Editor,—We read with great interest the paper by Nishimori et al(Gut1999;44:259–263) that reported the presence of two cationic trypsinogen gene mutations, the N21I in exon 2 and the R117H in exon 3, in two Japanese families with hereditary pancreatitis. Furthermore, this report showed that these mutations are worldwide and that the defect in cationic trypsinogen is important in the pathogenesis of hereditary pancreatitis. However, these authors questioned whether the N21I mutation could alter the molecular nature of the enzyme dramatically to produce pancreatitis, based on the observation that the first 38 N-terminal amino acids of the secreted cationic trypsinogen containing this mutation are identical to that of the native anionic trypsinogen isoform. This comparison may be important, but any possible functional effect of the mutation should be evaluated in the context of the tertiary structure of the native cationic trypsinogen itself.

The N21I mutation was first identified in two American families with hereditary pancreatitis by Gorry and colleagues1 in 1997; it has now has been shown to be present in three German2and two French3 families. The proposal that N21I also causes disease is strongly supported by the huge body of available research data, the linkage analysis, its perfect segregation with the dominant disease phenotype in several unrelated families with hereditary pancreatitis worldwide, its absence in the control chromosomes, and the previously identified R117H mutation in the same gene.4 ,5

At issue is how to interpret the mutation and, in this regard, it is important to note that the physicochemical characteristics of isoleucine are quite different from those of asparagine. These differences could significantly affect the local conformational structure. Through computer analysis, Nishimori et al predicted that the N21I substitution could change the native turn structure of the flanking region of wild type cationic trypsinogen to sheet structure, on mutant type enzyme. Moreover, the N21 is located on the surface of the molecule and near the R117 site, revealed by the crystallographic structure of trypsin.6 Thus, taking into account of the self destructive model of R117H,4 it is reasonable to predict that the N21I mutation may impair trypsin inactivation through altering the accessibility of trypsin-like enzymes to R117, and/or protecting the adjacent C22-C157 disulphide bond to prolong survival of trypsin after limited hydrolysis.1