December days: Cool science
Dec. 24th, 2014 10:35 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
livjack asked about Moar cool science (stuff that's obvious to you as well as stuff that's new to you). So for the old stuff I want to introduce the idea of apoptosis, because that's the thing so cool that it made up my mind I was actually going to study science to high level, when I was dithering about it in my late teens, and I do in fact work on it now. And because when I was babbling about transcription factors I clarified a few points in the comment discussion, and
silveradept mentioned I had no idea there was a self-destruct button for cells, so maybe this is something other people also don't know.
And for the new stuff, I know I've been babbling about it already to some people, especially if you were at our Christmas party at the weekend, but I came across a guy recently who is basically making a piece of kit which is very nearly a Star Trek medical tricorder.
I went to a very good talk when I was 16 or so at one of the youth science events Cambridge is so good at. The reason I went there was because I was really really interested in molecular genetics in the mid-90s, which was a good time to be interested because there was serious work going on on sequencing the human genome, and lots of popular information about genes available. But I thought the rest of biology was sort of boring, just a load of unconnected facts that didn't excite me intellectually, and I resented having to learn it in order to know what genes are for. I think the talk probably had "genetics" in the title but i wasn't about genetics in the sense of how inheritance works or the structure of chromosomes, it was about the genes that are most important in cancer.
The speaker was an Israeli called Aviva somebody, and I really wish I could remember better than that because I would love to thank her for inspiring me to embark on what has become my career. Anyway, she explained that cells have to constantly make decisions whether to grow, remain static or die, and gave a really accessible introduction to how this works. There's been a whole lot of research in the past 20 years, and I have been directly involved in some of it so it would be impossible for me to reproduce what I learned back then, but let me try and give an overview.
In multicellular animals, there are times when some cells dying is good for the body as a whole. This happens in development where animals at different stages of life might need different structures. It's not very obvious in humans, much more so in creatures which undergo serious metamorphosis, like insects with a larval and an adult stage with totally different body patterns. (When a caterpillar forms a pupa, in real life it's a lot less cute than in children's stories, the whole animal just dissolves into a chemical soup. Which has the ability to reform into a butterfly because it contains asymmetrically distributed transcription factors, and these create a kind of coordinate system allowing different regions to turn into a different body parts. But anyway, the first stage of this transformation is that every single cell of the caterpillar dies in a regulated way.) But it happens in humans too, for example human embryoes have tails and webbed fingers and toes, and they need to lose those before birth, again in a regulated way.
And cells need to die if there are too many of them. Tissues are much more dynamic than you might naively think, constantly producing new cells and balancing this by removing any excess. This makes it possible to repair injury and damage, and to make sure that there are always plenty of young, healthy cells available. There isn't a way for the body to estimate in advance exactly how many cells will be needed to repair a particular injury, so sometimes there are too many and they have to be removed.
The third reason why death of a few cells can be important for the health of the organism as a whole is to remove damaged cells. This is particularly relevant to my work because when this process goes wrong, the damaged cells may spread and eventually cause cancer. It's far better for one problematic cell (out of trillions) to die, than for it to reproduce and give rise to other cells which share the same damage.
In all these cases, cells use a system called apoptosis, from the Greek for leaf fall, and pronounced with the second p silent, or programmed cell death. (Some popular accounts call this cell suicide, and I have been warning my students that you can sometimes run into elaborate suicide metaphors if you search for information on this.) In apoptosis, there is a controlled sequence of events which leads to the contents of the cell being packaged up for disposal. If for some reason a cell dies without apoptosis, everything that should be inside the cell can spill out and cause further damage to the surrounding tissue. Partly because cells contain nasty chemicals including acids, which in a healthy cell are stored safely, but it's bad news if they just slosh around because the cell has burst open. And partly because, in animals that have an immune system, the internal contents of cells can be antigens and trigger an auto-immune response, as they are supposed to be internal so the immune system can't "see" them. This immune problem is important in diseases such as lupus and quite likely rheumatoid arthritis and maybe other auto-immune conditions too.
In mammals at least, most cells die by default. Each and every cell has the complete machinery in place to carry out apoptosis, and this machinery is only held in check by active survival signalling. The cells contain proteins which can make holes in the mitochondria, the part of the cell that turns chemical energy into a form that can be directly used by the cell. These are balanced by a usually equal number of inhibitor proteins, but if the balance changes even slightly so that there are uninhibited pro-death proteins present, that's curtains for the mitochondria. Parts of the mitochondria are then released into the cell, and they form a kind of scaffold which allows lots of proteases, enzymes that break down proteins, to be concentrated in the same place. The ones involved in apoptosis are called caspases, and they usually exist in a weak form that doesn't do much unless it's highly concentrated. Once they are all stuck to bits of broken mitochondria, they cleave eachother and remove the bit that makes them weak, becoming strong proteases that can readily break down anything. These initiator caspases then go and cleave and activate other caspases, and so on in a massive positive feedback loop. This means the cell quickly passes the point of no return; it makes no sense at all for a cell to be in some intermediate state between alive and dying!
The caspases at the end of the chain are sometimes called executioner caspases, or more prosaically effector caspases, and they chop up the whole cell, the membrane, the DNA, the cell skeleton and all. But they don't just chop it up randomly, they do it in such a way that everything ends up neatly encased in bits of membrane, keeping all the internal contents safe, and these little sacs of contents of the ex-cell can be safely eaten and recycled by surrounding cells. The process of turning a healthy cell into lots of little bubbles of stuff has a very characteristic visual appearance, called by the beautifully onomatopoeic term blebbing.
Some cells of the immune system can detect virus-infected cells, make contact with them, and activate death receptors at the cell surface. Death receptors can aggregate weak proto-caspases together just as mitochondrial components can, and trigger the whole caspase cascade in a similar way. Also, tumour suppressors, such as my beloved p53, the one pictured in my icon, can detect cancer-related changes and cause apoptosis, for example by transcribing more of the genes for pro-death factors such that they outnumber the pro-survival inhibitory factors. This is one of the main answers to the question of why not everybody has cancer all the time, it's a really important protective mechanism. But it has to be really tightly controlled, because otherwise you just get runaway cell death which for fairly obvious reasons tends to be bad for functioning organs. So yeah, that's apoptosis, and a lot of what I work on is figuring out exactly how cells know when it's the right time to die, and how cancer cells sometimes manage to get round these protections.
Anyway, new cool thing: I went to a talk a couple of weeks back by a malaria researcher called Sanjeev Krishna, and he talked about work he's doing on some technology which basically blew my mind. The idea is that they're combining Moores law electronic miniaturiazation, with the latest advances in PCR and genotyping, to make a device about the size and cost of a smartphone, robust enough to use in the field, which in 15 minutes can tell you things like the exact species of malaria parasite which is causing someone's disease, and predict which drugs it's likely to be resistant or sensitive to, and calculate an appropriate treatment protocol. It's not quite a fully general tricorder, because the chemistry bit needs some actual chemistry to happen, in this somewhat less magical than Star Trek reality. So it needs single-use cartridges to actually carry out the genotyping part. But they are developing different cartidges for different applications apart from malaria, and they are aiming for affordably cheap to be used as consumables in the poor parts of the world where malaria is such a big problem.
This is the website for this thing. It's not quite there yet, and the site is a bit glossy and full of grandiose claims. But I'm pretty excited about this as a realistic possibility within a few years.
[December Days masterpost]
(no subject)
Date: 2014-12-25 07:57 am (UTC)(no subject)
Date: 2014-12-25 07:58 pm (UTC)Thanks for the information about cell death. Can cells that have been infected with viruses or pathogens still have their components recycled?
And how is cell death handled with methods like chemotherapy and radiation? Is it still creating signals or is it just destroying on contact?