Through the retina, darkly

The circadian clock in the retina controls rod-cone coupling.
http://www.ncbi.nlm.nih.gov/pubmed/18786362

Rating: Slam Dunk

Article Summary
The Hypothesis
We perceive light and images using a sheet-like structure stretched across the rear of our eyes called the retina. Within the retina there are two types of cells that detect light; the conventional view held that the retina used 'cone' cells to sense light under normal conditions, and 'rod' cells in low-light or dark conditions. In this study, researchers turn that view on its head by demonstrating that our use of retinal rods in darkness is not dependent upon the actual amount of ambient light, but instead depends on our body's sleep-wake cycle (aka. circadian rhythm). Even more exciting was their discovery that when active at night, rods actually usurp control of cones to (hypothetically) increase the amount of low-light information being sent to the brain.

The Setup
The question at the heart of this study is surprisingly straightforward, despite the complex hypothesis. First some brief background: we have two systems for perceiving light in our eyes, both located in the retina. The cone system involves cone-shaped cells of three types, that each respond to a different type of light; roughly, red, green, and blue. These cells are really fast; if you're familiar with movies, you could say they have a high framerate. That is, they are capable of prcoessing many different images per second. The rod system involves - you guessed it - cells shaped like rods that are great at detecting really small amounts of light - but they are slow as molasses. If we used them all the time we'd never be able to drive, play baseball, or watch our favorite YouTube videos. Also, there's only one kind of rod meaning that the images they generate are monochrome. The question is this: how does our brain decide which of these two systems to use? Is it based on the amount of ambient light around us? Do we mix the two systems together in some way? Do we turn off the cones at night, or the rods during the day? A predominant theory prior to this work held that there were separate circuits from rods and from cones, like different video feeds or cameras in a newsroom, that connected to the brain. The brain then decided which system to use after some processing... sounds complicated doesn't it? Well, the answer is actually quite elegant and simple, and today's researchers are going to show you how it all works.

The Experiment
To come up with a hypothesis, the researchers drew upon two recently established features of the retina. First, it was found that rods and cones are electrically coupled to one another via structures called gap junctions. Gap junctions are protein bridges that allow charged ions (and other small particles) to move between cells, as if some one had strung a wire between them. However, previous work had shown that these connections were very weak - perhaps there were not that many junctions, or many of them were closed. Critically, those experiments were performed during the daytime. The second feature of note is that dopamine is released within the retina according to a sleep-wake, or circadian cycle. Dopamine is chemical used within the nervous system to transmit information; such molecules are termed neurotransmitters. It is traditionally associated with enjoymentof pleasing activites, and pathalogically its abundance can lead to addiction while its absence is a hallmark of Parkinson's disease. Dopamine is released at high levels during the waking part of the cycle, or 'subjective day', and drops low during the sleep cycle, 'subjective night'. Moreover, proteins that respond to dopamine (termed dopamine 'receptors') are found in rod cells. Combining these two pieces of information allowed the researchers to hypothesize that circadian control of dopamine levels might serve to alter the electrical coupling between rods and cones.

The researchers used two methods to study this question. In the first, they loaded individual cone cells in goldfish or mouse retinas with a visible tracer dye. This dye is small enough to pass through open gap junctions and should therefore be able to spread from the initial cell to any connecting cells, as through a web. So they waited for a while to see how far this dye might travel away from the initial cell. During the subjective day, they observed that the dye didn't spread very far - 5 to 6 cells in addition to the original cell they loaded. Subjective night, however was quite a different story. At night, the dye spread to THOUSANDS of other cells. Yes, you just read that right. That's the kind of scale of effect we scientists dream about. This result demonstrated that gap junction electrical coupling was indeed weak during the day, but very strong at night. To figure out if dopamine was involved, the researchers performed the experiment again, but this time in the presence of chemicals that block dopamine receptors on rod cells. This time they found that the dye spread to thousands of cells at any time of day, indicating that dopamine was preventing or inhibiting these electrical interactions.

Next the researchers turned to my favorite technique, patch clamp, to figure out the consequence of this increased gap junction coupling at night. Patch clamp is a powerful technique that allows us to record the electrical currents used by cells in the nervous system to communicate information, through the use of tiny electrodes. The researchers isolated intact goldfish retinas and were able to record the live electircal signals from individual rods and cones within the entire network of retinal cells. They compared the electrical signals from rods and cones in both subjective day and night, and the differences were incredible. During the subjective day, the signals were what you would expect; cones looked like cones, rods looked like rods and there was little obvious cross-talk between them. At night, however, things were quite a different story. Rods still looked like rods, but amazingly the cones also now looked like rods! They had the same monochrome light sensing, and the same slow response time. It seemed as though the rods had take over the usual cone function and circuitry to amplify their own signals, while also supressing that of the cones. The researchers also took the study one step further, to see what would happen if they darkened the retina during the subjective day, or shone a bright light on it during the subjective night. What they found was that bright light was able to reverse the rod take-over and make cones act like cones again. But when they tried the reverse - to darken the retina during the subjective day - there was no effect. The rods were only able to usurp the regular functioning of cones when the animal's body thought that it should be night.

What does the pithy scientist think?
Disclaimer: what follows is merely opinion, possibly speculation, and occasionally hearsay. But it's the best part, darn it!

This study is the scientific equivalent of a home run: clearly demonstrable, huge effects that are definitive and reverse a conventional viewpoint. It's amazing to think of a cell type that can take over its neighbors during the night-time when they are not needed, and relinquish them during the day. It's a bit like distributed computing (some good reading there). In all honesty, I don't even need the pros and cons points for this study because it's pretty darn airtight. The only complaint you could possibly make would be that the study wasn't conducted in humans, which would of course be impossible with current technology. But I'll spit out some pros just to make the authors feel even better about themselves:

Let's review the points that support he hypothesis that rods take over cones during the night:

1. There is a vast increase in electrical connections between rods and cones at night.
2. Mucking about with dopamine during the daytime can evoke the same increase.
3. Cones behave exactly like rods during the night-time.
   

Although there aren't any obvious cons, at least to my mind, there certainly is room for more questions and further unique hypotheses - for example (and these ARE hypothetical):

1. The brain never pays attention to the rods - and to get noticed, they have to take over the cones and use the cone circuitry to convey their messages to the brian.
2. Rods suppress cones during the night-time because the cones become too noisy when confronted with low-light situations and the brain can't interpret them. This is like trying to listen to music if someone is jackhammering outside your window. There's so much ambient noise, that it's difficult to hear the 'signal', or music.
3. Cones actually DO become rods during the night... in other words, they innately change their responses to light. Okay, this one's a stretch, but it's possible.
   

I would love to see someone look into those.

This study underlines a fundamental feature of biology that is often lost in the details - that the body is incredibly economical in its use of resources. The same central circuit being used for different input systems is an impressive way to leverage an existing energy and structural investment. It certainly is reminiscent of a multitasking computer system... or is the other way round? It's also beautiful mathematics. Is it by necessity, or pure chance? We may never know. But we can certainly marvel at its elegant simplicty.