Their report in the Oct. 22 issue of Science, based on studies in genetically engineered mice, shreds a 30-year-old notion that genes in this region are largely responsible for the condition's characteristic facial features and some of its other common traits. Down syndrome, which affects roughly 1 in 700 live births, is the most common genetic cause of mental retardation and congenital heart disease.
"The simplistic idea that just one of the hundreds of genes on chromosome 21 affects development no longer holds up," says Roger Reeves, Ph.D., professor of molecular biology and genetics in Johns Hopkins' Institute for Basic Biomedical Sciences and McKusick-Nathans Institute of Genetic Medicine. "Now researchers can take a deep breath, accept that the syndrome is complex, and move forward."
Down syndrome occurs when three -- instead of two -- copies of chromosome 21 are present in a fertilized egg, although rare cases occur when a section of the chromosome -- rather than the whole chromosome -- is found in triplicate in a situation called segmental trisomy. A small region of this replicated segment is found in triplicate in all people with segmental trisomy and Down syndrome's facial features, and so it had been dubbed the "Down syndrome critical region" or DSCR. Proponents of DSCR's presumed role had focused on its consistency in people with segmental trisomy, but largely ignored the fact that no one with this condition has only that region in triplicate, Reeves says.
To see whether DSCR is as critical as many suggested, then-graduate student Lisa Olson, Ph.D., created "chromosomally" engineered mice, and found that mice with three copies of just their DSCR equivalent actually had facial and skeletal changes opposite of those seen in Down syndrome.
"These mice weren't normal, but they weren't Down syndrome mice, either," says Reeves, whose lab had already spent 15 years studying the mouse version of DSCR. "Their faces were longer and narrower than normal, but Down syndrome is characterized by shorter than normal facial bones."
DSCR doesn't seem to be required for Down syndrome-like features to result, either, the researchers report. Olson found that mice with just two copies of DSCR but three copies of the rest of the chromosome did have the shorter bones characteristic of Down syndrome.
To measure DSCR's effects in mice, former Hopkins professor Joan Richtsmeier, Ph.D., now at Pennsylvania State University, used mathematical models she developed to compare the length, angles and positions of the facial bones of the mice to Down syndrome's effects in people.
"Some genes in the region contribute to the effects on facial bones, but, in triplicate alone, this region produces different traits than those seen in Down syndrome," says Reeves. "If anyone is going to try to treat the problems seen in Down syndrome, we need to understand what is really happening and when in development it happens.
"Until very recently, we wouldn't have even thought it possible to 'treat' Down syndrome problems," he adds. "The task seems insurmountable -- the genetic problem is there from conception, it's in every cell. But now we're beginning to identify 'developmental cassettes' in mice in which specific problems caused by a triple genetic dose might be modifiable -- if we can figure out the key players."
As part of this effort, Olson used a technique to precisely duplicate a defined chromosome segment and applied it to the DSCR section on mouse chromosome 16, the analog of human chromosome 21. She also made a mouse chromosome 16 that lacked DSCR entirely.
Then scientists in Hopkins' Transgenic Mouse Facility inserted each of the engineered chromosomes into mouse embryonic stem cells, creating stem cells with either three copies of DSCR or only one copy of DSCR. These stem cells were then used to create chimeras -- animals whose make-up comes partially from their original cells and partially from the inserted, engineered stem cells.
Physical inspection of these animals' offspring showed that triple-DSCR mice were bigger than normal mice. Mice with only one copy of DSCR were smaller than normal, similar to a well-studied mouse version of Down syndrome that has three copies of many more of the genes found on human chromosome 21.
Breeding the single-DSCR with the well-studied Down syndrome mouse produced a mouse with only two copies of DSCR but three copies of all other genes on mouse chromosome 16. That "hybrid" mouse was similar to its Down syndrome parent but more mildly affected, the researchers report.
Reeves' lab is now testing another long-standing but poorly supported tenet of Down syndrome research by using the mouse models to study the involvement of neural crest cells, precursors to structures affected in Down syndrome, including the face, heart and the nerves that serve the intestines.
The research was funded by the National Institute of Child Health and Human Development. Authors are Olson and Reeves of Johns Hopkins, and Richtsmeier and Jen Leszl of Penn State. Olson was supported by a fellowship from the Howard Hughes Medical Institute and is now assistant professor of biology, University of Redlands, Calif. Sarah South and Gail Stetten, both of Hopkins, proved the DSCR status of the mice.
On the Web: