DNA molecules can be used as building elements of molecular robots. So far, however, only simple functions have been achieved with such DNA robots, for instance, walking in a controlled fashion and picking up nanoparticles. Now Lulu Qian and colleagues from Caltech have developed a DNA robot capable of performing more complex cargo-sorting tasks (Science 357, eaan6558; 2017) .
The cargo-sorting DNA robot is realized using a simple algorithm and three molecular building blocks. It is composed of a single-stranded DNA walker, one arm and one hand. The DNA walker consists of two foot domains of 6 nucleotides each and one leg domain of 15 nucleotides. It is capable of a random walk along a track of single-stranded extensions on a DNA origami surface via a reversible strand-displacement reaction, exploring the entire DNA origami surface. The arm and hand domains can pick up cargoes from randomly distributed locations, and drop them off at specified destinations. Once the cargo is dropped off, the DNA robot is free to visit other locations on the DNA origami surface and pick up new cargoes. This occurs repeatedly until all cargoes are sorted. The rigidity of the DNA origami, the DNA sequence of the foot domains of the walker and the purity of the DNA origami are crucial for the successful experimental implementation of the DNA robot.
This DNA robot can also execute two distinct cargo-sorting tasks simultaneously, though the task completion level is sacrificed by 16–18% compared to only sorting one type of cargo. Furthermore, the researchers have shown that more DNA robots can work cooperatively to speed up the cargo sorting. The DNA robot performs an average of 296 steps to sort cargoes, which is between one and two orders of magnitude larger than previous DNA robot designs.
In principle, this system can be generalized to sort multiple populations of cargoes of various chemical molecules, nanoparticles and proteins from arbitrary locations without any energy supply. Therefore, it may find potential applications in chemical synthesis, molecular device fabrications and programmable therapeutics. However, the speed of the DNA robot needs to be further improved and more programming trials will be required for molecular robots to carry out more sophisticated tasks in smarter ways.