Mad Science Monday, 7/13/2009

Dammit. Dammit dammit dammit.

You might think from that intro that I'm writing about one of the more widely talked about science stories from this week. I may get to that one eventually, but the actual source of my frustration is the realization this morning that I had gotten the wrong paper for this week's entry. Sure, the one I got is mad, but I didn't get the one by the same group that involves rats with frikkin' laser beams attached to their heads. I'm going to try to give as much of an overview of all of that group's research as I can, but I can't for the life of me figure out how they made the stuff work in the laser rat story.

Update: I got the other paper in the middle of writing this, but I'm still focusing mostly on the newer paper; the one with the actual laser beams appears to use many of the same techniques, and I haven't had as much time to digest it, so I'll stick to the one I'm mostly grokking.

But I'm getting ahead of myself. First you're going to need some...

Mad Observations: A lot of observations are necessary to culminate in this level of mad science. There's too much to cover it completely, but let's see what I can get.

First, there's an archaebacterium, Natronomonas pharaonis, that makes a protein that pumps ions across the cell's membrane in response to light (specifically certain wavelengths of orange light). This pump is called the Natronomonas pharaonis halorhodopsin chloride pump, or NpHR. This pump has been adapted to be expressed in mammalian cells.

Next, there's the known mechanism of epileptic seizures, namely that the'rey caused by cascades of electrical potentials; basically, one cell has more positive charge inside than outside, and that causes it to induce the next cell to switch to the same condition, etc down a line.

Mad Reference: "Optogenetic control of epileptiform activity." Jan Tønnesen, Andreas T. Sørensen, Karl Deisseroth, Cecilia Lundberg, and Merab Kokaia. PNAS, published online before print July 6, 2009.

Mad Hypothesis: This paper even included a direct reference to the hypothesis they were testing: "Therefore, we tested a hypothesis that epileptiform activity can be optically controlled by selective expression of NpHR in principal cells." In other words, they thought screwing with the potential differences across membranes (by pumping chloride ions into those cells) would stop the epileptic cascade, and they figured they could induce that change with light by putting those pumps into the right cells. They also tested whether putting in the pumps (but not inducing them with light) caused any changes in the behavior of brain cells, and whether turning on the pumps caused any other problems (like sucking up too many chloride ions, stopping other things that need them from functioning properly).

Just to make sure that's clear, what they were testing is whether shining a laser beam inside a brain would work to stop seizures. Obviously.

Mad Experiment: It just keeps getting better. To test whether shining a laser on brains might be useful for treating epilepsy, they infected rats with a virus. Ok, this sounds all kinds of mad scientist, but it's actually a fairly well-established technique. This virus, technically a lentivirus, was modified to incorporate the gene for NpHR into cells that it infected. That gene was put under control of the calcium/calmodulin-dependent protein kinase IIa (CaMKIIa) promoter, meaning that, no matter what cells the gene might get inserted into, the protein would only be expressed in certain cells--namely, brain cells. They also put the enhanced yellow fluorescent protein (EYFP) in the same virus, also under control of that promoter. That let them cut up some of the rat brains (and other rat bits) and confirm that the technique had worked to get the protein expressed in the right place, and not in the wrong places (because it'd be bad for light-sensitive proteins to be expressed on, say, the skin, where they'd be doing their thing all the time, not just in response to a frikkin' laser beam).

Most of the research, unfortunately, was done in cultured rat brain cells, using that same basic technique as I described above. But they showed that they could get the protein expressed in rat brains. That's important, since they'd already done other research with implanting frikkin' laser beams in rat brains, to stop Parkinson's tremors.

Yes, that's a photo of a rat having light beamed into its brain through a fiber optic cable attached to a laser. We truly live in amazing times.

So anyway, they got their protein into cells (both in rat brains and in cultured cells), and then they tested the cultured cells using established epilepsy tests. They did these tests on unaltered cells, altered cells without light, and altered cells with the correct wavelength of light shining on them.

They All Laughed, But: It worked. When they shined lasers on the altered cells, their idea worked; the seizures (well, technically simulated seizures, since it was just a plate full of cells) stopped. It looks like this would actually work. But the whole time I was reading it, I was thinking, "Um, so. You need a frikkin' laser beam implanted in your brain." But then I found out I was missing half the story, since they'd already implanted frikkin' laser beams in rat brains. So this whole thing would totally work, and all it takes is:

• infection by a virus to put an archaebacterial protein into your brain,
• glowing proteins engineered from jellyfish thrown in to make sure it worked,
• surgery to implant a laser (or lasers) in your head, and
• potentially a fiber optic system following you around (although I guess we're bigger than rats, so maybe the lasers could be worn directly).

I say this all jokingly, but apparently that's way better than the current system of curing intense seizures, namely cutting the hell out of the affected area, hoping you don't get too much that's useful.

Mad engineers: This one's all yours now. We scientists have shown it would work. Implementing this is all you.