Decision-Gated In Silico Framework and Prototype Chemotype for Molecular Glue Stabilization of the PDX1–MAFA–NKX6.1 Complex in Type 2 Diabetes

Abstract Introduction/Aims : Loss of mature pancreatic β-cell identity under chronic glucolipotoxicity and oxidative stress drives progressive insulin secretory failure in type 2 diabetes. Mature identity depends on cooperative nuclear activity of PDX1, MAFA, and NKX6.1, which undergo cytoplasmic mislocalization and turnover under metabolic stress. Because transcription factors lack deep catalytic pockets, no drug class stabilizes this assembly. We aimed to establish a decision-gated computational framework to nominate small-molecule stabilizers (molecular glues) of the PDX1–MAFA–NKX6.1 complex. Methods: We integrated human islet pathology, murine lineage-tracing evidence, and structural bioinformatics into a five-stage, gate-controlled pipeline targeting human PDX1 (P52945), MAFA (Q8NHW3), and NKX6.1 (P78426). The cascade comprised: (1) multimer modeling and explicit-solvent molecular dynamics benchmarked against scrambled-interface controls (Gate G1); (2) consensus pocket detection (Gate G2); (3) hierarchical virtual screening with free-energy and cooperativity rescoring, ΔΔG_coop (Gate G3); and (4) physicochemical, PAINS, and homeodomain/bZIP selectivity filtering (Gate G4), culminating in prospective biophysical and human β-cell assays (Gate G5). Using RDKit, we generated a reference chemotype to parameterize pharmacophoric boundaries. Results: Evidence grading confirmed strong human (Tier 1) and genetic (Tier 2) validation for the triad but absent druggability evidence (Tier 3). The protocol defined preregistered stopping rules and three falsifiable predictions paired with SPR/ITC, NanoBRET, and GSIS assays, anchored by a PAINS-free prototype scaffold (C22H25N5O; MW 375.48 Da; cLogP 2.97; QED 0.745). Conclusions: This framework converts β-cell identity preservation into an auditable, falsifiable discovery pipeline enforcing prescreening model deposition and cooperativity gates before nomination.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-17
DOI
https://doi.org/10.21203/rs.3.rs-11750670/v1
Primary Topic
Pancreatic function and diabetes
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Decision-Gated In Silico Framework and Prototype Chemotype for Molecular Glue Stabilization of the PDX1–MAFA–NKX6.1 Complex in Type 2 Diabetes

Gabriela Correia Matos de Oliveira, Luís Jesuíno de Oliveira Andrade, Bittencourt Al, Luís Jesuino De Oliveira Andrade et al.
1 links
Research Square
Pancreatic function and diabetes
preprint

Decision-Gated In Silico Framework and Prototype Chemotype for Molecular Glue Stabilization of the PDX1–MAFA–NKX6.1 Complex in Type 2 Diabetes

Gabriela Correia Matos de Oliveira, Luís Jesuíno de Oliveira Andrade, Bittencourt Al, Luís Jesuino De Oliveira Andrade, Osmário Jorge de Mattos Salles
preprint en

Abstract

Abstract Introduction/Aims : Loss of mature pancreatic β-cell identity under chronic glucolipotoxicity and oxidative stress drives progressive insulin secretory failure in type 2 diabetes. Mature identity depends on cooperative nuclear activity of PDX1, MAFA, and NKX6.1, which undergo cytoplasmic mislocalization and turnover under metabolic stress. Because transcription factors lack deep catalytic pockets, no drug class stabilizes this assembly. We aimed to establish a decision-gated computational framework to nominate small-molecule stabilizers (molecular glues) of the PDX1–MAFA–NKX6.1 complex. Methods: We integrated human islet pathology, murine lineage-tracing evidence, and structural bioinformatics into a five-stage, gate-controlled pipeline targeting human PDX1 (P52945), MAFA (Q8NHW3), and NKX6.1 (P78426). The cascade comprised: (1) multimer modeling and explicit-solvent molecular dynamics benchmarked against scrambled-interface controls (Gate G1); (2) consensus pocket detection (Gate G2); (3) hierarchical virtual screening with free-energy and cooperativity rescoring, ΔΔG_coop (Gate G3); and (4) physicochemical, PAINS, and homeodomain/bZIP selectivity filtering (Gate G4), culminating in prospective biophysical and human β-cell assays (Gate G5). Using RDKit, we generated a reference chemotype to parameterize pharmacophoric boundaries. Results: Evidence grading confirmed strong human (Tier 1) and genetic (Tier 2) validation for the triad but absent druggability evidence (Tier 3). The protocol defined preregistered stopping rules and three falsifiable predictions paired with SPR/ITC, NanoBRET, and GSIS assays, anchored by a PAINS-free prototype scaffold (C22H25N5O; MW 375.48 Da; cLogP 2.97; QED 0.745). Conclusions: This framework converts β-cell identity preservation into an auditable, falsifiable discovery pipeline enforcing prescreening model deposition and cooperativity gates before nomination.

Research Square
Universidade Federal da Bahia (BR), Escola Bahiana de Medicina e Saúde Pública (BR), Universidade de Ribeirão Preto (BR), Universidade Estadual de Santa Cruz (BR)
Good health and well-being
Pancreatic function and diabetes
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon