Trypsin activity governs increased susceptibility to pancreatitis in mice expressing human PRSS1R122H

Abstract

Currently, an effective targeted therapy for pancreatitis is lacking. Hereditary pancreatitis (HP) is a heritable, autosomal-dominant disorder with recurrent acute pancreatitis (AP) progressing to chronic pancreatitis (CP) and a markedly increased risk of pancreatic cancer. In 1996, mutations in PRSS1 were linked to the development of HP. Here, we developed a mouse model by inserting a full-length human PRSS1R122H gene, the most commonly mutated gene in human HP, into mice. Expression of PRSS1R122H protein in the pancreas markedly increased stress signaling pathways and exacerbated AP. After the attack of AP, all PRSS1R122H mice had disease progression to CP, with similar histologic features as those observed in human HP. By comparing PRSS1R122H mice with PRSS1WT mice, as well as enzymatically inactivated Dead-PRSS1R122H mice, we unraveled that increased trypsin activity is the mechanism for R122H mutation to sensitize mice to the development of pancreatitis. We further discovered that trypsin inhibition, in combination with anticoagulation therapy, synergistically prevented progression to CP in PRSS1R122H mice. These animal models help us better understand the complex nature of this disease and provide powerful tools for developing and testing novel therapeutics for human pancreatitis.

Keywords: Gastroenterology; Proteases.

Figure 1. Transgenic expression of human PRSS1^R122H in mice.

(A) Schema of generation of the transgenic human PRSS1^R122H mouse model. An R122H mutation was introduced into a BAC harboring the full-length human PRSS1 gene by GalK-mediated recombineering technology. (B) Sanger DNA sequencing confirmed a CGC>CAC mutation, which confers R122H mutation. (C) Western blot showed a high level of PRSS1^R122H expression in the pancreas of transgenic mice and no expression in WT C57BL/6J mice. Human pancreas lysate was used as a control. Representative blots from 3 independent assays are shown. (D) Higher and prolonged trypsin activity was observed in the pancreatic acinar cells isolated from transgenic PRSS1^R122H mice than was seen in those from C57BL/6J mice. Mean ± SEM (n = 3). **P < 0.01; ***P < 0.001; 2-way ANOVA with Tukey’s multiple comparisons test.

Figure 2. Transgenic expression of human PRSS1^R122H caused severe AP.

(A) Schema of cerulein-induced AP protocol. (B) Representative photos of pancreata from transgenic human PRSS1^R122H mice and C57BL/6J mice 24 hours after cerulein induction (n = 8). (C) Significant increase in pancreatic edema (pancreas-to-body weight ratio) in PRSS1^R122H mice 24 hours after cerulein induction. Mean ± SEM (n = 8). *P < 0.05; ***P < 0.001; 2-way ANOVA with Tukey’s test. (D) Serum amylase levels after 24 hours of cerulein induction. Mean ± SEM (n = 8). ***P < 0.001; 2-way ANOVA with Tukey’s test. (E) Representative images of H&E staining of the pancreata (n = 8). Scale bars: 300 μm. (F) Histology score evaluation of AP. Mean ± SEM (n = 8). ***P < 0.001; 2-way ANOVA with Tukey’s test. (G) Representative immunohistochemical staining for CD11b (pan leukocytes), F4/80 (macrophage), and Gr-1 (neutrophil) positive inflammatory cells (brown signal with hematoxylin purple counterstain) on sections from transgenic PRSS1^R122H mice and C57BL/6J mice 24 hours after cerulein induction (n = 8). Scale bars: 200 μm. (H) Immunohistochemical staining for analysis of p65 nuclear translocation, an indicator of NF-κB activation in the pancreata of PRSS1^R122H mice and C57BL/6J mice (n = 8). Scale bars: 200 μm. (I) Quantification of p65 nuclear translocation in the pancreata of PRSS1^R122H mice and C57BL/6J mice. Mean ± SEM (n = 8). ***P < 0.001; 2-way ANOVA with Tukey’s test. (J) Pancreatic mRNA expression levels of monocyte chemoattractant protein-1 (Mcp1), tumor necrosis factor alpha (Tnfa), interleukin 1β (Il1b), and Il6 in PRSS1^R122H mice and C57BL/6J mice were measured by real-time RT-PCR. Mean ± SEM (n = 4). *P < 0.05; ***P < 0.001; 2-way ANOVA with Tukey’s test.

Figure 3. Progressive pancreatic damage and activation of stress signaling pathways in PRSS1^R122H mice.

(A) H&E staining showed progressive pancreatic damage on sections from transgenic PRSS1^R122H mice after cerulein-induced AP. In contrast, similarly treated C57BL/6J mice recovered fully (representative of 5 independent samples). Scale bars: 400 μm. (B) Acinar cell death, pancreatic edema, and inflammation evaluation in AP after cerulein induction. Mean ± SEM (n = 5). **P < 0.01; ***P < 0.001; 2-tailed unpaired Student’s t test. (C) Using TUNEL staining, extensive pancreatic acinar cell apoptosis was detected in transgenic PRSS1^R122H mice after cerulein induction. In contrast, pancreatic acinar cell apoptosis in C57BL/6J mice was much less prevalent (representative of 5 independent samples). Scale bars: 200 μm. (D) Cleaved caspase 3 was upregulated in the pancreatic acinar cells of PRSS1^R122H mice (representative of 5 independent samples). Scale bars: 200 μm. (E) Western blot showed higher levels of p53 and its target gene BAX, DNA damage signaling p-CHK1, and abnormal ER stress-related proteins Bip and CHOP in the pancreata of PRSS1^R122H mice than in those from C57BL/6J mice. Representative blots from 3 independent assays are shown. (F) Immunohistochemical staining for 8-hydroxy-2′-deoxyguanosine (8-OH) and phospho-Histone H2A.X (Ser139) (p-H2A.X) indicated the presence of DNA damage in the pancreata of PRSS1^R122H mice. Representative of 5 independent samples. Scale bars: 200 μm. (G) mRNA levels of endoplasmic reticulum stress-related signaling molecules Atf4 (activating transcription factor 4) and Xbp1s (X-box binding protein 1 spliced), and ROS-related enzymes Duox1 (dual oxidase 1) and Gpx4 (glutathione peroxidase 4) were measured by real-time RT-PCR. Mean ± SEM (n = 4). *P < 0.05; **P < 0.01; ***P < 0.001; 2-way ANOVA with Tukey’s multiple comparisons test.

Figure 4. CP developed in transgenic PRSS1^R122H mice.

(A) Schema of cerulein administration protocol. (B) After cerulein induction, the body weight changes of the mice were monitored. Mean ± SEM (n = 5–10). (C) Representative pancreas images on day 70 from transgenic PRSS1^R122H mice and WT C57BL/6J mice (n = 5–10). (D) Pancreas weight was compared 70 days after cerulein induction. Mean ± SEM (n = 5–10). ***P < 0.001; 2-tailed unpaired Student’s t test. (E) CP with fat replacement, fibrosis, and pancreatic intraepithelial neoplasia (PanIN) lesions developed in PRSS1^R122H mice. These features were similarly observed in the pancreata of human patients with hereditary pancreatitis. (F) Overall histology score of CP. Mean ± SEM (n = 5–10). ***P < 0.001; 2-tailed unpaired Student’s t test.

Figure 5. PRSS1^R122H mice were more sensitive to induction of pancreatitis than were PRSS1^WT mice.

(A) Transgenic mice expressing WT human PRSS1 (PRSS1^WT) were generated for comparison with PRSS1^R122H mice. (B) Western blot showed similar expression levels of PRSS1 in these transgenic mice. Representative blots from 3 independent assays are shown. (C) Pancreatic acinar cells were isolated from the indicated mice and trypsinogen activation in response to low and high concentrations of CCK was measured. Mean ± SEM (n = 3). **P < 0.01; ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (D) Greater pancreatic edema (pancreas-to-body weight ratio) was seen in PRSS1^R122H mice compared with PRSS1^WT mice at 24 hours after a single-dose of cerulein. The cerulein dose ranged from 2.5 μg/kg to 50 μg/kg. Mean ± SEM (n = 5). *P < 0.05; **P < 0.01; ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (E) Greater serum amylase levels were seen in PRSS1^R122H mice than in PRSS1^WT mice. Mean ± SEM (n = 5). *P < 0.05; **P < 0.01; ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (F) Overall histology score of AP developed after various doses of cerulein insults. Mean ± SEM (n = 5). **P < 0.01; ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (G) Representative H&E staining showed that PRSS1^R122H mice were more sensitive to cerulein-induced AP than were PRSS1^WT mice (n = 5). Scale bars: 200 μm.

Figure 6. Increased trypsin activity was the cause of severe pancreatitis in PRSS1^R122H mice.

(A) Enzymatically inactive Dead-PRSS1^R122H (PRSS1^Dead) transgene construct was developed by introducing a S200T point mutation into PRSS1^R122H using recombineering technology. (B) In response to cerulein, Dead-PRSS1^R122H generated the same amount of active trypsin as C57BL/6 mice. Mean ± SEM (n = 3). ***P < 0.001; 1-way ANOVA with Tukey’s test. (C) Schema of cerulein-induced AP in PRSS1^Dead and PRSS1^R122H mice. (D) Enlarged pancreata were observed in PRSS1^R122H mice but not in PRSS1^Dead mice 24 hours after cerulein induction (n = 10). (E) Comparison of pancreatic edema (pancreas-to-body weight ratio) 24 hours after cerulein induction. Mean ± SEM (n = 10). ***P < 0.001; 2-tailed unpaired Student’s t test. (F) Comparison of serum amylase level 24 hours after cerulein induction. Mean ± SEM (n = 10). ***P < 0.001; 2-tailed unpaired Student’s t test. (G) Cerulein caused more severe AP in PRSS1^R122H mice than in PRSS1^Dead mice. Representative images of H&E staining 24 hours after cerulein induction are shown (n = 10). Scale bars: 200 μm. (H) Overall histology score of AP from PRSS1^R122H mice and PRSS1^Dead mice. Mean ± SEM (n = 10). ***P < 0.001; 2-tailed unpaired Student’s t test. (I) To investigate cerulein-induced CP in PRSS1^R122H and PRSS1^Dead mice, the mice were sacrificed 7 days after cerulein induction. (J) Seven days after cerulein induction, all the pancreata from PRSS1^R122H mice shrank (indicating CP), but all the pancreata from PRSS1^Dead mice appeared to be normal (n = 8). (K) The pancreas-to-body weight ratio 7 days after cerulein indicated that the pancreata in PRSS1^R122H mice were much smaller than those in PRSS1^Dead mice. Mean ± SEM (n = 8). ***P < 0.001; 2-tailed unpaired Student’s t test. (L) At day 7, histologic examination (H&E staining) demonstrated chronic damage in PRSS1^R122H mice, whereas the pancreata of PRSS1^Dead mice were normal (n = 8). Scale bars: 200 μm.

Figure 7. Effective experimental therapeutics with an FDA-approved anticoagulation agent in PRSS1^R122H mice.

(A) Schema of pancreatitis induction and treatment in PRSS1^R122H mice. Pancreatitis was induced by cerulein, and therapeutic drugs were administered 5 hours after the first cerulein injection. (B) After 7 days of treatment, pancreata in the untreated control group became smaller. In contrast, pancreata in dabigatran-treated groups were mostly normal. Camostat only exhibited intermediate effects at higher does (300 mg/kg). Both drugs were given twice daily for 7 days. Mean ± SEM (n = 10 per group). ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (C) Histology score evaluation showed that dabigatran significantly improved CP. Mean ± SEM (n = 10 per group). ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (D) Representative images of H&E staining after drug treatments over 7 days (n = 10 per group). Scale bars: 200 μm.

Figure 8. Anticoagulation and trypsin inhibition synergistically improved pancreatitis in PRSS1^R122H mice.

(A) Immunohistochemical analysis showed intrapancreatic fibrin deposition, an indicator of increased coagulation, in the pancreata of PRSS1^R122H mice 24 hours after cerulein induction (n = 5). Scale bars: 200 μm. (B) Western blot showed that active thrombin significantly increased in the pancreata of PRSS1^R122H mice compared with those from C57BL/6J mice. Representative blots from 3 independent experiments are shown. (C) Schema of pancreatitis induction and drug treatment with the anticoagulation specific agent (apixaban), trypsin specific inhibitor (camostat), or in combination. Starting 5 hours after pancreatitis induction, drugs were given twice daily by oral gavage over 7 days. (D) Combination of anticoagulation therapy with factor Xa inhibitor apixaban (100 mg/kg) and trypsin inhibitor camostat (200 mg/kg) greatly protected the pancreas, as manifested by preservation of pancreas mass. Mean ± SEM (n = 15 per group). ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test. (E) Representative histologic images of H&E staining from untreated, apixaban alone, camostat alone, and combination therapy–treated mice (n = 15 per group). Scale bars: 200 μm. (F) Overall histology score evaluation of the mice. Mean ± SEM. Representative results from 6 mice per group are shown. ***P < 0.001; 1-way ANOVA with Tukey’s multiple comparisons test.