What if you could engineer biology this way? What would you build? Physicist and scientific prophet Freeman Dyson would love to build genetically engineered pets and ornamental plants. Standford biologist Drew Endy envisions a collection of standardized biological parts called BioBricks, off-the-shelf modules that biological engineers can assemble like snap circuits into amazing biological machines. An annual undergraduate competition, the International Genetically Engineered Machine competition draws teams of biogeeks who design glowing microbes that spell "Hello World" on an agar plate and gut bacteria that smell like mint or bananas.
This all sounds exciting, but what's the reality? Do biological engineers, or synthetic biologists (as they are most commonly called) have anything close to the know-how of today's electrical or aerospace engineers? The answer, obviously, is no.
Bioengineering Today is Falling Short
Dr. Pamela Sliver, from Harvard's Department of Systems Biology, visited us here today at Washington University in St. Louis. Dr. Silver is, among other things, a synthetic biologist, eager push the limits of our biological engineering capability. Her lab recently designed a genetic memory circuit that can remember a past event - a biological signal.
It works like this: a signal flips on gene A, which in turn switches on gene B. Gene B, once on, acts on itself in a positive feedback loop and thus stays on. In other words, once A switches on B, B stays on, even if the original signal that induced A is removed.
One application of this circuit is monitoring what happens to cells when their DNA is damaged (a key part of cancer research): imagine that the signal to induce gene A is the temporary signal generated when DNA is damaged, and that gene B produces a green fluorescent protein. Thus, if a cell's DNA is damaged, the circuit is activated and the cells glow green, even after the DNA damage signal stops. The circuit 'remembers' that at one point, DNA was damaged, and using a microscope, you can follow the behavior of the green-glowing cells that experienced a DNA damage event.
This memory circuit works well, but this research is also a sign that the state-of-the-art hasn't changed much in almost 10 years. Synthetic biologists, for the most part, are still putting a lot of effort into making extremely simple parts that, so far, don't have much wider applicability.
Even more telling is the fact that, while undergraduates get together every year and build fun little biological widgets from a set of standardized parts, there are virtually no professional biologists out there who do this. We still can't engineer from scratch something useful that requires anything more than the most simple genetic feedback loops.
Biofuels by Tinkering
Nowhere is our lack of engineering ability more evident than in the effort to make designer microbes that generate biofuels. Imagine photosynthetic bacteria, harvesting energy from the sun and producing some sort of fuel like hydrogen or ethanol. Biological engineers would love to design microbes that would make this process cost effective, and if they were successful, cheap biofuels could transform our energy economy.
We would love to engineer these biofuel-producing bacteria, but in truth, biologists right now are tinkerers, not engineers. Like Edison, who tinkered with dozens (hundreds?) of filament types before successfully inventing a long-lasting light bulb, today's bioengineers are working by trial and error. Edison was a phenomenal tinkerer, but his methods could not produce most of today's engineering marvels like microchips, Mercedes engines, and Boeing 787s. Engineers today have a well-developed theoretical basis for their work - they can plug some equations and parameters into a computer and produce incredibly complex designs before ever setting foot in a machine shop. Bioengineers today are more like the Wright brothers than like 21st century engineers.
Here we get to the greatest challenge faced by synthetic biologists: they have no strong theoretical basis for their designs, and it's not clear yet what that theoretical basis should be. Today's electrical engineer has a set of solid physical laws for manipulating one thing - electricity. With electricity you can work wonders. But what is the biological engineer's equivalement of electricity? There is no biological Ohm's law. A biological signal is a much more nebulous (and non-quantitative) concept than an electrical signal. As of yet, there is no mathematical framework or formalism in biology that can match the physics-based framework for engineering.
Maybe the lesson here is that we shouldn't carry analogies between electrical engineering and biology too far. Synthetic biologists, instead of looking to other engineering fields for inspiration, need to rethink what engineering means in biological terms. What knowledge do we need to design a biofuel-synthesizing organism or any biological system from scratch? Can we do this with a set of standardized parts that are snapped together (PDF), or is it better to strategically rewire existing systems produced by evolution? Biologists still need to figure out how best to model, talk about, and predict the behavior of biological systems. The future of bioengineering is not here yet.
Front page image: The Repressilator, by Michael B. Elowitz&Stanislas Leibler, Nature. 2000 Jan 20;403(6767):335-8, copyright Nature Publishing Group.