Solution/solid-phase synthesis of partially modified retro-ψ[NHCH(CF3)]-peptidyl hydroxamates

Article abstract of DOI:10.1016/S0040-4039(01)00375-6

The synthesis of a novel family of partially-modified (PM) retropeptidyl hydroxamates incorporating a [CH(CF₃)CH₂CO] unit as a surrogate of the conventional malonyl group, has been accomplished both in solution and in solid-phase. The key step is the Michael-type N-addition of free or polymer bound α-amino hydroxamates to 3-(E-enoyl)-1,3-oxazolidin-2-ones, which takes place very effectively, although with low stereocontrol. A number of tri- and tetra-peptidyl hydroxamates were obtained either in diastereomerically pure form (by solution-phase synthesis, after chromatographic purification), or as mixtures of two epimers in very good chemical purity (by solid-phase, after release from the resin), demonstrating that this method is suitable for preparing combinatorial libraries of PM retro-ψ[NHCH(CF₃)]-peptidyl hydroxamates for screening as metalloprotease inhibitors.

Full text of DOI:10.1016/S0040-4039(01)00375-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3141–3144
Solution/solid-phase synthesis of partially modified
retro-´[NHCH(CF3)]-peptidyl hydroxamates
Alessandro Volonterio,b,* Pierfrancesco Bravoa,b and Matteo Zandaa,*
aDipartimento di Chimica del Politecnico di Milano, via Mancinelli 7, I-20131 Milan, Italy
bC.N.R.-Centro di Studio sulle Sostanze Organiche Naturali, via Mancinelli 7, I-20131 Milan, Italy
Received 23 February 2001; revised 26 February 2001; accepted 5 March 2001
Abstract—The synthesis of
a novel family of partially-modified (PM) retropeptidyl hydroxamates incorporating a
[CH(CF3)CH2CO] unit as a surrogate of the conventional malonyl group, has been accomplished both in solution and in
solid-phase. The key step is the Michael-type N-addition of free or polymer bound a-amino hydroxamates to 3-(E-enoyl)-1,3-oxa-
zolidin-2-ones, which takes place very effectively, although with low stereocontrol. A number of tri- and tetra-peptidyl
hydroxamates were obtained either in diastereomerically pure form (by solution-phase synthesis, after chromatographic purifica-
tion), or as mixtures of two epimers in very good chemical purity (by solid-phase, after release from the resin), demonstrating that
this method is suitable for preparing combinatorial libraries of PM retro-„[NHCH(CF3)]-peptidyl hydroxamates for screening as
metalloprotease inhibitors. © 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
2.1. Solution-phase synthesis
The search for high-affinity, selective and relatively
non-toxic inhibitors of metalloproteases is a very
important issue in the pharmaceutical field.1 These pro-
teolytic enzymes have a key role in a number of serious
pathological conditions, such as inflammatory diseases,
atherosclerosis and cancers, and their inhibition consti-
tutes a primary therapeutic target.2 The field of syn-
thetic metalloprotease inhibitors is dominated by com-
The backbone of PM retro-„[NHCH(CF3)]-peptidyl
hydroxamates was built through a Michael-type N-addi-
tion of amino acid-derived O-Bn hydroxamates 1 with
the chiral 3-(E-enoyl)-1,3-oxazolidin-2-one 2 (Scheme
1).6 Compounds 1 were prepared by 1-hydroxy-benzotri-
azole/diisopropyl/carbodiimide (HOBt/DIC) promoted
condensation of N-Boc a-amino acids with O-Bn-
hydroxylamine hydrochloride in DMF/triethylamine
(TEA), followed by N-Boc-cleavage with trifluoroacetic
pounds containing
a terminal hydroxamate func-
tion, since the HONHCO- end group is very effective in
coordinating the Zn2+ cofactor of metalloproteases.3
This explains the current interest in developing novel
synthetic routes and novel structural classes of hydrox-
amate peptidomimetics, possibly using solid-phase–
combinatorial techniques which provide ready access to
libraries of compounds for fast simultaneous screening,
from which the most potent inhibitors are selected.4
This paper describes the efficient solution and solid-
phase synthesis of partially-modified (PM) retro-
„[NHCH(CF3)]-peptidyl hydroxamates, a novel class of
hydroxamates incorporating a [CH(CF3)CH2CO] unit
as a surrogate of the malonyl moiety featured in con-
ventional PM retropeptides (Fig. 1).5
O
O
H
N
H
N
RHN
NHR3
R1
O
O
R2
Partially-modifed Malonate-based
retro- [NHCO]-peptide
O
O
H
N
H
N
HO
*
N
H
NHR3
R1 CF3
O
R2
Partially-modified
retro- [NHCH(CF3)]-peptidyl hydroxamate
* Corresponding authors. Fax: +039 02 2399-3080; e-mail: alessandro.
Figure 1.
volonterio@dept.chem.polimi.it; zanda@dept.chem.polimi.it
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00375-6

3142
A. Volonterio et al. / Tetrahedron Letters 42 (2001) 3141–3144
R1
HOBt/DIC
TEA, DMF
R1
.
H
N
R1
TFA, DCM
10 min, r.t.
H
N
HCl
NH2
+
O
.
BnO
NH2 TFA
(> 98%)
BnO
NHBoc
12 h, r.t.
HO2C
NHBoc
O
O
1
(65-76%)
R1
O
Bn
CF3
H
N
Bn
O
O
Products
R1
Ratio 3:4a Yield (%)b
BnO
BnO
N
H
N
F3C
N
O
O
3
3a and 4a Me
3b and 4b iso-Pr 1.0 : 1.2
3c and 4c Bn 1.5 : 1.0
1.0 : 2.0
94
97
O
2
+
O
R1
O
Bn
CF3
H
N
> 98
TMP, DCM, 72 h, r.t.
N
H
N
a Determined by 1H and 19F NMR of the crude reaction
mixture. b Isolated yields.
O
O
4
O
Scheme 1.
acid (TFA) in dichloromethane (DCM). The
diastereomeric adducts 3 and 4a–c were formed by
conjugate addition of 1 to 2 (2 equiv.) in DCM/sym-
collidine (TMP) (2 equiv.) for 72 h at rt. These reac-
tions were very clean and high yielding, although low
diastereocontrol was achieved in all cases.7 However, in
view of combinatorial applications the low stereocon-
trol is not necessarily a drawback, because both epimers
at the trifluoromethyl (Tfm) stereocentre can be pro-
duced. Diastereomerically pure 3 and 4a–c were iso-
lated by flash chromatography (FC).
The resulting resin-bound a-amino hydroxamate was
submitted to 1,4-conjugate addition with the achiral
oxazolidin-2-one 10 (3 equiv. in DCM, 3 days, rt). The
FT-IR spectrum of the resulting Tfm-resin 11 exhibited
a strong O(CO)N band at 1785 cm−1, totally absent in
the precursor 9. Treatment of 11 with lithium hydrox-
ide (1 equiv. based on the theoretical loading) and
hydrogen peroxide (4 equiv.) in THF/H2O (0°C, 2 h)
cleaved the oxazolidin-2-one with excellent chemoselec-
tivity. As a result, the FT-IR of the CO2H resin 12
showed disappearance of the O(CO)N band at 1785
cm−1, and transformation of the amide band (1700
cm−1) into an intense band at 1620 cm−1, which could
be assigned to the carboxyl residue. Coupling of 12 to
a-amino esters 13 (DIC/HOAt, TMP/DMAP11)
afforded the tripeptidyl resins 14, from which the
hydroxamates 15 were released in good yields and
purity upon treatment with TFA/DCM (1 h, rt).12 As
Exocyclic cleavage of the oxazolidin-2-one was per-
formed by treatment of 4a and 4b with in situ generated
lithium hydroperoxide (Scheme 2),8 which delivered the
expected acids 5 without affecting the hydroxamate
moiety. The target PM retro-tripeptidyl hydroxamates
7a and 7b were obtained via HATU/HOAt9 promoted
1
coupling with
(6b), respectively, followed by catalytic hydrogenolysis
of the terminal OBn groups.
L
-Val-OBn (6a) and
L
-Phe-O-tert-Bu
expected, H and 19F NMR spectroscopy showed that
15a–e were formed as nearly equimolar mixtures of
epimers at the Tfm-substituted centre. For R3=tert-
butyl, TFA treatment resulted in concomitant hydroly-
sis of the terminal ester function delivering the free
carboxyl derivatives 15a,b, which were found to be
indefinitely stable upon storage at 4°C.
2.2. Solid-phase synthesis
This method was also adapted to the solid-phase, with
the view of preparing combinatorial libraries of PM
retro-„[NHCH(CF3)]-peptidyl
hydroxamates
for
In contrast, the hydroxamates having a terminal benzyl
ester function, obtained directly from 14c–e by TFA
release, were found to be rather unstable under the
same storage conditions, producing complex mixtures
of products within a few days. Thus, the ester functions
of polymers 14c–e were first hydrolysed with lithium
hydroxide, then TFA treatment of the resins released
the stable carboxyl-free hydroxamates 15c–e, with very
good overall yields and purities.
screening as metalloprotease inhibitors. First, we
addressed the preparation of retro-tripeptidyl hydroxa-
mates 15, having a CO2H terminus (Scheme 3). The
hydroxylamine resin 8 was prepared in two steps from
commercial Wang resin, according to the method of
Floyd,10 and coupled to an excess of
L-Fmoc-Ala to
give the protected alanine polymer 9, from which the
Fmoc group was cleaved with 20% piperidine in DMF.
R2
CO2-R3
CF3
O
R2
R1
H
N
R1
CF3
.
CO2-R3
H N
2
HCl
4a
or
H
N
HO
N
H
N
H
LiOH/H2O2
6a,b
CO2H
O
7
BnO
N
H
a R1 = Me, R2 = iso-Pr, R3 = H
(R3 = Bn for 6a) (88%)
b R1 = iso-Pr, R2 = Bn, R3 = tert-Bu (85%)
0 °C, 30 min
1. HATU/HOAt
TMP, DMF
O
5
a R1 = Me (75%)
b R1 = iso-Pr (60%)
4b
2. Pd(OH)2/C, H2
Scheme 2.

A. Volonterio et al. / Tetrahedron Letters 42 (2001) 3141–3144
3143
Ref. 10
H
N
Fmoc-L-Ala-OH
1. Piperidine, DMF
O
OH
NH2
NHFmoc
O
O
HOBt/DIC, DMAP
DMF/DCM
O
2.
9
O
8
Wang resin
O
N
CF3
10
R2
HCl
13
CF3
O
O
.
H
CF3
*
R1HN CO2-R3
LiOH (1 equiv.),
H2O2 (4 equiv.)
H
N
N
CO2H
N
H
N
O
*
N
O
HOAt/DIC
TMP-DMAP (cat.)
DMF
O
11
O
H
12
O
O
CF3
*
R2
TFA
.
TFA/DCM
H
N
CF3
O
R2
H
N
N
H
N
R1
CO2R3
O
(for a and b)
N
H
N
CO2H
HO
*
14
O
R1
15
O
a, b R3 = tert-Bu; c, d, e R3 = Bn
a R1 = H, R2 = H, yield 60%, purity 87%
b R1 = H, R2 = CH3, yield 60%, purity 79%
c R1 = R2 = -(CH2)3- yield 80%, purity 74%
d R1 = H, R2 = iso-Bu yield >98%, purity 96%
e R1 = H, R2 = Bn yield not det., purity 73%
(for c-e)
1. LiOH (4 equiv.)
2. TFA/DCM
Scheme 3.
Next, in order to demonstrate the versatility of the
method, we addressed the solid-phase preparation of
tetrapeptidyl hydroxamates having a CO2H terminus
(20 and 21), and tripeptidyl hydroxamates having a
methylamide terminus (23), which is often encountered
in metalloprotase inhibitors (Scheme 4).1 Tripeptidyl
hydroxamate polymers 16, having a CO2H end-group,
were obtained from 12 following the established proto-
were released with TFA/DCM in excellent purity as
mixtures of two epimers.
Finally, the retro-tripeptidyl hydroxamic amide 23 was
prepared in good purity upon coupling of the polymer
16e to methylamine (HOAt/DIC, TEA/DMAP), fol-
lowed by the usual release with TFA/DCM.
col. Coupling of 16d and 16b to
L
-Phe-O-tert-Bu (6b)
In summary, we have developed both a solution and a
solid-phase protocol to prepare a novel structural fam-
ily of fluorinated retro-peptidyl hydroxamates.
Although in this work we did not perform a true
and -Val-O-tert-Bu (17) (HOAt/DIC, TMP/DMAP),
L
afforded the resin-bound tetrapeptidyl derivatives 18
and 19, respectively. The free hydroxamates 20 and 21
R2 = iso-Bu and Me
HCl CO Me(Bn)
H2N R2
1.
2
.
CF3
*
O
R2
H
N
(16d and 16b)
12
N
H
N
H
CO2H
O
HOAt/DIC
HOAt/DIC
R4
HCl
17, 6b
TMP-DMAP
O
TMP-DMAP (cat.), DMF
.
16b,d,e
DMF
H2N CO2-tert-Bu
2. LiOH
CH3NH2
R2 = Bn
(4 equiv.)
HOAt/DIC
DMAP (cat.), TEA
DMF
CF3
O
R2
H
N
H
N
(16e)
CO2-R5
N
H
N
H
O
*
O
O
R4
18 R2 = iso-Bu, R4 = Bn, R5 = tert-Bu
19 R2 = Me, R4 = iso-Pr, R5 = tert-Bu
CF3
O
Bn
H
N
H
N
N
H
N
H
CH3
O
*
22
O
O
TFA/DCM
TFA/DCM
CF3
.
R2
TFA
HO
CF3
*
O
H
N
H
N
.
H
N
CO2H
TFA
O
Bn
N
H
N
H
H
N
O
O
R4
N
*
N
H
CH3
HO
20 R2 = iso-Bu, R4 = Bn, yield 68%, purity >98%
21 R2 = Me, R4 = iso-Pr, yield 95%, purity 92%
H
O
O
yield not det., purity 73%
23
Scheme 4.

3144
A. Volonterio et al. / Tetrahedron Letters 42 (2001) 3141–3144
combinatorial approach to the title compounds, the
examples shown provide good evidence that the method
is general and could be applied for that purpose. Evalu-
ation of some of the title compounds as matrix metallo-
protease inhibitors is currently in progress, and will be
reported in a full paper.
Goodman, M.; Chorev, M. Acc. Chem. Res. 1979, 12,
1–7.
6. (a) Shibuya, A.; Kurishita, M.; Ago, C.; Taguchi, T.
Tetrahedron 1996, 52, 271–278; (b) See also: Yamazaki,
T.; Shinohara, N.; Kitazume, T.; Sato, S. J. Fluorine
Chem. 1999, 97, 91–96.
7. (a) These Michael-type additions with a-amino hydroxa-
mates are remarkably less stereoselective than those
involving the corresponding a-amino esters (see Ref.
Acknowledgements
5a,b). For example, L-Val-OBn reacted with 2 providing
an 86:14 ratio in favour of the 4b analogue. The reasons
for this rather surprising drop of stereoselectivity are
presently unclear, but we have found that the steric
hindrance of the CO2R has a minor influence on the
stereocontrol of a-amino ester additions. In fact, nearly
the same degree of stereoselectivity was observed when
We thank C.N.R.–C.S.S.O.N for partial financial
support.
References
R=benzyl or tert-butyl in the case of
L-Phe-OR. This
1. Beckett, R. P.; Whittaker, M. Exp. Opin. Ther. Patents
1998, 8, 259–282.
2. Leung, D.; Abbenante, G.; Fairlie, D. P. J. Med. Chem.
2000, 43, 305–341.
3. Babine, R. E.; Bender, S. L. Chem. Rev. 1997, 97, 1359–
1472.
suggests that the low stereocontrol observed in this work
could be due to the fact that a-amino acid-derived esters
and hydroxamates react with 2 through different confor-
mations, with a weaker stereodirecting effect in the case
of hydroxamates. (b) Excess of 2 could be recovered
quantitatively by FC. (c) The absolute stereochemistry of
hydroxamates 3 and 4, and their derivatives, was unam-
biguously assigned by means of X-ray diffraction and
chemical correlation with known „[NHCH(CF3)]-
retropeptides (see Ref. 5a). The details will be published
in a full paper.
4. For recent examples, see: (a) Hanessian, S.; Moitessier,
N.; Wilmouth, S. Tetrahedron 2000, 56, 7643–7660; (b)
Scozzafava, A.; Supuran, C. T. J. Med. Chem. 2000, 43,
1858–1865; (c) Wang, L.; Phanstiel, O., IV J. Org. Chem.
2000, 65, 1442–1447; (d) Thouin, E.; Lubell, W. D.
Tetrahedron Lett. 2000, 41, 457–460; (e) Robinson, D. E.;
Holladay, M. W. Org. Lett. 2000, 2, 2777–2779; (f)
Buchardt, J.; Bruun Schiødt, C.; Krog-Jensen, C.;
Delaisse´, J.-M.; Tækker Foged, N.; Meldal, M. J. Comb.
Chem. 2000, 2, 624-638.
5. (a) For the solution-phase synthesis of „[NHCH(CF3)]-
retropeptides, see: Volonterio, A.; Bravo, P.; Zanda, M.
Org. Lett. 2000, 2, 1827–1830; (b) For the solid-phase
synthesis, see: Volonterio, A.; Bravo, P.; Moussier, N.;
Zanda, M. Tetrahedron Lett. 2000, 41, 6517–6521; (c) For
a review about retropeptides, see: Fletcher, M. D.; Camp-
bell, M. M. Chem. Rev. 1998, 98, 763–795; (d) See also:
8. The chiral auxiliary could be recovered in high yields.
Evans, D. A.; Britton, T. C.; Ellman, J. A.; Dorow, R. L.
J. Am. Chem. Soc. 1990, 112, 4011–4030.
9. HATU, N,N,N%,N%-tetramethyl-O-(7-azabenzo-triazol-1-
yl)uronium hexafluorophosphate; HOAt, 1-hydroxy-7-
azabenzotriazole.
10. Floyd, C. D.; Lewis, C. N.; Patel, S. R.; Whittaker, M.
Tetrahedron Lett. 1996, 37, 8045–8048.
11. DMAP, 4-dimethylamino-pyridine.
12. Yields were calculated on the amount of final products
released from the resins, referring to the loading of the
starting Wang resin (1.3 mmol/g).
.
.