Cells' Collaborative Middle Management

Like corporate and governmental organizations, cells rely on middle managers to keep things running smoothly. These “middle managers” function as a critical bridge that controls the flow of information traffic. According to recent research, however, the middle managers often partner with one another, ensuring that the failure of one manager doesn’t bring down the entire organization. Moreover, this partnering becomes more extensive in more complex organisms.

“Understanding the system isn’t about the function of the individual parts. [It’s about] understanding the importance of these information flow bottlenecks and how natural systems get around them,” says Mark Gerstein, PhD, professor of bioinformatics at Yale University. He and his colleagues have been studying networks of genes and transcription factors to describe the information flow within cells.

The work serves as part of a larger effort within Gerstein’s group to develop real-world analogies to explain how biological systems use and process information. Previously, the group had shown that hierarchies in biological regulatory systems resemble directed social structures such as governments and corporations. That study, published in PNAS in 2006, found that “middle managers rule,” Gerstein says. Transcription factors in the middle layers of the networks have the most regulatory interactions with other genes. “The genes in the middle are much more essential. If you knock them out, the organism is much more likely to die.”

In a paper published online in PNAS in March 2010, Gerstein and his colleagues took that work a step farther, seeking to understand how cells avoid failure at the sites of middle manager bottlenecks in five species ranging from E. coli to humans. First, they identified which genes are regulated by other genes in each organism. They then stacked the levels of regulators in hierarchies and placed them between two extreme types of social hierarchies, autocratic and democratic, and showed cellular regulatory hierarchies have "intermediate" structures. They found that, in all five organisms, coregulation happens most at the middle level and least at the bottom. And more complex organisms exhibit more collaborative, “democratic” regulatory structures with more interconnections. For example, yeast has about one regulator for every 25 targets whereas in humans the ratio is much smaller, about one to 10.

“The parallels between government structure and regulatory network structure are provocative,” says Trey Ideker, PhD, associate professor of medicine and bioengineering at the University of California, San Diego, who was not involved with the study. One question, says M. Madan Babu, PhD, an investigator in the MRC-Laboratory of Molecular Biology at the University of Cambridge, is the function of these hierarchies within a cell. “Are they really important? Or are they something that is emergent because of the complexity of the system and has no consequence whatsoever?”

Regulatory networks are definitely important for organism function, Gerstein notes. So the question of whether the networks emerged in response to complex roles or the system’s complexity allows organisms to carry on these complex interaction is a “chicken and egg type of issue.”