As a medicinal chemist, I spend a good amount of time feeling inadequate in the face of biomolecules. Just a look at the lists of best-selling pharmaceuticals these days is enough to bring that on, considering the number of biologic products on them. But business considerations aside, there are plenty of other good reasons for feeling that way. A good start is comparing my chemical abilities, or anyone else’s, to what can be accomplished by living systems.

Enzymes are the first candidates that come to mind, of course, and the quest to make them for ourselves continues. Various groups have shown that it can be done, if you’re willing to spend enough time and money modifying a known starting point, but coming up with real enzyme-like activity from scratch remains beyond anyone’s reach. Looking at what they can do, though, brings up two nearly contradictory thoughts. The first one, “Man, those catalytic activities are really impressive” is the standard reaction, and fair enough, but the second is “Seems like they could do a lot more than that, though, doesn’t it?” And while that may seem a bit arrogant (or delusional), think about it a bit.

What we’re seeing, when we look at the existing world of enzyme activity, is the residue of a billion years of evolution. Blind tinkering, vast incomprehensible amounts of it, has given us some extraordinary things. But there are many other things that we’d like to be able to do that evolution has apparently never taken any interest in. Take a field like transition-metal catalyzed reactions, for example, which synthetic organic chemists have been exploiting for years.

There are more metal-driven coupling reactions than anyone can possibly keep track of, and any given transformation can probably be optimized somehow if you’re just willing to spend enough of your life doing it. Things like the Suzuki reaction are wonderfully useful—so wonderfully useful that we chemists have probably cranked out too many of its products for any screening collection’s best interests. A colleague of mine has half-seriously proposed a cap-and-trade system for aryl couplings like this. Every chemist in a department would get a set number of “Suzuki coupons” per year, and if you use yours up, you have make a deal with someone who hasn’t. (Resistance to this idea would probably be fierce).

You Don’t See Suzuki Couplings in Nature, Do You?
But you don’t see Suzuki couplings in nature, or if you do, I certainly haven’t heard about it. None of the palladium-catalyzed reactions seem to be found in living systems, although they could surely be useful. Palladium just isn’t an element that any living creature has come to depend on, and none of them seem likely to start doing so any time soon. But there’s no intrinsic reason, I presume, that living systems couldn’t use it. And synthetic biology might be the way to force the issue, via the rapidly-moving fields of genomic-code expansion and artificial organisms. One could certainly imagine an enzyme active site with a catalytically active Pd atom in it, coordinated with a few amino acids. There must surely be redox couples that could regenerate the active species as it turns over.

Palladium Surprises
Give some microorganisms sufficient palladium in their growth media and a permissive attitude towards point mutations, and we might be surprised at what turns up. We might want to change the system by engineering in some benefit to being able to produce biaryls, just to stack the deck a bit. Other metals might be exploited in similar ways—gold, silver, iridium and others all have interesting and unusual catalytic properties, and they’re biologically silent (to the best of my knowledge). No microorganism that tried to survive on iridium ever lasted very long out there in the real world, presumably, but we don’t have to work under those limitations.

Past metal-catalyzed couplings, there are many other reactions that just don’t get a hearing in living cells. The Diels-Alder is a leading candidate—there are a few out there in biology, although my impression is that for each example you can probably find someone who’s ready to argue that it’s not a “true” Diels-Alder. But there’s no doubt that it’s not a widely used reaction in cells, for reasons that can only be speculated on. Other pericyclic reactions are in the same category: dipolar cycloadditions, photochemical 2+2 reactions, and so on. Life at the molecular level seems to be carrying on fine without recourse to these, but if we were somehow able to restart biochemical evolution, would that still be true? And is there any reason that such things couldn’t be spliced in now?

Talk of genetic code expansion leads to thoughts of non-natural amino acids, and the possibilities here are limitless. The existing collection is probably semi-random. Amino acids themselves are formed by all sorts of abiotic reactions (as witness their presence in things like carbonaceous chondrite meteorite fragments). But there are far more different structures in such material than the ones that life on Earth uses today, so at some point the existing list apparently just sort of settled into general use. While life has made the most of that collection, different amino acid side chains could surely enable reactions that just aren’t touched under current conditions. Just the set of potential modified phenylalanines is vast enough on its own. I’m imagining some sort of arylphosphine side chains to help coordinate those catalytic palladium atoms.

Using chemistry to “cheat” is OK
We have the ability to cheat in getting these things accomplished, naturally. We get to define “fitness” any way we want, rather than the traditional “Survive, breed, or die out.” And we don’t have to require that organisms handle these new pieces all the way through their cellular machinery. If our new catalytic wonder yeast needs (say) perfluorophenylalanine as an essential nutrient, then fine. That certainly wouldn’t fly in the world outside the fermentation tank, but the catalytic wonder yeast isn’t going to have to deal with that (it better not, actually). And it won’t have to deal with competition from other organisms, either, although setting up such competitions artificially might yield some interesting results.

What all these ideas have in common is that they require us to get our hands on the higher-level routines of biochemistry, rather than the primary ones. Instead of laboriously stitching in a bizarre catalytic site by hand, we need to understand protein synthesis well enough to turn these things out by the uncounted thousands (since most of them won’t work, you know). Instead of trying to make our own tools, we need to devote that effort into hacking the tool-making machinery itself (which is just where a lot of work is going these days).

If I can wax philosophical (OK, more philosophical), the history of the human race has been its unwillingness to accept the hands that it’s been dealt. Intelligence itself is the root cause of that, and it leads to all the things we think of as human. Building shelters instead of finding them, using fire, making tools, clearing land and growing plants—by building synthetic organisms and optimizing their activity, we are (for better or worse!) carrying on a very old tradition indeed. Our ancestors optimized corn and other crops beyond recognition and domesticated animals, all without a fraction of our capabilities, so we have quite a heritage to live up to. What do you think?  Please write in with your opinion.