2-Arylbenzo[ d]oxazole Phosphinate Esters as Second-Generation Modulators of Utrophin for the Treatment of Duchenne Muscular Dystrophy

Article abstract of DOI:10.1021/acs.jmedchem.0c00807

Utrophin modulation is a promising therapeutic strategy for Duchenne muscular dystrophy (DMD), which should be applicable to all patient populations. Following on from ezutromid, the first-generation utrophin modulator, we describe the development of a second generation of utrophin modulators, based on the bioisosteric replacement of the sulfone group with a phosphinate ester and substitution of the metabolically labile naphthalene with a haloaryl substituent. The improved physicochemical and absorption, distribution, metabolism, and excretion (ADME) properties, further reflected in the enhanced pharmacokinetic profile of the most advanced compounds, 30 and 27, led to significantly better in vivo exposure compared to ezutromid and alleviation of the dystrophic phenotype in mdx mice. While 30 was found to have dose-limiting hepatotoxicity, 27 and its enantiomers exhibited limited off-target effects, resulting in a safe profile and highlighting their potential utility as next-generation utrophin modulators suitable for progression toward a future DMD therapy.

Full text of DOI:10.1021/acs.jmedchem.0c00807

Subscriber access provided by UCL Library Services
Article
2-Arylbenzo[d]oxazole phosphinate esters as second-generation
modulators of utrophin for the treatment of Duchenne Muscular Dystrophy
Arran Babbs, Adam Berg, Maria Chatzopoulou, Kay E. Davies, Stephen G. Davies, Benjamin Edwards, David
Elsey, Enrico Emer, Simon Guiraud, Shawn Harriman, Cristina Lecci, Lee Moir, David Peters, Neil Robinson,
Jessica Rowley, Angela J. Russell, Sarah Squire, Jonathan Tinsley, Francis Wilson, and Graham M. Wynne
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00807 • Publication Date (Web): 18 Jun 2020
Downloaded from pubs.acs.org on June 18, 2020
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 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 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.
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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2-Arylbenzo[d]oxazole phosphinate esters as
second-generation modulators of utrophin for the
treatment of Duchenne Muscular Dystrophy
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
Arran Babbs,1 Adam Berg,1 Maria Chatzopoulou,2 Kay E. Davies,1* Stephen G. Davies,2
Benjamin Edwards,1 David J. Elsey,3 Enrico Emer,2 Simon Guiraud,1 Shawn Harriman,3
Cristina Lecci,4 Lee Moir,1 David Peters,3 Neil Robinson,5 Jessica A. Rowley,2 Angela J.
Russell,2,6* Sarah E. Squire,1 Jonathon M. Tinsley,3 Francis X. Wilson,3 and Graham M.
Wynne2
1 MDUK Oxford Neuromuscular Centre, Department of Physiology, Anatomy and
Genetics, University of Oxford, Oxford OX1 3PT, UK
2 Department of Chemistry, University of Oxford, Chemistry Research Laboratory,
Mansfield Road, Oxford, OX1 3TA, UK
3 Summit Therapeutics plc, 136a Eastern Avenue, Milton Park, Abingdon, Oxfordshire,
OX14 4SB, UK
4 Evotec (UK) Ltd, 114 Innovation Dr, Milton Park, Milton, Abingdon OX14 4RZ
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 45
1
2
3
4
5 S.H.B. Enterprises Ltd, 55 Station Road, Beaconsfield, HP19 1QL
5
6
7
8
9
6 Department of Pharmacology, University of Oxford, Mansfield Road, Oxford, OX1 3PQ,
UK
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
KEYWORDS. Bioisosterism, Duchenne muscular dystrophy, ezutromid, phosphinate ester,
utrophin modulator
Abstract
Utrophin modulation is a promising therapeutic strategy for Duchenne Muscular
Dystrophy, which should be applicable to all patient populations. Following on from
ezutromid, the first generation utrophin modulator, we describe the development of a
second generation of utrophin modulators, based on bioisosteric replacement of the
sulfone group with a phosphinate ester and substitution of the metabolically labile
naphthalene with a haloaryl substituent. The improved physicochemical and ADME
properties, further reflected in the enhanced pharmacokinetic profile of the most
advanced compounds, 30 and 27, led to significantly better in vivo exposure compared
to ezutromid, and alleviation of the dystrophic phenotype in mdx mice. While 30 was
found to have dose-limiting hepatotoxicity, 27 and its enantiomers exhibited limited off-
target effects, resulting in a safe profile and highlighting their potential utility as next
generation utrophin modulators suitable for progression towards a future DMD therapy.
2
ACS Paragon Plus Environment

Page 3 of 45
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
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 45
1
2
3
4
Introduction
5
6
7
8
9
Duchenne muscular dystrophy (DMD) is an X-linked recessive progressive muscle-
wasting disorder that affects 1 in 5000 male births.1 DMD is caused by mutations in the
dystrophin gene, leading to the functional loss of the cytoplasmic protein, dystrophin.
Dystrophin (427 kDa) is a vital component of the dystrophin associated protein complex
(DAPC), providing structural stability, strength and flexibility to skeletal muscle by
connecting the internal cytoskeleton to the extracellular matrix. Absence of dystrophin
results in muscle necrosis and regeneration, leading to replacement of muscle fibres with
adipose and fibrotic tissue. Symptoms include progressive muscle weakness, motor
development delay and abnormal gait from early childhood, and patients usually lose
ambulation by 8-12 years of age. Premature death typically results at age 20-30 due to
respiratory and cardiac failure.2, 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
For the majority of the patient population, current treatments involve alleviating
downstream effects (inflammation, fibrosis) with the use of clinical management and
corticosteroids.2, 3 Although symptomatic and supportive treatments have increased the
life expectancy of DMD patients raising median survival from 26 to 41 years,4 no effective
disease modifying treatment currently exists. Therapies addressing patient sub-
populations through mutation specific approaches such as stop-codon read through
(ataluren) and exon skipping (eteplirsen and golodirsen, for exons 51 and 53, respectively)
have received conditional approval, and gene therapy treatments are being developed.2,
4
ACS Paragon Plus Environment

Page 5 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
5 Consequently there is an urgent need for a therapy that is applicable to the whole patient
population.
7
8
9
For many years we and others have been investigating an alternative strategy to treat
DMD: to upregulate the expression of utrophin, a structural and functional paralog of
dystrophin, to compensate for the lack of dystrophin in patients.6 Partial alleviation of
DMD pathology has been achieved through expression of a truncated utrophin minigene
in transgenic mdx mice, providing evidence for utrophin modulation as a viable treatment
strategy.7 Expression of a full length utrophin gene in the mdx model also prevented
functional signs of DMD pathology in a dose dependent manner.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
Ezutromid 1 (formerly SMT C1100) was a first-in-class small molecule utrophin
modulator developed by Summit Therapeutics which was well tolerated in Phase 1a and
1b clinical trials following oral administration.2, 9, 10 In a multi-centre, open-label Phase 2
clinical trial (PhaseOut DMD), ezutromid demonstrated a statistically significant and
meaningful decrease in muscle regeneration and muscle inflammation after 24 weeks of
treatment. However these effects were not observed after 48 weeks and development of
ezutromid was discontinued.11 Although found to be safe and well tolerated in patients,
exposure of ezutromid was found to be lower than anticipated in a Phase 1 healthy
volunteer trial and a Phase 1b trial of 12 DMD patients, and drug levels reduced
10
significantly following repeated dosing.9,
Ezutromid also has sub-optimal
physicochemical properties and is extensively metabolised in humans to the
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 45
1
2
3
4
5
6
7
8
9
corresponding 1,2-dihydrodiol metabolites DHD 1 and DHD 3, both of which show
reduced or no activity.1, 10, 12 Taken together these findings can rationalise the initial
positive benefit followed by the lack of sustained efficacy in the Phase 2 clinical trial.
As a result of these findings, a medicinal chemistry program was conducted to develop
second-generation utrophin modulators, with improved physicochemical properties and
metabolic stability compared to ezutromid whilst maintaining a good safety profile.
Optimisation was focused around replacing the sulfone moiety (Region A, Figure 1), in
order to improve the compound’s aqueous solubility, and the naphthalene moiety
(Region B, Figure 1), which was found to be the metabolic hotspot.12
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 1. The clinical candidate ezutromid (1, SMT C1100). Regions A and B denote sites
of modifications to improve solubility (A) and metabolic stability (B) whilst maintaining a
good efficacy and safety profile.
Results and Discussion
Chemistry
The general preparation of amide benzo[d]oxazoles, discussed extensively in previous
publications,13 was applied here to generate naphthyl amide derivatives 2, 3, and cyclic
6
ACS Paragon Plus Environment

Page 7 of 45
Journal of Medicinal Chemistry
1
2
3
4
sulfonamide 4 (Scheme 1). Sulfonamide 5 was derived from the respective 2-aminophenol
5
6
7
via a modification of Method B (Schemes 1,2).
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
Scheme 1. Synthesis of amide and sulphonamide derivatives 2, 3 and 4.α
O
N
OH
O
N
(i)
(ii) or (iii)
R1
N
H2N
NH2
H2N
R2
2a
2
3
R1: MeCO R2: H
R1: iPrCO R2: H
2 HCl
O
OH
O
4
R1, R2
O
N
S
N
(iv)
H
N
S
O O
H
N
NH2
S
5
O O
α
o
Reagents and conditions: (i) naphthoic acid, PPA, 180 C, 3 h, 63%; (ii) acylchloride,
pyridine, rt, 24 h, 67-83% (2-3); (iii) 1. 3-chloropropanesulfonyl chloride, Et3N, CH2Cl2,
reflux, 1 h 2. NaH, DMF, rt to 90 oC, 16 h, 22% over two steps; (iv) 1. ArCHO, MeOH, 50 oC,
45 min, 2. DDQ, CH2Cl2, rt, 2 h.
A range of methods were used to access the phosphinate esters of the benzo[d]oxazoles
based on a common general route (Scheme 2) and these methods are described in detail
in the supporting information. In brief, intermediate 2-arylhalobenzo[d]oxazoles (5, 6-8a,
15-26a, 28-29a, 31-45a, 51a, 55a) were derived either from condensation of 2-
aminophenols with the respective carboxylic acid chloride followed by dehydration
(Methods A and D), or from the respective aldehyde with a consecutive oxidation step
(Methods B and C). The final products were obtained from the halo-derivatives via a Pd-
catalysed functionalisation with phosphinate esters or phosphinic acids (Methods E-G, 6-
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 45
1
2
3
4
5
6
7
8
9
27, 28-29, 31-55). Phosphinic acids were converted to the corresponding esters through
formation of the phosphinic acid chloride and reaction with the appropriate alcohol
(Method H, 26b, 55b). Where scale-up synthesis was required for in vivo studies, the
procedures were further optimised.14
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 2. General synthetic route to phosphinate estersα
O
O
N
(ii)
Ar
P
R1O
R2
6-26, 28-29, 31-55
OH
O
N
(i)
Ar
X
X
(iv)
NH2
O
O
N
6-8a, 15-26a, 28-29a,
31-45a, 51a, 55a
(iii)
Ar
P
HO
X: Br, I
R2
26b, 52, 55b
α Reagents and conditions: (i) Method A: Step 1. ArCOCl, dioxane, 210 oC, 3 h, MW, or
Method B: ArCHO, MeOH, 50 oC, 45 min, Step 2. DDQ, CH2Cl2, rt, 2 h, (yields 7-83% and
19-82% for Methods A and B, respectively) or Method C: Step 1. ArCHO, EtOH, 80 oC, 90
min, Step 2. PhI(OAc)2, MeOH, 60 oC, 30 min, (yields 44-57%) or Method D: Step 1. ArCOCl,
o
o
xylenes, 155 C, 1 h, Step 2. MsOH, 155 C, 2-3 h, (yields 12-81%); (ii) Method F:
o
HP(O)OR1R2, Pd(PPh3)4, Et3N, toluene, 110 C, 3 h, (yields 5-81%) or Method G:
HP(O)OMeEt, Pd2(dba)3, Cs2CO3, THF, 140 oC, 30 min, MW, (yields 6-58%); (iii) Method E:
o
EtPH(O)OH, Pd(OAc)2, Xantphos, DIPEA, DME/toluene, 90 C, 2 h, (yields 46-98%); (iv)
Method H: 1. SOCl2, DMF, 70 oC, 1 h, 2. DIPEA, MeOH, DMAP, CH2Cl2, rt, 30 min, (yields
36-85%)
Detailed synthetic procedures for analogues 27 and 30 are reported elsewhere.14, 15 The
synthesis of 26 was described previously,16 however, a significant improvement in yield
was observed when the compound was synthesised via the phosphinic acid route.
8
ACS Paragon Plus Environment

Page 9 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Structure Activity Relationship (SAR) 1: Replacement of the sulfone moiety (Region
A)
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
Sulfones are known to form strong intermolecular interactions in the solid state often
leading to low solubility.17 For this reason, replacements of the sulfone group of ezutromid
(1) with other hydrogen bond donor groups such as amides (2, 3), sulphonamides (4, 5)
and phosphinates (6, 7) were investigated in order to establish the impact of altering
region A on activity and solubility (Table 1). The ability of the compounds to increase
utrophin levels was investigated using the previously reported cell line H2K-mdx utrnA-
luc (H2K), a murine myoblast cell line that contains a stably integrated 8.4 kb fragment of
the human utrophin promoter linked to a firefly luciferase gene.1, 12 Replacement of the
sulfone moiety was well tolerated in all cases and even led to an apparent increase in
activity in the cases of amides (2, 3) and phosphinates (6, 7) (Table 1). Amides (2, 3) and
sulfonamide 4 had similar solubility to ezutromid, while phosphinate 7 was found to be
significantly more soluble (Table 1).
Table 1. SAR on Region A: Replacement of the sulfone moiety and preliminary SAR
investigation of phosphinate esters
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 45
1
2
3
4
R3
O
5
6
7
8
N
R2
R1
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
Cpd
1
R1
R2
R3
H2K EC50 ± SDa (μM) Solb (μM)
H
S
O O
H
0.187 ± 0.030
3-10
3-10
O
2
3
4
H
H
H
H
H
0.010 ± 0.002
0.038 ± 0.012
0.110 ± 0.029
N
H
O
1-3
N
H
O
3-10
O
S
N
S
H
H
5
6
H
H
0.146 ± 0.023
0.004 ± 0.002
NDc
ND
H
N
O O
O
H
H
O
P
O
P
7
H
H
0.071 ± 0.008
IAd
30-100
ND
O
O
P
8
H
H
O
O
P
9
H
H
0.026 ± 0.003
0.108 ± 0.022
ND
O
O
P
10
H
ND
O
10
ACS Paragon Plus Environment

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
P
11
12
13
H
H
H
H
0.034 ± 0.003
0.234 ± 0.057
0.148 ± 0.030
ND
ND
ND
O
O
P
H
O
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
O
P
H
H
O
O
14
H
IA
ND
O
P
Ph
a
firefly luciferase reporter gene assay reported as EC50 ± SD of 3 biological
replicates (3 technical replicates per experiment), b Cyprotex turbidimetric solubility,
c ND: not determined, d IA: inactive
Phosphinates are an underrepresented class of molecules as drugs, with most
phosphorous-containing drugs containing the phosphate or phosphonate structure
(nucleoside prodrugs, bisphosphonates).18, 19 The ACE inhibitor fosinopril includes a
phosphinate motif as a prodrug, the active species being the phosphinic acid
(fosinoprilat).20 The opportunity to study underexplored chemical space in combination
with the improved solubility observed from the preliminary studies, led to the selection
of phosphinate as sulfone replacement which was carried forward into more detailed SAR
studies (Table 1, analogues 6-14). It was found that the phosphinate ester substituent was
well tolerated in positions 5 and 6 of the benzoxazole ring (6 and 7, respectively), while
moving it to position 4 (8) led to loss of activity. Whilst bulky P-alkyl substituents including
i-Pr and t-Bu were tolerated (11-13), the corresponding P-phenyl derivative (14) showed
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 45
1
2
3
4
5
6
7
8
9
no activity. Ethyl phosphinate esters (9, 10) showed similar potency to their respective
methyl ester counterparts (6, 7). Subsequent optimization studies were performed using
the methyl ethylphosphinate scaffold.
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
SAR 2: Replacement of the naphthalene substituent to reduce hydrophobicity
(Region B)
As the naphthalene ring of ezutromid had been identified early on as the main site of
metabolism,12 variation of this region was investigated next (Region B, Figure 1) in order
to improve metabolic stability in the new series. Replacement of naphthalene with
heterobiaryls such as quinolines and isoquinolines was an approach to mitigate
cytochrome P450 oxidative metabolism12 (Table 2). In a parallel approach, halogen
substituents were used to block the metabolic hotspots in the phenyl ring (Table 2).
A range of quinolines and isoquinolines were explored in the 2-position of the
benzo[d]oxazole ring and the physicochemical properties, metabolic stability and readout
of phenotypic activity were investigated in vitro using the H2K-mdx reporter assay, and
were compared to that of ezutromid (1) and 6 or 7 (Table 2). The SAR appeared to be
tolerant of this variation. Similarly, the haloaryl and styryl derivatives also appeared to be
active, with the exception of o-trifluoromethylphenyl derivative 43.
Table 2. SAR on Region B: Replacement of the naphthalene ring with heterobiaryls or
haloaryls/styryl
12
ACS Paragon Plus Environment

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
P
O
N
O
O
N
O
Ar
O
P
Ar
O
Ar
O
N
S
8
9
O O
A
B
C
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
Cpd Scaffold
Ar
H2K EC50 ± SDa (μM)
1
A
B
B
0.187 ± 0.030
6
0.004 ± 0.002
0.062 ± 0.011
N
15
16
17
B
B
0.164 ± 0.044
0.157 ± 0.016
N
N
18
B
0.332 ± 0.049
N
19
7
B
C
C
1.14 ± 0.13
N
0.071 ± 0.008
0.085 ± 0.017
N
20
21
C
0.928 ± 0.164
N
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 45
1
2
3
4
5
6
7
8
9
22
23
24
C
C
C
0.085 ± 0.003
N
1.14 ± 0.34
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
N
1.76 ± 0.13
N
25
26
27
28
B
B
B
B
0.163 ± 0.037
0.112 ± 0.019
Cl
0.244 ± 0.020
F
0.823 ± 0.020
F
29
30
B
B
0.548 ± 0.125
CF3
1.28 ± 0.25b
CF3
31
B
0.861 ± 0.116
CF3
32
33
B
B
0.777 ± 0.083
F
F
F
0.198 ± 0.031
Cl
14
ACS Paragon Plus Environment

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
34
35
36
B
B
B
0.118 ± 0.029
Cl
Cl
0.244 ± 0.057
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
Cl
Cl
F
1.77 ± 0.38
F
Cl
Cl
37
38
B
B
0.168 ± 0.017
S
S
1.51 ± 0.26
Cl
CF3
39
B
1.77 ± 0.26
O
40
41
42
C
C
C
0.766 ± 0.083
1.23 ± 0.26
F
2.25 ± 0.76
F
43
C
IA
CF3
44
45
C
C
1.76 ± 0.22
F
F
1.95 ± 0.41
Cl
Cl
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 45
1
2
3
4
5
6
a firefly luciferase reporter gene assay reported as EC50 ± SD of
3 biological replicates (3 technical replicates per experiment);
Ref 14.
b
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
Next, we further characterised representative analogues and evaluated some basic
physicochemical, plasma protein binding and metabolic stability properties. Replacement
of the 2-naphthyl moiety (1, 7) with quinolinyl, styryl and halo-phenyl derivatives
(Structure-Property Relationships (SPR) ,Table 3) improved the solubility (in most cases,
to greater than 100 µM) compared to 2-naphthyl phosphinate (7, 30-100 µM) and
especially compared to ezutromid (1, 3-10 µM).
Table 3. Structure-Property Relationships (SPR) on Region B: Replacement of the
naphthalene ring with heterobiaryls or haloaryls/styryl
PPB b
Huma
Microsomal Stability
Mouse Human
Clint Clint
(µL/min/mg) (min) (µL/min/mg) (min)
Cpd
Sola
(µM)
Mouse
(% bound)
n
T1/2
T1/2
1
3-10
>99
99
99
ND
90
21.0
8.45
8.46
<1
66
14.4
15.8
3.14
96
7
30-100 99
164
164
87.7
441
141
15
20
>100
>100
NDc
77
>440 9.81
0
21
25
30-100 81
>100 ND
80
ND
ND
9.51
14.6
146
95
ND
11.9
116
16
ACS Paragon Plus Environment

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
26
27
28
30
32
33
35
40
30-100 ND
ND
12.2
16.1
292
5.12
14.4
18.2
27.2
<1
114
6.47
3.62
6.34
8.33
1.59
5.85
0.65
214
383
219
166
869
236
2120
2550
>100
>100
75
81
88
94
95
85
91
98
93
86
4.75
271
96
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
30-100 92
>100
>100
72
89
76
30-100 98
51
>100
83
>440 0.543
0
41
>100
75
93
24.1
58
4.43
313
a turbidimetric solubility; b PPB, plasma protein binding; c ND, not determined
The binding of analogues 7, 15, 20, 21, 28, 30, 32, 40, and 41 to human and mouse
plasma proteins varied, ranging between 80 to 99% bound fraction in humans and 72 to
99% bound fraction in mice (Table 3). The fraction of compound bound to plasma proteins
did not appear to be linked to phosphinate regiochemistry (5- or 6-), nor to the
substituent in position 2. The most highly bound compound appeared to be the
naphthalene analogue 7, and the least were quinolines 20 and 21 and fluorophenyl
derivatives 27 and 32.
The modifications greatly improved metabolic stability in most cases, apart from 28
which was cleared very quickly in mouse microsomes (Table 3). No patterns were apparent
in terms of the phosphinate or aryl/heterobiaryl/styryl substituent, nor for the
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 45
1
2
3
4
5
6
7
8
9
phosphinate regiochemistry. The most metabolically stable compound in both mouse and
human liver microsomes was the styryl analogue 40, while quinoline 20 had the lowest
intrinsic clearance in mouse liver microsomes and the 2,3-dichloro analogue 35 in human
liver microsomes.
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 4. Metabolic profile of a selection of haloaryls and 27 close analogues in different
species
Microsomal stability
Cpd
Clint (µL/min/mg)
T1/2 (min)
d
r
m
h
4
d
r
m
h
27
28
16
86 383
7.5 4.7
6.3
4
<1 184 292
Sa
219
2
5
30
32
41
7.8 7.0 5.1 8.3
177 197 271 166
2
2
2
3
1.1 3.3 14. 1.5
96.
1270 419
S 159
869
313
0
1
4
9
1
8.7 24. 4.4
57.
6
<1
4
1
3
a S: stable (100% after 45 min); d: dog; r: rat; m: mouse; h: human
microsomes
The metabolic profile of a selection of analogues, which had an encouraging
combination of activity and ADME properties, was further evaluated in different species
18
ACS Paragon Plus Environment

Page 19 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in order to select compounds that would have a favourable profile in an in vivo model,
but might also translate to an appropriate PK profile in humans (Table 4). On the basis of
this evaluation, compounds 27, 30 and 41 were selected and evaluated for their ability to
increase utrophin protein in LUmdx cells.
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
Phosphinate selection: utrophin protein upregulation in dystrophic mouse muscle
cells
22
Acknowledging the limitations of reporter gene assays,21,
we sought further
confirmation of utrophin modulation activity. Quantification of utrophin protein levels by
western blot in mouse mdx cells (LUmdx) treated with the selected analogues 27, 30 and
41 was used as an orthogonal method to support progression of these molecules to in
vivo evaluation. We observed an increase in utrophin protein in all three cases.
Phosphinates 27 and 30 increased utrophin up to 2.5-fold in a concentration-dependent
manner, while 41 increased utrophin 2-fold at 1 μM, but at 10 μM utrophin came down
to basal levels (Figure 2). 27 and 30 were therefore selected for progression into murine
in vivo PK and efficacy studies.
Figure 2. Treatment with 27, 30 and 41 increases utrophin protein in dystrophin deficient
mouse muscle cells. Relative utrophin expression was determined by western blot,
standardised for α-actinin loading and is shown as the mean of fold increase relative to
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 45
1
2
3
4
5
6
DMSO ± SD (two biological replicates, 3 technical replicates each) * p<0.05, ** p<0.01, ***
p<0.001
7
8
9
Western Blot analysis
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
3
***
O
*
**
O
P
F
**
**
27
30
N
O
*
O
N
2
1
0
O
P
*
CF3
O
O
P
O
N
O
F
41
0.1 1.0 10.0 0.1 1.0 10.0 0.1 1.0 10.0
Concentration (µM)
Characterisation and in vitro evaluation of phosphinates 27, 30 and their
enantiomers
As phosphinates have a P-stereogenic centre, some insight into the profile of the two
enantiomers was required in order to assess whether to progress the racemic mixture or
a single enantiomer. Separation of the enantiomers was performed using preparative
chiral HPLC,14 the specific optical rotations of 27 enantiomers were measured and the
enantiomers were assigned as (–) and (+), respectively. The absolute configuration of the
enantiomers was later inferred through an enantioselective synthesis of (+)-30 and (–)-
30.14 The enantiomers of 30 were found to have similar levels of activity to the racemate
(H2K EC50s were 1.28, 1.39 and 1.20 μM for 30, (+)-30 and (–)-30, respectively).14 In
20
ACS Paragon Plus Environment

Page 21 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
contrast, (+)-27 appeared to be slightly more active in the H2K-mdx reporter gene assay
than enantiomer (–)-27 and the corresponding racemic mixture, with a eudismic ratio of
5 (Table 5).
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 5. Determination of optical rotation value and assignment, and SAR for 27
enantiomers
풂ퟐ푫ퟓ
a
#
H2K EC50 ± SDb (μM)
1
0.187 ± 0.030
0.244 ± 0.020
27
(–)-
27
– 40.7 0.711 ± 0.155
(+)-
27
+ 40.7 0.124 ± 0.015
a Optical rotation [c= 1.0, THF]; b firefly
luciferase reporter gene assay reported
as EC50 ± SD of 3 biological replicates (3
technical replicates per experiment)
The metabolic stability of the pairs of enantiomers of 27 and 30 were next evaluated in
murine and human hepatocytes (Table 6). No significant differences were noted between
the enantiomers of each compound in either case. No detectable racemisation was
observed over the course of the 1-hour incubation. Consistent with the metabolic stability
determined in microsomes, all analogues were found to be more stable than ezutromid
in human hepatocytes. The enantiomers of 27, were cleared rapidly in mouse hepatocytes,
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 45
1
2
3
4
5
6
7
8
9
while the enantiomers of 30 were significantly more stable (Clint >75 for 27 enantiomers
vs 16 and 15 for (+)-30 and (–)-30). In human hepatocytes all analogues had low
clearance (Clint <4.9 μL/min/106 cells). All analogues were found to have a higher bound
fraction with human plasma proteins than mouse plasma proteins.
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 6. ADME evaluation of 27 and 30 enantiomers in mouse and human
Hepatocyte stabilitya
PPB
Clint
T1/2
#
m
h
m
h
m
h
(% bound)
(μL/min/106 cells)
(min)
1
>99
99
85
85
94
91
54
141
75
31
<3
<3
<3
4.9
26
9.8
19
89
94
45
(–)-27
(+)-27
(+)-30
(–)-30
77
80
83
87
>460
>460
>460
283
16
15
a
<1% interconversion to the opposite enantiomeric form was observed in
mouse hepatocytes after 60 min incubation; m: mouse; h: human hepatocytes
PK and safety profile for phosphinates 27, 30 and their enantiomers
The pharmacokinetics of the enantiomeric pairs and racemic mixtures of 27 and 30 were
investigated in the mdx mouse (Table 7). In general exposure was found to be low to
moderate for 27 and enantiomers at 5 mg/kg, but higher for 30. The half-life was found
to be short in all cases, but 27 and its enantiomers were cleared more rapidly than 30.
Interestingly, the less lipophilic 27 had a higher volume of distribution, potentially
22
ACS Paragon Plus Environment

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
attributed to its lower PPB and, hence, greater unbound fraction, however in all cases a
favourable distribution to tissues was predicted. The oral bioavailability of 30 and
enantiomers was also found to be superior to that of 27. In general, oral bioavailability
was good (>50%), apart from (–)-27, where it was 39 and 45% at day 1 and day 7,
respectively. Overall, this suggested that phosphinate 30 had a preferable
pharmacokinetic profile to 27. The reason for the observed decrease in AUC for (+)-30
and increase in AUC for (-)-30 after 7 days dosing was not clear. Despite this apparent
difference, as discussed later, (+)-30, (-)-30 and the racemic mixture showed no significant
differences in activity and efficacy in vivo in the mdx mouse.14
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 7. Pharmacokinetic properties of 27, 30 and enantiomers dosed i.v. or p.o. at 1
mg/kg and 5 mg/kg, respectively, in mdx mice*
Cpd/route
Cmax
(ng/mL)
562
AUC
(ng/mL/h
)
T1/2
(h)
CL
Vd
(L/kg)
1.46
app. F%
(mL/min/kg
)
27/i.v.
431
0.64
38.6
(–)-27/i.v.
(+)-27/i.v.
27/p.o. D1
27/p.o. D7
477
615
734
566
226
510
0.27
0.34
2.25
1.58
73.5
32.7
1.58
1.41
1663
1157
77.2
53.7
(–)-27/p.o.
D1
352
444
1.89
39.4
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 45
1
2
3
4
5
6
(–)-27/p.o.
D7
299
542
736
513
1569
1622
1.21
1.74
3.09
45.4
61.6
63.7
7
8
9
(+)-27/p.o.
D1
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
(+)-27/p.o.
D7
30/i.v.
1365
1520
1788
2341
3482
1611
1079
2196
6974
6910
0.61
0.31
0.77
2.08
1.24
10.3
15.3
7.57
0.61
0.56
0.56
(+)-30/i.v.
(–)-30/i.v.
30/p.o. D1
30/p.o. D7
86.5
85.8
(+)-30/p.o.
D1
3488
2796
1939
4063
11613
5171
3.71
1.03
1.55
0.79
>100
95.8
34.1
91.7
(+)-30/p.o.
D7
(–)-30/p.o.
D1
3741
(–)-30/p.o.
D7
10064
* n = 3 mice per datapoint; Abbreviations: AUC: area under the curve, CL: clearance,
cpd: compound, D1: 1 day, D7: 7 days, i.v.: intravenous, p.o.: per os
The mutagenic potential of the leading phosphinates was evaluated in an Ames assay
using genetically modified Salmonella typhimurium LT2 strains TA1535, TA1537, TA98 and
TA100, and Escherichia coli WP2 strain uvrA/pKM101 pKM101 as indicator organisms. In
most cases, no consistent increase in numbers of revertant (histidine or tryptophan-
independent) colonies was seen with any of the five indicator strains either in the presence
24
ACS Paragon Plus Environment

Page 25 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
or absence of S9 microsomes. It was thus concluded that none of racemic 27, 30, (–)-27,
(+)-30 and (–)-30 appeared mutagenic under the conditions used in this study. Curiously
(+)-27 showed a clear concentration-related increase in numbers of revertant colonies
(histidine-dependent) for indicator strain TA98 at and above a dose of 160 μg (0.501 μmol)
per plate in the presence of S9 microsomes. The increase reached 3.5 times the solvent
control value at 500 μg (1.57 μmol) per plate. For this reason, 30 was initially selected for
progression to a murine in vivo efficacy trial.
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
Phosphinate 30 is identified as an advanced molecule but proves to be
hepatotoxic in rats
The in vitro and in vivo profile of 30 (also termed SMT022357) has been described in
detail previously.1, 14 In brief, 30 was found to increase the utrophin protein in vitro and in
vivo, ameliorating the pathology in skeletal, respiratory and cardiac muscle in the mdx
mouse.1 No significant difference was observed in the activity and efficacy of the two
enantiomers, and activity was shown to be comparable to that of the racemic mixture.14
However, 30 was found to be hepatotoxic during a maximum tolerated dose (MTD) study
in rat (data not shown). Although the 100 mg/kg dose level was tolerated for both
enantiomers, adverse findings were apparent based on evidence of liver toxicity that
included liver enlargement and correlative histopathology findings. As a result, the
therapeutic window for this compound was deemed too small to progress further.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
A more detailed analysis provided evidence of hepatobiliary toxicity as gamma
glutamyltransferase (GGT) and total bilirubin (TBILI) were significantly elevated at both
dose regimens of 30 (300 and 100 mg/kg, Table S1). This was in good agreement with
the cholestatic injury observed during the MTD study. Enantiomer (–)-30 appeared to be
less toxic, however this was attributed to the difference in toxicokinetics between the two
enantiomers (Table S2), as (+)-30 showed greater exposure than (–)-30. It was noted that
the two enantiomers had similar PK profiles in the mdx mouse,14 although this was not
the case in rat.
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
In order to gain insight into the mechanism of hepatotoxicity and to determine whether
it could be mitigated, representative analogues were tested in HepRG spheroids, a 3D
hepatocyte model where several cell health parameters can be measured simultaneously
(Table S3). Although the limited number of analogues tested did not allow clear structure-
toxicity relationships to be observed, 30 and its enantiomers were amongst the most toxic
analogues tested, inducing the same type of cell damage as shown by an increase in
oxidative stress and a decrease in glutathione, DNA structure and cellular ATP. Another
trifluorosubstituted analogue 29 induced similar types of damage as 30. Analogue 27 and
its enantiomers were found to have only minimal hepatotoxicity. From the results
obtained, it was apparent that some analogues were significantly less toxic than others,
and there was no pattern in inducing the same type of cell damage.
26
ACS Paragon Plus Environment

Page 27 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The nature and location of the damage caused in rat liver during the MTD study
suggested P450 bioactivation to a hepatotoxic metabolite. As a result, metabolite
identification (MetID) studies were undertaken in samples from this study. It was found
that 30 was metabolised mainly to a glucuronide adduct of an oxidation product (M1)
and to the corresponding phosphinic acid (M3). Further metabolites identified were GSH-
conjugates (M7-8, Figure 3, Table S4).
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. Major 30 metabolites identified by MetID in samples from the MTD rat study
(representative fragment ions are given)
CF3
O
O
OGluc
N
H
P
O
m/z 386
M1
m/z 562
CF3
O
CF3
m/z 264
O
P
HO
O
N
H
O
P
N
H
m/z 310
O
M3
m/z 356
m/z 310
m/z 370
CF3
30
O
O
P
GSH product
N
H
O
M7 M8
,
m/z 500.8 (2+)
The metabolic profile of 30 was further studied in vitro in 4 different species to evaluate
if this profile was unique to rat and if there were significant differences in the metabolites
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 45
1
2
3
4
5
6
7
8
9
produced in primates. The metabolites of the two enantiomers in human hepatocytes
were also determined. In all cases, the main metabolic routes were oxidation and
conjugation with glucuronic acid (Table S5). Glucuronides were detected in larger
quantities in rat and dog hepatocytes. No significant differences were observed between
the two enantiomers (data not shown). Interestingly, the corresponding phosphinic acid
M3 was not detected in the in vitro incubations.
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
SAR 3: Further modifications on phosphinates 27 and 30
As phosphinates 27 and 30 had otherwise proved to be promising properties, some
closely related structural analogues were synthesised and evaluated for their
physicochemical and ADME properties. Varying the phosphinate ester and P-alkyl
substituents did not appear to significantly affect activity in analogues of 27, however the
combination of a P-phenyl substituent and changing the regiochemistry of the
phosphinate ester in position 6 of the benzoxazole ring (scaffold G) led to an inactive
analogue (50) (Table 8). Interestingly, changing the regiochemistry of phosphinate ester
to position 4 appeared to retain activity (51, scaffold H), in contrast to the previous SAR
observed with the naphthalene scaffold (8, Table 1).
Table 8. SAR on analogues of phosphinate 27 (Region A) and bulkier phosphinic esters
of 30
28
ACS Paragon Plus Environment

Page 29 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
R2
P
O
N
O
O
N
R1O
O
O
R1O
R3
R3
R3
N
7
P
8
9
F
G
H
R2
P
O
OR1
R2
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
Scaffol R1
d
R2
R3
#
H2K EC50 ± SDa (μM)
2
7
F
Me
Et
4-F
0.244 ± 0.020
4
6
F
Me
Me
Et
iPr
Ph
Et
4-F
0.303 ± 0.084
0.826 ± 0.136
1.25 ± 0.34
1.23 ± 0.26
1.51 ±0.28
IA
4
7
F
4-F
4
8
F
4-F
4
1
G
G
G
H
F
Me
Me
Me
Me
Me
H
Et
4-F
4
9
iPr
Ph
Et
4-F
5
0
4-F
5
1
4-F
1.19 ± 0.36
1.28 ± 0.25b
IA
3
0
Et
3-CF3
3-CF3
3-CF3
5
2
F
Et
5
3
F
cyclopropy Et
l
IA
29
ACS Paragon Plus Environment