Practical asymmetric synthesis of a potent Cathepsin K inhibitor. Efficient palladium removal following Suzuki coupling

Article abstract of DOI:10.1021/jo0205614

A large-scale, chromatography-free synthesis of a potent and selective Cathepsin K inhibitor 1 is reported. The key asymmetric center was installed by addition of (R)-pantolactone to the in situ-generated ketene 4a. The final step of the convergent synthe

Full text of DOI:10.1021/jo0205614

P r a ctica l Asym m etr ic Syn th esis of a P oten t Ca th ep sin K In h ibitor .
Efficien t P a lla d iu m Rem ova l F ollow in g Su zu k i Cou p lin g
Cheng-yi Chen,† Philippe Dagneau,‡ Edward J . J . Grabowski,† Renata Oballa,§ Paul O’Shea,*,‡
Peppi Prasit,§ J oe¨l Robichaud,§ Rich Tillyer,† and Xin Wang‡
Department of Process Research and Department of Medicinal Chemistry, Merck Frosst Centre for
Therapeutic Research, P.O. Box 1005, Pointe-Claire-Dorval, Que´bec, H9R 4P8, Canada, and Department
of Process Research, Merck Research Laboratories, P.O. Box 2000, Rahway, New J ersey 07065
paul_oshea@merck.com
Received August 26, 2002
A large-scale, chromatography-free synthesis of a potent and selective Cathepsin K inhibitor 1 is
reported. The key asymmetric center was installed by addition of (R)-pantolactone to the in situ-
generated ketene 4a . The final step of the convergent synthesis of 1 was completed via Suzuki
coupling of aryl bromide 7a with unprotected aryl piperazine boronic acid 13. Residual palladium
and iron generated in the Suzuki coupling were efficiently removed from crude 1 via a simple
extractive workup using lactic acid.
In tr od u ction
as a potent, selective, and reversible inhibitor of Cathe-
psin K.6 To enable further study of the pharmacological
properties of this compound, we sought an efficient,
chromatography-free, large-scale synthesis suitable for
the preparation of bulk quantities of 1. We envisioned
that the compound could be assembled in a highly
convergent manner via Suzuki coupling of an aryl pip-
erazine boronic acid 10 with an aromatic halide 7 to form
the biaryl linkage. The key asymmetric center could be
introduced by deracemization of acid 3 via addition of a
chiral alcohol to the corresponding in situ-generated
ketene 4 (Scheme 1).
Bone is a dynamic living tissue that undergoes con-
stant remodeling throughout adult life. This equilibrium
process begins with resorption of old bone by osteoclasts
followed by formation of new bone by osteoblasts. Os-
teoporosis is a disease caused by an imbalance in the bone
remodeling process leading to bone loss due to excessive
resorption. In the United States, 10 million people have
the disease and an estimated 34 million are at risk due
to low bone mass. Osteoporosis also accounts for more
than 1.5 million fractures each year.1 Cathepsin K is a
cysteine protease and a member of the papain superfam-
ily. It is highly expressed in osteoclasts, the cells respon-
sible for bone resorption.2,3 Cathepsin K has been impli-
cated in the bone resorption process due to its localization4
and because it is one of the few proteases that can
efficiently hydrolyze native collagen, which makes up
90% of the protein in bone.5 Selective inhibitors of
Cathepsin K could therefore be potential therapeutic
agents for the treatment of diseases characterized by
excessive bone loss such as osteoporosis.
Resu lts a n d Discu ssion
Racemic 2-alkyl-arylacetic acids have been the subject
of extensive research, and a number are commercially
available as antiinflammatory and analgesic drugs.7 The
asymmetric synthesis of alkyl-arylacetic acids has also
received considerable attention, and many elegant ap-
proaches have been developed.8 Following an earlier
discovery in our laboratories,9 we envisioned that the
asymmetric center of 1 could be installed via chiral
protonation of the in situ-generated alkyl-aryl ketene 4a
with (R)-pantolactone (Scheme 2).
Thus, commercially available 3-bromophenyl acetic
acid 2 was alkylated with 1-iodo-2-methylpropane in THF
using LHMDS as a base to provide 3a . The order of
reagent addition proved to be crucial for achieving a high
yield. For example, treatment of a THF solution of acid
As part of an ongoing drug discovery program at our
laboratories, substituted biphenyl 1 has been identified
† Department of Process Research, Merck Research Laboratories.
‡ Department of Process Research, Merck Frosst Centre for Thera-
peutic Research.
§ Department of Medicinal Chemistry, Merck Frosst Centre for
Therapeutic Research.
(1) “Osteoporosis Overview” Fact Sheet; National Institutes of
Health Osteoporosis and Related Bone Diseases National Resource
Center: Washington, DC, February, 2002.
(2) Bossard, M. J .; Tomaszek, T. A.; Thompson, S. K.; Amegadzie,
B. Y.; Hanning, C. R.; J ones, C.; Kurdyla, J . T.; McNulty, D. E.; Drake,
F. H.; Gowen, M.; Levy, M. A. J . Biol. Chem. 1996, 271, 12517.
(3) Bromme, D.; Okamoto, K.; Wang, B. B.; Biroc, S. J . Biol. Chem.
1996, 271, 2126.
(4) Yamaza, T.; Goto, T.; Kamiya, T.; Kobayashi, Y.; Sakai, H.;
Tanaka, T. Bone 1998, 23, 499.
(5) Garnero, P.; Borel, O.; Byrjalsen, I.; Ferras, M.; Drake, F. H.;
McQueney, M. S.; Foged, N. T.; Delmas, P. D.; Delaisse, J . M. J . Biol.
Chem. 1998, 273, 32347.
(6) Oballa, R.; Prasit, P.; Robichaud, J .; Isabel, E.; Mendonce, R.;
Venkatraman, S.; Setti, E.; Wang, D.-X. WO 01/49288. IC50 ) 3 nM.
(7) Ferreira, S. H.; Vane, J . R. Antiinflammatory Drugs; Springer-
Verlag: New York, 1979; p 321.
(8) Castaldi, G.; Cavicchioli, S.; Giordano, C.; Uggeri, F. J . Org.
Chem. 1987, 52, 3018 and references therein.
(9) Larsen, R. D.; Corley, E. G.; Davis, P.; Reider, P. J .; Grabowski,
E. J . J . J . Am. Chem. Soc. 1989, 111, 7650.
10.1021/jo0205614 CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/04/2003
J . Org. Chem. 2003, 68, 2633-2638
2633

Chen et al.
SCHEME 1
SCHEME 2
2 at -20 °C with LHMDS followed by addition of the
iodide gave acid 3a in ∼50% isolated yield. However,
addition of LHMDS to a mixture of acid 2 and the iodide
in THF improved the yield to 85%. Treatment of acid 3a
with oxalyl chloride and catalytic DMF in toluene at 25
°C generated the corresponding acid chloride. This mix-
ture was cooled to 0 °C, and addition of N,N′-dimethyl-
ethylamine provided ketene 4a . Addition of (R)-(-)-
pantolactone to the ketene at <-65 °C gave the desired
(R,R)-ester 5 in 96% yield and 87% diastereomeric excess
as determined by chiral HPLC.
83% ee acid was upgraded to 98.5% by a single crystal-
lization of the R-methylbenzylammonium salt in 83.5%
yield. The formation of amide 7a without racemization
was possible using N,N′-carbonyldiimidazole (CDI) or
PyBOP. CDI provided a cleaner workup; thus, acid 6a
was treated with CDI at room temperature to generate
the corresponding acyl imidazole. Addition of amino-
acetonitrile hydrochloride and Hunig’s base yielded
amide 7a in 98% yield and 98.1% ee as measured by
chiral HPLC.
Our initial strategy for the formation of the biaryl
moiety involved Suzuki coupling11 of N-Boc-boronic acid
10a with aryl bromide 7a as outlined in Scheme 3. Thus,
the amino moiety of bromophenylpiperazine 8 was pro-
tected using di-tert-butyl dicarbonate in acetonitrile to
give N-Boc-piperazine 9 in 97% yield. Treatment of 9 with
n-BuLi in 1:1 THF-toluene at -65 °C followed by
addition of triisopropyl borate gave N-Boc-boronic acid
10a after acidic workup. Suzuki coupling was completed
in 3 h using 5 mol % PdCl2(dppf)‚CH2Cl2 and aqueous
Na2CO3 in DMF at 90 °C to yield N-Boc-biaryl 11. Many
conditions were screened for removal of the N-Boc group;
however, in all cases, amide 12 was observed as a side
product (Scheme 4). Amide formation was minimized
using methanesulfonic acid; however, 12 was still formed
in ∼5% yield. Amide 12 was not rejected by crystalliza-
tion and could only be separated from 1 by column
chromatography.
The overall success of this synthetic strategy was
dependent on the development of reaction conditions for
removal of the ester group and subsequent amide forma-
tion, without racemization. Initially we investigated
acidic conditions (AcOH/HCl) for the hydrolysis of ester
5 to give (R)-acid 6a . The reaction was sluggish and was
not complete after 30 h at 85 °C. Increasing the temper-
ature shortened the reaction time but led to ∼10%
reduction in ee. Under alkaline conditions (LiOH/MeOH
or THF/water), hydrolysis was complete in 24 h at 20 °C
but led to significant racemization (25%). To our delight,
10
the use of LiOH/H2O2 at 0 °C led to minimal racemiza-
tion and the reaction was complete in less than 2 h, giving
chiral acid 6a in 93% yield. The optical purity was
upgraded by crystallization of the acid as its (R)-(+)-R-
methylbenzylammonium salt from 2-propanol/water. Thus,
(10) Evans, D. A.; Britton, T. C.; Ellman, J . A. Tetrahedron Lett.
1987, 28, 6141.
(11) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
2634 J . Org. Chem., Vol. 68, No. 7, 2003

Potent Cathepsin K Inhibitor
SCHEME 3
SCHEME 4
be very difficult, especially as standard workups or
crystallizations often fail to reduce levels to the low parts
per million range that is essential for the preparation of
an active pharmaceutical ingredient (API). Therefore, it
is generally undesirable to use a homogeneous transition
metal catalyst in the final synthetic step of an API. Such
reactions are often positioned early in the synthesis, thus
allowing subsequent processing to help in reducing
residual catalyst levels.
SCHEME 5
In an effort to address the cost of palladium and
minimize the amount of palladium in solution, a number
of reports have recently appeared describing heteroge-
neous recyclable catalysts. Thus, palladium catalysts on
silica,13 Pd/C,14 and palladium on polymer supports15 have
been investigated in the Suzuki coupling reaction. Un-
fortunately, many of these heterogeneous catalysts suffer
from lower catalyst activity than their homogeneous
counterparts. In addition, the activity of the catalyst
diminishes through recycling because palladium leaches
from the solid support. Parrish and Buchwald have
reported16 polymer-supported dialkylphosphinobiphenyl
ligands that upon treatment with a palladium source
provide highly active recyclable catalysts for the Suzuki
coupling reaction. However, for the Suzuki coupling of
aryl bromide 7a and boronic acid 13, these ligands were
unsuccessful when examined under homogeneous condi-
tions.
We were therefore not surprised to find that as a
consequence of our synthetic change, the levels of Pd and
Fe in isolated crude 1 were ∼8000 ppm. Neither chro-
matography nor treatment of crude 1 with n-Bu3P17
significantly reduced residual Pd/Fe levels. Consequently,
an efficient workup procedure was required to remove
the residual Pd and Fe from crude 1. Due to the lability
of the nitrile group of 1 under acidic conditions, aqueous
solutions of strong acids could not be used to extract the
product. Various organic acids were screened, and we
were delighted to find that lactic acid formed a water-
To avoid chromatography on a large scale, we turned
our attention to a Suzuki coupling using unprotected
boronic acid 13 (Scheme 5). The N-Boc group was
effectively removed from piperazine 10a using 4 N HCl
in dioxane-ethyl acetate to give boronic acid 13, isolated
as a hydrochloride salt in 86% yield. Various reaction
conditions were investigated for the Suzuki coupling of
unprotected boronic acid 13 with aryl bromide 7a .
Limited success was achieved using Pd2(dba)3 or Pd-
(OAc)2 with various solvent/phosphine/base combinations.
Switching to 5 mol % PdCl2(dppf)‚CH2Cl2 in DMF, the
reaction was complete in 3 h at 80 °C. Upon optimization,
the best conditions were found to be 3 mol % PdCl2-
(dppf)‚CH2Cl2 as a catalyst with aqueous K2CO3 as a base
in 10:1 toluene/DMF at 80 °C. Under these conditions,
the reaction was complete within 2 h to afford 1 in 89%
yield.
Despite the widespread use of palladium-mediated
reactions on a research scale, their use on an industrial
scale has had limited success. This is in part due to the
high cost of palladium, and equally importantly, to the
cost and difficulty of separating the palladium catalyst
from the product. The removal of homogeneous palladium
in an organic chemistry context is an issue that has been
generally neglected in the literature.12 Removal of even
small amounts (parts per million levels) of catalyst can
(13) Bedford, R. B.; Cazin, C. S. J .; Hursthouse, M. B.; Light, M. E.;
Pike, K. J .; Wimpers, S. J . Organomet. Chem. 2001, 633, 173.
(14) Sakurai, H.; Tsukuda, T.; Hirao, T. J . Org. Chem. 2002, 67,
2721.
(15) (a) J ang, S.-B. Tetrahedron Lett. 1997, 38, 1793. (b) Fenger, I.;
Le Drian, C. Tetrahedron Lett. 1998, 39, 4287. (c) Zhang, T. Y.; Allen,
M. J . Tetrahedron Lett. 1999, 40, 5813. (d) Uozumi, Y.; Danjo, H.;
Hayashi, T. J . Org. Chem. 1999, 64, 3384. (e) Buchmeiser, M. R.;
Schareina, T.; Kempe, R.; Wurst, K. J . Organomet. Chem. 2001, 634,
39.
(16) Parrish, C. A.; Buchwald, S. L. J . Org. Chem. 2001, 66, 3820.
(17) Larsen, R. D.; King, A. O.; Chen, C. Y.; Corley, E. G.; Foster,
B. S.; Roberts, E. F.; Yang, C.; Liberman, D. R.; Reamer, R. A.; Tschaen,
D. M.; Verhoeven, T. R.; Reider, P. J . J . Org. Chem. 1994, 59, 6391.
(12) Rosso, V. W.; Lust, D. A.; Bernot, P. J .; Grosso, J . A.; Modi, S.
P.; Rusowicw, A.; Sedergran, T. C.; Simpson, J . H.; Srivastava, S. K.;
Humora, M. J .; Anderson, N. G. Org. Proc. Res. Dev. 1997, 1, 311.
J . Org. Chem, Vol. 68, No. 7, 2003 2635

Chen et al.
mixture was aged at room temperature for 1.5 h. The solution
was cooled to 8-10 °C and N,N′-dimethylethylamine (744 g,
10.17 mol) added over ∼10 min. The temperature rose to 18
°C during the addition, and the batch was aged for 2 h at room
temperature. The mixture was cooled to -69 °C, and a solution
of (R)-(-)-pantolactone (529 g, 4.06 mol) in toluene (15.9 L)
was added over 3 h while maintaining the temperature <-64
°C. The mixture was aged at <-40 °C for 16 h, warmed to
room temperature, and quenched with water (9 L). The
aqueous layer was separated, and the toluene layer was
washed with saturated sodium bicarbonate (2 × 5 L) and water
soluble salt of 1 and did not hydrolyze the nitrile group.
Thus, crude 1 (∼8000 ppm Pd/Fe) was dissolved in ethyl
acetate and treated with 20 mol % n-Bu3P. The mixture
was extracted with aqueous lactic acid and the aqueous
layer washed with ethyl acetate. The lactic acid salt was
broken by addition of Na2CO3 and the product extracted
into ethyl acetate. The ethyl acetate layer was concen-
trated and the product crystallized from toluene to give
1 with <50 ppm Pd and Fe. A final swish with acetoni-
trile gave 1 of excellent purity.
(5 L). The organic layer was concentrated under reduced
25
pressure to give ester 5 (1.25 kg, 96% yield, 87% de). [R]D
)
Su m m a r y
+7.9 (c 1.48, MeOH). HPLC: Zorbax SB C18 250 mm × 4.6
mm column. Eluents: A, 0.1% aqueous H3PO4; B, acetonitrile;
2 mL/min. Gradient: A:B from 95:5 to 20:80 in 16 min, held
at 20:80 for 4 min. λ ) 220 nm, temperature 35 °C. tR: 3a )
14.3 min, 5 ) 17.6 min. Chiral HPLC: Chiralcel OD-R, 250
mm × 4.6 mm column. Eluents: A, 72% 0.5 M NaClO4/HClO4
pH ) 2; B, 28% acetonitrile; 2.0 mL/min. λ ) 220 nm,
temperature 50 °C. tR: (R,R)-5 ) 33.2 min, (S,R)-5 ) 36.2 min.
1H NMR (CDCl3): δ 7.52 (t, 1H, J ) 1.6 Hz), 7.41 (bd, 1H, J
) 7.8 Hz), 7.29 (bd, 1H, J ) 7.8 Hz), 7.20 (t, 1H, J ) 7.8 Hz),
5.32 (s, 1H), 4.02-3.97 (m, 2H), 3.78 (t, 1H, J ) 7.8 Hz), 2.02
(m, 1H), 1.73 (m, 1H), 1.54 (m, 1H), 1.15 (s, 3H), 1.04 (s, 3H),
0.95 (d, 3H, J ) 6.5 Hz), 0.94 (d, 3H, J ) 6.5 Hz). 13C NMR
(CDCl3): δ 172.2, 171.6, 140.2, 130.9, 130.4, 129.9, 126.6, 122.4,
75.9, 75.1, 49.9, 41.1, 40.0, 25.7, 22.9, 22.2, 22.1, 19.6. IR: 2961,
1795, 1745, 1146, 995, 760 cm-1. HRMS (FAB+) m/z 383.0857
[M + H+, calcd for C18H23BrO4, 383.0858].
(2R)-2-(3-Br om op h en yl)-4-m eth ylp en ta n oic Acid (6a ).
Pantolactone ester 5 (1.92 kg, 6.72 mol) was dissolved in
methanol (22 L). The mixture was cooled to 0 °C, and hydrogen
peroxide (30% in water, 1.9 L, 18.4 mol) was added. Aqueous
lithium hydroxide (4 N, 1.85 L, 7.4 mol) was added over 45
min while maintaining the temperature at <5 °C. The mixture
was aged for 2 h to complete the hydrolysis. The pH of the
solution was adjusted to pH ) 1 by addition of HCl (6 N, 1.4
L). Aqueous sodium sulfite (2 M, 8.74L, 17.5 mol) was added
over 45 min, maintaining the temperature at <20 °C. The pH
was adjusted to pH ) 6 by addition of HCl (6 N, 4 L). Toluene
(30 L) was added and the lower aqueous layer separated. The
toluene layer was washed with water (2 × 5 L and 2 × 3 L) to
give 1.7 kg of acid 6a (93% yield, 83% ee) as assayed by HPLC.
Chiral HPLC: Chiralcel OD-R, 250 mm × 4.6 mm. Eluent:
A, 72% 0.5 M NaClO4/HClO4 pH ) 2; B, 28% acetonitrile, 2.0
mL/min, λ ) 220 nm, temperature 50 °C. tR: (R)-6 ) 37.6 min,
(S)-6 ) 40.6 min. The toluene solution of acid 6a was
concentrated under reduced pressure and the remaining oil
dissolved in a mixture of 2-propanol (51 L) and water (2.5 L).
(R)-(+)-R-Methylbenzylamine (760 g. 6.27mol) dissolved in
2-propanol (1 L) was added in one portion. A precipitate formed
immediately, and the mixture was heated to 77 °C at which
point all the precipitate dissolved. The mixture was cooled to
20-25 °C over 16 h during which time the salt precipitated.
The salt was filtered, washed with 2-propanol (2 × 5 L), and
dried under vacuum at 35 °C for 24 h to yield 2.05 kg (77.6%
yield from 5, 98.5% ee). Mp: 180-183 °C. Chiral HPLC:
Chiralcel OD-R, 250 mm × 4.6 mm column. Eluents: A, 72%
0.5 M NaClO4/HClO4 pH ) 2; B, 28% acetonitrile; 2.0 mL/min.
λ ) 220 nm, temperature 50 °C. tR: (R)-6a ) 37.6 min, (S)-6a
) 40.6 min. 1H NMR (methanol-d4): δ 7.57 (t, 1H, J ) 1.7
Hz), 7.43-7.36 (m, 5H), 7.31 (bt, 2H, J ) 7.8 Hz), 7.16 (t, 1H,
J ) 7.8 Hz), 4.37 (q, 1H, J ) 6.9 Hz), 3.51 (t, 1H, J ) 6.8 Hz),
1.91 (m, 1H), 1.58 (d, 3H, J ) 6.9 Hz), 1.54-1.46 (m, 2H), 0.91
(app d, 6H). 13C NMR (methanol-d4): δ 181.4, 146.8, 140.3,
132.0, 130.9, 130.2, 130.1, 130.0, 128.0, 127.6, 123.1, 54.3, 52.2,
44.5, 27.4, 23.1, 23.0, 21.0. Anal. Calcd for C20H26BrNO2: C,
61.23; H, 6.68; N, 3.57. Found: C 61.48; H 6.42; N 3.85.
An efficient, large-scale, chromatography-free synthesis
of cathepsin K inhibitor 1 has been developed. The key
steps include deracemization of acid 3a via in situ
trapping of ketene 4a with (R)-pantolactone and a Suzuki
coupling of boronic acid 13 with bromide 7a to generate
the biaryl linkage and complete the convergent synthesis.
A simple and efficient workup was developed using lactic
acid to remove residual palladium and iron from 1
following the Suzuki coupling.
Exp er im en ta l Section
Gen er a l P r oced u r es. All commercially available sub-
strates, reagents, and solvents were used without further
purification. 1H NMR spectra were run at 500 MHz and 13C
NMR spectra at 125 MHz. Onedia Research Services, Inc.,
Whitesboro, NY, provided elemental analyses. High-resolution
mass spectra (HRMS-FAB+) were obtained at the Biomedical
Mass Spectrometry Unit, McGill University, Montre´al, Que´bec,
Canada. The enantiomeric excess was determined by HPLC
using chiral analytical columns. Melting points are uncor-
rected. Chiral assay of 1 was measured by SFC.
2-(3-Br om op h en yl)-4-m eth ylp en ta n oic Acid (3a ). To a
solution of 3-bromophenylacetic acid 2 (1.8 kg, 8.37 mol),
DMPU (1.8 L, 14.88 mol), and 1-iodo-2-methylpropane (1.6 kg,
8.69 mol) in THF (9 L) at -20 °C was added LHMDS (1 M in
THF, 17.6 L, 17.6 mol) over ∼3 h maintaining a temperature
between -19 and -22 °C. The mixture was aged for 4 h at a
temperature between -15 and -20 °C and warmed to room
temperature over 16 h. The batch was cooled to -5 °C, and
HCl (6 N, 5.5 L) was added over 2 h maintaining the
temperature at <20 °C. The lower aqueous layer was sepa-
rated and the THF layer concentrated under vacuum to yield
a viscous oil. The oil was dissolved in a premixed solution of
methanol (7.2 L) and water (2.1 L). The temperature was
adjusted to 20-22 °C, and water (0.5 L) was added, followed
by seed (10 g). The batch was aged for 1 h during which time
crystallization occurred. Water (8.1 L) was added over 3 h and
the batch aged overnight at room temperature. The batch was
filtered and the cake washed with 2/3 v/v methanol/water (2
× 3 L). The cake was dried under vacuum at 30 °C for 24 h to
yield 3a (1.92 kg, 85% yield). An analytically pure sample was
obtained by recrystallization from heptane. HPLC: Zorbax SB
C18 250 mm × 4.6 mm column. Eluents: A, 0.1% aqueous H3-
PO4; B, acetonitrile; 2 mL/min. Gradient: A:B from 95:5 to
20:80 in 16 min, held at 20:80 for 4 min. λ ) 220 nm,
temperature 35 °C. tR: 2 ) 10.1 min, 3a ) 14.3 min. Mp: 84-
84.7 °C. 1H NMR (CDCl3): δ 7.48 (t, 1H, J ) 1.6 Hz), 7.25 (app
d, 1H), 7.40 (bd, 1H, J ) 7.8 Hz), 7.19 (t, 1H, J ) 7.8 Hz), 3.62
(t, 1H, J ) 7.8 Hz), 1.94 (m, 1H), 1.66 (m, 1H), 1.48 (m, 1H),
0.91 (app d, 6H). 13C NMR (CDCl3): δ 179.9, 140.5, 131.0,
130.4, 130.0, 126.6, 122.5, 49.0, 41.7, 25.5, 22.3, 22.0. IR: 2957,
1702, 1571, 1212, 783, 690 cm-1. Anal. Calcd for C12H15BrO2:
C, 53.16; H, 5.59 Found: C, 53.24; H 5.61.
(3R)-4,4-Dim et h yl-2-oxot et r a h yd r ofu r a n -3-yl-(2R)-(3-
br om op h en yl)-4-m eth yl P en ta n oa te (5). To a solution of
3a (920 g, 3.39 mol) and DMF (5.1 mL, 0.066 mol) in toluene
(13.8 L) was added oxalyl chloride (516 g, 4.06 mol). The
To a suspension of acid 6a as its I-methylbenzylammonium
salt (1.994 kg, 5.08 mol) in toluene (25 L) was added HCl (1
N, 12.7 L). The mixture was agitated until dissolution was
2636 J . Org. Chem., Vol. 68, No. 7, 2003

Potent Cathepsin K Inhibitor
complete. The layers were separated, and the toluene layer
was washed with HCl (1N, 5 L) and water (5 L). The mixture
was solvent switched to THF to give 1.38 kg as assayed by
HPLC (75.5% yield from 5, 98.5% ee) of free acid 6a . A pure
sample of chiral acid 6a was prepared by crystallization from
the slurry. The slurry was filtered, and the wet cake was
washed with heptane (2 L) and dried under vacuum/nitrogen
for 12 h to give N-Boc boronic acid 10a (940 g, 92%). Mp: 187-
1
189 °C. H NMR (DMSO-d6, 3 drops of D2O): δ 7.67 (d, 2H, J
) 7.9 Hz), 6.88 (d, 2H, J ) 8.0 Hz), 3.43 (s, 4H), 3.12 (s, 4H),
1.40 (s, 9H). 13C NMR (DMSO-d6, 3 drops of D2O): δ 154.5,
152.6, 135.9, 123.6, 114.8, 79.7, 48.1, 43.8, 28.5. IR: 2976,
hexanes, mp 68-70 °C, [R]D25 ) -41.1 (c 1.19, MeOH). For H
1
and 13C NMR data, see 3a above.
1695, 1602, 1366, 1227, 737 cm-1. Anal. Calcd for C15H23
-
(2R)-2-(3-Br om op h en yl)-N-(cya n om et h yl)-4-m et h yl-
p en ta m id e (7a ). To a solution of acid 6a (1.38 kg, 5.09 mol)
in THF (20 L) was added N,N′-carbonyldiimidazole (988 g, 6.09
mol) portionwise over ∼5 min, maintaining the temperature
∼20 °C. After 30 min, aminoacetonitrile hydrochloride (938 g,
10.13 mol) was added over ∼5 min followed by addition of N,N-
diisopropylethylamine (1.97 kg, 15.21 mol) over ∼5 min. The
mixture became homogeneous after an additional 10 min. The
batch was aged for 2.5 h at room temperature to complete the
reaction. The reaction was quenched by addition of phosphoric
acid (3 M, 6.8 L), followed by addition of toluene (18 L). The
aqueous layer was separated, and the toluene layer was
washed successively with aqueous phosphoric acid (1 M, 2 L),
aqueous sodium bicarbonate (2 × 5 L), and water (3 L). The
organic layer was concentrated under vacuum to yield amide
BN2O4: C, 58.84; H, 7.57; N, 9.15. Found C, 58.94; H, 7.27; N,
9.08.
4-P ip er a zin -1-ylp h en ylb or on ic Acid H yd r och lor id e
(13). To N-Boc boronic acid 10a (930 g, 3.38 mol) in ethyl
acetate (15 L) was added a solution of HCl in dioxane (4 N,
2.4 L, 9.6 mol). The resultant slurry was aged at ambient
temperature for 6 h and filtered. The cake was washed with
ethyl acetate (4 L) and dried under vacuum/nitrogen for 12 h
to give phenylpiperazine boronic acid hydrochloride 13 (634
g, 86% yield). This material was used directly in the Suzuki
coupling reaction. Mp: 161-162 °C. 1H NMR (DMSO-d6, 3
drops of D2O): δ 7.69 (d, 2H, J ) 8.5 Hz), 6.97 (d, 2H, J ) 8.5
Hz), 3.46-3.42 (m, 4H), 3.23-3.19 (m, 4H). 13C NMR (DMSO-
d6, 3 drops of D2O): δ 151.2, 136.4, 122.8, 115.8, 46.2, 43.0.
IR: 3343, 3285, 2405, 1397, 845, 741 cm-1. An analytically
pure sample of the boronic acid as a free base was prepared
by treatment with NaOH followed by column chromatography.
1H NMR (DMSO-d6, 3 drops of D2O): δ 7.63 (d, 2H, J ) 8.7
Hz), 6.85 (d, 2H, J ) 8.7 Hz), 3.11 (s, 4H), 2.83 (s, 4H). 13C
NMR (DMSO-d6, 3 drops of D2O): δ 153.6, 136.2, 123.4, 114.7,
48.8, 45.7. Anal. Calcd for C10H15BN2O2: C, 58.29; H, 7.34; N,
13.60. Found C, 58.06; H, 7.03; N, 13.53.
(2R)-N-Cya n om eth yl-4-m eth yl-2-(4′-p ip er a zin -1-yl-1,1′-
bip h en yl-3-yl)p en ta n a m id e (1). To a mixture of toluene (12
L), DMF (1.2 L), and H2O (3.6 L) were added bromide 7a (569
g, 1.84 mol), boronic acid 13 (669.3 g, 2.18 mol), K2CO3 (1.27
kg, 9.2 mol), and PdCl2(dppf)‚CH2Cl2 (50 g, 0.0612 mol). The
slurry was degassed by bubbling N2 through for 30 min. The
mixture was heated to 80-85 °C for 2 h. The mixture was
cooled to 30 °C. EtOAc (12 L), H2O (2 L), and tri-n-butylphos-
phine (75 g, 0.37 mol) were added. The layers were separated,
and the organic layer was washed with H2O (12 L). The organic
layer was concentrated under reduced pressure to a volume
of 1-2 L. The residue was washed with hexanes (5 L), and
the solid that formed (crude 1) was filtered and washed with
hexanes (3 × 4 L).
25
7a as a pale yellow oil (1.53 kg, 97.5% yield, 98.1% ee). [R]D
) -47.7 (c 1.49, MeOH). HPLC: Zorbax SB C18 250 mm ×
4.6 mm column. Eluents: A, 0.1% aqueous H3PO4; B, aceto-
nitrile; 2 mL/min. Gradient: A:B from 95:5 to 20:80 in 16 min,
held at 20:80 for 4 min. λ ) 220 nm, temperature 35 °C. tR:
6a ) 14.3 min, 7a ) 13.8 min. Chiral HPLC: Chiralcel OD-
R, 250 mm × 4.6 mm column. Eluents: A, 72% 0.5 M NaClO4/
HClO4 pH ) 2; B, 28% acetonitrile; 2.0 mL/min. λ ) 220 nm,
temperature 50 °C. tR: (R)-7a ) 56.9 min, (S)-7a ) 65.2 min.
1H NMR (CDCl3): δ 7.45 (t, 1H, J ) 1.6 Hz), 7.40 (bd, 1H, J
) 7.8 Hz), 7.23 (bd, 1H, J ) 7.8 Hz), 7.20 (t, 1H, J ) 7.8 Hz),
6.76 (bt, 1H), 4.12 (dd, 1H, J ) 17.5, 6.0 Hz), 3.97 (dd, 1H, J
) 17.5, 5.6 Hz), 3.51 (t, 1H, J ) 7.7 Hz), 1.96 (m, 1H), 1.70
(m, 1H), 1.44 (m, 1H), 0.90 (d, 3H, J ) 6.5 Hz), 0.89 (d, 3H, J
) 6.5 Hz). 13C NMR (CDCl3): δ 173.4, 141.2, 131.0, 130.8,
130.5, 126.4, 122.7, 115.9, 50.0, 41.7, 27.5, 25.4, 22.6, 21.8.
IR: 3304, 2957, 2262, 1656, 1529, 768, 690 cm-1. HRMS
(FAB+) m/z 309.0603 [M + H+, calcd for C14H17BrN2O,
309.0602].
t er t -Bu t yl-4-(4-b r om op h e n yl)p ip e r a zin e -1-ca r b ox-
yla te (9). To a slurry of 1-(4-bromophenyl)-piperazine hydro-
chloride (1.86 kg, 7.71 mol) in acetonitrile (16 L) were added
triethylamine (2.03 kg, 20.06 mol), DMAP (81.8 g, 0.67 mol),
and di-tert-butyl dicarbonate (1.75 kg, 8.01 mol). The resultant
slurry was aged at 20-25 °C for 3 h. Water (40 L) was added
over 20 min. The slurry was aged at 20-25 °C for 30 min and
filtered. The wet cake was washed with water (10 L) and dried
under vacuum/nitrogen at 55 °C for 12 h to give N-Boc
piperazine 9, 2.19 kg (95% yield). Mp: 146.3-147.5 °C. 1H
NMR (acetone-d6): δ 7.38 (t, 2H, J ) 9.1 Hz), 6.96 (d, 2H, J )
9.1 Hz), 3.56-3.54 (m, 4H), 3.18-3.15 (m, 4H), 1.48 (s, 9H).
13C NMR (acetone-d6): δ 155.0, 151.5, 132.6, 119.0, 112.1, 79.8,
To crude 1 (1010 g) in ethyl acetate (20 L) was added tri-
n-butylphosphine (56 g, 0.277 mol) and the mixture aged for
1 h at room temperature. Water (15 L) and lactic acid (85%,
808 g, 7.6 mol) were added. The layers were separated, and
the aqueous layer was washed with ethyl acetate (3 × 5 L).
The combined organic layers were washed with water (5 L).
The combined aqueous layers were added to a suspension of
solid sodium carbonate (805 g, 7.6 mol) in ethyl acetate (15 L)
and the mixture agitated for ∼15 min until the carbonate was
dissolved. The aqueous layer was separated and the ethyl
acetate layer washed with water (5 L). To the organic layer
Darko KB® (200 g) and Solka Floc 40NF (200 g) were added.
The mixture was aged overnight at room temperature, filtered
through a bed of Solka Floc 40NF, and the cake washed with
ethyl acetate (4 L). The batch was concentrated under reduced
pressure and solvent switched to toluene. The volume was
adjusted to bring the concentration to 10 mL/g. The batch was
heated to 75 °C until completely dissolved, cooled to 65 °C and
seeded. The slurry was cooled to room temperature over ∼8
h, filtered, and washed with toluene (2 L). The filter cake was
dried under vacuum to yield 1 (640 g, 89% yield from amide
7).
49.7, 44.3, 28.6. IR: 2948, 1687, 1494, 1442, 1266, 733 cm-1
.
HRMS (FAB+) m/z 341.0865 [M + H+, Calcd for C15H21
-
BrN2O2, 341.0865]. Anal. Calcd for C15H21BrN2O2: C, 52.77;
H, 6.20, N, 8.21. Found: C, 52.85; H, 5.90; N, 8.17.
4-[4-(ter t-Bu toxycar bon yl)piper azin -1-yl]ph en ylbor on -
ic Acid (10a ). To a mixture of THF (7 L) and toluene (7 L)
was charged N-Boc piperazine 9 (1.14 kg, 3.34 mol). The
solution was cooled to -70 °C, followed by dropwise addition
of n-BuLi (1.6 M in hexanes, 2.3 L, 3.67 mol) while maintaining
the temperature below -60 °C. The mixture was aged at -70
°C for 20 min followed by the addition of triisopropyl borate
(752 g, 3.99 mol). The mixture was warmed to 0 °C, and the
reaction was quenched with saturated aqueous NH4Cl (4 L)
and water (1 L). Phosphoric acid (85 wt %, 461 g, 4.0 mol) was
added and the mixture agitated for 30 min. The organic layer
was separated and concentrated under vacuum to give a dark
blue slurry. The volume of the slurry was adjusted to ∼2 L
with toluene. Heptane (8 L) was added dropwise over 2 h to
Compound 1 (590 g) was suspended in acetonitrile (5.9 L)
and the mixture agitated for 16 h at room temperature. The
slurry was filtered and washed with cold (0-5 °C) acetonitrile
(2 L). The filter cake was dried under vacuum at 25 °C for 48
h to yield pure 1 (522 g, 88.5% recovery, 79% overall yield from
25
amide 7a , Fe/Pd <30 ppm, >99% ee). [R]D ) -47.9 (c 2.14,
J . Org. Chem, Vol. 68, No. 7, 2003 2637

Chen et al.
MeOH). Mp: 144-146 °C; HPLC: Zorbax SB C18 250 mm ×
4.6 mm column. Eluents: A, 0.1% aqueous H3PO4; B, aceto-
nitrile; 2 mL/min. Gradient: A:B from 95:5 to 20:80 in 16 min,
held at 20:80 for 4 min. λ ) 220 nm, temperature 35 °C. tR: 1
) 8.5 min. Chiral SFC-HPLC Chiralpak AD 250 × 4.6 mm
column, gradient 4-40% MeOH at 2%/min, held for 10 min,
1.5 mL/min, 300 bar, λ ) 215 nm, 35 °C. Sample preparation:
∼1 mg of 1 was dissolved in 1 mL of acetonitrile, and 3-5
drops of acetic anhydride were added; assay. tR: (R)-1 ) 20.4
Hz), 4.12 (dd, 2H, J ) 5.4, 3.3), 3.65 (t, 1H, J ) 7.7 Hz), 3.11-
3.08 (m, 4H), 2.86-2.84 (m, 4H), 1.96-1.90 (m, 1H), 1.59-
1.52 (m, 1H), 1.44-1.38 (m, 1H), 0.90 (app d, 6H). 13C NMR
(DMSO-d6): δ 174.3, 151.9, 141.7, 141.1, 130.9, 129.7, 128.0,
126.4, 126.1, 125.2, 118.5, 116.3, 50.0, 49.9, 46.5, 42.9, 28.0,
26.5, 23.3, 23.2. IR: 3296, 2957, 2262, 1679, 1521, 1239, 733
cm-1. HRMS (FAB+) m/z 391.2497 [M + H+, calcd for
C
24H31N4O, 391.2498]. Anal. Calcd: C, 73.81; H, 7.74; N, 14.35.
Found C, 73.87; H, 7.64; N, 14.28.
1
min, (S)-1 ) 17.4 min. H NMR (DMSO-d6): δ 8.78 (t, 1H, J
) 5.6 Hz), 7.53-7.48 (m, 3H), 7.45 (d, 1H, J ) 7.7 Hz), 7.34 (t,
1H, J ) 7.7 Hz), 7.20 (d, 1H, J ) 7.6 Hz), 7.01 (d, 2H, J ) 8.6
J O0205614
2638 J . Org. Chem., Vol. 68, No. 7, 2003