Fumarranol, a rearranged fumagillin analogue that inhibits angiogenesis in vivo

Article abstract of DOI:10.1021/jm060559v

The fumagillin family of natural products inhibits angiogenesis through the irreversible inhibition of the type 2 methionine aminopeptidase (MetAP2). Herein is reported a novel fumagillin analogue named fumarranol. It is shown that, like fumagillin, fumarranol selectively inhibits MetAP2 and endothelial cell proliferation. It is also active in a mouse model of angiogenesis in vivo. Unlike TNP-470, fumarranol does not covalently bind to MetAP2. Fumarranol may serve as a lead for a new class of angiogenesis inhibitors.

Full text of DOI:10.1021/jm060559v

© Copyright 2006 by the American Chemical Society
Volume 49, Number 19
September 21, 2006
Letters
Fumarranol, a Rearranged Fumagillin Analogue
That Inhibits Angiogenesis in Vivo
Jun Lu,† Curtis R. Chong,†,‡ Xiaoyi Hu,† and
Jun O. Liu*,†,§,#
Department of Pharmacology and Molecular Sciences, Medical
Scientist Training Program, Solomon H. Snyder Department of
Neuroscience, and Department of Oncology, The Johns Hopkins
UniVersity School of Medicine, Baltimore, Maryland 21205
ReceiVed May 11, 2006
Abstract: The fumagillin family of natural products inhibits angio-
genesis through the irreversible inhibition of the type 2 methionine
aminopeptidase (MetAP2). Herein is reported a novel fumagillin
analogue named fumarranol. It is shown that, like fumagillin, fumarranol
selectively inhibits MetAP2 and endothelial cell proliferation. It is also
active in a mouse model of angiogenesis in vivo. Unlike TNP-470,
fumarranol does not covalently bind to MetAP2. Fumarranol may serve
as a lead for a new class of angiogenesis inhibitors.
Figure 1. Structures of fumagillin and synthetic analogues.
with a completely rearranged skeleton, named fumarranol, that
inhibits MetAP2 and blocks angiogenesis (Figure 1).
Angiogenesis, the formation of new blood vessels, has been
implicated in the pathogenesis of several important human
diseases, including cancer, diabetic retinopathy, and age-related
macular degeneration. Inhibition of angiogenesis is emerging
as an effective new strategy for the treatment of angiogenesis-
dependent diseases.1,2 One of the most potent classes of small
molecule inhibitors is from the fumagillin family.3 Fumagillin,
its synthetic analogue TNP-470, and ovalicin have been shown
to specifically bind to type 2 methionine aminopeptidase
(MetAP2).4,5 In a mechanism that remains to be completely
elucidated, inhibition of MetAP2 by these small molecule
inhibitors led to the transcriptional activation of p53, which in
turn activates the expression of p21 that inhibits cyclinE‚Cdk2,
accounting for the cell cycle blockade by these inhibitors.6,7
Since the identification of MetAP2 as the target for fumagillin
and ovalicin, a number of attempts have been made to find new
and reversible inhibitors of this enzyme through either the
structural modification of fumagillin8-12 or high-throughput
screening.13-15 Herein, we report a new fumagillin analogue
Fumarranol is an unexpected product of a reaction intended
to produce another fumagillin analogue when fumaginone (6)
was reacted with potassium hydroxide in an attempt to open
the spiroepoxide group to produce a diol. A single product
distinct from the starting material (6), as judged by thin-layer
chromatography, was isolated. The molecular mass as deter-
mined by high-resolution mass spectrometry was 280 Da, 18
Da smaller than that for the expected diol, suggesting that the
hydroxide group might have failed to be incorporated into the
1
final product. Careful analysis of H and 13C NMR suggested
that the product is more consistent with the bicyclic structure 5
(Scheme 1). To verify the assigned structure of 5, a number of
derivatives were made. One derivative, 7, formed crystals. The
crystal structure of 7 (Figure 2) confirmed the existence of the
bicyclic ring system in 5, which can be explained by the
formation of a carbanion at the R-position of the 6-ketone group
in the starting material (6), which undergoes an intramolecular
SN2 reaction to open the spiroepoxide group (Scheme 1). As
is apparent from the crystal structural of 7, the presence of the
cyclopropyl ring and the concurrent loss of the spiroepoxide
group in 5 rendered it structurally distinct from fumagillin and
TNP-470. As a result, it was not expected that fumarranol would
† Department of Pharmacology and Molecular Sciences.
‡ Medical Scientist Training Program.
§ Solomon H. Snyder Department of Neuoscience.
# Department of Oncology.
10.1021/jm060559v CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/24/2006

5646 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Letters
Table 1. Inhibition of BAEC Proliferation and Human MetAP2
Scheme 1. Formation of Fumarranol and Its Conversion to
Analogue 7
Enzymatic Activity by Fumarranol and Analoguesa
retain the ability to bind human MetAP2 (hMetAP2) or to inhibit
endothelial cell proliferation.
Surprisingly, however, fumarranol remained active against
hMetAP2 in vitro, albeit with lower potency than fumagillin.
Its IC50 value is 3.2 µM (Table 1). Importantly, it is also active
at the cellular level, inhibiting the proliferation of bovine aortic
endothelial cells (BAEC) with an IC50 of 34 nM (Table 1). The
apparently higher cellular activity of fumarranol is similar to
those seen with TNP-470 and fumagalone (4) and can be
attributed to both the likely lower cellular concentration of
MetAP2 in comparison to that required for in vitro enzymatic
assays and the accumulation of fumarranol inside cells. More-
over, fumarranol remained specific for hMetAP2, as it did not
inhibit the enzymatic activity of recombinant hMetAP1 at 100
µM, suggesting that the inhibition of BAEC was likely to be
caused by the inhibition of hMetAP2.
A structure/activity (SAR) study of fumarranol analogues was
then pursued to determine the key structural elements in
fumarranol that are essential for its activity. Different substitu-
tions on the bicyclic ring system were designed and synthesized,
and the resultant analogues were tested against hMetAP2 and
in the BAEC proliferation assay (Table 1). The primary alcohol
group at C1′ is essential, as its conversion into an aldehyde (10)
caused a 10-fold increase in IC50 value against recombinant
hMetAP2. Not surprisingly, the replacement of the alcohol group
with the bulky phthalyl group (7) or methoxy group (12) led to
complete inactivation. In contrast to the C1′ position, which
requires an alcohol group, reduction of the C2 carbonyl group
to an alcohol (11) caused a complete inactivation. Similar
decreases in activity upon reduction of the carbonyl group to
an alcohol in the fumagillin and ovalicin core structures have
also been observed.4 Like fumagillin, the C6 side chain in
fumarranol is also sensitive to chemical modifications. Thus,
removal of the side-chain epoxide (8, 9) caused a significant
a Inactive, no activity at 100 µM.
decrease in activity. Further extension of the side chain ending
with a carboxylic ester (analogue 14) resulted in complete
inactivation, which is likely due to the inability of the enzyme
active site to accommodate the excessively long side chain.
Together, these results indicate that fumarranol remains an
optimal structure among the analogues tested with this unique
bicyclic core for the inhibition of hMetAP2 or endothelial cell
proliferation.
We next determined whether fumarranol is capable of
inhibiting angiogenesis in vivo in a mouse matrigel plug assay.
Figure 2. X-ray crystal structure of fumarranol analogue 7.

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5647
Given the significant structural difference between fumarranol
with a bicyclic core and fumagillin containing a cyclohexyl core,
it is surprising that fumarranol retains binding affinity as well
as isoform specificity of fumagillin. From the SAR data (Table
1), it is clear that several positions on the bicyclic core of
fumarranol are quite sensitive to further chemical modifications,
including the hydroxyl group at C1′ and the carbonyl group at
C6. It can be envisaged that a key structural element involved
in the binding of fumarranol to hMetAP2 is the epoxide-
containing side chain that interacts with the same hydrophobic
pocket in the enzyme as its counterpart in fumagillin.17 However,
the bicyclic core of fumarranol may have to reorient itself
relative to the cyclohexyl core of fumagillin to fit the remainder
of the active site of the enzyme. This may allow the primary
alcohol group to form a new interaction with either the active
site metal ion or a hydrogen-bond acceptor in the active site.
The reorientation will also necessarily lead to the repositioning
of the C6 carbonyl group of fumarranol in the active site of
MetAP2, making any modification at this position, including
reduction to an alcohol, intolerable. It will be interesting to see
whether the fumarranol core can be further modified to enhance
its affinity for MetAP2 and its potency as an angiogenesis
inhibitor.
Figure 3. Effects of fumarranol on angiogenesis in mouse matrigel
plug assays in vivo: (A) control; (B) TNP-470 at 30 mg/kg/day; (C)
fumarranol at 90 mg/kg/day.
Acknowledgment. This work was supported by NCI, the
Keck Foundation. C.R.C. and X.H. were supported by the
Congressionally Directed Breast Cancer Research Program
Predoctoral Fellowship, and C.R.C. was also supported by the
National Institutes of Health Medical Scientist Training Program.
In this assay, matrigel containing VEGF and basic FGF were
injected subcutaneously into mice and allowed to incubate for
10 days, during which new blood vessels grew into the matrigel
in response to stimulation by VEGF and bFGF. As shown in
Figure 3A, matrigel from control animals contained new blood
vessels as judged both by the red color of the matrigel and by
staining the matrigel slices in MAS-trichrome to visualize blood
vessels (Figure 3A). As expected, treatment of the animals with
TNP-470 at 30 mg/kg almost completely inhibited the growth
of new blood vessels into the matrigel implant (Figure 3B).
Importantly, administration of fumarranol at 90 mg/kg also
blocked the growth of new blood vessels (Figure 3C). The higher
amount of fumarranol that is required to achieve a comparable
degree of inhibition of angiogenesis in vivo is consistent with
the lower potency of fumarranol in comparison to TNP-470.
TNP-470, despite its reduced toxicity compared to fumagil-
lin,3 still has side effects. In animal experiments, this is exhibited
in part by the ulceration in animals given TNP-470 (Figure S1,
Supporting Information). Unlike TNP-470, however, administra-
tion of fumarranol at 90 mg/kg did not result in the same
ulceration of skins as seen with TNP-470. These observations
suggest that fumarranol is likely to be less toxic than TNP-
470. It remains to be determined whether fumarranol also
possesses a better pharmacokinetic profile than TNP-470, i.e.,
with a longer half-life in humans, because the most reactive
spiroepoxide group in TNP-470 has been eliminated in fumar-
ranol.
Fumagillin irreversibly inhibits MetAP2 through covalent
modification of His231 in the active site of the enzyme using
its spiroepoxide group.16,17 In fumarranol, the spiroepoxide group
is replaced by the cyclopropyl group and a primary alcohol. It
is therefore unlikely for fumarranol to form a covalent bond
with the His231 side chain in the same manner as fumagillin
and TNP-470. In addition, the chemically reactive chloroacetyl
side chain at C6 in TNP-470 (Figure 1) is absent from
fumarranol. These structural differences may account for the
lower toxicity of fumarranol in mice compared with that of TNP-
470.
Supporting Information Available: Figure S1, synthetic
procedures and the procedure for the determination of X-ray crystal
structure of analogue 7. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Folkman, J. Tumor angiogenesis. AdV. Cancer Res. 1985, 43, 175-
203.
(2) Risau, W. Mechanisms of angiogenesis. Nature 1997, 386, 671-
674.
(3) Ingber, D.; Fujita, T.; Kishimoto, S.; Sudo, K.; Kanamaru, T.; Brem,
H.; Folkman, J. Synthetic analogues of fumagillin that inhibit
angiogenesis and suppress tumour growth. Nature 1990, 348, 555-
557.
(4) Griffith, E. C.; Su, Z.; Turk, B. E.; Chen, S.; Chang, Y.-W.; Wu, Z.;
Biemann, K.; Liu, J. O. Methionine aminopeptidase (type 2) is the
common target for angiogenesis inhibitors AGM-1470 and ovalicin.
Chem. Biol. 1997, 4, 461-471.
(5) Sin, N.; Meng, L.; Wang, M. Q. W.; Wen, J. J.; Bornmann, W. G.;
Crews, C. M. The anti-angiogenic agent fumagillin covalently binds
and inhibits the methionine aminopeptidase, MetAP-2. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 6099-6103.
(6) Zhang, Y.; Griffith, E. C.; Sage, J.; Jacks, T.; Liu, J. O. Cell cycle
inhibition by the anti-angiogenic agent TNP-470 is mediated by p53
and p21WAF1/CIP1. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 6427-
6432.
(7) Yeh, J. R.; Mohan, R.; Crews, C. M. The antiangiogenic agent TNP-
470 requires p53 and p21CIP/WAF for endothelial cell growth arrest.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 12782-12787.
(8) 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. Design and synthesis of highly potent fumagillin
analogues from homology modeling for a human MetAP-2. Bioorg.
Med. Chem. Lett. 2000, 10, 39-43.
(9) Zhou, G.; Tsai, C. W.; Liu, J. O. Fumagalone, a reversible inhibitor
of type 2 methionine aminopeptidase and angiogenesis. J. Med. Chem.
2003, 46, 3452-3454.
(10) Fardis, M.; Pyun, H. J.; Tario, J.; Jin, H.; Kim, C. U.; Ruckman, J.;
Lin, Y.; Green, L.; Hicke, B. Design, synthesis and evaluation of a
series of novel fumagillin analogues. Bioorg. Med. Chem. 2003, 11,
5051-5058.
(11) Bernier, S. G.; Lazarus, D. D.; Clark, E.; Doyle, B.; Labenski, M.
T.; Thompson, C. D.; Westlin, W. F.; Hannig, G. A methionine

5648 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Letters
aminopeptidase-2 inhibitor, PPI-2458, for the treatment of rheumatoid
arthritis. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 10768-10773.
(12) Mazitschek, R.; Huwe, A.; Giannis, A. Synthesis and biological
evaluation of novel fumagillin and ovalicin analogues. Org. Biomol.
Chem. 2005, 3, 2150-2154.
(13) Wang, J.; Sheppard, G. S.; Lou, P.; Kawai, M.; Park, C.; Egan, D.
A.; Schneider, A.; Bouska, J.; Lesniewski, R.; Henkin, J. Physi-
ologically relevant metal cofactor for methionine aminopeptidase-2
is manganese. Biochemistry 2003, 42, 5035-5042.
(14) Garrabrant, T.; Tuman, R. W.; Ludovici, D.; Tominovich, R.;
Simoneaux, R. L.; Galemmo, R. A., Jr.; Johnson, D. L. Small
molecule inhibitors of methionine aminopeptidase type 2 (MetAP-
2). Angiogenesis 2004, 7, 91-96.
(15) Kallander, L. S.; Lu, Q.; Chen, W.; Tomaszek, T.; Yang, G.; Tew,
D.; Meek, T. D.; Hofmann, G. A.; Schulz-Pritchard, C. K.; Smith,
W. W.; Janson, C. A.; Ryan, M. D.; Zhang, G. F.; Johanson, K. O.;
Kirkpatrick, R. B.; Ho, T. F.; Fisher, P. W.; Mattern, M. R.; Johnson,
R. K.; Hansbury, M. J.; Winkler, J. D.; Ward, K. W.; Veber, D. F.;
Thompson, S. K. 4-Aryl-1,2,3-triazole: a novel template for a
reversible methionine aminopeptidase 2 inhibitor, optimized to inhibit
angiogenesis in vivo. J. Med. Chem. 2005, 48, 5644-5647.
(16) Griffith, E. C.; Su, Z.; Niwayama, S.; Ramsay, C. A.; Chang, Y. H.;
Liu, J. O. Molecular recognition of angiogenesis inhibitors fumagillin
and ovalicin by methionine aminopeptidase 2. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 15183-15188.
(17) Liu, S.; Widom, J.; Kemp, C. W.; Crews, C. M.; Clardy, J. Structure
of human methionine aminopeptidase-2 complexed with fumagillin.
Science 1998, 282, 1324-1327.
JM060559V