INPP5K controls the dynamic structure and signaling of wild-type and mutated, leukemia-associated IL-7 receptors.

The downstream signaling of the interleukin-7 (IL-7) receptor (IL-7R) plays important physiological and pathological roles, including the differentiation of lymphoid cells and proliferation of acute lymphoblastic leukemia cells. Gain-of-function mutations in the IL-7Rα chain, the specific component of the receptor for IL-7, result in constitutive, IL-7-independent signaling and trigger acute lymphoblastic leukemia. Here, we show that the loss of the phosphoinositide 5-phosphatase INPP5K is associated with increased levels of the INPP5K substrate phosphatidylinositol 4,5-bisphosphate (PtdIns[4,5]P2) and causes an altered dynamic structure of the IL-7 receptor. We discovered that the IL-7Rα chain contains a very conserved positively charged polybasic amino acid sequence in its cytoplasmic juxtamembrane region; this region establish stronger ionic interactions with negatively charged PtdIns(4,5)P2 in the absence of INPP5K, freezing the IL-7Rα chain structure. This dynamic structural alteration causes defects in IL-7R signaling, culminating in decreased expressions of EBF1 and PAX5 transcription factors, in microdomain formation, cytoskeletal reorganization, and bone marrow B-cell differentiation. Similar alterations after the reduced INPP5K expression also affected mutated, constitutively activated IL-7Rα chains that trigger leukemia development, leading to reduced cell proliferation. Altogether, our results indicate that the lipid 5-phosphatase INPP5K hydrolyzes PtdIns(4,5)P2, allowing the requisite conformational changes of the IL-7Rα chain for optimal signaling.

Altered chondrocyte differentiation, matrix mineralization and MEK-Erk1/2 signaling in an INPPL1 catalytic knock-out mouse model of opsismodysplasia.

Opsismodysplasia (OPS) is a rare but severe autosomal recessive skeletal chondrodysplasia caused by inactivating mutations in the Inppl1/Ship2 gene. The molecular mechanism leading from Ship2 gene inactivation to OPS is currently unknown. Here, we used our Ship2Δ/Δ mouse expressing reduced amount of a catalytically-inactive SHIP2 protein and a previously reported SHIP2 inhibitor to investigate growth plate development and mineralization in vivo, ex vivo and in vitro. First, as observed in OPS patients, catalytic inactivation of SHIP2 in mouse leads to reduced body length, shortening of long bones, craniofacial dysmorphism, reduced height of the hyperthrophic chondrocyte zone and to defects in growth plate mineralization. Second, intrinsic Ship2Δ/Δ bone defects were sufficient to induce the characteristic OPS alterations in bone growth, histology and mineralization ex vivo. Third, expression of osteocalcin was significantly increased in SHIP2-inactivated chondrocyte cultures whereas production of mineralized nodules was markedly decreased. Targeting osteocalcin mRNA with a specific shRNA increased the production of mineralized nodules. Fourth, levels of p-MEK and p-Erk1/2 were significantly increased in SHIP2-inactivated chondrocytes in response to serum and IGF-1, but not to FGF2, as compared to control chondrocytes. Treatment of chondrocytes and bones in culture with a MEK inhibitor partially rescued the production of mineralized nodules, the size of the hypertrophic chondrocyte zone and bone growth, raising the possibility of a treatment that could partially reduce the phenotype of this severe condition. Altogether, our results indicate that Ship2Δ/Δ mice represent a relevant model for human OPS. They also highlight the important role of SHIP2 in chondrocytes during endochondral ossification and its different differentiation steps. Finally, we identified a role of osteocalcin in mineralized nodules production and for the MEK-Erk1/2 signaling pathway in the OPS phenotype.

The lipid 5-phoshatase SHIP2 controls renal brush border ultrastructure and function by regulating the activation of ERM proteins.

The microvillus brush border on the renal proximal tubule epithelium allows the controlled reabsorption of solutes that are filtered through the glomerulus and thus participates in general body homeostasis. Here, using the lipid 5-phosphatase Ship2 global knockout mice, proximal tubule-specific Ship2 knockout mice, and a proximal tubule cell model in which SHIP2 is inactivated, we show that SHIP2 is a negative regulator of microvilli formation, thereby controlling solute reabsorption by the proximal tubule. We found increased PtdIns(4,5)P2 substrate and decreased PtdIns4P product when SHIP2 was inactivated, associated with hyperactivated ezrin/radixin/moesin proteins and increased Rho-GTP. Thus, inactivation of SHIP2 leads to increased microvilli formation and solute reabsorption by the renal proximal tubule. This may represent an innovative therapeutic target for renal Fanconi syndrome characterized by decreased reabsorption of solutes by this nephron segment.

Model Matters: Differences in Orthotopic Rat Hepatocellular Carcinoma Physiology Determine Therapy Response to Sorafenib.

PURPOSE: Preclinical model systems should faithfully reflect the complexity of the human pathology. In hepatocellular carcinoma (HCC), the tumor vasculature is of particular interest in diagnosis and therapy. By comparing two commonly applied preclinical model systems, diethylnitrosamine induced (DEN) and orthotopically implanted (McA) rat HCC, we aimed to measure tumor biology noninvasively and identify differences between the models. EXPERIMENTAL DESIGN: DEN and McA tumor development was monitored by MRI and PET. A slice-based correlation of imaging and histopathology was performed. Array CGH analyses were applied to determine genetic heterogeneity. Therapy response to sorafenib was tested in DEN and McA tumors. RESULTS: Histologically and biochemically confirmed liver damage resulted in increased (18)F-fluorodeoxyglucose (FDG) PET uptake and perfusion in DEN animals only. DEN tumors exhibited G1-3 grading compared with uniform G3 grading of McA tumors. Array comparative genomic hybridization revealed a highly variable chromosomal aberration pattern in DEN tumors. Heterogeneity of DEN tumors was reflected in more variable imaging parameter values. DEN tumors exhibited lower mean growth rates and FDG uptake and higher diffusion and perfusion values compared with McA tumors. To test the significance of these differences, the multikinase inhibitor sorafenib was administered, resulting in reduced volume growth kinetics and perfusion in the DEN group only. CONCLUSIONS: This work depicts the feasibility and importance of in depth preclinical tumor model characterization and suggests the DEN model as a promising model system of multifocal nodular HCC in future therapy studies.

The Ras/Rap GTPase activating protein RASA3: from gene structure to in vivo functions.

RASA3 (or GTPase Activating Protein III, R-Ras GTPase-activating protein, GAP1(IP4BP)) is a GTPase activating protein of the GAP1 subfamily which targets Ras and Rap1. RASA3 was originally purified from pig platelet membranes through its intrinsic ability to bind inositol 1,3,4,5-tetrakisphosphate (I(1,3,4,5)P4) with high affinity, hence its first name GAP1(IP4BP) (for GAP1 subfamily member which binds I(1,3,4,5)P4). RASA3 was thus the first I(1,3,4,5)P4 receptor identified and cloned. The in vitro and in vivo functions of RASA3 remained somewhat elusive for a long time. However, recently, using genetically-modified mice and cells derived from these mice, the function of RASA3 during megakaryopoiesis, megakaryocyte adhesion and migration as well as integrin signaling has been reported. The goal of this review is thus to summarize and comment recent and less recent data in the literature on RASA3, in particular on the in vivo function of this specific GAP1 subfamily member.

Mdm2 promotes systemic lupus erythematosus and lupus nephritis.

Systemic lupus erythematosus (SLE) is a polyclonal autoimmune syndrome directed against multiple nuclear autoantigens. Although RNA and DNA seem to have identical immunostimulatory effects on systemic and intrarenal inflammation, each seems to differ with regard to the propensity to induce mitogenic effects such as lymphoproliferation. To identify potential mechanisms by which DNA specifically contributes to the pathogenesis of lupus nephritis, we stimulated cells with immunostimulatory DNA or RNA in vitro and used microarray to compare the transcriptomes of RNA- and DNA-induced genes. Immunostimulatory DNA, but not RNA, induced Mdm2, which is a negative regulator of p53. In vivo, we observed greater expression and activation of Mdm2 in the spleen and kidneys in a mouse model of lupus (MRL-Fas(lpr) mice) than healthy controls. Treatment of MRL-Fas(lpr) mice with the Mdm2 inhibitor nutlin-3a prevented nephritis and lung disease and significantly prolonged survival. Inhibition of Mdm2 reduced systemic inflammation and abrogated immune complex disease by suppressing plasma cells and the production of lupus autoantibodies. In addition, nutlin-3a suppressed the abnormal expansion of all T cell subsets, including CD3(+)CD4(-)CD8(-) T cells, which associated with attenuated systemic inflammation. However, inhibiting Mdm2 did not cause myelosuppression or affect splenic regulatory T cells, neutrophils, dendritic cells, or monocytes. Taken together, these data suggest that the induction of Mdm2 promotes the expansion of plasma cells and CD3(+)CD4(-)CD8(-) T cells, which cause autoantibody production and immune complex disease in MRL-Fas(lpr) mice. Antagonizing Mdm2 may have therapeutic potential in lupus nephritis.

Dual blockade of the homeostatic chemokine CXCL12 and the proinflammatory chemokine CCL2 has additive protective effects on diabetic kidney disease.

Monocyte/ chemoattractant protein-1/chemokine ligand (CCL) 2 and stromal cell-derived factor-1/CXCL12 both contribute to glomerulosclerosis in mice with type 2 diabetes mellitus, through different mechanisms. CCL2 mediates macrophage-related inflammation, whereas CXCL12 contributes to podocyte loss. Therefore, we hypothesized that dual antagonism of these chemokines might have additive protective effects on the progression of diabetic nephropathy. We used chemokine antagonists based on structured l-enantiomeric RNA (so-called Spiegelmers) ie, the CCL2-specific mNOX-E36 and the CXCL12-specific NOX-A12. Male db/db mice, uninephrectomized at the age of 6 weeks, received injections of Spiegelmer, both Spiegelmers, nonfunctional control Spiegelmer, or vehicle from the age of 4 months for 8 weeks. Dual blockade was significantly more effective than monotherapy in preventing glomerulosclerosis. CCL2 blockade reduced glomerular leukocyte counts and renal-inducible nitric oxide synthase or IL-6 mRNA expression. CXCL12 blockade maintained podocyte numbers and renal nephrin and podocin mRNA expression. Consistently, CXCL12 blockade suppressed nephrin mRNA up-regulation in primary cultures of human glomerular progenitors induced to differentiate toward the podocyte lineage. All previously mentioned parameters were significantly improved in the dual-blockade group, which also suppressed proteinuria and was associated with the highest levels of glomerular filtration rate. Blood glucose levels and body weight were identical in all treatment groups. Dual chemokine blockade can have additive effects on the progression of diabetic kidney disease when the respective chemokine targets mediate different pathomechanisms of disease (ie, inflammation and progenitor differentiation toward the podocyte lineage).

An orally active chemokine receptor CCR2 antagonist prevents glomerulosclerosis and renal failure in type 2 diabetes.

The progression of diabetic nephropathy is associated with an infiltration of macrophages expressing different phenotypes. As classically activated chemokine receptor CCR2+ macrophages are thought to drive tissue inflammation and remodeling, we tested whether blocking CCR2 could reduce intrarenal inflammation and prevent glomerulosclerosis in type 2 diabetes. This was achieved with RO5234444, an orally active small-molecule CCR2 antagonist that blocks ligand binding, its internalization, and monocyte chemotaxis. Male type 2 diabetic db/db mice were uninephrectomized to increase glomerular hyperfiltration to accelerate the development of glomerulosclerosis. From 16 weeks until killing at 24 weeks of age, mice were chow fed with or without admixed antagonist to achieve a trough plasma concentration above IC50 for binding in the mouse. CCR2 blockade reduced circulating monocyte levels, but did not affect total leukocyte or neutrophil numbers, and was associated with a reduction in the number of macrophages and apoptotic podocytes in the glomerulus. This treatment resulted in a higher total number of podocytes, less glomerulosclerosis, reduced albuminuria, and a significantly improved glomerular filtration rate. This successful pre-clinical trial suggests that this antagonist may now be ready for testing in humans with the nephropathy of diabetes mellitus.

4SC-101, a novel small molecule dihydroorotate dehydrogenase inhibitor, suppresses systemic lupus erythematosus in MRL-(Fas)lpr mice.

Immunosuppressive treatments of systemic lupus (SLE) remain associated with significant toxicities; hence, compounds with better toxicity profiles are needed. Dihydroorotate dehydrogenase (DHODH) inhibition with leflunomide has proven to be effective in autoimmune diseases including SLE, but leflunomide can cause a variety of side effects. We hypothesized that 4SC-101, a novel DHODH inhibitor with a more favorable toxicity profile, would be as effective as high-dose cyclophosphamide (CYC) in controlling experimental SLE of female MRL(Fas)lpr mice. Daily oral gavage of 30, 100, and 300 mg/kg 4SC-101 from 12 to 22 weeks of age was compared with either vehicle or CYC treatment (30 mg/kg/week, i.p.) in terms of efficacy and toxicity. Three hundred milligrams per kilogram 4SC-101 was as effective as CYC in depleting spleen autoreactive T cells, B cells, and plasma cells as well as the respective DNA and RNA serum autoantibodies. This was associated with a comparable amelioration of the renal, dermal, and pulmonary SLE manifestations of MRL(Fas)lpr mice. However, even the highest dose of 4SC-101 had no effect on bone marrow neutrophil counts, which were significantly reduced in CYC-treated mice. Together, the novel DHODH inhibitor 4SC-101 is as effective as high dose CYC in controlling SLE without causing myelosuppression. Hence, DHODH inhibition with 4SC-101 might be suitable to treat active SLE with fewer side effects than CYC.

Progressive glomerulosclerosis in type 2 diabetes is associated with renal histone H3K9 and H3K23 acetylation, H3K4 dimethylation and phosphorylation at serine 10.

BACKGROUND: Distinct histone modifications regulate gene expression in certain diseases but little is known about histone epigenetics in diabetic nephropathy. The current study examined the role of histone epigenetics in development and progression of nephropathy in db/db mice. METHODS: We studied kidney damage in 6-month-old non-diabetic mice and type 2 diabetic db/db mice that underwent either sham surgery or uninephrectomy at 6 weeks of age which accelerates glomerulosclerosis in db/db mice via glomerular hyperfiltration. Histone H3K9 and H3K23 acetylation, H3K4 and H3K9 dimethylation and H3 phosphorylation at serine 10 was explored by western blotting of renal histone extracts. RESULTS: Uninephrectomy in C57BL/6 mice or onset of diabetes in type 2 diabetes reduced renal H3K23 acetylation, H3K4 dimethylation and H3 phosphorylation at serine 10. In contrast, H3K9 and H3K23 acetylation, H3K4 dimethylation and H3 phosphorylation at serine 10 were significantly increased in uninephrectomized db/db mice. The disease pattern of these mice is characterized by an increased glomerular cell proliferation, severe glomerulosclerosis, albuminuria and glomerular filtration rate reduction. Treating uninephrectomized db/db mice with a Mcp-1/Ccl2 antagonist prevented the histopathological damage and the aforementioned histone modification abnormalities of advanced diabetic glomerulosclerosis. CONCLUSION: We conclude that advanced diabetic nephropathy is associated with increased renal H3K9 and H3K23 acetylation, H3K4 dimethylation and H3 phosphorylation at serine 10 that enhance chromatin unfolding and gene expression.

Renal failure increases cardiac histone h3 acetylation, dimethylation, and phosphorylation and the induction of cardiomyopathy-related genes in type 2 diabetes.

The combination of diabetes and renal failure is associated with accelerated cardiomyopathy, but the molecular mechanisms of how renal failure drives diabetic heart disease remain elusive. We speculated that the metabolic abnormalities of renal failure will affect the epigenetic control of cardiac gene transcription and sought to determine the histone H3 modification pattern in hearts of type 2 diabetic mice with several degrees of renal dysfunction. We studied the histone H3 modifications and gene expression in the heart of 6-month-old nondiabetic mice and type 2 diabetic db/db mice that underwent either sham surgery or uninephrectomy at 6 weeks of age, which accelerates glomerulosclerosis in db/db mice via glomerular hyperfiltration. Western blotting of hearts from uninephrectomized db/db mice with glomerulosclerosis, albuminuria, and reduced glomerular filtration rate revealed increased acetylation (K23 and 9), phosphorylation (Ser 10), dimethylation (K4), and reduced dimethylation of (K9) of cardiac histone H3 as compared with db/db mice with normal renal function or nondiabetic wild-type mice. This pattern suggests alterations in chromatin structure that favor gene transcription. In fact, hearts from uninephrectomized db/db mice revealed increased mRNA expression of multiple cardiomyopathy-related genes together with cardiomyocyte hypertrophy. These data suggest that renal failure alters cardiac histone H3 epigenetics, which foster cardiomyocyte hypertrophy in type 2 diabetes.

Questions about chemokine and chemokine receptor antagonism in renal inflammation.

Chemokines remain attractive therapeutic targets for modulating inflammatory diseases in all areas of medicine including acute and chronic kidney disease. Industry has launched huge programs for the development of chemokine antagonists, and clinical trials with chemokine and chemokine receptor antagonists are ongoing. However, chemokine biology remains an area of unexpected discoveries. Here we discuss a number of questions which need to be addressed to further explore the potential of chemokine antagonism in renal inflammation: Why does renal expression of chemokines and chemokine receptors not always correlate with their functional significance? Why does chemokine antagonism only partially reduce renal leukocyte counts? Will antagonist combinations be more effective in reducing renal inflammation? What are the functional roles of homeostatic chemokines and atypical, nonsignaling chemokine receptors in renal inflammation? And finally, what classes of chemokine antagonists are available to address these questions experimentally?

18F9 (4-(3,6-bis (ethoxycarbonyl)-4,5,6,7-tetrahydrothieno (2,3-c) pyridin-2-ylamino)-4-oxobutanoic acid) enhances insulin-mediated glucose uptake in vitro and exhibits antidiabetic activity in vivo in db/db mice.

Insulin resistance is central to the pathogenesis of type 2 diabetes mellitus. Previous studies have demonstrated that compounds that cause adipogenesis and improve glucose uptake in 3T3-L1 cells are potential insulin sensitizers. Therefore, we evaluated one such compound, 18F9, for (1) adipogenesis in human subcutaneous preadipocyte (SQ) cells, (2) glucose uptake in human skeletal muscle myotubes and SQ cells, and (3) antidiabetic activity in db/db mice. We also investigated its effect on ex vivo glucose uptake in soleus muscle isolated from continuously treated db/db mice. Gene expression profiling in soleus muscle and epididymal fat of db/db mice was performed to understand its effect on glucose metabolism, lipid metabolism, and thermogenesis. 18F9 enhanced adipogenesis in SQ cells and increased glucose uptake in SQ and human skeletal muscle myotubes cells. In db/db mice, 18F9 exhibited dose-dependent reduction in plasma glucose and insulin level. Interestingly, 18F9 was as efficacious as rosiglitazone but did not cause body weight gain and hepatic adverse effects. In addition, 18F9 demonstrated no change in plasma volume in Wistar rats. Furthermore, it enhanced ex vivo glucose uptake in soleus muscles in these mice, which substantiates our in vitro findings. Human peroxisome proliferator activated receptor-gamma transactivation assay revealed a weak peroxisome proliferator activated receptor-gamma transactivation potential (44% of rosiglitazone at 10 mumol/L) of 18F9. Gene expression profiling indicated that 18F9 increased insulin sensitivity mainly through a phosphoinositide 3-kinase-dependent mechanism. 18F9 also up-regulated genes involved in lipid transport and synthesis at par with rosiglitazone. Unlike rosiglitazone, 18F9 elevated the expression of Pdk4. In addition, 18F9 elevated the expression of glycogen synthase and adiponectin significantly higher than rosiglitazone. Taken together, these observations suggest that 18F9 is a safer and potent insulin sensitizer that demonstrates promising antidiabetic activity and is worth further development.

Acute administration of GPR40 receptor agonist potentiates glucose-stimulated insulin secretion in vivo in the rat.

Recently, several in vitro studies have shown that GPR40 receptor activation by free fatty acids (FFAs) results in glucose-dependent insulin secretion. However, whether GPR40 receptor activation results in glucose-dependent insulin secretion in vivo in rats is not known. Therefore, we evaluated the effect of synthetic GPR40 receptor agonist (compound 1) on glucose tolerance test (GTT) in fed, fasted, and insulin-resistant rats. In oral GTT, intraperitoneal GTT, and intravenous GTT, GPR40 receptor agonist improved glucose tolerance, which was associated with increase in plasma insulin level. Interestingly, in GTTs, the rise in insulin levels in agonist-treated group was directly proportional to the rate of rise and peak levels of glucose in control group. Although glibenclamide, a widely used insulin secretagogue, improved glucose tolerance in all GTTs, it did not display insulin release in intraperitoneal GTT or intravenous GTT. In the absence of glucose load, GPR40 receptor agonist did not significantly change the plasma insulin concentration, but did decrease the plasma glucose concentration. Fasted rats exhibited impaired glucose-stimulated insulin secretion (GSIS) as compared with fed rats. Compound 1 potentiated GSIS in fasted state but failed to do so in fed state. Suspecting differential pharmacokinetics, a detailed pharmacokinetic evaluation was performed, which revealed the low plasma concentration of compound 1 in fed state. Consequently, we examined the absorption profile of compound 1 at higher doses in fed state; and at a dose at which its absorption was comparable with that in fasted state, we observed significant potentiation of GSIS. Chronic high-fructose (60%) diet feeding resulted in impaired glucose tolerance, which was improved by GPR40 receptor agonist. Therefore, our results demonstrate for the first time that acute GPR40 receptor activation leads to potentiation of GSIS in vivo and improves glucose tolerance even in insulin-resistant condition in rats. Taken together, these results suggest that GPR40 receptor agonists could be potential therapeutic alternatives to sulfonylureas.

Effects of trolox on nerve dysfunction, thermal hyperalgesia and oxidative stress in experimental diabetic neuropathy.

1. Diabetic neuropathy is one of the most common complications of diabetes and oxidative stress has been implicated to play a major role in its pathophysiology. 2. In the present study, we targeted oxidative stress using trolox, an anti-oxidant, in streptozotocin-induced diabetic neuropathy in rats. 3. Compared with control rats, diabetic rats showed significant deficits in motor nerve conduction velocity (MNCV; 49.91 +/- 1.94 vs 42.77 +/- 1.39 m/s, respectively) and nerve blood flow (NBF; 107.98 +/- 8.22 vs 38.9 +/- 2.7 arbitarary perfusion units, respectively) after 8 weeks of diabetes. Tail flick latencies for cold and hot immersion tests were also significantly reduced in diabetic rats, indicating thermal hyperalgesia. These observations indicate development of diabetic neuropathy. 4. A significant decrease in the activity of anti-oxidant enzymes (superoxide dismutase and catalase) and an increase in lipid peroxidation were observed in sciatic nerves from diabetic rats compared with age-matched control rats. Alterations in the activity of anti-oxidant enzymes and lipid peroxidation in diabetic rats indicate oxidative stress in diabetic neuropathy. 5. Two weeks treatment with trolox (10 and 30 mg/kg, i.p.) started on completion of the 6th week of diabetes significantly improved MNCV, NBF and inhibited thermal hyperalgesia. Trolox treatment also improved the activity of anti-oxidant enzymes and inhibited lipid peroxidation in sciatic nerves of diabetic rats. 6. The results of the present study suggest the beneficial effects of trolox in experimental diabetic neuropathy.

Effects of U83836E on nerve functions, hyperalgesia and oxidative stress in experimental diabetic neuropathy.

Oxidative stress has been implicated to play an important role in the pathogenesis of diabetic neuropathy, which is the most common complication of diabetes mellitus affecting more than 50% of diabetic patients. In the present study, we have investigated the effect of U83836E [(-)-2-((4-(2,6-Di-1-pyrrolidinyl-4-pyrimidinyl)-1-piperazinyl)methyl)-3,4-dihydro-2,3,7,8-tetramethyl-2H-1-benzopyran-6-ol, 2HCl], a potent free radical scavenger in streptozotocin (STZ)-induced diabetic neuropathy in rats. STZ-induced diabetic rats showed significant deficit in motor nerve conduction velocity (MNCV), nerve blood flow (NBF) and thermal hyperalgesia after 8 weeks of diabetes induction, indicating development of diabetic neuropathy. Antioxidant enzyme (superoxide dismutase and catalase) levels were reduced and malondialdehyde (MDA) levels were significantly increased in diabetic rats as compared to the age-matched control rats, this indicates the involvement of oxidative stress in diabetic neuropathy. The 2-week treatment with U83836E (3 and 9 mg/kg, i.p.) started 6 weeks after diabetes induction significantly ameliorated the alterations in MNCV, NBF, hyperalgesia, MDA levels and antioxidant enzymes in diabetic rats. Results of the present study suggest the potential of U83836E in treatment of diabetic neuropathy.