Daily Endocrinology Research Analysis
Analyzed 144 papers and selected 3 impactful papers.
Summary
Mechanistic and translational endocrinology advanced on three fronts: liraglutide’s mode of action varies by metabolic state with distinct brain versus islet contributions; an interpretable machine-learning classifier using LC-MS/MS steroid profiles accurately subtyped congenital adrenal steroidogenesis disorders; and adipocyte myoglobin emerged as a determinant of energy expenditure and obesity susceptibility. Together these studies point to precision therapeutics, AI-enabled diagnostics, and new bioenergetic targets in metabolic disease.
Research Themes
- Metabolic-state-dependent mechanisms of incretin therapies
- AI-assisted endocrine diagnostics using LC-MS/MS profiles
- Adipocyte bioenergetics as therapeutic targets for obesity
Selected Articles
1. Metabolic state determines the brain and direct islet effects of liraglutide on enhanced insulin secretion.
Using human pancreatic islets across metabolic states, the authors show that liraglutide enhances insulin secretion in glucose-intolerant but not normoglycaemic donors, while GLP-1R expression declines with rising HbA1c in T2D islets. Complementary mechanisms predominate by state: brain tanycyte-mediated actions in healthy conditions and direct islet effects during glucose intolerance, suggesting stratified use of GLP-1RAs.
Impact: Provides mechanistic evidence that the site and efficacy of GLP-1RA action depend on metabolic state, informing precision pharmacology and response prediction.
Clinical Implications: Supports metabolically stratified selection and counseling for GLP-1RAs; patients with glucose intolerance may derive greater islet-secretory benefits, whereas brain-mediated mechanisms may dominate in early disease.
Key Findings
- Liraglutide increased glucose-stimulated insulin secretion in glucose-intolerant donor islets (n=7, p=0.021) but not in normoglycaemic islets (n=7).
- GLP-1R mRNA levels declined progressively with rising HbA1c in T2D donor islets, indicating receptor-level changes with metabolic deterioration.
- Mechanistic framework: tanycyte-mediated brain actions predominate in healthy states, while direct islet effects emerge during glucose intolerance; insulin-independent mechanisms help maintain efficacy.
Methodological Strengths
- Use of human pancreatic islets spanning distinct metabolic states enables translational relevance.
- Integration of central (tanycyte-mediated) and peripheral (islet) mechanistic interrogation.
Limitations
- Human donor sample sizes per metabolic stratum are small, limiting precision and subgroup analyses.
- Not a clinical outcomes study; mechanistic inferences require prospective validation in patients.
Future Directions: Prospective stratified trials to test whether metabolic-state-informed selection of GLP-1RAs improves glycaemic outcomes and durability; biomarker development (e.g., GLP-1R expression proxies) to guide therapy.
AIMS/HYPOTHESIS: Liraglutide, a glucagon-like peptide-1 receptor (GLP-1R) agonist for type 2 diabetes and obesity management, shows variable patient responses. We investigated the metabolic state-dependent mechanisms underlying this heterogeneity and how liraglutide's mode of action shifts across stages of metabolic dysfunction. METHODS: We employed human pancreatic islets from donors across metabolic states (normoglycaemic [HbA RESULTS: Liraglutide (25 nmol/l) enhanced glucose-stimulated insulin secretion specifically in donors with glucose intolerance (n=7, p=0.021), with no effect in normoglycaemic islets (n=7), despite preserved GLP-1 (7-36) responsiveness. In type 2 diabetes islets, GLP-1R mRNA levels progressively decreased with rising HbA CONCLUSIONS/INTERPRETATION: Liraglutide operates through complementary, metabolic state-dependent pathways: tanycyte-mediated brain actions predominate in healthy conditions, direct islet effects emerge during glucose intolerance and insulin-independent mechanisms maintain efficacy across metabolic states. This mechanistic framework enables potential patient stratification in type 2 diabetes therapy, suggesting that matching liraglutide's predominant mechanism to individual metabolic profiles could optimise treatment outcomes.
2. Machine Learning Algorithms to Accelerate Etiological Diagnosis of Congenital Disorders of Adrenal Steroidogenesis.
A LightGBM-driven feature selection combined with an optimized decision tree classified eight CDAS subtypes using LC-MS/MS steroid panels with 97–99% overall accuracy across development and independent validation cohorts. SHAP highlighted a small set of steroids (e.g., 11-deoxycortisol, 17-OHP, 21-deoxycortisol, corticosterone) as primary discriminators, and dimensionality reduction confirmed biologically coherent subtype clusters.
Impact: Demonstrates an accurate, interpretable ML pipeline for etiologic CDAS classification that can accelerate triage and genetic testing decisions in pediatric endocrinology.
Clinical Implications: Integration into diagnostic pathways could reduce time-to-diagnosis, prioritize genotype confirmation, and support targeted management, especially in settings with LC-MS/MS capacity.
Key Findings
- Development cohort (n=1027) performance: accuracy 97.1%, sensitivity 99.5%, specificity 93.7%, macro-AUC 0.97; subtype accuracy >98% for most major subtypes.
- Independent validation cohort (n=507): accuracy 98.9%, sensitivity 93.6%, specificity 99.8%.
- SHAP identified 11-deoxycortisol, 17-hydroxyprogesterone, 21-deoxycortisol, and corticosterone as strongest discriminators; PCA/UMAP showed distinct biologically coherent subtype clusters.
Methodological Strengths
- Large development cohort with independent external validation and five-fold cross-validation.
- Model interpretability via SHAP, with dimensionality reduction confirming biological coherence.
Limitations
- Clinical impact (time-to-diagnosis, outcomes) not tested prospectively.
- Generalizability may depend on LC-MS/MS availability and population differences; further multi-ethnic validation needed.
Future Directions: Prospective clinical utility studies assessing workflow integration, cost-effectiveness, and patient outcomes; expansion to low-resource settings with simplified panels.
BACKGROUND: Early and accurate etiological diagnosis of congenital disorders of adrenal steroidogenesis (CDAS) is critical as timely targeted management can prevent life-threatening complications and improve long-term outcomes. OBJECTIVE: To develop and validate a machine learning-assisted decision tree model for classifying CDAS using plasma steroid hormone profiles quantified by liquid chromatography-mass spectrometry (LC-MS/MS). METHODS: A development cohort of 1027 participants (325 genetically confirmed CDAS patients representing eight subtypes/702 controls) was used for model construction. The Light Gradient Boosting Machine algorithm identified key discriminatory steroid hormones, which were integrated into an optimized decision-tree classifier. Internal performance was assessed through five-fold cross-validation. The performance of the model was further evaluated using a validation cohort comprising 507 independent LC-MS/MS steroid profiles. Additional analyses included Shapley additive explanations (SHAP), confusion matrix visualization, Principal Component Analysis (PCA), and Uniform Manifold Approximation and Projection (UMAP). RESULTS: In the development cohort, the model achieved a mean overall accuracy of 97.1%, sensitivity of 99.5%, and specificity of 93.7%, with a macro-AUC (area under curve) of 0.97. Subtype-level accuracy exceeded 98% for most major CDAS subtypes. In the validation cohort, overall accuracy was 98.9%, sensitivity 93.6%, specificity 99.8%. Feature importance analysis and SHAP identified 11-deoxycortisol, 17-hydroxyprogesterone, 21-deoxycortisol, and corticosterone as the strongest discriminators. PCA and UMAP revealed distinct clustering of CDAS subtypes, confirming the biological coherence of model predictions. CONCLUSION: Machine learning-assisted steroid profiling provides an accurate and highly interpretable diagnostic approach for CDAS, with potential for integration into pediatric endocrine diagnostics and decision-support systems.
3. Adipocyte Myoglobin Is a Determinant of Energy Expenditure and a Potential Target to Limit Obesity.
Adipose-specific myoglobin knockout reduced whole-body energy expenditure, impaired thermoregulation, and increased diet-induced obesity, with elevated circulating lipids and coordinated downregulation of oxidative metabolism pathways. Restoring or overexpressing myoglobin improved metabolic phenotypes and enhanced thermogenic activity, including in primary human adipocytes, nominating myoglobin as a therapeutic target.
Impact: Identifies adipocyte myoglobin as a key regulator of thermogenic lipid oxidation and systemic energy expenditure with human cell validation, opening a new avenue for anti-obesity strategies.
Clinical Implications: While preclinical, strategies to upregulate adipocyte myoglobin or mimic its effects could augment energy expenditure and complement existing obesity therapies.
Key Findings
- Adipose tissue-specific myoglobin knockout mice showed reduced energy expenditure, impaired thermoregulation, and increased susceptibility to diet-induced obesity.
- Circulating triglycerides and fatty acids increased, and omics analyses revealed coordinated downregulation of oxidative phosphorylation and fatty acid metabolism programs.
- Restoration/overexpression of myoglobin improved metabolism in vivo and enhanced thermogenic activity in primary human white adipocytes.
Methodological Strengths
- Use of adipose tissue-specific knockout and rescue strategies to establish causality.
- Cross-species validation including primary human adipocytes with functional readouts.
Limitations
- Preclinical study; no human in vivo intervention data.
- Long-term safety and feasibility of targeting myoglobin in adipose tissue are unknown.
Future Directions: Develop pharmacologic or gene-based approaches to upregulate adipocyte myoglobin; test metabolic efficacy and safety in large-animal models and early-phase human studies.
Nutritional overflow and a positive energy balance are hallmarks of metabolic diseases including obesity and type 2 diabetes. Brown and beige adipocytes maintain systemic metabolic homeostasis by clearing and oxidizing energy-rich nutrients during thermogenic activation. Myoglobin (MB) is classically regarded as a muscle-associated oxygen-binding protein; however it is also expressed in brown and beige adipocytes, where it contributes to intracellular lipid handling and oxidative metabolism. Here, we report that loss of MB exclusively in adipose tissue (AT) lowers whole-body energy expenditure, impairs thermoregulation, and increases susceptibility to diet-induced obesity. AT-specific MB knockout (ATMBKO) mice exhibit elevated circulating triglycerides (TG) and fatty acids, indicating defective lipid clearance and utilization. Omics analyses reveal coordinated downregulation of oxidative phosphorylation, fatty acid metabolism, and myogenic programs. Conversely, restoration of MB in MB knockout (MBKO) mice improves metabolism in vivo. MB expression determines the capacity for mitochondrial fatty acid oxidation in brown adipocytes, whereas MB overexpression in primary human white adipocytes enhances thermogenic activity, confirming functional relevance of MB in human AT. Together, these findings establish MB as a key determinant of thermogenic lipid metabolism and energy expenditure in vivo and increasing adipocyte MB expression could increase energy expenditure and complement obesity treatment strategies.