Design and synthesis of β-amino-α-hydroxy amide derivatives as inhibitors of MetAP2 and HUVEC growth

Article abstract of DOI:10.1016/j.bmcl.2004.04.004

The rational design and synthesis of β-amino-α-hydroxy amide derivatives as reversible inhibitors of methionine aminopeptidase-2 (MetAP2) with anti-proliferative activity against human umbilical vein endothelial cells (HUVECs) is described.

Full text of DOI:10.1016/j.bmcl.2004.04.004

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3181–3184
Design and synthesis of b-amino-a-hydroxy amide derivatives as
inhibitors of MetAP2 and HUVEC growth
Martin Sendzik,* James W. Janc, Ronnel Cabuslay, Lee Honigberg,
Richard L. Mackman, Catherine Magill, Neil Squires and Nancy Waldeck
Departments of Medicinal Chemistry and Biology, Celera, 180 Kimball Way, South San Francisco, CA94080, USA
Received 6 December 2003; revised 1 April 2004; accepted 2 April 2004
Abstract—The rational design and synthesis of b-amino-a-hydroxy amide derivatives as reversible inhibitors of methionine
aminopeptidase-2 (MetAP2) with anti-proliferative activity against human umbilical vein endothelial cells (HUVECs) is described.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
vein endothelial cells (HUVECs) had not been disclosed
to our knowledge.7 Therefore, our studies focused on
the design and synthesis of reversible bestatin-based
inhibitors of MetAP2, which showed growth inhibition
properties against HUVECs.
Folkman proposed in 1974 that the inhibition of angio-
genesis is potentially a promising approach for the
treatment of cancer.1 The antibiotic fumagillin (1a),
along with related natural analogues, was shown to have
potent anti-angiogenic activity and the synthetic ana-
logue TNP-470 (1b) has been in clinical trials against a
variety of cancer types.2 The target for fumagillin has
been identified as a type 2 methionine aminopeptidase
(MetAP2), a metal-dependent enzyme that is involved in
post-translational protein processing.3
The structure–activity relationship (SAR) around be-
statin-based analogues revealed that the hydrophobic
side chain (P1) contributes significantly to the bind-
ing. For example, the ethyl sulfide 2 (Fig. 1) with an
IC50 ¼ 100 nM against MetAP2 is one of the more po-
tent analogues in its class.5c;8 SAR around fumagillin
The natural product fumagillin irreversibly binds to
MetAP2 and has been shown to inhibit the growth of
endothelial cells.4 Binding is highly selective towards
MetAP2 over MetAP1. However, the irreversible bind-
ing of TNP-470 to MetAP2 could be responsible for the
toxicity that was observed in patients. Consequently, a
variety of different reversible MetAP2 inhibitors has
been investigated.5;6
O
OMe
1
O
R
O
O
O
1a Fumagillin
1b TNP-470
1c CKD-731
R =
R =
R =
OH
H
N
b-Amino-a-hydroxy amide based (bestatin-based)
inhibitors of methionine aminopeptidases have been
reported as reversible inhibitors in the literature.5 Their
activity towards growth inhibition of human umbilical
Cl
O
OMe
OMe
OMe
NH2
O
OEt
2
Keywords: Angiogenesis; MetAP2; Reversible inhibitor; HUVEC.
* Corresponding author. Tel.: +1-650-866-6533; fax: +1-650-866-6655;
e-mail: martin.sendzik@celera.com
S
N
H
OH
O
Figure 1. Known MetAP2 inhibitors.
Present address: Gilead Sciences, Foster City, CA 94404, USA.
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.004

3182
M. Sendzik et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3181–3184
Boc
Boc
Boc
HN
HN
NH
NH2
a, b, c, d
g
e, f,
+
COOMe
OH
COOR
OH
CN
OH
D
O
OH
4
3
5
6 : R=CH3
7 : R=H
h
O
l, m
OMe
OMe
HCl
11
OMe
.
HO
H2N
OMe
OMe
O
H
OMe
OMe
i, j, k
N
N
OMe
NH2
O
H
8
+
O
N
H
OMe
OH
BocHN
NH2
10
11
9
Scheme 1. Exemplary synthesis of b-amino-a-hydroxy amide derivatives. Reagents and conditions: (a) Boc2O, NEt3, THF; (b) HOBtꢀH2O, EDC,
(MeO)NHMeꢀHCl, NEt3; (c) 1 M LiAlH4, Et2O; (d) acetone cyanohydrin, Et3N, CH2Cl2; (e) conc. HCl, dioxane, reflux; (f) 4 M HCl/dioxane,
MeOH; (g) i. Boc2O, Et3N, THF; ii. separation of isomers; (h) 1 M aqNaOH, THF, MeOH; (i) 8, (COCl)2, cat. DMF, EtOAc; (j) 9, Et3N, THF;
(k) 4 M HCl/dioxane, MeOH; (l) 10, HOBtꢀH2O, EDC, Et3N; (m) 4 M HCl/dioxane, MeOH.
uncovered the acyl group as essential for biological
activity.9 The trimethoxy cinnamic ester CKD-731 (1c)
shows remarkably high anti-proliferative activity
against calf endothelial cells (IC50 ¼ 0.03 pg/mL).10
3. Structure–activity relationships
The amino alcohol analogues 12–31 synthesized in this
study were tested for their ability to inhibit recombinant
human MetAP2 (rhMetAP2). The activity assays were
performed using a modification of a literature assay.14
Furthermore, anti-proliferative activities of the ana-
logues were evaluated against HUVECs.15 The in vitro
inhibitory enzyme and cellular activities are summarized
in Table 1.
We hypothesized that a molecule bearing both structural
fragments, the b-amino-a-hydroxy acyl entity of 2 and
cinnamoyl moiety of 1c, joined by an appropriate linker
should have good inhibitory activity against both
MetAP2 and endothelial cell growth. Amides with a
variety of diamines linkers including para-phenylene-
diamine as a linker were investigated.11 Several com-
pounds of this scaffold have been prepared and
evaluated against MetAP2 and growth of HUVECs.
Compound 12 shows, as we hypothesized, inhibitory
activity against MetAP2 and endothelial cells. In com-
parison to 12 structural modifications at the P1 side
(e.g., 11–15) revealed significant changes in the enzyme
activity but not towards activity of endothelial cell
growth. The ethyl sulfide derivative 12 and phenyl-
ethylene derivative 15 showed highest binding affinities.
Loss of potency was also observed for the R-amino-R-
alcohol 16 compared to its R,S-epimer 11.
2. Chemistry
The synthesis of compound 11, as a representative
example of b-amino-a-hydroxy amide derivatives, is
outlined in Scheme 1. The required N-Boc protected
b-amino-a-hydroxy acid (7) was obtained from the
commercially available a-amino acid in an 8-step
Changes with additional substituents like methoxy and
methyl in the linker are tolerated with respect to activity
(e.g., 17, 18 and 20, 21). However, replacement of the
phenylenediamine in 15 with a saturated cyclohexyl-
enediamine linker (22) is accompanied by a lack of
activity towards endothelial cell growth. Modifications at
the P10 side did not alter the enzyme activity (e.g., 23–28),
but decreased significantly the activity towards endo-
thelial cells. The 3,4,5-trimethoxy cinnamic amides 12–
21, with the exception of 19, showed the most potent
inhibitory activity towards endothelial cells. Complete
removal of the cinnamic amide in 12 led to a compound
(28) without any detectable cellular activity despite
demonstrating significant potency against MetAP2. A
direct correlation between cellular and enzyme assay
results could not be established. Therefore, we hypothe-
size that the activity against HUVECs is the consequence
of a more complex mechanism of action. Similar SAR
towards calf pulmonary artery endothelial cells have
been described for fumagillin/CKD-731 analogues.9
sequence.12;13 Starting from
D-3-cyclohexyl alanine (10),
N-Boc protection, Weinreb amide formation, and
reduction with LiAlH4 gave the aminoaldehyde. Addi-
tion of acetone cyanohydrin to the aldehyde provided a
mixture of two diastereomeric cyanohydrines (4), which
could not be separated at this stage of the synthesis.
Hydrolysis of 4 to the acid was accompanied by the
deprotection of the amine. Esterification and N-Boc
protection proceeded in good yields after which dia-
stereomeric 5 and 6 could be readily separated by flash
chromatography. The desired acid 7 was obtained by
saponification of the corresponding ester 6.
Linker 9 was coupled first to the cinnamic acid 8 and
following N-Boc deprotection then coupled to 7. Final
deprotection of the amine gave 11 as its hydrochloride
salt, which was purified by HPLC or alternatively by
recrystallization.

M. Sendzik et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3181–3184
Table 1. Enzyme and cellular activities of derivatives
3183
O
NH2
linker
P1'
P1
OH
P1
linker
P1'
O
O
O
O
H
N
a
S
OMe
OMe
N
H
a
a
d
e
N
H
g
h
H
b
c
d
e
OMe
CH3
H
N
OMe
O
O
OMe
b
N
H
N
H
N
H
b
c
OMe
OEt
OEt
OMe
H
N
OMe
c
O
N
H
N
H
H
N
f
OMe
d
Compd#
P1–linker–P10
IC50 MetAP2 (nM)
GI50 HUVEC (lM)
GI @ 5 lM HUVEC (%)
11
12
13
14
15
b–a–b
a–a–b
c–a–b
d–a–b
e–a–b
177
56
0.67
nda
>98
89
755
356
67
nd
93
0.80
0.85
>98
>98
16
epi–b–a–b
2240
2.0
96
17
18
19
20
21
22
b–b–b
b–c–b
b–d–b
e–b–b
e–c–b
e–d–b
308
230
2200
158
82
1.2
1.8
nd
>98
>98
52
0.70
2.3
nd
>98
92
98
46
23
24
25
26
27
28
29
30
31
a–a–c
a–a–e
a–a–d
a–a–g
a–a–h
a–a–a
b–a–e
b–a–f
e–a–e
33
46
nd
nd
nd
nd
nd
nd
1.0
nd
1.8
43
91
35
43
48
20
45
58
15
0
217
209
169
>98
42
>98
a nd: not determined.
4. Conclusion
References and notes
1. (a) Folkman, J. Adv. Cancer Res. 1974, 19, 331–358; (b)
Folkman, J. Adv. Cancer Res. 1985, 43, 175–203; (c)
Folkman, J. Nature Med. 1995, 1, 27–31.
2. Ingber, D.; Fujita, T.; Kishmoto, S.; Sudo, K.; Kanamaru,
T.; Folkman, J. Nature 1990, 348, 555–557.
3. Reviews: (a) Lowther, W. T.; Matthews, B. W. Biochim.
Biophys. Acta 2000, 1477, 157–167; (b) Lowther, W. T.;
Matthews, B. W. Chem. Rev. 2002, 102, 4581–4607;
(c) Vaughan, M. D.; Sampson, P. B.; Honek, J. F. Curr.
Med. Chem. 2002, 9, 385–409.
In conclusion, we have successfully designed and
synthesized reversible inhibitors of MetAP2 with anti-
proliferation activity against HUVECs by introducing
the 3,4,5-trimethoxycinnamoyl moiety into b-amino-a-
hydroxyamide derivatives. The cellular assay results do
not correlate with the inhibition of MetAP2. Further
investigation is required to gain more insight into the
mechanism of action of these compounds.
4. Wang, J.; Lou, P.; Henkin, J. J. Cell. Biochem. 2000, 77,
465–473.
5. (a) Keding, S. J.; Dales, N. A.; Lim, S.; Beaulieu, D.; Rich,
D. H. Synth. Commun. 1998, 28, 4463–4470; (b) Lim, S.;
Rich, D. H. In Peptides, Chemistry, Structure and Biology;
Hodges, R. S., Smith, J. A., Eds.; ESCOM, 1994; pp 625–
627; (c) Craig, R. A.; Henkin, J.; Kawai, M.; Lynch, L.
M.; Patel, J.; Sheppard, G. S.; Wang, J.; U.S. Patent
Acknowledgements
The authors would like to acknowledge the support of
Dr. M. Venuti, Dr. M. Green, Dr. J. R. Spencer, Dr. P.
A. Sprengeler, and Dr. K. E. Wesson.

3184
M. Sendzik et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3181–3184
6,242,494, 2001; (d) Craig, R. A.; Kawai, M.; Lynch, L.
11. Unpublished results.
M.; Patel, J.; Sheppard, G. S.; Wang, J.; Yang, F;
BaMaung, N. Y., Abbot Lab., Intl. Pat. Appl. PCT/
US01/12274, WO 01/79157, April 13, 2001; (e) Disclosed
after submission of our manuscript: Sheppard, G. S.;
Wang, J.; Kawai, M.; BaMaung, N. Y.; Craig, R. A.;
Erickson, S. A.; Lynch, L.; Patel, J.; Yang, F.; Searle, X.
B.; Lou, P.; Park, C.; Kim, H. H.; Henkin, J.; Lesniewski,
R. Bioorg. Med. Chem. Lett. 2004, 14, 865–868.
12. Review: Andruszkiewicz, R. Polish J. Chem. 1998, 72, 1–48.
13. Rich, H. D.; Moon, B. J.; Boparai, A. S. J. Org. Chem.
1980, 45, 2280–2290.
14. (a) Yang, G.; Kirkpatrick, R. B.; Ho, T.; Zhang, G.-F.;
Liang, P.-H.; Johanson, K. O.; Casper, D. J.; Doyle, M.
L.; Marino, J. P.; Thomson, S. K.; Chen, W.; Tew, D. G.;
Meek, T. D. Biochemistry 2001, 40, 10645–10654; (b)
Enzyme inhibition assays: The assays were carried out at
room temperature (21–25 °C) in 96-well clear round-
bottom plates (Corning Inc., Corning, NY) in a final
volume of 100 lL. Assays were performed in a buffer
containing in 50 mM MOPS (pH 7.5), 100 mM KCl,
0.001% bovine serum albumin, and 2% DMSO. Rh-
MetAP-2 (30 nM, estimated by active site titration using
fumagillin) was incubated with varying amounts of
6. (a) Several non-bestatin based reversible inhibitors have
been disclosed while our research progressed. Some
examples: Marino, J. P.; Ryan, M. D.; Thompson, S.
K.; Veber, D. F. PCT Int. Appl. 2001, WO 00/1024796; (b)
Kallander, L. S.; Thompson, S. K. PCT Int. Appl. 2001,
WO 00/178723; (c) Marino, J. P.; Thompson, S. K.; Veber,
D. F. PCT Int. Appl. 2002, WO 00/205804; (d) Perron-
ꢀ
Sierra, F. M.; Pierre, A.; Burbridge, M.; Guilbaud, N.
inhibitor. Following 30 min incubation, 300 lM L-methio-
Bioorg. Med. Chem. Lett. 2002, 10, 1463–1466; (e) Zhou,
G.; Tsai, C. W.; Liu, J. O. J. Med. Chem. 2003, 3452–3454;
(f) Towbin, H.; Stolz, B.; Wood, J.; DeCaprio, J. A.;
Phillips, P. E. 93rd Annual Meeting of AACR, April 6–10,
2002; Abstr. 4732; (g) Luo, Q.-L.; Li, J.-Y.; Liu, Z.-Y.;
Chen, L.-L.; Li, J.; Qian, Z.; Shen, Q.; Li, Y.; Lushington,
G. H.; Ye, Q.-Z.; Nan, F.-J. J. Med. Chem. 2003, 2631–2640.
7. A reversible bestatin-based inhibitor of MetAP2 with
activity towards HUVEC proliferation has been disclosed
only a few days before the submission of this manuscript:
Wang, J.; Sheppard, G. S.; Lou, P.; Kawai, M.; BaMaung,
N.; Erikson, S. A.; Tucker-Garcia, L.; Park, C.; Bouska,
J.; Wang, Y.-C.; Frost, D.; Tapang, P.; Albert, D. H.;
Morgan, S. J.; Morowitz, M.; Shusterman, S.; Maris, J.
M.; Lesniewski, R.; Henkin, J. Cancer Res. 2003, 63,
7861–7869.
8. Note that 2 is a mixture of two diastereomers. It has been
shown that the R,S-amino alcohol derivatives have higher
binding affinities than their corresponding R,R isomers;
see also Ref. 3.
9. Han, C. K.; Ahn, S. K.; Choi, N. S.; Hong, R. K.; Moon,
S. K.; Chun, H. S.; Lee, S. J.; Kim, J. W.; Hong, C. I.;
Kim, D.; Yoon, J. H.; No, K. T. Bioorg. Med. Chem. Lett.
2000, 10, 39–43.
nine-7-amido-4-methylcoumarin was added to initiate the
reaction. The rate of substrate hydrolysis was continu-
ously monitored for 6 min (kex ¼ 355 nm, kem ¼ 460 nm)
using an fmax fluorescent plate reader (Molecular Devices,
Sunnyvale, CA); (c) Data analysis methods: Apparent
inhibition constants, k0i, were calculated from the velocity
data generated at the various inhibitor concentrations
using the software package, Batch Ki (Biokin Ltd.,
Pullman, WA): Kuzmic, P.; Sideris, S.; Cregar, L. M.;
Elrod, K. C.; Rice, K. D.; Janc, J. W. Anal. Biochem. 2000,
281, 62–67; Batch Ki provides a parametric method for the
determination of inhibitor potency using a transformation
of the tight binding inhibition model as described in
Morrison, J. F. Biochem. Biophys. Acta 1969, 185, 269–
286.
15. HUVEC proliferation assay: Cells were seeded at a density
of 1500 cells/well in 96-well Wallac flat-bottom isoplates,
cultured for 24 h @ 37 °C and then treated with com-
pounds (dose-response ranging from 0.04 to 10 lM) for
24 h. Cell proliferation was measured by 3H-thymidine
incorporation using the Scintillation proximity assay (3H-
thymidine uptake assay systemTM; Amersham Life Sci-
ence). Plates were counted in the Wallac 1450 MicroBeta
scintillation counter. GI50 values for compounds were
3
10. The activity is towards calf pulmonary artery endothelial
cells. This IC50 corresponds to IC50 ¼ 58 fM.
determined by plotting the percent control H-thymidine
uptake against the compound concentration.