Discoveries Reveal That Gene Regulation Is Bipolar

In this composite image by Greg Grieco, Frank Pugh holds a glass slide with a DNA microarray (magnified to fill the image background)

Two new studies, one to be published on 5 March 2004 in the journal Cell and the other published on 27 February 2004 in Molecular Cell, reveal a surprising relationship among the hordes of gene regulatory molecules that are the ultimate controllers of life processes. The surprise is that only a small portion of all genes — those needed to respond to emergencies — within a simple organism such as baker's yeast are heavily regulated. Most other genes, in contrast, typically control more routine housekeeping functions of the cell and appear to require much less regulation. "It appears that the cell's strategy is analogous to the way people run their lives—we focus more attention on emergencies like an asthma attack rather than on routine but essential housekeeping chores, like laundry," explains Frank Pugh, associate professor of biochemistry and molecular biology at Penn State and the leader of the research teams that made the discoveries.

In addition to Pugh, the researchers include graduate students Andrew D. Basehoar and Kathryn L. Huisinga, and Sara J. Zanton , a senior research technologist. "Often only a select few genes are intensively studied because they undergo lots of exciting regulation," Pugh says. "These highly regulated genes tend to respond to acute stresses like environmental toxins, heat, and viral infection, and are often taken as representative of the types of regulation governing most genes — but this appears not to be the case."

Now with the advent of DNA microarray technology, the regulation of all genes within an organism can be studied simultaneously. "Genome-wide approaches allow us to see the whole 'forest' of genes rather than focusing on just a few of the 'trees’,” Pugh says. By comparing the dependencies of every gene on the hordes of molecular regulators, Pugh noticed that most regulators tended to seek out the same small set of genes — those that typically respond to emergencies — while a select few regulators targeted the vast majority of the genome. When Pugh's team examined some of these regulators in more detail, several additional surprises jumped out.

Basehoar focused on a gene regulatory sequence call the TATA box. While it has long been known that a TATA box is important for proper gene regulation, its exact DNA sequence had remained elusive, in large part due to the prevailing view that any DNA sequence consisting of a random arrangement of As and Ts — two of the four letters in the DNA code — would suffice to function as a TATA box. Basehoar took advantage of recent comparisons of the entire DNA sequence of several related yeast species. Such species have evolved sufficiently that only the DNA sequence of their genes and associated control regions have remain unchanged over time. Using a powerful statistical approach, Basehoar was able to fish out the sequence of the TATA box since it had remained unchanged at many genes. Other sequences that also were rich in As and Ts changed from one species to another, indicating that they have little importance.

As Pugh explains, “It was reassuring that the proposed TATA box sequence passed two additional tests. First, the sequence often resided just upstream of genes, which is where gene regulatory sequences are found. Other A/T-rich sequences were scattered more or less randomly throughout the genome. Second, the expression of genes that contain a TATA box was impaired by genetic mutations along the DNA-binding surface of a protein that normally interacts with the TATA box. We reasoned that genes that have a TATA box are likely to depend on its interaction with its protein-binding partner.”

Pugh’s study reveals that a TATA box is associated with only a small portion of all yeast genes, which goes against the prevailing view that the TATA box is essential to all genes. "This result, plus the knowledge that, all genes are regulated by the TATA binding protein, even those lacking a TATA box, lead us to the second discovery in this study," Pugh says. Guided by hints from recent studies that the TATA binding protein is delivered to genes by either of two massive protein complexes called SAGA and TFIID, Pugh decided to use microarray experiments to investigate the effect of eliminating one complex or the other. Huisinga, who performed these experiments, found that a small fraction of all yeast genes depend primarily on SAGA to deliver the TATA binding protein, while the vast majority of genes depend upon TFIID for delivery. Strikingly, as Pugh puts it, “Genes that used SAGA typically had a TATA box, while genes that used TFIID lacked a TATA box. This was surprising in that it has long been thought that TFIID delivers the TATA binding protein to genes that have a TATA box.”

One final question remained: Was there any connection between the SAGA-TATA relationship and the highly regulated set of emergency-response genes? Indeed, the researchers discovered that there was a strong overlap between the two groups. “Emergency-response genes are designed to be turned on when needed and to be turned off when not needed, which requires a lot of regulation," Pugh explains. "On the other hand, housekeeping genes may not need as much attention, although steady expression of these genes is essential. TFIID may be particularly suited for this role. “

These studies may help guide researchers who are trying to understand a gene's function and its regulation by giving them some useful clues about where to start looking. "If your favorite gene has a TATA box then there is a good chance that it may be subjected to a lot of regulation, and there is a good chance that it may be responding to environmental stress or other transient needs of the cell," Pugh says.

In addition, his lab's findings likely are applicable to genetic studies of higher eukaryotes, including humans, because the regulatory processes involved are highly conserved throughout evolution. "Perhaps we can apply this research to the human genome to study other types of highly regulated responses in addition to stress, such as embryonic development," Pugh says.

This research was supported by the National Institutes of Health.

Contacts:

Frank Pugh : (+1) 814-863-8252, bfp2@psu.edu

Barbara Kennedy : (+1) 814-863-4682, science@psu.edu