Novel reversible methionine aminopeptidase-2 (MetAP-2) inhibitors based on purine and related bicyclic templates

Article abstract of DOI:10.1016/j.bmcl.2016.12.019

The natural product fumagillin 1 and derivatives like TNP-470 2 or beloranib 3 bind to methionine aminopeptidase 2 (MetAP-2) irreversibly. This enzyme is critical for protein maturation and plays a key role in angiogenesis. In this paper we describe the synthesis, MetAP-2 binding affinity and structural analysis of reversible MetAP-2 inhibitors. Optimization of enzymatic activity of screening hit 10 (IC₅₀: 1?μM) led to the most potent compound 27 (IC₅₀: 0.038?μM), with a concomitant improvement in LLE from 2.1 to 4.2. Structural analysis of these MetAP-2 inhibitors revealed an unprecedented conformation of the His339 side-chain imidazole ring being co-planar sandwiched between the imidazole of His331 and the aryl-ether moiety, which is bound to the purine scaffold. Systematic alteration and reduction of H-bonding capability of this metal binding moiety induced an unexpected 180° flip for the triazolo[1,5-a]pyrimdine bicyclic template.

Full text of DOI:10.1016/j.bmcl.2016.12.019

Accepted Manuscript
Novel reversible methionine aminopeptidase-2 (MetAP-2) inhibitors based on
purine and related bicyclic templates
Timo Heinrich, Hans-Peter Buchstaller, Bertram Cezanne, Felix Rohdich, Jörg
Bomke, Manja Friese-Hamim, Mireille Krier, Thorsten Knöchel, Djordje Musil,
Birgitta Leuthner, Frank Zenke
PII:
S0960-894X(16)31286-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.12.019
BMCL 24508
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 September 2016
6 December 2016
7 December 2016
Please cite this article as: Heinrich, T., Buchstaller, H-P., Cezanne, B., Rohdich, F., Bomke, J., Friese-Hamim, M.,
Krier, M., Knöchel, T., Musil, D., Leuthner, B., Zenke, F., Novel reversible methionine aminopeptidase-2
(MetAP-2) inhibitors based on purine and related bicyclic templates, Bioorganic & Medicinal Chemistry Letters
(2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.12.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Novel reversible methionine aminopeptidase-2 (MetAP-2)
inhibitors based on purine and related bicyclic templates
Timo Heinricha,*, Hans-Peter Buchstallera, Bertram Cezanneb, Felix Rohdichc, Jörg
Bomkea, Manja Friese-Hamimd, Mireille Kriera, Thorsten Knöchele, Djordje Musila,
Birgitta Leuthnera, Frank Zenkef
aDiscovery Technologies, bProcess Technologies, cDrug Disposition, din vivo
Pharmacology, eProject Management, fCellular Pharmacology
Merck KGaA, Frankfurter Str. 250, D-64293 Darmstadt, Germany.
*Corresponding author Tel.: +49-6151-726589, fax: +49-6151-723129, E-mail:
timo.heinrich@merckgroup.com

In 1990 Judah Folkman´s laboratory published a seminal discovery on the anti-
angiogenic activity of the natural compound fumagillin 1, which was isolated in 1951
from the fungus Aspergillus fumigatus.1 Subsequently, intensive activities were started
to improve the pharmacological and pharmaceutical properties of this compound leading
to the development of a highly potent derivative TNP-470 2 which entered clinical
development in 1992 as one of the first anti-angiogenic drugs in oncology.2,3 Although
the clinical development stopped in 2001 this approach has received quite some
attention ever since.4
1
2
3
Figure 1. Irreversible MetAP-2 inhibitors: 1 (natural product fumagillin), 2 (TNP-470,
former clinical development compound), 3 (beloranib, until recently under clinical
investigation as anti-obesity drug).
The molecular target of fumagillin and derivatives was identified in 1997 as methionine
aminopeptidase 2 (MetAP-2) by the laboratory of Craig Crews.5 Methionine
aminopeptidases are enzymes which remove the N-terminal amino acid methionine from
proteins during or after the translation process. Methionine is cleaved from some but not

all cellular proteins, especially proteins that bear a small amino acid at the second
position. As part of protein maturation, cleavage is assumed to be relevant for protein
stability (N-end rule) and proper function (protein folding, N-terminal myristoylation).
Although, the functional relevance of N-terminal processing may be different for each
affected protein and remains to be fully clarified. Three MetAPs exist in humans, MetAP-
1, MetAP-2, and MAP1D. MetAP-2 is widely expressed in all human tissues, although
differences exist in expression levels based on mRNA expression data. Complete
knockout of MetAP-2 in mice is embryonic lethal with a pronounced block at the
gastrulation stage. A tissue-restricted knockout in the endothelial cell compartment was
embryonic lethal as well and showed a pronounced abnormality in vascular
development,6 indicating that MetAP-2 is important during development and required for
the formation of the vascular system. Interestingly, the MetAP-2 knockout in the yeast S.
cerevisiae is viable, but is lethal in combination with a MetAP-1 knockout, suggesting
some redundancy between MetAP-1 and MetAP-2.7 Pharmacological inhibition of
MetAP-2 using different classes of inhibitors blocked proliferation of vascular endothelial
cells, explaining, at least in part, the anti-angiogenic mechanism of action.8-10
Beside the mentioned irreversible MetAP-2 inhibitors diverse compounds were
described in the literature, which inhibit MetAP-2 reversibly (Figure 2).

4 (A-832234)
5 (LAF389)
6 (JNJ-4929821)
9 (Indazole)
7 (A-357300)
8 (Triazole)
Figure 2. Reversible MetAP-2 inhibitors from different structural classes: 4 (A-832234,
Anthranilic acid-sulfonamide),11 5 (LAF389, Bengamide),12 6 (JNJ-4929821, Triazole-
sulfonamide),13 7 (A-357300, Bestatin),17 8 (Triazole),14 9 (indazole).15
The activity of more derivatives from each series together with their respective binding
modes has been discussed in recent reviews.16,17 Subsequently, anti-proliferative activity
on some tumor cell lines has been demonstrated for MetAP-2 inhibitors indicating
potential for direct anti-tumor effects as well.8,18 The anti-proliferative effect of MetAP-2
depletion using antisense or siRNA could be confirmed by some laboratories6,9,19 but not
by others.20 The molecular mechanism for anti-proliferative activity in endothelial and
selected tumor cells is not well understood. Several intracellular events have been noted
upon treatment with MetAP-2 inhibitors, such as induction of the tumor suppressor p53
and cell cycle inhibitor p21, modulation of phosphorylation of the translation initiation
factor eIF-2, reduced phosphorylation of the mitogen-activated protein kinase ERK, or
inhibition of non-canonical WNT signaling.21-25 However, a consistent causality between

target inhibition and the anti-proliferative activity could not be demonstrated so far. A
wealth of published data demonstrated that MetAP-2 inhibitors have potent anti-tumor
efficacy in vivo, predominantly shown in mouse models. Whether the anti-tumor efficacy
in these models was indeed mediated by anti-angiogenic and/or anti-proliferative effects
is not fully understood.8,15,17,26,27 The mechanism of action, i.e. the inhibition of N-
terminal protein processing, is a unique cellular response not seen with other targeted
inhibitors and that’s why we considered MetAP-2 as an interesting and promising tool to
interfere in tumor biology and stop tumor cell proliferation.
Accordingly, the corporate compound collection was screened against the target and the
2-Cl-phenyl-ether 10 was identified as a low micromolar hit (table 1). Subsequent efforts
were focused on improving potency via substitution of the phenyl ring. The synthesis of
the 2-(aryloxymethyl) pyrrolidine analogues is outlined in Scheme 1 and began with the
conversion of either R- or S-proline into the respective methyl esters.
a - d
e - g
Scheme 1. Synthesis of pyrrolidine derivatives. Reagents and conditions (a) SOCl2,
MeOH, DMF; 0 – RT; 2h; (b) BOC2O, TEA, tBuOH; 25°C, 21h; (c) LiCl, NaBH4, EtOH,
THF; -20°C – RT, 19 h; (d) MeSO2Cl, TEA, DCM; (e) R1-OH, Cs2CO3, DMF; 60°C, 16h;
(f) TFA, DCM; RT, 1h; (g) R2-Cl, for conditions see tables 1 and 2.
Protection of the nitrogen with a tert-butoxycarbonyl (Boc) group followed by reduction of
the ester yielded the N-Boc-protected amino alcohols. The alcohol was converted into

the corresponding mesylate, which was then substituted by the required phenol
derivative (R1) in the presence of cesium carbonate. After removal of the Boc group the
resulting secondary amines were coupled with different commercially available bicyclic
aryl-chlorides (R2) under alternative conditions as referenced in the tables 1 and 2.
Especially for the bicyclic aryl chlorides without NH-function (22 – 25) an alternative
approach worked well: Commercially available prolinol is directly coupled to the
respective bicyclic aryl chloride (conditions indicated in table 2) followed by mesylate
formation of the alcohol and substitution of the latter with appropriate phenolic building
blocks (R1).
Table 1
In vitro evaluation of synthetic compounds 10 – 1828
6
7
9
Compound
IC50* [µM]
5
4
1
2
8
3
10 a)
11 b)
12 c)
13 c)
14 d)
15 c)
2-Cl-Phe (R)
2-Cl-Phe (S)
2,3-di-F-Phe
3-F-Phe
1 ± 0.1
> 50
1.6 ± 0.1
2 ± 0.1
4-F-Phe
3.8 ± 2.5
1.3 ± 0.3
4-CN-Phe

16 b)
17 a)
18 a)
2-Me-Phe
1-naphthyl
0.7 ± 0.2
0.45 ± 0.1
5-isoquinolinyl 0.23 ± 0.1
* Activity in recombinant MetAP-2 enzyme assay. Mean of three determinations.
Coupling conditions of the pyrrolidine to the bicyclic aryl-chloride (R2-Cl): a) DIPEA,
1-butanol, 6.5 h, 120 °C, MW; b) TEA, 1-butanol, 1 h, 100 °C; c) acetone, 20 min,
150 °C, MW; d) diglyme, 20 min, 150 °C, MW;
The S-enantiomer 11 of the screening hit is inactive and hence further optimization work
focused on R-isomers. The biochemical activity of the halogen bearing hit-analogs 12,
13 or 14 is in the same range as initial hit 10, which suggests that changes in substituent
size and different inductive electron withdrawing residues on the aryl moiety have no
impact on ligand binding. Increasing electron density in the phenyl ring with a small
aliphatic moiety like in 16 provided an improvement in activity, while further extension of
the residue to the naphthalene system (17) resulted in even greater enzyme potency.
However, when considering lipophilic ligand efficiency (LLE),29 it is obvious that the
larger molecule 17 only shows minimal improvement (LLEs: 16 = 2.11; 17 = 2.06).
Nevertheless, the naphthalene moiety was used as the basis for further optimization of
enzyme potency, initially via increasing molecular polarity by introduction of a nitrogen
into the bicyclic ring. An isoquinoline derivative (18) validated these efforts with an
improved enzyme activity, which is also reflected in an improvement of the LLE (2.71).

Explorations of the ether-linker between the pyrrolidine bridge and the extended aryl
moiety itself did not improve activity. Alkane, alkene, amine, amide and ester were
tested, but turned out to be less potent.30
In a parallel approach the initial hit 10 was used as template to investigate the structural
activity relationship of the purine core. Even subtle substitutions like introduction of a
methyl group in positions 2 or 8 led to completely inactive compounds. Same is true for
the introduction of fluorine or amine in position 2.30 Structural analysis (vide infra) had
shown, that the purine nitrogens N3 and N9 are in contact to the metal ions in the active
site of the enzyme. Accordingly, it was anticipated that manipulations of the -electron
distribution of the bicyclic template should influence metal ion interaction and hence
activity (table 2).
Table 2.
In vitro evaluation of synthetic compounds 19 – 27 28
Compound
Structure
IC50* [µM]
19 a)
> 50
20 b)
2.2 ± 0.7

21 c)
8.6 ± 2.1
22 d)
23 e)
24 e)
25 f)
26 f)
27 g)
2.8 ± 0.6
> 50
1.8 ± 0.2
0.56 ± 0.1
0.11 ± 0.01
0.038 ± 0.001
* Activity in recombinant MetAP-2 enzyme assay. Mean of three determinations.
Coupling conditions of the pyrrolidine to the bicyclic aryl-chloride (R2-Cl): a) Acetone,
20 min, 150 °C, MW; b) TEA, 1-butanol, 40 min, 150 °C, MW; c) TEA, 1-butanol, 1.5
h, 100 °C; d) neat 100 °C, 12 h;31 e) Cs2CO3, ACN, 6 h, RT; e) generation from

alcohol with benzyltributylammoniumbromide, isopropyl acetate, ACN and POCl3, 1.5
h, 65 °C; f) TEA, 1-butanol, 100 min, 120 °C, MW; g) Cs2CO3, DMF, 80 °C, 12 h.
For the 7H-pyrrolo[2,3-d]pyrimidine derivative 19, the N9 atom of the purine system is
replaced by a carbon atom, fixing thereby the proton to the remaining nitrogen in the five
membered ring. As this is incompatible with coordination to the active site cation, the
compound is inactive as expected. Compound 20 is a 3H-imidazo[4,5-b]pyridine where
N1 of the original purine was replaced by a carbon atom. This modification has no
significant effect on binding affinity as both compounds 10 and 20 are equally active.
Moving purine nitrogen N7 into position 8 to get the 1H-pyrazolo[3,4-d]pyrimidine core in
21 reduces affinity to 8.6 µM. Probably only the 2H-tautomer can efficiently bind to the
active site, and this tautomer state might be energetically unfavorable per se. Compound
22 represents a [1,2,5]thiadiazolo[3,4-b]pyridine bearing a sulfur in the five membered
ring between the nitrogens. This compound is equipotent as its carbon analog 20.
Widening the angle of the five membered ring and increasing -electron density by the
sulfur atom has no positive impact on MetAP-2 binding. In addition, this is the first
scaffold replacement where not other tautomeric form is expected as no hydrogen atom
to a heteroatom needs to be allocated appropriately. Also, the pyrido[2,3-d]pyrimidine
derivative 23 bears no heteroatom-hydrogen and only lone pair donors. However, for
this molecule, the second six membered ring is not tolerated and the compound is
inactive. The introduction of a nitrogen atom in the bridge head position 5 of the bicyclic
system with a 6 and 5 membered ring allows the removal of the nitrogen bound
hydrogen without introducing an additional heteroatom. At this point, introduction of the

active naphthalene moiety (from compound 17) was evaluated in subsequent purine
core modifications to further drive enzymatic potency. Interestingly, it was observed that
the [1,2,4]triazolo[1,5-a][1,3,5]triazine analog 24 is 3 fold less active than compound 17.
It can be speculated that the high nitrogen loading of the aryl template induced a
reduced electron density on nitrogens 3 and 9 that could be responsible for the activity.
This consideration was confirmed when replacing the two nitrogens in positions 1 and 7
providing the imidazo[1,2-a]pyrimidine 25, a moleculde with sub-µM enzymatic activity.
Subsequent systematic exploration of the bicyclic core moiety and introduction of a
nitrogen in the five membered ring gave the [1,2,4]triazolo[1,5-a]pyrimidine 26 with
improved activity. Finally, increasing polarity in the naphthalene moiety by substitution
with the isoquinoline known from 18 resulted in compound 27 which possessed two digit
nM MetAP-2 activity.32 This improvement is confirmed with the LLE of 4.2 for 27.
Extended profiling of the most potent derivatives revealed that in vitro properties are in
an acceptable range. The thermodynamic solubility of the optimized hit 27 is 171 µg/ml
and its permeability in the Caco-2 assay is 21E-6 cm/s with an efflux ratio of 2.9.
The crystallization of derivatives from this series with the MetAP-2 protein helped to
rationalize the impact of the extended aryl moiety on affinity. As shown in figure 3 (PDB:
5LYW), the 2-methyl-phenyl ether of 16 is in -stacking interaction with His339 and
Tyr444. Obviously, the contribution of this -electron interaction to affinity is more
favorable for electron rich aryl-ethers of the ligand like in 16 and 17 than for electron
poor aromatic systems (13, 15).

Figure 3.33 XRay structure analysis of 16 bound to human MetAP-2. The distances
between the amino acid side chains and the aryl residue of the ether are given in
Ångstrom. The Mn2+ ions are depicted in purple. The distance from N3 to the closest ion
is 2.35 Å and from N9 2.39 Å. The distance between the metal cores is 3.24 Å.
Interestingly, the imidazole of His339 is sandwiched in a co-planar manner between the
aryl moiety of ligand 16 and imidazole of His331 (figure 4 A). Capturing the His339 side
chain in this conformation is unprecedented.34 For the published MetAP-2 structures e.g.
the apo form (PDB: 1BN5), a T-shaped orientation of the His339-imidazole edge to the
face of His331-imidazole is found (figure 4 B).

A
B
Figure 4.33 A. Depiction of imidazole from His339 sandwiched between the imidazole of
His331 and the 2-tolyl residue of 16. B. Apo-structure of human MetAP-2 shows T-
shaped orientation between the two imidazoles of His339 and His331.
For other ligands with structural data in the PDB, the imidazole of His339 is pushed out
of the active site as shown in figure 5 for the irreversible MetAP-2 inhibitor 2 (TNP-470,
figure 5 A; PDB: 1B6A) or the anthranilic acid derivative 4 (A-849519, figure 5 B; PDB:
1YW9).11

A
B
Figure 5.33 Xray structure analyses of human MetAP-2 with A. TNP-470 (shortest
distance between imidazoles of His339 and His331 is 4.52 Ångstrom – not shown here;
the covalent bond between the imidazole of His231 and the ligand is indicated with the
black arrow) B. A849519 (shortest distance between imidazoles is 4.55 Ångstrom – not
depicted explicitly)
Beside the analysis of interactions at the entrance to the active site, we investigated the
impact of core modifications on activity and therefore analyzed the structure of 16 bound
to MetAP-2. The result is shown in figure 6. From figure 6A one could assume that there
is some space in the back of the binding pocket. Small substituents on the purine core in
position 2 or 8 (indicated by the yellow arrow in figure 6) like fluorine or methyl give only
inactive compounds suggesting that the embodiment of the pocket does not leave room
for any substitution or does disrupt a water cluster in an unfavorable manner.

A
B
Figure 6.33 16 bound to MetAP-2 with protein surface is shown. The yellow arrows
indicate where additional space around the purine template (A) or the pyrrolidine (B) is
anticipated
Also, the methionine binding pocket of the enzyme where the pyrrolidine of the inhibitor
is positioned suggests additional space (figure 6B). Neither mono- nor di-fluoro
substitution or hydroxy- or keto-functionality in position 4 of the saturated ring improved
activity.30 The pyrrolidine moiety itself was examined in more detail i.e. ring size was
varied from azetidines via piperidines up to azepines. In all cases, no improvement of
activity could be found.30 This was considered remarkable, especially after analysis of
the binding mode of this compound class. Fluorine substitution on the pyrrolidine is filling
the methionine binding pocket and should increase the contact surface with the protein
resulting in higher affinity. Steric clashes were not considered to occur. The proximate
binding site above the metal ions is quite rigid as well as the more distal methionine side
chain binding pocket. But the latter can be addressed via an alternate vector as the

analysis of the crystal structure of 27 proved (figure 7; PDB code for structure of 27:
5LYX).
Figure 7.33 Superposition of the purine based derivative 16 (magenta) and the
[1,2,4]Triazolo[1,5-a]pyrimidine heterocycle of 27 (green). The bicyclic scaffold is flipped
by 180° for the latter compared to the orientation of the purine core.
The depiction in figure 7 illustrates the 180° turn of the metal binding template when
modifying the NH bearing purine (magenta in figure 7) into the triazolo[1,5-a]pyrimidine
without any heteroatom bound proton (green in figure 7). This was remarkable since the
initial assumption has been that a six-membered ring in the back of the pocket was not
favorable since compound 23 was inactive. From the triazolo[1,5-a]pyrimidine scaffold,
the pyrrolidine moiety enters the methionine pocket from a slightly more buried position
compared to the situation when the pyrrolidine is bound to the purine core. Another

notable change in orientation can be observed in figure 7: the imidazole side chain of
His339 is pushed out of - interactions discussed in figure 3. The closest contact
between the imidazole of His339 and the isoquinoline of 27 is 4.56 Å (not indicated in
the figure) and the methylene-carbon of the amino acid side chain is close to C3 of the
isoquinoline (3.72 Å). The linking C5 of the isoquinoline is 3.47 Å away from the closest
carbon of Tyr444, whereas the toluene of 16 is in 3.21 Å distance to the phenol of
Tyr444 (figure 3). Accordingly, the hydrophobic interactions of the ether moiety
presented on the triazolo[1,5-a]pyrimidine (27) contribute to lesser extend to the binding
affinity than the purine bound ether residue. This binding mode of the triazolo[1,5-
a]pyrimidine scaffold had been observed also by others using our MetAP-2 inhibitors in
the context of their investigations.15
In this study the original purine based MetAP-2 screening hit 10 was used as the basis
for multiple optimization approaches. Only the R-configuration at the pyrrolidine is
tolerated in the active site of MetAP-2. Pyrrolidine decorations did not improve activity.
Hence the saturated five-membered ring fills the lipophilic methionine binding-pocket
optimally. Moreover, variations of the -CH2-O- bridge had no beneficial effect on
potency. Optimization of the aryl moiety attached to the ether contributed to an
increased activity. Extension of the monocyclic ring to the bicyclic moiety proved to be
most favorable to capture His339 side chain in an unprecedented fashion and improve
binding by sandwiching this imidazole of His339 between the imidazole of His331 and
the aryl-ether of the ligand. The purine core is suitable when the 7H tautomer is
accessible. Small substituents on this template are not tolerated. An optimized nitrogen-
positioning within the indene-scaffold contributes to an improved activity. The best
activity was obtained with the [1,2,4]triazolo[1,5-a]pyrimidine template which was

identified via systematic investigations. Remarkably, the orientation of the bicyclic metal
binding template flipped by 180° when the original purine core was optimized into a
scaffold without any nitrogen-bound proton. Hydrophobic interactions of the isoquinoline
ether of 27 with Tyr444 and His339 are less pronounced compared to the VdW contacts
of the tolyl ether of 16. The outlined optimization enabled the generation of biochemical
active compounds with acceptable to favorable in vitro properties and paved the way for
further optimizations that will be discussed in due course.

References and notes
1.
Ingber, D.; Fujita, T.; Kishimoto, S.; Sudo, K.; Kanamaru, T.; Brem, H.; Folkman, J.
Nature 1990, 348, 555.
2.
3.
Kruger, E. A.; Figg, W. D. Expert. Opin. Investig. Drugs 2000, 9, 1383.
Bhargava, P.; Marshall, J. L.; Rizvi, N.; Dahut, W.; Yoe, J.; Figuera, M.; Phipps, K.; Ong,
V. S.; Kato, A.; Hawkins, M. J. Clin. Cancer Res. 1999, 5, 1989.
4.
5.
6.
Selvakumar , P.; Lakshmikuttyamma, A.; Dimmock, J. R.; Sharma, R. K. Biochim.
Biophys. Acta 2006, 1765, 148.
Sin, N.; Meng, L.; Wang, M. Q. W.; Wen, J. J.; Bornmann, W. G.; Crews, C. M. Proc.
Natl. Acad. Sci. USA 1997, 94, 6099.
Yeh, J.-R. J.; Ju, R.; Brdlik, C. M.; Zhang, W.; Zahng, Y.; Matyskiale, M. E.; Shotwell, J.
D.; Crews, C. M. Proc. Natl. Acad. Sci. USA 2006, 103, 10379.
7.
8.
Li, X.; Chang, Y.-H. Proc. Natl. Acad. Sci. USA 1995, 92, 12357.
Wang. J.; Tucker, L. A.; Stavropoulos, J.; Zhang, Q.; Wang, Y.-C.; Bukofzer, G.;
Niquette, A.; Meulbroek, J. A.; Barnes, D. M.; Shen, J.; Bouska, J.; Donawho, C.;
Sheppard, G. S.; Bell, R. L. Proc. Natl. Acad. Sci. USA 2008, 105, 1838.
Bernier, S. G.; Taghizadeh, N.; Thompson, C. D.; Westlin, W. F.; Hannig, G. J. Cell
Biochem. 2005, 95, 1191.
9.
10.
11.
Ehlers, T.; Furness, S.; Robinson, T. P.; Zhong, H. A.; Goldsmith, D.; Aribser, J.; Bowen,
J. P. Curr. Top. Med. Chem. 2016, 16, 1478.
Sheppard, G. S.; Wang, J.; Kawai, M.; Fidanze, S. D.; BaMaung, N. Y.; Erickson, S. A.;
Barnes, D. M.; Tedrow J. S.; Kolaczkowski, L.; Vasudevan, A.; Park, D. C.; Wang, G. T.;
Sanders, W. J.; Mantei, R. A.; Palazzo, F.; Tucker-Garcia, L.; Lou, P.; Zhang, Q.; Park,
C. H.; Kim, K. H.; Petros, A.; Olejniczak, E.; Nettesheim, D.; Hajduk, P.; Henkin, J.;
Lesniewski, R.; Davidsen, S. K.; Bell, R. L. J. Med. Chem. 2006, 49, 3832.
Xu, W.; Lu, J.-P.; Ye, Q.-Z. J. Med. Chem. 2012, 55, 8021.
12.
13.
Garrabrand, T.; Tuman, R. W.; Ludovici, D.; Tominovich, R.; Simoneaux, R. L.;
Galemmo, R. A. Jr.; Johnson, D. L. Angiogenesis 2004, 7, 91.
14.
Marino, J. P. Jr.; Fisher, P. W.; Hofmann, G. A.; Kirkpatrick, R. B.; Janson, C. A.;
Johnson, R. K.; Ma, C.; Mattern, M.; Meek, T. D.; Ryan, M. D.; Schulz, C.; Smith, W. W.;
Tew, D. G.; Tomazek, T. A. Jr.; Veber, D. F.; Xiong, W. C.; Yamamoto, Y.; Yamashita,
K.; Yang, G.; Thompson, S. K. J. Med. Chem. 2007, 50, 3777.
15.
Cheruvallath, Z.; Tang, M.; McBride, C.; Komandla, M.; Miura, J.; Ton-Nu, T.; Erikson,
P.; Feng, J.; Farrell, P.; Lawson, J. D.; Vanderpool, D.; Wu, Y.; Dougan, D. R.;
Plonowski, A.; Holub, C.; Larson, C. Bioorg. Med. Chem. Lett. 2016, 26, 2774.
Yin, S.-Q.; Wang, J.-J.; Zhang, C.-M.; Liu, Z.-P. Curr. Med. Chem. 2012, 19, 1021.
Mauriz, J. L.; Martin-Renedo, J.; Garcia-Palomo, A.; Tunon, M. J.; Gonzalez-Gallego, J.
Curr. Drug Targets 2010, 11, 1439.
16.
17.
18.
Wang, J.; Sheppard, G. S.; Lou, P.; Kawai, M.; BaMaung, N.; Erickson, 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.
19.
20.
Datta, B.; Datta, R. Exp. Cell Res. 1999, 246, 376.
Kim, S.; LaMontagne, K.; Sabio, M.; Sharma, S.; Versace, R. W.; Yusuff, N.; Phillips, P.
E. Cancer Res. 2004, 64, 2984.

21.
Zhang, Y.; Griffith, E. C.; Sage, J.; Jacks, T.; Liu, J. O. Proc. Natl. Acad. Sci. USA 2000,
97, 6427.
22.
23.
Datta, B.; Majumdar, A.; Datta, R.; Balusu, R. Biochemistry 2004, 43, 14821.
Zhang, Y.; Yeh, J. R.; Mara, A.; Ju, R.; Hines, J. F.; Cirone, P.; Griesbach, H. L.;
Schneider, I.; Slusarski, D. C.; Holley, S. A.; Crews, C. M. Chem.Biol. 2006, 13, 1001.
Datta, B.; Ghosh, A.; Majumdar, A.; Datta, R. Biochemistry 2007, 46, 3465.
Sundberg, T. B.; Darricarrere, N.; Cirone, P.; Li, X.; McDonald, L.; Mei, X.; Westlake, C.
J.; Slusarski, D. C.; Beynon, R. J.; Crews, C. M. Chem.Biol. 2011; 18, 1300.
Chun, E.; Han, C. K.; Yoon, J. H.; Sim, T. B.; Kim, Y.-K.; Lee, K.-Y. Int. J. Cancer 2005,
114, 124.
24.
25.
26.
27.
28.
Tucker, L. A.; Zhang, Q.; Sheppard, G. S.; Lou, P.; Jiang, F.; McKeegan, E.; Lesniewski,
R.; Davidsen, S. K.; Bell, R. L.; Wang, J. Oncogene 2008, 27, 3967.
Biochemical activity testing of MetAP-2: MetAP-2 activity was determined by an enzyme
coupled assay using the tripeptide Met-Ala-Ser (MAS) as substrate and recombinant
human MetAP-2 (His-Tev-MetAP-2, prepared in-house). The released methionine is
converted by L-amino acid oxidase (AAO) to Met ox and hydrogen peroxide is released.
In a second step horse radish peroxidase catalyses the oxidation of the leuko dye
dianisidine to dianisidine ox with hydrogen peroxide as co-substrate. The produced
dianisidine ox was detected photometrically as increase in absorbance at 450 nm.
MetAP-2 activity was determined in a kinetic measurement mode. The release of one
molecule methionine corresponds to the production of one molecule dianisidine ox. The
MetAP-2 enzymatic activity is directly corresponding to the increase in absorbance per
time. In detail, the assay was performed in 384 well microtiter plate (Greiner 78110 MTP,
transparent) in a total reaction volume of 50 µl at 22 °C. 0.35 µg of N-terminal His tag
human rec MetAP-2 (prepared in house, aa 2-478, fc 123 nM), 1 u horse radish
Peroxidase (Roche, Mannheim), 0.02 u L-amino acid oxidase (Sigma-Aldrich, St.Louis),
0.6 mM Dianisidine (Sigma-Aldrich, St. Louis, dissolved in 50 mM HCl, 10% DMSO)
were incubated in the absence or presence of the test compound (10 dilution
concentrations) in 100 mM Hepes, 50 mM NaCl, 50 µM MnCl2 at pH 7.0 for 15 min at 22
°C (According to the discussion in the literature we considered manganese as the
relevant metal co-factor Wang, J.; Sheppard, G. S.; Lou, P.; Kawai, M.; Park, C.; Egan,
D. A.; Schneider, A.; Bouska, J.; Lesniewski, R.; Henkin, J. Biochem. 2003, 42, 5035).
The reaction was started by the addition of 500µM (fc) MAS peptide (Sigma-Aldrich, St.
Louis). After mixing the first measurement of absorbance was performed on an Envision
multimode reader (Perkin –Elmer LAS Germany GmbH) at wavelength of 450 nm. The
reaction was incubated at 22 °C for additional 45 min and the second absorbance
measurement was performed. The increase of absorbance per time was determined.
The control value used was the inhibitor- free reaction with 0.5 % DMSO (fc). As
pharmacological inhibitor control Fumagillin (Sigma-Aldrich, St.Louis) in a final
concentration of 5 µM was used. The inhibitory values (IC50) were determined using the
program Condosseo® from GeneData.
29.
Ryckmans, T.; Edwards, M. P.; Horne, V. A.; Correia, A. M.; Owen, D. R.; Thompson, L.
R.; Tran, I.; Tutt, M. F.; Young, T. Bioorg. Med. Chem. Lett. 2009, 19, 4406.
Because of missing enzymatic activity compounds are not discussed in more detail.
Sharma, K. S. Indian J. Chem., Section B, 1979, 17, 13.
Synthesis of 27: To 2 °C cold H2SO4 (270 ml) was first added racemic malic acid (80 g,
596 mmol) without temperature increase above 10 °C and in the same way 3-amino-1H-
1,2,4-triazole (50 g, 595 mmol). The mixture was stirred for 12 h at RT and finally heated
30.
31.
32.

to 100 °C for 1h, giving a clear yellow solution. The cold mixture was poured on ice, pH
was adjusted to 6 with conc. NaOH solution and the precipitate was filtered off, washed
with cold water and dried, giving 35 g colorless crystals of [1,2,4]triazolo[1,5-a]pyrimidin-
7-ol. At RT the alcohol (6.5 g, 48 mmol) was given slowly to POCl3 (30 ml) and then
heated to reflux for 1.5 h, giving a clear solution. All volatile parts were removed in
vacuum and the remainder was poured on ice-water cautiously. The well stirred mixture
was treated with conc. NaOH solution to reach pH = 8. The precipitate was filtered off
and the liquid fraction were re-extracted with ethylacetate. The combined solid fractions
were dried to 5 g of 7-chloro-[1,2,4]triazolo[1,5-a]pyrimidine. A solution of 1-butanol (120
ml) with 7-chloro-[1,2,4]triazolo[1,5-a]pyrimidine (6.2 g, 40 mmol) and D-prolinol (4 g, 39
mmol) was heated for 3 h to 120 °C in the microwave with increased pressure up to 10
bar. After solvent removal and silica gel chromatography 8.4 g of yellow crystals were
obtained. [(2R)-1-([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)pyrrolidin-2-yl]methanol (4 g, 18
mmol) was dissolved in CDM (40 ml) and TEA (5.8 ml) was added. At 0 °C a solution of
methanesulfonyl chloride (2.8 ml; 36 mmol) in CDM (10 ml) was added slowly. After
complete addition the mixture was stirred at RT for 12 h. After aqueous extraction the
organic phase was evaporated to dryness and purified by chromatography giving the
product as brown oil (2.2 g). A suspension of [(2R)-1-([1,2,4]triazolo[1,5-a]pyrimidin-7-
yl)pyrrolidin-2-yl]methyl methanesulfonate (250 mg, 0.8 mmol), isoquinolin-5-ol (120 mg,
0.8 mmol) and cesium carbonate (1 g, 3 mmol) in DMF (20 ml) was heated to 80 °C for
12 h. After solvent removal and aqueous work up the organic phase was evaporated
and purified by chromatography to give 5-[[(2R)-1-([1,2,4]triazolo[1,5-a]pyrimidin-7-
yl)pyrrolidin-2-yl]methoxy]isoquinoline 27 as beige solid (120 mg). 1H NMR (500 MHz,
DMSO-d6) δ 9.24 (s, 1H), 8.47 – 8.39 (m, 2H), 8.31 (d, J = 5.7 Hz, 1H), 7.64 (d, J = 8.2
Hz, 1H), 7.54 (m, 2H), 7.23 (d, J = 7.7 Hz, 1H), 6.43 (d, J = 5.7 Hz, 1H), 5.72 (s, 1H),
4.44 (dd, J = 9.8, 5.0 Hz, 1H), 4.25 (dd, J = 9.8, 6.4 Hz, 1H), 4.00 (br. s, 1H), 3.80 (br. s,
1H), 2.27 (m, 3H), 2.11 (m, 1H). Mp.: 158 – 159 °C.
33.
34.
PDB visualization tool: Benchware® 3D Explorer 2.7; JPR Technologies
Unconstrained flex docking with CCDC Gold Suite v. 5.4.1 did not predict the observed
XRay pose.

Both, the purine and the triazolopyrimidine based
MetAP-2 inhibitors coordinate the manganese ions in
the active site but the 5/6 membered bicyclic ring
system is flipped by 180°C. When binding the latter the
imidazole side chain of His339 is pushed out of its
conformation with the purine MetAP-2 inhibitor.