![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
REVIEW
Dapartment of Medicine (Gastrointestinal), Department of Genetics, and Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA
 |
Abstract |
|---|
[Keywords: Colon cancer; hereditary; polyposis; syndromes]
900,000 cases worldwide (http://www.who.int/en). Colorectal cancer-related mortality comprises nearly 55,000 cases annually in the United States (http://www.cancer.org/docroot/stt/stt_0.asp). The lifetime risk of colorectal cancer in the average-risk person, defined as without personal history or family history of colorectal cancer and above the age of 50, is 5%–6%. This increases anywhere up to 20% when there is involvement of first-degree and/or second-degree relatives with colorectal cancer, and reaches a lifetime risk of 80%–100% in hereditary colorectal cancer syndromes, such as in hereditary nonpolyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP), respectively. An explosion of information and insights into the molecular pathogenesis of sporadic colorectal cancer dates back to the late 1980s, and since has served as a paradigm for the investigation of cancer genetics in general, and the emergence of novel diagnostics and therapeutics in cancers as well. Much of this was fueled by the identification, characterization, and elucidation of probands and families with hereditary forms of colorectal cancer. In parallel fashion, an appreciation of the biological underpinnings of colorectal cancer has been transformed through mouse models and delineation of molecular pathways that were predicated upon how these pathways operate to foster different forms of hereditary colorectal cancer.
In assessing the annual cases of colorectal cancer in the United States, it is clear that a distinction should be made for what is truly hereditary and what is in actuality familial. The former connotes a distinct genetic basis that has been defined, whereas the latter comprises an increased predisposition to cancer but without determination, as of yet, as whether there is a hereditary basis with discovery of pertinent tumor suppressor genes that are inactivated in the germline or whether the predisposition is stochastic. In that context, it is estimated that perhaps 20%–30% of all colorectal cancers have a familial basis. A study of 34 kindreds revealed that there is genetic susceptibility to sporadic colorectal adenomatous polyps and colorectal cancer (Cannon-Albright et al. 1988
, 1989). Ongoing efforts should be fruitful in deciphering the potential polygenic basis for familial colorectal cancer. These will be achieved through careful selection of families for genome-wide studies. An example is found in the study of nearly 60 kindreds in which siblings less than age 65 had colorectal cancer but without evidence of hereditary colorectal cancer syndromes. This led to genetic linkage to human chromosome 9q22.2-31.2 (Wiesner et al. 2003).
Hereditary colorectal cancer syndromes, especially with an emphasis on biology and genetics and their relationship to phenotypic manifestations, form the basis for this review (Table 1). Approximately 3%–4% of colorectal cancer cases are attributable to HNPCC or Lynch syndrome, and the slight variation is due likely to geography and ethnicity. Nearly 1% of colorectal cancer cases are due to FAP. Less than 1% of cases are due to a panoply of conditions, namely, MYH-associated polyposis (MAP), the hamartomatous polyposis syndromes, and hyperplastic polyposis.
|
|
FAP |
|---|
). The most compelling feature is the onset and progression of hundreds to thousands of small adenomatous polyps throughout the colon. Such polyps typically trace their emergence to the second decade of life, but have been noted to occur until age 40. Nevertheless, the inevitable course is the development of colorectal cancer unless this distinctive natural history is interrupted by surgical intervention, typically in the form of total proctocolectomy with ileoanal anastomosis (that is, removal of the colon with anastomosis of the terminal ileum of the small bowel with the anal canal via creation of a J-pouch). Certain situations may mandate a subtotal colectomy with ileorectal anastomosis, but the rectal stump or remnant must be monitored for polyp recurrence. Extraintestinal manifestations include gastric fundic polyps, small bowel (especially duodenal) adenomatous polyps, congenital hypertrophy of the retinal pigment epithelium (CHRPE), supernumerary teeth, osteomas, cutaneous lipomas and cysts, thyroid tumors, desmoid tumors, adrenal cortical adenomas, and hepatoblastomas. While many of these features are benign, FAP patients may develop thyroid cancer, gastric adenocarcinoma (<1% lifetime risk), duodenal adenocarcinoma (5%–10% lifetime risk), and/or ampullary adenocarcinoma. Desmoid tumors, which are benign fibroblastic neoplasms (typically intra-abdominal), present a particularly vexing problem due to their proclivity to occur after colectomy and/or to recur after surgical removal of the desmoid tumor itself.
Cytogenetic analysis associated FAP with an interstitial deletion on human chromosome 5q21 (Herrera et al. 1986), which was further expanded on by independent genetic linkage analyses to 5q21. Positional cloning verified the gene responsible for FAP to be the Adenomatous polyposis coli (APC) gene, a landmark effort (Groden et al. 1991; Kinzler et al. 1991; Nishisho et al. 1991). The APC gene contains 15 exons (ORF of 8538 nucleotides) with exon 15 being the largest coding region (6.5 kb). Sundry studies yielded information that germline APC mutations were distributed throughout its 15 exons (especially exon 15, however), and the preponderance of these were nonsense mutations, resulting in a truncated protein with a broad range of molecular masses less than the predicted 310-kDa (2843 amino acids) wild-type APC protein. Some of these germline mutations correlated with phenotypic manifestations, such as CHRPE, the severity of polyposis, and an attenuated form of FAP, designated as attenuated APC (AAPC) or attenuated FAP (Fig. 1). Alternative splicing of the APC gene 5' to exon 1 generates isoforms that may regulate cell growth and differentiation (Carson et al. 2004); additionally, other exon(s) can be alternatively spliced. Attenuated FAP is highlighted by mutations in either the extreme 5' or 3' ends of the APC exons (Spirio et al. 1993), and this may influence potentially the stability of the APC protein with markedly delayed onset of and a diminished range of clinical manifestations (e.g., fewer colonic adenomatous polyps in patients in their 50s or 60s). An APC mutation (T to A at nucleotide 3920 or APC I1307K) was found in 6% of Ashkenazi Jews and 28% of Ashkenazim with a family history of colorectal cancer (Laken et al. 1997). Interestingly, this mutation creates a small hypermutable region, indirectly causing cancer predisposition (Laken et al. 1997; Syngal et al. 2000). However, this mutation has low penetrance in the general population.
|
). Domains of the APC protein have been dissected and shown to correlate with functional relationships (Fig. 1). The APC protein, together with glycogen synthase kinase-3
, phosphorylates cytoplasmic -catenin, thereby leading to -catenin’s degradation (Fig. 1; Su et al. 1993; Korinek et al. 1997; Morin et al. 1997). Most of the -catenin is bound to E-cadherin to facilitate cell–cell contact through the adherens junctions (Gumbiner and McCrea 1993). However, germline or somatic APC mutations render cytoplasmic -catenin stable, resulting in nuclear translocation, where in concert with T-cell factors (TCF), certain target genes are activated transcriptionally. A partial list of the target genes is c-myc, cyclin D1, MMP-7, Axin2/conductin (Leung et al. 2002), and EphB/ephrinB (Batlle et al. 2002).
Other functions have been ascribed to the APC protein, including participation in transcription-independent-mediated apoptosis that involves caspase’s cleavage of APC itself (Qian et al. 2007). APC is involved also in chromosomal segregation. In this context, APC localizes to the ends of kinetochores during mitosis and forms a complex with the checkpoint proteins Bub1 and Bub3 (Kaplan et al. 2001). However, APC mutations disrupt microtubule binding and kinetochore–microtubule binding. This may help to explain only partially the chromosomal instability with aneuploidy and enhancement of loss of heterozygosity observed in sporadic colon cancers initiated by APC mutations (for review, see Nathke 2006) and also the chromosomal and spindle errors noted in embryonic stem cells homozygous for ApcMin/+ (multiple intestinal neoplasia) or Apc1638T alleles (Fodde et al. 2001). It should be emphasized that the chromosomal instability observed in colon cancers involves also loss of p53 function and telomere dysfunction.
Mouse models that have recapitulated features of FAP have a storied history. Mutagenesis studies revealed that the murine intestinal neoplasia (Min) mouse harbors many small intestinal polyps but also large intestinal polyps, although to a smaller extent (Moser et al. 1990). These mice eventually die due to anemia from gastrointestinal (GI) hemorrhage. The gene responsible for this phenotype was identified as the murine homolog of the APC gene (Su et al. 1992). Gene targeting strategies have also led to the widespread use of the Apc1638N (Fodde et al. 1994) and Apc(
716) (Oshima et al. 1995) knockout mice. The Apc1638N mice appear to have a shift of polyps to the colon and certain extraintestinal manifestations that are observed in humans. The Apc(716) mice have microadenomas preceding mature adenomas in the small intestine. Both mouse models, along with the ApcMin/+ mice, have proven useful through numerous studies in colorectal carcinogenesis related to the influence or role of environmental factors (e.g., high-fat diet), chemoprevention agents (aspirin, nonsteroidal anti-inflammatory agents, selective Cox-2 inhibitors), diagnosis (evolving proteomics), therapy, and establishment of intersecting pathways through crossbreeding with other mouse models. Particularly illuminating has been the role of COX-2 (Oshima et al. 1996) and the EP2 prostaglandin receptor (Sonoshita et al. 2001) in the background of Apc deficiency in the mouse and the manner in which this work has stimulated chemoprevention efforts in the human.
Genetic testing has exploited the notion that the APC gene is subjected to truncating germline mutations. Initial approaches were based on an in vitro transcription/translation assay to identify truncated APC proteins (Powell et al. 1993), but complete DNA sequencing is now standard. Mutations are detected in 80% of FAP cases. Upon identification of the APC mutation in an affected family member, then the specific APC mutation can be screened for in at-risk members with very high accuracy. Southern blotting may also unravel uncommon APC deletions. Recently, FAP genetic testing has permitted under appropriate circumstances the potential application of this knowledge to preimplantation in vitro fertilization strategies (Motou et al. 2007). Clinical monitoring commences between ages 10 and 12 with flexible sigmoidoscopy or colonoscopy every year, upper endoscopy every 1–2 yr starting at ages 20–25, annual thyroid exam supplemented as needed with ultrasound, and consideration of other modalities to evaluate other organs where cancer(s) may emerge. In aggregate, the genetic and clinical underpinnings of FAP have served as a platform for the dissection of the molecular properties of the APC protein and its roles in sporadic colon cancer.
|
HNPCC (Lynch) syndrome |
|---|
, and refined through consensus conferences (Vasen et al. 1991, 1999), this syndrome is marked by an autosomal dominant mode of inheritance, early onset of colon cancer often with a predilection for the right colon, and an 80% lifetime risk of colorectal cancer. The syndrome is noteworthy for a spectrum of extracolonic tumors, such as those originating from the endometrium, ovary, stomach, bile duct, kidney, bladder, ureter, and skin (in this context, referred to as the Muir-Torre syndrome with sebaceous adenomas, basal cell cancers, and keratoacantomas). The clinical hallmarks of HNPCC syndrome resulted in a classification scheme designated as the Amsterdam I criteria (Vasen et al. 1991), later modified as Amsterdam II to incorporate the importance of the extracolonic cancers (Table 2; Vasen et al. 1999).
|
). To that end, HNPCC patients have germline mutations in DNA mismatch repair (MMR) genes. In colonic or extracolonic tumors, the wild-type MMR allele is lost through somatic genetic alterations. As a result, DNA replication errors occur in repeat sequences (Aaltonen et al. 1993; Fishel et al. 1993; Ionov et al. 1993), typically in dinucleotide repeats, and are designated as the mutator phenotype. This phenotype can be revealed by PCR-based interrogation of genome-wide microsatellite sequences to evaluate for the possibility of length changes in the microsatellite sequences called microsatellite instability (MSI). The tumor is referred to as MSI-high or MSI-H when two or more markers of the recommended panel by the National Cancer Institute (BAT26, BAT25, D5S346, D2S123, and D17S250 markers) demonstrate MSI, and MSI-low or MSI-L with one unstable marker, or MS-stable or MSS when no microsatellite loci are unstable. Of note, MSI may be found in 15% of sporadic colorectal tumors (Thibodeau et al. 1998). Target genes of MSI include TGFIIR, E2F4, and Bax, among others. Germline mutations may be found in MLH1, MSH2, and MSH6 MMR genes, and account for perhaps 60%–80% of all detectable germline mutations (Fishel et al. 1993; Parsons et al. 1993; Lynch and de la Chapelle 2003; Vasen and Boland 2005). However, somatic MLH1 promoter methylation or MSH2 mutation may be found in MSI-H sporadic colorectal tumors. It is the discovery of MSI that led to the re-evaluation of the Amsterdam classification schema to evolve into the Bethesda, and subsequently revised Bethesda, criteria (Table 2; Boland et al. 1998; Umar et al. 2004). There are also unique histopathological features of MSI-H colorectal cancers, thereby potentially obviating the need for MSI analysis in evaluating whether a family might satisfy the criteria for HNPCC (Jenkins et al. 2007). It has been also advocated that the PREMM (Prediction of Mutations in MLH1 and MSH2) model based on a personal and family history might provide an estimate of the likelihood of finding mutations (Balmana et al. 2006).
The MMR system recognizes and corrects base-pair mismatches and small nucleotide (1–4 base pairs [bp]) insertion/deletion mutations that can occur during DNA replication (Chung and Rustgi 2003; Lynch and de la Chapelle 2003; Edelmann and Edelmann 2004; Vasen and Boland 2005). Eukaryotic MMR mechanisms are more evolved and complex than bacterial MMR. The eukaryotic system, best studied in yeast and in mammalian cells, involves more proteins. These include the homologs of bacterial MutS and MutL, namely, MSH (MSH1–MSH6) and MLH (MLH1–MLH3)/Post-meiotic segregation or PMS (PMS1–PMS2), respectively. In eukaryotes, MSH2 and MSH6 (called MutS
) form a heterodimer and recognize 1-bp mismatches and 1-bp insertion/deletion mutations (Fig. 2). MSH2 and MSH3 form a heterodimer (called MutS) and recognize not only what is recognized by MSH2–MSH6 but expand that capability to recognize 2- to 4-bp insertion/deletion mutations. MSH4 and MSH5 participate in the regulation of meiotic recombination (Ross-Macdonald and Roeder 1994). Upon recognition of DNA mismatches, members of the MLH and PMS gene families are recruited. MLH1 and PMS2 (yeast PMS1) complex with each other (called MutL) and are recruited to form a complex with MSH2–MSH6 or MSH2–MSH3, and this is mediated via MLH1. MLH1–PMS2 trigger a cascade of events that lead to the excision of the DNA strand carrying the mismatched base, which culminates in DNA resynthesis and ligation. The functional significance of MLH1–MLH3 (called MutL
) and MLH1–PMS1 (called MutL) in MMR remains to be elucidated. Excision and resynthesis of 1-bp mutations/insertions/deletions in part involve EX01, a 5'–3' exonuclease that binds MSH2, MSH3, and MLH1 and functions likely through the MSH2–MSH6 correction system (Tishkoff et al. 1997; Schmutte et al. 2001; Tran et al. 2001).
|
). Exo1 deficiency in the mouse has a limited effect on polyp number and size in Apc(1638N) mice (Kucherlapati et al. 2007). Collectively, these data would indicate that other exonucleases are involved, but their identification and characterization are the subject of ongoing investigation. Proteins autonomous from exonucleases may be involved in the later stages of DNA repair. Proliferating cell nuclear antigen (PCNA) binds MLH1, MSH3, and MSH6 (Umar et al. 1996) and potentially may serve to bridge MMR with DNA replication. Other functions have been attributed to MMR proteins, which include involvement in apoptosis (Chung and Rustgi 2003; Edelmann and Edelmann 2004).
Much work has underscored the development and characterization of mouse models that knock out different MMR genes (Edelmann and Edelmann 2004). Msh2–/– mice have reduced life span and a high incidence of lymphomas (T-cell) and small intestinal adenomas and adenocarcinomas with MSI-H in the tumors (de Wind et al. 1995; Reitmair et al. 1995, 1996; Smits et al. 2000). A subset of the mice has sebaceous gland tumors, reminiscent of Muir-Torre syndrome. Msh3–/– mice have a low incidence of late-onset GI tumors with moderate-high MSI (de Wind et al. 1999; Edelmann et al. 2000). Msh6–/– mice develop tumors but with delayed onset compared with Msh2–/– mice (Edelmann et al. 1997). However, the tumors in the Msh6–/– mice have little or no MSI in their tumors. Interestingly, consistent with the phenotype of msh6–/– mice, it was found that a subset of a slightly older cohort of patients with familial non-HNPCC colorectal cancers had germline MSH6 mutations (Kolodner et al. 1999). As might be expected, Msh3–/–, Msh6–/– mice develop tumors similar to Msh2–/– mice and have evidence of MSI-H (de Wind et al. 1999; Edelmann et al. 2000). Mlh1–/– mice behave in a manner similar to Msh2–/– mice with reduced life span and a similar spectrum of tumors (Baker et al. 1996; Edelmann et al. 1996). Their tumors display MSI-H. Msh4 or Msh5 knockout mice have no tumors. Pms1 knockout mice have no tumors (Prolla et al. 1998), and Pms2 knockout mice develop lymphomas and sarcomas but no GI tumors (Baker et al. 1995; Prolla et al. 1998). Collectively, these mouse models emphasize the central importance of MSH2, MLH1, and MSH6 in MMR and HNPCC germline mutations and point to potential functional redundancies of the other MMR proteins.
|
Genetic testing in HNPCC |
|---|
), but the precise algorithm may need to be done in accordance with local practice and recommendations. Fulfillment of Amsterdam I criteria, but without evidence of MSI-H status in the tumor(s), prompts the consideration of familial colorectal cancer X syndrome where germline mutations in the known MMR genes do not appear to exist (Lindor et al. 2005). Clinical monitoring of individuals at risk for HNPCC involves colonoscopy every 2 yr starting between ages 20 and 25, and annually above age 40. Women at risk merit annual endometrial aspiration biopsies and transvaginal ultrasonography to visualize the ovaries, commencing between ages 25 and 35 and occurring annually. Other measures need to be individualized. Patients who are gene mutation carriers should receive counseling about subtotal colectomy given the prominence of colon cancer, and for women, prophylactic hysterectomy and bilateral oophorectomies, especially since endometrial cancer is a defining feature of the disease. |
MAP |
|---|
). These mutations may be missense or nonsense, the latter yielding protein truncation. Two common mutations are Y165C and G382D, accounting for >80% of known mutations. Transmitted in an autosomal recessive inheritance pattern, MAP is defined as involving biallelic inactivation and patients harboring multiple colorectal adenomatous polyps without evidence of FAP or attenuated FAP (Sieber et al. 2003). Monoallelic carriers do not carry an increased risk of colorectal cancer (Balaguer et al. 2007). Myh deficiency has been demonstrated to enhance intestinal tumorigenesis in ApcMin/+ mice (Sieber et al. 2004). However, analyses of families with FAP have not supported the hypotheses that MYH is a disease modifier in FAP (Plasilova et al. 2004).
Clinical findings of an increased number of polyps may trigger suspicion of either MAP or attenuated FAP in the appropriate setting. However, the polyp number can be quite variable in MAP, and it has been demonstrated that if the number of polyps is used as the sole entry criterion, then nearly 25% of cases may be missed in the general population (Jo et al. 2005). Furthermore, these individuals may have a family history consistent with HNPCC, and thus, MYH gene testing may be necessary for patients who meet clinical criteria for HNPCC and who do not have evidence of DNA MMR gene mutations. These patients warrant regular interval screening and surveillance colonoscopy.
|
Peutz-Jeghers syndrome |
|---|
). Other phenotypic features include macules with broad but focal distribution: peroral, buccal, periocular, palmar/plantar surfaces, and anogenital surfaces. These macules wane during puberty but persist in the buccal mucosa. Sinus, bronchial, and bladder polyps may also be evident. As patients survive into their third and fourth decades of life, there is increased risk of other cancers (Giardiello et al. 2001): gastric, pancreatic, breast, ovarian, uterine, cervical, and lung, as well as Sertoli cell tumors (males) and sex cord tumors with annual tubules (females). The latter two tumors actually arise in adolescence and young adulthood.
|
). Shortly thereafter, the actual gene was discovered to be a novel serine–threonine kinase 11 (STK11), or LKB-1 (Hemminki et al. 1998). The tumors associated with Peutz-Jeghers syndrome have 19p LOH or somatic mutations in LKB1, consistent with the notion that LKB1 is a tumor suppressor gene. Mouse models have demonstrated that Lkb1–/– is embryonically lethal due to vascular abnormalities in the yolk sac and placenta (Bardeesy et al. 2002; Jishage et al. 2002). Lkb1+/– mice develop the expected polyps in the GI tract with some predilection for the stomach. As these mice age, they develop liver cancers (Miyoshi et al. 2002; Nakau et al. 2002). Interestingly, Lkb1 may be haplosufficient in polyps (Miyoshi et al. 2002; Nakau et al. 2002); however, Lkb1 LOH has been found also to occur in the epithelium of some of the polyps arising in the Lkb+/– mouse (Bardeesy et al. 2002). When Lkb1+/– mice are crossed into a p53-deficient background, the mice develop hamartomatous polyps at a faster rate and also evolve to contain hepatic adenomas and liver cancers (Takeda et al. 2006). The hepatic adenomas have evidence of Lkb1 LOH, suggesting that p53 loss may stimulate Lkb1 LOH (Takeda et al. 2006). Lkb1 deficiency prevents culture-induced senescence without concordant loss of Ink4a/Arf or p53 (Bardeesy et al. 2002). Lkb1–/– mouse embryonic fibroblasts do not undergo transformation by activated Ha-Ras either alone or with known cooperating transformation-associated oncogenes (Bardeesy et al. 2002).
The biological properties of LKB1 have emerged in recent years, with evidence of unexpected functions in the process. One original theme was that LKB1 is important in regulating cell proliferation and growth, perhaps in part through induction of apoptosis. To that end, ectopic LKB1 overexpression in cells leads to G1 cell cycle arrest (Tiainen et al. 1999). Phosphorylation of LKB1 (Ser 431) by PKA or p90 ribosomal S6 is essential for suppression of cell growth (Collins et al. 2000; Sapkota et al. 2001). LKB1 is needed for brahma-related gene-1-induced growth arrest (Marignani et al. 2001), and LKB1 plays a role in p53-dependent apoptosis (Karuman et al. 2001). These initial findings have been supplanted by the discoveries that LKB1 controls cellular polarity (as its Par4 homologs in model organisms) (Martin and St. Johnston 2003) and also cellular metabolism (Woods et al. 2003; Shaw et al. 2004a, b). Seminal studies have enlightened a role for LKB1 as being involved as a sensor for energy stress and nutrient deprivation (Fig. 4). LKB1 binds and activates AMP (, , subunits) (Shaw et al. 2004b) and related kinases to induce AMP kinase (AMPK), which, in turn, regulates the gene product of TSC2 or tuberin. Tuberin facilitates the generation of Rheb-GDP from Rheb-GTP. Rheb-GTP activates mTOR, which is critical in the regulation of protein synthesis, cell growth, and proliferation through the phosphorylation of S6K1 and rS6 and the inhibition of 4E-BP1 that allows the liberation of eIF4E. Germline LKB1 mutations and somatic LKB1 alterations result in the down-regulation of tuberin, up-regulation of Rheb-GTP, and induction of mTOR (Shaw et al. 2004a). Indeed, the hamartomatous polyps of Lkb1-deficient mice display enhanced activation of downstream effectors of mTOR (Shaw et al. 2004a). It is tempting to speculate that mTOR inhibitors that exist (e.g., Rapamycin), or those in development, might be useful as chemopreventive or therapeutic measures in patients with Peutz-Jeghers syndrome.
|
|
Juvenile polyposis |
|---|
). While reports of gastric, small bowel, and pancreatic cancers are found in the literature, it is unclear if these are true associations with juvenile polyposis. It is important to bear in mind that juvenile polyps can be sporadic.
|
) are reported in FJP (Howe et al. 1998, 2001; Sweet et al. 2005), suggesting that the TGF- pathway is critical in the pathogenesis of FJP. Conditional inactivation of Bmpr1a in mice impairs normal intestinal homeostasis with an expansion of the stem and progenitor cell populations, eventually leading to intestinal polyposis resembling FJP (He et al. 2004). Inhibition of BMP signaling by transgenic expression of noggin results in the formation of numerous ectopic crypt units (Haramis et al. 2004). These changes phenocopy the polyps observed in FJP (Haramis et al. 2004). Targeted disruption of Smad4 was pursued by crossbreeding the conditional Smad4 knockout mice and transgenic mice expressing Cre-recombinase under T-cell-specific promoters (Kim et al. 2006). The T-cell-specific homozygous Smad4 knockout mice developed normally, but life span was shortened. There is thickened mucosa of both the large and small intestines with polyps as well as rectal prolapse. Histology revealed effacement of villus architecture and expansion of the stromal compartment containing cystic lesions lined by columnar epithelial cells and plasma cell infiltrates in the intestine (Kim et al. 2006). In contrast, conditional Smad4 deletion in the intestinal epithelia does not yield intestinal tumors. This highlights the importance of the inflammatory response in the stromal compartment in the regulation of epithelial tumorigenesis. Genetic testing involves the exploration of germline mutational analysis of BMPR1A, SMAD4, or ENG, recognizing that a subset of FJP patients do not harbor these mutations. Clinical monitoring is similar to that recommended for PJS, but the risk for extracolonic cancers is not observed as noted.
|
Cowden’s syndrome |
|---|
). Pten deficiency in the mouse causes a multitude of tumors, including in the thymus, endometrium, liver, prostate, and GI tract, although the GI tract neoplasms are associated with lymphoid tissue (Podsypanina et al. 1999).
|
; Marsh et al. 1997). Comprising lipomas, pigmented macules of the penis, and macrocephaly, BRR also harbors germline PTEN gene mutations (Arch et al. 1997; Marsh et al. 1997), as is observed with Cowden’s. Clinical suspicion merits PTEN genetic testing in at-risk Cowden’s families with careful screening for thyroid and breast lesions, along with that for GI polyps. |
Other hamartomatous polyposis syndromes |
|---|
; Whitelaw et al. 1997). Other potential gene loci have been noted, namely, on chromosomes 15q13-14 (Jaeger et al. 2003) and 10q23 (Cao et al. 2006), although the latter might represent a variant of juvenile polyposis since loss of BMPR1A function was noted in one of the families. Thus, in the consideration of this syndrome, it would appear that the clinical features and types of polyps are important to define, and more studies are needed to establish the genetic underpinnings. Other conditions have involvement of their cognate polyps in different anatomic sites of the GI tract and include neurofibromatosis type I, multiple endocrine neoplasia type 2b, and the Gorlin syndrome (also referred to as the Gorlin-Goltz syndrome or the nevoid basal cell carcinoma syndrome). In each of these aforementioned conditions, the GI tract is one of multiple sites of involvement and not the cardinal feature.
The Gorlin syndrome has an autosomal dominant mode of transmission with multiple basal cell carcinomas; epidermoid cysts; "pits" on the palms and soles; odontogenic keratocysts; jaw, rib, and vertebral abnormalities; ovarian fibromas; short metacarpals; and hypertelorism. There is increased risk of medulloblastoma, and to a lesser extent, fibrosarcoma of the jaw. The underlying genetic basis is attributable to germline PTCH gene (chromosome 9q22.3) mutations, which is critical in sonic hedgehog signaling. One is struck by the overlap between some features of Gorlin syndrome with FAP (e.g., epidermoid cysts, jaw abnormalities), Turcot syndrome (medulloblastoma with germline APC mutations), and Muir-Torre syndrome (basal cell cancers with MSI). This raises the possibility of functional interactions between aberrant wnt signaling and sonic hedgehog signaling in some of the hereditary polyposis syndromes, or that MSI may be detectable in tumors with aberrant sonic hedgehog signaling. Of note, MSI has been detected in some sporadic basal cell carcinomas (Sardi et al. 2000).
|
Hyperplastic polyposis |
|---|
|
|
). It is conceivable that SSAs may be the precursors of MSI-H sporadic colon cancers. Apart from this consideration, SSAs harbor mutations in the BRAF gene, often V600E (Spring et al. 2006). Similarly, BRAF mutations are observed in MSI-H colon cancers (Tanaka et al. 2006) but are uncommon in MSS colon cancers or in sporadic adenomas. Thus, when encountered, SSAs, TSAs, and mixed polyps should be resected endoscopically since the time interval to colorectal cancer, and the frequency in which this occurs, is not known currently. Nor is it defined whether these types of polyps may, in turn, have their own intermediate lesions before evolving into colon cancer. Interestingly, BRAF mutations may be found in microvesicular and goblet cell-rich subtypes of serrated polyps, raising the intriguing possibility that they may progress to SSAs in the correct milieu (Lauwers and Chung 2006). |
Summary and future perspectives |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
E-MAIL anil2{at}mail.med.upenn.edu; FAX (215) 573-5412. 
Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1593107
|
References |
|---|
Amos, C.I., Bali, D., Thiel, T.J., Anderson, J.P., Gourley, I., Frazier, M.L., Lynch, P.M., Luchtefeld, M.A., Young, A., McGarrity, T.J., et al. 1997. Fine mapping of a genetic locus for Peutz-Jeghers syndrome on chromosome 19p. Cancer Res. 57: 3653–3656.
Arch, E.M., Goodman, B.K., Van Wesep, R.A., Liaw, D., Clarke, K., Parsons, R., McKusick, V.A., and Geraghty, M.T. 1997. Deletion of PTEN in a patient with Bannayan-Riley-Ruvalcaba syndrome suggests allelism with Cowden disease. Am. J. Med. Genet. 71: 489–493.[CrossRef][Medline]
Baker, S.M., Bronner, C.E., Zhang, L., Plug, A.W., Robatzek, M., Warren, G., Elliott, E.A., Yu, J., Ashley, T., Arnheim, N., et al. 1995. Male mice defective in the DNA mismatch repair gene PMS2 exhibit abnormal chromosome synapsis in meiosis. Cell 82: 309–319.[CrossRef][Medline]
Baker, S.M., Plug, A.W., Prolla, T.A., Bronner, C.E., Harris, A.C., Yao, X., Christie, D.M., Monell, C., Arnheim, N., Bradley, A., et al. 1996. Involvement of mouse Mlh1 in DNA mismatch repair and meiotic crossing over. Nat. Genet. 13: 336–342.[CrossRef][Medline]
Balaguer, F., Castellvi-Bel, S., Castells, A., Andreu, M., Munoz, J., Gisbert, J.P., Llor, X., Jover, R., de Cid, R., Gonzalo, V., et al. 2007. Identification of MYH mutation carriers in colorectal cancer: A multicenter, case-control, population-based study. Clin. Gastroenterol. Hepatol. 5: 379–387.[CrossRef][Medline]
Balmana, J., Stockwell, D.H., Steyerberg, E.W., Stoffel, E.M., Deffenbaugh, A.M., Reid, J.E., Ward, B., Scholl, T., Hendrickson, B., Tazelaar, J., et al. 2006. Prediction of MLH1 and MSH2 mutations in Lynch syndrome. JAMA 296: 1469–1478.
Bardeesy, N., Sinha, M., Hezel, A.F., Signoretti, S., Hathaway, N.A., Sharpless, N.E., Loda, M., Carrasco, D.R., and DePinho, R.A. 2002. Loss of the Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 419: 162–167.[CrossRef][Medline]
Batlle, E., Henderson, J.T., Beghtel, H., van den Born, M.M., Sancho, E., Huls, G., Meeldijk, J., Rob
Top
Abstract
FAP
