(From Columbia University Health Sciences Division, Office of External Relations)

Researchers at Columbia University College of Physicians & Surgeons led an international team that identified the first human genetic syndrome caused by defective DNA methylation. The failure to add methyl groups (CH3) to specific sections of DNA produces unstable chromosomes that break apart and rearrange abnormally, resulting in the syndrome Immunodeficiency, Centromere instability, and Facial anomalies (ICF). Associate Professor of Genetics and Development Dr. Timothy Bestor and his colleagues identify the gene responsible for this syndrome in the Nov. 11, 1999, issue of Nature.

The findings show that methylation plays an essential role in the organization and stability of at least one compartment of the human genome. They also suggest that classical satellite DNA, an abundant and repetitive class of DNA of previously unknown function, silences selected genes when methylated by forming small gene "prisons" within the nucleus.

"Because DNA methylation increases the information content of the genome, it can in principle have many functions. But the real functions have been difficult to identify and the subject of much controversy," says Dr. Bestor. "DNA methylation is known to be involved in genomic imprinting and inactivation of one X chromosome in females. This is the first time it has been associated with human disease."

ICF is a rare syndrome that was recognized only in 1988. While the diagnosis has been made in only about 40 people, it is likely to be under-diagnosed. People with the syndrome have mild facial abnormalities, similar in some respects to those of Down syndrome. Patients produce few antibodies, which causes most of them to die of infectious disease before adulthood. The immunodeficiency appears to arise from defective lymphocytes, white blood cells that fight disease.

The lymphocytes have extreme chromosomal abnormalities that result in improper breaking and rejoining of chromosomes 1, 9, and 16. Some of the resulting chromosomes have as many as 12 arms, six times the normal number. The abnormalities are centered around classical DNA satellites, sections of DNA with repeating sequences of either six or 23 nucleotide bases. Although classical satellite DNA represents more than 3 percent of the human genome, it is not transcribed into RNA and has no known function.

Dr. Bestor and his colleagues identified a gene, DNMT3B, both copies of which were mutated in five ICF patients. They showed that DNMT3B codes for an enzyme that adds methyl groups to cytosine nucleotides in the classical satellite DNA. Healthy people have heavily methylated classical satellite DNA. The classical satellite DNA in ICF patients is unmethylated.

In addition to finding the causative gene for ICF and demonstrating a role for DNA methylation in human disease and genome stability, Dr. Bestor and his colleagues believe their findings suggest a role for classical satellite DNA. Dr. Bestor and his colleagues believe that the methylated classical satellite DNA may form prison-like domains within the nucleus that silence genes at different stages of cellular differentiation.

Other laboratories had shown that when the protein product of a gene is no longer needed it becomes associated with domains formed around satellite DNA. Dr. Bestor and his colleagues showed that the classical satellite DNA in these domains is methylated.

They hypothesize that various genes are drawn into these domains where their expression is prevented by the numerous silencing proteins commonly associated with the methylated DNA. The affected gene itself is not methylated, but the proximity of methylated satellite DNA causes it to be inactive.

The lack of DNA methylation in ICF appears to prevent the formation of these gene "prisons" and allows some genes to express themselves continuously rather than only briefly during a specific stage of lymphocyte development. As a result, cellular differentiation progresses abnormally, producing the defective lymphocytes seen in ICF. The researchers are now conducting experiments to test that hypothesis.

Other researchers who made major contributions to this paper were Drs. Evani Viegas-Péquignot and Déborah Bourchis from Hôpital Necker-Enfants Malades in Paris and Drs. James Russo and Xiaoyan Qu of the Columbia Genome Center. The research was supported by grants from the National Institute of Child Health and Human Development and the Leukemia Society of America.