Efficient synthesis of AMP579, a novel Adenosine A1/A2 receptor agonist

Full text of DOI:10.1021/jo001066f

8114
J . Org. Chem. 2000, 65, 8114-8118
Sch em e 1a
Efficien t Syn th esis of AMP 579, a Novel
Ad en osin e A1/A2 Recep tor Agon ist
Adam W. Sledeski,* Gregory G. Kubiak,
Michael K. O’Brien, Matthew R. Powers,
Tory H. Powner,† and Larry K. Truesdale‡
Process Chemistry, Rhoˆne-Poulenc Rorer,§
Collegeville, Pennsylvania
adam.sledeski@aventis.com
Received J uly 14, 2000
In tr od u ction
AMP579 (1) is a novel adenosine analogue with high
affinity for both the A1 (Ki ∼5 nM) and A2a (Ki ∼56 nM)
receptor subtypes and has recently been demonstrated
in a clinically relevant manner to have cardioprotective
effects in various animal models of myocardial infarc-
tion.1,2 It is currently under development for potential
use in cardiovascular therapeutics.3 Structurally, 1 is
distinguished from adenosine by: 5′-N-ethylcarboxamide
(NECA)4 replacement for the hydroxyl group, carbocyclic
modification of the ribose unit, and 1-deaza N6-substi-
tuted purine5,6 system.
a
Reagents: (a) 3, NEt3, iPrOH, 47%; (b) 5, NEt3, BuOH, 92%;
(c) H2, Pt, 87%; (d) formamidine acetate, BuOAc, 55%; (e) aq HCl,
THF, 79%; (f) HCl, iPrOH; (g) formamidine acetate.
The synthetic route used to generate preclinical mate-
rial supplies is summarized in Scheme 1.3 Assembly of
two optically pure synthons, thienylamine 35,7 and car-
bosugar derivative 5,8,9 was followed by a sequential
condensation of the commercially available 2,4-dichloro-
3-nitropyridine 2 with 3 and 5 and then reduction of
the nitro group and cyclization to the desired 1-deazapu-
rine heterocycle. Although straightforward and triply
convergent, this route suffered from some major draw-
backs. First, the preparation of 3 necessitated seven
steps, several of which presented scale-up challenges
(Scheme 2). Second, the initial condensation step was
poorly selective, producing the desired 4-substituted
pyridine 4 in ca. 50% yield along with the mixture of the
starting materials, 2-thienylamino regioisomer and the
Sch em e 2a
a
Reagents: (a) NaNO2, HCl, 95%; (b) BH3, THF, 90%; (c) KOH,
56%; (d) 3-chlorothiophene, BuLi, 81%; (e) MsCl, NEt3; (f) NaN3,
EtOH, 94%; (g) PPh3, toluene, 99%.
diaminated compound A. The monoaminated pyridine 4
could either be purified by chromatography prior to use
in the next step, or used crude, in which case impurities
B and C were formed. The second amine introduction
* Corresponding author. Current address: Aventis Pharmaceuticals,
N2342, P.O. Box 6800, Bridgewater, NJ 08807.
† Present address: Boehringer Ingelheim Chemicals, Inc., Peters-
burg, VA.
‡ Present address: Alanex Corp., San Diego, CA.
§ Currently: Aventis Pharmaceuticals.
(1) Merkel, L.; Rojas, C. J .; J arvis, M. F.; Cox, B. F.; Fink, C.; Smits,
G. J .; Spada, A. P.; Perrone, M. H.; Clark, K. L. Drug. Dev. Res. 1998,
45, 30-43.
(2) McVey, M. J .; Smits, G. J .; Cox, B. F.; Kitzen, J . M.; Clark, K.
L.; Perrone, M. H. J . Cardiovasc. Pharmacol. 1999, 33, 703-710.
(3) Spada, A. P.; Fink, C. A.; Myers, M. R. PCT Int. Appl. WO
9528160, 1995.
(4) Chen, J .; Grim, M.; Rock, C.; Chan, K. Tetrahedron Lett. 1989,
41, 5543-5546.
and subsequent reduction step proceeded efficiently,
however formation of the deazapurine system was again
impaired by a regioselectivity issue. A 5:1 ratio of the
2,3:3,4 (D) fused systems was obtained, from which the
desired 2,3 regioisomer was isolated by chromatography.
The hydrolysis of the acetonide moiety was then followed
by the isolation of AMP579 (1) as the amorphous HCl
salt. The overall yield starting from the dichloroami-
nopyridine 2 was 17%, and the overall yield from the
chiral amino acid 8 was only 6%. This route was
(5) Fink, C. A.; Spada, A. P.; Colussi, D.; Rivera, L.; Merkel, L.
Nucleosides Nucleotides 1992, 11, 1077-1088.
(6) For other examples of carbocyclic and N6-substituted nucleosides
in the adenosine area, see: Siddiqi, S. M.; J acobson, K. A.; Esker, J .
L.; Olah, M. E.; J i, X.-d.; Melman N.; Tiwari, K. N.; Secrist, J . A., III;
Schneller, S. W.; Cristalli, G.; Stiles, G. L.; J ohnson, C. R.; Ijizerman,
A. P. J . Med. Chem. 1995, 38, 1174-1188 and references therein.
(7) Grondard, L.; Casimir, J .-P.; Leon, P.; O’Brien, M. K.; Powers,
M. R.; Robin, D. PCT Int Appl. WO 9811064, 1998.
(8) Bannister, R..; Hanson, C.; Henderson, N.; McCague, R.; Rue-
croft, G. Org. Process Res. Dev. 1997, 1, 415-419.
(9) O’Brien, M.; Leon, P.; Largeau, D.; Powers, M.; Durand, T. US
patent 5684159, 1997.
10.1021/jo001066f CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/24/2000

Notes
J . Org. Chem., Vol. 65, No. 23, 2000 8115
Sch em e 3a
fluxing acetonitrile for 27 h, the desired 4-thienylami-
nopyridine derivative 12 was formed in ca. 65% yield.
Equally encouraging was the lack of evidence at this
stage for the formation of the undesired 2-regioisomer17
or the diaminated pyridine. It is noteworthy that replac-
ing 10 with the corresponding trifluoromethylsulfon-
amide led to extensive â-elimination, producing the olefin
E and its double bond isomer in a total of 40% yield.
Although methane- and benzene- sulfonamides under-
went condensation with 11 with similar efficiency, unlike
the toluenesulfonamide, they did not offer crystallinity
in the starting material or product. At this point, we
decided to focus on using 10 as nucleophile and 11 as
electrophile and briefly screened some alternative reac-
tion conditions. The reaction time was drastically reduced
by the addition of a catalytic amount of 18-crown-6, or
by employing potassium tert-butoxide in THF. The latter
system turned out to be the most effective, allowing the
isolation of crystalline 12 in 83% yield. Although use of
isolated 2,4-difluoro-3-nitropyridine led to superior re-
sults in terms of impurity profile, from the process
development point of view in situ fluorination of 2
followed by the condensation was a viable and attractive
alternative. The fully substituted pyridine intermediate
13 was obtained quantitatively by condensing 12 with
the carbosugar synthon 5 in the presence of potassium
carbonate in N-methylpyrrolidinone (NMP) at 80 °C for
1.5 h. Other reaction conditions including potassium
carbonate in acetonitrile or in ethyl acetate, triethyl-
amine in acetonitrile or diisopropylethylamine in NMP
at 70-80 °C for 1-2 h worked equally well. Reduction
of the nitro functionality was then achieved via hydro-
genation of 13 in the presence of a catalytic amount of
platinum on carbon, affording crude triamine 14 in
quantitative yield. Although purification via flash chro-
matography or recrystallization yielded pure triamine 14
in 95 and 73% respectively, the crude material could be
used directly in the next step (Scheme 4).
a
Reagents: (a) 2.1 equiv of TsCl, aq NaOH, MTBE, 88%; (b)
3-chlorothiophene, BuLi, 90%.
subsequently rendered more practical, after the identi-
fication of a crystalline HCl salt of the triamine 7, and
ultimately a crystalline form of AMP579 itself,10 enabling
the elimination of all chromatographic purification steps.
Nevertheless, despite extensive optimization, the chemi-
cal yield of the first amination step did not exceed 60%
and the cyclization proceeded in a 9:1 ratio, indicating
inherent limitations of this approach.11 Attempts to
prevent the unwanted cyclization by protecting the 4-NH
functionality in 4 failed due to its inertness toward
common electrophiles such as Ac2O, Boc2O, or TsCl.
An alternative preparation of thienylamine 3 from
commercially available and inexpensive (R)-(-)-2-amino-
1-butanol 9 would only require three steps (Scheme 3).7
While aziridine formation and the ring opening by the
chlorothienyl anion were straightforward, removal of the
toluenesulfonyl group was problematic.12 Since sulfona-
mides of secondary amines are generally cleaved more
readily than those of primary amines,13 we were intrigued
by the idea of reversing the order of the detosylation and
aromatic nucleophilic substitution steps. Apart from a
net savings of four synthetic steps, this approach would
secure the regiospecific formation of the deazapurine if
the detosylation was delayed past the cyclization step.
The toluenesulfonamide moiety was also likely to impart
crystallinity13 to a larger number of intermediates, which
could lead to practical and scalable isolation protocols.
Resu lts a n d Discu ssion
To the best of our knowledge, N-alkylsulfonamides
have not been previously used as nucleophiles in aromatic
nucleophilic substitution.14 Our attempts to react the
tosylaziridine 10 with 2 using a variety of bases did not
result in any detectable substitution. The required
breakthrough occurred only when the electrophile was
switched to the more reactive 2,4-difluororo-3-nitropyri-
dine 11.15,16 Using potassium carbonate as base in re-
The stage was now set for the regiospecific formation
of the requisite deazapurine ring system. Whereas the
parent system 7 (or 8) underwent formamidine acetate
promoted closure in less than 4 h at 90 °C in n-butyl
acetate, the tosylated derivative 14 required over 40 h
at 120 °C to reach approximately 70% conversion when
the same reagent/solvent system was employed. Fortu-
nately, an alternative system consisting of a 1:1 (v/v)
mixture of triethylorthoformate and acetic anhydride
allowed complete conversion in 1 h at 120 °C.18 The pure
deazapurine 15 was isolated directly from the reaction
mixture as crystalline solid in 75% yield based on the
crude material from the preceding step.
(10) O’Brien, M. K.; Garcia, H.; Leon, P.; Powner, T.; Reilly, L. W.;
Shah, H. C.; Thompson, M. D.; Tsuei, C. T.; Vanasse, B. J .; Walther,
F. PCT Int Appl. WO 9825921, 1998.
(11) An alternative route avoiding the regioselectivity issue in the
first amination step is described in ref 10.
(12) Subsequently, a practical detosylation method was developed.7
(13) Greene T. W.; Wuts P. G. M. Protective Groups in Organic
Synthesis, 3rd ed.; J ohn Wiley & Sons: New York 1999; pp 603-616.
(14) (a) The closest related literature case involves stoichiometric
complexes of Na- and K-phthalimides and saccharinates: Rasshofer,
W.; Voegtle, F. Tetrahedron Lett. 1979, 1217-18. Other related
processes: (b) Goldberg coupling reaction between sulfonamide and
aryl bromide: Hall, R. J .; Marchand, J .; Oliveira-Campos, A. M. F.;
Queiroz, M.-J . R. P.; Shannon, P. V. R. J . Chem. Soc. Perkin Trans. 1
1992, 3439-3450. (c) Coupling of sulfonamide with phenylboronic acid
in the presence of cupric aceate: Chan, D. M. T.; Monaco, K. L.; Wang,
R.-P.; Winters, M. P. Tetrahedron Lett. 1998, 39, 2933-2936.
(15) Kroon, C.; Maaseeen van den Brink, A.; Vlietstra, E. J .;
Salemink, C. A.. Recl. Trav. Chim. Pays-Bas 1976, 95, 127-156. A
modified procedure for the large scale prep of 11 is described in the
Experimental Section.
Liberation of AMP579 (1) was then accomplished in
two steps. First, the toluenesulfonyl protective group was
removed using magnesium in a methanol/toluene mix-
ture. The reaction required 5 equiv of magnesium and
was completed in 3 h at 40 °C. At higher temperatures
(17) An indirect evidence for the formation of the small amount of
the unwanted 2-regioisomer during the condensation of 2 with 10 was
obtained after the next step. When crude (untriturated) 4 was used in
condensation with 5, a ca. 2A% of an isomer of 5 was detected by LC-
MS in the crude reaction mixture.
(18) The likely active reagent is diethoxymethyl acetate (DEMA):
Montgomery, J . A.; Fitzgibbon, W. E., J r. In Nucleic Acid Chemistry
vol. 2; Townsend, L. B., Tipson, R. S., Eds.; J ohn Wiley and Sons: New
York, 1978; pp 995-997. Neat DEMA (Aldrich) at 70-80 °C also
allowed ring closure of 14 to 15.
(16) For a recent example of a use of aryl fluoride in aromatic
nucleophilic substitution with a secondary alkylamine, see: Brown,
G. R.; Foubister, A. J .; Ratcliffe, P. D. Tetrahedron Lett. 1999, 40,
1219-1222.

8116 J . Org. Chem., Vol. 65, No. 23, 2000
Notes
ing the workup and crystallization,10 the desired AMP579
(1) was isolated as a free base in 75% yield.23
Sch em e 4a
In summary, we have demonstrated an efficient, highly
regioselective route to the adenosine agonist AMP579.
The route involves an unprecedented condensation of an
alkylsulfonamide with a pyridine synthon, then takes
advantage of the sulfonamide protecting group to prevent
the undesired cyclization pathway during the deazapu-
rine ring formation. Starting from the (R)-(-)-2-amino-
1-butanol (9), this process affords AMP579 in eight steps
and 35% overall yield, whereas the previous one neces-
sitated 12 steps from the amino acid 8 and the overall
yield was 6%. In addition, the approach described herein
allows concatenation of multiple steps, while offering the
flexibility of implementing isolation steps by means of
crystallization. The requisite cleavage of the N-Ts bond
was facilitated by the aryl substitution and was ac-
complished using magnesium in methanol or Super-
Hydride in THF. The latter constitutes a novel detosy-
lation protocol and should find further applications, since
it is well recognized that existing methods are substrate
dependent.24,25
a
Reagents: (a) KF, 18-c-6, NMP, 83%; (b) 10, t-BuOK, THF,
83%; (c) 5, K2CO3, NMP, 99%; (d) H2, Pt, EtOAc/MeOH, 97%; (e)
HC(OEt)3, Ac2O, 84%; (f) Mg, EtOH, 99% or LiEt3BH, THF, 96%
(crude); (g) concd HCl, THF, 75%.
Exp er im en ta l Section
Gen er a l Meth od s. All substrates and reagents were obtained
commercially and used without purification. Nuclear magnetic
resonance (NMR) chemical shifts (δ) are reports in parts per
million (ppm) downfield from tetramethylsilane (TMS) with
reference to internal solvent [1H NMR, CHCl3 (7.25), CHD2-
SOCD3 (2.50); 13C NMR, CHCl3 (77.0), CD3SOCD3 (39.43)]. Flash
chromatography was carried out using Merck Kiesegel 60 F254
(230-400 mesh). HPLC analysis was performed using following
conditions: Kromasil C8 column (250 × 4.6 mm), flow 1 mL/
min, eluent: acetonitrile 70%, water 20%, 20 mM potassium
phosphate buffer, pH 3.0, 10%, detection: UV, 230 nm. HPLC
results are presented as area percent of the peak (A%) for a
particular compound relative to the total area of all the peaks
integrated. Melting points are uncorrected.
more magnesium was necessary due to the competitive
formation of magnesium methoxide.19 The estimated
reaction yield was 84% (99% yield of crude material with
a purity of 85%) and the principal side reaction was the
cleavage of the N-C(aryl) bond leading to the release of
the tosylthienylamine 10 and the deazapurine analogue
F . Alternatively, the deprotection was cleanly achieved
2,4-Diflu or o-3-n itr op yr id in e15 (11). A mixture of 300 g
(1.56 mol, 1 equiv) of 2,4-dichloro-3-nitropyridine (2), 1.0 L of
1-methyl-2-pyrrolidinone, 270.5 g (4.65 mol, 2.97 equiv) of
potassium fluoride, and 66.0 g (0.25 mol, 0.16 equiv) of 18-
crown-6 was heated with stirring at 100 °C for 2 h. The reaction
mixture was cooled to 20 °C, and then it was partitioned between
1.75 L of water and 1.75 L of tert-butyl methyl ether (TMBE).
The organic layer was washed with 300 mL of brine, dried over
magnesium sulfate, and concentrated under reduced pressure.
11 (208 g, 83% yield) was obtained as a brown oil with a purity
of 97 A% (HPLC). The oil may be used as is or distilled. A 42.61
g portion of the 2,4-difluoro-3-nitropyridine was distilled at 61-
64 °C and 4.0 Torr to give the title compound (38.15 g, 90% yield,
with lithium triethylborohydride (Super-Hydride)20,21 in
THF at 4 °C to room temperature. We are unaware of
any precedent for sulfonamide cleavage using this re-
agent. In contrast, attempts to use the bis(2-methoxy-
ethoxy)aluminum hydride (Red-Al)22 in toluene at tem-
peratures between 40 and 108 °C led to mixtures of
starting material and several products including only
minor amounts of the desired material. Lithium and
sodium naphthalenides13 were also found to be ineffec-
tive. It is noteworthy that when either magnesium or
Super-Hydride protocols were applied to thienylsulfona-
mide substrate 10, no deprotection product was detected.
Finally, the acetonide moiety in 16 was hydrolyzed in
an aqueous hydrochloric acid/THF mixture, then follow-
1
99.6 A% (HPLC)) as a colorless oil. H NMR (200 MHz, CDCl3):
δ 8.45 (1H, app t), 7.28 (1H, dd). MS (EI, 70 eV, relative
intensity): 160 (M+, 75), 114 (25). Alternatively, the reaction may
be performed with 0.3 equiv of tetrabutylammonium bromide,
in place of 16-crown-6, in toluene at 110 °C.
(R)-N-[1-[(3-Ch lor oth ien -2-yl)m eth yl]pr opyl]-N-(2-flu or o-
3-n itr op yr id -4-yl)-4-m eth ylben zen esu lfon a m id e (12). To
the solution of (R)-N-[1-[(3-chlorothien-2-yl)methyl]propyl]-4-
methylbenzenesulfonamide (10)5,7 (43.55 g, 0.127 mol) in 71 mL
of tetrahydrofuran (THF) at 4 °C was added dropwise 1.0 M
solution of potassium tert-butoxide in tetrahydrofuran (138 mL,
0.138 mol). The resulting mixture was stirred for 15 min at 20
°C, and then 23.7 g (0.149 mol) of 2,4-difluoro-3-nitropyridine
(19) In a control experiment 15 was treated with a 10% magnesium
methoxide in methanol solution between room temperature and boiling
point. No detosylation product under these conditions was detected.
(20) Brown, H. C.; Kim, S. C.; Krishnamurthy, S. J . Org. Chem.
1980, 45, 1-12.
(21) For other synthetic applications of this reagent, see: Zaidlewicz,
M.; Brown, H. C. In Encyclopedia of Reagents for Organic Synthesis;
Paquette, L. A., Ed.; Wiley: Chichester, 1995; Vol. 5, pp 3180-3182.
(22) Gold, E. H.; Babad, E. J . Org. Chem. 1972, 37, 2208.
(23) Promising results were also obtained using a one step detosy-
lation/acetonide hydrolysis protocols such as methanesulfonic acid (see
the Acknowledgment) or trimethylsilyl trifluoroacetate in conjunction
with trifluoroacetic acid and thioanisole at 70-85 °C. The yields based
on compound 15 ranged between 50 and 70%.
(24) Vedejs, E.; Lin, S. J . Org. Chem. 1994, 59, 1602-1603.
(25) An, H.; Cook, P. D. Tetrahedron Lett. 1996, 40, 7233.

Notes
J . Org. Chem., Vol. 65, No. 23, 2000 8117
in 10 mL of tetrahydrofuran was added and the mixture was
heated at 65 °C for 1 h. The mixture was cooled to room
temperature and partitioned between 150 mL of water and 300
mL of ethyl acetate. The organic layer was washed with water
(2 × 150 mL) and brine (100 mL), dried (MgSO4), and concen-
trated under reduced pressure. The residue was successively
slurried with 100 mL of methanol, 150 mL of heptane, and 60
mL of methanol and vacuum-dried at room temperature with a
nitrogen bleed to afford the title compound (12) (51.1 g, 83%
yield, 97.6 A% (HPLC)) as a white solid. MS (ion spray) (relative
intensity): 484/486 (75), 312 (100).
m), 4.78 (m), 4.60 (m), 4.39 (m), 3.99 (s), 3.75 (s, min. conf.), 3.27
(dt), 3.12 (dd, min. conf.), 2.90-2.45 (m), 2.42 (s), 2.41 (1s, min.
conf.), 2.04 (apparent t), 1.81 (s), 1.48 (s), 1.47 (s, min. conf.),
1.55-1.22 (m), 1.30 (s), 1.29 (s, min. conf.), 1.16 (t), 1.15 (t, min.
conf.), 1.04 (t), 0.73 (t). 13C NMR (50 MHz, CDCl3) (mixture of
conformational isomers, all signals reported): δ 174.97, 174.84,
149.88, 149.53, 143.77, 143.60, 137.42, 136.47, 136.32, 133.45,
131.55, 131.08, 129.66, 129.38, 127.94, 127.73, 127.59, 125.99,
125.88, 123.59, 123.23, 123.10, 114.47, 114.05, 110.71, 86.36,
85.94, 85.13, 84.72, 64.46, 64.20, 56.90, 56.75, 53.99, 53.74, 34.92,
34.83, 32.58, 32.13, 31.23, 26.84, 26.78, 26.05, 24.96, 24.44, 21.59,
14.70, 12.57, 12.08.
A small sample of the crude product was recrystallized from
THF/CH3CN/H2O. Mp: 139 °C. 1H NMR (200 MHz, CDCl3): δ
8.30 (1H, d broad), 7.73 (2H, d broad), 7.38-7.15 (3H, m), 7.16
(1H, d), 6.82 (1H, d), 3.95 (1H, m), 3.27 (1H, dd), 3.02 (1H, dd),
2.45 (3H, s), 1.6 (2H, m broad), 0.70 (3H, t broad). 13C NMR (50
MHz, CDCl3): δ 157.565, 152.696, 148.771, 148.445, 145.131,
132.348, 129.945, 128.224, 127.757, 125.370, 125.267, 123.966,
123.768, 67.078, 32.346, 26.081, 21.648, 11.617. Anal. Calcd for
C20H19ClFN3O4S2: C, 49.64; H, 3.96; N, 8.68. Found: C, 49.68;
H, 3.96; N, 8.76.
[3a R-[3a r,4r,6a (R*),6a r]]-6-[7-[[1-[(3-ch lor ot h ien -2-yl)-
m eth yl]p r op yl] [4-Meth ylben zen esu lfon yl]a m in o]-3H-im i-
d a zo[4,5-b]p yr id -3-yl] N-Eth yl Tetr a h yd r o-2,2-d im eth yl-
4H-cyclop en ta -1,3-d ioxole-4-ca r boxa m id e (15). A mixture
of the crude triamine 14 (94.5 g, 0.13 mol), triethylorthoformate
(141.0 mL, 0.85 mol), and acetic anhydride (114.0 mL, 1.21 mol)
was heated at 100-105 °C for 2 h. The reaction mixture was
cooled to room temperature, diluted with 1.0 L of heptane, and
stirred for 1 h. The solids formed were isolated by filtration,
washed with cold heptane (500 mL), and dried under vacuum
with a nitrogen bleed for 2 days to provide the title compound
as a white solid (65.0 g, 75% yield, 97.0 A% (HPLC)). MS (ion
[3a R-[3a r,4r,6a (R*),6a r]]-6-[4-[[1-[(3-Ch lor oth ien -2-yl)-
m eth yl]p r op yl] [4-Meth ylben zen esu lfon yl]a m in o]-3-n itr o-
p yr id -2-yla m in o] N-E t h yl Tet r a h yd r o-2,2-d im et h yl-4H -
cyclop en ta -1,3-d ioxole-4-ca r boxa m id e (13). A mixture of the
monoaminated fluoronitropyridine 12 (123.0 g, 0.254 mol), 3aR-
[3aR,4R,6a,6aR]-6-amino-N-ethyltetrahydro-3,3-dimethyl-2,4-di-
oxabicyclo [3.3.0]octane-8-carboxamide benzoate (5)8,9 (123.0 g,
0.254 mol), potassium carbonate (325 mesh) (105.3 g, 0.762 mol),
and 1-methyl-2-pyrrolidinone (1.0 L) was heated at 85 °C with
stirring for 1.5 h. The reaction mixture was cooled to room
temperature and poured into 2.5 L of water with stirring. Solid
that formed was isolated by filtration, and the filtercake was
washed with 10 L of water. The filtercake was air-dried on the
filter overnight and then dried under vacuum with a nitrogen
bleed for 4 days to provide the title compound (179 g, 100%, 98
A% (HPLC)) as a yellow solid. Mp 90-113 °C. HRMS (FAB):
calcd for C31H38ClN5O7S2 (M + H)+ 692.1979, found 692.1931.
1H NMR (500 MHz, CDCl3) (mixture of conformational isomers,
ratio ca. 7:3, all signals reported): δ 8.33 (d), 8.25 (d, minor
conformer), 7.94 (d), 7.87 (d), 7.76 (d), 7.34 (d), 7.30 (d), 7.13
(m), 6.81 (m), 6.56 (d), 6.47 (d, min. conf.), 5.66 (m), 4.81 (d, min.
conf.), 4.74 (m), 4.58 (d, min. conf.), 4.56 (d), 3.99 (m), 3.87 (m),
3.35 (m), 3.21 (dd, min. conf.), 3.10 (dd), 2.90 (dd), 2.82 (dd), 2.55
(m), 2.41 (s), 2.13 (d), 1.96 (d, min. conf.), 1.64 (m), 1.60 (s), 1.52
(s), 1.30 (s), 1.22 (s), 1.40-1.20 (m), 1.15 (t), 0.88 (t), 0.64 (t,
min. conf.), 0.51 (t). 13C NMR (50 MHz, CDCl3) (mixture of
conformational isomers, as above): δ 173.781, 151.181, 144.504,
139.049, 135.972, 133.204, 129.715, 128.772, 128.089, 127.654,
123.641, 123.411, 114.768, 110.787, 86.284, 84.833, 64.913,
57.380, 53.328, 34.788, 32.361, 31.085, 30.680, 26.692, 24.297,
21.632, 14.718, 11.395.
1
spray): 672/674 (100). H NMR (200 MHz, CDCl3): δ 8.38 (1H,
d), 8.15 (1H, s), 7.85 (2H, d), 7.25 (1H, s), 7.22 (1H, d), 7.15 (2H,
dd), 6.78 (1H, d), 5.85 (1H, t), 5.05 (3H, m), 4.25 (1H, m), 3.28
(3H, m), 2.85 (3H, m), 2.60 (1H, q), 2.40 (3H, s), 1.60 (3H, s),
1.45 (2H, q), 1.25 (3H, s), 1.20 (3H, t), 0.90 (3H,t). 13C NMR (500
MHz CDCl3): δ 171.36, 149.04, 144.51, 143.53, 142.90, 137.70,
136.37, 135.34, 133.86, 129.45, 128.11, 127.46, 123.39, 123.21,
121.08, 114.22, 83.86, 81.88, 64.99, 60.69, 50.24, 34.63, 34.38,
33.35, 27.45, 25.26, 25.07, 21.58, 14.81, 11.64.
A sample of the crude material was recrystallized from ethyl
acetate/CH3CN. Mp: 197-199 °C. The structure of [3aR-
[3aR,4R,6a(R*),6aR]]-6-[7-[[1-[(3-chlorothien-2-yl)methyl]pro-
pyl] [4-methylbenzenesulfonyl]amino]-3H-imidazo[4,5-b]pyrid-
3-yl] N-ethyl tetrahydro-2,2-dimethyl-4H-cyclopenta-1,3-dioxole-
4-carboxamide was confirmed by single-crystal X-ray analysis
(see the Supporting Information).
[3a R-[3a r,4r,6a (R*),6a r]]-6-[7-[[1-[(3-Ch lor oth ien -2-yl)-
m eth yl]p r op yl]a m in o]-3H-im id a zo[4,5-b]p yr id -3-yl] N-Eth -
yl Tet r a h yd r o-2,2-d im et h yl-4H -cyclop en t a -1,3-d ioxole-4-
ca r boxa m id e (16). A. Usin g Ma gn esiu m in Meth a n ol. One
or two small crystals of iodine were added to a mechanically
stirred suspension of magnesium turnings (3.9 g, 0.16 mol) in
methanol (30 mL) under nitrogen. Bubbles rising from the
magnesium were observed as the mixture started to warm. A
solution containing tosylated compound 15 (20 g, 0.03 mol) in a
mixture of 65 mL of methanol and 55 mL of toluene was added
over 20 min, while the reaction temperature was maintained at
∼40 °C. The mixture was stirred at ∼40 °C for approximately 3
h, and then it was cooled to room temperature, concentrated to
one half of the original volume, and filtered through ca. 50 g of
Celite. The filter was washed with 150 mL of ethyl acetate, and
the combined filtrate was washed with 150 mL of water and
150 mL of brine. The organic phase was dried (MgSO4) and
concentrated under reduced pressure to provide the title com-
pound (15.4 g, 99% (crude), 85 A%, (HPLC)). MS (FAB): 518/
520 (100). 1H NMR (200 MHz, CDCl3): δ 8. 02 (1H, d), 7.90 (1H,
s), 7.09 (1H, d), 6.85 (1H, d), 6.34 (1H, d), 6.16 (1H, m), 5.32
(1H, d broad), 5.10-4.85 (3H, m), 3.90 (1H, m), 3.33 (2H, m),
3.09 (2H, d), 2.95-2.50 (3H, m), 2.0-1.5 (2H, m), 1.57 (3H, t),
1.31 (3H, s), 1.17 (3H, t), 1.05 (3H, t).
B. Usin g Lith iu m Tr ieth ylbor oh yd r id e. A solution of
lithium triethylborohydride (1.0 N in THF, 18.6 mL, 1.86 mmol)
was added dropwise to a magnetically stirred solution containing
tosylated compound 15 (5.0 g, 7.43 mmol) in 66 mL of THF at
about 4 °C under nitrogen. Following the end of the addition,
the mixture was stirred at room temperature for 2 h. HPLC
showed e3 A% of starting material and ca. 90 A% of the
detosylated material. Thirty millilters of water and 20 mL of
ethyl acetate were added, and the phases were separated. The
organic phase was washed with water and with brine, dried
(MgSO4), and concentrated under reduced pressure to provide
the title compound (3.7 g, 96% (crude), 85 A% (HPLC)) as a light
yellow solid.
[3a R-[3a r,4r,6a (R*),6a r]]-6-[4-[[1-[(3-ch lor ot h ien -2-yl)-
m eth yl]p r op yl] [4-Meth ylben zen esu lfon yl]a m in o]-3-a m i-
n op yr id -2-yla m in o] N-Eth yl Tetr a h yd r o-2,2-d im eth yl-4H-
cyclop en ta -1,3-d ioxole-4-ca r boxa m id e (14). A solution of the
diaminated nitropyridine 13 (90.0 g, 0.13 mol) in 1.1 L of ethyl
acetate was washed with water (3 × 500 mL), and then it was
diluted with 300 mL of methanol and hydrogenated in the
presence of platinum on carbon (8.8 g, Degussa type F101 RA/
W, 5% Pt, 50% water) under 50-55 psi of hydrogen for 24 h at
room temperature. The reaction mixture was filtered through a
glass microfiber filter, and the filter was washed with 300 mL
of hot ethyl acetate. The combined filtrates were evaporated
under reduced pressure to give a solid which was then slurried
with 800 mL of heptane and isolated by filtration. The filtercake
was dried under vacuum with a nitrogen bleed for 2 days to
provide the title compound (94.5 g, 109% (crude), 89 A% (HPLC),
97% corrected yield).
A small sample of the oil was triturated with Et2O to give
the title compound as an off-white solid. Mp: 175-177 °C.
HRMS (FAB): calcd for C31H41ClN5O5S2 (M + H)+ 662.2237,
found 662.222. 1H NMR (200 MHz, CDCl3) (mixture of confor-
mational isomers, ratio ca. 7:3, all signals reported): δ 7.75 (d),
7.63 (d), 7.54 (d), 7.27 (d), 7.21 (s), 7.15 (dd), 6.82 (dd), 6.64 (d),
6.48 (d, minor conformer), 5.95 (d, min. conf.), 5.83 (d), 5.80 (br.

8118 J . Org. Chem., Vol. 65, No. 23, 2000
Notes
[1S-[1r,2â,3â,4r(S*)]]-4-[7-[[1-[(3-Ch lor oth ien -2-yl)m eth -
yl]p r op yl]a m in o]-3H-im id a zo[4,5-b]p yr id -3-yl] N-Eth yl 2,3-
Dih yd r oxycyclop en ta n eca r boxa m id e (1). A 9.4 mL portion
of concentrated (37%) HCl was added to a magnetically stirred
mixture of crude detosylated acetonide 16 (14.2 g, ca. 0.027 mol)
in THF (32 mL) over 10 min while the mixture temperature was
maintained below 35 °C. Following the end of the addition, the
mixture was stirred for 3 h at room temperature, after which
time a precipitated solid was isolated by filtration. The solid was
washed with cold (5-10 °C) ethyl acetate and dried to constant
weight to give the title compound as its bis HCl salt. The solid
was partitioned between 100 mL of n-butyl acetate and 50 mL
of a saturated aqueous sodium carbonate solution. The organic
phase was washed with water and brine, cooled to room
temperature, and stirred for 16 h. The reaction mixture was
filtered and the solid was dried at 40-50 °C under reduced
pressure to give the title compound (10 g, 75% yield) as the
hemihydrate. Anal. Calcd for C22H28ClN5O3S‚0.5(H2O): C, 54.26;
H, 6.00; N, 14.38; Cl, 7.28. Found: C, 54.06; H, 5.94; N, 14.38;
m), 2.39 (1H, dt), 2.09 (1H, dt), 1.63 (2H, m), 1.05 (3H, t), 0.92
(3H, t). 13C NMR (50 MHz, DMSO-d6): δ 172.73 (C), 147.42 (C),
146.12 (C), 144.37 (CH), 138.74 (CH), 134.079 (C), 126.671 (CH),
124.516 (C), 122.779 (C), 121.626 (CH), 96.938 (CH), 75.073
(CH), 73.07 (CH), 58.61 (CH), 54.20 (CH), 48.81 (CH), 33.45
(CH2), 32.19 (CH2), 30.06 (CH2), 27.23 (CH2), 14.66 (CH3), 10.40
(CH3).
Ack n ow led gm en t. We thank our colleagues Luc
Grondard, Benoˆıt Viguier, and Laurence Pailleres-
Hubert from Rhoˆne-Poulenc Rorer, France, for their
contributions expanding the scope of this work, and in
particular for developing an alternative in situ tetrabu-
tylammonium bromide promoted fluorination/amination
and methanesulfonic acid/TFA/thioanisole detosylation
procedures.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
compounds 12-16; single-crystal X-ray data for compound 15.
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
Cl, 7.28. MS (ion spray): 478/480 ((M + H)+, 100), 346 (55). H
NMR (200 MHz, DMSO-d6): δ 8.23 (1H, s), 8.06 (1H, t), 7.84
(1H, d), 7.42 (1H, d), 6.96 (1H, d), 6.63 (1H, broad d), 6.34 (1H,
d), 5.16 (1H, broad d), 4.94(1H, d), 4.76 (1H, q), 4.33 (1H, m),
4.11 (1H, q), 3.12 (2H, q), 3.07 (2H, m), 2.73 (1H, m), 2.51 (1H,
J O001066F