Gastroenterology
Volume 126, Issue 4 , Pages 1190-1193, April 2004

IGF2 loss of imprinting: a potential heritable risk factor for colorectal cancer☆

  • Randy L. Jirtle

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, North Carolina, USA
    • Corresponding Author InformationAddress requests for reprints to: Randy L. Jirtle, Ph.D, Department of Radiation Oncology, Duke University Medical Center, Box 3433, Durham, North Carolina 27710 USA fax: (919) 684-5584

Article Outline

 

Colorectal cancer (CRC) affects more than 100,000 people in the United States each year, and results in more than 50,000 deaths. Therefore, it would be beneficial to have a diagnostic test for identifying those people with an elevated risk of developing this disease. In this issue of Gastroenterology, Cruz-Correa et al.1 provide further support for the novel concept that abnormal insulin-like growth factor 2 (IGF2) imprinting in peripheral blood leukocytes (PBL) may provide the basis for an epigenetic blood test to screen individuals early in life to identify those who are highly susceptible to developing CRC.

Progression from the earliest histologically identifiable colorectal lesion, the aberrant crypt focus, to a frank adenocarcinoma is driven by a progressive accumulation of both genetic and epigenetic alterations. Because the histologically distinct stages of tumor development occur in parallel with discrete molecular changes, CRC has become a powerful model system for dissecting out the molecular mechanisms involved in human cancer formation.

Oncogenes and tumor suppressor genes known to promote CRC when mutated include K-RAS (Kirsten rat sarcoma viral oncogene homolog), TP53 (tumor protein p53), and genes involved in transforming growth factor (TGF)-β activation (M6P/IGF2R [mannose 6-phosphate/insulin-like growth factor 2 receptor])2 and signaling (TGFβR2 and SMAD4 [mothers against decapentaplegic homolog 4]).3, 4 Moreover, the hereditary colon cancer predisposition syndromes, familial adenomatous polyposis and Gardner’s syndrome, result from germline mutations in the adenomatosis polyposis coli (APC) gene. However, it was clear from early on that epigenetic changes are also required to cause human CRC.

Epigenetic changes in the genome are stable but potentially reversible alterations that do not directly involve the DNA sequence, and are heritable through cell division. Epigenetic modifications include methylation of cytosine located within a CpG dinucleotide, and acetylation, methylation, and phosphorylation of histones.5 The first evidence that epigenetics played a role in cancer was the discovery that the genome of CRC is hypomethylated, relative to that of normal colonic epithelia.6 Subsequently, it was shown that coupled with this global hypomethylation of the bulk of the genome is a dense hypermethylation of normally unmethylated CpG islands associated with the promoters of genes involved in cell cycle control (p16INK4a [p16 inhibitor of cyclin-dependent kinase 4a]), DNA repair (hMLH1 [human mutL homolog 1]), and apoptosis (DAPK [death-associated protein kinase]).7, 8, 9, 10

Although tumor-suppressor gene inactivation by promoter hypermethylation occurs frequently in cancer, genome-wide hypomethylation is one of the earliest events to occur in the genesis of CRC.8, 9, 10 Demethylation of CpG islands can lead to the activation of oncogenes such as H-RAS (Harvey rat sarcoma viral oncogene homolog).8, 9 Demethylation can also reactivate transposable elements, thereby altering the transcription of adjacent genes and disrupting normal gene function by promoting the integration of these parasitic DNA elements into other genomic regions. DNA hypomethylation can also affect nuclear structures other than genes, resulting in chromosomal instability. This would favor mitotic recombination, loss of heterozygosity, and chromosomal aneuploidy. Moreover, the loss of DNA methylation can cause the dysregulation of imprinted genes.

Genomic imprinting results from an epigenetic modification in the germ line that leads to parent-of-origin dependent, monoallelic gene expression in somatic cells.11 The first experimental evidence that mammalian maternal and paternal genomes are not equivalent came from mouse nuclear transplantation experiments.12, 13 These elegant studies in the mid-1980s showed that diploid androgenotes derived from 2 male pronuclei and gynogenotes formed from 2 female pronuclei failed to develop properly during embryogenesis. Similarly, complete hydatidiform moles in humans, containing only paternal chromosomes, produce primarily placental tissue from which choriocarcinomas can develop, whereas dermoid cysts, containing only maternal chromosomes, produce primarily embryonic tissue from which teratomas can develop.14

One of the first imprinted genes identified was mouse Igf2,15 and soon thereafter, human IGF2 was also found to be expressed only from the paternal allele.9 Presently, more than 70 imprinted genes have been identified (http://www.geneimprint.com),16 and it is postulated that 100–500 imprinted genes may exist in the human genome. Marsupials and eutherian mammals are imprinted at the IGF2 locus, whereas the egg-laying monotremes are not imprinted.16, 17 Thus, IGF2 imprinting evolved approximately 150 million years ago in the late Jurassic period with the advent of live birth.

Multiple theories have been proposed to explain the origins of imprinting early in mammalian evolution.18, 19 According to the most debated of these theories, the “conflict hypothesis,‘20 imprinting is viewed not as an adaptation that presently benefits species survival, but rather as a consequence of an ancient reproductive battle between the sexes that involved polyandry, viviparity, and a skewed maternal versus paternal investment in the offspring. According to this provocative theory, imprinting arose because of a genetic tug-of-war between the parents to control the amount of nutrients extracted from the mother by her offspring. The conflict theory predicts that paternally expressed genes promote prenatal growth to benefit offspring fitness whereas maternally expressed genes suppress growth to maximize reproductive performance.21

Regardless of which hypothesis correctly describes why imprinting evolved, it is now clear that, because imprinted genes are functionally haploid, they markedly increase our susceptibility to cancer. This was first shown with the discovery that pathologic biallelic expression of IGF2 occurs early in the genesis of sporadic Wilms’ tumors.9 The oncogenic role of IGF2 loss of imprinting in Wilms’ tumor formation was further substantiated with the finding that the incidence of this juvenile kidney tumor is greatly increased in patients with Beckwith-Wiedemann syndrome in which IGF2 loss of imprinting arises either in the germline or early in development.9 IGF2 dysregulation in Wilms’ tumor is highly associated with inappropriate maternal methylation of the imprint control region (ICR) upstream of the H19 gene (Figure 1).9, 22, 23 Biallelic expression of IGF2 is now known to occur not only in Wilms’ tumor but also in a large number of adult human cancers, including CRC.24

  • View full-size image.
  • Figure 1. 

    Genomic structure of the human IGF2/H19 imprinted domain. IGF2 (9 exons; black boxes) and H19 (4 exons; stippled boxes) are reciprocally imprinted. IGF2 is expressed only from the paternal allele, whereas H19 is expressed only from the maternal allele. Regulation of IGF2 and H19 imprinting is controlled by allele-specific methylation at differentially methylated regions (DMR0) and the imprint control region (ICR). Black (methylated) and white (unmethylated) boxes above the genes indicate regions of preferential maternal (M) or paternal (P) CpG methylation. IGF2 has 4 promoters, P1–P4, that are located 5′ to exons 1, 4, 5, and 6, respectively. Enhancers (E) involved in regulating the reciprocal imprinting of IGF2 and H19 are also shown.

In contrast to Wilms’ tumor, IGF2 loss of imprinting in CRC unexpectedly involves hypomethylation rather than hypermethylation of the H19 ICR.25 In addition, the differentially methylated region upstream of IGF2 exon 3 (DMR0) is hypomethylated (Figure 1), and only this epigenetic alteration is tightly linked with IGF2 loss of imprinting in CRC.25, 26 Surprisingly, IGF2 loss of imprinting is also found in normal endoderm-derived colonic mucosa of patients with CRC,26, 27 and even in 10% of people who are healthy.26 Moreover, IGF2 loss of imprinting is present in mesoderm-derived PBL, and abnormal IGF2 imprinting in this tissue is highly correlated with both a familial and personal history of CRC.26 Taken together, these findings suggest that the epigenetic alteration responsible for abnormal IGF2 imprinting in these patients is an early event in embryogenesis.

In this issue of Gastroenterology, Cruz-Correa et al.1 add to these intriguing findings by assessing whether IGF2 loss of imprinting in PBL is associated with known environmental risk factors for CRC. A total of 172 individuals were examined for IGF2 loss of imprinting in PBL. Individuals with colorectal neoplasia (adenomas/cancer) were found to have a 5.1-fold increased risk of IGF2 loss of imprinting in PBL than people without colorectal tumors. In contrast, tobacco smoking; alcohol consumption; nonsteroidal antiinflammatory agent use; and the nutrient ingestion of calcium, folate, selenium, and fat were not correlated with IGF2 loss of imprinting. These findings provide compelling support for the postulate that abnormal IGF2 imprinting in PBL is not environmentally acquired during adulthood, but rather occurs early in embryonic development. Consequently, detection of this epigenetic perturbation could possibly be performed early in life, allowing for cancer-preventive measures to be started in high-risk patients before the early stages of CRC are first visually evident.

The most obvious unanswered question remaining is whether IGF2 loss of imprinting in PBL results from an inherited genetic mutation, and/or an epigenetic alteration induced by an environmental perturbation early in embryogenesis.29 Furthermore, it is important to know if the frequency of IGF2 loss of imprinting in PBL depends on factors such as (1) ethnicity; (2) geographical location; (3) the smoking, drinking, and nutritional habits of the birth mother; and (4) birth weight. It is necessary to also determine if abnormal imprinting of IGF2 is present in other normal tissues, and, if so, whether or not it predisposes an individual to cancer in those tissues. It would even be fascinating to determine if other imprinted genes are deregulated in individuals with IGF2 loss of imprinting, particularly other reciprocally imprinted genes, such as DLK1 (delta-like 1 homolog) and (maternally expressed gene 3) MEG3.30 Finally, it will need to be determined if the incidence of other pathologies such as obesity, diabetes, cardiovascular disorders, and even behavioral disorders are higher in individuals with abnormal IGF2 imprinting than in people who do not have this epigenetically induced perturbation in imprinting regulation.

The role of epigenetics and imprinting in carcinogenesis has long been in the shadow of human cancer genetics9; however, the compelling findings of this study have the potential to reverse this trend. It now appears possible to identify individuals early in life who are at high risk of developing sporadic CRC by screening the general population for epigenetic changes that result in IGF2 loss of imprinting. Thus, it may someday be possible not only to reduce their risk of CRC by increasing colonoscopy surveillance but also to directly modify their cancer risk with the use of dietary and/or therapeutic agents developed to reverse the epigenetic alterations that resulted in abnormal IGF2 imprinting. Because imprinting regulation often varies markedly between species,31, 32, 33 the best model for these important future studies remains the human.

28

Back to Article Outline

References 

  1. Cruz-Correa M, Cui H, Giardiello FM, Powe NR, Hylind L, Robinson A, et al. Loss of imprinting of Insulin Growth Factor-II gene (a potential heritable biomarker for colon neoplasia predisposition). Gastroenterology. 2004;126:964–970
  2. Jirtle RL. Genomic imprinting and cancer. Exp Cell Res. 1999;248:18–24
  3. Grady WM, Markowitz SD. Genetic and epigenetic alterations in colon cancer. Ann Rev Genomics Hum Genet. 2002;3:101–128
  4. Kahlenberg MS, Sullivan JM, Witmer DD, Petrelli NJ. Molecular prognostics in CRC. Surg Oncol. 2003;12:173–186
  5. Momparler RL. Cancer epigenetics. Oncogene. 2003;22:6479–6483
  6. Feinberg AP, Vogelstein B. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature. 1983;301:89–92
  7. Veigl ML, Kasturi L, Olechnowicz J, Ma AH, Lutterbaugh JD, Periyasamy S, et al. Biallelic inactivation of hMLH1 by epigenetic gene silencing, a novel mechanism causing human MSI cancers. Proc Natl Acad Sci U S A. 1998;95:8698–8702
  8. Esteller M, Herman JG. Cancer as an epigenetic disease (DNA methylation and chromatin alterations in human tumours). J Pathol. 2002;196:1–7
  9. Feinberg AP, Tycko B. Timeline (the history of cancer epigenetics). Nat Rev Cancer. 2004;4:1–11
  10. Herman JG, Baylin SB. Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med. 2003;349:2042–2054
  11. Reik W, Walter J. Genomic imprinting (parental influence on the genome). Nat Rev Genet. 2001;2:21–32
  12. McGrath J, Solter D. Completion of mouse embryogenesis requires both the maternal and paternal genomes. Cell. 1984;37:179–183
  13. Surani MA, Barton SC, Norris ML. Development of reconstituted mouse eggs suggests imprinting of the genome during gametogenesis. Nature. 1984;308:548–550
  14. Mutter GL. Role of imprinting in abnormal human development. Mutat Res. 1997;396:141–147
  15. DeChiara TM, Robertson EJ, Efstratiadis A. Parental imprinting of the mouse insulin-like growth factor II gene. Cell. 1991;64:849–859
  16. Murphy SK, Jirtle RL. Imprinting evolution and the price of silence. Bioessays. 2003;25:577–588
  17. Killian JK, Byrd JC, Jirtle JV, Munday BL, Stoskopf MK, MacDonald RG, et al. M6P/IGF2R imprinting evolution in mammals. Mol Cell. 2000;5:707–716
  18. Reik W, Walter J. Genomic imprinting (parental influence on the genome). Nat Rev Genet. 2001;2:21–32
  19. Sleutels F, Barlow DP. The origins of genomic imprinting in mammals. Adv Genet. 2002;46:119–163
  20. Moore T, Haig D. Genomic imprinting in mammalian development (a parental tug-of-war). Trends Genet. 1991;7:45–49
  21. Haig D, Graham C. Genomic imprinting and the strange case of the insulin-like growth factor II receptor. Cell. 1991;64:1045–1046
  22. Steenman MJC, Rainier S, Dobry CJ, Grundy P, Horon IL, Feinberg AP. Loss of imprinting of IGF2 is linked to reduced expression and abnormal methylation of H19 in Wilms’ tumour. Nat Genet. 1994;7:433–439
  23. Moulton T, Crenshaw T, Hao Y, Moosikasuwan J, Lin N, Dembitzer F, et al. Epigenetic lesions at the H19 locus in Wilms’ tumour patients. Nat Genet. 1994;7:440–447
  24. Falls JG, Pulford DJ, Wylie AA, Jirtle RL. Genomic imprinting (implications for human disease). Am J Pathol. 1999;154:635–647
  25. Cui H, Onyango P, Brandenburg S, Wu Y, Hsieh CL, Feinberg AP. Loss of imprinting in CRC linked to hypomethylation of H19 and IGF2. Cancer Res. 2002;62:6442–6446
  26. Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, Brandenburg S, et al. Loss of IGF2 imprinting (a potential marker of CRC risk). Science. 2003;299:1753–1755
  27. Cui H, Horon IL, Ohlsson R, Hamilton SR, Feinberg AP. Loss of imprinting in normal tissue of CRC patients with microsatellite instability. Nat Med. 1998;4:1276–1280
  28. Uejima H, Lee MP, Cui H, Feinberg AP. Hot-stop PCR (a simple and general assay for linear quantitation of allele ratios). Nat Genet. 2000;25:375–376
  29. Waterland RA, Jirtle RL. Transposable elements (targets for early nutritional effects on epigenetic gene regulation). Mol Cell Biol. 2003;23:5293–5300
  30. Wylie AA, Murphy SK, Orton TC, Jirtle RL. Novel imprinted DLK1/GTL2 domain on human chromosome 14 contains motifs that mimic those implicated in IGF2/H19 regulation. Genome Res. 2000;10:1711–1718
  31. Sullivan MJ, Taniguchi T, Jhee A, Kerr N, Reeve AE. Relaxation of IGF2 imprinting in Wilms tumours associated with specific changes in IGF2 methylation. Oncogene. 1999;18:7527–7534
  32. Killian JK, Nolan CM, Wylie AA, Li T, Vu TH, Hoffman AR, et al. Divergent evolution in M6P/IGF2R imprinting from the Jurassic to the Quaternary. Hum Mol Genet. 2001;10:1721–1728
  33. Paulsen M, El-Maarri O, Engemann S, Strodicke M, Franck O, Davies K, et al. Sequence conservation and variability of imprinting in the Beckwith-Wiedemann syndrome gene cluster in human and mouse. Hum Mol Genet. 2000;9:1829–1841

☆ Supported by National Institutes of Health Grants CA25951 and ES08823.

PII: S0016-5085(04)00298-7

doi:10.1053/j.gastro.2004.02.026

Gastroenterology
Volume 126, Issue 4 , Pages 1190-1193, April 2004

Article Tools

* Image Usage & Resolution