Joseph Martens, Lisa Laprade, and Fred Winston found that by switching on the new gene, they could stop the neighboring structural gene from being expressed. "It is the active transcription of another gene that is regulating the process," said Martens, HMS research fellow in genetics and lead author of the June 3 Nature study .
"I cannot think of another regulatory gene such as this one," said Winston, HMS professor of genetics. The researchers have evidence that the new gene, SRG1, works by physically blocking transcription of the adjacent gene, SER3. They found that transcription of SRG1 prevents the binding of a critical piece of SER3's transcriptional machinery.
The discovery raises tantalizing questions. How does this gene-blocking occur? Do other regulatory genes work in this fashion? Does the same mechanism occur in mammals, including humans?
At the same time, SRG1 provides clues to a recent puzzle. Researchers have lately begun to suspect that the long stretches of apparently useless, or junk, DNA might possess a hidden function. In the past year, evidence has been pouring in, not just from yeast but from mammals, that these apparently silent regions produce RNAs, which are associated with transcriptional activity (see Focus, March 5, 2004 http://focus.hms.harvard.edu/2004/March5_2004/biological_chemistry.html). Yet no one has found associated protein products. "For us it is easy to look at those findings and say, 'Well maybe those are examples of what is going on here in yeast,'" said Martens.
If so, the findings would carry an important message for the post-human genome era-namely, that researchers' attempts to turn the masses of data churned out by the Human Genome Project into an understanding of what actually happens in the human body may be even more complex than they anticipated. One of the main challenges for that effort is to figure out how and when genes are turned on and off during normal development and disease. Rather than look only at how genes are regulated by proteins, they would have to turn their attention to a new, and possibly more-difficult-to-detect form of control. And given that junk DNA makes up 95 percent of chromosomes, the mechanism could be fairly common.
"I think if nothing else, this sends up an alert that this likely occurs in other cases," said Winston. "When people are looking to understand regulation of genes from whatever organism-humans, flies, mice, yeast-they cannot just look for proteins that are acting there. It might be that it is simply the act of transcribing that is causing regulation."
Like many researchers, Winston and his colleagues may have known in the back of their minds that someday they would have to contend with junk DNA, but it was not their intention to map a new gene in those wild and relatively uncharted regions of the chromosome. If anything, the yeast SER3 gene was their lodestar. What intrigued them about the gene, which is involved in the synthesis of the amino acid serine, was its unusual expression pattern. To be turned on, genes must first be bound by an activator molecule. A common activator in yeast is a molecule called Switch/Sniff. While most genes are turned on by Switch/Sniff, SER3 is turned off by the complex.
In the course of exploring how this repression happens, Martens came across an even more surprising result. "The usual story when a gene is transcriptionally repressed is that RNA polymerase, TATA binding protein and a host of other factors associated with active transcription, will not be there," he said. He, Laprade, a research associate, and Winston conducted a series of experiments and found that the factors were all present and active, and they were located just upstream of the SER3 promoter-as was a jot of DNA needed for the onset of transcription, the TATA element.
Thinking that the TATA element might signify the beginning of a new gene, one associated with both the active RNA polymerase and SER3 repression, Martens mutated it. "We no longer saw the RNA, and we found transcription of SER3 was de-repressed," he said. "That is when we thought, 'OK, we have got a new regulatory gene.'" After characterizing SRG1, which turned out to be 550 base pairs long, they tackled the question, How is it regulating SER3? They put the question on the table during a lab retreat atop a downtown skyscraper. "Everybody talks, and they are not allowed to show any data," said Winston. Out of that intellectual free-for-all, three models emerged.
The first held that RNA transcripts produced from SRG1 were being recruited to SER3 and were somehow repressing transcription. The researchers assumed that if this were true, it would not matter where the RNA came from. As it turned out, SER3 was repressed only when the RNA was produced by an adjacent SRG1. The second model, which proposed that the SRG1 promoter outcompeted the SER3 promoter for transcription factors, also did not hold up to experimental scrutiny.
There had been hints all along favoring the third model. In this one, transcription of the nearby SRG1 somehow prevents an activator from binding the SER3 promotor. Using chromatin immunoprecipitation, a powerful method for imaging the location of molecules in living cells, the researchers found that this was exactly what happened: a well-known activator fell off the SER3 promotor when SRG1 was turned on. In fact, when SER3 was replaced by a reporter gene, the same thing happened-the turning on of SRG1 prevented the activator from binding to that gene as well.
As for how this interference actually occurs, one possibility is that the machinery required to transcribe SRG1 -RNA polymerase, TATA-binding proteins and other factors-somehow spills over to the nearby SER3 promotor, physically preventing it from being approached by an activator. "It is also possible that active transcription alters chromatin structure and modifies things in other ways," said Winston. As for the molecule that got them started in the first place, Switch/Sniff, the researchers now think it may activate SRG1 and in that way bring about SER3's anomalous repression. "That is our current thinking," Winston said. It is a view he expects will be revised. "Every time we thought we understood everything going on here, we have been wrong. There are additional layers of complexity."
HARVARD MEDICAL SCHOOL
Harvard Medical School has more than 5,000 full-time faculty working in eight academic departments based at the School's Boston quadrangle or in one of 47 academic departments at 18 Harvard teaching hospitals and research institutes. Those Harvard hospitals and research institutions include Beth Israel Deaconess Medical Center, Brigham and Women's Hospital, Cambridge Hospital, The CBR Institute for Biomedical Research, Children's Hospital Boston, Dana-Farber Cancer Institute, Forsyth Institute, Harvard Pilgrim Health Care, Joslin Diabetes Center, Judge Baker Children's Center, Massachusetts Eye and Ear Infirmary, Massachusetts General Hospital, Massachusetts Mental Health Center, McLean Hospital, Mount Auburn Hospital, Schepens Eye Research Institute, Spaulding Rehabilitation Hospital, VA Boston Healthcare System.
Journal
Nature