Synthesis and SAR of tetrahydroisoquinolines as Rev-erbα agonists

Article abstract of DOI:10.1016/j.bmcl.2012.04.023

The design and synthesis of a novel series of Rev-erbα agonists is described. The development and optimization of the tetrahydroisoquinoline series was carried out from an earlier acyclic series of Rev-erbα agonists. Through the optimization of the scaffo

Full text of DOI:10.1016/j.bmcl.2012.04.023

Bioorganic & Medicinal Chemistry Letters 22 (2012) 3739–3742
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of tetrahydroisoquinolines as Rev-erba agonists
Romain Noel, Xinyi Song, Youseung Shin, Subhashis Banerjee, Douglas Kojetin, Li Lin, Claudia H. Ruiz,
⇑
Michael D. Cameron, Thomas P. Burris, Theodore M. Kamenecka
Department of Molecular Therapeutics and Translational Research Institute, The Scripps Research Institute, Scripps Florida, 130 Scripps Way #A2A, Jupiter, FL 33458, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design and synthesis of a novel series of Rev-erba agonists is described. The development and opti-
Received 23 February 2012
Accepted 3 April 2012
Available online 13 April 2012
mization of the tetrahydroisoquinoline series was carried out from an earlier acyclic series of Rev-erb
a
agonists. Through the optimization of the scaffold 1, several potent compounds with good in vivo profiles
were discovered.
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Nuclear hormone receptor
Agonist
Structure–activity relationship
The nuclear hormone receptor superfamily (NR) has proven to
be a rich source of targets for development of drugs that target a
myriad of human diseases.1–3 Of the 48 members of the NR that ex-
ist in humans, roughly half have identified endogenous ligands
which modulate the receptors via binding to the ligand binding do-
main. The other half remain orphan receptors with unidentified or
is the first synthetic Rev-erb
date. In our internal cell-based assay, this compound dose-depen-
dently increased the interaction of Rev-erb with a CoRNR box
peptide derived from the transcriptional corepressor NCoR with
an EC50 = 2.3
M.15
a ligand described in the literature to
a
l
Unfortunately, this compound displayed poor rodent pharma-
cokinetics limiting its use as an in vivo probe. We set out to opti-
mize both the in vitro and in vivo profile of Rev-erb agonists so
as to provide novel tools to interrogate the function of these recep-
tors in vivo. This will hopefully help to validate Rev-erb as a poten-
tial drug target for treatment of disorders of the CNS and periphery.
In an effort to restrict free rotation in 1 and possibly improve
activity, we investigated restricting two arms of the molecule into
a ring as described in Figure 2. Constraining the glycine side chain
to the phenyl ring would afford tetrahydroisoquinoline (TIQ) scaf-
fold A. Given the simplicity and commercial availability of the TIQ
starting material, chemical structure–activity relationships (SAR)
began. A few simplifications were made based on SAR that was
learned in the acyclic series (1) (to be reported elsewhere). The
chlorine atom at the para-position of the benzyl group was not
needed for potency, and we began with the ethyl ester rather than
t-butyl, full well knowing that we would eventually have to modify
or remove this functional group if we had hopes of investigating
the compounds in vivo efficacy. Lastly, we looked to replace the
no known natural ligands. The nuclear hormone receptors Rev-erb
a
and Rev-erbb were recently deorphanised when the porphyrin
heme was identified as the physiological ligand for these recep-
tors.4–7 Heme binding is required for these receptors to recruit
corepressors such as NCoR so that they may actively repress tran-
scription of their target genes. The observation that Rev-erbs are
ligand regulated suggests that development of synthetic ligands
may be possible. Both Rev-erbs have particularly high expression
in the liver, adipose tissue, skeletal muscle and brain.8–10 The
Rev-erbs repress the transcription of key genes involved in a
number of physiological processes including circadium rhythm
and lipid and glucose metabolism.11,12 Dysregulation of the circa-
dian rhythm is associated with several disorders of the nervous sys-
tem including bipolar disorder, depression, anxiety, and sleep
disorders whereas metabolic effects have roles in diabetes and
obesity.
Recently, a novel ligand of Rev-erba (Fig. 1, GSK4112/SR-6452/
1) was identified in a fluorescence resonance energy transfer
(FRET) assay that significantly and specifically enhances the Rever-
ba
-NCoR interaction with an EC50 value of 400 nM.13,14 The com-
Cl
pound was reported to show no activity on related nuclear
hormone receptors (LRH1, SF1, FXR, or ROR ) using the same FRET
or LXRb in reporter-gene assays. This
GSK4112/SR6452/1
a
assay and no activity on LXR
a
O2N
N
S
CO2tBu
⇑
Corresponding author.
E-mail address: kameneck@scripps.edu (T.M. Kamenecka).
Figure 1. Rev-erb lead identified in FRET assay.
a
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.04.023

3740
R. Noel et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3739–3742
Table 2
Cl
Cl
Tertiary TIQ amides as Rev-erb
a
agonists
R
O
N
cyclize
O2N
O2N
N
N
OEt
S
S
CO2tBu
CO2tBu
A
1
Compound
R
aMax Inh
EC50
bNT
(lM)
Figure 2. Cyclization approach to tetrahydroisoquinoline analogs.
3g
0.64
0.73
0.51
0.81
N
O
O
O
R
X
a or b or c
or d or e
3h
3i
NT
NH
N
OEt
OEt
O
O
2
3a-x
>3.0
NT
O
X = CH2, SO2, CO, CO2, CONH
O
3j
O
Scheme 1. Synthesis of tetrahydroisoquinoline analogs. Reagents: (a) RCOCl, TEA,
CH2Cl2; (b) RSO2Cl, pyr, CH2Cl2; (c) RNCO, toluene; (d) ClCO2R, DIEA, CH2Cl2; (e)
RCHO, NaBH(OAc)3, ClCH2CH2Cl.
O
O
3k
3l
0.97
0.78
NT
NT
Br
nitrothiophene group, which could potentially lead to toxicity is-
sues in vivo.
Cl
O
O
Starting with commercially available TIQ-ethyl ester 2, the
amine was functionalized with a variety of groups in a single step
(Scheme 1). Reductive aminations gave tertiary amines and acyla-
tion with the appropriate electrophile afforded amides, ureas, car-
bamates and sulfonamides. This led to rapid chemical SAR.
Compounds were screened in a cell based luciferase assay in a
two-step format. Cells were co-transfected with an expression
3m
3n
0.97
1.0
NT
NT
N
O
O
3o
3p
0.41
0.53
2.5
2.1
plasmid harboring full-length Rev-erb
driven by the Bmal1 promoter. Compounds were first screened at
two concentrations (1 and 10 M) to determine the effect on
a and a luciferase reporter
O
l
repression of Bmal1 transcription. Rev-erb is a transcriptional
repressor. Rev-erb agonists lead to recruitment of co-repressors,
which leads to repression of transcription. Maximum inhibition
O
a
Results are average of two or more experiments. Value = fold change relative to
DMSO control at 10 M compound. The lower the number, the more efficacious the
l
at 10 lM is reported. A value of 1.0 effectively means no repres-
agonist in terms of repressing transcription.
b
NT = not tested.
Table 1
Tertiary amine TIQ Rev-erb
a
agonists
sion. Compounds that appeared efficacious were then fully titrated
in an eleven-point dose response format to generate EC50 values. In
our in-house assay, 1 showed only modest potency and efficacy
(Table 1).
R
O
N
OEt
The initial round of analogs investigated were tertiary amines
synthesized by reductive amination of the TIQ ester 2 with a vari-
ety of aryl and heteroaryl carboxaldehydes (Table 1). The com-
pound closest in structure to acyclic lead 1 was 3a, and exhibited
similarly poor relative efficacy with regards to repression of tran-
scription. The other analogs made (3b–3f) displayed equally poor
efficacy at a single concentration and therefore, concentration re-
sponse curves (CRC’s) were not generated.
Acylation of the TIQ nitrogen atom was investigated next with a
series of aryl and heteroaryl carboxylic acid chlorides (Table 2).
Amide groups bearing smaller groups (mono-aryl, and small het-
erocycles) showed no improvement in activity (3k–n), whereas
several biaryl-type derivatives showed modest improvements in
Compound
R
aMax Inh
EC50
2.3
(lM)
1
0.82
0.78
3a
bNT
S
O2N
3b
3c
1.0
NT
NT
S
HO2C
O2N
0.90
O
3d
1.0
NT
F3C
3e
3f
1.0
NT
NT
N
efficacy at 10 lM (3g–h, 3o–p). Unfortunately, improved efficacy
N
did not necessarily translate into an improved EC50, as those ana-
logs titrated showed similar potency to acyclic lead 1. The 1-naph-
thyl analog (3o) and the tetrahydronaphthyl analog 3p exhibited
the best in vitro activity amongst the amides synthesized.
0.95
S
a
Results are average of two or more experiments. Value = fold change relative to
DMSO control at 10 M compound. The lower the number, the more efficacious the
l
agonist in terms of repressing transcription.
The corresponding sulfonamides, carbamates, and ureas were
also investigated (Table 3). In this case, both mono-aryl and bi-aryl
b
NT = not tested.

R. Noel et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3739–3742
3741
Table 3
Table 4
Sulfonamide, carbamate and urea TIQ Rev-erb
a
agonists
Naphthylamide TIQ Rev-erba agonists
R
N
OEt
O
N
O
Compound
R
aMax Inh
EC50
bNT
(lM)
R
Compound
R
aMax Inh
EC50 (lM)
3q
0.75
0.50
Br
S
CO2Et
3n
4
0.41
0.67
0.61
0.59
0.55
0.71
0.65
0.37
0.41
0.41
0.74
0.40
0.53
0.43
2.5
bNT
NT
O2
CO2H
CONH2
5a
5b
5c
6a
6b
6c
6d
6e
7a
7b
7c
7d
3r
1.7
S
CONHPh
CONHCH2Ph
CH2OH
O2
NT
>3.0
NT
3s
3t
0.51
1.0
1.1
NT
SO2
O
CH2OCOCH3
CH2OPh
CH2OCH2Ph
NT
F3C
Cl
O
0.84
0.65
1.2
NT
O
3u
3v
1.3
1.0
NT
NT
CH2O-1-Naphthyl
CH2NH2
O
O
O
CH2NHSO2Ph
NT
O
CH2NHCO2CH2Ph
CH2NHCH2Ph
NT
3w
3x
1.5
1.0
NT
NT
N
H
>3.0
a
Results are average of two or more experiments. Value = fold change relative to
DMSO control at 10 M compound. The lower the number, the more efficacious the
agonist in terms of repressing transcription.
O
l
N
H
b
NT = not tested.
a
Results are average of two or more experiments. Value = fold change relative to
DMSO control at 10 M compound. The lower the number, the more efficacious the
l
agonist in terms of repressing transcription.
Table 5
b
NT = not tested.
Substituted phenyl ethers as TIQ Rev-erb
a
agonists
sulfonamides exhibited good potency and efficacy (3q–s). Interest-
ingly, all of the carbamates and ureas synthesized were essentially
inactive (3t–x). Nonetheless, this was the first time in the cyclic
series that compounds with improved potency and efficacy relative
to 1 had been identified.
N
O
O
R
Compound
R
aMax Inh
EC50 (lM)
Having somewhat optimized the nitrogen substituent, we
turned our attention to replacing the ester group (Table 4). Chem-
istry was carried out under standard conditions as described in
Scheme 2. The ester was hydrolyzed to the acid and converted to
various amides (5a–c). The ester could also be reduced to the pri-
mary alcohol (6a). This could then be further derivatized to a vari-
ety of ethers either by O-alkylation or through Mitsunobu
reactions with phenols (6b–r). In turn, the primary alcohol could
be converted to the primary amine via Mitsunobu reaction, and
then further functionalized (7a–d). While this chemistry was con-
ducted in racemic fashion, both enantiomers of the parent TIQ
ethyl ester are commercially available. This may lead to further
improvements in potency.
6c
6f
6g
6h
6i
6j
6k
6l
6m
6n
6o
6p
6q
6r
H
4-F
4-Ph
4-OMe
4-OEt
0.37
0.43
0.40
0.38
0.35
0.45
0.50
0.40
0.35
0.40
0.45
0.30
0.55
0.45
0.84
0.80
0.65
0.12
0.097
0.077
0.14
0.60
0.12
2.0
4-O-n-Pr
4-O-t-Bu
4-OCF3
4-1,2,4-Triazole
2-OMe
3-OMe
3-N(CH3)2
3,4-Di-OMe
2,4-Dif
0.62
0.78
0.33
0.71
a
As anticipated, the ester was not required for potency, as the
corresponding acid (4), and amide derivatives were equally effica-
cious (5a–c). Derivatives of the primary alcohol were also quite po-
tent and efficacious. Ether analogs 6c and 6d were the first agonists
Results are average of two or more experiments. Value = fold change relative to
DMSO control at 10 M compound. The lower the number, the more efficacious the
agonist in terms of repressing transcription.
l
synthesized in this cyclic series with EC50’s <1 lM. Primary amine
analogs (7b–7d) showed improvement in efficacy at 10
lM, how-
synthesized with an EC50 = 0.077 lM. Larger groups (t-Bu) were
ever EC50’s were right shifted.
also tolerated (6k). ortho-Methoxy substitution (6n) led to a loss
in potency, however meta-substituted analogs (6o–6q) retained
good activity.
These analogs show improved potency and efficacy relative to
the lead (1), and are devoid of a nitro group. Having improved the
potency and efficacy through modification of the TIQ side chain,
and given the iterative nature of medicinal chemistry, future rounds
of SAR will go back and reinvestigate the TIQ nitrogen group as well
as substitution of the TIQ aryl ring. The p-toluenesulfonyl analog 3s
Efforts to further improve in vitro activity were explored by
synthesizing substituted analogs of 6c using substituted phenols
(Table 5). Substitution at the 4-position with fluorine (6f, 6r) or a
phenyl ring (6g) had little effect on potency, however, 4-substitu-
tion with a polar residue showed a dramatic increase in activity (as
much as sevenfold, 6h). Slight improvement in potency was ob-
served as the alkyl group increased in size from methyl (6h) to
ethyl (6i), to n-propyl (6j) which was the most potent compound

3742
R. Noel et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3739–3742
In summary, a novel series of TIQ Rev-erb
a agonists has been
developed based on the lead 1. Optimization of this scaffold by
modification of the nitrogen substitution as well as the C-3 posi-
a
b
N
N
N
tion side chain led to a series of potent and efficacious Rev-erb
a
O
6a
O
agonists. Several compounds from this class have improved phar-
macokinetics and brain penetration relative to the lead structure
1. Evaluation of compounds of this type in vivo in models of circa-
dian rhythm is ongoing and will be reported in due course.
OH
OR1
O
CO2Et
c,d
3n
6b-r
e,f
Acknowledgement
This work was supported, in whole or in part, by National Insti-
tutes of Health Grants DK080201 and MH093429 (to T. P. B.).
N
N
CONHR4
O
O
5a-c
NR2R3
7a-d
References and notes
Scheme 2. Modifications to C-3 ester. Reagents: (a) LiBH4, MeOH; (b) R1Br, NaH or
R1OH, PPh3, DIAD or AcCl, pyr; (c) LiOH, THF; (d) R4NH2, HATU, DIEA; (e) (1)
(PhO)2P(O)(N3), PPh3, THF; (2) Pd/C, H2; (f) R2OCOCl or R3SO2Cl, Et3N or R2CHO,
NaBH(OAc)3.
1. Schulman, I. G. Adv. Drug Delivery Rev. 2010, 62, 1307.
2. Levi, M.; Wang, X.; Choudhury, D. Contrib. Nephrol. 2011, 170, 209.
3. Burris, T. P.; Busby, S. A.; Griffin, P. R. Chem. Biol. 2012, 19, 51.
4. Raghuram, S.; Stayrook, K. R.; Huang, P.; Rogers, P. M.; Nosie, A. K.; McClure, D.
B.; Burris, L. L.; Khorasanizadeh, S.; Burris, T. P.; Rastinejad, F. Nat. Struct. Mol.
Biol. 2007, 14, 1207.
5. Yin, L.; Wu, N.; Curtin, J. C.; Qatanani, M.; Szwergold, N. R.; Reid, R. A.; Waitt, G.
M.; Parks, D. J.; Pearce, K. H.; Wisely, G. B.; Lazar, M. A. Science 2007, 318, 1786.
6. Burris, T. P. Mol. Endocrinol. 2008, 22, 1509.
Table 6
In vivo data of selected Rev-erb
a
agonists
7. Kumar, N.; Solt, L. A.; Wang, Y.; Rogers, P. M.; Bhattacharyya, G.; Kamenecka, T.
M.; Stayrook, K. R.; Crumbley, C.; Floyd, Z. E.; Gimble, J. M.; Griffin, P. R.; Burris,
T. P. Endocrinology 2010, 151, 3015.
c
Compound
at1/2
(h)
Clp (mL/
min/kg)
Vd (L/
kg)
% F
b[plasma]
[Brain]
%
l
M
lM
b.p.
8. Dumas, B.; Harding, H. P.; Choi, H. S.; Lehmann, K. A.; Chung, M.; Lazar, M. A.;
Moore, D. D. Mol. Endocrinol. 1994, 8, 996.
9. Forman, B. M.; Chen, J.; Blumberg, B.; Kliewer, S. A.; Henshaw, R.; Ong, E. S.;
Evans, R. M. Mol. Endocrinol. 1994, 8, 1253.
10. Bonnelye, E.; Vanacker, J. M.; Desbiens, X.; Begue, A.; Stehelin, D.; Laudet, V.
Cell Growth Differ. 1994, 5, 1357.
1
0.4
1.9
1.7
NT
NT
353
39
43
NT
NT
9.9
3.8
3.8
NT
NT
0
0
0
NT
NT
0.25
1.2
NT
0.61
0.45
0.35
0.3
NT
0.40
1.5
140
25
NT
67
6d
7d
6j
6k
333
11. Feng, D.; Liu, T.; Sun, Z.; Bugge, A.; Mullican, S. E.; Alenghat, T.; Liu, X. S.; Lazar,
M. A. Science 2011, 331, 1315.
a
b
c
Rat pharmacokinetics, 1 mg/kg iv, 2 mg/kg po.
Compound dosed 10 mg/kg ip in mice, plasma and brain levels at t = 2 h.
b.p. = brain penetration.
12. Solt, L. A.; Kojetin, D. J.; Burris, T. P. Future Med. Chem. 2011, 3, 623.
13. Meng, Q. J.; McMaster, A.; Beesley, S.; Lu, W. Q.; Gibbs, J.; Parks, D.; Collins, J.;
Farrow, S.; Donn, R.; Ray, D.; Loudon, A. J. Cell Sci. 2008, 121, 3629.
14. Grant, D.; Yin, L.; Collins, J. L.; Parks, D. J.; Orband-Miller, L. A.; Wisely, G. B.;
Joshi, S.; Lazar, M. A.; Willson, T. M.; Zuercher, W. J. ACS Chem. Biol. 2010, 5, 925.
15. Unpublished in-house data.
16. Rat pharmacokinetics and mouse brain penetration: Pharmacokinetics of test
compounds was assessed in Sprague–Dawley rats (n = 3). Compounds were
dosed intravenously at 1 mg/kg and orally by gavage at 2 mg/kg. Blood was
taken at eight time points (5, 15, 30 min, 1, 2, 4, 6, 8 h) and collected into EDTA
containing tubes and plasma was generated using standard centrifugation
techniques. Plasma proteins were precipitated with acetonitrile and drug
concentrations were determined by LC–MS/MS. Data was fit by WinNonLin
was twice as potent as the corresponding naphthoyl amide (3o), so
perhaps there is more potency to be gained by mixing and matching
optimal substituents at both positions. These efforts are on-going.
Several compounds from this series were profiled in vivo (Table
6).16 The initial lead 1 is a poor tool for in vivo use given its poor
in vivo pharmacokinetic (PK) profile. Its PK profile is characterized
by very high plasma clearance, high volume of distribution, and
poor oral bioavailability in rats. However, it has a fairly low polar
surface area, leading to good brain penetration in mice despite
poor plasma exposure. TIQ analogs 6d and 7d have a much lower
clearance rate, volume of distribution, and longer half-life,
although they too, have poor oral absorption. They do, however,
have decent brain penetration. The t-butoxy analog 6k has the best
plasma and brain exposure of the compounds investigated.
Following a single 10 mg/kg dose, there is 20-fold drug concentra-
tion in brain at t = 2 h relative to its EC50. Further refinements in
structure will look at improving in vivo exposure including plasma
and brain penetration.
using
a noncompartmental model and basic pharmacokinetic parameters
including peak plasma concentration (Cmax), oral bioavailability, exposure
(AUC), half-life (t1/2), clearance (CL), and volume of distribution (Vd) were
calculated. CNS exposure was evaluated in C57Bl6 mice (n = 3). Compounds
were dosed at 10 mg/kg intraperitoneally and after 2 h blood and brain were
collected. Plasma was generated and the samples were frozen at À80 °C. The
plasma and brain were mixed with acetonitrile (1:5 v:v or 1:5 w:v,
respectively). The brain sample was sonicated with a probe tip sonicator to
break up the tissue, and samples were analyzed for drug levels by LC–MS/MS.
Plasma drug levels were determined against standards made in plasma and
brain levels against standards made in blank brain matrix. All procedures were
approved by the Scripps Florida IACUC.