1,321 words · auto-generated from the episode video
1:06:34escape. And also if it gets poked it'll reconstruct. So it's it sort of has security in that context. So the point being you want to be careful when you start doing things like this to not introduce >> you know new variables that the system will attack or not prepared for. But the the the structure we've already walked through really identifies that this is not going to interrupt regularly scheduled programming while still enabling the capability of capturing the memory of a cell over this long much longer period of time. >> Exactly. And so now you can ask this the mRNA that's inside the vault protein, right? How long does that last? Well, the thing is lasting for about 17 days.
1:07:14Okay, the stuff that's inside the vault protein. And you ask, okay, why is it lasting that long? Well, that's actually the inherent physics of RNA. >> Okay. >> Okay. RNA is ribboucleic acid versus DNA is deoxxyribboucleic acid. The one of the big differences is of course the DNA is double stranded and RNA is single stranded. So DNA is a bit more stable obviously because you've got you know a fullyfledged ladder that's sort of protecting the inner bases. The other big thing is that the ribos sugar which RNA has and DNA has deoxyibbos sugar. The rival sugar has an extra O. >> Okay. It's got an extra hydroxal group on the on the two of that pentagon.
1:07:54Right. Yep. >> And that extra O that oxygen oxygen is always just bad. Okay. It just attacks random crap. And so the RNA, that extra oxygen goes and attacks the phosphate backbone. >> And that process >> has a characteristic time scale of 17 days. So, we're reaching the theoretical limit of how long RNA can just hang around before the chemical RNA itself starts attacking itself. >> We we're we're literally at the what did we call the the ed the Edington limit of your of your in this case it would be the thermodynamic limit of of RNA stability. But what it's what it's showing is that there's no enzyatic
1:08:36degradation of mRNA, right? There's no other stuff. There's no exoomes that are coming in trying to like take it out. This is just pure physics. It's going to last about 17 to 18 days. >> It's locked down but for the fact that uh the structure of uh ribos itself will degrade >> will degrade the iron >> which we that's a bigger that's a whole >> Yeah. That's like I don't know what to do about that. Right. [laughter] >> Yeah. So so here you can show that like you know this >> this is really happening right. So the critical step is once you have once you have like the RNA that's present in the cytool let's say versus RNA that's present in the time vault. How do you
1:09:17tell the two apart? >> Right. Okay. >> Right. Cuz if I were to now take the cell and I were to say, okay, I'm I want you to record now and then in 5 days I want to read it. >> Right. How do I how do I >> tell what's already there versus what's in the vault? Well, I can introduce something called RNA, which is an enzyme that eats RNA. >> If I make it eat RNA, the stuff that's inside the time vault is protected from this enzyatic degradation. >> Oh my god, bro. This is so good. >> It's so good. And and so now the if you add RNAs, the only stuff that we're going to be able to read afterwards is the stuff [clears throat] that's inside the vault protein. >> God, that's [sighs] so good.
1:09:58>> It's so good. I this is this is this is um you're watching masters at work here. >> Yeah, this is great. >> I I can see why you started off with such a high high bar. >> High bar, right? There are some there are some caveats. For example, it's not going to it's not going to take in mitochondrial RNA because the mitochondrial RNA is inside the mitochondria, right? And it's not going to take in >> um any RNAs that lack a polyatail. There's some there's always some that are exceptions to the rule. And so, you know, the the interaction protein is not going to be able to grab that mRNA because there's no polyatil to actually grab. Um, the capture efficiency is also quite low. It only takes in about 3% of
1:10:40the total mRNA, right? Because there's so much mRNA. So, it's only going to take in a small sample. And so, that low capture rate means that you really need to pull together thousands of cells to prevent this statistical bias, right? where if you were to do do just one cell, how do you know that you got all of it if it's only taking 3%. But if you do a thousand cells, then you're pretty sure that you've got all of the mRNA covered. >> That's well that's a solvable. We'll solve that problem. >> And and there's there's ideas later that'll that'll help you solve it. But I just want to let you know that in at least in the first iteration of this, this is still pretty >> No, this this is this is >> cuz now that this base is there, like I I'm just shocked at how clever that is.
1:11:20>> Yeah, it's it's really nice. Okay. And so the way so how do we know that it works? Well, they did a really simple experiment. What they did was heat shock these cells after um after and then right after the heat shock they had the RNA go into the vault proteins. >> Okay? And then they compared the vault proteins right after the heat shock to when you and then you know you wait five days and you sequence the mRNA that's in the vault proteins and then you compare that to the heat shock and then right after you you lice [clears throat] it and then you do it right. If this whole thing worked, >> yes, >> then the mRNA that's in the vault proteins right after the heat shock should match
1:12:01>> normal cells right after the heat shock. The transcriptto should match. And that's exactly what they found, right? Is that the present transcriptto, if you look at >> 5 days later, the present transcriptto forgot that it was heat shock, >> but the stuff that's inside the vault remembers that it got heat shocked because you did the time machine thing. >> You did the backup >> and you did the backup right then. >> Right. Very beautiful. And the figures, I got to say, the aesthetics of the figures are also really nice. Like the color coordination of the blue and the red, I liked it. I don't know. >> Look, look at we like design and aesthetic, both experimental and visual. >> Yeah. Yeah. So, so I thought this was really cool. So now applications, right? They actually went further and went into applications. This is the one where
1:12:42people talk about, you know, real world applications. So Harvard when when it came out with the story in their press, they talked about how did that cancer cell become drugresistant? Mhm. >> One of the first things you can do is apply this thing to cancer cells. Okay. >> Whenever you do cancer cells and you treat the cancer cells with a drug, there's always 1% of cancer cells that are drug tolerant persisters. >> You know how like you have like the soap that's like kills 99% of germs. Well, there's always 1% of germs that get away, right? And the same thing happens