[SciShow intro plays]   Michael: CRISPR. Its name kind of sounds like an ad for fresh vegetables, it’s changing the entire field of biology. Using this technique, scientists can edit genes with a precision, accuracy, and speed they never had before. That’s great news for research and human health! But if humans can now manipulate the genome of any organism, there are ethical questions that we need to consider.   Any molecular biologist will tell you that genetic engineering is a tricky pain in the butt. The goal is to do one of a few things: You could change the amount of a gene in a cell, or delete it completely. You can also make a mutation in a gene, by altering, adding, or taking out pieces of DNA called base pairs. Or you can introduce a gene from a totally different organism. Like, how we made bacteria that synthesize insulin, and how we use the gene for GFP, the glowing green protein from jellyfish, as a biomarker in lots of experiments.   Traditionally, there are a handful of techniques you can use to do these things, but none of them are really convenient. For example, you can use a solution with special chemicals or particles to get DNA into lab-grown cells – generally called transfection. The cells will express the gene for a few days, but you get way too much of the gene floating around, and way too much expression.   Plus, it isn’t part of a chromosome with other genes and regulatory bits, so you can’t be sure it’s behaving normally. You can also use a modified virus to deliver a gene, because they normally stick their genetic material into cells, and integrate it with their host’s DNA. So just kill off the make-you-sick parts, give it a gene you want to study, and let the virus do what it does best. However, this method has problems too: the virus doesn’t care where it sticks the new gene, so it could be smack in the middle of another one, and cause a bunch of problems. And that’s just individual cells.   We also have these proteins called nucleases, that can target and cut certain sequences of DNA – kind of like tiny pairs of scissors. One type, called zinc finger nucleases, can be customized to cut near specific sequences of bases in grooves in the DNA double helix. They’ve even been used to make a few human patients more resistant to HIV infection, by breaking a gene that helps the virus enter our immune cells.   But when you change the building blocks of a protein, it’s really hard to predict what the 3D structure will be. So it takes a lot of time, and you have to engineer a brand new protein for each cut. Changing a whole organism, like knocking out a gene in a mouse, can take years, thousands of dollars, generations of breeding, not to mention a bit of a luck.   Or, it used to. Enter CRISPR, or more correctly, CRISPR-Cas9. CRISPR was originally described as a sort of immune system in archaea and bacteria. See, when some of these organisms get infected by a virus, they save a chunk of its genome in this long charm-bracelet-like string of DNA called a CRISPR locus.   They use each DNA chunk as a template to synthesize a similar molecule called RNA, which does lots of things in cells. Here, the RNA is a guide to find more matching DNA chunks. Cas9 is another one of those nuclease proteins, and it cuts wherever the guide RNA tells it to. So it’s all about that RNA-protein duo: The next time that virus shows up, the guide RNA finds the matching DNA, and calls over Cas9 to chop it up and destroy it.   And here’s the thing: a system where you can send one protein to cut anywhere in the genome, just by giving it a chunk of RNA, is really useful for molecular biologists. Making a specific sequence of RNA is much, much easier than making a specific 3D protein shape, like for those zinc finger nucleases. CRISPR lets you make a precise cut in DNA, so you can use it to knock out a gene, or even trick the cell into inserting a gene. It just depends on how the cell fixes the broken DNA strands.   Sometimes the cell just jams the ends of the DNA together as best it can. That method is prone to mistakes, so it’s good if you want to break a gene. Other times, the cell uses a template to repair DNA. And you can send in a template along with the RNA and Cas9, so it adds whatever you want, like a whole new gene!   Since the template’s guided to a specific part of the genome, you’re going to have just the right amount of your gene, integrated into a chromosome – unlike older techniques, where you get way too many copies, or even too few. So CRISPR makes gene editing faster, cheaper, and more accurate than ever before. Cue an explosion in genetic engineering research.   CRISPR is mostly being used to edit genomes in cells at lab benches – to learn how they function without a gene, or with a new one. Some scientists figured out if you blunt the “scissors” of Cas9 so it can’t cut, the protein will go where the guide RNA tells it to and just sit there, turning a gene off temporarily instead of deleting it. That way, Cas9 acts kind of like a light switch: turn the gene off, see what happens; turn it on again, see what happens then.   But, like any technique, the CRISPR-Cas9 system isn’t perfect. Sometimes, it cuts in the wrong place. It might look at a sequence that almost matches the guide RNA, and go, “yeah, that’s probably right.” It can also insert sections of DNA the wrong way around, or delete sections. In lab-grown cells, this isn’t a huge problem. You can just use the technique a couple times and check for cells that got it right, and it’s still way better than inaccuracies from other methods.   But doctors are also beginning to develop CRISPR as a gene therapy for patients. And making a wrong cut in the genome of an already-sick human could be disastrous. If they have cancer, for instance, an extra-damaged gene in the already-mutated cells might make the disease worse. That’s why scientists have been carefully engineering the guide RNA and the amino acid sequence of Cas9 – to try and make the technique more accurate, and cut the right sequence every time.   They’ve made enough progress that human clinical trials involving CRISPR-Cas9 are making their way through the approval process: one at the University of Pennsylvania in the U.S. and one at Sichuan University in China. Both are combining CRISPR with another hot cancer treatment called immunotherapy.   Immunotherapy is based on the idea that a person’s immune system can find and fight cancer, with some help. Both trials will take some of the patients’ immune cells, and use CRISPR to give them a genetic cancer-spotting boost. Then you put the immune cells back in the patient so they can get to work. One of the genes they’re altering was also targeted by a drug in the cancer treatment of former U.S. President Jimmy Carter. And researchers expect the CRISPR method to work even better.   Editing an adult’s genes in their own cells to treat them isn’t that controversial, but CRISPR-Cas9 gives scientists the power to edit human embryos. And as you can imagine, this is caused quite a debate in the scientific community. The good news is, scientists are aware of the “just because you can doesn’t mean you should” principle, and everyone is thinking very carefully about how to proceed.   We’ve also thought about using CRISPR to kill all the mosquitoes. Or, at the very least, alter their genomes, because some species transmit lots of deadly diseases. We’d do it by taking advantage of a naturally-occurring genetic quirk in some insects called a gene drive. Since chromosomes come in pairs, most genes have a 50/50 chance of being passed on to offspring.   Not gene drives, though: they code for a protein that cuts DNA, specifically to cut the other chromosome that doesn’t have the gene drive. The cell goes, “crap, how do I fix this?” And the gene drive goes, “look, chromosomes should match, so you should just copy me!” So, through the normal repair process, the cell ends up with a copy of the gene drive on both chromosomes – so it’s definitely going to be passed down to any offspring.   By doing this, gene drives can pretty much spread through an entire insect population after multiple generations, regardless of natural selection. And Cas9 can be engineered into a gene drive pretty easily: it already cuts DNA, so you just stick the gene for Cas9 in a mosquito chromosome, plus a sequence that’ll become the RNA guide, and any extra template sequence.   The RNA guide can target any mosquito gene you want, but only when the gene for Cas9 isn’t present. So any time that targeted gene is in the mosquito’s DNA, the CRISPR-Cas9 gene drive will essentially cut and paste itself over it. In theory, we could target genes that affect fertility in mosquito species, like the ones that carry malaria. As the gene drive spread over multiple generations, this would exterminate them forever.   A less extreme idea would be to target genes that allow things like malaria parasites or the Zika virus to survive inside certain mosquitoes – so they can still bite people, but diseases aren’t transmitted. By the way, this wouldn’t really work in humans. Mosquito generations are short enough that a gene drive could spread in a few years. Human generations... are not. You’d probably have to wait millennia to see any widespread effect.   For the most part, CRISPR is revolutionary because it’s a day-to-day tool for scientists. It’s a way for them to learn more about their favorite gene, and do more experiments. That’s good for the public because it speeds up research, and it’s cheap. And we all want effective pharmaceutical drugs faster, like for cancer treatments. None of those experiments are going to escape and start a genetically-modified zombie apocalypse, because those petri dishes of cells are pretty hapless – they’ll die if you leave them on the counter. Moving forward, research institutions and regulatory agencies are keeping a close eye on what CRISPR-Cas9 can do. Because this technique can do a lot of genetic engineering stuff, and it’s accessible to a lot of people. And, mostly, that’s just really exciting.   Thanks for watching this episode of SciShow, and thank you especially to our patrons on Patreon who help us make these videos. If you want to help support this show, just go to Patreon.com/SciShow. And don’t forget to go to YouTube.com/SciShow and subscribe!