Novel highly selective peroxisome proliferator-activated receptor δ (PPARδ) modulators with pharmacokinetic properties suitable for once-daily oral dosing

Article abstract of DOI:10.1016/j.bmcl.2017.10.037

Optimization of benzamide PPARδ modulator 1 led to (E)-6-(2-((4-(furan-2-yl)-N-methylbenzamido)methyl)phenoxy)-4-methylhex-4-enoic acid (18), a potent selective PPARδ modulator with significantly improved exposure in multiple species following oral administration.

Full text of DOI:10.1016/j.bmcl.2017.10.037

Accepted Manuscript
Novel highly selective peroxisome proliferator-activated receptor δ (PPARδ)
modulators with pharmacokinetic properties suitable for once-daily oral dosing
Bharat Lagu, Arthur F. Kluge, Ross A. Fredenburg, Effie Tozzo, Ramesh S.
Senaiar, Mahaboobi Jaleel, Sunil K. Panigrahi, Nirbhay K. Tiwari, Narasimha
R. Krishnamurthy, Taisuke Takahashi, Michael A. Patane
PII:
S0960-894X(17)31031-4
https://doi.org/10.1016/j.bmcl.2017.10.037
BMCL 25366
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
21 August 2017
26 September 2017
18 October 2017
Please cite this article as: Lagu, B., Kluge, A.F., Fredenburg, R.A., Tozzo, E., Senaiar, R.S., Jaleel, M., Panigrahi,
S.K., Tiwari, N.K., Krishnamurthy, N.R., Takahashi, T., Patane, M.A., Novel highly selective peroxisome
proliferator-activated receptor δ (PPARδ) modulators with pharmacokinetic properties suitable for once-daily oral
dosing, Bioorganic & Medicinal Chemistry Letters (2017), doi: https://doi.org/10.1016/j.bmcl.2017.10.037
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Novel highly selective peroxisome proliferator-activated receptor δ
(PPARδ) modulators with pharmacokinetic properties suitable for
once-daily oral dosing
Bharat Lagu,a* Arthur F. Kluge,a Ross A. Fredenburg,a Effie Tozzo,a Ramesh S. Senaiar,b
Mahaboobi Jaleel,b Sunil K. Panigrahi,b Nirbhay K. Tiwari,b Narasimha R. Krishnamurthy,b
Taisuke Takahashi,c and Michael A. Patanea
aMitobridge, Inc., 1030 Massachusetts Ave., Cambridge, Massachusetts 02138, United States
bAurigene Discovery Technologies, Ltd., Hyderabad – 500049, Telangana, India
cAstellas Pharma, Tsukuba, Japan
*e-mail: blagu@mitobridge.com
ABSTRACT
Optimization
of
benzamide
PPARδ
modulator
1
led
to
(E)-6-(2-((4-(furan-2-yl)-N-
methylbenzamido)methyl)phenoxy)-4-methylhex-4-enoic acid (18), a potent selective PPARδ modulator with
significantly improved exposure in multiple species following oral administration.
Keywords:
PPARδ modulator; GW501516; plasma exposure, ligand binding domain, x-ray structure.
Three known isoforms of a family of nuclear receptors called peroxisome proliferator-
activated receptors (PPARs) manage the biosynthesis, oxidation, transport and storage of lipids
by governing gene regulation of mRNAs responsible for proteins that coordinate these
activities.1 Modulators of PPARα (fibrates) and PPARγ (thiazolidinones) have been approved as
treatments for dyslipidemia and diabetes, respectively. PPARδ is highly expressed in liver,
skeletal muscle, intestine and adipose tissue.2 Therefore, selective PPARδ modulators could
potentially be useful treatments for metabolic disorders and conditions that would benefit from
muscle regeneration.3,4
Recently, Evans and co-workers have disclosed a chemical series of potent selective PPARδ
modulators, such as compound 1 (Figure 1)5 which are structurally distinct from the known
PPARδ modulator, GW501516.6 Herein, we describe structural modifications of 1 that were
intended to improve exposure following oral dosing. This effort culminated in the identification
of a potent selective PPARδ modulator, compound 18, which is suitable for further in vivo
characterization.
1

Compound 1 was found to have high clearance in mice and served as a starting point for
medicinal chemistry.7,8 We modified regions of its structure based on in vivo metabolism studies
and PPARδ receptor data. Through iterative optimization using structure-activity relationships
(SAR),compound 18 was identified; it displayed significantly improved in vivo exposure
(following i.v. and p.o. dosing) when compared to compound 1. The SAR, pharmacokinetic data
in multiple species and the protein bound x-ray structure of 18 are presented.
Figure 1. PPARδ modulators
A general synthesis for the compounds in Table 1 is shown in Scheme 1.9 Reductive
amination of salicylaldehyde with methylamine hydrochloride provided 2-((methylamino)
methyl)phenol 2. Suzuki coupling of iodobenzoic acid with furan-2-yl boronic acid produced 4-
(furan-2-yl)benzoic acid 3. The coupling of 2 and 3 led to amide 4, which after alkylation
reaction with a 6-bromohexanoic ester or a 6-bromohexenoic ester 5 yielded the corresponding
ester 6. Hydrolysis of these esters formed the corresponding carboxylic acids 7 - 18.
2

Scheme1. General synthesis of compounds in Table 1
As shown in Figure 2, the benzamide series as exemplified by the structure of compound
1 has been divided into three regions (“head”, “core” and “tail”) for discussion purpose. Oral
dosing of 1 in mice revealed that the clearance rate of 1 (195 mL/min/kg) exceeded the rate of
mouse hepatic blood flow (90 mL/min/kg).10,11 Therefore, 1 was unsuitable for in vivo mouse
efficacy studies and further optimization was necessary in order to identify a better candidate for
pharmacological evaluation. Incubation of 1 with rat liver microsomes revealed multiple
metabolic sites (Figure 2) that included oxidation in the aromatic groups in the “tail” and “core”
region, N-dealkylation and hydrolysis of the amide bond. Incubation of other compounds in this
series with rat and human hepatocytes showed acyl glucoronidation as a route for metabolism.12
Noticing that the trend for clearance rate was similar in rat and mouse, we screened the new
analogs for mouse i.v. exposure.13 Later, exposure following oral and i.v. dosing in mice and
rats were measured for the more interesting compounds.
3

We initially attempted to decrease clearance by modifying the furan in the “tail” region or
the phenyl group in the “core” region (Figure 2) by: 1) introducing substituents on the aryl rings
and 2) replacing the furan ring with cyclopropylethynyl, trifluoromethylethynyl and 4-
trifluoromethoxyphenyl groups. However, while these analogs maintained PPARδ potency (data
not shown), the clearance rate was unaffected. Serendipitously, we discovered that the N-methyl
amide, compound 7, had a decreased clearance rate compared to the clearance rate for 1 (44
mL/min/kg versus 195 mL/min/kg) after i.v. dosing in mice. Based on this finding, we kept the
N-methyl amide and the 2-furyl group (“tail” region) constant and systematically modified the
“core” and “head” regions. While the “core” region modifications did not affect the clearance
rate (data not shown), the “head” region modifications significantly improved the PK properties
of the series. The results are summarized in Table 1.
Figure 2. Sites for Phase I and Phase II metabolism of 1.
4

Table 1. SAR for PPARδ activity and i.v. exposure data from studies with CD-1 mice.
“Head” Region
Modification
t
CL
AUC
PPARδ
PPARδ
½
Cpd
7
(hr)c (mL/min/kg) (ng*hr/mL)
EC50 (nM)a EC50 (nM)b
13
47
1.7
2.8
44
3
1100
O
O
O
HO
HO
230
640
11400
8
9
Me
O
510
>10,000
4.6
1.3
48400
Me Me
O
Me
O
O
270
417
3.9
2.0
0.3
1.3
77700
30800
10
11
HO
HO
O
Me
>10,000
>10,000
12d
270
330 10
2.7
1.7
30100
13d
14
210
80
1630 880
NA
3.5
1.9
0.6
67
40500
750
O
O
HO
5

1850
690
>10,000
>10,000
1.2
1.6
4
1300
4700
15
16
O
O
Me
O
O
11
HO
HO
40
990
1.7
4
11800
17
Me
O
O
HO
5
37 5
2.3
15
3300
18
Me
The numbers are n=1 unless SEM is shown. aProtein interaction assay; bTransactivation assay. cCompounds dosed at 3 mg/kg;
dRelative stereochemistry determined by 1H NMR for racemic compound. NA = Not Available.
The PPARδ potency was determined in two assays, a protein interaction assay14 and a trans-
activation assay.15 Typically lower EC50 numbers (2 - 8 fold) were obtained in the protein
interaction assay as compared to the trans-activation assay, but the rank order of activities was
similar in both assays. Compounds 19 and 20 contain etheryl and amido linkers, which was
associated with loss of PPARδ activity and suggests that there is a lack of tolerance for polar
groups in the “head” region. Therefore, non-polar groups were incorporated at the α, β, γ or δ
positions of the hexanoic acid “head” region in the subsequent analogs. Analogs with
substitution at the α and β positions of the carboxylic acid (8 - 13) displayed significantly
improved half-lives, clearances and exposures. This observation may be related to the impact of
these substituents on β-oxidation and/or glucuronide formation of the carboxylic acid, which
could be principle component(s) of the high clearance observed for this series. However, the
improvements in the PK parameters for 8 - 13 came with diminished PPARδ potency compared
to 7. One analog, the α,β-unsaturated compound 14, showed only 6-fold loss in PPARδ potency
as compared to 7. The methyl substituent on the double bond at α or β position was associated
with improved clearance and exposure values for compounds 15 and 16; however, these analogs
also exhibited significantly lower PPARδ potency. Incorporating a methyl group in the γ-
position led to 17, which showed significantly reduced clearance and improved exposure with a
small loss in PPARδ potency. Compound 18, which combined the γ-methyl group and a γ,δ
6

trans double bond maintained PPARδ activity and improving clearance and exposure. The
plasma exposure after intravenous dosing in mice for all the compounds with EC50 under 100nM
are illustrated graphically in Figure 3.
Plasma exposure for selected compounds @3mpk i.v. dosing
100000
10000
1000
Compound 17
100
Compound 18
Compound 7
Compound 1
Compound 14
10
1
0
1
2
3
4
Time (h)
Figure 3: Plasma exposure for compounds with EC50 <100 nM after 3 mg/kg i.v. dosing in CD-1 mice.
Thus, compound 18 became a pivotal compound for the program and was profiled in more
detail as summarized in Table 2. Compound 18 displayed good ADME profile and good
exposure in mice, rats and monkeys following oral dosing.
7

Table 2. Potency, selectivity, ADME, DMPK, and PK data for 18
Assay
Results
EC50 = 5 nMa; EC50 = 37 5 nMb
EC50 >10,000 nMa; EC50 = 6100 nMb
EC50 >10,000 nMa,b
Human PPARδ
Human PPARα
Human PPARγ
Direct binding to hPPARδ by SPR16
Thermodynamic solubility
Caco-2 permeability
KD = 57 1 nM
190 µM
A to B = 4.58E-05; B to A= 1.03E-04 (Efflux ratio 2.24)
CYP450 inhibition
>10 µM for CYPs 3A4, 2C9, 2C19, 2D6, 1A2
1 mg/kg IV: t½ = 5.8 h; CL = 8.4 mL/min/kg; Vss= 2.6 L/kg ;
AUC = 1,980 ng*h/mL
10 mg/kg PO: t½ = 4.3 h; Cmax = 11,100 ng/mL; AUC =
16,600 ng*h/mL; %F = 84
3 mg/kg IV: t½ = 4.5 h ; Vss= 0.9 L/kg ;CL = 7.2 mL/min/kg;
AUC = 4,650 ng*h/mL
Mouse PKc
Rat PKd
10 mg/kg PO: t½ = 2.5h; Cmax = 3,805 ng/mL; AUC = 12,500
ng*h/mL; %F = 82
3 mg/kg IV: t½= 1.8 h ; CL = 8 mL/min/kg; Vss= 1.3 L/kg ;
AUC = 6,200 ng*h/mL
Monkey PKe
10 mg/kg PO: t½= 1.1 h: Cmax = 4,190 ng/mL; AUC = 6,050
ng*h/mL; %F = 29
aProtein interaction assay; bTransactivation assay; cMale CD1 mice (n = 3); dMale Wistar rats (n = 3); eMale
cynomolgus monkeys (n = 3).
An x-ray crystal structure of 18 bound to the ligand binding domain (LBD) of the
hPPARδ receptor was obtained with resolution of 2.0Å.17 Compound 18 was tightly integrated
with various amino acids throughout the LBD, totaling a series of twenty contacts with residues
within 4Å. As shown in Figure 3, it formed H-bond interactions of its carboxylic acid with side
chains of His287, His413, Thr253 and Tyr437, and proceeded to curl around helix 3 similar to
the reported structure of 1.5 The benzamide carbonyl was observed to be 34⁰ out of the plane of
the benzene ring. However, it appears that loss of conjugation with the benzene might have been
8

compensated for by a hydrogen bonding interaction of the carbonyl oxygen with a water
molecule that mediated interactions between it and the carbonyl of Leu304. Additional
hydrophobic contacts were seen with hydrophobic residues such as Leu294, Phe291, Leu303,
and Leu304. The hydrophobic contacts with Val245, Val305, Val312, Phe316 and Leu317 were
strong at the entrance of the Y-shaped binding pocket near the furan moiety.
Figure 3. X-ray structure of compound 18 (yellow) bound to LBD of PPARδ
In summary, we successfully optimized the pharmacokinetic properties of the benzamide
carboxylic acid PPARδ modulator 1, which led to identification of compound 18. The in vivo
efficacy, gene regulation profile and safety studies for 18 are discussed in an accompanying
paper.18 The protein bound structure of 18 enabled further refinements of the structure and
properties of 18, which will be presented in due course.
Acknowledgements
The authors thank Dr.s Masanori Miura and Susumu Watanuki (Astellas Pharma.), Susanta Samajdar and
Chetan Pandit (Aurigene) for helpful discussions.
References and Notes
1. Pirat, C.; Farce, A.; Lebègue, N.; Renault, N.; Furman, C.; Millet, R.; Yous, S.; Speca,S.; Berthelot,
P.; Desreumaux, P.; Chavatte, P. J. Med. Chem. 2012, 55, 4027
2. Girroir, E. E,; Hollingshead, H. E,; He, P,; Zhu, B,; Perdew, G. H,; Peters, J. M. Biochem. Biophys.
Res. Commun. 2008, 371, 456.
9

3. Barish, G. D.; Narkar, V. A,; Evans, R. M. J Clin Invest 2006, 116, 590.
4. Fan, W.; Waizenegger, W.; Lin, C. S.;Sorrentino,V.; He, M.; Wall, C. E.; Li, H.; Christopher Liddle,
C.; Yu, R. T.; Atkins, A. R.; Auwerx, J.; Downes,M.;Evans, R. M. Cell Metabolism 2017, 25, 1186.
5. Wu, C.-C.; Baiga, T. J.; Downes, M.; La Clair, J. J.; Atkins, A. R.; Richard, S. B.; Fan, W.; Stockley-
Noel, T. A.; Bowman, M. E.; Noel, J. P.; Evans, R. M. Proc. Natl. Acad. Sci. 2017, 114, E2563.
6. Oliver, Jr., W. R.; Shenk, J. L.; Snaith, M. R.; Russell, C. S.; Plunket, K. D.; Bodkin, N. L.; Lewis, M.
C.; Winegar; D. A.; Sznaidman, M. L.; Lambert, M. H.; Xu, H. E.; Sternbach, D. D.; Kliewer, S. A.;
Hansen, B. C.; Willson, T.M. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 5306.
7. The maximum rate for clearance/extinction of a drug given i.v. to a mouse is the cardiac output (CO),
which is approximately 500 mL/min/kg: see Janssen, B.; Debets, J.; Leenders, P.; Smits, J. Amer. J.
Physiol. 2002, 282, R928.
8. In this letter plasma clearance (CL) is categorized as high, medium or low with respect to its
percentage of total CO. Clearance is ranked as high if it is above 0.35 CO, medium if it is around 0.15
CO and low if it is around 0.05 CO: see Toutain, P. L.;Bousquet-Melou, A. J. Vet. Pharmacol. Therap.
2004, 27, 415.
9. Detailed synthetic schemes and procedures for preparing the compounds described in this letter can be
found in PCT Int. Appl. WO 2016057660..
10. Davies, B.; Morris, T, Pharmaceutical Res. 1993, 10, 1093
11. All the animal experiments were carried out as per the guidelines of the Committee for the Purpose of
Control and Supervision of Experiments on Animals (CPCSEA), Government of India and approved
by the Institutional Animal Ethics Committee (IAEC), Aurigene Discovery Technologies Ltd,
Bengaluru, India.
12. Compound (10 mmol/L) was incubated with cryopreserved rat (male and female) and human (gender
mixed) hepatocytes (1 X 106 cells/mL) at 37⁰C in for 30 and 120 min. The metabolites were analyzed
by LC-MS. The authors thank Dr. Keitarou Kadono (Astellas Pharma) for the data.
13. Compounds were dosed as a solution in 2% DMA , 20% HPβCD intravenously.at 3 mg/kg to male
CD-1 mice. A standard elimination curve was generated by measuring plasma exposure at 10 points
between 0-24h. The values shown are an average of three determinations at each time point.
14.
15. In the protein interaction assay the PPAR ligand complexes with a PPAR ligand binding domain fused
to an inactive fragment of galactosidase. This complex complements another inactive galactosidase
fragment to form an active galactosidase enzyme that is read out in a fluorescence assay. See
www.discoverx.com for details. All the compounds in Table 1 were inactive (EC50 >10,000 nM) as
modulators of PPARα and PPARγ. Compound 1, with a mean EC50 of 9.8 2.5 nM, served as a
positive control for PPARδ in each run.
16. In the transactivation assay CV-1 cells are transfected with a PPAR ligand binding domain fused to a
GAL4 promoter to generate a hormone-inducible activator. A test ligand is added and activity is
measured in a luciferase assay. See WO2016057660 for further details.
17. SPR binding studies were performed with PPARδ His-tagged (33.2 kD)at 25ºC using a one-shot
kinetics approach. BioRad ProteOn XPR36 optical biosensor equipped with a nickel-charged, NTA-
derivatized GLH sensor chip and equilibrated with running buffer (10 mM HEPES, 150 mM NaCl, 5%
glycerol, 1 mM TCEP, 0.005% Tween-20, pH 7.5) was used for analysis.
18. Coordinates of the X-ray structure have been deposited in the Protein Data Bank (accession code PDB
5XMX). The residue numbering follows the UniProt residue numbering convention in accession
number Q03181 (PPARD_HUMAN).
19. Lagu, B.; Kluge, A. F.; Goddeeris, M. M.; Tozzo, E.; Fredenburg, R. A.; Chellur, S.; Senaiar, R. S.;
Jaleel, M.; Babu, D. R. K.; Tiwari, N. K.; Takahashi, T.; Patane, M. A. Bioorg. Med. Chem. Lett. 2017
(accompanying paper).
Graphical Abstract
10

11