Few studies have demonstrated reproducible gene-diet interactions (GDIs) impacting metabolic disease risk factors, likely due in part to measurement error in dietary intake estimation and insufficient capture of rare genetic variation. We aimed to identify GDIs across the genetic frequency spectrum impacting the macronutrient-glycemia relationship in genetically and culturally diverse cohorts. We analyzed 33,187 participants free of diabetes from 10 National Heart, Lung, and Blood Institute Trans-Omics for Precision Medicine program cohorts with whole-genome sequencing, self-reported diet, and glycemic trait data. We fit cohort-specific, multivariable-adjusted linear mixed models for the effect of diet, modeled as an isocaloric substitution of carbohydrate for fat, and its interactions with common and rare variants genome-wide. In main effect meta-analyses, participants consuming more carbohydrate had modestly lower glycemic trait values (e.g., for glycated hemoglobin [HbA1c], -0.013% HbA1c/250 kcal substitution). In GDI meta-analyses, a common African ancestry-enriched variant (rs79762542) reached study-wide significance and replicated in the UK Biobank cohort, indicating a negative carbohydrate-HbA1c association among major allele homozygotes only. Simulations revealed that >150,000 samples may be necessary to identify similar macronutrient GDIs under realistic assumptions about effect size and measurement error. These results generate hypotheses for further exploration of modifiable metabolic disease risk in additional cohorts with African ancestry.

High-throughput proteomics allows researchers to simultaneously explore the roles of thousands of biomarkers in the pathophysiology of diabetes. We conducted proteomic association studies of incident type 2 diabetes and physiologic responses to an intravenous glucose tolerance test (IVGTT) to identify novel protein contributors to glucose homeostasis and diabetes risk. We tested 4,776 SomaScan proteins measured in relation to 18-year incident diabetes risk in participants from the Cardiovascular Health Study (N = 2,631) and IVGTT-derived measures in participants from the HERITAGE Family Study (N = 752). We characterize 51 proteins that were associated with longitudinal diabetes risk, using their respective 39, 9, and 8 concurrent associations with insulin sensitivity index (SI), acute insulin response to glucose (AIRG), and glucose effectiveness (SG). Twelve of the 51 diabetes associations appear to be novel, including β-glucuronidase, which was associated with increased diabetes risk and lower SG, suggesting an alternative pathway to insulin for glucose disposal; and plexin-B2, which also was associated with increased diabetes risk, but with lower AIRG, and not with SI, indicating a mechanism related instead to pancreatic dysfunction. Other novel protein associations included alcohol dehydrogenase-1C, fructose-bisphosphate aldolase-B, sorbitol dehydrogenase with elevated type 2 diabetes risk, and a leucine-rich repeat containing protein-15 and myocilin with decreased risk.

Chronic hyperglycemia increases pancreatic β-cell metabolic activity, contributing to glucotoxicity-induced β-cell failure and loss of functional β-cell mass, potentially in multiple forms of diabetes. In this perspective we discuss the novel paradoxical and counterintuitive concept of inhibiting glycolysis, particularly by targeted inhibition of glucokinase, the first enzyme in glycolysis, as an approach to maintaining glucose sensing and preserving functional β-cell mass, thereby improving insulin secretion, in the treatment of diabetes.

In this study we examine the instrument selection strategies currently used throughout the type 2 diabetes and HbA1c Mendelian randomization (MR) literature. We then argue for a more integrated and thorough approach, providing a framework to do this in the context of HbA1c and diabetes. We conducted a literature search for MR studies that have instrumented diabetes and/or HbA1c. We also used data from the UK Biobank (UKB) (N = 349,326) to calculate instrument strength metrics that are key in MR studies (the F statistic for average strength and R2 for total strength) with two different methods ("individual-level data regression" and Cragg-Donald formula). We used a 157-single nucleotide polymorphism (SNP) instrument for diabetes and a 51-SNP instrument (with partition into glycemic and erythrocytic as well) for HbA1c. Our literature search yielded 48 studies for diabetes and 22 for HbA1c. Our UKB empirical examples showed that irrespective of the method used to calculate metrics of strength and whether the instrument was the main one or included partition by function, the HbA1c genetic instrument is strong in terms of both average and total strength. For diabetes, a 157-SNP instrument was shown to have good average strength and total strength, but these were both substantially lesser than those of the HbA1c instrument. We provide a careful set of five recommendations to researchers who wish to genetically instrument type 2 diabetes and/or HbA1c. In MR studies of glycemia, investigators should take a more integrated approach when selecting genetic instruments, and we give specific guidance on how to do this.

Immune checkpoint inhibitors (ICIs) could cause type 1 diabetes (T1D). However, the underlying mechanism remains unclear. We immunohistochemically analyzed pancreatic specimens from three individuals with ICI-related T1D, and their histopathological data were compared those from three patients who had received ICI therapy but did not develop T1D (non-T1D) and seven normal glucose-tolerant subjects as control subjects. All ICI-related T1D patients had susceptible HLA haplotypes. In ICI-related T1D, the β-cell area decreased and the α-cell area increased compared with non-T1D and control subjects. The number of CD3-positive cells around islets increased in ICI-related T1D and non-T1D compared with control subjects, while the number of CD68-positive cells around islets increased in ICI-related T1D compared with non-T1D and control subjects. The expression ratios of programmed death-ligand 1 (PD-L1) on islets decreased in non-T1D and almost completely disappeared in ICI-related T1D, while PD-L1 expression was observed in most cells of pancreatic islets in control subjects. This study, therefore, indicates that ICI therapy itself could reduce PD-L1 expression on islets in all subjects, which may be related to β-cell vulnerability. In addition, we showed that absence of PD-L1 expression on β-cells, genetic susceptibility, and infiltration of macrophages as well as T lymphocytes around islets might be responsible for T1D onset.

Obesity is postulated to independently increase chronic kidney disease (CKD), even after adjusting for type 2 diabetes (T2D) and hypertension. Dysglycemia below T2D thresholds, frequently seen with obesity, also increases CKD risk. Whether obesity increases CKD independent of dysglycemia and hypertension is unknown and likely influences the optimal weight loss (WL) needed to reduce CKD. T2D remission rates plateau with 20-25% WL after bariatric surgery (BS), but further WL increases normoglycemia and normotension. We undertook bidirectional inverse variance weighted Mendelian randomization (IVWMR) to investigate potential independent causal associations between increased BMI and estimated glomerular filtration rate (eGFR) in CKD (CKDeGFR) (<60 mL/min/1.73 m2) and microalbuminuria (MA). In 5,337 BS patients, we assessed whether WL influences >50% decline in eGFR (primary outcome) or CKD hospitalization (secondary outcome), using <20% WL as a comparator. IVWMR results suggest that increased BMI increases CKDeGFR (b = 0.13, P = 1.64 × 10-4; odds ratio [OR] 1.14 [95% CI 1.07, 1.23]) and MA (b = 0.25; P = 2.14 × 10-4; OR 1.29 [1.13, 1.48]). After adjusting for hypertension and fasting glucose, increased BMI did not significantly increase CKDeGFR (b = -0.02; P = 0.72; OR 0.98 [0.87, 1.1]) or MA (b = 0.19; P = 0.08; OR 1.21 [0.98, 1.51]). Post-BS WL significantly reduced the primary outcome with 30 to <40% WL (hazard ratio [HR] 0.53 [95% CI 0.32, 0.87]) but not 20 to <30% WL (HR 0.72 [0.44, 1.2]) and ≥40% WL (HR 0.73 [0.41, 1.30]). For CKD hospitalization, progressive reduction was seen with increased WL, which was significant for 30 to <40% WL (HR 0.37 [0.17, 0.82]) and ≥40% WL (HR 0.24 [0.07, 0.89]) but not 20 to <30% WL (HR 0.60 [0.29, 1.23]). The data suggest that obesity is likely not an independent cause of CKD. WL thresholds previously associated with normotension and normoglycemia, likely causal mediators, may reduce CKD after BS.

Proteomics has been used to study type 2 diabetes, but the majority of available data are from White participants. Here, we extend prior work by analyzing a large cohort of self-identified African Americans in the Jackson Heart Study (n = 1,313). We found 325 proteins associated with incident diabetes after adjusting for age, sex, and sample batch (false discovery rate q < 0.05) measured using a single-stranded DNA aptamer affinity-based method on fasting plasma samples. A subset was independent of established markers of diabetes development pathways, such as adiposity, glycemia, and/or insulin resistance, suggesting potential novel biological processes associated with disease development. Thirty-six associations remained significant after additional adjustments for BMI, fasting plasma glucose, cholesterol levels, hypertension, statin use, and renal function. Twelve associations, including the top associations of complement factor H, formimidoyltransferase cyclodeaminase, serine/threonine-protein kinase 17B, and high-mobility group protein B1, were replicated in a meta-analysis of two self-identified White cohorts-the Framingham Heart Study and the Malmö Diet and Cancer Study-supporting the generalizability of these biomarkers. A selection of these diabetes-associated proteins also improved risk prediction. Thus, we uncovered both novel and broadly generalizable associations by studying a diverse population, providing a more complete understanding of the diabetes-associated proteome.

The molecular and functional heterogeneity of pancreatic β-cells is well recognized, but the underlying mechanisms remain unclear. Pancreatic islets harbor a subset of β-cells that co-express tyrosine hydroxylase (TH), an enzyme involved in synthesis of catecholamines that repress insulin secretion. Restriction of the TH+ β-cells within islets is essential for appropriate function in mice, such that a higher proportion of these cells corresponds to reduced insulin secretion. Here, we use these cells as a model to dissect the developmental control of β-cell heterogeneity. We define the specific molecular and metabolic characteristics of TH+ β-cells and show differences in their developmental restriction in mice and humans. We show that TH expression in β-cells is restricted by DNA methylation during β-cell differentiation. Ablation of de novo DNA methyltransferase Dnmt3a in the embryonic progenitors results in a dramatic increase in the proportion of TH+ β-cells, whereas β-cell-specific ablation of Dnmt3a does not. We demonstrate that maintenance of Th promoter methylation is essential for its continued restriction in postnatal β-cells. Loss of Th promoter methylation in response to chronic overnutrition increases the number of TH+ β-cells, corresponding to impaired β-cell function. These results reveal a regulatory role of DNA methylation in determining β-cell heterogeneity.

Increased saturated fatty acid (SFA) levels in membrane phospholipids have been implicated in the development of metabolic disease. Here, we tested the hypothesis that increased SFA content in cell membranes negatively impacts adipocyte insulin signaling. Preadipocyte cell models with elevated SFA levels in phospholipids were generated by disrupting the ADIPOR2 locus, which resulted in a striking twofold increase in SFA-containing phosphatidylcholines and phosphatidylethanolamines, which persisted in differentiated adipocytes. Similar changes in phospholipid composition were observed in white adipose tissues isolated from the ADIPOR2-knockout mice. The SFA levels in phospholipids could be further increased by treating ADIPOR2-deficient cells with palmitic acid and resulted in reduced membrane fluidity and endoplasmic reticulum stress in mouse and human preadipocytes. Strikingly, increased SFA levels in differentiated adipocyte phospholipids had no effect on adipocyte gene expression or insulin signaling in vitro. Similarly, increased adipocyte phospholipid saturation did not impair white adipose tissue function in vivo, even in mice fed a high-saturated fat diet at thermoneutrality. We conclude that increasing SFA levels in adipocyte phospholipids is well tolerated and does not affect adipocyte insulin signaling in vitro and in vivo.

Mitochondria experience genomic and functional instability in diabetes, and mitochondrial dysfunction has a critical role in the development of diabetic retinopathy. Diabetes also alters expressions of many long noncoding RNAs (LncRNAs), the RNAs with >200 nucleotides and no open reading frame. LncRNAs are mainly encoded by the nuclear genome, but mtDNA also encodes three LncRNAs. Our goal was to investigate the effect of hyperglycemia on mtDNA-encoded LncRNA cytochrome B (LncCytB) in mtDNA stability in diabetic retinopathy. Retinal endothelial cells, transfected with LncCytB-overexpressing plasmids or siRNA, incubated in 5 mmol/L d-glucose (normal glucose [NG]) or 20 mmol/L d-glucose (high glucose [HG]) for 4 days, were analyzed for LncCytB expression by strand-specific PCR and its mitochondrial localization by RNA fluorescence in situ hybridization. Damage-sensitive mtDNA regions were examined by micrococcal nuclease (MNase) digestion sequencing and LncCytB occupancy at mtDNA by chromatin isolation by RNA purification. Protective nucleoids in mtDNA were analyzed by SYBR Green-MitoTracker Red staining and confirmed in isolated mitochondria by flow cytometry. Compared with NG, HG downregulated LncCytB by >50% but had no significant effect on the other mtDNA-encoded LncRNAs. mtDNA packaging was impaired, MNase sensitivity was increased, and LncCytB occupancy at mtDNA was decreased. While LncCytB overexpression ameliorated mtDNA damage and decrease in nucleoids and copy numbers, LncCytB-siRNA exacerbated damage and further reduced nucleoids. Retinal microvessels from streptozotocin-induced diabetic mice and human donors with diabetic retinopathy presented a similar decrease in LncCytB and mtDNA nucleoids. Thus, LncCytB has a major role in maintaining mitochondrial genomic stability, and its downregulation in the hyperglycemic milieu contributes to increased vulnerability of mtDNA to damage.

Data from transgenic rodent models suggest that glucagon acts as an insulin secretagogue by signaling through the glucagon-like peptide 1 receptor (GLP-1R) present on β-cells. However, its net contribution to physiologic insulin secretion in humans is unknown. To address this question, we studied individuals without diabetes in two separate experiments. Each subject was studied on two occasions in random order. In the first experiment, during a hyperglycemic clamp, glucagon was infused at 0.4 ng/kg/min, increasing by 0.2 ng/kg/min every hour for 5 h. On one day, exendin-9,39 (300 pmol/kg/min) was infused to block GLP-1R, while on the other, saline was infused. The insulin secretion rate (ISR) was calculated by nonparametric deconvolution from plasma concentrations of C-peptide. Endogenous glucose production and glucose disappearance were measured using the tracer-dilution technique. Glucagon concentrations, by design, did not differ between study days. Integrated ISR was lower during exendin-9,39 infusion (213 ± 26 vs. 191 ± 22 nmol/5 h, saline vs. exendin-9,39, respectively; P = 0.02). In the separate experiment, exendin-9,39 infusion, compared with saline infusion, also decreased the β-cell secretory response to a 1-mg glucagon bolus. These data show that, in humans without diabetes, glucagon partially stimulates the β-cell through GLP-1R.

Catecholamine neurotransmitters dopamine (DA) and norepinephrine (NE) are essential for a myriad of functions throughout the central nervous system, including metabolic regulation. These molecules are also present in the pancreas, and their study may shed light on the effects of peripheral neurotransmission on glycemic control. Though sympathetic innervation to islets provides NE that signals at local α-cell and β-cell adrenergic receptors to modify hormone secretion, α-cells and β-cells also synthesize catecholamines locally. We propose a model where α-cells and β-cells take up catecholamine precursors in response to postprandial availability, preferentially synthesizing DA. The newly synthesized DA signals in an autocrine/paracrine manner to regulate insulin and glucagon secretion and maintain glycemic control. This enables islets to couple local catecholamine signaling to changes in nutritional state. We also contend that the DA receptors expressed by α-cells and β-cells are targeted by antipsychotic drugs (APDs)-some of the most widely prescribed medications today. Blockade of local DA signaling contributes significantly to APD-induced dysglycemia, a major contributor to treatment discontinuation and development of diabetes. Thus, elucidating the peripheral actions of catecholamines will provide new insights into the regulation of metabolic pathways and may lead to novel, more effective strategies to tune metabolism and treat diabetes.

Although there are some epigenome-wide association studies (EWAS) of insulin resistance, for most of them authors did not replicate their findings, and most are focused on populations of European ancestry, limiting the generalizability. In the Epigenetics in Pregnancy (EPIPREG; n = 294 Europeans and 162 South Asians) study, we conducted an EWAS of insulin resistance in maternal peripheral blood leukocytes, with replication in the Born in Bradford (n = 879; n = 430 Europeans and 449 South Asians), Methyl Epigenome Network Association (MENA) (n = 320), and Botnia (n = 56) cohorts. In EPIPREG, we identified six CpG sites inversely associated with insulin resistance across ancestry, of which five were replicated in independent cohorts (cg02988288, cg19693031, and cg26974062 in TXNIP; cg06690548 in SLC7A11; and cg04861640 in ZSCAN26). From methylation quantitative trait loci analysis in EPIPREG, we identified gene variants related to all five replicated cross-ancestry CpG sites, which were associated with several cardiometabolic phenotypes. Mediation analyses suggested that the gene variants regulate insulin resistance through DNA methylation. To conclude, our cross-ancestry EWAS identified five CpG sites related to lower insulin resistance.

The objective of this work was to investigate whether impaired insulin secretion can be restored by lifestyle intervention in specific subphenotypes of prediabetes. We assigned 1,045 participants from the Prediabetes Lifestyle Intervention Study (PLIS) to six recently established prediabetes clusters. Insulin secretion was assessed by a C-peptide-based index derived from oral glucose tolerance tests and modeled from three time points during a 1-year intervention. We also analyzed the change of glycemia, insulin sensitivity, and liver fat. All prediabetes high-risk clusters (cluster 3, 5, and 6) had improved glycemic traits during the lifestyle intervention, whereas insulin secretion only increased in clusters 3 and 5 (P < 0.001); however, high liver fat in cluster 5 was associated with a failure to improve insulin secretion (Pinteraction < 0.001). Thus, interventions to reduce liver fat have the potential to improve insulin secretion in a defined subgroup of prediabetes.

Genome-wide significant loci for metformin response in type 2 diabetes reported elsewhere have not been replicated in the Diabetes Prevention Program (DPP). To assess pharmacogenetic interactions in prediabetes, we conducted a genome-wide association study (GWAS) in the DPP. Cox proportional hazards models tested associations with diabetes incidence in the metformin (MET; n = 876) and placebo (PBO; n = 887) arms. Multiple linear regression assessed association with 1-year change in metformin-related quantitative traits, adjusted for baseline trait, age, sex, and 10 ancestry principal components. We tested for gene-by-treatment interaction. No significant associations emerged for diabetes incidence. We identified four genome-wide significant variants after correcting for correlated traits (P < 9 × 10-9). In the MET arm, rs144322333 near ENOSF1 (minor allele frequency [MAF]AFR = 0.07; MAFEUR = 0.002) was associated with an increase in percentage of glycated hemoglobin (per minor allele, β = 0.39 [95% CI 0.28, 0.50]; P = 2.8 × 10-12). rs145591055 near OMSR (MAF = 0.10 in American Indians) was associated with weight loss (kilograms) (per G allele, β = -7.55 [95% CI -9.88, -5.22]; P = 3.2 × 10-10) in the MET arm. Neither variant was significant in PBO; gene-by-treatment interaction was significant for both variants [P(G×T) < 1.0 × 10-4]. Replication in individuals with diabetes did not yield significant findings. A GWAS for metformin response in prediabetes revealed novel ethnic-specific associations that require further investigation but may have implications for tailored therapy.

Perifornical hypothalamus (PFH) orexin glucose-inhibited (GI) neurons that facilitate arousal have been implicated in hypoglycemia awareness. Mice lacking orexin exhibit narcolepsy, and orexin mediates the effect of the antinarcolepsy drug modafinil. Thus, hypoglycemia awareness may require a certain level of arousal for awareness of the sympathetic symptoms of hypoglycemia (e.g., tremors, anxiety). Recurrent hypoglycemia (RH) causes hypoglycemia unawareness. We hypothesize that RH impairs the glucose sensitivity of PFH orexin GI neurons and that modafinil normalizes glucose sensitivity of these neurons and restores hypoglycemia awareness after RH. Using patch-clamp recording, we found that RH enhanced glucose inhibition of PFH orexin GI neurons in male mice, thereby blunting activation of these neurons in low-glucose conditions. We then used a modified conditioned place preference behavioral test to demonstrate that modafinil reversed hypoglycemia unawareness in male mice after RH. Similarly, modafinil restored normal glucose sensitivity to PFH orexin GI neurons. We conclude that impaired glucose sensitivity of PFH orexin GI neurons plays a role in hypoglycemia unawareness and that normalizing their glucose sensitivity after RH is associated with restoration of hypoglycemia awareness. This suggests that the glucose sensitivity of PFH orexin GI neurons is a therapeutic target for preventing hypoglycemia unawareness.

Dysregulated lipid droplet accumulation has been identified as one of the main contributors to liver steatosis during nonalcoholic fatty liver disease (NAFLD). However, the underlying molecular mechanisms for excessive lipid droplet formation in the liver remain largely unknown. In the current study, hepatic E4 promoter-binding protein 4 (E4BP4) plays a critical role in promoting lipid droplet formation and liver steatosis in a high-fat diet (HFD)-induced NAFLD mouse model. Hepatic E4bp4 deficiency (E4bp4-LKO) protects mice from HFD-induced liver steatosis independently of obesity and insulin resistance. Our microarray study showed a markedly reduced expression of lipid droplet binding genes, such as Fsp27, in the liver of E4bp4-LKO mice. E4BP4 is both necessary and sufficient to activate Fsp27 expression and lipid droplet formation in primary mouse hepatocytes. Overexpression of Fsp27 increased lipid droplets and triglycerides in E4bp4-LKO primary mouse hepatocytes and restored hepatic steatosis in HFD-fed E4bp4-LKO mice. Mechanistically, E4BP4 enhances the transactivation of Fsp27 by CREBH in hepatocytes. Furthermore, E4BP4 is modified by SUMOylation, and HFD feeding induces deSUMOylation of hepatic E4BP4. SUMOylation of five lysine residues of E4BP4 is critical for the downregulation of Fsp27 and lipid droplets by cAMP signaling in hepatocytes. Taken together, this study revealed that E4BP4 drives liver steatosis in HFD-fed mice through its regulation of lipid droplet binding proteins. Our study also highlights the critical role of deSUMOylation of hepatic E4BP4 in promoting NAFLD.

Complex bidirectional cross talk between adipocytes and adipose tissue immune cells plays an important role in regulating adipose function, inflammation, and insulin responsiveness. Adipocytes secrete the pleiotropic cytokine IL-6 in response to both inflammatory and catabolic stimuli. Previous studies have suggested that IL-6 secretion from adipocytes in obesity may promote adipose tissue inflammation. Here, we investigated catabolic stimulation of adipocyte IL-6 secretion and its impact on adipose tissue immune cells. In obesity, catecholamine resistance reduces cAMP-driven adipocyte IL-6 secretion in response to catabolic signals. By restoring adipocyte catecholamine sensitivity in obese adipocytes, amlexanox stimulates adipocyte-specific IL-6 secretion. We report that in this context, adipocyte-secreted IL-6 activates local macrophage STAT3 to promote Il4ra expression, thereby sensitizing them to IL-4 signaling and promoting an anti-inflammatory gene expression pattern. Supporting a paracrine adipocyte to macrophage mechanism, these effects could be recapitulated using adipocyte conditioned media to pretreat bone marrow-derived macrophages prior to polarization with IL-4. The effects of IL-6 signaling in adipose tissue are complex and context specific. These results suggest that cAMP-driven IL-6 secretion from adipocytes sensitizes adipose tissue macrophages to IL-4 signaling.

Microvascular pathology in the brain is one of the suggested mechanisms underlying the increased incidence and progression of neurodegenerative diseases in people with type 2 diabetes (T2D). Although accumulating data suggest a neuroprotective effect of antidiabetics, the underlying mechanisms are unclear. Here, we investigated whether two clinically used antidiabetics, the dipeptidyl peptidase-4 inhibitor linagliptin and the sulfonylurea glimepiride, which restore T2D-induced brain vascular pathology. Microvascular pathology was examined in the striatum of mice fed for 12 months with either normal chow diet or a high-fat diet (HFD) to induce T2D. A subgroup of HFD-fed mice was treated with either linagliptin or glimepiride for 3 months before sacrifice. We demonstrate that T2D caused leakage of the blood-brain barrier (BBB), induced angiogenesis, and reduced pericyte coverage of microvessels. However, linagliptin and glimepiride recovered the BBB integrity and restored the pericyte coverage differentially. Linagliptin normalized T2D-induced angiogenesis and restored pericyte coverage. In contrast, glimepiride enhanced T2D-induced angiogenesis and increased pericyte density, resulting in proper vascular coverage. Interestingly, glimepiride reduced microglial activation, increased microglial-vascular interaction, and increased collagen IV density. This study provides evidence that both DPP-4 inhibition and sulfonylurea reverse T2D-induced BBB leakage, which may contribute to antidiabetic neurorestorative effects.