"We've known for a while that the protein coding genes of humans and chimpanzees are about 99 percent the same," said senior author Michael Snyder the Cullman Professor of Molecular Cellular and Developmental Biology at Yale."The challenge for biologists is accounting for what causes the substantial difference between the person and the chimp."

Conventional wisdom has been that if the difference is not the gene content, the difference must be in the way regulation of genes produces their protein products.

Comparing gene regulation across similar organisms has been difficult because the nucleotide sequence of DNA regulatory regions, or promoters, are more variable than the sequences of their corresponding protein-coding regions, making them harder to identify by standard computer comparisons.

"While many molecules that bind DNA regulatory regions have been identified as transcription factors mediating gene regulation, we have now shown that we can functionally map these interactions and identify the specific targeted promoters," said Snyder. "We were startled to find that even the closely related species of yeast had extensively differing patterns of regulation."

In this study, the authors found the DNA binding sites by aiming at their function, rather than their sequence. First, they isolated transcription factors that were specifically bound to DNA at their promoter sites. Then, they analyzed the sequences that were isolated to determine the similarities and differences in regulatory regions between the different species.

"By using a group of closely and more distantly related yeast whose sequences were well documented, we were able to see functional differences that had been invisible to researchers before," said Snyder. "We expect that this approach will get us closer to understanding the balance between gene content and gene regulation in the question of human-chimp diversity."

yale

In the last decade, scientists have studied whether mice could be genetically engineered and bred to grow human liver cells. Early results since 2004 showed it could be done, but the mice were difficult to breed, the time window for transplanting human liver cells into the mice was narrow, and the mouse liver, despite efforts to make the animal immunodeficient, often rejected the human cells.

Grompe's laboratory now has a system in which those disadvantages have been engineered out. It has created a severely immunodeficient mouse strain that develops liver disease only when the animals don't receive a protective drug called NTBC, allowing liver disease to be turned on and off.

"Our mice on this medicine are perfectly healthy, normal mice, and only when we take them off the NTBC do they get liver disease," Grompe said. "It's an easy system that any research lab should be able to set up, which is very different from what's around now."

In fact, the human liver cells from the repopulated mouse livers are indistinguishable from normal human liver cells, according to the study. "The healthy human liver cells take over and replace the sick mouse liver cells," Grompe said. "You end up with a healthy mouse that makes human blood clotting factors, all the proteins the liver makes, human bile, everything."

The mice also retain their unique traits for multiple generations, and each mouse can be implanted with human liver cells at least four times. Grompe estimates that each round of implantation can generate more than 20 million viable human liver cells.

"We think we will have a real edge in terms of quality and availability of cells," Grompe said. "We have a product. All we need to do is scale up and start selling it to anyone who wants to buy it."

In the coming months, Grompe's lab will develop a library of human liver cells from common variations of human drug metabolism. "Different humans metabolize drugs differently. So we want to create a library of cells from different humans to capture some of that variability," Grompe said.

ohsu