Excised polyps were histopathologically classified as hamartomatous in the cases A-1, B-1, B-2, and B-3, as adenomatous in the case C-1, and as hyperplastic in the case D-1. Case E-1 was free of any polyps (Table ). All studied individuals had pigmentation of the lips and buccal mucosa, while it was most visible in the children patients. None of the probands had pigmentation of extremities. A positive family history of cancer was only noted in the case D-1. Mutation analysis revealed three different germline mutations. In the family A, a germline mutation (c.350dupT) in exon 2 (Fig. ) was detected. This mutation is predicted to introduce a frameshift at codon Leu117, 46 novel amino acid residues, and a premature termination codon (p.Leu117PhefsX46). It was found in heterozygosity in both examined patients (A-1 and A-2).
Sequencing chromatogram from the analysis of STK11 gene in the family A showing a duplication of thymine in the position c.350, leading to a frameshift and a premature stop codon.
Individuals from the family B (cases B-1, B-2, and B-3) harboured deletion of a part of the promoter region and exon 1 (Fig. ). Case C-1 was a carrier of deletion of the whole STK11 gene (Fig. ). In the cases D-1 and E-1, we have not revealed any variation of STK11 gene by using the aforementioned methods.
Representative chromatogram from MLPA analysis of STK11 in the family B showing the relative reduction in the peak area of probes hybridising to exon 1 and a part of the promotor region (arrows mark the deleted regions).
Representative chromatogram from MLPA analysis of STK11 in the family C showing the relative reduction in the peak area of probes hybridising to exons 1–10 and the promotor region.
PJS is a relatively very well characterized disorder with a clear cut phenotype [22
]. However, in sporadic cases, the diagnosis of PJS may be uncertain. Although multiple hamartomatous polyps of the GIT are pathognomonic of the PJS, hyperplastic and adenomatous polyps are commonly present [17
]. Recently, it has been reported that STK11
deletions are not a rare cause of Peutz-Jeghers syndrome and account for up to 30% of patients with PJS [26
]. There was no difference in the clinical phenotype between the patients with point mutations or with large genomic deletions [26
]. However, the detailed phenotype of patients with different types of mutations was not reported.
Members of the family B (cases B-1, B-2, and B-3) had almost uniform clinical symptoms with variable age of onset of the first symptoms and detection of polyps. They are carriers of a germline mutation (deletion of a part of the promoter region and exon 1) of STK11. All three affected individuals had mucocutaneous hyperpigmentation predominantly on the lips and on the buccal mucosa, being the most prominent in the youngest patient (B-3) and very pale in the mother (B-1). Sons of the latter patient, i.e. the patients B-2 and B-3, were initially classified as PJS-suspected because of striking hyperpigmentation of the lips. Both of them are under careful surveillance. A few endoscopic examinations of the GIT have been performed so far, with polypectomy of hamartomatous polyps. This confirmed the diagnosis of definite PJS. Polyps were localized in the small intestine and colon. No polyps were detected in the stomach.
Deletion of the promoter region and exon 1 was reported in three independent studies in 12 PJS families overall [26
]. It could be a recurrent mutation, probably a consequence of an unequal recombination mediated by repetitive Alu (SINE elements) sequences. In accord with UCSC Genome Browser, the region of chr19:1,129,999–1,196,665, where STK11
gene is located, is rich for these repetitive sequences, which can be involved in large chromosome rearrangements. The implication of Alu repetitive elements in unequal genomic recombinations were described for another tumor suppressor gene, MSH2
, implicated in Lynch syndrome (HNPCC) [29
We failed to find any variation of STK11
gene in sporadic cases D-1 and E-1, which would explain their phenotype. On the other hand, they did not fulfill criteria for the diagnosis of definite PJS [3
]. We included these cases to the study on the basis of the result from the case C-1. Especially in case E-1 PJS polyps could develop later on. Studies with more individuals not fulfilling PJS diagnostic criteria were reported. None of the patients harboured a germline mutation of STK11
]. Some studies suggested there could be another locus responsible for PJS phenotype [31
]. Other authors stated according to their results that another locus is unlikely and the causative variation could be in regulatory regions such as promoter, enhancers, or splicing sites deep in introns, which are not detectable by conventional methods [26
The risk of developing various types of GIT cancers (in the esophagus, stomach, small bowel and colon) was determined in several studies [3
]. The cumulative risk for stomach cancer was 29% [8
]. Amos et al. noted that gastric polyps are very common among individuals with PJS [30
]. However, they did not specify the proportion of patients with a detectable PJS germline mutation and the gastric polyps/cancer. There are several case-reports and reviews reporting gastric cancer in PJS patients [3
]. In our group of probands, the case A-1 had developed gastric cancer at 28 years of age and died one year later. No genotype-phenotype correlations were published in PJS patients with gastric cancer [7
]. Konishi et al. reviewed 103 PJS patients with malignancy from literature and found out that the mean age of 8 cases with gastric cancer was 31.2 years as compared to 39.7 years in duodenal carcinoma (9 cases), and 48 years in colorectal carcinoma (13 cases). According to the literature and our results we suppose that gastric cancer has very aggressive course in some individuals with PJS and despite the very frequent endoscopic examinations with relevant treatment the next course is usually poor. Therefore, more attention should be paid to patients with molecularly confirmed PJS, especially those who have polyposis of the stomach. It would be particularly interesting to find out if there is a correlation between the genotype and phenotype in relation to the development of gastric cancer. There have been only two reports dealing with gastric cancer in PJS patients and mutational analysis of STK11
gene so far [24
]. Shinmura et al. described two PJS females (sisters) with gastric cancer in whom a STK11
germline mutation (c.890delG) was identified [24
]. Takahashi et al. reported a 14-year-old girl with sporadic PJS and early-onset gastric cancer harboring a frameshift (c.757_758insT) STK11
]. Similarly, as in our family A, the mutations led to a truncated protein lacking the kinase domain. These results suggest that the truncation mutations leading to loss of STK11
kinase domain could act in a dominant negative fashion and be responsible of tumor development. Schumacher et al. summarized clinical and mutational data from 132 PJS cases (83 without and 49 with cancer) to find correlation between the type/site of mutation and cancer. They proposed two different mechanisms of tumor development. One is based on the loss of STK11
functions due to truncation mutations and subsequent LOH as a second hit. This hypothesis is not in accord with findings of other authors, where a second hit was not a requisite condition for tumor development [24