Stat3 and MMP7 contribute to pancreatic ductal adenocarcinoma initiation and progression

Abstract

Chronic pancreatitis is a well-known risk factor for pancreatic ductal adenocarcinoma (PDA) development in humans, and inflammation promotes PDA initiation and progression in mouse models of the disease. However, the mechanistic link between inflammatory damage and PDA initiation is unclear. Using a Kras-driven mouse model of PDA, we establish that the inflammatory mediator Stat3 is a critical component of spontaneous and pancreatitis-accelerated PDA precursor formation and supports cell proliferation, metaplasia-associated inflammation, and MMP7 expression during neoplastic development. Furthermore, we show that Stat3 signaling enforces MMP7 expression in PDA cells and that MMP7 deletion limits tumor size and metastasis in mice. Finally, we demonstrate that serum MMP7 level in human patients with PDA correlated with metastatic disease and survival.

Figure 1. Persistent Stat3 Activation After Pancreatitis in Setting of Activated Kras

A-H, Immunohistochemistry (IHC) for p-Stat3 in control (A-D) or Ptf1a-Cre; Kras^G12D pancreata (E-H) after PBS and caerulein treatments. (A) PBS. (B) 1 day post-caerulein. (C) 7 days post-caerulein. (D) 21 days post-caerulein. (E) PBS. Stat3 is activated only in the epithelial cells of spontaneously formed PanINs and surrounding stromal cells. (F) 1 day post-caerulein. (G) 7 days post-caerulein. Stat3 activation is maintained in metaplastic ducts in mice with activated Kras. (H) 21 days post-caerulein. Stat3 activation is maintained in PanINs in mice with activated Kras. (I) Stat3 target gene Socs3 expression in pancreata by qPCR after caerulein injection. Socs3 expression is maintained in mice with activated Kras signaling 7 and 21 days after caerulein. Means ± SD (N = 3). * p < 0.05 and ** p < 0.01. (J) Western blot. p-Stat3 is observed only 1 day post-caerulein in control pancreata. (K) Western blot. Sustained aberrant p-Stat3 post-caerulein in Ptf1a-Cre; Kras^G12D pancreata. Scale bars, 50μm. See also Figure S1.

Figure 2. Stat3 Deletion Protects Against Spontaneous and Acute Pancreatitis-Accelerated PanIN Formation

Mice with activated Kras with or without Stat3 signaling (Ptf1a-Cre; Kras^G12D and Ptf1a-Cre; Kras^G12D; Stat3^f/f, respectively), were injected with PBS (A-C) or caerulein (D-V) and sacrificed either 7 or 21 days post-injection. The following stains were performed. (A-B) Alcian blue stain showing spontaneously formed PanINs. (C) Alcian blue positive lesion in total pancreas area of Ptf1a-Cre; Kras^G12D (N = 6) and Ptf1a-Cre; Kras^G12D; Stat3^f/f (N = 3) mice. Means ± SD. ** p < 0.01. (D-G) H&E. Fewer ductal and more acinar structures observed in Stat3-deficient mice. (H-K) Amylase. Stat3-deficient mice retained larger areas of amylase-expressing epithelium. (L-O) CK19. Mice with Stat3-intact display more CK19-positive ductal structures. (P-S) p-Stat3. (T-U) Alcian blue stain, 21 days post caerulein. (V) Alcian blue positive lesion in total pancreas area of Ptf1a-Cre; Kras^G12D and Ptf1a-Cre; Kras^G12D; Stat3^f/f mice. Means ± SD (N = 3). ** p < 0.01. Scale bars, 100μm. See also Figure S2.

Figure 3. Stat3 Supports Cell Proliferation During Kras-driven PanIN Development Following Acute Pancreatitis

Co-staining of Ki67 with amylase (A-F), cytokeratin 19 (I-N), or Alcian blue (Q-V) in pancreata from control, Ptf1a-Cre; Kras^G12D, and Ptf1a-Cre; Kras^G12D; Stat3^f/f mice 2 or 7 days post-injection. (A-C, I-K, Q-S) 2 days post-caerulein. (D-F, L-N, T-V) 7 days post-caerulien. Quantification of double positive cells displaying Ki67 with amylase (G-H), cytokeratin19 (O-P), or Alcian blue (W-X), per 400x field. Means ± SD (N = 3). ** p < 0.01. (G, O, W) 2 days post-caerulein. (H, P, X) 7 days post-caerulein. Scale bars, 50μm.

Figure 4. Elimination of Pancreatic Stat3 Alters Inflammatory Response of Mutant Kras Mice After Injury

(A) The percentage of CD45⁺ immune cells among total live cells present in the pancreas was quantified using flow cytometry. Data from 3 independent experiments are shown using control (N = 9), Ptf1a-Cre; Kras^G12D (N = 8) and Ptf1a-Cre; Kras^G12D; Stat3^f/f (N = 5) mice. (B) Relative expression of inflammatory mediators (IL-6, IL-11, IL-1β, Cox-2, ICAM1, Tnfα, and CCL2) in pancreata as measured by qPCR from control (N = 5), Ptf1a-Cre; Kras^G12D (N = 8), and Ptf1a-Cre; Kras^G12D; Stat3^f/f (N = 8) mice one day after caerulein injection. Means ± SD. (C) Relative expression of inflammatory mediators and Stat3 target genes (IL-6, IL-11, and Cox-2) in cultured primary acini isolated from wild-type and Ptf1a-Cre; Stat3^f/f mice 20 hours after treatment with BSA or 10nM caerulein. Means ± SD (N = 3). * p < 0.05, ** p < 0.01. (Ptf1a-Cre; Kras, Ptf1a-Cre; Kras; Stat3, and Ptf1a-Cre; Stat3 represent Ptf1a-Cre; Kras^G12D, Ptf1a-Cre; Kras^G12D, and Ptf1a-Cre; Kras^G12D; Stat3^f/f mice, respectively). See also Figure S3.

Figure 5. Stat3 Regulates MMP7 Expression During Kras-driven PanIN Development Following Acute Pancreatitis

A-F, IHC for MMP7 in pancreata from control, Ptf1a-Cre; Kras^G12D, and Ptf1a-Cre; Kras^G12D; Stat3^f/f mice, 2 or 7 days post-caerulein injections. (A-C) 2 days post-caerulein. (D-F) 7 days post-caerulein. (G-H) qPCR for MMP7 expression. Means ± SD (N = 3). * p < 0.05. I, Alcian blue positive lesion in total pancreas area of Ptf1a-Cre; Kras^G12D; MMP7^Het (N = 7) and Ptf1a-Cre; Kras^G12D; MMP7^KO (N = 4) mice. Means ± SD. Scale bars, 50μm. See also Figure S4.

Figure 6. MMP7 Expression is Mediated by Stat3 in PDA Cells and Contributes to Tumor Size and Metastasis

(A) Colo357 cells treated with compound S3I-201, a specific Stat3 inhibitor, or DMSO. Both MMP7 (qPCR) and p-Stat3 (Western blot) expression showed dose-dependent reduction. Means ± SD (N ≥ 3). ** p < 0.01, *** p < 0.001 (B) Number of Pdx1-Cre; Kras^G12D; MMP7^Het (N = 8) or Pdx1-Cre; Kras^G12D; MMP7^KO (N = 11) mice that were free of PDA found at time of death or at time of sacrifice at censor date. * p ≤ 0.05 (Chi-square). C-I, Characterization of Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^Het and Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^KO mice. (C) Size of primary tumor at death (cm). Solid bar denotes mean. *** p < 0.001. (D) Representative photograph of disease seen in Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^Het cohort. Large pancreatic head mass is outlined in yellow dashed line. Yellow arrow points to mesenteric lymphadenopathy. Liver nodules and ascites were also seen at necropsy in this mouse. (E) Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^KO mouse. Yellow dashed line outlines relatively smaller pancreatic mass. * denotes massively dilated stomach and ** denotes dilated proximal duodenum. Arrowhead points to transition point (located next to tumor) from dilated proximal duodenum to decompressed distal bowel. There was no liver disease or obvious lymphadenopathy. (F-G) H&E stains. Histology from both genotypes is similar. (H) IHC for MMP7 Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^Het, seen in patchy distribution. (I) No MMP7 staining in Pdx1-Cre; Kras^G12D; p53^f/+; MMP7^KO tumors. Scale bars, 100μm. See also Figure S5 and Table S1.

Figure 7. Serum MMP7 Level Predicts Metastatic Disease in Humans

A-B, Serum MMP7 levels in PDA patients. Means ± SEM. (A) Patients stratified by class. N0: no regional lymph node involvement; N1: regional lymph node involvement; M1: distant metastasis. ** p < 0.01. (B) Patients stratified by stage. (See Table S1 for definition of stage). * p < 0.05 vs. stage IIA, IIB, and III. (C) Kaplan-Meier curves showing significantly lower survival in patients with elevated serum MMP7 (red) than with low MMP7 (blue). Threshold is 20 ng/mL. *** p < 0.0001 by Mantel-Cox log-rank test. See also Table S2-S5.