Over many posts now we have discussed the definition and impact of synthetic biology, and also where and how technological surprise in biology might arise. I now want to bring together and summarize a few strings of thought in one place. First, the genome is the unit of engineering. Not parts, not circuits, but the genome as a whole. In this regard, systems biology eats synthetic biology. Similarly, we’ve also seen that technology surprise may arise not from assemblies of known parts, but from the unknown: for example, from random proteins interfacing with systems, or perhaps from the largely unplumbed depths of information and function that are being revealed by metagenomic analyses. Again, the real threat is not from so-called synthetic biology, but from systems biology.
I think that at root this is a biomimetic analysis. Where does new function arise in biology? Sometimes it arises from novel assemblies of pre-existing parts, and from the construction of interesting new circuits. It can be argued that yeast lysine biosynthesis was a great achievement in synthetic biology, in its day. Almost every other organism makes lysine via the diaminopimelate pathway, but not yeast. In some ancient catastrophe or adaptation, yeast lost the diaminopimelate pathway, and instead devised a new route to lysine by duplicating parts lying around the genome: citrate synthase and isocitrate dehydrogenase became homocitrate synthase and homoisocitrate dehydrogenase, ending up at keto-adipate. From there it was just a reduction and a few simple transaminations to lysine.
But I would assert this is the exception rather than the rule. In general, when micro-organisms acquire new function they acquire it from other micro-organisms, by horizontal transfer. Indeed, bacteria seem to have whole swaths of their genomes that serve as landing pads for new function, whether it be pathogenicity islands or CRISPR loci.
If we accept this analysis, then the question becomes: how do we engineer via horizontal transfer? Now, that’s sort of a silly statement, because what are plasmids, if not horizontal transfer? We’ve been engineering by horizontal transfer for a good long time now. But even with facile plasmid construction methods, like Gateway vectors, this engineering has been decidedly low throughput. This has led me to think about how we might come up with a new approach, a new method, which I am nominally (and in a spirit of political uncorrectness) calling miscegenesis: the breeding of everything with everything. Such a method would go well beyond genome shuffling, which to date has involved a relatively limited number of genomes. It would be horizontal transfer on steroids, so to speak, with organisms freely and continuously exchanging information.
Clearly biology does this over long evolutionary timescales, but how could we do this quickly? Homology-based methods are almost certainly out; site-directed recombination presents a fairly steep barrier to organisms even a few steps away in the tree of life.
I have some ideas along these lines. Indeed, I woke up a few nights ago not from a dream of a snake eating its own tail, but rather from a dream about pirates. You know those great old (and continuing) cartoons in TIBS, that sort of showed how scientists anthropomorphized molecules and processes? It was like that, a dream of molecular piracy. What kind of vector could I make that would steal the genes from another organism? Viruses do this, of course, but again in a sort of inefficient way. Viral particles will often misload random mRNAs into themselves, and these escape capsules may reach another organism and be reverse transcribed into the genome, but again, slow, slow, slow. No, I want a sleek vector vessel that sails into its host and grapples the genome, grabbing a chunk and chortling “Yo ho ho!” as it sails out again. Indeed, in our lab conversations we have begun to call this vector “The Grapple.”
We haven’t done it yet, but the blueprints are laid in, and the minions are conjuring. A first order, single-step Grapple is within our grasp, I believe. The more ambitious continuous Grapple, that truly does sail between organisms, continuously grabbing and dropping off chunks of genetic information is still just a bit beyond the horizons of my imagination. But I am encouraged by biology, which does the oddest things in the oddest ways. I am encouraged by old literature, from a time before restriction enzymes, when microbiologists mouth-pipetted their samples and liked it, by God. I am encouraged by a Fata Morgana that in pure evolutionary terms should probably not exist, a selfish element that has learned to do unselfish things. I am encouraged by the F’.