Genetics of Hamartomatous Polyposis Syndromes

By John M. Carethers, MD

Hamartomatous polyposis syndromes are a group of clinically distinct but perhaps genetically related disorders in which the predominant finding is multiple hamartomatous polyps in the gastrointestinal tract. All of these syndromes are transmitted in an autosomal dominant fashion; however, sporadic forms with germline mutations of a gene that is absent from the biological parents have been described. Hamartomatous polyps are characterized by mature but disorganized tissues that are indigenous to the site of origin, and that are not dysplastic. In spite of the benign appearance of the histologic characteristics, each hamartomatous syndrome has an elevated risk for cancer formation at specific organ sites that cannot be fully explained by germline mutations in specific genes because of genetic overlap between the syndromes.

Clinical Features of the Hamartomatous Polyposis Syndromes

Bannayan-Riley-Ruvalcaba syndrome (BRRS), also known as Bannayan-Zonana syndrome, Ruvalcaba-Myhre-Smith syndrome, and Riley-Smith syndrome because of its phenotypic variability, is a rare, congenital syndrome with features that include intestinal juvenile polyps, macrocephaly, subcutaneous and visceral lipomas and hemangiomas, cognitive and motor developmental delay, lipid storage myopathy, Hashimoto’s thyroiditis, and pigmentary spotting of the penis in males.1 Although the prevalence of BRRS syndrome is unknown, patients reported with this syndrome represent sporadic as well as familial occurrences with an autosomal dominant pattern of inheritance.2

Juvenile polyposis syndrome (JPS) is a congenital syndrome in which 10 or more juvenile polyps occur in the gastrointestinal tract. Unlike solitary sporadic juvenile polyps, familial juvenile polyps almost always recur after removal. Patients present by age 30, with a mean age of presentation of 9.5 years.3 The classical symptom is rectal bleeding, but because of the large number of polyps, patients can present with protein loss, malnutrition, cachexia, and failure to thrive.

Both BRRS and JPS patients harbor juvenile polyps within their intestines. Cowden’s disease, a syndrome in which multiple hamartomas develop on the skin and mucous membranes (its hallmark is facial trichilemmomas), causes polyps that are distinct from juvenile polyps. While polyps in Cowden’s disease may demonstrate a broad range of histology, the most common polyp appears to be a protuberance of cytological normal epithelium that is indigenous to the region from which the polyps arose.4,5 Hyperplastic polyps also are common. There may be cystic dilation of the glands, fibrosis of the lamina propria, and extension of the muscularis mucosae into the lamina propria. The occurrence of these features allows Cowden’s disease polyps to often be histologically confused with juvenile polyps.

Peutz-Jeghers syndrome, with its characteristic mucocutaneous pigmentary spots that appear in association with intestinal hamartomatous polyps, causes a polyp that is distinctive. Peutz-Jeghers polyps demonstrate an arborizing pattern of growth, with the muscularis mucosa extending into branching fronds of the polyp.3 Benign glands within the polyp may be surrounded by smooth muscle and may extend into the submucosa or muscularis propria (pseudoinvasion). Hereditary mixed polyposis syndrome is a recently described syndrome in which affected family members have atypical juvenile polyps, hyperplastic polyps, colonic adenomas, and colonic adenocarcinomas.6 Although this syndrome can present with atypical juvenile polyps, its gene has been linked to chromosome 6q in one extended family,7 and not to the chromosomal sites that have been implicated for BRRS, JPS, and Cowden’s disease.

Genes Involved with the Hamartomatous Polyposis Syndromes

Germline mutations in three genes are associated with the hamartomatous polyposis syndromes. The first is SMAD4, which encodes a key intracellular signal transducer and transcriptional regulator for the TGFb superfamily of ligands and receptors. In the colon, the effect of TGFb action is growth suppression. Thus, uncoupling TGFb action by mutational inactivation of SMAD4 would have the net effect of cellular proliferation. The second is PTEN, which encodes a dual-specific phosphatase (phosphoserine/phosphothreonine and phosphotyrosine residues) that can dephosphorylate proteins (FAK and others), but more importantly lipids (PIP3 and PI3,4P2). PTEN is a tumor suppressor protein which, by its dephosphorylating ability, contributes to programmed cell death and inhibits the cell’s ability to migrate and invade. Thus, mutational inaction of PTEN would remove these phenotypes, and have the net effect of cellular proliferation and enhancing cell migration and invasion. The third is STK11, which encodes a protein with homology to serine/threonine kinases. The pathway(s) that involve STK11 have not been elucidated.

A total of 87 germline mutations of PTEN have been described for Cowden’s disease, BRRS, and JPS at the end of the year 2000.8 Fewer mutations of SMAD4, all of them causing truncation of the protein, have been demonstrated solely in the germline of JPS patients. Mutations in STK11 exclusively occur in the germline of patients with Peutz-Jeghers syndrome. There is no clear genotype-phenotype correlation in these syndromes.

BRRS, JPS, and Cowden’s Disease: One Disease or Multiple Syndromes?

BRRS and JPS might be variants of each other because they share the common feature of intestinal juvenile polyposis, and are both transmitted in an autosomal dominant fashion to offspring. Patients with JPS are predisposed to juvenile hamartomatous polyps and gastrointestinal cancer, with a 15% incidence of colorectal carcinoma in young patients and a cumulative risk of 68% by age 60.9-12 Since the lifetime risk for sporadic colorectal cancer in the United States is about 5%, patients with JPS have a four- to 12-fold elevated risk of developing colorectal carcinoma. Peutz-Jeghers and Cowden’s disease do not carry an increased risk for colorectal cancers, although Peutz-Jeghers syndrome does have a higher incidence of early-onset cancers of the stomach, duodenum, and pancreas, and Cowden’s disease patients have a propensity for breast and thyroid cancers.3

DNA microsatellite markers were used to genetically map a BRRS patient by deletional analysis with an aberration involving chromosome 10q, between 10q23.2 and 10q24.1.2 This area matched an area of deletion also described for a JPS patient with multiple congenital abnormalities,13 and the Cowden critical region linked to chromosome 10q23.14 Some phenotypic features of BRRS have been described that overlap with those commonly found in Cowden’s disease. Facial trichilemmomas have been described in BRRS,15 and three BRRS kindreds have been described in which some members developed thyroid follicular neoplasms.15,16 These phenotypic descriptions, along with Cowden’s disease and BRRS and some patients with JPS commonly mapping to chromosome 10q23, originally supported the possibility that the same gene may be responsible for all of these described syndromes.

The mapped region on chromosome 10q23 that is common to these syndromes was found to contain the PTEN tumor suppressor gene, also known as MMAC1 and TEP1.17-19 Thereafter, two BRRS families demonstrated germline mutations in PTEN (one family demonstrated a missense mutation, and one showed a truncating mutation).20 Thus, loss of chromosome 10q23 and germline mutations of PTEN can cause the phenotype of BRRS. Germline PTEN mutations also have been identified in four of five families with Cowden’s disease,21 and in four families with JPS.22,23 Genetically, germline mutations in PTEN make BRRS, Cowden’s disease, and JPS allelic with each other, at least in some families. PTEN encodes a dual-specificity phosphatase with homology to tensin and auxillin, and has two potential tyrosine phosphate acceptor domains.17,19 Identical germline mutations in PTEN at codon 233, which cause a truncation mutation at one potential tyrosine phosphate acceptor site, have been reported in one family with BRRS and one with Cowden’s disease.20,21 Additionally, a mother with Cowden’s disease had a child with BRRS, with expected identical germline mutations in PTEN.24 As mentioned, BRRS, JPS, and Cowden’s disease appear clinically distinct. Although BRRS and JPS share juvenile polyps as common features and may be variants of each other, juvenile polyps are not common lesions in Cowden’s disease.4,5 In addition, the risk of colon cancer is elevated four- to 12-fold in JPS but not in Cowden’s disease,10,11 and has not been ascertained in BRRS.

While the majority of Cowden’s disease families demonstrate germline mutations in PTEN,25 this has not been the case for JPS and BRRS. Linkage mapping in eight informative JPS families excluded chromosome 10q22-24 as the susceptibility locus for JPS. Furthermore, 14 families with JPS and 11 sporadic JPS cases lacked PTEN mutations by denaturing gradient gel electrophoresis and direct DNA sequencing.26,27 A large Iowa kindred with JPS demonstrated linkage to chromosome 18q21.1.27 Subsequently, three familial and two sporadic JPS cases out of a total of nine demonstrated germline mutations in the SMAD4 gene at chromosome 18q21.1,28 whose gene product is a critical component of TGFb1 signal transduction. The most common SMAD4 mutation was a four-bp deletion from codons 414-416; a 2 bp deletion from codon 348 and a 1-bp insertion at codon 229 also were reported. All of these mutations are predicted to cause a truncated SMAD4 protein and prevent homotrimerization at its carboxyl terminus. Moreover, we failed to demonstrate any deletion of chromosome 10q23 and any germline mutation in PTEN in three BRRS patients.29 The frequency of finding PTEN germline mutations in BRRS is much lower than that observed for Cowden’s disease.25 Thus, the reports of BRRS and JPS patients demonstrating germline PTEN mutations constitute only a portion of the genetic etiology for these diseases. The discovery of a perturbation in the TGFb1 signaling pathway in JPS implicates other components of this transduction pathway in causing JPS and BRRS. SMAD2, another component of the TGFb signaling pathway and one that heterodimerizes with SMAD4, and SMAD4 appear not to be mutated in the three BRRS patients without germline PTEN mutations above (our unpublished observations).

Conclusions

The hamartomatous intestinal polyposis syndromes are inherited as autosomal dominant diseases. The genes involved in these syndromes are predicted to act in a recessive manner in the "target" tissue (i.e., they are tumor suppressor genes and require inactivation of both alleles), in accordance with the Knudson hypothesis. Germline mutations reported for the hamartomatous polyposis syndromes are listed in the Table.

Table-Hamartomatous polyposis syndromes and their associated germline-mutated genes
Hamartomatous Syndrome Chromosomal Location Mutated Gene Frequency in Germline
Hereditary Mixed Polyposis Syndrome 6q unknown -
Juvenile Polyposis (JPS) 18q21.1 SMAD4 21-50%
10q22.3-24.1 PTEN/MMAC1/TEP1 < 10%
Bannayan-Riley-Ruvalcaba (BRRS) 10q23 PTEN/MMAC1/TEP1 50%
Cowden’s Disease (CD) 10q22-23 PTEN/MMAC1/TEP1 > 80%
Peutz-Jeghers Syndrome 19p13.3 STK11 > 90%

Germline mutations in STK11 appear to cause the phenotype of Peutz-Jeghers syndrome exclusively. Likewise, germline mutations of SMAD4 appear to only cause the phenotype of JPS. Germline mutation of PTEN, however, is associated with the phenotypes seen in Cowden’s disease, BRRS, and in some JPS patients. Thus, genetic similarity exists and causes phenotypic heterogeneity (i.e., mutations in PTEN are associated with three hamartomatous syndromes), and genetic heterogeneity exists and causes phenotypic similarity (i.e., mutations in PTEN and SMAD4 are associated with JPS). There are likely other factors that affect these genes to modify the phenotype and each syndrome’s cancer risk. Additional genes likely are involved in at least BRRS and JPS families, as the described mutations in PTEN and SMAD4 do not account for all of the patients with these syndromes. (Dr. Carethers is Staff Gastroenterologist, VA San Diego Healthcare System, GI Section, University of California, San Diego.)

Acknowledgements

This study was supported by a grant from the U.S. Public Health Service (NCI R01 90231).

References

1. Gorlin RJ, Cohen Jr MM, Condon LM, et al. Bannayan-Riley-Ruvalcaba syndrome. Am J Med Genet 1992;44:307-314.

2. Zigman AF, Lavine JE, Jones MC, et al. Localization of the Bannayan-Riley-Ruvalcaba syndrome to chromosome 10q23. Gastroenterol 1997;113:1433-1437.

3. Burt R. Polyposis syndromes. In: Yamada T, ed. Textbook of Gastroenterology, 2nd ed. 1995:1944-1966.

4. Haggitt, RC, Reid BJ. Hereditary gastrointestinal polyposis syndromes. Am J Surg Pathol 1986;10:871-887.

5. Marra G, Armelao F, Vecchio FM, et al. Cowden’s disease with extensive gastrointestinal polyposis. J Clin Gastroenterol 1994;18:42-47.

6. Whitelaw SC, Murday VA, Tomlinson IPM, et al. Clinical and molecular features of the hereditary mixed polyposis syndrome. Gastroenterol 1997;112:327-334.

7. Thomas HJW, Whitelaw SC, Cottrell SE, et al. Genetic mapping of the hereditary mixed polyposis syndrome to chromosome 6q. Am J Hum Genet 1996;58:770-776.

8. Bonneau D, Longy M. Mutations of the human PTEN gene. Hum Mutat 2000;16:109-122.

9. Heiss KF, Schaffner D, Ricketts RR, et al. Malignant risk in juvenile polyposis coli: Increasing documentation in the pediatric age group. J Ped Surg 1993;28:1188-1193.

10. Jass JR, Williams CB, Bussey HJR, et al. Juvenile polyposis—A precancerous condition. Histopathol 1988;13:619-630.

11. Giardiello FM, Hamilton SR, Kern SE, et al. Colorectal neoplasia in juvenile polyposis or juvenile polyps. Arch Dis Child 1991;66:971-975.

12. Jarvinen HJ, KO Franssila. Familial juvenile polyposis coli; increased risk of colorectal cancer. Gut 1984;25: 792-800.

13. Jacoby RF, Schlack S, Sekhon G, et al. Del (10)(q22.3q24.1) associated with juvenile polyposis. Am J Med Genet 1997;70:361-364.

14. Nelen MR, Padberg GW, Peeters EAJ, et al. Localization of the gene for Cowden disease to chromosome 10q22-23. Nature Genet 1996;13:114-116.

15. Fargnoli, MC, Orlow SJ, Semel-Concepcion J, et al. Clinicopathologic findings in the Bannayan-Riley-Ruvalcaba syndrome. Arch Dermatol 1996;132:1214-1218.

16. Miles JH, Zonana J, Mcfarlane J, et al. Macrocephaly with hamartomas: Bannayan-Zonana syndrome. Am J Med Genet 1984;19:225-234.

17. Li J, Yen C, Liaw D, et al. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast and prostate cancer. Science 1997;275:1943-1947.

18. Li D-M, Sun H. TEP1, encoded by a candidate tumor suppressor locus, is a novel protein tyrosine phosphatase regulated by transforming growth factor b. Cancer Res 1997;57:2124-2129.

19. Steck PA, Pershouse MA, Jasser SA, et al. Identification of a candidate tumour suppressor gene, MMAC1, at chromosome 10q23.3 that is mutated in multiple advanced tumors. Nat Genet 1997;15:356-362.

20. Marsh DJ, Dahia PLM, Zheng Z, et al. Germline mutations in PTEN are present in Bannayan-Zonana syndrome. Nat Genet 1997;16:333-334.

21. Liaw D, Marsh DJ, Li J, et al. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet 1997;16:64-67.

22. Olschwang S, Serova-Sinilnikova OM, Lenoir GM, et al. PTEN germ-line mutations in juvenile polyposis coli. Nat Genet 1998;18:12-13.

23. Huang SC, Chen CR, Lavine JE, et al. Genetic heterogeneity in familial juvenile polyposis. Cancer Res 2000;60:6882-6885.

24. Zori RT, Marsh DJ, Graham GE, et al. Germline PTEN mutation in a family with Cowden syndrome and Bannayan-Riley-Ruvalcaba syndrome. Am J Med Genet 1998;80:399-402.

25. Marsh DJ, Coulon V, Lunetta KL, et al. Mutations spectrum and genotype-phenotype analyses in Cowden disease and Bannayan-Zonana syndrome, two hamartoma syndromes with germline PTEN mutations. Hum Mol Genetics 1998;7:505-517.

26. Marsh DJ, Roth S, Lunetta K, et al. Exclusion of PTEN and 10q22-24 as the susceptibility locus for juvenile polyposis syndrome. Cancer Res 1997;57:5017-5021.

27. Howe JR, Ringold JC, Summers RW, et al. A gene for familial juvenile polyposis maps to chromosome 18q21.1. Am J Hum Genet 1998;62:1129-113.

28. Howe JR, Roth S, Ringold JC, et al. Mutations in the SMAD4/DPC4 gene in juvenile polyposis. Science 1998;280:1086-1088.

29. Carethers JM, Furnari FB, Zigman AF, et al. Absence of PTEN/MMAC1 germline mutations in sporadic Bannayan-Riley-Ruvalcaba syndrome. Cancer Res 1998;58:2724-2726.