• chronic pancreatitis

In Gut, Hegyi and Sahin-Tòth report the first reliable mouse model of chronic pancreatitis (CP).1 More than six decades ago, the first pedigree of a family with inherited CP was published. Nevertheless, it lasted until now that the first reliable animal model of the disease was established. Whereas the first genetic associations were found to reside in the cationic trypsinogen gene (PRSS1), Hegyi and Sahin-Tóth developed an animal model with the frequent p.N256K mutation in Carboxypeptidase A1 (CPA1).2 3 Besides PRSS1 alterations, mutations in CPA1 harbour the second largest effect size for the development of CP (see also http://www.pancreasgenetics.org/) and therefore represent reasonable candidates for an animal model. In addition, the detection of CPA1 mutations in CP provided new mechanistic insights into the pathogenesis, but also reinforced the current notion that digestive proteases play a critical role in CP. The presented model is a knock-in of the human mutation into the Cpa1 mouse locus and exhibits spontaneous hallmarks of CP like fibrosis and acinar-ductal metaplasia. The mechanism of action most probably is endoplasmatic reticulum (ER) stress induced by misfolding of Cpa1, as the ER stress markers Hspa5 (BiP) and Ddit3 (CHOP) were elevated. These results are in line with in vitro data of the initial description of CPA1 in CP.3

One might argue that this animal model is not too thrilling as it confirms expected and partly known disease mechanisms. Moreover, therapeutic options in CP thus far are scarce and the disease is in most cases caused by harmful alcohol consumption and smoking and not by mutations in genes like chymotrypsinogen C (CTRC), CPA1, PRSS1 or the serine protease inhibitor Kazal type 1 (SPINK1).4 5 In alcoholic CP, however, recent genome-wide association studies demonstrated, that in addition to CTRC, PRSS1 and SPINK1 genetic susceptibility of novel loci like Claudin 2 (CLDN2) or Chymotrypsinogen B1 and B2 (CTRB1, CTRB2) alters the disease risk.6–8 This implies that distinct pancreatitis forms share common pathomechanisms and that we are or will be capable to decipher the risk for individuals to develop CP with genetic means, but still lack therapeutic options.

As such, this first animal model of the disease is of greatest importance, as it will enable researchers to develop approachable targets and change the course of the disease in different stages. To highlight this need, we briefly can summarise the few options we have.

The current therapeutic efforts discussed in CP tackle the influence of variants in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. Here, it is speculated that some of the many agents like potentiators developed for patients with cystic fibrosis might be used in patients with CP. There is no doubt that CFTR variants associate with CP, although the strength of the association and thereby most likely the biological effect is small compared with other genetic alterations.9 10 Apart from the impact of the genetic association that might influence the effect of a therapeutic approach, the timing of the intervention seems to be crucial for its outcome. Here, it is most likely necessary to intervene before the fibrotic remodelling of the pancreas has taken place to prevent chronic manifestation of the disease. This clearly demonstrates that apart from reliable mouse models as developed by Hegyi and Sahin-Tóth, a reliable early diagnosis of CP might be the clue to target the disease in the future. In conclusion, whether targeting CFTR is a feasible and promising approach for patients with CP will most probably be evident in the near future.

The presented work of Hegyi and Sahin-Tóth is the first in vivo evidence that mechanisms other than the premature activation of trypsin or the lack of its inhibition contribute to pancreatitis development. The authors convincingly demonstrated that mutation p.N256K induced CPA1 misfolding and elicited ER stress. As stated by the authors, this model has the potential to close the so far present knowledge gap and enables translational approaches, due to several reasons. First, in p.N256K homozygous mice, CP develops spontaneously (with first signs of CP already detectable after 1 month) and this process follows hallmarks we can recapitulate in the human situation. Second, although detailed data have not been shown, a similar but slower course of the disease is seen in heterozygous mice offering the option to enlarge the therapeutic window of new targets in a different setting. Third, the authors observed high trypsin activity throughout the disease course pointing out that this model additionally encompasses other important mechanisms apart from misfolding of CPA1 and consecutive ER stress.

In conclusion, this first reliable in vivo model has the potential to open avenues to decipher and target the underlying disease causing mechanisms for inflammatory pancreatic diseases. Since acute and CP seem to share parts of this mechanisms and since patients with CP have a higher risk to develop pancreatic cancer, the spectrum of the implications of the presented model will hopefully be even larger than expected.