Ketone bodies, particularly β-hydroxybutyrate (3HB), are often elevated in type 1 diabetes (T1D); however, their physiological roles remain unclear. In a low-carbohydrate diet study, patients with insulin-deficient diabetes exhibited reduced fasting blood glucose and increased fasting blood ketone levels, negatively correlated. Another clinical study using continuous glucose and ketone monitoring confirmed inverse glucose-ketone fluctuations. To test causality, we conducted animal and cellular studies. In streptozotocin-induced T1D mice, 7-week oral 3HB administration improved glucose metabolism and alleviated glycogenic hepatopathy. Imaging with 2-deoxy-2-[18F]-fluoro-d-glucose positron emission tomography/computed tomography demonstrated reduced hepatic and intestinal glucose uptake. Western blotting confirmed 3HB suppressed glucose transporter (sodium-glucose cotransporter 1, GLUT2, GLUT5) overexpression and normalized glycogen metabolism. In vitro, 3HB dose-dependently inhibited glucose transporter expression and glucose uptake in primary hepatocytes and IEC-6 cells. G protein-coupled receptor 109A (GPR109A) serves as the primary receptor for 3HB. Mechanistic studies using the GPR109A inhibitor mepenzolate bromide, the mTOR inhibitor rapamycin, and siRNA-mediated gene silencing revealed that these effects were GPR109A dependent and linked to inhibition of the PI3K/AKT/mTOR pathway. Overall, this study provides new insights into the role of ketone bodies in T1D, establishing 3HB as a modulator of glucose homeostasis through GPR109A-mediated suppression of glucose transporters in the liver and intestine.

Type 2 diabetes (T2D) impairs antiviral immunity; however, the causal link between T2D and interferon-α2 (IFN-α2) deficiency remains unclear. This study used genome-wide association study-based Mendelian randomization (MR) to investigate this relationship and validated the findings in an H1N1-infected diabetic mouse model. MR analysis of 26 single nucleotide polymorphisms showed a significant negative association between T2D and IFN-α2 levels (inverse variance weighted odds ratio 0.667; P = 0.000116) without heterogeneity or pleiotropy. In vivo experiments confirmed that db/db mice exhibited more severe H1N1-induced lung injury, higher viral loads, and lower survival rates compared with nondiabetic controls. However, exogenous IFN-α2 treatment significantly reversed these pathologic outcomes. Inflammatory cytokine profiling showed that IFN-α2 downregulated 21 elevated cytokines and restored Fas ligand levels in lung tissue. Mechanistically, Western blotting demonstrated that IFN-α2 inhibited the phosphorylation of JAK1/2 and STAT3, thereby suppressing excessive inflammation. In conclusion, our findings indicate that T2D leads to IFN-α2 deficiency, contributing to susceptibility to viral infection. Supplementation with IFN-α2 effectively attenuated virus-induced lung injury by inhibiting JAK/STAT3 signaling and cytokine storms, positioning IFN-α2 supplementation as a promising therapeutic strategy for managing influenza complications in patients with diabetes.

Large-scale multiancestry genome-wide association studies have identified hundreds of loci associated with type 2 diabetes (T2D) and glycemic traits, yet imputed genotyping arrays limit the detection of low-frequency and rare variants. Whole-genome sequencing (WGS) offers a more complete view of genetic variation, especially across diverse populations. We analyzed high-coverage (38×) WGS data from 21,913 T2D case subjects, 61,036 control subjects, and up to 50,011 individuals with no diabetes with fasting glucose, fasting insulin, and HbA1c from the National Heart, Lung, and Blood Institute Trans-Omics for Precision Medicine Program. We performed single-variant association testing, conditional analysis, fine-mapping, and Bayesian colocalization to identify genetic signals and assess regulatory relevance in diabetes-related tissues. We identified 76 distinct association signals across 34 loci, including novel variants at DUSP9 for T2D, and ROBO1, NDN, and MYT1 for HbA1c. Fine-mapping narrowed credible sets and improved causal variant resolution. Colocalization highlighted 80 expression signals in diabetes-related tissues, linking genetic associations to functional regulatory mechanisms. Our findings demonstrate the utility of WGS to uncover novel variants in diverse populations, enhance locus resolution, and link regulatory variation to disease-relevant tissues. This work refines the genetic architecture of T2D and glycemic traits and supports precision medicine efforts targeting diverse populations.

One-third of patients with diabetes will develop end-stage kidney disease, and approximately one-half of the patients requiring kidney transplantation in the U.S. suffer from diabetes. In 2002, Yamada, Sachs, and their colleagues explored the renal subcapsular site for islet implantation in a partially inbred miniature swine model. All pancreatectomized pigs that received autologous islets under the renal capsule maintained normoglycemia. The transplantation of a composite islets-kidney (i.e., the kidney with the revascularized autologous islets) into another pancreatectomized, bilaterally nephrectomized pig provided successful renal and islet function. In 2011, this model was extended successfully to nonhuman primates (NHPs). There are several hurdles to be overcome if this model is to be successfully translated into clinical practice, and xenotransplantation might resolve this problem. Pancreases harvested from gene-edited pigs could be sources of islets that could be isografted beneath the renal capsule of an identical cloned littermate. After allowing the islets to become vascularized over a period of 6-12 weeks, during which no immunosuppressive therapy would be required, the composite islets-kidney would then be transplanted into an immunosuppressed NHP or human recipient. On reperfusion of the graft, both renal and islet function should rapidly return. If sufficient islets have been transplanted with the kidney and the immunosuppressive regimen is successful, normoglycemia and normal renal function should be quickly achieved. However, a large number of genetically engineered neonatal donor piglets will be required to manufacture one therapeutic patient dose of islets, which may prove to be a logistical problem. Alternative approaches are discussed.

Infections represent major but underrecognized complications of type 1 diabetes (T1D), type 2 diabetes (T2D), and prediabetes. This recent overview includes use of linked U.K. primary care, hospitalization, and mortality data to examine infection risk for >800,000 individuals with diabetes or prediabetes, compared with an age-, sex-, and ethnicity-matched group without diabetes. We assessed 1) infections in primary care and requiring a prescription, 2) infections leading to hospitalization, and 3) infection-related mortality, over 5 years (2015-2019). Infection risks were consistently elevated across all forms of diabetes, compared with risk for those without diabetes. The highest relative risks were observed in T1D and lowest in prediabetes. While relative risks were similar across different ethnicities, the population burden of infections attributable to diabetes was highest among South Asians. There were independently strong graded associations for average HbA1c level and visit-to-visit HbA1c variability with infection risk, particularly for hospitalization infections. Among individuals with T1D, the strongest associations with infection risk were seen for elevated HbA1c levels, whereas for T2D, variability contributed more to the excess burden of infections. Infection-related mortality across all ICD-10 chapters was substantial, representing the third leading cause of death in T2D after cardiovascular disease and cancer. Further, it might be underreported, as sepsis is rarely coded as the underlying cause of death. These findings highlight the importance of improved glycemic control, earlier recognition and treatment of infections, and stronger emphasis on infection management in clinical guidelines. Incorporation of HbA1c variability into diabetes risk algorithms, and evaluation of interventions that might stabilize control, such as continuous glucose monitoring, may enhance infection prevention and reduce complications.

Worldwide metabolic dysfunction-associated steatotic liver disease (MASLD) is the leading cause of chronic liver disease. Despite high global prevalence, MASLD is not yet recognized as a noncommunicable disease, although all of the criteria are fulfilled for such a classification. MASLD is strongly associated with type 2 diabetes, cardiovascular disease, chronic kidney disease, and certain extrahepatic cancers. At 65% and 37% the prevalence of MASLD is extremely high in adults and children with type 2 diabetes, respectively. The pathogenesis of MASLD is closely related to that of type 2 diabetes, and diabetes-associated hyperglycemia, hyperinsulinemia, and hyperlipidemia promote progression from simple steatosis to hepatic inflammation and fibrosis. Thus, MASLD is now considered a complication of diabetes. However, there is still a great deal of work to be done to implement screening for and treatment of MASLD in everyday clinical diabetes management. In this article, the major mechanisms involved in the pathogenesis of MASLD and type 2 diabetes are discussed. Furthermore, the heterogeneity in the pathophysiology of MASLD and clusters in MASLD that may be relevant for future stratification of MASLD-associated risk of diseases are addressed. Finally, because of their strong hepato-, cardio-, and nephroprotective effects this article provides support as to why sodium-glucose cotransporter 2 inhibitors and glucagon-like peptide 1 receptor (co)agonists should be used as first-line pharmacotherapies in people with MASLD and type 2 diabetes.

Adipose triglyceride (TG) lipase (ATGL) is the rate-limiting enzyme of TG catabolism at the surface of lipid droplets (LDs). In pancreatic β-cells, lipolysis by ATGL increases with glucose and produces metabolites that support glucose-stimulated insulin secretion (GSIS). Upregulation of lipolysis by glucose is blunted in human islets affected by type 2 diabetes. However, how glucose regulates lipolysis in β-cells is unknown. We found that glucose recruits ATGL coactivator α/β-hydrolase domain containing 5 (ABHD5) to LDs in INS-1 cells and human β-cells. ABHD5 recruitment to LDs was reduced by inhibitors of cAMP-dependent protein kinase (PKA) or the expression of PKA-resistant mutant ABHD5, indicating that PKA phosphorylation contributes to ABHD5 recruitment to LDs. Importantly, ABHD5 was indispensable for glucose-stimulated lipolysis in INS-1 cells and human β-cells. ABHD5 knockdown increased LDs and TG content in INS-1 cells and human islets, indicating that ABHD5 is a critical regulator of ATGL activity in β-cells. Additionally, GSIS was impaired in ABHD5-downregulated INS-1 cells and human pseudoislets, which agrees with a role of ATGL in supporting GSIS in β-cells. Thus, ABHD5 plays an important role in conferring glucose responsiveness in lipolysis and supporting insulin secretion in β-cells.

We investigated whether lowering dietary carbohydrate content and glycemic index (GI) levels altered deep lipidomic profiles and whether these changes were associated with improved diurnal postprandial glucose response (PPGR). In the OmniCarb trial, 59 adults completed 5-week controlled feeding interventions (low carbohydrate/low GI vs. high carbohydrate/high GI) and 12-h meal tests. Comprehensive lipidomic profiling was performed to measure lipid species and lipid class-specific fatty acids (FAs) at the end of each intervention. We found that lowering carbohydrate content and GI levels significantly decreased triacylglycerols (TAGs) and phosphatidylcholines, while increasing lactosylceramides and phosphatidylethanolamines (PEs). Of 731 lipid species analyzed, 521 (71%) were significantly modified, including TAGs (n = 398), PEs (n = 45), and ceramides. Of 199 FAs analyzed across and within lipid classes, 89 showed significant changes (false discovery rate-adjusted P < 0.05), including decreases in saturated FAs (total and TAG FA12:0 and FA14:0) and palmitoleic acid and increases in very-long-chain saturated FAs, when lowering carbohydrate and GI levels. Between-diet changes in six total FAs and 17 lipid class-specific FAs were associated with half-day PPGR changes; greater decreases in joint FA score changes were linked to improved PPGRs. Our study suggests the potential importance of dietary carbohydrate-responsive lipidomic signatures in explaining individual variability in half-day PPGRs and may encourage future intervention studies to target these signatures.

The second-meal phenomenon refers to the improved glycemic response to a subsequent identical meal. We previously showed that morning hyperinsulinemia is a key mediator, priming the liver for enhanced net hepatic glucose uptake (NHGU) and glycogen storage during an afternoon hyperinsulinemic-hyperglycemic clamp. Postprandial NHGU is regulated by three primary mechanisms: insulin action, initiated by hyperinsulinemia; glucose effectiveness (GE), driven by hyperglycemia; and the portal glucose signal (PGS), a neurally mediated signal activated by glucose delivery into the hepatoportal circulation. It remains unclear, however, which of these mechanisms govern the increase in afternoon NHGU following morning insulin exposure. To address this, dogs underwent a morning clamp with either a 4-h hyperinsulinemic prime (prime group) (n = 8) or basal insulin delivery (no-prime group) (n = 8). After a 1.5-h rest, both groups underwent an afternoon hyperglycemic clamp with portal glucose delivery under basal insulin conditions to isolate the effects of a morning insulin prime on afternoon glucose-mediated hepatic signals (GE and the PGS). Mean afternoon NHGU was significantly greater in the prime group (2.2 ± 0.3 mg/kg/min) compared with the no-prime group (0.1 ± 0.3 mg/kg/min; P = 0.005), accompanied by augmented net glycolytic and glycogen flux. These findings indicate that morning insulin can enhance glucose-mediated afternoon NHGU independently of a rise in afternoon insulin. However, maximal second-meal NHGU also requires elevated afternoon insulin. Together, these findings suggest that strategically timed early-day insulin or insulinotropic interventions could potentially improve hepatic responsiveness in settings of impaired postprandial glycemic control, such as insulin resistance or diabetes.

Reduction in intestinal type 3 innate lymphoid cells (ILC3) frequency is associated with type 1 diabetes (T1D) pathogenesis in humans and in animal models. The current study showed that, after T1D induction by multiple-low-dose streptozotocin in male C57BL/6 mice, ILC3 were reduced in both blood and pancreas and produced less IL-22 and IL-2, and a decreased proportion expressed the gut-homing α4β7 integrin in the pancreas. Additionally, T1D induction in ILC3-impaired mice produced exacerbated hyperglycemia. To activate ILC3, Compound 1 (Cpd1), a synthetic selective free fatty acid receptor 2 (FFAR2) agonist, was administered orally in a prophylactic regimen, at early therapeutic stages, or during established disease, resulting in significant amelioration of T1D symptoms. Cpd1 upregulated ILC3 and activated regulatory T cells (Treg) within the small intestine lamina propria and the pancreas, enhanced intestinal barrier integrity, and increased microbial diversity. Cpd1 also reduced α4β7-expressing inflammatory cells in the pancreas while promoting the accumulation of gut-imprinted ILC3. IL-22 was important for this anti-inflammatory response, as treatment efficacy was inhibited when mice were additionally given neutralizing anti-IL-22 antibodies. Of note, NOD mice receiving Cpd1 exhibited postponed disease initiation, but the incidence of disease was similar to that of the control group. Overall, activation of intestinal ILC3 and Treg via FFAR2 engagement, and translation of these anti-inflammatory effects to the pancreas, provided significant benefit in a mouse T1D model.

Diabetic kidney disease (DKD) and diabetic cardiomyopathy continue to drive excess morbidity and mortality in diabetes, underscoring a critical gap between mechanistic insight and clinical translation. Growth differentiation factor-15 (GDF-15), a stress-inducible cytokine of the transforming growth factor-β superfamily, has emerged as a critical biomarker and putative modulator of metabolic inflammation. Yet the field remains divided on a fundamental question: is GDF-15 simply reporting tissue distress, or does it shape disease trajectories? In this article, we explore how GDF-15 may both signal and shape DKD and cardiovascular disease. Drawing on evidence from experimental models, longitudinal clinical studies, and multi-omics analyses, we highlight the context-dependent biology of GDF-15, protective during acute metabolic or inflammatory stress but potentially pathogenic when chronically elevated in diabetes. We examine its regulation via the GFRAL-RET signaling axis, its segment-specific expression across renal tubular compartments, and its emerging role in cardiac remodeling and metabolic inflammation. Recent clinical data position circulating GDF-15 as an early and sensitive indicator of DKD progression and cardiovascular events. At the same time, mechanistic studies increasingly implicate sustained GDF-15 signaling in mitochondrial dysfunction, inflammatory amplification, and maladaptive tissue remodeling. Together, these observations place GDF-15 at a critical inflection point between risk stratification and disease mechanism. A key unresolved challenge is defining when, where, and how GDF-15 signaling exerts adaptive versus maladaptive effects-knowledge that will be essential for determining whether GDF-15 should be targeted, harnessed, or restrained in diabetes.