Smart DNA: Programming the Molecule of Life for Work and Play

Scientific American takes a look at the possibilities of programming DNA. While there are many possibilities that can be beneficial to humans, there might also be some issues where we begin to "select" who we are at the molecular level. Some serious ethics issues will arise. For now, however, there seem to be some great health uses.

Smart DNA: Programming the Molecule of Life for Work and Play

Logic gates made of DNA could one day operate in your bloodstream, collectively making medical decisions and taking action. For now, they play a mean game of in vitro tic-tac-toe

By Joanne Macdonald, Darko Stefanovic and Milan N. Stojanovic

Tic-tac-toe-playing-computer consisting of DNA strands in solution demonstrates the potential of molecular logic gates. Jean-Francois Podevin

Key Concepts

• DNA molecules can act as elementary logic gates analogous to the silicon-based gates of ordinary computers. Short strands of DNA serve as the gates’ inputs and outputs.
• Ultimately, such gates could serve as dissolved “doctors”—sensing molecules such as markers on cells and jointly choosing how to respond.
• Automata built from these DNA gates demonstrate the system’s computational abilities by playing an unbeatable game of tic-tac-toe.

From a modern chemist’s perspective, the structure of DNA in our genes is rather mundane. The molecule has a well-known importance for life, but chemists often see only a uniform double helix with almost no functional behavior on its own. It may come as a surprise, then, to learn that this molecule is the basis of a truly rich and strange research area that bridges synthetic chemistry, enzymology, structural nanotechnology and computer science.

Using this new science, we have constructed molecular versions of logic gates that can operate in water solution. Our goal in building these DNA-based computing modules is to develop nanoscopic machines that could exist in living organisms, sensing conditions and making decisions based on what they sense, then responding with actions such as releasing medicine or killing specific cells.

We have demonstrated some of the abilities of our DNA gates by building automata that play perfect games of tic-tac-toe. The human player adds solutions of DNA strands to signal his or her moves, and the DNA computer responds by lighting up the square it has chosen to take next. Any mistake by the human player will be punished with defeat. Although game playing is a long way from our ultimate goals, it is a good test of how readily the elementary molecular computing modules can be combined in plug-and-play fashion to perform complicated functions, just as the silicon-based gates in modern computers can be wired up to form the complex logic circuits that carry out everything that computers do for us today.

Dissolved Doctors
Near the end of 1997 two of us (Stojanovic and Stefanovic) decided to combine our individual skills in chemistry and computer science and work on a project together. As friends from elementary school in Belgrade, Serbia, we happened to be having dinner, and, encouraged by some wine, we considered several topics, including bioinformatics and various existing ways of using DNA to perform computations. We decided to develop a new method to employ molecules to compute and make decisions on their own.

We planned to borrow an approach from electrical engineering and create a set of molecular modules, or primitives, that would perform elementary computing operations. In electrical engineering the computing primitives are called logic gates, with intuitive names such as AND, OR and NOT. These gates receive incoming electrical signals that represent the 0s and 1s of binary code and perform logic operations to produce outgoing electrical signals. For instance, an AND gate produces an output 1 only if its two incoming inputs are both 1. Modern-day computers have hundreds of millions of such logic gates connected into very complex circuits, like elaborate structures built out of just a few kinds of Lego blocks. Similarly, we hoped that our molecular modules could be mixed together into increasingly complex computing devices.

We did not aim, however, to compete with silicon-based computers. Instead, because Stojanovic had just finished a brief stint with a pharmaceutical company, we settled on developing a system that could be useful for making “smart” therapeutic agents, such as drugs that could sense and analyze conditions in a patient and respond appropriately with no human intervention after being injected. For example, one such smart agent might monitor glucose levels in the blood and decide when to release insulin. Thus, our molecular logic gates had to be biocompatible.

Such molecular modules could have innumerable functions. For instance, in diseases such as leukemia, numerous subpopulations of white blood cells in the immune system display characteristic markers on their cell surfaces, depending on the cells’ lineage and their stage of development. Present-day therapies using antibodies eliminate large numbers of these subpopulations at once, because they target only one of the surface markers. Such indiscriminate attacks can suppress the patient’s immune system by wiping out too many healthy cells, leading to serious complications and even death. Molecular modules capable of working together to sense and analyze multiple markers—including performing logical operations such as “markers A and either B or C are present, but D is absent”—might be able to select the specific subpopulations of cells that are diseased and growing out of control and then eliminate only those cells.

Go read the whole article.

Tags: Smart DNA, Programming the Molecule of Life for Work and Play, biology, genetics, health, Scientific American, Joanne Macdonald, Darko Stefanovic, Milan N. Stojanovic