Investigating the reaction mechanism and organocatalytic synthesis of α,α′-dihydroxy ketones

Article abstract of DOI:10.1039/c2ob06939c

A biomimetic TK one-pot reaction using hydroxypyruvate and aldehydes to generate α,α′-dihydroxy ketones in water has recently been described. To investigate this tertiary-amine mediated reaction mechanism two approaches were used. Firstly, ¹³C

Full text of DOI:10.1039/c2ob06939c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 2621
www.rsc.org/obc
PAPER
Investigating the reaction mechanism and organocatalytic synthesis of
α,α′-dihydroxy ketones†
James L. Galman, David Steadman, Lisa D. Haigh and Helen C. Hailes*
Received 18th November 2011, Accepted 31st January 2012
DOI: 10.1039/c2ob06939c
A biomimetic TK one-pot reaction using hydroxypyruvate and aldehydes to generate α,α′-dihydroxy
ketones in water has recently been described. To investigate this tertiary-amine mediated reaction
mechanism two approaches were used. Firstly, 13C labelled lithium hydroxypyruvate was synthesised and
used to establish where hydroxypyruvate is incorporated in the product. In separate experiments reaction
intermediates were also successfully intercepted and structurally identified using ESI-MS with tandem
mass spectrometry ESI-MS/MS. These studies indicated that two mechanisms appear to be operating, one
involving the addition of the tertiary amine catalyst to hydroxypyruvate, the other an aldol-based
mechanism. Since the first mechanism may enable facial stereodifferentiation in the addition of
intermediates to the aldehyde, a preliminary study on the use of chiral catalysts was performed and the
first asymmetric organocatalytic synthesis of α,α′-dihydroxy ketones in aqueous media achieved, in up to
50% ee, using a quinine ether catalyst.
Introduction
Transketolase (TK) is a versatile enzyme that requires thiamine
diphosphate (ThDP) and magnesium(II) ions as cofactors to
reversibly transfer a two carbon ketol unit from D-xylulose-5-
phosphate to D-ribose-5-phosphate, in the oxidative branch of
the pentose phosphate pathway.1 To render the reaction irrevers-
ible, Srere et al. demonstrated in vitro the use of β-hydroxypyru-
vate (HPA 1) as a ketol donor for TK, with the evolution of
carbon dioxide.2 Since then, TKs have been used by several
groups with a range of acceptor aldehydes 2 and HPA 1 for the
construction of (3S)-α,α′-dihydroxy ketones 3 in moderate to
excellent yields and stereoselectivities (Scheme 1).3–10 Although
in vivo α-hydroxylated aldehydes are utilised, wild-type TKs
have been reported to tolerate non-α-hydroxylated aliphatic alde-
hydes, but lower activities are observed.8–10 More recently,
E. coli TK variants possessing single-point active site mutations
have been reported that are able to accept non-α-hydroxylated
aliphatic aldehydes and generate either the (3S)- or (3R)-α,α′-
dihydroxy ketone 3 in good yields.11–13 E. coli TK mutants have
also been described that are interestingly able to accept aromatic
aldehydes such as benzaldehyde, although ees were variable and
Scheme 1 Formation of α,α′-dihydroxy ketones using TK or the bio-
mimetic TK reaction using N-methylmorpholine.
dependent on the substrate and variant used.14 In addition, yields
were low and side products formed in several cases. However,
there is clearly significant interest in the synthesis of compounds
possessing the α,α′-dihydroxy ketone functionality, which is
found in several natural products and can be readily converted to
a range of compounds including ketosugars and 2-amino-1,3-
diols.5,6,15–19
Non-enzymatic routes to α,α′-dihydroxy ketones have been
described, including a five step procedure to aromatic analogues
and an enantioselective chiral auxiliary strategy reported by
Enders et al., but these are multistep procedures, highlighting the
value of the enzyme TK in accessing these compounds using a
succinct sustainable approach.20,21
Department of Chemistry, University College London, 20 Gordon Street,
London WC1H 0AJ, UK. E-mail: h.c.hailes@ucl.ac.uk; Fax: +44 (0)20
7679 7463; Tel: +44 (0)20 7679 4654
Notably, a recent preliminary study has described a non-enzy-
matic one-pot tertiary amine catalysed reaction in water, using
Li-1 as the ketol donor and aldehyde acceptor to give 3
(Scheme 1).13,14,22 This biomimetic TK reaction typically used
†Electronic supplementary information (ESI) available: Conversion of
3d, 3e, 3f, 3i, 3k, 3l into Mosher’s esters. Also spectral data (1H and 13
C
NMR) for 3a–3l and chiral HPLC data where appropriate. See DOI:
10.1039/c2ob06939c
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2621–2628 | 2621

View Article Online
Scheme 2 Proposed mechanisms A–D for the formation of 3 from 1 and 2, and the position of incorporation of [3-13C]-labelled Li-HPA. Intermedi-
ates 4–7 are also indicated. For the discussion below 2a, 2b, 2c are where R is cy-C6H11, Ph and thienyl, respectively.
stoichiometric amounts of N-methylmorpholine (NMM) and
yields were dependent upon the acceptor aldehyde used, giving
variable yields of 3 from 2% to 63%, but in the absence of
amine the reaction did not proceed.13,14,22 More recently, this
transformation has been described using HPA in stereoselective
decarboxylative aldols, where NMM was identified as the
optimal catalyst and was used in sub-stoichiometric or stoichio-
metric quantities, together with single isomer protected
α-hydroxylated aldehydes. Anti : syn diastereoselectivities were
observed of between approximately 1 : 1 and 2 : 1.23
There is currently no one-step non-enzymatic asymmetric syn-
thesis of α,α′-dihydroxy ketones using non-chiral aldehydes,
only multistep strategies. However, the biomimetic TK reaction
requirement for a tertiary amine catalyst might enable develop-
ment as an asymmetric reaction by the use of chiral catalysts.
Herein we report studies to provide mechanistic insights into the
reaction, and also investigations to determine whether an asym-
metric version of the TK biomimetic reaction can be achieved.
were not discussed.23 In principle either the direct aldol addition
of 1a to the aldehyde could occur to give 6 followed by decar-
boxylation, protonation, and tautomerisation to give
3
(Scheme 2; mechanism B). An alternative aldol addition, of 1a
to the aldehyde to give 7, with subsequent tautomerisation then
decarboxylation and tautomerisation would generate the dihy-
droxy ketone 3 (Scheme 2; mechanism C).
Finally, a further mechanism (Scheme 2; mechanism D) invol-
ving initial decarboxylation of 1b, generated from 1, would give
a symmetrical enediolate intermediate, which upon an aldol
addition to 2 and subsequent tautomerisation would give 3. To
provide insights into the role of the tertiary amine, which may be
important in designing an asymmetric version of the reaction,
two mechanistic studies were performed: one using 13C-labelled
Li-HPA, and in separate experiments mass spectrometry was
used to identify any key reaction intermediates. Depending on
the mechanism of the reaction there are two positions that
[3-13C]-Li-1 could be incorporated as shown in Scheme 2. If
mechanisms A or B are involved then the 13C-label would be
found at [1-13C]-3. However, for the aldol mechanism C then
[2-13C]-3 would be generated. For mechanism D a mixture of
13C labelling would be observed at positions C-1 and C-2.
[3-13C]-Li-1 was prepared via the bromination of [3-13C]-sodium
pyruvate (99 atom% 13C) using neat bromine at 50 °C to give
the monobrominated product [3-13C]-bromopyruvate in 70%
yield.24,25 Then, following a literature protocol to synthesise
Li-1, 1 M lithium hydroxide was added slowly to [3-13C]-bromo-
pyruvate to afford [3-13C]-Li-1 in 54% yield.26
Results and discussion
Mechanistic studies using 13C labelled Li-HPA
In a preliminary publication, due to the requirement for the ter-
tiary amine and the ability of the reaction to proceed at pH 7–8 it
was proposed that the mechanism could be reminiscent of the
Baylis–Hillman reaction, with addition of the amine to carbanion
enolate 1a in a conjugate fashion or tartronic acid semialdehyde
1b to give a quaternary amine enolate 4 (Scheme 2; mechanism
A).22 Then aldol addition of the aldehyde, proton transfer to give
5 and elimination of the tertiary amine and decarboxylation, with
subsequent tautomerisation of the enediol, gives 3.
Cyclohexanecarboxaldehyde 2a and benzaldehyde 2b were
used as representative aliphatic and aromatic aldehydes in the
TK biomimetic reaction with [3-13C]-Li-1 and NMM, affording
[1-13C]-3a and [1-13C]-3b in 15% and 10% isolated yields,
respectively (Scheme 3). The 13C NMR spectra of [1-13C]-3a
and [1-13C]-3b indicated that the 13C label was incorporated at
Alternatively, aldol-based reactions could be envisaged as
mentioned by Rohr and Mahrwald, although mechanistic details
2622 | Org. Biomol. Chem., 2012, 10, 2621–2628
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 3 Synthesis of [1-13C]-3a and [1-13C]-3b using [3-13C]-Li-1.
C-1 in both cases. This was confirmed from the intensity of the
C-1 signal compared to unlabelled natural abundance 3a and 3b
1
13C NMR spectra previously reported.13,14 In addition, the H
NMR spectra confirmed this with a 1JCH at C-1 of approximately
145 Hz. The [3-13C]-Li-1 labelling experiment therefore indi-
cated that mechanisms C and D were not involved in this TK
biomimetic reaction, but mechanisms A or B may occur with
addition of the aldehyde at C-2 (Scheme 2; mechanisms A
and B).
Fig. 1 ESI(−)-MS for the biomimetic TK reaction with 2c and Li-1
catalysed by DABCO.
ESI-MS studies
ESI-MS and tandem ESI-MS/MS have proven to be useful
methods to probe mechanistic studies in solution, by characteris-
ing reaction intermediates and the selection and characterisation
of specific ions, as well as minor products formed.27–29 Of par-
ticular relevance to our current study, it has been used to investi-
gate organocatalytic reactions including the Baylis–Hillman
reaction, the asymmetric intramolecular Michael reaction of alde-
hydes, and organocatalytic cascades.30–32 With mechanisms A or
B implicated in the TK biomimetic mechanism ESI-MS (and
tandem ESI-MS/MS) was used to intercept and characterise reac-
tion intermediates. Initially, the reaction of cyclohexanecarbox-
aldehyde 2a with Li-1 catalysed by NMM in water was
monitored by ESI-MS with the aim of detecting reaction inter-
mediates in solution in the positive or negative ion mode. Unfor-
tunately no intermediary ions were detected, presumably due to
instabilities. Alternative aldehydes and tertiary amines were
explored and when utilizing 2-thiophenecarboxaldehyde 2c
(which has previously been used with TK variants) Li-1 and the
tertiary amine DABCO in water, intermediates were detected.14
The reaction was continuously monitored by direct infusion of
the aqueous solution into the ESI probe at a flow rate of 3 μL
min−1. A high cone voltage (85 V) was used to de-clutter and
eliminate intermediate ions defined by non-covalent interactions
in the gaseous phase; under these conditions highly charged
covalent species were detected (Fig. 1).
(favoured under the negative-ion ESI), side reaction products or
corresponded to fragmentation of the product.
These reactive anionic intermediate ions were individually
mass selected and collided with nitrogen at a high collision ion-
isation dissociation voltage (CID). This lowered the abundance
of the intermediate ions with the generation of fragments for
further individual structural characterisation in tandem mass
spectrometric (MS/MS) analysis. The relatively high CID (17
eV) produced less rearranged products and the elimination of
loosely held dimerisation adducts with exclusive higher energy
fragments were observed. ESI-(−)MS/MS spectra confirmed
[6c]− of m/z 215 as the aldehydic anionic intermediate with a
loss of carbon dioxide to form the product ion [3c-H]− of m/z
171. Other fragmented ions resulted from the loss of carbon
dioxide and methanol radical to yield [8-H]− of m/z 141 and the
loss of carbon dioxide and water to yield [9-H]− of m/z 153. The
same CID (17 eV) was also used to fragment intermediate [5c-
H]− of m/z 327 by the loss of carbon dioxide to yield [10-H]− of
m/z 283. Other dissociation routes revealed the loss of carbon
dioxide and water to yield [11-H]− of m/z 265, and the most
stable detectable fragment via the cleavage of DABCO and
methanol radical to yield [12-H]− of m/z 185 (Fig. 2).
The ESI(−)-MS spectra revealed signals at m/z 215 and 327
which correspond to the covalently bonded anionic intermediates
[6c]− (mechanism B) and [5c-H]− (mechanism A) respectively,
and therefore could potentially be of mechanistic importance.
The m/z of 215 could also correspond to 4c although later MS/
MS studies indicated this was not the case (see below). In
addition, an accurate mass of the signal (m/z 215.0004; calcd for
C8H7O5S, 215.0014), confirmed it corresponded to the molecu-
lar formula of 6c. The relative abundance of m/z 215 in compari-
son with m/z 327 might suggest pathway B is preferred, however
the low abundance of m/z 327 may be due to instability and
decomposition of the fragment at the ion source before it is
detected. Other ions were considered to be either impurities
Preliminary asymmetric TK biomimetic reactions
These ESI-MS studies therefore indicated that both mechanisms
A and B operate, indeed the predominant mechanism may be
dependent on the aldehyde and tertiary amine combination used,
together with the reaction conditions. No further mechanistic
insights were obtained from NMR experiments to detect reaction
intermediates. However, investigating the reaction between 1 and
1
2a in D2O over 24 h by H NMR with different mol% of NMM
(0, 5, 25, 50, 100) at pH 8 (see ESI†), indicated that the rate of
the reaction was proportional to the concentration of NMM
used (at <50 mol%), further supporting the involvement of
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2621–2628 | 2623

View Article Online
Table 1 Use of tertiary amine catalysts 13 and 14 in the biomimetic
TK reaction with Li-1 and 2b
Entry
Catalyst
Solvent
Time [h]
Yield [%]
ee [%]
1
2
3
4
5
6
7
8
9
13aa
13ba
13ca
13da
14aa
14ba
14ca
14da
14db
14dc
H2O
H2O
H2O
H2O
H2O–THF
H2O–THF
H2O–THF
H2O–THF
H2O–THF
H2O–THF
12
12
12
12
24
24
24
24
24
24
5
6
5
5
6
5
4
4
6
6
3
2
2
0
0
0
18
22
16
0
10
Reaction conditions: 2b (1.0 equiv) and Li-1 (1.0 equiv) in THF (1 mL)
and H2O (1 mL). a Amine catalyst (5 mol%). b Amine catalyst (50 mol
%). c Amine catalyst (1.0 equiv).
intermediates, altering facial stereoselectivities. The absolute
stereochemistry of the major isomer of 3b generated was estab-
lished as (3R) by formation of the Mosher’s ester at the primary
hydroxyl with (S)-MTPACl as previously described.42 Having
demonstrated some asymmetric induction with the quinine ether
catalysts, 14d was used further with other aromatic aldehydes
and the highest ees achieved under a range of reaction conditions
explored are presented in Table 2.
Fig. 2 ESI(−)-MS/MS spectra of m/z 215 [6c]− and m/z 327 [5c-H]−.
mechanism A. Studies were then performed to investigate
whether an asymmetric TK biomimetic reaction could be estab-
lished. If the reaction goes via mechanism A, the use of a single
isomer chiral tertiary amine would give an intermediate 4 which
could direct re- or si-addition to the aldehyde by facial stereodif-
ferentiation. For mechanism B a chiral tertiary amine cation with
an electrostatic interaction to the carboxylate of 1a might also
result in some asymmetric induction. However, asymmetric reac-
tions in water are challenging as water can interrupt ionic and
hydrogen bonding interactions critical for stabilising transition
states of the reaction or inhibit the catalyst’s activity.33 Initial
experiments were performed with Li-1 and benzaldehyde 2b and
a range of selected chiral catalysts that have previously been suc-
cessfully used as organocatalysts. Notably the pyrrolidine di-
amines 13a–d have recently been used in asymmetric Baylis–
Hillman reactions in protic solvents, and asymmetric Michael
additions and aldol reactions in aqueous media.34–38 The cinch-
ona alkaloids are well established asymmetric catalysts and com-
pounds 14 were selected including the hydroquinine analogues
14c and 14d.39–41
Stereoselectivities were determined via dibenzoylation of 3b
and chiral HPLC as previously described for TK reaction pro-
ducts.14 The pyrrolidine diamines 13 (Table 1; entries 1–4) gave
3b in poor yields with negligible ees. The addition of water
stable Lewis acids Sc(OTf)3 and Yb(OTf)3 to the reactions with
13 resulted in no increase in yield or ee. The bulkier catalysts
quinine and quinidine (Table 1; entries 5 and 6) gave no asym-
metric induction with varied catalyst loading, and water–THF
was required as a solvent system to ensure solubilisation of the
amine catalysts. Interestingly, the highest selectivities were
observed with the quinine ether derivatives 14c and 14d when
used in a (1 : 1) H2O–THF solvent system. When the catalyst
loading was increased from 5 mol% to one equivalent no ee was
observed (Table 1; entries 8–10). This is possibly due to the
complexation of several catalyst molecules to reaction
In all cases the yields were low, comparable to those observed
with the TK mutant enzymes reported,14 however, the stereo-
selectivities were promising. The yields observed were lower
than those achieved with the less bulky tertiary amine catalyst
NMM, perhaps reflecting steric hindrance upon the addition of
the amine in mechanism A. The para-substituted F- and Cl-
2624 | Org. Biomol. Chem., 2012, 10, 2621–2628
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 The asymmetric biomimetic TK reaction with various
aldehydes
Entry
Aldehyde
Time [h]
Yield [%]
ee [%]
1
2
3
4
5
6
7
8
9
2da R = 4-Me
2ea R = 4-F
12
12
48
24
72
12
48
24
24
2
3
2
3
2
2
3
3
4
21
36
50
0
0
38
0
2fb R = 4-Cl
2gb R = 4-OMe
2hb R = 4-Br
2ia R = 3-F
2ja R = 2-F
Fig. 3 Proposed rationale for stereoselectivities observed with the
addition of 4 to aldehyde 2.
2kc R = 3-OMe
2lc R = 3-CO2Me
36
25
Reaction conditions: 2d–2l (1.0 equiv) and Li-1 (1.0 equiv), solvent
(2 mL). a 2d, 2e, 2i, 2j 5 mol% 14d, 1 : 1 THF–H2O. b 2f, 2g, 2h 20 mol
% 14d, 1 : 2 THF–H2O. c 2k, 2l 10 mol% 14d, 1 : 1 THF–H2O.
Conclusions
In conclusion, we have used 13C labelling experiments to
provide evidence to support that mechanisms A and B operate in
the tertiary amine catalysed TK biomimetic reaction in aqueous
media. Also, some mechanism A and B reaction intermediates
have successfully been intercepted and structurally identified
using ESI with tandem mass spectrometry. Finally, we have
demonstrated the first asymmetric organocatalytic synthesis of
α,α′-dihydroxy ketones in aqueous media catalysed by a quinine
ether catalyst 14d with moderate ees obtained. Further tertiary
amine catalyst design that increases the stereoselectivities and
yield, and are applicable to a wider range of substrates, can now
be investigated.
aromatic aldehydes gave moderate ees in comparison to the
larger p-methoxy- and p-bromo-substituents where no ee was
observed. This decrease in optical purity suggested that the
increased size of the para-group may detrimentally influence
stereoselectivities. The meta-substituted aromatic aldehydes con-
taining both electron withdrawing and donating groups, F–,
MeO– and MeO2C–, gave products with ees in the range
25–38%, indicating that the reactivity or the steric bulk of these
substituents do not contribute significantly to the asymmetric
induction observed. The ortho-analogue 2j gave 3j in 0% ee,
and also 2,4-dichlorobenzaldehyde gave no hydroxy ketone
product when using amine 14d, so ortho-substituted benzal-
dehydes were not explored further with these catalysts. Aliphatic
aldehydes such as cyclohexanecarboxyaldehyde and hexanal
were investigated using amine 14d. Higher yields of 15–20%
were observed, but no enantioselectivities achieved. As for 3b
the optical purities of the dihydroxy ketone products (Table 2)
were determined by derivatisation (as the dibenzoate or mono-
benzoate) and chiral HPLC. The low yields observed when
using 14d may be due to the bulky nature of the quinine catalysts
facilitating the stereoselective reaction. As discussed above there
are two mechanisms A and B that both result in the formation of
3. Route A may more readily enable asymmetric inductions to be
achieved due to the formation of zwitterion 4 with the tertiary
amine covalently bound to HPA, enabling facial stereodifferen-
tiation upon addition to the aldehyde. Mechanism B would
require a strong electrostatic interaction between 14d and 1,
which in aqueous media will be disrupted by hydrogen bonding.
When considering the zwitterionic intermediate 4 and addition
to the aldehyde, as shown in Fig. 3, the lower face of the enolate
will not be accessible to the aldehyde. Then addition to the si-
face of the aldehyde (underside as drawn) may result from a
favourable interaction between Ar and methylquinoline ring and
would generate the (R)-product. Lower or negligible ees may
either reflect predominance of mechanism B with certain sub-
strate and catalyst combinations or that the aldehyde equally pre-
sents the re-face to 4 due to unfavourable steric interactions.
Experimental
General methods
Unless otherwise noted, solvents and reagents were reagent
grade from commercial suppliers (Sigma-Aldrich) and used
without further purification. Dry CH2Cl2 was obtained using
anhydrous alumina columns.43 All moisture-sensitive reactions
were performed under a nitrogen or argon atmosphere using
oven-dried glassware. Reactions were monitored by TLC on Kie-
selgel 60 F254 plates with detection by UV, potassium permanga-
nate and phosphomolybdic acid (PMA) [PMA hydrate (12 g)
and ethanol (250 mL)] stains. Flash column chromatography was
1
carried out using silica gel (particle size 40–63 μm). H NMR
and 13C NMR spectra were recorded using a Bruker Avance-
600 MHz machine. Coupling constants are measured in Hertz
(Hz) and NMR spectra were recorded at 298 K. Mass spectra
(EI/CI) were recorded on a Thermo Finnegan MAT 900XP, ESI
data (MS1 and MS2) on a Thermo Finnegan LTQ and HR-ESI
data on a Waters LCT Premier XE. Infrared spectra were
recorded on a Shimadzu FTIR-8700 and Perkin Elmer Spectrum
100 FTIR spectrometer. Chiral HPLC analysis was performed on
a Varian Prostar instrument equipped with a Chiracel OD
column (Daicel; Chiral Technologies Europe, France) 25 cm ×
0.46 cm.
[3-13C]-Bromopyruvate
and lithium
hydroxypyruvate
(13C-labelled and unlabelled) were synthesised as previously
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2621–2628 | 2625

View Article Online
described.24–26 The pyrrolidine diamines 13a–d were prepared
as previously reported.38,44,45 The cinchona alkaloids 14a–d
were commercially available (Sigma-Aldrich).
source voltage, 5 kV; capillary voltage, −1 V; and tube lens,
−37.52 V.
Synthesis of racemic α,α′-dihydroxyketones. The correspond-
ing aldehyde 2b–2k (1.00 mmol) was added to a solution of Li-
1 (110 mg, 1.00 mmol) and N-methylmorpholine (110 μL,
1.00 mmol) in water (20 mL) at pH 7 (adjusted with 10% HCl).
The reaction was stirred for 24–48 h at rt and monitored by TLC
analysis. Upon concentration in vacuo, the crude material was
dry loaded and purified using flash silica chromatography.
[1-13C]-Cyclohexyl-1,3-dihydroxy-2-propanone [1-13C]-3a.13
To a stirred solution of [3-13C]-Li-1 (0.110 g, 0.991 mmol) in
water (10 mL) at rt was added cyclohexanecarboxaldehyde
(120 μL, 0.991 mmol) and N-methylmorpholine (110 μL,
1.00 mmol). The reaction mixture was stirred at rt for 48 h, the
solvent was removed in vacuo and the crude product dry loaded
onto a flash silica chromatography column (EtOAc–hexane,
1 : 1) affording [1-13C]-3a (0.026 g, 15%) as a white solid. Rf
0.21 (EtOAc–hexane, 1 : 1); Mp 110–120 °C (EtOAc–hexane,
1 : 1); 1H NMR (600 MHz; CDCl3) δ 1.10–1.31 (5H, m),
1.35–1.85 (6H, m), 4.16 (1H, d, J 2.9 Hz, CHOH), 4.36 (1H,
Synthesis of α,α′-dihydroxyketones using catalyst 14d. The
corresponding aldehyde 2b, 2d–2l (1.00 mmol) was added to a
solution of Li-1 (110 mg, 1.00 mmol) and hydroquinine 4-
methyl-2-quinolyl ether 14d (5–20 mol%) in H2O–THF at pH
7–8 (Table 2). The reaction was stirred for 24–48 h at rt and
monitored by TLC analysis. Upon concentration in vacuo, the
crude material was dry loaded and the product purified using
flash silica chromatography.
1
2
1
dd, JCH 144 and JHH 19.4 Hz, 13CHHOH), 4.48 (1H, dd, JCH
144 and JHH 19.4 Hz, 13CHHOH); 13C NMR (150 MHz;
2
CDCl3) δ 25.8 (signals superimposed), 26.4, 28.9, 29.6, 42.3
1
and 42.7, 66.2 (13CH2OH), 79.3, 211.8 (CvO, JCC 39.0 Hz);
m/z (CI) 174 (MH+, 100%), 156 (54), 138 (60), 95 (25); m/z
(HR-CI) calcd for MH+ C813CH17O3 174.12112, found
174.12031.
Chiral HPLC analysis of 3b, 3d–3l to determine ees. Com-
pounds 3b and 3k were dibenzoylated (following a previously
reported protocol)14 and 3d–3j and 3l monobenzoylated and the
products analysed by chiral HPLC to determine ees, using a
Chiralpak OD column and the hexane–2-propanol solvent
system given. The absolute stereochemistry of 3 (where appropri-
ate) generated using 14d was determined using a Mosher’s deri-
vatisation method (see ESI†).42
[1-13C]-3-Dihydroxy-1-phenyl-2-propanone [1-13C]-3b.14 To a
stirred solution of [3-13C]-Li-1 (0.110 g, 0.991 mmol) in water
(10 mL) at rt was added benzaldehyde (102 μL, 1.00 mmol) and
N-methylmorpholine (110 μL, 1.00 mmol). The reaction mixture
was stirred at rt for 48 h, the solvent was removed in vacuo and
the crude product dry loaded onto a flash silica chromatography
column (EtOAc–hexane, 1 : 1) affording [1-13C]-3b (0.016 g,
10%) as a white solid. Rf 0.30 (EtOAc–hexane, 1 : 1); Mp
110–120 °C (EtOAc–hexane, 1 : 1); 1H NMR (600 MHz;
CDCl3) δ 2.70 (1H, br s, OH), 3.83 (1H, br s, OH), 4.23 (1H,
1,3-Dihydroxy-1-phenyl-2-propanone (3b).14 Racemic 3b was
dibenzoylated and HPLC analysis of the product (82 : 18,
1.0 mL min−1) gave retention times of 10.5 min (3R-isomer) and
13.4 min (3S-isomer). The stereoselective reaction was per-
formed using 5 mol% of 14d in a 2 mL mixture of H2O–THF to
give 3b (7 mg, 4%) in 22% ee (3R-isomer).
1
2
1
dd, JCH 145 and JHH 19.5 Hz, 13CHHOH), 4.34 (1H, dd, JCH
145 and JHH 19.5 Hz, 13CHHOH), 5.25 (1H, s, CHOH),
2
7.32–7.42 (5H, m, Ph); 13C NMR (150 MHz; CDCl3) δ 65.2
(13CH2OH), 77.6, 127.0, 129.4 (signals superimposed), 137.4,
1,3-Dihydroxy-1-(4-methylphenyl)-2-propanone
(3d). The
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3d as
a colourless oil (14 mg, 8%); νmax(neat)/cm−1 3403, 2924, 1727,
208.9 (CvO, JCC 40.5 Hz); m/z (CI) 168 (MH+, 40%), 150
1
(100), 122 (41); m/z (HR-CI) calcd for MH+ C813CH11O3
1
1513; H NMR (600 MHz; CDCl3) δ 2.35 (3H, s, CH3), 4.23
168.07417, found 168.07468.
(1H, d, J 19.5 Hz, CHHOH), 4.34 (1H, d, J 19.5 Hz, CHHOH),
5.21 (1H, s, CHOH), 7.27 (2H, d, J 8.0 Hz, ArH), 7.51 (2H, d, J
8.0 Hz, ArH); 13C NMR (150 MHz; CDCl3) δ 21.3, 65.3, 77.6,
127.0, 130.1, 134.4, 139.3, 209.2; m/z (CI) 181 (MH+, 18%),
163 (100), 145 (25), 135 (36); m/z (HR-CI) calcd for MH+
C10H13O3 181.08592, found 181.08624. Racemic 3d was mono-
benzoylated and HPLC analysis of the product (90 : 10, 1.0 mL
min−1) gave retention times of 15.1 min (3S-isomer) and
16.6 min (3R-isomer). The stereoselective reaction was per-
formed using 5 mol% of 14d in a 2 mL mixture of H2O–THF to
give 3d (4 mg, 2%) in 21% ee (3R-isomer).
MS protocols. The reaction between Li-1 and 2-thiophenecar-
boxaldehyde 2c to give 1,3-dihydroxy-1-(thiophen-2-yl)-2-
propanone has previously been described.14 2-Thiophenecarbox-
aldehyde 2c (75 μL, 0.802 mmol) was added to a solution of Li-1
(110 mg, 1.00 mmol) and DABCO (56 mg, 0.50 mmol) in water
(2 mL) for 2 h. The solution was then directly infused at 3 μL
min−1 for 5 min and was detected by ESI(−)-MS.
Tandem MS/MS. For signal identity confirmation, collision-
induced dissociation MS/MS experiments were performed on
precursor ions selected from MS1 using information-dependent
acquisition: MS1 was performed in the full-scan mode (m/z
100–500). MS2 was performed on the most intense MS1 peak at
m/z 215. The MS1 settings were the same as for MS2 and were
as follows: sheath gas, nitrogen at flow rate of 60 arbitrary units;
capillary temperature, 300 °C; source voltage, 5 kV; capillary
voltage, 15 V; and tube lens, 85 V. MS2 was performed on MS1
peak m/z 327 with the following conditions: sheath gas, nitrogen
at flow rate of 80 arbitrary units; capillary temperature, 300 °C;
1-(4-Fluorophenyl)-1,3-dihydroxy-2-propanone (3e). The re-
action was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3e as
a colourless oil (9 mg, 5%); νmax(neat)/cm−1 3377, 2924, 1725,
1
1603, 1509; H NMR (600 MHz; CDCl3) δ 2.73 (1H, s, OH),
3.85 (1H, s, OH), 4.23 (1H, d, J 19.5 Hz, CHHOH), 4.35 (1H,
d, J 19.5 Hz, CHHOH), 5.24 (1H, s, CHOH), 7.09 (2H,
m, ArH), 7.32 (2H, m, ArH); 13C NMR (150 MHz; CDCl3) δ
2626 | Org. Biomol. Chem., 2012, 10, 2621–2628
This journal is © The Royal Society of Chemistry 2012

View Article Online
2
3
65.2, 77.4, 116.4 (d, JCF 21.3 Hz), 128.9 (d, JCF 8.0 Hz),
d, J 3.9 Hz, CHOH), 7.04–7.10 (2H, m, ArH), 7.14 (1H, d, J 7.5
Hz, ArH), 7.38 (1H, m, ArH); 13C NMR (150 MHz; CDCl3) δ
133.2, 163.2 (d, JCF 247 Hz, CF), 208.9; 19F NMR (282 MHz;
1
CDCl3) δ −112.4; m/z (HR-EI) calcd for M+ C9H9FO3
184.05302, found 184.05270. Racemic 3e was monobenzoylated
and HPLC analysis of the product (90 : 10, 1.0 mL min−1) gave
retention times of 13.0 min (3S-isomer) and 15.3 min (3R-
isomer). The stereoselective reaction was performed using 5 mol
% of 14d in a 2 mL mixture of H2O–THF to give 3e (6 mg, 3%)
in 36% ee (3R-isomer).
65.2, 75.0, 114.0 (d, JCF 22.1 Hz), 116.3 (d, JCF 21.0 Hz),
2
2
3
3
122.7, 131.0 (d, JCF 8.6 Hz), 139.7 (d, JCF 6.8 Hz), 163.3 (d,
1JCF 247 Hz, CF), 208.5; 19F NMR (282 MHz; CDCl3) δ
−112.4; m/z (CI) 185 (MH+, 35%), 167 (100), 139 (95), 125
(29); m/z (HR-CI) calcd for MH+ C9H10FO3 185.06140, found
185.06101. Racemic 3i was monobenzoylated and HPLC analy-
sis of the product (90 : 10, 1.0 mL min−1) gave retention times
of 11.8 min (3S-isomer) and 14.6 min (3R-isomer). The stereo-
selective reaction was performed using 5 mol% of 14d in a 2 mL
mixture of H2O–THF to give 3i (4 mg, 2%) in 38% ee (3R-
isomer).
1-(4-Chlorophenyl)-1,3-dihydroxypropan-2-one
(3f). The
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3f as a
colourless oil (8 mg, 5%); νmax(neat)/cm−1 3344, 2923, 1728,
1
1-(2-Fluorophenyl)-1,3-dihydroxy-2-propanone
(3j). The
1598; H NMR (600 MHz; CDCl3) δ 2.68 (1H, s, OH), 3.84
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3j as a
colourless oil (12 mg, 7%); νmax(neat)/cm−1 3350, 2925, 1725;
1H NMR (600 MHz; CDCl3) δ 2.78 (1H, t, J 4.5 Hz, CH2OH),
3.93 (1H, d, J 4.5 Hz, CHOH), 4.28 (1H, dd, J 19.5 and 4.5 Hz,
CHHOH), 4.35 (1H, d, J 19.5 and 4.5 Hz, CHHOH), 5.52 (1H,
d, J 4.5 Hz, CHOH), 7.12 (1H, m, ArH), 7.19 (1H, m, ArH),
7.35 (2H, m, ArH); 13C NMR (150 MHz; CDCl3) δ 65.2, 71.6,
(1H, s, OH), 4.24 (1H, d, J 19.4 Hz, CHHOH), 4.35 (1H, d, J
19.4 Hz, CHHOH), 5.24 (1H, s, CHOH), 7.29 (2H, d, J 8.5 Hz,
ArH), 7.38 (2H, d, J 8.5 Hz, ArH); 13C NMR (150 MHz;
CDCl3) δ 65.2, 77.3, 128.4, 129.6, 135.3, 135.8, 208.7; m/z
(HR-EI-) calcd for [M − H]− C9H835ClO3 199.0162, found
199.0151. Racemic 3f was monobenzoylated and HPLC analysis
of the product (90 : 10, 1.0 mL min−1) gave retention times of
13.6 min (3S-isomer) and 15.9 min (3R-isomer). The stereo-
selective reaction was performed using 20 mol% of 14d in a
2 mL mixture of H2O–THF to give 3f (4 mg, 2%) in 50% ee
(3R-isomer).
2
2
116.1 (d, JCF 21.2 Hz), 124.7, 125.2 (d, JCF 20.3 Hz), 128.8,
3
1
131.0 (d, JCF 8.4), 160.3 (d, JCF 246 Hz, CF), 208.4; 19F
NMR (282 MHz; CDCl3) δ −112.4; m/z (HR-EI) calcd for M+
C9H9FO3 184.05302, found 184.05270.
1,3-Dihydroxy-1-(4-methoxyphenyl)-2-propanone (3g). The
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3g as
a colourless oil (12 mg, 6%); νmax(neat)/cm−1 3340, 2925, 1720,
1603; 1H NMR (600 MHz; CDCl3) δ 2.70 (1H, t, J 5.4 Hz,
CH2OH), 3.75 (1H, d, J 3.9 Hz, CHOH), 3.81 (3H, s, OCH3),
4.22 (1H, dd, J 19.2 and 5.4 Hz, CHHOH), 4.33 (1H, dd, J 19.2
and 5.4 Hz, CHHOH), 5.19 (1H, d, J 3.9 Hz, CHOH), 6.92 (2H,
d, J 8.7 Hz, ArH), 7.23 (2H, d, J 8.7 Hz, ArH); 13C NMR
(150 MHz; CDCl3) δ 55.5, 65.2, 77.3, 114.8, 128.5, 129.4,
160.4, 209.3; m/z (HR-CI) calcd for MH+ C10H13O4 197.08138,
found 197.08189.
1,3-Dihydroxy-1-(3-methoxyphenyl)-propan-2-one (3k). The
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3k as
a colourless oil (13 mg, 7%); νmax(neat)/cm−1 3325, 2928, 1722;
1H NMR (600 MHz; CDCl3) δ 3.81 (3H, s, OCH3), 4.24 (1H, d,
J 19.2 Hz, CHHOH), 4.35 (1H, d, J 19.2 Hz, CHHOH), 5.22
(1H, s, CHOH), 6.92 (1H, app. t, J 2.1 Hz, ArH), 6.90 (2H, m,
ArH), 7.31 (1H, app. t, J 8.1 Hz, ArH); 13C NMR (150 MHz;
CDCl3) δ 55.5, 65.2, 77.7, 112.4, 114.8, 119.3, 130.5, 138.8,
160.4, 209.0; m/z (CI) 197 (MH+, 14%), 279 (100), 151 (28), 84
(31); m/z (HR-CI) calcd for MH+ C10H13O4 197.08138, found
197.08189. Racemic 3k was dibenzoylated and HPLC analysis
of the product (90 : 10, 1.0 mL min−1) gave retention times of
32.6 min (3S-isomer) and 37.1 min (3R-isomer). The stereo-
selective reaction was performed using 10 mol% of 14d in a
2 mL mixture of H2O–THF to give 3k (6 mg, 3%) in 36% ee
(3R-isomer).
1-(4-Bromophenyl)-1,3-dihydroxy-2-propanone
(3h). The
reaction was carried out for 48 h and the product purified using
flash silica chromatography (EtOAc–hexane, 1 : 1) to give 3h as
a colourless oil (7 mg, 3%); νmax(neat)/cm−1 3390, 2923, 1725,
1
1590; H NMR (600 MHz; CDCl3) δ 2.69 (1H, s, OH), 3.84
(1H, s, OH), 4.24 (1H, d, J 19.5 Hz, CHHOH), 4.35 (1H, d, J
19.5 Hz, CHHOH), 5.22 (1H, s, CHOH), 7.23 (2H, d, J 8.4 Hz,
ArH), 7.54 (2H, d, J 8.4 Hz, ArH); 13C NMR (150 MHz;
CDCl3) δ 65.2, 77.3, 123.5, 128.7, 132.5, 136.4, 208.6; m/z (CI)
247 (M(81Br)+, 3%), 245 (M(79Br)+, 3), 187 (52), 185 (55), 85
(100); m/z (HR-CI) calcd for MH+ C9H10O379Br 244.98133,
found 244.98091.
3-(1,3-Dihydroxy-2-oxo-propyl)-benzoic acid methyl ester
(3l). The reaction was carried out for 48 h and the product
purified using flash silica chromatography (EtOAc–hexane, 1 : 1)
to give 3l as a colourless oil (11 mg, 5%); νmax(neat)/cm−1 3436,
1
2954, 1719, 1605; H NMR (600 MHz; CDCl3) δ 2.69 (1H, t, J
4.8 Hz, CH2OH), 3.89 (1H, d, J 3.9 Hz, CHOH), 3.93 (3H, s,
OCH3), 4.24 (1H, dd, J 19.5 and 4.8 Hz, CHHOH), 4.39 (1H,
dd, J 19.5 and 4.8 Hz, CHHOH), 5.32 (1H, d, J 3.9 Hz,
CHOH), 7.49 (1H, app. t, J 7.5 Hz, ArH), 7.54 (1H, m, ArH),
8.03 (2H, m, ArH); 13C NMR (150 MHz; CDCl3) δ 52.5, 65.3,
76.0, 128.2, 129.5, 130.5, 131.3, 131.4, 137.9, 166.5, 208.6; m/z
(CI) 225 (MH+, 45%), 207 (100), 179 (40), 165 (95); m/z
(HR-CI) calcd for MH+ C11H13O5 225.07630, found 225.07563.
Racemic 3l was monobenzoylated and HPLC analysis of the
1-(3-Fluorophenyl)-1,3-dihydroxy-2-propanone (3i). The reac-
tion was carried out for 48 h and the product purified using flash
silica chromatography (EtOAc–hexane, 1 : 1) to give 3i as a col-
1
ourless oil (9 mg, 5%); νmax(neat)/cm−1 3345, 2920, 1728; H
NMR (600 MHz; CDCl3) δ 2.68 (1H, t, J 4.7 Hz, CH2OH), 3.85
(1H, d, J 3.9 Hz, CHOH), 4.26 (1H, dd, J 19.5 and 4.7 Hz,
CHHOH), 4.38 (1H, dd, J 19.5 and 4.7 Hz, CHHOH), 5.26 (1H,
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2621–2628 | 2627

View Article Online
product (90 : 10, 1.0 mL min−1) gave retention times of
24.5 min (3S-isomer) and 28.1 min (3R-isomer). The stereo-
selective reaction was performed using 10 mol% of 14d in a
2 mL mixture of H2O–THF to give 3l (9 mg, 4%) in 25% ee
(3R-isomer).
17 K. Smithies, M. E. B. Smith, U. Kaulmann, J. L. Galman, J. M. Ward
and H. C. Hailes, Tetrahedron: Asymmetry, 2009, 20, 570–574.
18 M. E. B. Smith, B. H. Chen, E. G. Hibbert, U. Kaulmann, K. Smithies, J.
L. Galman, F. Baganz, P. A. Dalby, H. C. Hailes, G. J. Lye, J. M. Ward, J.
M. Woodley and M. Micheletti, Org. Process Res. Dev., 2010, 14, 99–
107.
19 H. C. Hailes, P. A. Dalby, G. J. Lye, F. Baganz, M. Micheletti, N. Szita
and J. M. Ward, Curr. Org. Chem., 2010, 14, 1883–1893.
20 M. Fetizon, P. Goulaouic and I. Hanna, Tetrahedron Lett., 1985, 26,
4925–4928.
Acknowledgements
21 D. Enders, M. Voith and A. Lenzen, Angew. Chem., Int. Ed., 2005, 44,
1304–1325.
22 M. E. B. Smith, K. Smithies, T. Senussi, P. A. Dalby and H. C. Hailes,
Eur. J. Org. Chem., 2006, 1121–1123.
This work was supported by an Engineering and Physical
Sciences Research Council (EPSRC) PhD Plus award (J.L.G.)
and the UCL Department of Chemistry (D.S.).
23 K. Rohr and R. Mahrwald, Org. Lett., 2011, 13, 1878–1880.
24 U. Hollstein, D. L. Mock, R. R. Sibbitt, U. Roisch and F. Lingens, J.
Labelled Compd. Radiopharm., 1980, 17, 289–296.
25 G. D. Dotson, R. K. Dua, J. C. Clemens, E. W. Wooten and R.
W. Woodard, J. Biol. Chem., 1995, 270, 13698–13705.
26 F. Dickens and D. H. Williamson, Biochem. J., 1958, 68, 74–84.
27 J. B. Fenn, M. Mann, C. K. Meng, S. F. Wong and C. M. Whitehouse,
Science (Washington, D. C.), 1989, 246, 64–71.
Notes and references
1 E. Racker, in The Enzymes, ed. P. D. Boyer, H. Lardy and K. Myrzback,
Academic Press, New York, 1961, vol. 5, pp. 397–412.
2 P. Srere, J. R. Cooper, M. Tabachnick and E. Racker, Arch. Biochem.
Biophys., 1958, 74, 295–305.
28 S. R. Wilson and Y. Wu, Organometallics, 1993, 12, 1478–1480.
29 S. R. Wilson, Q. Lu, M. L. Tulchinsky and Y. Wu, Chem. Commun.,
1993, 664–665.
30 L. S. Santos, C. H. Pavam, W. P. Almeida, F. Coelho and M. N. Eberlin,
Angew. Chem., Int. Ed., 2004, 43, 4330–4333.
31 M. T. Hechavarria Fonseca and B. List, Angew. Chem., Int. Ed., 2004, 43,
3958–3960.
32 W. Schrader, P. P. Handayani, J. Zhou and B. List, Angew. Chem., Int.
Ed., 2009, 48, 1463–1466.
33 U. M. Lindström, Chem. Rev., 2002, 102, 2751–2772.
34 Y. Hayashi, T. Tamura and M. Shoji, Adv. Synth. Catal., 2004, 346,
1106–1110.
35 J. M. Betancort, K. Sakthivel, R. Thayumanavan, F. Tanaka and C.
F. Barbas III, Synthesis, 2004, 1509–1521.
36 N. Mase, Y. Nakai, N. Ohara, H. Yoda, K. Takabe, F. Tanaka and C.
F. Barbas III, J. Am. Chem. Soc., 2006, 128, 734–735.
37 A. Moyano and R. Rios, Chem. Rev., 2011, 111, 4703–4832.
38 M. Bartok, Chem. Rev., 2010, 110, 1663–1705.
39 H. Li, B. Wang and L. Deng, J. Am. Chem. Soc., 2006, 128, 732–
733.
3 J. Bolte, C. Demuynck and H. Samaki, Tetrahedron Lett., 1987, 28,
5525–5528.
4 F. Effenberger, V. Null and T. Ziegler, Tetrahedron Lett., 1992, 33, 5157–
5160.
5 Y. Kobori, D. C. Myles and G. M. Whitesides, J. Org. Chem., 1992, 57,
5899–5907.
6 L. Hecquet, J. Bolte and C. Demuynck, Tetrahedron, 1994, 50, 8677–
8684.
7 R. J. Mitra, J. M. Woodley and M. D. Lilly, Enzyme Microb. Technol.,
1998, 22, 64–70.
8 C. Demuynck, J. Bolte, L. Hecquet and V. Dalmas, Tetrahedron Lett.,
1991, 32, 5085–5088.
9 G. R. Hobbs, M. D. Lilly, N. J. Turner, J. M. Ward, A. J. Willets and J.
M. Woodley, J. Chem. Soc., Perkin Trans. 1, 1993, 165–166.
10 K. G. Morris, M. E. B. Smith, N. J. Turner, M. D. Lilly, R. K. Mitra and
J. M. Woodley, Tetrahedron: Asymmetry, 1996, 7, 2185–2188.
11 E. G. Hibbert, T. Senussi, S. J. Costelloe, M. E. B. Smith, J. M. Ward, H.
C. Hailes and P. A. Dalby, J. Biotechnol., 2008, 134, 240–245.
12 M. E. B. Smith, E. G. Hibbert, A. B. Jones, P. A. Dalby and H. C. Hailes,
Adv. Synth. Catal., 2008, 350, 2631–2638.
40 K. Kacprzak and J. Gawronski, Synthesis, 2001, 961–998.
41 T. Mandal, S. Samanta and C. G. Zhao, Org. Lett., 2007, 9, 943–
945.
42 J. L. Galman and H. C. Hailes, Tetrahedron: Asymmetry, 2009, 20, 1828–
1831.
43 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and F.
J. Timmers, Organometallics, 1996, 15, 1518–1520.
44 T. Mukaiyama, N. Iwasawa, R. W. Stevens and T. Haga, Tetrahedron,
1984, 40, 1381–1390.
13 A. Cázares, J. L. Galman, L. G. Crago, M. E. B. Smith, J. Strafford,
L. Ríos-Solís, G. J. Lye, P. A. Dalby and H. C. Hailes, Org. Biomol.
Chem., 2010, 8, 1301–1309.
14 J. L. Galman, D. Steadman, S. Bacon, P. Morris, M. E. B. Smith, J.
M. Ward, P. A. Dalby and H. C. Hailes, Chem. Commun., 2010, 46,
7608–7610.
15 D. C. Myles, P. J. Andrulis III and G. M. Whitesides, Tetrahedron Lett.,
1991, 32, 4835–4838.
16 C. U. Ingram, M. Bommer, M. E. B. Smith, P. A. Dalby, J. M. Ward, H.
C. Hailes and G. J. Lye, Biotechnol. Bioeng., 2007, 96, 559–569.
45 S. K. Hendrie and J. Leonard, Tetrahedron, 1987, 43, 3289–3294.
2628 | Org. Biomol. Chem., 2012, 10, 2621–2628
This journal is © The Royal Society of Chemistry 2012