The synthesis and biological evaluation of 2-amino-4,5,6,7,8,9- hexahydrocycloocta[b]thiophenes as allosteric modulators of the A1 adenosine receptor

Article abstract of DOI:10.1016/j.bmcl.2011.04.080

A series of 2-amino-4,5,6,7,8,9-hexahydrocycloocta[b]thiophenes were prepared and evaluated as potential allosteric modulators of the A₁ adenosine receptor (AR). The structure-activity relationships of the 3-position were explored along with va

Full text of DOI:10.1016/j.bmcl.2011.04.080

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3704–3707
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The synthesis and biological evaluation of 2-amino-4,5,6,7,8,9-hexahydrocyclo-
octa[b]thiophenes as allosteric modulators of the A1 adenosine receptor
Luigi Aurelio a, , Arthur Christopoulos b, Bernard L. Flynn a, Peter J. Scammells a, Patrick M. Sexton b,
⇑
Celine Valant b,
⇑
a Medicinal Chemistry and Drug Action, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Parade, Parkville, Vic. 3052, Australia
b Drug Discovery Biology Laboratory, Monash Institute of Pharmaceutical Sciences and Department of Pharmacology, 381 Royal Parade, Parkville, Vic. 3052, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2-amino-4,5,6,7,8,9-hexahydrocycloocta[b]thiophenes were prepared and evaluated as poten-
tial allosteric modulators of the A1 adenosine receptor (AR). The structure–activity relationships of the
3-position were explored along with varying the size of the cycloalkyl ring. 2-Aminothiophenes with
amide and hydrazide groups in the 3-position were completely inactive in an A1-AR-mediated ERK1/2
phosphorylation assay, yet most of the 3-benzoyl substituted compounds exhibited allosteric effects
on responses mediated by the orthosteric agonist, R-PIA. Despite finding an increase in both agonistic
and allosteric activities by going from a cyclopentyl ring to a cyclohexyl ring in the 3-benzoyl series,
decreases were observed when further increasing the ring size. Varying the substituents on the phenyl
ring of the 3-benzoyl group also affected the activity of these compounds.
Received 8 March 2011
Revised 15 April 2011
Accepted 19 April 2011
Available online 4 May 2011
Keywords:
Allosteric modulator
A1 adenosine receptor (A1-AR)
2-Amino-3-benzoylthiophene
G protein-coupled receptors (GPCRs)
Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
During hypoxic, ischemic and inflammatory events, adenosine
is endogenously released to act as a cardio- and neuro-protective
agent via interaction with multiple subtypes of GPCRs. Efforts to
selectively target these GPCR subtypes using modified adenosine
analogs have been of limited therapeutic value due to poor subtype
selectivity. One way to increase the selectivity of an endogenous li-
gand for a receptor is by targeting an allosteric site on that receptor
with an allosteric enhancer (AE). The positive allosteric modulation
of the A1-AR by 2-aminothiophenes is well documented and has
been explored by a number of research groups.1–3 The pioneering
work of Bruns et al.4,5 investigated compound motifs that were
crucial for activity, and ultimately yielded four potent and selective
allosteric enhancers (the ‘PD series’) that also displayed weak
antagonism of the A1-AR at high concentrations (Fig. 1), with PD
81,7236 showing the most favourable ratio of allosteric enhance-
ment to antagonism.4 Subsequently, the discovery of two allosteric
modulators LUF 5484 and T-62 that are structurally related to PD
71,605, were shown to have improved potency compared to PD
81,723.2,7,8
in an in vitro dissociation kinetic binding assay of the A1-AR.9 By
extrapolation, further increases in ring size may improve allosteric
ligand activity, and thus the aim of the current study was to inves-
tigate this hypothesis by exploring the effect of introducing eight-
membered rings into 2-aminocycloalkyl[b]thiophenes.
To the best of our knowledge, the allosteric effects of 2-amino-
cycloocta[b]thiophenes at the A1-AR have only been investigated
once previously,10 and this study only included one example. We
In a previous study,9 it was shown that increasing the ring size
of 2-aminocycloalkyl[b]thiophenes from five to seven membered
rings improved the ability of these 2-aminothiophenes to act as
AEs to stabilize the agonist-receptor-G protein ternary complex
⇑
Corresponding authors. Tel.: +61 (0) 3 9903 9617 (L.A.); tel.: +61 (0) 3 9903
9091 (C.V.).
E-mail addresses: Luigi.aurelio@monash.edu (L. Aurelio), Celine.Valant@monash.
Figure 1. The PD series of allosteric enhancers discovered by Bruns and co-workers,
edu (C. Valant).
and the second generation allosteric enhancers LUF 5484 and T-62.
0960-894X/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.080

L. Aurelio et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3704–3707
3705
have also revisited some compounds with smaller ring sizes (i.e.,
12a–f) to enable a direct comparison of the effects of ring size to
be made using data obtained from the same functional assay. We
have previously observed that certain compounds which have pro-
duced allosteric effects in kinetic binding assays at higher concen-
trations, act as orthosteric antagonists in functional assays which
measure A1-AR-mediated phosphorylation of ERK1/2 (pERK1/2)
in intact CHO cells.11,12
The synthesis of the 2-aminocycloalkyl[b]thiophenes 2a,b be-
gan with the condensation of cycloalkanones 1c,d with activated
nitriles. When utilising ethyl cyanoacetate the classical Gewald13
conditions with ethanol solvent, nitrile, elemental sulphur and
base did not provide the compounds 2a,b and therefore alternate
conditions were employed. In the case of cycloheptanone 1c, a
1:1 mixture of DMF:Et3N provided the 2-aminothiophene 2a in
adequate yield, but in the case of cyclooctanone 1d no reaction oc-
curred under these conditions. The two step procedure by initially
forming the Knöevenagel condensation product under microwave
irradiation and then cyclising with sulphur provided the crude thi-
ophene 2b (Scheme 1) as a viscous resin, that is, contaminated
with the starting ketone 1d. Purification via column chromatogra-
phy on silica gel failed to separate the thiophene 2b from the ke-
tone 1d as they co-elute. In previous work,14 thiophene 2a was
hydrolysed to the acid 10a under basic conditions and then used
for subsequent amidation reactions. In the case of 2b, hydrolysis
under the same conditions resulted in decomposition and there-
fore the alternate route depicted in Scheme 1 was employed. The
thiophenes 2a,b were readily carbamoylated with Boc anhydride
and then hydrolysed to the acids 3a,b under basic conditions,
which were amidated with BOP reagent to the carboxamides 4 in
adequate yield and purity. The crude carbamates 4 were treated
with TFA to remove the Boc group and the crude 2-aminothioph-
enes were chromatographed and then recrystallized to purity.
The carboxamides 7 and 9 (Scheme 2) required a different route
in their synthesis and it was envisaged that simple hydrolysis from
the nitriles 6 and 8 under acidic conditions would provide the
amides. This was the case with 6, which was dissolved in concen-
trated sulphuric acid giving clean conversion to the amide 7, yet ni-
trile 8 was insoluble and even in 85% phosphoric acid only
returned 8. The carboxamide 9 was ultimately synthesised under
the same conditions as for 2a except ethyl cyanoacetate was
replaced with cyanoacetamide.
Scheme 3. Reagents: (i) TiCl4, pyridine, EtOCOCH2CN then S8, Et2NH, THF; (ii) EtOH,
morpholine, AcOH, S8.
Table 1
Effect of test compounds on A1-AR-mediated stimulation of ERK1/2 phosphorylation
in intact CHO FlpIn cells in the presence of an EC50 concentration of R-PIA
Scheme 1. Reagents: (i) 1c, DMF, Et3N, EtOCOCH2CN, S8 and 1d, PhMe, EtOC-
OCH2CN, AcONH4, MW then S8, EtOH, morpholine; (ii) dioxane, DMAP, Boc2O; (iii)
EtOH, H2O, NaOH; (iv) BOP, collidine, amine, CH2Cl2; (v) TFA, CH2Cl2.
Compound
R
n
Activitya
3
lM
10
l
M
5a
5b
5c
5d
5e
5f
7
9
12a
12b
12c
12d
12e
12f
12g
12h
12i
12j
NH2NH
NH2NH
PhNHNH
PhNHNH
PHCH2NH
3-ClPhCH2NH
NH2
3
4
3
4
4
4
4
3
1
1
2
2
3
3
4
4
4
4
52
48
44
47
48
49
48
47
50
75
70
73
55
53
3
51
48
42
43
50
47
48
50
56
98
91
88
68
74
75
1
1
4
4
1
2
1
3
2
3
5
3
3
7
1
1
3
1
2
1
1
1
3
6
8
3
2
4
NH2
4-CH3Ph
3-CF3Ph
4-CH3Ph
3-CF3Ph
4-CH3Ph
3-CF3Ph
4-CH3Ph
3-CF3Ph
4-ClPh
56 12
57
56
49
2
8
3
65 11
63 10
36
Ph
9
a
Effect of two concentrations (3 and 10 lM) of the test compound on A1-AR-
mediated stimulation of ERK1/2 phosphorylation in intact CHO FlpIn cells, in
presence of an EC50 concentration of R-PIA (determined on the same day as each
assay). Data represent the mean standard deviation of two experiments con-
ducted in triplicate. 0% is defined as basal ERK1/2 phosphorylation in the absence of
R-PIA, whereas 100% is defined as the maximal response to a saturating concen-
tration (100 nM) of R-PIA.
Scheme 2. Reagents: (i) DMF, Et3N, NCCH2CN, S8; (ii) concd H2SO4; (iii) DMF, Et3N,
H2NCOCH2CN, S8.

3706
L. Aurelio et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3704–3707
The aroylthiophenes 12a–f were synthesised via the method of
or hydrazide functionality in the 3-position (compounds 5a–f, 7
and 9) were completely inactive, irrespective of the size of the
cycloalkyl ring. Regarding the 2-amino-3-benzoyl compounds, it
is interesting to observe that most of them exhibited a positive
allosteric effect on R-PIA-mediated response. In this series of
compounds 12a–j, 4-CH3 and 3-CF3 substituents were the main
groups investigated. Within the 4-CH3 substituted family, increas-
ing the size of the cycloalkyl ring induced bi-phasic behaviour,
whereby an increase in both the agonistic and allosteric activities
of the compounds was noted when going from cyclopentyl 12a
to cyclohexyl 12c, whereas a decrease of activity was observed
when extending further to the cycloheptyl 12e and cyclooctyl
12g. In contrast, within the 3-CF3 substituted family, increasing
the size of the cycloalkyl ring seemed to have no particular effect
on the allosteric activity of the compounds, as evidenced by the
fact that no substantial variation of activity was observed between
12b, 12d, 12f and 12h. As a control, the compound presented in
Baraldi’s study, 12j, was also investigated and appeared to show
modest allosteric enhancing activity in ERK1/2 phosphorylation.
Finally, comparison of the unsubstituted benzoyl group in 12j, re-
vealed that substitution with either a methyl (12g), trifluoro-
methyl (12h) or chloro (12i) group was beneficial for the
allosteric activity, while not influencing the agonist properties of
the compounds.
Tormyshev et al.15 with slight modification (Scheme 3). In this
method, the Knöevenagel intermediate (not shown) is preformed
in EtOH, morpholine and acetic acid without removing water in
the process and finally adding elemental sulphur in the one pot.
In this fashion low to moderate yields of products were obtained
with the best yields derived from cyclohexanone 1b and cyclohep-
tanone 1c (12c–e). However, this method was unfruitful when
cyclooctanone 1d was used. Therefore the Lewis acid based con-
densation of ketones and malonates with titanium tetrachloride
devised by Lehnert16 was employed in the synthesis of 12g–j.
Although poor yields were obtained sufficient material was iso-
lated for further work.
In order to assess the biological activity of the novel series of
compounds, we initially screened all compounds using a plate-
based assay of A1-AR-mediated ERK1/2 phosphorylation (pERK1/
2) in intact CHO cells.17 For each compound, two concentrations
(3 and 10 lM) were tested alone to assess intrinsic agonism, and
against an EC50 concentration of the orthosteric agonist R-PIA, to
assess enhancement or inhibition of the A1-AR agonist activity
(Table 1). With the exception of 12b–d (each yielding approxi-
mately 50% of the maximum R-PIA response at 10 lM; Fig. 2A),
none of the compounds displayed any substantial ability to acti-
vate the receptor on their own. In the presence of an EC50 concen-
tration of R-PIA, a large number of compounds were found to be
inactive in modulating the response of the orthosteric agonist at
It is interesting to observe that the best allosteric enhancers are
also the compounds that exhibit agonist properties in their own
right, such that 12b–d are strong allosteric agonists, while 12i
3 and 10 lM (Fig. 2B). In fact, all 2-aminothiophenes with amide
Figure 2. Effect of two different concentrations (3 lM, left bar; 10 lM right bar) of test ligands on A1-AR-mediated stimulation of ERK1/2 phosphorylation in intact CHO FlpIn
cells, in absence (A) or in presence (B) of an EC50 concentration (0.3 nM) of R-PIA (determined on the same day as each assay); dashed line denotes 50% response level. 0% is
defined as basal ERK1/2 phosphorylation in the absence of R-PIA, whereas 100% is defined as the maximal response to a saturating concentration (100 nM) of R-PIA. Data
represent the mean standard deviation of two experiments conducted in triplicate.

L. Aurelio et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3704–3707
3707
ric enhancers at the A1-AR. As a consequence the assumption that
increasing the cycloalkyl ring size to improve the allosteric effects
did not hold. Also, as evidenced by the AlphaScreen plate-based as-
say, that increased allosterism is accompanied by antagonism. It is
also evident that 3-benzoyl substitution is highly desirably and
that methyl or trifluoromethyl groups on the 3-benzoyl are bene-
ficial for the allosteric activity. All of the 2-aminothiophenes with
amide and hydrazide functionality in the 3-position proved to be
inactive in this functional assay.
Acknowledgments
This research was supported by Discovery grant DP0877497 of
the Australian Research Council, Program grant 519461 of the Na-
tional Health and Medical Research Council (NHMRC) of Australia.
Arthur Christopoulos is a Senior, and Patrick Sexton a Principal, Re-
search Fellow of the NHMRC. We are grateful to Dr Michael Crouch,
TGR Biosciences, Adelaide, for generously providing the ERK Sure-
Fire Alphascreen kit reagents.
Supplementary data
Supplementary data (full experimental procedures including 1H
and 13C NMR data of compounds 5a–f, 7, 9 and 12a–j) associated
with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2011.04.080.
References and notes
1. Gao, Z.-G.; Kim, S.-K.; IJzerman, A. P.; Jacobson, K. A. Mini-Rev. Med. Chem. 2005,
5, 545.
2. Baraldi, P. G.; Iaconinoto, M. A.; Moorman, A. R.; Carrion, M. D.; Cara, C. L.; Preti,
D.; López, O. C.; Fruttarolo, F.; Tabrizi, M. A.; Romagnoli, R. Mini-Rev. Med. Chem.
2007, 7, 559.
Figure 3. R-PIA mediated stimulation of ERK1/2 phosphorylation in the absence or
presence of the indicated concentrations of test compounds in intact CHO FlpIn A1-
AR cells. Data represent the mean standard deviation of three experiments
conducted in duplicate. Curves drawn through the points represent the best global
fit of an allosteric model of ligand–receptor interaction.
3. La Motta, C.; Sartini, S.; Morelli, M.; Taliani, S.; Da Settimo, F. Curr. Top. Med.
Chem. 2010, 10, 976.
4. Bruns, R. F.; Fergus, J. H. Mol. Pharmacol. 1990, 38, 939.
5. Bruns, R. F.; Fergus, J. H.; Coughenour, L. L.; Courtland, G. G.; Pugsley, T. A.;
Dodd, J. H.; Tinney, F. J. Mol. Pharmacol. 1990, 38, 950.
6. Aurelio, L.; Flynn, B. L.; Scammells, P. J. Aust. J. Chem. 2009, 62, 402.
7. Van der Klein, P. A. M.; Kourounakis, A. P.; Ijzerman, A. P. J. Med. Chem. 1999, 42,
3629.
8. Baraldi, P. G.; Pavani, M. G.; Leung, E.; Moorman, A. R.; Varani, K.; Vincenzi, F.;
Boreac, P. A.; Romagnoli, R. Bioorg. Med. Chem. 2006, 16, 1402.
9. Tranberg, C. E.; Zickgraf, A.; Giunta, B. N.; Luetjens, H.; Figler, H.; Murphree, L.
J.; Falke, R.; Fleischer, H.; Linden, J.; Scammells, P. J.; Olsson, R. A. J. Med. Chem.
2002, 45, 382.
10. Baraldi, P. G.; Zaid, A. N.; Lampronti, I.; Fruttarolo, F.; Pavani, M. G.; Tabrizi, M.
A.; Shryock, J. C.; Leung, E.; Romagnolia, R. Bioorg. Med. Chem. Lett. 2000, 10,
1953.
11. Aurelio, L.; Valant, C.; Flynn, B. L.; Sexton, P. M.; Christopoulos, A.; Scammells,
P. J. J. Med. Chem. 2009, 52, 4543.
12. Aurelio, L.; Valant, C.; Figler, H.; Flynn, B. L.; Linden, J.; Sexton, P. M.;
Christopoulos, A.; Scammells, P. J. Bioorg. Med. Chem. 2009, 17, 7353.
13. Gewald, K.; Schinke, E.; Bottcher, H. Chem. Ber. 1966, 99, 94.
14. Nikolakopoulos, G.; Figler, H.; Linden, J.; Scammells, P. J. Bioorg. Med. Chem.
2006, 14, 2358.
15. Tormyshev, V. M.; Trukhin, D. V.; Rogozhnikova, O. Y.; Mikhalina, T. V.;
Troitskaya, T. I.; Flinn, A. Synlett 2006, 2559.
16. Lehnert, W. Tetrahedron 1973, 29, 635.
appears as a weak allosteric agonist. To investigate this observa-
tion, we constructed complete R-PIA concentration–response
curves in the absence or presence of increasing concentrations of
two of the compounds. We chose 12c as an allosteric enhancer
with strong agonism and 12h as an allosteric enhancer with weak
agonsim. Increasing concentrations of 12c promoted a leftward
shift of the orthosteric agonist concentration–response curve, with
concomitant strong (allosteric) agonist activity (Fig. 3A). In con-
trast, increasing concentrations of 12h more modestly potentiated
the R-PIA concentration–response curve, while exhibiting minimal
allosteric agonism (Fig. 3B). Application of an operational model of
allosterism17,18 to the data yielded pKB estimates of the affinity of
the compounds 12c and 12h for the allosteric site of 5.10 0.14
and 5.07 0.17, respectively, and estimates of the positive cooper-
ativity factors of the interactions with log
ab = 0.98 0.13 (i.e.,
a
b = 10) and 0.46 0.07 (i.e., b = 3), respectively. It appears, there-
a
fore, that, neither the type of substituent on the 3-benzoyl group
nor the size of the cycloalkyl ring are the key points for engender-
ing affinity to the allosteric site, but rather influence the coopera-
tivity of these compounds exerted on R-PIA.
17. Aurelio, L.; Valant, C.; Flynn, B. L.; Sexton, P. M.; White, J. M.; Christopoulos, A.;
Scammells, P. J. J. Med. Chem. 2010, 53, 6550.
18. Leach, K.; Sexton, P. M.; Christopoulos, A. Trends Pharmacol. Sci. 2007, 28, 382.
In conclusion, a novel series of 2-amino-4,5,6,7,8,9-hexahydro-
cycloocta[b]thiophenes were synthesised and evaluated as alloste-