Systematic Single-Cell Dissection of Cell Cycle and TGFβ-Induced State Transitions Underlying Gemcitabine Resistance in 3D Pancreatic Tumor Tissue Models

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background: Pancreatic ductal adenocarcinoma (PDAC) is highly lethal due to late diagnosis and rapid resistance to chemotherapy (e.g. Gemcitabine, GEM) with poor outcome. We investigate resistance mechanisms in a 3D tissue model. Methods: We established a three dimensional PANC-1 model on a decellularized porcine intestinal tissue matrix, recapitulating patient-like features. Niche-specific drug response and resistance induction after TGF-β1 stimulation motivated single-cell RNA sequencing (scRNA-seq) and systems biological analysis comparing GEM treatment with or without TGF-β1 stimulation. Results: Moderate GEM resistance was associated with E2F1, mTOR, and checkpoint signaling, including upregulation of CDK1, AURKA, TPX2, TOP2A, and BIRC5. TGF-β1 promoted additional resistance through EMT, KRAS signaling, glycolysis, and hypoxia pathways, with early EMT drivers followed by late induction of SPOCK1, MBOAT2, COL5A1, ADAMTS6, THBS1, and FN1. Decision level and trajectory analysis resolved G1–S progression and TGF-β1–induced EMT and revealed enrichment of GEM-surviving cells along a unique bottleneck centered on the CDK1–CDKN1A (p21) axis. This "S-phase persistence" state represented proliferative drive (CDK1–Wee1) counterbalanced by TGF-β1/SMAD3 checkpoint signaling, producing a poised but arrested phenotype. Hybrid EMT/S-phase–persistent cells co-expressed EMT markers and cell-cycle regulators (e.g., RRM1/2, MYBL2, CLSPN, DTL), conferring replication-stress tolerance and hallmarks of resistance. Conclusions: Our study maps the dynamic emergence of GEM resistance in PDAC at single-cell resolution. Beyond PDAC, this work underscores the value of matrix-based 3D scRNA-seq models and provides a generalizable framework for dissecting cancer resistance.

Article activity feed