α-Diketones as acyl anion equivalents: A non-enzymatic thiamine-promoted route to aldehyde-ketone coupling in PEG400 as recyclable medium

Article abstract of DOI:10.1016/j.tet.2011.08.056

By mimicking the peculiar behavior of thiamine diphosphate-dependent acetylacetoin synthase, it has been demonstrated that thiamine hydrochloride 2a and its simple analogue thiazolium salt 2b are able to activate α-diketones as acyl anion equivalents in nucleophilic acylations, such as the homo-coupling of α-diketones and the hitherto unreported cross-coupling between α-diketones and α-ketoesters. These carboligation reactions were optimized under stoichiometric (2a) and catalytic conditions (2b) by using eco-friendly PEG₄₀₀ as the reaction medium, thus allowing both solvent and thiazolium salt recycling.

Full text of DOI:10.1016/j.tet.2011.08.056

Tetrahedron 67 (2011) 8110e8115
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
a-Diketones as acyl anion equivalents: a non-enzymatic thiamine-promoted route
to aldehydeeketone coupling in PEG400 as recyclable medium
,
a
b
b
Olga Bortolini a, Giancarlo Fantin a , Marco Fogagnolo , Pier Paolo Giovannini , Valentina Venturi ,
*
Salvatore Pacifico a, Alessandro Massi a
Dipartimento di Chimica, Laboratorio di Chimica Organica, Universita di Ferrara, Via L. Borsari 46, I-44121 Ferrara, Italy
,
*
a
ꢀ
b
ꢀ
Dipartimento di Biologia ed Evoluzione, Universita di Ferrara, Corso Ercole I D’Este, 32, I-44121 Ferrara, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 May 2011
Received in revised form 2 August 2011
Accepted 22 August 2011
Available online 27 August 2011
By mimicking the peculiar behavior of thiamine diphosphate-dependent acetylacetoin synthase, it has
been demonstrated that thiamine hydrochloride 2a and its simple analogue thiazolium salt 2b are able to
activate
a-diketones as acyl anion equivalents in nucleophilic acylations, such as the homo-coupling of
a-
diketones and the hitherto unreported cross-coupling between
a
-diketones and -ketoesters. These
a
carboligation reactions were optimized under stoichiometric (2a) and catalytic conditions (2b) by using
eco-friendly PEG400 as the reaction medium, thus allowing both solvent and thiazolium salt recycling.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
a
a
-Diketones
-Hydroxyketones
Polyethylene glycol
Thiazolium salts
N-Heterocyclic carbenes
1. Introduction
already disclosed enzymatic homo-coupling of 1,2-diketones3
(virtually an intermolecular aldehydeeketone coupling) was ini-
Thiamine diphosphate (ThDP)-dependent enzymes, such as
pyruvate decarboxylase (PDC), benzaldehyde lyase (BAL), and
benzoylformate decarboxylase (BDF), are a group of biocatalysts
involved in a variety of reactions including the formation and
cleavage of carbonecarbon bonds.1 These enzymes have in com-
mon the capability to generate an ‘active aldehyde’ species,2 that is,
a ThDP-bound carbanioneenamine intermediate, which reacts
tially considered. Subsequently, the more challenging thiamine-
catalyzed direct cross-coupling reaction between diketone donors
and
a-ketoester acceptors was also investigated to explore the
potential of the proposed carboligation methodology (Fig. 1). In-
deed, while significant advances have been recently made in N-
heterocyclic carbene (NHC) catalysis7 to promote intramolecular
aldehydeeketone couplings,8 examples of the non-enzymatic in-
termolecular variant are limited to the study by Enders and Hens-
eler,9 which described the cross-coupling between aldehydes and
with an aldehyde acceptor to form a
a-hydroxyketone in an
acyloin-type condensation.2 Within this area of research, a recent
study from our group demonstrated that 1,2-diketones may serve
as acyl anion equivalents when ThDP-dependent acetylacetoin
synthase (AAS) is utilized as catalyst in the homo-coupling of 1,2-
diketones to form chiral
triggered us to investigate a possible biomimetic route, reminiscent
of ThDP activity,5 to nucleophilic acylations involving
-diketones
a
-hydroxyketones (Fig. 1).3,4 This finding
a
as acyl anion donors. By considering the peculiar mechanism of
action of AAS,3 it was envisaged that the sole thiamine coenzyme
(pre-catalyst) in the presence of a suitable base could display the
same ability to activate a
-diketones as acyl anion sources.6 Hence,
the optimization of an organocatalytic (racemic) version of the
* Corresponding authors. E-mail addresses: giancarlo.fantin@unife.it (G. Fantin),
Fig. 1. Synthesis of
a-hydroxy-1,3-diketones and a-hydroxy-1,3-ketoesters using
msslsn@unife.it (A. Massi).
a-diketones as acyl anion equivalents.
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.08.056

O. Bortolini et al. / Tetrahedron 67 (2011) 8110e8115
8111
highly electrophilic trifluoromethyl ketones using a bicyclic tri-
catalytic conditions (2a, 20 mol %), although a modest conversion
(45%) was achieved within a reasonable reaction time (48 h, entry
5). Fortunately, the use of the commercially available thiazolium
salt 2b, that is, a simplified analogue of thiamine 2a, greatly in-
creased the reaction rate and 3a was recovered in almost quanti-
tative yield after 12 h stirring and standard work-up (entry 6). The
recyclability of 2b in PEG400 was finally demonstrated as before
(entry 7).
We then examined the tolerance of the procedure for different
donoreacceptor substrates with stoichiometric thiamine 2a
(Method A, Table 2) and found that 3,4-hexanedione 1b was suc-
cessfully converted to the expected product 3b (88%, entry 1).
When the protocol was extended to the unsymmetrically
azolium salt as the pre-catalyst.10
2. Results and discussion
In analogy with our previous enzymatic study,3 we initially in-
vestigated the activity of the thiamine hydrochloride 2a-Et3N
couple in water using 2,3-butanedione (biacetyl) 1a as
donoreacceptor substrate (Table 1, entry 1). Triethylamine was
chosen as the base to reproduce the basic strength of active sites in
the catalytic pocket of ThDP-dependent enzymes.11 Against the
almost complete consumption of 1a as determined by TLC and 1H
NMR analyses, only small amounts of the expected product 3a
survived to these conditions, in agreement with an old report of
Mizuhara and Handler6 asserting that biacetyl 1a, in aqueous al-
kaline medium and in the presence of thiamine, is cleaved to ac-
etaldehyde and 2-(1-hydroxyethyl)-thiamine. On the other hand,
substituted dialkyl a-diketones 1cee (entries 2e4), it was observed
that acetyl anion transfer (MeCOꢀ) was predominant over migra-
tion of its higher carbanion counterparts, in full agreement with the
the
a-hydroxy-1,3-diketone 3a product has been reported to be
Table 2
unstable under basic and thermal conditions due to a complex
sequence of equilibra involving the corresponding oxy-anions.12
These multiple rearrangements are also active in the presence of
mild bases as sodium bicarbonate. Our findings, however, suggest
that the instability of 3-like products is related not only to the
presence of a base but, and more importantly, to the concomitant
use of an aqueous medium, as demonstrated by the results found
for the condensation of 1a promoted by the thiamine hydrochloride
2a-Et3N couple in EtOH (entry 2). In this case, the expected product
3a was in fact obtained in 75% isolated yield. It is worth stressing
that crucial for a successful work-up was the purification of 3a by
distillation. Encouraged by these promising results, we decided to
widen the selection of protic solvents with polyethylene glycol
(PEG400), a non-toxic and eco-friendly medium that should allow
for both solvent and pre-catalyst recycling.13 Hence, when the
mixture of 1a (0.5 M), thiamine hydrochloride 2a (1 equiv), and
Short study on the applicability of the optimized stoichiometric and catalytic homo-
coupling procedures to selected a
-diketonesa
Entry
1
a
-Diketone
a
-Hydroxy-1,3-diketone
(Method A (%); Method B (%))b
Et3N (2 equiv) in PEG400 was stirred for 3 h, the target a-hydrox-
yketone 3a was recovered in almost quantitative yield (95%) after
extraction with Et2O and subsequent product distillation (entry 3).
Rewardingly, the PEG solution containing the thiamine promoter
could be reloaded with fresh diketone 1a and Et3N to afford 3a in
90% isolated yield (entry 4). Additional recycles (up to five) resulted
equally effective with minimal loss of product yield. No change in
the reaction outcome occurred on moving from stoichiometric to
2
3
Table 1
Optimization of the homo-coupling of 2,3-butanedione 1aa
4
5
Entry
Catalyst (mol %)
Solvent
Time (h)
3a Yield (%)b
1c
2a (100)
2a (100)
2a (100)
2a (100)
2a (20)
H2O
EtOH
5
5
3
<5
75
95
90
45
95
91
2c
3c
PEG400
PEG400
PEG400
PEG400
PEG400
a
Reactions performed with 2.0 mmol of diketone (0.5 M).
Isolated yield.
4c,d
5e
3
b
48
12
12
cLower amounts (<10%) of coupling products arising from propionyl anion transfer
were also detected (GCeMS and 1H NMR analyses).
dYield determined by 1H NMR analysis of the crude reaction mixture using bro-
moform as internal standard.
6e
2b (20)
2b (20)
7d,e
a
Reactions performed with 2.0 mmol of 1a (0.5 M).
Isolated yield.
Et3N: 200 mol %.
Reaction performed with recycled 2 and PEG400 after addition of fresh Et3N.
Et3N:100 mol %.
b
c
eLower amounts (<10%) of coupling products arising from butyryl anion transfer
were also detected.
d
e
fLower amounts (<10%) of coupling products arising from pentanoyl anion transfer
were also detected.

8112
O. Bortolini et al. / Tetrahedron 67 (2011) 8110e8115
reactivity pertaining to the AAS enzymatic system.3 Accordingly,
the corresponding -hydroxyketones 3cee (65e71%) were isolated
a
together with lower amounts (11e24%) of the isomeric derivatives
4cee (entries 2e4). As previously observed, the results obtained in
the above transformations with catalytic 2b (Method B) closely
paralleled those detected with stoichiometric thiamine 2a although
with a longer reaction time (12e24 h).
The mechanism that accounts for all these findings may invoke
the formation of the intermediate II resulting from the addition of
the thiazolin-2-ylidene I to the a-diketone 1, and its evolution to
the Breslow intermediate III by attack of the PEG solvent to the
carbonyl of II. Subsequent addition of the Breslow intermediate to
the acceptor, that is, a second molecule of a-diketone 1, leads to the
formation of the product 3 and regeneration of the catalyst I
(Scheme 1).14 It is worth noting that formation of the key in-
termediate III by the above postulated mechanism seems to be
confirmed by the isolation of 1 equiv of acetylated PEG solvent
(PEG-OAc) from the crude reaction mixture.15
Scheme 2. Proposed reaction pathway for the reduction of 1f.
Table 3
Homo-coupling of ethyl pyruvate 6 and cross-couplings between 1a and selected
-ketoesters 6e9a
a
Entry
1
Donor
Acceptor
a-Hydroxy-1,3-ketoesters
(Method A(%); Method B (%))b
6
No reaction
Scheme 1. Proposed reaction pathway for the homo-coupling of alkyl diketones 1aee
(substrate 1a as representative example).
2
3
1a
The overall reactivity of the thiazolium-promoted homo-cou-
pling of a-diketones 1 can be fully explained by Scheme 1, except
when the formation of a phenyl-substituted Breslow intermediate
of type V occurs (Scheme 2). This is demonstrated by the results
observed in the homo-coupling of 1-phenyl-1,2-propanedione 1f
(Table 2, entry 5). In that case, the fate of intermediate V seems to
be different with generation, through prototropic rearrangement to
VI,16 of a hydride equivalent, that is, transferred to the acceptor 1f.
1a
1a
Once the a
-diketone undergoes reduction to 5f,17 the thiazolin-2-
ylidene I is regenerated by transfer of the benzoyl group from the
4
acyl thiazolium intermediate VII to PEG (Scheme 2).15,18
In a next set of experiments we investigated the donoreacceptor
attitudes of
a-ketoesters toward the thiazolium-Et3N system. In
a control experiment (Table 3, entry 1), ethyl pyruvate 6 was not
reactive under the previously optimized homo-coupling conditions
(Methods A and B). This finding, however, may be exploited to per-
5
1a
form cross-coupling reactions between
a suitable acyl anion donor, such as 1a. To explore this hitherto un-
reported carboligation reaction, the commercially available
a-ketoester acceptors and
a
-
a
ketoester esters 6e9 (3 equiv) were coupled with 1a under optimal
stoichiometric and catalytic conditions (entries 2e5). Gratifyingly,
All reactions performed with 3 equiv of
Isolated yield.
a-ketoester acceptor.
b

O. Bortolini et al. / Tetrahedron 67 (2011) 8110e8115
8113
the corresponding
a
-hydroxy-1,3-ketoesters 10e13 were obtained
stirred for 5 min, allowed to separate out and the ethereal solution
was decanted. This process was repeated twice to obtain the crude
in fair yields (around 50%), thus demonstrating the feasibility of this
approach. While the side homo-coupling reaction of 1a could not be
suppressed under these conditions, the isolation of the target
products 10e13 was facilitated by the volatile nature of the homo-
coupling by-product 3a.
a-hydroxyketone 3 in Et2O, whereas the mother liquor (PEG400-
thiamine hydrochloride 2a) was kept aside for further runs. The
extraction solvent was then removed under a nitrogen stream and
the residue containing the target a-hydroxyketone purified by ei-
ther bulb-to-bulb distillation (compounds 3a and 3b) or flash
chromatography (compounds 3cee). Product yields are reported in
the next paragraphs for Method A.
3. Conclusion
In summary, wehave demonstrated that 1,2-diketones mayserve
Method B. To a vigorously stirred mixture of 3-benzyl-5-(2-
as acyl anion equivalents in thiamine-promoted
a-diketone homo-
hydroxyethyl)-4-methylthiazolium
chloride
2b
(54
mg,
couplings in analogy with a previously reported ThDP-dependent
enzymatic system. In addition, a novel substrate combination, that
0.20 mmol), Et3N (140 L, 1.00 mmol), and PEG400 (4 mL)
tone 1 (2.00 mmol) was added in one portion. The mixture was
m
a
-dike-
is,
a
-diketone to
a
-ketoester, has been investigated by using the
stirred at room temperature until TLC analysis revealed the disap-
same thiamine promoter. The setup of a catalytic procedure for the
above transformations has also been optimized by employing
a more active analogue of thiamine as the pre-catalyst. Polyethylene
glycol (PEG400) has been shown to be an effective and reusable re-
action medium for the above transformations. Further in-
vestigations of the generality of the disclosed mode of acyl anion
generation and the development of NHC-promoted asymmetric
variants are currently underway in our laboratories.
pearance of the starting a-diketone (12e24 h). The reaction me-
dium was then diluted with Et2O (5 mL), vigorously stirred for
5 min, allowed to separate out and the ethereal solution was dec-
anted. This process was repeated twice to obtain the crude
hydroxyketone 3 in Et2O, whereas the mother liquor (PEG400
a
-
-
methylthiazolium 2b) was kept aside for further runs. The extrac-
tion solvent was then removed under a nitrogen stream and the
residue containing the target
a-hydroxyketone purified as de-
scribed above. Product yields for Method B are reported in Table 2.
4. Experimental section
4.1. General remarks
4.2.1. 3-Hydroxy-3-methylpentane-2,4-dione (3a). The crude re-
action mixture was bulb-to-bulb distilled (50 ꢂC, 5 mmHg) to give
3a3,19 (124 mg, 95%) as a colorless liquid. Lit.:19b bp 43e44 ꢂC
(4 mmHg). ESI MS (130.1): 153.5 (MþNaþ). Compound 3a partially
decomposes on silica gel.
Reactions were monitored by TLC on silica gel 60 F254 with de-
tection by charring with phosphomolybdic acid. Flash column
chromatography was performed on silica gel 60 (230e400 mesh).
€
Bulb-to-bulb distillation was performed with a Buchi Glass Oven B-
4.2.2. 4-Ethyl-4-hydroxyheptane-3,5-dione (3b). The crude reaction
mixture was bulb-to-bulb distilled (84 ꢂC, 5 mmHg) to give 3b3,20
580 apparatus. 1H (300 MHz) and 13C (75 MHz) NMR spectra were
recorded in CDCl3 solutions at room temperature. Peaks assignments
were aided by 1He1H COSY and gradient-HMQC experiments.
GCeMS spectra were recorded using a Varian 4000 GC/MS/MS sys-
(151 mg, 88%) as
a colorless liquid. ESI MS (172.1): 195.8
(MþNaþ). Compound 3b partially decomposes on silica gel.
tem equipped with
(25mꢁ0.25 mm) containing dimethyl-n-pentyl-
a
fused capillary column Megadex
5
4.2.3. (R/S)-3-Hydroxy-3-methylhexane-2,4-dione
(3c). Column
b-cyclodextrin on
chromatography with 12:1 cyclohexane/AcOEt afforded 3c3,21
(102 mg, 71%) as a colorless oil. ESI MS (144.1): 145.1 (MþHþ). A
chromatographic fraction containing 4c but slightly contaminated
by 3c was collected for 4c3 identification (1H NMR analysis).
Compound 4c partially decomposes on silica gel.
OV 1701. Analyses were carried out with a gradient of 1.5 ꢂC/min
(from 80 ꢂC up to 200 ꢂC); retention times (tR) are given in minutes.
ESI MS analyses were performed in positive ion mode with samples
dissolved in 10 mM solution of ammonium formate in 1:1 MeCN/
H2O. For accurate mass measurements the compounds were ana-
lyzed in positive ion mode by electrospray ionization (ESI) hybrid
quadrupole orthogonal acceleration time-of-flight mass spectrom-
eter(Q-TOFMS)fittedwithaZ-sprayelectrosprayionsource(Waters,
Manchester, UK). The capillary source voltage and the cone voltage
were set at 3200 V and 45 V, respectively; the source temperature
was kept at 120 ꢂC; nitrogenwas used as a drying gas at a flow rate of
ca. 80 L/h. The time-of-flight analyzer was externally calibrated with
NaI from m/z 300 to 2000 to yield accuracy near to 5 ppm. Accurate
4.2.4. (R/S)-3-Hydroxy-3-methylheptane-2,4-dione
(3d). Column
chromatography with 12:1 cyclohexane/AcOEt afforded 3d3,21
(102 mg, 65%) as a colorless oil. ESI MS (158.1): 181.7 (MþNaþ). A
chromatographic fraction containing 4d was collected for 4d3
identification (1H NMR analysis). Compound 4d partially de-
composes on silica gel.
4.2.5. (R/S)-3-Hydroxy-3-methyloctane-2,4-dione
(3e). Column
mass data were collected by directly infusing samples (10 pmol/
m
L in
chromatography with 12:1 cyclohexane/AcOEt afforded 3e21
(115 mg, 67%) as a colorless oil. GCeMS (70 eV, EI): tR 23.65 (first
enantiomer); 23.84 (second enantiomer), m/z 172 (Mþ,<1%), 130
(20), 88 (100), 43 (45). ESI MS (172.1): 195.5 (MþNaþ). A chro-
matographic fraction containing 4e was collected for 4e identifi-
cation. 3-Butyl-3-hydroxypentane-2,4-dione 4e: GCeMS (70 eV,
1:1 MeCN/H2Oþ0.1% ammonium formate) into the system at a flow
rate of 5 m
L/min. Spectroscopic data of compounds 3a,3,19 3b,3,20
3c,d,3,21 3e,21 4c,3 4d,3 5f,22 10,23 and 1324 were identical to those
reported in the literature. Copies of the 1H spectra of 3aee, 4d, 5f,10,
and 13 are reported as purity and identity documentation.
EI): tR 20.90, m/z 172 (Mþ,<1%), 130 (100), 43 (68). 1H NMR:
d¼4.65
4.2. Optimized procedure for the homo-coupling of
a-
(br s, 1H, OH), 2.25 (s, 6H, CH3(CO)), 2.04e1.95 (m, 2H, 2 H-40),
1.40e1.19 (m, 4H, 2 H-20, 2 H-30), 0.90 (t, 3H, J¼7.0 Hz, CH3). 13C
diketones 1aee
NMR:
d
¼207.6 (2C), 91.0 (C), 36.1 (CH2), 25.3 (CH2), 25.2 (2 CH3),
Method A. To a vigorously stirred mixture of thiamine hydro-
chloride 2a (337 mg, 1.00 mmol), Et3N (279 L, 2.00 mmol), and
PEG400 (4 mL) -diketone 1 (2.00 mmol) was added in one portion.
The mixture was stirred at room temperature until TLC analysis
revealed the disappearance of the starting -diketone (3e5 h). The
reaction medium was then diluted with Et2O (5 mL), vigorously
22.7 (CH2), 13.8 (CH3). ESI MS (172.1): 173.4 (MþHþ). HRMS (ESI/Q-
TOF): calcd m/z for C9H17O3 [MþH]þ, 173.1178; found, 173.1170.
Compound 4e partially decomposes on silica gel.
m
a
a
4.2.6. (R/S)-1-Hydroxy-1-phenylpropan-2-one (5f). Method A. To
a vigorously stirred mixture of thiamine hydrochloride 2 (337 mg,

8114
O. Bortolini et al. / Tetrahedron 67 (2011) 8110e8115
1.00 mmol), Et3N (279
m
L, 2.00 mmol), and PEG400 (4 mL)
a
-dike-
(s, 3H, CH3). 13C NMR:
d
¼203.6 (C), 170.8 (C), 130.1 (C), 128.8 (2 CH),
tone 1f (293
mL, 2.00 mmol) was added in one portion. The mixture
128.6 (2 CH), 126.3 (CH), 84.6 (C), 53.6 (CH3), 26.9 (CH3). ESI MS
(208.1): 209.7 (MþHþ). HRMS (ESI/Q-TOF): calcd m/z for C11H13O4
[MþH]þ, 209.0814; found, 209.0821.
was stirred at room temperature for 3 h and then diluted with Et2O
(5 mL). The resulting mixture was vigorously stirred for 5 min,
allowed to separate out and the ethereal solution was decanted.
This process was repeated twice. The collected ethereal fractions
were concentrated and the resulting residue was eluted from
a column of silica gel with 4:1 cyclohexane/AcOEt to give 5f22
(234 mg, 78%) as a white amorphous solid. ESI MS (150.1): 173.7
(MþNaþ).
4.3.3. (R/S)-Ethyl 2-acetyl-2-hydroxy-4-phenylbutanoate
(12). Column chromatography with 8:1 cyclohexane/AcOEt affor-
ded 12 (125 mg, 50%) as a yellow foam. 1H NMR:
d¼7.38e7.15 (m,
5H, Ph), 4.28 (br s, 1H, OH), 4.25 (q, 2H, J¼7.0 Hz, OCH2CH3),
2.70e2.60 (m, 2H, 2 H-4), 2.45 (ddd, 1H, J3a,4a¼8.0 Hz, J3a,4b¼8.5 Hz,
J3a,3b¼14.0 Hz, H-3a), 2.30 (s, 3H, CH3), 2.24 (ddd, 1H, J3b,4a¼7.5 Hz,
J3b,4b¼8.0 Hz, J3a,3b¼14.0 Hz, H-3b), 1.27 (t, 3H, J¼7.0 Hz, OCH2CH3).
The subsequent elution with AcOEt afforded a mixture of PEG-
OBz and PEG-OAc. PEG-OBz: 1H NMR:
d
¼8.20e8.05, 7.60e7.50,
and 7.48e7.40 (3m, Ph), 4.50e4.40 and 3.90e3.80 (2m, OCH2
13C NMR:
d
¼204.9 (C), 171.0 (C), 141.1 (C), 128.7 (4 CH), 126.4 (CH),
-
CH2OBz), 3.70e3.50 (m, OCH2CH2Oe). PEG-OAc: 1H NMR:
84.1 (C), 63.0 (CH2), 37.2 (CH2), 29.8 (CH2), 24.8 (CH3),14.3 (CH3). ESI
MS (250.1): 251.6 (MþHþ). HRMS (ESI/Q-TOF): calcd m/z for
C14H18O4 [MþH]þ, 251.1283; found, 251.1277.
d
¼4.30e4.20 and 3.60e3.50 (2m, OCH2CH2OAc), 3.70e3.50 (m,
OCH2CH2O-), 2.08 (s, CH3).
Method B. To a vigorously stirred mixture of 3-benzyl-5-(2-
hydroxyethyl)-4-methylthiazolium
0.20 mmol), Et3N (279 L, 2.00 mmol), and PEG400 (4 mL)
tone 1f (293
was stirred at room temperature for 12 h and then diluted with
Et2O (5 mL). The resulting mixture was vigorously stirred for 5 min,
allowed to separate out and the ethereal solution was decanted.
This process was repeated twice. The collected ethereal fractions
were concentrated and the resulting residue was eluted from
a column of silica gel with 4:1 cyclohexane/AcOEt to give 5f22
(225 mg, 75%) as a white amorphous solid.
chloride
2b
(54
mg,
-dike-
4.3.4. (R/S)-Ethyl 2-acetyl-2-hydroxy-3-methylbutanoate
(13). Column chromatography with 8:1 cyclohexane/AcOEt affor-
ded 1324 (96 mg, 51%) as a yellow foam. ESI MS (188.1): 189.3
(MþHþ).
m
a
m
L, 2.00 mmol) was added in one portion. The mixture
Acknowledgements
We gratefully acknowledge the University of Ferrara (Progetto
FAR 2010) and the Italian Ministry of University and Scientific Re-
search (Progetto FIRB Chem-Profarma-Net Grant RBPR05NWWC
008) for financial supports. Thanks are also given to Mr. Paolo
Formaglio for NMR experiments.
4.3. General procedure for the cross-couplings of 1a with a-
ketoesters 6e9
Supplementary data
Method A. To a vigorously stirred mixture of thiamine hydro-
chloride 2 (337 mg, 1.00 mmol), Et3N (279 L, 2.00 mmol),
ketoester 6e9 (3.00 mmol), and PEG400 (4 mL) 2,3-butanedione 1a
(84 L, 1.00 mmol) was added in one portion. The mixture was
stirred at room temperature until TLC analysis revealed the disap-
pearance of 1a (3e8 h). The reaction medium was then diluted with
Et2O (5 mL), vigorously stirred for 5 min, allowed to separate out
and the ethereal solution was decanted. This process was repeated
twice. The collected ethereal fractions were concentrated and the
resulting residue was eluted from a column of silica gel with the
m
a-
1H and 13C spectra for new compounds. Supplementary data
associated with this article can be found in the online version, at
doi:10.1016/j.tet.2011.08.056. These data include MOL files and
InChIKeys of the most important compounds described in this
article.
m
References and notes
1. (a) Sprenger, G. A.; Pohl, M. J. Mol. Catal. B: Enzym. 1999, 6, 145e159; (b) Pohl,
suitable elution system to give the corresponding a-hydroxy-1,3-
ketoester 10e13. Product yields are reported in the next para-
graphs for Method A.
€
M.; Lingen, B.; Muller, M. Chem.dEur. J. 2002, 8, 5288e5295.
2. Enders, D.; Balensiefer, T. Acc. Chem. Res. 2004, 37, 534e541.
3. Giovannini, P. P.; Pedrini, P.; Venturi, V.; Fantin, G.; Medici, A. J. Mol. Catal. B:
Enzym. 2010, 64, 113e117.
4. In a parallel investigation, Muller and co-workers demonstrated that
Method B. To a vigorously stirred mixture of 3-benzyl-5-(2-
€
a
-hydroxy
hydroxyethyl)-4-methylthiazolium
0.20 mmol), -ketoester 6e9 (3.00 mmol), and PEG400 (4 mL) 2,3-
butanedione 1a (84 L, 1.00 mmol) was added in one portion. The
chloride
2b
(54
mg,
methyl ketones can be obtained by using a different ThDP-dependent enzyme
€
catalyst and pyruvate as acetyl anion equivalent. Lehwald, P.; Richter, M.; Rohr,
a
€
C.; Liu, H.-W.; Muller, M. Angew. Chem., Int. Ed. 2010, 49, 2389e2392.
m
5. Kluger, R.; Tittmann, K. Chem. Rev. 2008, 108, 1797e1833.
mixture was stirred at room temperature until TLC analysis
revealed the disappearance of 1a (12e48 h). The reaction medium
was then diluted with Et2O (5 mL), vigorously stirred for 5 min,
allowed to separate out and the ethereal solution was decanted.
This process was repeated twice. The collected ethereal fractions
were concentrated and the resulting residue was eluted from
a column of silica gel with the suitable elution system to give the
6. Pioneering observations by Mizuhara and Handler established that the
thiamine-promoted reaction of biacetyl and acetaldehyde involved scission of
the biacetyl into two moieties to form acetoin and acetate. Mizuhara, S.;
Handler, P. J. Am. Chem. Soc. 1954, 76, 571e573.
7. For recent reviews, see: (a) Moore, J. L.; Rovis, T. Top. Curr. Chem. 2010, 291,
77e144; (b) Enders, D. J. Org. Chem. 2008, 73, 7857e7870; (c) Rovis, T. Chem.
Lett. 2008, 37, 2e7; (d) Enders, D.; Niemeier, O.; Henseler, A. Chem. Rev. 2007,
107, 5606e5655.
8. For representative examples, see: (a) Enders, D.; Niemeier, O.; Balensiefer, T.
Angew. Chem., Int. Ed. 2006, 45, 1463e1467; (b) Takikawa, H.; Hachisu, Y.; Bode,
J. W.; Suzuki, K. Angew. Chem., Int. Ed. 2006, 45, 3492e3494; (c) Enders, D.;
Niemeier, O.; Raabe, G. Synlett 2006, 2431e2434; (d) Li, Y.; Feng, Z.; You, S.-L.
Chem. Commun. 2008, 2263e2265.
corresponding a-hydroxy-1,3-ketoester 10e13. Product yields are
reported in Table 3.
4.3.1. (R/S)-Ethyl 2-hydroxy-2-methyl-3-oxobutanoate (10). Column
chromatography with 3:1 cyclohexane/AcOEt afforded 1023 (88 mg,
55%) as a yellow oil. ESI MS (160.1): 183.5 (MþNaþ).
9. Enders, D.; Henseler, A. Adv. Synth. Catal. 2009, 351, 1749e1752.
10. For a recent study on highly chemoselective crossed condensations between
aliphatic and ortho-substituted aromatic aldehydes, see: O’Toole, S. E.; Rose, C.
A.; Gundala, S.; Zeitler, K.; Connon, S. J. J. Org. Chem. 2011, 76, 347e357.
11. (a) Sheehan, J. C.; Hunneman, D. H. J. Am. Chem. Soc. 1966, 88, 3666e3667; (b)
For a recent review, see: Nemeria, N. S.; Chakraborty, S.; Balakrishnan, A.; Jor-
dan, F. FEBS J. 2009, 276, 2432e2446.
12. (a) Rubin, M. B.; Inbar, S. J. Org. Chem. 1988, 53, 3355e3558; (b) Rubin, M. B.;
Inbar, S. Tetrahedron Lett. 1979, 20, 5021e5024.
13. Chen, J.; Spear, S. K.; Huddleston, J. G.; Rogers, R. D. Green Chem. 2005, 7, 64e82.
4.3.2. (R/S)-Methyl 2-hydroxy-3-oxo-2-phenylbutanoate
(11). Column chromatography with 8:1 cyclohexane/AcOEt affor-
ded 11 (87 mg, 42%) as a yellow foam. 1H NMR:
7.48e7.35 (2m, 5H, Ph), 4.78 (br s, 1H, OH), 3.88 (s, 3H, OCH3), 2.26
d
¼7.60e7.50 and

O. Bortolini et al. / Tetrahedron 67 (2011) 8110e8115
8115
14. Formation of the Breslow intermediate III may also be explained by release
from II of a putative ketene, which in turn is entrapped by the PEG solvent. A
similar mechanism was proposed by Mizuhara and Handler for the thiamine-
promoted reaction of biacetyl 1a with acetaldehyde, though no evidence of
ketene expulsion was provided (see Ref. 6).
15. The extractability with Et2O of acylated PEG byproducts (PEG-OAc, Schemes 1
and 2; PEG-OBz, Scheme 2) results in a partial loss of PEG solvent after each
run, thus limiting its reuse in repeated experiments.
(b) Phillips, E. P.; Chan, A.; Scheidt, K. A. Aldrichimica Acta 2009, 42, 55e66;
Other possible mechanisms of forming 5f are: (i) release of benzaldehyde
from intermediate VI followed by attack of this by an acetyl anion equivalent,
or (ii) formation of a product of type 3 or 4 followed by a retro-aldol reaction
to lose the extra acyl group. In order to gain a more detailed view of the re-
action mechanism, devoted studies are currently underway in our
laboratories.
€
19. (a) Christoffers, J.; Kauf, T.; Werner, T.; Rossle, M. Eur. J. Org. Chem. 2006,
16. For a detailed study on the 1,2-proton shift involved in this type of intermediate
species, see: Gronert, S. Org. Lett. 2007, 9, 3065e3068.
17. For a previous study on the preferential reduction of the benzoyl group of 1f,
see: Toukoniitty, E.; Franceschini, S.; Vaccari, A.; Murzin, D.-Y. Appl. Catal., A
2006, 300, 147e154 and references therein.
18. The mechanism of Scheme 2 is reminiscent of the reaction pathway proposed
by Scheidt and co-workers for NHC-catalyzed hydroacylation of activated
ketones: (a) Chan, A.; Scheidt, K. A. J. Am. Chem. Soc. 2006, 128, 4558e4559;
2601e2608; (b) Ramirez, F.; Bhatia, S. B.; Bigler, A. J.; Smith, C. P. J. Org. Chem.
1968, 33, 1192e1196.
20. Clausen, C.; Weidner, I.; Butenschon, H. Eur. J. Org. Chem. 2000, 3799e3806.
21. Naef, R.; Jaquier, A. PCT Int. Appl. 2006, WO 2006092749 A1 20060908.
22. Zhang, W.; Shi, M. Chem. Commun. 2006, 1218e1220.
23. Baucherel, X.; Levoirier, E.; Uziel, J.; Juge, S. Tetrahedron Lett. 2000, 41,
1385e1387.
€
24. Christoffers, J.; Werner, T.; Unger, S.; Frey, W. Eur. J. Org. Chem. 2003, 425e431.