A Look at the Inner Life of a Doomed Cell
by MikeGeneOne of the major complaints about Harvard's "Inner Life of the Cell" video was that it was too serene and uncrowded, creating an "illusion" of design that did not exist. What I have for you below is another award-winning animation which factors for those complaints. It shows the crowded, chaotic, Brownian world of the cell. The process that is illustrated is apoptosis, otherwise known as programmed cell death.
Here's the walk through: You'll see a T lymphocyte approach a diseased cell and the animation will switch to the surface of the diseased cell. Welcome to chaos of the molecular world! You'll then see the lymphocyte present its death ligand to the death receptors of the cell, resulting in trimerization and thus activation. In essence, the self-destruct button has been pushed. Then, you'll switch to the cytoplasmic face of the death receptors which can now attach adaptor proteins, which in turn, fish out procaspase 8. The procaspase 8s become activated and break off to activate other caspase 3s in the cytoplasm "“ the signal is being amplified and spread. The caspase 3s will go about cleaving other cytoplasmic proteins. At this point, you'll switch to a lower magnification to watch the signal spread, causing mitochondria to release their cytochrome c. This protein, which is normally part of the life-giving electron transport chain needed to fuel of the ATP synthase will now moonlight as the a death signal, where it will bind to a protein called Apf1 triggering the formation of the the death wheel, which will then sequentially bind caspase 9s to form the apoptosome, which will activate an army of caspase 3s, reinforcing the signal from the death receptor. The activated caspases will begin the process of orderly taking the cell apart and the cleavage of part of the actin cytoskeleton is shown. Then you switch to a picture of the cell to show the result of the buzzing caspases "“ the cell breaks apart into many small vesicles known as apoptotic bodies. What is not shown is that phagocytotic cells will then simply eat these, providing for the clean removal of diseased or worn out cells.
Are you ready for the show?
If you ask me, the chaos, crowding, and Brownian movement don't erase the wonder of this process, which looks like some sci-fi alien nanotechnology. On the contrary, that this purposeful, coordinated, machine-like activity occurs is in the midst of a molecular storm makes the cell even more amazing. Oh, and one more thing "“ each and every day, this process is happening in one out of every ten cells in your body. As you read this, millions of cells in your body are doing the very thing you just watched.
March 28th, 2008 at 6:57 pm
Cool video, thanks for putting it up.
In a cancerous cell is it business as usually up until the mitochondria is going to release its cytochrome c?
Comment by Doug — March 28, 2008 @ 6:57 pm
March 28th, 2008 at 7:53 pm
It depends. In some cancers, a protein known as bcl-2 is overexpressed. Bcl-2 is an anti-apoptic protein and its overexpression prevents the mitochondrial step.
Comment by MikeGene — March 28, 2008 @ 7:53 pm
March 28th, 2008 at 9:37 pm
Suppose a researcher were trying to defeat a particular disease by, say, regulating apoptosis. Would it be more efficient for that scientist to approach the cell as a designed machine, or as the product of RV + NS, or does it matter?
Comment by chunkdz — March 28, 2008 @ 9:37 pm
March 28th, 2008 at 9:50 pm
Both. You approach it as a mega-machine, looking for nodes that can be manipulated in order to redirect the process. Disease is a disruption in homeostasis, so you try to set it right. But cancer cells evolve and exist as quasi-foreign organisms, meaning that RV & NS can possibly circumvent your best strategy.
Comment by MikeGene — March 28, 2008 @ 9:50 pm
March 28th, 2008 at 10:10 pm
Well said, Mike. Indeed, the question of "evolutionary origins" doesn't generally enter into the question of cancer cells nor into the question of how such cells become cancerous. Rather, one attempts to determine what kind of cancer one is dealing with (i.e. what is the underlying genetic and developmental etiology for the cancer). Knowing this is highly relevant to the treatment of the cancer, as different forms of cancer (i.e. different genetic and developmental lines of cancer cells) respond differently to different treatments.
However, as Mike points out, evolution and natural selection is highly relevant to the outcome of cancer treatment, as natural selection very rapidly selects for cancer cell lines that are resistant to chemotherapy and radiation. This is why when a cancer returns following a period of remission, the treatment(s) that caused the remission generally no longer work.
Comment by Allen_MacNeill — March 28, 2008 @ 10:10 pm
March 28th, 2008 at 10:15 pm
By the way, most of the sources of phenotypic (and genotypic) variation that provide the raw material for natural selection are clearly and unambiguously not random. For a good explanation as to why this is the case, I recommend Jablonka and Lamb (2005) Evolution in Four Dimensions, MIT Press, Cambridge, MA, ISBN 0262101076, chapter 7, pages 245 to 283: "Interacting Dimensions: Genes and Epigenetic Systems".
Comment by Allen_MacNeill — March 28, 2008 @ 10:15 pm
March 29th, 2008 at 12:02 am
Mike,
Is there any way to use this information to predict what counter-strategy to use? IOW, is it more efficient to tweak the machines to be able to robustly defend against most conceivable threats, or is it more efficient to wait for the mutated vector to do it's worst, then adapt the treatment? I am wondering if the paradigm shouldn't be to take a faulty machine and program it better, rather than play catch up waiting for RV + NS to do it's worst. I hope this makes sense - I'm kinda tired..
Comment by chunkdz — March 29, 2008 @ 12:02 am
March 29th, 2008 at 12:02 am
Allen,
Does this mean we should now refer to it as CAUNRV + NS?
Comment by chunkdz — March 29, 2008 @ 12:02 am
March 29th, 2008 at 1:15 am
Or maybe this is a better way to put it:
Let's say I want to kill a cancer cell. But the darn thing keeps mutating and becoming resistant to treatment.
The evolutionary biology approach would begin with survival of the fittest - find something that kills the cancer, but not the host. But since evolutionary biology lacks any power to predict what mutation will occur, and how it may manifest itself, the evolutionary biology approach simply ends up chasing it's tail, finding new more potent treatments that can kill the cancer, but not too potent to kill the host. RV + NS generally wins this fight because it controls the fight.
Now the engineering approach would be different. An engineer looks at the cancer cell and sees faulty programming and heavy data corruption, leading to dangerous mechanical control problems. The engineers instinct is not to kill the broken machine (as in the evolutionary biology approach), but to FIX the machine.
If I'm the engineer, my primary task is going to be ramping up the cancer cell's error correction scheme big time. Let's reduce the data corruption to a minimum, then we won't be chasing our tails every time the damn thing mutates. Then we reprogram the cell. Maybe dial down the replication rate, stabilize the metabolism, etc.
Then when the sucker is fixed and stable, we can program it to kill itself in a way that does minimal damage to the surrounding tissue, and maybe program it to tell it's buddies to kill themselves too.
I realize that this is already going on in cancer research to varying degrees. But it appears to me that evolutionary biology seems relatively weak as a means to predict outcomes, whereas the engineering approach seems to be able to find ways to make the machines MORE predictable.
Engineering is a proactive approach, while evolutionary biology seems a reactive approach.
Comment by chunkdz — March 29, 2008 @ 1:15 am
March 29th, 2008 at 2:11 am
Nice. One of the things that keeps me going is you guys.
Comment by MikeGene — March 29, 2008 @ 2:11 am