DNA on the move

Deep inside a beaker in a humming chemistry lab in New York City, a spindly spider crawls over a jumble of origami. It’s not the colored-paper kind of origami, but rather is made of precisely designed segments of DNA. For that matter, so is the spider.

This spider wasn’t built to spin webs or eat bugs. It’s a DNA nanorobot, a primitive version of the machines that may someday perform tasks too small for humans to do.

For more than a decade, scientists have been developing DNA nanomachines, from tiny tweezers to two-legged “walkers” that can step to the left or right. Recent molecular robot research has gone a step further, aiming to get DNA molecules to organize themselves and move about, all without batteries or information storage in their nanobodies. These machines harness the power of natural DNA-DNA interactions programmed into the origami foundation.

“Right now there’s a molecular explosion going on in programmable behavior of molecules,” says biochemist William Shih of Harvard University. “Just like we’ve seen the evolution in electronics from the calculator to the iPhone 4, we’re going to see these things evolve into sophisticated vehicles that can sense their environment and target diseased tissue without harming healthy cells.”

The latest arachnoid nanobots have three to four legs and walk across expansive landscapes of exquisitely folded DNA. Some of these molecular machines can take 50 steps all by themselves. Others sport wiggly arms that can pick up and carry around nanoparticles.

DNA spiders aren’t going to take over the world anytime soon. They’re more like toddlers at this point, tentatively feeling their way across molecular territories as researchers work out the basics of getting the crawlers to move. But one day nanobot armies may tackle jobs at a scale too small for even the most sophisticated laboratory apparatus. Spiders might be able to seek and destroy cancers in the human body, assemble nanosized medical devices, and build tiny computers vastly smaller than the dot on the iPhone’s “i.”

“We’re pushing the envelope of what’s possible with DNA as a working material because we can understand, control and direct DNA more than any other material,” says chemist Lloyd Smith of the University of Wisconsin–Madison.

Built-in programming

Getting DNA to move on its own isn’t easy. Ordinarily, DNA exists in cells as a twisted double-stranded helix, the blueprint for all life’s materials. It’s stable and unreactive, untwisting only to be copied for making other molecules like proteins or to replicate itself.

But in recent years, scientists have figured out how to use DNA’s own code to set it in motion. DNA strands are made up of four basic chemical building blocks, abbreviated A, T, G and C. In regular DNA, those four letters spell out codes for building proteins. In spiders, those letters are the basis of propulsion.

Each of the spider’s legs is made of a single strand of DNA with a specifically engineered sequence of letters. Just as in regular DNA, the A’s of one strand are shaped just right to latch onto the T’s of another, and the C’s match up with G’s. By binding to the right partner letters, the legs can stick to single strands of DNA nearby.

That’s where the origami surface comes in. DNA origami was invented in 2006 by Caltech synthetic molecular biologist Paul Rothemund. He folded single strands of DNA back and forth until they filled complex two-dimensional shapes: nanosized triangles, stars and smiley faces. Then he designed smaller “staple strands” that matched up with adjacent DNA folds to latch on and hold the shapes in place. Just mixing the single-stranded pieces together in solution allowed the shapes to assemble themselves.

Rothemund’s self-assembling DNA origami makes an ideal walking track for DNA spiders, offering a large, 2-D surface into which scientists can program instructions for a spider’s movement. That way, the spider doesn’t have to carry any information onboard.

In the origami, select staple strands are elongated with extra DNA building blocks to form a crawling trail for the spider. These single strands add a third dimension to the flat surface, sticking upward from the origami like seaweed on the ocean floor. Since their DNA letter sequences match up with those on the spider legs, the staple strands hold the spider to the surface and form a path for walking on.

That part is easy. The tricky part is getting one of the nanobot’s legs to pick up and step forward to the next strand.

Forging a path

One solution is to use DNA enzymes in the spider leg to slice through the staple strand. Breaking that strand uproots the leg, allowing it to move on to a nearby strand that is still intact.

Chemist Milan Stojanovic of Columbia University uses this cutting method to get his three-legged DNA spiders to move on their own. They can take upwards of 50 steps without stumbling off the track, he and colleagues reported May 13 in Nature.

In the past, a problem has been that two-legged walkers sometimes pick up both legs at the same time and float away from the track. With three legs, the spider has a better chance of having at least one leg on the surface at all times.

“The more legs you have, the stickier spiders are and the more steps they can take,” says Stojanovic.

His spider has an extra appendage that works like an anchor to bind only to a “start” strand on the origami. When researchers add a piece of DNA that dislodges the spider’s anchor, the spider begins to crawl along the other strands of the track.

Since the spider cuts up the strands as it goes, the DNA track gets used up behind it. So the spider is most likely to step forward, not back. Following the track laid out on the 65-by-90 nanometer origami field, the spider can walk straight, turn a corner and make its way along the path with no outside help. After about 30 minutes, the spider reaches a “stop” strand that its foot enzymes can’t cut. Mission accomplished.

Stojanovic says his next goal with this spider will be to increase the number of steps it can take and program more complex movements into the origami. He also wants to design spiders that can link together and cooperate on a task. Spiders might also read each other’s trails, he says, the way ants or other social insects do.

One day in the distant future, these spiders may be able to crawl around on cell membranes, recognizing diseased cells and helping destroy them.

“That’s a dream, not something that is around the corner,” Stojanovic says.

For that dream to become a reality, the crawlers would need to graduate from their artificial DNA tracks and be able to traverse a more natural landscape, such as the surface of a cell. But since cell membranes aren’t covered in DNA, spiders would need to be designed to interact with a different molecule, perhaps an intermediate protein that scientists would insert onto a cell.

“It’s a slow process,” Stojanovic says. “If you ever want to see something like that happening, you have to go through the stage we are going through.”

Burned bridges

One problem with an autonomous spider that cuts up the track behind it is that the origami is used up after only one run.

“If your motors are forever destroying the tracks, you’ve got to rebuild the tracks, which would cost you a huge amount of energy,” says physicist Andrew Turberfield of the University of Oxford in England. “An automobile that chewed up the road behind it would be a bit unpopular.”

Turberfield is working on ways for walkers to move by themselves without destroying their tracks. His team has come up with a two-footed walker that walks along a reusable strand of DNA by flipping over itself, like a gymnast doing handsprings across a mat.

A “fuel” strand added to the surrounding solution raises the walker’s back foot. The walker then flips over and moves a step forward on the track. It can go backward simply by switching to a fuel strand that reacts with the front foot.

This type of nanobot takes its inspiration from kinesin, a natural molecular motor that carries cargo around the cell, Turberfield says. Kinesin’s two feet coordinate so that the back foot is always the one to pick up first and move forward.

“We’re looking at what cells do with motors and are trying to emulate the cell,” Turberfield says. “If you look at biology as inspiration, then you won’t go far wrong.”

Nano assembly lines

Another breed of DNA spider new to the nanotech world does more than walk. It also picks up cargo with three DNA arms.

A group led by DNA nanotechnology pioneer Ned Seeman of New York University has designed a four-legged, three-armed spider that picks up gold nanoparticles from stations along an origami track. The spider can’t walk by itself, but rather requires the scientists to add short, single strands of DNA into the surrounding solution at each step to coax the feet forward.

Researchers embed three stations into the origami. Each station holds a gold nanoparticle wrapped in a single strand of DNA that is complementary to the DNA in the spider’s arms. When a spider stops at a station, one of its DNA arms binds to this DNA leash, grabbing the nanoparticle and plucking it off the origami. As the spider moves away from that station, it carries the new cargo along to the next stop, where it will pick up another nanoparticle.

Seeman compares the spider to a car chassis moving along an assembly line.

Components are added to the spider “like you would add a door, a steering wheel or an engine to a chassis,” says Seeman, whose spider work also appears in the May 13 issue of Nature.

After two more stops, the spider might have up to three particles in its arms. But there may also be just one or two. That’s because the stations can be programmed to either give up or keep their cargo. Using the same track, the spider may pick up different combinations of nanoparticles.

In the future, Seeman says, a single assembly line may be able to work with many more than three building blocks. Such longer assembly lines could build more complex objects. He also plans to make these spiders autonomous so they can do the work without the scientists having to add new DNA strands for each move.

Finally, he wants to try to assemble molecules that will bind together into complexes, rather than make things that ultimately have to be held together by the DNA. He thinks spiders could pick up individual molecules at each station that would bind to each other. By bringing together molecules one at a time, spiders could fit together molecular puzzle pieces that don’t react well together in nature. This could be a help to chemists, he says.

“What we do right now in virtually all chemistry is throw in a bunch of things in the pot, and swirl them around so they collide with one another,” Seeman says. “In principle we could do reactions more easily.”

Crawling into the future

Researchers concede that DNA spiders can’t do anything useful yet, and most scientists are reluctant to project too far into the future about what these nano­robots could eventually do.

“It’s very appealing to be able to picture something that’s a billion times smaller than a human that can move,” says Caltech biological engineer Niles Pierce, who has done work on DNA walkers. “But to take that locomotion and put it to productive use for fabrication of nanoscale components, that’s still a futuristic goal.”

Yet engineers should continue to experiment with these nanomachines to pave the way for novel discoveries in nanorobotics, Pierce says. “The stuff you would learn along the way pays dividends in other areas.”

And the hope remains that decades down the road, swarms of DNA spiders might be deployed in people’s bodies searching for telltale signs of cancer or disease. Those spiders would signal to each other where to find the troubled tissue and work together to dive-bomb their targets with sacs of medicine. Such targeted anti­disease missiles could avoid the side effects that occur when drugs pervade the body and affect other tissues.

Likewise, DNA crawlers with many arms might skitter across origami assembly line factories, grabbing nanoparticles one by one and assembling them in a precise order. Nanospiders might build nanosized computer chips that cram more memory, power and speed into smaller and smaller spaces. Or perhaps arrange nanoparticles into new configurations and make new metamaterials for cloaking devices.

“One of our goals is to make DNA and protein systems that are as complex as a living cell,” says Rothemund. “But to get there, we’re going to have to make our DNA nanotechnology systems hundreds of times, thousands of times more complex.”

Research labs tinkering with DNA devices increase in number every year as scientists realize how they might harness DNA’s potential, he says.

“It’s gotten 10 times cooler in the last 10 years, and it got 10 times cooler in the 10 years before that,” Rothemund says. “At every stage, I think it will become more and more compelling.”