Stereoselective synthesis of the epoxysuccinyl peptide E-64c

Article abstract of DOI:10.1055/s-2006-948191

A highly diastereoselective PTC epoxidation is employed in the synthesis of the potent cysteine protease inhibitor E-64c. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-948191

LETTER
2063
Stereoselective Synthesis of the Epoxysuccinyl Peptide E-64c
Synthesis of
E
-64
a
c
rry Lygo,* Stuart D. Gardiner, Daniel C. M. To
School of Chemistry, University of Nottingham, Nottingham, NG7 2RD, UK
E-mail: B.Lygo@Nottingham.ac.uk
Received 6 June 2006
Abstract: A highly diastereoselective PTC epoxidation is em-
ployed in the synthesis of the potent cysteine protease inhibitor E-
64c.
O
H
N
RO2C
N
H
O
Key words: epoxysuccinates, asymmetric epoxidation, quaternary
ammonium salts, phase-transfer catalysis
O
1: R = H; E-64c
2: R = Et; E-64d (loxistatin)
Epoxysuccinyl peptides have received much attention due
to their propensity to act as potent and selective inhibitors
of cysteine proteases.1 In many instances these com-
pounds lead to irreversible inhibition via reaction of the
epoxysuccinyl moiety with the nucleophilic thiol present
in the enzyme active site.2 Since this mechanism is specif-
ic to cysteine proteases, epoxysuccinyl peptides generally
show low reactivity towards other types of protease.
O
R'
X
N
H
O
O
3
Figure 1
For the purposes of this study we opted to investigate the
diastereoselective epoxidation of benzoylacrylamides 4.
These compounds were chosen for study because the di-
rect precursors are commercially available, and because
the g-keto group of the acrylamide should activate the
substrate towards Weitz–Scheffer epoxidation. The g-
keto group should also ensure that initial addition of
hypochlorite occurs at the a-carbon (Scheme 1). An added
advantage of this type of substrate is that the phenyl
ketone can serve as a masked carboxylic acid. This avoids
the problem of needing to differentiate between two
carboxyl derivatives when applying this chemistry to
targets such as 1.
The naturally-occurring epoxysuccinyl peptide E-64, and
the closely related compounds E-64c (1) and E-64d (2,
loxistatin) are typical examples of this class of enzyme
inhibitor.3 They exhibit high activity towards a range of
cysteine proteases and E-64 is widely employed as a
means of characterising cysteine protease activity.
Most synthetic approaches to epoxysuccinyl peptides of
this type employ the coupling of an intact epoxysuccinate
derivative to the peptide fragment.2–5 This is a highly
effective strategy, but has limitations in terms of the
functionality that can be incorporated at the carboxy
terminus of the succinate.2 In this paper we report the
development of a novel stereoselective approach to
epoxysuccinyl peptide derivatives that also allows
straightforward access to keto-epoxide analogues. The
utility of this chemistry is demonstrated in the synthesis of
E-64c (1) and loxistatin (2).
O
R4N+OCl–
O
Ph
X
N
H
Ph
X
N
H
O–
O
O
Cl
R4N+
4
5
O
O
O
– R4N+Cl–
We have recently been involved in the development of
asymmetric phase-transfer-catalyzed (PTC) Weitz–
Scheffer epoxidations.6,7 This chemistry employs quater-
nary ammonium salts in conjunction with aqueous NaOCl
to effect the epoxidation of electron deficient alkenes and
we envisaged that it might offer an alternative route into
epoxysuccinyl peptides whilst also allowing access to a
range of epoxy-ketone analogues. To test this hypothesis
we decided to investigate whether compounds such as 3
(Figure 1) could serve as intermediates in the synthesis of
targets 1 and 2.
Ph
X
N
H
O
6
O
O
Scheme 1
Compounds 8a and 8b were prepared via coupling of (E)-
3-benzoylacrylic acid with the appropriate leucine ester
(Scheme 2). Epoxidation of these compounds was then in-
vestigated employing aqueous NaOCl in conjunction with
a quaternary ammonium phase-transfer catalyst. These
are conditions that we have shown to be effective for a
wide range of enones,6 however, as far as we are aware,
they have never previously been applied to peptide sub-
strates. A particular issue in applying this chemistry to
these substrates lies in the fact that aqueous NaOCl is
known to effect the N-chlorination of amides.8 Thus if this
SYNLETT 2006, No. 13, pp 2063–2066
1
7
.0
8
.2
0
0
6
Advanced online publication: 09.08.2006
DOI: 10.1055/s-2006-948191; Art ID: D15906ST
© Georg Thieme Verlag Stuttgart · New York

2064
B. Lygo et al.
LETTER
chemistry is to be successful, we need to identify condi-
tions that favor epoxidation over this alternative process.
Et
N
N
N+
N+
H
H
Initial studies into the PTC epoxidation of substrates 8
using conditions that were optimized for the epoxidation
of simple chalcones (1 mol% catalyst, 4 equiv 15% aq
NaOCl, 25 °C)6 met with limited success. The epoxida-
tions were fast (t1/2 = 20 min compared with t1/2 = 90 min
for chalcone under the same conditions), but significant
amounts of N-chlorination were observed. Control exper-
iments established that N-chlorination also occurred in the
absence of phase-transfer catalyst. Since epoxidation did
not occur without the catalyst we speculated that HOCl
rather than NaOCl might be responsible for this side reac-
tion. Since the equilibrium concentration of HOCl should
reduce as the pH of the aqueous phase increases, we
decided to examine the effect of adding KOH to maintain
a pH >12 in the aqueous phase. It was found that this
suppressed the N-chlorination, and by combining this
modification with a modest increase in catalyst loading (to
5–10 mol%), high selectivity for the desired epoxidation
could be achieved.
BnO
Br–
BnO
Br–
12
11
Figure 2
Table 1 Asymmetric PTC Epoxidation of 8a and 8b
Entry
Substrate
Catalyst
n-Bu4NBrc
11d
Yield (%)a Ratio 9:10b
1
2
3
4
5
6
8a
64
70
69
58
72
72
1:1
5:1
2:5
2:3
3:1
1:5
12d
8b
n-Bu4NBrc
11d
12d
a Yield after purification, see representative experimental procedure10
for further details.
b Estimated from the 1H NMR spectra of the crude products.
c 10 mol% catalyst.
a: R = Me
b: R = t-Bu
d 5 mol% catalyst.
H2N
CO2R
7
(i)
O
Our earlier studies on the asymmetric epoxidation of chal-
cones demonstrated that the cinchona alkaloid-derived
quaternary ammonium salts 11 and 12 (Figure 2) induce
high levels of stereocontrol. These two catalysts usually
give roughly equal but opposite levels of stereoselectivity
with a given substrate.6,9 Thus it was anticipated that they
should provide a means of achieving stereoselective
access to diastereoisomers 9 and 10.
Ph
N
H
CO2R
8
O
(ii)
O
O
Ph
Ph
+
N
H
CO2R
N
H
CO2R
O
O
9
10
O
O
The results obtained using these catalysts are given in
Scheme
2
Reagents and conditions: (i) PhCOC=CCO2H, i-
BuO2CCl, NMM (8a; 95%, 8b; 85%); (ii) 15% aq NaOCl (3 equiv), Table 1 (entries 2, 3, 5, 6). As expected they produced
pH >12, PhMe, PTC, 25 °C (for further details see Table 1).
opposite diastereoisomers, but for a given substrate the
magnitude of the diastereoselectivity differed somewhat.
For diastereoisomer 9, the methyl ester gave highest
selectivity (5:1), whereas for diastereoisomer 10, the tert-
butyl ester gave the best selectivity (again 5:1). As the na-
ture of the amino acid ester appeared to have a significant
influence on the level of diastereoselectivity we decided
to extend this study to include the amide substrate 14.
When the epoxidation was performed on enone 8a using
the achiral phase-transfer catalyst n-Bu4NBr, the corre-
sponding products 9a and 10a were obtained as a 1:1 mix-
ture (Scheme 2 and Table 1, entry 1). It is worth noting
that little or no ester hydrolysis was observed despite the
highly basic nature of the aqueous NaOCl solution. The
corresponding tert-butyl ester 8b gave a 2:3 mixture of
diastereoisomers with n-Bu4NBr (Table 1, entry 4), and
again the products 9b and 10b could be isolated in good
yield.
Enone 14 was prepared via coupling of (E)-3-benzoyl-
acrylic acid with leucine amide 13 (Scheme 3).11 Epoxi-
dation of 14 was then investigated as before. It was found
that all three catalysts resulted in successful epoxidation,
however, the levels of diastereoselectivity obtained using
the chiral catalysts 11 and 12 (Table 2) were significantly
lower than those obtained for the ester substrates 8.
Both reactions involving n-Bu4NBr resulted in little or no
diastereoselectivity indicating that the pre-existing stereo-
genic center in substrates 8a and 8b has little influence
over the stereochemical outcome of this epoxidation.
Consequently, in an effort to develop a diastereoselective
epoxidation we turned our attention to the use of chiral
phase-transfer catalysts.
These results reinforce the earlier observation that the
nature of the carboxyl terminus has a significant influence
on the levels of diastereoselectivity when the chiral cata-
lysts are employed. These results suggest that substrate 8a
Synlett 2006, No. 13, 2063–2066 © Thieme Stuttgart · New York

LETTER
Synthesis of E-64c
2065
KOH provided direct access to loxistatin (2). Simply by
adding water and extending the reaction time to two
hours, E-64c (1) could be obtained. The 1H NMR spectra
of loxisatin (2) and E-64c (1) prepared by the methods
described here were in agreement with those previously
reported for these materials.2,3
H
N
H2N
O
13
(i)
O
H
N
Ph
O
N
H
O
O
O
(i)
14
Ph
Ph
N
H
CO2H
(ii)
N
H
CO2Me
O
O
O
O
O
9a
17
O
H
H
N
(ii)
(iii)
Ph
N
Ph
N
N
H
16
O
O
H
O
O
O
15
O
O
+
O
(iv)
(v)
O
H
N
H
N
1
2
Ph
N
H
PhO2C
N
H
18
O
O
O
O
16
Scheme
3
Reagents and conditions: (i) PhCOC=CCO2H, i-
Scheme 4 Reagents and conditions: (i) LiOH, H2O, THF, MeOH,
0 °C (76%); (ii) H2N(CH2)2CH(CH3)2, EDCI, HOBt, NMM, CH2Cl2
(76%); (iii) MCPBA, CHCl3, 50 °C, (70%); (iv) EtOH, KOH, 0 °C,
30 min, (53%); (v) EtOH, aq KOH, 0 °C, 2 h, (65%).
BuO2CCl, NMM (84%); (ii) 15% aq NaOCl (3 equiv), pH >12,
PhMe, PTC, 25 °C (for further details see Table 2).
Table 2 Asymmetric PTC Epoxidation of 14
This study has demonstrated that the asymmetric phase-
transfer epoxidation of substrates such as 8 is an effective
means of preparing epoxysuccinyl peptides derivatives.
The chemistry described should allow access to a wide
range of E-64c analogues and application of this method-
ology in the search for new cysteine protease inhibitors is
currently underway in our laboratories.
Entry
Catalyst
n-Bu4NBrc
11d
Yield (%)a
Ratio 15:16b
1
2
3
67
72
64
1:1
2:1
2:3
12d
a Yield after purification.
b Estimated from the 1H NMR spectra of the crude products.
c 10 mol% catalyst.
Acknowledgment
d 5 mol% catalyst.
We thank EPSRC for studentship support (S.D.G. and D.C.M.T)
and GSK for additional financial support. We also acknowledge use
of the EPSRC Chemical Database Service at Daresbury.
is the best precursor for targets such as E-64c (1), and
crucially, this establishes that compound 9a can be ob-
tained in high yield and with good diastereoselectivity
using this approach.
References and Notes
(1) (a) Leung-Toung, R.; Zhao, Y. Q.; Li, W. R.; Tam, T. F.;
Karimian, K.; Spino, M. Curr. Med. Chem. 2006, 13, 547.
(b) Powers, J. C.; Asgian, J. L.; Ekici, O. D.; James, K. E.
Chem. Rev. 2002, 102, 4639.
This brief study suggests that the asymmetric PTC epoxi-
dation of substrates such as 8 is a useful way of accessing
both diastereoisomers of epoxy ketones such as 9 and 10.
To further probe the utility of this approach we next exam-
ined conversion of 9a into E-64c (1).
(2) Meara, J. P.; Rich, D. H. J. Med. Chem. 1996, 39, 3357.
(3) (a) Barrett, A. J.; Kembhavi, A. A.; Brown, M. A.; Kirschke,
H.; Knight, C. G.; Tamai, M.; Hanada, K. Biochem. J. 1982,
201, 189. (b) Tamai, M.; Hanada, K.; Adachi, T.; Oguma,
K.; Kashiwagi, K.; Omura, S.; Ohzeki, M. J. Biochem. 1981,
90, 255. (c) Hanada, K.; Tamai, M.; Ohmura, S.; Sawada, J.;
Seki, T.; Tanaka, I. Agric. Biol. Chem. 1978, 42, 529.
(4) See for example: (a) Verhelst, S. H. L.; Bogyo, M.
ChemBioChem 2005, 6, 824. (b) Chehade, K. A. H.;
Baruch, A.; Verhelst, S. H. L.; Bogyo, M. Synthesis 2005,
240. (c) James, K. E.; Asgian, J. L.; Li, Z. Z.; Ekici, O. D.;
Rubin, J. R.; Mikolajczyk, J.; Salvesen, G. S.; Powers, J. C.
J. Med. Chem. 2004, 47, 1553. (d) Detterbeck, R.; Hesse,
M. Helv. Chim. Acta 2003, 86, 222. (e) Roush, W. R.;
Hernandez, A. A.; McKerrow, J. H.; Selzer, P. M.; Hansell,
It was found that this could be achieved via the four-step
sequence shown in Scheme 4. Thus, ester hydrolysis fol-
lowed by carbodiimide-mediated coupling to 3-methyl-
butylamine gave amide 16 in good overall yield. We
found that the EDCI coupling was best performed in
dichloromethane rather than DMF, the latter solvent lead-
ing to significant amounts of epoxide-opening.12 Baeyer–
Villiger oxidation of the phenyl ketone then gave active
ester intermediate 18, which on exposure to ethanolic
Synlett 2006, No. 13, 2063–2066 © Thieme Stuttgart · New York

2066
B. Lygo et al.
LETTER
(9.0 mmol) was added and stirring continued for 4–16 h.
Then, H2O (100 mL) was added and layers separated. The
aqueous was extracted with EtOAc (2 × 100 mL) and the
combined organics were dried (Na2SO4), then concentrated
under reduced pressure. The residue was then purified by
chromatography on silica gel.
E.; Engel, J. C. Tetrahedron 2000, 56, 9747. (f) Goursalin,
B. J.; Lachance, P.; Bonneau, P. R.; Storer, A. C.; Kirschke,
H.; Broemme, D. Bioorg. Chem. 1994, 22, 227.
(g) Giordano, C.; Calabretta, R.; Gallina, C.; Consalvi, V.;
Scandurra, R.; Noya, F. C.; Franchini, C. Eur. J. Med. Chem.
1993, 28, 917. (h) Huang, Z. Y.; McGowan, E. B.; Detwiler,
T. C. J. Med. Chem. 1992, 35, 2048.
Selected NMR Data.
(5) For an alternative approach see: Sarabia, F.; Sanchez-Ruiz,
A.; Chammaa, S. Bioorg. Med. Chem. 2005, 13, 1691.
(6) (a) Lygo, B.; To, D. C. M. Chem. Commun. 2002, 2360.
(b) Lygo, B.; To, D. C. M. Tetrahedron Lett. 2001, 42,
1343. (c) Lygo, B.; Wainwright, P. G. Tetrahedron 1999,
55, 6289. (d) Lygo, B.; Wainwright, P. G. Tetrahedron Lett.
1998, 39, 1599.
(7) For a recent review on epoxidation via asymmetric catalysis,
see: Xia, Q.-H.; Ge, H.-Q.; Ye, C.-P.; Liu, Z.-M.; Su, K.-X.
Chem. Rev. 2005, 105, 1603.
(8) See for example: (a) Jensen, J. S.; Lam, Y.-F.; Helz, G. R.
Environ. Sci. Technol. 1999, 33, 3568. (b) Orton, K. J. P.;
Bradfield, A. E. J. Chem. Soc. 1927, 986.
(9) Lygo, B.; Andrews, B. I. Metal-Catalysed Carbon–Carbon
Bond-Forming Reactions, In Catalysts for Fine Chemical
Synthesis, Vol. 3; Roberts, S. M.; Whittall, J.; Mather, P.;
McCormack, P., Eds.; J. Wiley and Sons, Ltd.: Chichester,
2004, 27–33.
Compound 9a: 1H NMR (400 MHz, CDCl3): d = 7.99–7.95
(2 H, m, ArH), 7.64–7.46 (3 H, m, ArH), 6.60 (1 H, br d, J =
8.5 Hz, NH), 4.72–4.64 (1 H, m, NHCH), 4.27 (1 H, d, J =
2.0 Hz, NHCOCH), 3.78 (3 H, s, OMe), 3.72 (1 H, d, J = 2.0
Hz, ArCOCH), 1.78–1.58 [3 H, m, CH2, CH(CH3)2], 1.01 (3
H, d, J = 6.5 Hz, CH3), 0.99 (3 H, d, J = 6.5 Hz, CH3). 13
C
NMR (100 MHz, CDCl3): d = 191.5 (C), 172.5 (C), 166.2
(C), 134.9 (C), 134.4 (CH), 129.0 (CH), 128.5 (CH), 56.2
(CH), 54.9 (CH), 52.5 (CH3), 50.4 (CH), 41.3 (CH2), 25.0
(CH), 22.7 (CH3), 22.0 (CH3).
Compound 10a: 1H NMR (400 MHz, CDCl3): d = 7.99–7.95
(2 H, m, ArH), 7.64–7.46 (3 H, m, ArH), 6.50 (1 H, br d, J =
8.5 Hz, NH), 4.72–4.64 (1 H, m, NHCH), 4.41 (1 H, d, J =
2.0 Hz, NHCOCH), 3.78 (3 H, s, OMe), 3.73 (1 H, d, J = 2.0
Hz, ArCOCH), 1.78–1.58 [3 H, m, CH2, CH(CH3)2], 1.01 (3
H, d, J = 6.5 Hz, CH3), 0.99 (3 H, d, J = 6.5 Hz, CH3). 13
C
NMR (100 MHz, CDCl3): d = 191.3 (C), 172.9 (C), 166.4
(C), 134.9 (C), 134.4 (CH), 129.0 (CH), 128.5 (CH), 55.8
(CH), 54.9 (CH), 52.5 (CH3), 50.2 (CH), 41.0 (CH2), 24.8
(CH), 22.8 (CH3), 21.7 (CH3).
(10) General Procedure for Asymmetric Epoxidation.
A mixture of 15% aq NaOCl (9.0 mmol) and 12 M aq KOH
(1 mL) was added dropwise to a solution of enone (3.0
mmol) and the appropriate catalyst (0.15 mmol) in PhMe (70
ml). The resulting mixture was stirred vigorously (1000 rpm)
at r.t. for 30 min, then a second portion of 15% aq NaOCl
(11) Tamai, M.; Yokoo, C.; Murata, M.; Oguma, K.; Sota, K.;
Sato, E.; Kanaoka, Y. Chem. Pharm. Bull. 1987, 35, 1098.
(12) For further discussion of this, see: Roush, W. R.; Hernandez,
A. A.; Zepeda, G. Synthesis 1999, 1500.
Synlett 2006, No. 13, 2063–2066 © Thieme Stuttgart · New York