Discovery of imidazo[1,5-a]pyridines and -pyrimidines as potent and selective RORc inverse agonists

Article abstract of DOI:10.1016/j.bmcl.2015.05.055

The nuclear receptor (NR) retinoic acid receptor-related orphan receptor gamma (RORγ, RORc, or NR1F3) is a promising target for the treatment of autoimmune diseases. RORc is a critical regulator in the production of the pro-inflammatory cytokine interleukin-17. We discovered a series of potent and selective imidazo[1,5-a]pyridine and -pyrimidine RORc inverse agonists. The most potent compounds displayed >300-fold selectivity for RORc over the other ROR family members, PPARγ, and NRs in our cellular selectivity panel. The favorable potency, selectivity, and physiochemical properties of GNE-0946 (9) and GNE-6468 (28), in addition to their potent suppression of IL-17 production in human primary cells, support their use as chemical biology tools to further explore the role of RORc in human biology.

Full text of DOI:10.1016/j.bmcl.2015.05.055

Accepted Manuscript
Discovery of imidazo[1,5-a]pyridines and -pyrimidines as potent and selective
RORc inverse agonists
Benjamin P. Fauber, Alberto Gobbi, Kirk Robarge, Aihe Zhou, Adrian Barnard,
Jianhua Cao, Yuzhong Deng, Céline Eidenschenk, Christine Everett, Arunima
Ganguli, Julie Hawkins, Adam R. Johnson, Hank La, Maxine Norman, Gary
Salmon, Susan Summerhill, Wenjun Ouyang, Wei Tang, Harvey Wong
PII:
S0960-894X(15)00529-6
DOI:
http://dx.doi.org/10.1016/j.bmcl.2015.05.055
BMCL 22745
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 April 2015
18 May 2015
21 May 2015
Please cite this article as: Fauber, B.P., Gobbi, A., Robarge, K., Zhou, A., Barnard, A., Cao, J., Deng, Y.,
Eidenschenk, C., Everett, C., Ganguli, A., Hawkins, J., Johnson, A.R., La, H., Norman, M., Salmon, G., Summerhill,
S., Ouyang, W., Tang, W., Wong, H., Discovery of imidazo[1,5-a]pyridines and -pyrimidines as potent and selective
RORc inverse agonists, Bioorganic & Medicinal Chemistry Letters (2015), doi: http://dx.doi.org/10.1016/j.bmcl.
2015.05.055
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.

Discovery of imidazo[1,5-a]pyridines and -pyrimidines as
potent and selective RORc inverse agonists
Benjamin P. Fauber,a,* Alberto Gobbi,a Kirk Robarge,a Aihe Zhou,a Adrian Barnard,b Jianhua
Cao,c Yuzhong Deng,a Céline Eidenschenk,a Christine Everett,a Arunima Ganguli,b Julie
Hawkins,b Adam R. Johnson,a Hank La,a Maxine Norman,b Gary Salmon,b Susan Summerhill,b
Wenjun Ouyang,a Wei Tang,c and Harvey Wonga
a Genentech, Inc., 1 DNA Way, South San Francisco, California 94080, USA
b Argenta, Units 7-9 Spire Green Centre, Flex Meadow, Harlow, Essex, CM19 5TR, UK
c
ChemPartner, No. 5 Building, 998 Halei Road, Zhangjiang Hi-Tech Park Pudong New Area,
Shanghai China, 201203
Corresponding Author:
Benjamin P. Fauber
Genentech, Inc.
1 DNA Way, South San Francisco, CA 94080, USA
E-mail address: Fauber.Benjamin@gene.com
Telephone: +1-650-467-5773
1

Abstract.
The nuclear receptor (NR) retinoic acid receptor-related orphan receptor gamma (RORγ,
RORc, or NR1F3) is a promising target for the treatment of autoimmune diseases. RORc is a
critical regulator in the production of the pro-inflammatory cytokine interleukin-17. We
discovered a series of potent and selective imidazo[1,5-a]pyridine and -pyrimidine RORc inverse
agonists. The most potent compounds displayed >300-fold selectivity for RORc over the other
ROR family members, PPARγ, and NRs in our cellular selectivity panel. The favorable potency,
selectivity, and physiochemical properties of GNE-0946 (9) and GNE-6468 (28), in addition to
their potent suppression of IL-17 production in human primary cells, support their use as
chemical biology tools to further explore the role of RORc in human biology.
Keywords: RORc; RORγ; IL-17; PPARγ; GNE-0946; GNE-6468.
2

Inhibiting the function of interleukin (IL)-17 with anti-IL-17 antibodies has proven
successful in the treatment of psoriasis,1-3 rheumatoid arthritis (RA),4 ankylosing spondylitis,5
and uveitis.6 IL-17 is a validated target for the treatment of psoriasis. IL-23 lies upstream of
IL-17 and anti-IL-12/23 antibodies (e.g. ustekinumab)7 have demonstrated efficacy in the
treatment of moderate-to-severe psoriasis. The recent FDA approval of the anti-IL-17 antibody
secukinumab8 for the treatment of moderate-to-severe plaque psoriasis further highlighted the
importance of the IL-17 pathway in autoimmune diseases.9-10 The nuclear receptor (NR) retinoic
acid receptor-related orphan receptor gamma (RORγ or RORc, also known as NR1F3)11 is an
important regulator in the production of IL-17.12 RORc also plays a role in production of IL-2213
and cytokines from innate lymphoid cells (ILCs)14-15 and γδ T cells.16 Based on the influence of
RORc over multiple inflammatory pathways, it has been proposed that RORc is a valuable
molecular target for the treatment of inflammatory diseases.17-23
Our group has previously reported the discovery and optimization of tertiary sulfonamide
RORc inverse agonists.24-27 In 2012, Organon (Merck) disclosed an acyl-indazole series of RORc
inverse agonists.28 This disclosure led us to investigate an exemplary acyl-indazole compound
from Organon (1, Table 1) in a time-resolved fluorescence biochemical assay that monitored the
ability of the human RORc ligand binding domain (LBD) to bind to a co-activator peptide
derived from steroid receptor co-activator-1 (SRC1).29 Compounds that disrupted the recruitment
of the SRC1 co-activator peptide were RORc inverse agonists. Compound 1 was a potent
inverse agonist in the RORc SRC1 biochemical assay (EC50 = 16 nM).
Compounds with structures similar to 1 have been reported as potent peroxisome
proliferator-activated receptor (PPAR)-γ ligands,30 which led us to profile 1 in a biochemical
assay that monitored the ability of a compound to displace a fluorescent probe from the human
3

PPARγ-LBD.31 In this PPARγ competitive binding assay, 1 displayed a strong affinity for human
PPARγ (IC50 = 130 nM). The PPARγ potency of 1 was superior to the FDA-approved PPARγ
agonists pioglitazone32 (IC50 = 720 nM) and troglitazone33 (IC50 = 340 nM), as measured in our
PPARγ binding assay. Due to the potential risks associated with the modulation of PPARγ,34 we
designed new analogs with the goal of improving selectivity for RORc over PPARγ (as
compared with 1) through the exploration of alternative cores to the 1-acyl-indazole motif
embedded in 1.
In our exploration of different core motifs, we attempted to preserve the 1- and 3-position
vectors of the azole core for the 4'-benzoic acid and keto-arene groups, respectively. We also
prioritized cores that removed the potential metabolic liability of the 1-acyl-indazole moiety, as
related compounds have demonstrated in vivo instability.35 The imidazo[1,5-a]pyridine analog36
(2, Table 1) fulfilled these criteria and also provided an analog with improved RORc inverse
agonist potency (EC50 = 10 nM) in comparison to 1. Compound 2 also possessed a superior
ligand-lipophilicity efficiency (LLE)37-38 in comparison to 1 (2 LLE = 6.5, 1 LLE = 4.9) and an
improved selectivity for RORc over PPARγ (PPARγ IC50 = 1.5 µM). Saturation of the pyridine
ring in 2 provided a 5,6,7,8-tetrahydro-imidazo[1,5-a]pyridine analog (3) and resulted in a
reasonable RORc inverse agonist potency (EC50 = 58 nM) with an LLE value (LLE = 6.4) and
PPARγ affinity (PPARγ IC50 = 2.1 µM) comparable to 2.
Synthesis of an
imidazo[1,5-a]pyrimidine analog (4) provided a modest RORc inverse agonist (EC50 = 84 nM)
and further decreased the affinity for PPARγ (IC50 = 4.2 µM). The imidazo[1,5-a]pyrazine
analog (5) possessed a substantially diminished RORc inverse agonist activity (EC50 = 2.6 µM)
in comparison to 4. Further core changes such as the 3-acyl-tetrahydroindazole (6),
imidazo[1,2-a]pyrazine (7), and 5,6,7,8-tetrahydro-imidazo[1,2-a]pyran (8) were devoid of
4

detectable RORc inverse agonist activity at concentrations up to 10 µM. The difference in
RORc potency between 3 and 6 was striking given their structural similarities. We rationalized
this difference in that the nitrogen atoms in the core of 3 do not form a resonance contribution to
the ketone group, whereas, 6 does allow a resonance contribution into the ketone from the core
N1 nitrogen atom. These differences in resonance contributions between the cores of 3 and 6
could have an impact on the Lewis basicity of the respective ketone functional groups and thus,
how they complement the RORc-LBD ligand binding pocket.
5

Table 1
SAR of core changes
Compd
Core
RORc SRC1 EC50a (μM) [%eff] LLEb LogD (pH 7.4) PPARγ IC50c (µM)
0.016 [-99%]
0.010 [-99%]
0.058 [-99%]
0.084 [-99%]
2.6 [-75%]
>10
4.9
6.5
6.4
6.8
4.9
-
2.9
1.5
0.8
0.3
0.7
-0.3
1.0
1.0
0.13
1.5
1
2
3
4
5
6
7
8
2.1
4.2
ND
ND
ND
ND
>10
-
>10
-
See the Supplementary Data for experimental details associated with each assessment. All assay results are reported
a
as the geometric mean of at least two separate runs. ND = not determined. Inhibition of RORc-LBD recruitment of
the SRC1 co-activator peptide; negative percent efficacy denotes inverse agonism relative to the basal activity of
6

b
apo-RORc-LBD. Ligand-lipophilicity efficiency (LLE) was calculated using the equation: (RORc SRC1 pEC50) –
LogD (pH 7.4).37-38 cBinding assay monitored the displacement of a fluorescent probe from the PPARγ-LBD.
The favorable profile of 2 led us to focus on the imidazo[1,5-a]pyridine core. We
explored the replacement of the carboxylic acid group in 2 with a range of amides, heterocyclic
carboxylic acid isosteres, and N-acylsulfonamides. These analogs were devoid of detectable
RORc inverse agonist activity at concentrations up to 10 µM. Due to the importance of the
4'-carboxylic acid group in maintaining RORc inverse activity, we sought to explore the
substitution and modification of the 4'-benzoic acid moiety (Table 2). Introduction of a phenol at
the 3'-position of the ring (9, GNE-0946) resulted in an approximate fivefold improvement in
RORc inverse agonist potency (EC50 = 2 nM) over 2, but 9 also had an increased PPARγ affinity
(IC50 = 650 nM). Introduction of fluorine at 2'-position on the arene (10) provided an analog
with similar RORc inverse agonist activity as 2 and an increased affinity for PPARγ
(IC50 = 320 nM). Similar improvements in RORc potency were also noted in the Organon
(Merck) patent application28 with the installation of these 3'-phenol and 2'-fluorine groups,
respectively, into analogs of 1.
Fluorine was the preferred 2'-position substituent as chlorine, cyano, and methyl groups
were >100-fold less potent in the RORc SRC1 biochemical assay. Synthesis of the cyclohexene
carboxylic acid analog (11) resulted in an approximate tenfold loss of RORc inverse agonist
activity as compared with 2. The cyclohexane carboxylic acid analogs (12 and 13) resulted in a
further loss of RORc inverse agonist potency, with the 1',4'-trans diastereomer (12) as the
preferred stereoisomer (EC50 = 400 nM). Introduction of a piperidine 4'-carboxylic acid (14)
provided a compound of similar RORc inverse agonist potency to 12, with no detectable affinity
for PPARγ up to concentrations of 10 µM. Various hydroxy-piperidine analogs were explored in
7

an attempt to restore some RORc potency, as similar changes (i.e., 2 to 9) have previously led to
a fivefold improvement in potency. Introduction of an alcohol at the 4'-position (15) led to a
tenfold loss of potency (EC50 = 6.8 µM) as compared with 14. Placement of the alcohol group at
the 3'-position led to improvements in potency, with the trans diastereomers being tenfold more
potent than the cis diastereomers. Separation of the trans diastereomers into the two separate
enantiomers with unknown absolute stereochemistry led to compounds 16 and 17. Compound
17 possessed an approximate fivefold improvement in RORc inverse agonist potency
(EC50 = 97 nM) as compared with 14.40 Compound 17 also had no detectable affinity for PPARγ
up to concentrations of 10 µM. Although the RORc potency, LLE, and selectivity profile of 17
was encouraging, the approximate tenfold difference in RORc potency as compared with 2 led us
to focus on the 4'-benzoic acid motif in subsequent analogs.
8

Table 2
SAR of the imidazo[1,5-a]pyridine 1-position
Compd
R
RORc SRC1 EC50a (μM) [%eff] LLEb LogD (pH 7.4) PPARγ IC50c (µM)
0.002 [-99%]
0.009 [-99%]
0.10 [-99%]
0.40 [-90%]
4.3 [-73%]
6.2
6.2
5.5
5.1
3.5
5.9
5.1
6.1
6.6
2.5
1.8
1.5
1.3
1.8
0.4
0
0.65
0.32
1.8
9
10
11
ND
ND
>10
ND
>10
>10
12
13
0.54 [-94%]
6.8 [-56%]
14
15
16d
17d
0.44 [-95%]
0.097 [-98%]
0.3
0.4
See the Supplementary Data for experimental details associated with each assessment. All assay results are reported
a
as the geometric mean of at least two separate runs. ND = not determined. Inhibition of RORc-LBD recruitment of
the SRC1 co-activator peptide; negative percent efficacy denotes inverse agonism relative to the basal activity of
b
apo-RORc-LBD. Ligand-lipophilicity efficiency (LLE) was calculated using the equation: (RORc SRC1 pEC50) –
LogD (pH 7.4).37-38 cBinding assay monitored the displacement of a fluorescent probe from the PPARγ-LBD.
dSingle enantiomers of the trans diastereomers with unknown absolute stereochemistry; the stereochemistry at the 3'-
and 4'-positions of the piperidine ring was arbitrarily assigned.
9

In parallel to our exploration of the SAR on the 1-(4'-benzoic acid) group, we also
evaluated a series of changes to the 3-acylarene moiety (Table 3). For this endeavor, we chose
the 5,6,7,8-tetrahydro-imidazo[1,5-a]pyridine core (similar to 3) as our prior studies26 had shown
that minimizing the aromatic ring count of RORc ligands can lead to improvements in selectivity
and ADME properties. Removal of the 2'-chlorine on the 3-acylarene of 18 led to an
approximate 25-fold loss of potency (19, EC50 = 0.84 µM). We explored other analogs that
replaced the 2',6'-disubstitution with substituents elsewhere on the 3-acylring and these
compounds had no detectable potency (up to a 10 µM concentration) in the RORc biochemical
assay. Thus, we focused our efforts on 2',6'-disubstituted arenes with the goal of minimizing the
overall lipophilicity of the analogs. Replacement of the 6'-(trifluoromethyl) group on the arene
with a range of substituents including fluorine (20), chlorine (21), cyano (22), and methoxy (23)
all led to a loss of RORc potency with no improvements in LLE relative to 18. The calculated
LogP (cLogP)39 value of the 2-chloro-6-cyclopropylarene analog (24) suggested that this
compound was less lipophilic than 18 (24 cLogP = 6.2, 18 cLogP = 6.4). The measured LogD
(pH 7.4) values indicated that 24 was slightly more lipophilic than 18 (24 LogD = 1.6, 18
LogD = 1.3) and 24 was a more potent RORc inverse agonist (EC50 = 27 nM). Additionally, 24
possessed an improved PPARγ selectivity profile in comparison to 18. Replacement of the
2'-chloro group on the 3-acylarene with cyano (25) or methoxy (26) substituents provided
analogs that were significantly less potent in the RORc SRC1 biochemical assay. We also
explored a range of ortho- and ortho/ortho-disubstituted heteroaromatic rings and only the
3',5'-dichloro-pyridine analog (27) demonstrated a detectable level of RORc inverse agonist
10

activity (EC50 = 0.51 µM). In sum, there appeared to be a strict requirement for lipophilic
2',6'-disubstitution on the 3-acylarene group.
Table 3
SAR of the 3-acylarene group
Compd
R
RORc SRC1 EC50a (μM) [%eff] LLEb LogD (pH 7.4) PPARγ IC50c (µM)
2'-Cl-6'-CF3-Ph
2'-CF3-Ph
0.032 [-96%]
0.84 [-96%]
0.19 [-99%]
0.077 [-99%]
0.25 [-99%]
0.24 [-98%]
0.027 [-99%]
0.22 [-99%]
0.67 [-96%]
0.51 [-98%]
6.2
4.7
5.5
5.7
5.9
5.8
5.9
5.9
-
1.3
1.4
1.2
1.5
0.7
0.8
1.6
0.7
ND
1.1
0.76
5.6
2.9
1.1
7.7
5.5
1.3
>10
9.1
4.2
18
19
20
21
22
23
24
25
26
27
2'-Cl-6'-F-Ph
2',6'-diCl-Ph
2'-Cl-6'-CN-Ph
2'-Cl-6'-MeO-Ph
2'-Cl-6'-(c-Pr)-Ph
2'-CN-6'-CF3-Ph
2'-OMe-6'-CF3-Ph
3',5'-diCl-pyridyl
5.2
See the Supplementary Data for experimental details associated with each assessment. All assay results are reported
a
as the geometric mean of at least two separate runs. ND = not determined. Inhibition of RORc-LBD recruitment of
the SRC1 co-activator peptide; negative percent efficacy denotes inverse agonism relative to the basal activity of
b
apo-RORc-LBD. Ligand-lipophilicity efficiency (LLE) was calculated using the equation: (RORc SRC1 pEC50) –
LogD (pH 7.4).37-38 cBinding assay monitored the displacement of a fluorescent probe from the PPARγ-LBD.
After exploring the core, 1-position substituents, and the optimal substitution of the
3-acylarene group, we prioritized the optimal substituents that possessed favorable potency and
selectivity for RORc over PPARγ. Combining these optimal substituents into one compound
(28, GNE-6468, Table 4) provided a highly potent RORc inverse agonist (EC50 = 2 nM) with
>1,000-fold selectivity for RORc over PPARγ. Compound 28 also had the highest LLE value
11

(LLE = 7.2) of all the compounds we profiled. We then tested the potent RORc inverse agonists
with favorable RORc selectivity over PPARγ in a series of ROR and NR cell-based assays to
further explore their cellular potencies and selectivity profiles.
Table 4
Combination of favored substituents into one analog
RORc SRC1
LogD
PPARγ
LLEb
7.2
EC50a (μM) [%eff]
(pH 7.4) IC50c (µM)
0.002 [-99%]
1.5 3.4
See the Supplementary Data for experimental details associated with each assessment. All assay results are reported
as the geometric mean of at least two separate runs. aInhibition of RORc-LBD recruitment of the SRC1 co-activator
peptide; negative percent efficacy denotes inverse agonism relative to the basal activity of apo-RORc-LBD.
bLigand-lipophilicity efficiency (LLE) was calculated using the equation: (RORc SRC1 pEC50) – LogD
(pH 7.4).37-38 cBinding assay monitored the displacement of a fluorescent probe from the PPARγ-LBD.
We examined 2, 9, 10, 18, 24, and 28 in a series of HEK-293 cell Gal4-ROR construct
human NR-dependent transcriptional reporter assays (Table 5). We profiled the three isoforms
of human ROR (RORc, RORb, and RORa) by monitoring the suppression of their basal
transcriptional activity in the absence of any exogenous agonist.29 In order to assess the NR
cellular selectivity of the potent RORc inverse agonists, we also tested these compounds in a
panel of cellular reporter assays of human farnesoid X receptor (FXR), liver X receptor (LXR)-α,
LXRβ, and PXR in both agonist mode (no agonist ligand added) and antagonist mode (using
12

T0901317
[N-(2,2,2-trifluoroethyl)-N-[4-[2,2,2-trifluoro-1-hydroxy-
1-(trifluoromethyl)ethyl]phenyl]-benzenesulfonamide] as an exogenous ligand).29 All of these
compounds demonstrated favorable RORc inverse agonist cellular potency values
(EC50 = 4-36 nM), with no detectable activity against the other ROR family members (up to
10 µM concentration). In addition, none of these compounds demonstrated detectable activity
against the other NRs in our cell assay panel in agonist or antagonist mode (up to 10 µM
concentration). These results demonstrated the excellent cellular selectivity profiles of 2, 9, 10,
18, 24, and 28 for RORc over other NRs (>277- to >1,000-fold selectivity).
Table 5
RORc potency and selectivity profiles in Gal4 human transcription assaysa
RORc Cell RORb Cell RORa Cell
FXR Cell
EC50 (µM)
LXRα Cell LXRβ Cell
PXR Cell
EC50 (µM)
Compd
EC50 (µM)
EC50 (µM)
EC50 (µM)
EC50 (µM)
EC50 (µM)
0.013
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
2
0.004
>10
>10
>10
>10
9
0.010
>10
>10
>10
>10
10
18
24
28
0.036
>10
>10
>10
>10
0.026
>10
>10
>10
>10
0.013
>10
>10
>10
>10
See the Supplementary Data for experimental details associated with each assessment. All assay results are reported
as the geometric mean of at least two separate runs. aAll assays were conducted in HEK293-Gal4 cellular constructs.
All NR assays monitored the suppression of their respective basal transcriptional activities, an outcome consistent
with inverse agonist activity of ligands with these receptors.
Based on their favorable RORc cell potency values, we progressed 2, 9, 10, 18, 24, and
28 into human PBMC cytokine production assays29 to assess their abilities to inhibit the
production of IL-17 (Table 6). Compounds 2, 10, 18, and 24 displayed modest inhibition of
IL-17 production in the human PBMC assay (EC50 = 230, 120, 230, and 300 nM, respectively),
13

whereas 9 and 28 displayed more potent inhibition of IL-17 production (EC50 = 17 and 30 nM,
respectively). Compounds 9 and 28 possessed a unique substructure in contrast to 2, 10, 18, and
24, indicating that the 1-(3'-hydroxy-4'-benzoic acid) motif present in 9 and 28 may engender
enhanced potency in the IL-17 production assay. It was also noteworthy that none of the
compounds showed any activity in interferon (IFN)-γ or CellTiter-Glo® (CTG) counter screen
assays, demonstrating that the compounds were not indiscriminately suppressing cytokine
production, nor were they grossly cytotoxic.
Compounds 9 and 28 were potent inhibitors of IL-17 production in human PBMC cells
(EC50 values ≤ 30 nM, Table 6), yet they were no more effective in further suppressing the
maximum percent inhibition of IL-17 production (%max) than compounds that were an order of
magnitude less potent (i.e. 2, 10, 18, and 24). Indeed, we have noted in our previously-disclosed
human PBMC data with several RORc inverse agonist chemotypes that the IL-17 PBMC %max
values average 77% inhibition (±23% of the %max inhibition value for the 13 different human
IL-17 PBMC assay results we have disclosed where the IL-17 PBMC
EC50 = 0.044-3.0 µM)24-26,29 The lack of complete suppression of IL-17 production in human
PBMCs (i.e. IL-17 PBMC %max ≥ 99%) may be due to the role of RORa in the production of
IL-17. Previous mouse genetic knock-out studies have demonstrated that the production of
murine IL-17 is dependent on both murine RORα and RORγ.41-42 Potent and selective human
RORa inverse agonists would be required to explore this hypothesis in the context of human
IL-17 production. The deficiency of potent and selective human RORa tool compounds is the
focus of several research programs.21-22
14

Table 6
Potency in human IL-17 and IFNγ production assaysa
IL-17
PBMC
EC50
(µM)
IL-17
PBMC
%max.
IFNγ
EC50
(µM) (µM)
CTG
EC50
Compd
inhibition
0.23
0.017
0.12
69%
70%
63%
64%
72%
73%
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
2
9
10
18
24
28
0.23
0.30
0.030
See the Supplementary Data for experimental details associated with each assessment. All assay results
a
are reported as the mean of at least three separate runs. All assays were conducted using peripheral
blood mononuclear cells (PBMCs) isolated from human whole blood. Interferon gamma (IFNγ) and
CellTiter-Glo® (CTG) cell culture assays were used as positive controls to monitor for non-TH17 cell
cytokine activity and adverse off-target effects on cell physiology, respectively.
We also evaluated 9 and 28 in a suite of in vitro ADME assays (Table 7).43-44 Compound
9 displayed moderate predicted hepatic clearance (CLhep) values in human and rodent hepatocyte
metabolic stability assays (CLhep = 9 and 32 mL/min/kg, respectively), whereas 28 displayed
high predicted clearance values in human and rodent hepatocytes (CLhep = 17 and 42 mL/min/kg,
respectively). Further profiling of the compounds in human and rat plasma-protein binding
(PPB) assays demonstrated that 9 and 28 were highly protein bound in both species
(%bound = >99% and 99%, respectively). Compounds 9 and 28 displayed favorable apparent
permeability (Papp(A→B) = 12 and 8 × 10-6 cm/s, respectively) with minimal efflux
(Papp(A→B)/Papp(B→A) < 2) in a Madin-Darby canine kidney (MDCK) cellular permeability assay.45
Both 9 and 28 were highly soluble as assessed by a kinetic aqueous solubility assay (45-86 µM).
Thus, 9 and 28 were highly potent and selective RORc inverse agonists with favorable properties
and moderate-to-high clearance values.
15

Table 7
In vitro ADME profiles
Hu Hepa
CLhep
(mL/min/kg) (mL/min/kg)
Rat Hepb
CLhep
Human
PPB
(%bound)
Rat
PPB
(%bound)
MDCK
MDCK
Solubilityd
(µM)
Compd
Pappc A→B Pappc B→A
(10-6 cm/s) (10-6 cm/s)
9
32
42
>99
99
>99
99
12
8
10
15
86
45
9
17
28
a
See the Supplementary Data for experimental details associated with each assessment. Predicted human hepatic
b
clearance values extrapolated from in vitro human hepatocyte (Hu Hep) metabolic stability experiment. Predicted
rat hepatic clearance values extrapolated from in vitro rat hepatocyte (Rat Hep) metabolic stability experiment.
cMadin-Darby canine kidney (MDCK) cell permeability assay to assess membrane permeability (Papp); A→B:
apical-to-basolateral, B→A: basolateral-to-apical.45 dAqueous kinetic solubility at pH 7.4 (measured in a high-
throughput assay).
Compound 9 was profiled in a rodent in vivo pharmacokinetic (PK) experiment (Table 8)
to determine if the in vitro rat hepatic clearance estimate (Table 7) was predictive of its in vivo
clearance value. In a rat PK experiment, 9 demonstrated a high plasma clearance value
(CLp = 130 mL/min/kg) that did not correlate with its in vitro clearance value (Rat Hep
CLhep = 32 mL/min/kg). The in vivo plasma clearance of 9 was in excess of rodent liver blood
flow (55 mL/min/kg), indicating that extrahepatic clearance mechanisms46 may also be
participating in the metabolism of 9.
A very recent set of patent applications from Glenmark Pharmaceuticals47 describe two
series of compounds related to 1. It is unknown if the compounds in the Glenmark patent
applications possess similar metabolic stabilities as those observed with 9 and 28.
16

Table 8
Single dose rat in vivo PK profile of 9
a
b
Compd
CLp (mL/min/kg)
Vd (L/kg)
Cmaxc (μM)
AUC (μM*h)
t1/2 (h)
F%d
130
1.2
0.07
0.25
0.2
40%
9
See the Supplementary Data for experimental details associated with each assessment. Data reported are the
arithmetic means from the dosing cohorts (Sprague-Dawley rats, n = 3/dose). Dosed at 1.5 mg/kg po (37/63
a
suspension of DMSO/MCT) and 0.5 mg/kg iv (25/60/15 solution of DMSO/PEG400/saline). Observed plasma
b
c
d
clearance (CLp). Volume of distribution (Vd). Maximum plasma concentration (Cmax). Oral bioavailability (F%)
was calculated according to the equation F% = (dose normalized AUCpo)/(dose normalized AUCiv). CLp, Vd, and t1/2
were derived from an iv study and Cmax, AUC, and F% were derived from a po study.
In conclusion, we evolved a literature series of RORc inverse agonists (i.e. 1) into a series
of imidazo[1,5-a]pyridine and -pyrimidines with improved selectivity for RORc over PPARγ.
Several compounds within the imidazo[1,5-a]pyridine and -pyrimidine series demonstrated
potent RORc inverse agonist activity in biochemical and cellular assays. The most potent
compounds also displayed >300-fold selectivity for RORc over the other ROR family members
and NRs in our cellular selectivity panel. The favorable potency, selectivity, and physiochemical
properties of GNE-0946 (9) and GNE-6468 (28), in addition to their potent suppression of IL-17
production in human primary cells, support their use as chemical biology tools to further explore
the role of RORc in human biology.48
Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://INSERT URL HERE
17

Acknowledgements
We thank Drs. Krista Bowman and Jiansheng Wu, and their respective Genentech
research groups, for performing all required protein expression and purification activities.
18

References and Notes
1) Langley, R. G.; Elewski, B. E.; Lebwohl, M.; Reich, K.; Griffiths, C. E. M.; Papp, K.;
Luig, L.; Nakagawa, H.; Spelman, L.; Sigrgeirsson, B.; Rivas, E.; Tsai, F.-F.; Wasel, N.;
Tyring, S.; Salko, T.; Hamele, I.; Notter, M.; Karpov, A.; Helou, S.; Papavassilis, C. N.
Engl. J. Med. 2014, 371, 326.
2) Papp, K. A.; Leonardi, C.; Menter, A.; Ortonne, J. P.; Krueger, J. G.; Kricorian, G.; Aras,
G.; Li, J.; Russell, C. B.; Thompson, E. H.; Baumgartner, S. N. Engl. J. Med. 2012, 366,
1181.
3) Leonardi, C.; Matheson, R.; Zachariae, C.; Cameron, G.; Li, L.; Edson-Heredia, E.;
Braun, D.; Banerjee, S. N. Engl. J. Med. 2012, 366, 1190.
4) Kellner, H. Ther. Adv. Musculoskel. Dis. 2013, 5, 141.
5) Baeten, D.; Baraliakos, X.; Braun, J.; Sieper, J.; Emery, P.; van der Heijde, D.; McInnes,
I.; van Laar, J. M.; Landewé, R.; Wordsworth, P.; Wollenhaupt, J.; Kellner, H.;
Paramarta, J.; Wei, J.; Brachat, A.; Bek, S.; Laurent, D.; Li, Y.; Wang, Y.; Bertolino, A.
P.; Gseiger, S.; Wright, A. M.; Hueber, W. Lancet 2013, 382, 1705.
6) Hueber, W.; Patel, D. D.; Dryja, T.; Wright, A. M.; Koroleva, I.; Bruin, G.; Antoni, C.;
Draelos, Z.; Gold, M. H.; Durez, P.; Tak, P. P.; Gomez-Reino, J. J.; Foster, C. S.; Kim, R.
Y.; Samson, C. M.; Falk, N. S.; Chu, D. S.; Callanan, D.; Nguyen, Q. D.; Rose, K.;
Haider, A.; Di Padova, F. Sci. Transl. Med. 2010, 2, 52ra72.
7) Griffiths, C. E. M.; Strober, B. E.; van de Kerkhof, P.; Ho, V.; Fidelus-Gort, R.; Yeilding,
N.; Guzzo, C.; Xia, Y.; Zhou, B.; Li, S.; Dooley, L. T.; Goldstein, N. H.; Menter, A. N.
Engl. J. Med. 2010, 362, 118.
19

8) FDA
approves
new
psoriasis
drug
Cosentyx.
http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm430969.htm
(accessed Feb 9, 2015).
9) Bartlett, H. S.; Million, R. P. Nat. Rev. Drug Discov. 2015, 14, 11.
10) Pappu, R.; Ramirez-Carrozzi, V.; Ota, N.; Ouyang, W.; Hu, Y. J. Clin. Immunol. 2010,
30, 185.
11) Hirose, T.; Smith, R. J.; Jetten, A. M. Biochem. Biophys. Res. Commun. 1994, 205, 1976.
12) Ivanov, I. I.; McKenzie, B. S.; Zhou, L.; Tadokoro, C. E.; Lepelley, A.; Lafaille, J. J.;
Cua, D. J.; Littman, D. R. Cell 2006, 126, 1121.
13) Sabat, R.; Ouyang, W.; Wolk K. Nat. Rev. Drug Discov. 2014, 13, 21.
14) Cording, S.; Medvedovic, J.; Cherrier, M.; Eberl, G. FEBS Letters 2014, 588, 4176.
15) Eberl, G.; Marmon, S.; Sunshine, M.-J.; Rennert, P. D.; Choi, Y.; Littman, D. R. Nat.
Immunol. 2004, 5, 64.
16) Pantelyushin, S.; Haak, S.; Ingold, B.; Kulig, P.; Heppner, F. L.; Navarini, A. A.; Becher,
B. J. Clin. Invest. 2012, 122, 2252.
17) Williams, S. C. P. Nat. Med. 2013, 19, 1078.
18) Yang, J.; Sundrud, M. S.; Skepner, J.; Yamagata, T. Trends Pharmacol. Sci. 2014, 35,
493.
19) Isono, F.; Fujita-Sato, S.; Ito, S. Drug Disc. Today 2014, 19, 1205.
20) Fauber, B. P.; Magnuson, S. J. Med. Chem. 2014, 57, 5871.
21) Kojetin, D. J.; Burris, T. P. Nat. Rev. Drug Discov. 2014, 13, 197.
22) Kamenecka, T. M.; Lyda, B.; Chang, M. R.; Griffin, P. R. Med. Chem. Commun. 2013, 4,
764.
20

23) Dhar, T. G. M.; Zhao, Q.; Markby, D. W. Annu. Rep. Med. Chem. 2013, 48, 169.
24) René, O.; Fauber, B. P.; Boenig, G.; Burton, B.; Eidenschenk, C.; Everett, C.; Gobbi, A.;
Hymowitz, S. G.; Johnson, A. R.; Kiefer, J. R.; Liimatta, M.; Lockey, P.; Norman, M.;
Ouyang, W.; Wallweber, H. A.; Wong, H. ACS Med. Chem. Lett. 2015, 6, 276.
25) van Niel, M. B.; Fauber, B. P.; Cartwright, M.; Gaines, S.; Killen, J. C.; René, O.; Ward,
S. I.; Boenig, G.; Deng, Y.; Eidenschenk, C.; Everett, C.; Gancia, E.; Ganguli, A.; Gobbi,
A.; Hawkins, J.; Johnson, A. R.; Kiefer, J. R.; La, H.; Lockey, P.; Norman, M.; Ouyang,
W.; Qin, A.; Wakes, N.; Waszkowycz, B.; Wong, H. Bioorg. Med. Chem. Lett. 2014, 24,
5769.
26) Fauber, B. P.; René, O.; Boenig, G.; Burton, B.; Deng, Y.; Eidenschenk, C.; Everett, C.;
Gobbi, A.; Hymowitz, S. G.; Johnson, A. R.; La, H.; Liimatta, M.; Lockey, P.; Norman,
M.; Ouyang, W.; Wang, W.; Wong, H. Bioorg. Med. Chem. Lett. 2014, 24, 3891.
27) Fauber, B. P.; René, O.; Burton, B.; Everett, C.; Gobbi, A.; Hawkins, J.; Johnson, A. R.;
Liimatta, M.; Lockey, P.; Norman, M.; Wong, H. Bioorg. Med. Chem. Lett. 2014, 24,
2182.
28) Karstens, W. F. J.; van der Stelt, M.; Cals, J.; Azevedo, R. C. R. G.; Barr, K. J.; Zhang,
H.; Beresis, R. T.; Zhang, D.; Duan, X. WO 2012/106995, Aug 16, 2012.
29) See the Supplementary Data section of the following reference for the detailed
biochemical and cellular assay protocols: Fauber, B. P.; Boenig, G.; Burton, B.;
Eidenschenk, C.; Everett, C.; Gobbi, A.; Hymowitz, S. G.; Johnson, A. R.; Liimatta, M.;
Lockey, P.; Norman, M.; Ouyang, W.; René, O.; Wong, H. Bioorg. Med. Chem. Lett.
2013, 23, 6604.
21

30) (a) Acton, J. J.; Akiyama, T. E.; Chang, C. H.; Colwell, L.; Debenham, S.; Doebber, T.;
Einstein, M.; Liu, K.; McCann, M. E.; Moller, D. E.; Muise, E. S.; Tan, Y.; Thompson, J.
R.; Wong, K. K.; Wu, M.; Xu, L.; Meinke, P. T.; Berger, J. P.; Wood, H. B. J. Med.
Chem. 2009, 52, 3846. (b) Acton, J. J.; Black, R. M; Jones, A. B.; Moller, D. E.; Colwell,
L.; Doebber, T. W.; MacNaul, K. L.; Berger, J.; Wood, H. B. Bioorg. Med. Chem. Lett.
2005, 15, 357.
31) The human PPARγ-LBD fluorescence polarization binding assay was conducted using
the Life Technologies (ThermoFisher Scientific Corp., Grand Island, NY)
“Polar-ScreenTM PPAR Gamma Competitor Assay, Green” assay kit (catalog number
PV6136).
For
additional
details
on
this
assay,
see:
https://tools.lifetechnologies.com/content/sfs/manuals/Polar_Screen_PPARgamma_comp
etitor_Assay_Green_PI.pdf (accessed Feb 12, 2015).
32) Sohda, T.; Momose, Y.; Meguro, K.; Kawamatsu, Y.; Sugiyama, Y.; Ikeda, H.
Arzneim.-Forsch. 1990, 40, 37.
33) Fujiwara, T.; Yoshioka, S.; Yoshioka, T.; Ushiyama, I.; Horikoshi, H. Diabetes 1988, 37,
1549.
34) Graham, D. J.; Ouellet-Hellstrom, R.; MaCurdy, T. E.; Ali, F.; Sholley, C.; Worrall, C.;
Kelman, J. A. JAMA 2010, 304, 411.
35) Liu, K.; Black, R. M.; Acton, J. J.; Mosley, R.; Debenham, S.; Abola, R.; Yang, M.;
Tschirret-Guth, R.; Colwell, L.; Liu, C.; Wu, M.; Wang, C. F.; MacNaul, K. L.; McCann,
M. E.; Moller, D. E.; Berger, J. P.; Meinke, P. T.; Jones, B.; Wood, H. B. Bioorg. Med.
Chem. Lett. 2005, 15, 2437.
22

36) For representative synthetic protocols and characterization of the analogs, see Fauber, B.
P.; Gobbi, A. WO 2015/036411, Mar 19, 2015. The synthesis and characterization of 9
(GNE-0946) and 28 (GNE-6468) are described in the Supplementary Data section of this
manuscript.
37) Leeson, P. D.; Springthorpe, B. Nat. Rev. Drug Discov. 2007, 6, 881.
38) Leach, A. R.; Hann, M. M.; Burrows, J. N.; Griffen, E. J. Mol. BioSyst. 2006, 2, 429.
39) The partition coefficient (LogP) was calculated with internal software using the VolSurf
approach. For additional details, see Cruciani, G.; Crivori, P.; Carrupt, P.-A.; Testa, B.
THEOCHEM 2000, 503, 17.
40) A similar improvement in potency was noted in a recent 1-acyl-indazole patent
application from Merck. For additional details, see Barr, K. J.; MacLean, J.; Zhang, H.;
Beresis, R. T. WO 2014/028600, Feb 20, 2014.
41) Sundrud, M. S.; Rao, A. Immunity, 2008, 28, 5.
42) Yang, X. O.; Pappu, B. P.; Nurieva, R.; Akimzhanov, A.; Kang, H. S.; Chung, Y.; Ma,
L.; Shah, B.; Panopoulos, A. D.; Schluns, K. S.; Watowich, S. S.; Tian, Q.; Jetten, A. M.;
Dong, C. Immunity 2008, 28, 29.
43) In vitro assays to explore the themes of cellular permeability and predicted metabolic
clearance have been referred to as in vitro ADME assays. For additional discussion, see
Tarbit, M. H.; Berman, J. Curr. Opin. Chem. Biol. 1998, 2, 411.
44) See the Supplementary Data section of Ref. 26 for the in vitro ADME assay protocols.
45) Irvine, J. D.; Takahashi, L.; Lockhart, K.; Cheong, J.; Tolan, J. W.; Selick, H. E.; Grove,
J. R. J. Pharm. Sci. 1999, 88, 28.
46) Krishna, D. R.; Klotz, U. Clin. Pharmacokinet. 1994, 26, 144.
23

47) (a) Das, S.; Thomas, A.; Khairatkar-Joshi, N.; Shah, D. M.; Bajpai, M. WO 2015/052675,
Apr 16, 2015. (b) Chaudhari, S. S.; Thomas, A.; Shone, S. V.; Khairatkar-Joshi, N.;
Bajpai, M. WO 2015/008234, Jan 22, 2015.
48) GNE-0946 (9) and GNE-6468 (28) can be requested through the Genentech Research
Contracts
and
Reagents
Program
(https://www.gene.com/gene/reagents-
program/request.do).
24

Graphical Abstract.