This printer is made of 600 strands of DNA that put themselves together

If the molecules put themselves together as intended, the process can then be scaled up millions or billions of times. But if they don’t, it’s back to the drawing board.

After years of R&D, Benson and his colleagues have figured out exactly how to make a remarkable mini-machine. Their printer is “capable of moving a central printhead in two dimensions on top of a surface, and it’s completely constructed from DNA,” Benson says.

The research is described in a paper published Wednesday in the peer-reviewed journal Science Robotics.

DNA is a nanoscale building material 

“We love working with DNA,” Benson says. “It’s quite a simple molecule, just a string of these nucleotides.”

Those nucleotides — smaller molecules that join together to form RNA and DNA — are composed of a sugar molecule (either ribose in RNA or deoxyribose in DNA) attached to a phosphate group and a nitrogen-containing base. In DNA, the bases are adenine (A), thymine (T), guanine (G), and cytosine (C) (in RNA, the thymine is replaced with uracil (U)).

The A and T molecules are attracted to each other, and so are the G and C molecules. In living organisms that use DNA to communicate genetic information, nucleotides are the basic building block of these molecules. They are assembled by the cell one at a time and strung together by the process of replication (in DNA) or transcription (for RNA). 

The DNA and RNA are then, in turn, used as instructions for building proteins and directing other activities within a cell.

But that biological function isn’t an inherent property of DNA itself. A gene — a distinct sequence of nucleotides — can only cause a particular biological process to occur under very precise circumstances. Divorced from its biological support system, “it’s just a molecule,” Benson says. “It’s like a puzzle that’s also magnetic, so the pieces can find each other if you design it the right way.”

He says that’s not quite how the body uses DNA, but there are some resonances between how DNA works in its original environment and how researchers are using it in the lab. “Our bodies are composed of all these fantastic proteins that are built in a similar way, by folding linear polymers into more intricate shapes,” he says.

The idea behind the printer is fairly simple

The printer described in the paper is built from roughly 18,000 base pairs of DNA. The researchers didn’t start with such an ambitious goal, though. Their earlier work focused on a linear motor that could be used to drive a hydraulic arm.

“We really drew inspiration from the macroscale,” he says.

Once they’d figured out how to work on that more extensive scale, the researchers started on the more ambitious project. They realized that if they were to “take several linear motors, place them together, and control them independently, [they could] build something that has a complex function,” Benson explains. “We managed to combine these linear motors in a way where they can operate independently and sort of orthogonal to each other.”

The machine is fairly simple. A printhead moves along a crossbar, which moves along two perpendicular rails. “The core idea is to move a central piece in two dimensions at the nanoscale. And then we’ve added functionality to the central piece to make it function as a printhead to pattern a surface below, all through DNA interactions,” he adds.

Fabricating from DNA is a long process 

While the products that DNA-based machines might one day make are complex, Benson says DNA itself is pretty easy to work with, at least in principle.

If a scientist were to need a specific protein, “there’s probably no one in the world who can make it for you because it’s so complicated to understand the interactions,” Benson says. “But DNA is so simple you could design the shapes by hand.”

In practice, they use software to sketch up their DNA designs and see how they’ll behave in the real world. “Then we can do basically hand-tuning,” Benson says. “We say okay, this bit seems a little bit bent and this bit seems a little bit long. Then we change the design, put it back into simulation, and see what it looks like,” he says.

The final product isn’t a blueprint, though. It’s “basically a list of DNA sequences,” Benson says. Most of them are roughly 60 nucleotides long. A complex machine like the printer is made of 600 or 700 sequences, he says.

A supplier turns the digital file into DNA, which is delivered with each strand in its own well on a plate. The researchers mix them together under specific saline conditions and at high temperatures.

“The heat breaks up everything, no DNA is hybridized,” he says. “You can’t just mix them and put them on a bench at room temperature because you’re not going to get what you want … at low temperature, sequences that you’ve designed not to be complimentary are going to be a little bit complimentary. It’s going to be a mess probably.”

After mixing the strands at relatively high temperatures, the researchers slowly cool the concoction over the course of several hours. “When you slowly move down, you favor the correct base-pairing interactions. And that’s why self-assembly works,” says Benson.

For a complex assembly like the printer, different pieces have to be allowed to self-assemble separately. Those components are then mixed together and allowed to self-assemble into the final product.

Then it’s showtime. “We can put it into electron microscopes and super-resolution light microscopes and see it works,” Benson says.

This technology could revolutionize manufacturing… but it’s not there yet

Manufacturing complex molecules is a big business, and it’s vital to making pharmaceuticals and other products. Right now nearly all of that manufacturing relies on what Benson calls “hard robotics.”

“There’s a big line of trying to manufacture things in electronic ways, by lithography and even moving individual atoms by really heavy instrumentation. The resolution there is pretty good,” he says.

“We are trying to develop a parallel track manufacturing, which is based on soft robotics and more bio-inspired techniques,” he says. Benson describes his group’s work as the road less traveled. “I can’t really think of that many papers doing soft robotic manufacturing in this style… We are more similar to how the system works in nature, like the ribosomes that can sort of synthesize complex machinery,” he says.

Right now, Benson’s team can’t compete with conventional manufacturing approaches. “I wouldn’t say we’re really close to [having products] yet,” but the approach has a major advantage.

“We don’t have one printer making one thing. We aim to do billions of copies of whatever we do,” he says.

“Printing molecules would be our dream goal when the technology matures a little bit. I would imagine having some sort of template. Then you would go in and add functional groups to different sides by moving a printhead. Then, of course, you would need millions or billions of copies of the same molecule,” he says.

“That’s where I think the future of the technology is.”