Kyunghee Koh, William J. Joiner, Mark N. Wu, Zhifeng Yue, Corinne J. Smith, and Amita Sehgal
Science 18 July 2008 321: 372-376

www.sciencemag.org/cgi/cont.../5887/372

There was an article in Science last month that I found very exciting. In it, authors Koh et al. report the identification of a gene in flies which regulates sleep ... without the protein that the gene codes for, the flies had an 80% reduction in their amount of sleep! To quote the abstract: "We propose that SLEEPLESS is a signaling molecule that connects sleep drive to lowered membrane excitability."

The first thing about this article that I thought was really neat is that it addresses what is often thought of as a "cognitive" or psychological process with a molecular process. There are huge loads of work out there looking at sleep from a cognitive perspective ... what are the effects on cognition of sleep deprivation, what processes, such as memory consolidation, take place during sleep, etc. All of these studies are basically coming from the perspective that the "purpose" of sleep is essentially neural ... the brain needs to be shut off from external stimulation so that it can perform essential "housekeeping" functions like memory consolidation. While it is likely true that this is indeed the case, this fly study also points to another, very different potential function of sleep that possibly has very little to do with cognition at all. Clearly it is too early to generalize the result to humans, as it is often the case that a behavior or process that in flies is mediated by a single gene is often mediated by large sets of genes and interacting gene cascades in higher animals.

But it certainly is suggestive. The article mentions that several other sleep-related genes have been identified, and that the sleep-wake cycle is most likely regulated by an interaction between circadian rhythms (which may be more neural in nature, and hence body-wide) and sleep-promoting factors like the gene they have identified (which may be more local in nature).

So, here is my wild conjecture, pretty much totally unsupported by the article, but that was really fun to think about after perusing it quickly. Perhaps there are cellular processes that go on throughout our waking day which lead to the accumulation of certain "waste" products in the cell or the degradation of membrane molecules, etc. The accumulation of these within the cell may eventually trigger the release of sleep-promoting proteins into the extracellular fluid. Over the course of the day, as many of the bodies' cells accumulate waste products or undergo wear and tear, these sleep-promoting proteins accumulate in the bloodstream. When the circadian cycle begins to tip towards sleep at the end of the day, receptors in the brain for these sleep-promoting proteins are also highly active, and together, these processes lead the organism's nervous system to progress towards a sleep state. While in the sleep state, a number of processes are happening ... in the brain, all of those "neural housekeeping" functions are occurring, aiding in memory consolidation, etc. But at the cellular level, perhaps there are also a great many "housekeeping" functions that require the body to be mostly _still_ for them to function.

It is not hard to imagine that some membrane repair processes or waste trafficking inside individual cells require a near complete lack of acceleration to happen correctly (because they utilize gravity to move a chemical gradient, or require a minimal tension on membranes, etc). In other words ... maybe our body's individual cells need us to _stop_moving_ for an extended period each day in order to repair and clean themselves. No one individual cell has the ability to tell the entire organism to be completely still ... only by the secretion of this local chemical signal, accumulated and acting upon the nervous system, can bring the whole organism to a state where acceleration is at a minimum.

Although I find neural explanations for the function of sleep to both be really interesting and compelling, I can't help but smile at the idea that a process that affects our mental life as individuals as strongly as sleep does may, in part, be dictated by stuff going on in individual cells (and not even neurons). My conjecture may prove to be totally wrong, but its fun to think about the possible existence of such a compelling "local to global" signaling mechanism.

Yay science geekery. I stop now.