June 28, 2013 | 7
Can plants do math? That is the assertion of a new paper published in the journal eLife this week titled “Arabidopsis plants perform arithmetic division to prevent starvation at night.” The plants in question aren’t spitting out numerical answers to word problems on their leaves, but doing normal plant stuff: using energy stored as starch at different rates depending on environmental conditions. Plants get their energy from sunlight, so at night the rate of starch consumption has to be smooth in order to maintain energy until dawn and prevent a “sugar crash.” The researchers found in a previous study that that plants will consume their starch almost completely every night and that the rate of consumption will stay mostly constant after “sunset,” regardless of whether the lights go out earlier or later than the plant “expects” based on their circadian rhythm. Based on these results, the researchers proposed a mathematical model whereby the plants are “dividing” the level of starch stores by the number of hours until dawn in order to determine the proper rate of consumption.
It’s not that much of a leap of logic to assume that the availability of sunlight and the circadian rhythm control how photosynthetic plants behave, especially when it comes to the metabolism of starch. There is extensive and ongoing research on the mechanisms of circadian oscillations and the regulatory and metabolic consequences of these cycles, how genes are turned on and off and how metabolism is regulated. Saying that this kind of regulation is an arithmetic “calculation” is, as a friend of mine said, “like saying pizzas calculate pi because their area is always pi*r^2.”
The authors agree that it is “conceptually unclear how such a computation might be performed,” but it is conceptually and rhetorically very interesting at least how the paper develops this extended metaphor of cellular calculations. Living cells sense and respond to many signals external to the cell (like sunlight) or internal (like starch levels). The ways that these signals are sensed, processed, and integrated into coordinated responses in different conditions and in different cells has been one of the more rich areas of biological research for decades. This kind of cellular “information processing” is also of huge interest in synthetic biology, where biological “computations” are being built in simplified genetic networks to create cellular “logic” that can be used to create biosensors or precisely control engineered metabolic pathways.
I have metaphors, similes, and analogies on the brain these days (more than usual) because I’m reading China Miéville’s Embassytown, a science fiction novel set in the distant future on a planet of insect-like aliens with a unique Language. It’s impossible for these creatures to speak untruths, and in order to use similes they must be literally constructed–to say “this is like the rock that was broken in half and cemented together” there must exist a rock that was broken in half and cemented together, an object and an event that became part of the Language. I can’t help but analogize here again to synthetic biology, where the computational metaphors about biological systems and signal processing are literally being built in to genetic designs. We build metaphors to talk through them, to understand biology through the analogy to engineering.
Are plants “actually doing maths”? Are bacteria engineered with transcriptional cascades that respond to the ratio of two signals ”better at math than some people,” as some headlines claimed? These are the kinds of metaphors that can lead to some very philosophical deep thoughts about the nature of numbers and the appropriateness of machine metaphors in biology, but perhaps, like Embassytown’s similes, the construction of these models and “circuits” is a prelude to new biological Language. What will we say with these mathematical similes?