Second-generation process research towards eletriptan: A fischer indole approach

Article abstract of DOI:10.1021/op100251q

The development of a second-generation process for the synthesis of eletriptan via a Fischer indole cyclisation is described. The finalised process offers several potential advantages over the current route of manufacture including cost, throughput, and safety.

Full text of DOI:10.1021/op100251q

Organic Process Research & Development 2011, 15, 98–103
Second-Generation Process Research Towards Eletriptan: A Fischer Indole Approach
Christopher P. Ashcroft,*,† Paul Hellier,‡ Alan Pettman,† and Simon Watkinson†
Chemical Research and DeVelopment, Pfizer Global Research and DeVelopment, Sandwich, Kent CT13 9NJ, United Kingdom,
and Process DeVelopment Centre, Pfizer Global Manufacturing, Loughbeg API Site, Loughbeg, Cork, Ireland
Abstract:
aldehyde 8 or a suitably protected form would be required,
together with the hydrazine 10. Formation of 8 or 9 and indeed
10 however, proved to be quite challenging. The first successful
synthesis (scheme 3) of 9 was via lithiation of N-methylpyrolle5
11 and subsequent alkylation with 2-(2-bromoethyl)-1,3-diox-
olane, followed by hydrogenation of the pyrrole to yield the
racemic pyrrolidine 9a.
The development of a second-generation process for the synthesis
of eletriptan via a Fischer indole cyclisation is described. The
finalised process offers several potential advantages over the
current route of manufacture including cost, throughput, and
safety.
Eletriptan (R)-7 belongs to the class of drugs known as
triptans and was approved in the United States in December
2002. It is a potent and selective 5-HT agonist and is marketed
as Relpax for the treatment of migraine. It is currently a low
volume product; however, at the time the research was started,
the development of novel formulations and drug delivery
systems could have significantly increased bulk demand for the
drug. The current manufacturing route1 to eletriptan 7 is well
developed and robust, but does suffer some limitations (scheme
1). The key starting material 5 is extremely expensive and
contributes approximately 50% of the total cost of manufacture.
It uses the costly, unnatural isomer of proline as well as the
highly noxious and sensitizing reagent phenyl vinyl sulfone.
The synthesis also incorporates a lithium aluminium hydride
reduction, generating large waste aqueous streams. Thus an
alternative route of manufacture was sought that would be able
to deliver this increase in bulk demand.
During the assessment of alternative routes to eletriptan, the
Fischer indole synthesis was highlighted as a particularly
attractive way to build up the molecule. It had the potential to
deliver a convergent, highly efficient synthesis (scheme 2), as
well as being a well understood and reliable transformation.
The Fischer indole reaction was discovered in 1883 by Hermann
Emil Fischer,2 and despite its age, is still one of the most
commonly used methods to synthesise indole containing
molecules. Indeed several other members of the Triptan family
of drugs are manufactured via this method.3 If this approach
were successful, then many of the issues associated with the
current route of manufacture could be avoided.
This synthesis was not amenable to further scale up, due to
the low yields, cryogenic reaction conditions and chromatog-
raphy. However sufficient material was isolated to validate the
use of 9a in the Fischer indole reaction by condensation with
commercially available 4-bromophenyl hydrazine. This deliv-
ered rac-5, a racemic form of an intermediate in the original
synthesis, in good yield.
The need for a more robust and scalable synthesis, led to
the development of two further routes. The first starts from
4-methylaminobutyric acid 13 which is protected as the benzyl
carbamate and transformed to the Weinreb amide 14. Addition
of the Grignard reagent formed from 2-(2-bromoethyl)-1,3-
dioxane 15, followed by deprotection with concomitant cycli-
sation and imine reduction gave a high yielding, robust four-
step synthesis of compound 9b (Scheme 4). The acetal of choice
was changed from the five-membered 1,3-dioxolane to the six-
membered 1,3-dioxane due to the improved stability of the
respective Grignard reagent at the required reaction temperature.
This route was reproducible, reliable, and very easily scaled,
but suffers from the use of expensive reagents (all intermediates
being oils) and the need for a protecting group. However a 50-g
batch of the fumarate salt of 9b was easily made using this
chemistry.
The third successful synthesis of racemic pyrrolidine acetal
9 starts from N-methylpyrrole-2-carboxaldehyde 20 (scheme
5). This undergoes a key Wittig olefination6 with a phosphorous
ylide derived from bromoacetaldeyde, followed by global
hydrogenation and isolation as the fumarate salt. The acetal has
again been changed to the 5,5-dimethyl-1,3-dioxane as this gives
a crystalline intermediate 20 that was less prone to hydrolysis
than either the unsubstituted dioxane or dioxolane acetals. Wittig
reaction of Aldehyde 19 with the triphenylphosphonium ylide
derived from 18a failed to give any conversion to the olefin
20. However, switching to the trinbutylphosphonium ylide 18b
afforded complete conversion to olefin 20. Since the trin-
butylphosphine oxide byproduct is partially water-soluble, it can
be removed with multiple water washes of the product solution.
Despite the increased stability of the dimethyldioxane acetal,
olefin 20 still hydrolyses to the corresponding aldehyde upon
The Fischer reaction proceeds via condensation of an aryl
hydrazine with a carbonyl compound, followed by a 1,3-
sigmatropic rearrangement and subsequent elimination of am-
monia.4 In order to prove that the Fischer reaction was viable
for the preparation of eletriptan (R)-7, the synthesis of either
* christopher.ashcroft@pfizer.com.
† Pfizer Global Research and Development, U.K.
‡ Pfizer Global Manufacturing, Ireland.
(1) Ogilvie, R. J. New Process for the Preparation of Anti-migraine Drug,
Eletriptan; WO/2002/050063, 2002.
(2) Fischer, E.; Jourdan, F. Chem. Ber. 1883, 16, 2241.
(3) Li, J.-J.; Johnson, D. S.; Roth, B. D.; Sliskovic, D. R. Contemporary
Drug Syntheses; John Wiley and Sons: Chichester, 2004.
(4) Robinson, B. The Fischer Indole Synthesis; John Wiley and Sons:
Chichester, 1982.
(5) Brittain, J. M.; Jones, R. A.; Arques, J. S.; Saliente, T. A. Synth.
Commun. 1982, 12, 231.
(6) Spangler, C. W.; McCoy, R. K. Synth. Commun. 1988, 18, 51.
98
•
Vol. 15, No. 1, 2011 / Organic Process Research & Development
10.1021/op100251q  2011 American Chemical Society
Published on Web 12/01/2010

Scheme 1. Current manufacturing route to eletriptan (R)-7a
a Reagents and conditions: (a) EtMgBr, Et2O. (b) 3, DCM, 50% from 1. (c) LiAlH4, THF, 72%. (d) Ac2O, TEA, DMF. (e) Phenyl vinyl sulfone (PVS), Pd(OAc)2,
P(°Tol)3, TEA, DMF, 80% from 5. (f) H2, Pd/C, MeSO3H, acetone, 95%. (g) K2CO3, MeOH, 92%. (h) HBr, acetone 73%.
Scheme 2. Fischer indole approach to eletriptan (R)-7
Scheme 4. Improved synthesis of racemic pyrrolidine acetal
9ba
Scheme 3. First synthesis of racemic pyrrolidine acetal 9aa
a Reagents and conditions: (a) CbzCl, KOH, water, toluene. (b) CDI, TEA,
DCM, HN(OMe)Me · HCl, 82% from 13. (c) THF, 60 °C, 85%. (d) H2, Pd/C,
MeOH, 60 psi, 50 °C. (e) Fumaric acid, MeOH, EtOAc, 78% from 16.
synthesized in two steps from commercially available 4-nitro-
phenethyl bromide 21 (Scheme 6). However, conversion to the
hydrazine 10 via the diazonium salt 23 was unsuccessful. It
was clear from LC/MS analysis that the diazotisation was
working efficiently and that the problem lay with the stability
of the free hydrazine 10. Even the hydrochloride salt of the
hydrazine 10 degraded rapidly at ambient temperature. The best
results were obtained using tin(II) chloride as reductant and
isolating 10 as its hydrochloride salt, with storage of the material
at -20 °C. This yieded a meagre 13% of 10.
a Reagents and conditions: (a) nBuLi, TMEDA, hexane, 50 °C. (b) 2-(2-
Bromoethyl)-1,3-dioxolane, THF, -60 °C 25% from 11. (c) H2, Rh/C, 100 psi,
100 °C, 100%. (d) 4-Bromophenyl hydrazine hydrochloride, H2SO4 (4% w/v),
75%.
prolonged exposure to water and as a result, the best isolated
yield of 20 from 18b was only 35%. A significantly improved
procedure was developed using the triethylphosphonium ylide
derived from 18c. Again the conversion to olefin 20 is high,
but the triethylphosphine oxide byproduct is highly water-
soluble and so can be completely removed with a single water
wash. This minimizes the contact of the product with water
and a significantly improved yield is attained (77%). Trieth-
ylphosphine is extremely pyrophoric and difficult to handle, but
can be conveniently purchased as a solution in THF which is
much easier and safer to handle. The different pyrrolidine acetals
9a,b and c were shown to react identically under Fischer
reaction conditions.
It was at this point that a recent patent7 was found, detailing
the reduction of diazonium salts with L-ascorbic acid (vitamin
C). Of particular interest was the potential to form a stable,
isolable oxalyl hydrazide intermediate 24, which could be
cleaved to the free hydrazine on treatment with acid. It was
hoped that we could utilize 24 as a hydrazine equivalent in the
(7) Lambert, J. F.; Norris, T. Preparation of N-(5-cyclopropyl-1-quinolin-
5-yl-1H-pyrazole-4-carbonyl)guanidine and other sodium-hydrogen
exchanger type-1 (NHE-1) inhibitors, with improved product color, via
an improved preparation of arylhydrazines using ascorbic acid reduction;
WO/2002/044133, 2002.
With racemic 9 in hand, attention switched to the synthesis
of aryl hydrazine 10. The required aniline 22 was readily
Vol. 15, No. 1, 2011 / Organic Process Research & Development
•
99

Scheme 5. Alternative synthesis of racemic pyrollidine
Scheme 8. Resolution approach to give the desired (R)-7a
acetal 9ca
a Reagents and conditions: (a) Neopentyl glycol, H2SO4, toluene, 90%. (b)
n
PR3, MeCN, 100%. (c) NaOEt, DMF, EtOH R ) Ph ) 0%; R ) Bu ) 35%;
R ) Et ) 77%. (d) H2, Rh/C, EtOH, 100 psi, 70 °C. (e) Fumaric acid, MeOH,
EtOAc, 95% from 20.
a Reagents and conditions: (a) NaOH(aq), DCM. (b) Dibenzoyl-L-tartaric acid,
MEK, IPA, 35%, 94% ee from 9b. (c) 24a, H2SO4 (aq), MeCN, 59%, 94% ee.
Scheme 6. Synthesis of aryl hydrazine 10a
tages over the more traditional methods used for this process.
Ascorbic acid is extremely cheap, nontoxic, and environmentally
benign. It is added directly to the diazonium salt mixture as an
aqueous solution in a dose-controlled manner, avoiding the need
to transfer the diazonium salt solution between vessels. As such,
this process was deemed suitable for further scale up onto plant
scale.
Having successfully demonstrated an efficient synthesis of
racemic eletriptan rac-7, the final objective for the project was
to synthesis the single enaniomer of eletriptan (R)-7. This was
achieved via classical resolution of rac-9b, by formation of a
diastereoisomeric salt. Both enantiomers of 9b were isolated
as either the L or D-dibenzoyltartarate salt and reacted with
hydrazine oxalate 24a, under the Fischer reaction conditions,
to furnish both enantiomers of eletriptan (R)-7 and (S)-7.
Comparison with an authentic marker showed that the enanti-
omer of 9d isolated from the L-benzoyl tartaric acid gave the
desired enantiomer of eletriptan (R)-7 (scheme 8).
a Reagents and conditions: (a) PhSO2Na, IPA, H2O. (b) H2, Pd/C, THF 80%
from 21. (c) NaNO2, HCl, H2O, MeCN. (d) SnCl2, HCl, H2O 13% from 22.
Scheme 7. Formation of oxalyl hydrazide 24 and
subsequent Fischer reactiona
Conclusions
A scalable, commercially viable alternative route for the
synthesis of eletriptan (R)-7 has been developed.8 It is believed
that this new synthesis would have been better able to support
the potentially dramatic increase in bulk demand for this
compound, had alternative dosage formulations and drug
delivery systems been successful. The route uses benign
reagents, avoids the noxious phenyl vinyl sulfone and eliminates
the large waste streams derived from lithium aluminium hydride
reduction. A safe, scalable diazotisation and reduction protocol
was developed, via a stable, acid labile oxalyl hydrazide
intermediate. Despite the requirement for a classical resolution
of a key intermediate, the cost of the synthesis is predicted to
be significantly lower than the current route of manufacture.
a Reagents and conditions: (a) NaNO2, H2SO4, H2O, MeCN, L-ascorbic acid,
H2O, 83%. (b) KOH, H2O, MeCN, CaCl2 86%. (c) 9b, H2SO4, H2O, MeCN,
84%. (d) NaNO2, H2SO4, H2O, MeCN, L-ascorbic acid, H2O, 9b, 75%.
Fischer reaction, as the acidic conditions would reveal the free
hydrazine 10 in situ. Oxalyl hydrazide 24 is not a solid and is
tricky to isolate, but can be obtained as a crystalline calcium
salt 24a prior to running the Fischer reaction. Alternatively, a
one-pot procedure was developed to take the aniline 22 directly
through to racemic eletriptan (rac-7) (Scheme 7). The ascorbic
acid reduction of diazonium salts offers some general advan-
Experimental Section
1H NMR spectra were recorded on a Bruker Avance 400
MHz NMR spectrometer. LC/MS analysis was performed using
(8) Ashcroft, P. C. Modified Fischer Indole Synthesis of Eletriptan; WO/
2005/103035, 2005.
100
•
Vol. 15, No. 1, 2011 / Organic Process Research & Development

the following system; Hewlett-Packard 1100 with SB C18 3.0
mm ×50 mm, 1.8 µm particles; mobile phase consisting of
solvent A, 0.05% TFA in water, solvent B, 0.05% TFA in
acetonitrile. 0 min ) 5% solvent B; 3.5 min ) 100% solvent
B; 4.5 min ) 100% solvent B; 4.6 min ) 5% solvent B; run
time 5 min; column temperature 50 °C; λ ) 225 nm; with
Waters Micromass ZQ 2000/4000 mass detector.
give the free base as a clear oil. This was redissolved in ethyl
acetate (200 mL) and methanol (20 mL), and to this was added
a solution of fumaric acid (10.5 g, 95 mmol) in methanol (100
mL). The methanol was removed azeotropically by distillation
and replaced with ethyl acetate. The product was granulated
for 16 h then collected by filtration and dried under vacuum to
constant mass to yield 9b (20.2 g, 64.1 mmol, 64%) as a white
crystalline solid. 1H NMR ((CD3)2SO): δ ) 1.30-1.55 (m, 5H),
1.73-1.89 (m, 4H), 1.98-2.06 (m, 1H), 2.48 (s, 3H), 2.50-2.67
(m, 2H), 3.26-3.31 (m, 1H), 3.65-3.71 (m, 2H), 3.95-3.99
(m, 2H), 4.50-4.52 (t, 1H), 6.49 (s, 2H); Anal. Calcd for
C15H25NO6: C, 57.13; H, 7.99; N, 4.44. Found: C, 57.31; H,
8.03; N, 4.43.
(5,5-Dimethyl-[1,3]-dioxan-2-ylmethyl)triethylphosponi-
um bromide (18c). To a solution of 2-bromo-1,1-diethoxy-
ethane (11.7 g, 59.2 mmol) in toluene (55 mL) was added
neopentyl glycol (9.24 g, 88.8 mmol) followed by sulfuric acid
(98%, 3.2 mL, 59.2 mmol). The mixture was heated at 110 °C
for 16 h, then cooled to 4 °C and diluted with water (117 mL).
The mixture was extracted with TBME (2 × 117 mL), washed
with water (55 mL) and dried with magnesium sulfate. The
mixture was filtered and concentrated under reduced pressure.
The residue was redissolved in THF (55 mL) and triethylphos-
phine (1 M solution in THF) (88.8 mL, 88.8 mmol) was added
and heated at 65 °C for 48 h. Solvent was removed at reduced
pressure to give 18c (17.4 g, 53.3 mmol, 90%) as a brown oil.
1H NMR (CDCl3): δ ) 0.77 (s, 3H), 1.18 (s, 3H), 1.29-1.38
(m, 9H), 2.57-2.66 (m, 6H), 2.98-3.03 (dd, 2H), 3.51-3.53
(d, 2H), 3.63-3.65 (d, 2H), 4.97-5.03 (m, 1H).
Benzyl {4-[Methoxy(methyl)amino]-4-oxobutyl}methyl-
carbamate (14). To a solution of 4-(methylamino)butanoic acid
hydrochloride 13 (30.0 g, 195 mmol) in aqueous potassium
hydroxide solution (3 M, 260 mL, 780 mmol), was added the
benzyl chloroformate (33.3 g, 195 mmol) and stirred for 2 h.
The reaction was quenched into aqueous hydrochloric acid (5
M, 200 mL) and extracted twice with (tert)-butylmethyl ether
(TBME) (2 × 100 mL). The combined extracts were washed
with water (100 mL), dried with magnesium sulfate and filtered.
Solvent was removed by distillation and the residue redissolved
in dichloromethane (200 mL). Carbonyldiimidazole (32 g, 195
mmol) was added and stirred for 1 h. To the reaction was added
triethylamine (27 mL, 195 mmol) followed by N,O-dimethyl-
hydroxylamine hydrochloride, and this was stirred for 16 h. The
reaction was quenched into hydrochloric acid (2 M, 200 mL),
and the organic layer was washed with water (200 mL).
Distillation of solvent gave the title compound (47.0 g, 160
mmol, 82% from 13, 95.6% purity by LC/MS analysis) as a
clear oil. 1H NMR (CDCl3): δ ) 1.86-1.89 (m, 2H), 2.37-2.44
(m, 2H), 2.93 (s, 3H), 3.13-3.15 (br s, 3H), 3.33-3.37 (t, 2H),
3.59-3.63 (br s, 3H), 5.11 (s, 2H), 7.26-7.34 (m, 5H); LC/
MS: Rt ) 2.68 min; m/z 295 [MH]+.
Benzyl [6-(1,3-Dioxan-2-yl)-4-oxohexyl]methylcarbamate
(16). To a slurry of magnesium turnings (4.6 g, 190 mmol) in
tetrahydrofuran (THF) (100 mL), was added a crystal of iodine,
followed by 2-(2-bromoethyl)-[1,3]-dioxane (7.4 g, 38 mmol)
as a solution in THF (10 mL). The reaction was heated to 65
°C with stirring and further 2-(2-bromoethyl)-[1,3]-dioxane
(29.6 g, 152 mmol) was added as a solution in THF (40 mL).
After heating at 65 °C for a further 1 h, the reaction was cooled
to 20 °C. This freshly prepared Grignard solution was added
to a solution of 14 (40 g, 136 mmol) in THF (250 mL) at 4 °C.
The mixture was heated to 65 °C for 2 h, then quenched into
aqueous citric acid solution (10% w/v, 250 mL), and the organic
layer was concentrated at reduced pressure and redissolved in
TBME (200 mL). This was then washed with the original citric
acid solution followed by water (200 mL) and dried with
magnesium sulfate. The mixture was filtered and concentrated
under reduced pressure to give 16 (47.5 g, 136 mmol, 95%,
2-[(E)-2-(5,5-dimethyl-1,3-dioxan-2-yl)vinyl]-1-methyl-
1H-pyrrole (20). To a solution of 18c (7.0 g, 21.4 mmol) in
dimethylformamide (DMF) (47 mL) was added 1-methyl-1H-
pyrrole-2-carboxaldehyde 19 (4.73 g, 21.4 mmol). Sodium
ethoxide solution in ethanol (21%w/w, 8.8 mL, 23.5 mmol)
was added and heated at 80 °C for 3 h. The reaction was
quenched with water (100 mL) and extracted with TBME (2
× 50 mL). The combined organics were washed with water
(50 mL), dried (MgSO4), filtered and concentrated. The residue
n
was recrystallised from heptane (10 mL) to give 20 (3.65 g,
16.5 mmol, 77%, 97.9% purity by LC/MS analysis) as a beige
1
solid. H NMR (CDCl3): δ ) 0.79 (s, 3H), 1.26 (s, 3H),
3.55-3.58 (d, 2H), 3.66 (s, 3H) 3.70-3.73 (d, 2H), 5.01-5.03
(d, 1H), 5.97-6.02 (dd, 1H), 6.10-6.12 (m, 1H), 6.40-6.41
(m, 1H), 6.60-6.61 (m, 1H), 6.66-6.70 (d, 1H); LC/MS: Rt
) 2.02 min; m/z 223 [MH]+.
2-[2-(5,5-Dimethyl-1,3-dioxan-2-yl)ethyl]-1-methylpyrro-
lidine Fumarate (9c). To a solution of 20 (4.9 g, 22.1 mmol)
in ethanol (100 mL) was added 5% rhodium on carbon (50%
wet) (1 g), and the mixture was hydrogenated at 70 °C and
100 psi of hydrogen for 16 h with stirring. After this time the
catalyst was removed by filtration through a filter aid, and the
solvent was distilled at reduced pressure. The residue was
redissolved in ethyl acetate (25 mL), and to this was added
fumaric acid (2.57 g, 22.1 mmol) as a solution in methanol (25
mL). The methanol was distilled azeotropically and replaced
with ethyl acetate, maintaining a volume of 50 mL. The product
was granulated for 16 h then collected by filtration and dried
under vacuum to constant mass to yield 9c (7.2 g, 21.0 mmol,
1
90.2% purity by LC/MS analysis) as a clear oil. H NMR
(CDCl3): δ ) 1.31-1.35 (m, 1H), 1.83-1.88 (m, 4H),
2.03-2.08 (m, 1H), 2.38-2.54 (m, 4H), 2.92 (s, 3H), 3.27-3.31
(m, 2H), 3.71-3.77 (m, 2H), 4.05-4.09 (m, 2H), 4.54-4.56
(m, 1H), 5.12 (s, 2H), 7.29-7.36 (m, 5H); LC/MS: Rt ) 2.98
min; m/z 350 [MH]+.
2-[2-(1,3-Dioxan-2-yl)ethyl]-1-methylpyrrolidine Fuma-
rate (9b). To a solution of 16 (35 g, 95.5 mmol) in methanol
(200 mL) was added 5% palladium on carbon (50% wet) (3.5
g) and hydrogenated at 50 °C and 60 psi of hydrogen for 16 h
with stirring. The catalyst was removed by filtration through a
filter aid, and the solvent was distilled at reduced pressure to
Vol. 15, No. 1, 2011 / Organic Process Research & Development
•
101

1
95%) as a white crystalline solid. H NMR ((CD3)2SO): δ )
stirring for 16 h, and then quenched into aqueous potassium
hydroxide solution (2 M, 82 mL, 164 mmol). The product was
extracted twice with ethyl acetate (2 × 200 mL), washed with
water (200 mL), and dried with magnesium sulfate. The mixture
was filtered and concentrated under reduced pressure. The
residue was purified via flash chromatography, eluting with
dichloromethane (95) ethanol (5) ammonia solution (1) to give
rac-7 (10.2 g, 13.4 mmol, 84%, 96.8% purity by LC/MS
analysis). 1H NMR (CDCl3): δ ) 1.51-1.85 (m, 4H), 2.22-2.28
(m, 1H), 2.43-2.49 (m, 4H), 2.56-2.62 (m, 1H), 3.11-3.18
(m, 4H), 3.42-3.46 (m, 2H), 6.91-6.93 (s, 1H), 7.01 (s, 1H),
7.23-7.27 (d, 1H), 7.31 (s, 1H), 7.56-7.60 (m, 2H), 7.65-7.68
(m, 1H), 7.96-7.98 (d, 2H), 8.14 (s, 1H); LC/MS: Rt ) 2.30
min; m/z 383 [MH]+.
3-[(1-Methylpyrrolidin-2-yl)methyl]-5-[2-(phenylsulfonyl)-
ethyl]-1H-indole (rac-7) (Method B, One-Pot Procedure).
To a solution of 22 (1.0 g, 3.83 mmol) in acetonitrile (10 mL)
at 4 °C was added sulfuric acid (9.4 M, 7 mL, 65.8 mmol)
followed by sodium nitrite (0.29 g, 4.20 mmol) in aqueous
solution (1 mL). After stirring for 1 h at 4 °C, ascorbic acid
(0.74 g, 4.20 mmol) was added as an aqueous solution (1.5
mL). After stirring for a further 1 h at 4 °C, the reaction was
warmed to ambient temperature and stirred for 16 h. The
mixture was diluted with water (10 mL) and 9b (1.2 g, 3.83
mmol) was added. The reaction was heated at +80 °C for 16 h,
then neutralised with aqueous potassium hydroxide solution (15
mL, 5 M, 75.0 mmol), and diluted with water (50 mL). The
product was extracted twice with ethyl acetate (2 × 20 mL),
washed with water (20 mL), and dried with magnesium sulfate.
The mixture was filtered and concentrated under reduced
pressure. The residue was purified via flash chromatography,
eluting with dichloromethane (95 mL), ethanol (5 mL), and
ammonia solution (1 mL) to give rac-7 (1.1 g, 2.87 mmol, 75%,
95.9% by LC/MS analysis). 1H NMR (CDCl3): δ ) 1.51-1.85
(m, 4H), 2.22-2.28 (m, 1H), 2.43-2.49 (m, 4H), 2.56-2.62
(m, 1H), 3.11-3.18 (m, 4H), 3.42-3.46 (m, 2H), 6.91-6.93
(s, 1H), 7.01 (s, 1H), 7.23-7.27 (d, 1H), 7.31 (s, 1H), 7.56-7.60
(m, 2H), 7.65-7.68 (m, 1H), 7.96-7.98 (d, 2H), 8.14 (s, 1H);
LC/MS: Rt ) 2.30 min; m/z 383 [MH]+.
(2R)-2-[2-(1,3-Dioxan-2yl)ethyl]-1-methylpyrrolidine(2R,3R)-
2,3-Bis(benzyloxy)succinic Acid. 9b (2.37 g, 7.52 mmol) was
partitioned between DCM (20 mL) and 1 M potassium
hydroxide solution (20 mL, 20 mmol), and the aqueous was
extracted with further DCM (2 × 20 mL). The combined
extracts were dried (MgSO4), filtered, and concentrated. The
residue was redissolved in 2-butanone (7.5 mL) at 75 °C, and
dibenzoyl-L-tartaric acid (2.7 g, 7.53 mmol) was added as a
solution in 2-butanone (7.5 mL). The solution was cooled to 0
°C, and the product was granulated for 16 h and then collected
by filtration and dried under vacuum to constant mass. The solid
was recrystallised from 2-propanol (7.5 mL) to give 9d (1.3 g,
2.63 mmol, 35%, 94% ee) as a white crystalline solid. 1H NMR
((CD3)2SO): δ ) 1.30-1.55 (m, 5H), 1.73-1.89 (m, 4H),
2.02-2.13 (m, 1H), 2.48 (s, 3H), 2.80-2.93 (m, 1H), 2.96-3.08
(m, 1H), 3.37-3.46 (m, 1H), 3.61-3.71 (m, 2H), 3.95-3.99
(m, 2H), 4.51-4.54 (m, 1H), 5.64 (s, 1H), 7.42-7.54 (t, 2H),
7.60-7.64 (t, 1H) 7.92-7.99 (d, 2H); Anal. Calcd for
0.68 (s, 3H), 1.08 (s, 3H), 1.34-1.60 (m, 4H), 1.71-1.82 (m,
3H), 1.96-2.05 (m, 1H), 2.45 (s, 3H), 2.51-2.61 (m, 2H),
3.21-3.26 (m, 1H), 3.36-3.39 (d, 2H), 3.50-3.52 (d, 2H),
4.42-4.44 (t, 1H), 6.50 (s, 2H); Anal. Calcd for C17H29NO6:
C, 59.46; H, 8.51; N, 4.08. Found: C, 59.83; H, 8.07; N, 4.38.
4-(2-Benzenesulfonylethyl)phenylamine (22). To a mixture
of 4-nitrophenethyl bromide 21 (10 g, 43.5 mmol) in 2-propanol
(120 mL) and water (30 mL) was added sodium benzenesulfi-
nate (13.1 g, 65.2 mmol) and heated to 70 °C for 42 h. The
reaction mixture was cooled to ambient temperature and filtered.
The solid was dissolved in methanol (150 mL), and 5% Pd/C
(50% wet) (1 g) was added and the mixture hydrogenated at
60 °C and 60 psi of hydrogen for 18 h. The mixture was filtered
through a filter aid and concentrated. The residue was dissolved
in ethyl acetate (30 mL) and hexane (250 mL) added. Filtration
gave 22 (6.96 g, 34.8 mmol, 80%, 98.8% purity by LC/MS
analysis) as a white solid. 1H NMR (CDCl3): δ ) 2.87-2.96
(m, 2H), 3.25-3.35 (m, 2H), 3.32-3.69 (b, 2H), 6.57 (d, 2H),
6.87 (d, 2H), 7.52-7.60 (m, 2H), 7.62-7.69 (m, 1H), 7.92 (d,
2H); LC/MS: Rt ) 1.66 min; m/z 262 [MH]+.
Oxo(2-{4-[2-(phenylsulfonyl)ethyl]phenyl}hydrazino)ace-
tic Acid (24) (Method A). To a solution of 22 (20.2 g, 77.3
mmol) in acetonitrile (60 mL) at 4 °C was added sulfuric acid
(9.4 M, 60 mL, 563 mmol) followed by sodium nitrite (5.9 g,
85.0 mmol) in aqueous solution (11.8 mL). After stirring for
1 h at 4 °C, ascorbic acid (15.0 g, 85.0 mmol) was added as an
aqueous solution (30 mL). After stirring for a further 1 h at 4
°C, the reaction was warmed to ambient temperature and stirred
for 16 h. After this time the reaction was diluted with water
(40 mL) and extracted twice with ethyl acetate (2 × 100 mL).
The combined organic extracts were washed with water (100
mL) and dried (MgSO4). The mixture was filtered and con-
centrated under reduced pressure to give 24 (22.3 g, 64.2 mmol,
83%, 90.9% purity by LC/MS analysis) as a red oil. 1H NMR
((CD3)2SO): δ ) 2.75-2.78 (m, 2H), 3..50-3.53 (m, 2H),
6.59-6.61(d, 2H), 6.95-6.97 (d, 2H), 7.62-7.66 (m, 2H),
7.71-7.75 (m, 1H), 7.90-7.92 (m, 2H), 10.54 (s, 1H); LC/
MS: Rt ) 2.13 min; m/z 347 [M - H]-.
Calcium Bis[oxo(2-{4-[2-(phenylsulfonyl)ethyl]phenyl}-
hydrazino)acetate] (24a) (Method A). 22 (22.3 g, 64.0 mmol)
was dissolved in acetonitrile (100 mL) and aqueous potassium
hydroxide solution (0.64 M, 100 mL, 64.0 mmol) added and
stirred for 1 h. Aqueous calcium chloride solution (1 M, 32
mL, 32.0 mmol) was added, and the precipitate was granulated
for 16 h. The solid was collected by filtration and dried under
vacuum until constant volume to give the title compound (20.2
g, 27.5 mmol, 86%, 98.4% purity by LC/MS analysis) as a
white crystalline solid. 1H NMR ((CD3)2SO): δ ) 2.71-2.75
(m, 2H), 3..49-3.53 (m, 2H), 6.56-6.58 (d, 2H), 6.92-6.94
(d, 2H), 7.60 (br s, 1H), 7.63-7.67 (m, 2H), 7.73-7.75 (m,
1H), 7.90-7.92 (d, 2H), 10.13 (br s, 1H); LC/MS: Rt ) 2.13
min; m/z 347 [M - H]-.
3-[(1-Methylpyrrolidin-2-yl)methyl]-5-[2-(phenylsulfonyl)-
ethyl]-1H-indole (rac-7) (Method A). To a slurry of 24a
(11.7 g, 15.9 mmol) and 9b (10.0 g, 31.8 mmol) in acetonitrile
(84 mL) was added aqueous sulfuric acid (10% v/v, 1.88 M,
84 mL, 158 mmol). The solution was heated at 80 °C with
102
•
Vol. 15, No. 1, 2011 / Organic Process Research & Development

C29H35NO10: C, 62.47; H, 6.33; N, 2.51. Found: C, 62.64; H,
6.39; N, 2.46.
4H), 2.22-2.28 (m, 1H), 2.43-2.49 (m, 4H), 2.56-2.62 (m,
1H), 3.11-3.18 (m, 4H), 3.42-3.46 (m, 2H), 6.91-6.93 (s,
1H), 7.01 (s, 1H), 7.23-7.27 (d, 1H), 7.31 (s, 1H), 7.56-7.60
(m, 2H), 7.65-7.68 (m, 1H), 7.96-7.98 (d, 2H), 8.14 (s, 1H);
LC/MS: Rt ) 2.30 min; m/z 383 [MH]+.
(R)-3-[(1-Methylpyrrolidin-2-yl)methyl]-5-[2-(phenylsul-
fonyl)ethyl]-1H-indole ((R)-7). To a slurry of 24a (1.0 g, 1.35
mmol) and 9d (1.5 g, 2.7 mmol) in acetonitrile (8 mL) was
added aqueous sulfuric acid (10% v/v, 1.88 M, 8 mL, 15 mmol).
The solution was heated at 80 °C with stirring for 16 h, and
then quenched into aqueous potassium hydroxide solution (50
mL, 2 M, 100 mmol). The product was extracted twice with
ethyl acetate (2 × 50 mL), washed with water (50 mL), and
dried with magnesium sulfate. The mixture was filtered and
concentrated under reduced pressure. The residue was purified
via flash chromatography, eluting with dichloromethane (95
mL), ethanol (5 mL), and ammonia solution (1 mL) to give
(R)-7 (0.6 g, 0.797 mmol, 59%, 94% ee, 97.8% achiral purity
Acknowledgment
We thank Liam Tully, Alastair Denholm, Ian Moses, and
Dave Entwistle for useful scientific input. We also thank Erwan
Coat and Catriona Thom for analytical support and Neal Sach
for chiral salt screening.
Received for review September 16, 2010.
OP100251Q
1
by LC/MS analysis). H NMR (CDCl3): δ ) 1.51-1.85 (m,
Vol. 15, No. 1, 2011 / Organic Process Research & Development
•
103