Methylation pattern of the HLA-DR alpha gene in human tissues

Methylation state of the human HLA-DRA gene in T-lymphocytes and B-lymphocytes of transgenic mice. Lack of methylation at one 5'-GCGC site is not required for gene expression

A consistent lack of DNA methylation at one or both of two GCGC (Hha I) restriction sites in the 5' region of the HLA-DRA gene has been previously documented by the use of methyl-sensitive restriction enzymes in human cells and tissues, irrespectively of their expression of DR alpha products. Evidence presently available, however, does not exclude that a lack of methylation in this region, although not sufficient, might be necessary for gene expression. In this report, we show that only one of the 5'-GCGC sites is protected, although less efficiently than in man, from CG-->mCG modifications in tissues and cells of transgenic mice carrying an expressed single copy of the HLA-DRA gene/diploid genome. We demonstrate that the two 5' GCGC sites of the HLA-DRA transgene are fully methylated in DR alpha- splenocytes (more than 80% T-lymphocytes), while one of them (the most 5' site) is not methylated in a fraction of DR alpha+ splenocytes (more than 95% B-lymphocytes). These results provide evidence that absence of DNA methylation in the 5' region is not necessary for, but might be associated with and possibly secondary to the expression of the DRA gene.

Hypomethylation of MHC class II Eb gene is associated with expression

Methylation patterns in the major histocompatibility complex (MHC) class II Eb locus have been analyzed in cell lines representative of different cell types; in particular those with phenotypes found at various stages of B cell development. A series of variant B cell lymphoma lines which serves as a model in which to investigate mechanisms regulating class II gene expression in normal peripheral B cells has been examined. Eb methylation patterns have also been determined in various healthy mouse tissues. The pattern of methylation of the Eb locus varies between different cell lines and tissue types such that hypomethylation is associated with gene expression. There appears to be a methylation pattern which is permissive for class II gene expression and which is characteristic of a variety of cell lines, but is lost in cell lines representing terminally differentiated class II nonexpressing plasma cells. Another methylation pattern has been identified which is found in cloned cell lines selected for expression of very high levels of cell surface class II product. The patterns of methylation associated with MHC class II expression involve changes in methylation sites located within the first intron and several kilobase pairs 5' of the promoter, but no changes were observed in the 3' end of the locus. Moreover, the different methylation patterns do not map to the prominent CpG rich cluster located 5' of the Eb promoter and which remains completely methylated regardless of transcriptional status.

Transgenic mice mimic the methylation pattern of the human HLA-DR alpha gene

The methylation pattern of the human HLA-DR alpha gene has been studied in different tissues of transgenic mice. Offspring from two transgenic lines was selected for this analysis, carrying the integrated HLA-DR alpha gene in either single or multiple (8-10) copies per diploid genome. In transgenic animals two distinct methylation patterns of the HLA-DR alpha gene are generated, due to a complete methylation of all the GCGC and CCGG sites the former, and to unmethylation restricted to one or both the GCGC sites located in the 5' portion of the HLA-DR alpha gene, the latter. Unmethylation restricted to the 5' portion of the HLA-DR alpha gene is a highly conserved feature in human tissues and in vitro cultured cell lines; therefore, it is concluded that the methylation pattern of the human HLA-DR alpha transgene may be faithfully reconstituted in transgenic animals. Northern blotting analysis of the RNA isolated from tissues of the transgenic mouse carrying single-copy HLA-DR alpha transgene demonstrates its tissue specific expression, suggesting that transgenic mice may represent an "in vivo" experimental system to study the relationship between methylation state and transcriptional activation.