Genomic imprinting and DNA epigenetic modification in human liver and carcinogenesis : regulation of the IGF2 and H19 genes

Abstract

Genomic imprinting, which is an non-classical genetic event discovered approximately a decade ago, has been shown to be involved in the mammalian development and several kinds of human diseases. Epigenetic modifications, such as DNA methylation, are believed to be markers functioning as the imprinting signals. With more and more studies focused in this field, imprinting alteration has been found to be involved in some of the genetic diseases inherited tumor syndromes and sporadic tumors. However, the molecular mechanism involved is still unclear.

The human IGF2 gene is located on chromosome llpl5.5 and encodes a protein product of 67 amino acids functioning as a fetal growth factor and a cell mitogen. In fetal tissues, only the paternal allele is expressed while the maternal one is silent. The Hl9 gene is located approximately 200 kb downstream of IGF2. In most of the tissues, only the maternal allele is expressed. The Hl9 and IGF2 genes are believed to be not only physically linked but also correlated by utilizing the same enhancer. Among different forms of epigenetic modifications of DNA, methylation is believed to be the best candidate for the imprinting signal, and evidence has been shown that DNA methylation is involved in the regulation of both IGF2 and Hl9.

In this study, promoter activities, general and promoter-specific imprinting, expression levels and DNA methylation status of the two genes were investigated in normal human livers and primary liver tumors, hepatoblastoma and hepatocellular carcinoma. Results show that the four promoters of IGF2 are under a tight but dynamic control in a developmental-specific way. The P3 and Pl promoters are activated in fetal and adult livers respectively, each producing approximately 70% and 50% of the total IGF2 transcripts. Methylation studies show that developmental and sequence-specific DNA methylation may be involved in the regulation of the promoter activities. The P2 and the P4 promoters are active throughout development with relatively minor changes in expression levels before and after birth. In adult, a relaxation of the imprinting from the P4 promoter can be seen by usage of the two parental alleles randomly. In both hepatoblastoma and hepatocellular carcinoma, promoter Pl is found significantly silenced and this turns out to be an important feature of liver tumors. P3 activity is upregulated in subgroups of both tumors with altered total expression levels of IGF2 and Hl9. The loss of imprinting of IGF2 can be found occasionally in both kinds of tumors.

In conclusion, the aberrant IGF2 regulation, including disrupted promoter control by silencing of P1 and altered promoter-specific DNA methylation, is an frequent phenomenon in human liver cancers. Since the IGF2 gene is regulated in a tight but dynamic, developmental-specific way, any disruption during this procedure may affect the normal liver development. Therefore, the molecular basis of the aberrant IGF2 regulation is worth looking into to understand the molecular tumorigenesis of human liver cancers.