Development of a novel Pd-catalyzed N-acyl vinylogous carbamate synthesis for the key intermediate of ICE inhibitor VX-765

Article abstract of DOI:10.1021/ol702532h

(Chemical Equation Presented) A novel Pd-catalyzed coupling of Cbz-protected proline amide with 4-bromo-5-ethoxyfuran-2(5H)-one was developed for the synthesis of the P1-P2 unit (5) of VX-765. The process afforded quantitative coupling in the presence of water, providing a 1:1 mixture of 5 and its ethoxy epimer epi-5. Compound 5 was isolated as a single diastereomer via fractional crystallization, which was stereoselectively converted to 17 via hydrogenation, and subsequently transformed to VX-765. Nine examples of the Pd coupling are presented with yields ranging from 76-98%.

Full text of DOI:10.1021/ol702532h

ORGANIC
LETTERS
2008
Vol. 10, No. 2
185-188
Development of a Novel Pd-Catalyzed
N-Acyl Vinylogous Carbamate Synthesis
for the Key Intermediate of ICE Inhibitor
VX-765
Gerald J. Tanoury,* Minzhang Chen,§ Yong Dong, Raymond E. Forslund, and
Derek Magdziak
Chemical DeVelopment, Vertex Pharmaceuticals Inc., 130 WaVerly Street,
Cambridge, Massachusetts 02139
jerry_tanoury@Vrtx.com
Received October 17, 2007
ABSTRACT
A novel Pd-catalyzed coupling of Cbz-protected proline amide with 4-bromo-5-ethoxyfuran-2(5H)-one was developed for the synthesis of the
P1 P2 unit (5) of VX-765. The process afforded quantitative coupling in the presence of water, providing a 1:1 mixture of 5 and its ethoxy
epimer epi-5. Compound 5 was isolated as a single diastereomer via fractional crystallization, which was stereoselectively converted to 17 via
−
hydrogenation, and subsequently transformed to VX-765. Nine examples of the Pd coupling are presented with yields ranging from 76−98%.
Interleukin-1 (IL-1) is a pro-inflammatory and immunoregu-
latory protein which is involved in the pathogenesis of acute
and chronic inflammatory and autoimmune diseases.1 IL-
1â, one of two isoforms of IL-1, is converted into its
biologically active form by interleukin-1â converting enzyme
(ICE) and has been shown to be involved in apoptosis and
inflammation.2 Hence, ICE inhibitors represent a class of
compounds useful for the control of such events. VX-765 is
an inhibitor of ICE and has been developed for the treatment
of ICE-related diseases.3 VX-765 is a triamide containing
Scheme 1. Conventional Retrosynthesis of VX-765
§ Current department: Technical Operations, Vertex Pharmaceuticals.
(1) (a) Wood, D. D.; Ihrie, E. J.; Dinarello, C. A.; Cohen, P. L. Arthritis
Rheum. 1983, 26, 975. (b) Pettipher, E. J.; Higgs, G. A.; Henderson, B.
Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 8749. (c) Bataille, R.; Chappard,
D.; Bernard, K. Int. J. Clin. Lab. Res. 1992, 21, 283. (d) Vidal-Vanaclocha,
F.; Amezaga, C.; Asumendi, A.; Kaplanski, G.; Dinarello, C. A. Cancer
Res. 1994, 54, 2667.
one proline (Pro) and one tert-Leu (Tle) residue, a 4-amino-
3-chlorobenzoic acid unit, and a tetrahydroethoxyfuranone
unit. Previous strategies for the synthesis of VX-765 focused
exclusively on amide bond formations, leading to the
10.1021/ol702532h CCC: $40.75
© 2008 American Chemical Society
Published on Web 12/15/2007

retrosynthetic analysis shown in Scheme 1. Although the Pro,
Tle, and the benzoic acid units are readily available starting
materials, preparation of aminotetrahydroethoxyfuranone 4
remains difficult and impractical due to its instability and
proclivity for undesired side reactions under amide bond-
forming conditions. As such, and in order to circumvent the
problems associated with 4, a new approach to VX-765 based
on the P1-P2 key intermediate 5 was developed via the
disconnections shown in Scheme 2. The process relies on
substrate and N,N,N′,N′-tetramethyl-1,2-diaminocyclohexane
as the ligand. In our case, the practicality and ease of
accessibility of 7 required a coupling method amenable to
(E)-3-bromoacrylates, substrates not conducive to Cu-cata-
lyzed conditions. Our initial investigations began with the
typical conditions used for Pd-catalyzed amide N-arylations
(eq 1). Under these conditions, Z-Pro-NH2 was coupled to
7 to give 5 and epi-5 in yields ranging from 58 to 78%,
capricious results at best. Control experiments showed the
irreproducibility arose from the instability of 7 under basic
conditions.
Scheme 2. Retrosynthesis of VX-765 via Pd-Catalyzed
Coupling
To solve the instability problem, a series of experiments
were conducted to determine the best base, solvent, temper-
ature, and precatalyst combination to effect the desired
coupling reaction. A rapid evaluation of solvents (nonpolar
and polar aprotic solvents) and bases showed 7 was stable
in the presence of K2CO3 or Cs2CO3 in toluene at 22-25
°C for 8 h and began to decompose (HPLC) at temperatures
of 45-50 °C. Next, we turned our attention to determining
the best ligand and palladium source for the coupling.
formation of the crucial C-N bond of 5 via coupling of the
amide Z-Pro-NH2 and bromoethoxyfuranone, a process that
has not been previously demonstrated. We present here a
novel target-oriented methodology for the synthesis of 5 via
development of a Pd-catalyzed N-acyl vinylogous carbamate
synthetic methodology.
The coupling of â-halo enoates with amides to generate
N-acyl vinylogous carbamates is a new process and has been
demonstrated previously by Porco using allyl (E)-3-iodoacryl-
ates and (E)-3-iodoacrylamides in the presence of 1-5 mol
% of CuI2.4-6 In this case, iodides were required as the
Table 1 shows the results from evaluating two practical
Table 1. Ligand Evaluation for Coupling in Equation 1a
(2) (a) Sleath, P. R.; Hendrickson, R. C.; Kronheim, S. R.; March, C. J.;
Black, R. A. J. Biol. Chem. 1990, 265, 14526. (b) Howard, A. D.; Kostura,
J. R.; Thornberry, N.; Ding, G. J.; Limjuco, G.; Weidner, J.; Salley, J. P.;
Hogquist, K. A.; Chaplin, D. D.; Mumford, R. A. J. Immunol. 1991, 147,
2964. (c) Black, R. A.; Kronheim, S. R.; Sleath, P. R. FEBS Lett. 1989,
386. (d) Kostura, M. J.; Tocci, M. J.; Limjuco, G.; Chin, J.; Cameron, P.;
Hillman, A. G.; Chartrain, N. A.; Schmidt, J. A. Proc. Natl. Acad. Sci.
U.S.A. 1989, 86, 5227. (e) Steller, H. Science 1995, 267, 1445.
(3) For leading references to VX-765, see: (a) Wannamaker, W.; Davies,
R.; Namchuk, M.; Pollard, J.; Ford, P.; Ku, G.; Decker, C.; Charifson, P.;
Weber, P.; Germann, U. A.; Keisuke, K.; Randle, J. C. R. J. Pharmacol.
Exp. Ther. 2007, 32, 509. (b) Stakc, J. H.; Beaumont, K.; Larsen, P. D.;
Straley, K. S.; Henkel, G. W.; Randle, J. C. R.; Hoffman, H. M. J. Immunol.
2005, 175, 2630. (c) Ravizza, T.; Lucas, S.-B.; Silvia, B.; Bernardino, L.;
Ku, G.; Noe, F.; Malva, J.; Randle, J. C. R.; Allan, S.; Vezzani, A. Epilepsia
2006, 47, 1160. For leading references to caspase inhibitors, see: (d)
Cornelis, S.; Kerrse, K.; Festiens, N.; Lamkanfi, M.; Vandenabeele, P. Curr.
Pharm. Des. 2007, 13, 367. (e) Linton, S. D. Curr. Top. Med. Chem. 2005,
5, 1697.
conversion (%)
entry
ligand
Xantphos
BINAP
DPEPhos
DPPB
DPPE
DPPF
Cy2P-biphenyl
t-Bu2P-biphenyl
Pd(OAc)2
Pd2(dba)3
1
2
3
4
5
6
7
8
>85
21
47
0.0
16
72
>85
23
72
7.2
7.4
70
9.8
8.5
30
42
a Conditions: 1.1 equiv of 7, 10 mol % of Pd source, 15 mol % of ligand,
Cs2CO3 (1.5 equiv), toluene, 60 °C, 2 h.
(4) (a) Han, C.; Shen, R.; Su, S.; Porco, J. A., Jr. Org. Lett. 2004, 6, 27.
Also see: (b) Buckler, R. T.; Hartzler, H. E. J. Med. Chem. 1975, 18, 509.
(c) Boger, D. L.; Wysocki, R. J., Jr. J. Org. Chem. 1989, 54, 714. (d)
Hosokawa, T.; Takano, M.; Kuroki, Y.; Murahashi, S. I. Tetrahedron Lett.
1992, 33, 6643. (e) Suh, E. M.; Kishi, Y. J. Am. Chem. Soc. 1994, 116,
11205.
(5) For related Cu-catalyzed coupling of amides, see: (a) Jiang, L.; Job,
G. E.; Kalpars, A.; Buchwald, S. L. Org. Lett. 2003, 5, 3667. (b) Klapars,
A.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 7421. (c)
Klapars, A.; Antilla, J. C.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc.
2001, 123, 7727. (d) Shen, R.; Porco, J. A. Org. Lett. 2000, 2, 1333. (e)
Guo, X.; Rao, H.; Fu, H.; Jiang, Y.; Zhao, Y. AdV. Synth. Catal. 2006,
348, 2197. (f) Lee, J. M.; Ahn, D.-S.; Jung, D. Y.; Lee, J.; Do, Y.; Kim, S.
K.; Chang, S. J. Am. Chem. Soc. 2006, 128, 12954.
(6) Also see: (a) Pan, X.; Cai, Q.; Ma, D. Org. Lett. 2004, 6, 1809-
1812. (b) Hu, T.; Li, C. Org. Lett. 2005, 7, 2035-2038. (c) Willis, M. C.;
Brace, G. N.; Holmes, I. P. Synthesis 2005, 3229-3234. (d) Lee, J. M.;
Ahn, D.-S.; Jung, D. Y.; Lee, J. G.; Do, Y.; Kim, S. K.; Chang, S. J. Am.
Chem. Soc. 2006, 128, 12954. (e) Villa, M. V. J.; Targett, S. M.; Barnes,
J. C.; Whittingham, W. G.; Marquez, R. Org. Lett. 2007, 9, 1631.
palladium sources and several ligands. The reaction condi-
tions used high catalyst loadings (10 mol %) for rapid
coupling in order to minimize bromide decomposition. For
Pd(OAc)2, Xantphos, DPEPhos, and DPPF gave the best
conversions after 2 h at 60 °C (entries 1,3, and 6); Cy2P-
biphenyl and t-Bu2P-biphenyl gave less than 10% conversion
(entries 7 and 8), and BINAP gave only 21% conversion
(entry 2). Alkylidene diphenylphosphino ligands gave poor
results with DPPE and DPPB, providing 16 and 0%
conversion, respectively (entries 4 and 5). Replacing Pd-
(OAc)2 with Pd2dba3 gave comparable results: Xantphos,
DPEPhos, and DPPF were the best ligands (entries 1, 3, and
6). Interestingly, BINAP gave nearly identical results for
186
Org. Lett., Vol. 10, No. 2, 2008

Pd2(dba)3 and Pd(OAc)2, but Cy2P-biphenyl and t-Bu2P-
biphenyl gave much higher conversions for Pd2(dba)3. From
the data, Xantphos was chosen as the best ligand.
base, but not too much to allow for excess water in solution.
Additionally, the PTC was used to maintain efficient mass
transfer between phases. The superior features of this system
can be seen explicitly in Figure 1. After the addition of 7
was complete (3 h), the coupling in the presence of 3 equiv
of water continued unabated until complete conversion was
obtained (circles). However, in the case of a 1:1 toluene/
water mixture, catalyst deactivation began soon after the
addition of 7 was complete (diamonds). Although the reac-
tion proceeded to 99% completion, the data show that at
85-90% conversion the rate of reaction significantly de-
creased.
The success of the Pd-catalyzed coupling in the presence
of 3 equiv of water and PTC prompted further investigations
into the nature of the reaction. The optimized conditions for
the coupling are shown in eq 2, and the results from
monitoring the reaction progress are shown in Figure 2. Of
In conjunction with optimization studies, the robustness
of the reaction was evaluated by performing the reaction on
100 g scale. Using Pd2(dba)3 (2.5%), Xantphos (7.5%), Cs2-
CO3 (2 equiv), and 7 (1.1 equiv) in toluene, less than 70%
conversion was observed after 3 h of reaction time. Charging
an additional 0.5 equiv of 7 and warming the reaction mixture
to 105 °C for 8 h resulted in less than 80% conversion. HPLC
analysis of the reaction mixture showed the presence of large
amounts of 7. The stability of 7 under these conditions was
expressly contrary to our results from the stability study.
After cooling the reaction mixture to room temperature and
upon adding water for workup, copious amounts of gas
evolution (CO2) was observed, indicative of HBr quenching
with Cs2CO3. This observation led to the conclusion that the
poor solubility of the inorganic base in anhydrous toluene,
and possible coating of the base by CsBr generated during
the reaction, prevented neutralization of HBr, which conse-
quently inhibited the catalyst and provided the acidic
conditions necessary for 7 to remain intact even at high
temperatures. The inability of the inorganic base to neutralize
HBr was addressed by investigating aqueous conditions.
Two scenarios were envisaged for successful development
of the Pd-catalyzed coupling under aqueous conditions. In
the first scenario, a 1:1 v/v toluene/water solvent mixture
was examined. Under these conditions, using 2.5 equiv of
K2CO3 as base, the precatalyst generated from Pd(OAc)2 (1
mol %) and Xantphos (1.5 mol %) provided 99% conversion
of 6 and 7 to 5 and epi-5 at 50 °C in 5.5-6 h. The use of
aqueous conditions resulted in a more rapid decomposition
of 7, which was obviated by slow addition of 7 over 3-3.5
h. The use of Pd2(dba)3 gave inferior results. However, when
the 1:1 toluene/water mixture was replaced by 3 equiv of
water and 1 mol % of a phase transfer catalyst (PTC) in
toluene, the reaction performed better, giving a 99.5%
conversion of 6 to 5 and epi-5 in 3.5-4 h. The use of 3
equiv of water, 1.5 equiv relative to K2CO3, was chosen
specifically to generate the sesquihydrate K2CO3(H2O)1.5.
Thus, enough water was present to fully wet the anhydrous
Figure 2. Mol % of components during the course of the reaction.
significant note is that, during the first 2 h of addition, the
concentration of 5 and its ethoxy epimer epi-5 rose to 65%,
while the concentration of 7 remained below 1%, indicating
that decomposition of 7 was significantly slower than
coupling to 6. After the addition of 7 was complete, HPLC
analysis showed about 4% of 7 in solution, signifying that
6% of 7 decomposed during the course of the reaction. If
the reaction mixture was allowed to stir at 50 °C overnight,
7 was consumed. The conditions in eq 2 were further
validated by successfully performing the coupling on a 1 kg
scale.
The utility of this coupling process was extended to other
systems (Table 2). Coupling of arylamides and alkylamides
to bromoethoxyfuranone occurred readily (entries 1 and 2),
as well as the coupling of carbamates (entry 3). The reaction
also worked well for the coupling of amides, carbamates,
and heterocyclic amides to â-bromoacrylates (entries 4 and
Figure 1. Effect of water content on reaction rate.
Org. Lett., Vol. 10, No. 2, 2008
187

its 1:1 mixture with ethoxy epimer epi-5) and stereoselective
reduction of the double bond to give 17 (Scheme 3).
Table 2. Pd-Catalyzed Coupling of Amides and Carbamates to
â-Bromoacrylates, â-Bromoacrylamides, and 7a
Scheme 3. Stereoselective Hydrogenation of 5 and Coupling
to Z-Tle-OH
Gratifyingly, after a simple aqueous workup of the crude
reaction mixture and concentration, 5 fractionally crystallized
from the reaction mixture in 37% isolated yield as a 93:7
mixture of diastereomers. After recrystallization from toluene
(90% recovery), diastereomerically pure 5 was isolated as a
white solid. Subsequent hydrogenation effected stereoselec-
tive reduction of the double bond and concomitant removal
of the Cbz protecting group to give 17. Compound 17 was
transformed to VX-765 via coupling with Z-Tle-OH (to give
18), removal of the Cbz protecting grouping, and coupling
to benzoic acid 1.
In conclusion, we have developed a novel Pd-catalyzed
coupling process for the conversion of â-bromoenoates to
â-amidoenoates within the context of a target-oriented
synthesis of the key intermediate of ICE inhibitor VX-765.
The process demonstrates the ability to perform the coupling
under conditions mild enough to allow efficient conversion
of base-sensitive bromides. Additionally, the methodology
was shown to be successful for the coupling of carbamates
and a range of amides to â-bromoacrylates and â-bro-
moacrylamides.
a Conditions: 1.1 equiv of bromide, 1 mol % of Pd(OAc)2, 1.5 mol %
of Xantphos, K2CO3 (2 equiv), water (3 equiv), toluene, 60 °C, 3-3.5 h.
Acknowledgment. G.J.T. thanks Prof. Brian Stoltz for
helpful discussions.
5, respectively) and to â-bromoacrylamides (entries 6-8).
The reactions were comparable in rates. The acrylate products
were susceptible to isomerization if allowed to remain at
elevated temperatures for extended periods after reaction
completion.
Supporting Information Available: Characterization
data, NMR spectra, detailed experimental procedures, and
structures of phosphine ligands. This material is available
free of charge via the Internet at http://pubs.acs.org.
Having demonstrated an efficient coupling process, the
next task required isolation of the single diastereomer 5 (from
OL702532H
188
Org. Lett., Vol. 10, No. 2, 2008