Discovery of aminopiperidine-based Smac mimetics as IAP antagonists

Article abstract of DOI:10.1016/j.bmcl.2011.12.109

A series of structurally unique Smac mimetics that act as antagonists of inhibitor of apoptosis proteins (IAPs) has been discovered. While most previously described Smac mimetics contain the proline ring (or a similar cyclic motif) found in Smac, a key feature of the compounds described herein is that this ring has been removed. Despite this, compounds in this series potently bind to cIAP1 and elicit the expected phenotype of cIAP1 inhibition in cancer cells. Marked selectivity for cIAP1 over XIAP is observed for these compounds, which is attributed to a slight difference in the binding groove between the two proteins and the resulting steric interactions with the inhibitors. XIAP binding can be improved by constraining the inhibitor so that these unfavorable steric interactions are minimized.

Full text of DOI:10.1016/j.bmcl.2011.12.109

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1690–1694
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of aminopiperidine-based Smac mimetics as IAP antagonists
⇑
Edward J. Hennessy , Jamal C. Saeh, Li Sha, Terry MacIntyre, Haiyun Wang, Nicholas A. Larsen,
Brian M. Aquila, Andrew D. Ferguson, Naomi M. Laing, Charles A. Omer
AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of structurally unique Smac mimetics that act as antagonists of inhibitor of apoptosis proteins
(IAPs) has been discovered. While most previously described Smac mimetics contain the proline ring
(or a similar cyclic motif) found in Smac, a key feature of the compounds described herein is that this ring
has been removed. Despite this, compounds in this series potently bind to cIAP1 and elicit the expected
phenotype of cIAP1 inhibition in cancer cells. Marked selectivity for cIAP1 over XIAP is observed for these
compounds, which is attributed to a slight difference in the binding groove between the two proteins and
the resulting steric interactions with the inhibitors. XIAP binding can be improved by constraining the
inhibitor so that these unfavorable steric interactions are minimized.
Received 29 November 2011
Revised 20 December 2011
Accepted 21 December 2011
Available online 31 December 2011
Keywords:
Apoptosis
Inhibitor of apoptosis proteins
Smac
Ó 2011 Elsevier Ltd. All rights reserved.
Protein–protein interactions
Isothermal titration calorimetry (ITC)
In multicellular organisms, apoptosis (programmed cell death)
is a crucial mechanism for maintaining homeostasis and for elim-
inating damaged cells.1 Under normal circumstances, both extrin-
sic apoptotic stimuli and intrinsic cellular stresses can lead to a
highly coordinated signaling cascade that terminates in the death
of the cell. Cancer cells, however, often do not respond to apoptotic
cues; this resistance to cell death is widely recognized as one of the
hallmarks of cancer.2 As a result, a number of elements of the
apoptotic machinery have received attention as potential targets
for cancer therapeutics.
The inhibitor of apoptosis proteins (IAPs) are one of the most
widely investigated classes of apoptosis regulators.3 Certain IAPs
(e.g., XIAP, cIAP1, cIAP2), which are frequently overexpressed in
cancer cells, can directly bind to and inhibit proteases within the
cell that are essential for apoptosis (caspases). The IAP family is
characterized by the presence of one to three repeating domains
of ꢀ70 amino acids known as BIR (baculovirus IAP repeat)
domains, and it is the interaction between these BIR domains
and caspases that results in apoptosis inhibition.
interaction occurs through the four N-terminal residues of Smac
(Ala-Val-Pro-Ile, AVPI),6,7 and numerous small molecule IAP antag-
onists that mimic interactions with key hotspots in the Smac bind-
ing groove (e.g., the hydrophobic pockets occupied by the methyl
side chain of the alanine residue and the sec-butyl side chain of
the isoleucine residue) have been described in the patent and
chemical literature.8,9 Moreover, a number of these IAP inhibitors
are currently under investigation as anticancer agents in clinical
trials.10
The proline ring of AVPI (1) is thought to contribute to the over-
all binding affinity of Smac towards XIAP-BIR3 by both restricting
the conformation of the surrounding peptide chain and by estab-
lishing favorable non-bonding interactions between the ring meth-
ylene groups with Trp323.11 Thus, most previously described
peptidic Smac mimetics maintain a cyclic structure at that position
(Fig. 1). However, several proapoptotic proteins (related to Smac)
derived from Drosophila do not contain proline at this position
(Reaper, Grim),12,13 and synthetic tetrapeptides based on these
have been shown to have modest binding affinity for
XIAP-BIR3.14 Moreover, it is has been previously demonstrated that
cIAP1-BIR3 is tolerant of Smac-like peptides containing a valine
residue in place of the proline.15 As part of our medicinal chemistry
efforts directed at finding novel IAP antagonists, we devised a
strategy to excise the proline ring from examples of previously
published Smac mimetics, with the hope that target potency would
be maintained. Our intention was that this approach would result
in synthetically tractable scaffolds amenable to rapid exploration
of structure–activity relationships, and that our learnings could
be extrapolated to other series of Smac mimetics. As a result, we
It has been established that the interaction between IAP BIR do-
mains and the proapoptotic caspases can be disrupted by an
endogenous protein normally localized in the mitochondria, Smac
(second mitochondrial activator of caspases).4,5 Smac is released
into the cytoplasm in response to apoptotic stimuli, and binds to
the IAP BIR domains in preference to caspases, thus releasing the
caspases and allowing apoptosis to occur. The bulk of this binding
⇑
Corresponding author.
E-mail address: edward.hennessy@astrazeneca.com (E.J. Hennessy).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.12.109

E. J. Hennessy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1690–1694
1691
Figure 1. Examples of previously described Smac mimetics.
Table 1
have discovered a series of aminopiperidine-based inhibitors
(structurally related to known compound 2) that do not contain
a proline residue or a similar cyclic structure at that position.
Examples of these compounds potently bind to the BIR3 domain
of cIAP1, and also produce the phenotype expected from cIAP1
inhibition in cancer cell lines. Interestingly, this series also demon-
strates unexpectedly high binding selectivity for cIAP1 over XIAP.
Herein, we describe our efforts and the preliminary SAR for the
series.
SAR around ring size and substituent length
Compound 4 (Fig. 2), in which the absolute stereochemistry of
the bridgehead carbon of 2 is maintained, was prepared from com-
mercially available BOC-3-(S)-aminopiperidine and was found to
potently bind to the BIR3 domain of cIAP1 (IC50 = 40 nM). More-
over, since binding of antagonists to cIAP1 is known to result in
autoubiquitination and subsequent protein degradation in certain
cancer cell lines,16,17 we were pleased to observe this activity in
MDA-MB231 cells following treatment with this compound
(EC50 = 113 nM). Encouraged by this promising result, we set about
exploring the SAR around this hit.
Initial work was directed at varying the ring size of the cyclic
amine, as well as modifying the substituent appended to the ring
nitrogen (Table 1). It was found that phenethyl substitution affords
the most potent compounds regardless of ring size, although the
phenethyl-substituted pyrrolidine (6) and homopiperidine (17)
analogs were found to be less potent than the analogous piperidine
(4). Interestingly, however, the benzyl-substituted homopiperidine
(16) is 6 times more potent than the analogous piperidine (11) and
12 times more potent than the corresponding pyrrolidine (5), sug-
gesting that the larger ring better positions the benzyl group to-
wards the hydrophobic pocket occupied by the isoleucine residue
of Smac. Interestingly, acylation of the ring nitrogen results in only
modest drops in potency, implying that the basic piperidine nitro-
gen of 4 may not be optimally interacting with Gly306, a residue
involved in hydrogen bonding interactions with Smac6,7 and many
Smac mimetics. This data also suggests that planarization of the
ring nitrogen resulting from acylation adversely affects the orien-
tation of the aromatic group such that interactions with the hydro-
phobic pocket are suboptimal.
Compound
n
1
0
R
cIAP1-BIR3 IC50 (lM)
4
5
0.04
3.38
6
7
0
0
0.50
0.74
8
0
0
0
5.02
0.78
0.74
9
10
11
12
1
1
1.69
0.79
13
14
15
1
1
1
3.36
0.13
0.45
16
17
2
2
0.27
0.08
Having established that incorporation of the 3-aminopiperidine
motif in these Smac mimetics affords potent inhibitors of cIAP1, we
further investigated the effects of varying different regions of the
scaffold (Table 2). Replacement of the cyclohexylglycine residue
with tert-leucine (18) maintained binding affinity, as was expected
based on work previously conducted in the area.11,14 We also found
that methylation of the amide nitrogen at the aminopiperidine res-
idue (19) resulted in a potent inhibitor, even though steric interac-
tions between this methyl group and the aminopiperidine ring are
likely to result in a different conformation than would be expected
Figure 2. Structure of lead aminopiperidine-based Smac mimetic.

1692
E. J. Hennessy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1690–1694
Table 2
SAR of scaffold modifications
(Supplementary data). It was found that a number of aromatic sub-
stitution patterns are tolerated (Table 3), with an apparent slight
preference for small, lipophilic groups in the ortho- and para- posi-
tions. Despite the relatively flat SAR in this area of the molecule,
however, it was found that more polar groups or larger aromatic
systems (37–39) provided less potent inhibitors.
An unexpected feature of this series of compounds is the
remarkable lack of potency towards the BIR3 domain of XIAP (all
compounds tested had an IC50 greater than 5 micromolar). We
attribute this selectivity to unfavorable steric interactions between
the piperidine ring of these compounds with Tyr324 of XIAP-BIR3.
In cIAP1-BIR3 the corresponding residue is a phenylalanine, and
thus there is more room for Smac mimetics with substituents at
this position to bind. Indeed, this structural difference has previ-
ously been exploited in the design of cIAP1-selective inhibitors.18
Given the current understanding of the interplay between the IAPs
and the different caspases involved in apoptosis, and the expecta-
tion that pan-IAP inhibition is necessary for single agent anticancer
activity, we directed our efforts at increasing the binding affinity of
this series towards XIAP-BIR3.
We surmised that if the piperidine ring of compound 4 was
tethered into a bicyclic system, thus locking it into a boat confor-
mation, then the steric interactions between the inhibitor and
Tyr324 of XIAP-BIR3 would be minimized. Analysis of a docked
pose of 4 overlaid with a proposed bicyclic inhibitor (Fig. 3) dem-
onstrates this principle. Compound 47 was thus synthesized fol-
lowing the route described in Scheme 1. The chiral pyridine-
derived bromolactone 40 was prepared as previously described
in the literature,19 and was converted to the bicyclic Cbz-protected
amino acid 41 by treatment with zinc in acetic acid. Conversion to
acyl azide 42 followed by Curtius rearrangement in the presence of
trifluoroacetic acid afforded the trifluoroacetamide 43. Deacylation
in the presence of methanolic NaOH liberated the primary amine,
which was then coupled to BOC-protected cyclohexylglycine to
give amide 44. Deprotection and coupling to BOC-N-Me-Ala-OH
gave compound 45, which was subjected to concomitant hydroge-
nation and Cbz removal to give bicyclic piperidine 46. Alkylation
with phenethyl bromide followed by final BOC deprotection gave
the target 47.
Compound
R1
R2
H
R3
cIAP1-BIR3 IC50
0.026
(lM)
18
19
t-Bu
Cy
Me
0.044
20
21
Cy
Cy
H
H
0.062
1.99
22
23
t-Bu
t-Bu
H
H
5.75
3.89
in Smac mimetics containing a proline ring. Incorporation of
hydrophobic groups commonly found in Smac mimetics at the C-
terminus yielded mixed results, with the diphenyl-substituted
chain (20) maintaining potency but the (R)-tetrahydronapthyl
(21) giving a poorer inhibitor. Interestingly, while acyl groups of
the appropriate length did not completely abrogate binding affin-
ity, the corresponding sulfonamides (22, 23) were markedly less
potent.
Being that most changes to our initial hit 4 resulted in dimin-
ished potency, we next turned our attention to studying the effects
of substitution on the aromatic ring of the phenethyl group. A
number of substituted phenethylpiperidines were synthesized by
alkylation of a suitably protected piperidine with commercially
available phenethyl bromides, followed by deprotection
Table 3
SAR of phenethyl-substituted piperidines
Ex
R1
Ar
cIAPi-BIR3 IC50
(
lM)
cIAP1 cell EC50 (l
M)a
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Cy
Cy
2-Fluorophenyl
3-Fluorophenyl
4-Fluorophenyl
4-Fluorophenyl
2-Methoxyphenyl
3-Methoxyphenyl
4-Methoxyphenyl
4-Chlorophenyl
4-Bromophenyl
4-Methylphenyl
4-Hydroxyphenyl
4-Nitrophenyl
4-Trifluoromethylphenyl
2-Pyridyl
0.033
0.068
0.018
0.020
0.050
0.052
0.090
0.024
0.024
0.030
0.042
0.032
0.049
0.301
0.593
0.115
0.104
0.279
0.058
0.060
0.156
0.164
0.158
0.117
0.071
0.060
0.068
0.060
0.300
>0.37
>0.37
0.185
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
4-Pyridyl
3-Pyridyl
a
EC50 for cIAP1 degradation in MDA-MB231 cells.

E. J. Hennessy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1690–1694
1693
Table 4
Isothermal titration calorimetry measurements (XIAP-BIR3) for compounds 2 and 47.
Compound
Kd
(lM)
D
H (kcal/mol)
TDS (kcal/mol)
2
47
0.9
0.9
À1.3
À1.9
7.0
6.2
increase in XIAP-BIR3 and cIAP1-BIR3 potency, compound 47 has
improved single agent antiproliferative activity in MDA-MB231
cells (GI50 = 100 nM) relative to compound 4 (GI50 = 1.01 lM).
The profile of compound 47 is comparable to that of the known
Smac mimetic 2 (XIAP-BIR3 IC50 = 280 nM).20 Given the similar
binding affinities towards XIAP-BIR3 of compounds 2 and 47, we
questioned if there are differences in the enthalpic (
pic S) contributions to the Gibbs free energy of binding
G = S) between the two. This is not uncommon, as an in-
DH) and entro-
(D
(D
D
H À TD
crease in the enthalpy of binding (achieved through introduction of
functional groups capable of forming favorable hydrogen bonding
interactions with the protein) is often opposed with a correspond-
ing entropic penalty, resulting in little net change in binding affin-
ity.21 Using isothermal titration calorimetry (ITC), we were able to
Figure 3. Overlay of docked poses of compounds 4 (gray) and 47 (yellow) in the
Smac binding groove of XIAP-BIR3. The inset demonstrates how compound 47
might relieve the steric clash between 4 and Tyr324.
determine that compound 47 shows a significant increase in
DH
relative to 2 (Table 4), albeit with a decrease in the entropic contri-
bution to the binding energy. Thus, we postulate that relieving the
steric clash with Tyr324 results in a tighter hydrogen bonding
interaction between the piperidine basic nitrogen and the back-
bone of Gly306. The corresponding decrease in entropic contribu-
tion for 47 may be a result of a decrease in lipophilicity relative
to 2 (logD of 1.64 vs 2.61) or perhaps due to an increase in
rotational freedom in 47 relative to the more rigid bicyclic 2.
Gratifyingly, compound 47 was found to be a potent inhibitor of
cIAP1-BIR3 (IC50 = 4 nM, a 10-fold increase relative to compound 4)
and effectively induced cIAP1 degradation in MDA-MB231 cells
(EC50 = 5 nM). Moreover, as expected based on our modeling stud-
ies, binding to the BIR3 domain of XIAP was markedly improved
(>20 fold) relative to compound 4 (IC50 of 47 = 365 nM, while
IC50 of 4 is greater than 8 lM). Furthermore, consistent with the
Scheme 1. Synthesis of bridged piperidine 47. Reagents and conditions: (i) Zn, HOAc, 80 °C; (ii) 4-methylmorpholine, ethyl chloroformate, acetone, 0 °C, then NaN3, H2O; (iii)
TFA, EtOAc, 80 °C; (iv) NaOH, MeOH, H2O; (v) Boc-cyclohexylglycine, EDCI, HOBt, DMF, 0 °C to rt; (vi) 4 N HCl/dioxane, rt; (vii) Boc-N-Me-Ala-OH, EDCI, HOBt, 4-
methylmorpholine, DMF, 0 °C to rt; (viii) H2 (1 atm), 10% Pd/C, EtOH, rt; (ix) phenethyl bromide, NaHCO3, CH3CN, 80 °C; (x) TFA, CH2Cl2, rt.

1694
E. J. Hennessy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1690–1694
3. For a recent review, see: Gyrd-Hansen, M.; Meier, P. Nat. Rev. Cancer 2010, 10,
In conclusion, we have discovered a series of novel aminopiperi-
561.
dine-based inhibitors of IAP proteins. These compounds are struc-
turally distinct from most previously described Smac mimetics in
that the cyclic portion of the AVPI scaffold has been deleted. De-
spite this, many members of this series are effective at inducing
cIAP1 degradation in cancer cells. We attribute the high selectivity
for cIAP1 (relative to XIAP) to the subtle structural differences in
the Smac binding groove of the proteins, and have been able to
augment XIAP binding affinity by constraining the piperidine ring
into a boat conformation through the introduction of a bridging al-
kyl chain. Further work is needed to ascertain the potential of this
series to deliver anticancer therapeutics.
4. Du, C.; Fang, M.; Li, Y.; Li, L.; Wang, X. Cell 2000, 102, 33.
5. Verhagen, A. M.; Ekert, P. G.; Pakusch, M.; Silke, J.; Connolly, L. M.; Reid, G. E.;
Moritz, R. L.; Simpson, R. J.; Vaux, D. L. Cell 2000, 102, 43.
6. Liu, Z.; Sun, C.; Olejniczak, E. T.; Meadows, R. P.; Betz, S. F.; Oost, T.; Herrmann,
J.; Wu, J. C.; Fesik, S. W. Nature 2000, 408, 1004.
7. Wu, G.; Chai, J.; Suber, T. L.; Wu, J.-W.; Du, C.; Wang, X.; Shi, Y. Nature 2000,
408, 1008.
8. Mannhold, R.; Fulda, S.; Carosati, E. Drug Discovery Today 2010, 15, 210.
9. Flygare, J. A.; Fairbrother, W. J. Exper. Opin. Ther. Patents 2010, 20, 251.
10. http://www.clinicaltrials.gov.
11. Oost, T. K.; Sun, C.; Armstrong, R. C.; Al-Assaad, A.-S.; Betz, S. F.; Deckwerth, T.
L.; Ding, H.; Elmore, S. W.; Meadows, R. P.; Olejniczak, E. T.; Oleksijew, A.;
Oltersdorf, T.; Rosenberg, S. H.; Shoemaker, A. R.; Tomaselli, K. J.; Zou, H.; Fesik,
S. W. J. Med. Chem. 2004, 47, 4417.
12. White, K.; Grether, M. E.; Abrams, J. M.; Young, L.; Farrell, K.; Steller, H. Science
1994, 264, 677.
Acknowledgments
13. Chen, P.; Nordstrom, W.; Gish, B.; Abrams, J. M. Genes Dev. 1996, 10, 1773.
14. Sharma, S. K.; Straub, C.; Zawel, L. Int. J. Pept. Res. Ther. 2006, 12, 21.
15. Kurakin, A.; Bredesen, D. E. J. Mol. Recognit. 2007, 20, 39.
16. Varfolomeev, E.; Blankenship, J. W.; Wayson, S. M.; Federova, A. V.; Kayagaki,
N.; Garg, P.; Zobel, K.; Dynek, J. N.; Elliott, L. O.; Wallweber, H. J. A.; Flygare, J.
A.; Fairbrother, W. J.; Deshayes, K.; Dixit, V. M.; Vucic, D. Cell 2007, 131, 669.
17. Vince, J. E.; Wong, W. W.-L.; Khan, N.; Feltham, R.; Chau, D.; Ahmed, A. U.;
Benetatos, C. A.; Chunduru, S. K.; Condon, S. M.; McKinlay, M.; Brink, R.;
Leverkus, M.; Tergaonkar, V.; Schneider, P.; Callus, B. A.; Koentgen, F.; Vaux, D.
L.; Silke, J. Cell 2007, 131, 682.
We thank Galina Repik and David Whitston for their work in
helping to conduct the biochemical and cell assays described in
this paper. We also thank Kelly Theriault, Erin Code, Xiahui Zhu,
Ian Hampton, and Hannah Pollard for supplying protein for the as-
says and for the ITC work. Ziling Lu is also acknowledged for assis-
tance in compound purification.
18. Ndubaku, C.; Varfolomeev, E.; Wang, L.; Zobel, K.; Lau, K.; Elliott, L. O.; Maurer,
B.; Federova, A. V.; Dynek, J. N.; Koehler, M.; Hymowitz, S. G.; Tsui, V.;
Deshayes, K.; Fairbrother, W. J.; Flygare, J. A.; Vucic, D. ACS Chem. Biol. 2009, 4,
557.
19. Satoh, N.; Akiba, T.; Yokoshima, S.; Fukuyama, T. Angew. Chem. Int. Ed. 2007, 46,
5734.
20. Chauhan, D.; Neri, P.; Velankar, M.; Podar, K.; Hideshima, T.; Fulciniti, M.;
Tassone, P.; Raje, N.; Mitsiades, C.; Mitsiades, N.; Richardson, P.; Zawel, L.; Tran,
M.; Munshi, N.; Anderson, K. C. Blood 2007, 109, 1220.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.12.109.
References and notes
21. Lafont, V.; Armstrong, A. A.; Ohtaka, H.; Kiso, Y.; Amzel, L. M.; Freire, E. Chem.
Biol. Drug Des. 2007, 69, 413.
1. Kerr, J. F. R.; Wyllie, A. H.; Currie, A. R. Br. J. Cancer 1972, 26, 239.
2. Hanahan, D.; Weinberg, R. A. Cell 2011, 144, 646.