KRAS-mediated upregulation of CIP2A promotes suppression of PP2A-B56α to initiate pancreatic cancer development

Abstract

Oncogenic mutations in KRAS are present in ~95% of patients diagnosed with pancreatic ductal adenocarcinoma (PDAC) and are considered the initiating event of pancreatic intraepithelial neoplasia (PanIN) precursor lesions. While it is well established that KRAS mutations drive the activation of oncogenic kinase cascades during pancreatic oncogenesis, the effects of oncogenic KRAS signaling on regulation of phosphatases during this process is not fully appreciated. Protein Phosphatase 2A (PP2A) has been implicated in suppressing KRAS-driven cellular transformation and low PP2A activity is observed in PDAC cells compared to non-transformed cells, suggesting that suppression of PP2A activity is an important step in the overall development of PDAC. In the current study, we demonstrate that KRAS^G12D induces the expression of an endogenous inhibitor of PP2A activity, Cancerous Inhibitor of PP2A (CIP2A), and phosphorylation of the PP2A substrate, c-MYC. Consistent with these findings, KRAS^G12D sequestered the specific PP2A subunit responsible for c-MYC degradation, B56α, away from the active PP2A holoenzyme in a CIP2A-dependent manner. During PDAC initiation in vivo, knockout of B56α promoted KRAS^G12D tumorigenesis by accelerating acinar-to-ductal metaplasia (ADM) and the formation of PanIN lesions. The process of ADM was attenuated ex vivo in response to pharmacological re-activation of PP2A utilizing direct small molecule activators of PP2A (SMAPs). Together, our results suggest that suppression of PP2A-B56α through KRAS signaling can promote the MYC-driven initiation of pancreatic tumorigenesis.

Fig. 1. Oncogenic KRAS signaling promotes suppression of the PP2A-B56α complex.

A Representative western blot of human pancreatic ductal epithelial (HPDE) cells with tet-inducible KRAS^G12D (HPDE-iKRAS) treated with 50 ng/mL doxycycline for 0–12H in a time course. B–D Quantification of CIP2A, MYC, and pERK/tERK protein expression at 0, 8, 12H post-KRAS^G12D induction in HPDE-iKRAS cells. E Quantification of average puncta per cell of PLA between B56α and PP2Ac after 0H (OFF) or 8H (ON) of KRAS^G12D induction and treated with siNT siRNA or siRNA to CIP2A. Quantification is normalized to the average signal from the siNT OFF condition. F Representative images of PLA quantified in E. Scale bars represent 10 μm, white dashed lines represent edges of cells as identified through actin staining. G Representative images of three independent biological replicate stains of WT, 1MO, 2MO, and 5MO KC pancreatic tissue (left to right). IF stains indicate PanCK, CIP2A, pS62 MYC, and Merge of all stains (top to bottom). Scale bar represents 100 μm. All experiments were repeated as three independent biological replicates and statistical analysis was performed using one-way ANOVA. *p < 0.05 **p < 0.01 n.s. = not significant.

Fig. 2. Loss of oncogenic KRAS signaling promotes activation of PP2A-B56α.

Representative western blot (A) and quantification (B) of protein signaling in time-matched tet-inducible KRAS^G12D cells after 0–72H with or without Dox (loss of KRAS^G12D) in iKRAS3 cells. C Representative images of pancreatic tissue from iKRAS* mouse models after 3W of Dox-induced KRAS^G12D signaling (top) or after 1W of Dox withdrawal (bottom). Immunofluorescent staining of PanCK, CIP2A, pS62 MYC, and all stains merged (left to right). Statistical analysis was performed using one-way ANOVA. All data are representative of 3–4 biological replicates. *p < 0.05 ***p < 0.001 ****p < 0.0001 n.s. = not significant. Scale bar represents 100 μm.

Fig. 3. Overexpression of B56α abrogates oncogenic phenotypes.

A Soft agar assay of stable iKRAS3 cell line with empty vector control or overexpression of B56α, stained with crystal violet and scanned after two weeks of growth. B Quantification of four biological replicates of the soft agar assay represented as fold change relative to the number (left) and size (right) of colonies formed in the iKRAS3 empty vector control. C Clonogenic assay of iKRAS3 cells with empty vector control or overexpression of B56α stained with crystal violet after six days. D Quantification of three replicates of clonogenic assay, measured by fold change in the percent area covered by cells. Quantified using FIJI. E Growth curve of orthotopic pancreatic allograft of iKRAS3 EV or B56αOE cell lines in 8-week-old NRG mice, volume of tumor in mm³ and +/−SEM. F Endpoint tumor weight in grams of each condition. (EV n = 8, B56α OE n = 9). G Picture of orthotopic tumors at endpoint displayed from largest (right) to smallest (left) in each condition (top: EV, bottom: B56αOE). Scale bar represents 1 cm. H Representative images of immunofluorescent staining from EV (left) or B56αOE (right) tumors. Stains in images represent PanCK, HA-tag, pS62 MYC, and merged images (top to bottom). All experiments were performed at least three times independently and statistical analysis was done using student’s t-test. *p < 0.05 **p < 0.01 ***p < 0.001 ****p < 0.0001 Scale bar represents 100 μm.

Fig. 4. Loss of B56α promotes acceleration of acinar-to-ductal metaplasia.

A Representative images of acinar cells isolated from WT, B56α^hm/hm, KC, and KCB^hm/hm genetic mouse models cultured in collagen for 1d, 2d, and 3d. Arrows indicate duct-like structures. Inset represents digital zoom of acinar and duct-like structures. B Time course of acinar-to-ductal transdifferentiation kinetics in each condition over three days, quantified every 6H. Time course includes data from one representative biological replicate. C mRNA expression of ductal cell marker, ck19, in each genotype at 1d and 2d timepoints. *p < 0.05 **p < 0.01 ***p < 0.001 ****p < 0.0001 Scale bar represents 100 μm.

Fig. 5. Loss of B56α leads to altered chromatin accessibility and transcriptional programs associated with oncogenesis.

A Volcano plot of DEGs in 2d KCB^hm/hm condition compared to 2d KC condition. Log2 FC indicates the mean expression level for each gene. Each dot represents one gene. B Pathways identified from upregulated DEGs in 2d KCB^hm/hm versus 2d KC condition. C Gene Set Enrichment Analysis (GSEA) of MYC regulated genes comparing KCB^hm/hm to KC conditions at 2d. Gene set derived from Muthalagu et al. NES: normalized enrichment score. FDR: false discovery rate. D Differentially accessible peaks (DAPs) identified in 2d KCB^hm/hm versus 2d KC. E Pathways identified from DAPs as more accessible in KCB^hm/hm at 2d. F Binding motifs identified as more accessible in 2d KCB^hm/hm versus 2d KC ADM. G Venn diagram showing the overlap of MYC ChIP binding sites with DAPs that are differentially more accessible in KCB^hm/hm at 2d. H Pathways identified from the 2183 MYC ChIP binding sites that are more accessible in KCB^hm/hm at 2d compared to KC conditions. Statistical analysis performed using either a t-test of the area under the curve or one-way ANOVA. The padj for pathway analysis was calculated using enrichr for 2020_MSigDB_hallmark gene sets.

Fig. 6. Loss of B56α promotes progression of PDAC precursor lesions.

A Representative 2MO or 5MO timepoints from pancreatic tissue in respective genetic mouse models stained with H&E (top), Alcian Blue (middle), and Masson’s Trichrome (bottom). B–D Quantifications of healthy acinar area, stromal content, and number of PanINs quantified from stains represented in A, respectively, at 2MO and 5MO timepoints. n = 6–9 mice per genotype. E–G Immunofluorescent stains in pancreata harvested 2MO KC or KCB^hm/hm mice. Representative images depicting (E) Amylase/PanCK/CIP2A, (F) Amylase/CIP2A/pS62 MYC, and (G) PanCK/CIP2A/pS62 MYC. n = 4 mice per genotype. *p < 0.05 **p < 0.01 n.s. = not significant Scale bar represents 100 μm.

Fig. 7. Re-activation of PP2A-B56α prevents initiation of pancreatic cancer.

A Representative images of acinar to ductal differentiation in KC acinar cells after 72H in collagen and treated with increasing doses of SMAP, DT061. Scale bar represents 100 μm. B Percentage of duct-like structures over total structures present at each concentration of DT061 treatment. C Average ductal lumen size of duct-like structures present in each condition. D Distribution of the ductal lumen measurements at each concentration of DT061. E–H qPCR of mRNA isolated from the organoids in collagen after 72H of DT061 treatment at each concentration dosed. ADM data representative of one biological replicate. Assay was repeated with same trends in three independent biological replicates. I Survival curve of PDX1-Cre; LSL-Tp53^R172H; LSL-KRAS^G12D (KPC) mice enrolled in a survival experiment when 100 mm³ tumor was identified and then treated twice daily with 15 mg/kg DT061 via oral gavage or with vehicle until mice succumb to PDAC disease burden. *p < 0.05 ***p < 0.001 ****p < 0.0001.