Folding DNA Into Tiny Circuits
IRA FLATOW, host:
You're listening to SCIENCE FRIDAY from NPR News. I'm Ira Flatow.
Up next, new technology that turns DNA into parts of computer chips.
Of course, going back, going 50 years now, the holy grail of electronics is really to shrink the parts, to make them smaller and smaller, and that's one of the reasons why the early days of solid-state electronics were funded by the military so they could build these tiny circuits that fit into military satellites.
And that search for shrinkage is still a driving force in digital technology. Scientists have taken a logical step - sure makes sense, doesn't it? You take two of the smallest objects that you can find, nanoparticles and DNA molecules, you combine them together to make DNA computer circuits.
And here to tell us how to build a circuit using DNA is my next guest, Greg Wallraff. He is a research scientist at the IBM Almaden Research Center in San Jose, California. Thanks for being with us today.
Mr. GREG WALLRAFF (Research Scientist, IBM Almaden Research Center): Yeah, thanks. It's good to be on the show.
FLATOW: Tell us what you have found. What did you do? How do you make DNA and non-living things like nanoparticles work together?
Mr. WALLRAFF: Well, what we're trying to do is actually use the DNA molecule as a template to assemble these nanoparticles. The DNA we're using is an engineered DNA, you might call it. It's called DNA origami. It was - the design of this was developed by our collaborator at Caltech, Paul Rothemund, and what it amounts to is a single molecule of DNA that's actually relatively large. It can be formed into a shape, folded back and forth on itself so it's in a two-dimensional shape. The size is about between 100 and 200 nanometers on an edge and it's about two nanometers thick.
Mr. WALLRAFF: And depending on how you make it, this thing can - the DNA can be shaped into things like triangle, or rectangle, or five-pointed star - even things like a smiley face.
And that's all very interesting, but what we want to use it for, there's a second attribute of this DNA origami and that is you can put patterns on the surface of this DNA structure. And those patterns consist of little posts and you can bind nanoparticles to those posts.
And the thing that's interesting to us is the posts can be as close as six nanometers apart, so that the nanoparticles you'd put there, which we hope will ultimately be made into transistors, can be put at very fine pictures, very close together. And that's the interest in trying to use DNA to fabricate microchips.
FLATOW: So you don't care about the chemical composition. You care about the fact that you can use the DNA as sort of a building block, as sort of a skeleton to put your nanoparticles on top of.
Mr. WALLRAFF: That's exactly right.
FLATOW: So it's like the girders inside the building. You don't see it, but they hold the building up.
Mr. WALLRAFF: Yeah. Another analogy people use, it's this DNA origami structure is like a miniature circuit board.
Mr. WALLRAFF: And the circuit board is fairly big, but you can assemble much smaller components on it.
FLATOW: You're not wire wrapping, right? No. Or something like that around there?
1-800-989-8255 is our number if you'd like to talk about DNA and nanotechnology.
Have you actually made a working part yet?
Mr. WALLRAFF: No, we - that's in the future. What we've done so far is the first step.
As I mentioned, we want to use this as - to assemble the components, but really important is that we're able to take these - this miniature circuit board and then place that on a larger structure to make the connections.
So we've - in the paper we recently published, that's where we described: a way of positioning individual origami structures on silicon wafer surface. And then, once we get the components built up, we have a way of connecting things to the outside world and basically wiring things up.
FLATOW: Let me ask you about the mechanics of how this works. How do you get a molecule of DNA to behave the way you want it to behave, to make this shape that you want it to make?
Mr. WALLRAFF: Yeah, that's - the way it's done is you start off with a long single strand of DNA. As you recall, DNA normally exists as a duplex, two single strands wrapped together in the famous double helix. And what Paul's technique involves, is starting with a fairly long single strand of DNA that's isolated from a virus. And then, he has prepared very short pieces of DNA. He calls them staples.
And what the staples do is interact with a long single stand in the same way that normally happens when you make double-stranded DNA, but they're designed so that the staple - half of the staple reacts with one segment of the long strand, another half reacts with the other, and that forms a loop. And then he has several hundred of these strands together and they just self-assemble. You mix the single strand of DNA, you mix all 200-odd strands that he's designed using a - the composition of which he designed using computer program - you mix them together, you heat them up, the DNA hybridizes, he's precisely to find nanostructures form and that's it. You'll have billions of these in a single drop of water when you're done with the experiment.
FLATOW: So do you then go in and fish out the shapes you want? Are there all different kinds of shapes? You can say, I'll take one of that and one of that, one of that?
Mr. WALLRAFF: No, no. He's - the staples are designed to form one shape.
FLATOW: One shape, billions of them.
Mr. WALLRAFF: Yes. And then, the next step then is the thing we talk about, is we have our - we use lithographic techniques, standard semiconductor processing techniques to make another sort of template. But this template is on the silicon wafer. And there - we've designed it so that their binding sites that match the shape of the origami, and they have a property such that origami sticks to them.
So, we've, sort of, got this silicon wafer covered with tens of millions, we can, of these sticky patches. And then, you just pour the solution containing these shapes onto the wafer. And shapes that match up match up, and then you have all these DNA nanostructures lined up on the wafer like ducks in a row.
FLATOW: And they just know how to do that?
Mr. WALLRAFF: Yeah. It's just - this is a kind of - it's a templated self-assembly. And it's - I think it's even a familiar topic on - you've talked about it before…
FLATOW: Yeah, yeah.
Mr. WALLRAFF: …in your show, and this is an example of that.
FLATOW: It just sounds astounding is why, you know, to people who have - to think that the DNA can - to link up and do those things. And then, at what point do you put the nanoparticles in there?
Mr. WALLRAFF: Yeah, so the sequence is you prepare the DNA nanostructures and solution.
Mr. WALLRAFF: And then, you prepare our lithographically fabricated template with the sticky patches.
Mr. WALLRAFF: Then, you pour the DNA nanostructures, and they line up where you want. And then the next step is to put the nanoparticles on. And I think -maybe they didn't say enough about these sites on the origami. There are actually also little segments of DNA, single strands of DNA. And you can design those so that you can take the compliment of that single strand of DNA and put it on the nanoparticle.
And then those match up. They do this - the familiar Watson-Crick base pairing. And then, these - the nanoparticle will find the site that it's basically coded for. And the advantage of the origami, in addition to being able to put these sites very close together, you can make them, sort of, individually addressable, because in principle each site can be different, and so each site would be specific for putting a different kind of nanoparticle down or a different sort of nanorod or carbon nanotube.
FLATOW: And would you then use these sites to make transistors out of them or memory circuits?
Mr. WALLRAFF: That's exactly right. That's the ultimate goal.
FLATOW: Mm-hmm. So you basically can make memory at the size of a DNA molecule.
Mr. WALLRAFF: Yeah. It would be - the memory would be made at the size - that's right, because the DNA molecule is two nanometers. But the size would be determined by the - actually, the size of the component you put on, the nanoparticle you put on.
FLATOW: Mm-hmm. And what - well, let me get some questions on. I'm hogging the conversation at 1-800-89-8255.
(Soundbite of laughter)
Let's go to the phones. Let's go to - I'm just fascinated. Jim(ph) of Boca Raton.
JIM (Caller): Hi, Ira. I love your show.
FLATOW: All right, go ahead. Thank you.
JIM: Years ago, I read in Discover magazine how they were willing to try to take the, you know, the DNA, the ACTG. And instead of using it as scaffolding that they were going to try and go from, you know, like binary code and then do, like, a four code which I guess would - I'm not sure what that would be called, like, a tetra or a quaternary - I don't know what it would be called. But that they would use that, that they, you know, instead of the, you know, on and off, would be four.
And I just wondered - and it was the Japanese researchers that made some headway. And I just wondered what ever happened to that? I mean, where is that going?
Mr. WALLRAFF: Yeah, that's - this is sort of specifically designed at the fabrication end. And you're right, my understanding anyway, DNA computing was very popular thing a number of years ago. But I personally haven't seen much about it lately.
JIM: They just, kind of, died on the vine?
Mr. WALLRAFF: I think - I'm sure people are still looking at it. I'm just not that familiar with the most recent developments.
FLATOW: All right, Jim. Thank…
FLATOW: Thanks for calling.
JIM: Thank you.
FLATOW: So what's the next logical step now? How do you go about then assembling real working electronic component out of it?
Mr. WALLRAFF: Well, one of the things we have to address still - as I said, we needed to be able to put these individual structures on the surface of the silicon wafer. And one of the things we're working at now is still trying to optimize that process. We have a couple of different kinds of templates that we use to line the structures up, but we're still trying to optimize it because, as you know, if we ultimately do try to get this into - make a chip, the performance tolerances and the alignment overlay issues are very stringent.
And so, we're trying to make sure that we can line these structures up is - with the best precision we possibly can. But then, the next step then, and also we're looking at better understanding how these DNA structures stick to the wafers and see if we can design other templates built out of other materials for use in the fabrication process.
But the next step is to start coming up with ways to put the components on the chips, on the DNA rather.
Mr. WALLRAFF: And that's demonstrated. It's already been demonstrated by us and others, and that's one of the things we're working out now is trying to optimize our process.
FLATOW: I heard of a new component. I think I read about it this week, like a nano-level laser beam? A laser - spazer(ph), I think it was called. Are you familiar with that?
Mr. WALLRAFF: Not really, but that's one of the advantages of, I think, this approach. And our ultimate goal is to use it to make computer chips. But I think it can be used for lots of other things, just sort of in its current form. And I think anybody who wants to, for example, put down molecules or nanoparticles at precisely spaced dimensions and portently(ph) being able to find them on a substrate can use this.
Mr. WALLRAFF: So you could use things, like you could put down individual molecules and study their spectroscopic properties. If you have a nanoparticle that can leave, you could put that down and study it. I think it can be useful for a lot of other scientific applications as well as some technological applications too that we may not have thought of yet.
FLATOW: Mm-hmm. Talking about DNA and microchips on SCIENCE FRIDAY from NPR News, talking with Greg Wallraff that works at IBM in San Jose, California. 1-800-989-8255 is our number.
Julia(ph) in Chicago. Hi, Julia.
JULIA (Caller): Hi.
FLATOW: Hi there.
JULIA: Hi. I have a question. Usually, when I think of nanotechnology I think of things that are not alive or nonbiological. But now that you've incorporated the DNA structure, are you actually making things that are biological with nonbiological, or, like, pharmaceutical entities, or exactly what is it?
FLATOW: Is it alive, is I thought she wanted to say.
Mr. WALLRAFF: Well…
JULIA: It sounds like it's alive.
(Soundbite of laughter)
Mr. WALLRAFF: It's not really alive, your analogy, but using it as girders for assembly is more accurate. And there's a whole field of DNA science, it's called DNA nanotechnology, and that's exactly what people do. They use DNA, and this origami is an example. You know, the DNA is sort of a construction material. So we're using it to make these flat, two-dimensional objects people can make…
FLATOW: Is it alive, Greg, or not, cause I…
Mr. WALLRAFF: Oh, no. It's not alive.
FLATOW: Not alive. Okay.
(Soundbite of laughter)
Mr. WALLRAFF: No, it's not.
FLATOW: All right. And I don't want to interrupt because - but we ran out of time. But I want to make sure we got that answer because folks will be wanting - thank you, Greg, for taking time to be with us today, and good luck with you.
Mr. WALLRAFF: Okay. Thank you.
FLATOW: You're welcome.
Greg Wallraff, a research scientist at IBM Research Center in San Jose, California.