Statistics from Altmetric.com
Editor,—In a recent issue (Gut1999;45: 317–322), David Whitcomb reviewed the current knowledge on molecular genetics and pathogenetic concepts in hereditary pancreatitis. We were delighted by this review from an expert who has recently published landmark papers in which the most frequent mutations of the cationic trypsinogen in hereditary pancreatitis were described.
However, there is considerable confusion concerning the nomenclature of the identified mutations, as in the literature two different systems were used in parallel: the chymotrypsin system and the trypsinogen system. The two frequent mutations R117H and N21I were reported in the chymotrypsin nomen- clature1 2 whereas for the other mutations (A16V, K23R) the trypsinogen system was used.3 4 By mistake, the N21I mutation could be assumed to be located two amino acids proximal to K23R but is in fact six amino acids distal. Remarkably, in the paper of Gorry and colleagues2 the N21I mutation was reported in the chymotrypsin nomenclature whereas for the frequent polymorphism D162D in exon 4 the trypsin nomenclature was used.
The chymotrypsinogen nomenclature favoured in the above review is based on the static alignment of the amino acid sequences at serine 195 of the chymotrypsin molecule, as Ser195 is the catalytic residue of all serine proteases. However, use of this nomenclature for mutations of the cationic trypsinogen has several drawbacks:
The start codon (Met) is not codon 1.
The amino acids of the signal peptide as well as five amino acids in the active enzyme are not numbered. Mutations of these amino acids cannot be referred to adequately.
There are 11 numbers without corresponding amino acids.
A direct alignment of chymotrypsinogen and trypsinogen to numerate mutations is not useful as homologies are rare.
At summary meetings on “Mutation Databases” in San Francisco, USA and “Locus-specific Mutation Databases” in Heidelberg, Germany, the International Nomenclature Working Group pointed out that the nomenclature of mutations needs to be accurate and unambiguous. A list of clear recommendations for mutation names has been published.5
To avoid further confusion in the numbering of trypsinogen mutations, we suggest that the trypsinogen numbering system should be used as this system undoubtedly identifies all amino acids in the sequence of trypsinogen.6 This nomenclature also fulfils the criteria proposed by the Nomenclature Working Groups5:
Numbering of the protein starts at methionine (codon 1).
Each amino acid in the protein can be identified by a unique number.
As a consequence, however, the frequent mutations R117H and N21I will have to be called R122H and N29I whereas the nomenclature of the other mutations (A16V, K23R) can be maintained.
To sequence trypsinogen genes in DNA from patients suspected of having hereditary pancreatitis is a simple task and will probably become a standard diagnostic procedure for early onset chronic pancreatitis. Comparable with cystic fibrosis, which is the most common congenital cause of chronic pancreatitis, more mutations of trypsinogens are expected to be found in the future. The standardised gene based nomenclature of trypsinogen mutations that is proposed here should prevent future confusion and clearly identify all mutations of the molecule.
Editor,—I appreciate the comments of Keimet al in response to the leading article on hereditary pancreatitis. It is clear that there is confusion about the proper numbering system for cationic trypsinogen (PRSS1) and that consensus on a uniform numbering system should be adopted.1-1 We certainly agree that the numbering systems proposed by the nomenclature working groups are useful and should be adapted, but the chymotrypsinogen numbering system (chy#), at times, is also useful. However, we also suggest that if the original R122H (chy #R117H) and N29I (chy #N21I) mutations are mentioned, that both nomenclatures are given for reasons of clarity and consistency. Allignment of the serine proteases with both numbering systems clarifies this problem.1-2
We also noted a problem in figure 1 of our review on hereditary pancreatitis1-3 that reflects our own difficulty with the two systems. The problem occurred when we attempted to align the chymotrypsinogen in exon III with trypsinogen by back numbered from the Calpha backbone of trypsinogen structure in the MAGE image,1-4 which shows a two amino acid numbering deletion (#67 and #68). Inserting the two spaces in trypsinogen sequence resulted in a frame shift of chymotrypsinogen through chy#128 (for example, the aspartic acid (D) at chy#102 should be aligned with trypsinogen), although the numbering of cationic trypsinogen, anionic trypsinogen, and mesotrypsinogen remains correct in the figure (allowing for the various chy# given for the trypsinogen amino acids coded by codons 71 to 73). However, alignment of trypsinogen and chymotrypsinogen using the figure 3 in the pancreatic elastase chapter of Hartley and Shotton1-5 provides a proper alignment, as recently published by Whitcomb.1-2 We regret any further confusion the frame shift in a portion chymotrypsinogen exon III may have caused.