Pyrrolidine-5,5-trans-lactams 3. Alternative regiochemical outcome of the acyl-iminium coupling reaction

Article abstract of DOI:10.1055/s-2003-40879

Enantio- and regioselective syntheses of the pyrrolidine 5,5-trans-lactams benzyl (3aS,6aR)-6,6-dimethyl-5-oxohexahydropyrrolo[3,2-b]pyrrole-1(2H)-carboxylate and benzyl (3aS,6aR)-6S-ethyl-5-oxohexahydropyrrolo[3,2-b]pyrrole-1(2H)-carboxylate from a common intermediate 2-ethoxy-3S-(2,2,2-trifluoro-acetylamino)-pyrrolidine-1-carboxylic acid benzyl ester are reported. The key step in both syntheses is the acyl iminium ion mediated condensation of the pyrrolidine with a ketene acetal.

Full text of DOI:10.1055/s-2003-40879

1722
PAPER
Pyrrolidine-5,5-trans-lactams 3. Alternative Regiochemical Outcome of the
Acyl-Iminium Coupling Reaction.
P
yrrolidine-5,5-tra
a
ns-lactam
s
v
3
id M. Andrews,* Alan D. Borthwick, Helene Chaignot, Paul S. Jones, J. Ed Robinson, Pritom Shah,
Martin J. Slater, Richard J. Upton
GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Fax +44(1438)763620; E-mail: david.m.andrews@astrazeneca.com
Received 9 May 2003
This paper is dedicated to the memory of Helene Chaignot, who died on the 13th May 2003.
Abstract: Enantio- and regioselective syntheses of the pyrrolidine
5,5-trans-lactams benzyl (3aS,6aR)-6,6-dimethyl-5-oxohexahydro-
pyrrolo[3,2-b]pyrrole-1(2H)-carboxylate and benzyl (3aS,6aR)-
6S-ethyl-5-oxohexahydropyrrolo[3,2-b]pyrrole-1(2H)-carboxylate
from a common intermediate 2-ethoxy-3S-(2,2,2-trifluoro-acety-
lamino)-pyrrolidine-1-carboxylic acid benzyl ester are reported.
The key step in both syntheses is the acyl iminium ion mediated
condensation of the pyrrolidine with a ketene acetal.
Key words: antiviral agents, lactams, neighboring-group effects,
regioselectivity, steric hindrance
The synthesis of the ethyl-substituted pyrrolidine-5,5-
trans-lactams 5 and 6 and their activity as inhibitors of
hepatitis C virus (HCV) NS3/4A protease has been recent-
ly reported (see Scheme 1).1
The initial synthesis was based upon the application of the
flexible acyliminium ion methodology developed by
Macdonald et al.2,3 When tested in a biochemical assay
measuring inhibition of the HCV NS3/4A protease, the
(3aS,6aR,6S) compound 5 was found to be two- to four-
fold more potent than the alternate diastereomer 6. Dia-
stereomer 5 and analogues thereof thus became the focus
of our synthetic efforts. Under the initial reaction condi-
tions explored in the production of 2, no stereoinduction
was seen for the introduction of the ethyl group at position
6, and so the route was a very inefficient means of produc-
ing the desired single isomer 5. In particular, the two dias-
tereomers produced in the condensation step could only
be efficiently separated by first derivatising the lactam 4
with the tert-butyldimethylsilyl group. The silyl lactams
so produced were completely separated by Biotage chro-
matography (cyclohexane–EtOAc, 6:1). Fluoride depro-
tection using tetrabutyl ammonium fluoride buffered with
acetic acid, yielded the separated isomers of 4 in less than
10% overall yield from the precursor 1. This was in
marked contrast to the synthesis of the elastase inhibitor
intermediate 7 (Figure 1), which was produced by analo-
gous methodology from the enantiomer of 1 in 44% over-
all yield.2 The difference in yield was attributable to three
main factors: no stereoinduction for the introduction of
the ethyl side chain in 2; the poor isolated yield of 3; and
Scheme 1 (i) BF3·OEt2, CH2Cl2; (ii) K2CO3, EtOH, 37% from 1;
(iii) t-BuMgCl, THF; (iv) Multiple steps as described in ref.1
the cumbersome silylation/desilylation step to separate
the isomers of 4.
Closer inspection revealed that the most inefficient step in
the reaction sequence was the coupling of the acylimini-
um precursor 1 with the ketene acetal reagent 10a ([1-
ethoxy-but-1-enyloxy]-trimethylsilane). Work-up and
analysis of the potassium carbonate hydrolysis step indi-
cated that in addition to the desired product 3 (formed in
50% theoretical yield and as a 1:1 mixture of diastere-
omers), an isomeric oxazolidine 8a (Figure 2) was formed
in 40% yield – also as a potential mixture of diastereo-
mers.
Synthesis 2003, No. 11, Print: 05 08 2003.
Art Id.1437-210X,E;2003,0,11,1722,1726,ftx,en;C03003SS.pdf.
© Georg Thieme Verlag Stuttgart · New York

PAPER
Pyrrolidine-5,5-trans-lactams
1723
diastereoselectivity in this reaction at the position of the
ethyl substituent.
Figure 1
Scheme 2 Regiochemical outcome observed with hindered and un-
hindered silyl ketene acetals.
The alkylation of the unsubstituted trans-lactam template
12 has been previously described,5 and we postulated that
by accessing 12 via the unsubstituted tert-butyldimethyl-
silyl protected ketene acetal 10b and the acyl iminium
methodology described herein, we would be able to syn-
thesize the desired ‘SRS’ alkylated product in good yield
(Scheme 3). An important advantage is that this route uti-
lizes stereoselective alkylation for the introduction of the
ethyl side-chain and avoids the complication of the forma-
tion of a mixture of diastereomers in the iminium ion con-
densation.
Figure 2 Oxazolidine 8a. Key HMBC’s connectivities are marked
with arrows. In addition, strong NOE’s are observed between the pro-
tons at C4 and C5; a coupling constant of 5.5 Hz is also observed for
these protons, providing additional confirmatory evidence. Oxazoli-
dine 8b. NOE’s were observed between the bridgehead carbons (po-
sitions C4 and C5) and the methylene of the sidechain (marked
position 1), confirming the configuration of the CF3 substituent.
We postulated that the oxazolidine 8a arises as a conse-
quence of the ketene acetal reacting between the nitrogen
and oxygen atoms of the oxazoline ring, as shown in
Scheme 2. Condensation of a trifluoromethyl ketone es-
ter with aminoethanol yields the corresponding monocy-
clic oxazolidine system,4 but this is the first report of the
synthesis of the bicyclic system as a consequence of imi-
nium ion-mediated condensation. The 1H NMR spectrum
of 8a was complex due to the presence of rotamers at
room temperature, but these coalesced at 100 °C into a
single, major set of peaks. It appeared that the major prod-
uct was a single diastereoisomer, but the configuration of
the CF3 group could not be directly established due to the
absence of diagnostic NOE’s. A related minor diastereo-
meric component (ca. 10%) was also present, most likely
the opposite stereoisomer at the ethyl sidechain. The con-
figuration of the CF3 group in this diastereoisomer could
also not be established due to the absence of diagnostic
NOE’s. We postulated, therefore, that the ketene acetal
would most likely approach the oxazoline from the least
hindered alpha face, with the result that in both the major
and minor isomers, the CF3 substituent adopts a config-
uration (c.f. 8b, vide infra). The potential 10:1 diastereo-
meric NMR ratio for 8a implies further
Scheme 3 (i) BF3·OEt2, CH2Cl2; (ii) Multiple steps as described;
(iii) LHMDS (1.1 equiv)/THF –78 °C, then EtI (2 equiv).
Disappointingly, the isolated product of initial experi-
ments was shown to be exclusively the oxazolidine 8b
(Scheme 2). Spectrum doubling at room temperature
Synthesis 2003, No. 11, 1722–1726 © Thieme Stuttgart · New York

1724
D. M. Andrews et al.
PAPER
again hampered characterization of 8b by NMR, but at over two steps from 1 (Scheme 2 and Table 1). Com-
100 °C in DMSO-d6, spectra coalesced into a single set of pound 9c was converted by standard methodology to the
sharp peaks. On this occasion, HMBC and NOE studies
, -dimethyl trans-lactam 13. The comparison of overall
provided complete confirmation of relative stereochemis- yield dramatically illustrates the gain in synthetic efficien-
try, confirming that the ketene acetal did indeed approach cy derived by working in the disubstituted series; the -
the oxazoline from the least-hindered face, giving rise to ethyl diastereomer 5 was produced in less than 9% overall
the single diastereoisomer 8b (Figure 2).
yield from 1 c.f. the dimethyl analogue 13, furnished in
74% overall yield (Table 2)
We attempted to draw upon analogous experience with
the substituted trans-lactams, which indicated that it Compounds were tested in a biochemical hepatitis C NS3/
might be possible to influence the regiochemical outcome 4A protease assay.7 Further to the greatly enhanced syn-
of the reaction by replacing dichloromethane with aceto- thetic accessibility, the initial hypothesis that disubstitu-
nitrile as solvent. Some improvement was noted by HPLC tion would be tolerated biologically was amply confirmed
reaction monitoring (in that the reaction proceeded to a when the potency of 5, 6, and 13 were compared
1:1 mixture of 11 and 8b), but as this tactic did not greatly (Scheme 4 and Table 2).
influence reaction outcome in the case of the ethyl-substi-
This interesting chemical problem – and the solutions ex-
tuted compound, this was not pursued further (see
plored – illustrate a useful stratagem available to the me-
Table 1).
dicinal chemist (but not often exploitable by the natural
products chemist): on some occasions, difficulties in syn-
thesis can be obviated by the removal or modification of
those moieties which present current methodology with
particular challenges.
Table 1 Influence of Solvent and Substituents on the Yield and Re-
giochemistry
R1
R2
Solvent
Aminoester
Oxazoline
Isolated Yield (%) Isolated Yield (%)
Et
Et
H
H
CH2Cl2
MeCN
CH2Cl2
MeCN
CH2Cl2
51
51
0
39
21
43
H
H
H
H
1:1a
90
Me
Me
0
a HPLC ratio of products – products not isolated.
Observing that the hindered isobutyryl-derived silyl
ketene acetal gave rise to excellent yields en route to the
isopropyl trans-lactam 7, we postulated that reaction out-
come was determined by steric bulk, as indicated in
Scheme 2. The poor results demonstrated with the unsub-
stituted ketene acetal are entirely consistent with this hy-
pothesis. It thus became apparent that the medicinal
chemical search for an optimal P1 substituent either need-
ed to be refocused away from the problematic ethyl sub-
stituent, or that the acyl iminium methodology needed to
be dramatically modified, or abandoned.
Modeling studies (based upon crystal structures of ethyl
trans-lactams soaked into NS3 protease) and our knowl-
edge of the relative potency of diastereomers 5 and 6 sug-
gested that a very attractive target for synthesis was the
Scheme 4
Table 2 Comparison of the Potency of 5, 6, and 13
, -dimethyl trans-lactam.6 From a synthetic perspective, Compound Yielda
the advantages of working with the corresponding dime-
IC50 ( M, 4 h preincubation) HCV protease
biochemical assay
thyl silyl ketene acetal 10c were two-fold: firstly, we pos-
5
6
9%
8%
30
99
34
tulated that since 10c was the most sterically-hindered
ketene acetal examined to date, it would give rise to the
most favorable regiochemical outcome; secondly, since
the newly introduced acetic acid ethyl ester side-chain is
, -dimethyl substituted, no diastereomer problem exists.
Gratifyingly, 10c did indeed perform well in the conden-
sation step, and 9c was produced in 90% isolated yield
13
74%
a Yield over multiple steps 1 to 5, 6, and 13.
Synthesis 2003, No. 11, 1722–1726 © Thieme Stuttgart · New York

PAPER
Pyrrolidine-5,5-trans-lactams
1725
Benzyl (3aS,6aS)-2-(2-Ethoxy-2-oxoethyl)-2-(trifluoromethyl)-
hexahydro-4H-pyrrolo[3,2-d][1,3]oxazole-4-carboxylate (8b)
A mixture of 2-ethoxy-3S-(2,2,2-trifluoro-acetylamino)-pyrroli-
dine-1-carboxylic acid benzyl ester (1) (1.0 g, 2.89 mmol) and [(1-
ethoxyvinyl)oxy](tert-butyldimethyl)silane (10b) (1.75 g, 8.65
mmol) in CH2Cl2 (10 mL) was cooled to 5 °C (ice bath) and treated
with BF3·OEt2 (814 mg, 5.74 mmol, 705 L). After 1 h more [(1-
ethoxyvinyl)oxy](tert-butyldimethyl)silane (10b) (583 mg, 2.89
mmol) was added and the reaction allowed to stir for a further 3 h at
r.t. The mixture was quenched with sat. aq NaHCO3 solution (5 mL)
and the organic layer separated using a phase separation cartridge.
Removal of the solvent and silica gel chromatography of the residue
(eluent, CH2Cl2–Et2O; 25:1) yielded 8b as a colorless oil (501mg,
43%).
1H NMR (DMSO-d6, 100 °C): = 7.34–7.30 (m, 5 H, C6H5), 5.19–
5.11 (m, 2 H, C6H5CH2), 5.92 [m, 1 H, OCONCH(O)CH], 4.40 (m,
1 H, OCONCHCH), 4.08 [br m, 1 H, CHNHC(CF3)CH2], 4.20–4.10
(m, 2 H OCH2CH3), 3.52 (m, 1 H, OCONCHHCH2), 3.46 [m, 1 H,
OCONCHHCH2], 2.80 [s, 2 H OCCH2(CF3)COO], 1.92 (m, 1 H,
CONCH2CHHCH), 1.83 (m, 1 H, CONCH2CHHCH), 1.24 (t, J = 7
Hz, 3 H, OCH2CH3).
NMR spectra (1H 400MHz, 13C 100MHz) were recorded on a Bruk-
er DRX400, Bruker DPX400 or Varian Inova400 spectrometer.
Chemical shifts ( ) are reported in ppm and coupling constants (J)
in Hz. In CDCl3 chemical shifts were referenced using TMS as an
internal standard or in DMSO-d6 using the residual solvent signal
(1H: = 2.50ppm, 13C: = 39.5ppm). NMR Spectra were recorded
at room temperature unless otherwise noted. HPLC analysis was
achieved using a Hewlett Packard Series 1050 with a Phenomenex
Prodigy 5 OD5-2 column (150 × 4.6mm). The mobile phase was A
(H2O + 0.1% TFA) and B (MeCN + 0.05% TFA) used as a linear
gradient of 15–95% of B over 14 min with a flow rate of 1.5 mL/
min; detection was at 215nm. Elution times are quoted as tR in min.
Values are 0.2 min. Thermospray mass spectra were recorded on
an HP5989B Engine, using aqueous ammonium acetate as solvent,
with the filament in positive ion mode. Accurate positive ion mass
spectra were acquired as accurate mass centroided data using a Mi-
cromass Q-Tof 2 hybrid quadrupole time-of-flight mass spectrome-
ter, equipped with a Z-spray interface, over a mass range of 80–
1200 Da, with a scan time of 0.95 s and an interscan delay of 0.07
s. Reserpine was used as the external mass calibrant ([M +
H]+ = 609.2812 Da). The Q-Tof 2 mass spectrometer was operated
in W reflectron mode to give a resolution (FWHM) of 16000-
20000. Ionization was achieved with a spray voltage of 3 kV, a cone
voltage of 30 V, with cone and desolvation gas flows of 5–10 and
500 L/min respectively. The source block and desolvation tempera-
tures were maintained at 120 °C and 300 °C respectively. The ele-
mental composition was calculated using MassLynx v3.5 for the
[M + H]+ and the mass error quoted as ppm. Syntheses of 1–6 and
13 have been previously reported.1 Compound 10b was prepared as
previously reported.8 Compound 10c was purchased from Aldrich.
All compounds are racemic unless otherwise indicated.
13C NMR (DMSO-d6, 100 °C):
= 13.3 (OCH2CH3), 30.1
(ONCHCH2CO), 30.7 (CONCH2CH2CH), 38.1 [OCCH2(CF3)CO],
43.6 (OCONCH2CH2), 59.9 (OCH2CH3), 60.8 (OCONCHCH),
65.6 (C6H5CH2), 92.0 [OCONCH(O)CH], 94.7 (2J19F-13C = 30.5 Hz,
CCF3), 123.5 (1J19F-13C = 287Hz, CF3), 126.6, 127.1, 127.7, 136.4
(C6H5), 153.1 (OCON), 167.3 (CO2CH2CH3).
MS: m/z = 420 (M + NH4)+.
MS (thermospray): m/z = 420 (M + NH4)+.
MS-ES: m/z calcd for C18H21F3N2O5 (MH+), 403.1463; found
403.1476.
Benzyl (3aS,6aS)-2-[1-(Ethoxycarbonyl)propyl]-2-(trifluoro-
methyl)hexahydro-4H-pyrrolo[3,2-d][1,3]oxazole-4-carboxy-
late (8a)
Benzyl (2R,3S)-3-Amino-2-(2-ethoxy-1,1-dimethyl-2-oxoeth-
yl)pyrrolidine-1-carboxylate (9c)
Prepared as reported previously7 as a yellow oil, 8.48 g, 65%Th.
To a solution of 1 (5.2 g, 14.4 mmol) in anhyd CH2Cl2 (160 mL) at
0 °C under nitrogen was added 10a (5 mL, 14.4 mmol) dropwise,
followed by BF3·OEt2 (8.8 mL, 72 mmol). The mixture was stirred
at 0 °C for 8 h and then at r.t. overnight. The reaction was quenched
by the addition of HCl (1 M, 160 mL) and extracted with CH2Cl2.
The organic layer was washed successively with NaHCO3 solution
(100 mL) and brine (100 mL), dried over MgSO4 and the solvent
was evaporated to afford a mixture of 2 and 8a as a yellow oil (6.18
g). The mixture was dissolved in EtOH (56 mL), a solution of
K2CO3 (13.54 g, 98 mmol) in H2O (110 mL) was added and the so-
lution heated at 60 °C for 3 h. The EtOH was removed by evapora-
tion in vacuo, Et2O was added to the resulting orange oil, and the
ethereal layer separated. The etheral fraction was washed with HCl
solution (1 M), dried, and evaporated to afford 8a as a yellow oil
(1.84g, 30%).
1H NMR (CDCl3): = 7.35 (m, 5 H, C6H5), 5.10 (s, 2 H, C6H5CH2),
3.80–4.00 (m, 2 H, NHCHCH2CH2), 3.65 (s, 3 H, CO2CH3), 3.45
(m, 1 H, NH2CH), 3.35 (m, 1 H, NCHCMe2) 2.07 (m, 1 H,
NCH2CHH), 1.60 (m, 1 H, NCH2CHH) 1.15–1.25 [br s, 6 H,
C(CH3)2].
MS (ES, Pos.): m/z = 321 (MH+).
{[(1E)-1-Ethoxybut-1-enyl]oxy}(trimethyl)silane (10a)
To a solution of i-Pr2NH (7.7 mL, 55 mmol) in anhyd THF (50 mL)
at 0 °C was added n-BuLi (55 mmol) and the mixture was stirred for
10 min and then cooled to –78 °C, ethyl butyrate (6.6 mL, 50 mmol)
in anhyd THF (50 mL) was slowly added. The mixture was stirred
at –78 °C for 30 min, chlorotrimethylsilane (6.34 mL, 55 mmol) in
anhyd THF (50 mL) was added and the temperature of the mixture
allowed to return to r.t. and stirred for 1 h. The mixture was
quenched using sat. NaHCO3 solution (50 mL) and the mixture ex-
tracted with hexanes, washed with H2O (100 mL), brine, dried
(MgSO4) and concentrated, then distilled in vacuo to yield a color-
less oil (6.4g, 68%) bp.98 °C at ca. 10 mm.
1H NMR (DMSO-d6,100 °C): = 7.28–7.40 (m, 5 H, C6H5), 5.81 (d,
J = 5.5 Hz, 1 H, OCONCHOC), 5.16, 5.10 (2 d, J = 13 Hz, 2 H,
C6H5CH2), 4.23 (m, 1 H, NHCHCH2), 4.08–4.16 (m, 2 H,
OCH2CH3), 3.51 (m, 1 H, OCONCHHCH2) 3.44 (m, 1 H,
OCONCHHCH2), 2.74 [dd, J = 11.5 Hz, 3.5Hz,
1
H,
CH(C2H5)CO], 1.91 (m, 1 H, OCONCH2CHH), 1.83 (m, 1 H,
OCONCH2CHH), 1.73 [m, 1 H, CH(CHHCH3)], 1.67 [m, 1 H,
CH(CHHCH3)], 1.20 (t, J = 7 Hz, 3 H, OCH2CH3), 0.86 [t, J = 7 Hz,
3 H, CH(CH2CH3)].
1H NMR (CDCl3): = 3.83 (q, J = 7.3 Hz, 2 H, OCH2CH3), 3.75 (t,
J = 7.3Hz, 1 H, CCHCH2), 1.96 (dt, J = 7.3 Hz, 2 H, CH2CH3), 1.22
(t, J = 7.3 Hz, 3 H, OCH2CH3), 0.92 (t, J = 7.3 Hz, 3 H, CH2CH3),
0.22 [s, 9H, Si(CH3)3].
13C NMR (DMSO-d6, 80 °C):
= 10.5 [CH(CH2CH3)], 12.5
(OCH2CH3), 30.0 (OCONCH2CH2), 43.0 (OCONCH2CH2), 43.0
[CH(CH2CH3)], 51.0 (NHCHCH2), 51.0 [CH(C2H5)CO], 59.5
(OCH2CH3), 65.0 (C6H5CH2), 91.5 (OCONCHOC), 97.5 (C(CF3),
124.5 (1J19F-13C = 287 Hz, CF3), 125.5–127.0, 137.5 (C6H5), 154.0
(OCON), 171.0 (CO2CH2CH3).
Benzyl (2S,3R)-2-(2-Ethoxy-2-oxoethyl)-3-[(trifluoroacetyl)-
amino]pyrrolidine-1-carboxylate (11)
A reference sample of 11 (to allow the product ratio of 8b:11 to be
quantified by HPLC) was synthesized as follows. Benzyl (2S,3R)-
3-[(tert-butoxycarbonyl)amino]-2-(2-ethoxy-2-oxoethyl)pyrroli-
Synthesis 2003, No. 11, 1722–1726 © Thieme Stuttgart · New York

1726
D. M. Andrews et al.
PAPER
dine-1-carboxylate,5 (500 mg, 1.23 mmol) was dissolved in CH2Cl2
(2 mL) and treated with trifluoroacetic acid (2 mL). The reaction
mixture was left to stand at r.t. overnight. Trifluoroacetic anhydride
(2 mL) was added and the reaction mixture left to stand at r.t. over-
night. The volatiles were removed in vacuo and the residue dis-
solved in CH2Cl2 (10 mL). Sat. Na2CO3 solution (10 mL) was added
and the mixture stirred for 5 min. The organic phase was separated
using a phase separation cartridge and concentrated to yield 11 as a
colorless oil (437 mg, 88%).
1H NMR (DMSO-d6, 100 °C): = 9.30 (br m, 1 H, NHCOCF3),
7.28–7.30 (m, 5 H, C6H5), 5.15–5.07 (m, 2 H, C6H5CH2), 4.30 (m,
1 H, OCONCHCHCH2), 4.11 (m, 1 H, OCONCHCH), 4.06 (q,
J = 7 Hz, 2 H, OCH2CH3), 3.60 (m, 1 H, OCONCHHCH2), 3.41 (m,
1 H, OCONCHHCH2), 2.72 (m, 1 H, CONCHCHHCO), 2.57 (m, 1
H, CONCHCHHCO), 2.22 (m, 1 H, CONCH2CHHCH), 1.88 (m, 1
H, CONCH2CHHCH), 1.18 (t, J = 7 Hz, 3 H, OCH2CH3).
References
(1) Slater, M. J.; Andrews, D. M.; Baker, G.; Bethell, S. S.;
Carey, S.; Chaignot, H.; Clarke, B.; Coomber, B.; Ellis, M.;
Good, A.; Gray, N.; Hardy, G.; Jones, P.; Mills, G.;
Robinson, E. Bioorg. Med. Chem. Lett. 2002, 12, 3359.
(2) 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.
(3) For recent reviews, see: (a) Speckamp, W. N.; Moolenaar,
M. J. Tetrahedron 2000, 56, 3817. (b) Hiemstra, H.;
Speckamp, W. N. In Comprehensive Organic Synthesis,
Vol. 2; Trost, B. M.; Fleming, I., Eds.; Pergamon: Oxford,
1991, 1047–1082.
(4) Slusarczuk, G. M. J.; Joullie, M. M. J. Org. Chem. 1971, 36,
37.
(5) Borthwick, A. D.; Crame, A. J.; Davies, D. E.; Exall, A. M.;
Jackson, D. L.; Mason, A. M.; Pennell, A.; M, K.;
Weingarten, G. G. Synlett 2000, 504.
(6) Andrews, D. M.; Chaignot, H.; Coomber, B. A.; Good, A.
C.; Hind, S. L.; Johnson, M. R.; Jones, P. S.; Mills, G.;
Robinson, J. E.; Skarzynski, T.; Slater, M. J.; Somers, D.
O’N. Org. Lett. 2002, 4, 4479.
(7) Andrews, D. M.; Carey, S. J.; Chaignot, H.; Coomber, B. A.;
Gray, N. M.; Hind, S. L.; Jones, P. S.; Mills, G.; Robinson,
J. E.; Slater, M. J. Org. Lett. 2002, 4, 4475.
13C NMR (DMSO-d6, 100 °C):
= 13.4 (OCH2CH3), 28.2
(CONCH2CH2CH), 37.0 (CONCHCH2CO), 44.0 (OCONCH2CH2),
54.2 (OCONCHCHCH2), 58.8 (OCONCHCH), 59.4 (OCH2CH3),
65.6 (C6H5CH2), CCF3 115.2 (1J19F-13C = 288Hz, CF3), 126.7,
127.1, 127.6, 136.3 (C6H5), 153.1 (OCON), 155.3 COCF3
(2J19F-13C = 36Hz, CF3), 169.3 (CO2CH2CH3).
MS: m/z = 403 (M + H)+.
MS (thermospray): m/z = 403 (M + H)+.
(8) Mikami, K.; Matsumoto, S.; Ishida, A.; Takamuku, S.;
Suenobu, T.; Fukuzumi, S. J. Am. Chem. Soc. 1995, 117,
11134.
Acknowledgment
We thank Drs. Graham Baker, Sue Bethell and Malcolm Ellis for
provision of NS3 protease protein and initial assay systems; and
Derek Evans for provision of intermediates.
Synthesis 2003, No. 11, 1722–1726 © Thieme Stuttgart · New York