Numerous studies indicate that genetically programmed changes in methylation of cytosine residues in vertebrate DNA modulate gene expression during development. This may be due to cytosine methylation affecting gene expression either by favoring or by inhibiting specific DNA protein interactions. A number of sequence-specific DNA-binding proteins from vertebrates have been found to have their binding to DNA decreased by methylation of C residues at their DNA recognition sites. Only three vertebrate proteins have thus far been reported to bind better to DNA sequences when they contain 5-methylcytosine (m5C) residues. We discovered the first of these proteins to be described, namely, methylated DNA- binding protein 1 (MDBP-1). Its methylation-specific interactions with three human gene regions will be studied in this proposal. One of these genes is expressed to a very different extent in embryos vs. in adults. The other two genes are expressed from an active X chromosome in female mammals but not from the X chromosome inactivated during embryogenesis in the female. All three genes appear to be down-regulated by methylation. MDBP-1 is a ubiquitous mammalian protein implicated in the control of gene expression. It binds specifically to a set of related 14 base-pair (bp) sites when they are CpG-methylated and to another set of similar 13- or 14-bp sequences in a cytosine methylation-independent mode. We showed that MDBP-1 is indistinguishable from previously described and variously named DNA-binding activities that have been studied in at least eight different labs and that recognize similar sequences in MHC class II gene promoters, a murine ribosomal protein promoter, the human c-myc intron I, the hepatitis B virus (HBV) enhancer l, polyomavirus enhancer B, and a human cytomegalovirus immediate early enhancer. These sites do not require cytosine methylation for MDBP-1 binding. In the proposed research the functional significance of methylation-dependent MDBP-1 sites will be addressed. This is a class of sites described so far only by our lab. Such sites at the beginning of an MHC class I gene, the hypoxanthine phosphoribosyl transferase gene, and the alpha-galactosidase A gene in human cells will be studied by comparing levels of gene expression in transient transfection or in vitro transcription assays. For these assays we will use DNA constructs containing mutant versus wild-type, unmethylated or methylated MDBP-1 sites at naturally occurring strategic locations. Also, the methylation status of the corresponding wild-type sites in vivo will be examined. We will continue trying to clone MDBP-1 cDNA. Lastly, we will determine the effect on gene expression of changing a methylation-independent site in the HBV enhancer to a methylation- dependent site.
Showing the most recent 10 out of 22 publications
