or with human DN due to experimental differences including background strain, type of diabetes, method of induction and duration of diabetes, animal age and gender. This review describes currently used DN animal models. We followed a standardized diabetes induction protocol and designed and implemented a set of phenotyping parameters to classify the development and severity of DN. By applying standard protocols, we hope to facilitate the comparison and characterization of DN across different background strains in the hope of discovering the most human like model in which to test potential therapies.

Effects of ALCAR on the fast axoplasmic transport in cultured sensory neurons of streptozotocin-induced diabetic rats

The effects of acetyl-L-carnitine (ALCAR) on fast axoplasmic transport were studied in cultured dorsal root ganglion (DRG) neurons of diabetic rats. Three-month-old male rats were used 7 days after streptozotocin injection. Neurons obtained from ganglia were cultured with a high concentration of glucose. The amount and the mean velocity of retrogradely transported particles, reduced in the diabetic animal, were transiently recovered by 1 mM ALCAR. The number of particles moving at 0.8-1.2 microm/s, considered to be lysosomes, increased in the velocity distribution. ALCAR did not modify the amount and mean velocity of anterograde particles which were unaffected by diabetes, or of bidirectional particles in neurons of control rats. This study suggests that diabetic neuropathy may be relieved by ALCAR via recovering retrograde axoplasmic transport.

Diabetic neuropathy in the rat: 1. Alcar augments the reduced levels and axoplasmic transport of substance P

This study examined the sciatic nerve axonal transport of substance P-like immunoreactivity (SPLI) and its basal content in stomach, sciatic nerve and lumbar spinal cord of 8- and 12-week alloxan-diabetic rats, respectively. One group of diabetic rats received acetyl-l-carnitine (ALCAR) throughout the experimental period. Alloxan treatment caused hyperglycemia and reduced boy growth. Axonal transport of SPLI was studied by measurement of 24-hour accumulation at a ligature on the sciatic nerve. There was a marked reduction (from 50% to 100% according to the nerve segment examined) of anterograde and retrograde accumulation of SPLI in the constricted nerve of 8-week diabetic rats. In the sciatic nerve of ALCAR-treated diabetic rats, the accumulation of SPLI was comparable to control values. In the sciatic nerve, lumbar spinal cord and stomach of 12-week diabetic rats, there is a significant reduction of SPLI content. ALCAR treatment prevented SPLI loss in these tissues. Sciatic nerves showed the typical sorbitol increase and myo-inositol loss that were significantly counteracted by ALCAR. This study suggests that ALCAR treatment prevents diabetes-induced sensory neuropathy by improving altered metabolic pathways such as polyol activity and myo-inositol synthesis, and by preventing the reduction of synthesis and axonal transport of substance P.

Time-course of nociceptive disorders induced by chronic loose ligatures of the rat sciatic nerve and changes of the acetylcholinesterase transport along the ligated nerve

Changes in the axonal transport of acetylcholinesterase (AChE) were studied in the painful mononeuropathy induced by setting 4 loose ligatures around the right sciatic nerve of the rat. Since changes in the axonal transport of AChE can be used to assess axonal degeneration/regeneration, we used this marker to investigate whether the time course of pain-related behavioral disorders observed following chronic constriction injury (CCI) to the sciatic nerve are related to the time course of the regeneration of the injured axons. In addition, a comparison was made between changes in AChE observed in this model of nerve injury and those observed after sciatic nerve crush. The rats were examined for pain-related disorders daily during the first postoperative week then at 7, 14 and 21 days after nerve ligation. The pain-related disorders, only detected from 7 days after ligation, were maximal at 14 days postinjury, and began to lessen at the end of the 3rd postoperative week. Within the first 3 days after loose ligation, the AChE transport dropped to 40% of its normal value, but recovered rapidly during the 3rd week post-surgery, indicating that most of the injured neurons were reconnecting their target cells. Thus, the injury produced by the loose ligatures was registered by the neurons several days before the first nociceptive manifestations of the injury, and the pain-related disorders lasted after most of the re-elongating axons had reconnected their target.(ABSTRACT TRUNCATED AT 250 WORDS)

Defective activity of Na+,K(+)-ATPase in peripheral nerve of diabetic rats is independent of the axonal transport of the enzyme

This study addressed the question as to whether the reduced activity of Na+,K(+)-ATPase reported to occur in diabetic nerves and to play a crucial role in the pathogenesis of diabetic neuropathy could be due to derangements in the axonal transport of the enzyme. A micromethod was developed to evaluate the ATPase accumulation in individual segments of ligated sciatic nerves from streptozotocin-induced diabetic rats. The results confirmed a approximately 40% decrease in the background activity, but showed that the enzyme was transported at similar rates in both anterograde and retrograde directions, suggesting that the decrease in its activity does not depend on an altered delivery along the axons.

Aldose reductase activity and myo-inositol levels in sciatic nerve and dorsal root ganglia of the diabetic mutant mouse [C57/BL/Ks (db/db)]

Glucose, polyol (sorbitol, fructose), and myo-inositol levels were estimated in peripheral nerve tissue of the diabetic mutant mouse [C57/BL/Ks (db/db)]. At 26 and 40 weeks of age, there was significant accumulation of glucose, sorbitol, and fructose. Tissue myo-inositol levels were lower than those in age-matched control animals at 40 weeks, but not at 26 weeks. Polyol changes in mouse nerve were less marked than those in rat nerve. In dorsal root ganglia, there was also an increase in glucose, sorbitol, and fructose, with a decrease in myo-inositol concentration. These findings show that there is increased aldose reductase activity in peripheral nerve tissue of the diabetic mutant mouse.

Diabetic neuropathy in db/db mice develops independently of changes in ATPase and aldose reductase. A biochemical and immunohistochemical study

ATPase activity was investigated in sciatic and optic nerves of female mutant diabetic C57Bl/Ks (db/db) mice and age-matched control mice (db/m and m/m). Nerves from animals aged 50, 70, 125, 180 and 280 days were assayed in vitro for ATPase activity in the presence or absence of ouabain: the ouabain-sensitive fraction contained Na+,K(+)-ATPase. Enzymatic activity was compared within and between age-matched groups. No significant difference in Na+,K(+)-ATPase activity was detected between the diabetic and control mice, whether expressed as mumol Pi/h-1 formed per gramme wet weight or per nerve (protein content). The activity decreased by about 25% in both the sciatic and optic nerves of the oldest animals. These results were strikingly similar in all groups, regardless of the type of nerve examined, confirming that the development of neuropathy in this animal model is unrelated to the postulated derangement of Na+,K(+)-ATPase activity. Among possible explanations, a lack of polyol pathway activation was investigated by staining the sciatic nerves of animals from all groups with the peroxidase-antiperoxidase procedure using a polyclonal antiserum raised against the enzyme aldose reductase. Histological sections of all nerves were consistently negative, suggesting that these animals actually lack the enzyme involved in activating the self-perpetuating metabolic cycle leading to deranged nerve function. The db/db mouse appears to present particular biochemical changes which merit attention with a view to clarifying the pathogenesis of diabetic neuropathy.

Denervation and hyperinnervation in the nervous system of diabetic animals: III. Functional alterations of G proteins in diabetic encephalopathy

G protein-mediated effects on cAMP production were evaluated in the corpus striatum of diabetic rats 5 and 14 weeks after alloxan injection by measuring both D1-receptor-induced stimulation and D2-receptor-mediated inhibition of adenylate-cyclase activity. At 5 weeks of diabetes, no obvious alterations of G protein functions were detected. Both dopamine-stimulated adenylate cyclase and bromocriptine-induced inhibition of enzyme activity were indeed similar in control and diabetic animals. Fourteen weeks after alloxan injection, profound alterations were observed. Dopamine-stimulated cAMP production was markedly increased in diabetic rats, whereas bromocriptine ability to reduce cAMP formation was almost abolished at this late stage of diabetes. Hypoactivity of Gi/Go proteins was also confirmed by the reduced ability of the GTP non-hydrolyzable analog GTP-gamma-S to inhibit forskolin-stimulation of adenylate cyclase. These results show an apparent functional imbalance between Gs and Gi/Go-mediated transduction mechanisms, with an increased efficacy of Gs activity likely due to the loss of Gi/Go inhibitory functions. Concomitantly with such transductional alteration detected in chronic diabetes, we observed a marked increase of the striatal content of met-enkephalin, which is known to utilize Gi/Go proteins for inhibition of adenylate cyclase. The measurement of other transmitters (vaso-active intestinal peptide, substance P, serotonin, noradrenaline, and dopamine) did not reveal any difference with respect to controls. The observed transductional defect in diabetic animals and the increased content and/or hyperinnervation by the metenkephalinergic system could be correlated as mutual compensatory mechanisms.

Denervation and hyperinnervation in the nervous system of diabetic animals. I. The autonomic neuronal dystrophy of the gut

Peripheral neuropathy is a correlate of experimental diabetes induced in rats by means of a single injection of alloxan. The autonomic and enteric innervation of the gut are profoundly affected in the small intestine of such animals. A complex process of denervation and hyperinnervation of the gut wall of diabetic animals is observed. It was previously reported that the cholinergic parasympathetic innervation of the intestine is markedly reduced. We have found that noradrenergic sympathetic axons hyperinnervate the duodenum of diabetic rats, whereas noradrenaline levels are significantly reduced in the jejunum. The putative enteric neurotransmitter dopamine is also present in higher levels in the duodenum. The intrinsic peptidergic neurons of the gut are deeply affected as well in diabetic rats. Substance P and met-enkephalin content are remarkably reduced throughout the small intestine, whereas vasoactive intestinal polypeptide levels (VIP) are significantly increased in the duodenum. Indeed, immunocytochemical staining of the ileum did reveal hypertrophy of VIP-positive axons in diabetic rats. The intrinsic serotoninergic innervation of the gut is apparently unaffected. Our results indicate that the changes of gut innervation observed in experimental diabetes are consistent with increased content and also likely with hyperinnervation by the neuronal systems involved in smooth muscle relaxation and decreased content and with denervation by those systems with smooth muscle contraction properties. Such a perturbed gut innervation may be responsible of the gastrointestinal dysfunctions that are among the most common complications of diabetes.

Denervation and hyperinnervation in the nervous system of diabetic animals. II. Monoaminergic and peptidergic alterations in the diabetic encephalopathy

The monoaminergic innervation of the central nervous system (CNS) is characterized by long and short projecting neurons. The neurological correlates of diabetes are usually referred to as processes of degenerative atrophy affecting motor and sensory peripheral nerves. We have found that the long serotoninergic axons innervating the spinal cord and the cerebral cortex are unaffected in diabetic animals and that the noradrenergic innervation of the cortex is normal as well. The serotonin content is doubled in the hypothalamus with no apparent alteration of 5-HIAA levels, suggesting a supernumerary innervation that is accompanied by a reduced release. In pons medulla oblongata, serotonin and dopamine with the relative metabolites 5-HIAA and DOPAC are significantly reduced, whereas noradrenaline is markedly increased. In the hippocampus, there is a reduction of serotonin content. The serotoninergic alterations are peculiar as suggested by the sparing of the most distal projections that is accompanied by hyperinnervation of the hypothalamus and the loss of shorter collaterals in the pons medulla oblongata. In the hypothalamus and in the striatum of diabetic rats, there are significant higher levels of substance P and met-enkephalin, respectively. The abundance of proenkephalin A mRNA is also increased in the striatum. Conversely, in the lumbar cord of diabetic animals, the levels of substance P and met-enkephalin are significantly reduced. Such alterations likely reflect retrograde degeneration of the peripheral sensory input. The CNS changes are unlikely due to vascular abnormalities in the brain of diabetic rats; rather, we suggest that the persistent lack of insulin is the major factor involved as a trigger of the monoaminergic changes in the diabetic brain.

Slow axonal transport impairment of cytoskeletal proteins in streptozocin-induced diabetic neuropathy

The impairment of slow axonal transport of cytoskeletal proteins was studied in the sciatic nerves of streptozocin-diabetic rats. [35S]Methionine was unilaterally injected into the fourth lumbar ganglion and spinal cord, to label the sensory and motor axons, respectively, and then the polymerized elements of the cytoskeleton and the corresponding soluble proteins were analyzed separately. In addition, the pellet/supernatant ratio for tubulin and actin was also assessed. Our results indicate that the velocity of slow component a (SCa) of axonal transport, particularly that of neurofilaments, was strongly reduced (by 60%) in sensory axons. At the same time, a decreased pellet/supernatant ratio of tubulin, possibly owing to a depolymerization of stable microtubules, was also observed. The transport of slow component b (SCb) of axonal transport was also impaired, but the extent of this impairment could not be precisely evaluated. In contrast, motor axons showed little or no impairment of both SCa and SCb at the time studied, a result suggesting a delayed development of the neuropathy in motor axons.

Axonal transport in mdx mouse sciatic nerve

Anterograde and retrograde flows of acetylcholinesterase (AChE) in sciatic nerves of adult mdx mice were compared with those of normal mice. Specific molecular forms of AChE were resolved by high-performance liquid chromatography such that slow anterograde (G1 + G2), fast anterograde and fast retrograde (G4 and A12) flows could be simultaneously studied. Although we found no difference in the total AChE activity and the molecular forms in non-ligated nerves between mdx and the normal mice, ligated nerves showed significant differences. The total AChE activity accumulated at the proximal segment of ligated nerve was higher in mdx mice than in normal mice after 24 h ligation. The G1 + G2 molecular forms were accumulated more in the proximal segment of mdx than the normal. A12, on the other hand, was more abundant in both segments of mdx mice than the normal. No statistically significant difference in the accumulated amount of G4 molecular form was present between mdx and the normal mice at either proximal or distal segment. These results indicated that axonal flow in sciatic nerve likely plays a role in muscle regeneration, and that the transport machinery in dystrophin-deficient mdx neuron is probably normal.

Reduced axonal transport of the G4 molecular form of acetylcholinesterase in the rat sciatic nerve during aging

Aging in the sciatic nerve of the rat is characterized by various alterations, mainly cytoskeletal impairment, the presence of residual bodies and glycogen deposits, and axonal dystrophies. These alterations could form a mechanical blockade in the axoplasm and disturb the axoplasmic transports. However, morphometric studies on the fiber distribution indicate that the increase of the axoplasmic compartment during aging could obviate this mechanical blockade. Analysis of the axoplasmic transport, using acetylcholinesterase (AChE) molecular forms as markers, demonstrates a reduction in the total AChE flow rate, which is entirely accounted for by a significant bidirectional 40-60% decrease in the rapid axonal transport of the G4 molecular form. However, the slow axoplasmic flow of G1 + G2 forms, as well as the rapid transport of the A12 form of AChE, remain unchanged. Our results support the hypothesis that the alterations observed in aged nerves might be related either to the impairment in the rapid transport of specific factor(s) or to modified exchanges between rapidly transported and stationary material along the nerves, rather than to a general defect in the axonal transport mechanisms themselves.

Axonal transport of choline acetyltransferase and 6-phosphofructokinase activities in genetically diabetic mice

This study examined the anterograde axonal transport of activities of the cytoplasmic enzymes choline acetyltransferase and 6-phosphofructokinase in genetically diabetic C57BL/Ks (db/db) mice and their nondiabetic (+/?) littermates. Diabetic mice exhibited marked reductions in the accumulation of both choline acetyltransferase and 6-phosphofructokinase activity against a constriction of the left sciatic nerve (38% and 51% of nondiabetic values, respectively). Enzyme activities per unit length of unconstricted nerve were not different from those of nondiabetic mice. The nerves of diabetic mice did not accumulate measurable amounts of sorbitol or fructose and showed no myo-inositol depletion. Thus this study concludes that, in diabetic mice, the deficits in anterograde axonal transport of these two enzymes do not arise from the accumulation of sorbitol and fructose nor from depletion of nerve free myo-inositol.

Statil-sensitive polyol formation in nerve of galactose-fed mice

The sciatic nerve of diabetic C57BL/Ks mice did not accumulate either sorbitol or fructose despite elevated nerve glucose levels, nor did they show a depletion of free myo-inositol. Galactose feeding of both nondiabetic and diabetic mice for five days resulted in marked accumulations of dulcitol in nerve indicating polyol forming activity. Levels of myo-inositol showed small elevations in nerves of galactose-fed mice which reached significance when compared to diabetic mice fed the standard diet. Dulcitol accumulation was prevented by concurrent administration of the aldose reductase inhibitor Statil, but this did not prevent the increased levels of myo-inositol.

Axonal transport of the molecular forms of acetylcholinesterase in rats at the onset of diabetes induced by streptozotocin

During the development of streptozotocin-induced diabetic neuropathy in the rat, the axonal transport of 4 acetylcholinesterase molecular forms was studied by measuring their accumulation on both sides of transected sciatic nerves. Our results indicate that both the anterograde and retrograde axonal transport of all these forms remain normal between 2 and 5 weeks after the induction of diabetes by streptozotocin injection.

Slow and fast axonal transport of acetylcholinesterase molecular forms in polyarthritic rats

Acetylcholinesterase (AChE) activity and its distribution among different molecular forms were studied in the sciatic nerve of normal and polyarthritic rats. Axonal transport of each form was investigated on the basis of its accumulation on both sides of a transection. Although an increase in total AChE activity could be detected in the sciatic nerves of polyarthritic animals, both anterograde and retrograde axonal transport of all the molecular forms investigated were similar in normal and polyarthritic rats. This suggests that neither slow nor fast axonal transport is impaired in polyarthritic rats. Hence, the neurophysiological modifications observed at the spinal, thalamic and cortical levels of the CNS are presumably not a consequence of peripheral axonal disability.

Impaired axonal transport of acetylcholinesterase in the sciatic nerve of alloxan-diabetic rats: effect of ganglioside treatment

The anterograde and retrograde axonal flow of acetylcholinesterase were studied in the sciatic nerve of alloxan-diabetic rats after five weeks of experimental diabetes. A slight reduction of the anterograde axonal flow of the enzyme was found in alloxan-diabetic compared to control rats. Sedimentation analysis revealed a major reduction of anterograde axonal flow of the light globular forms of the enzyme (G1 + G2), which are probably conveyed by slow transport. There was also a minor reduction of the anterograde flow of the globular form G4, while no modification of the axonal flow of the heavy asymmetric form A12 was found. Both G4 and A12 molecular forms are conveyed by fast axonal transport. In contrast, no abnormality of the retrograde axonal flow of acetylcholinesterase was observed. Ganglioside treatment antagonized the decline of the anterograde axonal flow of the enzyme in alloxan-diabetic rats. These results are consistent with the view that experimental diabetic neuropathy is associated with axonal transport defects, and suggest a protective effect of ganglioside treatment against neuronal damage(s) related to the diabetic syndrome.

Age-related changes of neurites in Meissner corpuscles of diabetic mice

Light touch and low-frequency vibration sense deteriorate during the development of diabetic neuropathy. As Meissner corpuscles are mechanoreceptors that respond to these sensations, this study explored the structural changes of neurites in Meissner corpuscles of diabetic mice C57BL/Ks(db/db). We evaluated silver impregnated neurites from forepaw digital pads from 46 diabetic and 46 nondiabetic female mice which ranged in age from 2.5 to 17 months. Light microscopically, neurites from diabetic mice were less coarse, less tortuous, and exhibited decreased varicosity and decreased corpuscle size compared with those from nondiabetic mice. Number of corpuscles per area and neurite intraepidermal continuations showed a statistically significant decrease with age and with diabetes. Projected area and width of neurites, measured within a fixed interval on camera lucida tracings, showed both a statistically significant increase with age and a decrease with diabetes. Neurite complexity was unchanged between diabetics and nondiabetics. These findings suggest that axonal dwindling, a characteristic of peripheral nerves in diabetes, extends to the receptors and occurs throughout the lifespan of the mouse.

Experimental diabetic neuropathy: impairment of slow transport with changes in axon cross-sectional area

Analysis of slow axonal transport in the sciatic and primary visual systems of rats with streptozotocin-induced diabetes of 4-6 weeks duration showed impairment of the transport of neurofilament subunits, tubulin, actin, and a 30- and a 60-kDa polypeptide in both systems. The degree of impairment was not uniform. Transport of polypeptide constituents of the slow component b, such as the 30- and 60-kDa polypeptides, appeared to be more severely affected than the transport of constituents of the slow component a, such as neurofilaments. Morphometric analysis of sciatic axons revealed a proximal increase and a distal decrease of axonal cross-sectional area. It is proposed that impairment of axoplasmic transport and changes of axonal size are related. Transport impairment results in a larger number of neurofilaments, microtubules, and other polypeptides in the proximal region of the axon, which increases in size, whereas fewer neurofilaments, microtubules, and other polypeptides reach the distal axons that show a size decrease. Such changes in axonal transport and area are likely to occur in other diabetic animal models and in human diabetes.

Retrograde effect of muscle on forms of acetylcholinesterase in peripheral nerves

In the peripheral nerves of birds and mammals, acetylcholinesterase (AChE) exists in four main molecular forms (G1, G2, G4, and A12). The two heaviest forms (G4 and A12) are carried by rapid axoplasmic transport, whereas the two lightest forms (G1 and G2) are probably much more slowly transported. Here we report that nerves innervating fast-twitch (F nerves) and slow-twitch (S nerves) muscles of the rabbit differ both in their AChE molecular form patterns and in their anterograde and retrograde axonal transport parameters. Since we had previously shown a selective regulation of this enzyme in fast and slow parts of rabbit semimembranosus muscle, we wondered whether the differences observed in the nerve could be affected by the twitch properties of muscle. The results reported here show that in F nerves that reinnervate slow-twitch muscles, both the AChE molecular form patterns and axonal transport parameters turn into those of the S nerve. These data suggest the existence of a retrograde specific effect exerted by the muscles on their respective motoneurons.

Altered axonal transport of cytoskeletal proteins in the mutant diabetic mouse

Polypeptides in the motor axons of the sciatic nerve in 120-day-old normal and diabetic mice C57BL/Ks (db/db) were labeled by injection of [35S]methionine into the ventral horn of the spinal cord. At 8, 15, and 25 days after the injection, the distribution of radiolabeled polypeptides along the sciatic nerve was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Four major radiolabeled polypeptides, tentatively identified as actin, tubulin, and the two lightest subunits of the neurofilament triplet, were studied in both diabetic and control mice. In the diabetic animals, the two polypeptides identified as actin and tubulin showed a reduction of average velocity of migration along the sciatic nerve, resulting in a higher fraction of radioactivity in the proximal part of the sciatic nerve, whereas the front of radioactivity (advancing at maximal velocity) moved at a normal rate. In contrast, both the average and maximal velocities of the two neurofilament subunits were slower in the diabetic mice than in the control mice. These results indicate that the axonal transport of the cytoskeletal proteins is differentially affected in the course of diabetic neuropathy, and may suggest that the impairment concerns mainly the proteins carried by the slowest component of axonal transport.

Quantitative analysis of myelin and axolemma particle distribution in C57BL/Ks diabetic mice and the effects of ganglioside treatment

By freeze-fracture technique we estimated myelin and axolemma intramembranous particle density in C57BL/Ks mice. A decrease in myelin particle content compared to controls is present in both 180 and 280 day old genetic diabetic mice. In addition, the axolemma of myelinated axons is affected in interparanodal regions while no modification was detected at nodal level. Such alterations of myelin membrane structure may also be responsible for the lower motor nerve conduction velocity (MNCV) observed in these diabetic mice; however this hypothesis cannot be taken into consideration for the reduction in MNCV at the early stage of the neuropathy (prior to 180 days of life). Therefore the structural changes of both myelin sheath and interparanodal axolemma as visualized by freeze-fracture are most likely related to late complications of the disease instead of being responsible for the changes in excitability. The low myelin and axolemma particle density of diabetic mice was found normal after 30 days' treatment with gangliosides. Such findings are in agreement with previous results on a significant effect of ganglioside treatment on MNCV and axonal area alterations in 180 and 280 day old genetic diabetic mice.

Development of diabetic neuropathy in the C57BL/Ks (db/db) mouse and its treatment with gangliosides

We studied the development of diabetic neuropathy and its treatment with gangliosides using the C57BL/Ks mouse. The results of axonal morphometry showed the presence of a progressive axonal atrophy which was maximal at 180 days of age. To 400 days of age there was no longer any significant difference, perhaps due to aging processes. Nerve conduction velocity changed significantly from the early days of life. Thirty-day treatment with gangliosides significantly improved nerve conduction velocity and axonal morphometry at 180 and 280 days of life. No effect was observed with treatments at 30 or 60 days. It was previously shown that the early phase of the C57BL/Ks mouse neuropathy was reversed by insulin, whereas the late phase (180 days) was not. We showed elsewhere that at 180 days of age in the C57BL/Ks mouse there was a drastic decrease in slow transport of AChE (G1 and G2 molecular forms) indicating a shift in neuronal metabolism and suggesting that the disease was then more intrinsically neuronal. Using the suggestion of Robertson and Sima (Diabetes 29: 60-67, 1980) we label the first phase of the neuropathy "metabolic" (treatable with insulin) and the second phase "neuronal" (treatable with gangliosides). This "neuronal" phase could be related to the degenerative stage of human diabetic neuropathy.

Ganglioside treatment of genetic and alloxan-induced diabetic neuropathy

Peripheral neuropathy is a common complication of diabetes. Using the mutant diabetic mouse C57BL/ks (db/db) and alloxan-treated rats, 30 days after intoxication, we investigated development and treatment with gangliosides of such a disease. The db/db mouse develops a neuropathy characterized by a loss in conduction velocity shown as early as 80-90 days after birth and maintained throughout life. At later stages (5-6 months of age) there is a drop in slow transport and myelin particle density. These changes are correlated by a lack of response to insulin treatment, which, prior to this stage, is capable of improving nerve conduction velocity (NCV). On the other hand gangliosides became effective, improving NCV, myelin particle density and sensory perception (auditory deficit) at 5 months of age in the db/db mouse. We presume that this differential neuronal response to insulin and gangliosides indicates a change of the neuropathy from a metabolic stage to neuronal. Alloxan induced diabetic neuropathy is treatable with gangliosides even 30 days after intoxication.