2011 Diabetes  
The aim of this study was to evaluate whether urinary markers of tubular injury are predictors of early glomerular fi ltration rate (GFR) loss in diabetes. Our study group consisted of 255 Joslin subjects with type 1 diabetes, no proteinuria and normal GFR at baseline. This cohort was followed for 5-7 years and multiple determinations of serum cystatin C were performed to identify patients with signifi cant early GFR loss. As the outcome of the study rapid GFR loss during 5-7 years of follow-up
more » ... was defi ned as an annual GFR decrease >4%. Baseline urinary concentrations of tubular markers, N-acetyl-b-D-glucosaminidase (NAG) and Kidney Injury Molecule-1 (KIM1) were measured by nephelometry and on the Luminex platform, respectively. Rapid GFR loss occurred in 49 subjects (20%) and was more frequent in those who had higher concentrations of KIM1 in urine, p value for trend: 0.0001. Proportion of cases stratifi ed by KIM1 increasing quartiles is presented in the table below. KIM1-Q1 KIM1-Q2 KIM1-Q3 KIM1-Q4 GFR loss [%] 8 11 27 31 number at risk 64 63 64 64 Multivariate logistic analysis included baseline age, albumin excretion rate (AER), hemoglobin A1c and body mass index. Odds ratio of rapid GFR loss for subjects with KIM1 concentrations above median were OR (CI): 3.9 (1.9-7.9), p value 0.0002 in the univariate, and 2.7 (1.3-5.8), p value 0.009, in the multivariate model, respectively. Association of NAG with rapid GFR loss was borderline and did not remain signifi cant in the multivariate analysis. Urinary concentrations of KIM1, but not of NAG, predict early rapid GFR loss in subjects with type 1 diabetes. These fi ndings also provide strong evidence that tubular injury is involved in the disease process, which underlies the development of early progressive renal function loss. ADA-Funded Research Our results indicated that the Pro/Pro genotype and smoking were signifi cant independent risk factors for diabetic nephropathy. The possible synergistic effects of genotype and smoking may aggravate oxidative stress and contribute to the development of diabetic nephropathy. were exposed to 4% patient sera for 24 hours, and exposure to insulin (100ng/ ml) was utilized in selected experiments. Actin remodeling was evaluated by confocal images of phalloidin stained cells and western blotting for synaptopodin and RhoA. Cellular insulin sensitivity was measured as insulin ability to phosphorylate AKT. Apoptosis was determined by quantifi cation of cleaved caspase 3 and PARP. Changes in gene expression profi les were also evaluated using microarray analysis. Podocytes exposed to D-MA sera but not D-NA and NHC sera experienced actin cytoskeleton remodeling characterized by severe cell blebbing (90% of cells when compared to 20% in D-NA and 3% in NHC, p<0.001). Cell blebbing was associated with decreased synaptopodin and RhoA protein levels and increased cleaved caspase 3 and PARP (p<0.01) when compared to NHC. These fi ndings were associated with a loss of the ability of insulin to phosphorylate AKT (p=0.04 in D-MA versus NHC) and with the signifi cant modulation of several genes involved in cytokines action, actin remodeling and insulin signaling. In conclusion, our study suggests the presence of a circulating factor responsible for podocyte malfunction in patients with diabetes developing albuminuria. Lactobacillus species have been negatively correlated with diabetes onset. Recently the strains Lactobacillus reuteri TD1 and Lactobacillus johnsonii N6.2 were isolated from Biobreeding diabetes resistant rats (BB-DR). Feeding strain N6.2, but not strain TD1, to BioBreeding diabetes prone rats (BB-DP) slowed the onset of diabetes. Comparative genomic approaches were used to investigate the differences in these two genomes and how their unique characteristics relate to the onset of diabetes. Shotgun sequencing was performed using the Illumina Genome Analyzer IIx. De novo and reference assemblies were done using the CLC Genomics Workbench. Optical maps of both strains verifi ed the accuracy of the de novo assemblies. Strain TD1 was missing approximately 300 genes from each of the reference strains DSM 20016 and JCM 1112. Strain TD1 lacked genes in two pathways, glycerol metabolism and vitamin B12 synthesis. Both of these pathways are necessary for the production of reuterin, a broad-spectrum bacteriocin thought to contribute to the probiotic activity of L. reuteri strains. Strain N6.2 was missing 247 genes, compared to the reference strain F19785, but no whole pathways were missing or signifi cantly affected. However, in strain N6.2 known probiotic genes involved in bacteriocin production, such as those required for lactacin F and hydrogen peroxide production were found. Several stress resistance genes were also observed in strain N6.2, including those required for bile salt hydrolases. The probiotic factors present in strain N6.2 may contribute to its negative correlation with the onset of diabetes. The absence of reuterin production in strain TD1 may account for its ineffectiveness as a probiotic. Reuterin defi ciency in strain TD1 could be responsible for the differential effectiveness seen in the two strains in mitigating the onset of diabetes. The GENIE (GEnetics of Nephropathy: an International Effort) consortium was launched as a large study of diabetic nephropathy of 7,359 subjects with type 1 diabetes (T1D), with the aim to study the role of common genetic variation and the biological mechanisms contributing to diabetic nephropathy (DN). All subjects had T1D diagnosed before 35 years of age, were at least 18 years old and provided written consent. Our primary analysis evaluated the A4 For author disclosure information, see page 785. ADA-Funded Research ORALS GENETICS-TYPE 1, TYPE 2, AND COMPLICATIONS association of genetic loci with DN cases (n=2877), defi ned by the presence of macroalbuminuria or end-stage renal disease (ESRD). Controls in this analysis (n=3315) were individuals with no sign of DN despite 15 years of T1D. We further investigated the association of genetic loci with ESRD (n=1399) using different control group defi nitions. Genome-wide association scans (GWAS) with our standardized quality control (QC) procedures were conducted separately for FinnDiane (3370 samples, 549,530 SNPs) and the All Ireland-Warren 3-UK GoKinD collection (1687 samples, 791,687 SNPs). The same QC steps were also applied to the US GoKinD data retrieved from dbGAP (1595 samples, 360,899 SNPs). Imputation (MACH, HapMapII CEU) resulted in ∼2.4 million SNPs common to all cohorts. Association analysis was performed with PLINK and individual study results were meta-analyzed using METAL. The primary DN analysis revealed fi ve independent loci with P<10 -5 at 19q12, 20q11.21, 5q33.1, 4q34, and 2q33.3-q34. A genome-wide signifi cant (P<5x10 -8 ) association was detected in the AFF3 gene (P=4.8x10 -9 , OR 0.74) when cases with ESRD were compared with all other subjects. This fi nding is further supported by an in vitro model of DN, where AFF3 is downregulated in proximal tubular epithelial cells after TGF-β1 stimulation. Additionally, sexstratifi ed analysis of ESRD cases compared to normoalbuminuric controls revealed a signifi cant association at 2q31.1 in women (P=1.8x10 -8 , OR 1.9), with no association detected in men. Follow-up of our main fi ndings for DN and ESRD-related analyses is completed for 15% of the total 4450 planned samples. In summary, we have identifi ed loci associated with ESRD and expect to do likewise for DN in T1D. The interaction between the genetic and environmental factors that contribute to risk for cardiovascular complications among individuals with type 2 diabetes mellitus (T2DM) is not well understood. Selenoprotein S (SelS), one member of the selenprotein family, has recently been investigated for its association with a range of infl ammatory markers especially in the context of cardiovascular disease (CVD). The aim of the current study was to examine the role of SELS genetic variants in risk for CVD in T2DM. The association of 10 single nucleotide polymorphisms (SNPs) tagging the SELS gene was evaluated with subclinical CVD phenotypes including coronary, carotid and aortic calcifi ed plaque and intimal-medial thickness, C-reactive protein concentrations, and HbA1C in 1220 European Americans from the family-based Diabetes Heart Study (DHS). Of the 10 SNPs examined, four showed signifi cant association with at least one of the vascular calcifi cation phenotypes (p=0.0044-0.0362). The strongest evidence of association was found for rs34713741 and rs12917258 which were associated with coronary (p=1.25x10 -5 , β=-0.547 and p=3.79x10 -6 , β=-0.542 respectively), carotid (p=9.94x10 -6 , β=-0.561 and p=1.75x10 -7 , β=-0.626) and aortic calcifi ed plaque (p=1.41x10 -4 , β=-0.501 and p=5.68x10 -6 , β=-0.547) under an additive model of inheritance, as well as with both CRP (p=0.0279, β=-0.154 and p=0.0343, β=-0.142) and HbA1C (p=0.0022, β=-0.033 and p=6.45x10 -5 , β=-0.041). Further, these two SNPs were also associated with all-cause mortality (p=0.0279, β=-0.154 and p=0.0343, β=-0.142 respectively) over the follow up period of 7.3 ± 2.3 years (mean ± SD). These fi ndings support a role of SelS in contributing to CVD development in T2DM. Understanding the interplay between glycemic control, infl ammatory pathways and vascular calcifi cation may prove important in predicting and managing the macrovascular complications of T2DM. It is critical to understand the molecular mechanisms that underlie insulin resistance to develop new therapeutic approaches. We previously demonstrated that the mTORC1-S6K1 pathway is activated by insulin and nutrient excess (e.g. amino acids [AA]), which leads to the inhibition of the PI3K-Akt pathway via inhibitory serine phosphorylation of IRS-1, notably on serine 1101 (S1101). However, even in the absence of AA, insulin still promote IRS-1 S1101 phosphorylation by other kinases that remain to be fully characterized. We have now uncovered a new negative feedback loop arising from the MEK-ERK pathway, which activates p90 ribosomal S6 kinase (RSK). Using an antibody that detects a phosphorylation site (S221) located in the activation segment of RSK, we found that insulin activate both RSK1 and RSK3 in L6 myocytes even in the absence of AA overload. Computational analyses revealed that S1101 within IRS-1 falls into the consensus motif of RSK. Furthermore, RSK inhibition using either the pharmacological inhibitor BI-D1870 or after adenoviral expression of a dominant negative (DN) RSK1 mutant showed that RSK promotes the phosphorylation of IRS-1 on S1101 but not on S636/639. Accordingly, expression of the DN-RSK1 mutant also improved insulin action on muscle glucose transport and glucose production in hepatic cells. Importantly, RSK inhibition did not impede insulin's ability to activate the mTORC1-S6K1 pathway indicating that RSK does not promote S1101 phosphorylation by activation of this pathway. Finally, we found a close correlation between the activation of RSK and the phosphorylation of IRS-1 S1101 in high-fat fed obese insulin-resistant mice. We propose that RSK is a novel regulator of insulin signaling and glucose metabolism by promoting IRS-1 S1101 phosphorylation. This work could lead to new therapeutic strategies to alleviate insulin resistance in obese patients. This effect was age-dependent. Our results suggest that hyperinsulinemia cause IR by disrupting the Aktmediated pathway. Our results suggest a new theory for the etiology of diabetic neuropathy i.e. that, similar to metabolic tissues, neurons develop IR, and, in turn, cannot respond to the neurotrophic properties of insulin, resulting in neuronal injury and the development of neuropathy. Adipose tissue macrophage (ATM) accumulation through CCR2 and its ligand MCP-1 is considered pivotal in the development of insulin resistance. Recently, we demonstrated that loss of a different C-C chemokine receptor, CCR5, prevents insulin resistance induced by high fat (HF) feeding or leptin defi ciency independently of MCP-1/CCR2. Expression of mRNA for CCR5 and its all ligands (MIP-1α, MIP-1β, RANTES, MCP-2) was markedly increased in stromal vascular (SV) fraction compared to adipocyte fraction of epididymal white adipose tissue (eWAT) of genetically (ob/ob) and diet-induced obese (DIO) mice. Immunohistochemistry on eWAT in DIO localized CCR5 + cells to F4/80 + macrophages in crown-like structure. To quantify CCR5 + ATMs in lean and obese mice, fl uorescence-activated cell sorter (FACS) analysis was performed on SV cells isolated from eWAT. ATMs identifi ed as CD45 + CD11b + F4/80 + cells were increased in DIO mice by 12.2-fold compared with WT mice. Only a small percentage of ATMs coexpressed CCR5 in WT mice. However, DIO mice had 11.9-fold increase in CCR5 + cells within ATMs, indicating that CCR5 + macrophages accumulate in eWAT of obese mice. On a chow diet, no differences were observed in either CD11c + MGL1 -(M1) or CD11c -MGL1 + (M2) expression within ATMs from WT and Ccr5 -/mice. However, on a HF diet, in addition to reduction of total ATM content, Ccr5 -/mice had 39% fewer M1 ATMs whereas and 33% more M2 ATMs than WT mice, resulting in predominance of M2 over M1 ATM population. Importantly, chimeric mice lacking CCR5 only in myeloid cells (bone marrow transplant of Ccr5 -/into WT) were protected from HF diet-induced hyperinsulinemia and glucose intolerance. In conclusion, CCR5 defi ciency causes M2 dominant phenotypic shift in ATM, which contributes to attenuation of obesityinduced insulin resistance. Chronic infl ammation and accumulation of adipose tissue macrophages (ATMs) play a pivotal role for generation of insulin resistance in obesity. Forkhead box-containing protein O subfamily 1(Foxo1) has important roles for metabolic regulation in several insulin-responsive tissues. However, little is known about physiological roles of Foxo1 in ATMs. Foxo1 expression is signifi cantly increased in F4/80(+) CD11c (+) CD206 (-) macrophages in adipose tissues of mice under high fat-diet (HFD). Immunofl uorescence of ATMs using both anti-F4/80 and anti-Foxo1 antibodies revealed that nuclear localization of Foxo1 in ATMs decreased at 16-week and increased at 24week of HFD. In order to investigate roles of Foxo1 in ATMs, we generated macrophage-specifi c constitutively nuclear (CNFoxo1 LysM ) or transactivationdefective Foxo1 (Δ256), which has no C-terminal domain after amino acid 256, transgenic (Δ256 LysM ) mice using Rosa26-CNFoxo1 or Rosa26-Δ256 and M lysozyme Cre (LysM-Cre) mice. The CNFoxo1 LysM mice exhibited elevated insulin secretion during intraperitoneal glucose tolerance test, decreased insulin sensitivity in insulin tolerance test, and increased triglyceride accumulation in liver under HFD. FACS analysis revealed that F4/80(+) cells increased by about 50% and F4/80(+) CD11c (+) CD206 (-) cells increased by about 100% in stromal vascular fraction (SVF) from CNFoxo1 LysM compared with control mice. Analysis of gene expression of markers of M1 and M2 macrophages in SVF from epididymal fat and in ATMs collected by FACS revealed that Ccr2 expression level was signifi cantly increased in CNFoxo1 LysM mice. Transduction of adenovirus encoding CNFoxo1 in RAW264.7 cells also increased Ccr2 expression. Luciferase assay demonstrated that CNFoxo1 increased Ccr2 promoter activity and chromatin immunoprecipitation (ChIP) assay also demonstrated that Foxo1 bound to Ccr2 promoter. In contrast, the Δ256 LysM mice exhibited signifi cantly improved glucose tolerance and increased insulin sensitivity compared with control mice under HFD. These data suggest that Foxo1 induces Ccr2 expression in ATMs, increases infi ltration of macrophages to adipose tissue, and causes insulin resistance. Chronic infl ammation mediated by NF-κB may contribute to the pathogenesis of type 2 diabetes (T2D). Salicylates inhibit NF-κB activity and improve glycemia in T2D, although effects on insulin resistance are uncertain. 70 predominantly male veterans with IFG and/or IGT (mean age 58±8 yrs, BMI 32±6 kg/m 2 , A1c 6.0±0.3%) were randomized in a double-masked 12 week trial of salsalate (SAL) up to 4.0 g/d vs placebo (P) to examine effects on glucose metabolism in "prediabetes". Baseline characteristics were generally similar between groups. The mean SAL dose was 3.7 ± 0.5 g/d (reduced for tinnitus in 32% SAL and 17% P treated patients). SAL signifi cantly lowered fasting glucose (FG) by 8% (Figure) and total OGTT glucose area under the curve (AUC) by 5%, (p=0.05). Weight did not change during the study. No signifi cant changes in fasting or AUC insulin, or AUC C-peptide occurred, although fasting C-peptide was reduced 20% (p<0.01), suggesting that FG was maintained in the setting of reduced insulin secretion. Notably, reductions in FG were strongly and inversely correlated (r= -0.64, p=0.004; Figure) with average serum salicylate levels. Glucose utilization (M) and insulin levels during euglycemic-hyperinsulinemic clamps were similar in SAL and P groups at baseline and did not change with therapy in either group; however, there was a modest positive association between end of study M values and salicylate levels. In summary, SAL lowers FG and C-peptide in patients with IFG/IGT, possibly indicating reduced beta cell burden. The modest effects on M suggest that improvement in FG may be due to reduced endogenous glucose production. Higher salicylate concentrations are associated with greater declines in FG, and may modestly enhance glucose utilization. Targeting infl ammation may provide a new strategy for diabetes prevention. A16 For author disclosure information, see page 785. ADA-Funded Research ORALS OBESITY-HUMAN craving in the stress condition correlated positively with activation in the caudate, putamen, insula, and parahippocampus; conversely, in NW subjects, food craving correlated inversely with activation in the caudate, putamen, and insula (p<.05). We conclude that in response to food cue and stress exposure, OB individuals have increased activation in limbic-striatal neural pathways; furthermore, activation in these regions is correlated with food craving. These fi ndings suggest that food craving in OB individuals is linked to the activation of brain regions underlying motivation and emotion, and such activation may in turn contribute to over consumption of highly desirable foods. Recombinant human hyaluronidase (rHuPH20) is FDA-approved to increase the dispersion and absorption of other injected drugs. rHuPH20 accelerates the exposure to and action of RHI and rapid insulin analogs. We compared glucose control in well-controlled T1DM using insulin lispro or RHI+rHuPH20 (INSULIN-PH20) as prandial insulin. After a 1 month run-in using bid insulin glargine and prandial lispro, 46 patients (age 42±13 years, BMI 26±4 kg/m 2 , A1C=6.9±0.6 and 24 males) were randomized in an open-label crossover
doi:10.2337/db11-1-378 fatcat:x6jz522funauvb7ajpgeatiu7e