Family 1 (Pedigree 1, Figure 2). Two JP patients were reported in this Finnish family. Multiple colonic juvenile polyps were observed in both of these patients, one of them was also diagnosed with colorectal and pancreatic carcinomas at ages 42 and 50, respectively (JP 4/1 in study II, family 1 in study III).
Family 2. This Finnish family (Pedigree 2, Figure 2) includes four members with JP, two of whom were diagnosed with colorectal carcinoma (at ages 40 and 50). Among all JP cases, the juvenile polyps were observed in the large intestine and in one patient also in the stomach. One additional member of this family had been diagnosed with a colonic adenoma without other evidence of JP. This individual had been operated on for aortic stenosis of unknown etiology at the age of 25. One JP patient and two at risk individuals (ages 11 and 17) displayed a ventricular septal defect (family 2 in study III).
Family 3. Family 3 (Pedigree 3, Figure 2) originates from Texas, the United States, and includes five members with JP. One of these JP patients was diagnosed with a colorectal carcinoma at the age of 30. Multiple colonic polyps including adenomatous changes were reported in all five JP cases (family 3 in study III).
Family 4 (Pedigree 4, Figure 2). This family originates from Finland and includes six members with JP, two of whom are also diagnosed with colorectal carcinoma (at ages 34 and 53). Among all JP cases, juvenile polyps were observed in colon and in two patients also in the stomach (JP2/13 in study II, family 4 in study III).
Family 5 (Pedigree 5, Figure 2). Family 5 originates from Finland and includes eight members with JP. One of them is also diagnosed with colorectal carcinoma. There are two other cancer cases in this family, one with acute myeloid leukemia and other with gastric cancer (JP5/1 in study II).
Sporadic case 1 (female). At birth, a hamartoma of the renal pelvis was diagnosed and removed. At the age of seven the patient was diagnosed with Wilms' tumor, and the right kidney was removed. At the age of 12, Ebstein's anomaly (displacement of the tricuspid valve) was diagnosed and surgically corrected. At the age 22, the patient was operated on for bowel obstruction caused by a poorly differentiated adenocarcinoma in the colon ascendens. In addition to the carcinoma, 20 to 30 polyps were observed in the proximal colon up to hepatic flexure. Several polyps were examined by an experienced pathologist, and all were hamartomatous polyps, which could be designated as juvenile polyps. In one polyp, some features of Peutz-Jeghers polyp were also seen. This patient died at the age of 25 (sporadic case 1 in study III).
Sporadic case 2 (female). This patient has congenital panhypopituitarism. JP was diagnosed at the age of 38. Juvenile polyps were observed in the large and small intestine and in stomach. Biological parents and relatives are unknown and the patient has no children.
Sporadic case 3 (male). JP was diagnosed at the age of 13. Juvenile polyps were observed in the large and small intestine. This patient has also been diagnosed with empty cella-syndrome, Osler's disease, and epilepsy. There is no family history of JP (JP 1/1 in study II, sporadic case 3 in study III).
Sporadic case 4 (male). This patient was diagnosed with 30 to 40 colonic juvenile polyps at age of 6, but there is no family history of JP (JP 10/1 in study II).
Sporadic case 5 (female). At the age of 18 ulcerative colitis (UC) was diagnosed, later the diagnosis was confirmed as JP (age of 29). In addition to JP, this patient is also mentally retardated. There is no clear evidence of family history of JP, however the patient's father has a history of gastrointestinal symptoms, but has not been clinically evaluated.
In addition to patients listed above, studies I and II included JP families and sporadic cases from United Kingdom (study I) and United States (studies I and II). Clinical features and pedigree of the Iowa kindred (family I-13 in study II) have been previously reported; first by Stemper and colleagues (1975) and later by Howe and colleagues (1998a).
This study was approved by the Ethical Committee of the Department of Medical Genetics, University of Helsinki.
Figure 2. Pedigrees of JP families. Those samples where DNA was available are marked with an asterisk.
Fourteen Finnish hereditary non-polyposis colorectal cancer (HNPCC) kindreds from whom lymphoblastoid cell lines were available were selected for this study. One affected individual per family was included in the study. Six kindreds fulfilled the Amsterdam criteria for HNPCC (Vasen et al., 1991). Other patients represent familial HNPCC-like colorectal cancer (CRC). The number of patients with CRC or endometrial cancer ranged from 2 to 6 per family. All kindreds selected for this study have been previously shown to be MLH1 and MSH2 mutation negative (Nyström-Lahti et al., 1996; Holmberg et al., 1998). All except three kindreds displayed microsatellite stable tumors (MSS). In these three kindreds DNA from tumor tissue has not been available.
Between May 1994 and June 1998 over one thousand fresh-frozen specimens of colorectal adenocarcinoma have been collected at the Department of Medical Genetics, Haartman Institute, University of Helsinki (Aaltonen et al., 1998; Salovaara et al., in press). Among those, 26 microsatellite instable (MSI) and 16 MSS tumors were selected for SMAD4 methylation analysis and 15 MSI and 7 MSS tumors for SMAD4 mutation screening, respectively.
In studies I and II, DNA was extracted from JP patient cell lines (cell pellets) or blood samples with standard procedure. DNA extraction from paraffin-embedded tumor or normal tissue was performed using the phenol and chloroform procedure described in Kannio et al. (1996).
In studies III and IV, total cellular RNA was extracted from lymphoblasts by RNA extraction kit (QIAGEN).
In study V, the tumor DNA was extracted from fresh frozen tumor specimens by the standard procedure described by Lahiri and Nürnberger (1991). The corresponding normal DNA was extracted from normal mucosa or blood.
20 µl of cDNA was created from 0.8 µg of RNA using standard random priming methods with 200 units of M-MLV reverse transcriptase (Promega), 1 ´ reaction buffer (Promega), 10 µM random hexamer and 60 units of RNAse inhibitor (Promega). The reaction was carried out at 42°C for 1 h and then 95°C for 10 min.
Primers for genomic PTEN amplification have been previously described (Liaw et al., 1997; Steck et al., 1997) except for primers for amplification and sequencing of exons 2 and 4 which are shown in study I. The detailed PCR conditions are described in study I.
Primers used for genomic amplification of the SMAD4 gene have been previously described (Moskaluk et al., 1997) except for new primers for exons 4, 7 and 8, which are described in study III. Those primers were designed by using the Primer3 program. PCR conditions for genomic SMAD4 amplification are shown in study III.
The cDNA sequence for SMAD4 was derived from the GenBank (accession number U44378). The gene was amplified in five fragments and PCR primers for cDNA amplification were designed using the Primer3 server (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3.cgi) and they are listed in study IV. The PCR reactions were carried out as described in studies II, III, IV and V.
The cDNA sequences for SMAD2, SMAD3 and SMAD7 were derived from GenBank database (accession numbers U65019, U76622 and AF010193, respectively). The genes were amplified in five fragments and PCR primers for cDNA amplification were designed using the Primer3 server. Primer sequences and PCR conditions are shown in studies III and IV.
The genomic sequences for ALK1 and endoglin were derived from the GenBank database (accession numbers U77707-U77713 for ALK1 and U37439-47, U17156-7, and AF036969-71 for endoglin, respectively). The primers and PCR conditions for amplification of all exons of ALK1 and endoglin genes have been previously published. Those primers and conditions were also used here (Berg et al., 1997; Gallione et al., 1998).
DGGE was performed for all exons of PTEN, with the exception of exons 2 and 4, in probands from 10 JPS families and 8 sporadic cases. The rest of the samples (4 JP families and 3 sporadic cases) and exons 2 and 4 in all cases, were directly sequenced. For DGGE conditions, see study I.
In study II, the germline DCC and SMAD4 mutations were initially screened by SSCP assay. The SSCP procedure is described in study II.
The PCR products were purified using the QIAquick PCR purification Kit (QIAGEN). Direct sequencing of the purified PCR products was performed using the ABI PRISM Dye Terminator or ABI PRISM dRhodamine cycle sequencing kits (PE/ABI). Cycle sequencing products were electrophoresed on 6% Long Ranger gels (FMC Bioproducts) and analyzed on an Applied Biosystems model 373A or 377 DNA sequencer (PE/ABI).
In study III, restriction enzyme digestion was used to screen for the presence of two base substitutions in SMAD4 in control individuals. EcoRI (New England BioLabs) digestion was used to detect a C to G change at codon 177 (exon 4). For the analysis of an A to C change at codon 353 AlwI (New England BioLabs) digestion was performed. The detailed digestion procedure is described in study III.
In study IV, the presence of an A to G change at codon 170 (SMAD3 exon 3) in control individuals was analyzed with HgaI (New England BioLabs) digestion. For the detailed procedure, see study IV.
In study V, NsiI (New England BioLabs) digestion was used to detect a G to A change at codon 118 (SMAD4 exon 2). The PCR was performed as described is study V. The detailed digestion procedure is described in study V.
In study III, PAGE analysis was performed to analyze the presence of 4-base-pair deletion in control individuals. A new set of primers was designed for amplification of SMAD4 exon 9. Primers were forward: GGTTGCACATAGGCAAAGGT and reverse: TTGGGTAGATCTTATGAACAGCA (5' to 3'). With these primers, a 156-bp fragment, containing the site of the 4-base-pair deletion, was amplified from exon 9. For PCR conditions, see study III. 10 µl of denaturing loading buffer (95% formamide, 20 mM EDTA, 0.05 % bromphenol blue, 0.05% xylene cyanole FF) was added to 10 µl of PCR sample, and the sample was denaturated for 5 min at 80°C. A 5 µl aliquot of the mixture was loaded in 6% polyacrylamide gels containing 8.3 M urea and run at 2.5 kV for 50 min. Finally, the gels were dried and autoradiographed.
Eight JP families originating from the United States and the United Kingdom were included in the linkage analysis of study I. The microsatellite markers, PCR conditions and statistical methods used for this analysis are described in detail in the original study (study I).
In study III, those two JP families, in which the germline mutations of SMAD4 gene were not detected, were tested for linkage to 18q21. Microsatellite markers D18S970, D18S474, D18S1099, D18S851, D18S484, D18S858 and D18S977 (SMAD4 is located close to markers D18S474 and D18S1099) were used for the analysis. PCR reactions were carried out in a volume of 10 µl containing 50 ng DNA, 1 ´ PCR buffer, 1.5 mM MgCl2, 200 µM each of dATP, dGTP, dTTP and 0.7 µl of [α-32P] dCTP (3000Ci/mmol, Amersham), 0.5 µM of each primer and 0.25 units of AmpliTaqGOLD polymerase (PE/ABI). After PCR amplification in standard conditions, PAGE analysis was performed (see above).
Multipoint linkage analyzes were performed using the GENEHUNTER program (Kruglyak et al. 1996). The JP locus was defined as an affection status locus with dominant inheritance. Three liability classes and age dependent penetrances were set as follows: liability class 1 assigned to age <21 years, with heterozygote penetrance of 50%; liability class 2, age 21-50 years, heterozygote penetrance 70%; liability class 3, >50 years heterozygote penetrance 80%. The JP locus frequency was assumed to be 1/50,000 and the marker loci frequencies and their genetic distances were obtained from the CEPH database V8.1 (http://www.cephb.fr and http://www-genome.wi.mit.edu/cgi-bin/contig/phys_map).
In study V, the methylation status of the SMAD4 promoter was studied in tumors from patients with CRC. The fragment selected for this analysis was CG-rich region, including the non-coding exon 1. To determine whether this particular promoter region was hypermethylated, a PCR-based HpaII and MspI restriction enzyme assay was used. Both tumor and normal DNA were digested and the reactions contained either no enzyme, 25 units of HpaII, or 20 units of MspI. Samples were incubated for 16 h at 37°C. To analyze cleavage of the SMAD4 promoter region, 12.5 ng of DNA from each digest was analyzed by PCR in 25 µl reaction volume. Primers were designed (Primer3) to amplify a 408 bp fragment of SMAD4 promoter containing six HpaII/MspI restriction sites and the primer sequences were: forward: 5'-CAAGTTGGCAGCAACAACAC; and reverse: 5'- ACATGGCGCGGTTACCT.