Article Text

Dysregulated KLF4 expression plays a pivotal role in the pathogenesis of pancreatic intraductal papillary mucinous neoplasms
  1. Xiangsheng Zuo1,
  2. Liang Wang2,
  3. Yi Liu1,
  4. Huamin Wang3,
  5. Margarete Hafley2,
  6. Mihai Gagea4,
  7. Ru Chen5,
  8. Yun Xiong6,
  9. Sheng Pan7,
  10. Imad Shureiqi1,8,
  11. Robert S Bresalier2,
  12. Daoyan Wei2
  1. 1Department of Gastrointestinal Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
  2. 2Department of Gastroenterology, Hepatology and Nutrition, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
  3. 3Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
  4. 4Department of Veterinary Medicine and Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
  5. 5Section of Gastroenterology and Hepatology, Department of Medicine, Baylor College of Medicine, Houston, TX, USA
  6. 6Department of Bioinformatics and Computational Biology, and Proteomics Core Facility, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
  7. 7The Brown Foundation Institute of Molecular Medicine, and McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA
  8. 8Departmento of Internal Medicine, Rogel Cancer Center, University of Michigan, Ann Arbor, MI, USA
  1. Correspondence to Dr Daoyan Wei, Department of Gastroenterology, Hepatology, and Nutrition, Unit 1644, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA; dwei{at}; Dr Robert S Bresalier, Department of Gastroenterology, Hepatology, and Nutrition, Unit 1644, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA; rbresali{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Intraductal papillary mucinous neoplasms (IPMNs) carry a significant risk of progressing to invasive pancreatic cancer (PC), yet the lack of sensitive and specific biomarkers hampers accurate prognostication and clinical intervention. The prevalence of somatic KLF4 hotspot mutations, notably KLF4K409Q and KLF4S411Y, in over 50% of cases, predominantly in low-grade (LG) regions of IPMNs, suggests a potential role of KLF4 in IPMN pathogenesis. However, critical knowledge gaps persist regarding KLF4 expression in IPMNs, its causal role in pathogenesis and the impact of mutations on KLF4 function. In this study, we conducted an immunohistochemical analysis to characterise KLF4 expression in IPMNs and PC. We found elevated KLF4 levels in IPMNs but reduced expression in PC. Using an established Klf4 conditional transgenic mouse model, we demonstrated that KLF4 overexpression induced pancreatic IPMN-like cystic lesions under Kras mutant conditions. Gene transduction and functional assays revealed that the KLF4K409Q mutant protein maintains its DNA binding ability to regulate gene expression, while losing the K-site-specific ubiquitination, resulting in a prolonged half-life compared with wild-type KLF4 protein. This suggests a gain-of-function mechanism for the KLF4K409Q mutation. These findings significantly enhance our understanding of IPMN pathogenesis, providing insights for potential interventions and risk assessment in IPMN management.

Fujikura et al1 highlight the high prevalence (>50%) of KLF4 hotspot mutations KLF4K409Q and KLF4S411Y in low-grade (LG) regions of human intraductal papillary mucinous neoplasms (IPMNs). This observation and another report2 suggest the potential involvement of KLF4 in IPMN pathogenesis. However, significant knowledge gaps exist regarding KLF4 expression in IPMNs, its causal role in IPMN pathogenesis, and the impact of somatic mutations on KLF4 protein function.

We conducted immunohistochemical analysis using a highly specific antibody capable of detecting both wild-type and mutant KLF4 proteins (online supplemental figure S1) and found significantly elevated KLF4 expression in both LG and high-grade (HG) IPMNs when compared with adjacent non-tumour or normal tissues (figure 1A,B, online supplemental figure S2A-B), while in invasive pancreatic cancer (PC), especially in poorly differentiated cases, exhibited reduced or absent KLF4 expression (figure 1C,D and online supplemental table S1). Changes in KLF4 expression observed at the protein level mirror those at the mRNA level (figure 1D vs 1E).

Supplemental material

Figure 1

KLF4 IHC staining in human pancreatic tissues with IPMN and invasive PCs. (A–C) Representative images of KLF4 staining in LG-IPMN (A, left panel, tissue scan; right panels, microscopic images of the indicated areas with LG-IPMN lesions (R1, R2) and adjacent normal tissue (R3). Red arrows indicate normal ducts, and yellow arrowhead indicates normal acini) and a case with LG/HG-IPMN (B, left panel, tissue scan; microscopic images of the indicated areas with LG-IPMN (R1), HG-IPMN (R2) lesions and adjacent non-tumour tissue (R3), and in invasive tumour tissues (C, images derived from a TMA: C1, mucinous adenocarcinoma; C2, ductal adenocarcinoma, grade 2; C3, ductal adenocarcinoma, grade 1). (D) Statistical analysis of KLF4 IHC staining in different pancreatic lesions (normal, n=47; LG-IPMN, n=19; HG-IPMN, n=18; invasive PC, n=65). (E) Relative levels of KLF4 mRNA expression in different human pancreatic lesions (data retrieved from GEO public dataset GSE19650/GDS3836). *p<0.05; ****p<0.0001 vs normal tissues or as indicated. (Also refer to online supplemental figure S2 and table S1). HG, high grade; IPMN, intraductal papillary mucinous neoplasm; LG, low grade; PC, pancreatic cancer.

To establish a causal link between KLF4 alterations and IPMN formation, we reused our Klf4 transgenic mice (KLF4OE or KOE) (figure 2A). KLF4 overexpression in PKOE mice results in a significant decrease in pancreatic acini and an increase in ductal-like structures by postnatal day 10 compared with KOE alone3 with the emergence of typical pancreatic cystic lesions in PKOE mice at the age of 15–20 weeks (figure 2B1). Extensive replacement of pancreatic tissues by multiple cystic lesions is observed (figure 2B2-3), with associated acinar-to-ductal metaplasia (ADM) (figure 2B4). Cysts are lined by single-layer epithelial cells expressing cytokeratin 19 (CK-19) with no atypia or dysplasia (figure 2B5).

Figure 2

Effect of KLF4 overexpression on pancreatic lesion development. (A) Diagram illustrating the generation of mouse strains with KLF4 transgenic overexpression. (B) Representative images from a PKOE mouse at 19.4 weeks (B1, gross pancreas, arrows indicate cystic lesions; B2, whole-mount pancreatic tissue section; B3, low magnification of indicated areas showing cystic lesions (black arrows); B4, high magnification of indicated area showing acinar-to-ductal metaplasia (ADM) lesion (green arrow) and B5, image of immunofluorescence (IF) staining of the cystic lesion). (C) Representative images from PRKOE mice at different ages (C1, gross pancreas; C2 and 3, whole-mount pancreatic tissue section and low magnification of indicated area showing cystic lesions from a 16.4-week-old mouse; C4–6, whole-mount pancreatic tissue section (middle panel) and images of low and high magnification of indicated areas showing mixed lesions of IPMN, PanIN and local pancreatic ductal adenocarcinoma (PDAC) from a 22-week-old mouse). (D) Representative images from an SRKOE mouse at 26.1 weeks following tamoxifen (TAM) injection (D1, gross abdominal image at necropsy (arrow indicates cystic lesion in pancreas); D2, whole-mount pancreatic tissue section; D3, low magnification of indicated area showing IPMN-like lesion; D4, high magnification of indicated area showing dysplastic cells (yellow arrow) and D5, image of IF staining of the pancreatic tissue with PanIN and IPMN-like cystic lesion (insert: showing dysplastic cells)).(E–F) Representative images of whole-mount pancreatic tissue section and low magnification of indicated areas from an SR mouse (23.8 weeks) (E) or an SKOE mouse (26.1 weeks) (F) following TAM injection. (G–H) Quantification of percent of pancreatic lesion-affected areas (G) in SRKOE, SR and SKOE mice. **p<0.01, ****p<0.0001. (H) Quantification of incidence of pancreatic lesions in SRKOE, SR, and SKOE mice, *p<0.05, comparison between SR and SRKOE; # p<0.05, comparison between SKOE and SRKOE. ADM, acinar-to-ductal metaplasia; Amy, amylase; CAG, the CMV early enhancer/chicken ß actin (CAG) promoter; CK-19, cytokeratin 19; GFP, green fluorescent protein; IPMN, intraductal papillary mucinous neoplasm; PanIN, pancreatic intraepithelial neoplasm; TAM, tamoxifen.

KLF4OE potentiates mutant Kras-induced pancreatic intraepithelial neoplasia (PanIN) formation in PRKOE mice at a younger age (~9 weeks)3 with a significant increase in the incidence of IPMN-like cystic lesions (>80%) mixed with PanINs and/or localised PC in PRKOE mice at the age around 19–22 weeks (figure 2C) compared with PR mice (<10%). Targeting KLF4OE specifically to ductal epithelial cells through Sox9-CreERT2 4 5 resulted in the development of broad, large duct-interconnecting cystic structures, morphologically consistent with IPMN in SRKOE mice at 24–27 weeks following tamoxifen (TAM) injection at 10 weeks (figure 2 D1-3). Adjacent to the pancreatic cystic lesions, numerous PanIN lesions and residual acinar cells/acini were observed (figure 2 D3-5). SR mice subjected to similar TAM treatment exhibited sporadic ADM and LG PanIN lesions affecting smaller areas, consistent with previous findings (figure 2E).6 The pancreases from SKOE mice appeared nearly normal (figure 2F), suggesting that while KLF4 overexpression alone has minimal impact on pancreatic ductal epithelial cells, it significantly promotes IPMN and PanIN formation under Kras mutant conditions (figure 2G–H). Similar to the findings in human tissues (figure 1), KLF4 protein expression tended to be reduced or lost in PDAC or in poorly differentiated PDAC as the lesion progressed (online supplemental figure S3).

To better understand how the KLF4 hotspot mutation influences IPMN pathogenesis, we conducted gene transduction assays. Transfection of the KLF4K409Q mutant vector led to substantially elevated protein expression in both PANC-1 PC and HPNE normal pancreatic ductal progenitor cells compared with those transfected with KLF4WT vector. The KLF4K409Q vector induced increased CK-19 or reduced CD44 levels in PANC-1 or HPNE cells (online supplemental figure S4). Both KLF4WT and KLF4K409Q proteins recognised and bound similarly to the known KLF4 consensus sequence of its target genes in electrophoresis mobility shift and chromatin immunoprecipitation assays (online supplemental figure S5),7 8 indicating that KLF4K409Q protein retains its binding capacity to regulate related gene expressions.3 8

Considering the relatively short half-life (t1/2) of the KLF4 protein9 and the crucial lysine (K) residues in proteasome-based degradation and protein stability,10 we investigated the half-life of the KLF4K409Q protein, observing a significant prolongation of the half-life of the KLF4K409Q protein (online supplemental figure S6A-B). Further mass spectrometry analysis identified that the K409 residual in KLF4WT protein exhibited ubiquitination while the mutant Q409 residual did not (online supplemental figure 6C-D), suggesting a novel gain-of-function mechanism for the KLF4K409Q mutant protein in LG-IPMN. However, the enigma surrounding the declining incidence of the KLF4K409Q mutation and KLF4 expression in HG-IPMN and invasive PC warrants further exploration.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.


We thank Dr Anirban Maitra for his valuable support and advice. We are grateful to Dr Philip L Lorenzi and Li Li for their valuable advice and technical support for the mass spectrometry analysis. This work was supported by MD Anderson’s Histopathology Core Lab, Award Number P30CA016672 from the NIH National Cancer Institute.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • XZ and LW contributed equally.

  • Contributors Conceptualisation: DW, XZ, LW, RSB; acquisition and analysis of data: LW, DW, MH, XZ, HW, RC, MG, YX; resources and intellectual input and feedback: HW, RC, IS, SP, RSB; supervision and coordination: DW, RSB; writing original draft: DW, XZ, LW. All authors provided discussion, participated in revising the manuscript and agreed to the final version.

  • Funding The work was supported in part by the NIH-funded R01 grants (R01CA236905 to XZ, R01CA195651 to HW and R01CA198090 to RSB) and Texas Medical Center Digestive Disease Center Research Core Center Program P30DK056338 (to RSB and DW), and the grants from The University of Texas MD Anderson Cancer Center Duncan Family Institute for Cancer Prevention and Risk Assessment, the MD Anderson Cancer Center Institutional Research Program, and The Elsa U Pardee Foundation (to DW). This work was also supported by the Sheikh Khalifa bin Zayed Foundation.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.