Discovery and in Vivo Evaluation of the Potent and Selective PI3Kδ Inhibitors 2-((1S)-1-((6-Amino-5-cyano-4-pyrimidinyl)amino)ethyl)-6-fluoro-N-methyl-3-(2-pyridinyl)-4-quinolinecarboxamide (AM-0687) and 2-((1S)-1-((6-Amino-5-cyano-4-pyrimidinyl)amino)ethyl)-5-fluoro-N-methyl-3-(2-pyridinyl)-4-quinolinecarboxamide (AM-1430)

Article abstract of DOI:10.1021/acs.jmedchem.6b00827

Optimization of the potency and pharmacokinetic profile of 2,3,4-trisubstituted quinoline, 4, led to the discovery of two potent, selective, and orally bioavailable PI3Kδ inhibitors, 6a (AM-0687) and 7 (AM-1430). On the basis of their improved profile, these analogs were selected for in vivo pharmacodynamic (PD) and efficacy experiments in animal models of inflammation. The in vivo PD studies, which were carried out in a mouse pAKT inhibition animal model, confirmed the observed potency of 6a and 7 in biochemical and cellular assays. Efficacy experiments in a keyhole limpet hemocyanin model in rats demonstrated that administration of either 6a or 7 resulted in a strong dose-dependent reduction of IgG and IgM specific antibodies. The excellent in vitro and in vivo profiles of these analogs make them suitable for further development.

Full text of DOI:10.1021/acs.jmedchem.6b00827

Subscriber access provided by UNIV HEIDELBERG
Article
Discovery and in Vivo Evaluation of the Potent and Selective PI3K#
Inhibitors 2-((1S)-1-((6-Amino-5-cyano-4-pyrimidinyl)amino)ethyl)-6-
fluoro-N-methyl-3-(2-pyridinyl)-4-quinolinecarboxamide (AM-0687)
and 2-((1S)-1-((6-Amino-5-cyano-4-pyrimidinyl)amino)ethyl)-5-
fluoro-N-methyl-3-(2-pyridinyl)-4-quinolinecarboxamide (AM-1430)
Felix Gonzalez-Lopez de Turiso, Xiaolin Hao, Youngsook Shin, Minna Bui, Iain David Grant Campuzano,
Mario G. Cardozo, Michelle C. Dunn, Jason Duquette, Benjamin Fisher, Robert S. Foti, Kirk R Henne, Xiao He,
Yi-Ling Hu, Ron C. Kelly, Michael G Johnson, Brian S Lucas, John D McCarter, Lawrence R McGee, Julio C
Medina, Daniela P. Metz, Tisha San Miguel, Deanna Mohn, Thuy Tran, Christine Vissinga, Sharon Wannberg,
Douglas A. Whittington, John Whoriskey, Gang Yu, Leeanne Zalameda, Xuxia Zhang, and Timothy D. Cushing
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00827 • Publication Date (Web): 14 Jul 2016
Downloaded from http://pubs.acs.org on July 17, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 46
1
2
3
4
5
6
7
8
9
Discovery and in Vivo Evaluation of the Potent and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Selective PI3Kδ Inhibitors 2ꢀ((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ
4ꢀpyrimidinyl)amino)ethyl)ꢀ6ꢀfluoroꢀNꢀmethylꢀ3ꢀ(2ꢀ
pyridinyl)ꢀ4ꢀquinolinecarboxamide (AMꢀ0687) and 2ꢀ
((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ4ꢀpyrimidinyl)amino)ethyl)ꢀ5ꢀ
fluoroꢀNꢀmethylꢀ3ꢀ(2ꢀpyridinyl)ꢀ4ꢀquinolinecarboxamide
(AMꢀ1430)
Felix GonzalezꢀLopez de Turiso,ǁ,* Xiaolin Hao,† Youngsook Shin, ‡,* Minna Bui,† Iain D. G. Campuzano,‡
Mario Cardozo,† Michelle C. Dunn,† Jason Duquette,† Benjamin Fisher,† Robert S. Foti, ǁǁ Kirk Henne,§ Xiao
He,† YiꢀLing Hu,# Ron C. Kelly,⊥ Michael G. Johnson,† Brian S. Lucas,† John McCarter,‡ Lawrence R. McGee,†
Julio C. Medina,† Daniela Metz,# Tisha San Miguel,‡ #Deanna Mohn, # Thuy Tran,§ Christine Vissinga,# Sharon
Wannberg,# Douglas A. Whittington,ǁ John Whoriskey,# Gang Yu, # Leeanne Zalameda,‡ Xuxia Zhang, # Timothy
D. Cushing.†
†Department of Therapeutic Discovery, §Department of Pharmacokinetics and Drug Metabolism, Amgen Inc.,
1120 Veterans Boulevard, South San Francisco, California 94080, United States.
ACS Paragon Plus Environment

Page 3 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
‡Department of Therapeutic Discovery, #Department of Inflammation Research, ⊥Drug Product Technologies,
Amgen Inc., One Amgen Center Drive, Thousand Oaks, California 91320, United States.
ǁDepartment of Therapeutic Discovery, ǁǁDepartment of Pharmacokinetics and Drug Metabolism, Amgen Inc.,
360 Binney Street, Cambridge, Massachusetts 02142, United States.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABSTRACT
Optimization of the potency and pharmacokinetic profile of 2,3,4ꢀtrisubstituted quinoline 4, led to the discovery
of two potent, selective, and orally bioavailable PI3Kδ inhibitors, 6a (AMꢀ0687) and 7 (AMꢀ1430). On the
basis of their improved profile, these analogs were selected for in vivo pharmacodynamic (PD) and efficacy
experiments in animal models of inflammation. The in vivo PD studies, which were carried out in a mouse
pAKT inhibition animal model, confirmed the observed potency of 6a and 7 in biochemical and cellular assays.
Efficacy experiments in a keyhole limpet hemocyanin (KLH) model in rats demonstrated that administration of
either 6s or 7 resulted in a strong doseꢀdependent reduction of IgG and IgM specific antibodies. The excellent
in vitro and in vivo profiles of these analogs make them suitable for further development.
INTRODUCTION
Phosphoinositide 3ꢀkinases (PI3Ks) are lipid kinases that phosphorylate the 3´ position of the inositol ring in
phosphatidylinositol.1 The resulting phosphoinositides (PIs) are secondary messengers that play central roles in
the regulation of cellular processes including signal transduction, cell survival and control of membrane
trafficking. Thus, inhibition of PI3K activity has been regarded as a suitable strategy to modulate these key
cellular processes.2,3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 46
1
2
3
4
5
6
7
8
9
Among the different types of PI3Ks, class I PI3Ks play a central role in the regulation of phosphatidylinositol
(3,4,5)ꢀtriphosphate (PIP3) levels. Importantly, PIP3 functions as a membrane anchoring site for key protein
kinases including serine/threonine kinases such as AKT (or PKB), PDK1 and tyrosine kinases of the Tec
family. Regulation of the PIP3 level in the cell is controlled by two different subclasses of class I PI3Ks: class
IA, which is comprised of PI3Kα, β and δ isoforms and class IB, which is comprised of PI3Kγ.4,5 In terms of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
tissue distribution, PI3Kα and PI3Kβ are ubiquitously expressed whereas PI3Kγ and PI3Kδ are primarily found
in leukocytes.6 The key role of class I PI3Ks in signal transduction in combination with its tissue distribution
and mouse genetic studies have made small molecule inhibition of class I PI3Ks a prominent area of research
within the pharmaceutical industry.7,8 Despite this interest, only one PI3Kδ inhibitor (Idelalisib) has been
approved for the treatment of certain types of cancer. The limited success in this field is a testimony to not only
the vast amount of research necessary to bring a new medicine to patients, but also the importance of flexibility
during clinical trial design and execution. In particular, within the PI3K field, mouse genetic studies and
preclinical animal models of disease established PI3Kα and PI3Kβ as plausible targets for the treatment of
human cancer8 whereas PI3Kγ and PI3Kδ, which have a central role in leukocyte biology, were viewed as
attractive targets for the treatment of inflammatory diseases.9 Despite this data, clinical trials in cancer patients
with PI3Kα and PI3Kβ inhibitors are still in progress and results from pivotal Phase 2 and Phase 3 clinical trials
for drug approval are still pending.10 Concurrent with these studies, the realization that PI3K mediated human
B cell modulation could be beneficial for the treatment of certain types of leukemias led to the repositioning of
PI3Kδ inhibitors as potential therapeutics for the treatment of oncologic diseases.11 This change in
development strategy proved critical for the successful outcome of the PI3Kδ field and has resulted in the
accelerated approval of Idelalisib for the treatment of patients suffering from relapsed chronic lymphocytic
leukemia (in combination with Rituximab),12 relapsed follicular nonꢀHodgkin lymphoma and relapsed small
lymphocytic lymphoma.13
ACS Paragon Plus Environment

Page 5 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Our own research efforts in this field have focused on a rational design approach to identify PI3K inhibitors
with built in kinase and isoform selectivity for the treatment of both oncologic and inflammatory human
diseases. Some of the initial work in this area described the identification and in vivo evaluation of PI3Kα, dual
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PI3Kβ/δ and selective PI3Kδ inhibitors (Figures 1a and 1b).14 Recently, the identification of the clinical
candidate 3 (AMG 319), which is currently being evaluated in Phase 2 clinical trials for the treatment of human
papillomavirus (HPV) and negative head and neck squamous cell carcinoma (HNSCC) has also been reported.15
Herein we describe the discovery and in vivo preclinical studies of 6a and 7, two potent and selective PI3Kδ
inhibitors containing a unique 4ꢀcarboxamide quinoline pharmacophore. These studies are part of an effort to
identify a clinical candidate with suitable profile for the treatment of inflammatory diseases.
Figure 1. Structures of PI3K inhibitors. a) Dual PI3Kβ/δ inhibitor, b) Selective PI3Kδ inhibitor, c) Selective PI3Kδ inhibitor.
RESULTS AND DISCUSSION
Optimization Strategy
Efforts to identify a second clinical candidate within the PI3Kδ program focused on further improving the
isoform selectivity profile of 3 while maintaining its favorable PK parameters. A number of different strategies
were pursued to achieve this goal, including exploration of new chemical series (e.g., benzimidazoles,14a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 46
1
2
naphthyridines,16 pyridopyrimidinones17 or thienopyridines18) as well as small modifications of the quinoline
ring present in the 3 series (Figure 2).19 The latter approach led to the identification of the 2,3,4ꢀsubstituted
quinoline pharmacophore exemplified by 4, a reasonably potent and selective PI3Kδ inhibitor.20 At the outset,
this compound was viewed as a suitable lead for further optimization and work focused on improving its
intrinsic microsomal stability and PXR activation profile (Figure 2). To address these issues the strategy
focused on reducing the clogP of these analogues by replacing the pyridine ring at C4 with more polar groups.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NH2
HN
N
CN
NH
N
N
N
N
NH
N
2
3
N
F
4
N
F
N
N
4
3
PI3K a IC50 = 33000 nM
PI3K a IC50 = 2700 nM
PI3K a IC50 = 18 nM
PI3K a IC50 = 850 nM
PI3K a IC50 = 7900 nM
PI3K a IC50 = 2800 nM
PI3K a IC50 = 2.4 nM
PI3K a IC50 = 850 nM
pAKTb IC50 = 1.5 nM
pAKTb IC50 = 8.6 nM
HLMc Clint = <14 L/min.mg
HLMc Clint = 47 L/min.mg
hPXRd POC @ 2,10 M = 3, 8%
hPXRd POC @ 2,10 M = 8, 51%
Figure 2. Profile of PI3Kδ inhibitors 3 and lead compound 4. aBiochemical: Alphascreen assay. bCellular: In vitro HTRF assay,
phosphorylatedꢀAKT.cSingle experimental values. %Turnover was measured by LC/MS after incubation of parent compound (1 ꢀM)
in liver microsome (0.25 mg/mL) in potassium phosphate (66.7 mM) buffered with NADPH (1 mM) at 37 oC for 30 min. dhuman
Pregnane X Receptor. Data are shown as Percent of Control.
Due to the synthetic accessibility of a C3 phenylꢀsubstituted quinoline penultimate intermediate (carboxylic
acid 15, Scheme 1), initial optimization of the C4 group was performed within this series. Preliminary work
focused on replacing the pyridine ring at C4 with small functional groups that would also contain a Hꢀbond
acceptor (Figure 3a, general structure 5).21 After a suitable replacement of the pyridine ring was identified, our
ACS Paragon Plus Environment

Page 7 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
attention was refocused on improvement of the PXR activation profile by further reducing clogP. Among the
compounds synthesized, the series in which the C3 pyridine substituent had been reintroduced was found to
display the best overall profile (Figure 3b, general structure 6). Finally, modification of the 6ꢀFꢀquinoline
substitution pattern was also explored since previous optimization of the 3 series suggested that this
modification could improve the overall profile of these inhibitors (Figure 3c, representative structure 7).15
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Strategy for optimization of quinoline analogs. a) Introduction of a small, polar group with a Hꢀbond acceptor at C4. b)
Introduction of a pyridine ring to reduce clogP. c) Introduction of a 5ꢀFꢀsubstituted quinoline.
Chemistry
With this strategy in mind, a rapid synthesis of 3ꢀphenylquinoline derivatives was developed (Scheme 1).
Pfitzinger condensation between 5ꢀfluoroisatin (8) and 1ꢀphenylbutanꢀ2ꢀone provided 4ꢀquinoline acid 9.22
Next, exposure of 9 to thionyl chloride followed by a methanol quench afforded methyl ester 10. Treatment of
10 with 1,3ꢀdibromoꢀ5,5ꢀdimethylhydantoin in the presence of a radical initiator provided benzyl bromide 11,
which was converted to the corresponding amine intermediate 12 in two steps including: 1) substitution using
sodium azide and 2) hydrogenation under Pd/C catalysis. Key carboxylic acid 14 was obtained through a SNAr
reaction between 12 and 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile followed by LiIꢀmediated methyl ester
hydrolysis. Chiral acid 15 was obtained by SFC purification of this racemate (using isopropanol/hexane
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 46
1
2
3
4
gradient on an AD column) and the final analogs 5aꢀc and 5fꢀi were synthesized through treatment of 15 with
5
6
various amines and the coupling reagent PyBop in DMF.
7
8
9
The synthesis of analogs 5d and 5e began with the Bocꢀprotection of racemic amine 12. The resultant
intermediate was purified by chiral SFC to afford chiral carbamate 16. The methyl ester substituent of 16 was
then hydrolyzed under aqueous basic conditions to give the corresponding acid, which was subsequently
coupled with either ethylamine or isopropylamine to provide 17d and 17e, respectively. The final analogs 5d
and 5e were obtained after HClꢀmediated Bocꢀdeprotection and SNAr of the corresponding amines with 4ꢀ
aminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1a
ACS Paragon Plus Environment

Page 9 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Br
NH2
N
O
N
N
F
iii
iv
i
O
F
F
F
N
CO2R
H
CO2Me
11
CO2Me
8
9
, R = H
12
ii
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
, R = Me
NH2
N
NH2
NH2
N
CN
NH
CN
NH
5a, R = NH2
5b, R = NHMe
5c, R = NMe2
CN
NH
N
N
N
N
vii
viii
O
O
N
N
N
S
5f, R =
5g, R =
5h, R =
N
N
NH
OH
v
F
F
F
N
CO2R
O
R
O
OH
15
5i, R = NH(CH2)2NHMe
13, R = Me
vi
14
, R = H
12
NH2
N
CN
NH
O
O
O
N
ix
HN
O
HN
O
N
N
x
xi
N
F
F
F
CO2Me
R
O
R
16
17d, R = NHEt
17e, R = NHiPr
5d, R = NHEt
5e, R = NHiPr
a
Reagents and conditions: (i) 1ꢀphenylbutanꢀ2ꢀone, KOH, EtOH, reflux, 96%; (ii) SOCl2, MeOH, 21%; (iii) 1,3ꢀdibromoꢀ5,5ꢀ
dimethylhydantoin, benzoyl peroxide (cat.), CCl4, reflux, 56%; (iv) 1. NaN3, DMF; 2. H2, Pd/C, MeOH, 95% over two steps; (v) 4ꢀ
aminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile, nꢀBuOH, Hunig's base, 110 °C; (vi) LiI, pyridine, 100 °C, 81% over two steps; vii) chiral
separation, 45%; (viii) amine, PyBop, Hunig's base, DMF, 13ꢀ99%; (ix) 1. Boc2O, TEA, THF, r.t., 77%; 2. chiral separation, 44%; (x)
1. NaOH, MeOH, 60 °C, 99%; 2. amine, PyBop, Hunig's base, DMF, 69ꢀ86%; (xi) 1. HCl, r.t.; 2. 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀ
carbonitrile, nꢀBuOH, Hunig's base, 100 °C, 31ꢀ68% over 2 steps.
The synthesis of the C3 pyridine compounds was performed following two different routes (Schemes 2 and 3).
Analog 6d was synthesized following a synthetic route similar to that described in Scheme 1 (albeit in lower
yield). The methyl ester 19 was obtained by Pfitzinger condensation of 8 with 1ꢀ(pyridinꢀ2ꢀyl)butanꢀ2ꢀone and
subsequent esterification with TMSCHN2. An analogous three step benzylic amination procedure was
followed, which included: 1) bromination, 2) substitution, and 3) azide reduction to yield amine 21. Coupling
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 46
1
2
3
4
5
6
7
8
9
of this intermediate with 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile and hydrolysis of the resultant methyl ester
provided acid 23. Finally, the analog 6d was obtained via PyBopꢀmediated amide coupling with azetidinꢀ3ꢀol
as a racemate.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2a
aReagents and conditions: (i) 1ꢀ(pyridinꢀ2ꢀyl)butanꢀ2ꢀone, KOH, EtOH, reflux, 99%; (ii) TMSCHN2, MeOH, r.t., 29%; (iii) 1,3ꢀ
dibromoꢀ5,5ꢀdimethylhydantoin, benzoyl peroxide (cat.), CCl4, reflux, 80%; (iv) 1. NaN3, DMF; 2. H2, Pd/C, 86% over two steps; (v)
4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile, nꢀBuOH, Hunig's base, 100 °C, 98%; (vi) LiI, pyridine, 100 °C, 98%; (vii) amine, PyBop,
Hunig's base, DMF, 13%.
The low yield of the previous route prompted the development of an alternative approach for the synthesis of
analogs 6a,b and 6e (Scheme 3). This route commenced with known ketone 24 (prepared in one step from the
corresponding Weinreb amide), which when refluxed with 5ꢀfluoroisatin 8 and potassium hydroxide in ethanol
gave acid 25. This intermediate was then coupled to a variety of different amines under the previously
described PyBop conditions to deliver the corresponding amides 26a,b and 26e. Amides 26a,b and 26e were
ACS Paragon Plus Environment

Page 11 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
then treated with 4 N HCl to remove the Boc protecting group. Subsequent SNAr reaction of the crude amine
with 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile provided the final products 6a, 6b and 6e after chiral separation
(Scheme 3ꢀtop). The synthesis of 6c (Scheme 3ꢀbottom) began with the coupling of key acid intermediate 25
with benzyl piperazineꢀ1ꢀcarboxylate under standard conditions. Deprotection of the Bocꢀgroup and a SNAr
reaction of the resultant amine with 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile afforded 27. Final deprotection
of the Cbz group in 27 provided analog 6c.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3a
aReagents and conditions: (i) 8, KOH, EtOH, reflux, 78%; (ii) R1R2NH, PyBop, Hunig's base, DMF, 75%; (iii) 1. 4N HCl in dioxane;
2. nꢀBuOH, Hunig's base, 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile, 110 °C; c) chiral separation, 13ꢀ42% from 26a,b,e; iv) amine,
PyBop, Hunig’s base, DMF, 24%; v) 1. 4N HCl in dioxane; 2. nꢀBuOH, Hunig's base, 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile, 110
°
C, 100%; (vi) H2, 10% Pd/C, MeOH, 26%.
An analogous synthetic route to that described in Scheme 3 enabled the synthesis of the 5ꢀfluoro analog 7. This
sequence started with commercially available 4ꢀfluoroisatin and NꢀBocꢀprotected ketoamino intermediate 24,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 46
1
2
3
4
5
6
7
8
9
which provided acid intermediate 28 via the previously described Pfitzinger condensation (Scheme 4).
Coupling of intermediate 28 with methylamine using PyBop and Hunig’s base provided Bocꢀprotected
intermediate 29. Deprotection of the Bocꢀgroup with 4N HCl in dioxane and subsequent SNAr reaction of the
amino intermediate with 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile provided the desired racemic 7. Chiral
separation by SFC led to the isolation of 7.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4a
aReagents and conditions: (i) 4ꢀfluoroisatin, KOH, EtOH, reflux, 72%; (ii) MeNH2, PyBop, Hunig's base, DMF, 86%;
(iii) a) 4N HCl in dioxane; b) 1. nꢀBuOH, Hunig's base, 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile, 110 °C; 2. chiral separation, 13%
from 29.
Structure Activity Relationship (SAR)
Initial replacement of the pyridine ring at the C4 position in 4 with a primary amide led to an improvement in
both biochemical and cellular potencies (5a, Table 1). This analog also displayed improved in vitro microsomal
stability and PXR activation profile relative to the parent bisꢀpyridine compound 4. A further improvement in
biochemical and cellular potency was achieved by replacing the primary amide in 5a with a methylamide (5b,
Table 1). Compound 5b also had improved PI3Kα/γ isoform selectivity relative to 5a. The low microsomal
intrinsic clearance correlated well with a favorable PK profile (Cl = 0.5 L/hr/kg) and the Clu (clearance
corrected by free fraction) was comparable to that observed in 3 (4.5 vs 2.7 L/hr/kg, Table 1).23 Introduction of
ACS Paragon Plus Environment

Page 13 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a dimethylamide at the C4 position (analog 5c, Table 1) led to a marginal improvement in biochemical potency,
but the PXR activation profile was unfavorable (75 and 151 POC at 2 and 10 ꢀM, respectively). The secondary
ethylamide substitution (analog 5d, Table 1) was significantly less potent that the corresponding methylamide
5b (51 vs 1.3 nM) in the biochemical assay. Cyclic tertiary amides were also evaluated (analogs 5f, 5g, 5h,
Table 1). Among these, the piperazine amide 5g displayed the most favorable potency profile (0.2 nM against
PI3Kδ in the biochemical assay), but the rat PK was significantly worse (Clu = 60 L/hr/kg).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1: SAR of 3ꢀphenyl analogs 5aꢀi.
NH2
CN
NH2
HN
N
N
CN
NH
N
N
N
N
NH
N
N
NH
N
N
F
F
N
F
N
N
O
R
3
4
5a-h
Calculated
properties
Biochemical potencya
PI3K IC50 (nM)
Cellular potency
ADME Properties
CYP3A4 /
Mouse B
Human B
cellb
RLM /
HLMd
clogP /
LipE
Cmpd
R
cell
hPXRf
Cl / Clug
L/hr/kg
2D6e
(pAKT)c
POC @
δ
α
β
γ
ꢀL/(min.mg)
IC50 (ꢀM)
IC50 (nM)
IC50 (nM)
2/10 ꢀM
3
4
ꢀ
18
2.8
2
33000
7900
5760
12600
703
2700
2800
3540
1240
545
850
850
301
634
146
4650
853
8.6
4.6
1.8
1.9
2.2
1.2
1.3
1.5
8.6
ꢀ
<14 / <14
16 / 47
>27 / >27
>27 / >27
13 / >27
14 / >27
19 / >27
>27 / >27
13 / >27
3 / 8
8 / 51
0.3/2.7
2.3 / 4.9
2.3 / 6.6
2.1 / 6.8
2.5 / 7.1
2.3 / 7.7
3.0 / 4.6
3.3 / 5.5
ꢀ
ꢀ
5a
5b
5c
5d
5eh
–NH2
–NHMe
–NMe2
–NHEt
–NHiPr
<14 / <14
15 / <14
426 / 198
<14 / 22
<14 / 51
5 / 33
0.6/ꢀ
1.3
0.8
51
3.2
0.3
ꢀ
15 / 52
75 / 151
4 / 61
0.5/4.5
ꢀ
ꢀ
ꢀ
6230
12700
17000
8050
ꢀ
ꢀ
38 / 128
5f
1.2
0.2
444
949
143
35.4
22.5
ꢀ
78 / 62
50 / 82
>27 / >27
>27 / >27
50 / 63
7.4 / ꢀ
ꢀ
1.5 / 6.7
2.4 / 7.5
ꢀ
ꢀ
5g
1100
1.2
0.9/60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 46
1
2
3
4
5
6
5h
5i
1.3
9.5
ꢀ
ꢀ
2720
507
ꢀ
49 / 40
>27 / >27
>27 / >27
3.3 / 17
ꢀ
2.3 / 6.8
2.3 / 5.7
ꢀ
–NH(CH2)2NHMe
15300
9550
2.9
ꢀ
<14 / <14
ꢀ4.1 / 5.6
7
8
9
aBiochemical: Alphascreen assay. bCellular: In vitro antiꢀIgM/CD40Lꢀinduced human B cell proliferation (as measured by thymidine incorporation) assay. cCellular:
Ability of compound to inhibit antiꢀIgM induced AKT phosphorylation (pAKTSer473) in mouse B cells; phosphoꢀAKT (pAKT) expression on B220+gated B cells was
determined by flow cytometry in mouse splenocytes. dSingle experimental values. % Turnover was measured by LC/MS after incubation of parent compound (1 ꢀM) in
liver microsome (0.25 mg/mL) in potassium phosphate (66.7 mM) buffered with NADPH (1 mM) at 37 oC for 30 min. eCYP: cytochrome P450 assay competitive
(midazolam as probe substrate, 5 ꢀM) reported as % inhibition at 3 ꢀM (LC/MS); fhuman Pregnane X Receptor. Data are shown as Percent of Control. gPK experiments
were carried out using male SpragueꢀDawley rats (n=3). Test compounds formulated at appropriate concentrations for either IV (0.5 mL/kg) or oral (2ꢀ10 mL/kg)
administration. Unbound clearance equals the total clearance divided by the rat fraction unbound (CLu = CL/fu). hRacemic compound.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With the goal of improving the unfavorable PXR activation profile observed in some of the most promising
analogs (e.g., 5aꢀc), a strategy directed towards introducing polarity in 5aꢀh was pursued. Consistent with the
structure of bisꢀpyridine analog 4, it was predicted that replacement of the phenyl group at C3 in analogs 5aꢀh
with a pyridine ring would not be detrimental to potency and could potentially address the PXR activation
liability. To test this hypothesis analogs 6aꢀe, and 7 were synthesized.
Table 2: SAR of analogs 6aꢀe and 7.
NH2
NH2
HN
N
CN
NH
N
CN
NH
N
N
N
N
N
NH
N
N
N
N
F
N
F
F
O
NH
N
R
O
3
6a-e
7
Calculated
properties
Biochemical potencya
PI3K IC50 (nM)
Cellular potency
ADME Properties
Mouse B
Human B
cellb
RLM /
HLMd
CYP3A4 /
2D6e
clogP /
LipE
Cmpd
R
cell
hPXRf
Cl/ Clug
L/hr/kg
(pAKT)c
POC @ 2
/ 10 ꢀM
IC50 (nM)
IC50 (nM)
δ
α
β
γ
ꢀL/(min.mg)
IC50 (ꢀM)
3
ꢀ
18
2.9
3.6
33000
17600
16800
2700
2840
850
3530
5360
8.6
0.8
1.6
1.5
0.7
0.5
<14 / <14
<14 / <14
<14 / <14
>27 / >27
>27 / >27
>27 / >27
3 / 8
1 / 12
ꢀ1 / 5
0.3/2.7
0.5/2.3
0.2/6.1
2.3 / 4.9
1.3 / 7.6
1.8 / 6.9
6a
6b
–NHMe
–NHEt
11000
6ch
4.3
5750
7160
964
1.6
ꢀ
17 / 24
>27 / 27
0 / 7
ꢀ
1.2 / 7.4
ACS Paragon Plus Environment

Page 15 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
6dh
12
10900
4610
3590
7.4
ꢀ
<14 / <14
>27 / >27
1 / 4
ꢀ
1.2 / 7.1
7
8
9
6e
7
–NH(CH2)2NHMe
ꢀ
4.8
4.6
29200
14180
15600 23200
2200 3220
1.2
0.2
8.5
0.8
<14 / 33
>27 / >27
>27 / >27
2.4 / 4.9
2 / 3
1.0/28.1
0.7/1.7
1.7 / 6.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
<14 / <14
0.95 / 7.4
aBiochemical: Alphascreen assay; bCellular: In vitro antiꢀIgM/CD40Lꢀinduced human B cell proliferation (as measured by thymidine incorporation) assay. cCellular: Ability
of compound to inhibit antiꢀIgM induced AKT phosphorylation (pAKTSer473) in mouse B cells; phosphoꢀAKT (pAKT) expression on B220+gated B cells was determined by
flow cytometry in mouse splenocytes. dSingle experimental values. %Turnover was measured by LC/MS after incubation of parent compound (1 ꢀM) in liver microsome
(0.25 mg/mL) in potassium phosphate (66.7 mM) buffered with NADPH (1 mM) at 37 oC for 30 min. eCYP: cytochrome P450 assay competitive (midazolam as probe
substrate, 5 ꢀM) reported as % inhibition at 3 ꢀM (LC/MS). fhuman Pregnane X Receptor. Data are shown as Percent of Control. gPK experiments were carried out using
male SpragueꢀDawley rats (n=3). Test compounds formulated at appropriate concentrations for IV (0.5 mL/kg). Unbound clearance equals the total clearance divided by the
rat fraction unbound (CLu = CL/fu). hRacemic compound.
Table 2 illustrates the in vitro and in vivo data for the C3 pyridine analogs; consistent with our hypothesis, the
PXR activation profile was significantly improved. The reduction in the clogP of these analogs did have an
impact in their biochemical potency although the overall ligand efficiency was generally very good (lipE > 7 for
most analogs). Among these PI3K inhibitors, the methyl amide substituted compound (6a, Table 2) displayed
excellent levels of biochemical and cellular potency and similar PK profile relative to 3 (Clu = 2.3 vs Clu =
2.7). The potency profile of the ethyl amide analog 6b was also very good although the Clu was about 3ꢀfold
higher that the parent inhibitor 6a. The overall parameters of tertiary amides 6c and 6d were also satisfactory
although the cellular potency of these analogs was reduced relative to 6a. Having established the methyl amide
in 6a as the optimal substituent, the fluorination pattern of the quinoline ring was also explored.
Contemporaneous work within the 3 series suggested that the C5 fluoroꢀquinoline substitution might be
beneficial. Accordingly, 7 was prepared and had excellent potency and selectivity profile in biochemical and
cellular assays as well as the best rat PK profile ever observed in this series (Clu = 1.7 L/hr/kg).
The potency and selectivity profiles of 3, 6a and 7 in biochemical, cellular and human whole blood assays are
shown in Table 3. This data highlights the improved potency and isoform selectivity of 6a and 7 relative to 3 in
both biochemical and cellular assays. In the case of 6a, the selectivity ratios over other isoforms were found to
be improved relative to 3. The intrinsic human whole blood assay data for 6a and 7 was also determined and it
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 46
1
2
3
4
was found to be within 2.7 and 4.5ꢀfold relative to 3 respectively, which was deemed suitable for further
5
6
progression into in vivo studies.
7
8
9
Table 3: Profile of selected analogs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PI3K Biochemical potencya
PI3Kδ Cellular potency
IC50 (nM)
IC50 (nM) and PI3Kδ selectivity
Mouse B
cell
Human B
cellb
HWB
HWB_unbound
(pAKT)e
δ
α/δ
β/δ
γ/δ
(pAKT)d
(pAKT)c
Cmpd
(nM)
(ratio)
(ratio)
(ratio)
IC50 (nM)
IC50 (nM)
IC50 (nM)
IC50 (nM)
3
6a
7
18
2.9
4.6
1833
6068
3082
150
979
478
47
8.6
0.8
0.2
1.5
0.7
0.8
10.9±2.5
42±34
1.7
4.6
7.7
1217
700
18.5±11.5
aBiochemical: Alphascreen assay. bCellular: In vitro antiꢀIgM/CD40Lꢀinduced human B cell proliferation (as measured by
thymidine incorporation) assay. cCellular: Ability of compound to inhibit antiꢀIgM induced AKT phosphorylation (pAKTSer473) in
mouse B cells; phosphoꢀAKT (pAKT) expression on B220+gate. d B cells was determined by flow cytometry in mouse
splenocytes; compound pretreated human whole blood (HWB) was stimulated with antiꢀIgD to induce phosphorylation of AKT
(pAKTSer473). eThe unbound human whole blood (HWB_unbound) potency was derived by multiplying the human plasma protein
f
binding (PPB) fraction unbound by the total HWB potency ( fu × HWB). Cellular: Compound pretreated human whole blood
(HWB) was stimulated with antiꢀIgD to induce CDꢀ69 expression on B cells (6 hours) and was evaluated by flow cytometry.
The physicochemical properties and oral PK of these analogs is shown in Table 4. The solubility of both 6a and
7 in PBS was comparable to that of 3 and these compounds also exhibited good passive permeability. The oral
bioavailability of 6a in the rat was found to be superior to both 3 and 7, while the Clu remained low for all three
analogs.
ACS Paragon Plus Environment

Page 17 of 46
Journal of Medicinal Chemistry
1
2
3
Table 4: Physicochemical properties and PK profile of selected analogs.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Solubilitya
Permeabilityb
Protein Binding (fu)
IVe
Clu (L/h/Kg)f
PO
%F
Cl
(L/h/Kg)
Cmpd
Papp x 10ꢀ6 cm/s
humanc
ratd
ꢀg/mL
3
6a
7
>200
200
30
14.6
8
0.16
0.11
0.42
0.12
0.22
0.18
0.34
0.51
0.72
2.8
2.3
4.0
54
70g
29h
200
aSingle experimental values. Aqueous equilibrium solubility assay: compound was equilibrated at room temperature for 72 hours in
PBS buffer (pH 7.4), supernatant was centrifuged and analyzed by HPLC for reporting solubility as ꢀg/mL. bPassive permeability in
LLCPK1 cells, apparent permeability (Papp) measured by mass balance (LC/MS). c5 ꢀM concentration in human plasma, protein
binding measured by ultracentrifugation and LC/MS. d5 ꢀM concentration in rat plasma, protein binding measured by
ultracentrifugation and LC/MS. ePharmacokinetic parameters following administration in male SpragueꢀDawley rats: two animals
per study, dosed at 0.5 mg/kg as a solution in 100% DMSO. fUnbound clearance equals the total clearance divided by the rat fraction
unbound (CLu = CL/fu). gDosed at 2.0 mg/kg in 0.5% Tween 80, 1% methylcellulose, 98.5% water. hDosed at 2.0 mg/kg in 0.5%
methylcellulose, 1% Tween 80, 98.5% water.
Kinase selectivity profile and crystallography
Consistent with the profile of the 3 series, both quinolines, 6a and 7, were found to retain excellent kinase
selectivity (KINOMEscan's selectivity score: SꢀScore(35)=0.013 for 6a, SꢀScore(35)=0.008 for 7) in a large
panel of 442 protein kinases tested at 10 ꢁM drug concentration.24 This selectivity can be attributed to a
common propellerꢀshaped conformation displayed by these molecules, illustrated by the Xꢀray coꢀcrystal
structure of analog 5g with PI3Kγ (Figure 4). The improved biochemical potency of these analogs can also be
rationalized by this coꢀcrystal structure; the C4 substituent of the inhibitor makes additional interactions with
the face of the Trp residue adjacent to the quinoline moiety (Trp812 in PI3Kγ, Trp760 in PI3Kδ).25 Added
flexibility in the position of this Trp residue in PI3Kδ relative to the other class I isoforms likely allows for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 46
1
2
3
4
optimization of these additional interactions, thereby enhancing selectivity primarily through a greater potency
5
6
7
boost to PI3Kδ compared to the other isoforms.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Crystal structure of PI3Kg in complex with compound 5g (PDB code: 5KAE). Residues that differ in PI3Kd are labeled.
Dashed lines indicate hydrogen bonds.
In Vivo Pharmacology
Given the improved selectivity of 6a and 7 these analogs were advanced for evaluation in our in vivo animal
models of inflammation. The in vivo experiments were performed with analogs 6a and 7, which were orally
dosed to IgMm mice26 15 minutes before an intravenous (IV) injection of FITCꢀlabeled antiꢀIgM to stimulate
circulation of B cells. The mice were sacrificed 30 minutes after stimulation to measure pAKT levels in spleen
and whole blood of FITCꢀIgM positive B cells by flow cytometry (Figure 5). Both analogs 6a and 7 showed
doseꢀdependent inhibition of AKT phosphorylation in antiꢀIgM stimulated B cells isolated from the whole
blood and spleen compartments of IgMm mice. The calculated ED50 values for 6a in spleen and whole blood
were <0.01 mg/kg and <0.04 mg/kg, respectively (Figures 5a and 5b). The calculated ED50 values for 7 in
splenocytes and whole blood were <0.003 mg/kg and 0.005 mg/kg, respectively (Figures 5c and 5d).
ACS Paragon Plus Environment

Page 19 of 46
Journal of Medicinal Chemistry
1
2
3
4
(a) pAKT Levels in Mouse Splenocytes (6a)
5
1400
1200
1000
800
600
400
200
0
6
7
8
9
ED50 <0.01 mg/kg
*
*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
*
*
*
*
Figure 5. Inhibition of activation induced AKT phosphorylation in B cells in vivo. Vehicle, 6a, 7 in 2% HPMC + 1% Tween 80, pH 2.0, was
administered to IgMm mice orally 15 minutes before intravenous (IV) injection of FITCꢀlabeled antiꢀIgM to stimulate B cells. Mice were sacrificed
30 minutes after stimulation to measure pAKT in the spleen and whole blood of FITCꢀIgM positive B cells by flow cytometry. (a) pAKT of 6a in
splenocytes. (b) pAKT of 6a in whole blood. (c) pAKT of 7 in splenocytes. (d) pAKT of 7 in whole bloodꢀTest was used to evaluate differences
between groups. The asterisk (*) denotes statistical significance (p < 0.05) when compared to the control group.
In a second in vivo experiment, the data from the in vivo pAKT assay was utilized to select doses for multiꢀdose
rat proteinꢀantigen (Keyhole Limpet Hemocyanin, KLH) studies. A multiple dose rat study was performed to
determine whether 6a and 7 could impact B cell function in vivo. For this purpose, the effect of 6a and 7 on a
humoral immune response, which is a measure of whether a compound inhibits antigenꢀspecific Ig secretion by
B cells, was studied via oral administration at multiple doses (Figures 6 and 7). These experiments
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 46
1
2
3
4
5
6
7
8
9
demonstrated that analog 6a inhibited KLHꢀspecific IgG and IgM in a doseꢀdependent manner (Figure 6a and
Figure 6b). The calculated ED50 values were 0.026 mg/kg and 0.016 mg/kg, respectively. Exposure at these
doses were expected to significantly inhibit PI3Kδ given the free drug concentrations achieved relative to the in
vitroꢀderived pAKT_unbound IC50 (0.61 nM) over the 24 hours dosing interval (Figure 6c). Analog 7 inhibited
KLHꢀspecific IgG and IgM in a doseꢀdependent manner (Figure 7a and Figure 7b). The calculated ED50 values
were 0.016 mg/kg and 0.015 mg/kg, respectively. Exposure at these doses were expected to significantly inhibit
PI3Kδ given the free drug concentrations achieved relative to in vitroꢀderived pAKT_unbound IC50 (0.67 nM)
over the 24 hours dosing interval (Figure 7c).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Exposure at all doses of 6a and 7 were below in vitro PI3Kβ pAKT_unbound IC50 (2882 nM and 3500 nM,
respectively) over 24 hours dosing interval (Figure 6c and Figure 7c). Additionally, both compounds, 6a and 7,
were well tolerated at all doses. Ultimately, the KLHꢀmediated humoral immune response allowed us to select
6a and 7 as candidates for further development.
b) KLH-specific IgM at day 10
(6a)
5000
ED50 = 0.016 mg/kg
4000
3000
2000
1000
0
90%
97%
mg/kg
+ KLH
Mean SEM
* p<0.05 vs Vehicle
(Dunnett's multiple comparisons)
ACS Paragon Plus Environment

Page 21 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Inhibition of KLHꢀspecific antibodies and deltaꢀspecific coverage. Vehicle or 6a in 1% Pluronic F68, 2% HPMC, 10% Captisol, 87%
water, was administered (0.001, 0.003, 0.01, 0.03, and 0.1 mg/kg) q.d. po for 10 days in female Lewis rats (N = 8/dose group). 2 hours after the first
dosing, 200 ꢁL of PBS containing 60 ꢁg of KLH was administered to each rat intravenously. Ten days after the KLH priming, blood was collected
for the measurement of KLH specific IgG (a) and IgM (b) by ELISA. The yꢀaxis is represented as a mean serum dilution factor. Error bars represent
the standard error of the mean (SEM) of eight rats. (c) After administration of 6a, plasma was also harvested at day 10 to assess exposures in each
dose group. Unbound drug concentrations were measured by LC−MS/MS and plotted relative to: PI3Kδ in vitro mouse pAKT_unbound IC50 0.61
nM [calculated from 0.7 nM (PI3Kδ in vitro mouse pAKT IC50) × 0.876 (fu in pAKT assay media)] represented as a blue solid line; PI3Kβ in vitro
human pAKT_unbound IC50 2882 nM [calculated from 3290 nM (PI3Kβ in vitro human pAKT IC50) × 0.876 (fu in pAKT assay media)] represented
as a red line.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 46
1
2
3
4
5
6
7
8
9
c) 7 Unbound Plasma Concentration vs Time
10
1
0.01 mg/kg
0.03 mg/kg
0.1 mg/kg
0.3 mg/kg
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PI3K pAKT_u IC50
PI3K pAKT_u IC50
0.1
0.01
0.001
0.0001
0
2
4
6
8
10 12 14 16 18 20 22 24
Time (h)
Figure 7. Inhibition of KLHꢀspecific antibodies and deltaꢀspecific coverage. Vehicle or 7 in 1% Pluronic F68, 2% HPMC, 10% Captisol, 87% water,
was administered (0.003, 0.01, 0.03, 0.1, and 0.3 mg/kg) q.d. po for 10 days in female Lewis rats (N = 8/dose group). 2 hours after the first dosing,
200 ꢁL of PBS containing 60 ꢁg of KLH was administered to each rat intravenously. Ten days after the KLH priming, blood was collected for the
measurement of KLH specific IgG (a) and IgM (b) by ELISA. The yꢀaxis is represented as a mean serum dilution factor. Error bars represent the
standard error of the mean (SEM) of eight rats. (c) After administration of 7, plasma was also harvested at day 10 to assess exposures in each dose
group. Unbound drug concentrations were measured by LC−MS/MS and plotted relative to: PI3Kδ in vitro mouse pAKT_unbound IC50 0.67 nM
[calculated from 0.8 nM (PI3Kδ in vitro mouse pAKT IC50) × 0.842 (fu in pAKT assay media)] represented as a blue solid line; PI3Kβ in vitro human
pAKT_unbound IC50 3500 nM [calculated from 4110 nM (PI3Kβ in vitro human pAKT IC50) × 0.842 (fu in pAKT assay media)] represented as a red
line.
CONCLUSION
A novel series of potent, selective and metabolically stable PI3Kδ inhibitors have been identified by optimizing
the properties of the trisubstituted quinoline 4 (Figure 8). The successful strategy to improve the profile of this
inhibitor involved modification of the pyridine ring at C4 and initial optimization within the C3 phenyl
quinoline scaffold leading to the identification of a new series of amide analogs (e.g. 5b). These new
compounds had improved microsomal stability and isoform selectivity but unfavorable PXR activation profile.
The latter issue was addressed by reducing the clogP of these molecules, which was accomplished by replacing
the phenyl group at C3 (Figure 8, compound 5b) with a pyridine ring. This strategy resulted in the
ACS Paragon Plus Environment

Page 23 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
identification of 6a, which upon further modification of the halogenation pattern led to the discovery of 7. Both
of these analogs showed satisfactory activity in both mice pAKT and rat KLH efficacy studies, which in
combination with their optimal pharmacokinetic and selectivity profile, made them suitable candidates for
further development.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Evolution of the SAR leading to the identification of 6a and 7.
EXPERIMENTAL
All solvents and chemicals used were reagent grade. Anhydrous solvents were purchased from SigmaꢀAldrich
and used as received. Analytical thin layer chromatography (TLC) and silica gel column chromatography were
performed on Merck silica gel 60 (230ꢀ400 mesh). Removal of solvents was conducted by using a rotary
evaporator and residual solvents were removed from nonꢀvolatile compounds using a vacuum manifold
maintained at approximately 1 Torr. All yields reported are isolated yields. Preparative reverse–phase high
pressure liquid chromatography (RP–HPLC) was performed using an Agilent 1100 series HPLC and
Phenomenex Gemini C18 column (5 ꢀm, 100 mm × 30 mm i.d.), eluting with a binary solvent system, A and B,
using a gradient elution [A, H2O with 0.1% TFA, B, CH3CN with 0.1% TFA] with UV detection at 220 nm.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 46
1
2
3
4
5
6
7
8
9
All final compounds were purified to ≥95% purity as determined by an Agilent 1100 series HPLC with UV
detection at 220 nm using the following method: Zorbax SBꢀC8 column (3.5 ꢀm, 150 mm × 4.6 mm i.d.),
eluting with a binary solvent system, A and B, using a 5ꢀ95% B (0ꢀ15 min) gradient elution [A, H2O with 0.1%
TFA, B, CH3CN with 0.1% TFA]; flow rate 1.5 mL/min. Mass spectral data was recorded on an Agilent 1100
series LCꢀMS with UV detection at 254 nm. All accurate mass data (highꢀresolution mass spectra: HRMS)
were acquired on a Synapt G2 QꢀToF instrument operating in positive electrospray ionization mode, over the
m/z range 50ꢀ1200. Lock mass correction was performed on the leucineꢀenkephalin ion m/z 556.2771. The
instrument resolution was 28,000 at FWHM. The compounds were introduced into the mass spectrometer using
an Agilent 1200 operated with a C4 Bridgedꢀethylꢀhybrid (BEH) analytical column (2.1 x 50 mm) at 0.25
mL/min. NMR spectra were recorded on a Bruker Avance 400 MHz and 500 MHz NMR spectrometer.
Chemical shifts (δ) are reported in parts per million (ppm) relative to residual undeuterated solvent as internal
reference and coupling constants (J) are reported in hertz (Hz). Splitting patterns are indicated as follows: s =
singlet; d = doublet; t = triplet; q = quartet; qn = quintet; dd = doublet of doublets; dt = doublet of triplets; tt =
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
triplet of triplets; m = multiplet; br = broad peak.
Preparative SFC method for chiral separation. Sample was dissolved in DCMꢀMeOH (2:1, 35 mg/mL).
Column: ADꢀH (250 × 21 mm, 5 ꢀm), ASꢀH (250 × 30 mm, 5 ꢀm), IAꢀH (250 × 21 mm, 5 ꢁm), Lux column
(250 × 30 mm, 5 mm). Mobile Phase: A=Liquid CO2; B=MeOH, EtOH, or isopropanol, Flow Rate: 70
o
mL/min. Column/Oven temperature: 40 C. UV detection wavelength 220 nm. 20.7 mg/injection. 200 ꢀ 206
bar inlet pressure.
2ꢀEthylꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid (9). A mixture of 5ꢀfluoroisatin 8 (5.00 g, 30.3 mmol),
1ꢀphenylbutanꢀ2ꢀone (4.99 mL, 33.3 mmol) and potassium hydroxide (5.10 g, 91 mmol) in EtOH (100 mL) was
heated to reflux for 3 h. After cooling to r.t., the reaction mixture was concentrated in vacuo. The residue was
ACS Paragon Plus Environment

Page 25 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
diluted with water and acidified with 1 N HCl to pH 5. The resulting solid was collected by filtration, washed
with water, and dried to give 2ꢀethylꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid 9 (8.60 g, 29.1 mmol, 96 %
yield) as a fine tan powder. Mass spectrum (ESI) m/z 296.2 [M+H]+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl 2ꢀethylꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (10). To a solution of 2ꢀethylꢀ6ꢀfluoroꢀ3ꢀ
phenylquinolineꢀ4ꢀcarboxylic acid 9 (8.60 g, 29.1 mmol) in DCM (100 mL) was added thionyl chloride (11.1
mL, 151 mmol) and the resulting mixture was stirred at r.t. overnight. The solvent was removed and MeOH (50
mL) was added and the resulting solution was stirred at r.t. for 2 h. The solvent was removed and crude residue
was purified by column chromatography on silica gel using a 0ꢀ40% gradient of EtOAc/hexane as eluent to give
methyl 2ꢀethylꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 10 (2.00 g, 6.47 mmol, 21% yield) as an offꢀwhite
solid. 1HꢀNMR (400 Hz, CDCl3) δ 8.14 (dd, J=8.0, 4.0 Hz, 1H), 7.40ꢀ7.54 (m, 5H), 7.26ꢀ7.35 (m, 2H), 3.62 (s,
3H), 2.85 (q, J=8.0 Hz, 2H), 1.23 (t, J=8.0 Hz, 3H). Mass Spectrum (ESI) m/z 310 [M+H]+.
Methyl 2ꢀ(1ꢀbromoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (11).
Methyl 2ꢀethylꢀ6ꢀfluoroꢀ3ꢀ
phenylquinolineꢀ4ꢀcarboxylate 10 (1.00 g, 3.2 mmol) and 1,3ꢀdibromoꢀ5,5ꢀdimethylhydantoin (647 mg, 2.26
mmol) were suspended in carbon tetrachloride (30 mL) and treated with benzoyl peroxide (78 mg, 0.32 mmol)
and the mixture was heated at reflux for 3 h. The reaction mixture was cooled to r.t. and treated with a saturated
aqueous sodium bicarbonate solution. The layers were separated and the aqueous layer was extracted with
DCM twice. The combined organic layers were washed with brine, dried over sodium sulfate, filtered, and
concentrated in vacuo to give methyl 2ꢀ(1ꢀbromoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 11 (1.40 g,
3.60 mmol, 56% yield) as a yellow solid. Mass Spectrum (ESI) m/z 388.0 [M+H (79Br)]+ and 390.0 [M+H
(81Br)]+.
Methyl 2ꢀ(1ꢀaminoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (12). To a solution of methyl 2ꢀ(1ꢀ
bromoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 11 (1.17 g, 3.0 mmol) in DMF (6 mL) was added NaN3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 46
1
2
3
4
5
6
7
8
9
(0.293 g, 4.5 mmol) at r.t. After 2 h, the reaction mixture was diluted with water and extracted twice with
EtOAc. The organic layers were combined, washed with water, brine, dried over Na2SO4, filtered, and
concentrated in vacuo to give methyl 2ꢀ(1ꢀazidoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate as a tan oil.
Mass Spectrum (ESI) m/z 351 [M+H]+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A solution of methyl 2ꢀ(1ꢀazidoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (3.0 mmol) in MeOH (10 mL)
was treated with 10% Pd/C (100 mg) and the mixture was stirred under an atmosphere of hydrogen gas at r.t.
overnight. The reaction mixture was filtered through a CeliteTM pad and concentrated in vacuo to give methyl
2ꢀ(1ꢀaminoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 12 (0.92 g, 2.85 mmol, 95% yield) as a yellow oil.
Mass Spectrum (ESI) m/z 325.2 [M+H]+.
Methyl 2ꢀ(1ꢀ(6ꢀaminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (13).
To the solution of methyl 2ꢀ(1ꢀaminoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 12 (660 mg, 2.04 mmol)
in nꢀBuOH (6 mL) was added 4ꢀaminoꢀ6ꢀchloropyrimidineꢀ5ꢀcarbonitrile (314 mg, 2.04 mmol) and DIPEA
(0.43 mL, 2.44 mmol) and the resulting mixture was heated to 120 °C for 3 h. The reaction mixture was cooled
to r.t., diluted with water, and extracted with EtOAc (2 × 20 mL). The organic layers were combined, washed
with water, brine, dried over Na2SO4, filtered, concentrated and purified by column chromatography on silica
gel (DCM/MeOH/NH3, 20/1/0.1) to give methyl 2ꢀ(1ꢀ(6ꢀaminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀ
1
phenylquinolineꢀ4ꢀcarboxylate (747 mg, 1.69 mmol, 83% yield) as an offꢀwhite solid. HꢀNMR (400 Hz,
DMSOꢀd6) δ 8.16 (dd, J=8.0, 4.0 Hz, 1H), 7.91 (s, 1H), 7.81 (td, J=8.0, 4.0 Hz, 1H), 7.59 (dd, J=8.0, 4.0 Hz,
1H), 7.45ꢀ7.52 (m, 5H), 7.35 (dd, J=8.0, 4.0 Hz, 1H), 7.25 (s, br, 2H), 5.30ꢀ5.40 (m, 1H), 3.59 (s, 3H), 1.30 (t,
J=8.0 Hz, 3H). Mass Spectrum (ESI) m/z 443.2 [M+H]+.
2ꢀ(1ꢀ(6ꢀAminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid (14). To
a suspension of methyl 2ꢀ(1ꢀ(6ꢀaminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀ
ACS Paragon Plus Environment

Page 27 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
carboxylate 13 (1.09 g, 2.47 mmol) in pyridine (20 mL) was added lithium iodide (0.99 g, 7.40 mmol) and the
resulting mixture was stirred at 100 °C overnight. The mixture was concentrated in vacuo. The residue was
suspended in water and the aqueous mixture was acidified with 1 N HCl to pH 5. The resulting solid was
collected by filtration, washed with water, and dried to give 2ꢀ(1ꢀ(6ꢀaminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid (1.04 g, 2.43 mmol, 98% yield) as a white solid. HꢀNMR (400
Hz, DMSOꢀd6) δ 8.16 (dd, J=8.0, 4.0 Hz, 1H), 7.97 (s, 1H), 7.81 (td, J=8.0, 4.0 Hz, 1H), 7.72 (dd, J=8.0, 4.0
Hz, 1H), 7.38ꢀ7.52 (m, 8H), 5.30ꢀ5.40 (m, 1H), 1.30 (t, J=8.0 Hz, 3H). Mass Spectrum (ESI) m/z 429.2
[M+H]+.
(S)ꢀ2ꢀ(1ꢀ(6ꢀAminoꢀ5ꢀcyanopyrimidinꢀ4ꢀylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid (15).
The racemic mixture 14 was purified with an ADꢀH column using 22% isopropanol (0.2% diethylamine) as
additive B in supercritical CO2. Compound 15 was the first eluting enantiomer and isolated as a yellow solid
(45% yield). Mass spectrum (ESI) m/z 429.2 [M+H]+.
General Procedure for the Synthesis of 5aꢀc and fꢀi. To a solution of (S)ꢀ2ꢀ(1ꢀ(6ꢀaminoꢀ5ꢀcyanopyrimidinꢀ4ꢀ
ylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylic acid 15 (0.16 mmol) in DMF (1 mL) was added amine
(1.5 equiv.), DIPEA (1.1 equiv.), and PyBop (2.2 equiv.) and the resulting mixture was stirred at r.t. for 1 h.
The mixture was concentrated in vacuo. The residue was dissolved in EtOAc and washed with water, brine,
dried over sodium sulfate, filtered, and concentrated in vacuo. The crude product was purified by column
chromatography on silica gel using 0ꢀ100% EtOAc/hexane as eluent or preparative HPLC to give 5aꢀc, fꢀi.
2ꢀ((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ4ꢀpyrimidinyl)amino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylꢀ4ꢀquinolinecarboxamide (5a).
Obtained in 20% yield as a white solid. (400 MHz, CD3OD) δ ppm 8.19 (1 H, dd, J=9.2, 5.3 Hz), 8.10 (1 H, s),
7.64 ꢀ 7.71 (1 H, m), 7.59 (1 H, dd, J=9.4, 2.7 Hz), 7.43 ꢀ 7.52 (5 H, m), 5.58 (1 H, q, J=6.7 Hz), 1.43 (3 H, d,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 46
1
2
J=6.8 Hz). Mass spectrum (ESI) m/z 428.2 [M+H]+. HRMS (ESI) m/z calculated for C23H18FN7O [M+H]+:
3
4
5
6
428.1635; mass measured, 428.1629.
7
8
9
2ꢀ((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ4ꢀpyrimidinyl)amino)ethyl)ꢀ6ꢀfluoroꢀNꢀmethylꢀ3ꢀphenylꢀ4ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
quinolinecarboxamide (5b). Obtained in 27% yield as a white solid. H NMR (400 MHz, DMSOꢀd6) δ ppm
8.34 ꢀ 8.43 (1 H, m), 8.13 (1 H, dd, J=9.2, 5.5 Hz), 7.95 (1 H, s), 7.78 (1 H, td, J=8.8, 2.7 Hz), 7.36 ꢀ 7.54 (7 H,
m), 7.29 (2 H, br s), 5.35 (1 H, quin, J=6.7 Hz), 2.53 (3 H, d, J=4.7 Hz), 1.26 (3 H, d, J=6.7 Hz). Mass
Spectrum (ESI) m/z 442.0 [M+H]+. HRMS (ESI) m/z calculated for C24H20FN7O [M+H]+: 442.1792; mass
measured, 442.1784.
2ꢀ((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ4ꢀpyrimidinyl)amino)ethyl)ꢀ6ꢀfluoroꢀN,Nꢀdimethylꢀ3ꢀphenylꢀ4ꢀ
quinolinecarboxamide (5c). Obtained in 26% yield as a white solid. 1H NMR (400 MHz, CD3OD) δ ppm 8.21
(1 H, dd, J=9.2, 5.3 Hz), 7.98 (1 H, s), 7.64 ꢀ 7.71 (1 H, m), 7.57 ꢀ 7.62 (1 H, m), 7.48 ꢀ 7.56 (3 H, m), 7.40 ꢀ
7.47 (1 H, m), 7.36 (1 H, dd, J=9.3, 2.6 Hz), 5.61 (1 H, q, J=6.6 Hz), 2.84 (3 H, s), 2.60 (3 H, s), 1.26 (3 H, d,
J=6.7 Hz). Mass Spectrum (ESI) m/z 456.2 [M+H]+. HRMS (ESI) m/z calculated for C25H22FN7O [M+H]+:
456.1948; mass measured, 456.1946.
4ꢀAminoꢀ6ꢀ(((1S)ꢀ1ꢀ(4ꢀ((1,1ꢀdioxidoꢀ4ꢀthiomorpholinyl)carbonyl)ꢀ6ꢀfluoroꢀ3ꢀphenylꢀ2ꢀ
quinolinyl)ethyl)amino)ꢀ5ꢀpyrimidinecarbonitrile (5f). Obtained in 13% yield as a white solid. Mass
Spectrum (ESI) m/z 546.2 [M+H]+. HRMS (ESI) m/z calculated for C27H24FN7O3S [M+H]+: 546.1724; mass
measured, 546.1807.
4ꢀAminoꢀ6ꢀ(((1S)ꢀ1ꢀ(6ꢀfluoroꢀ3ꢀphenylꢀ4ꢀ(1ꢀpiperazinylcarbonyl)ꢀ2ꢀquinolinyl)ethyl)amino)ꢀ5ꢀ
1
pyrimidinecarbonitrile (5g). Obtained in 40% yield as a white solid. H NMR (400 MHz, CD3OD) δ ppm
8.22 ꢀ 8.27 (1 H, m), 8.21 (1 H, s), 7.72 (1 H, td, J=8.9, 2.8 Hz), 7.64 ꢀ 7.68 (1 H, m), 7.47 ꢀ 7.63 (6 H, m), 5.80
(1 H, q, J=6.8 Hz), 3.87 (1 H, ddd, J=14.5, 7.5, 3.4 Hz), 3.66 (1 H, ddd, J=14.5, 6.8, 3.5 Hz), 3.34 ꢀ 3.39 (1 H,
ACS Paragon Plus Environment

Page 29 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
m), 3.22 ꢀ 3.29 (1 H, m), 3.10 ꢀ 3.20 (1 H, m), 2.89 ꢀ 3.01 (1 H, m), 2.63 ꢀ 2.75 (1 H, m), 2.45 ꢀ 2.54 (1 H, m),
1.28 (3 H, d, J=6.7 Hz). Mass Spectrum (ESI) m/z 497.0 [M+H]+. HRMS (ESI) m/z calculated for C27H25FN8O
[M+H]+: 497.2214; mass measured, 497.2216.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4ꢀAminoꢀ6ꢀ(((1S)ꢀ1ꢀ(6ꢀfluoroꢀ4ꢀ((3ꢀhydroxyꢀ1ꢀazetidinyl)carbonyl)ꢀ3ꢀphenylꢀ2ꢀquinolinyl)ethyl)amino)ꢀ5ꢀ
1
pyrimidinecarbonitrile (5h). Obtained in 71% yield as a white solid. H NMR (400 MHz, CD3OD) δ ppm
8.17 ꢀ 8.25 (1 H, m), 8.00 (1 H, d, J=1.4 Hz), 7.85 (1 H, d, J=5.3 Hz), 7.39 ꢀ 7.73 (6 H, m), 5.42 ꢀ 5.70 (1 H, m),
4.01 ꢀ 4.65 (2 H, m), 3.40 ꢀ 3.99 (3 H, m), 1.20 ꢀ 1.58 (3 H, m). Mass Spectrum (ESI) m/z 484.2 [M+H]+.
HRMS (ESI) m/z calculated for C26H22FN7O2 [M+H]+: 484.1897; mass measured, 484.1899.
2ꢀ((1S)ꢀ1ꢀ((6ꢀAminoꢀ5ꢀcyanoꢀ4ꢀpyrimidinyl)amino)ethyl)ꢀ6ꢀfluoroꢀNꢀ(2ꢀ(methylamino)ethyl)ꢀ3ꢀphenylꢀ4ꢀ
1
quinolinecarboxamide (5i). Obtained in 99% yield as a white solid. H NMR (400 MHz, CD3OD) δ ppm 8.21
(1 H, dd, J=9.3, 5.4 Hz), 8.12 (1 H, s), 7.68 (1 H, ddd, J=9.3, 8.4, 2.8 Hz), 7.41 ꢀ 7.58 (6 H, m), 5.62 (1 H, q,
J=6.7 Hz), 3.35 ꢀ 3.48 (2 H, m), 2.64 ꢀ 2.69 (2 H, m), 2.62 (3 H, s), 1.45 (3 H, d, J=6.7 Hz). Mass Spectrum
(ESI) m/z 485.2 [M+H]+. HRMS (ESI) m/z calculated for C26H25FN8O [M+H]+: 485.2214; mass measured,
485.2213.
(S)ꢀMethyl 2ꢀ(1ꢀ(tertꢀbutoxycarbonylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (16). To a
solution of methyl 2ꢀ(1ꢀaminoethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate 12 (2.45 g, 7.55 mmol) in THF
(30 mL) was added Bocꢀanhydride (1.65 g, 7.55 mmol) followed by triethylamine (1.05 mL, 7.55 mmol) and
the reaction mixture was stirred at r.t. After 30 min, the solvent was concentrated in vacuo and the crude
product was purified by column chromatography on silica gel using 0ꢀ30% EtOAc/hexane to afford methyl 2ꢀ
(1ꢀ(tertꢀbutoxycarbonylamino)ethyl)ꢀ6ꢀfluoroꢀ3ꢀphenylquinolineꢀ4ꢀcarboxylate (2.47 g, 5.82 mmol, 77% yield)
as a white powder. Mass Spectrum (ESI) m/z 425.2 [M+H]+.
ACS Paragon Plus Environment