1,308 words · auto-generated from the episode video
1:13:12with cancer. If you target therapy with something called called oymeib which is a drug for cancer 1% of these cancer cells survive. Now the debate is the following. Were these cancer cells pre-addapted >> or was the fact that I introduced the drug causing them to adapt? >> This is a debate that goes all the way back to Charles Darwin and Jean Baptiste Lamar. Um there was the idea of how do organisms gain traits? The Darwinian model was the one at the top that you see, which is let's take a population of giraffes, right? Why do giraffes have long necks where well, you could have a population that has a bunch of different
1:13:53neck sizes and then the longer necks could reach the trees and eat the trees and so then the shorter necks died off and then the subsequent populations got longer and longer necks. That's actually the one at the bottom. Okay. Now the Lamarian hypothesis is the hereditary nature of acquired characteristics. Meaning the giraffes like kept stretching their necks and then that stretching of the necks made the subsequent population stretch their necks >> and so now that's why we have longer neck giraffes. >> Would it be almost like an epigenetic
1:14:33kind of theory? Like more of an epigenetic theory? >> Exactly. Yeah. And for what we know from large scale populations, it's really a Darwinian thing. But the Lamar hypothesis is not completely off ground, right? At the cellular level, we have examples of that. For example, with bacteria, when we heat shock bacteria, the bacteria just start taking in DNA from their environment, right? So that's kind of Lamarian because I've introduced a stress and then that stress is causing the genetics to change. So the question is what's happening with these cancer cells, right? And currently we can only record the bulk population which is mostly dead >> and [clears throat] then we can record the survivors which is after the fact. So to answer this question in cancer biology you have to be able to record
1:15:14what's happening before the drug was introduced >> and then probe what the survivors were doing >> doing. Yeah. Yes. >> Does that make sense? And this is perfect for the time vault. So what they did was they had a protocol where you record for 24 hours with no drug and then you introduce this cancer drug oimon. You record this cancer drug for 4 days 99% die. Yes. >> And then the ones that survive, let's read out what's inside their vaults. >> Yes. Yes. >> And it was definitive evidence for the selection model, meaning that these cancer cells had previously adapted. >> The cancer cells have a very high mutation rate. And so they're exploring
1:15:55all of these different possibilities and the 1% of cancer cells that survived >> had a distinct transcriptional signature before the drug treatment. >> Right? And so now the focus for therapeutics can shift from adaptation to targeting that pre-existing state. And what they did was they they looked at some of the genes that were expressed in those vaults, specifically FN1. >> They knocked that out >> and then now none of the cancer cells survive that drug. >> This is so it's like literally working. This this is actually un unreal. I want people to understand how unreal this is because now you're B because we can now DVR ourselves. We can
1:16:35>> we can MC time machine backup ourselves. It means we can actually see what happens before like we can understand after you introduce some external factor >> like what the changes are and particularly in the cancer research example you just brought up. This is so fascinating. You have all of them you you you do the recording in the time vault. You're doing it over 24 hours. You have all of them there. Then you introduce the current treatment. >> Yeah. 99% die. >> 99 are gone. 1% are left over. You can look at the 1% that are left over and say, "What is it about these guys that's different?" And then now we can attack the thing like the FN1 that's different because we are able to now actually see
1:17:16over time do it again and now they all die. That's incredible. >> Yeah, >> that's incredible. Do you understand what we can do with that? >> Yeah, that's so cool. You can you can use this for like and the fact that these vault proteins are just ubiquitous in malian cells means that you can just apply this all over the place. >> Exactly. This almost feels uh structural conceptually as I try to put it into my brain. It feels crisperesque as being a platform versus a single point solution. >> Exactly. Yeah. >> This is really a really big deal. Once it, you know, all the once it percolates and people poke a hole, reviewer two pokes a hole that >> No, but people people are very excited
1:17:56from the from the press coverage that I've seen, people are very excited. >> This is I mean, holy moly. >> Yeah, it's very cool. >> And what a great name. Time vaults. >> Yeah, great name. Well done. Well done. Well done on that one. >> Whoever was doing the branding, right, >> that the branding on that one is great. >> So, what's the future? Well, we want to overcome this 3% capture efficiency, right? So, one thing that I thought was really cool, the idea was you can put in inside the vault protein, you can put in something called reverse transcriptise, and what that's going to do is take the mRNA that's getting stuck inside the vault protein and reverse transcribe that into DNA. >> Now, the DNA is going to last much longer than the 17 days that's the thermodynamic limit for RNA. And so, you
1:18:38could like have weeks >> of time vaults, right? This so it's it's almost like there's a a short-term memory in the mRNA storage that get and then sort of like how we record. We record onto a small hard drive and then we dump it onto our archive drive for long-term storage. >> It's exactly that. >> Yeah. The other thing could be like you could have molecular identifiers per vault >> per cell >> and then it's kind of like a barcode for each cell. And then you could be like these mRNA came from this cell and these mRNA came from this cell. So you get the aggregate but you also get single cell resolution. kind of cool, right? >> To think about >> you could combine this time vaults with super resolution and expansion microscopy. So then you could locate
1:19:20where the time vaults were in a cell. Right now they're just sort of len in a test tube. But imagine, you know, you stick it under a microscope. You could locate where in the cell it is. Is it next to the nucleus? Is it next to the endopplasmic reticulum? So on and so forth. Is it next to the the cell membrane? >> Synthetic organels. You could have programmable delivery vehicles >> that will just like go through. >> I think it's it's it's very very cool. >> My mind is just >> Yeah, there's so much that you can do with this. >> This this this was a good Again, we don't always like to give credit to our two of our arch rivals in Harvard and MIT. >> But kudos, guys. You know, >> this was a good one.
1:20:00>> Still not Princeton, but you know. [laughter] >> Yeah, we'll give credit where credit's due. >> Yeah, this one's a good one.