New Steroidal 4-Aminoquinolines Antagonize Botulinum Neurotoxin Serotype A in Mouse Embryonic Stem Cell Derived Motor Neurons in Postintoxication Model

Article abstract of DOI:10.1021/acs.jmedchem.7b01710

The synthesis and inhibitory potencies against botulinum neurotoxin serotype A light chain (BoNT/A LC) using in vitro HPLC based enzymatic assay for various steroidal, benzothiophene, thiophene, and adamantane 4-aminoquinoline derivatives are described. In addition, the compounds were evaluated for the activity against BoNT/A holotoxin in mouse embryonic stem cell derived motor neurons. Steroidal derivative 16 showed remarkable protection (up to 89% of uncleaved SNAP-25) even when administered 30 min postintoxication. This appears to be the first example of LC inhibitors antagonizing BoNT intoxication in mouse embryonic stem cell derived motor neurons (mES-MNs) in a postexposure model. Oral administration of 16 was well tolerated in the mouse up to 600 mg/kg, q.d. Although adequate unbound drug levels were not achieved at this dose, the favorable in vitro ADMET results strongly support further work in this series.

Full text of DOI:10.1021/acs.jmedchem.7b01710

Subscriber access provided by READING UNIV
Article
New Steroidal 4-Aminoquinolines Antagonize Botulinum
Neurotoxin Serotype A in Mouse Embryonic Stem Cell
Derived Motor Neurons in Post-intoxication Model
Jelena Konstantinovic, Erkan Kiris, Krishna P Kota, Johanny Kugelman-Tonos,
Milica Videnovic, Lisa H. Cazares, Natasa Terzic Jovanovic, Tatjana Z Verbic,
Boban Andjelkovic, Allen J. Duplantier, Sina Bavari, and Bogdan A. Solaja
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01710 • Publication Date (Web): 31 Jan 2018
Downloaded from http://pubs.acs.org on February 2, 2018
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 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
New Steroidal 4ꢀAminoquinolines Antagonize
Botulinum Neurotoxin Serotype A in Mouse
Embryonic Stem Cell Derived Motor Neurons in
Postꢀintoxication Model
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
Jelena Konstantinović¶, Erkan Kiris ¥, Krishna P. Kota†, Johanny KugelmanꢀTonos †, Milica
Videnović#, Lisa H. Cazares †, Nataša Terzić Jovanović∇, Tatjana Ž. Verbić ¶, Boban Andjelković
¶, Allen J. Duplantier†, Sina Bavari*,‡ and Bogdan A. Šolaja*,¶,§
¶ University of Belgrade, Faculty of Chemistry, Studentski trg 16, P.O. Box 51, 11158,
Belgrade, Serbia
¥Mouse Cancer Genetics Program, Center for Cancer Research, National Cancer Institute,
Frederick, Maryland 21702, United States
†Molecular and Translational Sciences Division, United States Army Medical Research Institute
of Infectious Diseases, 1425 Porter Street, Frederick, Maryland 21702, United States
#Faculty of Chemistry Innovative Centre, Studentski trg 12ꢀ16, 11158 Belgrade, Serbia
∇ University of Belgrade, Institute of Chemistry, Technology, and Metallurgy, Njegoševa 12,
11000 Belgrade, Serbia
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
4
5
‡United States Army Medical Research Institute of Infectious Diseases, 1425 Porter Street,
Frederick, Maryland 21702, United States
6
7
8
9
§Serbian Academy of Sciences and Arts, Knez Mihailova 35, 11158 Belgrade, Serbia
ABSTRACT
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
The synthesis and inhibitory potencies against botulinum neurotoxin serotype A light chain
(BoNT/A LC) using in vitro HPLC based enzymatic assay for various steroidal, benzothiophene,
thiophene and adamantane 4ꢀaminoquinoline derivatives is described. In addition, the
compounds were evaluated for the activity against BoNT/A holotoxin in mouse embryonic stem
cell derived motor neurons. Steroidal derivative 16 showed remarkable protection (up to 89% of
uncleaved SNAPꢀ25) even when administered 30 minutes post intoxication. This appears to be
the first example of LC inhibitors antagonizing BoNT intoxication in mouse embryonic stem cell
derived motor neurons (mESꢀMNs) in a postꢀexposure model. Oral administration of 16 was well
tolerated in the mouse up to 600 mg/kg, qd. Although adequate unbound drug levels were not
achieved at this dose, the favorable in vitro ADMET results strongly support further work in this
series.
INTRODUCTION
Botulinum neurotoxins (BoNTs) are proteins produced by the Gramꢀpositive anaerobic
bacterium Clostridium botulinum. They are amongst the most potent toxins known and are
causative agents of botulism, a serious and lifeꢀthreatening illness in humans and animals.1 There
are at least seven distinct serotypes (AꢀG), however, three of them (A, B and E) are considered
the most noxious in humans.2,3 The majority of efforts have been focused on identification of
BoNT/A inhibitors with intracellular activity, because antibodyꢀbased treatments were found
ACS Paragon Plus Environment
2

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
successful only before toxin enters a neuron.4 Recent reviews have covered the in vitro ꢀ in vivo
gap in this field,4 focusing on small molecule nonꢀpeptide inhibitors (SMNPI), potential drug
targets and the MOA of these SMNPIs.5,6 Very recently, several approaches using SMNPIs have
been explored by targeting the host intoxication pathways not by directly inhibiting the toxins’
proteolytic activity but focusing on Src family kinase signaling,7 thioredoxin reductase system8,9
and by the development of phosphatase inhibitors that would have an impact on BoNT
intoxication in motor neurons.10 The alternative approach was also developed, the inhibitors that
interact with BoNT/A LC; among them the hydroxamates11 and aminoquinolines showed high
inhibitory activities against BoNT/A LC in cellꢀfree assays.12ꢀ16 12,13,14,15,16 Diverse cellꢀbased
assays have been used as more relevant methods for evaluation of new drugs capable of
protecting SNAPꢀ25 from BoNT/A LC cleavage.17ꢀ20 17,18,19,20 More importantly, these methods
simulate all key steps in intoxication process, starting from holotoxin binding to the cell surface
to cleavage of the SNARE proteins.21,17 Since there is a continuous need for discovering new
therapeutics for treatment of BoNT/A intoxication, this method is considered to be a very helpful
tool for narrowing the spectrum of compounds for testing in animal models. Ex vivo assays, such
as the mouse hemidiaphragm assay for BoNT/A induced muscle paralysis are also widely used
for evaluating new drug candidates (e.g. quinolinꢀ8ꢀol inhibitors22 and EGA23). Despite several
attempts,23ꢀ27 23,24,25,26,27 there is still no inhibitor that is significantly effective in an animal model
of intoxication, especially when administered post toxin injection. Recently published complexes
of Cu(II) with dithiocarbamate and bis(thiosemicarbazone) are also only capable of extending
time to death of BoNT/A intoxicated mice.28 Similarly, newly published mercaptoacetamide
inhibitor ABS 252 proved to be effective in extending survival of BoNT intoxicated mice.29
Presently, the only SMNPI with antiꢀBoNT/A activity in the mouse model is Dyngo-4a which
administered 3 h postꢀintoxication provides 30% mice survival for 24 h.30 For the treatment of
BoNT intoxication, it is necessary for inhibitors to reach the neuromuscular junction and enter
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
3

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
6
7
8
9
the neuronal cell. Thus, adequate drug exposure at the target tissue is key, and in vitro ADMET
parameters such as solubility, microsomal stability, permeability and plasma protein binding can
be used as an effective means to down select active compounds capable of providing acceptable
PK properties.
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
We designed our new BoNT/A LC inhibitors according to structureꢀbased docking
simulations.16 Since docking simulations indicated that besides steroids the benzothiophene
derivatives of aminoquinoline also fit into the binding cleft of BoNT/A LC (keeping the main
interactions with amino acid residues within the active site, c.f. Supporting information), we
decided to investigate the contribution of other carriers to the inhibitory activity (Chart 1).
Here, we report on the synthesis, pharmacokinetic analysis and detailed evaluation of
new steroidal, benzothiophene, thiophene and adamantane 4ꢀaminoquinoline inhibitors of
BoNT/A LC and their respective inhibitory potencies against BoNT/A holotoxin in mouse
embryonic stem cell derived motor neurons (mESꢀMNs).
Chart 1. General structures of investigated steroidal, benzothiophene, thiophene and adamantane
derivatives
ACS Paragon Plus Environment
4

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
OAc
previous
work
Cl
N
1
NH
Inh BoNT/A LC: 96%
Ki = 0.285 µM
Cl
NH
HN
OAc
NH
H
OAc
N
8
9
NH
2
NH
Inh BoNT/A LC: 89%
IC50 = 0.81 µM
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
Ki = 3.22 µM
H2N
OAc
H
BENZOTHIOPHENES
Cl
R2
R1
N
R'
N
R
()n
NH
varying R1, R2, R3, X and n
this work
R''
X
S
THIOPHENES
X
OAc
R2
NH
N
R3
S
()n
N
this work
()n
NH
R
R1
N
R1
change of
the carrier
N
H2N
OAc
ADAMANTANES
H
X
R
H
N
X
N
()n
R1 R2
()m
STEROIDS
N
CHEMISTRY
Continuing our search for efficient BoNT/A inhibitors, we report on new steroidal derivatives
with improved inhibitory activities against BoNT/A LC. Here, we focus on steroidal derivatives
with a basic amino group at C(3), varying the linker connecting the steroidal and aminoquinoline
components. Also, we examined three other classes of aminoquinoline derivatives with
benzothiophene, thiophene and adamantane carriers for their inhibitory activities against
BoNT/A LC. All steroidal, benzothiophene, thiophene and adamantane derivatives were
synthesized according to our established procedures (Schemes 1 and 3).16,31,32 All tested
inhibitors were fully characterized and their purities were >95% (as determined by HPLC,
Supporting Information).
Based on our earlier findings,16 the newly synthesized steroidal inhibitors were chosen for
essential SAR studies within the series. To that purpose, we varied the substitution at carbon
atoms next to nitrogens, at NꢀC(24) and substitution at quinoline C(7) position. Steroidal
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
4
5
6
7
8
9
derivatives were obtained by reductive amination starting from C(24) alcohol derivative 3.16
After removal of protecting group, compounds 14, 16, 17 and 19ꢀ23 were obtained in moderate
to excellent yield (48ꢀ97%). In addition, subjecting the Bocꢀprotected derivatives to reductive
amination with 37% formaldehyde and NaBH3CN followed by removal of the protecting group
afforded Nꢀmethylated derivatives 15 and 18 in acceptable yields (56ꢀ60%). Due to the presence
of the methyl group at NꢀC(α), compounds 17ꢀ21 were obtained as mixture of diastereomers and
were tested as such (vide infra). According to our docking model only C(7) (and not (C12))
acetoxy group is directing structural element featuring the Zn2+ chelation so favorably
contributing to inhibitor’s diving into catalytic cleft (Fig. S11ꢀS13).16 The inhibitor 24 (68%)
was obtained by hydrolysis of acetoxy groups in 16 with potassium hydroxide in methanol
(Scheme 1).
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 novel steroidal derivatives 14-24
OAc
R2
4: n=1, R=Boc, R1,R2,R3,X=H (60%)
12: n=1, R=Boc, R1,R2,X=H, R3=Me (87%)
5: n=1, R=Boc, R1,R2,R3=H, X=Cl (66%)
6: n=1, R=Boc, R1=Me, R2,R3,X=H (59%)
13: n=1, R=Boc, R1,R3=Me, R2,X=H (82%)
7: n=1, R=Boc, R1,R3,X=H, R2=Me (65%)
8: n=1, R=Boc, R1=Me, R2,R3=H, X=Cl (67%)
9: n=1, R=Boc, R1,R3=H, R2=Me, X=Cl (77%)
10: n=3, R=Boc, R1,R2,R3,X=H (54%)
11: n=3, R=Boc, R1,R2,R3=H, X=Cl (58%)
14: n=1, R,R1,R2,R3,X=H (48%)
OAc
ii)
ii)
N
R3
OH
NH
i)
R1
N
RHN
OAc
H
BocHN
OAc
316
H
X
OAc
OH
iv)
N
H
N
H
NH
iii)
NH
15: n=1, R,R1,R2,X=H, R3=Me (56%)
16: n=1, R,R1,R2,R3=H, X=Cl (97%)
17: n=1, R1=Me, R,R2=H, R3,X=H (71%)
18: n=1, R,R2,X=H, R1,R3=Me (60%)
19: n=1, R,R1,R3,X=H, R2=Me (94%)
20: n=1, R,R2,R3=H, R1=Me, X=Cl (97%)
21: n=1, R,R1,R3=H, R2=Me, X=Cl (72%)
22: n=3, R,R1,R2,R3,X=H (76%)
N
H2N
OAc
N
H2N
OH
H
H
Cl
16
24: 68%
Cl
23: n=3, R1,R2,R3=H, X=Cl (92%)
i) 1) PCC, CH2Cl2; 2) aminoquinoline, MeOH, NaBH4; ii) HCHO, MeOH, ZnCl2, NaBH3CN; iii) CF3COOH, CH2Cl2;
iv) KOH, MeOH, 5 days, 60 °C
Steroidal derivative 30 with the amide functional group next to quinoline was prepared in order
to explore the effect of changes in pKa and conformation on inhibitory activity. NꢀBoc protected
derivative 29 was obtained in 43% yield by reductive amination of steroidal C(24) aldehyde and
amine 28 (prepared in few steps starting from 4,7ꢀdichloroquinoline). It has been noticed that
ACS Paragon Plus Environment
6

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
amine 28 decomposes to 25 and pyrrolidineꢀ2ꢀone, therefore, it was used in next step without
detailed characterization. The final compound 30 was obtained in good yield (80%) after
removal of the Boc protecting group (Scheme 2).
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 2. Synthesis of steroidal derivative 30
O
O
Cl
N
NH2
Cl
N3
HN
N
HN
N
i)
ii)
iii)
iv)
Cl
N
Cl
Cl
Cl
25: 80%
26: 89%
27: 75%
OAc
O
N
H
NH2
NH
HN
v)
RHN
O
N
OAc
H
N
Cl
28: 66%
Cl
29: R= Boc (43%)
30: R=H (80%)
vi)
o
o
i) (NH4)2CO3, phenol, 110 to 165 C, 3.5 h; ii) 4-chlorobutanoyl chloride, Et3N, CH2Cl2, 0 C to r.t.; iii) NaN3,
DMF, 80 oC; iv) Ph3P, THF, H2O, 65 oC; v) 1) aldehyde, dry MeOH, 2) NaBH4; vi) CF3COOH, CH2Cl2
Synthesis of novel benzothiophene and thiophene derivatives is presented in Scheme 3.
Benzothiophene and thiophene derivatives 33, 34, 42ꢀ45 and 50 were obtained by reductive
amination from corresponding aldehydes (31, 39ꢀ41 and 49) in moderate yield (12ꢀ77%).
Compound 47 was obtained from 45 after removal of TMSꢀgroup in moderate yield.
Benzothiophene and thiophene Nꢀmethyl tertiary amino derivatives, 37, 38 and 48 were
synthesized by reaction of formaldehyde with secondary amines 35, 36 and 46 respectively,
reported in our previous work.31,33
Scheme 3. Synthesis of novel benzothiophene (33, 34, 37, 38) and thiophene (42-44, 47, 48, 50)
derivatives
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
4
5
6
7
8
R1
33: R=F, R1,R2,R3=H, X=Cl (56%)
34: R=F, R1=Me, R2,R3=H, X=H (77%)
35: R=F, R1,R2,R3=H, X=H (50%)33
37: R=F, R1,R2=H, R3=Me, X=H (74%)
CHO
R
R
NH
i)
N
R3
R2
N
ii)
ii)
31
S
S
36:
R=CN, R1,R3=H, R2=Me, X=H (21%)
31: R=F31
32: R=CN31
38: R=CN, R1=H, R2,R3=Me, X=H (74%)
X
9
42: n=1, R=CN, R1=H, X=H (65%)
43: n=9, R=CN, R1=H, X=H (40%)
N
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
R1
N
i)
R
NH
44:
n=1, R=CH2CN, R1=H, X=H (12%)
45: n=1, R=C CTMS, R1=H, X=Cl (48%)
47:
S
()n
S
X
R
iii)
ii)
CHO
n=1, R=C CH, R1=H, X=Cl (49%)
39: R=CN31
46: n=5, R=CN, R1=H, X=H (26%)31
40: R=CH2CN
48: n=5, R=CN, R1=Me, X=H (50%)
R=C CTMS31
41:
NH
i)
NC
CHO
NC
50:
(54%)
S
NH
S
N
49
i) 1) aminoquinoline/aldehyde, AcOH glac, MeOH/CH2Cl2, rt, 2 h 2) NaBH4, rt, 12 h; ii) HCHO, MeOH, ZnCl2, NaBH3CN; iii) K2CO3, MeOH, rt
Syntheses of novel adamantane derivatives are presented in Scheme 4. Adamantane derivatives
with tertiary nitrogen 56ꢀ65, 68 and 69 were obtained using the same procedures as mentioned
above in moderate to excellent yield (55ꢀ93%) after coupling of prepared amines (51ꢀ 55, AQ11,
AQ7, AQ8)13,32,34 to corresponding aldehydes. Key reaction for synthesis of adamantane
derivatives 77ꢀ82 was the Pdꢀmediated Buchwald coupling of previously prepared amines (70ꢀ
75)32 to 4ꢀchloroquinoline in moderate to excellent yield (44ꢀ91%). Compounds 84 and 85 were
obtained in high yield by methylation of compounds 78 and 83, respectively.
Scheme 4. Synthesis of novel adamantane derivatives 56-69, 77-82, 84 and 85
ACS Paragon Plus Environment
8

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
R2 R1
56: n=0, m=1, R=Me, R1,R2=H, X=Cl (82%)
R2 R1
()n
()m
57: n=1, m=1, R=Me, R1,R2=H, X=Cl (87%)
58: n=2, m=1, R=Me, R1,R2=H, X=Cl (93%)
59: n=0, m=1, R,R2=Me, R1=H, X=Cl (91%)
60: n=1, m=1, R,R2=Me, R1=H, X=Cl (87%)
61: n=0, m=2, R=Me, R1,R2=H, X=Cl (89%)
62: n=1, m=2, R=Me, R1,R2=H, X=Cl (84%)
63: n=2, m=2, R=Me, R1,R2=H, X=Cl (86%)
64: n=0, m=2, R,R2=Me, R1=H, X=Cl (93%)
65: n=1, m=2, R,R2=Me, R1=H, X=Cl (80%)
66: n=1, m=1, R,R1,R2,X=H (60%)
HN
N
N
R
()n
HN
N
NH2
1) i) or iii)
2) ii)
(except for
66 and 67
X
)
X
51: n=0, R1,R2=H, X=Cl13
52: n=1, R1,R2=H, X=Cl13
53: n=2, R1,R2=H, X=Cl13
54: n=0, R2=Me, R1=H, X=Cl32
55: n=1, R2=Me, R1=H, X=Cl32
AQ11: n=0, R1,R2=H, X=H34
AQ7: n=1, R1,R2=H, X=H32
AQ8: n=2, R1,R2=H, X=H32
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
67: n=1, m=2, R,R1,R2,X=H (70%)
ii)
68: n=1, m=1, R=Me, R1,R2,X=H (72%)
69: n=1, m=2 R=Me, R1,R2,X=H (55%)
R1
R2
77: n=0, m=1, R,R2=H, R1=Me (60%)
78: n=1, m=1, R,R2=H, R1=Me (91%)
()m
N
iv)
HN
()n
H2N
NH
()m
()n
ii)
R
84: n=1, m=1, R,R1=Me, R2=H (89%)
79: n=2, m=1, R,R2=H, R1=Me (44%)
80: n=0, m=2, R,R2=H, R1=Me (50%)
81: n=1, m=2, R,R2=H, R1=Me (51%)
82: n=2, m=2, R,R2=H, R1=Me (53%)
83: n=1, m=1, R,R1=H, R2=Me (73%)32
85: n=1, m=1, R,R2=Me, R1=H (70%)
R2 R1
N
70: n=0, m=1, R2=H, R1=Me32
71: n=1, m=1, R2=H, R1=Me32
72: n=2, m=1, R2=H, R1=Me32
73: n=0, m=2, R2=H, R1=Me32
74: n=1, m=2, R2=H, R1=Me32
75: n=2, m=2, R2=H, R1=Me32
76: n=1, m=1, R2=Me, R1=H32
ii)
i) 1) aminoquinoline/aldehyde, AcOH glac, MeOH/CH2Cl2, rt, 2 h 2) NaBH4, rt, 12 h; ii) HCHO, MeOH, ZnCl2, NaBH3CN; iii)
aldehyde, NaBH(OAc)3, CH2Cl2, rt, 24 h; iv) 4-chloroquinoline, Pd(OAc)2, SPhos, K3PO4, dioxane, 85 oC, 24 h
RESULTS
Recently, we discovered steroidal bisꢀaminoquinoline (bisꢀACQ) inhibitors with 90ꢀ97%
BoNT/A LC inhibition and Ki values within the range 0.103ꢀ0.389 ꢁM (such as 1, Chart 1).16
However, several comments regarding our hit compounds should be considered. Simplifying the
original structure should be desired since the synthesis of bisꢀACQ derivatives and purification
thereof is rather demanding task. In addition, the compounds with MW > 900, although of
natural product origin, are not commonly considered drugꢀlike and could have difficulties in
reaching the target as well. Therefore, we have chosen the compound 2 (Chart 1),16 with IC50 =
0.81 ꢁM and Ki = 3.22 ꢁM in HPLCꢀbased assay and 77% of SNAPꢀ25 protection at 30 ꢁM in
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
preꢀintoxication scenario in embryonic chicken spinal primary neurons as very promising
starting point for further improvement of the steroidal inhibitors.
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
Evaluation of Inhibitory Activity Against BoNT/A LC.
Inhibitory activities against BoNT/A LC in proteolytic and against holotoxin in cellꢀbased assay
are presented in Table 1. In vitro activities for all other compounds that were evaluated only
against BoNT/A LC are given in Supporting Information (Table S1).
We employed a wellꢀdefined 17ꢀmer peptide (termed P39, acetylꢀSNKTRIDEANQRATKMLꢀ
amide, that contains the SNAPꢀ25 scissile bond)35 HPLCꢀbased proteolytic assay. Synthesized
steroidal derivatives have shown BoNT/A LC inhibition up to 85% at a standard 20 µM
concentration (compound 18) and IC50 values ranging from 0.7ꢀ7.1 µM (Table 1, Table S1). Our
control compound 1, in the current proteolytic HPLCꢀbased assay showed 90% inhibition of
BoNT/A LC at 20 ꢁM, in excellent agreement with results reported previously (95% of
inhibition).16 Compound 24 with hydroxy groups at C(7) and C(12), with 48% inhibition of
BoNT/A LC appeared to be much less potent than other steroidal derivatives. The negative effect
of the amide functionality on BoNT/A LC inhibition (30, 37%) was also observed. Introduction
of a methyl group into the linker or secondary nitrogen and changing its position, as well as the
variations in the presence of chlorine atom at the quinoline moiety influenced the inhibitory
activity to certain extent.
Benzothiophene derivatives showed good inhibition activities too, with IC50 values ranging
within 3.3ꢀ10.2 µM, thus being comparable to the steroidal series. By changing the position of
cyano group from C(6) to C(5) the inhibition at 20 ꢁM concentration was sharply improved
(Table 1, Table S1, compounds 36 (61%), 87 (75%), 86 (84%) vs 88 (23%), 89 (8%)).
ACS Paragon Plus Environment
10

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Derivatives with cyano group instead of fluorine atom at C(5), showed higher degree of
inhibition (Table 1, Table S1, compounds 86 (84%), 87 (75%) vs 35 (34%), 90 (56%)), while the
compounds without tertiary nitrogen appeared to be more active at 20 ꢁM concentration than
their methylated analogs (Table 1, Table S1, compounds 91 (69%), 36 (61%) vs 92 (48%), 38
(27%)). In thiophene series, the most potent inhibitors proved to be 46 and 93 with cyano
substituent and long methylene linker (6 and 8 methylene groups and 70ꢀ80% and 77% of
inhibition, respectively). Compound 46 (MV150) with 70ꢀ80% of inhibition has been utilized as
positive control, for evaluating the performance of the 17ꢀmer HPLCꢀbased proteolytic assay.
Changing the position of aminoquinoline linker connection to phenyl moiety from paraꢀ to metaꢀ
had no significant effect on inhibitory activity (Table 1, 46 (70ꢀ80%) vs 50 (68%)). Improvement
of the inhibitory activity was observed in derivatives without chlorine atom at C(7) position of
aminoquinoline or without tertiary nitrogen (Table 1, Table S1, compounds 93 (77%), 46 (70ꢀ
80%) vs 94 (14%), 95 (30%) and 96 (35%), 48 (55%)). In addition, cyano derivatives were found
to be superior inhibitors than their ethynyl analogs (Table 1, Table S1, compounds 97 (36%), 46
(70ꢀ80%) vs 47 (19%), 98 (53%)). Examined adamantane derivatives bearing chlorine atom on
aminoquinoline moiety proved to be more active in the HPLCꢀbased proteolytic assay than their
desꢀchloro analogs (Table 1, Table S1, compounds 57 (71%), 60 (67%), 99 (64%) vs 69 (31%),
85 (51%), 100 (42%) and other analogs), and that differs from the previously mentioned
thiophene series. Tertiary nitrogen and chlorine atom at C(7) of aminoquinoline moiety together
had the most favorable effect on the inhibitory activity (57 and 60, 71% and 67% of inhibition,
respectively, Table 1).
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 addition, experimental logD values are presented in Table 1. LogD values are obtained using
reversedꢀphase thinꢀlayer chromatography at two different pH: pH=1 (MeOH/HCl (70/30)) for
compounds 14ꢀ19, 24, 46, 50, 93, 96 and 101 and pH=10 (acetone/NH3/H2O (85/5/10)) for
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
compounds 34, 36-38, 57, 60, 86, 87, 91, 102ꢀ106.36 For compound 16 logD was also determined
4
5
6
at physiological pH=7.3, using shakeꢀflask method (octanol/TBS buffer).37ꢀ39 37,38,39
7
8
9
Table 1. Inhibitory activities against BoNT/A LC and holotoxin in proteolytic and cellꢀbased
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
assaya
In vitro
proteolytic
assay
mES-MNs pre
intoxication
% of full
length SNAP-
25 (10 µM; 20
µM)
mES-MNs post
intoxication at
20 µM % of full
length SNAP-25
(30 min; 60 min)
Compoundb exp LogDc
% inh
BoNT/A LC
at 20 µM; IC50
(µM)
2.14
(pH=1)
80; 5.7
52; 1.5
28; 67
70; 69
48; 49
14
2.11
(pH=1)
15
ꢀ
2.55
(pH=1);
66; 4.5
72; 88
64; 45
16
2.07
(pH=7.3)d
2.28
(pH=1)
71; 2.7
85; 0.7
75; 3.0
48
67; 69
20; 41
86; 87
46; 58e
58; 68
34; 53
44; 55
59; 64
21; 26
36; 43
38; 22
17
2.38
(pH=1)
18
ꢀ
2.21
(pH=1)
50; 22
19
2.97
(pH=1)
49;e ꢀ
24
3.85
(pH=10)
75; 7.4
61; 10.2
46
16; 17
34
3.67
(pH=10)
36
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
3.98
(pH=10)
37
3.42
(pH=10)
27
38
3.67
(pH=10)
84; 4.6
75; 3.3
86
3.85
(pH=10)
87
3.98
(pH=10)
69; 8.8
62; 62
22; 48
91
ꢀ
ꢀ
3.08
70ꢀ80; 3.4
46
ACS Paragon Plus Environment
12

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(pH=1)
3.08
(pH=1)
3.42
(pH=1)
3.33
(pH=1)
2.93
(pH=1)
3.24
(pH=10)
3.47
(pH=10)
68; 8.7
77; 6.8
35
46; 71
50
93
ꢀ
31; 70
54; 60
18; 39
51; 56
63; 72
20; 19
34; 22
ꢀ
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
67; 9.3
71; 8.8
67; 11.7
101
57
46; 39
15; 16
60
30
ꢀ
51; 5.2
30; 31
23; 24
47; 46
68
81
85
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
3.40
(pH=10)
4.33
(pH=10)
3.24
(pH=10)
3.09
(pH=10)
4.13
(pH=10)
65; 2.7
29; 30
30; 29
31; 35
24; 23
30; 32
102
103
104
105
106
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
2
15
ꢀ
ꢀ
Negative
control
DMSO
Positive
control
1
100
100
ꢀ
ꢀ
ꢀ
90; 12.4f
70; 84
36; 20
aResults are given as mean value of three independent experiments. Syntheses of compounds
36, 46, 86, 87, 91, 93, 96 and 101 were reported in our previous work.31 Syntheses of compounds
102ꢀ106 were reported in our previous work.32 cExperimental LogD using reversedꢀphase thinꢀ
layer chromatography at pH=1 (MeOH/HCl (70/30)) or pH=10 (acetone/NH3/H2O (85/5/10)).
dExperimental LogD in octanol/TBS buffer at pH=7.3 using shakeꢀflask method. eCompound 24
was tested at 8 and 16 μM in preꢀintoxication model and at 16 μM in 30 minutes postꢀ
intoxication model only; fIn previous test the inhibition was 95.46% and Ki = 0.285ꢁM.16
b
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
Inhibitory Activity Against Holotoxin in mES-MNs.
4
5
6
7
Preꢀintoxication model
8
9
Based on the activity in primary screen and structural diversity (Table 1, Table S1), 30
compounds were chosen for secondary screening in cellꢀbased assay in mouse ESꢀcell derived
motor neurons (mESꢀMNs) in preꢀintoxication model at two concentrations – 10 and 20 ꢁM
(Table 1). SNAPꢀ25 protein cleavage was measured by Westernꢀblot analysis. In this model,
compounds were added 30 minutes prior to holotoxin (500 pM) and intoxicated for 4 hours.
Inhibitors tested during BoNT/A challenge in mESꢀMNs in preꢀintoxication model were found to
afford uncleaved SNAPꢀ25 up to 88% at 20 ꢁM concentration (compound 16), with steroidal
compounds being generally the most promising.
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 contrast to in vitro results, using this assay we can substantially differentiate our steroidal
inhibitors from benzothiophene derivatives, which were capable to protect SNAPꢀ25 from
cleavage up to 68% at 20 ꢁM concentration. Another issue is important to note – benzothiophene
derivatives (36, 86 and 87) with cyano instead of fluorine substituent were much less active
despite the high percent of inhibition in HPLCꢀbased assay. In addition, in this model the
compound 38, shows significantly higher protection of SNAPꢀ25 in comparison with nonꢀ
methylated analogues, although it would be eliminated based on its poor inhibitory activity
(27%) in primary screening. Three thiophene derivatives have shown to be capable of protecting
SNAPꢀ25 from cleavage more than 60% at 20 ꢁM concentration (50, 93 and 96), while
adamantane derivatives were found to be less active, despite promising results obtained in
primary screening. Only one adamantane 60, showed 72% protection of SNAPꢀ25 at 20 ꢁM.
Twelve compounds with good results obtained in preꢀintoxication model (56ꢀ88% SNAPꢀ25
protection at 20 ꢁM concentration, Table 1) were subjected to preꢀexposure doseꢀresponse
experiment in concentration range from 0.1 to 20 ꢁM (Figure 1). The results obtained in doseꢀ
ACS Paragon Plus Environment
14

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
response experiment at 20 ꢁM concentration are similar to those obtained in the initial
experiment at two concentrations, given in Table 1. The obtained doseꢀresponse results clearly
indicate that introduction of chlorine atom at C(7) position of quinoline moiety highly improved
the activity of steroidal inhibitors (16 vs 14; IC50 ~ 10 ꢁM vs IC50 = 10ꢀ20 ꢁM, respectively).
Another pair of derivatives, 19 with methyl group next to NꢀC(24) and 17 with methyl group
next to quinoline moiety, also merit attention. Significantly higher protection of SNAPꢀ25 is
observed with 19 at 20 ꢁM concentration, while at lower concentrations both derivatives
exhibited comparable inhibitory activity. Except mentioned four steroidal derivatives, compound
1 showed remarkable activity in preꢀintoxication scenario (IC50 < 10 ꢁM). As the inhibitor 24
could arise as putative metabolite in vivo, despite its lower activity in HPLCꢀproteolytic assay
compared to other steroidal derivatives, we were intrigued by moderate activity of this
compound in preꢀintoxication model (58% at 16 μM, Table 1). Tested at nine concentrations
0.25 to 64 ꢁM it also showed doseꢀdependent behavior, with IC50 =8ꢀ16 ꢁM (Figure S1, Table
S5). Benzothiophene derivatives 34 and 38 showed IC50 values in 10ꢀ20 ꢁM range. In addition,
three thiophene (50, 93 and 96) and two adamantane (57 and 60) 4ꢀaminoquinoline derivatives
were evaluated in doseꢀdependent preꢀintoxication model, in order to examine the scope of
carriers coupled to aminoquinoline moiety. To our pleasure, not only steroidal and
benzothiophene aminoquinolines showed remarkable activities in doseꢀdependent manner, but
also thiophene and adamantane derivatives. As one can find in Figure 1, 60 has shown the
highest protection of SNAPꢀ25 at 20 ꢁM concentration (90%) and IC50 < 10 ꢁM. Good doseꢀ
response and high percent of inhibition is also seen for 93 and 96 (IC50 = 10ꢀ20 ꢁM and IC50 = 5ꢀ
10 ꢁM, respectively).
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
15

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Protection of SNAPꢀ25 in mESꢀMNs in preꢀintoxication model (results are given as
mean value of three independent experiments +/ꢀ SEM, values given in Table S2).
Postꢀintoxication model
The most effective derivatives were tested postꢀsymptomatically (in triplicate, 30 and 60 minutes
following 500 pM holotoxin administration) at 20 ꢁM concentration (Figure 2). Compound 60,
which was the foremost candidate in preꢀintoxication model, unfortunately failed in postꢀ
exposure model with only 15% of intact SNAPꢀ25. From all tested derivatives, compound 16
highlights with 64% of SNAPꢀ25 cleavage protection when administered 30 minutes postꢀ
intoxication, and 45% protection when neurons were treated 60 minutes after BoNT/A
administration. Other derivatives were significantly less active, with steroidal series still being
the most promising.
ACS Paragon Plus Environment
16

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Protection of SNAPꢀ25 in mESꢀMNs in postꢀintoxication model at 20 ꢁM (results are
given as mean value of three independent experiments +/ꢀ SEM, values given in Table S3).
Since compound 16 showed excellent behavior in postꢀintoxication scenario (Figure 2 and Figure
S2, Table S6), and compound 24 is considered as a potential metabolite which is reasonable to
expect to be formed in mouse gut and liver, we subjected both 16 and 24 to 30 minutes postꢀ
exposure doseꢀresponse experiment. Compounds were tested in duplicate at 9 concentrations,
starting from 0.25 ꢁM → 64 ꢁM (Figure 3). In this run 16 exhibited a doseꢀdependent protection
of SNAPꢀ25, with IC50 ~8 ꢁM. To the best of our knowledge, this is the most active smallꢀ
molecule inhibitor of BoNT/A LC in postꢀintoxication scenario in mESꢀMNs (89% of full length
SNAPꢀ25 at 32 μM). The activity of compound 24 in mESꢀMNs (69% full length SNAPꢀ25 at 64
ꢁM) was somewhat surprising; however, observation that both, the compound 16 and its putative
metabolite were active in mESꢀMNs strongly supports further examination of biological activity.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Protection of SNAPꢀ25 in mESꢀMNs by compounds 16 and 24 administered 30 min
postꢀintoxication – dose response experiment 0.25 to 64 ꢁM (results are given as mean value of
two independent experiments +/ꢀ SEM, values given in Table S4). IC50 ~8 ꢁM. Compound 16
showed potential toxicity at 64 μM, since GAPDH (used as loading control) levels were lower.
In the independent doseꢀresponse experiment, 16 showed protection of SNAPꢀ25 up to 64% at
20 ꢁM (Figure S2, values given in Table S6).
Toxicity studies.
In vivo toxicity studies of several derivatives subjected to postꢀintoxication assay was estimated
in a mouse model. As can be seen from Table 2, 16, 34, 93 and 96 proved to be completely nonꢀ
toxic at the given dose (all 5 mice survived 30 days after administration and showed normal
appearance and behavior).
ACS Paragon Plus Environment
18

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
Table 2. Toxicity study in micea
4
5
6
Compound
Mice alive/
7
8
9
(at 3 × 160 mg/kg dose)
Total mice
5/5
16
19
34
57
60
93
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
3/5
5/5
4/5
3/5
5/5
5/5
aGroups of five healthy mice were treated per os (p.o.) for three consecutive days with
aminoquinolines suspended in 0.5% hydroxyethylcellulose ꢀ 0.1% Tween 80. Individual mouse
behavior and appearance was monitored two times a day for 30 days.
In the separate host toxicity studies, two groups of 5 healthy mice were subjected to oral
administration of the compound 16 at higher concentrations (400 and 600 mg/kg, single dose),
and even at the highest applied concentration, 16 proved to be nonꢀtoxic (all 5 mice survived 30
days after administration and showed normal appearance and behavior).
Pharmacokinetic Analysis of Compound 16.
ADMET parameters
Given its ability to antagonize BoNT/A in mESꢀMNs in postꢀintoxication model (in full
accordance with previous step analyses), 16 appeared as the best candidate for further testing.
ADMET parameters for this compound are presented in Table 3. Compound 16 showed good
stability in both human and mouse liver microsomes (halfꢀlife >60 min) and very good stability
in human plasma (>85% remaining at 1 h). It showed solubility >50 ꢁg/mL (Table 3) as
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
9
determined by laser nephelometry and confirmed by more precise shakeꢀflask method with
spectrofluorimetric detection (640 ꢁg/mL). In MDR1ꢀMDCK permeability assay, 16 showed
excellent tissue penetration (45.3×10ꢀ6 cm/s in A→B direction and 24.6×10ꢀ6 cm/s in B→A
direction). It acts as moderate to poor inhibitor of five CYP450 enzymes (which play important
role in the drug metabolism). One in vitro parameter possibly limiting this compound as potential
drug is high plasma protein binding, >99% according to equilibrium dialysis method. However,
the efficacy of the drug transport could not be attributed to the PPB only; instead, the binding
constants with major transporters in the blood should be considered. In addition to dialysis, the
interaction between 16 and HSA and AGP proteins has been studied by monitoring the changes
in fluorescence spectrum of HSA and AGP upon addition of increasing amounts of 16
(Ksv=(4.56±0.27)×104 for HSA; Ksv=(6.99±0.25)×105 for AGP).
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 3. ADMET parameters for compound 1640
ADMET Properties
Purity of the sample (%)a
>90
>50
>60
<23
>85
1ꢀ10
45.3
0.54
>99
Solubility at pH 7.4 (ꢁg/mL)b
T1/2 (min)
Stability – Microsomesc
CLint (ꢁL/min/mg)
Stability – Human plasma (% remaining at 1 h)d
CYP450 Inhibition IC50 (ꢁM)e,f
Papp (aꢀb, 10ꢀ6 cm/s)
MDR1ꢀMDCKg
Pgp Efflux Ratio
Plasma protein binding (%)h
b
aPurity analysis was determined by LCꢀMS/MS. Solubility (n=3) was determined using laser
c
nephelometry to measure light scattering. Substrate depletion experiments were performed by
incubating test compound with liver microsomes (human and mouse) for 1 hour at 37°C. dPlasma
e
stability was determined following incubation for 1 hour at 37°C. The IC50 value was
determined from the net fluorescent signal from incubation at 37°C for a set time with an active
f
cytochrome P450 enzyme and a fluorescent probe substrate. For CYP3A4/BQ (1.1 ꢁM),
ACS Paragon Plus Environment
20

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
CYP3A4/DBF (1.2 ꢁM), CYP3A4/BFC (1.4 ꢁM), CYP2C19/CEC (5 ꢁM) and CYP2D6/AMMC
(>10 ꢁM). The MDR1/MDCK apparent permeability constant (Papp) was determined after 1
g
hour incubation at 37 °C of the compound on the apical compartment (A to B) or basolateral
compartment of the cell monolayer (B to A). The Efflux Ratio is the ratio of the Papp in the B to
A direction divided by the Papp in the A to B direction. According to equilibrium dialysis
5
6
7
8
h
method.
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
In Vivo Mouse Pharmacokinetics
Aminoquinoline 16 was selected for determination of concentration levels in mouse serum and
evaluation of plasma protein binding due to its promising activity in mESꢀMNs, nonꢀtoxicity,
and very good ADMET properties. Compound 16 was dosed orally, in a single dose, at two
different concentrations – 400 and 600 mg/kg to groups of 7 and 6 mice, respectively. For both
doses, maximal concentration of the drug in the blood was determined by UPLCꢀMS/MS in
samples previously treated with acetonitrile (15 → 120 min, Table 4 and Figure 4).
Table 4. Concentration levels of 16 in mouse serum at different time points following oral
administration at two different doses.a
400 mg/kg dose
µg/mL (µM)
2.3 (3.3)
600 mg/kg dose
µg/mL (µM)
8.5 (11.9)
Time (min)
15
22
30
ꢀb
5.5 (7.8)
2.8 (4.0)
2.3 (3.3)
2.0 (2.8)
<LOD
4.2 (5.9)
2.2 (3.1)
<LOQc
2.8 (4.0)
<LOQd
45
60
90
120
<LOD
aConcentration ꢁg/mL of compound was determined by UPLCꢀMS/MS. Concentration in ꢁM
is given in parentheses. Results are given as a single value for one mouse for each time point.
LOD = limit of detection; LOQ = Limit of quantitation. bSample was not taken. Concentration
c
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
d
was estimated to 1.8 ꢁg/mL (2.5ꢁM)<LOQ. Concentration was estimated to 1.7 ꢁg/mL
(2.4ꢁM)<LOQ.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Pharmacokinetics of 16 in mice serum after oral dosing at two concentrations – 400
and 600 mg/kg.
Obtained in vivo pharmacokinetic properties are summarized in Table 5. In addition, mice
samples were analyzed for compound 24, monoacetylated analogue and corresponding M+16
and M+32 metabolites, in order to examine possible deacetylation in the mouse gut. None of
these metabolites were detected.
Table 5. In vivo pharmacokinetic properties for 16 in mice after oral administration at two
different concentrations
parameter
400 mg/kg dose
600 mg/kg dose
Tmax (min)
22
7.8
253
43
15
11.9
526
23
Cmax (ꢁM)
AUC0ꢀ120 (ꢁMꢂmin)
t1/2 (min)a
ACS Paragon Plus Environment
22

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
time Tc > 3ꢁM (min)
>45
>45
4
5
Plasma protein binding (%)b
>99%
>99%
6
7
8
a
b
t1/2 (min) values calculated from graph on Fig. 4 using GraphPad Prism 6; According to no
detection of free concentration (ultracentrifugation method, see Materials and Methods)
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
DISCUSSION
Logical step in the investigation of drug’s antiꢀBoNT/A potential was the evaluation of the
candidates in a cellꢀbased assay using mESꢀMNs, since they mimic the whole intoxication
process. We proved that our inhibitors protect SNAPꢀ25 from cleavage in a doseꢀdependent
manner, when administered prior to holotoxin (Figure 1). Thiophene derivative 96 has shown the
IC50 = 5ꢀ10 ꢁM, benzothiophene derivatives 34 and 38 IC50 = 10ꢀ20 ꢁM and among steroidal
derivatives 16 singled out with IC50 ~ 10 ꢁM. Adamantane derivative 60 seemed to be the
favorable candidate with 90% of SNAPꢀ25 protection at 20 ꢁM and IC50 < 10 ꢁM. The
discrepancy between results obtained in proteolytic HPLCꢀbased assay and in mESꢀMNs for
cyano derivatives 36, 86 and 87 (61ꢀ84% inhibition vs 26ꢀ53% SNAPꢀ25 protection at 20 ꢁM)
could be attributed to their inability to enter the neurons. Regarding compound 38 different MoA
might be considered since this compound does not act as LC inhibitor (27% inhibition, Table 1)
but it moderately protects SNAPꢀ25 in mESꢀMNs in preꢀintoxication scenario (IC50 = 10ꢀ20 ꢁM,
Figure 1).
Preꢀexposure model for BoNT/A intoxication is very convenient for discovering drugs with good
permeability, and the results of inhibitory activities could be useful for prophylactic purposes.
More significant challenge is postꢀexposure model, which demonstrates the ability of compound
to protect SNAPꢀ25 from BoNT/A cleavage after holotoxin endocytosis. Noteworthy results of
our inhibitors in postꢀintoxication model in mESꢀMNs offered us valuable information that
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
9
newly synthesized derivatives not only enter the cells, but also very probably inhibit to great
extent the BoNT/A LC inside the neuronal cytosol. From all examined derivatives, we managed
to select steroidal compound 16 which antagonizes BoNT/A holotoxin up to 89% at 32 ꢁM when
administered 30 minutes postꢀintoxication. More importantly, the essential difference between
our previous hit bisꢀACQ compounds came out – compound 1 with < 40% protection failed in
postꢀintoxication scenario, thus justifying the synthesis of novel C(3)ꢀamino derivatives.
In the early stages of drug discovery, the optimization of physicochemical properties (e.g.
lipophilicity) can greatly improve a compound’s chance in obtaining the proper balance among
permeability, solubility and metabolism; and the experimental LogD of 2.07 (pH=7.3) obtained
for compound 16 was considered optimal in that regard.41 Subsequent ADMET and PK studies
were performed in order to determine if compound 16 was capable of providing adequate drug
exposure. As supported by its LogD value, compound 16 displayed favorable ADMET
properties. In an MDR1/MDCK assay it showed permeability constant Papp>20, predictive for
good tissue penetration and the efflux ratio << 2.5, thus not being a substrate for Pꢀgp. PPB for
this compound was estimated to >99% according to equilibrium dialysis method, however,
spectrofluorimetric measurements of 16 binding for HSA and AGP, indicate that interaction
between 16 and plasma proteins is optimal to enable transport of the drug and to release it at its
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
target (Ksv=104ꢀ106). This experimental observation is supported by detailed reviews on albuminꢀ
42,43,44
drug and AGPꢀdrug interactions.42ꢀ44
Due to its nontoxicity in vivo at very high
concentration (up to 600 mg/kg), we provided pharmacokinetic analysis in mice. Compound 16
administered at 600 mg/kg dose provides Cmax up to 11.9 ꢁM (8.5 ꢁg/mL, Table 4) and is
detectable in mouse serum up to 120 min after administration (Figure 4). Unfortunately, detailed
PK analysis was published only for few inhibitors submitted to in vivo studies, so detailed
analysis and comparison with 16 is not possible.4
ACS Paragon Plus Environment
24

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
CONCLUSION
4
5
6
7
8
Synthesized compounds antagonize BoNT/A LC and BoNT/A in mESꢀMNs in a doseꢀdependent
manner in both preꢀ and postꢀintoxication models. To the best of our knowledge, this is the first
example of LC inhibitors antagonizing BoNT intoxication in mouse ESꢀcell derived motor
neurons in a postꢀexposure model (few SMNPIs of BoNT/A LC active in embryonic chick spinal
motor neurons were reported previously45). Compound 16 proved to be potent inhibitor of
BoNT/A holotoxin in mESꢀMNs in postꢀintoxication scenario and well tolerated in the mouse up
to 600 mg/kg, p.o. Binding constants with major transporters in blood (HSA and AGP) are
within desirable values, suggesting that inspite of high PPB it could be delivered to its target.
With very good ADMET properties and a plasma Cmax >10 ꢁM in mice after oral
administration, we believe that compound 16 has distinguished itself from other lead BoNT
inhibitors in the literature.4 Current efforts are focused on reducing protein binding,
determining/mitigating the clearance route of 16 in mouse and determining if optimal
formulation and/or subcutaneous administration could improve its PK profile and merit testing in
a mouse lethality model. Noteworthy, there are no post symptomatically administered LC
inhibitors to date that are significantly efficacious in a mouse model of BoNT intoxication.4
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
EXPERIMENTAL SECTION
Chemistry. Melting points were determined on a Boetius PMHK apparatus and were not
corrected. IR spectra were recorded on a ThermoꢀScientific Nicolet 6700 FTꢀIR diamond crystal
1
spectrophotometer. H and 13C NMR spectra were recorded on a Varian Geminiꢀ200
spectrometer (at 200 and 50 MHz, respectively), and a Bruker Ultrashield Advance III
spectrometer (at 500 and 125 MHz, respectively) in the indicated solvent (vide infra) using TMS
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
5
6
7
8
9
as the internal standard. Chemical shifts are expressed in ppm (δ) values and coupling constants
(J) in Hz. ESI–MS (HRMS) spectra of the synthesized compounds were acquired on a Agilent
Technologies 1200 Series instrument equipped with Zorbax Eclipse Plus C18 (100 × 2.1 mm i.d.
1.8 ꢁm) column and DAD detector (190ꢀ450 nm) in combination with a 6210 TimeꢀofꢀFlight
LC/MS instrument in positive and negative ion mode. The samples were dissolved in MeOH
(HPLC grade). The selected values were as follows: capillary voltage 4 kV; gas temperature 350
°C; drying gas 12 L minꢀ1; nebulizer pressure 45 psig; fragmentator voltage: 70 V. Mass spectral
analyses were done using electrospray ionization in positive ion mode on a Surveyor separations
module coupled to a ThermoFinnigan TSQ AM triple quadrupole mass spectrometer. Gas
chromatography tandem mass spectrometry (GCꢀMS) analyses were performed on an Agilent
7890A GC (Agilent) system equipped with a 5975C inert XL EI/CI MSD and a flame ionization
detector (FID) connected by capillary flow technology through a 2ꢀway splitter with makeꢀup
gas. An HPꢀ5 MS capillary column (Agilent Technologies, 25 mm i.d., 30 m length, 0.25 μm
film thickness) was used. The flash chromatography was performed on Biotage SP1 system
equipped with UV detector and FLASH 12+, FLASH 25+ or FLASH 40+ columns charged with
KPꢀSIL (40 – 63 µm, pore diameter 60 Å), KPꢀC18ꢀHS (40 – 63 µm, pore diameter 90 Å) or KPꢀ
NH (40 – 63 µm, pore diameter 100 Å) as an adsorbent. Elemental analyses were realized with
an Elemental Vario EL III microanalyser. Compounds were analyzed for purity (HPLC) using a
Agilent 1200 HPLC system equipped with Quat Pump (G1311B), Injector (G1329B) 1260 ALS,
TCC 1260 ( G1316A) and Detector 1260 DAD VL+ (G1315C). Compound 42 was analyzed for
purity (HPLC) using Waters 1525 HPLC dual pump system equipped with an Alltech, Select
degasser system, and dual λ 2487 UVꢀVIS detector. All tested inhibitors were fully characterized
and their purities were >95% (as determined by HPLC, c.f. Supporting Information). HPLC
analysis was performed in two diverse systems for each compound. Specific HPLC methods are
as follows: Method A: Zorbax Eclipse Plus C18 4.6 × 150 mm, 1.8µ, S.N. USWKY01594 was
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
26

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
used as the stationary phase. Eluent was made from the following solvents: 0.2% formic acid in
water (A) and methanol (B). The analysis were performed at the UV max of the compounds (at
330 nm for compounds 14ꢀ16, 18, 19, 22, 23, 34, 37, 38, 50, 56ꢀ65, 66ꢀ69, 77ꢀ82, 84, 85 and 93,
and at 254 nm for compound 17) to maximize selectivity. Compounds were dissolved in
methanol, final concentrations were ~1 mg/mL. Flow rate was 0.5 mL/min. Method B: Zorbax
Eclipse Plus C18 4.6 x 150 mm, 1.8µ, S.N. USWKY01594 was used as the stationary phase.
Eluent was made from the following solvents: 0.2% formic acid in water (A) and acetonitrile
(B). The analysis were performed at the UV max of the compounds (at 330 nm for compounds
14ꢀ19, 22, 23, 34, 37, 38, 44, 50, 56ꢀ65, 66ꢀ69, 77ꢀ82, 84, 85 and 93) to maximize selectivity.
Compounds were dissolved in methanol, final concentrations were ~1 mg/mL. Flow rate was 0.5
mL/min. Method C: Zorbax Eclipse Plus C18 2.1 x 100 mm, 1.8µ, S.N. USUXU04444 was
used as the stationary phase. Eluent was made from the following solvents: 0.2% formic acid in
water (A) and methanol (B). The analysis was performed at the UV max of the compound (at
330 nm for compounds 20, 21, 24, 43, 44, 47; at 270 nm for compound 30, and at 254 nm for
compound 33) to maximize selectivity. Compound was dissolved in methanol, final
concentration was ~1 mg/mL. Flow rate was 0.2 mL/min. Method D: Zorbax Eclipse Plus C18
2.1 x 100 mm, 1.8µ, S.N. USUXU04444 was used as the stationary phase. Eluent was made
from the following solvents: 0.2% formic acid in water (A) and acetonitrile (B). The analysis
was performed at the UV max of the compound (at 254 nm for compounds 20, 24, 33; at 270 nm
for compound 30 and at 330 nm for compounds 21, 43 and 47) to maximize selectivity.
Compound was dissolved in methanol, final concentration was ~1 mg/mL. Flow rate was 0.2
mL/min. Method E: Poroshell 120 ECꢀC18, 4.6 x 50mm, 2.7ꢁ, S.N. USCFU07797 was used as
the stationary phase. Eluent was made from the following solvents: 0.2% formic acid in water
(A) and acetonitrile (B). The analysis was performed at the UV max of the compound (330 nm
for compound 48) to maximize selectivity. Compound was dissolved in methanol, final
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
27

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
4
5
6
7
8
9
concentration was ~1 mg/mL. Flow rate was 0.5 mL/min. Method F: Poroshell 120 ECꢀC18, 4.6
x 50mm, 2.7ꢁ, S.N. USCFU07797 was used as the stationary phase. Eluent was made from the
following solvents: 0.2% formic acid in water (A) and methanol (B). The analysis was performed
at the UV max of the compound (330 nm for compound 48) to maximize selectivity. Compound
was dissolved in methanol, final concentration was ~1 mg/mL. Flow rate was 0.5 mL/min.
Method G: Symmetry C18, 4.6 x 150 mm, 5 ꢁm, S.N. 021336278136 37 was used as the
stationary phase. Eluent was made from the following solvents: 0.2% formic acid in water (A)
and methanol (B). The analysis was performed at the UV max of the compound (340 nm for
compound 42) to maximize selectivity. Compound was dissolved in methanol, final
concentration was ~1 mg/mL. Method H: Nucleosil C18, 4 x 150 mm, 5 ꢁm was used as the
stationary phase. Eluent was made from the following solvents: 0.2% formic acid in water (A)
and methanol (B). The analysis was performed at the UV max of the compound (340 nm for
compound 42) to maximize selectivity. Compound was dissolved in methanol, final
concentration was ~1 mg/mL.
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
Procedure A: General procedure for the synthesis of N-Cbz protected aminoquinolines 107
and 109.13 The mixture of 4,7ꢀdichloroquinoline/4ꢀchloroquinoline (1 equiv) and monoꢀCbz
protected diaminoalkane (1.1 – 1.2 equiv) was slowly heated to 80 °C for 1 h, and the mixture
was continued for 6ꢀ8 h at 120ꢀ130 °C. After cooling to r.t., reaction mixture was transferred to
the separation funnel using CH2Cl2/1M NaOH. The organic layer was washed with 1M NaOH,
water and brine. The organic layer was dried over anhydrous Na2SO4 and solvent was evaporated
under reduced pressure. Crude product was purified using column chromatography.
Procedure B: General procedure for the obtainment of steroidal derivatives 4-11 and 29.16
Alcohol (1 equiv) was dissolved in CH2Cl2. PCC (1.5 equiv) was added, and the mixture was
stirred at r.t. for 3.5 h. Reaction mixture was filtered through a short column of SiO2 (eluent
ACS Paragon Plus Environment
28

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CH2Cl2/EtOAc = 7/3). Crude aldehyde was dissolved in dry MeOH, aminoquinoline (1.5 equiv)
was added, and mixture was stirred at r.t. overnight. NaBH4 (2 equiv) was added, and stirring
was continued at r.t. for 12 h. Solvent was removed under reduced pressure and crude mixture
was prepared for column purification.
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
Procedure C: General procedure for the removal of the Boc-protecting groups with TFA
for compounds 14-23 and 30. A solution of the NꢀBocꢀprotected amine in TFA/CH2Cl2 (v:v;
1:10), was stirred at r.t. for 6 h. Solvents were evaporated under reduced pressure and the residue
was treated with CH2Cl2/2.5M NaOH. The organic layer was dried over MgSO4, and the solvent
was evaporated under reduced pressure.
Procedure D: General procedure for N-methylated aminoquinolines 12, 13, 37, 38, 48, 56-
65, 68, 69 and 84, 85.46 To a stirred solution of aminoquinolines (1 equiv) in MeOH containing
37% aqueous formaldehyde (2 equiv), the mixture of ZnCl2 (2 equiv) and NaHB3CN (4 equiv) in
MeOH was added. After the reaction mixture was stirred at r.t. for 4 h, the solution was taken up
in 0.1 M NaOH and most of MeOH was evaporated under reduced pressure. Aqueous solution
was extracted with CH2Cl2, the combined extracts were washed with water and brine and dried
over anhydrous Na2SO4. The solvent was evaporated under reduced pressure.
Procedure E: General procedure for reductive amination to produce compounds 33, 34, 42-
45, 50, 66 and 67. Amine (1.5 equiv) and appropriate aldehyde (1 equiv) were dissolved in
MeOH/CH2Cl2 mixture (v:v; 2:1), glac. AcOH (1.5 equiv) was added, and the mixture was
stirred under Ar atmosphere at r.t. After 3 h, NaBH4 (6 equiv) was added, and stirring was
continued for another 18 h. Solvent was removed under reduced pressure, and the residue was
dissolved in CH2Cl2. The organic layer was washed with 2M NH4OH, water and then extracted
ACS Paragon Plus Environment
29