Engineered Artificial Carboligases Facilitate Regioselective Preparation of Enantioenriched Aldol Adducts

Article abstract of DOI:10.1021/jacs.0c02351

Controlling regio- A nd stereoselectivity of aldol additions is generally challenging. Here we show that an artificial aldolase with high specificity for acetone as the aldol donor can be reengineered via single active site mutations to accept linear and cyclic aliphatic ketones with notable efficiency, regioselectivity, and stereocontrol. Biochemical and crystallographic data show how the mutated residues modulate the binding and activation of specific aldol donors, as well as their subsequent reaction with diverse aldehyde acceptors. Broadening the substrate scope of this evolutionarily na?ve catalyst proved much easier than previous attempts to redesign natural aldolases, suggesting that such proteins may be excellent starting points for the development of customized biocatalysts for diverse practical applications.

Full text of DOI:10.1021/jacs.0c02351

Subscriber access provided by Uppsala universitetsbibliotek
Communication
Engineered artificial carboligases facilitate regioselective
preparation of enantioenriched aldol adducts
Duncan Stuart Macdonald, Xavier Garrabou, Cindy Klaus, Rebecca Verez, Takahiro Mori, and Donald Hilvert
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c02351 • Publication Date (Web): 19 May 2020
Downloaded from pubs.acs.org on May 19, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
Duncan S. Macdonald‡, Xavier Garrabou‡, Cindy Klaus, Rebecca Verez, Takahiro Mori, and Donald
Hilvert*
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Laboratory of Organic Chemistry, ETH Zurich, 8093 Zurich, Switzerland
‡ These authors contributed equally to this work
Supporting Information Placeholder
ABSTRACT: Controlling regio- and stereoselectivity of aldol
additions is generally challenging. Here we show that an artifi-
cial aldolase with high specificity for acetone as the aldol donor
can be reengineered via single active site mutations to accept
linear and cyclic aliphatic ketones with notable efficiency, regi-
oselectivity and stereocontrol. Biochemical and crystallo-
graphic data show how the mutated residues modulate the
binding and activation of specific aldol donors, as well as their
subsequent reaction with diverse aldehyde acceptors. Broad-
ening the substrate scope of this evolutionarily naïve catalyst
proved much easier than previous attempts to redesign natural
aldolases, suggesting that such proteins may be excellent start-
ing points for the development of customized biocatalysts for
diverse practical applications.
The aldol reaction represents a methodological cornerstone
of synthetic organic chemistry, allowing for direct C-C bond
formation under mild conditions. Achieving regio- and ste-
reocontrol in aldol additions of unmodified ketones, how-
ever, remains challenging with small-molecule catalysts.
Enzymes offer a potentially promising alternative, due to
their inherent chirality and precise substrate recognition,
but engineering of natural aldolases to meet synthetic needs
is far from trivial.1–4 Natural class I aldolases are highly
evolved for their native reaction and have proven resistant
to efforts to change donor specficity.1 Because artificial en-
zymes have little historical bias, they are potentially more
amenable to engineering by laboratory evolution.5–8
Figure 1: (A) Retroaldol cleavage of (R)-methodol,21 catalyzed
by RA95.5-8F.18 (B) Structure of RA95.5-8F in complex with a
diketone inhibitor (green). Mechanistically relevant residues
are shown in gray, residues mutated in this study in cyan. PDB
code: 5an7.
otherwise formidable challenge for conventional synthetic
methods requiring complex catalysts in relatively high load-
ing.19,20
A family of computationally designed carboligases has
shown remarkably high and tunable activity for diverse re-
actions that proceed via Schiff base intermediates.9–16 The
aldolase RA95.5-8F (Figure 1), obtained by design and ul-
trahigh-throughput evolution,17,18 offers native-like cata-
lytic rates, good to excellent stereoselectivity and high total
turnover numbers for the addition of acetone to a variety of
aliphatic and aromatic aldehydes. As for natural aldolases,
though, this artificial carboligase poorly tolerates alterna-
tive donors. Here we show that single active site mutations
can considerably expand the donor specificity of RA95.5-8F.
The resulting variants efficiently catalyze additions of cyclic
ketones and n-alkan-2-ones to aldehydes with complete re-
gioselectivity and control of up to three chiral centers, an
To create RA95.5-8F variants with broader substrate
scope, four residues lining the binding pocket for the aldol
donor (Phe112, Leu131, Ile133, and Leu159) and three ad-
ditional residues (Phe184, Leu210, and Met231) that pack
against the catalytic quartet were targeted independently
for cassette mutagenesis (Figure 1). The resulting libraries
were screened for cleavage of diastereomeric mixtures of
aldol adducts of cyclopentanone and cyclohexanone with 6-
methoxy-2-naphthaldehyde (Scheme 1A). Reaction pro-
gress in crude cell lysates was monitored spectrophotomet-
rically at 350 nm in multiwell plates. Although all of the li-
braries proved inactive toward the cyclopentanone ad-
ducts, the F112 library afforded several hits for the cyclohe-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
Scheme 1: (A) Screening reactions (B) Aldol adducts pro-
duced biocatalytically with F112I/L/V RA95.5-8F. (C) Ste-
reoselective deuteuration of cyclohexanone.
trobenzaldehyde afforded 1b as the dominant product ste-
1
2
3
4
5
6
7
8
reoisomer. In this case, the anti/syn and enantiomeric ratios
were 83:17 and >99:1, respectively. For the reaction of cy-
clohexanone and 3-phenylpropanal, depending on the
catalyst employed, 1c was obtained with anti/syn and
enantiomeric ratios up to 91:9 and 98:2. For comparison,
the parental enzyme achieved only modest yields (15%)
and diastereoselectivities (85:16) for this reaction.18 The
F112V variant, which has the smallest side chain, even ac-
cepts substituted cyclohexanones as aldol donors. Reaction
of 4-methyl- and 4-ethycyclohexanone with 4-nitrobenzal-
dehyde afforded aldol adducts 2 and 3 with excellent stere-
ocontrol at three chiral centers (Scheme 1B).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because the engineered aldolases exploit amine cataly-
sis, cyclohexanone should undergo facile deuteration via its
Schiff base and enamine adducts with Lys83 (Scheme 1C).22
As expected, the F112I/L/V variants all promote H/D-ex-
change in buffered D2O (Figure S17). Isolation of the result-
ing monodeuterated cyclohexanone and determination of
its specific optical rotation ([훼]2퐷6= +2.2° (c = 0.41, CDCl3))
revealed that one of the two pro-R -hydrogens exchanges
first.23 The enzyme-bound enamine is thus preferentially
deuterated on the same face as the aldehyde acceptors add
in the synthetic aldol reactions. Unexpectedly, though, the
rates of H/D exchange determined by 1H-NMR spectroscopy
were at most twofold faster than that observed for the par-
ent enzyme RA95.5-8F, indicating that the enhanced syn-
thetic ability of the variants derives from a beneficial effect
on a later, slower step in the catalytic mechanism, most
likely C-C bond formation.
xanone derivatives. Rapid cleavage of the larger substrate
was made possible by replacement of Phe112 with the
smaller aliphatic amino acids valine, leucine, and isoleucine.
Having successfully engineered RA95.5-8F to accept
bulky, cyclic ketones as aldol donors, our attention turned
to extended linear ketones. The regioselective preparation
of linear aldol adducts from asymmetric ketones has been
described, for example using stoichiometric amounts of pol-
ymer-bound lithium dialkylamides.24 However, simultane-
ous control over regio- and stereochemistry remains chal-
lenging,25–27 requiring costly organocatalysts at high cata-
lyst loading.28,29 We anticipated that the inexpensive and tai-
lorable enzyme would be selective in both properties at
sub-stoichiometric catalyst loadings. HPLC screening of the
seven aldolase libraries directly for the reaction of 2-butan
one and 3-phenylpropanal yielded a single hit, which had
the I133F mutation located at the bottom of the donor bind-
The F112V, F112L and F112I variants were overpro-
duced and purified. The crystal structure of the apo F112L
variant, solved to 2.16 Å resolution, confirmed that the ef-
fects of the large-to-small mutation are largely local. The
protein backbone aligns well with the parent enzyme (c
r.m.s. deviation of 0.336 Å) and is only distinct in the previ-
ously unresolved L1 loop (Figure S28). The leucine substi-
tution, combined with subtle repositioning of Phe89, opens
up the binding pocket, providing space for sterically more
demanding substrates.
Kinetic measurements with aldol adduct 1a (Scheme
1B), prepared enzymatically, showed that mutation of
Phe112 improved catalytic efficiency by more than an order
of magnitude, largely through improvements in kcat (Table
S5). Indeed, the steady-state parameters for this substrate
(kcat = 7 to 9 s-1 and kcat/KM = 7 x 104 to 1.1 x 105 M-1 s-1) are
comparable to those for the RA95.5-8F-catalyzed cleavage
of (R)-methodol18 (kcat = 11 s-1 and kcat/KM = 3.4 x 104 M-1 s-
1).
Scheme 2: (A) Aldol synthesis with linear ketones. (B)
Aldol adducts prepared at synthetic scale with I133F
RA95.5-8F.
In addition to aldol cleavage, all three variants promote
aldol synthesis with high stereoselectivity. For example,
F112I RA95.5-8F catalyzes the addition of cyclohexanone to
6-methoxy-2-naphthaldehyde, the microscopic reverse of
the screening reaction, to produce 1a with an anti/syn ratio
of 97:3 and an enantiomeric ratio (e.r.) of 98:2. As for the
parent enzyme, the observed stereochemistry can be ra-
tionalized by attack of the cyclohexanone donor, activated
as an enamine by Lys83, on the re-face of the aldehyde ac-
ceptor. Similarly, the reaction of cyclohexanone with 4-ni-
2
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
differential enolization cannot explain why I133F is the only
1
2
variant that is useful for aldol synthesis with the extended
linear ketones.
3
4
5
6
7
8
9
To understand the synthetic prowess of this variant, we
determined its crystal structure in complex with 2-penta-
none to 1.90 Å resolution. Comparison with the parent en-
zyme showed how the I133F mutation reshaped the donor
binding pocket (Figure 2B). Although phenylalanine is
larger than the original isoleucine at position 133, its
sidechain adopts a rotameric conformation that creates ad-
ditional space at the bottom of the cavity. Accompanying ad-
justments in the position of neighboring residues, particu-
larly Phe184 which shifts by 2.5 Å, further opens the pocket.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the complex, 2-pentanone is bound to Lys83 as a
Schiff base, but its propyl group extends toward the mouth
of the pocket where the acceptor aldehyde must bind (Fig-
ures 2B & S31). A water molecule, bound by the catalytic
residues Tyr51, Asn110, and Tyr180, is well placed (3.2 Å)
to either hydrolyze the Schiff base or abstract the C-3 pro-R
hydrogen (Figure S29). However, since we only observed al-
dol additions at C-1 of this ketone, the enamine resulting
from C-3 deprotonation cannot be productive for C-C bond
formation, presumably because the propyl chain sterically
blocks access of the aldehyde acceptor (Figure S31). To
form 5, the bound 2-pentanone would have to flip by ~180°
to position the C-1 methyl for deprotonation and subse-
quent attack on the incoming 3-phenylpropanal. The same
applies for the smaller 2-butanone, where the relative rates
of H/D-exchange at C-1 and C-3 (Table S7) indicate that the
ratio of productive to unproductive poses is ~1:3.
Figure 2: (A) Mechanism of regioselective enamine formation
by the I133F enzyme, as observed in deuteuration studies. (B)
Structure of I133F RA95.5-8F (PDB code: 6TF8) covalently
bound to pentan-2-one (purple) with observed water mole-
cules (red). The overlaid parental enzyme (gray) is complexed
with a diketone inhibitor (green).
Modeling the Schiff base adducts of 4c and 5 using the
Rosetta software suite30 confirmed that both linear prod-
ucts are readily accommodated in the remodeled active site.
These molecules can adopt extended conformations similar
to that seen for the diketone inhibitor in the parent enzyme
(Figure S32-S33), with slight adjustments in the position of
the catalytic lysine helping to accommodate the bulkier do-
nor in the expanded pocket. The docking models further in-
dicate that a productive hydrogen bond between the prod-
uct hydroxyl group and Tyr51 likely dictates the enzyme’s
stereochemical preferences. Although the disfavored enan-
tiomer also fits in the active site, it cannot make an analo-
gous interaction for steric reasons.
ing pocket. Biocatalytic transformations with the purified
enzyme confirmed that the I133F variant confers complete
regiocontrol over the addition of 2-butanone to 3-phe-
nylpropanal, providing exclusively the normally disfavored
linear regioisomer, adduct (S)-4a, with an e.r. of 95:5
(Scheme 2). Even higher e.r. values (98:2) were obtained
with aromatic aldehydes such as 6-methoxy-2-naphthalde-
hyde and 4-nitrobenzaldehyde. Again, only the linear regi-
oisomers 4b and 4c were observed. Notably, the stereose-
lectivities achieved are similar to those reported for acetone
using RA95.5-8F as catalyst. With larger aldol donors, how-
ever, stereoselectivity erodes. 2-Pentanone reacts with
3-phenylpropanal to afford the linear aldol adduct (S)-5
with an e.r. of 88:12, whereas 2-hexanone showed no con-
version, suggesting that the engineered active site sterically
restricts donor chain length.
Carboligases are attractive catalysts for preparing chiral
building blocks for complex natural products and other
molecules.31–36 In practical terms, however, the application
of natural aldolases in industrial processes has been limited
relative to other enzyme classes such as hydrolases, oxi-
doreductases, and transaminases,8 owing in part to their
narrow donor specificity.2,37 Despite some successes,3,38 ex-
panding the substrate scope of these catalysts has proved
challenging because substrate recognition and catalysis are
usually entangled in complex networks of polar residues.1,39
The unprecedented promiscuity15,16,40 and mutational
tolerance14,18 of artificial aldolases created by design and
evolution offers a potential alternative. In these systems,
substrate recognition is largely independent of the catalytic
apparatus, so successful discrimination between different
aldol adducts only requires creation of complementary
pockets of appropriate size, shape and polarity for the ap-
pended groups, a considerably easier task than redesigning
Steady-state kinetics showed that (R)-4b is efficiently
cleaved by I133F RA95.5-8F (kcat = 4.8 s-1, kcat/KM = 3.4 x 104
M-1 s-1). Although the parent enzyme and the F112I variant
are both poor catalysts for the synthesis of this aldol adduct,
they unexpectedly promote its cleavage with comparable or
even slightly higher catalytic efficiency (Table S6). Con-
sistent with this observation, deuteration experiments with
2-butanone indicated that both RA95.5-8F and the I133F
variant catalyze H/D-exchange at both C-1 and C-3 (Figures
2A and S25-S27). No difference was observed in the rate of
deuteration of C-1 between the two enzymes (Table S7), so
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
Grosser, S. T.; Halsey, H. M.; Hughes, G. J.; Jo, J.; Joyce, L. A.; Kolev, J. N.;
hydrogen bonding networks. In favorable cases, as shown
in the current study, large changes in substrate preference
can even be realized by single active site mutations. The en-
suing control over multiple stereocenters and regioselec-
tive formation of normally disfavored linear aldol adducts
from linear ketones adds to the repertoire of tools available
to synthetic chemists looking for efficient ways to control C-
C bond formation. Although more extensive engineering
may be needed to adapt RA95 aldolases for other sub-
strates, particularly polar ketones, ultrahigh-throughput
screening methods may expedite the search for useful vari-
ants.18,41–43
Liang, J.; Maloney, K. M.; Mann, B. F.; Marshall, N. M.; McLaughlin, M.;
Moore, J. C.; Murphy, G. S.; Nawrat, C. C.; Nazor, J.; Novick, S.; Patel, N. R.;
Rodriguez-Granillo, A.; Robaire, S. A.; Sherer, E. C.; Truppo, M. D.;
Whittaker, A. M.; Verma, D.; Xiao, L.; Xu, Y.; Yang, H. Design of an in Vitro
Biocatalytic Cascade for the Manufacture of Islatravir. Science 2019,
366, 1255–1259.
1
2
3
4
5
6
7
8
(5)
Kiss, G.; Çelebi-Ölçüm, N.; Moretti, R.; Baker, D.; Houk, K. N.
Computational Enzyme Design. Angew. Chemie Int. Ed. 2013, 52, 5700–
5725.
(6)
Computational Design. Curr. Opin. Chem. Biol. 2013, 17, 221–228.
(7) Hilvert, D. Design of Protein Catalysts. Annu. Rev. Biochem.
2013, 82, 447–470.
Kries, H.; Blomberg, R.; Hilvert, D. De Novo Enzymes by
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(8)
Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.;
Moore, J. C.; Robins, K. Engineering the Third Wave of Biocatalysis.
Nature 2012, 485, 185–194.
(9)
Jiang, L.; Althoff, E. A.; Clemente, F. R.; Doyle, L.;
Röthlisberger, D.; Zanghellini, A.; Gallaher, J. L.; Betker, J. L.; Tanaka, F.;
Barbas, C. F.; Hilvert, D.; Houk, K. N.; Stoddard, B. L.; Baker, D. De Novo
Computational Design of Retro-Aldol Enzymes. Science 2008, 319,
1387–1391.
(10) Althoff, E. A.; Wang, L.; Jiang, L.; Giger, L.; Lassila, J. K.; Wang,
Z.; Smith, M.; Hari, S.; Kast, P.; Herschlag, D.; Hilvert, D.; Baker, D. Robust
Design and Optimization of Retroaldol Enzymes. Protein Sci. 2012, 21,
717–726.
(11) Bjelic, S.; Kipnis, Y.; Wang, L.; Pianowski, Z.; Vorobiev, S.; Su,
M.; Seetharaman, J.; Xiao, R.; Kornhaber, G.; Hunt, J. F.; Tong, L.; Hilvert,
D.; Baker, D. Exploration of Alternate Catalytic Mechanisms and
Optimization Strategies for Retroaldolase Design. J. Mol. Biol. 2014,
426, 256–271.
(12) Wagner, J.; Lerner, R. A.; Barbas, C. F. Efficient Aldolase
Catalytic Antibodies That Use the Enamine Mechanism of Natural
Enzymes. Science (80-. ). 1995, 270, 1797.
(13) Garrabou, X.; Beck, T.; Hilvert, D. A Promiscuous De Novo
Retro-Aldolase Catalyzes Asymmetric Michael Additions via Schiff Base
Intermediates. Angew. Chemie Int. Ed. 2015, 54, 5609–5612.
(14) Garrabou, X.; Macdonald, D. S.; Wicky, B. I. M.; Hilvert, D.
Stereodivergent Evolution of Artificial Enzymes for the Michael
Reaction. Angew. Chemie Int. Ed. 2018, 57, 5288–5291.
(15) Garrabou, X.; Verez, R.; Hilvert, D. Enantiocomplementary
Synthesis of γ-Nitroketones Using Designed and Evolved Carboligases.
J. Am. Chem. Soc. 2017, 139, 103–106.
(16) Garrabou, X.; Macdonald, D. S.; Hilvert, D. Chemoselective
Henry Condensations Catalyzed by Artificial Carboligases. Chem. - A
Eur. J. 2017, 23, 6001–6003.
(17) Giger, L.; Caner, S.; Obexer, R.; Kast, P.; Baker, D.; Ban, N.;
Hilvert, D. Evolution of a Designed Retro-Aldolase Leads to Complete
Active Site Remodeling. Nat. Chem. Biol. 2013, 9, 494–498.
(18) Obexer, R.; Godina, A.; Garrabou, X.; Mittl, P. R. E.; Baker, D.;
Griffiths, A. D.; Hilvert, D. Emergence of a Catalytic Tetrad during
Evolution of a Highly Active Artificial Aldolase. Nat. Chem. 2017, 9, 50–
56.
(19) Fotaras, S.; Kokotos, C. G.; Tsandi, E.; Kokotos, G.
Prolinamides Bearing Thiourea Groups as Catalysts for Asymmetric
Aldol Reactions. European J. Org. Chem. 2011, 1310–1317.
(20) Cobb, A. J. A.; Shaw, D. M.; Longbottom, D. A.; Gold, J. B.; Ley,
S. V. Organocatalysis with Proline Derivatives: Improved Catalysts for
the Asymmetric Mannich, Nitro-Michael and Aldol Reactions. Org.
Biomol. Chem. 2005, 3, 84–96.
(21) List, B.; Barbas, C. F.; Lerner, R. A. Aldol Sensors for the Rapid
Generation of Tunable Fluorescence by Antibody Catalysis. Proc. Natl.
Acad. Sci. U. S. A. 1998, 95, 15351–15355.
(22) Zeymer, C.; Zschoche, R.; Hilvert, D. Optimization of Enzyme
Mechanism along the Evolutionary Trajectory of a Computationally
Designed (Retro-)Aldolase. J. Am. Chem. Soc. 2017, 139, 12541–12549.
(23) Dauphin, G.; Cramain, J. C.; Kergomard, A.; Renard, M. F.;
Veschambre, H. Microbiological Synthesis and Circular Dichroism of
Optically Active 2-Deuterio-Cycloalkanones. Tetrahedron Lett. 1980,
21, 4275–4278.
(24) Seki, A.; Ishiwata, F.; Takizawa, Y.; Asami, M. Crossed Aldol
Reaction Using Cross-Linked Polymer-Bound Lithium Dialkylamide.
Tetrahedron 2004, 60, 5001–5011.
The Supporting Information is available free of charge on
the ACS Publications website.
Complete experimental procedures, with sequence infor-
mation, additional kinetic data, crystallographic data, code
for running Rosetta modeling, as well as Figures S1-S33 and
Tables S1-S9 (PDF)
*hilvert@org.chem.ethz.ch
ORCID
Duncan S. Macdonald: 0000-0001-9467-6658
Xavier Garrabou: 0000-0003-3663-8420
Cindy Klaus: 0000-0001-5977-1158
Takahiro Mori: 0000-0002-2754-5858
Donald Hilvert: 0000-0002-3941-621X
Author Contributions
D.S.M. and X.G. contributed equally to this work.
The authors declare no competing financial interests.
The authors are grateful to the Swiss National Science Foun-
dation and the ETH Zurich for generous support of this
work. We also acknowledge Sabine Österle, Andreas Kaspar
and the students of OCPI class of 2019 for their contribu-
tions.
(1)
Aldolases as Biocatalysts. Curr. Opin. Chem. Biol. 2014, 19, 25–33.
(2) Hernández, K.; Szekrenyi, A.; Clapés, P. Nucleophile
Windle, C. L.; Müller, M.; Nelson, A.; Berry, A. Engineering
Promiscuity of Natural and Engineered Aldolases. ChemBioChem 2018,
19, 1353–1358.
(3)
Soler, A.; Gutiérrez, M. L.; Bujons, J.; Parella, T.; Minguillon,
C.; Joglar, J.; Clapés, P. Structure-Guided Engineering of D-Fructose-6-
Phosphate Aldolase for Improved Acceptor Tolerance in Biocatalytic
Aldol Additions. Adv. Synth. Catal. 2015, 357, 1787–1807.
(4)
Huffman, M. A.; Fryszkowska, A.; Alvizo, O.; Borra-Garske,
M.; Campos, K. R.; Canada, K. A.; Devine, P. N.; Duan, D.; Forstater, J. H.;
4
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
(25) Zhang, X.; Zhao, W.; Qu, C.; Yang, L.; Cui, Y. Efficient
Asymmetric Aldol Reaction Catalyzed by Polyvinylidene Chloride-
Supported Ionic Liquid/l-Proline Catalyst System. Tetrahedron:
Asymmetry 2012, 23, 468–473.
to Aromatic Aldehydes Catalyzed by Proline-Based Carboligases. ACS
Catal. 2020, 10, 2522–2527.
1
2
3
4
5
6
7
8
(36) Chen, Q.; Chen, X.; Cui, Y.; Ren, J.; Lu, W.; Feng, J.; Wu, Q.; Zhu,
D. A New D-Threonine Aldolase as a Promising Biocatalyst for Highly
Stereoselective Preparation of Chiral Aromatic β-Hydroxy-α-Amino
Acids. Catal. Sci. Technol. 2017, 7, 5964–5973.
(37) Clapés, P.; Garrabou, X. Current Trends in Asymmetric
Synthesis with Aldolases. Adv. Synth. Catal. 2011, 353, 2263–2283.
(38) Güclü, D.; Szekrenyi, A.; Garrabou, X.; Kickstein, M.; Junker,
S.; Clapés, P.; Fessner, W. D. Minimalist Protein Engineering of an
Aldolase Provokes Unprecedented Substrate Promiscuity. ACS Catal.
2016, 6, 1848–1852.
(39) Sugiyama, M.; Hong, Z.; Liang, P. H.; Dean, S. M.; Whalen, L. J.;
Greenberg, W. A.; Wong, C. H. D-Fructose-6-Phosphate Aldolase-
Catalyzed One-Pot Synthesis of Iminocyclitols. J. Am. Chem. Soc. 2007,
129, 14811–14817.
(40) Garrabou, X.; Wicky, B. I. M.; Hilvert, D. Fast Knoevenagel
Condensations Catalyzed by an Artificial Schiff-Base-Forming Enzyme.
J. Am. Chem. Soc. 2016, 138, 6972–6974.
(41) Debon, A.; Pott, M.; Obexer, R.; Green, A. P.; Friedrich, L.;
Griffiths, A. D.; Hilvert, D. Ultrahigh-Throughput Screening Enables
Efficient Single-Round Oxidase Remodelling. Nat. Catal. 2019, 2, 740–
747.
(42) Obexer, R.; Pott, M.; Zeymer, C.; Griffiths, A. D.; Hilvert, D.
Efficient Laboratory Evolution of Computationally Designed Enzymes
with Low Starting Activities Using Fluorescence-Activated Droplet
Sorting. Protein Eng. Des. Sel. 2016, 29, 355–366.
(43) Bunzel, H. A.; Garrabou, X.; Pott, M.; Hilvert, D. Speeding up
Enzyme Discovery and Engineering with Ultrahigh-Throughput
Methods. Current Opinion in Structural Biology. Elsevier Ltd February
1, 2018, pp 149–156.
(26) Bañõn-Caballero, A.; Guillena, G.; Nájera, C. Cross-Linked-
Polymer-Supported N-{2’-[(Arylsulfonyl)Amino][1,1’- Binaphthalen]-
2-Yl}prolinamide as Organocatalyst for the Direct Aldol Intermolecular
Reaction under Solvent-Free Conditions. Helv. Chim. Acta 2012, 95,
1831–1841.
(27) Peng, Y.-Y.; Ding, Q.-P.; Li, Z.; Wang, P. G.; Cheng, J.-P. Proline
Catalyzed Aldol Reactions in Aqueous Micelles: An Environmentally
Friendly Reaction System. Tetrahedron Lett. 2003, 44, 3871–3875.
(28) Guillena, G.; del Carmen Hita, M.; Nájera, C. BINAM-
Prolinamides as Recoverable Catalysts in the Direct Aldol
Condensation. Tetrahedron: Asymmetry 2006, 17, 729–733.
(29) Pluta, R.; Kumagai, N.; Shibasaki, M. Direct Catalytic
Asymmetric Aldol Reaction of α-Alkoxyamides to α-Fluorinated
Ketones. Angew. Chemie 2019, 131, 2481–2485.
(30) Rohl, C. A.; Strauss, C. E. M.; Misura, K. M. S.; Baker, D. Protein
Structure Prediction Using Rosetta. Methods Enzymol. 2004, 383, 66–
93.
(31) Schmidt, N. G.; Eger, E.; Kroutil, W. Building Bridges:
Biocatalytic C–C-Bond Formation toward Multifunctional Products.
ACS Catal. 2016, 6, 4286–4311.
(32) Rej, R. K.; Das, T.; Hazra, S.; Nanda, S. Chemoenzymatic
Asymmetric Synthesis of Fluoxetine, Atomoxetine, Nisoxetine, and
Duloxetine. Tetrahedron: Asymmetry 2013, 24, 913–918.
(33) Fátima, Â. de; Lapis, A. A. M.; Pilli, R. A. A Concise Total
Synthesis of (R)-Fluoxetine, a Potent and Selective Serotonin Reuptake
Inhibitor. J. Braz. Chem. Soc. 2005, 16, 495–499.
(34) Sabitha, G.; Padmaja, P.; Yadav, J. S. A Concise Total Synthesis
of Diospongins A and B. Helv. Chim. Acta 2008, 91, 2235–2239.
(35) Saifuddin, M.; Guo, C.; Biewenga, L.; Saravanan, T.; Charnock,
S. J.; Poelarends, G. J. Enantioselective Aldol Addition of Acetaldehyde
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1: (A) Retroaldol cleavage of (R)-methodol,21 catalyzed by RA95.5-8F.18 (B) Structure of RA95.5-8F
in complex with a diketone inhibitor (green). Mechanistically relevant residues are shown in gray, residues
mutated in this study in cyan. PDB code: 5an7.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1: (A) Screening reactions (B) Aldol adducts pro-duced biocatalytically with F112I/L/V RA95.5-8F.
(C) Stereoselective deuteuration of cyclohexanone.
390x437mm (96 x 96 DPI)
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: (A) Mechanism of regioselective enamine formation by the I133F enzyme, as observed in
deuteuration studies. (B) Structure of I133F RA95.5-8F (PDB code: 6TF8) covalently bound to pentan-2-one
(purple) with observed water molecules (red). The overlaid parental enzyme (gray) is complexed with a
diketone inhibitor (green).
83x110mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2: (A) Aldol synthesis with linear ketones. (B) Aldol adducts prepared at synthetic scale with I133F
RA95.5-8F.
404x243mm (96 x 96 DPI)
ACS Paragon Plus Environment