An efficient, practical approach to the synthesis of 2,4-disubstituted thiazoles and oxazoles: Application to the synthesis of GW475151

Article abstract of DOI:10.1021/op000086g

GlaxoWellcome Research and Development, Chemical Development Division, Medicines Research Centre,. A new method for the synthesis of 2,4-disubstituted oxazoles and thiazoles and 2,4,5-trisubstituted oxazoles from readily available starting materials is described. The methodology has been applied on multigram scale and involves transfer of oxidation state through a molecular framework. In particular the oxazole-containing amino acid fragment of the 5,5-transfused lactam GW475151, 1, has been prepared in excellent yield and purity.

Full text of DOI:10.1021/op000086g

Organic Process Research & Development 2001, 5, 37−44
An Efficient, Practical Approach to the Synthesis of 2,4-Disubstituted
Thiazoles and Oxazoles: Application to the Synthesis of GW475151
Stephen A. Hermitage,* Kevin S. Cardwell, Tim Chapman, Jason W. B. Cooke, and Rebecca Newton
GlaxoWellcome Research and DeVelopment, Chemical DeVelopment DiVision, Medicines Research Centre,
Gunnels Wood Road, SteVenage, Hertfordshire, SG1 2NY, UK
Abstract:
reaction of diazocarbonyl compounds with nitriles,7 oxidation
A new method for the synthesis of 2,4-disubstituted oxazoles
and thiazoles and 2,4,5-trisubstituted oxazoles from readily
available starting materials is described. The methodology has
been applied on multigram scale and involves transfer of
oxidation state through a molecular framework. In particular
the oxazole-containing amino acid fragment of the 5,5-trans-
fused lactam GW475151, 1, has been prepared in excellent yield
and purity.
of oxazolines,8 Hantzsch-type condensation of amides and
R-bromo ketones,9 acid-catalyzed cyclisation of substituted
iminoethers via the Cornforth protocol,10 cyclodehydration
of aldehydoamides using Ph3PI211 and acid-mediated conver-
sion of amide acetals.12
Many of these routes have limitations in terms of yield,
operation on kilogram scale, cost of reagents, and robustness.
We sought an efficient route into the oxazole acid 3 that
would be applicable to routine large-scale manufacture. We
surmised that oxazolines could be converted to oxazoles
through transfer of oxidation state through a molecular
framework13 (Figure 3). To this end we anticipated oxazolines
of type 4 (where X is a leaving group) when treated with
base would undergo deprotonation R to the ester followed
by a subsequent elimination of HX to form intermediates of
type 5. We postulated that the intermediate 5 would either
aromatise (pathway A) or be trapped by a nucleophile
(pathway B). Should the intermediate 7 be produced, we
hoped subsequent elimination of the nucleophile would
produce the oxazole 6. This concept is attractive since
environmentally unacceptable and high-cost reagents are
circumvented. Indeed, the concept could be applied to the
preparation of small-molecule building blocks used in more
complex natural product synthesis.
Introduction
The oxazole and thiazole ring systems are common
structural motifs in a number of biologically active mol-
ecules.1
These five-membered heterocyclic ring systems originate
in nature as a consequence of peptide modifications contain-
ing serine, threonine, and cysteine side chain residues and
are the product of cyclodehydration and redox reactions2
(Figure 1). All three ring oxidation states (the azoline,
azolidine, and azole) are observed and alter peptide backbone
connectivity, conformation, and electronic distribution, thus
offering opportunity as new recognition elements for inter-
molecular interactions.
Pyrrolidine 5,5-trans-fused ring systems have been of
particular interest to us as templates for the inhibition of
human neutrophil elastase (HNE).3,4 In particular we have
been interested in the pyrrolidine 5,5-trans-fused lactam,
GW475151,5 1, containing a 2,4-disubstituted oxazole as a
potential therapy for respiratory diseases such as acute
respiratory distress syndrome, cystic fibrosis, emphysema,
and chronic bronchitis.
In this report we disclose a novel route to oxazole and
thiazole ring systems, with application to the synthesis of
the oxazole acid 3 and other oxazole- and thiazole-containing
molecules.
Results and Discussion
Preparation of the methyl imidate 9 from dichloroaceto-
nitrile 8 was achieved by the dropwise addition of dichloro-
acetonitrile to a methanol solution containing sodium meth-
Retrosynthetic analysis of GW475151, 1, (Figure 2)
identified two coupling partners, the pyrrolidine 5,5-trans-
fused lactam, 2, and the 2,4-disubstituted oxazole acid, 3.
There have been many synthetic efforts surrounding the
preparation of oxazoles6 including rhodium(II)-catalysed
(7) Bagley, M. C.; Buck, R. T.; Hind, S. L.; Moody, C. J. J. Chem. Soc.; Perkin
Trans. 1 1998, 591.
(8) (a) Evans, D. L.; Minster, D. K.; Jordis, U.; Hecht, S. M.; Mazzu, A. L.,
Jr.; Meyers, A. I. J. Org. Chem. 1979, 44, 497. (b) Meyers, A. I.; Tavares,
F. X. J. Org. Chem. 1996, 61, 8207. (c) Williams, D. R.; Lowder, P. D.;
Gu, Y.-G.; Brooks, D. A. Tetrahedron Lett. 1997, 38, 331.
(9) (a) Liu, P.; Celatka, C. A.; Panek, J. S. Tetrahedron Lett. 1997, 38, 5445.
(b) Panek, J. S.; Beresis, R. T. J. Org. Chem. 1996, 61, 6496.
(10) (a) Cornforth J. W.; Cornforth R. H. J. Chem. Soc. 1947, 96. (b) Cornforth,
J. W.; Cookson, E. J. Chem. Soc. 1952, 1085.
(11) Wipf, P.; Miller, C. P. J. Org. Chem. 1993, 58, 3604.
(12) Swaminathan, S.; Singh, A. K.; Li, W.-S.; Venit, J. J.; Natalie, K. J., Jr.;
Simpson, J. H.; Weaver, R. E.; Silverberg, L. J. Tetrahedron Lett. 1998,
39, 4769.
(1) Lewis, J. R. Nat. Prod. Rep. 1999, 16, 389.
(2) Roy, R. S.; Gehring, A. M.; Milne, J. C.; Belshaw, P. J.; Walsh, C. T. Nat.
Prod. Rep. 1999, 16, 249.
(3) Macdonald, S. J. F.; Clarke, G. D. E.; Dowle, M. D.; Harrison, L. A.;
Hodgson, S. T.; Inglis, G. G. A.; Johnson, M. R.; Shah, P.; Upton, R. J.;
Walls, S. B. J. Org. Chem. 1999, 64, 5166.
(4) Macdonald, S. J. F.; Belton, D. J.; Buckley, D. M.; Spooner, J. E.; Anson,
M. S.; Harrison, L. A.; Mills, K.; Upton, R. J, Dowle, M. D.; Smith, R. A.;
Molloy, C. R.; Risley, C. J. Med. Chem. 1998, 41, 3919.
(5) Clarke, G. D. E.; Dowle, M. D.; Finch, H.; Harrison, L. A.; Inglis, G. G.
A.; Johnson, M. R.; Macdonald, S. J. F.; Shah, P.; Smith, R. A. WO9912933.
(6) (a) Boyd, G. V. Prog. Heterocycl. Chem.; 1997, 9, 207. (b) Boyd, G. V.;
Prog. Het. Chem. 1998, 10, 209.
(13) Cardwell, K. S.; Hermitage, S. A.; Sjolin, A. Tetrahedron Lett. 2000, 41,
4239.
10.1021/op000086g CCC: $20.00 © 2001 American Chemical Society and The Royal Society of Chemistry
Published on Web 11/17/2000
Vol. 5, No. 1, 2001 / Organic Process Research & Development
•
37

Figure 1.
Figure 2.
Figure 3.
oxide as catalyst at ∼0 °C. Condensation of the methyl
imidate 9 with serine methyl ester hydrochloride gave the
oxazoline 10 in excellent crude yield. Alternatively, the
thiazoline 11 was prepared when 9 reacted with cysteine
methyl ester hydrochloride. Upon treatment of 10 and 11
with 1 mol equiv of sodium methoxide in methanol,
rearrangement occurred as anticipated in Figure 3. Interest-
ingly, a subtle stereoelectronic effect led to the oxazoline
10 giving rise to the methoxy intermediate 14, whereas the
thiazolidine 11 gave rise to the thiazole 13 directly. Methoxy
oxazoline 14 could be converted to oxazole 12 by acid-
catalyzed elimination using CSA (camphor sulphonic acid)
in toluene at 70 °C. In both cases the oxazole 12 and thiazole
13 were prepared in respectable yield over the synthetic
sequence on a multigram scale without intermediate purifica-
tions. The structure of oxazole 12 was confirmed by X-ray
crystallography as depicted in Figure 4.
Figure 4. X-ray structure of 12.
oxazoline 16 was obtained with little stereoselectivity (3:2
ratio of diastereomers).
The monochloro derivative 20 was prepared from chloro-
acetonitrile 18 via the imidate 19. Transfer of oxidation state
was achieved via the methoxyoxazoline 21 to generate 22
(Scheme 3).
In a series of reactions analogous to that depicted in
Scheme 1, the more substituted oxazole 17 was prepared in
good overall yield of 48% from dichloroacetonitrile 8 and
threonine methyl ester hydrochloride (Scheme 2). In the key
“transfer of oxidation state” reaction, intermediate methoxy
38
•
Vol. 5, No. 1, 2001 / Organic Process Research & Development

Scheme 1a
a Key: a) NaOMe (10 mol %), MeOH, -10 to 0 °C; b) serine methyl ester hydrochloride, MeOH (10 88% crude yield) or cysteine methyl ester hydrochloride,
MeOH (11 84%, crude yield); c) NaOMe, MeOH, 10 °C (14 84% crude from 8, 13 66% yield after purification from 8); d) CSA, toluene, 70 °C (12 48% yield after
purification from 8).
Scheme 2 a
a Key: a) i) NaOMe (10 mol %), MeOH, -10 to 0 °C; ii) threonine methyl ester hydrochloride, MeOH (15 quantitative crude yield); b) NaOMe, MeOH, 10 °C (16
86% crude yield from 8); c) CSA, toluene, 70 °C (17 48% yield after purification from 8).
Scheme 3 a
a Key: a) i) NaOMe (10 mol %), MeOH, -10 to 0 °C; b) serine methyl ester hydrochloride, MeOH (20 70% crude yield from 18); c) NaOMe, MeOH, 10 °C (21
67% crude yield from 18); d) CSA, toluene, 70 °C (22 24% yield after purification from 18).
In an effort to establish a reliable process for the
manufacture of GW475151, 1, we investigated the robustness
of conversion of 10 to 12 via the intermediate 14. Our
intention was to run a “telescoped process” on pilot plant
from dichloroacetonitrile 8 without isolation of any inter-
mediates. Part of our “route robustness testing” was to treat
the dichlorooxazoline 10 with 2 equiv of sodium methoxide
in methanol to simulate the event of an overcharge of reagent.
Interestingly, we observed the conversion of 10 to a mixture
of 12 and 24 (1:6). Our proposed mechanism for the
formation of this interesting product 24 is outlined in Scheme
4. Reaction calorimetry revealed that the conversion of 10
into 14 is essentially instantaneous and controlled by the rate
of addition of sodium methoxide solution. In this process
the sodium methoxide is consumed and sodium chloride
formed as a by-product. When the second molar equivalent
of sodium methoxide is added, we propose that intermediate
23 can be regenerated. This regeneration of 23 from 14
however requires only a catalytic amount of sodium meth-
oxide (since the methoxide anion is renewed as a “leaving
group”), allowing the remaining sodium methoxide to act
as a nucleophile. Addition of methoxide to 23 followed by
elimination of chloride, addition of methanol, and ultimately
â-elimination/tautomerisation gives 24.
Improvements to the Synthetic Route. The observation
of dimethoxy derivative 24 led us to examine further our
Vol. 5, No. 1, 2001 / Organic Process Research & Development
•
39

Scheme 4
Scheme 5 a
i
a Key: a) NaOMe (10 mol %), MeOH, CH2Cl2 -10 to 0 °C; b) serine ethyl ester hydrochloride, MeOH, CH2Cl2; c) Pr2NEt, CH2Cl2; d) pyrrolidine, CH2Cl2; e)
maleic acid, IPA (28 74% yield after purification from 8).
new methodology for the formation of oxazoles. Our
investigations centered around chemical robustness with other
considerations such as streamlining of solvents, “work up”
requirements, the potential for telescoping stages and “end
game” isolations as other key areas. We believed that
attaining robustness in our chemical sequence through an
understanding of the chemical reactivity and selectivity
would increase the overall chemical yield.
methane:methanol volume solvent mixture was used. Reac-
tion of serine ethyl ester hydrochloride with methyl imidate
9 gave the dichlorooxazoline 25 in excellent crude yield and
1
purity by H NMR spectroscopy. Since the bulk of the
solution was dichloromethane at this stage, a simple aqueous
work up removed the ammonium chloride by-product leaving
the dichlorooxazoline 25 in solution in anticipation for the
elimination reaction. A base screen15 identified Hunig’s base
(iPr2NEt) as the best reagent for the smooth conversion of
25 to the oxazole 26. Evidence of quaternarisation of 26 in
the mass spectrum was observed when triethylamine was
used as base. The use of dichloromethane instead of methanol
as solvent for this reaction circumvented formation of the
methoxyoxazoline intermediate 14. We found it advantageous
Critical examination of the chemistry at each stage led
us to optimise the series of reactions depicted in Scheme 5
for the conversion of dichloroacetonitrile 8 through to the
pyrrolidinooxazole ethyl ester maleate salt 28. The conver-
sion of dichloroacetonitrile 8 to the methyl imidate 9 at high
concentration ( ∼2-3 vols in methanol with respect to the
dichloroacetonitrile) led to the formation of a dark red
solution of the imidate 9 containing unknown impurities.
Cleaner conversion of 8 to 9 was observed (monitoring the
(14) Galeazzi, E.; Guzman, A.; Nava, J. L.; Lui, Y.; Maddox, M. L.; Muchowski,
J. M. J. Org. Chem. 1995, 60, 1090.
(15) Bases includedEt3N, nBu3N, iPr2NEt, N-butylpyrrolidine, cyclohexyl-
diethylamine, N-phenyl morpholine, pyridine, dimethyl aniline, picoline.
1
reaction by H NMR spectroscopy) when a 9:1 dichloro-
40
•
Vol. 5, No. 1, 2001 / Organic Process Research & Development

Scheme 6 a
a Key: a) CH2Cl2, aq K2CO3; b) KOH, IMS, acetone (29 90% yield after purification from 28)
to remove the Hunig’s base at this point with an aqueous
acid wash since telescoping the series of reactions with
Hunig’s base present made for a problematic isolation of
maleate salt 28. When oxazoline 25 and oxazole 26 were
independently subjected to prolonged treatment with aqueous
acid, the oxazoline hydrolysed to aqueous soluble by-
products, whereas the oxazole was recovered in near
quantitative yield. This offered purification/rework oppor-
tunities should they be required in future batches. Simple
pyrrolidine displacement of 26 to give 27 required 2 equiv
of pyrrolidine to drive the reaction to completion. This effect
has been observed in a similar displacement16 and is
associated with the formation of HCl and protonation of the
pyrrolidine during the course of reaction. The use of the ethyl
ester 26 offers a significant advantage over the methyl ester
counterpart 12. Treatment of methyl ester 12 with pyrrolidine
gave amide 3017 as a major by-product, whereas the ethyl
industrial methylated spirit (IMS). The potassium salt was
isolated via precipitation from IMS/acetone in 90% yield
from 28.
Conclusions
This report summarises a novel and efficient route to
oxazoles 12, 17, 22, 26, and the thiazole 13. The approach
uses inexpensive, widely available starting materials and
provides another general synthetic approach to this important
class of compound. We have applied this chemistry to the
synthesis of a key building block, 3, of GW475151, 1. This
chemistry can be performed on multigram scale and avoids
the need for purification by chromatography.
General Experimental Section
Proton and carbon NMR were recorded on a Bruker
DPX400 spectrometer. Microanalytical data were performed
by Butterworth Laboratories Ltd. Infrared spectra were
recorded on a Nicolet 20SXC FTIR spectrophotometer and
mass spectra run on Micromass Q-TOF, hybrid quadrupole
time-of-flight MS -/+ve ion electrospray instrument. Re-
agents and solvents were obtained from commercial suppli-
ers.
Experimental Section
ester 26 does not undergo amidation even under forcing
conditions using 5 equiv of pyrrolidine at reflux. A screen
of acids (CSA, maleic, fumaric, MeSO3H, oxalic) against a
series of solvents (acetonitrile, ethanol, isopropylacetate,
acetone, dichloromethane, dimethylformamide, isopropyl
alcohol) with the pyrrolidinooxazole 27 was examined to
determine the best conditions for isolation of 27 as a salt.
Our criteria for success was that the salt formed had to be
insoluble in the chosen solvent, and the parent acid, soluble.
Maleic acid and isopropyl alcohol were chosen, and the
maleate salt 28 was isolated in 74% yield in a telescoped
process from dichloroacetonitrile 8. The salt 28 was routinely
isolated as a 1:1 stoichiometry of base and acid. In summary,
we have significantly improved our original synthetic
sequence by carefully considering the choice of solvents (and
reaction concentration), the choice of reagents (and their
stoichiometry), the ability to telescope reactions, the isolation
strategy, and overall robustness of the process.
Methyl 2-(dichloromethyl)-4,5-dihydro-1,3-oxazole-4-
carboxylate, 10. A solution of sodium methoxide in
methanol (25% w/w, 5.75 mL, 24.9 mmol) was diluted with
methanol (50 mL) and cooled to -10 °C. Dichloroacetonitrile
8 (20 mL, 249 mmol) was added dropwise over 25 min
whilst the temperature was maintained below 0 °C. The
mixture was stirred for a further 20 min at -5 °C, then DL-
serine methyl ester hydrochloride (38.7 g, 249 mmol) was
added along with methanol (40 mL). The mixture was stirred
overnight, gradually warming to room temperature. CH2Cl2
(140 mL) and water (80 mL) were added and the layers
separated. The aqueous layer was extracted with CH2Cl2 (80
mL) and the combined organic extracts concentrated in vacuo
to give dichlorooxazoline 10 (46.7 g, 88%, crude from
dichloroacetonitrile) as an orange oil. Spectroscopic data
showed dichlorooxazoline 10 to be sufficiently pure to
proceed without further purification. Rf 0.30 (20:80 EtOAc:
iso-octane); IR (thin film) 1662, 1744 cm-1; 1H NMR (400
MHz, CDCl3) δH 3.83 (s, 3H), 4.64-4.69 (m, 1H), 4.73-
4.77 (m, 1H), 4.87-4.93 (m, 1H), 6.29 (s, 1H); 13C NMR
(100 MHz, CDCl3) δC 53.01, 60.83, 68.10, 71.25, 164.51,
170.18; MS calcd for C6H8NO3Cl2 (MH+) 211.9881, found
211.9889.
The target pyrrolidinooxazole potassium salt 29 was
prepared in two steps from the maleate salt 28 as outlined
in Scheme 6. The free base 27 of 28 was liberated using
potassium carbonate solution, extracted into dichloromethane,
and subsequently hydrolysed using potassium hydroxide in
(16) Nyce, P. L.; Gala, D.; Steinman, M. Synthesis 1991, 571.
(17) An authentic sample of 30 was prepared from 29 using CSA/CDI activation
followed by reaction with pyrrolidine.
Methyl 2-(chloromethyl)-4-methoxy-4,5-dihydro-1,3-
oxazole-4-carboxylate, 14. To a solution of crude di-
Vol. 5, No. 1, 2001 / Organic Process Research & Development
•
41

chlorooxazoline 10 (46.2 g, 218 mmol) in methanol (40 mL)
was added a solution of sodium methoxide in methanol (25%
w/w, 49.9 mL, 218 mmol) dropwise over 50 min, keeping
the temperature below 10 °C. The mixture was left to stir
overnight, gradually warming to room temperature. CH2Cl2
(140 mL) and water (80 mL) were added and the layers
separated. The aqueous layer was extracted with CH2Cl2 (80
mL), and the combined organic extracts were concentrated
in vacuo to give the methoxy intermediate 14 (43.3 g, 84%,
crude from dichloroacetonitrile 8 in the previous step) as an
oil. Rf 0.25 (25:75 EtOAc:iso-octane); IR (thin film) 1657,
dropwise over 50 min, keeping the temperature below 10
°C. The mixture was then stirred for 2 h. CH2Cl2 (65.5 mL)
and water (37.5 mL) were added and the layers separated.
The aqueous layer was extracted with CH2Cl2 (37.5 mL) and
the sample concentrated in vacuo to yield the crude chloro-
thiazole 13 (20.3 g) as an orange/brown solid. A sample of
the crude product (2.5 g) was removed and purified by flash
column chromatography (25:75 increasing to 30:70 EtOAc:
iso-octane) to yield the pure chlorothiazole 13 (1.71 g, 66%
isolated pure from dichloroacetonitrile 8 used in the synthetic
sequence) as a white solid. Rf 0.25 (25:75 EtOAc:iso-octane);
1
1
1750 cm-1; H NMR (400 MHz, CDCl3) δH 3.42 (s, 3H),
IR (Nujol mull) 1723 cm-1; H NMR (400 MHz, CDCl3)
3.85 (s, 3H), 4.19 (d, J ) 13, 1H), 4.23 (d, J ) 13, 1H),
4.39 (d, J ) 10.5, 1H), 4.59 (d, J ) 10.5, 1H); 13C NMR
(100 MHz, CDCl3) δC 36.03, 51.99, 53.13, 75.48, 101.86,
168.01, 169.28.
δH 3.97 (s, 3H), 4.89 (s, 2H), 8.23 (s, 1H); 13C NMR (100
MHz, CDCl3) δC 41.75, 53.01, 129.59, 147.18, 161.89,
168.00; MS calcd for C6H7NO2ClS (MH+) 191.9886, found
191.9870.
Methyl 2-(chloromethyl)-1,3-oxazole-4-carboxylate, 12.
To methoxy intermediate 14 (42.9 g, 207 mmol) in toluene
(100 mL) was added camphorsulphonic acid (7.21 g, 31.1
mmol) at room temperature, and the mixture was heated to
70 °C. This temperature was maintained for 50 min, and
then the mixture was cooled to room temperature and washed
with aqueous K2CO3 solution (10% w/v, 60 mL) followed
by water (80 mL). The combined aqueous extracts were
back-extracted with toluene (120 mL) and the combined
organic layers concentrated in vacuo to yield the crude
chlorooxazole 12 (30.1 g) as a brown solid. A sample of
crude chlorooxazole (2.5 g) was purified by flash column
chromatography (25:75 EtOAc:iso-octane) to yield the pure
chlorooxazole 12 (1.73 g, 48% isolated pure from dichloro-
acetonitrile 8 in this sequence) as a white solid. IR (Nujol
mull) 1578, 1715 cm-1; 1H NMR (400 MHz, CDCl3) δH 3.93
(s, 3H), 4.63 (s, 2H), 8.27 (s, 1H);13C NMR (100 MHz,
CDCl3) δC 35.31, 52.35, 133.85, 145.07, 159.91, 161.08; MS
calcd for C6H7NO3Cl (MH+) 176.0114, found 176.0127.
Methyl 2-(dichloromethyl)-4,5-dihydro-1,3-thiazole-4-
carboxylate, 11. A solution of sodium methoxide in
methanol (25% w/w, 2.67 mL, 11.65 mmol) was diluted with
methanol (23.5 mL) and cooled to -10 °C. Dichloroaceto-
nitrile 8 (9.35 mL, 116.5 mmol) was added dropwise over
25 min, keeping the temperature below 0 °C. The mixture
was stirred for a further 30 min, and then L-cysteine methyl
ester hydrochloride (20.0 g, 116.5 mmol) was added along
with methanol (18.7 mL). After the mixture stirred overnight,
CH2Cl2 (65.5 mL) and water (37.5 mL) were added and the
layers separated. The aqueous layer was then extracted with
CH2Cl2 (37.5 mL), and the combined organic extracts were
concentrated in vacuo to yield dichlorothiazoline 11 (22.0
g, 84% crude from dichloroacetonitrile 8) as an oil. IR (thin
Methyl 2-(dichloromethyl)- 5-methyl-4,5-dihydro-1,3-
oxazole-4-carboxylate, 15. A solution of sodium methoxide
in methanol (25% w/w, 0.67 mL, 2.95 mmol) was diluted
with methanol (5.9 mL) and cooled to -10 °C. Dichloro-
acetonitrile 8 (2.37 mL, 29.5 mmol) was added dropwise
over 15 min whilst the temperature was maintained below 0
°C. The mixture was stirred for a further 35 min at -5 °C,
and then L-threonine methyl ester hydrochloride (5.0 g, 29.5
mmol) was added along with methanol (4.7 mL). The
mixture was stirred for 2 h. CH2Cl2 (16.5 mL) and water
(9.5 mL) were added and the layers separated. The aqueous
layer was extracted with CH2Cl2 (9.5 mL), and the combined
organic extracts were concentrated in vacuo to give the
dichlorooxazoline 15 (6.65 g, quantitative) as an orange oil.
Spectroscopic data showed dichlorooxazoline 15 to be
sufficiently pure to proceed without further purification. IR
1
(thin film) 1658, 1744, 2956 cm-1; H NMR (400 MHz,
CDCl3)_δH 1.53 (d, J ) 6.3, 3H), 3.81 (s, 3H), 4.39 (d, J )
7.5, 1H), 5.03-5.09 (m, 1H), 6.27 (s, 1H); 13C NMR (100
MHz, CDCl3) δC 20.96, 53.29, 61.50, 74.79, 81.53, 164.13,
170.51; MS calcd for C7H10NO3Cl2 (MH+) 226.0038, found
226.0027.
Methyl 2-(chloromethyl)-4-methoxy-5-methyl-4,5-di-
hydro-1,3-oxazole-4-carboxylate, 16. To a solution of
dichlorooxazoline 15 (6.39 g, 28.3 mmol) in methanol (4.7
mL) was added a solution of sodium methoxide in methanol
(25% w/w, 6.46 mL, 28.3 mmol) dropwise over 20 min,
keeping the temperature below 10 °C. After stirring for 90
min, CH2Cl2 (16.5 mL) and water (9.5 mL) were added and
the layers separated. The aqueous layer was extracted with
CH2Cl2 (9.5 mL), and the combined organic extracts were
concentrated in vacuo to give the methoxy intermediate 16
(5.40 g, 86% crude from dichloroacetonitrile 8) as an oil.
IR (thin film) 1660, 1750, 2954, 2986 cm-1; 1H NMR (400
MHz, CDCl3) δH 1.29 (d, J ) 6.8, 1.8H,), 1.44 (d, J ) 6.6,
1.2H), 3.43 (s, 1.8H), 3.45 (s, 1.2H), 3.84 (s, 3H), 4.18 (s,
1.2H), 4.20 (s, 1.8H), 4.65 (q, J ) 6.8, 0.6H), 4.74 (q, J )
6.6, 0.4H).
1
film) 1610, 1743, 2850, 2955 cm-1; H NMR (400 MHz,
CDCl3) δH 3.64-3.78 (m, 2H), 3.84 (s, 3H), 5.19 (m, 1H),
6.50 (s, 1H); 13C NMR (100 MHz, CDCl3) δC 35.81, 53.07,
66.33, 77.46, 170.01, 172.41; MS calcd for C6H8NO2SCl2
(MH+) 227.9653, found 227.9652.
Methyl 2-(chloromethyl)-1,3-thiazole-4-carboxylate, 13.
A solution of dichlorothiazoline 11 (21.3 g, 93.5 mmol) in
methanol (20 mL) was treated with a solution of sodium
methoxide in methanol (25% w/w, 21.4 mL, 93.5 mmol)
Methyl 2-(chloromethyl)-5-methyl-1,3-oxazole-4-car-
boxylate, 17. To the methoxy intermediate 16 (5.07 g, 22.9
mmol) in toluene (15 mL) was added camphorsulphonic acid
(0.80 g, 3.4 mmol) at room temperature and the mixture
42
•
Vol. 5, No. 1, 2001 / Organic Process Research & Development

heated to 70 °C for 90 min. The solution was cooled to room
temperature and washed with aqueous K2CO3 solution (10%
w/v, 7 mL) followed by water (9.5 mL). The combined
aqueous extracts were back-extracted with toluene (14 mL)
and the combined organic layers concentrated in vacuo to
give the crude chlorooxazole 17 (3.83 g) as an orange solid.
A sample of crude chlorooxazole 17 (2.0 g) was purified by
flash column chromatography (25:75 EtOAc:iso-octane) to
yield the pure chlorooxazole product 17 (1.26 g, 48% from
dichloroacetonitrile 8 used in the synthetic sequence) as a
sulphonic acid (1.29 g, 5.6 mmol) was added. The mixture
was stirred for a further 2 h. The solution was cooled to
room temperature and washed with aqueous K2CO3 solution
(10% w/v, 12 mL) followed by water (16 mL). The combined
aqueous extracts were back-extracted with toluene (24 mL)
and the combined organic layers concentrated in vacuo to
yield the methyl oxazole 22 (2.28 g) as a yellow solid. A
sample of crude methyl oxazole (1.4 g) was purified by flash
column chromatography (25:75 EtOAc:iso-octane) to yield
the pure compound 22 (1.14 g, 24% isolated pure from
chloroacetonitrile 18) as a white solid. Rf 0.11 (25:75 EtOAc:
1
white solid. IR (Nujol mull) 1615, 1720, 2961 cm-1; H
1
NMR (400 MHz, CDCl3) δH 2.66 (s, 3H), 3.92 (s, 3H), 4.58
(s, 2H); 13C NMR (100 MHz, CDCl3) δC 12.40, 35.85, 52.45,
128.33, 157.41, 158.24, 162.60; MS calcd for C7H9NO3Cl
(MH+) 190.0271, found 190.0256.
iso-octane); IR (Nujol mull) 1734, 2853, 2923 cm-1; H
NMR (400 MHz, CDCl3) δH 2.52 (s, 3H), 3.91 (s, 3H), 8.14
(s, 1H); 13C NMR (100 MHz, CDCl3) δC 14.21, 52.46,
133.63, 144.14, 162.09, 162.80; MS calcd for C6H8NO3
(MH+) 142.0504, found 142.0487.
Methyl 2-(chloromethyl)-4,5-dihydro-1,3-oxazole-4-
carboxylate, 20. A solution of sodium methoxide in
methanol (25% w/w, 1.47 mL, 6.43 mmol) was diluted with
methanol (10 mL) and cooled to -6 °C. Chloroacetonitrile
18 (4.06 mL, 64.3 mmol) was added dropwise over 15 min
keeping the temperature below 2 °C. The mixture was stirred
for 45 min, then DL-serine methyl ester hydrochloride (10.0
g, 64.3 mmol) was added along with methanol (8 mL). The
reaction mixture was stirred overnight. CH2Cl2 (29 mL) and
water (16 mL) were then added and the layers separated.
The aqueous layer was extracted with CH2Cl2 (16 mL), and
the combined organic extracts were concentrated in vacuo
to give the chlorooxazoline 20 (8.02 g, 70% crude from
chloroacetonitrile 18) as an orange oil. Spectroscopic data
showed the sample to be sufficiently pure to proceed without
further purification. IR (thin film) 1663, 1743, 2957 cm-1;
1H NMR (400 MHz, CDCl3) δH 3.81 (s, 3H), 4.16 (s, 2H),
4.53-4.58 (m, 1H), 4.61-4.66 (m, 1H), 4.81-4.86 (m, 1H);
13C NMR (100 MHz, CDCl3) δC 36.37, 53.23, 68.60, 70.90,
165.90, 171.21; MS calcd for C6H9NO3Cl (MH+) 178.0271,
found 178.0253.
Methyl 2-methyl-4-methoxy-4,5-dihydro-1,3-oxazole-
4-carboxylate, 21. To a solution of chlorooxazoline 20 (7.48
g, 42.1 mmol) in methanol (8 mL) was added a solution of
sodium methoxide in methanol (25% w/w, 9.63 mL, 42.1
mmol) dropwise over 20 min, keeping the temperature below
10 °C. The reaction mixture was stirred for 90 min, and then
CH2Cl2 (29 mL) and water (16 mL) were added and the
layers separated. The aqueous layer was extracted with
CH2Cl2 (16 mL), and the combined organic extracts were
concentrated in vacuo to give methyl methoxy intermediate
21 (7.00 g, 40.4 mmol, 67% crude from chloroacetonitrile
18) as an oil. IR (thin film) 1655, 1750, 2956 cm-1; 1H NMR
(400 MHz, CDCl3) δH 2.12 (s, 3H), 3.38 (s, 3H), 3.83 (s,
3H), 4.26 (d, J ) 10.3, 1H), 4.47 (d, J ) 10.3, 1H); 13C
NMR (100 MHz, CDCl3) δC 14.68, 52.00, 53.32, 74.90,
102.46, 170.33, 171.11; MS calcd for C7H12NO4 (MH+)
174.0766, found 174.0758.
Dimethoxy intermediate, 24. To dichlorooxazoline 10
(2 g, 9.43 mmol) in MeOH (10 mL) was added NaOMe in
MeOH (4.3 mL, 18.87 mmol) at 0 °C. The mixture was
allowed to warm to room temperature and stirred for a further
1 h. CH2Cl2 (150 mL) and water (50 mL) were added
followed by brine (50 mL). The CH2Cl2 layer was separated
and concentrated in vacuo to leave an oil (1.66 g). Crude
1H NMR spectroscopy of the oil indicated a 6:1 ratio of 24:
12. Attempts to purify 24 by column chromatography
resulted in decomposition. Data on 24: 1H NMR (400 MHz,
CDCl3) δH 2.15 (s, 3H), 3.41 (s, 3H), 3.50 (s, 3H), 3.84 (s,
3H), 5.30 (s, 1H); 13C NMR (100 MHz, CDCl3) δC 15.00,
52.56, 52.81, 57.99, 109.87, 145.04, 167.67, 170.37; MS
calcd for C8H14NO4 (MH+) 204.0872, found 204.0869.
Ethyl 2-(pyrrolidin-1-ylmethyl)-1,3-oxazole-4-carbox-
ylate maleate salt, 28. To CH2Cl2 (495 mL) and MeOH
(55 mL) was added NaOMe in MeOH (1.61 mL, 0.007 mol,
0.01 equiv, 25% w/w solution) under nitrogen and the
mixture cooled to -5 °C. Dichloroacetonitrile 8 (75 g, 0.68
mol) was added dropwise over 45 min, maintaining the
temperature below 0 °C, and the mixture stirred for 60 min
at 0 °C. Serine ethyl ester HCl was added (115.3 g, 0.68
mol) and the mixture stirred overnight at 20 °C. Water (275
mL) was added and the CH2Cl2 layer separated. The aqueous
phase was back-extracted with CH2Cl2 (275 mL), and the
combined CH2Cl2 extracts were concentrated at atmospheric
pressure to 350 mL. Hunig’s base (178 mL, 1.02 mol) was
added and the mixture heated to 50 °C for 5 h and then
cooled to 20 °C and stirred overnight. CH2Cl2 (440 mL) was
added, the mixture was cooled to 5 °C, and then HCl (2 M,
500 mL) was added cautiously. The CH2Cl2 layer was
separated and washed with water (250 mL), and the solution
was concentrated to 600 mL at atmospheric pressure and
then cooled to 15 °C. Pyrrolidine (113.5 mL, 1.36 mol) was
added over 10 min, and the mixture was heated to reflux for
30 min and then chilled to -5 °C and stored overnight. The
temperature was adjusted to 20 °C and the CH2Cl2 solution
washed with aqueous K2CO3, (10% w/v, 1000 mL) followed
by water (200 mL). The CH2Cl2 extracts were concentrated
at atmospheric pressure to 350 mL and then IPA (300 mL)
was added. The mixture was further concentrated at atmo-
spheric pressure to 350 mL and then cooled to 0 °C. Maleic
Methyl 2-methyl-1,3-oxazole-4-carboxylate, 22. To the
methyl methoxy intermediate 21 (6.43 g, 37.1 mmol) in
toluene (20 mL) was added camphorsulphonic acid (1.29 g,
5.6 mmol) at room temperature. The mixture was heated to
70 °C for 2 h whereupon a second portion of camphor-
Vol. 5, No. 1, 2001 / Organic Process Research & Development
•
43

acid (78.9 g, 0.68 mol) was added as a solution in IPA (400
mL) over 5 min and the resulting precipitate stirred at 0-4
°C for a further 30 min. The solid was filtered off at suction
and washed with chilled (0 °C) isopropyl alcohol (200 mL).
The filter cake was dried at 50 °C in vacuo to give maleate
salt 28 (1:1 stoichiometry of free base and maleic acid) (170.7
g, 74%) as a pale yellow crystalline solid. IR (Nujol mull)
1724, 2853, 2923 cm-1; 1H NMR (400 MHz, d4-MeOH) δH
1.35 (t, J ) 7, 3H), 2.08-2.15 (m, 4H), 3.48-3.51 (m, 4H),
4.35 (q, J ) 7, 2H), 4.67 (s, 2H), 6.24 (s, 2H), 8.64 (s, 1H);
13C NMR (100 MHz, d4-MeOH) δC 14.47, 24.15, 50.72,
55.90, 62.56, 135.01, 136.52, 147.51, 158.11, 162.16, 170.78;
MS calcd for C11H17N2O3 (MH+) 225.1239, found 225.1223.
Anal. Calcd for C15H20N2O7: C, 52.94; H, 5.92; N, 8.23.
Found: C, 53.30; H, 5.84; N, 8.06.
Amide impurity, 30. Potassium salt 29 (4 g, 17.1 mmol)
was slurried in CH2Cl2 (26 mL) under nitrogen and cooled
to ∼8 °C. Camphorsulphonic acid (7.9 g, 34.2 mmol) was
added in one portion and a suspension of carbonyl diimid-
azole (CDI) (2.8 g, 17.1 mmol) in CH2Cl2 (14 mL) added
over ∼25 min. CH2Cl2 (7 mL) was used to rinse in excess
CDI solid. The mixture was allowed to stir overnight,
gradually warming to room temperature and then pyrrolidine
(2.2 mL, 25.7 mmol) added in one portion. The mixture was
stirred at room temperature for 4 h. Aqueous NaHCO3 (7%
w/w solution, 40 mL) was added and the mixture stirred for
15 min. The layers were separated and the aqueous layer
back-extracted with CH2Cl2 (10 mL). The combined CH2Cl2
extracts were washed with aqueous NaHCO3 (7%w/w
solution, 20 mL) and brine (20 mL) and then concentrated
in vacuo to give the amide 30 as a tan oil (2.04 g, 48%)
which solidified upon standing. IR (Nujol mull) 1608, 1633,
2825, 2924 cm-1; 1 H NMR (400 MHz, d4-MeOH) δH 1.81-
1.86 (m, 2H), 1.90-2.03 (m, 4H), 2.66-2.69 (m, 2H), 3.56-
3.59 (m, 2H), 3.85 (s, 2H), 3.88-3.92 (m, 2H), 8.35 (s, 1H);
13C NMR (100 MHz, d4-MeOH) δC 24.33, 24.75, 27.32,
40.03, 49.66, 52.07, 54.78, 137.78, 144.34, 162.04, 162.57;
MS calcd for C13H20N3O2 (MH+) 250.1556, found 250.1538.
2-(Pyrrolidin-1-ylmethyl)-oxazole-4-carboxylic acid po-
tassium salt, 29. Maleate salt 28 (20 g, 0.059mol) was
slurried in CH2Cl2 (100 mL) at 20 °C under nitrogen.
Aqueous K2CO3 (100 mL, 10% w/v) was added dropwise
over ∼20 min. (Caution! CO2 evolution.) The CH2Cl2 layer
was separated and washed with water (100 mL) and then
concentrated at atmospheric pressure to 40 mL. The solution
was cooled to ∼30 °C. IMS (100 mL) was added and
distillation continued at atmsopheric pressure to ∼40 mL.
The solution was cooled to 20 °C. Potassium hydroxide
pellets (4.3 g, 0.065 mol, 85% w/w) were dissolved in IMS
(40 mL). The potassium hydroxide solution was added to
the solution of free base ester over ∼15 min and the solution
allowed to stir at ∼20 °C overnight. The solution was
concentrated to ∼40 mL. Acetone (60 mL) was added and
the slurry stirred at 0 °C for 30 min and then filtered off at
suction. The solid was washed with chilled (0 °C) acetone
(40 mL) and dried at ∼50 °C in vacuo overnight to give
potassium salt 29 (12.41 g, 90%) as a white powder. IR
Acknowledgment
We are grateful to Athina Chatzi, Claire Greaney, Anne
Hubert, Chris Jones, Sean Lynn, Nisha Mistry, Andrew Ray,
Alec Simpson, and Dave Sugden for analytical support and
Karen Preston for process safety evaluation studies.
Supporting Information Available
Crystal structure information for 12; proton and carbon
NMR for compounds 10-17, 20-22, 24, 28-30. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
(Nujol mull) 2790, 2853, 2924 cm-1; H NMR (400 MHz,
d4-MeOH) δH 1.81-1.84 (m, 4H), 2.64-2.67 (m, 4H), 3.81
(s, 2H), 8.11 (s, 1H); 13C NMR (100 MHz, d4-MeOH) δC
23.23, 51.17, 53.69, 139.34, 141.87, 161.33, 167.48; MS
calcd for C9H13N2O3 (MH+) 197.0926, found 197.0914. Anal.
Calcd for C9H11N2O3K: C, 46.14; H, 4.73; N, 11.96.
Found: C, 46.12; H, 4.47; N, 11.68.
Received for review August 1, 2000.
OP000086G
44
•
Vol. 5, No. 1, 2001 / Organic Process Research & Development