Development of a practical synthesis of STA-5312, a novel indolizine oxalylamide microtubule inhibitor

Article abstract of DOI:10.1021/op6002852

An efficient synthesis of the novel microtubule inhibitor STA-5312(3-[(4-cyanophenyl)methyl]-N-3-methyl-5-isothiazolyl) -α-oxo-1-indolizineacetamide) was developed. A novel DMF/ Me ₂SO₄ directed regioselective synthesis of the 3-(4-c

Full text of DOI:10.1021/op6002852

Organic Process Research & Development 2007, 11, 246−250
Development of a Practical Synthesis of STA-5312, a Novel Indolizine
Oxalylamide Microtubule Inhibitor
Hao Li, Zhiqiang Xia, Shoujun Chen, Keizo Koya, Mitsunori Ono, and Lijun Sun*
Synta Pharmaceuticals Corp., 45 Hartwell AVenue, Lexington, Massachusetts 02421, U.S.A.
Abstract:
An efficient synthesis of the novel microtubule inhibitor STA-
5312 (3-[(4-cyanophenyl)methyl]-N-(3-methyl-5-isothiazolyl)-r-
oxo-1-indolizineacetamide) was developed. A novel DMF/
Me2SO4 directed regioselective synthesis of the 3-(4-cyano-
benzoyl)-indolizine (4) was a critical transformation within the
four-step process. Alternatively, a CuCl mediated synthesis of
3-(4-cyanobenzyl)indolizine (5) was also developed. All inter-
mediates were obtained in high quality and were used directly
for the next step without extensive purification. The drug
substance itself was purified by recrystallization from a mixture
of THF and water, resulting in a high purity product (HPLC
>98%). The process was applied successfully in the manufac-
turing of kilograms of GMP API.
Results and Discussion
The medicinal chemistry approach to the synthesis of
STA-5312 is depicted in Scheme 1. 4-Cyanophenacyl
bromide (2) was treated with 2 equiv of pyridine in
acetonitrile at room temperature.5 The resulting pyridinium
salt was easily isolated in quantitative yield by collecting
the precipitates from the reaction mixture, which, after drying
in vacuo, was treated with 3-butyn-2-one in the presence of
potassium carbonate as a base at room temperature in THF.6
Removal of inorganic salt, concentration of the organic
solution, and flash column purification with silica gel resulted
in 1-acetyl-3-(4-cyanobenzoyl)indolizine (3) in 72% yield.
All attempts to transform the acetyl group to the desired
oxalyl functionality or its equivalent were not successful.
However, when compound 3 was refluxed in benzene
together with 3 equiv of ethylene glycol and 2 equiv of
p-toluenesulfonic acid monohydrate an unprecedented regio-
selective deacetylation (retro Friedel-Crafts reaction) oc-
curred leading to the formation of 3-(4-cyanobenzoyl)-
indolizine 4. This interesting transformation, however, was
difficult to scale up and required silica gel chromatographic
purification to provide pure product 4 in yields ranging from
30% to 60%. Reduction of the carbonyl group of 4 with
NaBH3CN and TMSCl gave the key 3-(4-cyanobenzyl)-
indolizine 5 in 56% yield after purification by silica gel
chromatography.7 Oxalylation of compound 5 with 1.2 equiv
of oxalyl chloride was carried out in anhydrous diethyl ether
at 0 °C. The removal of excessive oxalyl chloride and solvent
required careful control of the external bath temperature to
be below 25 °C in order to avoid decomposition. The crude
oxalyl monochloride derivative 5A was then used in the
amidation reaction without purification. Thus, the THF
solution of compound 5A was treated with 2 equiv of
5-amino-3-methylisothiazole 0 °C. After silica gel chromato-
graphic purification, STA-5312 (1) was isolated as a yellow
Introduction
The development of multidrug resistance (MDR) is one
of the leading causes of chemotherapy failure in cancer
treatment.1 The search for new chemical entities that can
overcome MDR remains an active research area in anticancer
drug discovery and development.2
3-[(4-Cyanophenyl)methyl]-N-(3-methyl-5-isothiazolyl)-
R-oxo-1-indolizineacetamide, STA-5312 (1), possesses potent
antitumor activity both in vitro and in vivo through its
binding to a novel binding site in the microtubules.3 More
remarkably, STA-5312 demonstrates similar potency in
inhibiting tumor growth in tumor xenograph animal models
with both wild type and MDR type cancers.4 As a novel
microtubule inhibitor, STA-5312 was selected as a clinical
development candidate.
* To whom correspondence should be addressed. Telephone: 781-274-8200.
Fax: 781-274-8228. E-mail: lsun@syntapharma.com.
(1) For a recent review: Szaka´cs, G.; Paterson, J. K.; Ludwig, J. A.; Booth-
Genthe, C.; Gottesmandd, M. M. Nat. ReV. Drug DiscoVery 2006, 5, 219.
(2) (a) Simoni, D.; Grisolia, G.; Giannini, G.; Roberti, M.; Rondanin, R.;
Piccagli, L.; Baruchello, R.; Rossi, M.; Romagnoli, R.; Invidiata, F. P.;
Grimaudo, S.; Jung, M. K.; Hamel, E.; Gebbia, N.; Crosta, L.; Abbadessa,
V.; Di Cristina, A.; Dusonchet, L.; Meli, M.; Tolomeo, M. J. Med. Chem.
2005, 48, 723. (b) Oberlies, N. H.; Chang, C.-j.; McLaughlin, J. L. J. Med.
Chem. 1997, 40, 2102. (c) Antonini, I.; Polucci, P.; Kelland, L. R.; Menta,
E.; Pescalli, N.; Martelli, S. J. Med. Chem. 1999, 42, 2535.
(3) (a) James, D. A.; Koya, K.; Li, H.; Chen, S.; Xia, Z.; Ying, W.; Wu, Y.;
Sun, L. Bioorg. Med. Chem. Lett. 2006, 16, 5164. (b) James, D. A.; Koya,
K.; Li, H.; Chen, S.; Xia, Z.; Ying, W.; Wu, Y.; Sun, L. Abstr. of Papers,
232nd ACS National Meeting, San Francisco, CA, United States, Sept. 10-
14, 2006, MEDI-039.
(4) Sun, L.; Koya, K.; Li, H.; Przewloka, T.; James, D.; Chen, S.; Xia, Z.;
Liang, G.; Tatsuta, N.; Wu, Y.; Zhou, D.; Korbut, T.; Du, Z.; Ono, M.
Abstr. of Papers, 228th ACS National Meeting, Philadelphia, PA, United
States, August 22-26, 2004, MEDI-094.
(5) Miki, Y.; Kinoshita, H.; Yoshimaru, T. Heterocycles 1987, 26, 199.
(6) Miki, Y.; Hiroishi, Y.; Hachiken, H.; Takemura, S. J. Heterocycl. Chem.
1991, 28, 45.
(7) Box, V. G. S.; Meleties, P. C. Tetrahedron Lett. 1998, 39, 7059.
246
•
Vol. 11, No. 2, 2007 / Organic Process Research & Development
10.1021/op6002852 CCC: $37.00 © 2007 American Chemical Society
Published on Web 02/24/2007

Scheme 1a
Scheme 3a
a Reagents and conditions: (a) Br2, EtOAc, rt; (b) picoline, MeCN, rt, 83%
(two steps); (c) DMF(O-But)2 (10 equiv), DMF, 130-135 °C, 80%.
The Sonogashira coupling reaction9 of 6 with 2-bromo-
pyridine proceeded smoothly in THF at 60 °C to afford the
desired 2-alkynylpyridine derivative 7 as an oil in 80% yield
and excellent purity of >98% (HPLC, 230 nm) after routine
workup and removal of organic solvents. The CuCl mediated
cyclization reaction was carried out in TEA/DMA (1:7
volume ratio) at 130 °C with 1 equiv of CuCl. We found in
our hand that, by using 1 mol equiv of CuCl, the cyclization
reaction proceeded quickly and took only a few hours to
reach completion at a scale of 10-50 g. After cooling to
room temperature, the reaction mixture was added to water
and ethyl acetate. Treatment of the organic solution with
charcoal and removal of solvent afforded the desired key
indolizine compound 5 as an off-white solid in 90% yield
with an HPLC purity of >95%.
a Reagents and conditions: (a) pyridine, CH3CN, rt, quantitative; (b) K2CO3,
3-butyn-2-one, THF, rt, 72%; (c) HOCH2CH2OH, p-TsOH·H2O, benzene, reflux,
60%; (d) NaBH3CN, TMSCl, CH3CN, rt, 56%; (e) (COCl)2, Et2O, 0 °C; (f)
5-amino-3-methylisothiazole (2 equiv), THF, 0 °C, 78% (two steps).
Scheme 2a
a Reagents and conditions: (a) LDA, DMPU, propargyl bromide, THF, -78
to -40 °C, 92%; (b) 2-bromopyridine, Pd(PPh3)2Cl2, CuI, TEA, THF, 60 °C,
80%; (c) CuCl, TEA, DMA, 130 °C, 90%.
Being aware of the regulatory requirement to eventually
achieve a ppm level of transition metals in drug substances,
we decided that before we further optimized this route that
utilized Pd and Cu, we would also explore alternative routes
that do not employ transition metals. In our laboratory, we
were able to dramatically improve the chemoselectivity of a
reported method by using DMF di-tert-butyl acetal (Scheme
4).10 Under the newly established reaction condition, reaction
of picolinium salt 8, which was easily prepared by treating
picoline with 4-cyanophenacyl bromide (2),11 with a large
excess (10-15 equiv) of DMF di-tert-butyl acetal at 130
°C with DMF as a solvent, resulted in the isolation of the
indolizine 4 in good yield (Scheme 3). However, the high
cost of DMF di-tert-butyl acetal makes it less attractive for
large scale production. Hence we set out to explore a more
affordable reagent to replace the acetal and hopefully to
improve the selectivity further.12
The intermolecular cyclization promoted by DMF acetal
presumably proceeds via the formation of the iminium
intermediate (I). Apparently, the bulky tert-butyl group
accelerates the in situ formation of its corresponding iminium
intermediate (I, R ) t-Bu) and therefore facilitates the
intermolecular cyclization reaction. We rationalized that a
preformed iminium intermediate should further favor the
formation of the intermolecular cyclization. It is reported that
the formation of the DMF solution of methoxymethylene-
dimethylammonium methylsulfate (I, R ) Me) was achieved
via the adduction of DMF with methyl sulfate at elevated
temperatures.13 The adduct I had been applied in the
crystalline solid (78% yield, HPLC purity >98%). However,
we encountered significant scale-up challenges when the
reaction was carried out at >10 g scales. Only moderate yield
(30-40%) was achieved and often required multiple runs
of column purification to achieve a purity of >95%.
In order to implement a process that is transferable for
large scale GMP manufacturing, significant efforts were
required to resolve the major drawbacks such as the uses of
expensive reagents (i.e., 3-butyn-2-one), toxic reagent and
solvent (i.e., NaBH3CN and benzene, respectively), low yield,
and requirement of silica gel chromatographic purification
in key steps.
With the recognition that the transformation of compound
3 to the key intermediate 4, via the selective removal of the
pendent 1-acetyl group is not atomically economic, we
initiated our process development by exploring new routes
that could lead to more direct formation of the key
intermediate 4 or 5.
A publication from the group of Gevorgyan described a
novel synthesis of 3-alkylindolizines via a CuCl mediated
cycloisomerization of 2-alkynylpyridine.8 Thus, when 2-hex-
1-ynylpyridine was heated to 130 °C in the presence of 50
mol % of CuCl, 3-n-propylindolizine was synthesized in 91%
yield.8 We were able to successfully apply this novel
approach to the synthesis of compound 5 in up to a 50 g
scale (Scheme 2). R-Deprotonation of p-tolunitrile with
freshly prepared LDA (n-BuLi and diisopropylamine) at -78
°C in THF was followed by slow addition of neat propargyl
bromide. The reaction temperature was raised to -40 °C
and then quenched with saturated NH4Cl solution. 4-(Butyn-
3-yl)benzonitrile (6) was easily purified and isolated as a
solid in 92% yield following standard aqueous washings,
removal of organic solvents, and drying under a vacuum.
(9) Sonogashira, K.; Tohda, Y. Tetrahedron Lett. 1975, 4467.
(10) (a) Copar, A.; Stanovnik, B.; Tisler, M. J. Heterocycl. Chem. 1993. 30,
1577. (b) Xia, Z.; Przewloka, T.; Koya, K.; Ono, M.; Chen, S.; Sun, L.
Tetrahedron Lett. 2006, 47, 8817.
(11) Miki, Y.; Kinoshita, H.; Yoshimaru, T. Heterocycles 1987, 26, 199.
(12) Przewloka, T.; Xia, Z.; Koya, K.; Ono, M.; Sun, L. Abstr. of Papers, 228th
ACS National Meeting, Philadelphia, PA, United States, August 22-26,
2004, ORGN-682.
(8) Kel’in, A. V.; Sromek, A. W.; Gevorgyan, V. J. Am. Chem. Soc. 2001,
123, 2074.
(13) Bredereck, H.; Effenberger, F.; Simchen, G. Chem. Ber. 1963, 96, 1350
Vol. 11, No. 2, 2007 / Organic Process Research & Development
•
247

Scheme 4a
Finally, optimization of the oxalylation of compound 5
and subsequent amidation with 2-amino-4-methylthiazole led
to the establishment of a one pot process in THF that not
only significantly improved the reaction efficiency but also
simplified the isolation of high purity production while
eliminating the necessity of chromatographic purification.
When THF was used as solvent, treatment of 5 with oxalyl
chloride (1.2 equiv) at -40 °C was followed by sequential
addition into the reaction mixture of Et3N (1.2 equiv), DMAP
(10 mol %), and 2-amino-4-methylthiazole (1.2 equiv as in
THF). In order to minimize the formation of impurities, it
was imperative to maintain the reaction temperature below
-40 °C during the oxalylation reaction and below -25 °C
during the subsequent amidation reaction. The use of 10 mol
% of DMAP allowed the amidation to proceed smoothly
while the reaction temperature was maintained below -25
°C which significantly reduced the formation of impurities.
After the reaction proceeded to completion (monitored
by TLC), the reaction was quenched with about 1.5 volume
equiv of 3% NaHCO3 aqueous solution which led to the
precipitation of the product STA-5312 while most of the
impurities remained in the solution phase. After washing the
collected solid with THF/water (1:2) and 80% EtOH, the
crude product was then dissolved in THF. Attempts to
recrystallize the product with the addition of an organic
antisolvent, such as heptane, EtOH, i-PrOH, or EtOAc, led
to the formation of an oily paste. On the other hand, adding
water to the THF solution resulted in the precipitation of
the product as fine powders with a purity of >98%.
a Reagents and conditions: (a) Me2SO4/DMF, 60-80 °C; (b) 8, Et3N, DMF,
0 °C to rt, 76%.
Scheme 5a
a Reagents and conditions: (a) BH3-THF, MeOH/MeCN, 50 °C, 52%.
preparation of dioxolanes and dioxanes14 and in the Beck-
mann rearrangement of cyclohexanone oxime to its corre-
sponding ꢀ-caprolactam.15
To our delight, when the picolinium salt 8 was treated at
room temperature with a preformed DMF solution of
methoxymethylenedimethylammonium methylsulfate in the
presence of Et3N as a base, the desired indolizine product 4
was readily formed as the major product (Scheme 4). The
reaction was easy to scale up. We obtained reproducible
yields and purities at scales ranging from 20 to 100 g (75-
85%). Isolation was readily achieved by adding water to the
reaction mixture resulting in the precipitation of product 4
which was collected, washed with water, and vacuum-dried.
The reduction of the 3-acylindolizine 4 to the key
intermediate 5 was achieved by using BH3-THF as an
alternative reducing agent to the toxic sodium cyanoboro-
hydride used during the earlier stage of the program (Scheme
5).16 The use of MeCN as solvent was crucial in preventing
the reduction of the cyano functional group. Interestingly,
we found that a small amount of MeOH was beneficial to
the reaction and resulted in a much cleaner reaction. The
best ratio for 4/BH3-THF/MeOH was found to be 1/4/2.
The optimal reaction temperature was 50 °C. A higher
temperature often led to a much lower yield, while for
temperatures lower than 50 °C sluggish reactions occurred.
Subsequent investigation determined that contaminants from
boranic side products in compound 4 often caused messy
and unpredictable reactions when 4 was treated with oxalyl
chloride. Pure product 5 (HPLC purity >95%) was obtained
as an off-white solid by filtering the crude product through
a short silica gel plug using DCM and hexanes as eluents to
remove the contaminants. Although the yield of this step,
after purification, is moderate (50-60%), the low cost of
its precursor compound 4 made it attractive to large scale
manufacturing.
Conclusions
We developed an efficient four-step process for the
synthesis of STA-5312, a novel indolizine type microtubule
inhibitor. All reaction intermediates were obtained in high
purities which were used directly for the next step without
extensive purification. The drug substance itself was purified
by recrystallization from THF and water. The novel DMF/
Me2SO4 directed synthesis of mono 3-substituted indolizines
formed the key transformation of the process. This chemical
process was applied successfully to the multi-kilogram GMP
manufacturing of the API (Scheme 6). When comparing the
new route to the original route, we were able to not only
shorten the synthesis from six steps to four steps but also
eliminate the use of expensive reagents, hazardous solvents,
and chromatographic purifications.
Experimental Section
General Procedures: Melting points were obtained from
DSC Q100, TA Instrument (with heating ramp 10.0 °C/min,
nitrogen purge) and were uncorrected. NMR data were
obtained at 300 MHz using a Varian instrument with TMS
as an internal standard. FTIR analysis was performed with
a PERKIN ELMER FT-IR spectrometer (Spectrum 1000),
Norwalk, CT. HPLC analyses were performed on a Hewlett-
Packard series 1100 reversed-phase liquid chromatograph
equipped with a UV detector and an Agilent Eclipse XDB-
C18, 150 mm × 4.6 mm, 5 µm column. LCMS data were
acquired from a Finnigan LCQ LC/MS system (equipped
(14) Kantlehner, W.; Gutbrod, H. D. Liebigs Ann. Chem. 1979, 1362.
(15) Izumi, Y.; Fujita, T. J. Mol. Catal. A: Chem. 1996, 106, 43.
(16) Mlotkowska, B. L.; Cushman, James A.; Harwood, J.; Lam, W. W.; Ternay,
A. L., Jr. J. Heterocycl. Chem. 1991, 28, 731.
248
•
Vol. 11, No. 2, 2007 / Organic Process Research & Development

Scheme 6a
a Reagents and conditions: (a) picoline, MeCN, rt, 83%; (b) (i) Me2SO4, DMF, 60 °C; (ii) Et3N, 0 °C to rt, 76%; (c) BH3-THF, MeOH/MeCN, 50 °C, 52%; (d)
(i) oxalyl chloride, THF, -50 to -40 °C; (ii) 5-amino-3-methylisothiazole, TEA, DMAP, THF, -30 to -25 °C, 2 h, 50%.
with an Agilent 1100 series HPLC). Microanalysis was
performed by Atlantic Microlab, Inc., Norcross, GA.
All reagents and solvents were commercially available
and were used without further purification unless otherwise
stated. All anhydrous reactions were conducted in oven-dried
glassware under the protection of nitrogen. Air and moisture
sensitive reagents were transferred via syringes. Column
chromatographic purification was performed with EM silica
gel 60 under slightly pressured air or nitrogen.
1 h. The reaction mixture was cooled to ∼10 °C, quenched
with ice water (70 mL) carefully (∼8 mL/min). To the
mixture was then added EtOAc (200 mL). The organic layer
was dried with Na2SO4 and concentrated under reduced
pressure. The residue was dissolved in CH2Cl2 (100 mL)
and diluted with heptane (30 mL), filtered through a short
silica gel funnel (110 g of silica gel packed in 1:1 CH2Cl2/
hexanes), and eluted with 1:1 CH2Cl2/hexanes (∼600 mL),
and the product fraction was collected and evaporated to give
12.4 g of 3-(4-cyanobenzyl)indolizine (51% yield, 96% purity
by HPLC) as a white solid. Rf 0.3 (10% ethyl acetate in
hexanes); mp (DSC): 76.3 °C. IR (cm-1): 2224, 1604, 1503,
1434, 1363, 1316, 1174, 1026, 838, 819, 755. 1H NMR (300
MHz, CDCl3) δ (ppm), 4.09 (s, 2H), 6.36 (m, 2H), 6.56 (m,
2H), 7.15 (d, J ) 8.4 Hz, 2H), 7.32 (d, J ) 9.3 Hz, 1H),
7.45 (m, 3H); 13C NMR (CDCl3): δ (ppm) 143.8, 133.6,
130.7, 130.1, 128.1, 120.1, 118.4, 118.1, 117.2, 115.1, 112.4,
110.2, 99.8, 32.4. Anal. Calcd for C16H12N2 (232.1): C,
82.73; H, 5.21; N, 12.06. Found: C, 82.61; H, 5.10; N, 12.00.
ESMS calcd for C16H12N2, 232.1; found, 233.1 (M + H)+.
3-[(4-Cyanophenyl)methyl]-N-(3-methyl-5-isothiazolyl)-
R-oxo-1-indolizineacetamide (STA-5312). To a stirred
solution of oxalyl chloride (22.7 mL, 0.26 mol) in anhydrous
THF (750 mL) at -50 to -40 °C (inner temp) was added a
solution of 3-(4-cyanobenzyl)indolizine (50.24 g, 0.22 mol)
in anhydrous THF (100 mL). The addition rate was con-
trolled at ∼10 mL/min so that the inner temperature was
maintained below -40 °C. The resulting mixture was kept
at ∼ -40 °C for 1 h. To it were successively added
triethylamine (36.2 mL, 0.26 mol), DMAP (2.65 g, 0.02 mol),
and a THF solution (50 mL) of 5-amino-3-methylisothiazole
(33.18 g, 0.29 mol), while the inner temperature was
maintained below -30 °C. After the addition the mixture
was then stirred at ∼ -25 °C for 2 h, quenched with 3%
sodium bicarbonate (1.5 L), and stirred at 0 °C for 4 h. The
precipitate was collected and washed with THF/water (1:2,
0 °C, ∼3 L) and 80% ethanol (0 °C, 1 L) to give the crude
product (HPLC purity: 94%, 230 nm) as a brown solid. The
crude product was dissolved in THF (800 mL) and refluxed
for 15 min. After filtration to remove any insoluble particles,
the filtrate was cooled to 0 °C, and to it water (1 L) was
added with stirring. The mixture was let stand at 0 °C
overnight. The precipitate was collected, washed with THF/
water, (1:2, 0 °C) and dried under a vacuum to give 41.9 g
(47.6% yield, 99% purity by HPLC) of 2-[3-(4-cyanobenzyl)-
indolizin-1-yl]-N-(3-methylisothiazol-5-yl)-2-oxoacetamide
(STA-5312) as a yellow crystalline powder. Mp (DSC) 233.5
°C; IR (cm-1) 3292 (m), 3149 (m), 3028 (w), 2969 (w), 2841
(w), 2229 (m), 1670 (m), 1550 (s), 1505 (s), 1477 (s), 1354
3-(4-Cyanobenzoyl)indolizine (4). To a solution of
4-acetylbenzonitrile (34 g, 235 mmol) in EtOAc (320 mL)
was added Br2 (neat, 11.9 mL, 320 mmol) at room temper-
ature. The reaction mixture was stirred at room temperature
for 1 h. After the solvent was removed under reduced
pressure, the resulting residue was dissolved in CH3CN (210
mL). To it was added picoline (50 mL, 500 mmol). After
the reaction mixture was stirred for 2 h at room temperature
and then 0.5 h at 0 °C, EtOAc (70 mL) was added to the
mixture. The resulting precipitates were collected by filtration
and washed with EtOAc to give 2-methyl-1-(4-cyano)-
phenacylpyridinium bromide (60 g, 88%) which was used
directly in the next step. 1H NMR (300 MHz, DMSO-d6) δ
(ppm) 9.05-8.03 (m, 8H), 6.78 (s, 2H), 2.74 (s, 3H). To a
stirred suspension of the above picolinium salt (50 g, 120
mmol) in DMF (500 mL) was added DMF-Me2SO4 (400
mL) (pre-prepared in a separate reaction flask by stirring a
mixture of equal mole equivalents of DMF and Me2SO4 at
60-80 °C for 3 h, then cooled to room temperature). After
the addition, the reaction mixture was stirred at room
temperature for 15 min. To it was then added Et3N (500
mL) while the inner temperature was kept between 25 and
40 °C. After stirring at room temperature for 2 h, the reaction
mixture was charged with ice water (2000 mL) with stirring
which led to the formation of slurry. The precipitates were
collected, washed with water, and dried under a vacuum to
give 29 g (76%) of 3-(4-cyanobenzoyl)indolizine. Rf 0.3
(10% ethyl acetate in hexanes); mp 156-157 °C (recrystal-
1
lized from ethyl acetate); H NMR (300 MHz, CDCl3) δ
(ppm) 9.98 (d, J ) 6.6 Hz, 1H), 7.89-7.77 (m, 4H), 7.60
(d, J ) 11 Hz, 1H), 7.30-7.22 (m, 2H), 7.01 (m, 1H), 6.57
(d, J ) 6.1 Hz, 1H); 13C NMR (CDCl3) δ (ppm) 181.75,
144.91, 140.17, 132.12, 129.41, 129.23, 126.92, 125.43,
122.09, 118.88, 118.42, 114.60, 114.12, 103.61. ESMS calcd
for C16H10N2O, 246.1; found, 247.1 (M + H)+. Anal. Calcd
for C16H10N2O: C, 78.03; H, 4.09; N, 11.38. Found: C,
77.82; H, 3.92; N, 11.11.
3-(4-Cyanobenzyl)indolizine (5). To a solution of 3-(4-
cyanobenzoyl)indolizine (24.6 g, 100 mmol) in CH3CN (600
mL) containing MeOH (9 mL) was added BH3-THF (1 M,
450 mL), and the resulting mixture was stirred at 50 °C for
1
(s); H NMR (CDCl3) δ (ppm), 2.46 (s, 3H), 4.33 (s, 2H),
Vol. 11, No. 2, 2007 / Organic Process Research & Development
•
249

6.78 (s, 1H), 6.97 (t, 1H, J ) 6 Hz), 7.33 (d, 2H, J ) 9 Hz),
7.42 (t, 1H, J ) 7.5 Hz), 7.62 (d, 2H, J ) 9 Hz), 7.80 (d,
1H, J ) 6 Hz), 8.17 (s, 1H), 8.66 (d, 1H, J ) 9 Hz), 10.44
(s, 1H); 13C NMR (CDCl3) δ (ppm), 19.3, 32.4, 108.5, 109.8,
111.4, 115.6, 118.5, 119.9, 121.4, 124.0, 124.5, 127.1, 129.0,
132.8, 140.5, 141.8, 158.6, 160.8, 163.7, 174.9. ESMS calcd
for C22H16N4O2S, 400.1; found, 401.1 (M + H)+. Anal. Calcd
for C22H16N4O2S: C, 65.98; H, 4.03; N, 13.99; O, 7.99; S,
8.01. Found: C, 65.80; H, 4.12; N, 13.85; O, 8.06; S, 8.03.
3-(4-Cyanobenzyl)indolizine (5, Scheme 2). To a stirred
solution of N,N-diisopropylamine (62.0 mL, 0.22 mol) in
anhydrous THF (300 mL) in a one-neck round-bottom flask
(1 L) at -78 °C was added n-BuLi (122.0 mL, 3.6 M, 0.44
mol) over 10 min. The reaction temperature rose up to -65
°C. The mixture was stirred at -78 °C for 30 min. DMPU
(53.2 mL, 0.44 mol) was added in one portion. The resulting
mixture was stirred at -78 °C for 1 h. Some slurry was
observed. p-Tolunitrile (48.0 mL, 0.4 mol) was diluted with
THF (20 mL) and added dropwise to the above mixture at
-78 °C within 40 min. Stirring was continued for 1 h. The
reaction mixture became yellowish red. It took 20 min to
add propargyl bromide (80%, Aldrich, 22.2 mL, 0.2 mol) to
the reaction mixture, which was subsequently stirred at -78
°C for 1.5 h. Reaction was allowed to warm up to -40 °C
and quenched with saturated NH4Cl (120 mL), water (60
mL), and ethyl acetate (400 mL). In a 2 L separatory funnel,
the organic layer was separated and the aqueous layer was
extracted with ethyl acetate (2 × 200 mL). The combined
organic layers were washed with water (40 mL) and brine
(40 mL), dried over sodium sulfate, and concentrated in
vacuo to yield an oil residue, which gradually turned into a
yellow solid at room temperature. The solid was then washed
with hexanes to give 4-(butyn-3-yl)benzonitrile (28.6 g,
92.4%), which was used for the next step without further
purification (purity 97.4%, 230 nm HPLC). 1H NMR
(CDCl3): δ (ppm) 7.60 (m, 2H); 7.36 (m, 2H); 2.90 (t, 2H,
J ) 7.2 Hz, 2H); 2.51 (m, 2H); 1.99 (t, J ) 2.7 Hz, 1H).
ESMS calcd (C11H9N), 155.1; found, 156.1 (M + H)+. To
a 1 L round-bottom flask were added THF (250 mL),
4-(butyn-3-yl)benzonitrile (49.38 g, 0.318 mol), TEA (250
mL, 1:1 TEA/THF), bromopyridine (33.3 mL, 0.35 mol),
Pd (PPh3)2Cl2 (4.47 g, 6.0 mmol), and CuI (1.21 g, 6.0
mmol). The reaction mixture was stirred at 60 °C for 2 h.
The reaction was then cooled to room temperature and
quenched with water (100 mL) and ethyl acetate (100 mL).
The organic layer was separated and washed with water (50
mL) and brine (30 mL), dried over sodium sulfate, and
concentrated in vacuo to give a brown solid, which was
subsequently washed with heptane (2 × 50 mL) to afford
4-(4-(pyridin-2-yl)butyn-3-yl)benzonitrile as an off-white
solid (58.73 g, 79.6%) with a purity of 98.4% (HPLC, 230
nm). 1H NMR (CDCl3) δ (ppm) 8.54 (m, 1H); 7.59 (m, 3H);
7.39 (m, 2H); 7.31 (m, 1H); 7.19 (m, 1H); 3.01 (t, J ) 6.9
Hz, 2H); 2.76 (t, J ) 6.9 Hz, 2H). ESMS calcd (C16H12N2),
232.1; found, 233.1 (M + H)+. The crude product (58.73 g,
0.25 mol) was dissolved in DMA (490 mL) in a round-
bottom flask (1 L). To it were added TEA (70 mL, 1:7 TEA/
DMA) and CuCl (25.05 g, 0.25 mol). The resulting mixture
was stirred at room temperature for 10 min and then kept
stirring at 130 °C for an additional 3 h. After cooling to
room temperature, the reaction mixture was quenched with
water (300 mL) and ethyl acetate (300 mL). The dark brown
mixture was then filtered through Celite to give a clear two-
layer mixture. The organic phase was charged with decolor-
izing carbon (10 g) and stirred at room temperature for 20
min, followed by filtration through Celite and concentration
to give desired 3-(4-cyanobenzyl)indolizine as an oil, which
transformed into a solid after refrigeration (-4 °C) overnight
(52.63 g, 89.6%) with purity 94.9% (230 nm, HPLC).
Analytical data are identical to those of an authentic sample.
Received for review December 28, 2006.
OP6002852
250
•
Vol. 11, No. 2, 2007 / Organic Process Research & Development