00:00:00 Well, where did we start?

00:00:08 Well, our group and the group of Richard Lerner started in 1986 by looking at the notion whether

00:00:13 antibodies could selectively stabilize a transition state.

00:00:17 So let's consider a simple reaction, a hydrolysis of an ester to the corresponding acid.

00:00:24 This reaction proceeds through a rate-limiting transition state, which is a tetrahedral negatively

00:00:28 charged species.

00:00:30 So if you're a chemist, you've learned back in sophomore organic chemistry that the solvent

00:00:35 one should run this reaction in is a polar solvent, because the transition state is a

00:00:40 charged compound, the substrate is a neutral compound.

00:00:44 If one runs this in a polar solvent, one selectively stabilizes the transition state relative to

00:00:50 the ground state.

00:00:51 Well, I'd like to make the argument, in fact, that the immune system is capable of producing

00:00:55 an ideal solvent for the transition state, because after all, the antibody-combining

00:01:00 site can be viewed as a heterogeneous microsolvent that's exactly complementary to the transition

00:01:08 state in terms of steric and its electronic properties.

00:01:12 So the prediction, then, is if one generated an antibody to a transition state, one would

00:01:17 have a very effective and selective catalyst.

00:01:20 In fact, this notion was first tested back in 86 by Richard and myself independently

00:01:26 by looking at a series of rather simple reactions, the hydrolysis of activated esters and activated

00:01:32 carbonates.

00:01:33 Let's consider, for example, the hydrolysis of this nitrophenylcholine carbonate.

00:01:38 This reaction, the hydroxide ion-dependent reaction, proceeds from a neutral sp2 substrate

00:01:43 to a tetrahedral negatively charged transition state.

00:01:47 So the theory, then, in theory, if one generated an antibody that's selectively bound to this

00:01:51 transition state, one would get a catalyst.

00:01:53 Well, in fact, you can't do that, because transition states don't lie in a free energy

00:01:57 well and, therefore, one can't immunize with them.

00:02:01 So we decided to use a well-known trick of medicinal chemistry, and that's the notion

00:02:05 of a transition state analog.

00:02:07 If one substitutes the carbon in this tetrahedral negatively charged species with a phosphorous,

00:02:13 one generates a stable analog of the transition states whose steric and electronic properties

00:02:18 mimic those of the unstable transition state.

00:02:22 So if antibodies are generated to transition state analogs, those antibodies should then

00:02:27 act as catalysts.

00:02:29 Now when we generated monoclonal antibodies against these transition state analogs, in

00:02:34 fact, we did get selective chemical catalysts.

00:02:37 And there are a number of properties of these antibodies I'd like to just overview.

00:02:41 First of all, they showed saturation kinetics.

00:02:43 What do I mean by this?

00:02:44 What I mean by this is the reaction's at least a two-step process.

00:02:47 The first step is a binding step of the substrate in the antibody combining site, followed by

00:02:52 an intramolecular reaction in which the substrate bound to the antibody is converted to product

00:02:58 and released.

00:02:59 Well, what's the significance of these kinetics?

00:03:01 Well, the significance is that we can trade antibody binding affinity for a reduction

00:03:06 in the free energy of activation of reaction.

00:03:08 Second of all, these antibodies show high specificity.

00:03:12 I'll give illustrations of those later in the talk.

00:03:15 They accelerate the rate of the reaction ten to the third to a million times over background.

00:03:20 And consistent with the idea that they're selectively stabilizing the transition state,

00:03:24 they bind to transition state analogs more tightly than substrate.

00:03:27 Well, what's going on in these antibody active sites?

00:03:30 Well, in one case, in the case of the antibodies that hydrolyze the nitrophenylcholine esters

00:03:36 and carbonates, we actually have a crystal structure of this catalytic antibody.

00:03:41 The structure was solved by Davies and co-workers at the NIH.

00:03:45 Shown here is the structure of the antibody binding the transition state analog.

00:03:49 The transition state analog is a species in yellow.

00:03:53 The tetrahedral phosphate is bound via electrostatic interaction with arginine 52H of the heavy

00:04:01 chain shown in red.

00:04:02 So you can see this antibody is, in fact, complementary to the transition state and

00:04:07 this complementarity involves an electrostatic interaction.

00:04:11 I should point out that the tyrosine shown here in green is disposable, plays no role

00:04:17 in the enzymatic reaction.

00:04:18 Therefore, we've used modern protein engineering techniques to substitute this tyrosine with

00:04:23 a histidine.

00:04:24 A histidine, we think, acts as a general base to accelerate this antibody catalyzed reaction

00:04:31 and the mutant protein actually is about tenfold more efficient than the parent antibody.

00:04:36 So you can see that we can use modern molecular biological methods to increase the efficiency

00:04:42 of our antibody catalyst.

00:04:45 Well, what's happened since these early studies?

00:04:48 Well, we've gone on, our group and others have gone on, to show that you can actually

00:04:52 use antibodies to catalyze the hydrolysis of unactivated alkyl esters, in fact, the

00:04:58 stereospecific hydrolysis of unactivated alkyl esters.

00:05:02 Why is this interesting?

00:05:03 Well, this is interesting for a number of reasons.

00:05:06 First of all, this reaction allows you to do a chiral resolution of an ester to form

00:05:11 the corresponding alcohol or acid.

00:05:13 This is a reaction that's of interest to process chemists to form pure chiral intermediate.

00:05:18 Now, given the fact that there is no general chemical solution for stereospecific ester

00:05:24 hydrolysis, process chemists today are starting to use lipases in these chemical processes.

00:05:29 Now, in the absence of a lipase, an enzyme that has the appropriate specificity, I'd

00:05:34 like to argue that you could, in fact, generate an antibody that carries out this reaction.

00:05:38 In fact, Kitazumi has shown recently that one can use antibodies to separate a mixture

00:05:44 of four diastereomeric esters, alkyl esters.

00:05:48 He generated four different catalytic antibodies, each specific for each of the four transition

00:05:53 states.

00:05:54 And using these antibodies, he produced each chiral alcohol with a DE of greater than 97%

00:06:02 and almost quantitative yield.

00:06:05 We've recently applied these notions of hydrolyzing alkyl esters to a problem in therapy, okay,

00:06:13 in cancer chemotherapy, to selective prodrug activation.

00:06:17 As many of you know, 5-fluorodeoxyuridine, shown on this slide, is a potent cytotoxic

00:06:23 agent used in the treatment of certain forms of cancer.

00:06:26 Interestingly, if you esterify the 5'-hydroxyl group, you reduce the toxicity of this reagent

00:06:33 significantly.

00:06:34 So the idea, then, is if one could generate a catalytic antibody that selectively converts

00:06:40 the 5'-ester to the 5'-alcohol at a cancer cell, you would selectively dose the

00:06:47 cancer cell relative to all of the other cells in the body, making a much better chemotherapeutic

00:06:53 regime.

00:06:54 Well, in fact, we've generated antibodies that catalyze the hydrolysis of the valial

00:06:59 5'-ester by generating antibodies to the corresponding phosphonate.

00:07:04 These antibodies actually are quite efficient.

00:07:07 Now we're asking the question whether we can, in fact, create a bifunctional antibody

00:07:11 where one end is specific for the tumor cell, one end is catalytic, localize this antibody

00:07:16 at the tumor cell, and generate a high local concentration of 5'-FU.

00:07:21 Now I should point out, an important design aspect of this chemistry is that there's no

00:07:26 known enzyme that'll cleave this D-valial ester.

00:07:30 We are not going to get a nonspecific activation of the prodrugs throughout the body, only

00:07:35 at the site where the antibody's localized.

00:07:38 Well, in fact, notions of transition state stabilization have been used to generate a

00:07:43 number of catalytic antibodies up for other reaction, including amide bond cleavage, selective

00:07:49 phosphoryl transfer reactions of interest in chemistry and biology, as well as in the

00:07:55 selective cleavage of ethers, activated ethers, and enol ethers.

00:08:01 We've shown, in fact, that antibodies to transition states act as catalysts.

00:08:04 What else do antibodies against transition state analogs teach us?

00:08:08 In fact, these antibodies can teach us something about mechanisms of biological catalysis.

00:08:14 Let's consider, for example, the enzyme ferrochelatase, which catalyzes the methylation of the planar

00:08:21 neutral porphyrin shown here to the corresponding metalloporphyrin.

00:08:26 Now this enzyme has been the subject of considerable mechanistic investigation because it's thought

00:08:32 to function via a unique mechanism, a mechanism not involving acidic catalysis or covalent

00:08:38 catalysis, but simply involving distortion of the substrate.

00:08:42 Why is that?

00:08:43 Well, it turns out enalkylporphyrins are incredibly potent inhibitors of this enzyme, binding

00:08:48 significantly more tightly than the planar porphyrin substrate.

00:08:52 And in fact, enalkylporphyrins are bent porphyrins.

00:08:56 They're non-planar porphyrins in which the nitrogen-loam pairs are distorted out of the

00:09:00 planar porphyrin ring.

00:09:03 So based on this, it's been proposed that the mechanism whereby the enzyme operates

00:09:07 is to bind the planar neutral porphyrin, distort it out of planarity, the nitrogen-loam

00:09:13 pairs can now access the metal, it's metallated, it snaps back shut, and we get the metallated

00:09:19 porphyrin.

00:09:20 Well, if that's true and the enzymes function via distortion, if we make antibodies to a

00:09:26 distorted hapten molecule, those antibodies should act like ferrochelatases.

00:09:33 And in fact, when we carry out this experiment, we do generate catalytic antibodies, and

00:09:38 in fact, those catalytic antibodies have kinetic properties and specificities that are very,

00:09:44 very similar to the enzymes that evolved over hundreds of millions of years.

00:09:49 Okay, well, where do we stand?

00:09:52 Well, what we've shown is that we can generate catalytic antibodies by generating antibodies

00:09:57 to transition state analog.

00:09:59 In fact, Pauling, over 40 years ago, proposed that enzymes have evolved through the process

00:10:04 of natural selection to bind transition state.

00:10:07 Well, in fact, we've verified this notion by showing that, in fact, if we carry out

00:10:13 the process of immunological selection against the transition state, we get catalyst.

00:10:18 Okay, what other kinds of strategies can we use for the generation of a catalytic antibody?

00:10:24 Well, another notion that one can apply to this problem is the notion of catalysis by

00:10:28 approximation or proximity effects.

00:10:31 What do I mean by this?

00:10:33 What I mean is to form the activated complex between two reacting groups, one has to overcome

00:10:39 a lot of rotational and translational barriers to form this complex.

00:10:44 The question is, is do enzymes catalyze reactions by using their binding energy to bind these

00:10:50 two reacting groups and hold them in a fixed reactive conformation?

00:10:54 And if that's true, then we ought to be able to use the binding affinity and specificity

00:10:59 of antibodies to do the same thing.

00:11:01 Well, in fact, Jenks and Koshland and Bruce and others have argued that, in fact, this

00:11:06 may be the main mechanism whereby enzymes catalyze chemical reactions.

00:11:11 So to test that notion and see whether we can use this as a rule for generating antibody

00:11:15 catalysts, we first looked at an intramolecular reaction, the conversion of carismic acid

00:11:21 to prefinic acid.

00:11:23 This reaction is of interest for a number of reasons.

00:11:26 First of all, it's involved in the biosynthesis of aromatic amino acids in plants and bacteria,

00:11:31 and as such is catalyzed by the enzyme carismate mutase, which accelerates the reaction roughly

00:11:37 a million-fold over the background rate.

00:11:40 Second of all, this enzyme is a rare example of what's formally a pericyclic reaction,

00:11:45 a 3,3-sigmatropic rearrangement of carismate to prefinate.

00:11:49 Now, this reaction is known both in the uncatalyzed and enzyme-catalyzed forms to proceed through

00:11:55 a chair-like conformationally restricted transition state, and it costs approximately minus 13

00:12:02 entropy units to form this transition state because we're restricting the conformations

00:12:07 of the carismate substrate.

00:12:09 So then the idea is if we could make an antibody that binds carismate, the substrate, and locks

00:12:15 it in this chair-like transition state, we should get a catalyst, and that should be

00:12:18 pretty good.

00:12:19 Well, in fact, we've generated antibodies to the bicyclic diacid transition state analog

00:12:24 shown here, and in fact, one such antibody accelerates this reaction 10 to the fourth

00:12:29 fold over background.

00:12:31 So that's about 100 times slower than the naturally occurring enzyme.

00:12:35 How's the antibody work?

00:12:36 We've in fact recently measured the activation parameters for this antibody-catalyzed reaction,

00:12:41 and in fact have shown that the delta S double dagger for the reaction is approximately zero.

00:12:47 So the antibody is in fact functioning as an entropy trap.

00:12:50 Well, given we can catalyze an intramolecular reaction using this strategy, the obvious

00:12:55 next step is to go to a bimolecular reaction, and we chose to look at the Diels-Alder reaction,

00:13:01 which is of interest for a number of reasons.

00:13:03 First of all, there's no known enzyme that catalyzes a Diels-Alder reaction, so it was

00:13:08 of interest to see if we could generate a catalytic antibody that does this.

00:13:11 Second of all, this is a very important reaction in chemistry for forming carbon-carbon bonds.

00:13:16 Two carbon-carbon bonds are formed in a single step.

00:13:19 And finally, this reaction is known to have a very large negative entropy of activation.

00:13:25 So to test whether we can make antibodies that catalyze a Diels-Alder reaction, we looked

00:13:29 at the model system shown here, a Diels-Alder condensation of this water-soluble diene with

00:13:35 a molybdenum dienophile to form the corresponding cyclohexane derivative product.

00:13:41 This reaction, again, proceeds through a very conformationally restricted transition state.

00:13:46 And the question, then, one has to think about is what molecule does one synthesize to instruct

00:13:52 the immune system to make a Diels-Alderase catalyst?

00:13:55 Well, we chose to form the carbon-carbon sigma bonds and lock our hapten into the Diels-Alder-like

00:14:04 transition state with an ethanol bridge.

00:14:07 So we generated antibodies to this 2, 2, 2 bicyclic system.

00:14:11 And in fact, when we made antibodies to this hapten, we found that they accelerated the

00:14:15 reaction roughly 10 to the 6th over background.

00:14:18 Now, the next obvious step in this series is to ask whether we can use the specificity

00:14:23 of antibodies to catalyze the formation of the exo, or the disfavored product, over the

00:14:29 more favored endoproduct, something that's relatively difficult to do with chemical reagents.

00:14:36 We've also looked at another bimolecular reaction, a transesterification reaction shown here.

00:14:41 This reaction involves a transfer of an acyl group to a secondary alcohol of the climidine.

00:14:48 This reaction is a tough reaction for a number of reasons.

00:14:52 First of all, we're going to carry out this reaction in water.

00:14:54 So although we'd like to transfer the acyl group to a secondary alcohol, we have a lot

00:14:59 better nucleophile present in the reaction mixture.

00:15:02 Moreover, there's 10 to the 5th times more of this better nucleophile of water than the

00:15:07 secondary alcohol.

00:15:09 So you need a pretty good catalyst to transfer the acyl group to the thymidine versus water

00:15:14 itself, which would be just a simple hydrolytic reaction.

00:15:17 To do this, we generated antibodies against the hapten shown here, this phosphonate diester

00:15:24 which incorporates elements of both substrates and the leaving group in the appropriate configuration.

00:15:29 In fact, when we generated antibodies to this phosphonate diester, we isolated one and accelerated

00:15:36 the reaction roughly a billion times over the background rate.

00:15:40 Now that's the rate you see in highly evolved enzymes.

00:15:43 Moreover, there was absolutely no hydrolysis of the ester catalyzed by the antibody.

00:15:51 So the antibody has such high specificity that it selectively transferred the acyl group

00:15:55 to the secondary alcohol versus ubiquitous water.

00:15:59 Again, this is the kind of specificity one sees in a highly evolved enzyme.

00:16:04 I should point out Richard Lerner has also generated antibodies that carry out a similar

00:16:08 transesterification reaction using similar substrate.

00:16:12 But in his case, the antibodies function via a completely different mechanism.

00:16:16 He showed, in fact, that his antibodies, which accelerate the reaction again about

00:16:20 a billion fold, his mechanism involves an acyl antibody intermediate.

00:16:26 Our mechanism simply involves binding of the two substrates in a reactive configuration

00:16:31 in the antibody combining site.

00:16:34 So nature can use multiple mechanistic solutions for a single catalytic problem.

00:16:39 This is again reflected in enzymes in the serine, aspartyl, and metalloproteases.

00:16:45 What other kinds of reactions can you do using this idea of entropy and entropic restriction?

00:16:51 Well, one can imagine generating antibodies that catalyze macrocyclic lactinization reactions.

00:16:57 One can make a peptide ligase that pastes together peptides to make protein.

00:17:01 One could make an aldolase.

00:17:04 One could perhaps make catalysts that generate stereoregular polymers.

00:17:08 All of these issues are currently under investigation.

00:17:11 I think I should point out also that this exercise has taught us an important thing.

00:17:15 In fact, the simple notion of proximity effects or entropic restriction can lead to very,

00:17:22 very large rate accelerations in enzymatic catalysis.

00:17:25 And in fact, as Jens pointed out, this is probably one of the most important mechanisms

00:17:30 whereby enzymes operate.

00:17:32 So what other strategies can we use to generate catalytic antibodies?

00:17:36 We've seen two.

00:17:37 The third possible strategy is the notion of general acid, general base catalysis.

00:17:41 Let's consider for a minute the reaction shown here, the isomerization of this alpha-hydroxy

00:17:47 ketone to the corresponding aldehyde, alpha-hydroxy aldehyde.

00:17:53 What's interesting about this reaction is that it's catalyzed by an enzyme, and the

00:17:56 enzyme accelerates this reaction roughly 10 to the 8th of a background.

00:18:01 But the enzyme accelerates this reaction to such a degree that the rate-limiting step

00:18:05 in the enzymatic reaction is diffusion of the substrates into the active site.

00:18:10 Not chemistry, but simple physical phenomena.

00:18:13 So clearly we'd love for all our catalysts to have this kind of catalytic efficiency.

00:18:18 So how are we going to make antibodies if it functions this effectively?

00:18:21 Well, you've got to realize that this enzyme, tri-SPSM erase, does two things.

00:18:25 First of all, it stabilizes the negatively charged transition state, something we know

00:18:29 how to do already.

00:18:31 But it also has a general base in the active site that deprotonates the C-alpha position.

00:18:37 And by using these two things together, TS stabilization and a general base, it achieves

00:18:41 a very large rate acceleration.

00:18:44 So we'd clearly love to know how to generate general bases in the antibody-combining site.

00:18:48 So how do we do that?

00:18:49 We looked again at a deprotonation reaction, but when we deprotonate in this case, we're

00:18:54 carrying out the elimination of fluoride from a beta-fluoroketone to give the corresponding

00:18:59 enone.

00:19:00 So to catalyze this reaction with a general base, you'd like a base positioned such that

00:19:05 it can abstract a C-alpha proton to initiate the reaction.

00:19:08 Well, how do we get that base in the appropriate position?

00:19:11 Well, we take advantage of the complementarity of the antibody to the hapten against which

00:19:16 it's generated.

00:19:17 If we want a negatively charged carboxylate acting as a base in the combining site, what

00:19:21 we do is we make the antibody to a positively charged hapten.

00:19:26 In fact, we did that, generated antibodies to these benzylammonium haptens, and in fact

00:19:30 found antibodies that catalyze this elimination reaction roughly 10 to the fifth over the

00:19:34 background rate.

00:19:37 We've characterized these antibodies extensively, shown that there's a carboxylate, identified

00:19:42 the carboxylate, determined the rate-limiting step, and actually determined the stereochemistry.

00:19:48 Another interesting illustration of this notion of generating general acid and general basic

00:19:52 groups in antibody combining sites is a recent reaction published by Richard Lerner and Kim

00:19:58 Janda and co-worker.

00:20:00 This reaction involves the hydrolysis of an enol ether to the corresponding aldehyde.

00:20:06 What you need to carry out this reaction is an acidic group, such as a glutamate or aspartate

00:20:10 in the antibody combining site, and Lerner and Janda and others reason that if they generated

00:20:16 antibodies to a positively charged hapten, they might get an acidic residue in the combining

00:20:21 site.

00:20:23 In fact, that's exactly what happened.

00:20:24 They did isolate catalytic antibodies, and interestingly, these antibodies catalyzed

00:20:29 the conversion of the enol ether to the chiral aldehyde, and gave the chiral aldehyde a greater

00:20:35 than 98% EE.

00:20:37 These are extremely selective catalysts delivering a proton to only one face of the enol ether,

00:20:42 something that's very difficult for chemists to do.

00:20:46 What other kinds of reactions can you do using this strategy?

00:20:49 You can imagine generating antibodies that have a general base that catalyze RNA cleavage,

00:20:55 or antibodies that catalyze an oxy-cope rearrangement via an anionic substituent effect.

00:21:01 So that's a third strategy one can use to generate antibodies.

00:21:05 A fourth approach one can take involves exploiting chemical cofactors to carry out antibody catalyzed

00:21:12 reactions.

00:21:13 What do I mean by this?

00:21:14 Now, enzymes, when faced with a difficult chemical reaction, such as an oxidative reaction

00:21:19 or the hydrolysis of an amide or a glycosidic linkage, often recruit cofactors.

00:21:27 The natural cofactors are metal ions, pyridoxal, flavins, hemes, and so forth.

00:21:34 Question is, can we also generate antibodies that use cofactors?

00:21:37 Now with antibodies, we should not only be able to use the natural cofactors, but because

00:21:42 we can basically dial in any specificity we'd like, we should be able to use a huge

00:21:47 range of unnatural synthetic cofactors.

00:21:50 For instance, metal hydrides could be used in redox reactions, transition metals, rhodium

00:21:56 phosphine complexes, photosensitizers, and so forth.

00:22:00 As a demonstration of this idea, we asked whether we could generate antibodies that

00:22:05 catalyze a stereospecific reduction of an alpha-keto amide.

00:22:10 Now, if you were an enzyme and you were going to carry out this reaction, you would probably

00:22:13 choose nicotinamide as a cofactor.

00:22:15 If you look up the price of nicotinamides and borohydrides in Aldrich, you find out

00:22:20 nicotinamides cost about a thousand times more than a borohydride.

00:22:23 Borohydrides are more powerful reagents.

00:22:26 So we asked whether we could generate an antibody that uses borohydrides as a catalytic auxiliary

00:22:30 to carry out this reaction.

00:22:31 Well, in fact, in this case, we generated antibodies to a phosphonate, which we thought

00:22:36 might be a reasonable mimic of a hydride attack on a carbonyl.

00:22:39 And those antibodies actually did carry out the reaction, and they gave the corresponding

00:22:44 2S product with a DE of greater than 98%.

00:22:48 That's interesting since the background reaction gives the other diastereomer, the 2R diastereomer,

00:22:53 with a 56% DE.

00:22:57 More recently, we've begun to look at the stereospecific reduction of unactivated ketones.

00:23:02 For example, we've looked at the reduction of the nitrobenzyl ketone shown on this slide.

00:23:08 Now, in this case, we've used antibodies not to a phosphonate, but to an anoxide hapten.

00:23:14 And, in fact, we've isolated an antibody that catalyzes the reduction of this carbonyl group

00:23:19 in this benzylic system with the corresponding alcohol with an EE of greater than 98%.

00:23:26 More recently, we've gone on to look at the issue of regiospecificity.

00:23:29 So we've asked whether we can generate an antibody that catalyzes the reduction, say,

00:23:33 of the nitrobenzyl carbonyl versus a methoxybenzyl carbonyl group.

00:23:37 And, in fact, we've isolated one such antibody that does do this, that selectively reduces

00:23:42 the nitrobenzyl carbonyl.

00:23:45 So we don't have to use protecting groups to achieve selective reduction in this system.

00:23:50 We're also looking at natural cofactors, such as hemes, and our hope here is to use hemes

00:23:57 to catalyze the regiospecific hydroxylation of substrates, for example, a steroid.

00:24:02 We've, in fact, shown that antibodies elicited to an enalkylporphyrin will, in fact, bind

00:24:08 the ferricene and, in the presence of hydrogen peroxide as the oxidant, catalyze the oxidation

00:24:14 of a number of aromatic chromogenic substrates.

00:24:17 We now, the problem with these antibodies, although the rate is pretty good, 200 per

00:24:21 molar per second, the specificity for substrate is low.

00:24:25 In an attempt to build in a substrate binding site, we're making antibodies now to enalkylporphyrins

00:24:31 in which the substrate of interest, say a steroid, is linked to one of the pyrrol nitrogens

00:24:37 of the porphyrin ring.

00:24:38 This should give us two binding sites, one for the heme and one for the substrate.

00:24:44 Other kinds of cofactors that have been looked at are pyridoxals and flavins.

00:24:49 Other potential cofactors include metal ligand complexes, which Lerner and coworkers have

00:24:55 used to generate antibodies that catalyze the cleavage of a peptide bond.

00:25:00 Light has also been used as a cofactor in antibody catalyzed reactions in a 2 plus 2

00:25:05 cycloreversion reaction, again showing that, in fact, all of the rich excited state photochemistry

00:25:10 that's available to synthetic chemists is also available to catalytic antibodies.

00:25:15 So where are we going?

00:25:17 What are we going to see in the next few years in the area of catalytic antibodies?

00:25:20 Well, one thing we're going to see is exploiting combinatorial libraries for the generation

00:25:26 of catalytic antibodies.

00:25:28 Combinatorial technology is going to allow us to assay, rather than 50 antibodies for

00:25:32 catalytic activity, 100,000 antibodies for catalytic activity.

00:25:37 So we're very likely to see increased catalytic rates from this technology.

00:25:42 Also, the process of natural selection is going to be brought to bear on the design

00:25:46 of catalytic antibodies.

00:25:50 Many reactions, hydrolytic reactions of amides, esters, phosphodiesters, and sugars, can be

00:25:56 coupled to the survival of a living cell.

00:25:59 And by coupling a catalytic antibody reaction to the life of E. coli, we can bring the process

00:26:05 of natural selection to bear on the evolutions of these catalysts and increase the rates

00:26:09 dramatically this way as well.

00:26:11 Finally, a third important strategy that's currently being examined is to combine the

00:26:16 concepts we've talked about during this talk, combine transition state stabilization with

00:26:21 entropic reduction with a general base, bring all of these catalytic concepts to bear on

00:26:26 the design of a single catalytic antibody for a single reaction.

00:26:31 And this idea has been actually nicely illustrated by Richard Lerner in a relatively spectacular

00:26:35 reaction shown on the next slide.

00:26:38 Richard asked whether he could generate antibodies that catalyze an intramolecular ring opening

00:26:43 of an epoxide to give the corresponding cyclic ether.

00:26:47 Now, how are you going to generate an antibody that does this reaction?

00:26:51 Well, Richard made antibodies to a cyclic N-oxide.

00:26:55 The N-oxide functionality was thought to reflect the dipolar nature of the transition state

00:27:00 for CO bond cleaving.

00:27:02 The six-membered ring is thought to reflect the cyclic six-membered ring transition state

00:27:06 in this reaction.

00:27:07 So we're achieving two effects in this hapten.

00:27:10 Well, in fact, this reaction, this antibody catalyzed reaction, actually goes to give

00:27:15 the disfavored product.

00:27:17 The six-membered ring product is a major reaction.

00:27:20 Now, if you carry out this reaction in solution, what you get is a five-membered ring favored

00:27:26 cyclic ether, which is favored by Baldwin's rules.

00:27:29 So what we see here is an example where the specificity and affinity of the antibody have

00:27:34 been used to selectively stabilize a disfavored over a favored transition state.

00:27:40 So what have we learned over the past six years by looking at catalytic antibodies?

00:27:44 Well, one thing we've learned is that, in fact, one can generate catalytic antibodies

00:27:48 via a number of approaches, stabilizing transition states, reducing entropy of activation, general

00:27:54 acid-base catalysis, covalent catalysis, or many of these features simultaneously.

00:28:01 Antibodies can catalyze a large number of different reactions.

00:28:04 They can catalyze hydrolytic reactions, amide bond cleavage, paracyclic reactions, a Diels-Alder

00:28:10 reaction, substitution reactions, a cyclization of an epoxide, isomerization reactions, and

00:28:16 redox reactions.

00:28:18 Antibody catalyzed reactions are characterized also by high specificity and, in many cases,

00:28:23 very high rates rivaling those of the corresponding enzymes.

00:28:27 And finally, the characterization of these catalytic antibodies has, in fact, given

00:28:32 us greater insight into the mechanisms of biological catalysis, and I think we'll continue

00:28:37 to do so.

00:28:42 The final speaker is Dr. George M. Whitesides.

00:28:46 He is professor of chemistry at Harvard University, where his research interests include biochemistry,

00:28:52 material science, rational drug design, and molecular recognition.

00:28:57 Dr. Whitesides' presentation will focus on what reactions can be accomplished by enzymes

00:29:03 and how and where to use these reactions.

00:29:06 Dr. Whitesides' topic today is organic synthesis using enzymes.

00:29:12 I'd like to talk to you about enzymes.

00:29:16 Enzymes are proteins that are catalytic.

00:29:20 They're obtained from natural sources, and they offer the chemist unique opportunities

00:29:26 to bring selectivity and, in some cases, rate acceleration to certain types of reactions

00:29:33 that are very difficult to accomplish otherwise.

00:29:37 In the past, enzymes have not been a very important part of the catalytic repertoire

00:29:43 of synthetic chemists, at least in part because they were not very readily obtainable.

00:29:50 The advent of recombinant DNA technology has made it possible to produce almost any enzyme

00:29:56 now in substantial quantities.

00:29:59 So the availability of these species is no longer a limiting factor in their use in synthesis.

00:30:07 What I would propose to do is to discuss three topics with you.

00:30:12 First, I'd like to outline the characteristics of enzymes as catalysts.

00:30:18 What makes them interesting?

00:30:19 Why are they different from other kinds of catalysts?

00:30:22 Second, I'd like to give you a little sense for how they're used in the laboratory.

00:30:28 And third, and taking most of the time, I want to show you applications in which enzymes

00:30:34 provide unique solutions to problems in chemical catalysis.

00:30:40 Let's begin by thinking about some of the characteristics of enzymes broadly.

00:30:50 Enzymes are proteins.

00:30:52 They can be obtained either from microorganisms with the advantage of large-scale production

00:30:58 using recombinant techniques, or they can be obtained from animal or plant tissue.

00:31:06 Their characteristics as a catalyst are primarily that they are selective, but secondarily that

00:31:13 they give rate accelerations.

00:31:16 When you have a problem that requires selectivity, think about enzymes.

00:31:21 They also have the characteristic that they are regulated, that is to say their capability

00:31:27 to accelerate a reaction depends upon what other compounds are present in solution.

00:31:34 This capability for regulation is sometimes useful, but it is more often a nuisance.

00:31:40 So I'll point out one occasion in which regulation is important, but in general we will ignore

00:31:48 this part of enzymatic catalysis.

00:31:52 Where have enzymes actually been used?

00:31:54 There are three applications in which they have clearly established their value.

00:32:01 The first is in chiral synthesis, particularly the synthesis of small chiral fragments for

00:32:08 use in pharmaceutical synthesis.

00:32:11 The second is industrial chemistry, where their ability to work in room temperature

00:32:17 aqueous solution offers solutions to environmental problems and sometimes substantially interesting

00:32:24 solutions to problems in purification.

00:32:28 They have also begun to revolutionize the synthesis of sugars.

00:32:33 Glycobiology is one of the most exciting areas of sugar chemistry, and enzymes are providing

00:32:38 one of the key technologies for making sugars and their derivatives.

00:32:43 What about cost?

00:32:45 The tradition has it that enzymes are expensive as catalysts.

00:32:51 Is this true?

00:32:54 In fact, in some circumstances they can be quite economical as catalysts because they

00:32:59 have very high catalytic efficiency.

00:33:01 Let's look at a typical set of numbers.

00:33:07 For an enzyme operating under optimal conditions, it should be possible to achieve turnover

00:33:13 numbers, that is moles of product per mole of enzyme, in the order of 10 to the 6th to

00:33:20 10 to the 8th.

00:33:21 These are very high numbers for a catalytic process.

00:33:28 If one takes a typical cost of $10 per gram for an industrial enzyme, the cost of a mole

00:33:35 of enzyme would be in the order of a million dollars, admittedly expensive.

00:33:41 On the other hand, for a turnover number of 10 to the 6th to 10 to the 8th, the contribution

00:33:47 to the cost of the product based on the enzyme will be only in the order of a dollar to a

00:33:54 penny per mole.

00:33:55 Thus, under some circumstances at least, enzymes need not be expensive as catalysts.

00:34:02 One can take advantage of their very high turnover numbers to lower their contributed

00:34:08 cost to the product to the point where it may not even be significant.

00:34:12 How are enzymes manipulated in solution?

00:34:16 One of their virtues is that they're easy to use.

00:34:21 Probably the simplest thing to do with them is simply to add an enzyme to a solution containing

00:34:26 the reactant.

00:34:28 This is a common procedure with lipases and with certain amidases in which the enzymes

00:34:33 are quite inexpensive and very sturdy in solution.

00:34:38 In some circumstances, it's necessary to immobilize the enzymes or to somehow contain them either

00:34:45 to preserve their activity or to enable them to be recovered or sometimes to make it easy

00:34:51 to separate them from the products.

00:34:55 Let me show you several simple pieces of apparatus that we use in our laboratory when we're using

00:35:01 enzymes in catalysis.

00:35:07 A common procedure is to use the enzymes in a mobilized form.

00:35:14 Here I show you an enzyme system contained behind a dialysis membrane.

00:35:22 This technique called MEEC, M-E-E-C, for membrane enclosed enzymatic catalysis, provides a very

00:35:31 simple method of keeping the enzyme separated from the products and allowing it to be recovered.

00:35:38 In this kind of procedure, one adds the enzyme to the dialysis membrane, carries out the

00:35:44 reaction, the reactants diffuse in to the compartment, react, and then diffuse out.

00:35:51 At the end of the reaction, to separate the enzyme, one simply pulls out the dialysis

00:35:55 tubing.

00:35:56 It can be stored in a refrigerator and frequently reused on many occasions.

00:36:02 A second procedure for using an enzyme is to immobilize it on a solid support.

00:36:09 Glass beads are commonly used and glutaraldehyde is perhaps the most useful cross-linking agent.

00:36:15 A third procedure is to immobilize the enzyme by covalent reaction in a polyacrylamide gel

00:36:24 or some other soft diffusible substrate.

00:36:28 We have employed extensively a gel based on a polymer called PAN, and I show you here

00:36:36 a typical enzyme preparation.

00:36:38 The suspended particles are the polyacrylamide gel containing immobilized enzyme.

00:36:45 This kind of procedure can be used to recover the enzyme by centrifugation or filtration

00:36:51 at the conclusion of the process.

00:36:55 It's useful to mention two other aspects of enzymatic catalysis.

00:37:00 First, all enzymes are temperature sensitive, most are pH sensitive, and many are sensitive

00:37:07 to oxygen.

00:37:09 It is, therefore, useful to set up reactions in such a fashion that one can control pH

00:37:17 and control access of oxygen.

00:37:21 Here I show you a common apparatus that we use in the laboratory for this purpose.

00:37:26 It's simply a round bottom flask and it has attached to it a pH controller and a pH sensor.

00:37:33 Very simple system and one that is accessible in any laboratory on a small scale.

00:37:39 The final point that I would emphasize about enzymes is that they are proteins and hence

00:37:46 are subject to biodegradation by microorganisms.

00:37:50 It is annoying to carry out a reaction on Friday afternoon, come back on Monday, and

00:37:56 find that a yeast is eating your catalyst.

00:37:59 The procedures for preventing such biodegradation are straightforward, but they're a little

00:38:03 unfamiliar to many chemists.

00:38:06 Basically, one adds a fungicide or a material like azide that prevents the growth of most

00:38:14 microorganisms.

00:38:16 Let's turn now to examples of the uses of enzymes in organic synthesis.

00:38:23 I want to talk about three types of applications.

00:38:28 Simple applications in which we are concerned with hydrolysis or isomerization, more complex

00:38:36 systems, and then very complicated systems involving multiple enzymes.

00:38:44 The most important simple enzymatic reactions are those based on esterases or amidases for

00:38:52 hydrolysis of esters and amides or on simple isomerization reactions.

00:38:58 These are also the basis for the largest scale reactions.

00:39:03 The reactions of intermediate complexity either involve unusual enzyme systems, for example,

00:39:12 the use of aldolases for forming sugars, or cofactors.

00:39:18 Cofactors are reagents that are required by many enzymes to achieve phosphorylation or

00:39:23 to donate or accept hydrides in oxoreductase catalyzed systems.

00:39:30 Probably the most complex type of multi-enzyme system now used are those that are involved

00:39:37 in the Loire pathway for sugar biosynthesis.

00:39:40 And I will mention those at the end of the talk.

00:39:45 Let's begin by simple examples focused on hydrolases.

00:39:51 These are reactions that can be carried out in very large scale and are operationally

00:39:56 very straightforward.

00:40:01 Probably the ideal enzyme in many circumstances is acylase I.

00:40:08 Acylase is an enzyme that hydrolyzes acetylamino acids.

00:40:13 It has the characteristic that it will accept a very wide variety of amino acids, both natural

00:40:19 ones and unnatural ones.

00:40:21 It is effectively enantiospecific.

00:40:26 So acylase I has the characteristic that it combines two attractive characteristics,

00:40:32 very broad substrate specificity and very high enantioselectivity.

00:40:37 It also is very inexpensive and very sturdy.

00:40:43 Lipase is a representative of another very important class of enzymes.

00:40:48 Lipase hydrolyzes esters enantioselectively.

00:40:53 The example that I show here is the hydrolysis of glycidylbutyrate.

00:41:01 Let me give you an example in which I can sketch the procedure.

00:41:06 What one does in this hydrolysis is to take a beaker.

00:41:11 You fill it a quarter full with water and then another quarter full with glycidylbutyrate.

00:41:18 Lipase is added as a crude powder.

00:41:21 The reaction is stirred until it stops and what is left in the insoluble organic phase

00:41:27 is one enantiomer of unhydrolyzed glycidylbutyrate.

00:41:33 Separation involves simply pouring off the upper layer.

00:41:36 It's difficult to find anything easier.

00:41:39 What about large scale processes?

00:41:42 Enzymes have actually found several important applications in very large scale chemistry,

00:41:48 chemistry that qualifies as full industrial scale.

00:41:52 The value of these applications is either environmental friendliness, enantioselectivity,

00:41:58 or compatibility with food products.

00:42:02 The first example is the isomerization of glucose to fructose.

00:42:06 This is a process used in making high fructose corn syrup and is practiced in the United

00:42:12 States on a scale of approximately 10 to the 9th pounds per year.

00:42:18 A second interesting and surprising example is the hydrolysis of acrylonitrile to acrylamide.

00:42:27 In older industrial practice, this reaction was carried out using a copper chromite catalyst

00:42:33 in steam and there was a strong feeling that it would be difficult to find anything much

00:42:38 simpler.

00:42:39 In fact, overall, the enzyme catalyzed process developed by NITO has proved to be better.

00:42:49 The productivity of the enzyme and the stability of the enzyme are excellent and the purity

00:42:55 of the product is such that purification costs in this process are substantially lower than

00:43:02 they are in the steam hydrolysis.

00:43:05 Overall, it now seems that the enzyme catalyzed hydrolysis of acrylonitrile to acrylamide

00:43:12 is the best method of carrying out this process.

00:43:16 The third example is the hydrolysis of penicillin G to 6-amino penicillinic acid, 6-APA.

00:43:24 6-APA is used as an intermediate in the synthesis of semi-synthetic penicillins.

00:43:31 And the value of enzymatic catalysis in this case is the selective hydrolysis of the more

00:43:38 stable amide group in the presence of a more reactive amide group.

00:43:44 A final example is the enantiospecific addition of ammonia to fumaric acid to make optically

00:43:51 active aspartic acid.

00:43:53 This is one representative of a class of reactions in which enzymes are used to make optically

00:44:00 pure amino acids for food and parenteral use.

00:44:06 Let's consider more complicated reactions involving cofactors.

00:44:11 There are two classes of cofactors that are commonly used in current enzymatic synthesis.

00:44:18 One is based on ATP.

00:44:20 The second is based on the nicotinamide cofactors NAD, NADH, and their derivatives.

00:44:28 ATP is used in biological systems as the equivalent of tosyl chloride for oxygen activation.

00:44:36 Where you would use tosyl chloride, nature phosphorylates an oxygen to make it a better

00:44:41 leaving group.

00:44:44 The nicotinamide cofactors are used as hydride donors and hydride acceptors.

00:44:49 They are, in a sense, the equivalent of borohydride or lithium aluminum hydride for reductions

00:44:57 and of chromium trioxide or Swern reagent for oxidations.

00:45:04 These materials are too expensive to be used stoichiometrically, so it's necessary to recycle

00:45:09 them in situ in order to make a number of moles of product per starting mole of cofactor.

00:45:18 Let me show you a simple example of the use of NADH as a cofactor in an enzyme catalyzed

00:45:27 reaction leading to a chiral useful piece.

00:45:32 The reaction here is based on lactate dehydrogenase.

00:45:36 This is a widely available enzyme that has excellent enantioselectivity in the conversion

00:45:42 of pyruvate and its analogs to lactic acid and its analogs.

00:45:48 The reaction proceeds with conversion of NADH to NAD+.

00:45:54 The reconversion of NAD plus to NADH can be accomplished using a number of systems.

00:46:02 The one that is most widely used involves formate dehydrogenase as the catalyst and

00:46:07 formate as the hydride donor.

00:46:10 One can also find excellent systems based on glucose as the hydride donor and glucose

00:46:15 dehydrogenase as catalyst or, in some cases, glucose 6-phosphate dehydrogenase as the catalyst

00:46:22 and glucose 6-phosphate as the donor.

00:46:26 The example here is the conversion of chloropyruvate to optically active epoxy acrylic acid through

00:46:35 optically active chlorolactic acid.

00:46:38 The reaction proceeds very smoothly and has been carried out on scales approaching a mole.

00:46:46 What does enantioselectivity mean in this type of process?

00:46:50 Let me show you an example of a reduction using a deuterium label to provide an optically

00:46:59 pure labeled product.

00:47:04 This reaction is the conversion of trifluoroacetaldehyde to monoduterated trifluoroethanol.

00:47:12 Examination of the signals in the NMR spectrum labeled R and S give an indication of the

00:47:20 enantioselectivity of this process.

00:47:24 The use of enzymes in preparing labeled compounds is one of their widest specialty uses.

00:47:32 The use of enzymes for oxidation reactions is, for a number of reasons, somewhat more

00:47:39 complicated technically than reductions.

00:47:42 There are, nonetheless, some very valuable cases in which it is possible to make chiral

00:47:48 intermediates using enzymatic oxidations.

00:47:52 The example that I show here is one that represents work of Brian Jones, who has been one of the

00:48:00 pioneers in this area.

00:48:02 It is the conversion of a diol to an optically active lactone based on a NAD catalyzed oxidation.

00:48:13 The enzyme used here is alcohol dehydrogenase.

00:48:17 The NADH is reconverted to NAD using glutamate dehydrogenase.

00:48:24 Jones has also explored systems in which oxygen is the final oxidant using a system in which

00:48:33 there is an intermediate di-methylene blue as an electron transfer catalyst and an additional

00:48:40 enzyme, diaphorese, to accelerate the rates of these reactions.

00:48:47 The difficulty with oxidation reactions is that the product is typically more hydrophobic

00:48:52 than the starting material and tends to stick in the active site of the enzyme.

00:48:57 Hence, these reactions are often regulated, that is to say, inhibited by product.

00:49:04 Let's turn now to the use of ATP in organic synthesis.

00:49:09 A simple example is based on the enzyme glycerol kinase, which phosphorylates enantiospecifically

00:49:17 a wide variety of analogs of glycerol.

00:49:23 Glycerol kinase will take derivatives of glycerol as substrates that have small polar functionalities

00:49:31 in the 3 position, a variety of groups in the 2 position, and a limited number of substituents

00:49:37 in the 1 position.

00:49:39 These optically active glycerol phosphate analogs are useful in the synthesis of phospholipids,

00:49:46 phospholipid analogs, and related compounds.

00:49:50 In carrying out the types of reactions that require ATP regeneration, what are the best

00:49:59 phosphate donors to use?

00:50:02 There are now 2 very well developed systems for the synthesis of phosphates that can be

00:50:11 donated to ADP in situ for use in ATP cofactor recycling.

00:50:19 The first of these is acetyl phosphate.

00:50:22 Acetyl phosphate is easily prepared in very large quantities if needed by acetylation

00:50:28 of phosphoric acid with acetic anhydride.

00:50:34 Probably the more useful reagent is phosphoenolpyruvate.

00:50:39 There are 2 types of syntheses of phosphoenolpyruvate.

00:50:43 One involves starting with pyruvate, bromination, reaction with a phosphate, and hydrolysis

00:50:51 following a classical chemical procedure.

00:50:55 The second is based on enzymology.

00:51:00 This is a procedure that begins with 3-phosphoglyceric acid and uses 2 enzymes to convert 3-PGA to

00:51:10 phosphoenolpyruvate, PEP, in situ.

00:51:15 A reaction that utilizes the PEP, and one that I'm going to talk about in a moment,

00:51:22 the conversion of dihydroxyacetone to dihydroxyacetone phosphate, proceeds then in situ using 4 enzymes

00:51:34 overall, 2 to convert 3-PGA to PEP, 1 to recycle ADP into ATP, and the 4th to convert

00:51:46 the organic reactant, the dihydroxyacetone, to the product, dihydroxyacetone phosphate.

00:51:54 This set of reactions gives a hint of the extent to which enzymes can be coupled in

00:52:01 in situ reactions.

00:52:04 Let me turn now to a somewhat more sophisticated example of organic synthesis.

00:52:09 This is the synthesis of a sugar, KDO, keto-deoxy-octanoic acid.

00:52:18 This compound is important as a constituent in microbial cell walls.

00:52:22 It is a potential target for new types of antibiotics, and there is substantial interest

00:52:28 in KDO as a substrate in reactions in which inhibitors are being tested.

00:52:38 The synthesis proceeds in 3 steps.

00:52:43 The first involves phosphorylation of arabinose using hexokinase to give arabinose 5-phosphate.

00:52:52 There's an important lesson from this step.

00:52:56 The biochemical literature reports that arabinose is not a substrate for hexokinase.

00:53:03 In fact, in many circumstances, one should not be too closely guided by the biochemical

00:53:10 literature, because activities of enzymatic catalysts that are too low to be significant

00:53:16 in biochemical circumstances may, in fact, be very useful in chemistry.

00:53:22 In this particular instance, the activity of arabinose is less than 1% of the activity

00:53:28 of the natural substrate for hexokinase, which is glucose.

00:53:33 Nonetheless, by putting in a substantial excess of hexokinase, it is possible to get a smooth

00:53:40 conversion of arabinose to arabinose 5-phosphate without the use of protecting groups.

00:53:46 The ATP in this reaction is recycled by reaction with PEP catalyzed by pyruvate kinase.

00:53:58 The second step in the reaction is the carbon-carbon bond formation between arabinose 5-phosphate

00:54:05 and phosphoenolpyruvate catalyzed by KDO 8-phosphate synthetase.

00:54:12 This is a reaction that proceeds stereospecifically and provides a complex sugar skeleton, again

00:54:21 without the use in protecting groups.

00:54:24 The final step in the reaction is dephosphorylation using acid phosphatase, and once again illustrates

00:54:31 the use of enzymes in sugar chemistry to avoid protection and deprotection steps.

00:54:37 Let me talk now about another enzyme that has proved particularly useful in carbohydrate synthesis.

00:54:45 This is rabbit muscle aldolase, Rama, R-A-M-A.

00:54:50 This enzyme catalyzes the stereospecific reaction between the enol of dihydroxyacetone phosphate

00:54:58 and the aldehyde group of a variety of aldehydes.

00:55:02 It has been used extensively in forming carbon skeletons in sugars.

00:55:09 The example that I give here is one developed by Wong, leading to the compounds deoxynoduramicin

00:55:19 and deoxymanonoduramicin.

00:55:22 These are compounds which are useful as inhibitors in certain steps in the processing of oligosaccharides

00:55:31 in biochemical systems.

00:55:34 The crucial element in this synthesis is the carbon-carbon bond formation between dihydroxyacetone

00:55:43 phosphate and the aldehyde containing nitrogen protected in the form of an azide.

00:55:50 The reaction proceeds under very mild conditions and proceeds stereospecifically.

00:55:56 The final step of the reaction, the conversion of the azido compound to the aza sugar,

00:56:04 uses two steps, dephosphorylation using an enzyme and the catalytic heterogeneous reduction

00:56:13 of azide to amine.

00:56:16 A further and increasingly important application of enzymes in carbohydrate synthesis

00:56:23 is to the preparation of oligosaccharides.

00:56:28 The formation of glycosidic links remains one of the real challenges in carbohydrate chemistry.

00:56:35 And the enzymes of the so-called Loire pathway have proved enormously useful in preparing

00:56:42 glycosidic bonds without the use of protecting groups.

00:56:47 The example shown here is one of the first and simplest.

00:56:54 The overall process involves a number of steps.

00:56:58 Conversion of glucose to glucose-6-phosphate enzymatically, isomerization of glucose-6-phosphate

00:57:06 to glucose-1-phosphate, reaction of glucose-1-phosphate with UTP to make UDP-glucose.

00:57:18 UDP-glucose epimerase is then used to change the configuration of one of the hydroxyl groups.

00:57:26 And a glycosyltransferase is used to form the glycosidic bond that connects the glucose

00:57:33 moiety to N-acetylglucosamine.

00:57:38 The UDP that is produced in this final step is reconverted to UTP by phosphorylation using

00:57:46 phosphoenolpyruvate.

00:57:49 The overall process involves six enzymes working cooperatively in the same solution,

00:57:56 converting three starting materials into one complex final product without intermediate

00:58:04 protecting groups.

00:58:06 The final process that I want to talk about is one that integrates many of the steps that

00:58:12 we've talked about.

00:58:13 This is the conversion of glucose to ribulose diphosphate.

00:58:19 Ribulose diphosphate is an important intermediate in the fixation of carbon dioxide by plants.

00:58:28 This compound is a very unstable substance, and its availability has limited certain kinds

00:58:34 of studies in this important enzymatic system.

00:58:40 The process shown is one involving several steps with coordinated cofactor regeneration.

00:58:49 The initial conversion of glucose to glucose 6-phosphate requires ATP regeneration.

00:58:55 The oxidation of glucose 6-phosphate at C1 can be accomplished using bromine, but it

00:59:03 can also be accomplished using a nicotinamide cofactor and cofactor regeneration.

00:59:09 The oxidation of 6-phosphogluconic acid at C3 and decarboxylation to give ribulose 5-phosphate

00:59:20 requires another cycle of nicotinamide cofactor regeneration.

00:59:24 And the final step, the conversion of ribulose 5-phosphate to ribulose diphosphate, is again

00:59:31 ATP requiring.

00:59:33 Hence, overall, this is a process that requires four cofactor regeneration cycles to work

00:59:41 at the same time.

00:59:43 Let me conclude by making several remarks about the capabilities and future application

00:59:51 of enzymes.

00:59:54 Enzymes have not been commonly used in organic synthesis because they have not been readily

00:59:58 available and because many of the important targets, steroids, terpenes, and related compounds

01:00:05 are much better made using other methods.

01:00:08 As chemistry turns increasingly to products that are water-soluble, nucleic acids, peptides,

01:00:17 sugars, or to processes in which environmental impact is crucial, enzymes have a unique role

01:00:25 to play.

01:00:27 The availability of enzymes is no longer an issue.

01:00:31 It is possible to make almost any enzyme that one wants given sufficient effort by recombinant

01:00:37 techniques.

01:00:39 There are, however, a substantial number of technologies which must be learned in order

01:00:45 to use these enzymes effectively as catalysts in organic synthetic procedures.

01:00:51 I propose that the future of organic synthesis will require a substantially larger investment

01:00:59 in the technology of enzymology and in other applications involving biological catalysts

01:01:07 perhaps used or generated in situ.

01:01:11 The opportunities for chemists who understand how to use these classes of biological candidates

01:01:18 in real processes is enormous.

01:01:22 Thank you, gentlemen.

01:01:23 Our telephone lines are again open, so hop to it.

01:01:26 Go to your phones and give us a call.

01:01:27 The telephone numbers are 800-368-5781 and 5782 or in the Washington, D.C. area, 202-463-3170.

01:01:37 We have about 25 minutes to take your calls, so get them in now because we have this captive

01:01:42 audience of fine chemists here who are happy to answer your questions.

01:01:46 Now, if you get a busy signal, please hang up and try again.

01:01:49 Okay?

01:01:50 Now is the time to call.

01:01:51 You got those numbers in front of me or they should be all over me somewhere another year.

01:01:55 We have a question here which we could plumb the depths of probably for a long time to

01:01:59 speculate on it, but I'm going to throw it out to each of you, and I'd like each of you

01:02:02 to answer it in turn shortly if you can.

01:02:05 And that is, what is the future of non-biological versus biological catalysis?

01:02:11 In 25,000 words or less, right?

01:02:14 Who should start?

01:02:15 Okay, let's do it clockwise.

01:02:16 We'll go with Peter.

01:02:17 You go ahead.

01:02:18 Well, I don't think it's an issue of biological versus a biological.

01:02:22 I think it's an issue of what's the best approach for the specific transformation one's interested in.

01:02:28 If you're talking about a general reaction for small molecules and epoxidation that's

01:02:32 applicable to a wide class of substrates, then I think you want to go with an a-biological catalyst.

01:02:39 If you're looking at a large molecule and you want to do a transformation on a specific piece of it,

01:02:45 if you're going to selectively recognize a large molecule with many functional groups,

01:02:49 you're almost compelled to use a larger molecule such as protein to achieve that degree of recognition.

01:02:55 So I think what's going to happen is there's going to become an increasing appreciation

01:02:59 of what a-biological catalysis is good for and what biological catalysis is not.

01:03:04 So they're not necessarily at odds with one another.

01:03:06 No, they're complementary.

01:03:07 They're just how they use one another.

01:03:08 Yes, George.

01:03:09 There is going to be one place in which there is a real competition,

01:03:12 and that is in the area of biologically useful processes which have low environmental impact.

01:03:19 There are going to be many reactions, I believe,

01:03:21 in which the biological process enzyme catalyzed or catalytic monoclonal or whatever may actually be more expensive.

01:03:28 But in which environmental constraints will force an effort to remove organic solvents,

01:03:33 to limit byproducts, and to do related kinds of downstream optimization to lower overall systems cost.

01:03:41 So I think there may emerge a regulated aspect to this kind of catalysis

01:03:46 which may not make direct sense economically but which has value

01:03:49 because if you look at the overall relations with the community,

01:03:53 the impact on the environment, waste, purification cost,

01:03:57 where it might be simpler to use and less expensive to use a biological catalyst

01:04:01 than a non-biological catalyst in directly competitive reactions.

01:04:05 Okay. Do you disagree, though, with his basic theory?

01:04:08 Aside from the-

01:04:09 I absolutely agree with what he says.

01:04:11 It's just a question of how one does the calculation.

01:04:14 All right. Okay. Barry Sharpless?

01:04:16 Okay. I agree with my two colleagues here.

01:04:20 A bad sign.

01:04:22 But I guess the thing that George brought up earlier about water being such a friendly thing for life on this planet,

01:04:30 the more we're forced to use things like water,

01:04:33 the more than the enzyme approach to doing even the small molecule transformations

01:04:38 that Peter said were probably good candidates for a biological catalyst

01:04:42 will start to be more efficient with the enzymes.

01:04:47 But I guess there is a real problem right now.

01:04:50 If we're going to use water as a solvent, we do have a lot of reactions in organic chemistry

01:04:55 which are going to have to be totally revamped, and it's hard to see how we can do that.

01:04:59 So on the timescale of the next 30, 40 years, it's hard to imagine we're not going to be doing a lot of organic chemistry

01:05:07 and organic solvents.

01:05:09 But I do think right now the technologies are complementary because, for example,

01:05:14 Chi-Wei Wang's example I gave in my lecture where the alvalase chemistry, well, you started that.

01:05:20 And there is a case where the sugars are really, really beautifully handled.

01:05:25 It's hard to imagine any non-enzyme catalyst that could deal with that.

01:05:28 And so right now I see a clear advantage in that system, and the sugar area is huge and growing.

01:05:34 But I guess I like to think that we can do the propylene oxide with a non-enzyme catalyst.

01:05:40 Okay. Barry Trost?

01:05:42 I would tend to agree very much with the comments that have been made.

01:05:45 I think that Peter's comments regarding what are some of the special aspects of biological catalysis

01:05:53 are ones where we will see an ever-growing importance of the use of these things in a practical sense.

01:05:59 I also agree that the issue of environmental concerns is going to grow,

01:06:03 and in that sense both biological and abiological catalysis have a role to play,

01:06:09 especially from the point of view of trying to develop reactions where potentially

01:06:16 you could have the feature that the only thing that will be generated is the product and nothing else.

01:06:22 And that's one of the nice things that could come out of trying to invent some reactions

01:06:27 which will most likely be, at least in the future that I see,

01:06:31 probably going to be abiologically developed rather than biologically developed.

01:06:35 Okay, good. Some good comments there.

01:06:37 Once again, let me exhort you to call 800-368-5781 or 5782

01:06:42 in the Metropolitan, Washington, D.C. area, 202-463-3170.

01:06:47 Our first call is from South Seattle Community College in Seattle, Washington.

01:06:51 Go ahead, please.

01:06:53 Yes, I have questions for all four gentlemen.

01:06:56 Okay.

01:06:57 The first question is for the two Barrys.

01:07:00 At what point does it become practical to recycle used metal catalysts,

01:07:05 such as palladium, probably less so osmium, on a research scale?

01:07:10 And, for example, what does the Trost Group do?

01:07:14 Should I go on to the other questions?

01:07:16 No, hold up on them, and we'll get this out of the way.

01:07:18 Then we'll come back with the other ones. Go ahead.

01:07:21 On a research scale, especially the kind of research scale that we operate at,

01:07:25 the amounts of palladium are sufficiently small

01:07:28 that it doesn't make sense for us to involve ourselves in the actual recycling process.

01:07:34 So we would work with one of the major processors of palladium

01:07:39 and simply send them the spent waste, and they would then recycle the metal itself.

01:07:44 So we don't involve ourselves in it.

01:07:46 The methodology, the technology for recycling palladium is well worked out,

01:07:51 and it's something that can be done on a very large scale, if desirable,

01:07:54 but on a research, especially an academic research scale,

01:07:57 you rely on the commercial vendors, and you just send them the spent waste.

01:08:01 Okay. Was that same question for Barry Sharpless, too, or do you have a separate one for him?

01:08:05 Same question.

01:08:07 Okay.

01:08:08 The osmium catalyst is used in a very low amount, or can be.

01:08:13 One part in 50,000 is enough if you're going to be patient,

01:08:17 so you don't have too much in there.

01:08:19 But normally we use 0.2 percent in the lab, and sometimes 1 percent if we really want a fast reaction.

01:08:25 And we don't recover the osmium in small amounts,

01:08:30 but back 10, 15 years ago we worked with osmium stoichiometrically,

01:08:34 and we used to go through 200, 300 grams over a couple-year period,

01:08:39 and that we kept because there was a lot of osmium there,

01:08:42 and we gave it to a friend who was an osmium inorganic, organic metallic chemist,

01:08:47 and she recovered it in the traditional way, just oxidized with bleach and steamed to still.

01:08:52 And it's really an easy element to keep track of.

01:08:55 It's a little bit too volatile for some people's likes.

01:08:59 All right. Seattle, your questions for Peter and George. Go ahead, please.

01:09:03 Okay. For Professor Whitesides, you showed a pH controller there.

01:09:09 Can you tell us a little bit more about which of these enzymatic processes require such automatic control

01:09:15 and what it would cost for a small lab to set up such an apparatus?

01:09:20 Whenever the process is producing an acid or a base, in general, one has to have pH control.

01:09:26 A simple pH controller is a few hundred dollars.

01:09:29 They're quite inexpensive.

01:09:31 But if that's something which you don't want to do,

01:09:34 then we often simply run reactions by adding a little bit of an indicator

01:09:38 and dripping an acid and base from a burette.

01:09:41 It's not a very complicated issue.

01:09:43 The enzymes will tolerate a few pH units in swing with some loss in kinetic activity but no long-term damage.

01:09:50 So it's not a critical issue in most cases to keep track of the pH.

01:09:54 Okay.

01:09:56 Okay. Last question for Professor Schultz.

01:09:59 Could you just give us some idea of what kind of resources and time are needed

01:10:04 to do the kinds of catalytic antibody research that you do?

01:10:09 To do what we do, right now we do everything with hybridoma technology to produce the monoclonal antibodies.

01:10:15 So one has to synthesize the haptens, so you need a conventional organic chemistry lab,

01:10:20 and then one needs a cell culture facility and some animals to make the hybridomas.

01:10:26 That whole process takes about four months from start to finish,

01:10:30 and one has to be experienced in cell culture technique.

01:10:34 Currently there are recombinant methods coming online to produce these libraries of antibody monoclonal antibodies,

01:10:41 and once the system is working well, right now the levels of expression aren't quite what they have to be

01:10:48 to make this an everyday technique.

01:10:52 But once the expression levels get a little higher,

01:10:54 then all one has to do is be able to pour an agar plate and grow bacteria,

01:10:58 so you need a 37-degree room and an incubator, something like that, and that's it,

01:11:02 in addition to your own organic chemistry lab.

01:11:05 Okay, thank you. A caller from Seattle.

01:11:07 You've taken up our questions here from all the guys here.

01:11:11 Now we go back to Atlanta, Georgia, to the ACS local section in Georgia.

01:11:15 Go ahead, please.

01:11:17 The question is for Professor Whiteside.

01:11:21 Considering the important aspects of the cost of enzymes in enzymatic processes,

01:11:27 my question is what is your opinion of the current experimentation of use of enzymes in organic solvents?

01:11:34 There are places where enzymes are certainly valuable in what I will call mixed organic aqueous media.

01:11:41 I think there's no such thing as an enzyme that works in truly non-aqueous solution.

01:11:47 The non-aqueous enzymology, so-called, is enzymology that's carried out in an organic solvent

01:11:54 that contains enough water to coat the protein with a hydration shell or a number of layers of water.

01:12:01 But those are great media for certain types of reactions that involve, let's say, transesterification or dehydrations

01:12:08 or types of reactions in which one wants to manipulate the thermodynamic activity of water in the system.

01:12:14 I think there's no doubt that they're useful.

01:12:16 One has to be a little careful at this point about taking some of the information that's in the literature

01:12:22 about activities in non-aqueous solvents and really trying to use them in practical organic synthesis

01:12:27 because it turns out that some of those processes are more complicated

01:12:32 and the rates are lower than you can read from the literature.

01:12:35 But they're interesting. Just be a little cautious in getting into that kind of work.

01:12:39 All right. Our next call is from the South Texas local section ACS in Corpus Christi. Go ahead, please.

01:12:46 This is to Sharpless.

01:12:49 I was wondering whether sulfonamide is helping to increase the turnover number.

01:12:53 Is there any precedence in the literature or any rationale for using this?

01:13:03 That is actually described briefly in our publication in JOC, the most recent one,

01:13:11 where we had the thalazine ligands and also the sulfonamide improvement.

01:13:15 And the sulfonamide, it can be an aromatic sulfonamide or methyl sulfonamide

01:13:20 just happens to be one of the ones that work as well as any.

01:13:23 And what it does is it makes the anion in the carbonate phase, gets the sulfonamide anion,

01:13:28 it travels presumably to the interface or slightly into someplace near the interface

01:13:32 and helps to strip the glycol off the osmium.

01:13:35 That can be a slow step with many of these glycolates.

01:13:38 They're just too hindered and the osmium all hangs up at the osmate ester stage in those cases.

01:13:44 Normal olefins like styrene don't do that, so they're turning over at the rate of addition,

01:13:49 but the others need some help.

01:13:51 So we hydrolyte, we found this trick of adding sulfonamide,

01:13:55 and the idea came from earlier work we had done where we took hydroxide and phase transferred it,

01:14:00 but we couldn't do that here for various reasons because we phase transferred the oxidant.

01:14:04 And we had to transfer the anion by another means, which was to make it more lipophilic,

01:14:10 and that seems to work well.

01:14:12 It may not be the final solution, but it really makes it go up to 50 or 100 times faster than without it.

01:14:18 All right.

01:14:19 Our next call is from the King of Prussia, Pennsylvania, and SmithKline, Beecham.

01:14:23 Go ahead, please.

01:14:24 This is for Dr. Schultz.

01:14:25 What is your experience with mixed organic solvents and catalytic antibodies?

01:14:30 We've only done one experiment ourselves with reverse micelles,

01:14:36 and it was a hydrolytic reaction of an ester, and that worked very well.

01:14:41 Richard Lerner's group has done some work with putting these on solid supports

01:14:46 and trying to run these reactions with high percentages of organic solvents,

01:14:52 and those have actually worked quite well, too.

01:14:55 We've recently collaborated with a chemical engineering group

01:14:58 who has put these catalytic antibodies onto solid support,

01:15:01 and they've done a variety of biophysical measurements

01:15:04 to show that actually the active site is not perturbed at all

01:15:08 and the antibodies retain their full level of function.

01:15:12 All right.

01:15:13 Our next call is from the University of Wisconsin in Milwaukee.

01:15:16 Go ahead, please.

01:15:17 Hello.

01:15:18 I have two questions, one for the white slide, Professor, and one for Dr. Schultz.

01:15:23 I'll start with Professor George's white slide, please.

01:15:27 We understand the basic understanding of a reaction is important

01:15:31 to improve the method of scientific application.

01:15:36 So what about enzyme?

01:15:37 You can give any insight, understanding of the mechanism,

01:15:41 how it is giving a kind of asymmetric induction, something like that?

01:15:47 It's a very good question.

01:15:48 I think that one of the advantages to enzymology

01:15:51 is that in terms of a well-defined catalytic site,

01:15:55 you probably have as much information in an enzyme as you do in any catalyst

01:15:59 that we can work with, so that there are real opportunities with enzymology

01:16:03 to measure mechanisms, to measure kinetic constants,

01:16:06 and to use that information in designing the optimum conditions

01:16:10 for the enzymatic process.

01:16:13 In some of the other kinds of reactions that one works with,

01:16:16 for example, some of Barry's titanium or osmium catalyzed systems,

01:16:20 although the molecules are smaller,

01:16:22 the catalytic systems I think in some ways are actually more complicated.

01:16:26 So one of the keys in using enzymology is, in fact,

01:16:29 to take advantage of the mechanistic information

01:16:32 and the kinetic information that is relatively readily available.

01:16:36 All right.

01:16:37 My next question is for Barry Sharpless.

01:16:40 I know that you are working with designing a new catalyst with titanium.

01:16:45 You haven't talked about that in this presentation.

01:16:48 Do you have anything to add with that respect?

01:16:51 A new catalyst with titanium?

01:16:54 No, designing a new catalyst on the basis of your results

01:16:59 with your successful results with titanium catalyst.

01:17:04 Well, we're always screening for new reactions with metals and ligands these days,

01:17:09 but currently we don't have anything happening at all in the lab with titanium.

01:17:16 It would be better to improve that catalyst.

01:17:18 The catalyst is inadequate in its turnovers.

01:17:21 It's turned over 20 times, and that's not enough for a really good catalyst,

01:17:25 and we'd like to have some way of getting that system to turn over better.

01:17:30 The problem with it is it produces its own toxins.

01:17:33 It's one of the epoxy alcohols or suicide substrates for the catalyst.

01:17:38 The catalyst is very active in the first few seconds or minutes,

01:17:41 and it's dying constantly, and it doesn't help to add more catalyst

01:17:44 because there's something deadly produced in there

01:17:47 that just nails the catalyst faster than anything you can do to compensate.

01:17:51 So if anybody can improve this catalyst system,

01:17:55 they'd have a nice contribution to this process.

01:17:58 All right, there's your quest out there.

01:18:00 You can be famous, right?

01:18:02 All right, we'll go to St. Louis now, University of Missouri.

01:18:05 Yes, my question goes to all four gentlemen there.

01:18:09 For the design of non-biologic catalysts and empirical lures

01:18:13 to keep ligands to be C2 symmetry,

01:18:16 the enzyme or natural peptides, they usually don't have a C2 symmetry.

01:18:20 So according to you, why we need this C2 symmetry ligand

01:18:24 in making a good catalyst?

01:18:26 Do we really need it?

01:18:28 Okay.

01:18:30 Go ahead.

01:18:31 I've thought about this myself quite a bit

01:18:33 because the first C2 ligand ever made intentionally and used

01:18:38 in beautiful work and very original was by Henri Cagan-Diop,

01:18:43 is today still a very dominant concept in design.

01:18:48 Noyori's BINAP and Diop, BINAP.

01:18:52 The tartrate system is C2.

01:18:54 And I think that these things are coincidence to some extent.

01:18:58 C2 does simplify the mechanism, so you have a better chance of winning.

01:19:02 But let's face it, nature's catalysts are almost all C1

01:19:05 unless you have a dimeric subunit.

01:19:07 And the new osmium catalyst is C1, completely C1.

01:19:12 That system.

01:19:13 So I think there's more probability that you'll see more C1 systems

01:19:17 in the end than C2 or C3 or C4.

01:19:21 Okay.

01:19:22 That was directed to any and all.

01:19:23 Anybody else have a follow-up comment on that?

01:19:25 I would agree with Barry that a major component

01:19:28 in the early evolution of C2 catalyst was simply the fact

01:19:33 that if you're going to put two of something on that's asymmetric,

01:19:36 it's easy to do in the C2 system.

01:19:38 So there's an element of coincidence, simplicity,

01:19:41 and ease of synthesis that have come into those.

01:19:43 They're not necessarily deeply appropriate for these kinds of problems,

01:19:47 but they do work.

01:19:49 Okay.

01:19:50 Just a follow-up.

01:19:51 Obviously, in the enzymatic system, you're creating a chiral space

01:19:53 which is much, much larger than you are doing in a simple,

01:19:57 let's say, non-enzymatic-type reaction.

01:19:59 And so you're trying to simplify the chiral environment,

01:20:03 this much smaller chiral environment,

01:20:05 so that it is less ambiguous in the way that the substrate is going to dock

01:20:10 onto the metal.

01:20:12 Obviously, to the extent that you would build larger chiral space

01:20:15 in non-biological methods,

01:20:17 you would also have the opportunities for much more expansive types of molecules,

01:20:23 and C2 symmetry is not going to be a prerequisite.

01:20:26 All right.

01:20:28 We go now to Virginia Commonwealth University in Richmond

01:20:31 for our next question.

01:20:32 Go ahead, please.

01:20:33 Yeah, my question is to Barry Sharpless.

01:20:36 In your asymmetric dehydroxylation,

01:20:39 you have used osmium tetroxide and dihydroquinidine ligand.

01:20:43 Now my question is,

01:20:45 have you carried out this reaction using ruthenium tetroxide or any such compound?

01:20:51 And if so, what was the enantiomeric excess obtained?

01:20:55 Yes, we tried with ruthenium tetroxide.

01:20:58 The usual thing that we try to do is we screen around that part of the periodic table

01:21:02 where the catalyst metal shows its best activity

01:21:05 just to make sure we haven't missed something.

01:21:07 With ruthenium tetroxide, of course,

01:21:09 you have to accept the fact that it's going to cleave right through the double bond.

01:21:12 So we tried kinetic resolutions of chiral olefins in the presence of the alkaloid.

01:21:18 But frankly, I think the problem is that there the redox potential is so great

01:21:23 that it just oxidizes the quinucleidine nitrogen to the N-oxide

01:21:29 and it destroys the ligand as a potential ligand for the D0 ruthenium.

01:21:36 So you don't see any induction, or at least we don't see any induction.

01:21:40 We don't know if that's the explanation.

01:21:42 But there's technetium oxo's which work well.

01:21:45 Alan Davison's shown that.

01:21:47 You can make glycolates with those.

01:21:49 And that could be an interesting system because it has different geometry and ligands,

01:21:53 although people might not like working with technetium if they had a choice.

01:21:57 Then rhenium is showing some interesting reactions from several labs that are unpublished right now.

01:22:03 But osmium, unfortunately, is really the star element here for doing this hydroxylation chemistry.

01:22:11 And I think we're stuck with that unless there's a real surprise out there that I'm not betting on.

01:22:17 Okay, you have only a few more minutes now to get your call in.

01:22:20 So now's the time to do it.

01:22:21 800-368-5781 or 5782 in the D.C. area.

01:22:26 202-463-3170.

01:22:29 Our next question is from Allegheny College in Needville, Pennsylvania.

01:22:33 Go ahead, please.

01:22:34 Yes.

01:22:35 My question is for Peter Schultz.

01:22:38 As an organic chemist, if I designed what I think is a molecule for a transition state,

01:22:46 do you see in the future that we could send this away to be custom made for these antibodies?

01:22:55 Or will a wide variety of them be available?

01:23:00 If you have a transition state analog, I think in the near future it's going to be fairly straightforward

01:23:05 for you to make the monoclonal yourself if that's what you're asking about.

01:23:10 Right now there's a huge gap between,

01:23:12 and there's historically been a large gap between chemistry and immunology

01:23:16 because of the language barriers between the two fields.

01:23:19 But hybrid ELMA methods aren't that hard to learn.

01:23:23 They need special facilities that are available in any biology department.

01:23:28 Pretty soon the whole idea behind these phage displays of antibodies and so forth and so on

01:23:35 is to basically put the whole immune system of a mouse into a small test tube

01:23:40 that any lab can use for very low cost.

01:23:44 I think that technology is probably three years away

01:23:48 from perhaps replacing a lot of conventional hybrid ELMA methodology.

01:23:52 We have time for one more question, and I'll direct this to George Whitesides.

01:23:56 Is chiral synthesis really important in organic or pharmaceutical synthesis, and if so, why?

01:24:02 The argument for chiral synthesis I think is now overwhelming,

01:24:06 which is that the FDA is increasingly insisting on the testing and production of chirally pure compounds.

01:24:16 So I think all technology, whether it's biological or abiological,

01:24:20 that leads to improved methods of making chiral pieces

01:24:23 to go into the primary consumer of chiral pieces,

01:24:26 which is the pharmaceutical and the agricultural industries, is clearly going to be useful.

01:24:31 Okay, good. Before we close, we have some final comments right now

01:24:34 from Barry Trost, who organized this wonderful event today. Barry?

01:24:38 I'd like to say that we are trying to present to you the overview of what catalysis can do,

01:24:46 not from the point of view that we're discussing competition,

01:24:49 but from the point of view that these offer opportunities that are going to be complementary.

01:24:53 I really would like to take this opportunity of thanking publicly George Whitesides,

01:24:58 Peter Schultz, and Barry Sharpless in taking the amount of time

01:25:02 that all of us had to do to present both the tape and the live session here.

01:25:09 So to these three gentlemen, I am very thankful,

01:25:11 and I hope that you all have found this a useful session.

01:25:14 Thank you very much, Dr. Trost.

01:25:16 Well, we've come to the end of today's program,

01:25:18 the 10th American Chemical Society Satellite Television Seminar.

01:25:21 And thanks to our speakers, Peter Schultz, Barry Sharpless, George Whitesides,

01:25:25 for making this an interesting and informative program.

01:25:28 And a special thanks again to Barry Trost for organizing the program.

01:25:32 On behalf of the American Chemical Society Continuing Education Department,

01:25:35 I want to thank all of you for joining us today, too.

01:25:38 So please don't forget to complete and return your seminar evaluation forms.

01:25:42 We need your comments and suggestions.

01:25:44 Let us know if there are any particular topics that you'd like us to cover in future programs.

01:25:48 Our next program is on March 16, 1993,

01:25:52 on molecular modeling and the discovery of new drugs.

01:25:56 Please call the ACS Continuing Education Department for details.

01:25:59 Until then, I am Paul Anthony.

01:26:01 Have a very good day, and thank you.