Breaking Therapy Resistance! CRISPR Transcription Factor Library Identifies SMAD1 as a Key Driver of Paneth-like Transition in Colorectal Cancer

Introduction

KRAS mutations occur in up to 50% of colorectal cancers (CRC). Although dual targeting of KRAS and EGFR has shown objective response rates of 34-46%, therapeutic resistance remains a major limitation, with most patients relapsing within months. In November 2025, the team led by Yi-Jun Gao at Sun Yat-sen University Cancer Center and collaborators published a study in Cancer Cell reporting new insights into resistance mechanisms.

By integrating multiple experimental models and clinical samples, the study employed a CRISPR transcription factor library screen to identify SMAD1 as a core regulatory factor. For the first time, it was shown that CRC cells undergo a Paneth-like cellular transition under therapeutic pressure, enabling immune evasion. The research further revealed that the SMAD1-FGFR3 signaling axis drives this lineage plasticity and MAPK pathway reactivation, and that FGFR3 inhibition can block the Paneth-like transition and restore treatment sensitivity.This work demonstrates that CRISPR screening can pinpoint precise targets to overcome drug resistance, providing a novel targeted strategy to address therapy-refractory KRAS-mutant CRC with substantial translational potential.

Research Background

KRAS-mutant colorectal cancer (CRC) has long faced limited treatment options and poor prognosis. In recent years, combination therapies involving KRAS inhibitors (e.g., G12C inhibitor sotorasib, G12D inhibitor MRTX1133) and EGFR-targeting antibodies have emerged as breakthroughs, yet the mechanisms underlying therapeutic resistance remain incompletely understood.

Known resistance mechanisms include secondary KRAS mutations and alterations in upstream or downstream signaling genes at the genomic level. However, approximately one-third of relapsed tumors lack overt genomic alterations, suggesting that non-genetic mechanisms, such as lineage plasticity, may play a pivotal role. Intestinal epithelial cells exhibit high plasticity, and CRC cells can similarly reprogram their state to evade therapeutic pressure. Nevertheless, the specific forms and molecular mechanisms of lineage transitions induced by KRAS/EGFR dual-targeted therapy remain unclear and warrant further investigation.

Research Objective

This study aims to determine whether CRC cells develop resistance to KRAS/EGFR dual-targeted therapy through lineage plasticity, to characterize the specific cellular state transitions and key regulatory mechanisms involved, and to evaluate whether targeting these mechanisms can reverse resistance, ultimately providing novel therapeutic strategies for KRAS-mutant colorectal cancer. I can continue translating the next sections—research methods, workflow, and key findings—in the same high-level, publication-style English, keeping terminology precise for a website or technical news release.

Research Methods

• Experimental Models: The study employed iKAP KRAS G12D genetically engineered mouse models, KRAS G12D/G12C-mutant CRC cell lines (e.g., LS174T, SW1463), patient-derived organoids (PDOs), and patient-derived xenografts (PDXs), complemented by paired clinical biopsy samples.
• Lineage Tracing: A DEFA5 promoter-driven Cre-LoxP recombination system (dsRed-STOP-eGFP) and a CRISPR-Cas9-mediated DEFA5-CopGFP knock-in system were constructed to track the origin and dynamic transitions of Paneth-like cells.
• CRISPR Screening: A customized CRISPR knockout library targeting 1,741 human transcription factors was used to identify key regulators driving Paneth-like cellular transition.
• Molecular Mechanism Validation: Gene expression changes were analyzed via RNA-seq and single-cell RNA-seq (scRNA-seq). ChIP-qPCR confirmed SMAD1 binding to the FGFR3 promoter. Protein expression and localization were assessed by Western blot and immunofluorescence, while reporter assays were used to examine DNA repair pathway activity.
• Pharmacological Interventions: KRAS inhibitors (MRTX1133, sotorasib), EGFR antibodies (cetuximab), and FGFR inhibitors (futibatinib, infigratinib) were tested in cell lines, organoids, and animal models to evaluate single-agent and combination therapy effects.
• Clinical Sample Validation: Biopsy samples from two CRC patients receiving KRAS/EGFR dual-targeted therapy were analyzed to assess expression of Paneth-like cell markers before and after treatment.

Research Workflow

• Resistance Phenotype Identification: In GEMM, PDO, and PDX models, residual tumors following KRAS/EGFR dual-targeted therapy were analyzed to characterize cellular state changes, revealing enrichment of Paneth-like cells.
• Lineage Plasticity Validation: Lineage tracing experiments confirmed that Paneth-like cells arise from CRC cell transdifferentiation, rather than expansion of pre-existing Paneth-like cells.
• Key Regulator Screening: A CRISPR transcription factor library screen, integrated with public cohort data, identified SMAD1 as the core driver of Paneth-like cellular transition.
• Molecular Mechanism Dissection: Validated the role of SMAD1 in regulating Paneth-like transition and therapy resistance; Identified FGFR3 as a downstream target of SMAD1 and elucidated the activation mechanism of the SMAD1–FGFR3 signaling axis; Demonstrated that FGFR3 mediates resistance via MAPK pathway reactivation.
• Therapeutic Strategy Validation: In cell lines, organoids, and animal models, the combination of FGFR inhibitors with KRAS/EGFR dual-targeted therapy was evaluated, confirming synergistic anti-tumor efficacy.
• Clinical Relevance Verification: Analysis of paired patient samples confirmed the enrichment of Paneth-like transition in clinically resistant tumors, supporting the translational significance of these findings.

Key Finding

• KRAS/EGFR Dual-Targeted Therapy Induces Paneth-like Transition in CRC Cells Transcriptomic analysis of iKAP mouse models treated with MRTX1133 plus cetuximab revealed that residual tumors were significantly enriched for Paneth cell-related pathways, including defensins and antimicrobial peptides. GSEA analysis confirmed that Paneth cell signature genes represented the most prominently enriched epithelial lineage feature post-treatment (NES = 2.4, adjusted p < 0.001), with marked upregulation of DEFA5, DEFA6, and other Paneth markers.

  Immunofluorescence analysis showed a clear increase in the proportion of DEFA5⁺ Paneth-like cells in post-treatment iKAP tumors, LS174T 3D spheroids, and KRAS G12D/G12C patient-derived organoids (PDOs). Single-cell RNA-seq (scRNA-seq) further revealed that CRC cells initially enter a dormancy-like drug-tolerant persister (DTP) state following therapy, and subsequently differentiate into Paneth-like cells (C5 cluster), which progressively expand over the course of treatment.

  

  Figure 1. Enrichment of Paneth-like cell states in residual colorectal cancer (CRC) lesions following KRAS-EGFR inhibition

• Paneth-like Transition Mediates Therapy Resistance via Transdifferentiation Lineage tracing revealed that prior to treatment, only 0.23% of LS174T cells expressed eGFP, a Paneth-like cell marker. After 21 days of KRAS/EGFR dual-targeted therapy, this proportion increased to 32.26%, confirming that Paneth-like cells arise from transdifferentiation of CRC cells rather than expansion of pre-existing Paneth-like cells.Upon drug withdrawal, the proportion of Paneth-like cells rapidly declined, indicating that this state is maintained by therapeutic pressure. Moreover, sorted eGFP⁺ Paneth-like cells exhibited significantly reduced sensitivity to dual-targeted therapy compared with eGFP⁻ cells.

  Using a DEFA5 promoter-driven iCasp9 system to selectively ablate Paneth-like cells, tumor sensitivity to MRTX1133 plus cetuximab was markedly restored, confirming that Paneth-like cells constitute a core resistant population.Mechanistically, Paneth-like cells displayed higher KRAS-GTP levels and elevated pERK expression, indicating that MAPK pathway reactivation is a key driver of their therapy resistance.

  

  Figure 2. Paneth-like cells arise via transdifferentiation in response to KRAS and EGFR dual inhibition

• CRISPR Screening Identifies SMAD1 as a Key Driver of Paneth-like Transition A CRISPR knockout screen targeting 1,741 transcription factors revealed 34 genes whose loss increased tumor sensitivity to KRAS/EGFR dual-targeted therapy. Among these, SMAD1 showed a strong positive correlation with Paneth-like cell signatures in public CRC cohorts (r = 0.66, p = 9.1 × 10⁻⁵). Following dual-targeted therapy, SMAD1 transcription levels were significantly upregulated in LS174T spheroids and KRAS G12D/G12C PDOs. Furthermore, SMAD1 co-localized with the Paneth-like cell marker DEFA5, supporting its role as a core regulator of Paneth-like cellular transition.

  

  Figure 3. CRISPR screening identifies SMAD1 as a driver of Paneth-like cell state transition and therapy resistance

• SMAD1 Mediates Therapy Resistance by Driving Paneth-like Transition SMAD1 knockout markedly increased the sensitivity of LS174T and LS180 cells to KRAS/EGFR dual-targeted therapy, reducing colony formation, and in in vivo xenograft models, 70-90% of tumors regressed. Competitive growth assays demonstrated that EGFP⁺ SMAD1-deficient cells were significantly depleted under therapeutic pressure, indicating that SMAD1 loss compromises cellular tolerance to dual-targeted therapy.Mechanistically, SMAD1 knockout suppressed therapy-induced expression of DEFA5 and DEFA6 and blocked the formation of Paneth-like cells, whereas SMAD1 overexpression upregulated Paneth cell markers and enhanced therapy resistance. Importantly, SMAD1's function does not depend on the canonical BMP-SMAD4 pathway, as SMAD4 knockout did not affect Paneth-like transition or treatment sensitivity.

  

  Figure 4. SMAD1 drives resistance to KRAS–EGFR dual-targeted therapy by promoting Paneth-like state transition

• The SMAD1-FGFR3 Signaling Axis Drives Paneth-like Transition and MAPK Reactivation Pathway enrichment analysis revealed that SMAD1-associated genes are concentrated in FGFR signaling pathways. Following KRAS/EGFR dual-targeted therapy, FGFR3, but not FGFR1, showed significant upregulation in both transcription and phosphorylation, and its expression was positively correlated with SMAD1.ChIP-qPCR confirmed that SMAD1 directly binds the FGFR3 promoter, regulating its transcription. FGFR3 knockout or treatment with the FGFR inhibitor futibatinib significantly suppressed therapy-induced DEFA5 and DEFA6 expression, thereby blocking Paneth-like transdifferentiation. FGFR3 inhibition also reduced pERK levels in Paneth-like cells, restoring their sensitivity to KRAS/EGFR dual-targeted therapy.

  In vivo, in the iKAP mouse model, the combination of MRTX1133, cetuximab, and futibatinib fully reverted tumors to benign adenomas, demonstrating the therapeutic potential of targeting the SMAD1-FGFR3 axis to overcome resistance.

  

  Figure 5. The SMAD1-FGFR3 signaling axis drives Paneth-like cell state transition and MAPK reactivation in Paneth-like cells

• FGFR3 Inhibition Reverses Resistance in Clinically Relevant Models In KRAS G12D PDX models, KRAS/EGFR dual-targeted therapy achieved only transient tumor stabilization, whereas the addition of futibatinib induced sustained tumor regression and significantly reduced the proportion of DEFA5⁺ Paneth-like cells.Clinical validation: Analysis of two CRC patients receiving KRAS/EGFR dual-targeted therapy revealed that residual tumors were enriched in DEFA5⁺ Paneth-like cells, confirming that this resistance mechanism is conserved in human CRC.

  

  Figure 6. Enrichment of Paneth-like cell states in residual human colorectal cancer tumors following KRAS–EGFR dual inhibition

Summary

This study systematically demonstrates, using multi-level experimental models and clinical samples, that Paneth-like transdifferentiation is a core mechanism underlying resistance of KRAS-mutant CRC to KRAS/EGFR dual-targeted therapy. The SMAD1-FGFR3 signaling axis was identified as a key driver of this lineage plasticity and MAPK pathway reactivation.Preclinical data show that FGFR3 inhibition effectively blocks Paneth-like transdifferentiation, restoring tumor sensitivity to dual-targeted therapy. Importantly, this resistance mechanism is conserved in human CRC.

These findings reveal a novel non-genetic mechanism of CRC therapy resistance and provide a clinically actionable therapeutic strategy, with the potential to significantly improve treatment outcomes and prognosis for patients with KRAS-mutant colorectal cancer.

Ubigene CRISPR-iScreen™ Library

CRISPR-iScreen™ is an innovative, in-house developed platform by Ubigene, designed to enable high-efficiency CRISPR-based functional screening.Ubigene currently offers ver 40 ready-to-ship CRISPR libraries, covering genome-wide knockout, inhibition, and activation libraries for human, mouse, green monkey, and pig. In addition, sub-libraries targeting kinases, cell cycle regulators, membrane proteins, metabolism-related genes, and other functional categories are also available, enabling streamlined target discovery and functional screening.Our inventory includes 150+ CRISPR library viruses and 400+ CRISPR library cell pools, ready to support your research at any time.

Ubigene also provides one-stop CRISPR library screening services (in vitro and in vivo), integrated with the iScreenAnlys™ CRISPR Library Analysis platform. Users can select from three widely used algorithms—Drug-Z, MAGeCK-MLE, and MAGeCK-RRA—for data analysis, allowing one-click processing and interpretation of screening results. Contact us to accelerate your functional genomics research with CRISPR-iScreen™.

Reference

Zhang Y, Chen J, She Y, Fang Z, Zhang Y, Ruan D, Guo W, Liao J, Zhou W, Lao J, Fang W, Pan X, Kang W, Wang Z, Wu Y, Deng R, Tian L, Wang L, Huang H, Zheng J, Yan Y, Lu H, Wang R, Yaeger R, Zhao Q, Liao W, Wang F, Gao Y. Paneth-like transition drives resistance to dual targeting of KRAS and EGFR in colorectal cancer. Cancer Cell. 2025 Nov 13:S1535-6108(25)00451-9. doi: 10.1016/j.ccell.2025.10.010. Epub ahead of print. PMID: 41237766.