Scott Turner’s recent book, Purpose and Desire: What Makes Something “Alive” and Why Modern Darwinism Has Failed to Explain It, is an elegant reflection on the history of science and the nature of life — its perpetual mystery and marvelous stability in the face of change. Living things accomplish that remarkable stability by monitoring and responding to their environment with suitable physiological changes. The process by which this happens is called homeostasis.

According to Jerry Coyne at Why Evolution Is True, it’s well understood how homeostasis evolved — it’s been selected, it has evolved, and it has a genetic basis.

“Homeostasis” — the ability of an organism to maintain aspects of its function or morphology in the face of environmental changes — is not something mystical, but a result of selection itself: organisms face varying and often unpredictable environments, and have evolved ways to deal with these so they don’t lose reproductive output (growing fur when it’s cold, spines if you’re a rotifer in a pond with predatory fish, and so on). That this can happen is evidenced by our ability to select for greater or lesser degrees of homeostasis, showing that it has a genetic basis and thus could be subject to selection….

Again, homeostasis can easily evolve by natural selection, and needn’t reflect “purpose and desire”, which is either a teleological force within organisms or some external intelligence guiding the process. It’s not surprising that this book was recommended not by biologists, but by members of the Discovery Institute. [Emphasis added.]

I’d very much like to hear a detailed account from Coyne of the evolution of homeostasis, rather than a simple declaration that it can be selected.

The evolution of homeostasis requires the coevolution of multiple parts. Control systems in vertebrates can be quite baroque, so it would be unfair to ask questions about homeostasis there, so let’s go to one of the simplest kinds of homeostasis — just one part of the regulation of sugar metabolism in E coli.

Even the simplest of regulatory pathways requires a sensor that monitors the environment, an effector that produces the response needed, and negative feedback between the two. One of the best studied examples is the lac operon, which encodes the enzymes that break down lactose. Cells prefer glucose, but they can use lactose if there is no glucose. Thus the time to turn on the lac operon is when glucose is low and lactose is high. Normally, the operon is shut off by a lac repressor protein that binds a regulatory site in the DNA. When lactose is present, allolactose, an isomer of lactose, causes the repressor to change conformation and release from the DNA. When lactose is low, there is little allolactose, so the repressor stays bound to the DNA.

What if glucose is present? If it is high, and lactose is low, such that the lac repressor is bound, no transcription occurs from the lac operon. When glucose is low and lactose high, though, the cell makes cAMP, a starvation indicator, which binds to a protein called CAP (catabolite activator protein), which binds to a DNA binding site upstream of where the RNA polymerase usually binds in front of the lac operon. CAP causes the RNA polymerase to increase transcription from the lac operon; since lactose is high, the lac repressor is not bound and transcription proceeds. The cell is now able to break down lactose and turn it into energy.

Remember the basic components of a regulatory system? The sensors in this system are the lac repressor and cAMP. The effectors are lac repressor and CAP. Where is the negative feedback? lac repressor is sensitive to allolactose, so the presence of lactose causes the repressor to come off the lac operon’s repressor binding site. However, even with the repressor no longer bound, transcription of the lac operon is normally low. Under conditions with high lactose and high glucose, the lac operon will not be transcribed. The cell would rather use glucose than lactose. Thus the cell prevents the enzymes that break down lactose from being made. If glucose is low and lactose is high, however, the low-glucose “survival” mode kicks in. The cell makes cAMP, and that triggers the production of CAP, a DNA binding protein. CAP binds to a region of DNA near the RNA polymerase, which activates lac operon expression. Low glucose and high lactose activates; high glucose and low lactose inhibits.

How many moving parts does that represent? lac repressor, allolactose, lac repressor DNA binding site, interactions among the three; cAMP, CAP, RNA polymerase, the polymerase and CAP binding sites, and whatever turns on cAMP, and all their interactions. Each interaction means a stereospecific conformational match between the interacting proteins. They have to be adapted to fit one another.

Why might Coyne think it’s easy to evolve such a system? Well, he probably thinks things happened one at a time. At first everything’s on all the time. Then a protein develops the ability to bind a particular DNA upstream of the lac operon (never mind where the operon came from) or the DNA evolves a sequence that binds the protein. Now the lac operon is turned off all the time. Bad. But some version of the protein somehow develops the ability to bind lactose or its isomer, and somehow that changes the lac repressor’s conformation so that it lets go. Good. Now we have a functional system for lac homeostasis. But notice, there was no benefit until all pieces were in place. And selection doesn’t operate until the last step.

My only remaining comment is that this isn’t the first time we’ve seen reviews from people who haven’t read the book in question. I hope Jerry Coyne does. Maybe it will put a dent in his selectionist mentality.

Photo: E. coli, by Erbe, digital colorization by Christopher Pooley, both of USDA, ARS, EMU. [Public domain], via Wikimedia Commons.