Assessing DNA methylation in the developing human intestinal epithelium: potential link to inflammatory bowel disease.

DNA methylation is one of the major epigenetic mechanisms implicated in regulating cellular development and cell-type-specific gene expression. Here we performed simultaneous genome-wide DNA methylation and gene expression analysis on purified intestinal epithelial cells derived from human fetal gut, healthy pediatric biopsies, and children newly diagnosed with inflammatory bowel disease (IBD). Results were validated using pyrosequencing, real-time PCR, and immunostaining. The functional impact of DNA methylation changes on gene expression was assessed by employing in-vitro assays in intestinal cell lines. DNA methylation analyses allowed identification of 214 genes for which expression is regulated via DNA methylation, i.e. regulatory differentially methylated regions (rDMRs). Pathway and functional analysis of rDMRs suggested a critical role for DNA methylation in regulating gene expression and functional development of the human intestinal epithelium. Moreover, analysis performed on intestinal epithelium of children newly diagnosed with IBD revealed alterations in DNA methylation within genomic loci, which were found to overlap significantly with those undergoing methylation changes during intestinal development. Our study provides novel insights into the physiological role of DNA methylation in regulating functional maturation of the human intestinal epithelium. Moreover, we provide data linking developmentally acquired alterations in the DNA methylation profile to changes seen in pediatric IBD.

Specific tumour-associated methylation in normal human term placenta and first-trimester cytotrophoblasts.

Human placentation displays many similarities with tumourigenesis, including rapid cell division, migration and invasion, overlapping gene expression profiles and escape from immune detection. Recent data have identified promoter methylation in the Ras association factor and adenomatous polyposis coli tumour suppressor genes as part of this process. However, the extent of tumour-associated methylation in the placenta remains unclear. Using whole genome methylation data as a starting point, we have examined this phenomenon in placental tissue. We found no evidence for methylation of the majority of common tumour suppressor genes in term placentas, but identified methylation in several genes previously described in some human tumours. Notably, promoter methylation of four independent negative regulators of Wnt signalling has now been identified in human placental tissue and purified trophoblasts. Methylation is present in baboon, but not in mouse placentas. This supports a role for elevated Wnt signalling in primate trophoblast invasiveness and placentation. Examination of invasive choriocarcinoma cell lines revealed altered methylation patterns consistent with a role of methylation change in gestational trophoblastic disease. This distinct pattern of tumour-associated methylation implicates a coordinated series of epigenetic silencing events, similar to those associated with some tumours, in the distinct features of normal human placental invasion and function.

The marks, mechanisms and memory of epigenetic states in mammals.

It is well recognized that there is a surprising degree of phenotypic variation among genetically identical individuals, even when the environmental influences, in the strict sense of the word, are identical. Genetic textbooks acknowledge this fact and use different terms, such as 'intangible variation' or 'developmental noise', to describe it. We believe that this intangible variation results from the stochastic establishment of epigenetic modifications to the DNA nucleotide sequence. These modifications, which may involve cytosine methylation and chromatin remodelling, result in alterations in gene expression which, in turn, affects the phenotype of the organism. Recent evidence, from our work and that of others in mice, suggests that these epigenetic modifications, which in the past were thought to be cleared and reset on passage through the germline, may sometimes be inherited to the next generation. This is termed epigenetic inheritance, and while this process has been well recognized in plants, the recent findings in mice force us to consider the implications of this type of inheritance in mammals. At this stage we do not know how extensive this phenomenon is in humans, but it may well turn out to be the explanation for some diseases which appear to be sporadic or show only weak genetic linkage.

Extracellular ATP couples to cAMP generation and granulocytic differentiation in human NB4 promyelocytic leukaemia cells.

Priming of NB4 promyelocytic cells with all-trans retinoic acid, followed by extracellular ATP in the presence of a phosphodiesterase inhibitor, elevated cAMP and activated protein kinase A. The order of potency for cAMP production was ATP (EC50 = 95 +/- 13 micromol/L) > ADP > AMP = adenosine. The order of potency of ATP analogues was 2'- and 3'-O-(4-benzoylbenzoyl)-ATP (EC50 = 54 +/- 15 micromol/L) = adenosine 5'-O-(3-thio) triphosphate (EC50 = 66 +/- 4 micromol/L) > ATP > beta,gamma-methylene ATP (EC50 = 200 +/- 55 micromol/L). Adenosine 5'-O-thiomonophosphate and adenosine 5'-O-(2-thio) diphosphate inhibited ATP-induced cAMP production. Differentiation also occurred as measured by increased expression of CD11b and N-formyl peptide receptor and changes in cell morphology. UTP did not elevate cAMP or induce differentiation, indicating that P2Y2, P2Y4, and P2Y6 receptors were not involved. The P2Y11 receptor, a cAMP-linked receptor on promyelocytic HL-60 cells, was detected in NB4 cells by reverse transcription-polymerase chain reaction and northern blotting. This receptor has the same order of potency with respect to cAMP production as that observed in HL-60 cells.