Enzymes do things we can’t do. That’s one of the facts of life in organic synthesis, and it’s likely to be true for quite a while. (We can do things that enzymes can’t do, to be sure, but in many cases that’s probably because enzymes have never yet seen a need to do them). Cells, most especially, can do chemical transformations that we can’t do, because there the various enzymes are in their happy native environments, and entire enzymes systems are assembled and ready to get to work.
Biotechnology takes great advantage of this through recombinant cells, and any number of proteins (and protein-based drugs) are produced via live cell culture. Small molecules, not so much. The protein-making machinery is mighty and ubiquitous, but the small-molecule-making pathways, once you get away from the common molecules of life, tend to be little bespoke cottage industries inside a cell. Some natural-product-based compounds and starting materials are made this way (antibiotics, the starting material for taxol, etc.), but even these are taking advantage, for the most part, of specific pathways that the cells already had up and running, rather than having them make anything different.
You can always go in and try to modify and engineer enzymes, but this is already enough of a project with isolated enzymes, much less in living cells. An easier way to extend the concept is to give the cells other substrates and see what they make of them. Some interesting “unnatural products” have been made this way, but a more common use is to take advantage of a particular enzyme pathway, in most cases a chiral reduction. Live yeast has been a favorite system for such reactions for many decades now, and it’s one of those transformations in the “When it works, it works” category. When it doesn’t work, you can always try a different strain of yeast, and plenty have been developed over the years. You can buy collections of them, and collections of some of the isolated enzymes to see what might work for you.
The ideal would be to do such screening in as close to a single-cell manner as possible. That would give you some variety in screening, but it’s a significant technical challenge – isolating the cells, keeping them happy, getting the substrate to them, analyzing the products on such a small scale. Here’s a paper that’s working towards that goal, though: they’re work down to the several-hundred-cell level, using E. coli cells engineered with an Aspergillus enzyme. It’s done in an integrated microfluidic chip, with an ingenious built-in electrophoresis step, and a deep-UV fluorescence detector. None of these would be your first choice if you were doing the reaction in a flask, of course, but this is just the sort of thing that’s going to have to be worked through on the way to the single-cell world, and I’m glad to see people making the effort. You could imagine, eventually, growing up a batch of some test organism with an induced high mutation rate, screening huge numbers of individual cells, and quickly arriving at a good enzymatic starting point. Speed the day!
I can’t talk about yeast reductions without telling a story, though. Back when I was in graduate school, we had a Japanese post-doc in the lab who needed to do one of these on pretty large scale to get a chiral starting material. He had, though, apparently never done a yeast reaction, and had probably never seen yeast in action at all. At least, that’s what we gathered from what happened. I was down in that lab using the analytical balance (the only one for the whole group – those were the days), when I heard our normally quiet Japanese guy suddenly start making plenty of noise. Unless I missed my guess completely, those sounded like vigorous Japanese curses and sounds of alarm.
What he’d done, it turned out, was to go get a four-liter Erlenmeyer, dump in a goodly heap of dry baker’s yeast, pour some lukewarm water into it, and then give it a good slug of sugar to really get it going. It got going. When I came around to his hood, the post-doc was staring in dismay at a yeast volcano: an apparently endless stream of tan foam was swiftly overflowing the Erlenmeyer and spreading out into his hood. Still shouting in displeased Japanese, he located a bucket and put the flask into it, but that was only a holding measure. The foam kept on coming, and was obviously going to fill the bucket before finally winding down. By this time, he’d attracted a crowd of appreciative onlookers – the last thing he wanted, I’m sure – and we were encouraging him with helpful suggestions like adding more sugar, putting some green food coloring in the flask and calling the campus newspaper, and so on. He finally groaned, poured the lot down the sink, and went home.
A few days later, though, he spoke at group meeting, and had apparently gotten over the trauma. “I try yeast reduction”, he told he group, grinning. “But yeast is very dangerous.” He reached into a bag he had with him, and pulled out a dinner roll: “However, it makes very good bread!”