 We're back, we're live, two o'clock, rock. Think Tecawai, likeable science every Friday, two o'clock. And we have our likeable scientist here, Ethan Allen. Hi, Ethan. Hi. Nice to be here again. Yeah. Today's show is entitled Drilling Down on a Crisper. And that's not a breakfast cereal at all. So we should define it, we should understand it better, we should understand the incredible discovery involved in Crisper. And then we should talk about where is it going and what are the problems because we know that really powerful technologies, especially in biochemistry, have their problems. Right, right. Absolutely. So what does it stand for, Crisper? C-R-I-S-P-R. So it's a clustered, regularly interspaced, short palindromic repeat. And what that really means is, if you think about it, you've got to sort of go back a little step back into the biochemistry of our genes or our chromosomes. Your chromosomes contain all the instructions to build you, right? Every cell in your body has complete instructions to build you. Every living organism has this, yeah. Complete instructions, each cell in every living room, that is one single molecule that is several feet long, if it were stretched out, lives inside each and each and every cell in your body. And it's just basically a series of little individual units that are read off in groups of three, typically, to make sort of letters of an alphabet and spell out. And the part of it that we're most familiar with are the genes, which are segments that tell the body how to build proteins, because proteins really run a lot of our show. Basically, they have structure, they run processes. And way back when, eons ago, some clever little single-celled organisms discovered that if they had this crisper stuff, these clusters of regularly interspaced, short palindromic repeats, they could use these to cut DNA, to slice the DNA strand. And that's a useful thing to do if you've been invaded by a foreign being, if you can go and cut their DNA in half. It's part of the immune system then. Yeah, exactly. Sort of slows them down and all. And the beauty of this is it's a new technology that we've discovered that you can cut DNA very precisely. If you want to cut it at one specific point, you can set up a crisper system that will find that point, any DNA strand, and cut right there at that one point. At the gene. Yeah, so you can cut out a particular gene, if you can, using other versions of crisper and insert a new gene in its place. Even more powerfully, we think that genes is really being really important, but we really only have 21,000, 23,000 genes, and it's not really many. The power of the genes is that they are regulated by these flanking sequences. So lots and lots of this long strand of chromosome isn't genes at all, only a little bit of it really is genes, but a lot of the rested so-called junk DNA for a long time runs those genes and sort of says, turn on now, turn off now. Express yourself more now, express yourself less. And if you think about it, you've got to run that, right? Think of a little plant growing, right? A rose bush, say. It starts out, what does it got to do? It's got to put up some stems, it's got to put up some roots, it's got to put up some leaves. But every cell in the rose bush knows how to make a rose bud and open up into a rose flower, but you don't want to do that first, right? When your little sprout is high, you know, putting out a rose flower will be useless, will be worse than useless, it will be deadly. You never survive. You've got to wait until you've grown up and you're several years old and you've got leaves and roots and stems and then those genes can get turned on. Everything has to tell those genes, okay, now it's time to start making flowers, you know? And that's sort of really the power of CRISPR is you can tell genes when to turn on and off. You can not only just cut them out entirely, but you can actually control them in a very fine tuned way. Oh, this sounds like computer programming. Very much is. It's sort of the coding basically, the underlying coding of the whole of our whole genomics. The underlying coding, the underlying intelligence, the controlling mechanism, and then you have the functions that actually do things, right? And so it's a layered arrangement, right? In a sense, it's as if you could be driving your car down the road and not just have an accelerator, but you could, for instance, change the size of the pistons and cylinders and adjust the gap on your spark plugs and just the fuel mix all while you're driving down the road. In a sense, it's that level of control, you know, you could change your car from being a pickup truck to an XK Jag, you know, while you're driving basically. Yeah. I mean, it's sort of that kind of thing. Well, to me, one point you mentioned that is very important is we used to think that the strands, the chromosome strands on which the genes were laid were junk, what do you call them? Junk DNA. Much was thought to be junk DNA. Junk DNA. But now we may not do that. Right. Now are they smarter than we thought? There's repressors, enhancers, promoters, we've recognized all these other little sections that flank the genes are doing actually important work with the genes, although they themselves aren't coding for protein. They're not junk at all. No, no. They are. They are the fine tune mechanisms, you know. So can we talk about the, you know, the biochemistry of it? Sure. What is in a chromosome? So as I recall, there's 48 chromosomes in a human being, whatever, something like that. And the genes are on the chromosomes. They're parts of the chromosome. Part of them. Right. Okay. And you can look at the strand and you can see every so often you see a gene. And genes come in pairs, don't they? Well, the chromosomes typically are two strands wrapped around each other, yes. Okay. Okay. So what is it biochemically that we're looking at when we look at this chromosome that's, what, this long? Right. Right. It's one big DNA molecule, deoxyribonucleic acid. So it is a molecule with a bunch of little sort of ribose sugars, units held together in a spiral pattern with these, these little... The helix. Yes. Double helix, so-called. Right. And it's basically, it has four kinds of units on it, adenine, cysteine, guanine, and cysteine. Say that again, the four units. Acidine, guanine, thymine, and cysteine, I think. Okay. I'm not right. Those are elements on the DNA of the molecule. Yes, right. Those little so-called deoxyribo-sugar units, basically. Okay. And that four-letter code basically gets translated into, eventually into amino acids and then eventually from amino acids they get built into proteins and that's how we run the whole thing. So it's chemical. Oh, yeah. The whole thing is modern sophisticated chemistry. But it's all being read linearly. That whole DNA strand gets, there's a little, literally a little molecular machine that runs along that, reading that, and dumping out appropriate, all within one molecule. Runs that whole, many, many feet of molecule. So the parts of the molecule, which is really tiny, talk to each other. Right. Or some parts control other parts. It turns out, yes. It turns out DNA doesn't just squash itself into a messy ball in the middle of your nucleus. It actually folds in a very controlled pattern. They're just now beating to understand how its own structure helps it fold in the right kind of way. So the right parts are near the right other parts. When you say fold, you mean a physical fold? Yeah. Folds on itself. Right. So it's not naturally like that. It's folded up. It can't be that long. Because it's really tight. Yeah. It's inside of cells. I mean, the difference between a molecule and an atom, I mean, if I'm at the molecular level, can I see the atoms? How much smaller are they? And how do they form up to be the molecule? So atoms are measured, and molecules are measured in a scale called, typically called nanometers. Now nanometers comes from where nano is from the Greek nanostorph. Which means a nanometer is one billionth of a meter. Okay. Now that doesn't sound... You may have trouble grasping it. The analogy I like to use is your fingernails. They grow. You know your fingernails grow, right? You have to file them occasionally. A nanometer is the amount that your fingernails grow in one second. That's not much. Not much at all, right? Yeah. A nanometer is very tiny. And most atoms are... You measure an atom in roughly a nanometer or a couple nanometers size, depending upon the atom. And molecules are a few nanometers in size. So I mean, it's all tiny, tiny stuff. That's why a strand of it three feet long can pack into such a tiny thing as a cell that's only, you know, literally a few thousand nanometers. So a molecule is composed of atoms. Yes. But it's complex. There's a lot of atoms in a molecule, not just a few. Well, you know, there's very simple atoms like oxygen, I mean molecules like the oxygen molecule. There's just two oxygen atoms stuck together, and it's like water. The DNA is not simple. No, no. DNA is a very complex molecule. And it would be bigger, I suppose, than oxygen. Yes. Again, a single strand of DNA could be stretched out to several feet long. That wouldn't be the case with oxygen V. No, no. It wouldn't be that long. Okay. Okay. And now let's talk about genes for a minute so we get the picture complete. Okay. And every so often on the strand of the DNA molecule, the chromosome, we have genes. And they're just a section of the same kind of stuff that's already there. That particular section basically is an instruction book that says build this kind of protein, build this particular enzyme, you know, whatever it may be. Do they look different under a microscope? How can you tell when you're, as you're moving up and down this strand of chromosome, how can you tell you're at a gene rather than in the junk, the junk DNA part of the chromosome? You really couldn't look at the DNA molecule. It's sort of, it's the output from the little machine as it runs up and down it and reads it out basically just starts because of that particular sequence of letters as it spells out a meaningful word or builds a protein in this case. And then the next part doesn't actually do that. It makes a little other thing as it turned out to be very important how that protein may work, but it's not actually building a protein. You know, they had a program on 60 Minutes a couple of weeks ago about Colombia, the country of Colombia in South America where there are a remarkable high percentage of people who have Alzheimer's. This was valuable from a scientific point of view because then they could look at the genetic makeup of these people and see what it is about them as opposed to people elsewhere and find out what is it without Alzheimer's, how is it expressed itself in the genetic makeup. And they found there was one gene, this isn't beyond further laboratory testing, but they need to test it, but they found one gene that was common in all of them and came to believe that with certain scientific confidence that that gene was responsible for the Alzheimer's. Interesting. I had not heard this before. And they had, you know, some graphics that if not photographs through the microscope must have been an atomic microscope to see exactly what the gene, what the gene looked like on the strand of chromosome, which I found was very interesting because and I'm sure we'll go into this in the second part of the show because if they could fix that gene then logically you wouldn't have propensity toward Alzheimer's anymore. And that's exactly what CRISPR gives them the ability to do is take a picture of the gene literally and snip it out and then stick a new version that's correct in its place. This is getting more exciting every minute. And speaking of minutes, we're going to take a nano minute here. We're going to take a short break. I've already taken it. Really? Hello, Haka Ko. I'm Marsha Joyner and I'm inviting you to navigate the journey. We are discussing the end of life options and we would really love to have you every Wednesday morning at 11 a.m. right here. Happy holidays and Merry Christmas from Hibachi Talk and all of us here at St. St. Kauai. How are you doing? Welcome to Hibachi Talk. Gordo the techs are here. We're here every Friday from one o'clock till about 1.45 when we talk tech with many, many great guests. I got Andrew the security guy who helps me co-host and I got Poppy Chulow who comes in once in a while to help us through the show. So please come join Hibachi Talk every Friday. Angus will be here too. So remember, like we say at the end of every show, how are you doing? Okay, I'm Jay Fidel. I'm here with Ethan Allen, our likable scientist on likable science. We're talking about CRISPR. It's not the first time we've talked about CRISPR and it certainly will not be the last. This is revolutionary science. We're talking about how it works. We're talking about how we can use it to make life better or at least different. So let's talk about exactly what you would do with CRISPR to fix, for example, the Alzheimer's chain. So if you identified a particular gene that was associated with Alzheimer's, you could arrange, in theory through CRISPR, in an embryo, you would test and see if they had that gene. And in that embryo, then you would basically run your CRISPR technology, find that gene, literally snip it out and stick back in its place a correct version of the gene. It's not going to give you Alzheimer's. Okay, stop there for a minute. So you actually cut the chromosome like a film, like a celluloid film. You cut that part out and then splice it back together. And then you find a good gene, say an Alzheimer's or a gene that doesn't have the defect of this Alzheimer's gene. And then you hold the two ends of the film, like splicing a film, and you put the better gene back in that same spot. And now you have a chromosome with a better gene and presumably no Alzheimer's. Now, as you can imagine, it's not quite as simple as it sounds. Let's drill down. How do you do that? Actually, there was a very funny article in Science recently where this reporter was writing this whole article on the use of CRISPR because it's been a huge thing in the mouse lab business because now you can generate lines of mice with all these different genetic mutations very quickly and very precisely. And people say about CRISPR, oh, it's so easy, an idiot can do it. And so this reporter said, well, I'm going to try this. And he sat down and was taken through the process by a scientist who does a lot of this. And his version of the experiment did not actually work. The scientist guiding him was sort of doing the same experiment in parallel. His did work. It is just really sort of a cut-and-paste kind of thing. You pick up certain bits and pieces. What equipment do you use to do it? It's very sort of very simple. So if you're literally pipetting solutions back and forth and moving cells around, there's nothing very elaborate about it. It's all under a microscope. It's done a lot of it. Mind you, in the microscope, a lot of it is just you're doing this with fluids basically, full of aisles of fluids and moving stuff around. That's the thing. It really is a sort of cookbook, sort of simple stuff. There are subtleties to it. Okay, so I have my chromosome and I have this gene which I'm suspicious of. Cut the gene out, take it away, put a new, better one. Where do we get the better one from exactly? Get it from some other person? Yeah, another organism. Right, or you build it yourself. Oh, you build it yourself? Can I buy that at Tandy maybe? That's just about what it's almost like going out to your home depot. Except it's like you depot. It's scary. Yeah, I mean, they're beginning to make these kind of things. You can just buy off the shelf now and stick in. This reminds me of drones. Anybody can have a drone that can be made lethal. Okay, so you put in another gene that you got or made or scary, and then you sew it up. How do you sew it up? Does it sew up by itself? It basically does. Actually, all CRISPR really does is sort of take things apart and sets it up so that normal DNA repair mechanisms will work. DNA has a tendency to want to repair itself. And it's alive, right? Do you say that DNA is a living organism? I would not. But that's where you get into this funny thing. Our virus is alive. A virus particle is basically just RNA or DNA wrapped in a protein coat. And when it sits there like that, it basically isn't alive. It's not metabolizing. It's not doing anything. Only when it comes in contact with a proper host, then it links onto that host. It shoots its DNA into the host, uses the host's mechanisms to start reproducing its own DNA. I can't say it's alive, but it's a symbiotic kind of thing. Yeah, and some obliques are synergistic. So in the case of... But you don't need that for the DNA molecule we're talking about. So let's assume it's not alive, and you can do this without a host, right? You can do this with that single strand. And the properties of it, the biochemical properties, will heal the cut, and now it'll become one. Not always perfectly is one of the problems. Sometimes like an extra amino acid will get in the middle of it, or an extra rival sugar. And then, yes, your whole reading frame is now thrown off. Which could give you a completely unpredictable result. Probably not. Maybe a bad result. The other thing is, and this is a question I always forget to ask when you get these conversations. So congratulations, Ethan. You have spliced one gene on one DNA chromosome. That's great. But I have about 27 billion DNA molecules in my body. How does it get from the one, if at all, to the other? Well, one of the things you want to typically do, and it's often done, for instance, in breeding the mice that I was talking about earlier, you do this in an embryo, and once you put that in with that cell, you put it in a cell that's going to divide many, many times. That cell has the desired proper sequence in it. That sequence is just repeated and repeated and repeated. So in an embryonic situation, it's easy because you know there's going to be cell splitting. And it's going to multiply logarithmically. But in an ordinary sitting here at the table, it's not going to happen. So in the human test, they've actually have recently done CRISPR now over in China with some person with a very advanced incurable lung cancer. And they basically pulled some of their own cells out. They disabled a lot of them. They disabled the gene. And this gene usually sort of tamps down the cell's immune response. What is... It tamps down the immune response. You don't want... Your whole immune system needs to run at some level. It can't be too active. It's too active. It's going to react against your own cell as possible. Reject everything. Hurt, attack everything. Your own cells, harmless things that are floating around. But you don't want to be too passive. You don't want to be too passive. And one of the things cancer cells take advantage of, at least certain kinds of lung cancer, is they help keep your immune cells passive, basically. And so your immune cells don't fight the cancer. They help turn down, tamp down that cell. So you take that gene that has been tamped down and basically click it to an always-on position where it's always maximally on now. It's on the computer programming again. And then essentially clone that cell over and over and over and then inject a bunch of these cells back in around the tumor. Okay, so now you have a certain number of cells. It's a finite number of cells that you have managed to replicate in a laboratory setting. And you insert them. You inject them into the area of the tumor. In this one, Chinese investigation. But that's it. They're not going to further proliferate. They're just going to stay there and they're going to presumably have some effect on the cells in that area, but then I can replicate themselves. I suspect, and I don't know all the details, I suspect these cells actually do divide and continue to divide. And basically now you've got an army of cells that basically won't be turned off and tamp down by the cancer and should attack the cancerous cells and kill them. Well, it just seems to me, just sitting talking with you, you can program a cell to replicate itself. You can make a cell think it's in an embryo. Yes, you can. Many of your cells do. I mean, your skin is always growing and your liver is always replacing itself. So you can teach the new cell that you fixed to sort of become many cells. Yeah, the problem, of course, is, now if you think about it this individual, let's say those cells go and chew up all the cancerous cells but what's now going to turn them off? Now you've got this sort of hyperactive immune system. Which tends to reject things. Right, and it may start having some bad effects. Now in this case, this person has terminal lung cancer and got nothing to lose. If they get rid of the lung cancer, they've gained a lot. Even if they have problems down the road, hey, that's too bad. And again, this is just, what's really amazing about this is that four years ago, this was a technique that was known to a handful of scientists. People, nobody had heard of it. There was no thought of it being used. Here, in four years, they've moved this from this really obscure little sort of laboratory science into something that's being tested on people. On people. And the interesting thing is just a footnote is that in China, they did it before we did it in the U.S. because we had the FDA to deal with. And then I got to approve that until they're good and ready. And in effect, and the Chinese probably will figure out how to tamp them down, tamp them up, adjust them before, during, and after their engagement with the cancer cell so that you can come out with outside effects and be healthy, which could be knock wood, which could be the cure to cancer. And that's why they're doing it, I am sure. You got to give them credit. There's a whole issue of what they call conditional knockouts where you want to be able to turn off a gene or turn on a gene only under certain conditions and when you either have put another chemical into the vicinity or shine a wavelength of light on or something, you get a switch that basically takes them on and off. It's a computer program. It's an if-then statement or a case statement. And you can program them to do that. And you can program them to split and proliferate. My old boss, Leroy Hood, used to say... I remember that. I met him. I told you that. All right. And he said, biology is fundamentally information science. That's what it is. Well, I mean, it sounds very promising. Do you have any sense of what is happening in this country and what other advances are being made using CRISPR? And we don't have time for a complete discussion, but I'm just interested in your thought about that because I think we should come back to this again. It's not as if we're rehashing old stuff because this is moving so fast that we could discuss it every so often and always have new material to deal with. Exactly. So, you know, in a few months, they're going to be doing another larger-scale trial in China on more people. Probably by the end of 2017, we'll start some very preliminary tests in the U.S. because everyone sees that there's a great deal of potential in this technology to do a lot of good for people if we can sort of get a hold of it and know what to do. But you don't have to have a rich imagination to figure out how we could do bad, too. You could splice the gene the wrong way. You could splice the gene to have some really profound negative effect on a given organism, and then you could splice the gene, splice the whole DNA strand to replicate, and then you could have it replicate quickly because that's all within the programming here and then effectively kill large populations with it. So it's very scary, and he who controls CRISPR now will control much more in the future. It is one of the so-called dual-use technologies. Thank you. So many things are, right? They can be used for good, they can be used for ill. That's getting to be an increasingly recognized problem is how do you deal with these as there are more and more of such technologies and they're more and more sophisticated, but they're also getting sort of more widespread and easier to use. That's the real thing about CRISPR. It doesn't take a really big fancy lab. It doesn't take a gazillion dollars' worth of equipment. You can do it with stuff you can buy, stand at a commercial firm. You could set up in the studio, probably, a lab and start doing CRISPR technology. Yeah, it's like hacking. What does it take to get into hacking? The hacking programs you can buy on the Internet are free, most of them. Anyway, the point is that the answer to your question is nobody's watching. It's a life of its own and we have to learn to watch and learn to monitor it and learn to be smarter than the next guy. They are watching. People are getting together and thinking about these and discussing these very issues. Years ago, when recombinant DNA first came out, there was the big Asilomar conference. They brought together scientists, philosophers, theologians, lawyers, everyone, and sort of had them sit and say, look, this recombinant DNA has potential to do real bad things if we just let it go. But we can't stop it. It's basically sort of a Pandora's box kind of thing. It's been opened. What do we do? How do we regulate it? How do we control it as best we can? Okay. We're going to leave it there. Yeah. That's Ethan Allen. He's likable. He's a likable scientist. This is on likable science here on Think Tech. Every Friday at 2 p.m. Come back soon and we'll talk some more about CRISPR and many other technologies you need to know about. Aloha.