Shaking the tree of life

Friday, August 15, 2008
by Joe Caspermeyer

Click, click, clack. Click, click, clack. Click, click, clack.The sound pours through the doorway. Click, click, clack.

Welcome to Sudhir Kumar’s laboratory. Step inside. Be prepared for a shock. The familiar sights and sounds of modern research— clanking test tubes, technicians in white lab coats, winking LED displays— are nowhere to be found.

Click, click, clack. The rhythmic tapping is the sound of scientists pounding away on computer keyboards. The keyboards are connected to a large bank of networked computers. It is the vital piece of technology that sustains Kumar and his team as they work to solve some of the greatest unanswered questions in biology.

  • How and when did life on Earth evolve?
  • How can scientists identify the genes involved in diseases such as cancer?
  • How does an organism develop from a tiny, fertilized egg into an adult body made up of trillions of cells?

Kumar is director of the Center for Evolutionary Functional Genomics (EFG) at Arizona State University. Kumar and his research team are using new methods and tools to uproot the conventional scientific wisdom of biology. In the process, they are giving the tree of life a good shaking.

Kumar has training in genetics, evolutionary biology, and electrical engineering. He uses a new branch of science called bioinformatics as a tool to find answers to big questions.

Kumar defines bioinformatics as “any type of information processing that relates to biology.” Using bioinformatics requires a deep understanding of biology, computer science, and statistics.

Bioinformatics is a totally new way of studying biology. Kumar does not do lab experiments using live organisms (in vivo). He does not grow cultures in test tubes (in vitro). He does his work using only the silicon power of computer microprocessors. This type of work has been dubbed science in silico.

Working in a lab is often necessary, but it has some drawbacks. Labs require specialized equipment. And living specimens require a high level of care and maintenance.

Kumar’s group enjoys much more flexibility. They can take on many different questions that involve different organisms at the same time. He mines data banks for information to help answer his questions.

“That data usually exists in huge amounts. It almost always requires us to develop new analytical methods and tools,” he says.

“Data mining” got its name because it’s a bit like mining for gold. But the chinking sound of pickaxes on rock has been replaced by the crunching of raw data on computers. The valuable nuggets are not gold or silver, but useful information. The information is buried in national databases such as GenBank. GenBank contains more than 23 million DNA sequence records deposited over the years by scientists from around the world. The gemstones waiting to be found and understood are part of the immense streams of DNA sequence data.

Molecular Clocks
Kumar does not shy away from the tough questions. For instance, he has asked exactly when modern mammals first burst onto the evolutionary scene. The fossil record suggests that mammals emerged about 65 million years ago. Scientists refer to this period as the K/T boundary. K/T refers to a mineral layer deposited in rocks between the Cretaceous (K) and Tertiary (T) geologic periods.

Because scientists know when this layer was formed, they know the ages of fossils found in the layer. It was during this time when mass extinctions spelled doom for 75 percent of all life on Earth. It was also the end of the Age of Dinosaurs.

But the fossil record for early mammals was incomplete. It showed that mammals existed during the K/T boundary, but not whether they existed before that. To address his question, Kumar needed a new kind of watch, a new way to tell evolutionary time.

The ASU scientist built himself a “molecular clock” to study the problem. The molecules that make up Kumar’s clock are DNA. DNA is the chemical blueprint for life found in every cell in every living thing.

All the DNA information contained in an organism is called a genome. By comparing the DNA sequences from the genomes of different organisms, Kumar devised a new way to tell time.

“Copies of the genome are being made continuously and passed on through each new generation. Over time, mutations, or errors, are always occurring within a genome,” he explains.

Think about it in terms of a photocopier. The machine does not always work perfectly. In much the same way, the cellular machinery needed to copy a genome can jam and break down. On a photocopy, mistakes show up as streaks and smudges. On the genome, mistakes show up as DNA mutations.

Some mutations are bad enough to kill the organism that has them. These mutations obviously do not get passed on. But some mutations don’t have an effect on the organism. These mutations do get passed on generation after generation. “These mutations are known to accumulate more or less linearly with time. This is where we get the concept of molecular clocks,” says Kumar.

Molecular clocks are not perfectly accurate. “However, these clocks do provide a direct relationship between time and evolutionary distance,” Kumar says.

Kumar’s molecular clocks suggest that mammals appeared on Earth between 90 million and 110 million years ago. That is almost 30 million years earlier than the fossil record indicates.

“Our results showed for the first time that early mammals may have lived along with dinosaurs long before the extinctions that occurred at the K/T boundary,” Kumar says. “These early mammals were probably tiny creatures, perhaps no bigger than a mouse.”

Using the new timeline, the researchers were also able to compare the early history of mammals with the geological history of the Earth. Our planet was a place of violent upheaval around 100 million years ago.

“The continents were breaking apart 100 million years ago, just about the same time that mammal groups were being established,” Kumar says.

His group proposed an idea to fit the time and events. They call it the “Continental Breakup Hypothesis.” It says that when individual animals or large groups of mammals are stranded on an island or land mass that is split from the main population, over a long period of time those creatures will evolve into new species. In recent years, other scientists have supported the predictions from this hypothesis.

“In addition to studying the fossil record, the molecular clock technique is now commonly used by scientists,” Kumar says.

Solving Disease Riddles
Evolution is just one area of study. Kumar’s group is also using their tools to unravel the mysteries of cancer, cystic fibrosis, and other diseases. To find answers, Kumar looks at gene sequences from a virtual zoo of animal DNA.

One of those animals is the puffer fish. Japanese sushi chefs use the puffer fish to prepare a dish known as fugu. However, if fugu is prepared incorrectly, the chef can kill his customers. One puffer fish contains enough deadly toxin to poison all the guests in the restaurant.

When Kumar started using the DNA of puffer fish and other organisms to help identify genetic mutations, he caught the attention of Jeffrey Trent. Trent is the director of the Translational Genomics Research Institute (TGen) in downtown Phoenix.

Kumar’s work was key to the successful completion of the Human Genome Project. The project was a massive international effort to sequence the three billion chemical letters of DNA that make up our human genome. The sequencing work was finished in April 2003, exactly 50 years after scientists James Watson and Francis Crick solved the elegant spiral structure of the DNA molecule.

Scientists at TGen use genome science as a tool to solve the medical riddles of cancer. They want to transform the idea of cancer as an acute life-threatening disease into one that is more a manageable, chronic disease.

“Cancer is really 212 different diseases,” Trent says. “We are giving Kumar specific intervals within the human genome where we think there are likely to be genes for certain key diseases. His group will then use their tools to help us identify those genes.

“Then we can test the genes at TGen,” he explains.

The ASU scientists take these genome intervals and compare them with similar intervals from other animals such as puffer fish, chickens, mice, and cows. By studying the same gene across different species over time, Kumar can figure out which sequences are the most conserved—in other words, which sequences do not change.

Kumar has found that the most conserved DNA letters within a protein over evolutionary time are the most vital towards that gene functioning properly. Why? Because mutations that cause disease are found most often in these positions.

Throughout the course of evolution, nature holds onto the most important DNA information and discards the rest. By identifying the changes in the letters, Kumar’s group can now assist the scientists at TGen in choosing new directions for their experimental work.

In the future, the ASU researchers plan to remain f lexible enough to continue making discoveries in a wide range of areas such as cancer research, evolution, developmental biology, and software development.

“In every sense, development and evolution are intertwined. I think of it as all one project,” he explains. “I’ve always liked evolution because so much remains unknown and it is so challenging to infer history. I want to immerse myself in many different areas and learn to see the interconnections.”