Daily Endocrinology Research Analysis
Three studies push endocrinology forward across mechanism, prediction, and translation: a Science Advances study delineates transcriptome signatures linked to functional recovery of human islet β cells in type 2 diabetes remission; a mechanistic AJP-Endocrinology paper shows cAMP-pathway rescue of aquaporin-2 trafficking in a novel AVPR2 truncation causing nephrogenic diabetes insipidus; and a JMIR Diabetes study delivers CGM-only machine-learning models that accurately forecast exercise-related
Summary
Three studies push endocrinology forward across mechanism, prediction, and translation: a Science Advances study delineates transcriptome signatures linked to functional recovery of human islet β cells in type 2 diabetes remission; a mechanistic AJP-Endocrinology paper shows cAMP-pathway rescue of aquaporin-2 trafficking in a novel AVPR2 truncation causing nephrogenic diabetes insipidus; and a JMIR Diabetes study delivers CGM-only machine-learning models that accurately forecast exercise-related hypo- and hyperglycemia in type 1 diabetes.
Research Themes
- Beta-cell plasticity and transcriptomic signatures in T2D remission
- Precision endocrinology via receptor-bypass therapy in nephrogenic diabetes insipidus
- AI-driven decision support to prevent exercise-related glycemic events in type 1 diabetes
Selected Articles
1. Functional recovery of islet β cells in human type 2 diabetes: Transcriptome signatures unveil therapeutic approaches.
This mechanistic study investigates human islets to define transcriptome signatures associated with functional recovery of β cells in the context of type 2 diabetes remission after diet, surgery, or pharmacotherapy. The signatures are positioned to unveil therapeutic approaches that promote β-cell recovery.
Impact: It provides a human tissue-based molecular map of β-cell recovery, a critical step toward disease-modifying therapies in T2D. The transcriptomic framework can guide target discovery and translational strategies.
Clinical Implications: Although preclinical, the identified β-cell recovery signatures could prioritize therapeutic targets and biomarkers for interventions aiming at T2D remission and sustained β-cell function.
Key Findings
- Defines transcriptome signatures linked to functional recovery of human β cells in T2D remission contexts.
- Anchors therapeutic hypothesis generation for boosting β-cell function based on human islet molecular profiles.
- Positions diet-, surgery-, and drug-induced remission as contexts to study β-cell plasticity and repair programs.
Methodological Strengths
- Human islet tissue analysis directly relevant to T2D pathophysiology
- Transcriptomic profiling enabling pathway-level inference for therapeutic targeting
Limitations
- Abstract details on sample size and validation cohorts are not provided in the excerpt
- Translational efficacy requires functional validation and clinical studies
Future Directions: Functionally test candidate pathways and targets emerging from signatures in multi-system models and early-phase clinical studies; develop biomarkers to track β-cell recovery in remission.
Remission of type 2 diabetes (T2D) can occur after hypocaloric diet, bariatric surgery, or pharmacological treatments and associates with improved β cell function. Here, we studied islets from nondiabetic (
2. Functional characterization and cAMP-mediated rescue of a novel truncating AVPR2 mutation causing nephrogenic diabetes insipidus.
A novel truncating AVPR2 mutation (D191) causes receptor misprocessing and loss of cAMP signaling, blocking AQP2 trafficking and water reabsorption. Pharmacologic activation of the cAMP pathway (forskolin or 8-Br-cAMP) bypassed the defective receptor to restore downstream signaling and AQP2 apical localization.*
Impact: Reveals a precise pathomechanism for an AVPR2 truncation and demonstrates a receptor-bypass strategy that could be generalized to receptor loss-of-function disorders.
Clinical Implications: For NDI with nonfunctional AVPR2, therapies directly enhancing intracellular cAMP may restore AQP2 trafficking and water reabsorption independent of AVPR2. This supports precision diagnostics to guide pathway-targeted treatments.
Key Findings
- Identified a novel truncating AVPR2 variant (c.570dup; D191*) in pediatric NDI.
- Mutant receptor showed intracellular retention, rapid proteasomal degradation, absent cAMP response to dDAVP, and impaired AQP2 translocation.
- Forskolin or 8-bromo-cAMP restored cAMP signaling and rescued AQP2 apical localization independent of AVPR2.
Methodological Strengths
- Integrated molecular, signaling, and trafficking assays in a renal epithelial model
- Mechanistic rescue experiments pinpointing pathway-level therapeutic leverage (cAMP activation)
Limitations
- Single-patient mutation discovery with in vitro validation; no clinical trial of cAMP-directed therapy
- Rescue shown in cell systems; in vivo efficacy and safety are untested
Future Directions: Translate cAMP-bypass strategies to in vivo models and evaluate pharmacologic agents (e.g., PDE inhibitors, V2-independent cAMP enhancers) for NDI; explore allele-specific chaperones where residual AVPR2 function exists.
Vasopressin plays a central endocrine role in water homeostasis by activating the arginine vasopressin receptor 2 (AVPR2) receptor in renal collecting duct cells. Mutations in AVPR2 are a leading cause of X-linked nephrogenic diabetes insipidus (NDI), a disorder marked by renal insensitivity to vasopressin, leading to polyuria, polydipsia, and hypernatremia. We identified a novel truncating AVPR2 mutation (c.570dup; D191*) in a pediatric patient with NDI and investigated its molecular and functional consequences using a renal epithelial cell model. The D191* mutant exhibited marked reduction in total and surface receptor expression due to intracellular retention and rapid proteasomal degradation. Functional assays revealed that 1-deamino-8-d-arginine vasopressin (dDAVP) stimulation failed to elicit cAMP production or activate downstream signaling targets, including CREB and ERK1/2, in cells expressing the mutant receptor. Aquaporin-2 (AQP2) membrane translocation, essential for water reabsorption, was also impaired. Notably, treatment with forskolin or 8-bromo-cAMP restored cAMP levels, reactivated downstream signaling, and rescued AQP2 localization to the apical membrane, independent of AVPR2 activation. These findings uncover the pathophysiological mechanism by which D191* impairs vasopressin signaling and suggest that bypassing the receptor via direct cAMP pathway activation offers a promising therapeutic strategy for NDI. This study highlights the endocrine relevance of precision molecular diagnostics and supports functional rescue approaches for receptor-based disorders.
3. Managing Exercise-Related Glycemic Events in Type 1 Diabetes: Development and Validation of Predictive Models for a Practical Decision Support Tool.
Using 1901 exercise episodes from 329 adults with type 1 diabetes, nested cross-validated ML models achieved AUROCs of 0.88–0.99 for predicting hypo- and hyperglycemia during and after exercise. Critically, CGM-only models matched multi-modal models, were well-calibrated (Brier ≤0.08), and robust to noise, minimizing user burden for decision support.
Impact: Demonstrates that automatically captured CGM streams alone suffice for accurate, calibrated risk prediction of exercise-related glycemic events, enabling scalable, low-friction decision support tools.
Clinical Implications: Clinicians and patients can rely on CGM-only decision support to tailor insulin and carbohydrate strategies around exercise without manual entry of diet/exercise data, potentially improving safety and adherence to physical activity.
Key Findings
- Models achieved cross-validated AUROCs of 0.880–0.992 for predicting glycemic events during and 1 hour after exercise.
- CGM-only models performed statistically indistinguishably from models using demographics, insulin/carbs, and exercise features.
- CGM-only models were well-calibrated (Brier score ≤0.08) and robust to noisy inputs.
Methodological Strengths
- Repeated stratified nested cross-validation with calibration assessment
- Head-to-head modality contribution analysis informing deployment cost–benefit
Limitations
- Sample skewed toward White (94.5%) and female (74.8%), limiting generalizability
- External prospective validation and real-time clinical impact trials are needed
Future Directions: Prospective external validation, integration into closed-loop systems, and randomized trials to test behavioral and glycemic outcomes with CGM-only decision support.
BACKGROUND: Exercise is an important aspect of diabetes self-management. Patients with type 1 diabetes frequently struggle with exercise-induced hyperglycemia and hypoglycemia, decreasing their willingness to exercise. OBJECTIVE: We aim to build accurate and easy-to-deploy models to forecast exercise-induced glycemic events in real-world settings. METHODS: We analyzed free-living data from the Type 1 Diabetes Exercise Initiative study, where adults with type 1 diabetes wore a continuous glucose monitor (CGM) while performing video-guided exercises (30-minute exercises at least 6 times over 4 weeks), along with concurrent detailed phenotyping of their insulin program and diet. We built models to predict glycemic events (blood glucose ≤54 mg/dL, ≤70 mg/dL, ≥200 mg/dL, and ≥250 mg/dL) during and 1 hour post exercise with variables from 4 data modalities, such as demographic and clinical (eg, glycated hemoglobin; CGM (blood glucose value and their summary statistics); carbohydrate intake and insulin administration; and exercise type, duration, and intensity. We used repeated stratified nested cross-validation for model selection and performance estimation. We evaluated the relative contribution of the 4 input data modalities for predicting glycemic events, which informs the cost and benefit for including them in the decision support tool for risk prediction. We also evaluated other important aspects related to model translation into decision support tools, including model calibration and sensitivity to noisy inputs. RESULTS: Our models were built based on 1901 exercise episodes for 329 participants. The median age for the participants was 34 (IQR 26-48) years. Of the participants, 74.8% (246/329) are female and 94.5% (311/329) are White. A total of 182/329 (55.3%) participants used a closed-loop insulin delivery system, while the rest used a pump without a closed-loop system. Models incorporating information from all 4 data modalities showed excellent predictive performance with cross-validated area under the receiver operating curves (AUROCs) ranging from mean 0.880 (SD 0.057) to mean 0.992 (SD 0.001) for different glycemic events. Models built with CGM data alone have statistically indistinguishable performance compared to models using all data modalities, indicating the other 3 data modalities do not add additional information with respect to predicting exercise-related glycemic events. The models based solely on CGM data also showed outstanding calibration (Brier score ≤0.08) and resilience to noisy input. CONCLUSIONS: We successfully constructed models to forecast exercise-induced glycemic events using only CGM data as input with excellent predictive performance, calibration, and robustness. In addition, these models are based on automatically captured CGM data, thus easy to deploy and maintain and incurring minimal user burden, enabling model translation into a decision support tool.