"Ground control to Major Tom, your circuit's dead, there's something wrong…can you hear me Major Tom?"
Even if you can't hear David Bowie, you are reading the words on this page thanks to signaling sent from cells in your eyes to cells in your brain. Multicellular organisms need to sense signals from their environment and then interpret them accurately so that they can respond appropriately. Information must also transit into and ignite responses from single cells. Instead of radio waves or telephone wires, cells receive and transfer information through the collision of molecules. Ligand molecules outside of the cell bind to receptors on the cell surface and the signal is relayed to responsive elements deeper inside the cell through a series of molecular collisions. The response the cell makes to a given signal is frequently proportional to the number of receptors that bind the ligand molecule on the cell surface. This produces an interesting problem for biologists: how can a message, with a specific identity and amplitude, be accurately preserved as it is transferred from the cell surface to the analytical center of the cell, the nucleus?
Most biological signaling systems transfer information through a series of chemical reactions as different molecules collide, potentially generating errors and noise in the process. In a new study published in Cell Systems, computational biologist Steve Andrews and his colleagues in a team led by Fred Hutch biologist Roger Brent (Basic Sciences Division) report the results of their investigation searching for general arrangements of chemical reactions cells could use that would best produce an output response that mirrored the fraction of cell-surface receptors bound by ligand.
Rather than seeking to explain the function of a single biological signaling pathway, the scientists took a broad approach. They aimed instead to evaluate which, of all possible reaction mechanisms, could best generate a precise response that reflected the amount of bound-receptor on the cell surface, a phenomenon known as dose-response alignment. They used computational methods to evaluate how accurately abstracted pathways, represented as a collection of letters linked together by arrows representing either a stimulatory or inhibitory effect on the downstream output, could produce dose-response alignment over a wide range of input signal.
Biologists have identified countless examples of feedback. For example, in negative feedback, the active form of a molecule will inhibit the production of too much output by preventing the action of earlier active molecules in the pathway. To their surprise the Brent group found that feedback alone could not generate dose-response alignment but another arrangement of reactions, called Push-Pull, could. Usually, biologists characterize each molecule in a specific signaling pathway as being active or inactive. Only the active forms of molecules have influence on the downstream step. In Push-Pull, the supposedly "inactive" forms of the molecules can attenuate the signal by inhibiting, or "pulling down" the subsequent step.
The scientists have found a few specific examples of inactive molecules playing an active function in defining signaling outputs. Their collaborators have reported that a receptor, which when active promotes dissociation of a protein from the receptor, also has a function in its "inactive" form, promoting the association of a protein with the receptor. Their findings prompt reconsideration of the way that the activity of signaling molecules is defined in biological signaling pathways. Rather than playing a single type of role in a pathway, many molecules may either stimulate or inhibit the output.
Another type of signaling mechanism that the scientists found could produce dose-response alignment makes use something called closed-loop control, which requires a comparator-adjuster. "Imagine you are driving a car and you want to keep moving at a constant speed," said Brent. "When you encounter a hill you need to push harder on the pedal to keep the same speed while moving up the incline." "You as, the driver and comparator-adjuster, are comparing the input, your desired speed, with the output, the speed of the vehicle. And you are adjusting how far you press the accelerator until the output signal matches your desired speed. If the car goes up a hill, and the car begins to slow, you press down more on the accelerator so as to maintain a constant speed."
While there are many examples of man-made devices that use such a system, no known biological signaling system works that way. Indeed, the lack of this approach to control in cells may explain the efficacy of anti-cancer drugs that target intermediate steps in signaling pathways such as RAF-MEK-ERK. If this pathway contained a comparator-adjustor, inhibition of RAF or ERK would not lower the output, and the comparator adjuster would press down harder on the system to maintain constant output. But blocking intermediate steps in this pathway usually diminishes output, suggesting that this is not a strategy that cell's use to create dose-response alignment.
Overall, their study has uncovered a new lens through which to view signaling dynamics. Brent explains, "At Berkeley we called it 'finding Neptune.'" "When William Herschel accounted for the gravity of other planets…his work told him that he could expect to point his telescope at a certain spot in the sky and expect to see the new planet beyond Uranus." Similarly, knowing that supposedly inactive forms of signaling proteins can actually suppress downstream activity has already prompted scientists to look for suppression by "inactive" forms, and to find it.
Andrews SS, Peria WJ, Yu RC, Colman-Lerner A, Brent R. 2016. "Push-pull and feedback mechanisms can align signaling system outputs with inputs." Cell Systems. 3, 444-455.
This research was supported by the National Institutes of Health, MITRE, and the Simons Foundation.