Synthesis of valuable chiral intermediates by isolated ketoreductases: Application in the synthesis of α-alkyl-β-hydroxy ketones and 1,3-diols

Article abstract of DOI:10.1002/adsc.200606185

Regio- and stereoselective reductions of α-substituted 1,3-diketones to the corresponding β-keto alcohols or 1,3-diols by using commercially available ketoreductases (KREDs) are described. A number of α-monoalkyl- or dialkyl-substituted symmetrical as well as non-symmetrical diketones were reduced in high optical purities and chemical yields, in one or two enzymatic reduction steps. In most cases, two or even three out of the four possible diastereomers of α-alkyl-β-keto alcohols were synthesized by using different enzymes, and in two examples both ketones were reduced to the 1,3-diol. By replacing the α-alkyl substituent with the OAc group, 1-keto-2,3-diols, as well as 1,2,3-triols were synthesized in high optical purities. These enzymatic reactions provide a simple, highly stereoselective and quantitative method for the synthesis of different diastereomers of valuable chiral synthons from non-chiral, easily accessible 1,3-diketones.

Full text of DOI:10.1002/adsc.200606185

FULL PAPERS
DOI: 10.1002/adsc.200606185
Synthesis of Valuable Chiral Intermediates by Isolated Ketoreduc-
tases: Application in the Synthesis of a-Alkyl-b-hydroxy Ketones
and 1,3-Diols
Dimitris Kalaitzakis,a J. David Rozzell,b Ioulia Smonou,a,*
and Spiros Kambourakisb,*
a
Department of Chemistry, University of Crete, Voutes-Iraklio 71003, Crete, Greece
Fax: (+30)-2810–545001; e-mail: smonou@chemistry.uoc.gr
BioCatalytics Inc, 129 North Hill Ave, Suite 103, Pasadena, CA 91106, USA
b
Fax: (+1)-626-356-3999; e-mail: skambourakis@biocatalytics.com
Received: April 18, 2006; Accepted: July 24, 2006
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Regio- and stereoselective reductions of a- ples both ketones were reduced to the 1,3-diol. By
substituted 1,3-diketones to the corresponding b- replacing the a-alkyl substituent with the OAc
keto alcohols or 1,3-diols by using commercially group, 1-keto-2,3-diols, as well as 1,2,3-triols were
available ketoreductases (KREDs) are described. A synthesized in high optical purities. These enzymatic
number of a-monoalkyl- or dialkyl-substituted sym- reactions provide a simple, highly stereoselective and
metrical as well as non-symmetrical diketones were quantitative method for the synthesis of different
reduced in high optical purities and chemical yields, diastereomers of valuable chiral synthons from non-
in one or two enzymatic reduction steps. In most chiral, easily accessible 1,3-diketones.
cases, two or even three out of the four possible dia-
stereomers of a-alkyl-b-keto alcohols were synthe- Keywords: 1,2-diols; 1,3-diols; enzymatic reduction;
sized by using different enzymes, and in two exam- ketoreductase; b-hydroxy ketones
Introduction
Ru(II)[3] or rhodium[4] catalysts or hydroboration
using chiral Co(II) catalysts.[4] Methods requiring
Asymmetric reduction of double bonds (C=X where chiral starting precursors such as amides[5] and ke-
X=C, O, N) represents a very useful and widely ap- tones,[6] or aldol condensations starting from achiral
plicable method for introducing chirality into a pro- ketones,[7] through boron enolates in the presence of
chiral compound in high yield (up to the theoretical tertiary amines,[8] or through zirconium enolates[9] and
100%). Asymmetric reduction of ketones is particu- bases like LDA, LTMP and LHMDS[10] have also
larly important, since chiral alcohols are present in a been used. Compounds with high optical purities have
wide spectrum of pharmaceutical and other high- been obtained; however, these methods are frequent-
value compounds, or can be used as chiral building ly not generally applicable for the synthesis of other
blocks in the synthesis of larger, more complicated structurally related compounds or different diastereo-
compounds. Optically active a-alkyl-b-hydroxy ke- mers of the same compounds.
tones and their analogues represent important classes
Biocatalytic asymmetric reductions using whole
of chiral synthons that have been used in the synthesis cells of baker’s yeast[11a–d] or isolated enzymes[11e–i]
of many natural products and pharmaceuticals[1] and have also been reported. For the whole cell reactions,
have been identified in a spectrum of biologically yields and/or enantiomeric purities of products are
active compounds[2] including polyketides. We now low, because of the large number of ketoreductases
report a two-step chemo-enzymatic route to a number present in this yeast. An isolated reductase from Lac-
of these compounds by starting from prochiral a- tobacillus brevis (LBADH) was found to be selective
alkyl-1,3-diketones (Figure 1).
in the reduction of C-5 of a 3,5-dioxocarboxylate,[11e,f]
Several methods for the stereoselective synthesis of while one ketoreductase from Saccharomyces cerivi-
b-hydroxy ketones have been developed. They in- sae was selective in the reduction of the C-3 ketone of
clude asymmetric reductions using chiral ruthenium the same compound.[11i] The latter enzyme was active
1958
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
FULL PAPERS
Figure 1. Enzymatic reduction of a-alkyl-1,3-diketones with NADPH-dependent ketoreductases. KRED=ketoreductase,
GDH=glucose dehydrogenase.
against a number of substrates including two more two, and potentially three chiral centers are created
1,3-diketones.[11i] Recently, whole cells of lyophilized from a non-chiral diketone substrate in high yields (>
Rhodococcus ruber were found to be active against a 90%) in one or two enzymatic steps (Figure 1). The
number of ketones including some 1,2- and 1,3-dike- chiral center of the a-carbon bearing the alkyl
tones.[11g] One broad substrate and solvent tolerant group(s) is created by the selective reduction of one
dehydrogenase was isolated from this organism.[11h] of the two enantiotopic ketones when R1 =R2
Even though a few 1,3-diketones have been reduced (Figure 1). When R1 and R2 are different and R3 =H
by the previous biocatalysts, none of them contained both chiral centers are set in one enzymatic reduction
substituents at the a-position as in the examples pre- step and in high yield (up to the theoretical 100%)
sented in this report. In every case only one active due to the rapid equilibrium of two enantiomeric di-
enzyme was found giving one (of the possible two) ketones.[12a,b,14,15] In this latter case the enzyme shows
enantiomers.
both regio- and stereoselectivity.
It has been clearly shown in a number of recent re-
Preliminary results of the enzymatic reductions of
ports[12–15] that commercially available ketoreductases some mono-a-alkyl-substituted diketones that are dis-
are chemo-, stereo-, and regioselective catalysts that cussed in this report have already been published.[14,15]
often reduce a wide range of ketones and represent In particular the reduction of 4-methyl-3,5-heptadione
an important addition that complements and expands (entry 8, Table 1) by ketoreductase EXP-A1B forms
the existing technology of ketone reductions. Cloned the natural pheromone (+)-sitophilure, so the synthe-
and overexpressed enzymes (as in the case of all ke- sis of this compound was reported separately.[15] In
toreductases described herein) are easily and cost-ef- the present work the substrate complexity of the dike-
fectively synthesized on a commercial scale by fer- tones was expanded to include dialkyl-substituted as
mentations. The required cofactor is no longer a sig- well as non-symmetrical 1,3-diketones (Table 1). The
nificant addition to the cost of the process since possibility of increasing even more the number and
NAD(P)H can be added in catalytic amounts and be the structural diversity of the compounds that can be
efficiently recycled by various enzymatic methods, synthesized from the same a-alkyl 1,3-diketone scaf-
such as glucose dehydrogenase (GDH),[13a,b] formate fold was investigated by introducing an O-acetyl
dehydrogenase (FDH),[13b] or phosphite dehydrogena- group as one of the a-substituents, which upon reduc-
se.[13c]
In this report we present a regio- and stereoselec- triols (V).
tion would form a,b-dihydroxy ketones (IV) or 1,2,3-
tive synthesis of different diastereomers of a large
number of a-alkyl-b-hydroxy ketones (II) in good to
excellent optical purities, from readily available non- Results and Discussion
chiral a-alkyl-1,3-diketones (I) by the use of thirty-
four commercially available ketoreductase enzymes. Asymmetric synthesis is moving into an area where
All prochiral a-alkyl-1,3-diketones (I) were chemical- not only high enantiomeric purities of products are
ly synthesized from cheap and readily available start- required, but also in the development of methods that
ing materials in one or two steps in high yields. We give easy access to both enantiomers (or a number of
also show that various diastereomers of the corre- diastereomers if possible) of a targeted compound.
sponding 1,3-diols (III, V) can be synthesized by step- This is particularly important in the synthesis of bio-
wise reduction by using different enzymes. As a result logically active compounds (such as pharmaceuticals
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1959

Dimitris Kalaitzakis et al.
FULL PAPERS
Table 1. Enzyme-catalyzed stereoselective reduction of diketones to keto alcohols.
Entry Substrate
KRED Diastereomeric ratio%
Yield[d] (Time)
Product
A
B
C
D
102 and >99
-
-
-
-
-
100% (24 h)
106[a]
A1B
3R,4S
74
1a
1b
2a
26
100% (10 min)
1
3R,4S
3S,4R
127
3
94 3S,4S
-
3
90% (24 h)
102 and >99
100% and 90%
(12 h)
100% (3 h)
-
-
-
-
-
106[a]
112[a]
3R,4S
96
4
3R,4S
95
3S,4R
90
A1B
107[a]
-
-
-
-
5
100% (1 h)
100% (6 h)
2
2c
2b
3a
10
3S,4R
118 and
119[a]
>98
3S,4S
-
-
-
100% (24 h)
102 and >99
106
112
100% and 93%
(12 h)
100% (6 h)
-
-
-
-
-
3R,4S
97
3
3R,4S
96
3S,4R
79
A1B
107
-
-
4
100% (1 h)
82% (24 h)
3c
3
8
10
3S,4R
118 and
119
5
95 3S,4S
-
-
100% (24 h)
100% (6 h)
3b
96
3R,4S
101[a]
-
-
-
-
4
-
102 and >99
94% and 76%
(24 h)
100% (6 h)
4a
106[a]
123
3R,4S
96
3R,4S
-
-
1
3
4
90
3S,4R
72
3S,4R
A1B
107[a]
-
10
28
100% (24 h)
40% (24 h)
4c
5a
102 and >99
-
-
-
-
100 (3 h and 6 h)
100% (1 h)
106[a]
112[a]
3R,4S
95
2
3
3R,4S
99
3S,4R
90
A1B
107[a]
-
-
1
6
100 (20 min)
100% (12 h)
5
5c
2
2
3S,4R
-
-
108[a]
1
99 3S,4S
-
100% (24 h)
118 and 1 and 99 and 97
0 and 100% (24 h and
12 h)
5b
119[a]
1
3S,4S
2
1960
www.asc.wiley-vch.de
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
Table 1. (Continued)
FULL PAPERS
Entry Substrate
KRED Diastereomeric ratio%
Yield[d] (Time)
Product
A
B
C
D
-
102 and >99
100% (12 h and
24 h)
-
-
-
6a
106
3R,4S
92
3S,4R
84
A1B
107
-
8
100% (2 h)
100% (6 h)
6
6c
26
3S,4R
118 and 7 and 93 and 90
0 and
3
-
-
-
100% (24 h)
100% (24 h)
119
124
7
4
3S,4S
96 3S,4S
6b
102 and >99
-
-
-
-
-
100% (12 h)
100% (24 h)
106[a]
131
3R,4S
93
7
7a
3R,4S
97
107[a]
A1B
-
-
-
-
3
-
100% (12 h)
100% (0.5 h)
3S,4R
>98
3S,4R
7
7c
118 and 10 and 90 and 92
-
-
-
-
93% (24 h)
100% (24 h)
119[a]
124
8
5
3S,4S
95 3S,4S
7b
97
4S,5R
A1B[a]
-
3
-
100% (40 min)
8c
8
9
96
and>99
4S,5S
118 and
119[a]
4
93% and 100%
(24 h and 12 h)
-
8b
&<1
102 and >99
100% and 85%
(24 h)
85% (24 h)
-
-
-
-
-
-
106[a]
103[a]
2S,3R
>99
2S,3R
9a
9b
127
3
96 2S,3S
-
1
92% (24 h)
91
2S,3R
95
102
106
8
3
1
2
-
-
75% (24 h)
46% (24 h)
10a
10b
2S,3R
94
2S,3R
1
126
4
-
-
100% (12 h)
10
107
127
96 2S,3S
94 2S,3S
3
-
-
-
75% (24 h)
100% (24 h)
6
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1961

Dimitris Kalaitzakis et al.
FULL PAPERS
Table 1. (Continued)
Entry Substrate
KRED Diastereomeric ratio%
Yield[d] (Time)
100% (1 h)
Product
A
B
C
D
-
91
3R,4S
112
-
9
11a
108
118 and
119
1
-
99 3S,4S
>99
3S,4S
-
-
-
-
70% (24 h)
100% (24 h)
11
11b
11d
89
3R,4R
51
A1B
107
-
-
-
-
11
49
100% (1.5 h)
34% (24 h)
101 and 98
-
-
2
6
-
-
100% (1 h)
100% (6 h)
112
3R,4S
12a
113 and 94
115
3R,4S
90
3R,4R
A1B
-
-
-
10
100% (24 h)
100% (1 h)
12d
12c
12b
13a
12
65
3S,4R
107
-
35
108 and
119
118
>99
3S,4S
>99
3S,4S
-
-
-
-
-
-
-
100% (24 h)
100% (6 h)
112[c]
113[c]
98
97
2
3
-
-
100% (0.2 h)
100% (0.2 h)
-
13
[a]
Ref.[14]
[b]
Ref.[15]
[c]
The absolute stereochemistry of this reaction product has not been determined, the assignment of the major diastereomer
as A was based on its retention time in chiral GC.
The yields were calculated from the integration on GC of crude reaction extracts. Scaling of a few reactions and isolation
of keto-alcohols gave very similar yields as those reported in this Table.
[d]
and pesticides), when libraries of compounds are syn- substituted, 1–8, three disubstituted, 11, 12 and 13,
thesized in order to be tested for biological activity, and two unsymmetrical diketones, 9 and 10, were se-
or when improvement of the properties of an already lected as substrates for stereoselective reduction
active compound is made by the synthesis of closely (Table 1). All these compounds were synthesized
related analogues. As a result, the development and from the corresponding 1,3-diketone after alkylation
discovery of catalysts that are chemo-, regio- and ste- with the appropriate alkyl halide. High yields of iso-
reoselective and can give access to a number of iso- lated diketone were routinely obtained in reaction
mers of closely related compounds are highly desira- processes that require cheap and readily available
ble.
The results presented in this report show that ketor- conditions (details in Experimental Section).
eductases fulfill all the previous requirements. In ad- With only a couple of exceptions, all enzymes gave
starting materials and simple and scalable reaction
dition, enzymatic reductions using the isolated en- the keto alcohols shown in Table 1 as the only prod-
zymes described herein have been scaled to laborato- uct, after all diketone had completely reacted. Addi-
ry, preparative, and industrial scales for the synthesis tion of excess enzyme and long incubation times re-
of chiral intermediates.[13a,b] In order to demonstrate sulted in the formation of the diol in some cases, but
the wide range of keto alcohols that can be accessed in the majority of cases enzymes were completely un-
by using different ketoreductases, eight mono-alkyl- reactive towards the keto alcohol they produced. In
1962
www.asc.wiley-vch.de
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
FULL PAPERS
all but one substrate (13, Table 1) presented in this of the keto alcohol containing the S-hydroxy, R-alkyl
report, more than one diastereomer was synthesized configuration in all monoalkyl-substituted diketones
in good to excellent optical purities with the action of (1–7, 9 and 10, Table 1), while neither enzyme reacted
different enzymes. For example, in five out of the with the dialkyl-substituted substrates 11–13. The
eight symmetrical monoalkyl-substituted diketones same trend in the formation of the same major diaste-
(entries 2, 3, 5, 6, 7) and in two out of the three di- reomer, though without always perfect selectivity, was
alkyl-substituted diketones (entries 11 and 12), three also observed in most active enzymes that are not
out of the four possible diastereomers were synthe- shown in Table 1. With the exception of the a-methyl
sized in good to excellent optical purities with the diketone 1, where enzymes KRED118 and KRED119
action of different enzymes. In the rest of the cases gave an almost equimolar mixture of (3S,4S) and
two out of the four diastereomers were produced (3R,4S) keto alcohols (data not shown), both enzymes
(substrates 1, 4, 8, 9, 10). The data reported in Table 1 gave good to excellent optical purities of keto alco-
represent the best reactions obtained after screening hols containing the S-configuration for both the alco-
thirty-one ketoreductases of the commercially avail- hol and alkyl groups with all the monosubstituted and
able KRED30,000 kit (Biocatalytics) and three of the the disubstituted diketones (Table 1). Both enzymes
thirty-three newly discovered enzymes (EXP33,000 showed almost no activity for substrates 4, 9 and 10.
kit, BioCatalytics Inc). The only exception is Note also that even in the case of substrate 1, both
KRED107 where all its reaction products are report- KRED118 and KRED119 gave the S-alcohol upon re-
ed (even though this enzyme did not give high selec- duction of the ketone; however, both were not selec-
tivity with all reported substrates) as a representative tive for the configuration of the a-methyl group. In
example and in order to illustrate general enzyme se- contrast, KRED102 and KRED106 formed keto alco-
lectivities that will be discussed later.
hols in high optical purities with the same stereopre-
As predicted in our previous report[14] and now con- ference for all the substrates they reduced, including
firmed experimentally, in every previously reported 1 (Table 1). In general, substrate 1 was reactive with
diketone more diastereomers were synthesized, or an almost all enzymes tested; however, only three en-
improvement of the optical purities was achieved zymes (Table 1) showed good to excellent diastereose-
after the screening of fourteen newly discovered en- lectivity of product formation.
zymes. In particular, the keto alcohol (3S,4R) that was
As a general rule, most enzymes showed good to
accessed in moderate optical purities by KRED107 in excellent stereoselectivity for the reduction of the
substrates 2, 5 and 7 and in low optical purities in 3,4 keto group. However, the selectivity for the a-alkyl
and 6, can now be synthesized very efficiently in im- stereocenter was not always high. As a result, most
proved optical purities by EXP-A1B (Table 1). Addi- enzymatic reductions would either give a single dia-
tional diastereomers can now be synthesized from stereomer or a mixture of products where the two
substrates 1, 4 and 9. The utility of this methodology major (or sometimes the only two) diastereomers
was expanded by the reduction of non-symmetrical would contain the same absolute configuration for the
a,a-dialkyl-substituted 11, 12 and one a-alkyl-a-O- alcohol and opposite configurations for the alkyl sub-
acetyl-substituted diketone 13 (Table 1). Reduction of stituted compound. This selectivity has already been
the disubstituted ketones resulted in the production discussed for KRED118 and 119 with substrate 1. An-
of an otherwise difficult to synthesize chiral tertiary other example includes KRED107, an enzyme that
carbon, while replacing the a-alkyl group with other would always give the R-configuration in the synthesis
functionalities increased the number of important in- of the alcohol (except substrate 10) but mixed S- and
termediates that can be accessed. When the two keto R- configurations for the a-alkyl group, the R-config-
groups were not chemically equivalent, as in the case uration product always being the major one (Table 1).
of the non-symmetrical substrates 9 and 10, all active Although this selectivity was true for most enzymes
enzymes were selective for the reduction of the less and for all the substrates that were active (symmetri-
substituted ketone.
cal and non-symmetrical mono-substituted com-
Besides being regioselective, many enzymes showed pounds 1–10), there were a couple of exceptions
good to excellent stereoselectivities of reduction re- when comparing the reductions of mono- versus dis-
gardless of the a-alkyl group. In a previous report the ubstituted substrates. This trend is illustrated by the
general R or S preference of reduction for the en- reductions of EXP-A1B which gave keto alcohol
zymes of KRED30,000 kit was identified by using two (3S,4R) as the major diastereomer in the monosubsti-
standard ketones.[13a] Most enzymes retained this ste- tuted symmetrical diketones 2–8, (3R,4S) with the
reoselectivity for the reduction of the diketone dis- methyl-containing diketone 1, and (3R,4R) in the dis-
cussed herein, and most gave the same absolute con- ubstituted (11, 12) analogues (Table 1). These results
figuration in the formation of both chiral carbons of clearly show that prediction of a ketoreductase selec-
every keto alcohol reduced. For example, both tivity cannot always by done with absolute certainty,
KRED102 and KRED106 gave high optical purities even when the results of similar substrates are known.
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1963

Dimitris Kalaitzakis et al.
FULL PAPERS
Replacing the a-alkyl group, or one of the alkyl esterification (Figure 2). This reaction progress was
1
groups in the dialkyl substrates with side chains con- monitored by H NMR at three different pH values
taining heteroatoms such as oxygen can give access to (6.5, 6.9 and 7.5). In every case, the b-O-acetylated al-
an even larger class of important intermediates. For cohol 13a was the only observed final product even
example, we synthesized substrate 13 where one alkyl though in early time points of the reaction, a mixture
group was replaced with an O-acetyl group. Substrate of the a- and the b-oxo-acetylated alcohols (13i and
13 represents a very interesting and potentially very 13a) was observed. Many enzymes gave product with
useful class of substrates since its reduction product, substrate 13. However, only two gave high stereose-
a,b-dihydroxy ketone, contains a secondary and a ter- lectivity of product formation (Table 1). The absolute
tiary 1,2-diol. Stereoselective synthesis of such com- configuration of the product has not yet been deter-
pounds is very difficult and is usually achieved by mined. The assignment of the major product as A in
asymmetric epoxidation of an alkene followed by hy- Table 1 was based on the relative time of elution of
drolysis. Here the stereochemistry of both hydroxy this product from the chiral GC column (first to come
groups is produced in one step starting from a non- out).
chiral, easy to synthesize starting material (13,
Table 1).
High regioselectivity and, most of the times,
medium to excellent stereoselectivity were obtained
Upon enzymatic reduction of diketone 13, the a- in all reactions. Most enzymes gave the keto alcohol
oxo-acetyl keto alcohol 13i was forming first. Under as the only product, and only a couple of them (A1B
the reaction conditions however, this intermediate re- for example) were able to reduce further the keto al-
arranged very fast to the final product b-oxo-acetyl cohol they produced, upon addition of excess enzyme
keto alcohol 13a, probably by an intramolecular trans- and after longer reaction times. Even in these cases,
the keto alcohol could be isolated as a pure product
in quantitative yields, if the reaction progress was
monitored. This selectivity was quite surprising con-
sidering the wide substrate range that most of these
enzymes reduce.[12–15] In order to determine whether
such selectivity was dependent on the keto alcohol
stereochemistry or the keto alcohol chemical compo-
sition, we investigated the reduction of various diaste-
reomers of the same keto alcohol (substrate 2). This
keto alcohol was selected because three out of the
four possible diastereomers can be synthesized in
high optical purities. As a result, the dependence of
reactivity on absolute stereochemistry could be more
thoroughly checked. Each keto alcohol 2a–c (Table 2)
was synthesized and was then tested for reduction by
using selected KREDs. Most KREDs tested were
either inactive or had lower activities for the reduc-
tion of keto-alcohols 2a–c than for the reduction of
diketone 2.
In Table 2 are shown some representative examples
of enzyme reactivity. In one example, KRED101
showed small stereoselectivity for the formation of
Figure 2. Product formation and rearrangement of substrate
13.
Table 2. Enzymatic reduction of diketones 2 and 13 and subsequent formation of their corresponding 1,3-diols.
Substrate
First Reduction: keto alcohol (% de)
Second Reduction: 1,3-diol (% de)
101
102
112
118
A1B
Product
101
102
112
118
A1B
Product
2
58%
99%
96%
3R,4S 2a
99%
NR
NR
NR
NR
NR
99%
99%
NR
NR
NR
NR
NR
NR
NR
2S
2R
2S
2R
2R
A
99%
27%
99%
93%
3S,4S 2b
95%
99%
99%
99%
95%
30%
3S,4R 2c
13a
95%
97%
99%
98%
13
95%
95%
98%
>99%
B
NR=no reaction was detected.
1964
www.asc.wiley-vch.de
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
FULL PAPERS
keto alcohol by giving a mixture of 2a and 2b. How- yields in a timely fashion (<24 h) can be achieved in
ever, it reduced all three diastereomers 2a–c to the most cases. The parameters that need to be optimized
corresponding 1,3-diols in high optical purities. EXP- in order to scale enzymatic reductions to preparative
A1B on the other hand, gave products in high optical (10–50 g) and pilot (24 kg) scale have been discussed
purities after the reduction of both diketone 2 and all previously.[13a,b] In all the reactions presented here, the
diastereomers 2a–c. KRED102 and KRED118 that NADPH cofactor was very efficiently recycled by
led to optically pure 2a and 2b respectively, were in- using glucose and glucose dehydrogenase (GDH), an
active with all keto alcohols 2a–c (Table 2). Just as enzyme that has been cloned and overexpressed and
the enzyme reactivity was not easily predictable, the can regenerate NADPH from NADP+ by the con-
stereoselectivity of ketone reduction also varied de- comitant oxidation of glucose to gluconolactone.
pending on substrate and enzyme. For example, Product isolation and purification in all these reac-
KRED112 and KRED101 catalyzed the S reduction tions was very simple and consisted of an aqueous or-
of the keto group of 2a and 2b, and R- for the keto al- ganic extraction, where all the salts, cofactors and glu-
cohol 2c. Both enzymes showed S enantiopreference cose products remain in the aqueous layer and all the
for the first reduction of diketone 2 (Table 2). In con- organic products (and starting materials if present)
trast, enzyme A1B showed R stereoselectivity in the are extracted in the organic layer.
reduction of both diketones and all keto alcohols
(Table 2).
The diastereomeric ratio of products presented in
Tables 1 and 2 was derived by chiral GC analysis. The
We extended our studies to the reduction of keto relative configuration of keto alcohols 1, 8, 9 and 10
1
alcohol 13a which, upon further reduction, would give was found by comparing their H NMR spectra with
chiral 1,2,3-triols (Figure 1). Again many enzymes re- literature data. The relative configuration of keto al-
acted with keto alcohol 13a but only some are shown cohols 2–7 was found by measuring the Jvic(H,H) after
in Table 2. Looking at this Table, a few interesting ob- decoupling experiments. The relative configuration of
servations can be made. First, KRED102, which was keto alcohols 11 and 12 was found by comparing the
inactive towards all three diastereomers of keto alco- 13C NMR spectrum of the diastereomers and by ap-
hol 2, now gave a product of high optical purity. Al- plying the Heathcock rule.[16] The absolute configura-
though only one diastereomer could be accessed in tion of keto alcohols 1–12 was found by transforming
high optical purities in the first reduction of diketone them to their corresponding MPA (R or S methoxy-
13, both possible enantiomers were formed in the phenylacetic acid) esters.[17]
second step for the reduction of the keto alcohol (A
and B in Table 2). The absolute configuration of these
products was not determined. However on the basis Conclusions
of the previous stereoselectivity in the second reduc-
tion step, enantiomer A should have the 2S alcohol The reduction of a large number of a-monoalkyl and
configuration and enantiomer B should have the 2R dialkyl symmetrical as well as non-symmetrical 1,3-di-
configuration.
ketones by thirty four NADPH-dependent commer-
All reactions shown in Table 1 were performed on cially available ketoreductase enzymes (KREDs) is
a small scale (1 mL, 25 mM substrate) by using 10– presented. All diketones described in this report were
25% (g/g) of enzyme relative to the diketone and synthesized in high yields from readily available start-
10% (mol/mol) of cofactor. Selected reactions (#1 ing materials. Chiral keto alcohols and diols, such as
with KRED-102 and #8 with KRED-118 and KRED- the ones described herein, represent very useful syn-
A1B) were performed in larger scale (100 mL 50 mM thons in organic synthesis. Such compounds have
substrate) by using the same relative concentration of been used as precursors in the synthesis (or are struc-
KRED to substrate (10–20% g/g) and 1% (mol/mol) tural elements) of various biologically active com-
of NADPH relative to substrate. Quantitative yields pounds and pharmaceuticals.
were obtained within 24 h of reaction time, and isolat-
At least two, and in many cases three out of the
ed yields of >85% were obtained (details in Experi- four possible diastereomers of all the a-alkyl-substi-
mental Section). In general, high yields and purities tuted b-keto alcohols (symmetrical, non-symmetrical
of products are usually obtained from the enzymatic mono- or dialkyl-substituted) were selectively pro-
reductions since byproducts resulting from harsh reac- duced by the use of different enzymes. With only
tion conditions (i.e., excessive heating, use of strong two exceptions, all enzymes gave the same absolute
acid/bases) do not form under the mild enzymatic re- configuration for the reduction of the first keto
action conditions (near neutral pH and 25–358C).
group regardless of the a-alkyl substitution. The ste-
It is important to note at this point that these reac- reoselectivity of the second chiral center (that of the
tions have not been optimized in terms of substrate, a-alkyl position), however, was not always high and
enzyme and cofactor. Improvement of the relative was dependent on the enzyme and the diketone sub-
amounts of KRED to substrate to give high reaction strate.
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1965

Dimitris Kalaitzakis et al.
FULL PAPERS
Most enzymes were either inactive or possessed have, making them an even more appealing biocata-
much lower activity for the reduction of the keto-al- lyst for large-scale synthesis. The use of isolated en-
cohol that they produced. As a result, the isolation of zymes as catalysts for organic reactions is becoming a
chemically pure keto alcohols in high reaction yields more standardized and practical tool in the hands of
and optical purities was possible in every successful synthetic chemists, mainly because of the increasing
enzymatic reduction. This selectivity is impressive if need for environmentally benign methods, the com-
one considers that most diketones (with the exception mercial availability and low price of purified enzymes.
of non-symmetrical 9 and 10) possess a plane of sym-
metry making the two keto groups enantiotopic. To
our knowledge, no chemical catalysts exist that can Experimental Section
distinguish between so closely related keto groups
and give only one of the four potential diastereomeric General Analytical Methods
keto alcohols and without further reduction to the
Flash chromatography was carried out with 60 Å silica gel
diol. In the case of the two non-symmetrical diketones
9 and 10, the less hindered ketone was reduced by all
active enzymes.
according to the procedure of Still.[19] The progress of the
enzymatic reactions and the selectivities were determined
by gas chromatography of crude EtOAc extracts (for the
small scale reactions) and of isolated products (in the larger
scale reactions) using HP5890II gas chromatograph equip-
ped with an FID detector; (column: 30 m × 0.25 mm ×
0.25 mm chiral capillary column, 20% permethylated b-cy-
clodextrin HP Part No 19091G-8233). Some keto alcohols
were acetylated prior to GC analysis. The exact methodolo-
gy used to separate every racemic keto alcohol is presented
in the supplementary material. In addition representative
chiral GC chromatograms for racemic and enzyme-produced
keto alcohols 1, 2, 4, 5, 7, 9 are presented in the supplemen-
tary material of ref.[14]
Formation of the 1,3-diols was achieved in a second
enzymatic step. Almost every enzyme tested showed
lower activity in the reduction of the keto alcohol
compared to the corresponding diketone. In all cases
investigated, high stereoselectivity in the 1,3-diol for-
mation was achieved even by enzymes that were not
very stereoselective in the first reduction step of the
same diketone (Table 2).
By changing the functionality at the a-alkyl posi-
tion other useful classes of compounds can be synthe-
sized. For example, we replaced one alkyl group with
an acetyl group (compound 13), and upon enzymatic
reduction we obtained either the 1-keto-2,3-diol or
the 1,2,3-triols (IV, V, Figure 1). One can speculate
that by replacing of the acetyl with a free or protected
amine, various different 1,2-amino alcohols can be
synthesized by using the same methodology. In addi-
tion, other functionalities can be present in the a-
alkyl substituent as in substrates 5–7 and 12. Clearly,
a large number of useful and reactive chiral synthons
can be synthesized by modifications of the 1,3-dike-
tone scaffold. Experiments are underway to further
extend the applicability of this method.
The results that are presented in this report do not
only show an easy, high yielding and stereoselective
synthesis of various important classes of compounds
(Figure 1), but also represent an excellent example of
the usefulness of enzymes as catalysts in chiral organ-
ic synthesis. The general applicability of enzymatic re-
ductions in the synthesis of a wide range of chiral al-
cohols as well as their scalability has recently been
shown.[12–14] An additional advantage of using isolated
enzymes for screening that is not usually applicable to
chemical or whole cell microbial catalysts, is their
rapid identification of activity towards any substrate.
After an active enzyme has been identified, scaling
the process to laboratory-scale quantities is usually
fast and straightforward.[13a,b] Further improvement of
the enzymatic activity, stability or selectivity is also
possible using modern techniques of enzyme evolu-
1H NMR and 13C NMR spectra were recorded on a 300 or
500 MHz Bruker spectrometer in CDCl3 solutions, by using
Me4Si as internal standard. Chemical shifts are reported in
ppm downfield from Me4Si. Yields refer to isolated and
1
spectroscopically pure materials. All H NMR data and ab-
solute stereochemistry assignments data are included in the
Supporting Information.
Synthesis of 1,3-Diketones
a-Alkyl-1,3-diketones 1–7 and 11, 12 were prepared from
commercially available 2,4-pentanedione (acetylacetone) by
alkylation with the proper alkyl halide. a-Alkyl 1,3-diketone
10 was prepared from commercially available 1-phenyl-1,3-
butadione by alkylation with methyl iodide. Diketones 8
and 9 were prepared by oxidation of their corresponding
racemic keto alcohols. Diketone 13 was prepared from com-
mercially available 3-chloro-2,4-pentanedione by treatment
with sodium acetate following by alkylation with methyl
iodide.
All diketones presented in this report are literature com-
pounds.The 1H NMR spectra of the synthesized diketones
1–13 were compared to those reported in the literature.
Synthesis of a-Alkyl 1,3-Diketones, 1–7 and 10
Under a nitrogen atmosphere acetyl acetone (50 mmol) or
1-phenyl-1,3-butadione (50 mmol) were dissolved in anhy-
drous acetone (40 mL) and pre-dried potassium carbonate
(46.5 mmol) was added. After stirring the solution at room
temperature for 5 min, the proper alkyl iodide [1, 2, 10] or
alkyl bromide (61.5 mmol) [3, 4, 5, 6, 7] was added drop
wise over a period of 2 min and the reaction mixture was re-
tion. This is a unique advantage that only enzymes fluxed overnight. Complete reaction was observed by thin
1966
www.asc.wiley-vch.de
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
FULL PAPERS
layer chromatography analysis of reaction aliquots. After
condensation of the reaction mixture, diethyl ether (50 mL)
was added, the mixture was filtered and the organic layer
was evaporated to dryness. Pure products were obtained
using silica gel chromatography (hexane/EtOAc, v/v, 8/1),
70% to 90% isolated yield. Exact reaction yields and prod-
uct characterization for each diketone are presented in the
supplementary material.
preparation of diketones 10–12 , by alkylation of the pre-
pared 1-acetyl-2-oxopropyl acetate with methyl iodide in
90% isolated yield. Combined isolated yield for both steps:
54%. Without any further purification diketone 13 was sub-
jected to enzymatic reduction.
Synthesis of Racemic b-Keto Alcohols as Standards
for Chiral GC Analysis
Racemic keto alcohols 1–7 and 11–13 were prepared as
standards by reduction with sodium borohydride. The corre-
sponding keto alcohols 8 and 9 were prepared from 3-penta-
none by aldol condensation with the corresponding alde-
hyde. Keto alcohol 10 was prepared from 1-phenyl-1-propa-
none by aldol condensation with acetaldehyde.
Synthesis of Diketones, 11 and 12
Under a nitrogen atmosphere 3-ethyl-2,4-pentanedione 2
(39 mmol, 5 g) or 3-allyl-2,4-pentanedione
5 (39 mmol,
5.54 g) were dissolved in anhydrous acetone (40 mL) fol-
lowed by the addition of pre-dried potassium carbonate
(40 mmol, 5.5 g). After stirring the solution at room temper-
ature for 5 min, methyl iodide (45 mmol, 2.72 mL) was
added dropwise in each reaction over a period of 2 min and
the reaction mixture was refluxed overnight, where com-
plete reaction was observed by thin layer chromatography.
After condensation of the reaction mixture, diethyl ether
(50 mL) was added, the mixture was filtered and the organic
layer was evaporated to dryness. Pure diketone (4.5 g) were
obtained using silica gel chromatography (hexane/EtOAc, v/
v, 8/1), 90% isolated yield for both reactions.
Synthesis of Racemic b-Keto Alcohols 1–7 and 11, 12
and 13
Under
a
nitrogen atmosphere sodium borohydride
(0.098 mmol) was added in dry ethanol (10 mL) and the
mixture was cooled to 08C for 5 min. A solution of dry etha-
nol (5 mL) containing diketone (0.3 mmol) was then added
drop wise. After stirring at 08C for 2–4 h, TLC and GC
analysis confirmed reaction completion. The reaction mix-
ture was then cooled to 08C, saturated ammonium chloride
(1 mL) was added and the mixture was evaporated to dry-
ness. After addition of 15 mL water, the mixture was ex-
tracted twice with ethyl acetate (2×20 mL). The combined
organic layers were dried over MgSO4, and evaporated to
dryness. Pure products were obtained using silica gel chro-
matography (hexane/EtOAc, v/v, 9/1), 70% to 80% isolated
yield.
Synthesis of Diketones 8, and 9, by Oxidation of the
Corresponding Hydroxy Ketone with PCC
A typical protocol is described. Under a nitrogen atmos-
phere pyridinium chlorochromate (PCC, 4.9 g, 22.5 mmol)
was added in dry methylene chloride (10 mL) and the mix-
ture was cooled to 08C. Then a solution of hydroxy ketone
(15 mmol, 2.16 g of keto alcohol 8 or 1.95 g of keto alcohol
9, whose synthesis is described in section 2.2), in dry CH2Cl2
(5 mL) was slowly added and the reaction mixture was stir-
red at 08C for 2 h where completion of the reaction was ob-
served by TLC analysis. Diethyl ether (100 mL) was then
added followed by double filtration over solid Floricil.
Washing of the solids with Et2O, and evaporation of the
combined organic layers gave the product, which was used
for enzymatic reduction without any further purification.
Pure products were also obtained using silica gel chromatog-
raphy (hexane/EtOAc, v/v, 5/1), 70% isolated yield in both
cases.
The same method was used for the reduction of diketone
13. After purification with flash column chromatography
(hexane/EtOAc, v/v, 9/1), 30% of the racemic keto-alcohol
was isolated.
Synthesis of Racemic Keto Alcohols 8, 9 and 10 by
Aldol Condensation
These substrates were prepared by aldol condensation start-
ing from pentan-3-one or 1-phenyl-1-propanone and addi-
tion of propionaldehyde for compound 8 and acetaldehyde
for 9 and 10. A typical protocol for their synthesis is de-
scribed as follows. Under a nitrogen atmosphere, diisopro-
pylamine (4.2 mL, 30 mmol) was dissolved in dry THF
(15 mL). The mixture was cooled to 08C and BuLi 1.45m in
hexane (13.8 mL, 20 mmol) was added dropwise. After stir-
ring for 15 min at room temperature, the mixture was
cooled to À788C and a solution of pentan-3-one (2.1 mL,
20 mmol) in dry THF (10 mL) or 1-phenyl-1-propanone
(2.66 mL, 20 mmol) was added and the mixtures were stirred
Synthesis of Diketone 13
a) 1-Acetyl-2-oxopropyl acetate was prepared from commer-
cially available 3-chloro-2,4-pentanedione according to the
following procedure: under a nitrogen atmosphere dry
sodium acetate (902 mg, 11 mmol) and 3-chloro-2,4-pentane-
dione (1.25 mL, 11 mmol) were dissolved in 20 mL dry
DMF. The reaction mixture was left under stirring at 508C
overnight. After completion of the reaction, 50 mL H2O
were added and the mixture was extracted three times with
CHCl3 (60 mL). The organic layer was washed three times
with H2O (60 mL), dried over MgSO4 and evaporated to
dryness. The pure product, 1-acetyl-2-oxopropyl acetate, was
obtained using silica gel chromatography (hexane/EtOAc, v/
v, 8/1) 60% isolated yield.
for 20 min.
A solution of aldehyde (propionaldehyde
1.44 mL or acetaldehyde 1.12 mL, 20 mmol) in dry THF
(10 mL) was then added over a period of 20 min and the so-
lution was stirred for another 3 h (temperature was slowly
increased to 08C). Completion of the reaction was observed
by TLC analysis of reaction aliquots. The reaction mixture
was quenched with saturated NH4Cl (50 mL) and was then
extracted with diethyl ether (60 mL). The organic layer was
washed with saturated NaCl (100 mL) followed by distilled
water (100 mL). After drying over MgSO4 evaporation to
b) 1-Acetyl-1-methyl-2-oxopropyl acetate, diketone 13,
was prepared by following the same procedure as in the
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1967

Dimitris Kalaitzakis et al.
FULL PAPERS
dryness, and silica gel chromatography (hexane/EtOAc, v/v,
5/1), 70% of pure keto alcohols were isolated.
Enzymatic Reduction of Keto Alcohols 2a–c to the
Corresponding 1,3-Diols (Table 2)
After the enzymatic production of a-ethyl-b-hydroxy ketone
to each 2a–c diastereomers using KRED102 (2b),
KRED118 (2b) and EXP-A1B (2c) reduction to the 1,3-diol
was performed. In the same buffer and after the synthesis of
2a–c (1 mL, 200 mM, pH 6.9), 4 mg of a different ketoreduc-
tase, 2 mg of glucose dehydrogenase, 2 mg NADPH and
18 mg of glucose were added. The reactions were incubated
at 378C and reaction aliquots were taken at various time
points. After extraction with ethyl acetate they were ana-
Enzymatic Reductions
Thirty-four different ketoreductases (KRED-101–131 kit;
and three from the KRED-EXP-A1B kit; BioCatalytics,
Inc. Pasadena, CA USA) were screened for reduction. In
addition to the ketoreductases, both NADPH and glucose
dehydrogenase (GDH) that was used for cofactor recycling,
are products available from BioCatalytics.
1
lyzed by GC chromatography. H NMR and absolute config-
uration analysis data are included in the Supporting Infor-
mation.
Small-Scale Screening of Enzymatic Reactions (Table
1)
The screening was performed in small-scale reactions where
in a phosphate buffered solution (1 mL, 200 mM, pH 6.9)
each substrate (25 mM) was mixed with NADPH (2.5 mM,
2 mg) and 2 mgmLÀ1 of each ketoreductase. The mixture
also contained glucose (100 mM, 18 mg) and glucose dehy-
drogenase (GDH, 2 mgmLÀ1) for cofactor recycling. The re-
actions were incubated at 378C and reaction aliquots were
taken at various time intervals. After extraction with ethyl
acetate the crude extracts were analyzed by GC chromatog-
raphy. All the small-scale enzymatic reductions were per-
formed according to the previous protocol, except for dike-
tones 1 and 9, where the reaction was buffered at pH 6.5
and diketone 13 at pH 7.5. Reaction times and yields are
shown in Table 1. Isolated products were analyzed by
1H NMR and are shown in the supplementary material.
Enzymatic Reduction of Substrate 13 with KRED 112
In 8 mL phosphate buffer (200 mM, pH 7.5) were added
2 mg of KRED112, 2 mg GDH, 2 mg NADPH, 200 mg glu-
cose (140 mM) and then 200 mg substrate (145 mM) were
added periodically (10 mg every 5 min). The reaction was in-
cubated at room temperature for two hours until completion
of the reaction. The product was isolated by extracting the
crude reaction mixture with EtOAc (10 mL × 2). The com-
bined organic layers were then extracted with saturated
NaCl solution, dried over MgSO4 and evaporated to dryness
to give 180 mg (90%) of pure keto alcohol 13a.
Enzymatic Reduction of Keto Alcohol 13a
The enzymatic reduction of keto alcohol 13a was accom-
plished by following the same procedure as in enzymatic re-
ductions of keto alcohols in small scale.
Larger-Scale Enzymatic Reductions
Larger-scale enzymatic reductions were prepared according
to the following protocol. A phosphate-buffered solution
(100 mL, pH 6.5 or 6.9, 200 mM) containing 50 mM sub-
strate, NaCl (200 mM), glucose (120 mM), NADPH
(0.5 mM), glucose dehydrogenase (50 mg) and the appropri-
ate ketoreductase (50–150 mg) was stirred at 378C for 24 h,
until GC analysis of crude extracts showed complete reac-
tion. Periodically the pH was readjusted to 6.5 or 6.9 with
NaOH (2M). Product was isolated by extracting the crude
reaction mixture with EtOAc (70 mL × 2). Sometimes cen-
trifugation (6000 rpm, 10 min) was required for the aqueous
and the organic layers to separate as clear solutions. The
combined organic layers were then extracted with saturated
NaCl solution, dried over MgSO4 and evaporated to dryness.
Pure, optically active alcohols were obtained in 85% to
90% yield. The products were analyzed by NMR spectros-
copy. Their optical purity was determined by chiral GC
chromatography using 20% permethylated cyclodextrin
column.
Acknowledgements
IS thanks Prof Frances H. Arnold for her generous hospitali-
ty during her sabbatical stay at Caltech (2006). Financial sup-
port for this workcame from the GreekSecretariat of Re-
search and Technology (Pythagoras I, 2004 and PENED
2003). D. K. thanks the Greek National Scholarships Founda-
tion (IKY) for providing a three year fellowship. We also
thankProf. G. J. Karabatsos for his valuable comments.
References
[1] a) D. Sawada, M. Kanai, M. Shibasaki, J. Am. Chem.
Soc. 2000, 122, 10521; b) I. Paterson, D. Y.-K. Chen,
J. L. Acena, A. S. Franklin, Org. Lett. 2000, 2, 1513;
c) F. Eustache, P. I. Dalko, J. Cossy, Tetrahedron Lett.
2003, 44, 8823.
[2] a) M. Kobayashi, K. Higuchi, M. Murakami, Y. Tajima,
S. Aoki, Tetrahedron Lett. 1997, 38, 2859; b) J. D
White, R. Hanselmann, R. W. Jackson, W. J. Porter, Y.
Ohba, T. Tiller, S. Wang, J. Org. Chem. 2001, 66, 5217;
c) Y. Che, J. B. Gloer, D. T. Wicklow, Org. Lett. 2004, 6,
1249; d) M. Magnin-Lachaux, Z. Tan, B. Liang, E.-i.
Negishi, Org. Lett. 2004, 6, 1425; e) G. F. Liou, C.
Khosla, Curr. Opin. Chem. Biol. 2003, 7, 279; f) R. W.
According to this protocol diketone 1 (570 mg, 50 mM)
was reduced with KRED-102 (150 mg) at pH 6.5. Isolated
yield 513 mg, 90%. Diketone 8 (710 mg, 50 mM) was re-
duced with KRED-119 (100 mg) at pH 6.9. Isolated yield
618 mg, 87%. Diketone 8 was also reduced with KRED-
A1B (50 mg) at pH 6.9. Isolated yield: 604 mg, 85%.
1968
www.asc.wiley-vch.de
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1958 – 1969

Synthesis of a-Alkyl-b-hydroxy Ketones and 1,3-Diols
FULL PAPERS
Hoffmann, Angew. Chem. Int. Ed. 1987, 26, 489; g) I.
Paterson, V. A. Doughty, G. Florence, K. Gerlach,
M. D. McLeod, J. P. Scott, T. Trieselmann, ACS Sympo-
sium Series 2001, 783 (Organoboranes for Syntheses),
195–206; h) M. Chartrain, P. M. Salmon, D. K. Robin-
son, B. C. Buckland, Curr. Opin. Biotechnol. 2000, 11,
209; i) J. Staunton, K. J. Weissman, Nat. Prod. Rep.
2001, 18, 380; j) R. W. Hoffmann, W. Helbig, W.
Ladner, Tetrahedron Lett. 1982, 23, 3479; k) J. L. Vicar-
io, A. Job, M. Wolberg, M. Müller, D. Enders, Org.
Lett. 2002, 4, 1023; l) N. Arai, N. Chikaraishi, S.
Omura, I. Kuwajima, Org. Lett. 2004, 6, 2845; m) F. J.
Schmitz, S. P. Gunasekera, G. Yalamanchili, M. B. Hos-
sain, D. Van der Helm, J. Am. Chem. Soc. 1984, 106,
7251.
T. Harada, Tetrahedron:Asymmetry 2003, 14, 2659;
d) M. J. Homann, R. B. Vail, E. Previte, M. Tamarez, B.
Morgan, D. R. Dodds, A. Zaks, Tetrahedron 2004, 60,
789; e) M. Muller Angew. Chem. Int. Ed. 2005, 44, 362;
f) M. Wolberg, W. Hummel, C. Wandrey, M. Muller
Angew. Chem. Int. Ed. 2000, 39, 4306; g) K. Edegger,
W. Stampfer, B. Seisser, K. Faber, S. F. Mayer, R. Oehr-
lein, A. Hafner, W. Kroutil Eur. J. Org. Chem. 2006,
1904; h) B. Kosjek, W. Stampfer, M. Pogorevc, W.
Goessler, K. Faber, W. Kroutil Biotechnol. Bioeng.
2004, 86, 55; i) T. Ema, H. Moriya, T. Kofukuda, T.
Ishida, K. Maehara, M. Utaka, T. Sakai J. Org. Chem.
2001, 66, 8682.
[12] a) S. Kambourakis, J. D. Rozzell, Adv. Synth. Catal.
2003, 345, 699; b) S. Kambourakis, J. D. Rozzell, Tetra-
hedron 2004, 60, 663; c) D. Zhu, C. Mukherjee, J. D.
Rozzell, S. Kambourakis, L. Hua, Tetrahedron 2006, 62,
901; d) D. Zhu, B. E. Rios, J. D. Rozzell, L. Hua, Tetra-
hedron: Asymmetry 2005, 16, 1541; e) D. Zhu, C. Mar-
kherjee, L. Hua, Tetrahedron:Asymmetry 2005, 16,
3275.
[3] F. Eustache, P. J. Dalko, J. Cossy, Org. Lett. 2002, 4,
1263.
[4] Y. Ohtsuka, K Koyasu, T. Ikeno, T. Yamada, Org. Lett.
2001, 3, 2543.
[5] J. L. Vicario, D. Badia, E. Dominguez, M. Rodriguez,
L. Carrillo, J. Org. Chem. 2000, 65, 3754.
[6] N. A. Van Draanen, S. Arseniyadis, M. T. Crimmins,
H. C. Heathcock, J. Org. Chem. 1991, 56, 2499.
[7] a) H-X. Wei, R. L. Jasoni, H. Shao, J. Hu, P. W. Pare,
Tetrahedron 2004, 60, 11829; b) J. Jankowska, J. Mly-
narski, J. Org. Chem. 2006, 71, 1317; c) R. Mahrwald,
B. Schetter, Org. Lett. 2006, 8, 281; d) N. Mase, Y.
Nakai, N. Ohara, H. Yoda, K. Takabe, F. Tanaka, C. F.
Barbas III, J. Am. Chem. Soc. 2006, 128, 734.
[8] D. A. Evans, J. V. Nelson, E. Vogel, T. R. Taber, J. Am.
Chem. Soc. 1981, 103, 3099.
[13] a) I. A. Kaluzna, J. D. Rozzell, S. Kambourakis, Tetra-
hedron:Asymmetry 2005, 16, 3682; b) D. Pollard, M.
Truppo, J. Pollard, C. Chen, J. Moore Tetrahedron:A-
symmetry 2006, 17, 554; c) T. W. Johannes, R. D. Wood-
yer, H. Zhao Appl. Environ. Microbiol. 2005, 71, 5728.
[14] D. Kalaitzakis, J. D. Rozzell, S. Kambourakis, I.
Smonou, Org. Lett. 2005, 7, 4799.
[15] D. Kalaitzakis, J. D. Rozzell, S. Kambourakis, I.
Smonou, Eur. J. Org. Chem. 2006, 2309.
[16] C. H. Heathcock, M. C. Pirrung, J. E. Sohn, J. Org.
Chem. 1979, 44, 4294–4299.
[9] Y. Yamamoto, K. Maruyama, Tetrahedron Lett. 1980,
21, 4607;.
[17] J. M. Seco, E. Quinoa, R. Riguera, Chem. Rev. 2004,
[10] C. H. Heathcock, C. T. Buse, W. A. Kleshick, M. C. Pir-
rung, J. E. Sohn, J. Lampe, J. Org. Chem. 1980, 45,
1066.
[11] a) D. Buisson, S. Henrot, M. Larcheveque, R. Azerad,
Tetrahedron Lett. 1987, 28, 5033; b) A. Fauve, H. Ve-
schambre, J. Org. Chem. 1988, 53, 5215; c) For a recent
review see: K. Nakamura, R. Yamanaka, T. Matsuda,
104, 17.
[18] a) I. Paterson, J. M. Goodman, M. A. Lister, R. C.
Schumann, C. K. McClure, R. D. Norcross, Tetrahedron
1990, 46, 4663; b) A. Fauve, H. Veschambre, Tetrahe-
dron Lett. 1987, 28, 5037.
[19] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 14,
2923–2925.
Adv. Synth. Catal. 2006, 348, 1958 – 1969
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1969