Diabetologia最新文献

Aims/hypothesis: Brown adipose tissue (BAT) consumes excess energy through heat production by uncoupling protein 1 (UCP1) to regulate the metabolic profile, but the UCP1-independent mechanisms of BAT, such as in endocrine function, are largely unknown. Our previous study showed that BAT-derived neuregulin 4 (NRG4) displays anti-atherosclerotic properties. Thus, we hypothesised that BAT could regulate diabetic nephropathy, a diabetic microvascular complication, via NRG4.

Methods: To investigate the influence of NRG4 from BAT on podocyte apoptosis, both loss- and gain-of-function approaches were used in in vivo experiments. Diabetic nephropathy models were created using BAT-specific Nrg4-knockout (BKO) mice, global Nrg4-knockout (KO) mice and wild-type (WT) mice. In in vitro studies, podocytes (MPC5) were exposed to glucose and recombinant NRG4 (rNrg4). Additionally, brown adipocytes were co-cultured with MPC5 podocytes using a transwell system. The expression levels of proteins associated with podocyte apoptosis and signalling pathways were measured.

Results: BAT-specific NRG4 deficiency aggravated podocyte apoptosis (increased by 47.46%) and increased the urinary albumin/creatinine ratio (increased by 41.71%), decreased nephrin expression and increased desmin expression. As expected, these changes were reversed by NRG4 replenishment using adeno-associated virus-NRG4 interscapular BAT injection and BAT transplantation assays in KO mice. Additionally, co-culture experiments demonstrated that brown adipocytes from WT mice could alleviate high-glucose-induced podocyte apoptosis. In in vitro experiments, recombinant NRG4 inhibited high-glucose-induced podocyte apoptosis. Mechanistically, the Akt-glycogen synthase kinase 3 β (GSK-3β) pathway is crucial for the protection that BAT-derived NRG4 provides to podocytes in diabetic nephropathy.

Conclusions/interpretation: Our data show that BAT had a protective effect on podocyte apoptosis in diabetic nephropathy through BAT-derived NRG4, and the Akt-GSK‑3β signalling pathway may mediate the inhibition of BAT-derived NRG4 on podocyte apoptosis in diabetic nephropathy.

Aims/hypothesis

Growing evidence suggests that timing may be an important aspect of physical activity that influences cardiometabolic health. However, the current literature is inconclusive regarding the time of day that physical activity offers the greatest metabolic advantages. We investigated associations between hourly physical activity levels and diabetes mellitus and glycaemic biomarkers in a cross-sectional and nationally representative sample of US adults.

Methods

We studied 7074 adults (mean age 48 years; 52% women) from the National Health and Nutrition Examination Survey (2011–2014). Physical activity was measured by actigraphy. A monitor-independent movement summary (MIMS) unit was used to derive the total activity level (divided into quintiles) for hourly windows that were defined relative to sleep timing and according to clock time. The primary outcome was prevalent diabetes, and secondary outcomes included fasting glucose, fasting insulin, HOMA-IR and 2 h OGTT results.

Results

Physical activity levels in late morning and late afternoon were associated with lower adjusted odds of diabetes. Specifically, in late morning (8:01–9:00 h after the sleep midpoint), the highest quintile of activity was associated with a 35% decrease (OR 0.65; 95% CI 0.44, 0.96) in the odds of diabetes when compared with the lowest quintile, while in late afternoon (11:01–17:00 h after the sleep midpoint), the highest quintiles were associated with 56% and 36% lower odds (OR 0.44; 95% CI 0.29, 0.69 and OR 0.64; 95% CI 0.43, 0.95). Higher night-time activity was associated with higher odds of diabetes. Similar patterns of results were observed with OGTT data and across subgroups of age, gender, race/ethnicity, chronotype and sleep duration.

Conclusions/interpretation

Our findings suggest that the timing of physical activity may modulate its metabolic effects.

Graphical Abstract

Aims/hypothesis

Type 1 diabetes is categorised into autoantibody positive and autoantibody negative. Most type 1 diabetes research has focused on European populations, leaving a gap in understanding in relation to other ethnic groups, including Thai populations. This lack of data is significant given Thailand’s poor prevention and therapeutic management strategies. We aimed to investigate the frequency and distribution of islet autoantibodies among Thai individuals with long-standing type 1 diabetes diagnosed before the age of 30 years.

Methods

We conducted a nationwide population-based study involving 48 hospitals in Thailand from May 2020 to September 2023, enrolling 953 participants. Demographic and clinical characteristics of individuals with autoantibody-positive and -negative type 1 diabetes were analysed. The autoantibodies GAD65, IA-2 and ZnT8 were measured using ELISA. A random C-peptide level was detected by electrochemiluminescence immunoassay.

Results

Thai individuals with autoantibody-negative type 1 diabetes comprised 34.2% of the population. Among all individuals, the frequency of GAD65, IA-2 and ZnT8 was 56%, 37% and 33%, respectively. Autoantibody-negative individuals with type 1 diabetes were older at diagnosis, had higher BMI and had higher random C-peptide levels compared with autoantibody-positive individuals with type 1 diabetes. Female individuals had a higher prevalence of type 1 diabetes than male individuals (58% vs 42%; p=1.531 × 10⁻⁵). The southern region of Thailand exhibited a distinct pattern of autoantibody frequency compared with other regions (p=0.0001561).

Conclusions/interpretation

The frequency, distribution and characteristics of autoantibody-positive and -negative long-standing type 1 diabetes in Thailand showed uniqueness from other populations. This provides insight into the disease that may have implications for type 1 diabetes prediction, treatment and pathogenesis, especially in the Southeast Asian population.

Graphical Abstract

Aims/hypothesis

The management of GCK-MODY during pregnancy remains challenging. We evaluated the impact on pregnancy and neonatal outcomes of two standardised insulin strategies.

Methods

In this prospective observational study, participants chose (in agreement with their physician) to be treated with insulin either when maternal capillary blood glucose (CBG) ≥ thresholds for gestational diabetes (5.3 mmol/l before or 6.7 mmol/l 2 h after meals) (MG group) or when fetal abdominal circumference ≥75th percentile (FG group). In the FG group, insulin was also initiated if CBG ≥ safety levels (6.7 mmol/l before meals or 11.1 mmol/l 2 h after meals). Data on glycaemic management, modalities and timing of insulin therapy and maternal and neonatal outcomes were recorded.

Results

In the MG group (n=25), insulin was initiated more frequently (100% vs 75%, p=0.01) and earlier (p=0.001), with lower CBG and more frequent hypoglycaemic episodes compared with the FG group (n=21). However, there were no differences in pregnancy and neonatal outcomes. In the total cohort, the rate of large for gestational age (LGA) neonates, preterm deliveries and Caesarean sections was 22.2%, 2.2% and 40%, respectively. The rate of LGA was 0% among the neonates with the GCK variant vs 36% in those without (p=0.005). There were no associations between LGA and pregnancy characteristics, insulin therapy strategy or glycaemic management.

Conclusions/interpretation

In our study, the rate of LGA primarily depended on the fetal GCK genotype rather than the treatment strategy or glycaemic management. Our results suggest that a standardised strategy based on ultrasound monitoring of fetal growth and glycaemic safety thresholds, leading to delayed insulin initiation, offers a good fetal prognosis and minimises the risk of maternal hypoglycaemia.

Trial registration

ClinTrials.gov NCT02556840.

Graphical Abstract