Chronic progression of cerulein-induced acute pancreatitis in trypsinogen mutant mice

Abstract

T7K24R mice carry mutation p.K24R in mouse cationic trypsinogen (isoform T7), which is analogous to the human hereditary pancreatitis-associated mutation p.K23R. The mutation renders trypsinogen more prone to autoactivation. We recently reported that T7K24R mice exhibit increased severity of acute pancreatitis induced by repeated cerulein injections. The objective of the present study was to test whether trypsinogen mutant mice are prone to develop chronic pancreatitis, as observed in patients. We characterized the natural course of cerulein-induced pancreatitis in T7K24R mice and the C57BL/6N parent strain from the acute episode to 3 months post-attack. As expected, an acute episode of pancreatitis in C57BL/6N mice was followed by rapid recovery and histological restitution. In stark contrast, T7K24R mice developed progressive chronic pancreatitis with acinar cell atrophy, persistent macrophage infiltration, and diffuse fibrosis. The nadir of pancreas damage occurred on days 5-6 after the acute episode and was accompanied by digestive dysfunction. Remarkably, histological recovery was markedly delayed and permanent, chronic changes were still detectable 1-3 months after the acute pancreatitis episode. We conclude that during cerulein-induced acute pancreatitis in T7K24R mice, trypsin triggers an autonomous inflammatory program resulting in chronic disease progression, even after the cessation of cerulein-mediated injury. We propose that this uniquely trypsin-dependent mechanism explains the development of hereditary chronic pancreatitis in humans. Trypsin inhibition during acute attacks should prevent or delay progression to chronic disease.

INTRODUCTION

Chronic pancreatitis (CP) often develops due to underlying genetic susceptibility [1]. In a large majority of cases, inborn mutations result in the accelerated activation of the digestive protease precursor trypsinogen to trypsin [2]. In autosomal dominant hereditary pancreatitis, mutations in the human cationic trypsinogen (encoded by the serine protease 1, PRSS1, gene) exert their effect either by stimulating trypsinogen autoactivation or by blocking protective trypsinogen degradation. In sporadic CP cases with no family history (idiopathic CP), loss-of-function mutations in the serine protease inhibitor Kazal type 1 (SPINK1) gene and/or the gene encoding the trypsinogen-degrading protease chymotrypsin C (CTRC) can increase CP risk by weakening these anti-trypsin defenses. Conceptually, the disease-associated mutations in the PRSS1, SPINK1, and CTRC genes and their phenotypic characteristics define the so-called trypsin-dependent pathological pathway of genetic risk in CP [2]. The trypsin-dependent disease mechanism was also supported by the discovery of protective variants such as an autodegrading variant in the PRSS2 gene (encoding anion trypsinogen) and a common inversion at the chymotrypsin B1-B2 (CTRB1-CTRB2) locus that facilitates degradation of anionic trypsinogen [2, 3].

Recently, novel mouse models were developed that carried pancreatitis-associated mutations in the activation peptide of the mouse cationic trypsinogen (isoform T7). The T7D23A mice contained mutation p.D23A, which is similar to the human PRSS1 mutation p.D22G and the T7K24R mice harbored the p.K24R mutation that is analogous to the human PRSS1 mutation p.K23R [4, 5]. These two mutations stimulate trypsinogen autoactivation 50-fold and 5-fold, respectively, resulting in different pancreatitis phenotypes. Heterozygous T7D23A mice developed spontaneous, early-onset, progressive CP with complete penetrance. In contrast, heterozygous or homozygous T7K24R mice developed no spontaneous pancreatitis. However, when challenged with repeated cerulein injections, T7K24R mice exhibited more severe acute pancreatitis than C57BL/6N controls. Furthermore, cerulein-induced intrapancreatic trypsin activation was higher in T7K24R mice relative to C57BL/6N mice. Finally, sustained, 2-day stimulation with cerulein caused a CP-like disease in T7K24R mice while C57BL/6N mice recovered readily.

Based on the apparent lack of recovery of T7K24R mice from sustained cerulein-mediated injury, we hypothesized that higher intrapancreatic trypsin levels in these mice may trigger a progressive inflammatory program even after cessation of the cerulein challenge, resulting in a CP-like phenotype and delayed recovery. To better characterize this phenomenon, here we compared the natural course of cerulein-induced acute pancreatitis in T7K24R and C57BL/6N mice from the acute episode to 3 months after the attack. The findings demonstrate that besides increasing the severity of acute pancreatitis in T7K24R mice, trypsin also drives subsequent progression to CP and this latter effect is much more dramatic and sustained.

METHODS

Experimental animals.

The generation of the T7K24R mouse strain on the C57BL/6N genetic background was described recently [5]. The T7K24R knock-in allele contains the p.K24R mutation in exon 2 of the mouse cationic trypsinogen (isoform T7) gene [6]. Unless indicated otherwise, T7K24R mice were maintained and used for experiments in the homozygous state. C57BL/6N mice were purchased from Charles River Laboratories (Wilmington, MA) or produced in our breeding facility from the same stock. Both male and female mice were studied. The number of mice used in the experiments is shown in the figures. Experimental animals were 11-12 weeks old and weighed around 25 g (males) and 20 g (females). Details of genotyping for the T7K24R allele were reported previously [5].

Animal studies protocol approval.

Animal experiments were performed at the University of California Los Angeles with the approval and oversight of the Animal Research Committee, including protocol review and post-approval monitoring. Some of the initial studies were carried out at Boston University with the approval and oversight of the Institutional Animal Care and Use Committee. The animal care programs at these institutions are managed in full compliance with the US Animal Welfare Act, the United States Department of Agriculture Animal Welfare Regulations, the US Public Health Service Policy on Humane Care and Use of Laboratory Animals and the National Research Council's Guide for the Care and Use of Laboratory Animals. The University of California Los Angeles and Boston University have approved Animal Welfare Assurance statements (A3196-01 and A3316-01, respectively) on file with the US Public Health Service, National Institutes of Health, Office of Laboratory Animal Welfare. Both institutions are accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC).

Cerulein-induced pancreatitis.

Acute pancreatitis was induced in mice by 12 hourly intraperitoneal injections of the secretagogue peptide cerulein, unless indicated otherwise. Cerulein (catalog number C9026, Sigma-Aldrich, St. Louis, MO) was dissolved in normal saline and administered in a supramaximal stimulatory dose of 50 μg/kg. Mice were sacrificed on days 1, 2, 3, 4, 5, 6, 14, 30 and 90 after the last cerulein injection (day 0), and the pancreas and blood were harvested. Historic samples or data were used for mice euthanized 1 h after the last cerulein injection (day 0 values).

Plasma amylase.

Enzyme activity of amylase in blood plasma (1 μL assayed) was determined with the 2-chloro-p-nitrophenyl-α-D-maltotrioside substrate, as reported recently [5]. Rate of substrate cleavage was expressed in mOD/min units.

Histology and immunohistochemistry.

These experiments were performed with the help of the Boston University Experimental Pathology Laboratory Service Core, the Specialized Histopathology Core, Brigham and Women's Hospital, Boston, MA and the Translational Pathology Core Laboratory of UCLA. Pancreas tissue was fixed in 10% neutral buffered formalin; paraffin-embedded (FFPE); sectioned and stained with hematoxylin-eosin or Masson’s trichrome staining, as indicated. Immunohistochemistry staining for the leukocyte markers MPO and F4/80 was carried out as described previously [4].

Food intake measurement.

Mice were given 8 hourly intraperitoneal injections of saline (n=4) or cerulein (50 μg/kg, n=9) and body mass and food intake were recorded daily for 3 days before and for 4 days after the last injection. Mice had free access to regular chow and water. Most mice were housed in pairs; the average food intake was calculated per cage (i.e. for two mice) and this value was plotted for each mice.

RESULTS

Progression of cerulein-induced acute pancreatitis in T7K24R mice.

To examine the natural course of cerulein-induced pancreatitis in C57BL/6N and T7K24R mice, we induced an acute episode of pancreatitis with 12 hourly injections of cerulein and sacrificed mice on days 1, 2, 3, 4, 5, 6, 14, 30 and 90 after the last injection (day 0). We measured pancreas mass, plasma amylase activity and performed histological analysis of pancreas sections by hematoxylin-eosin staining. For comparison, historic data were used for acute parameters (day 0 values) measured from mice euthanized 1 h after the last injection [5]. When the pancreas mass was plotted as a function of time, the two strains exhibited strikingly different behavior. Thus, during the acute episode, T7K24R mice developed much more significant pancreatic edema resulting in a large increase in pancreas mass, as described previously [5]. Subsequently, however, the pancreas of T7K24R mice began to shrink and the atrophy proceeded until day 6, when slow recovery started (Figure 1). In contrast, C57BL/6N mice rapidly recovered from the acute episode and their pancreas mass normalized within 2 days.

Figure 1.

Changes in pancreas mass during chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in T7K24R and C57BL/6N mice with 12 hourly injections of cerulein, and mice were sacrificed one hour later (day 0) or after the indicated number of days. Control mice (day 0) were given saline injections. A, Individual data points were graphed as a function of time. The day 0 values represent historic data [5]. B, Descriptive statistics (mean, standard deviation) and comparison of the difference of means between the T7K24R and C57BL/6N mice by two-tailed unpaired t-test.

We monitored plasma amylase activity during the acute episode and for 6 days thereafter (Figure 2). As documented previously, amylase levels in C57BL/6N and T7K24R mice increased sharply during the acute attack with significantly higher values measured in T7K24R mice [5]. However, in both mouse strains plasma amylase activity declined rapidly and returned to baseline by day 3 after the last cerulein injection.

Figure 2.

Plasma amylase activity during chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in T7K24R and C57BL/6N mice with 12 hourly injections of cerulein, and mice were sacrificed one hour later (day 0) or after the indicated number of days. A, Individual data points are shown. The day 0 values were generated from historic plasma samples [5]. Although not shown, under the same conditions, plasma amylase activity of saline-treated mice were in the 15-30 mOD/min range (historic data). B, Descriptive statistics (mean, standard deviation) and comparison of the difference of means between the cerulein-treated T7K24R and C57BL/6N mice by two-tailed unpaired t-test.

Histological analysis of T7K24R pancreata revealed that after the acute episode progressive acinar atrophy set in, which reached its lowest point on day 5 after the last cerulein injection, when nearly all normal acini disappeared (Figure 3). Slow recovery started on day 6 but it remained incomplete on days 14 and 30 (Figure 4). Even on day 90, adipose infiltration and “scarring” was apparent among the re-formed acini of T7K24R mice. In contrast, the pancreas of C57BL/6N mice rapidly recovered from the acute episode and showed only minimal residual changes by day 3.

Figure 3.

Histology of chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Representative pictures of hematoxylin-eosin stained pancreas sections from C57BL/6N and T7K24R mice given 12 hourly injections of cerulein are shown. Mice were sacrificed after the indicated number of days relative to the last injection. The scale bar corresponds to 50 μm.

Figure 4.

Delayed recovery from cerulein-induced acute pancreatitis in T7K24R mice. Representative pictures of hematoxylin-eosin stained pancreas sections from C57BL/6N and T7K24R mice given 12 hourly injections of cerulein are shown. Mice were sacrificed after the indicated number of days relative to the last injection. The scale bar corresponds to 50 μm.

Persistent macrophage infiltration and fibrosis in T7K24R mice after cerulein-induced acute pancreatitis.

The analysis of the pancreas mass and histology indicated rapidly progressing, atrophic CP in T7K24R mice after a cerulein-induced acute episode. To evaluate the inflammatory cells during this process, we used immunohistochemistry to stain pancreas sections for the neutrophil granulocyte marker MPO and for the macrophage marker F4/80. As shown in Figure 5, neutrophils were abundant on day 1 after the last cerulein injection but their numbers rapidly decreased by day 2 and remained low on subsequent days. In contrast, the number of F4/80 positive macrophages seemed to increase throughout the 6-day time course studied. Some of this apparent increase was likely due to the diminishing pancreas mass, but even after taking this confounding factor in consideration, the sustained macrophage infiltration was striking. Macrophages, just like neutrophils, first appeared in the interlobular spaces, however, they soon migrated inside the lobules and surrounded the acini. By days 4-6, essentially all acini were enclosed by a macrophage ring.

Figure 5.

Inflammatory cell infiltration during chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in T7K24R mice with 12 hourly injections of cerulein and mice were sacrificed after the indicated number of days. Immunohistochemistry staining was performed for the neutrophil granulocyte marker myeloperoxidase (MPO, left column) and the macrophage marker F4/80 (right column). Representative sections are shown. The scale bar corresponds to 50 μm.

Fibrosis was assessed by Masson’s trichrome staining of pancreas sections, which colors collagen fibers blue (Figure 6). Diffuse fibrosis was evident on days 4 and 6 and it remained prominent during recovery on days 14 and 30. Even 90 days after the acute episode, diffuse and patchy collagen deposition was observed among the acini.

Figure 6.

Pancreas fibrosis during chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in T7K24R mice with 12 hourly injections of cerulein and mice were sacrificed after the indicated number of days. Pancreas sections were stained with Masson’s trichrome staining; blue color indicates fibrosis. Scale bars correspond to 200 μm (left column) and 50 μm (right column, higher magnification).

Digestive dysfunction in T7K24R mice due to acinar atrophy.

When T7K24R mice were sacrificed and autopsied, we noted that the stomach and intestines were distended, filled with undigested chow (Figure 7A). To assess whether T7K24R mice undergo transient digestive failure because of the rapid pancreas atrophy, we treated mice with 8 hourly injections of cerulein or saline and monitored their food intake (Figure 7B) and their body mass (Figure 7C). After saline injections, T7K24R mice slightly and transiently increased their food intake, with no change or a small increase in body mass. This was a compensatory increase after the period of fasting during which the injections were administered. In contrast, T7K24R mice given cerulein consumed slightly less chow on day 1 after the last injection followed by a marked increase in food intake on days 2, 3 and 4. Despite the increased food intake, body mass of T7K24R mice steadily declined on days 1 through 4 after the acute episode, indicating that food was not digested and/or absorbed properly.

Figure 7.

Digestive dysfunction during chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in T7K24R mice with 8 hourly injections of cerulein. Control mice were given saline injections. (A) Distention of the stomach and the intestines in T7K24R mice 4 days after the last cerulein injection. The gastrointestinal tract contained largely undigested chow. (B) Food intake of T7K24R mice before (pre-inj., n=4) and after (post-inj., n=9) cerulein or saline injections. Black symbols indicate male mice; gray symbols indicate female mice. Note that most mice were housed in pairs and individual data points represent the average food consumption per mouse within a cage. (C) Body mass of T7K24R mice before (pre-inj., n=4) and after (post-inj., n=9) cerulein or saline injections. Individual data points are shown.

Heterozygous and homozygous T7K24R mice exhibit similar disease progression.

To test the effect of gene dosage on the trypsin-dependent progression of pancreatitis, we compared disease course in heterozygous and homozygous T7K24R mice by monitoring pancreas mass after 12 hourly cerulein injections (Figure 8A). While the initial pancreas mass increase due to inflammatory edema was smaller in heterozygous mice, subsequent progression to CP was similar in heterozygous and homozygous T7K24R mice reaching the same degree of atrophy by day 6 after the last cerulein injection. Recovery was slightly more efficient in heterozygous T7K24R mice, as judged by the higher pancreas mass on day 30. Hematoxylin-eosin stained pancreas sections on day 6 showed comparable histology with loss of acini and infiltration of inflammatory cells (Figure 8B).

Figure 8.

Effect of gene dosage on the chronic progression of cerulein-induced acute pancreatitis in T7K24R mice. Acute pancreatitis was induced in heterozygous and homozygous T7K24R and C57BL/6N mice with 12 hourly injections of cerulein and mice were sacrificed one hour later (day 0) or after the indicated number of days. (A) Pancreas mass. Individual data points are shown; the values for the C57BL/6N and homozygous T7K24R mice were taken from Figure 1A. The mean values (±standard deviation) of pancreas mass (in mg units) of heterozygous T7K24R mice on days 0, 1, 2, 4, 6 and 30 were 329.0±86.3 (n=9), 286.6±54 (n=8), 272.2±30.7 (n=5), 145.6±36.3 (n=7), 91±50 (n=6), and 151.7±52.6 (n=6), respectively. When the means were compared between homozygous and heterozygous T7K24R mice by two-tailed unpaired t-test, significant differences were found on day 0 (p=0.0006), day 2 (p=0.0091), and day 30 (p=0.0358), but not on day 1 (p=0.1564), day 4 (p=0.1257) and day 6 (p=0.6824). (B) Hematoxylin-eosin stained pancreas sections from mice sacrificed 6 days after the last injection. Scale bars correspond to 100 μm (left column) and 50 μm (right column, higher magnification).

DISCUSSION

In the present study, we compared the natural course of cerulein-induced pancreatitis in T7K24R and C57BL/6N mice. Previously, we demonstrated that increased intrapancreatic trypsin activation in the T7K24R mice given cerulein was associated with more severe acute pancreatitis relative to C57BL/6N mice [5]. Here, we found that after the acute episode, T7K24R mice progressed to CP while C57BL/6N mice recovered rapidly. Progression of disease was characterized by massive acinar atrophy, diffuse fibrosis, pseudotubular complexes and sustained macrophage infiltration. The nadir of pancreas atrophy was observed on days 5-6 after the last cerulein injection. From day 6, new acini started to appear, however, recovery remained incomplete even at 3 months post-attack with signs of permanent parenchymal loss, adipose replacement and fibrosis. Remarkably, the rapid pace of acinar atrophy caused a transient digestive failure in the T7K24R mice indicated by temporary weight loss despite increased food intake and a distended gastrointestinal tract filled with undigested chow.

Since the only difference between C57BL/6N and T7K24R mice is the accelerated autoactivation of mutant cationic trypsinogen, it is reasonable to conclude that the observed phenomena are due to the higher trypsin activity in T7K24R mice generated by cerulein hyperstimulation. Furthermore, the progressive nature of the disease even after the cessation of cerulein injections indicates that trypsin triggered an autonomous inflammatory program, which involved massive macrophage infiltration with subsequent acinar cell atrophy and fibrosis. We note that a similar CP-like phenotype was observed previously in T7K24R mice after a 2-day cerulein hyperstimulation protocol [5]. Similarly, transgenic mice carrying human cationic trypsinogen with the p.R122H mutation or human anionic trypsinogen developed progressive CP after a cerulein-induced episode [7, 8]. In all these cases, the hallmark of the process was the marked acinar atrophy with essentially the entire parenchyma affected. The bulk of the acinar cell mass loss was in all likelihood due to macrophage-induced acinar-to-ductal metaplasia rather than cell death. Previously, we found that in the CP-like disease of T7K24R mice after 2 days of cerulein stimulation the extent of apoptosis was low, suggesting that cell death is not a major factor in parenchymal atrophy [5].

We speculate that release of intra-acinar trypsin and/or the activation of inappropriately secreted trypsinogen in the interstitial space initiates progressive disease, possibly through activation of proteinase-activated receptors on acinar cells and/or stellate cells. The subsequent macrophage infiltration then drives inflammation and atrophy with surprisingly slow resolution and quasi-permanent sequelae. Cerulein-induced basolateral secretion of zymogens has been documented before [9, 10]. In C57BL/6N mice interstitial trypsinogen activation is presumably low, as inappropriate secretion may be countered by defense mechanisms (i.e. inhibition by SPINK1 and degradation by chymotrypsin) and by lymphatic drainage. In contrast, the stronger propensity of cationic trypsinogen to autoactivate in T7K24R mice would overcome the anti-trypsin defenses and trigger the CP-like progression. A highly similar scenario could be observed with the PRSS1/PRSS2 transgenic mice where human trypsinogens autoactivate robustly [7, 8]. In case of the T7D23A mice, in which trypsinogen autoactivates at markedly accelerated rates, even minor basolateral leakage might be enough to cause onset and progression of pancreatitis [4].

Genetic deletion of T7 trypsinogen from C57BL/6 mice abolished cerulein-induced intrapancreatic trypsin activation but had only a modest effect on acute pancreatitis severity [11, 12]. These observations argue that cerulein-induced acute pancreatitis is not a trypsin-dependent pathological process primarily, even though severity may be increased by boosting trypsin activation, as seen during the acute episode of T7K24R mice [5]. Consequently, anti-trypsin therapies may work partially or may not work at all in the context of cerulein-induced acute pancreatitis and preclinical test results may be difficult to interpret. However, the progressive, trypsin-dependent pancreatitis of T7K24R mice can serve as a convenient preclinical model to test the effects of anti-trypsin therapeutic approaches. In this regard, we tested the utility of heterozygous T7K24R mice, which may be easier to obtain or maintain when crossed onto other backgrounds. When compared to homozygous T7K24R mice, heterozygous animals developed very similar CP, which progressed slightly slower but reached essentially the same degree of atrophy as in homozygous mice.

In summary, we demonstrated that after a cerulein-induced acute pancreatitis episode, T7K24R mice develop progressive pancreatitis with CP-like features and delayed recovery. We posit that in T7K24R mice cerulein-induced intrapancreatic trypsin activation triggers an autonomous inflammatory program which progresses even after the cessation of the cerulein-mediated injury. The observations provide a plausible explanation for the development of hereditary CP in humans and suggest that trypsin inhibition during acute attacks should prevent or delay progression to chronic disease.