DNA methylation is an important regulator of genetic information in species ranging from bacteria to humans. DNA methylation appears to be critical for mammalian development because mice nullizygous for a targeted disruption of the DNMT1 DNA methyltransferase die at an early embryonic stage. No DNA methyltransferase mutations have been reported in humans until now. We describe here the first example of naturally occurring mutations in a mammalian DNA methyltransferase gene. These mutations occur in patients with a rare autosomal recessive disorder, which is termed the ICF syndrome, for immunodeficiency, centromeric instability, and facial anomalies. Centromeric instability of chromosomes 1, 9, and 16 is associated with abnormal hypomethylation of CpG sites in their pericentromeric satellite regions. We are able to complement this hypomethylation defect by somatic cell fusion to Chinese hamster ovary cells, suggesting that the ICF gene is conserved in the hamster and promotes de novo methylation. ICF has been localized to a 9-centimorgan region of chromosome 20 by homozygosity mapping. By searching for homologies to known DNA methyltransferases, we identified a genomic sequence in the ICF region that contains the homologue of the mouse Dnmt3b methyltransferase gene. Using the human sequence to screen ICF kindreds, we discovered mutations in four patients from three families. Mutations include two missense substitutions and a 3-aa insertion resulting from the creation of a novel 3 splice acceptor. None of the mutations were found in over 200 normal chromosomes. We conclude that mutations in the DNMT3B are responsible for the ICF syndrome.DNA methylation ͉ somatic cell complementation ͉ consanguinity ͉ RNA splicing

Chromosomal abnormalities associated with hypomethylation of classical satellite regions are characteristic for the ICF immunodeficiency syndrome. We, as well as others, have found that these effects derive from mutations in the DNMT3B DNA methyltransferase gene. Here we examine further the molecular phenotype of ICF cells and report several examples of extensive hypomethylation that are associated with advanced replication time, nuclease hypersensitivity and a variable escape from silencing for genes on the inactive X and Y chromosomes. Our analysis suggests that all genes on the inactive X chromosome may be extremely hypomethylated at their 5' CpG islands. Our studies of G6PD in one ICF female and SYBL1 in another ICF female provide the first examples of abnormal escape from X chromosome inactivation in untransformed human fibroblasts. XIST RNA localization is normal in these cells, arguing against an independent silencing role for this RNA in somatic cells. SYBL1 silencing is also disrupted on the Y chromosome in ICF male cells. Increased chromatin sensitivity to nuclease was found at all hypomethylated promoters examined, including those of silenced genes. The persistence of inactivation in these latter cases appears to depend critically on delayed replication of DNA because escape from silencing was only seen when replication was advanced to an active X-like pattern.

The XIST gene, expressed only from the inactive X chromosome, is a critical component of X inactivation. Although apparently unnecessary for maintenance of inactivation, XIST expression is thought to be sufficient for inactivation of genes in cis even when XIST is located abnormally on another chromosome. This repression appears to involve the association of XIST RNA with the chromosome from which it is expressed. Reactivated genes on the inactive X chromosome, however, maintain expression in several somatic cell hybrid lines with stable expression of XIST. We describe here another example of an XIST-expressing humanhamster hybrid that lacks X-linked gene repression in which the human XIST gene present on an active X chromosome was reactivated by treatment with 5-aza-2-deoxycytidine. These data raise the possibility that human XIST RNA does not function properly in human-rodent somatic cell hybrids. As part of our approach to address this question, we reactivated the XIST gene in normal male fibroblasts and then compared their patterns of XIST RNA localization by subcellular fractionation and in situ hybridization with those of hybrid cells. Although XIST RNA is nuclear in all cell types, we found that the in situ signals are much more diffuse in hybrids than in human cells. These data suggest that hybrids lack components needed for XIST localization and, presumably, XIST-mediated gene repression.Stable expression of XIST is required on the inactive X chromosome for the establishment of mammalian X chromosome inactivation (reviewed in ref. 1). The role of XIST in the maintenance of repression has been questioned, however. Previous studies of inactive X chromosomes with XIST deletions indicate that XIST RNA is not necessary to maintain X inactivation (2, 3), presumably because other repressive systems, such as promoter methylation, histone deacetylation, and͞or late replication, are maintaining inactivation. Our studies of human-hamster hybrids containing an inactive X chromosome with azacytidine-reactivated genes indicate that XIST expression is not sufficient to prevent reactivation or to reinitiate silencing of these genes (4). A similar conclusion was reached by other workers studying reactivation of X-linked genes in another cell hybrid system (5).To examine this phenomenon further, we reactivated the silent XIST gene on the human active X chromosome in a human-hamster hybrid and in normal human male fibroblasts. The rationale for reactivation was based on the apparent regulation of XIST expression by 5Ј-CG-3Ј dinucleotide methylation. This region is hypermethylated on the silent, active X allele and is hypomethylated on the expressed, inactive X allele in both human (6, 7) and murine (8-11) somatic tissues. A further indication that 5Ј hypermethylation is important in Xist regulation is that the active X allele is expressed in somatic cells of male mice deficient in DNA methyltransferase (12, 13).Repression by 5Ј-CG-3Ј dinucleotide methylation commonly is found for X-inactivated genes, and reactivat...