Reactions

This molecule is called L-DOPA, and it is the most effective treatment we have for Parkinson's disease.

This molecule is called D-Dopa, and it does nothing for Parkinson's and causes a potentially deadly side-effect called granulocytopenia.

Now, it turns out that it is really easy to set up a chemical reaction that will make both of these molecules at the same time, but making just L-DOPA without D-DOPA, that was so hard that the person who did it won the Nobel Prize.

How did the first hammer get hammered in a world without hammers?

Is it turtles all the way down?

So now I'm gonna explain why these two are in fact different molecules.

If you already know about chirality, just skip ahead to the next section.

Now, the confusing thing here is that these look so similar.

In fact, they have the same boiling point, melting point, and almost every other chemical and physical property that you can think of is the same.

In fact, every atom in each of these structures is connected to the same other atoms.

But they are, in fact, different, because the atoms of these two molecules occupy different positions in 3D space.

But these molecules are related to each other, and they're related to each other in an interesting way, which I will show you by taking this mirror out of my microwave and using it to display L-DOPA's reflection.

The real molecule you're looking at, that's L-DOPA.

The molecule you're seeing in the mirror, that's D-DOPA.

So L-DOPA and D-DOPA are mirror images of each other, and the atom that's responsible for making those two molecules mirror images of each other is this one.

But the thing is, these mirror image molecules are not super imposable, and chemists call molecules that are non-super imposable mirror images of each other enantiomers.

And you can see that most clearly if I try to overlap them on each other.

So I'm gonna take this one and rotate it behind this one.

And you can see this oxygen lines up with this oxygen, this hydrogen lines up with this hydrogen, and so on all the way through the molecule, until we get to this carbon right here.

On this one, the nitrogen sticks out towards you.

On the back molecule, the nitrogen sticks out towards me.

And there is no amount of rotation or flipping or anything that I can do in three dimensions that will turn this front molecule into this back molecule.

These atoms will never overlap.

And so these in fact are two different molecules.

A 50-50 mixture of two enantiomers used to be considered a single drug, one drug, and some drugs still are this way today.

For example, ibuprofen is a 50-50 mixture of enantiomers.

And by the way, a 50-50 mixture of enantiomers takes a long time to say, so chemists call these kinds of mixtures racemic mixtures.

So that is the situation back in the 1960s when this paper in "The New England Journal of Medicine" comes out.

Now, this describes a small study that enrolled 21 patients with Parkinson's disease, and the doctors start off by giving the patients a racemic mixture of 50-50 L-DOPA and D-DOPA, and they see an incredible response to the treatment.

But they also see a problematic side-effect called granulocytopenia, which basically makes you more susceptible to infections, some of which can be dangerous or even life-threatening.

So the doctors decide to switch to the much harder-to-get and much more expensive pure L-DOPA, and what they see is that the granulocytopenia disappears.

So they conclude that the L-DOPA is responsible for the treatment, and the D-DOPA is responsible for the side-effect.

And this is great news.

Let's just give the patients pure L-DOPA.

The problem is, back then, you could not make pure L-DOPA at scale.

There were two options, and both were bad.

Option one was to get it from plants.

But if you wanna treat millions of patients with Parkinson's disease, that is not necessarily the most economically viable possibility.

Option two is to just go ahead and make the racemic mixture, make a 50-50 mixture of L-DOPA and D-DOPA, and then isolate the L-DOPA.

Let's take a closer look at option two.

The main chemical step involves a molecule derived from vanillin.

And I should mention here that I am simplifying a little bit.

The real molecule contains some protecting groups, which we don't need to show because there's a real important point to this.

If you take this and run it over a palladium catalyst in the presence of hydrogen, the hydrogen will add to this double bond right here.

But the problem is it can add from the top or it can add from the bottom.

Coming in from the top gives you L-DOPA, and coming in from the bottom gives you D-DOPA.

And because this molecule is flat, the hydrogen is equally likely to come in from the top as it is from the bottom, so you're equally likely to create L-DOPA as you are to create D-DOPA.

And that means that the product of this reaction is gonna be a 50-50 mixture of enantiomers, L-DOPA and D-DOPA.

And that means that if you wanted to make a hundred tons of L-DOPA, then you would also have to make a hundred tons of D-DOPA, a completely useless byproduct, and that is 50% waste right off the bat.

But it gets much worse.

Your L-DOPA and your D-DOPA are completely mixed together, completely intermingled in whatever giant metal tank you did the reaction in, and you cannot separate them using traditional purification techniques that chemists love to rely on, like simple chromatography, recrystallization or distillation.

In fact, the only way to separate these two molecules is by having them interact with a single enantiomer of some other molecule.

And that is key.

It has to be a single enantiomer.

You could take that single enantiomer and react it with both of these molecules to form two new compounds.

Now, the important thing to note here is that these are no longer mirror images of each other, which means they will have loads of different physical and chemical properties, which means you can now separate these two by recrystallization.

And this technique, by the way, is called diastereomeric recrystallization.

Now, once you do that and you have this molecule as a pure compound, you can crack off whatever single enantiomer... Enantiomer, enant... By the way, I've heard people mispronouncing it on YouTube as enatiomer, and that's not right.

It's enantiomer.

Why?

I don't know.

Once you have this molecule as a pure compound, you can crack off whatever single enantiomer you chose and you're left with pure L-DOPA.

Now, in addition to adding a bunch of steps, this ends up using tons and tons and tons of some other single enantiomer molecule, and that means that your overall yield for L-DOPA is gonna be low, maybe a third, maybe less, and that is not a good yield for a commercial drug.

If you were paying really close attention and happen to be a practicing organic chemist, I have no clue if this particular alcohol would react in this way.

I just picked a random chiral alcohol and did it for the purposes of demonstration.

Please don't come at me with a pitchfork in the comments.

And this yield rule, it is not just true of L-DOPA.

Before the chemistry that led to the Nobel was invented, if you wanted to set up a chemical reaction to make one molecule of a pure enantiomer, that meant that you had to throw away two molecules, at least, of other stuff.

Now, if you could figure out how to make L-DOPA catalytically, that would be a commercial hit and a chemical hit.

It'd be a commercial hit because you are making an incredibly necessary drug, and it'd be a chemical hit because you're solving a fundamental problem in chemistry.

And Bill Knowles was the first person to figure out how to do it.

This right here is the heart of the reaction that he discovered.

You start with this, you add this catalyst, and you get this molecule, which, after a quick deprotection step, is L-DOPA.

Now, the yield that Bill Knowles got was 97.5% L-DOPA, and only 2.5% D-DOPA, and that is incredible.

This is the first time that chemists had managed to get a non-enzyme catalyst, we'll come back to enzymes later, to make one enantiomer in such an incredible enantiomeric excess with such high yield.

So now, if you wanna make a hundred tons of L-DOPA, you don't need to waste 200 tons of other stuff.

You just get a hundred-ish tons of your starting material, a few kilograms of catalyst, and that's it.

I mean, some glassware.

You need some glassware.

And like, a big reaction vessel, not glassware, 'cause a hundred tons is a lot.

This is a large...

I'm simplifying a little bit, but you get the idea.

Pretty soon after his discovery, pharma companies started making single enantiomer drugs, as opposed to the racemic mixtures of the past.

And then, in 1992, the FDA published official guidance, saying that you could no longer consider two different enantiomers to be the same drug.

And today, only 5%, 5%, of all new drugs are racemic mixtures.

So how did Bill's catalyst make L-DOPA without making D-DOPA?

Well, let's look at it.

Okay, so here is a 3D structure of one part of the catalyst that Bill used.

Now, this thing can exist as two enantiomers.

Here's one, and here is the other.

This enantiomer makes L-DOPA, and this one makes D-DOPA.

So the reaction produces only L-DOPA by having only this enantiomer in the reaction vessel.

Here's the thing.

Even though Bill's technique is catalytic and therefore super efficient, it still relies on having one pure enantiomer of something else, just like the earlier technique we saw, diastereomeric recrystallization.

And it turns out that this is a universal inviolable... Inviolable?

You can't violate it, law of chemistry.

If you want to make one enantiomer of a molecule, you need to already have one enantiomer of some other molecule.

And this sets up a real chicken-and-egg problem.

Or to use a better analogy, it'd be like if you couldn't make a hammer without using a mallet, and you couldn't make a mallet without using a sledgehammer, and you couldn't make a sledgehammer without...

So if you need a hammer to make a hammer, how did the first hammer get hammered in a world without hammers?

Is it turtles all the way down?

As always with turtles, you have to just start at the top and work your way down.

So let's do that.

Let's start with L-DOPA.

To make L-DOPA, you need one enantiomer of this catalyst.

To make one enantiomer of this catalyst, you need menthol.

Menthol is chiral, and although there are eight different versions of menthol that could theoretically exist, in nature, there is just one version.

Nature produces just one of the eight versions of menthol.

And if you take that one version of menthol and react it with this molecule, instead of a 50-50 mixture of enantiomers, you get an 80-20 mixture.

Couple more steps, and you've got your Nobel Prize-winning catalyst.

Let us descend one more turtle.

To make just one of the eight possible variants of menthol, you need enzymes.

Now, every enzyme in nature exists as a single enantiomer, meaning its active site, the place where the chemistry happens, will be a very specific shape that only makes one enantiomer of whatever molecule it's making.

But how do you make the enzyme a single enantiomer?

Well, you have to make everything it's built out of, all the building blocks, all the amino acids, single enantiomers.

How do you do that?

Well, all the enzymes that build those building blocks have to also be single enantiomers.

And here, we enter a recursive loop of enzymes and small molecules and enzymes and small molecules that extends billions of years back to the very earliest life.

So one definition of life is the ability to do things, reproduce, stack blocks, run away, kill prey, whatever.

And doing things on the molecular scale means building reliable 3D structures, because something's 3D structure determines what it can or cannot do.

Because 3D structures are so critical to life, if we ever figure out what the very first enantioselective synthesis was, it might help us explain how non-living stuff became living stuff.

I had no idea when I started writing this video that I would end up following an unbroken chain of enantioselective syntheses from the 2001 recipient of the Nobel Prize all the way back to the very beginnings of life itself.

That's a lotta turtles.