Editor—Chronic idiopathic pancreatitis is a genetically heterogeneous disease.1-3 Mutations of the cationic trypsinogen (CT) gene underlie some cases of juvenile pancreatitis,3-5 and mutations of the cystic fibrosis transmembrane conductance regulator (CFTR) gene have been associated with chronic pancreatitis in adults.6-8 However, these genes account for only a relatively small proportion of cases. More recently, the serine proteinase inhibitor Kazal type 1 (SPINK1) gene, also calledPSTI, has attracted attention as a possible cause for chronic pancreatitis.9 10 One study by Chenet al 9 did not find disease causing mutations of SPINK1 among 14 families with hereditary and 30 patients with sporadic pancreatitis, apart from two rare amino acid substitutions which were observed at a comparable frequency to the general population. By contrast, another study by Witt et al 10 reported 23 out of 68 children and adolescents with chronic pancreatitis whose disease was associated with the occurrence ofSPINK1 mutations in the heterozygous or homozygous state. In particular, one founder mutation, N34S, was identified in 18/68 German patients but only in 1/279 controls.10 In the work presented here, we have addressed the role of SPINK1 mutations in a series of 20 adult German pancreatitis patients, a cohort that we had previously analysed for mutations in the CT andCFTR genes.8

The mean age of patients in our series was 32 years (range 19-46 years). All of them presented with either recurrent pancreatitis characterised by at least three episodes of pancreatitis at least 12 months apart or with chronic idiopathic pancreatitis.8Genomic DNA was extracted from white blood cells and the four exons of the SPINK1 gene were amplified by PCR using published primers9 10 and scanned for mutations by single strand conformation polymorphism (SSCP) analysis and direct sequencing.

Only two patients were found to carry sequence alterations of theSPINK1 gene (table 1). One 26 year old patient was homozygous for the previously reported missense mutation N34S.9 10 His parents were both heterozygotes and did not show any signs of pancreatitis, consistent with an autosomal recessive mode of inheritance. This patient had not been found to carry aCFTR gene alteration in our previous study8 (table 1). The second patient, a 35 year old male, was heterozygous for a new SPINK1 mutation, a G→A transition at nucleotide 194, that is, the last nucleotide of exon 3 (fig 1A). The 194G→A transition could be confirmed by restriction enzyme analysis as it abolishes a recognition site forHphI and creates a new site forTspRI (fig 1B). This substitution does not seem to affect splicing of SPINK1 mRNA, as assessed by nested RT-PCR from a rectal biopsy of the patient (not shown). However, it leads to a missense mutation R65Q at an amino acid position that is conserved in rat and mouse, although some variability exists in cattle.11 Interestingly, the same patient had also been found to be heterozygous for a nonsense mutation of theCFTR gene, Y1092X in exon 17b8(table 1). To elucidate the role of this double heterozygosity further, we performed a segregation analysis among the healthy family members of the patient. Both the SPINK1 andCFTR mutations were also found in the patient's mother and sister who did not show any signs of pancreatitis. These observations indicate that double heterozygosity for the SPINK1 R65Q and theCFTR Y1092X mutations is not sufficient to cause chronic pancreatitis or at least not fully penetrant.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

(A) Direct sequencing of exon 3 of the SPINK1 gene showing heterozygosity for the R65Q substitution (asterisk). (B) Screening for missense mutations N34S and R65Q by restriction enzyme analysis. Exon 3 PCR products were amplified using primers from Witt et al10 and were digested with TspRI and separated on a 2% agarose gel. Lane 1: size marker (kb ladder, BRL), lane 2: wild type control, lane 3: homozygous N34S, lane 4: heterozygous R65Q. Note the distinct patterns for mutations N34S and R65Q in this assay.

In summary, the frequency of SPINK1mutations is low in our cohort of adult patients with chronic idiopathic pancreatitis. This is in agreement with the previous report from Brittany,9 but seems to contrast with the study of German patients by Witt et al.10 One possible reason for the discrepancy may lie in a sampling bias. Similar to CT mutations,SPINK1 gene mutations might be predominantly responsible for juvenile early onset pancreatitis, a condition not preferentially selected for in our study. In total, mutations of theCT, CFTR, andSPINK1 genes constituted less than one third of our cases with adult pancreatitis suggesting that additional genes may be involved in the aetiology of this disorder. A pathogenic role of double heterozygosity for SPINK1 andCFTR gene variations could not be confirmed in the single family identified here, but such a possibility may deserve further investigation in larger cohorts.