A Mystery Solved

By Jason Socrates Bardi

That blessed mood,
In which the burden of the mystery,
In which the heavy and the weary weight
Of all this unintelligible world,
Is lightened—that serene and blessed mood...

—William Wordsworth, Lines Composed a Few Miles Above Tintern Abbey, July 13, 1798  

Now that the genomes of humans, yeast, and over a hundred other organisms have been solved and partially or completely annotated, one of the largest tasks left to science is to figure out what all those genes do.

For years, biologists looking to understand what certain genes do in humans have turned their attention to that fermenting foam of rising fame—yeast. There is one simple reason for this: species like the budding yeast Saccharomyces cerevisiae (baker’s yeast) are useful model organisms in which to study basic biology.

Yeast is much more malleable than human cells, and it lends itself to laboratory work. Compared to human cells, it can more easily be genetically altered, and its rapid lifecycle means experiments can be done in hours or days instead of weeks or months. Plus, yeast boasts much of the same basic biochemistry as human cells and shares many of the same types of proteins, mechanisms, and molecular controls.

This means that if a scientist discovers something about what a protein in yeast does, chances are there is a similar protein in human cells that does the same thing—perhaps it even has a similar amino acid sequence or the same three-dimensional structure. And indeed, many discoveries in yeast have led to discoveries in human cells.

There are also occasionally examples of discoveries of genes in humans that predate the same discoveries in yeast.

Such was the case in genes involved in the cell cycle initiation, the first step in one of the most basic and complicated processes in life where a single cell replicates its DNA and divides into two daughter cells, an area of focus for Associate Professor Curt Wittenberg in the Molecular Biology Department at The Scripps Research Institute. The ultimate goal of this research is to translate what is discovered into a better understanding of how to treat human disease.

“Cancer is a disease of cell cycle regulation,” says Wittenberg. “Cell cycle research is cancer research, by its very nature.”

What Was One Becomes Two

The cell cycles of humans and yeast are not identical, but they are very similar. When human cells divide, they undergo a complicated process whereby the membrane that envelops the DNA-containing nucleus is broken down, the genomic DNA is doubled and then pulled apart, the nuclear envelope is reformed around the two identical genomes, and then cytokinesis (the actual fission or division of the cell into two equal halves) occurs.

Yeast cells divide almost identically to human cells, with a few nuanced differences. For instance, budding yeast divides asymmetrically. When budding yeast divides, it does so by producing a small protruding “bud” into which one copy of the duplicated DNA is moved before the bud grows and pinches off.

Decades ago, when the cycles of a cell were first being worked out, scientists divided the cycle into distinct stages that could be distinguished by their outward appearance. In recent years, scientists have moved well beyond this simple description and have begun to identify the hundreds of genes and proteins active at particular times in the cell cycle and to map the coordinated action of these myriad molecules.

For instance, one of the initial steps in mitosis is that the DNA in the cell must be duplicated so that when the cell divides, each daughter cell will have its own DNA. Hundreds of different proteins carry out and control just this one step of the cell cycle, and once the cell enters the cell cycle, it must begin to express hundreds of genes that will help it along the way. But before any of this coordinated gene expression can occur, says Wittenberg, the cell must “decide” to enter the cell cycle. This decision—to progress from the G1 to the S phase in the jargon of cell biology—is a crucial moment in the life of a cell.

If a cell does not detect the right growth factors or nutrients, for instance, it will not initiate the cell cycle and begin replicating its DNA. Understanding how the decision to enter the cell cycle is made on the molecular level is one of the keys to understanding control of the cell cycle.

For a number of years, scientists have known that a human protein called the retinoblastoma tumor suppressor (Rb) is involved in the signaling pathway that controls the cell cycle initiation in human cells. In human cells, Rb blocks transcription until the cell is ready to enter the cell cycle.

“Rb is the critical element that restrains cell division in normal cells,” says Wittenberg.

Rb is what is known technically as a transcription repressor. It represses the expression of a large number of genes involved in the cell cycle by sitting on their transcription factors, the proteins that promote the expression of these genes. This keeps them inactive until Rb receives the right signal to release. Once Rb gets the right signal, it releases the transcription factors under its control, the floodgates of gene expression are thrown open, and cell division is initiated.

This function can make Rb something of a dangerous protein. If it is not regulated correctly, or if it is mutated and cannot repress transcription at all, the cell may lose its control over the timing of the cell cycle. Cancer is a disease of cell cycle regulation, and cells with malfunctioning Rb may divide uncontrollably—one of the hallmarks of cancer.

For this reason, Rb is what is known as a tumor suppressor protein, and mutant forms of Rb causes retinoblastoma, a devastating type of retinal cancer in children. Moreover, says Wittenberg, Rb is known to be involved in a wide variety of cancer tumors. “Rb is found to be either mutated or misregulated in essentially all human cancers,” he says.

But while the identity and function of Rb have been known for several years, the equivalent protein was not known in yeast. In yeast, says Wittenberg, scientists have not known which molecule controls the initiation of the cell cycle. In fact, he says, the best explanation of how this initiation occurred has been something like that old Far Side comic strip where a physics teacher is questioning his student’s proof, which reads “and then a miracle occurs.”

Miracles, or mysteries, rarely last indefinitely against the onslaught of science. A number of laboratories, including Wittenberg’s, began in the mid-1990s to make concerted efforts to identify these factors—the transcription repressor and the other proteins that initiate the cell cycle.

“Now” says Curt Wittenberg, “what we’ve discovered in yeast is an analogous pathway [to retinoblastoma].”

A Mystery Explained

This discovery arose from work on a large family of enzymes known as the cyclin-dependent kinases, a class of enzymes that are formed from the union of two different types of proteins—the cyclins and the kinases.

Cyclin-dependent kinases are a huge family of workhorse proteins in human and yeast cells that carry out a wide variety of duties by attaching chemical groups known as phosphates to other proteins. These phosphates modify the action of the other proteins, turning them on or off, for instance, and the cyclin dependent kinases provide a handy way to rapidly control the action of these other proteins in a cell.

Cyclin-dependent kinases have a “CDK” or kinase end, which carries out the chemistry of hanging the phosphate groups on other proteins, and a cyclin end, which acts as a regulatory unit that controls where and when the CDK can phosphorylate.

In humans, multiple CDKs and cyclins come together to make active cyclin-dependent kinases that regulate the cell cycle. In yeast, only one CDK is essential for cell cycle regulation, but there are nine different cyclins that are involved. Each cyclin drives the CDK to different targets in different parts of the cell at different times in the cell cycle, and one class, known as the “Cln” family, are important at the beginning of the cell cycle in yeast.

“Virtually all cell cycle transitions involve regulation by cyclin-dependent kinases,” says Wittenberg.

Wittenberg’s laboratory looks at how different cyclins work with yeast’s sole CDK at various stages in the cell cycle. Nearly a decade ago, Wittenberg and his colleagues showed that one of these binary enzymes, known as Cln3/CDK, was required for budding yeast to enter the cell cycle. “It turns on a number of transcription factors that are already present on the genes,” says Wittenberg, “And then the events of the cell cycle ensue.”

In a recent interview, Wittenberg described the mechanism by which this occurs. If Cln3/CDK is overexpressed in yeast, he says, the cells become uniformly small—about two-thirds of their regular size. The reason for this is that excess Cln3/CDK prematurely pushes cells into the cell cycle. Conversely, says Wittenberg, if you inhibit Cln3/CDK, the yeast cells become larger than normal because cell cycle initiation is inordinately delayed.

But even though scientists have understood for a few years that the cell cycle is controlled by Cln3/CDK, Wittenberg says, nobody has known how Cln3/CDK exerted that control. A couple of years ago, Scripps Research Associate Rob de Bruin came to Wittenberg’s laboratory and immediately set out to discover how. “We wanted to know what happened right as the cells entered the cell cycle and became licensed to divide,” says Wittenberg. They knew that Cln3/CDK was not acting alone, since Cln3/CDK is not itself a transcription activator, and they knew it must act through some intermediary much like Rb in humans. But no obvious yeast equivalent of Rb emerged.

Now, in an article featured on the cover of a recent issue of the journal Cell, Wittenberg and his colleagues report that the yeast equivalent of Rb is a protein called Whi5.

Whi5 had been identified about two years earlier in yeast mutagenesis experiments. These are the biological equivalent of a fishing expedition in which scientists separately mutate every gene in an organism like yeast, look for interesting characteristics in the survivors, and then map the characteristics (the phenotypes) to the responsible genes (genotypes).

When the Whi5 gene was mutated, the yeast cells became uniformly small—due to the fact that the cells enter division before they are full size. This struck Wittenberg and de Bruin as being similar to what they had seen when Cln3/CDK was overexpressed. Could Whi5 be the mysterious transcription repressor that initiates the cell cycle under the control of Cln3/CDK?

In some ways, Whi5 seemed like a poor candidate. It looks nothing like Rb in terms of its sequence of amino acids.

The scientists showed, however, that control of cell cycle initiation is ultimately in the hands of Whi5 and that in budding yeast, Whi5 carries out the same function as retinoblastoma in human cells.

Wittenberg and de Bruin showed in vitro and in vivo that Whi5 is a target for phosphorylation by Cln3/CDK.

When Cln3/CDK phosphorylated Whi5, this phosphorylation resulted in the dissociation of Whi5 from the DNA, where it was repressing transcription. This led to the expression of the genes Whi5 was repressing and the initiation of the cell cycle.

Cln3/CDK is what makes Whi5 come off, and when Whi5 does come off, transcription becomes active and the cell is driven into the cell cycle and division. When they mutated the target phosphorylation sites on Whi5, entry into the cell cycle was delayed.

“This is a big step in understanding the relationship between human cell cycles and yeast cell cycles.”

The Technical Prerogatives

Wittenberg and de Bruin were able to do this work by employing a number of techniques that allowed them to halt the cell cycle progression of the yeast cells. This allowed them to create pools of cells that were all at the same place in the cell cycle.

The other innovation was a proteomics technology known as MudPIT. MudPIT, a acronym for multidimensional protein interaction technology, had been pioneered at Scripps Research by Professor John Yates, an author on the paper with de Bruin and Wittenberg.

MudPIT basically combines liquid chromatography (which is like a molecular "sieve" that separates a complex mixture into its component parts) with tandem mass spectrometry (which identifies the components based on their masses). Scientists take the gene encoding a protein and fuse it with a molecule that can tightly bind to two other molecules—calmodulin and a type of immunoglobin. This allows the researchers to perform a two-step purification and identify proteins that bind to their protein of interest. In this case, the researchers analyzed transcription complexes. The mass spectrometry instrument then detects these masses and uses sophisticated software to identify thousands of separate proteins.

This technique found a number of hits, and the scientists are studying several of these. One of them was Whi5, which turned out to be the molecule they were looking for.

While this discovery is a big piece of the puzzle, there are still many details that must be uncovered before the complete picture of cell cycle control is known, says Wittenberg. What does Whi5 do when it is bound to the transcription factors? Which sites on Whi5 are phosphorylated by Cln3/CDK? What other proteins interact with Whi5 and what do they do?

The researchers are also looking at these details in the context of the greater topic of how the cell cycle is regulated. They are asking questions such as how proteins are generated at distinct periods during the process and how environmental influences are brought to bear on the cell cycle.

In the end, says Wittenberg, understanding these regulatory systems in yeast will be beneficial because it will reveal more about the analogous human system of regulating the cell cycle, which goes awry in diseases like retinoblastoma.

“Hopefully by learning something about the details of the yeast system, we can learn something that will help us understand the development of human cancer,” he says.

Send comments to: jasonb@scripps.edu


"What we've discovered in yeast is an analogous pathway [to
restinoblastoma]," says Associate Professor Curt Wittenberg (right), pictured here with colleague Research Associate Rob de Bruin.
Photo by Kevin Fung.
















The Wittenberg lab's research was published in a recent issue of the journal Cell. Cover micrographs by D.J. Clarke and J.F. Gimenez-Abian. Concept and art by M. Giesbertz, M. Pique, and C. Wittenberg. Image courtesy of Cell Press.