SNP creates transcriptional elements that interfere with globin expression; similar mechanism could trigger other genetic disorders
By Melissa Lee Phillips | May 26, 2006
Single nucleotide polymorphisms (SNPs) can cause genetic disease by creating entirely new transcriptional elements, according to a report in this week's Science.
The study shows that a SNP in noncoding DNA near the alpha globin gene cluster creates a gain-of-function promoter-like element that interferes with normal globin expression, causing the blood disease alpha thalassemia. Similar mechanisms could be at the heart of other genetic disorders, the authors suggest.
"It's really a novel mechanism by which a disease can be created genetically," said Gerd Blobel of the University of Pennsylvania in Philadelphia, who was not involved in the study.
In alpha thalassemia, reduced expression of any of four alpha globin genes causes anemia and red blood cell abnormalities. Many different mechanisms have been shown to underlie this down-regulation.
Led by Marco De Gobbi of John Radcliffe Hospital in Oxford, UK, and Vip Viprakasit of Mahidol University in Bangkok, Thailand, the researchers studied 148 Melanesian people who had alpha thalassemia but didn't show any of the molecular defects previously known to cause the disease. Through association studies, the researchers found that affected individuals shared noncoding DNA sequences upstream of the globin cluster, suggesting this region might be the source of the causative mutation.
They resequenced the globin cluster plus 213 kb of surrounding DNA in one affected individual, and identified SNPs and other fragments that differed from the wild-type sequence. To uncover SNPs associated with functional changes, the authors constructed a tiled microarray from DNA near the alpha globin cluster. They found that one spot in the DNA from the thalassemic person showed a peak of RNA transcription that was not seen in normal individuals.
"They basically applied a systematic approach to look at all the gene expression activity across the whole locus -- and found something completely unexpected," said Thomas Hudson of McGill University in Montréal, Québec, who was not involved in the study.
The novel transcribed region in the alpha globin locus contained poorly conserved, mostly noncoding sequence. Among 17 SNPs in this region, the authors found that just one segregated with the disease in affected families and associated completely with the stretch of DNA found in people with the unexplained form of thalassemia. This SNP changed the sequence AATA to GATA, thereby creating a potential binding site for GATA-1, a transcription factor essential for erythroid development and gene regulation.
The authors then used chromatin immunoprecipitation (ChIP) to show that the transcription factor GATA-1 does in fact bind to the novel allele. They found that the mutated site attracts a complex often found at erythroid regulatory elements. The site also binds RNA polymerase II and is associated with a new peak of active chromatin.
"These densely tiled arrays that they used and ChIP profiling are great tools, in combination with SNP analysis, to identify mutations" in noncoding regions, said Blobel. "That's one of the highlights" of the study, he said.
The results show that this SNP creates a new "promoter-like" element in the alpha globin locus, the authors say. They also showed that an individual with this mutation has alpha globin expression levels 80 times below normal -- the ultimate cause of the thalassemia. The authors propose that the novel gain-of-function element down-regulates alpha globin genes by out-competing the genes' normal promoters.
Many studies have shown that promoters can arise from a new SNP, said senior author Douglas Higgs, also of John Radcliffe Hospital, "but that always occurs in preexisting elements," he said. In this case, the mutation "creates a transcription factor binding site and a promoter, and apparently creates it out of nothing. We know that that region otherwise doesn't normally do anything -- it's not conserved throughout evolution and we know that you can normally delete it from the alpha globin cluster without it having any effect at all," Higgs said.
"I think that SNPs like the one we've shown here, by and large, people just don't know what to think about them," Higgs told The Scientist. "What I hope we've shown is that people should start thinking about these SNPs because they might be doing something."
"This is a beautiful piece of work," Ross Hardison of Pennsylvania State University in College Station, who was not involved in the study, told The Scientist. "It has implications for how we interpret genome evolution and try to find functional elements within it." Trying to understand how noncoding SNPs function or even trying to map their genomic locations "is a fairly new challenge," Hardison said. And while some studies focus only on regions conserved between species, these data "show that you should not ignore anything."
Melissa Lee Phillips
Links within this article
M. De Gobbi et al., "A regulatory SNP causes a human genetic disease by creating a new transcriptional promoter," Science, May 26, 2006.
C. Choi, "Regulatory DNAs may be missed," The Scientist, March 24, 2006.
A. Munnich, "Treating genetic disease today," The Scientist, May 1, 2006.
D.R. Higgs et al., "A review of the molecular genetics of the human alpha-globin gene cluster," Blood, April 1989.
J.M. Perkel, "Chromatin immunoprecipitation," The Scientist, May 1, 2006.