Directed Evolution of Enantioselective Enzymes
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 99 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/18705 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Chymiatrie30 / 99
2
3
4
5
6
7
8
10
12
13
14
18
19
20
21
22
23
24
25
26
28
29
33
36
38
41
45
46
47
48
49
50
52
53
57
58
59
61
63
64
65
67
70
72
73
74
75
77
78
79
82
83
84
89
91
94
96
00:00
EnzymeAreaComputer animation
00:15
EnzymeEnantiomereStereoselektive SyntheseChemistryGene expressionGeneLibrary (computing)Repeated sequence (DNA)DNS-SyntheseSetzen <Verfahrenstechnik>MutageneseStructural steelHost (biology)StereoselectivityNutrientBrothTube (container)AgarChemistProcess (computing)Germanic peoplesPotenz <Homöopathie>HomologisierungOrganische ChemieGeneAreaHost (biology)CollectingChemical reactionChain (unit)BET theoryStereoselectivityOptical coherence tomographyWine tasting descriptorsAmino acidResistenzDNS-SyntheseBiogenesisPeptide sequenceMutageneseProteinStereoselektive SyntheseFood additiveUmweltchemikalieEnzymeHigh-throughput screeningMotion (physics)Sample (material)Tube (container)Library (computing)Molecular biologyHuman subject researchFoodHarvester (forestry)Structural steelCell (biology)BiosynthesisSeparation processScreening (medicine)GenotypeBlock (periodic table)Atom probeMan pageStoichiometryMolecularityEnantiomereSynthetic oilAntibacterialAgarProteaseSetzen <Verfahrenstechnik>Program flowchart
09:20
AcidEnzymeActive siteOxygenierungMutageneseChemistryAmineDNS-SyntheseChlorofluorocarbonBioavailabilitySubstrat <Chemie>EnantiomereHydrolaseEpoxideHamElectronic cigaretteAmino acidGoldEnzymeSubstrat <Chemie>Combine harvesterX-ray crystallographyActive siteEnantiomerentrennungWursthülleMixtureErdrutschOrganische ChemieSetzen <Verfahrenstechnik>BiotechnologySystemic therapyLactitolMutageneseChemical structureSolventReaction mechanismFood additiveLibrary (computing)OxygenierungWine tasting descriptorsBinding energyStereoselectivitySunscreenAageChemical reactionThermoformingPotenz <Homöopathie>PeriodateLeft-wing politicsEnzymkinetikHydrolaseGeneData conversionAldehydeTumorOctane ratingEnantiomereRacemizationEpoxideDiolMetabolic pathwayScreening (medicine)Lecture/ConferenceMeeting/Interview
18:24
Electronic cigaretteChemical experimentComputer animation
Transcript: English(auto-generated)
00:00
What made me go into a research area which is far away from synthetic organic chemistry?
00:23
You may know that I've worked in and done research in synthetic organic chemistry, developing methods for decades. So some people have asked me, what made you go into this completely different area which involves molecular biology,
00:41
enzymology and so on? Well, I thought about this question, and it turns out that the roots can be traced actually to my work here. When I came from Bonn to this university in 1980, my colleague Reinhard Hoffmann suggested that we offer,
01:04
to the Gesellschaftdeutsche Chemieke, the German Chemical Society, a course on stereoselective synthesis, methods in asymmetric catalysis and stoichiometric reactions. So I thought that's a nice idea, and I said yes, I'll do it.
01:23
And we asked another colleague, Professor Geis, at that time at Darmstadt University to also participate. And we divided the subjects up. And Professor Geis was responsible for enzymes as catalysts in synthetic organic chemistry.
01:44
And we offered this course every two years for a whole decade, and we were all present as the speakers, as the three speakers during the whole course. It was always a one-week course, eight hours a day.
02:00
So I learned a lot myself, not just the participants, which were students, industrial people, and so on, from all over Europe. It was really one of the best, in German, Vortbelungskorze der Gesellschaftdeutsche Chemieke. So I learned something about enzymes, and I had previously no knowledge of that subject whatsoever.
02:22
So I was sensitized. And then in 1994, I read a paper by chance in Nature entitled DNA Shuffling, written by a molecular biologist by the name of Wim Stemme. So I was curious, I read it, I didn't really understand the details,
02:41
but it was clear this has to do with directed evolution. And he was interested in antibiotic resistance, beta-lactamases, and so on, so I started to read the literature more on directed evolution. And I read the seminal paper by Frances Arnold at Caltech,
03:07
written and appeared in 1993. She went through several cycles of mutagenesis in order to increase the stability of a protease. So what does this actually mean?
03:22
What is directed evolution? That's the second question I have on my little list here. Everybody knows what evolution in nature is. It's a continuous cycle of mutagenesis, gene mutagenesis selection, gene mutagenesis selection.
03:42
It's a powerful driving force in nature. Nothing can be understood in biology in the absence of evolution. And it has been the dream of enzymologists and evolutionary researchers to simulate this process in the laboratory.
04:01
In other words, to perform evolution in the test tube. And this is what Stemme did, this is what Arnold performed, and we read this at a very early stage, and then this posed the question, can we harness this powerful force, namely evolution,
04:24
put it in the test tube in order to control a parameter which admittedly is not trivial, namely asymmetric catalysis, enantioselectivity. So let me begin with a cartoon. Here you see our new approach to asymmetric catalysis,
04:44
directed evolution of enantioselective enzymes. And on the right, upper part, you see a circle. It symbolizes a wild type enzyme, in other words, the enzyme that occurs in nature. It has poor selectivity, poor enantioselectivity
05:02
in the reaction that you or we may be interested in. So we take the gene, the square to the left, which encodes this enzyme and subject it to a gene mutagenesis method. And there are a number of these methods available,
05:21
which were developed in the 1980s, 1990s, even to this day. And such things as error prone, polymerase chain reaction, a shotgun method is the most popular method used to this day. Then we have DNA shuffling, which is a recombinant method.
05:43
I mentioned it already. Stemma, you take a gene or two genes, you slice them enzymatically into pieces, and you reassemble them. So this is simulating sexual evolution. Let me now show you some details, what is actually done in the laboratory.
06:05
Once you perform one of those methods on some gene which encodes an enzyme of interest to you, you have it in the test tube here, and then you transfer this collection of mutated genes
06:23
into a bacterial host such as E. coli, and you plate out on agar plates, on many agar plates. Here is symbolically the first plate, and after a while you see little colonies growing, each coming from a single cell, producing a mutant. You collect them, you harvest them, you give them food,
06:44
and they feel good. You put them individually into the wells of microtiter plates, and then you suddenly have hundreds and thousands of little factories producing potentially enantioselective enzymes.
07:05
So we used some of these methods such as error-prone PCR and also DNA shuffling in proof of principle studies using lipases, and we were able to increase the enantioselectivity to a notable degree.
07:23
If you do something completely new for the first time, it doesn't matter how efficient it is. So we were not concerned about practical ramifications or efficiency of what we're actually doing.
07:40
Now the challenge in directed evolution is to develop methods which allow you to probe protein sequence space efficiently. Now what do I mean by that? Consider, for example, an enzyme composed of 300 amino acids. If you introduce one mutation, one point mutation,
08:03
randomly, everywhere, at every position, and remember there are 20 building blocks, 20 different amino acids, you can calculate there are about 7,000 different mutants possible. If you introduce two mutations simultaneously, this jumps up to about 15 million, and three, 30 billion,
08:24
which are impossible to screen. And remember, how are you going to screen even 1,000 or 5,000 samples for enantiomeric purity? So this was one challenge, to develop high-throughput methods for EE determination.
08:42
I will not go into any details there. The other intellectual challenge is what I just addressed, namely methodology development. We published our proof of principle paper, others joined us, industrial companies also used this method
09:01
to create new catalysts for asymmetric catalysis, but we were not happy, really, with the methods. So let me now show you how to beat the so-called numbers problem in directed evolution, how to make small libraries, high-quality libraries, with less efforts.
09:22
That is the challenge. And our answer is shown on this slide here. We call it iterative saturation mutagenesis, ISM. You first make a decision regarding sites in the enzyme where you want to randomize.
09:40
So saturation mutagenesis is a method that I have not introduced yet, but it is as follows. You can choose, for example, one position anywhere in the enzyme. You define it and introduce randomly all 20 proteogenic amino acids there, and you get a library of all mutants.
10:02
There are 20 then. If you saturate or randomize, as people call it, each two positions simultaneously, it's 20 to the power of 2, 400, and so on. Three amino acid position sites would be 8,000.
10:20
So you need a decision where to randomize. So it's knowledge-driven, mechanism-driven, structure-driven, and we have developed criteria with which you can choose the right appropriate positions where to randomize.
10:42
But let's say, and I will show you in a minute what those criteria are. Let's say you have analyzed your system, and there are four sites, A, B, C, D. And just to make it less abstract, let's say A and B are sites composed
11:02
of two amino acid positions and C and D, three amino acid positions. And you can see on the slide that in the case of three, as I said, there are 8,000 mutants. It doesn't mean that you harvest the first 8,000 bacterial colonies and you have all of those.
11:21
There's a statistical argument and statistics according to which you have to do so-called oversampling. So you have to harvest many, many, many more if you insist on really screening all of those 8,000, for example. But if you look at the scheme, it looks a little complicated.
11:41
We make four libraries, screen them for enantioselectivity, and put the winner. As you can see, in each case, A, B, C, and D. Then the iterativity comes into play. We have changed the catalyst, the enzyme, structurally. And we take the gene that encodes this mutant enzyme
12:04
and then visit the other sites in the case of A, B, C, and D. And then we continue until we have visited all four sites. And then it converges. So when we set up this scheme, we did not dream how successful it would be.
12:25
This is the most efficient way to do directed evolution. Now, let's look at the criteria. We have to make the decisive choice where to randomize.
12:40
So it is a combination of, let's say, rational design and randomization. And that's shown on the next slide. We call this combinatorial active site saturation test, CAST. So it's a nice acronym, CASTING. For substrate scope and enantioselectivity.
13:01
And you simply look at the binding pocket and see which amino acid residues align this binding pocket. Those are our A, B, C, and Ds. And then we perform this systematically in the sense of iterativity, according to the scheme that I showed you on the last slide.
13:21
Now let's briefly look at our first example. And that's shown on the next slide here. This concerns a so-called kinetic resolution of a racemate. In this case, an epoxide. And we use an epoxide hydrolase.
13:42
So it's a racemate one-to-one mixture of R and S. And we only want one of the enantiomers to react to the diol. And that would leave, after 50% conversion, the other starting material untouched. And those can then be separated. So we perform the CAST analysis based on the X-ray structure.
14:04
And we came up with six sites, A, B, C, D, E, and F. Each composed of two or three amino acid positions. On the right side, you can see a cartoon which picture these six sites.
14:22
So the student performed the saturation mutagenesis six times, got six libraries, and the best hit came out of library B. And this was then used as a starting point. People call it a template to visit another site.
14:41
So the question is, where should you go? If you remember, the dendritic scheme was somewhat complicated. Today, we know it really doesn't matter which pathway you take. But on the next slide, you see the results. We have the wild type. This is the selectivity factor,
15:01
the relative rate of one with respect to the other enantiomer. And you see the best came out of B. This has a selectivity factor of 14. Then the student visited C, D, F, and E. And we came up with a mutant, LW202,
15:21
which has a selectivity factor of 115. And we only had to look at 20,000 reactions, which happens to be the same number that we had already screened using the old strategies, error-prone PCR, in an older study. But the results there were very, very poor.
15:42
We could only double it, and here we have multiplied it by a factor of 20 or 25. An interesting question concerns the problem of identifying the reason for enhanced enantioselectivity.
16:02
So we just recently published a paper concerning the question in this specific case, and it's a long story. You can read it in the Journal of American Chemical Society, 2009, just a few months ago. But I only want to show you one little thing
16:22
that is very important, namely the X-ray structure of the wild type, the starting enzyme, and the X-ray structure of the evolved enantioselective one. If you look at the two X-ray structures, they're essentially identical. But if you zoom in into the binding pocket,
16:42
and let's take a look at that here. On the left you see the wild type, and these are our sites, and here's the binding pocket. It's kind of a narrow tunnel. And if you just take a quick look on the right, you see the best mutant, LW2O2, and I think you'll admit it's completely different.
17:02
So the shape of the binding pocket has been changed to such an extent that only one enantiomer fits in and reacts. The other one does not react and does not fit into the binding pocket.
17:20
I could show you now a movie of this whole thing. I'm going to leave that off, and we've done kinetics and many other experiments. You can find details here. So now I'm more or less at the end of this video. I hope that you enjoyed this little adventure accompanying me.
17:43
I hope the basic principles are clear now. The ramifications are far-reaching, and it's not just an antioselectivity substrate scope, but we can also handle thermal stability and stability against hostile solvents.
18:03
So those are the most important parameters for real applications. Those are the traditional, the historical limitations of enzymes as catalysts in biotechnology and also in synthetic organic chemistry.
18:22
I hope you enjoyed this little trip.