Toward chemistry-based design of the simplest metalloenzyme-like catalyst that works efficiently in water

Article abstract of DOI:10.1002/asia.201403004

Enzymes exhibit overwhelmingly superior catalysis compared with artificial catalysts. Current strategies to rival enzymatic catalysis require unmodified or minimally modified structures of active sites, gigantic molecular weight, and sometimes the use of harsh conditions such as extremely low temperatures in organic solvents. Herein, we describe a design of small molecules that act as the simplest metalloenzyme-like catalysts that can function in water, without mimicking enzyme structures. These artificial catalysts efficiently promoted enantioselective directtype aldol reactions using aqueous formaldehyde. The reactions followed Michaelis-Menten kinetics, and heat-resistant asymmetric environments were constructed in water.

Full text of DOI:10.1002/asia.201403004

DOI: 10.1002/asia.201403004
Communication
Aldol Reaction
|Very Important Paper|
Toward Chemistry-Based Design of the Simplest Metalloenzyme-
Like Catalyst That Works Efficiently in Water
Taku Kitanosono and Shu Kobayashi*[a]
¯
of surfactants in water or the use of organic solvents as cosol-
Abstract: Enzymes exhibit overwhelmingly superior catal-
ysis compared with artificial catalysts. Current strategies to
vents.[8–13] The staggering difficulties that thwart the realization
of the ultimate goal are mainly attributed to the problems of
rival enzymatic catalysis require unmodified or minimally
aqueous conditions: weak, noncovalent interactions among
modified structures of active sites, gigantic molecular
substrates, a chiral ligand, and a metal ion under competitive
weight, and sometimes the use of harsh conditions such
polar conditions.[14] In addition, the solubility of a designed cat-
as extremely low temperatures in organic solvents. Herein,
alyst in water is another problem. The utilization of host mole-
we describe a design of small molecules that act as the
cules, such as protein scaffolds,[15] cyclodextrins,[16] crown
simplest metalloenzyme-like catalysts that can function in
ethers,[17] and calixarenes,[18,19] has often enabled researchers to
water, without mimicking enzyme structures. These artifi-
construct a hydrophobic cavity in water and to accelerate the
cial catalysts efficiently promoted enantioselective direct-
reaction rate as enzymes do.[20] However, the modified host
type aldol reactions using aqueous formaldehyde. The re-
molecules still have high molecular weights, and there are
actions followed Michaelis–Menten kinetics, and heat-re-
many limitations to the application of “chemozymes” involving
sistant asymmetric environments were constructed in
host molecules. Furthermore, little valuable rationale has ever
water.
been obtained for the resulting catalytic activity despite recent
advances in computational design.[21] We therefore considered
that an understanding of the principles of enzyme catalysis
Enzymatic reactions in existing organisms are completely
chemo-, regio-, and stereoselective, even at low concentra-
tions, without any protection methodology under remarkably
mild conditions, and in neutral aqueous solutions at atmos-
pheric pressure and temperature.[1] Even though a variety of
enzymatic reactions have been used extensively in industry,[2]
there are some practical drawbacks: enzymes are basically
labile and it is difficult to preserve their catalytic activity.[3] Until
recently, to obtain new enzymes, natural enzymes have been
taken and evolved by mutation or modification of their active
sites (protein engineering)[4] or by catalytic antibodies.[5] How-
ever, these methodologies cannot offer a fundamental pre-
scription to overcome the disadvantages of enzymes. Hence,
chemists can dream of outdoing 3.8 billion years of evolution
and natural selection by a clean-sheet design that is not based
on enzymatic motifs from nature. It is the ultimate goal for
chemists to design “privileged” catalysts that outdo enzymes.
For example, such catalysts might perform predefined tasks via
natural or unnatural reactions in water.[6]
must surmount those problems and accomplish the formidable
task. As one example, the active site of class II fructose-1,6-bi-
sphosphate (FBP) aldolase, which is a zinc-dependent catalyst
of reversible direct-type asymmetric aldol reaction in organ-
isms, can construct a highly systematized reaction environment
that produces prodigious catalysis.[22] Inspired by their three-di-
mensional elaboration, we focused on the direct-type asym-
metric aldol reaction as a model reaction.
The proposed catalytic mechanism of class II FBP aldolase
bears eloquent testimony to the fact that the aldol reaction is
a formidable and attractive subject because of its many re-
quirements.[23] Shibasaki et al. conducted pioneering work in
this field in organic solvents.[9] While some organocatalysts as
class I aldolase mimics work in water, such reactions have gen-
erally resulted in poor reactivity and selectivity in the absence
of any organic solvent, even in the case of polymer-type orga-
nocatalysts.[20,24]
We envisaged that compaction of the independent factors
crucial for catalysis into one structure could be an attractive
highway to the realization of the simplest enzyme-like cataly-
sis.[25] We focused on a catalyst system consisting of a Sc(OTf)3-
chiral N-oxide ligand complex, an excess amount of surfactant,
and a catalytic amount of pyridine for a direct-type enantiose-
lective aldol reaction in water.[13] Based on this speculation,
concreteness was added to the conceptual structure
(Scheme 1a). To address the solubility issue, an amphiphilic
structure was designed, which was modeled after the architec-
ture of a membrane protein. At the active site, a chiral N-oxide
ligand was expected to form a chelate complex with a scandi-
Despite such a deep yearning for “artificial enzymes,”[7] there
have been few successes without the use of excess amounts
[a] T. Kitanosono, Prof. Dr. S. Kobayashi
Department of Chemistry, School of Science
The University of Tokyo
Hongo, Bunkyo-ku, Tokyo 113-0033 (Japan)
Fax: (+81)3-5684-0634
E-mail: shu_kobayashi@chem.s.u-tokyo.ac.jp
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/asia.201403004.
Chem. Asian J. 2014, 9, 1 – 7
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
ing effect on the amide NÀH bonds, which arise from intermo-
1
lecular hydrogen bonds with N-oxide moieties, in the H NMR
spectra.[27] The fixation of the amide plane inferred from these
intermolecular hydrogen bonds suggests intermolecular and
intramolecular aggregation of “bipod” moieties. Self-assembly
in water is especially desired to enable the functional opera-
tion by the addition of an external metal ion as enzymes do.
Based on our initial design, the solubility of L1 in water might
be dramatically improved by chelation with Sc3+ ion and the
formation of ion pairs at the pyridinyl moieties. As expected,
the catalyst solution prepared from just mixing Sc(OTf)3 and L1
seemed to be a colloidal dispersion. No sooner was an excess
amount of Sc(OTf)3 added to the aqueous solution of L1 than
the resultant solution turned transparent; the complex solution
turned out to be acidic[28] (Figure 1a).
Scheme 1. (a) Design of the highly systematized catalytic system; (b) synthe-
sis of L1. Reactions and conditions: [a] 1) nBuLi, THF, À788C, 3 h, 2) Br-
(CH2)10Br (1.5 equiv), THF, À258C,10 h, 61%; [b] KCN (2.0 equiv), DMSO, 808C,
3 h, 87%; [c] LiAlH4 (0.75 equiv), THF, rt, 3 h, 90%; [d] K2CO3 (2.1 equiv), Br-
(CH2)3Br (0.45 equiv), MeCN, reflux, 8 h, 88%; [e] Pd/C, H2, EtOH, 3 h, Quant.;
[f] 1) ClCO2iBu (2.2 equiv), CH2Cl2, 1 h 2) 4 (2.0 equiv), CH2Cl2, 2 h, 72%;
[g] mCPBA (2.0 equiv), CH2Cl2, 2 h, 99%.
um ion to promote intramolecular electron transfer, which
would result in a drastic change in the free energy of electron
transfer. The terminal pyridine moieties were needed to form
pyridinium salts to regulate the acidity in the whole system
and to enhance proton transfer across water in conjunction
with amide protons as Brønsted acids.[26] The two polarized
bonds and the two ion pairs in the structure were expected to
improve the solubility of the catalyst in water as an amphi-
pathic molecule. We also assumed that noncovalent interac-
tions between the metal center and substrates, which are es-
sential for the reaction, would be more stable within the sur-
rounding shield of hydrophobic “bipods.”
Figure 1. (a) Appearance of the reaction solution before (left) and after
(right) addition of substrates; (b) ESI-HRMS spectrum of the Sc(OTf)3–L1
complex.
Spectroscopic investigation of the catalyst was then per-
formed to gain insight into the catalyst structure and the struc-
1
ture–function relationship. H NMR spectroscopic analysis sup-
The designed ligand L1 was synthesized according to
Scheme 1b. The synthesis commenced with the construction
of two segments, 4 and 7. Lithiation of a-picoline 1 and cou-
pling with 1,10-dibromodecane furnished compound 2, which
underwent cyanidation to elongate the carbon chain. Success-
ful reduction of 3 afforded 4 in 48% overall yield. The synthe-
sis of fragment 7 involved appropriate coupling of l-proline
benzyl ester 5 with 1,3-dibromopropane and subsequent de-
benzylation. Finally, condensation between fragments 7 and 4
followed by selective N-oxidation produced L1 in an enantio-
pure form (Table S1 and Scheme S1 in the Supporting Informa-
tion). The N-oxide moieties were characterized by the deshield-
ports the formation of an Sc(OTf)3–L1 complex (Figure S2, Sup-
porting Information). When Sc(OTf)3 and L1 were combined, all
of the signals corresponding to the pyridinyl moieties were
shifted downfield. Similar peak shifts were observed when 2-
ethylpyridine and Sc(OTf)3 were mixed in a ratio of 3:1, which
indicated the formation of ion pairs rather than Sc–pyridine
complexes. It is assumed that the formation of ion pairs can
be ascribed to the ionic character of Sc-OTf bonds in Sc(OTf)3
(Figure S3, Supporting Information). The signal observed in the
19F NMR spectrum indicated the formation of triflic anion in
the reaction system, which might contribute to the formation
of a pyridinium cation in the catalyst structure. This pyridinium
&
&
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
cation seems to play a vital role in the solubility or construc-
tion of an efficient reaction environment, resulting in good
yield and enantioselectivity. For metal salts with a small hydrol-
ysis constant, the generation of Brønsted acids through cation
hydrolysis in an aqueous environment is well documented.[29]
The positive-mode ESI-HRMS spectrum showed a major
signal at m/z=492.27 corresponding to [L1+Sc(OTf)]2+, which
suggested the formation of a 1:1 complex of chiral N-oxide L1
and Sc(OTf)3 (Figure 1b). Because of its highly oxophilic charac-
ter, the scandium ion might be coordinated by the four
oxygen atoms of the carbonyl and N-oxide groups.[30] The
equal intervals of signals from m/z=983.54 to 1433.42 corre-
spond to TfOH, arising from the formation of trifluoromethane-
sulfonic acid salts at the pyridinyl moieties. The observation of
H–D exchange between the Sc–L1 complex and CD3OD corro-
borated the high acidity of the amide protons as Brønsted
acids and the assumed coordination environment.
As a model reaction to evaluate the catalytic activity of the
Sc–L1 complex, we conducted the enantioselective direct-type
aldol reaction of 2-methyl-1-in-
ment was maintained constantly in water by the catalyst
system, which is chemically much simpler than natural en-
zymes. Intriguingly, the functional integration also produced
significant rate acceleration as well as enhanced the enantiose-
lectivity compared with an independently functioning system
composed of 10 mol% Sc(OTf)3, 12 mol% L0, 20 mol% pyri-
dine, and 20 mol% SDS; the reaction rate was increased by
about 16 times.[35] To identify suitable structures, several modi-
fied chiral ligands (L2–L11) complexed with Sc(OTf)3 were in-
vestigated, as shown in Scheme 2. The use of L2, which is
linked at the m-position of pyridine, instead of L1 gave almost
the same result as L1. Ligand L3 is linked at the p-position of
pyridine; both reactivity and enantioselectivity were decreased
significantly. Changing the chain length of the surfactant
moiety did neither improve the reactivity nor selectivity. A pyri-
dine substituent on the amide moiety (L7) exhibited no cata-
lytic activity. Without the pyridine moiety, the chiral induction
was not retarded irrespective of lower reactivity. On the other
hand, when half ligand L9 or a precursor of N-oxide ligand L1’
danone using aqueous formal-
dehyde in water.[31] To our de-
light, the reaction proceeded
smoothly to afford the desired
product with a chiral quaternary
carbon in 75% yield with 60%
enantiomeric excess (ee). While
direct-type aldol reactions of
monodentate ketones that have
no activated functional groups
generally have very low catalyt-
ic turnover,[32,33] the catalytic ef-
ficiency of this catalyst system is
anomalous. Furthermore, the
yield improved to 91% (catalyt-
ic turnover number, CTN=18.2),
amazingly without damaging
the enantioselectivity at 408C.[34]
The enantiopurity of the prod-
uct was maintained even at
1008C (Figure S6, Supporting In-
formation). This property is in
complete contrast to that of an
enzyme, as an enzyme generally
has an optimum temperature at
which it works best. Although
thermal
rearrangement
of
amine oxide is generally easy,
the Sc(OTf)3–L1 catalyst has
high stability and identical cata-
lytic activity even at high tem-
peratures. It is assumed that the
strong Lewis acidity of the scan-
dium ion reinforces and stabiliz-
es complexation with L1. In par-
ticular, we could prove that an
efficient asymmetric environ-
Scheme 2. Effect of ligand structures.
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
was employed, no chiral induc-
tion was observed. Proline
amide-derived L10, which is the
basic backbone of L1, afforded
the product with lower selectivi-
ty. Attachment of a surfactant
moiety to the upper fragment
had negative effects on both re-
activity and selectivity. It is
noted that both surfactant and
pyridine moieties at precise po-
Scheme 3. Unoptimized examples of the Sc–L1 complex-catalyzed reaction.
sitions perform cooperative
roles in defining the reactivity
and enantioselectivity of this reaction. The reaction could be
extended to several other ketone and thioester derivatives
(Scheme 3).[36]
signed catalyst system possesses no cavity to completely cap-
ture a hydrophobic substrate, which makes microscopic heter-
ogeneity inevitable, especially in the initial stage before the
dispersion of an added substrate comes to equilibrium. To reit-
erate, the microscopically heterogeneous system in which hy-
drophobic and hydrophilic substrates were reacted with each
other in water thwarts direct spectrometric measurements.
Therefore, simple quenching at regular intervals should be im-
plemented without relying on general enzymological and
chemical methodologies in kinetic investigations. The analytical
method for obtaining reliable kinetic profiles using high-per-
formance liquid chromatography (HPLC) measurements was
established through a continuous process of trial and error.
The plotted progress curve consists of a brief initial transient
In contrast to organocatalytic aldol reactions, metal-depen-
dent direct-type aldol reactions are assumed to proceed via
the formation of enolate intermediates. When the scandium
catalyst was released into deuterium oxide in coexistence with
a ketone, the formation of a reactive enolate was confirmed by
the observation of an a-deuterated ketone.[37] The assumed
mechanism based on this information was thus attributed to
the existence of a stable substrate-binding intermediate (Fig-
ure 2a). To deepen the insight into the reaction mechanism of
this metalloenzyme-like reaction in water, we then attempted
to characterize it through kinetic analysis. The elaborately de-
Figure 2. (a) Schematic representation of Sc–L1-catalyzed direct-type aldol reaction; (b) plots of yield versus time (top) and rate versus concentration
(bottom) for the kinetic experiments; (c) Hanes–Woolf plot; (d) asymmetric hydroxymethylation of silicon enolate in aqueous media.
&
&
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
and an ensuing period during which the reaction is proceeding
at the maximum velocity. Eventually, the reaction approached
the final equilibrium asymptotically. Because the reaction is
viewed as being in a kind of quasi-equilibrium that follows
a quasi-steady-state approximation during the second phase,
the slope in this phase corresponds to the initial reaction rate
(Figure S10, Supporting Information). The reaction follows first-
order kinetics in the catalyst, whereas saturation kinetics was
recorded on the concentration of 2-methyl-1-indanone (Fig-
ure 2b). The saturated correlation observed in this reaction
system indicates the existence of a substrate-binding step in
the catalytic cycle, implying a rigid interaction between the
catalyst and the substrate. Displaying these experimental data
on a Hanes–Woolf plot[38] demonstrated a first-order depend-
ence (Figure 2c). Because of the sophisticated kinetic profile of
the reaction, other kinetic models[39] using different mathemati-
cal treatments were applied. We used a logistic approach
based on the probability of the reaction that enabled us to ap-
proximate the reaction profile as a linear expression against
time (Figure S11, Supporting Information). This analysis indi-
cates that the heterogeneity in the initial stage of the reaction
did not affect the reaction rate. Thus, it has been shown that
asymmetric hydroxymethylation catalyzed by the Sc–L1 com-
plex follows Michaelis–Menten kinetics, as enzymatic reactions
do. In addition, the observation of burst kinetics in the initial
period, which did not converge to zero, could be interpreted
as resulting from a steady-state rate-determining step follow-
ing product release. The kinetic parameters obtained were
computed using the least-squares method: maximum velocity
Experimental Section
Typical experimental procedure for hydroxymethylation
A mixture of Sc(OTf)3 (15.7 mg, 0.032 mmol) and chiral N,N’-dioxide
L1 (15.8 mg, 0.02 mmol) in water (1.2 mL) was stirred at room tem-
perature for 1 h. A ketone (55 mL, 0.4 mmol) was added to the re-
sulting 16.7 mm catalyst solution and an aqueous solution of form-
aldehyde (150 mL, 35% w/w, 2.0 mmol). After stirring for 24 h at
408C, the reaction mixture was quenched with sat. aqueous
NaHCO3 and brine. The aqueous layer was extracted thrice with di-
chloromethane, and the combined organic layers were washed
with brine and dried over anhydrous Na2SO4. After removal of the
solvents under reduced pressure, the residue was purified by prep-
arative TLC (n-hexane/EtOAc=3:1) to give the corresponding hy-
droxymethylated adduct 2a (64.0 mg, 91% yield) as a white
powder.
Acknowledgements
This work was partially supported by a Grant-in-Aid for Science
Research from the Japan Society for the Promotion of Science
(JSPS), Global COE Program, The University of Tokyo, MEXT,
Japan, and the Japan Science Technology Agency (JST). T.K.
thanks the JSPS for a Research Fellowship for Young Scientists.
The authors also thank Dr. Katsuaki Kawasumi for his contribu-
tion at an initial stage of this work.
Keywords: aldol reaction · asymmetric catalysis · enzyme
models · metalloenzymes · supramolecular chemistry
V
max =k2[catalyst]0 =0.0160 mmolhÀ1, k2 =2.18 hÀ1
Toward an in-depth understanding of the mode of action in
.
[1] Enzyme Catalysis in Organic Synthesis (Eds.: K. Drauz, H. Waldmann),
Wiley-VCH, Weinheim, 2002.
[2] Enzymes in Industry 3 (Ed.: W. Aehle), Wiley-VCH, Weinheim, 2007.
[3] A. M. Klibanov, Science 1983, 219, 722–727.
[4] a) M. T. Reetz, Angew. Chem. Int. Ed. 2001, 40, 284–310; Angew. Chem.
2001, 113, 292–320; b) G. Winter, A. R. Fersht, A. J. Wilkinson, M. Zoller,
M. Smith, Nature 1982, 299, 756–758.
[5] a) A. Tramontano, K. D. Janda, R. A. Lerner, Science 1986, 234, 1566–
1569; b) S. J. Pollack, J. W. Jacobs, P. G. Schultz, Science 1986, 234,
1570–1573.
[6] M. D. Toscano, K. J. Woycechowsky, D. Hilvert, Angew. Chem. Int. Ed.
2007, 46, 4468–4470; Angew. Chem. 2007, 119, 4552–4554.
[7] The term “artificial enzyme” first appeared in: R. Breslow, L. E. Overman,
J. Am. Chem. Soc. 1970, 92, 1075–1077.
[8] Racemic reaction, see: T. Darbre, C. Dubs, E. Rusanov, H. Stoeckli-Evans,
Eur. J. Inorg. Chem. 2002, 3284–3291.
[9] a) A. Matsuzawa, C. R. Opie, N. Kumagai, M. Shibasaki, Chem. Eur. J.
2014, 20, 68–71; b) N. Yoshikawa, Y. M. A. Yamada, J. Das, H. Sasai, M.
Shibasaki, J. Am. Chem. Soc. 1999, 121, 4168–4178; c) Y. M. A. Yamada,
N. Yoshikawa, H. Sasai, M. Shibasaki, Angew. Chem. Int. Ed. Engl. 1997,
36, 1871–1873; Angew. Chem. 1997, 109, 1942–1944.
[10] a) B. M. Trost, H. Ito, E. R. Silcoff, J. Am. Chem. Soc. 2001, 123, 3367–
3368; b) B. M. Trost, E. R. Silcoff, H. Ito, Org. Lett. 2001, 3, 2497–2500;
c) B. M. Trost, H. Ito, J. Am. Chem. Soc. 2000, 122, 12003–12004.
[11] a) S. Itoh, T. Tokunaga, M. Kurihara, S. Aoki, Tetrahedron: Asymmetry
2013, 24, 1583–1590; b) S. Itoh, T. Tokunaga, S. Sonoike, M. Kitamura, A.
Yamano, S. Aoki, Chem. Asian J. 2013, 8, 2125–2135; c) S. Itoh, M. Kita-
mura, Y. Yamada, S. Aoki, Chem. Eur. J. 2009, 15, 10570–10584.
[12] J. Kofoed, T. Darbre, J.-L. Reymond, Org. Biomol. Chem. 2006, 4, 3268–
3281.
aldol reactions catalyzed by a chiral Lewis acid complex, the ki-
netic behavior of this reaction was compared with that of
a chiral scandium-catalyzed aldol reaction using a silicon eno-
late (Figure 2d). The formation of the same product over time
exhibited a linear correlation in the concentration of both cata-
lyst and substrate, indicating that the overall rate is equal to
k[silicon enolate][catalyst].[40] The rate constant k was calculat-
ed to be 0.0192 hÀ1 mmolÀ2 using the least-squares method.[41]
We therefore assumed that the reaction of a silicon enolate
proceeded through a Lewis acid activation involving a weak in-
teraction. Consequently, active species in the reaction of the
silicon enolate are based on only a weak interaction, such as
coordinating or ionic bonds and generated through equilibri-
um with the starting substrate.
In summary, we have designed and synthesized small mole-
cules toward the simplest metalloenzyme-like catalyst system,
which catalyzed enantioselective direct-type aldol reactions
using aqueous formaldehyde (formalin) in water. This catalyst
system constructed an efficient reaction environment by coop-
erative interaction between several functional moieties as well
as well-regulated basicity. It is noted that this catalyst has
a much smaller molecular weight than enzymes, including al-
dolases. The simple design, including the precise replacement
of minimal functions, is also remarkable. We believe that this
report provides guidelines for the artificial design of enzyme-
like catalysts in water.
[13] S. Kobayashi, M. Kokubo, K. Kawasumi, T. Nagano, Chem. Asian J. 2010,
5, 490–492.
[14] S. Kobayashi, Pure Appl. Chem. 2007, 79, 235–245.
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
[15] J. Collot, J. Gradinaru, N. Humbert, M. Skander, A. Zocchi, T. R. Ward, J.
Am. Chem. Soc. 2003, 125, 9030–9031.
[30] X-ray crystallography of a chiral Sc complex with N-oxide backbone de-
rived from l-proline has been reported, see: Y. Liu, D. Shang, X. Zhou,
X. Liu, X. Feng, Chem. Eur. J. 2009, 15, 2055–2058.
[16] For instance, see: a) W. B. Motherwell, M. J. Bingham, Y. Six, Tetrahedron
2001, 57, 4663–4686; b) P. Tastan, E. U. Akkaya, J. Mol. Catal. A 2000,
157, 261–263; c) R. Breslow, N. Nesnas, Tetrahedron Lett. 1999, 40,
3335–3338; d) R. Breslow, S. D. Dong, Chem. Rev. 1998, 98, 1997–2012.
[17] S. Sasaki, K. Koga, Chem. Pharm. Bull. 1989, 37, 912–919.
[18] Y. K. Agrawal, H. Bhatt, Bioinorg. Chem. Appl. 2004, 2, 237–274.
[19] Examples of enzyme mimics based on polymers, see: J. Matsui, I. A.
Nicholls, I. Karube, K. Mosbach, J. Org. Chem. 1996, 61, 5414–5417.
[20] J. Bjerre, C. Rousseau, L. Marinescu, M. Bols, Appl. Microbiol. Biotechnol.
2008, 81, 1–11.
[31] Less acidic ketones show a reluctance to generate active enolates
under aqueous and catalytic conditions. Several trials of organocatalysts
in pure water have been reported, see: a) J. Paradowska, M. Stodulski, J.
Mlynarski, Angew. Chem. Int. Ed. 2009, 48, 4288–4297; Angew. Chem.
2009, 121, 4352–4362; b) N. Mase, Y. Nakai, N. Ohara, J. Am. Chem. Soc.
2006, 128, 734–735; c) Y. Hayashi, T. Sumiya, J. Takahashi, Angew. Chem.
Int. Ed. 2006, 45, 958–961; Angew. Chem. 2006, 118, 972–975; d) H.
Torii, M. Nakadai, K. Ishihara, Angew. Chem. Int. Ed. 2004, 43, 1983–
1986; Angew. Chem. 2004, 116, 2017–2020.
[32] For reported catalytic turnover number (CTN=yield/equivalent of cata-
lyst) values for direct-type aldol reactions without any directing group,
see: 4.8 (chiral organocatalyst in water/DMF=8/92, ref. 24a), 9.9 (chiral
organocatalytic resin, ref. 24d), 11.1 (chiral metal catalyst with 16 mol%
of water, ref. 9b and 9c), 16.8 (chiral dimetallic catalyst under anhy-
drous conditions, ref. 10b and 10c), 19.2 (chiral metal catalyst with sol-
vent amount of a substrate in NMP-MeOH cosolvent, ref. 11a–c).
[33] B. List, R. A. Lerner, C. F. Barbas, III, J. Am. Chem. Soc. 2000, 122, 2395–
2396.
[21] O. Khersonsky, D. Rothlisberger, O. Dym, S. Albeck, C. J. Jackson, D.
Baker, D. S. Tawfik, J. Mol. Biol. 2010, 396, 1025–1042.
[22] a) P. Clapꢁs, W.-D. Fessner, G. A. Sprenger, A. K. Samland, Curr. Opin.
Chem. Biol. 2010, 14, 154–167; b) H. J. M. Gijsen, L. Qiao, W. Fitz, C.-H.
Wong, Chem. Rev. 1996, 96, 443–473; c) D. Voet, J. G. Voet, Biochemistry,
Wiley-VCH, Weinheim, 1999.
[23] a) J. Schmidt, C. Ehasz, M. Epperson, K. Klas, J. Wyatt, M. Hennig, M. For-
coni, Org. Biomol. Chem. 2013, 11, 8419–8425; b) B. M. Trost, C. S. Brin-
dle, Chem. Soc. Rev. 2010, 39, 1600–1632.
[34] The CTN value reached more than 50 even at 258C in the presence of
1 mol% of catalyst, although this is an unoptimized result. For compari-
son with reported values, see ref. 32.
[35] Sc-L1 complex: 16% yield in 2 h; independent system: 12% yield in
24 h.
[36] See the Supporting Information for further details.
[37] T. Imamoto, Lanthanides in Organic Synthesis, Academic Press, London,
1994.
[24] a) A. H. Henseler, C. Ayats, M. A. Pericꢂs, Adv. Synth. Catal. 2014, 356,
1795–1802; b) N. Mase, C. F. Barbas, III, Org. Biomol. Chem. 2010, 8,
4043–4050; c) S. Bertelsen, K. A. Jørgensen, Chem. Soc. Rev. 2009, 38,
2178–2189; d) D. Font, S. Sayalero, A. Bastero, C. Jimeno, M. A. Pericꢂs,
Org. Lett. 2008, 10, 337–340; e) S. S. Chimni, D. Mahajan, Tetrahedron:
Asymmetry 2006, 17, 2108–2119; f) J.-L. Reymond, Y. Chen, J. Org.
Chem. 1995, 60, 6970–6979.
[25] Class II aldolases have molecular weights of approximately 70000; see:
W. J. Rutter, T. Rajkumar, E. Penhoet, M. Kochman, Ann. N. Y. Acad. Sci.
1968, 151, 102–117.
[26] T. Kitanosono, M. Sakai, M. Ueno, S. Kobayashi, Org. Biomol. Chem.
2012, 10, 7134–7147.
[27] I. A. O’Neil, N. D. Miller, J. V. Barkley, C. M. R. Low, S. B. Kalindjian, Synlett
1995, 617–618.
[28] The pH value is 3.55 (average of 10 measurements).
[29] Representative examples, see: a) S. Kobayashi, T. Ogino, H. Shimizu, S.
Ishikawa, T. Hamada, K. Manabe, Org. Lett. 2005, 7, 4729–4731; b) S. Ko-
bayashi, S. Nagayama, T. Busujima, J. Am. Chem. Soc. 1998, 120, 8287–
8288.
[38] Among the several linear plots of the Michaelis–Menten equation, this
method is preferable on statistical grounds because of smaller and
more consistent errors across the plot.
˘
˘
[39] M. V. Putz, A. M. Lacrama, V. Ostafe, J. Optoelectron. Adv. Mater. 2007, 9,
2910–2916.
[40] a) T. Kitanosono, S. Kobayashi, Chem. Rec. 2014, 14, 130–143; b) C. Mu-
kherjee, T. Kitanosono, S. Kobayashi, Chem. Asian J. 2011, 6, 2308–2311.
[41] The calculated standard deviation is 0.00180.
Received: August 27, 2014
Published online on && &&, 0000
&
&
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
COMMUNICATION
Water Music: We have designed the
simplest metalloenzyme-like catalyst
system model that can function in
water, without any aid from an enzy-
matic structure itself. By compaction
and precise orchestration of essential
functions, this artificial system efficiently
catalyzed enantioselective direct-type
aldol reactions using aqueous formalde-
hyde.
&
Aldol Reaction
T. Kitanosono, S. Kobayashi*
&& – &&
Toward Chemistry-Based Design of
the Simplest Metalloenzyme-Like
Catalyst That Works Efficiently in
Water
Chem. Asian J. 2014, 9, 1 – 7
www.chemasianj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ