Visualizing the therapeutic possibilities

If a fledgling mechanic wants to understand how cars run, she needs to become familiar with the individual working parts of an engine. And to really grasp how these parts work together, she'd have to study a running engine. Only then would she be able to comprehend what's happening under the hood.

It's much the same challenge for Dr. Adrian Ferré-D'Amaré and members of his Basic Sciences Division lab. They aim to understand riboswitches, genetic elements that may be key for creating new antibiotics to fight bacterial infections. These elements are made of RNA, a molecular relative of the genetic blueprint of life, DNA. But in order to understand the function and potential of this genetic switch, they must first see its structure and then see it in action.

Ferré-D'Amaré and postdoctoral fellows Drs. Daniel Klein and Thomas Edwards succeeded in visualizing the structures of two riboswitches, glmS and thi-box, in various states. Klein's glimpse of the glmS structure was a first. The structures of only three riboswitches have ever been visualized so far, by researchers elsewhere. Klein's and Edwards' findings were published in the September issues of the journals Science and Structure, respectively.

It's not simply an esoteric pursuit, said Ferré-D'Amaré, but in fact, may have clinical relevance. Riboswitches are genetic controls that help regulate the expression of genes. A portion of a bacterium's genes is controlled by riboswitches. Given the significant role these natural RNA switches play in bacterial genetic control, it's possible that drugs could be created to disrupt them.

"Currently, most therapies are based on protein function," he said. "Riboswitches are changing this because they are essential for the lives of bacteria. Critters like anthrax, listeria and staph are potential applications because they depend on riboswitches for survival."

The Hutchinson Center has filed a provisional patent application for using the glmS structure solved by Ferré-D'Amaré's group for antibiotic development. "Now that we know what it looks like, we may be able to create new antibiotics," Klein said. "Because the glmS riboswitch is not used by human cells, an antibiotic that targets this RNA might have fewer side effects."

While this is promising, Ferré-D'Amaré admits that new antibiotics are a future hope. "It's a tall order to get a drug that works in patients," he said, "but it makes sense for the Center to patent the concept."

Not too long ago, most people considered RNA to be just a disposable copy of the really important nucleic acid, DNA. RNA — ribonucleic acid — simply functioned as a messenger. But discoveries in the 1980s found that RNA could fold into complex shapes and catalyze biochemical reactions, a function previously thought to be restricted to protein enzymes. It is now known that all proteins in all living things on earth are made by an RNA enzyme or "ribozyme."

Solving the structure

The recent discovery of riboswitches demonstrated that in addition to essential catalysis in the cell, RNAs regulate how genes are expressed. Each cell must regulate the expression of hundreds of different genes in response to changing environmental or cellular conditions. The majority of these sophisticated genetic control factors are proteins, but RNA also can form precision genetic switches, and these elements can control fundamental biochemical processes.

Most riboswitches are thought to regulate gene expression through a structural rearrangement caused by binding to a small molecule metabolite. The glmS riboswitch Klein and Ferré-D'Amaré studied is unique in being both a riboswitch and a ribozyme. This RNA cleaves to itself in the presence of glucosamine-6-phosphate, an essential building block of the bacterial cell. Klein and Ferré-D'Amaré discovered that this building block functions as a coenzyme of the riboswitch, rather than inducing a change in the structure of the riboswitch.

The work on thi-box, the riboswitch Edwards is studying, also revealed remarkable features. Though Edwards was not the first to show its basic structure, he was the first to see how the riboswitch changes structure as it functions. "We wanted to see what happens when it's working," Edwards said.

His work showed how the thi-box can select the correct small molecule from a complex mixture of metabolites. "Inside cells, there are a lot of these molecules all around," Edwards said. "It's very important if you're trying to regulate a certain pathway that it picks the correct small molecule. These riboswitches need to be very specific."

The researchers used crystallographic techniques to solve the structures of the riboswitches. The goal is to learn where the thousands of atoms that make up a riboswitch are in space. Because a single RNA molecule cannot be held stationary while using X-rays to illuminate it, researchers coax the RNAs to form a crystal, in which billions of molecules arrange themselves into a regular array. The resulting crystals (which are still too small to see with the naked eye) can be manipulated and used for X-ray experiments. "Crystallization has an element of luck to it," Edwards said. "Growing crystals so you can solve the structure is a matter of trial and error. It's one of the biggest hurdles we had to get over."

Once crystals were formed, Ferré-D'Amaré's group used Lawrence Berkeley National Lab's particle accelerator to collect the data. The accelerator has very high intensity, tunable X-ray beams, so experiments can be performed that are impossible with the X-ray source available at the Center. Analysis of how the accelerator's X-rays interacted with the crystals, together with further experiments and computation, allowed Ferré-D'Amaré, Klein and Edwards to visualize the structures.

"Few other labs in the country are in business of solving RNA structures," Ferré-D'Amaré said. "It's a big gamble of time and resources, but the payoff is huge in understanding how a molecule works."