Discovery of (R)-8-(1-(3,5-difluorophenylamino)ethyl)- N, N -dimethyl-2-morpholino-4-oxo-4 H -chromene-6-carboxamide (AZD8186): A potent and selective inhibitor of PI3Kβ and PI3Kδ for the treatment of PTEN-deficient cancers

Article abstract of DOI:10.1021/jm501629p

Several studies have highlighted the dependency of PTEN deficient tumors to PI3Kβ activity and specific inhibition of PI3Kδ has been shown activity against human B-cell cancers. We describe the discovery and optimization of a series of 8-(1-anilino)ethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxamides as PI3Kβ/δ inhibitors, which led to the discovery of the clinical candidate 13, also known as AZD8186. On the basis of the lower lipophilicity of the chromen-4-one core compared to the previously utilized pyrido[1,2-a]pyrimid-4-one core, this series of compounds displayed high metabolic stability and suitable physical properties for oral administration. Compound 13 showed profound pharmacodynamic modulation of p-Akt in PTEN-deficient PC3 prostate tumor bearing mice after oral administration and showed complete inhibition of tumor growth in the mouse PTEN-deficient PC3 prostate tumor xenograft model. 13 was selected as a clinical candidate for treatment of PTEN-deficient cancers and has recently entered phase I clinical trials.

Full text of DOI:10.1021/jm501629p

Subscriber access provided by UPSTATE Medical University Health Sciences Library
Article
Discovery of (R)-8-(1-(3,5-difluorophenylamino)ethyl)-N,N-
dimethyl-2-morpholino-4-oxo-4H-chromene-6-carboxamide
(AZD8186): A Potent and Selective Inhibitor of PI3K#
and PI3K# for the treatment of PTEN-deficient cancers
Bernard Barlaam, Sabina Cosulich, Sebastien Degorce, Martina Fitzek, Stephen Green, Urs Hancox,
Christine Lambert-Van Der Brempt, Jean-Jacques Lohmann, Mickael Maudet, Remy Morgentin, Marie-Jeanne
Pasquet, Aurelien Peru, Patrick Plé, Twana Saleh, Michel Vautier, Mike Walker, Lara Ward, and Nicolas Warin
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501629p • Publication Date (Web): 16 Dec 2014
Downloaded from http://pubs.acs.org on December 18, 2014
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 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of (R)-8-(1-(3,5-difluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo-
4H-chromene-6-carboxamide (AZD8186): A Potent and Selective Inhibitor of PI3Kβ and
PI3Kδ for the treatment of PTEN-deficient cancers
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
,
b
b
b
b
b
Bernard Barlaam,* Sabina Cosulich, Sébastien Degorce, Martina Fitzek, Stephen Green,
b
a
a
a,1
Urs Hancox, Christine Lambertꢀvan der Brempt, JeanꢀJacques Lohmann, Mickaël Maudet,
a,2
a
a
a
a
Rémy Morgentin, MarieꢀJeanne Pasquet, Aurélien Péru, Patrick Plé, Twana Saleh, Michel
a
b
b
a
Vautier, Mike Walker, Lara Ward, Nicolas Warin
a
AstraZeneca, Centre de Recherches, Z.I. La Pompelle, B.P. 1050, Chemin de Vrilly, 51689
Reims, Cedex 2, France.
b
AstraZeneca Pharmaceuticals, Oncology iMed, Alderley Park, Macclesfield, Cheshire SK10
4
TG, United Kingdom.
Keywords: PI3Kβ, PI3Kδ, kinase, PTENꢀdeficient tumours
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Several studies have highlighted the dependency of PTEN deficient tumours to PI3Kβ activity
and specific inhibition of PI3Kδ has been shown activity against human Bꢀcell cancers. We
describe the discovery and optimization of a series of 8ꢀ(1ꢀanilino)ethyl)ꢀ2ꢀmorpholinoꢀ4ꢀoxoꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4Hꢀchromeneꢀ6ꢀcarboxamides as PI3Kβ/δ inhibitors, which led to the discovery of the clinical
candidate 13, also known as AZD8186. On the basis of the lower lipophilicity of the chromenꢀ4ꢀ
one core compared to the previously utilised pyrido[1,2ꢀa]pyrimidꢀ4ꢀone core, this series of
compounds displayed high metabolic stability and suitable physical properties for oral
administration. Compound 13 showed profound pharmacodynamic modulation of pꢀAkt in
PTENꢀdeficient PC3 prostate tumour bearing mice after oral administration and showed complete
inhibition of tumour growth in the mouse PTENꢀdeficient PC3 prostate tumour xenograft model.
13 was selected as a clinical candidate for treatment of PTENꢀdeficient cancers and has recently
entered phase I clinical trials.
ACS Paragon Plus Environment

Page 3 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The PI3KꢀAkt signalling pathway plays a critical role in cell growth, proliferation, motility and
survival. In human cancer, this pathway is activated by several mechanisms, including somatic
mutations, deletions and amplifications. A number of compounds targeting components of the
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
ꢀ4
PI3K pathway are currently in early clinical trials, but despite the significant efforts invested in
the generation of a number of inhibitors, it remains unclear which cancers will benefit from such
therapeutic approaches.
The phosphatidylinositolꢀ3ꢀkinases (PI3Ks) are a class of enzymes that catalyse the
5
phosphorylation of phosphoinositides at the 3ꢀhydroxyl position. PI3Ks are grouped into three
6
classes, based on their substrate specificity and structural features. Class I PI3Ks primarily
phosphorylate phosphatidylinositolꢀ4,5ꢀdiphosphate (PIP2) to produce phosphatidylinositol
3
,4,5ꢀtriphosphate (PIP3) whilst the tumour suppressor PTEN (phosphatase and tensin homolog)
7
reverses this process. PIP3 acts as a second messenger and activates the serine/threonine kinase
Akt and other downstream effectors, triggering critical pathways in cell growth, proliferation,
motility and survival.
Class I PI3Ks are further divided into class IA enzymes (PI3Kα, PI3Kβ and PI3Kδ) and class IB
enzymes (PI3Kγ). Class IA PI3Ks are heterodimeric, with a p110 catalytic unit and a p85
8
regulatory unit. The different class I isoforms of PI3K were originally thought to be redundant in
function and, as such, most of the PI3K inhibitors currently in clinical trials are panꢀPI3K
inhibitors, inhibiting all four PI3K isoforms. However, the main concern associated with pan
PI3K inhibitors is their tolerability profile due to the inhibition of all PI3K isoforms. Moreover,
some of the early pan PI3K clinical candidates also have offꢀtargets activity on other members of
the PI3Kꢀrelated kinase (PIKK) family. On the other hand, specific inhibition of different PI3K
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
isoforms may offer potential strategies for targeting a number of different tumour types and
maximising potential therapeutic window with potential for combination, while eliminating the
9
more deleterious effects which may arise from panꢀisoform inhibition. Individual distinct
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
functions for the individual PI3K isoforms have started to emerge. The expression of PI3Kα and
PI3Kβ is common in all tissues, but expression of PI3Kγ and PI3Kδ is mainly restricted to the
10
immune system. The PI3Kα isoform has been the most extensively studied to date. The gene
encoding for the p110α subunit, PIK3CA, is mutated in several cancers and amplified in a variety
of other cancer types. Therefore, inhibition of PI3Kα is considered as a promising approach for
1
1ꢀ12
cancer therapy.
The PI3Kδ isoform has been shown to play a critical role in Bꢀcell signalling
in response to a number of cytokines and chemokines. Specific inhibition of PI3Kδ has been
shown to have activity in human Bꢀcell cancers such as chronic lymphocytic leukemia (CLL) and
13ꢀ14
indolent nonꢀHodgkin lymphoma (iNHL).
Specific inhibition of PI3Kβ has been reported to block platelet responses to shear stress and to
15
regulate the formation of arterial thrombi. In the oncology field, no activation mutations have
been identified in p110β to date, but examples of gene amplification in breast and ovarian
cancers have emerged. Moreover, PTEN deficient tumours have been shown to depend on
PI3Kβ activity: conditional knockout of the PIK3CB, the gene coding for the p110β subunit but
not PIK3CA coding for p110α, blocks the formation of prostate tumours in a model of mouse
prostate tumour driven by PTEN loss. Additional studies have highlighted the dependency of
1
6ꢀ18
many PTEN deficient tumour cell lines to PI3Kβ activity.
Until recently, very few isoformꢀselective PI3K inhibitors, exhibiting appropriate drugꢀlike
properties, have been described. 1 (idelalisib, also known as GSꢀ1101 and CALꢀ101) was the first
isoform selective compound entering clinical trial for the treatment of cancers and, based on its
ACS Paragon Plus Environment

Page 5 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13,19
selective PI3Kδ pharmacology, is currently developed in hematological malignancies.
Most
pharmacology studies of PI3Kβ selective inhibition have relied on a series of compounds
20
21
described by Thrombogenix, as exemplified by 2 (TGXꢀ221) and 3 (PIKꢀ108). Subsequent
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
2
2
efforts at AstraZeneca resulted in the discovery of 4 (AZD6482), a PI3Kβ/δ inhibitor showing
inhibition of platelet aggregation suitable for short intravenous infusion in humans. Another
group showed that 4 (alternatively named KINꢀ193) displayed excellent tumour growth inhibition
associated with modulation of downstream markers in PTEN deficient tumours models after
23
intraperitoneal (i.p.) administration.
All these results have fueled a renewed interest in the area of PI3K isoform selective inhibitors,
2
4ꢀ28
including PI3Kβ selective inhibitors. Recently, many PI3Kβ selective inhibitors
have been
disclosed, including 2ꢀmethylꢀ1ꢀ[2ꢀmethylꢀ3ꢀ(trifluoromethyl)benzyl]ꢀ6ꢀ(morpholinꢀ4ꢀyl)ꢀ1Hꢀ
2
9
30,31
benzimidazoleꢀ4ꢀcarboxylic acid (GSK2636771) and 5 (SAR260301)
which are in early
clinical trials.
We previously reported the identification of a new series of pyrido[1,2ꢀa]pyrimidꢀ4ꢀones and our
3
2ꢀ33
efforts
to optimize this series as orally active PI3Kβ/δ inhibitors, leading to compounds 6 and
7
(Figure 1). Combining high potency, high solubility and high metabolic stability was a
significant challenge in the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone series, possibly because of the intrinsic
lipophilicity of the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone core. Therefore, we considered the replacement of
pyrido[1,2ꢀa]pyrimidꢀ4ꢀone core, ideally by more hydrophilic cores as a way to combine higher
solubility and metabolic stability of the series while maintaining potency. We previously reported
an homology model of PI3Kβ and the key interactions of pyrido[1,2ꢀa]pyrimidꢀ4ꢀone based
3
3
inhibitors with the PI3Kβ enzyme. In this model, the carbonyl group of pyrido[1,2ꢀa]pyrimidꢀ4ꢀ
oneꢀbased inhibitors makes a key Hꢀbond interaction with the protein. The exact residue involved
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
is not fully clarified: initially, from our homology model, we anticipated that the catalytic Lys805
3
3
was involved in the Hꢀbond interaction, but another group also reported that their homology
model suggested that the carbonyl of pyrido[1,2ꢀa]pyrimidꢀ4ꢀoneꢀbased 2 made an Hꢀbond
interaction, not with catalytic Lys805, but with the backpocket Tyr839 via a network of hydrogen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
bonds involving a bridge water molecule. Our crystal structure of 4 in PI3Kγ, did not shed
3
2
clarity of the exact residue involved. Regardless of the exact residue on the protein involved in
the interaction, we considered that the presence of a carbonyl in an alternative core at the same
position as in the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone core was essential for activity. More recently, the
crystal structure of the chromenꢀ4ꢀoneꢀbased inhibitor 3 in PI3Kα was reported, supporting that
the binding mode of chromenꢀ4ꢀoneꢀ and pyrido[1,2ꢀa]pyrimidꢀ4ꢀoneꢀbased inhibitors are
34
similar. Among the different cores we considered, the chromenꢀ4ꢀone core was seen as a viable
option, despite contradicting lipophilicity predictions by clogP, ACDlogP and AstraZeneca’s
global lipophilicity model AZlogD. cLogP and ACDlogP consistently predicted the chromenꢀ4ꢀ
ones to be more lipophilic than pyrido[1,2ꢀa]pyrimidꢀ4ꢀones, whereas AZlogD predicted them to
be isoꢀlipophilic, with the exception of the acidic matched pair of 4 (Table 1).
In this article, we report the optimization of a series of 8ꢀ(1ꢀanilinoethyl)ꢀ2ꢀmorpholinoꢀ4ꢀoxoꢀ
chromeneꢀ6ꢀcarboxamides and the discovery of 13, also known as AZD8186, as a clinical
candidate currently in phase I clinical trials for the treatment of PTEN deficient tumours.
ACS Paragon Plus Environment

Page 7 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHEMISTRY
General methods of synthesis of the 2ꢀmorpholinoꢀchromenone core reported by Morris
3
5,37
or
alternatively via the 2ꢀalkylthiochromenone intermediate were adapted to make compounds 8-
3. More specifically, compounds 8-43 were prepared as shown in Schemes 1ꢀ7 or by analogous
3
8
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
4
methods. The synthesis of compounds 8-34 and 37-43 shares the same intermediate 48. Two
routes to 48 were developed, both starting from methyl 3ꢀbromoꢀ4ꢀhydroxybenzoate 44,
commercially available and readily accessible from methyl 4ꢀhydroxybenzoate by bromination
(Scheme 1). The first route developed relied on the Fries rearrangement of 45 to give 46 after reꢀ
3
5
esterification. 46 was then reacted with known (dichloromethylene)morpholinꢀ4ꢀium chloride
3
5
4
7 in the presence of BF .Et O, following a method developed by Morris, to give 48 in good
3 2
yield. However, this route suffered from drawbacks for synthesis on larger scale (hundreds of
grams or more): the Fries rearrangement of 45 involves high temperature and solventꢀfree
conditions. The shelf life of moistureꢀsensitive 47 is questionable and its synthesis involved use
of chlorine gas and release of obnoxious sulphur derived sideꢀproducts. Therefore, we developed
an alternative synthesis which also started from 44. Heck coupling with 1ꢀ(vinyloxy)butane
3
6
afforded previously known 49, which, after bromination, gave 46, a intermediate in the first
sequence. Enolate formation with LHMDS and trapping with morpholineꢀ4ꢀcarbonyl chloride
afforded the βꢀketoamide 50, which was cyclised in the presence of triflic anhydride to give 48,
3
7
according to an alternative method developed by Morris. This sequence was scaled up to multiꢀ
39
tens of kilogramme quantities of 48. With this key intermediate in hand, several routes were
developed to the final compounds. The first route (Schemes 2 and 3), which allowed variation of
the amide at the final step, used the following sequence. 48 was transformed into 51 via a Stille
reaction with tributyl(1ꢀethoxyvinyl)stannane. Conversion of 51 to the bromide 53 and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
subsequent halide displacement with the corresponding anilines gave the esters 54 and 65, which
were then readily converted to the corresponding amides 8, 23-34. Compounds 27 and 32
(isolated as a single enantiomer) were made from the enantiomerically pure carboxylic acid 66a,
obtained by chiral purification from racemic carboxylic acid 66.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The second route allowed variation of the aniline at the last step (Scheme 4). Again, starting from
51, the dimethylamide functionality was installed first to give the key intermediate 58, which was
then reacted with the desired aniline to give 9-16, 18-20 and 22. Compounds 12, 13, 15, 22
(isolated as a single enantiomer) were isolated by separation of the enantiomers at the last step.
Alternatively, an enantioselective route was developed (Scheme 5). This route relied on having
access to the enantiomerically pure amine 64. The key intermediate 48 was converted into the
methyl ketone 61, following standard group manipulation. Reaction of 61 with (R)ꢀ2ꢀtertꢀbutylꢀ2ꢀ
40
sulfinamide in the presence of titanium tetraisopropoxide gave imine 62. Reduction of 62 with
4
1
sodium cyanoborohydride and cleavage of the sulfinyl group afforded amine 64 of (R)ꢀ
4
2
configuration. Buchwald or ChanꢀLam reaction of 64 respectively with the corresponding aryl
bromide or the boronic acid, afforded 17 and 21. The configuration of 64 and therefore, of 17 and
21 was proven by Buchwald reaction of 64 with 1ꢀbromoꢀ3,5ꢀdifluorobenzene, giving 13 of
known configuration.
Compounds 35-1 and 36-1 were made according to Scheme 6 via the alkylthiochromenone
method. The dianion of 46 was treated with carbon disulfide to give the thione 67. SꢀAlkylation
with ethyl iodide followed by oxidation with mꢀCPBA afforded the key sulfone 69. Displacement
of the ethyl sulfonyl group with (R)ꢀ or (S)ꢀ2ꢀmethylmorpholine gave 70a and 70b respectively.
70a was subjected to a similar sequence as in Scheme 3 to give 35-1 of (R,R) configuration and
35-2 of (R,S) configuration. Similarly, 70b gave 36-1 of (S,R) configuration and 36-2 of (S,S)
configuration.
ACS Paragon Plus Environment

Page 9 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 39 was made by two different methods (Scheme 7). The first method started from 53.
Bromide displacement with Nꢀmethylꢀ3,5ꢀdifluoroaniline followed by standard functional group
manipulation and chiral chromatography at the final step afforded 39. Alternatively, methylation
of 13 with dimethyl sulfate in the presence of LHMDS afforded 39, proving its (R) configuration.
The other compounds 37-38 and 40-43 were made by analogy with the Schemes presented.
Detailed schemes and procedures are available in the Supporting Information.
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
For all compounds evaluated as a pure enantiomer (i.e. compounds 12, 13, 15, 17, 21, 22, 27, 32,
35-1, 36-1, 39, 40, 42, 43), the configuration of the carbon bearing the aniline was determined to
be (R) as follows. Compounds 13, 25, 32, 39, 42 and 43 were made from 66a. The absolute
configuration of 66a was determined to be (R) by solving of the structure of 76, the (S)ꢀ1ꢀ
phenylethylamide of 66a. The structure of 76 was determined by Xꢀray singleꢀcrystal diffraction. 17
and 21 were made from enantiomerically pure amine 64 of (R) configuration. The configuration of 64
was determined by synthesis of 13 of known configuration from 64 by Buchwald coupling with 1ꢀ
bromoꢀ3,5ꢀdifluorobenzene. The configuration of 12, 15, 22, 35-1, 36-1, 40 was determined by
vibrational circular dichroism (VCD). The configuration of 17 and 39 was also confirmed by
VCD
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Compounds 8-43 were evaluated either as a racemate or a pure enantiomer, as indicated in Tables
2ꢀ6. Their biological activity is reported in the same Tables. Selectivity among the PI3K isoforms
was routinely measured with enzyme assays. α/β selectivity was also monitored in cell assays
(inhibition of Akt phosphorylation at Thr308 in PIK3CA mutant human breast ductal carcinoma
BT474 cells sensitive to PI3Kα inhibition and at Ser473 in PTENꢀnull breast adenocarcinoma
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MDAꢀMBꢀ468 cells sensitive to PI3Kβ inhibition). When compounds were isolated as a pure
enantiomer by chiral separation at the final step (i.e. compounds 12, 13, 15, 22, 32, 35-1, 36-1,
39, 40), the PI3Kβ biological activity resides preferably in one isomer, with the other being in
general more than 100 fold less active both in PI3Kβ enzyme and cell assays (unpublished data).
2
6
22
33
As previously reported for the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone series, i.e. (R)ꢀ2, 4 and 6, the
configuration of the carbon bearing the aniline was determined as (R) for all compounds
evaluated as a pure enantiomer in this series (i.e. compounds 12, 13, 15, 17, 21, 22, 27, 32, 35-1,
36-1, 39, 40, 42, 43).
The first chromenꢀ4ꢀone 8, bearing a 4ꢀfluoroaniline and evaluated as a racemate, maintained
3
3
similar PI3Kβ potency as its pyrido[1,2ꢀa]pyrimidꢀ4ꢀone matched pair (e.g. PI3Kβ cell: IC
5
0
0.128 ꢀM for 8 vs. 0.062 ꢀM for the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone matched pair, (–)ꢀenantiomer),
but 8 had lower lipophilicity (log D 1.5 for 8 vs. 2.3 for its pyrido[1,2ꢀa]pyrimidꢀ4ꢀone matched
7
.4
pair). The reduced lipophilicy is a general feature of this series: an analysis involving 54 matched
pairs of chromenꢀ4ꢀones and pyrido[1,2ꢀa]pyrimidꢀ4ꢀones made in this project showed an
average log D7.4 decrease of –0.69 (+/- SEM: 0.02) (see Figure 2). This retroꢀanalysis also
confirmed the counterꢀperformance of clogP and ACDlogP which predicted the opposite trend,
and of our global model AZlogD which predicted an isoꢀlipophilic transformation. As an
ACS Paragon Plus Environment

Page 11 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
illustration, the difference of measured log D for the third pair in Table 1 (i.e. compounds 7 and
7
.4
12) is –0.8.
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In a similar way as for 2, this series of chromenꢀ4ꢀones adopts a Tꢀshape conformation and the
aniline portion sits in a small lipophilic pocket limited by Trp787 and Met779. Based on the
homology model of PI3Kβ we have previously reported, we anticipated that further substitution
at the 4ꢀposition of the aniline would be very limited because of the absence of space at the
bottom of this pocket, whereas some space would be available from the 2ꢀ and 3ꢀpositions (see
Figure 3). This hypothesis is supported by the SAR in the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone series,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
33
where a substituent larger than a fluorine in the 4ꢀposition showed reduced potency.
With the SAR we had developed in the pyrido[1,2ꢀa]pyrimidꢀ4ꢀone series in mind, we first
concentrated our work on varying the substituents on the aniline while keeping the
dimethylamide at Cꢀ6. Overall, the same general trends were observed between the two series.
The 3ꢀfluoro 9 or 3ꢀchloro 10 both increased PI3Kβ potency on cell compared to 4ꢀfluoro 8. Very
little potency increase was seen in the PI3Kβ enzyme assay, because the assay, as configured, has
43
reached the tight binding limit for inhibitors with this potency level. Bisꢀsubstitution of the
aniline was very fruitful: whereas substitution with 2,3ꢀdifluoro 11, 3,4ꢀdifluoro 12 or 2.5ꢀ
difluoro 14 did not increase PI3Kβ potency, the 3,5ꢀfluoro 13 showed a significant increase of
potency vs. the 3ꢀfluoro 9 (PI3Kβ cellular IC 0.003 ꢀM for 13 vs. 0.032 ꢀM for 9). This
5
0
pattern was confirmed starting from the 3ꢀchloro 10: whereas the 2ꢀfluoro 3ꢀchloro 15 or the 4ꢀ
fluoro 3ꢀchloro 16 had similar potency in the PI3Kβ cellular assay as 10, the 3ꢀchloro 5ꢀfluoro 17
also showed improved potency in the PI3Kβ cellular assay (IC 0.001 ꢀM). We further explored
5
0
the 3,5ꢀsubstitution pattern with additional compounds: small halogens such as fluoro and chloro
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were preferred over methyl or methoxy (see 13 and 17 vs. 18 and 19) and the 3,5ꢀdichloro 20 had
reduced potency compared to the 3,5ꢀdifluoro 13 or the 3ꢀchloro 5ꢀfluoro 17.
The reduced lipophilicity of the chromenꢀ4ꢀone core would appear to correlate with the lower
intrinsic clearance (Cl ) in human hepatocytes and the reduced apparent permeability (P ) seen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
int
app
for many examples of this series. We had reported that mediumꢀhigh intrinsic clearance in human
3
3
hepatocytes had plagued our optimization of the pyrido[1,2ꢀa]pyrimidꢀ4ꢀones series. Based on
an analysis of 27 matched pairs (Figure 4, right panel), most compounds in the chromenꢀ4ꢀone
series showed lower Clint than their pyrido[1,2ꢀa]pyrimidꢀ4ꢀone matched pair. Only more
lipophilic compounds in the chromenꢀ4ꢀone series, such as 17 (3ꢀCl 5ꢀF; log D 2.8) or 20 (3,5ꢀ
7
.4
6
Cl; log D 3.2) showed medium/high Cl (Cl resp. 12 and 18 ꢀL/min/10 cell for 17 and 20).
7
.4
int
int
Because of the reduced lipophilicy of the chromenꢀ4ꢀones, permeability was a parameter we
monitored. In a similar matched pair analysis as above from 9 matched pairs, permeability as
measured by Cacoꢀ2 Papp was reduced in the chromenꢀ4ꢀones vs. the pyrido[1,2ꢀa]pyrimidꢀ4ꢀones
(Figure 4, left panel). Moderate permeability was seen with compounds 12 and 13 (Cacoꢀ2 Papp
ꢀ6
ꢀ1
ꢀ6
ꢀ1
6
.5.10 cm.s and 8.10 cm.s resp.). Introduction of a fluorine at the ortho position increased
ꢀ6
ꢀ1
ꢀ6
ꢀ1
ꢀ6
ꢀ1
permeability (i.e. 13.10 cm.s and 12.10 cm.s vs. 6.5.10 cm.s for 11 (2,3ꢀF) and 14 (2,5ꢀF)
ꢀ6
ꢀ1
vs. 12 (3,4ꢀF)). High permeability (Cacoꢀ2 Papp 21.10 cm.s ) was seen for 15 (2ꢀF 3ꢀCl),
combining the effect of the ortho fluorine and the increased lipophilicity due to the chlorine. This
observation was further utilized with 22 (2,3,5ꢀF) which exhibited high permeability (Cacoꢀ2 Papp
ꢀ6
ꢀ1
1
6.10 cm.s ), while keeping the high potency and high metabolic stability of 13 (3,5ꢀF).
Overall, at this stage the 3,5ꢀdifluoroaniline (as in 13) was seen as a suitable aniline combining
high potency, human metabolic stability and sufficient permeability.
ACS Paragon Plus Environment

Page 13 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
Next, we turned our attention to the amide portion of the molecule. Whereas the dimethyl amide
1
3 was highly potent, a variety of tertiary amides (23-26) were tolerated, but none showed any
potency improvement. Rings were also tolerated, as illustrated by azetidine 27 and pyrrolidine 28
having similar potency as dimethylamide 13. But larger rings such as piperidines 29 and 31 and
morpholine 30 had slightly reduced potency. Secondary amides were also tolerated (i.e. 32-34).
But permeability was reduced to an unsatisfactory level, when adding an additional Hꢀbond
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ6
ꢀ1
ꢀ6
donor (either hydroxyl or amidic NH): for instance, Cacoꢀ2 Papp resp. 1.0.10 cm.s , 1.0.10
ꢀ1
ꢀ6
ꢀ1
cm.s and 3.5.10 cm.s for 25 (CON(Me)CH CH OH), 31 (COꢀpiperidinꢀ1ꢀylꢀ4ꢀOH) and 32
2
2
ꢀ6
ꢀ1
(
CONHMe) and plummeted to 0.28.10 cm.s for 33 (CONHCH CH OH), having two
2 2
additional Hꢀbond donors.
According to our homology model of PI3Kβ, the morpholine at the 2ꢀposition of the chromenꢀ4ꢀ
one forms a critical interaction in the kinase hingeꢀregion with Val854, in a similar way as for
other morpholine containing inhibitors of the PIKK family. We considered the potential effects
on the PI3K activity when adding substituents on the morpholine. However, incorporating chiral
methylꢀsubstituted morpholines and/or bridged morpholines was a significant synthetic effort in
the chromenꢀ4ꢀone series, as previously shown in the Synthetic Chemistry Section. Therefore, an
initial exploration was performed on a synthetically more readily accessible core (unreported
results) and only the most interesting substituted morpholine (i.e. (R)ꢀ and (S)ꢀ 2ꢀMeꢀmorpholine)
are reported in the chromenꢀ4ꢀone series here. After the course of this work, other groups have
also reported the effects of morpholine substitution in some members of the PIKK family
4
4
45
26
(ATR , mꢀTOR , PI3Kβ ).
In the chromenꢀ4ꢀone series, the (R)ꢀ2ꢀMeꢀmorpholine 35-1 showed a significant increase of
potency against PI3Kβ compared to the unsubstitued morpholine 13. On the other hand, the (S)ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
ꢀMeꢀmorpholine 36-1 had slightly reduced activity against PI3Kβ compared to 13. As the hinge
region is a highly conserved region between the four PI3K isoforms, we did not expect selectivity
changes among the four PI3K isoforms. Indeed, it is noteworthy that these modifications had
similar effects on all isoforms, with the limitation that the PI3Kβ and PI3Kδ and even PI3Kα
enzyme assays reached ‘tight binding conditions’ for some of the most potent inhibitors, in
particular for 35-1. At the cellular level, the results were striking: 35-1 showed increased potency
on PI3Kα and PI3Kβ (~7 fold) compared to 13 whereas 36-1 was slightly less potent both on
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PI3Kα and PI3Kβ (~4 fold). All three compounds maintained the same high degree of selectivity
(~ 200ꢀ300 fold) between the two isoforms.
We also explored the linker between the aniline and the chromenꢀ4ꢀone core. Compound 37, the
analogue of 13 where the methyl on the carbon bearing the aniline is removed, or 38 (Nꢀ
methylation of 37) lost significant affinity for PI3Kβ (~10 fold at the cellular level) and some
degree of selectivity vs. PI3Kα (PI3Kβ/ PI3Kα ratio at the cellular level: ~100 fold for 37 and
~25 fold for 38). Introducing an ether linker instead of a nitrogen linker reduced potency against
PI3Kβ significantly (see 40 vs. 13). However, an interesting aspect was the excellent
permeability observed for compounds 38 and 40, certainly due to the removal of an Hꢀbond
ꢀ6
donor. Indeed, 39 (Nꢀmethylation of 13) showed very high permeability too (Cacoꢀ2 Papp 42.10
ꢀ1
cm.s ). Compound 39 exhibited high potency in the PI3Kβ cell assay, although it was slightly
reduced compared to 13. Selectivity against PI3Kα (cellular PI3Kβ/ PI3Kα ratio ~200 fold) and
PI3Kγ (PI3Kγ enzyme IC 2.8 ꢀM) and high metabolic stability were maintained. It is difficult
50
to fully rationalize the role of the methyl on the carbon bearing the aniline for selectivity, but the
constraints induced by the presence of this methyl may limit the conformational flexibility of the
molecule and favour a Tꢀshape conformation over a flatter conformation. The greater plasticity of
ACS Paragon Plus Environment

Page 15 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PI3Kδ and probably of PI3Kβ may enable these isoforms to more readily accommodate rigid Tꢀ
4
6,47
shaped molecules, by opposition to PI3Kα and PI3Kγ.
Amide modifications of 39 were also
well tolerated, giving similar cell activity than the dimethyl amide (see 41-43). In addition, in this
case, it is worth noting that amides containing a hydroxyl group (i.e. 42-43) maintained
acceptable permeability.
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
Pharmacokinetics of representative compounds from this series were evaluated in mice (Table 7).
In most cases, the observed in vivo clearance was well predicted from mouse hepatocytes in vitro.
Selected compounds (13, 22, 35-1 and 39) were further profiled in dog pharmacokinetics,
confirming that in vivo clearance in dogs was essentially predicted by in vitro hepatocyte
clearance. Physical properties and safety properties of 13 and 39 were further investigated. For
instance, 13 (from crystalline material) had suitable aqueous solubility: 108 ꢁg/mL in FASSIF
(Fasted Simulated Intestinal Fluid) and 80 ꢁg/mL in aqueous buffer (pH 6.5). 39 also displayed
good solubility: 350 ꢀM in aqueous phosphate buffer (pH 7.4), but as an amorphous material.
Despite numerous attempts, all our attempts of crystallization of 39 failed. 13 did not inhibit the
hERG channel at 100 ꢀM whereas some inhibition (IC 19 ꢀM) was seen with 39. Previous
5
0
reports had highlighted the risk of formation of glutathione adducts with compounds containing
anilines, especially with the 4ꢀfluoroanilino motif, via a mechanism of oxidative defluorination of
4
8,49
the aniline leading to the intermediate reactive quinoneimine.
Trapping experiments of 13
with glutathione did not show any evidence of formation of glutathione adducts in human liver
microsomes.
The overall kinase selectivity of 13 was evaluated in several panels of recombinant protein and
lipid kinase assays. 13 did not show significant activity against a panel of 59 protein kinases
tested at 1 ꢁM (Dundee protein kinase panel). Furthermore, there was no significant activity seen
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in a panel of 11 lipid kinases (Dundee lipid kinase panel) and 4 lipid kinases (InVitrogen), apart
from the class I PI3K family. The kinase selectivity of 13 was confirmed in the commercial
TM
KINOMEscan
at Ambit against a panel of 442 kinases, showing the compound to be a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
selective inhibitor of PI3Kβ and PI3Kδ with no significant binding at 10µM against the 442
kinases tested outside the PI3K family. The selectivity profile of 13 among the PI3K isoforms
was tested in cell based assays by assessing the ability to inhibit Akt phosphorylation: 13 gave
potent inhibition of pꢀAkt in cells sensitive to PI3Kβ inhibition (IC 0.003 ꢀM (+/- SEM: 0.001)
5
0
in PTENꢀnull breast adenocarcinoma MDAꢀMBꢀ468 cells) and in cells sensitive to PI3Kδ
inhibition (IC 0.017 ꢀM (+/- SEM: 0.004) in Jeko B cell line), but not to cells sensitive to
5
0
PI3Kα inhibition (IC50 0.752 ꢀM (+/- SEM: 0.293) in PIK3CA mutant human breast ductal
carcinoma BT474).
13 was selected for further in vivo evaluation. 13 had a short half life in mouse because of its
medium/high mouse clearance associated with a low volume of distribution. The good in vitro-in
vivo clearance correlation seen in preclinical species (mouse and dog) suggested that clearance
was essentially metabolic in the liver, presumably CYP P450 mediated. Indeed, coꢀadministration
of 1ꢀaminobenzotriazole (ABT), an irreversible inhibitor of cytochrome P450 metabolizing
5
0,51
enzymes known to reduce CYP P450ꢀmediated clearance in vivo,
increased AUC and
provided more prolonged exposure (Figure 5).
In an acute pharmacodynamic experiment following a single oral dose (100 mg/kg and 25 mg/kg
of 13) in nude mice bearing PTENꢀdeficient PC3 prostate tumour xenografts, target modulation
was assessed by measuring Akt phosphorylation levels at Ser473 at 30 minutes, 4 and 8 hours. At
both doses, strong inhibition of pꢀAkt was observed at early time points, with still significant
inhibition at 8 hours (see Figure 6).
ACS Paragon Plus Environment

Page 17 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The antitumour activity of 13 was evaluated in the PTENꢀdeficient PC3 prostate tumour
xenograft model in nude mice, either at 100 mg/kg b.i.d. without ABT coꢀadministration, or at
60, 30 and 10 mg/mg b.i.d. with ABT coꢀadministration: complete inhibition of tumour growth
was seen at 60 mg/kg with ABT, with evidence of doseꢀresponse (Figure 7). Higher inhibition
was seen in the 60 mg/kg and 30 mg/kg groups with ABT coꢀadministration than in the 100
mg/kg group without ABT coꢀadministration, consistent with the increased, more prolonged
exposure.
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 18 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSION
We have designed a series of 8ꢀ(1ꢀanilino)ethyl)ꢀ2ꢀmorpholinoꢀ4ꢀoxoꢀ4Hꢀchromeneꢀ6ꢀ
carboxamide as PI3Kβ/δ inhibitors. Optimization of the aniline and the amide substituents led to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
identification of potent PI3Kβ/δ inhibitors with excellent selectivity vs. PI3Kα and PI3Kγ. Based
on the lower lipophilicity of the chromenꢀ4ꢀone core compared to the previously described
pyrido[1,2ꢀa]pyrimidꢀ4ꢀone core, higher metabolic stability and suitable physical properties for
oral administration was seen across the series. Compound 13 showed profound
pharmacodynamic modulation of pꢀAkt in PTENꢀdeficient PC3 prostate tumour bearing mice
after oral administration and showed complete inhibition of tumour growth in the mouse PTENꢀ
deficient PC3 prostate tumour xenograft model. 13 was selected as a clinical candidate and has
recently entered phase I clinical trials. Further in vitro and in vivo biological characterization of
5
2
13 has recently been reported.
ACS Paragon Plus Environment

Page 19 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MATERIALS AND METHODS
General methods
All experiments were carried out at ambient temperature under an inert atmosphere. Evaporations
were carried out by rotary evaporation or utilising Genevac equipment in vacuo and workꢀup
procedures were carried out after removal of residual solids by filtration. Flash chromatography
purifications were performed on an automated Armen Glider Flash: Spot II Ultimate (Armen
Instrument, SaintꢀAve, France) using prepacked Merck normal phase Si60 silica cartridges
(granulometry: 15ꢀ40 or 40ꢀ63 µm) obtained from Merck, Darmstad, Germany. Preparative
chromatography was performed on a Waters instrument (600/2700 or 2525) fitted with a ZMD or
ZQ ESCi mass spectrometers and a Waters XꢀTerra or a Waters XꢀBridge or a Waters SunFire
reverseꢀphase column (Cꢀ18, 5 µm silica, 19 mm diameter, 100 mm length, flow rate of 40
mL/min) using decreasingly polar mixtures of water (containing 0.2% ammonium carbonate) and
acetonitrile as eluent. Intermediates were not fully purified but their structure and purity were
assessed by TLC, NMR, HPLC and mass techniques and are consistent with the proposed
structures. The purities of the compounds for biological testing were assessed by NMR, HPLC
and mass spectral techniques and are consistent with the proposed structures characterized; purity
was at least 95%. Unless stated otherwise proton magnetic resonance spectra were determined
using a Bruker Avance 500 (500 MHz) instrument. Measurements were taken at ambient
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
temperature unless otherwise specified. H NMR chemical shift values are reported as δ values
(ppm) downfield from internal TMS in appropriate organic solutions. The following
abbreviations have been used: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; dd, doublet
of doublets; ddd, doublet of doublet of doublet; dt, doublet of triplets; bs, broad signal. Analytical
HPLC was carried out using an Waters Alliance HT (2790 & 2795) fitted with a Waters ZQ ESCi
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
or ZMD ESCi mass spectrometer and an X Bridge 5ꢁm Cꢀ18 column (2.1 x 50 mm) at a flow
rate of 2.4 mL/min, using a solvent system of 95% A + 5% C to 95% B + 5% C over 4 minutes,
where A = water, B = methanol, C = 1:1 methanol:water (containing 0.2% ammonium
carbonate). Detection was by Electrospray Mass Spectrometry and by UV absorbance at a wave
length of 254 nm. Accurate mass spectra were recorded on a Thermo LTQꢀFT in +ve ion mode
with a thermo Accela pump and a Surveyor PDA + with a CTC autosampler. Combustion
analyses (C, H, N) were performed with a CarloErba EA 1108 analyser. Water was measured by
the KarlꢀFisher method using a Mettler DL 18.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Methyl 3-bromo-4-hydroxybenzoate (44)
Bromine (0.185 L, 3.6 mol) was added dropwise a stirred suspension of methyl 4ꢀ
hydroxybenzoate (500 g, 3.29 mol) in DCM (4 L) at 0°C under nitrogen. The mixture was stirred
for 24 h at rt. A solution of sodium metabisulfite (62.5 g, 329 mmol) in 2L of water was then
added, while keeping the temperature around 15°C, followed by 500 mL of MeOH. The organic
layer was washed with water, brine, dried over magnesium sulfate, filtered and concentrated to
1
dryness to afford 44 (710 g, 94%) as a white solid. H NMR (CDCl ): 3.89 (s, 3H), 5.95 (s, 1H),
3
ꢀ
7.05 (d, 1H), 7.92 (dd, 1H), 8.19 (d, 1H). MSꢀESI m/z 229 [MꢀH]
Methyl 4-acetoxy-3-bromobenzoate (45)
To a stirred suspension of 44 (270 g, 1168 mmol) in DCM (1.5 L) was added pyridine (150 mL).
Acetyl chloride (87 mL, 1227 mmol) was then added dropwise at room temperature and under
nitrogen. The mixture was left to stir for 2 h at rt. Water (1 L) was then added followed by aq.
2N HCl until pH 1. The organic layer was then washed with water, brine, dried over magnesium
1
sulfate, filtered and evaporated to dryness to afford 45 (300 g, 94 %) as a white powder. H NMR
ACS Paragon Plus Environment

Page 21 of 79
Journal of Medicinal Chemistry
1
2
3
(
DMSOꢀd ): 2.34 (s, 3H), 3.87 (s, 3H), 7.47 (d, 1H), 8.01 (dd, 1H), 8.20 (d, 1H). MSꢀESI m/z 229
6
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ
[
MꢀAc]
Methyl 3-acetyl-5-bromo-4-hydroxybenzoate (46)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A mixture of 45 (150 g, 549 mmol) and aluminum trichloride (220 g, 1648 mmol) was heated
and stirred at 140°C in the absence of solvent for 3 h. Upon cooling to room temperature the
solid was crushed and cautiously added to water (1.5 L) with stirring. HCl (250 mL, 12N) was
then added and stirring was maintained for 30 min. The solid obtained was collected by
filtration, washed with water (2 x 2 L) and dried overnight to afford crude 3ꢀacetylꢀ5ꢀbromoꢀ4ꢀ
ꢀ
hydroxybenzoic acid (120 g, 84 %) as a yellow powder. MSꢀESI m/z 258 [MꢀH] . To a
suspension of 3ꢀacetylꢀ5ꢀbromoꢀ4ꢀhydroxybenzoic acid (240 g, 926 mmol, obtained by repeated
the reaction above several times) in MeOH (2 L) was added dropwise thionyl chloride (68 mL,
9
26 mmol) under nitrogen. The mixture was heated at 80°C for 3 h. The reaction mixture was
cooled to room temperature concentrated, diluted with DCM. The organic layer was washed with
brine, dried over magnesium sulfate, filtered and concentrated to afford a crude solid, which was
purified on silica, eluting with 70% of DCM in petroleum ether. The solvents were evaporated to
1
dryness to afford 46 (108 g, 42.7 %) as a white powder. H NMR (DMSOꢀd ): 2.76 (s, 3H), 3.88
6
ꢀ
(
s, 3H), 8.31 (s, 1H), 8.45 (s, 1H). MSꢀESI m/z 271 [MꢀH] .
An alternative procedure is as follows. To a stirred solution of 49 (240 g, 1236 mmol) in DCM (2
L) was added pyridine (0.400 L, 4944 mmol) followed by a dropwise addition of bromine (0.070
L, 1360 mmol) at 0°C. The reaction mixture was stirred at rt for 2 h then cooled to 5°C and aq.
4
N HCl (0.927 L, 3708 mmol) was added dropwise. The organic phase was separated, dried over
magnesium sulfate, filtered and concentrated to afford a brown solid which was stirred in
ether/petroleum ether (1:1, 1 L) for 1 h. The solid was collected by filtration and dried to afford
4
6 (270 g, 80%) as a beige powder, characterised above.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Methyl 8-bromo-2-morpholino-4-oxo-4H-chromene-6-carboxylate (48)
To a stirred solution of 46 (106 g, 388 mmol) in toluene (1L) was added dropwise BF .Et O
3
2
(0.201 L, 1630 mmol) under nitrogen. The resulting solution was stirred for 18 h at rt, then 4ꢀ
dichloromethylene)morpholinꢀ4ꢀium chloride 47 (143 g, 698 mmol) was added and the mixture
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
heated at 90°C for 12 h. Upon cooling to room temperature, diethyl ether (1.5 L) was added and
the solid was collected by filtration. This solid was then suspended in MeOH (1 L) and the
mixture was heated at 50°C for 2 h. Upon cooling to room temperature, the solid was collected
by filtration then solubilised in DCM (1 L) and washed with water and sat. aq. sodium
bicarbonate. The organic layer was dried over magnesium sulfate, filtered and evaporated to
1
dryness to afford 48 (68.0 g, 47.6 %) as an offꢀwhite solid. H NMR (DMSOꢀd ): 3.60 (m, 4H),
6
+
79
3.85 (m, 4H), 3.90 (s, 3H), 5.63 (s, 1H), 8.32 (s, 1H), 8.41 (s, 1H). MSꢀESI m/z 368 [MH , Br]
+
81
and 370 [MH , Br].
An alternative procedure is as follows. Trifluoromethanesulfonic anhydride (0.755 L, 4487
mmol) was added to a stirred solution of 50 (433 g, 1122 mmol, pooled material from several
batches) dissolved in 1,2ꢀdichloroethane (1 L) at rt (exotherm). The resulting solution was stirred
at 50°C for 18 h. The mixture was partially evaporated, and the residue was diluted with MeOH
(1.6 L) at 0°C (exotherm) and stirred for 1 h at rt. The solvent was evaporated again and the
residue was diluted in DCM, quenched with saturated aqueous sodium carbonate and extracted
with DCM. The combined organic phases were washed with brine, dried over magnesium sulfate
and concentrated to afford the crude product. The crude was triturated under MTBE (2x), ethyl
acetate (1x) and MTBE (1x). The solid was dried to afford 48 (208 g, 50%) characterised above
as a beige solid.
Methyl 3-acetyl-4-hydroxybenzoate (49)
ACS Paragon Plus Environment

Page 23 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To a degassed solution of 44 (350 g, 1.51 mol) in ethanol (3 L) were added triethylamine (0.528
L, 3.79 mol), 1ꢀ(vinyloxy)butane (0.588 L, 4.54 mol), 1,1'ꢀbis(diphenylphosphino)ferrocene (33.1
g, 60.6mmol) and palladium(II) acetate (8.50 g, 37.9 mmol) under nitrogen. The mixture was
heated at 70°C overnight. The reaction was cooled down, filtered and the filtrate concentrated.
The resulting solid was solubilised with DCM (2 L) and 4N aq. HCl (1.14 L, 4.54 mol) was
added under stirring. Stirring was maintained for 2 h, the organic phase was separated, dried over
magnesium sulfate, filtered and concentrated to afford a solid which was stirred in diethyl ether
(5 L) for 2 h. The solid was filtered off and the filtrate concentrated to dryness to afford 49 (240
g, 82%) as a beige powder.
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 3-bromo-4-hydroxy-5-(3-morpholino-3-oxopropanoyl)benzoate (50)
To a solution of lithium bis(trimethylsilyl)amide (1.41 L, 1.41 mol, 1M in THF) at ꢀ65°C under
nitrogen was added dropwise a sol. of 46 (120 g, 0.44 mol) in THF (1.2 L). The solution was
allowed to warm to 0°C, and maintained at this temperature for 1 h. The solution was cooled
back to ꢀ65°C and morpholineꢀ4ꢀcarbonyl chloride (0.055 L, 483 mmol) was added. The mixture
was stirred at rt for 2 h then cooled to ꢀ30°C, DCM (1.5 L) and water (1 L) were added followed
by dropwise addition of aq. 6N HCl (500 mL) then aq. 2N HCl (300 mL) until pH 7, the aqueous
solution was extracted with DCM (3x). The combined extracts were dried over magnesium
sulfate and evaporated. The crude product was triturated in MTBE to obtain 50 (153 g, 90%) as a
1
beige solid. H NMR (DMSOꢀd ): 3.45 (m, 4H), 3.5ꢀ3.6 (m, 4H), 3.85 (s, 3H), 4.39 (s, 2H), 8.28
6
+
79
+ 81
(
s, 1H), 8.38 (s, 1H). MSꢀESI m/z 386 [MH , Br] and 388 [MH , Br].
Methyl 8-acetyl-2-morpholino-4-oxo-4H-chromene-6-carboxylate (51)
To a stirred suspension of 48 (40 g, 108 mmol) in dioxane (300 mL) was added tributyl(1ꢀ
ethoxyvinyl)stannane (38.5 mL, 114 mmol) and bis(triphenylphosphine) palladium (II) chloride
(3.05 g, 4.4 mmol). The mixture was purged with nitrogen and heated at 90°C for 18 h. More
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tributyl(1ꢀethoxyvinyl)stannane (20 mL) and bis(triphenylphosphine) palladium(II) chloride (1.5
g, 2.1 mmol) were added and the reaction mixture heated for an additional 3 h. The mixture was
cooled to room temperature, 2N aq. HCl (81 mL, 163 mmol) was added and the dark suspension
heated at 45°C for 30 min. The dioxane was evaporated under vacuum, the residue was dissolved
in DCM and saturated aq. sodium bicarbonate was added until pH 4. The organic layer was
washed with brine, dried over magnesium sulfate, filtered and concentrated to afford a crude
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compound which was triturated in diethylether, filtered and dried to afford 51 (25.0 g, 69.5 %) as
1
an offꢀwhite solid. H NMR (CDCl ): 2.73 (s, 3H), 3.62 (m, 4H), 3.84 (m, 4H), 3.98 (s, 3H), 5.53
3
+
(
s, 1H), 8.66 (s, 1H), 9.02 (s, 1H). MSꢀESI m/z 332 [MH ].
Methyl 8-(1-hydroxyethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxylate (52)
Sodium tetraborohydride (82 mg, 2.2 mmol) was added to a solution of 51 (650 mg, 2.0 mmol) in
methanol (20 mL) and DCM (10 mL) at ꢀ10°C. After 15 min of stirring at ꢀ10°C, the reaction
mixture was quenched with water (25 mL). The volatiles were removed and the aqueous layer
extracted twice with DCM. The combined organics phases were washed with brine, dried over
magnesium sulfate and concentrated. The residue was triturated with diethyl ether and the solid
collected by filtration to give 52 (500 mg, 76 %) as a brown solid, which was used for the next
1
step without further purification. H NMR (CDCl ): 1.55 (d, 3H), 3.22 (d br, 1H), 3.50 (m, 4H),
3
+
3
.82 (m, 4H), 5.30 (m, 1H), 8.31 (s, 1H), 8.58 (s, 1H). MSꢀESI m/z 334 [MH ].
Methyl 8-(1-bromoethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxylate (53)
Phosphorous tribromide (1.65 mL, 1.65 mmol) was added to a stirred suspension of 52 (500 mg,
1
.5 mmol) in DCM (10 mL) in an ice bath under nitrogen. The resulting solution was stirred at
room temperature for 24 h. The reaction was incomplete and further phosphorous tribromide
0.30 mL, 0.3 mmol) was added and the reaction mixture stirred for an additional 12 h. The
(
solvent was evaporated, the residue was suspended in water and ice and aq. sat. sodium carbonate
ACS Paragon Plus Environment

Page 25 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was carefully added until pH 6. The precipitate was collected by filtration, washed with water
1
then with diethyl ether and dried to afford 53 (470 mg, 79 %) as a dark beige solid. H NMR
(CDCl ): 2.19 (s, 3H), 3.61 (m, 4H), 3.88 (m, 4H), 3.97 (s, 3H), 5.54 (s br, 1H), 8.38 (s, 1H), 8.82
3
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
+ 79
+ 81
(
s, 1H). MSꢀESI m/z 396 [MH , Br] and 398 [MH , Br].
Methyl 8-(1-(4-fluorophenylamino)ethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxylate
54)
ꢀFluoroaniline (0.21 mL, 2.2 mmol) was added to a stirred suspension of 53 (215 mg, 0.54
mmol) dissolved in DCMꢀMeOH (4 ml, 3:1) at rt. The resulting suspension was stirred at rt for
6 h and 50°C for 16 h. The crude product was purified by flash chromatography on silica gel
(
4
1
eluting with 5% methanol in dichloromethane. The solvent was evaporated to dryness. The
1
resulting solid was triturated in Et O and dried to give 54 (230 mg, 99%) as a white solid. H
2
NMR (DMSOꢀd ): 1.50 (d, 3H), 3.58 (m, 4H), 3.75 (m, 4H), 3.85 (s, 3H), 4.93 (m, 1H), 5.62 (s,
6
1
H), 6.4 (d br, 1H), 6.47 (m, 2H), 6.86 (m, 2H), 8.14 (s, 1H), 8.38 (s, 1H). MSꢀESI m/z 427
+
[MH ].
8
-(1-(4-Fluorophenylamino)ethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxylic acid (55)
Aqueous 2N sodium hydroxide (0.28 mL, 0.56 mmol) was added to a stirred solution of 54 (240
mg, 0.56 mmol) in MeOH (2 mL). The resulting solution was stirred at rt for 16 h. Water (2 mL)
was added and the pH was adjusted to 3 with 2N aq. HCl. The solid was filtered, washed with
water and dried to give 55 (83 mg) as a white solid. The filtrate was evaporated to give another
1
additional batch of 55 (81 mg) as a white solid. H NMR (DMSOꢀd ): 1.50 (d, 3H), 3.58 (m, 4H),
6
3
.75 (m, 4H), 4.93 (m, 1H), 5.60 (s, 1H), 6.38 (d br, 1H), 6.47 (m, 2H), 6.83 (m, 2H), 8.15 (s,
+
1H), 8.37 (s, 1H). MSꢀESI m/z 413 [MH ].
8
-(1-(4-fluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-chromene-6-
carboxamide (8)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TBTU (77 mg, 0.24 mmol) was added to a stirred solution of 55 (82 mg, 0.20 mmol), 4ꢀ
methylmorpholine (0.055 mL, 0.50 mmol) and 2M solution of dimethylamine (0.119 mL, 0.24
mmol) dissolved in NMP (1.2 mL). The resulting solution was stirred at rt for 2 hours. The
reaction mixture was purified by preparative HPLC using a Waters SunFire reverseꢀphase
column (Cꢀ18, 5 microns silica, 19 mm diameter, 100 mm length, flow rate of 40 ml / minute)
and decreasingly polar mixtures of water (containing 0.2% ammonium carbonate) and
acetonitrile as eluent. The fractions containing the desired compound were evaporated, triturated
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
with Et O and dried to afford 8 (53 mg, 60.7 %) as a white solid. H NMR (DMSOꢀd ): 1.50 (d,
2
6
3H), 2.66 (s, 3H), 2.92 (s, 3H), 3.50ꢀ3.65 (m, 4H), 3.70ꢀ3.81 (m, 4H), 4.92ꢀ5.02 (m, 1H), 5.60 (s,
1
3
1
H), 6.27 (d, 1H), 6.46 (dd, 2H), 6.84 (dd, 2H), 7.56 (d, 1H), 7.77 (d, 1H). C NMR (DMSOꢀd ):
6
22.0, 34.7, 38.6, 44.4, 46.5, 65.2, 86.1, 113.5 (d), 115.1 (d), 121.9, 122.5, 127.6, 132.2, 133.3,
+
1
43.8, 151.1, 153.7, 155.0, 162.1, 169.1, 174.6. HRMSꢀESI m/z: 440.19791 [MH ] C H FN O
24
26
3
4
requires 440.19801.
8-(1-hydroxyethyl)-2-morpholino-4-oxo-4H-chromene-6-carboxylic acid (56)
To a suspension of 51 (193 mg, 0.41 mmol) in methanol (2 mL) was added at ꢀ15°C sodium
borohydride (15.4 mg, 0.41 mmol). The resulting suspension was stirred at ꢀ15°C for 20 minutes.
The reaction mixture was quenched with 2N aqueous NaOH (0.41 mL, 0.82 mmol) and allowed
to warm to room temperature for 1 h. More NaOH (0.41 mL, 0.82 mmol) was added and stirring
was maitained for another 15 min. Aq. HCl (0.92 mL, 1.8 mmol) was added to adjust pH to 2ꢀ3.
The resulting precipitate was diluted with a about 1 mL of water, collected by filtration, washed
with ethyl acetate, diethyl ether and dried to a constant weight to afford 56 (122 mg, 94 %) as a
1
orange solid which was used without further purification. H NMR (DMSOꢀd ): 1.42 (d, 3H),
6
3.55 (m, 4H), 3.75 (m, 4H), 5.22 (m, 1H), 5.52 (d br, 1H), 5.59 (s, 1H), 8.30 (s, 1H), 8.40 (s, 1H).
+
MSꢀESI m/z 320 [MH ].
ACS Paragon Plus Environment

Page 27 of 79
Journal of Medicinal Chemistry
1
2
3
8
-(1-Hydroxyethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-chromene-6-carboxamide (57)
TSTU (108 mg, 0.36 mmol) was added portionwise to a mixture of 56 (115 mg, 0.36 mmol) and
DIPEA (0.069 mL, 0.40 mmol) in DCM (1 mL) at 25°C under nitrogen. The resulting mixture
was stirred at 25°C for 1.5 h. Dimethylamine (2M in THF, 0.180 ml, 0.36 mmol) was then added
and stirring was maintained for an additional 30 min. The mixture was poured onto a silica gel
column and purified by flash chromatography eluting with 5% methanolic ammonia (7 N) in
dichloromethane. The solvent was evaporated to dryness. The residue was triturated in diethyl
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
ether, collected by filtration and dried to afford 57 (123 mg, 99%) as an offꢀwhite solid. H NMR
(DMSOꢀd ): 1.41 (d, 3H), 2.9ꢀ3.0 (m, 6H), 3.54 (m, 4H), 3.75 (m, 4H), 5.21 (m, 1H), 5.48 (d br,
6
+
1H), 5.57 (s, 1H), 7.76 (s, 1H), 7.79 (s, 1H). MSꢀESI m/z 347 [MH ].
8-(1-Bromoethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-chromene-6-carboxamide (58)
1M Phosphorous tribromide in DCM (0.35 mL, 0.35 mmol) was added dropwise to 57 (120 mg,
0.35 mmol) suspended in DCM (1 mL). The resulting suspension was stirred at rt for 4 days.
The resulting gum was triturated in ether to give a precipitate which was collected by filtration,
washed with ether and dried to a constant weight to afford 58 as the hydrobromide salt (140 mg,
+
8
2 %) which was used without further purification. MSꢀESI m/z 409 [MH ].
(R)-8-(1-(3,5-difluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-chromene-
6-carboxamide (13)
To a suspension of 58 (3.63 g, 7.41 mmol) in DMF (35 mL) under nitrogen was added 3,5ꢀ
difluoroaniline (3.82 g, 29.6 mmol). The resulting yellow solution was stirred at 50°C for 5 h.
The reaction mixture was concentrated to dryness, then purified by flash chromatography on
silica gel eluting with 1 to 7 % methanol in DCM. The solvent was evaporated to dryness to
afford 8ꢀ(1ꢀ(3,5ꢀdifluorophenylamino)ethyl)ꢀN,Nꢀdimethylꢀ2ꢀmorpholinoꢀ4ꢀoxoꢀ4Hꢀchromeneꢀ6ꢀ
carboxamide (2.32 g, 68.5 %) as a pale yellow solid. This racemic compound was separated by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
chiral preparative HPLC (column: Merck 50mm 20ꢁm Chiralpak IC, eluent: MeCN / MeOH /
DEA 90/10/0.2) to give 13, second eluting enantiomer 0.923 g. Analytical HPLC: flow: 1
mL/min, column Chiralpak IC 5 ꢁm 4.6 x 250 mm, eluting with MeCN / IPA 85:15, t 14.2 min
R
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
D
20°C
1
(
other enantiomer t 9.7 min). [α]
= ꢀ122.6° in EtOH. H NMR (DMSOꢀd ): 1.52 (d, 3H),
R
6
2.74 (s, 3H), 2.95 (s, 3H), 3.50ꢀ3.64 (m, 4H), 3.70ꢀ3.79 (m, 4H), 4.97ꢀ5.05 (m, 1H), 5.60 (m,
1
3
1
H), 6.15 (dd, 2H), 6.22 (tt, 1H), 6.96 (d, 1H), 7.54 (d, 1H), 7.81 (d, 1H). C NMR (DMSOꢀd ):
6
21.5, 38.5, 39.7, 44.4, 46.4, 65.3, 86.2, 90.8 (t), 95.7 (d), 122.2, 122.6, 127.4, 132.2, 132.4, 150.0
+
(t), 151.4, 162.1, 163.3 (dd), 169.1, 174.5. HRMSꢀESI m/z: 458.18820 [MH ] C H O N F
24
25
4
3
2
requires 458.18859. Anal. Found (%w/w): C, 62.44; H, 5.46; N, 8.84; C H O N F , H O (0.3 %
24
25
4
3
2
2
w/w), EtOAc (6.1% w/w), CH Cl (0.1% w/w) requires C 62.27 H 5.75 N 8.59.
2
2
(R)-8-(1-(3,4-Difluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-chromene-
6-carboxamide (12)
Using similar conditions as for 13 except that the reaction was run in NMP at 45°C and the
reaction mixture was purified by preparative HPLC, was obtained racemic 8ꢀ(1ꢀ(3,4ꢀ
difluorophenylamino)ethyl)ꢀN,Nꢀdimethylꢀ2ꢀmorpholinoꢀ4ꢀoxoꢀ4Hꢀchromeneꢀ6ꢀcarboxamide
(35.0 mg, 28.8 %) as a white solid. A larger batch of this racemic compound (2.19 g) was
separated by chiral preparative HPLC (column: CelluCoat 250 x 50 mm, 10µm, eluent:
Heptane/EtOH/TEA 50/50/0.1) to give 12, second eluting enantiomer: 1.09 g. Analytical HPLC,
flow: 1 mL/min, column: CelluCoat 250 x 4.6 mm, 10µm eluting with Heptane/EtOH/TEA
D
2
1
50/50/0.1. [α]
: ꢀ120° in MeCN. H NMR (DMSOꢀd ): 1.50 (d, 3H), 2.70 (s, 3H), 2.94 (s,
0°C
6
3H), 3.50ꢀ3.63 (m, 4H), 3.71ꢀ3.79 (m, 4H), 4.92ꢀ5.01 (m, 1H), 5.60 (s, 1H), 6.25 (d, 1H), 6.46
(ddd, 1H), 6.54 (d, 1H), 7.05 (dd, 1H), 7.54 (d, 1H), 7.78 (d, 1H). HRMSꢀESI m/z: 458.18814
+
[
MH ] C H O N F requires 458.18859.
2
4
25
4
3 2
ACS Paragon Plus Environment

Page 29 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
Similarly was obtained:
(R)-8-(1-(3-Chloro-2-fluorophenylamino)ethyl)-N,N-dimethyl-2-morpholino-4-oxo-4H-
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
chromene-6-carboxamide (15)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
622 mg; chiral preparative HPLC (column: Chiralpak IC 21 x 250mm, 5µm, eluent:
DCM/isopropanol 1:1), second eluting enantiomer. Analytical HPLC, flow: 1 mL/min, column:
Chiralpak IC 21 x 250mm, 5µm, eluent: DCM/isopropanol, t 14.8 min (other enantiomer t 10.8
R
R
D
2
1
min). [α]
: ꢀ159° in MeCN. H NMR (DMSOꢀd ): 1.58 (d, 3H), 2.70 (s, 3H), 2.94 (s, 3H),
6
0°C
3.50ꢀ3.64 (m, 4H), 3.70ꢀ3.80 (m, 4H), 5.05ꢀ5.15 (m, 1H), 5.60 (m, 1H), 6.37 (dd, 1H), 6.44 (d,
+
1H), 6.65 (ddd, 1H), 6.83 (dd, 1H), 7.60 (d, 1H), 7.78 (d, 1H). HRMSꢀESI m/z: 474.15894 [MH ]
C H O N ClF requires 474.15904
24 25
4
3
8-Bromo-2-morpholino-4-oxo-4H-chromene-6-carboxylic acid (59)
Sodium hydroxide (4.4 mL, 8.7 mmol) was added to 48 (1.6 g, 4.4 mmol) in MeOH (30 mL).
The resulting suspension was stirred at rt for 16 h. The mixture was diluted with water and the
pH adjusted to 3 with aq. 2N HCl. The precipitate was collected by filtration, washed with water
1
and dried over phosphorus pentoxide overnight to give 59 (1.30 g, 84 %) as a beige solid. H
NMR (DMSOꢀd ): 3.60 (m, 4H), 3.76 (m, 4H), 5.65 (s, 1H), 8.34 (s, 1H), 8.43 (s, 1H). MSꢀESI
6
+ 79
+ 81
m/z 354 [MH , Br] and 356 [MH , Br].
-Acetyl-2-morpholino-4-oxo-4H-chromene-6-carboxylic acid (60)
8
Bis(triphenylphosphine)palladium (II) chloride (12.78 mg, 0.02 mmol) was added to a stirred
mixture of 59 (215 mg, 0.61 mmol) and tributyl(1ꢀethoxyvinyl)stannane (0.226 mL, 0.67 mmol)
in 1,4ꢀdioxane (5 mL). The mixture was purged with nitrogen and was stirred at 100°C for 3 hrs.
Aq. 2N HCl (0.5 mL) was added and the reaction mixture was stirred at 50°C for 25 minutes then
allowed to cool to room temperature and concentrated under vacuum. The product was diluted
ACS Paragon Plus Environment