A chemical switch for transforming a purine agonist for toll-like receptor 7 to a clinically relevant antagonist

Article abstract of DOI:10.1021/acs.jmedchem.0c00011

Toll-like receptor 7 (TLR7) is an established therapeutic target in myriad autoimmune disorders, but no TLR7 antagonist is available for clinical use to date. Herein, we report a purine scaffold TLR7 antagonist, first-of-its-kind to our knowledge, which was developed by rationally dissecting the structural requirements for TLR7-targeted activity for a purine scaffold. Specifically, we identified a singular chemical switch at C-2 that could make a potent purine scaffold TLR7 agonist to lose agonism and acquire antagonist activity, which could further be potentiated by the introduction of an additional basic center at C-6. We ended up developing a clinically relevant TLR7 antagonist with favorable pharmacokinetics and 70.8% oral bioavailability in mice. Moreover, the TLR7 antagonists depicted excellent selectivity against TLR8. To further validate the in vivo applicability of this novel TLR7 antagonist, we demonstrated its excellent efficacy in preventing TLR7-induced pathology in a preclinical murine model of psoriasis.

Full text of DOI:10.1021/acs.jmedchem.0c00011

Subscriber access provided by UNSW Library
Article
A Chemical switch for transforming a purine agonist for
Toll-like receptor 7 to a clinically relevant antagonist
Ayan Mukherjee, Deblina Raychaudhuri, Bishnu Prasad Sinha, Biswajit Kundu, Mousumi
Mitra, Barnali Paul, Purbita Bandopadhyay, Dipyaman Ganguly, and Arindam Talukdar
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00011 • Publication Date (Web): 17 Apr 2020
Downloaded from pubs.acs.org on April 17, 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 50
Journal of Medicinal Chemistry
A chemical switch for transforming a purine agonist for Toll-like
receptor 7 to a clinically relevant antagonist
1
2
3
4
5
6
7
8
9
Ayan Mukherjee,†,ψ,§ Deblina Raychaudhuri,‡,ψ Bishnu Prasad Sinha,‡,ψ Biswajit Kundu,† Mousumi
Mitra,‡ Barnali Paul,†,§ Purbita Bandopadhyay,‡,§ Dipyaman Ganguly,*,‡ Arindam Talukdar*,†
†Department of Organic and Medicinal Chemistry, CSIR-Indian Institute of Chemical Biology, 4 Raja S.
C. Mullick Road, Kolkata 700032, WB (India).
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
‡IICB-Translational Research Unit of Excellence, CSIR-Indian Institute of Chemical Biology, Kolkata,
India.
§Academy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, India
ψThese authors contributed equally to the study
*Email: atalukdar@iicb.res.in; dipyaman@iicb.res.in
Supporting Information Placeholder
ABSTRACT: Toll-like receptor 7 (TLR7) is an established therapeutic target in myriad autoimmune
disorders, but no TLR7 antagonist is available for clinical use to date. Herein we report a purine scaffold
TLR7 antagonist, first-of-its-kind to our knowledge, which was developed by rationally dissecting the
structural requirements for TLR7-targeted activity for a purine scaffold. Specifically, we identified a
singular chemical switch at C-2 that could make a potent purine scaffold TLR7 agonist to loose agonism
and acquire antagonist activity, which could further be potentiated by the introduction of an additional
basic center at C-6. We ended up developing a clinically relevant TLR7 antagonist with favorable
pharmacokinetics and 70.8% oral bioavailability in mice. Moreover, the TLR7 antagonists depicted
excellent selectivity against TLR8. To further validate the in vivo applicability of this novel TLR7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 50
antagonist we demonstrated its excellent efficacy in preventing TLR7-induced pathology in a preclinical
murine model of psoriasis.
1
2
3
4
5
6
7
8
INTRODUCTION
9
Toll-like receptors (TLRs) are the critical first line of immune defense that are mostly expressed
on antigen presenting cells, e.g. dendritic cells and recognize molecular signature patterns associated
with pathogens.1-3 The members of the TLR family of receptors that can recognize nucleic acid
molecules, both of self and non-self origins, viz. TLR7, TLR8 and TLR9, are located inside the acidic
endosomal compartments of the immune cells.2,4-6 Among them, TLR7, which recognizes single-
stranded RNA, is expressed in humans selectively in plasmacytoid dendritic cells (pDCs) and B
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
lymphocytes, and when activated in pDCs drives type I interferon production from pDCs.4 DC-derived
P
type I IFNs are critical innate immune events in a number of autoimmune diseases, viz. systemic lupus
erythematosus, psoriasis, Sjögren’s syndrome, systemic sclerosis etc.7-11 Thus the endosomal TLRs in
pDCs, e.g. TLR7, are important therapeutic targets for these clinical contexts.12-14 Application of the
TLR7 agonist imiquimod on the skin leads to cutaneous autoimmunity in mice, which mimics the
human skin autoimmunity in psoriasis, and consequently is an established preclinical model for
psoriasis.15 But a suitable TLR7 antagonist is yet to be available for clinical use.16
Both receptor agonists and antagonists are important therapeutic molecules in different clinical
contexts and in many cases agonists and antagonists share overlapping binding region in their target
molecules. Thus, agonistic chemical scaffolds are often used as a template for designing antagonists
with shared region-specificities for the receptors.17-19 For example, development of β-adrenergic
receptor antagonists from the endogenous agonist adrenalin,20 histamine receptor 2 antagonists from
histamine21 and opioid receptor antagonists from opioids.22 TLR7 mediated immune activation has
implication in autoimmune diseases like systemic lupus erythematosus (SLE), psoriasis. In SLE
autoreactivity against RNA and RNA binding proteins are very prevalent and these immune complexes
ACS Paragon Plus Environment

Page 3 of 50
Journal of Medicinal Chemistry
activate TLR7 in both pDCs and B cells, thereby perpetuating the inflammatory cascade.10 In psoriasis
self-RNA molecules released by dying keratinocytes form complexes with cationic antimicrobial
peptides. These complexes have been shown to activate both plasmacytoid and conventional dendritic
cells thus fuelling the cutaneous inflammation.8,10 Hence, blocking TLR7 activation with suitable
antagonist is sought for, which may intercept one of the key pathogenic nodes in these diseases. Purine
scaffold is extensively studied for its potent TLR7 agonist activity.23-26 Interestingly, guanosine, an
endogenous purine class molecule, is also an agonist for TLR7.27 The mechanism of TLR7 activation by
guanosine has been shown through crystallographic study which is of great value in design and
development of TLR7 ligands.28 Previous efforts directed towards the purine scaffold have led to potent
agonists’ thorough understanding about the region-specificities of critical substituents. From
crystallographic evidence and structural development of TLR7 agonist, it is understandable that -NH2
group at C-6 position is crucial for scaffold recognition by TLR7.28 Thus, logically we did not change -
NH2 group at C-6 position in our initial SAR development. Development of 2-substituted-8-
hydroxyadenine as IFNα inducer via TLR7 pathway revealed the profound effect of four carbon long
alkoxy chain at C-2 position on the agonistic activity.25,29 Among various substituents, O-alkyl groups at
C-2 position have been found to have superior ability to modulate the agonistic activity in combination
with N-benzyl or N-alkylamine at N-9 position.30-32 Previous study on pteridinone core based TLR7
agonist showed that C-2 position butoxy linker bind to the conserved hydrophobic pocket of TLR7.33
The N-9 substitutions modulate both activity and pharmacokinetic profile of purine,23,24 perhaps because
the weak base gets protonated and trapped in the acidic endosomal compartment,34,35 harbouring TLR7.
Throughout the literature the C-2 position was substituted and subsequently modification was done in
N-9 position but not the other way round. Keeping this in mind, we intended to evaluate the fate of the
agonist where C-2 substituent is absent keeping the known relevant N-9 substituents and –NH2 group at
C-6 position. Although extensive focus has been devoted towards development of TLR7 agonist, there
is no reported TLR7 antagonist of purine scaffold, to our knowledge. Imiquimod belongs to
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 50
imidazoquinoline class of molecules, which also includes other well-known TLR7 agonists like
gardiquimod and resiquimod.36 TLR7 agonist gardiquimod was successfully transformed into TLR7
antagonist by converting to its 3H regioisomer.18
1
2
3
4
5
6
7
8
9
We have initiated our study from a representative purine agonist 7 containing essential TLR7
agonistic features at C-2, C-6, C-8 and N-9. We’ve explored the minimal structural features that impart
antagonism in compound 7, for mapping the synthetic path to a novel purine scaffold antagonist. We
have depicted that a small structural modification in the ligand can lead to major changes in their
functional activity. For assessing the TLR7-targeted activity (both agonism and antagonism) of the
molecules, we used engineered HEK293 cells that expressed exogenous human TLR7 and reported
TLR7-induced NF-κB activation in a secreted alkaline phosphatase-driven colorimetric assay. We have
used a non-selective TLR7 antagonist hydroxychloroquine as a positive control. Through SAR
development, we identified clinically relevant TLR7 antagonist 23 that has favourable pharmacokinetic
properties with 70.8% oral bioavailability in mice. The nature of antagonism was determined by
subjecting these antagonists to agonism assay. The TLR7 antagonists reported herewith showed
excellent selectivity against TLR8. In vivo efficacy of 23 was demonstrated by its ability to prevent
TLR7-induced pathology in a preclinical murine model of psoriasis.
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
CHEMISTRY
Chemical synthesis of TLR7 agonist 7 and 9. C-6 position of compound 130 was substituted
with n-butoxy linkage via sodium t-butoxide treatment to obtain compound 2. The methoxy group at C-
8 was introduced by bromination of 2 using N-bromosuccinimide followed by refluxing with a 5 M
solution of sodium methoxide in methanol to provide compound 3 (Scheme 1). Deprotection of pyran
ring from compound 3 using 10% TFA followed by 3-chloropropyl substitution in compound 4 using 1-
bromo-3-chloropropane yielded compound 5. Chlorine group in compound 5 was then substituted with
ACS Paragon Plus Environment

Page 5 of 50
Journal of Medicinal Chemistry
either piperidine or ethylpiperazine to give compound 6 and 8. Finally, C-8 position methoxy group was
deprotected using 4 M HCl in 1,4-dioxane to get compound 7 and 9.
1
2
3
4
5
6
Scheme 1a
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
aReagents and conditions: (i) Sodium t-butoxide (4.0 equiv.), n-BuOH, 100 ºC, 12 h, 77%; (ii) a. N-
bromosuccinimide (1.05 equiv.), CHCl3, 3 h, rt, b. sodium methoxide (2.7 equiv.), CH3OH, reflux, 4 h,
80%; (iii) 10% TFA in methanol, rt, 72 h, 73%; (iv) 1-bromo-3-chloropropane (2.0 equiv.), potassium
carbonate (2.6 equiv.), rt, 16 h, 67%; (v) piperidine for compound 11 or ethylpiperazine for compound
12 (3.0 equiv.), potassium carbonate (1.5 equiv.), rt, 6 h, 67% - 73%; (vi) 4 M HCl in 1,4-dioxane, 3 h,
87% - 91%.
Chemical synthesis of weak TLR7 antagonist 14. In the present scheme 2, we successfully
standardized a protocol for the synthesis of compound 14 by avoiding protection-deprotection chemistry
at N-9 position in purine scaffold. Bromo group of 8-bromo adenine 1037 was substituted with methoxy
group by refluxing with a 5 M solution of sodium methoxide in methanol to obtain compound 11, which
on treatment with 1-bromo-3-chloropropane provided compound 12. Chlorine group was substituted by
ethylpiperazine to yield compound 13 followed by deprotection of methoxy group gave compound 14.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 50
Scheme 2a
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
aReagents and conditions: (i) 5 M solution of sodium methoxide in methanol (10 equiv.), reflux, 12 h,
72%; (ii) 1-bromo-3-chloropropane (2.0 equiv.), potassium carbonate (2.6 equiv.), rt, 16 h, 78%; (iii)
ethylpiperazine (3.0 equiv.), potassium carbonate (1.5 equiv.), rt, 6 h, 63%; (iv) 4 M HCl in 1,4-dioxane
3 h, 86%.
Chemical synthesis of potent TLR7 antagonist. Substitution of 3-carbon linker with
ethylpiperazine group at N-9 position of 6-chloropurine (15) resulted in the intermediate compound 16
(Scheme 3). In the formation of compound 16 there is a mechanistic possibility of getting two structural
isomers namely N-9 isomer and N-7 isomer. Here we obtained exclusively N-9 isomer and structure
was confirmed by the X-ray crystallographic structure of hydrochloride salt of intermediate compound
16 (SI, Fig S6). Thereafter, the chloride group at C-6 position was replaced by free –NH2 group by
treating with ammonia in ethanol at 60 ºC to derive compound 17. Reduction of C-6 chloro group using
palladium-carbon in 16 gave compound 18. The chloro group at C-6 of intermediate 16 was substituted
with different organic amine such as dimethylamine, piperidine, morpholine, piperazine,
ethylpiperazine, and cyclopentylpiperazine yielded compound 19, 20, 21, 22, 23 and 24 respectively.
The N-9 position of 25 was treated with 1-bromo-3-chloropropane to get chloro intermediate 26, which
was subsequently reacted with various azacyclic bases such as piperazine, morpholine, piperidine and
pyrroline to yield compound 27, 28, 29 and 30 respectively. Compound 31 with propyl substitution at
N-9 position was made from 25 using the same protocol used for the preparation of 26. Structural
ACS Paragon Plus Environment

Page 7 of 50
Journal of Medicinal Chemistry
conformation of compound 23 was made by the analysis of NOESY, HMBC spectra, the detail is in
1
2
3
supplementary information.
4
5
6
Scheme 3a
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
aReagents and conditions: (i) 1-(3-chloropropyl)-4-ethylpiperazine (2.0 equiv.), potassium carbonate
(2.0 equiv.), DMF, 60 ºC, 12 h, 84%; (ii) ammonia in ethanol (7.0 equiv.), 60 ºC, 12 h, 67%; (iii) H2,
Pd/C, THF, rt, 2 h, 91%; (iv) respective amine, potassium carbonate (2.0 equiv.), CH3CN, reflux, 12 h,
73% - 95%; (v) 1-bromo-3-chloropropane for (26) and 3-bromopropane for (31) potassium carbonate
(2.0 equiv.), DMF, rt, 12 h, 71% - 78%; (vi) respective amine, potassium carbonate (2.0 equiv.), DMF,
60 ºC, 12 h, 78% - 93%.
RESULTS AND DISCUSSION
The importance of butoxy at C-2, –NH2 at C-6 and hydroxyl at C-8 in TLR7-agonism by purine
chemotypes is established.38,39 We selected compound 7 as a starting point as it contains all the essential
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 50
features of a TLR7 agonist. The synthetic strategy (Scheme 1) for compound 7 allowed us to readily
introduce an additional basic center at N-9. Replacing piperidine with ethylpiperazine, led to the
enhancement in agonism in 9 (Scheme 1), which had an EC50 of 0.9 µM as opposed to 2.6 µM for 7 (Fig
1a, 1b. Table 1). Based on the previous SAR studies it has been established that substituents such as
butoxy at C-2 position are a key feature for TLR7 agonism.23,25 As discussed before, we hypothesized
that C-2 butoxy may be targeted to transform the potent TLR7 agonist 9 into a TLR7 antagonist. Indeed,
removal of C-2 butoxy in 14 (Scheme 2), led to the disappearance of TLR7 agonism and acquisition of
considerable antagonism (Fig 1c, 1d. Table 1). Thus, we were able to identify a ‘chemical switch’ at C-2
that could make a potent purine scaffold TLR7 agonist to loose agonistic properties and acquire
antagonistic properties towards TLR7.
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
a
b
c
d
9
14
7
9
7
9
14
9
f
e
14
23
14
SAR
23
Figure 1. Discovery of TLR7 antagonist from known TLR7 agonist. (a,c,e) Conversion of TLR7
agonist 7 (EC50 = 2.6 µM) to a more potent TLR7 agonist 9 (EC50 = 0.9 µM) by the addition of
ethylpiperazine group (a) compound 9 to weak TLR7 antagonist 14 (IC50 = 43.1 µM) by the elimination
of butoxyl group (c) 3 to potent TLR7 antagonist 23 (IC50 = 4.7 µM) by systematic SAR study (e).
(b,d,f) HEK-Blue™ TLR7 reporter cells were incubated overnight in the presence of indicated
concentrations of given compounds and on the following day the culture supernatants were subjected to
a colorimetric assay to measure the degree of TLR activation/inhibition. Data is represented as Mean ±
SEM.
ACS Paragon Plus Environment

Page 9 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
This novel purine scaffold molecule 14, despite showing significant antagonism to human TLR7,
had an IC50 (43.1 µM) too high for a therapeutically relevant antagonist. So, we explored if omission of
other essential agonist-specific molecular features from 14 can help to develop more potent antagonists
with translational potential. We found that removal of the hydroxyl group at C-8 led to rather a loss of
all activity against TLR7 in compound 17 at 50 µM. The importance of the –NH2 group was evident
from the crystal structure of guanosine as well as the chemical ligands from imidazoquinoline class
too.27 Also, the SAR study at C-6 position of purines has depicted the significance of –NH2 group.40
Accordingly, we focused on the modification at C-6 position. Neither complete deletion of –NH2 group
from C-6 in compound 18 nor converting the –NH2 at C-6 to dimethylamino in compound 19 brought
antagonistic activity towards TLR7 (Table 1) at 50 µM. Next, we extended the dimethylamino group of
compound 19 with different azacylic moieties like piperidine, morpholine and piperazine. Interestingly,
piperidine substitution at C-6 position in compound 20 regained the TLR7 antagonistic effect and the
compound shows moderate TLR7 antagonism with an IC50 value of 23.9 µM. Replacement of
piperidine with morpholine in compound 21 showed comparable TLR7 antagonism with an IC50 of 25
µM. Since morpholine cannot be further extended at 4’ position we introduced piperazine in compound
22 with an aim for further derivatization, but the TLR7 antagonistic potency decreased to 39.3 µM.
However, ethyl substitution on piperazine NH group in compound 23 regained the TLR7 antagonistic
effect in single digit micromolar level with an IC50 value of 4.7 µM. Of note here, an ethylpiperazine
group at N-9 led to enhancement of TLR7 agonism, while introducing the same group at C-6 potentiated
TLR7 antagonism. Further logical extension of ethyl group in compound 23 with more hydrophobic
cyclopentane group in compound 24 led to slight decrease in the TLR7 antagonist activity (IC50 value of
6.2 µM, Table 1). Figure 1 is a snapshot of our SAR development depicting minimal structural
requirements and a singular chemical switch that could transform a TLR7 agonist into a TLR7
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
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 50
antagonist in the same purine scaffold, which is represented in a TLR7 reporter assay done in parallel
using an established TLR7 agonist (CL264), the purine agonists 7 and 9 as well as the TLR7 antagonist
23 which has been shown to inhibit CL264-induced TLR7 activation (Fig 1g).
1
2
3
4
5
6
7
8
9
Table 1. SAR depicting the conversion of agonist to antagonist
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
TLR7
TLR7
Antagonism
(µM) IC50
Comp.
No.
R2
R3
R4
Agonism
(µM) EC50
2.6±0.05
0.9±0.01
R1
-OBu
-OBu
H
-NH2
-NH2
-NH2
-NH2
H
-OH
-OH
-OH
H
7
a
a
9
14
17
18
19
20
21
22
b
b
b
b
b
b
b
43.1±4.1
H
a
H
H
a
H
NMe2
H
a
H
H
23.9±3.1
25.0±3.5
39.3±3.7
H
H
H
H
H
H
23
b
4.7±0.4
H
H
H
H
H
H
H
H
H
H
H
H
24
27
28
29
30
31
b
b
b
b
b
6.2±1.5
37%C
40.0±4.2
15.8±1.8
48.3±4.9
H
b
NT
a
Hydroxychloroquine
8.6±1.2
aNo significant inhibition at 50 µM. No significant activation at 50 µM. % Inhibition at 50 µM. NT-
b
c
not tested. Hydroxychloroquine is a positive control
Initially, in compound 9 it was observed that the introduction of weak base at N-9 can modulate
TLR7 activity. Thus in our next development, we further explored the role of weak base at N-9 position
10
ACS Paragon Plus Environment

Page 11 of 50
Journal of Medicinal Chemistry
with different groups such as morpholine, pyrrolidine and piperidine. Replacing ethylpiperazine group
in compound 23 with piperazine in 27 led to drastic fall in TLR7 antagonist activity (37% inhibition at
50 µM). The result suggests that hydrophobic ethyl group is essential for TLR7 antagonistic activity.
The result correlates well with our first set of SAR results where unsubstituted piperazine at C-6 in
compound 22 showed 8 fold less potency than ethyl substituted piperazine group in 23. Further
substitution with morpholine, piperidine and pyrrolidine yielded less potent TLR7 antagonist compound
28, 29 and 30 with IC50 40.0, 15.8, 48.27 µM respectively. Expectedly, elimination of weak base at N-9
in compound 31 led to complete loss of TLR7 antagonism at 50 µM (Table 1). To ensure the validity of
TLR7 antagonism assay, the ability of different doses of hydroxychloroquine to inhibit TLR7 activation
was studied as a positive control. Comparison between compound 17 with 23 suggests that introduction
of ethylpiperazine group with an additional basic center and an extended hydrophobic attachment at C-6
had profound potentiating effect on TLR7 antagonism in purine scaffold, whereas additional basic
center at N-9 can modulate the TLR7-targeted activity (either agonism as in 7 and 9, or antagonism as in
23 and 29).
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
None of the TLR7 antagonists showed any TLR7 agonism upto 50 µM dose (SI, Fig S2) which
ruled out the possibility of partial agonism. As TLR7 and TLR8 show high degree of structural
similarity,19 next we monitored the efficacy of our TLR7 antagonists in TLR8 antagonistic assay and we
found that none of the TLR7 antagonists showed significant TLR8 inhibition (SI, Fig S3) up to 50 µM.
The result strongly suggests that the TLR7 antagonists reported in the manuscript are selective for TLR7
when compared with TLR8 antagonism.41 TLR7 antagonistic effect of compound 23 was further
assessed by evaluating the ability of our compound to inhibit the production of TLR7 triggered
proinflammatory cytokines such as IL-6 and TNFα.41,42 Compound 23 showed 42.9 % inhibition (Fig
2a) of IL-6 gene transcription at IC50 dose (5 µM) in HEK-blue TLR7 cell line. Inhibition of
proinflammatory cytokines was measured by real time PCR using selective TLR7 agonist CL264 as
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 50
positive control. For further validation employing ex vivo human blood-derived model, we observed the
inhibition of TLR7 driven TNFα induction in primary human peripheral blood mononuclear cells
(hPBMCs) by ELISA (Fig 2b). These results led us to conclude that our compound of interest is a potent
TLR7 antagonist capable of inhibiting TLR7 mediated immune responses as depicted by the inhibition
of IL-6 and TNFα production in response to TLR7 activation. Thus, we have successfully identified
hitherto unknown site-specific substitutions that can switch a well-established purine scaffold TLR7
agonist into a TLR7 antagonist, as demonstrated through the TLR7 reporter assay.
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
TLR7 and TLR8 share almost similar ligand binding domain. Both the TLRs have one small
molecule ligand binding site and another ssRNA binding site.27 Zhang et al.19 reported a third binding
site for the inhibition of TLR8 in the rest state. The binding model was extrapolated for finding new
antagonist binding zone in TLR7. Karroum et al. validated this new antagonist binding site in TLR7
through systematic computational analysis of Imidazo[1,2-a]pyrazines, Imidazo[1,5-a]quinoxalines, and
Pyrazolo[1,5-a]quinoxalines class TLR7 antagonist.41 Due to unavailability of human TLR7 crystal
structure we have built homology model structure of TLR7 (SI, Fig S7). Molecular docking study and
validation was performed using Discovery Studio suite 4.1 (SI, Fig S8a), which shows that our TLR7
antagonists also bind in the same region attaining similar pattern of interactions along with newer
interactions (Fig 2c, 2d). The molecular docking indicates that the purine ring of the lead molecule 23 is
sandwiched between two hydrophobic residues Phe329 and Tyr1374 forming π-π stacking interactions
(Fig 2c). Hydrogen bonding between hydroxyl group of Ser452 with nitrogen of C-6 position of
ethylpiperazine was observed in compound 23 as one of the key important interaction signifying the
importance of terminal nitrogen in piperazine group as observed in the structure-activity relationships
between compounds 19, 20 and 21 with dimethylamino, piperidine and morpholine group respectively.
N-9 Position of the alkyl linker with pendent hydrophobic amine was stabilised by the hydrophobic
interaction of Leu331 and Phe329. The significance of ethyl group can be observed by comparing the
12
ACS Paragon Plus Environment

Page 13 of 50
Journal of Medicinal Chemistry
inhibitory potency of compound 23 and 27. Compound 23 showed IC50 value of 4.7 µM whereas
compound 27 without the ethyl group showed 37% inhibition at 50 µM. Similar binding interactions
were observed for other potent TLR7 antagonist compound 24 (Fig 2d). From the molecular docking
analysis of our other TLR7 antagonists we can conclude that along with hydrogen bonding with Ser452,
hydrophobic interactions play a vital role in determining TLR7 antagonism (SI, Fig S8).
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. Inhibition of proinflammatory cytokines and docking study of compound 23 with
homology modelled hTLR7 (a) HEK-Blue™ TLR7 reporter cells were treated with 5 µM compound
23, one hour prior to stimulation with CL264, 6 hours after which, the cells were harvested and
subjected to gene expression studies to monitor the mRNA levels of the IL6 gene relative to that of 18S
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 50
(housekeeping gene) in the presence and absence of compound treatment. Data is represented as Mean ±
SEM, *p<0.05, n=6. Unpaired Student’s t-test was done to measure statistical significance. (b) Primary
human PBMCs were treated with 5 µM compound 23, one hour prior to stimulation with Gardiquimod
and incubated overnight. On the following day the culture supernatants were subjected to ELISA to
measure the degree of TLR activation/inhibition in terms of TNFα production. Data is represented as
Mean ± SEM, *p<0.05, n=4. Paired Student’s t-test was done to measure statistical significance. (c)
Interaction between TLR7 ectodomain (homology model template PDB ID: 5WYZ) and compound 23.
(d) Interaction between TLR7 ectodomain and compound 24. Hydrogen bonds and hydrophobic
interactions are shown by green dash line and grey dash line respectively. *marks residues are in chain
B.
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
Selectivity of TLR7 antagonist 23 over TLR8 can be anticipated through docking analysis.
TLR8 and TLR7 share similar antagonistic binding site except for one single residue change.41 Polar
residue Thr384 for TLR7 is replaced by nonpolar bulky Ile403 in TLR8. The molecular modeling of
compound 23 in TLR7 indicates that the N-9 flexible linker is 5 Å away from Thr384 (Fig 2c). It can be
anticipated that due to the steric bulk of Ile403 compound 23 is unable to bind to TLR8, which could be
the probable reason for selectivity of our antagonists towards TLR7 over TLR8.
Table 2. In vitro pharmacokinetics of compound 23.
Caco-2 Permeability
Plasma stability
humand miced
Microsomal stability
human
−6
P
(10 cm/s)
app
PSA
Aq. sol
μg/ml
2
clogP
NRBa
mice
Å
efflux
ratio
A→Bb
B→Ac
T1/2(mi
n)
T1/2(min
)
QH%
QH%
200.3 1.38 52.6
7
3.57
9.28
2.6
95% 99.2% 90.3
26.5
111.8
14.4
aNumber of rotatable bolds. bApical to basal. cBasal to apical. dremaining after 2h.
14
ACS Paragon Plus Environment

Page 15 of 50
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 3. Pharmacokinetics profiling of potent TLR7 antagonist 23. (a) Doses of 15 mg/kg and 10
mg/kg body weight of compound 23 was orally administered to 3 sets of male C57BL/6 mice. (b) The
dose of 15 mg/kg body weight of compound 23 was intravenously administered to 3 male C57BL/6
mice. (c) Calculated pharmacokinetics parameter in tabulated formed. 2% Ethanol in PBS (0.01M, pH =
7.4) buffer used as vehicle. The concentration of compound present in serum at indicated time points
was quantified by LC-MS/MS (Figure S4).
We selected compound 23 for further investigation. Initially we aimed at gathering insights
about in vitro pharmacokinetics of 23 through a series of established assays. The aqueous solubility of
23 was found to be considerable (200 μg/ml) at pH 7.4 and calculated clogP value was 1.38 (Table 2).
The polar surface area (52.6 Å2) and seven rotatable bonds were also good indicators that compound 23
would have favourable oral bioavailability (Table 2).43 On Caco-2 permeability assay the efflux ratio of
23 was found to be 2.6, suggesting that this molecule would have moderate membrane permeability and
is not likely to be a substrate for efflux transporters, viz. P-glycoprotein. In a nut shell these results
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 50
reflect that compound 23 may have good adsorption in the gastrointestinal tract. Compound 23 also
showed acceptable plasma stability as well as metabolic stability (liver microsomal assay), assessed in
case of both mice and human (Table 2). Moreover, compound 23 showed no cytotoxicity on HEK293
cells in vitro up to a concentration of 100 µM (Fig S4).
1
2
3
4
5
6
7
8
9
Finally, to assess in vivo applicability of this novel TLR7 antagonist 23 we checked the in vivo
pharmacokinetics of the molecule on oral and intravenous administration to C57BL/6 mice in two doses
of 10 mg/kg and 15 mg/kg. The maximum plasma concentration (Cmax) surpassed the in vitro IC50
against TLR7 in 15 mg/kg oral dose whereas in 10 mg/kg dose Cmax remained below the IC50. It was
found that compound 23 was rapidly absorbed into body (Fig 3a) with a half-life of 4.6 hours upon 15
mg/kg oral dose (Fig 3c). As compound 23 is basic nature it shows good Vss (4.2 L/kg, Fig 2c) which
reflect the high duration of drug in target site.44 Finally all the favourable range of in vitro and in vivo
pharmacokinetics (Fig 3) of compound 23 adds up to give excellent oral bioavailability (F = 70.8%, Fig
3a, 3b).
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
16
ACS Paragon Plus Environment

Page 17 of 50
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 4. Efficacy of TLR7 antagonist in rodent psoriasis model. (a) Schematic representation of
experimental protocol followed- Compound 23 or vehicle was orally administered for 7 consecutive
days starting one day before (Day 0) topical application of imiquimod on the dorsal skin surface, which
was done for 6 consecutive days. The following day mice were sacrificed, normal and psoriatic skin
harvested and subjected to histopathological evaluation. (b) (left) Shaved normal dorsal skin before
imiquimod application, (middle) shaved imiquimod treated dorsal skin of vehicle treated mouse on day
7 showing significant scale formation on skin, (right) shaved imiquimod treated dorsal skin of
compound 23 treated mouse on day 7 showing markedly lesser scale formation compared to previous
group. (c) Thickness of skin fold of vehicle treated mice (n=9) and compound treated mice (n=9) were
recorded and plotted every day (3 readings each) to monitor the gradual changes taking place upon daily
imiquimod application. Data is represented as Mean ± SEM. 2% Ethanol in PBS (0.01 M, pH = 7.4)
buffer was used as vehicle.
To validate in vivo efficacy of 23 for TLR7 antagonism in a clinically relevant context, we
utilized an established preclinical model of the cutaneous autoimmune disease psoriasis. Daily
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 50
application of the TLR7 agonist imiquimod to the skin of C57BL/6 mice leads to a psoriasiform
inflammation in 6 days, which closely mimics the human disease in terms of morphology,
histopathology and the underlying immunocellular involvements.15 This preclinical model has not only
been the key support for the pathogenetic role of TLR7 in this clinical context, but also has been
instrumental for discovery of other downstream pathogenetic mechanisms for the disease as well as
validation of most of the clinically used targeted biologic therapies used clinically.45 We administered
15 mg/kg body weight compound 23 by oral gavage to the mice daily from the first day of application
of imiquimod cream and on the 6th day assessed disease development, in comparison with vehicle-
administered control mice having developed imiquimod-driven psoriasis (Fig 4a). We found that oral
administration of our TLR7 antagonist 23 could totally prevent the cutaneous pathology in terms of
macroscopic morphology (Fig 4b). There was also significant retardation in skin thickening in the
antagonist-treated mice (Fig 4c).
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
18
ACS Paragon Plus Environment

Page 19 of 50
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 5. Histopathologic scoring of the H&E-stained skin sections of mice with psoriatic disease.
(a) (upper) H&E stained, imiquimod exposed, dorsal skin section of vehicle treated mouse showing
marked thickening of epidermis and presence of hyperkeratosis, papillomatosis and acanthosis, (lower)
H&E stained, imiquimod exposed, dorsal skin section of compound 23 treated mouse showing
appreciably lesser thickening of epidermis. (b) Representative images showing the key pathognomonic
histopathologic features used for scoring. (c) All histological features were scored based on their
severity and preponderance, 1 being least severe and 3 being most severe. (d) Imiquimod exposed dorsal
skin sections from compound treated or vehicle treated mice were graded on the basis of different
clinically relevant histopathological parameters such as hyperkeratosis, acanthosis, papillomatosis,
suprapapillary thinning, Monros microabscess, spongiosis, pustule of Kogoj, vascular dilatation of
control group (n=9) and compound treated group (n=9), as a measure of disease severity. (e) Data on
scores for individual pathognomonic features compared between control and treatment groups.
(**p<0.01).
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 50
1
2
3
4
5
6
7
8
9
Histopathologic scoring of the lesional skin in the treated mice revealed nearly complete absence
of the pathognomonic features like hyperkeratosis, parakeratosis, acanthosis, spongiosis, papillomatosis,
suprapapillary thinning, vascular dilatation, microabscesses and pustule formation (Fig 5a, 5b). On
scoring the histopathology based on the listed features (Fig 5c), there was total amelioration of disease
on oral administration 23 (Fig 5d, 5e).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Inhibition of pro-inflammatory TLR7 mediated gene expression in rodent model of
psoriasis. (a-e) Dorsal skin sections obtained from C57BL/6 mice treated with vehicle (n=6) or treated
with 15 mg/kg compound 23 (n=6) for 7 consecutive days were processed and subjected to gene
expression studies to measure expression of indicated genes- TNFα (a), MX1 (b), MX2 (c), ISG15 (d)
and IFIT1 (e). (*p<0.05, **p<0.01, ns = non significant).
20
ACS Paragon Plus Environment

Page 21 of 50
Journal of Medicinal Chemistry
In a repeat experiment too, the amelioration of cutaneous pathology with compound 23 was
1
2
evident in terms of reduction in skin fold thickness (Figure S5) as well as in terms of reduction in the
tissue expression of the proinflammatory cytokine TNFα (widely implicated in the pathogenesis of
psoriasis) and four major type I IFN signature genes (ISGs) MX1, MX2, ISG15, IFIT1 (surrogate
markers for type I IFN induction in situ) (Fig 6a-e).
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 7. Short-term visceral toxicity assessment. (a-f) Representative images of H&E stained
sections of liver specimens (a,b), lung specimens (c,d) and kidney specimens (e,f) obtained from
C57BL/6 mice treated with vehicle (a,c,e) or 15 mg/kg compound 23 (b,d,f) for 7 consecutive days.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 50
Moreover, histomorphology of lung, liver, kidney did not show any remarkable pathologic
changes following oral administration of compound 23 for 7 consecutive days at a dose rate of 15 mg/kg
as compared to vehicle administration (Fig 7). No notable toxicity-induced pathology was discernible,
viz. changes in parenchymal architecture of the liver, hepatocyte survival, fatty infiltration or
inflammation in compound treated mice as compared to vehicle treated mice (Fig 7a, 7b). Similarly, no
discernable changes were observed in lung microanatomy as well. No evidence of immune infiltration
or alveolar architecture was observed in the lungs (Fig 7c, 7d). In kidneys no glomerular hypertrophy,
mesangial proliferation, capsular space obliteration or inflammation was observed in the compound
treated mice as compared to control mice (Fig 7e, 7f). We also did not observe any behavioral or
physiological abnormalities in experimental animals within the study period, thus indicating that oral
administration of compound 23 does not cause any apparent short-term visceral cytotoxity in C57BL/6
mice. Thus this novel TLR7 antagonist 23 was very efficient in preventing a TLR7-driven TNFα and
type I IFN induction and the autoimmune pathology in vivo.
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
CONCLUSION
In summary, we have been able to define minimal structural requirements in a purine scaffold
TLR7 agonist 1 and identify a singular ‘chemical switch’ in the scaffold that could transform the
molecule to a TLR7 antagonist. We established that a small structural modification in TLR7 ligand can
lead to reversal in their functional activity. We further developed on that and could design a potent
TLR7 antagonist 23, which had favourable in vitro and in vivo pharmacokinetics with excellent oral
bioavailability (70.8%) in mice. Interestingly, ethylpiperazine group at N-9 led to enhancement of TLR7
agonism, while introducing the same at C-6 led to potentiation of TLR7 antagonism. Introduction of an
additional basic center at C-6 has profound potentiating effect on TLR7 antagonism in purine scaffold.
Our SAR studies successfully identified site-specific substitutions that can switch a well-established
22
ACS Paragon Plus Environment

Page 23 of 50
Journal of Medicinal Chemistry
purine scaffold TLR7 agonist into TLR7 antagonist. None of the TLR7 antagonist showed any TLR7
agonism, which signifies that they are pure antagonists. It is worthwhile to mention that all the TLR7
antagonists reported herein showed excellent selectivity against TLR8. The docking analysis of our
TLR7 antagonists showed that the purine ring is sandwiched between two hydrophobic residues Phe329
and Tyr1374 forming π-π stacking interactions along with hydrogen bonding with Ser452, which plays
vital role in determining TLR7 antagonism. Finally, we could validate the in vivo efficacy of this TLR7
antagonist in a clinically relevant murine model of psoriasis. The histomorphology of lung, liver, kidney
did not show any remarkable pathologic changes following oral administration. The TLR7 antagonist 23
thus appears to be a very promising lead molecule for further development into clinically translatable
TLR7 antagonists that can be first-in-class therapeutic agents in the relevant clinical contexts, which
include a number of autoimmune diseases accounting for significant morbidity worldwide.
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
EXPERIMENTAL
Aqueous solubility assay. 6 µl of 50 mM DMSO stock was added to the deep well plate
containing 594 µl of pH 7.4 0.25 M phosphate buffer (including DMSO control), mixed and incubated
for 18 hours at room temperature with constant mixing at 300 rpm. The plate was sealed well during the
incubation process. After incubation, the samples were centrifuged for 20 min at 4000 rpm. 400 µL of
supernatant was analyzed by HPLC-UV. The DMSO content in the sample was 1.0% and the final
concentration of the compound in deep-well plate was 500 µM. the spectrum was read using Shimadzu
UV spectrophotometer at 240 nm. (Calculation is in Table S1).
Caco-2 permeability assay. 10 mM stock of compound in DMSO was diluted with HBSS
buffer pH 7.4 to a final concentration of 10 μM. For the proliferation of cells, 250 μl of DMEM was
added to the basal compartment of 96 well multi-screen Caco-2 plate and seeded 12000 cells/well (0.16
× 106 cells/ml) in all the apical wells placed in CO2 incubator at 37 ºC. On the day of assay, the medium
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 50
was removed and washed twice with HBSS Buffer. 250 μl of HBSS buffer with 2% BSA was added to
basal wells and 75 μl of the test compound was added to apical wells. 25 μl of basal samples were
collected at 120 min and processed as mentioned below. 250 μl of the test compound and 75 μl of HBSS
buffer with 2% BSA was added to basal wells and apical wells respectively. At 120 min 25 μl of apical
samples were collected and processed as stated below. Single point calibration curve in HBSS buffer
with 2% BSA was used. Donor samples and receiver samples were diluted 1:1 with HBSS separately
while former contain 2% BSA. It was precipitated with 200 μl of acetonitrile containing internal
standard and vortexed for 5 min @ 1000 rpm, centrifuged at 4000 rpm for 10 min. Finally, 100 μl of
supernatant was diluted with 200 μl of water and submitted for LC-MS/MS analysis.(Calculation is in
Table S2)
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
Plasma stability assay. 25 μM stock of test compound was prepared in acetonitrile from the
stock solution of 10 mM in DMSO. By diluting acetonitrile:water (50:50) from the previously prepared
10 mM stock, 125 µM sample of 200 µl was prepared. For 0 min samples, plasma was heat inactivated
at 56 ºC for 45 min and to it 3 μl of 25 μM test compound was added to make the final volume 75 μl. In
200 μl acetonitrile, a 25 μl aliquot of the mixture was taken and crashed. At 0, 15, 30, 60 and 120 min, a
25 μl aliquot of the sample was precipitated immediately with 200 μl of acetonitrile, centrifuged at 4000
X RCF, 4 ºC for 20 minutes and analyzed on LC-MS/MS. (Calculation for human plasma stability in
Table S3 and mice plasma stability Table S4).
Microsomal stability. 10 mM stock of test compound in DMSO was diluted with 1:1 water
acetonitrile to get 200 µl 1 mM stock solution. 100 µM working concentration of test compound was
prepared by diluting 1 mM stock solution of test compound in DMSO with acetonitrile: water (1:1). 2.5
mM NADPH cofactor solution was prepared from 16 mM stock solution and 3.33 mg/ml working stock
of liver microsomes was prepared from 20 mg/ml liver microsome master stock solution. 85 µl of
phosphate buffer of pH 7.4 was mixed with 2.5 µl of test compound (100 µM) and 75 µl 3.33 mg/ml
24
ACS Paragon Plus Environment

Page 25 of 50
Journal of Medicinal Chemistry
HLM in a deep well plate for pre-incubation of 10 min at 37 ºC. 32.5 µl mixture was added to 17.5 µl
buffer incubate for 60 min without NADPH cofactor. For zero minute sample, 16.5 µl preincubated
mixture was crashed with 200 µl acetonitrile and 8.75 µl cofactor was added to it. 62 µl NADPH-
cofactor was added to remaining incubation mixture and kept for 60 min at 37 ºC. For other time point
samples, 25 μl of incubation mixture at 5, 10, 30, 60 and 60 min without cofactor was precipitated
immediately with 200 μl of acetonitrile containing internal standard and vortex for 5 min at 1200 RPM.
The samples were centrifuged at 4000 x RCF, 4 °C for 10 min. Supernatant (150 μl) was diluted with
150 μl of water and analyzed on LC-MS/MS (Calculation for human microsomal stability in Table S5
and mice microsomal stability Table S6).
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
TLR7 agonism and antagonism assay using HEK-Blue™ hTLR7 reporter cell line. The
TLR7 agonistic or antagonist activity of synthesized compounds was assayed using a HEK293 cell line
engineered to express human TLR7 as well as a NF-кB induced secreted alkaline phosphatase (SEAP).
The assay was based on the principle that TLR activation with TLR agonist leads to downstream NF-кB
activation which in turn induces SEAP production and secretion into culture supernatant, causing colour
change upon interaction with the detection media. 7×104 cells/well were seeded in 100 mg/ml Normocin
supplemented complete DMEM medium and allowed to adhere to the substrate for 5 hours at 37 ºC and
5% CO2. For antagonism assays, CL264 (TLR7 agonist, 5 μg/ml) was added to the cells pre-incubated
for 1 hour with indicated doses of compounds, whereas for agonism assays, cells were only treated with
indicated doses of compounds. Following overnight incubation, 50 μl of culture supernatant from each
well, was added to 200 μl of Quanti-Blue detection media and constantly monitored for colour change.
After 1-2 hours of incubation, spectrophotometric reading at 655 nm wavelength was recorded. To
ensure the validity of TLR7 antagonism assay, the ability of different doses of hydroxychloroquine to
inhibit TLR7 activation was studied as a positive control. Two-tailed Paired Student’s t-test was
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 50
performed using GraphPad Prism software version 5.0. HEK-Blue™ hTLR7 reporter cell line,
Normocin and CL264 were purchased from Invivogen, USA.
1
2
3
4
5
6
7
8
9
TLR8 antagonism assay using HEK-Blue™ hTLR8 reporter cell line. To ensure the
specificity of the synthesized antagonists for human TLR7 and exclude the possibility of nonspecific
antagonism toward human TLR8 which is structurally very similar to human TLR7, we performed
reporter assays using HEK-Blue™ hTLR8 reporter cell line. The protocol used to perform this
experiment was exactly same as that for the antagonism assay in HEK-TLR7 cell line, described above.
The only difference was the usage of a TLR8 specific agonist CL075 (3 μg/ml) as the stimulant.
Cytotoxicity assay in HEK293 cell line. 1 µM, 5 µM, 10 µM, 20 µM, 40 µM, 60 µM, 80 µM
and 100 µM of indicated compounds were added to 6×104 HEK293 cells cultured incomplete DMEM at
37 ºC and 5% CO2 overnight, following which cells were harvested in PBS (0.01 M, pH = 7.4), stained
with 2.5 mg/ml PI for 1 min and percentage of cells showing PI uptakein a flow cytometer (BD LSR
Fortessa) was measured.
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
IL-6 mRNA expression measurement through RT-PCR Analysis. HEK-BlueTM TLR7 cells
were seeded in a 96 well plate at a density of 1×105 cells/well and allowed to attach overnight. Next
morning, cells were pre-incubated with 5 µM of compound 23 for one hour before being stimulated with
5 µg/ml of TLR7 agonist-CL264 for 6 hours, following which cells were harvested in TRIzol reagent
(Invitrogen). Total RNA was isolated using the manufacturer’s protocol followed by cDNA preparation
using the Applied Biosystems kit and real time PCR was done using BioRad SYBR Green Master Mix
to measure the expression levels of IL6 gene relative to that of 18S (housekeeping gene). Primer
sequences
used,
Hu-18S:
Forward:
5’-GTTGGTTTTCGGAACTGAGG-3’Reverse:5’-
TCGTTTATGGTCGGAACTACG-3’ and Hu-IL6: Forward: 5’-CACAGACAGCCACTCACCTCTTC-
3’ Reverse: 5’-TTTGCTGCTTTCACACATGTTACTC-3’.
26
ACS Paragon Plus Environment

Page 27 of 50
Journal of Medicinal Chemistry
TLR7 antagonism assay in primary human PBMCs. Primary human peripheral blood
1
2
mononuclear cells (PBMCs) were isolated from blood taken from healthy individuals after obtaining
their informed consent as well as ethical approval from the Institutional Ethics Committee, in
accordance with the guidelines mentioned in the Declaration of Helsinki. PBMCs were obtained by
performing density gradient centrifugation and seeded at a density of 2 × 105 cells/well in a 96 well
plate. PBMCs were pre-incubated with 5 µM of compound 23 for one hour before being stimulated with
5 µg/ml of TLR7 agonist-Gardiquimod overnight. Next day, culture supernatant was collected and
subjected to TNFα ELISA according to the protocol outlined by the Manufacturer (Mabtech).
Homology modeling and docking study. Homology structure of human TLR7 ectodomain was
build using sequence Q9NYK1 retrieve from UNIPROT. Homology model structure of hTLR7 was
built using TLR8 (PDB 5WYZ) as a template in discovery studio modeler suite, according to the
reported procedure.41 Docking experiment was performed using Discover Studio 4.1 C-DOCKER
module. Docking method was optimized by analyzing reported TLR7 antagonist binding pose. Random
conformers of compounds were searched using dynamics of 1000 steps. Docking simulation was
performed using default parameter of C-DOCKER module in Discover Studio 4.1. The docked results
were fully minimized using CHARM force field with conjugate gradient 1.0. Pictorial view of docking
pose was made using PyMOL.
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
In-vivo pharmacokinetics and bioavailability study. Nine 8-10 weeks old male C57BL/6 mice
were recruited for the experiment. They were divided into 3 groups each having three mice. The
compound was reconstituted in 2% ethanol in PBS (0.01 M, pH = 7.4). The mice belonging to the first
group were given oral gavage with 15 mg/kg compound, the mice belonging to the second group were
given oral gavage with 10 mg/kg compound and the mice belonging to the third group were given
intravenous injections of 15 mg/kg compound. Blood samples were drawn at 0 h, 0.5 h, 1 h, 2 h, 4 h, 6
h, 12 h and 24 h after compound administration and serum was isolated for the mice administered orally
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 50
with the compound. On the other hand, blood samples were drawn at 0 h, 0.25 h, 0.5 h, 1 h, 2 h, 4 h, 6 h,
12 h and 24 h after compound administration and serum was isolated from the mice injected
intravenously with the compound.10 μl of serum from each mouse was extracted with 30 μl of LC/MS-
grade acetonitrile + 0.1% formic acid (J.T. Baker) by alternate vortexing and chilling. Next, the
supernatant was collected by centrifugation at 14,000 rpm for 10 min and analyzed by LC-MS/MS to
measure the concentration of compound 23 in serum. Standards of compound 23 ranging from 74.688
pg to 5975 pg was prepared in serum and extracted in acetonitrile + 0.1% formic acid. Mass
spectrometry was carried out in LTQ ORBITRAP XL using Hypersil Gold C18 column having a
diameter of 100 X 2.1 mm with particle size being 1.9 μ. The column formed the stationary phase and a
mixture of solution A (acetonitrile + 0.1% formic acid) and solution B (HPLC grade H20 + 0.1% formic
acid) formed the mobile phase. Liquid chromatography (injection volume 2 μl) was done using an
isocratic solution containing 40% acetonitrile and 0.1% formic acid. The intact compound had a
retention time of 0.76 mins and (m/z) of 387.2988, while (m/z) of fragmented compound after MS/MS
was 273.1852. All data was analyzed using the Thermo Xcalibur software.
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
In-vivo efficacy study. 18 wild type C57BL/6 male mice aged 8-10 weeks, provided by CSIR-
IICB animal house facility were divided into two groups; 1. Control group-containing 9 mice with
induced psoriasis, treated with 2% ethanol in sterile PBS (0.01 M, pH = 7.4) (vehicle) and 2. Treatment
group- containing 9 mice with induced psoriasis treated with compound 23. Psoriasis was induced by
daily topical application of 62.5 mg 5% w/w imiquimod cream (Imiquad; Glenmark, India) on a patch
of shaved dorsal skin of the mice for 6 consecutive days. Vehicle and compound (15 mg/kg) were
administered orally daily for 7 days, beginning one day prior to the start of imiquimod application. Skin
fold measurement of the affected area was performed by a person blind to the experimental details and
outcomes, with a standard Vernier caliper on a daily basis starting prior to imiquimod application and
continuing till the end of the experiment. The experimental animals were sacrificed on the 7th day of
28
ACS Paragon Plus Environment

Page 29 of 50
Journal of Medicinal Chemistry
Imiquimod application and portions of affected skin as well as normal skin were collected from each
animal (in 4% paraformaldehyde) for histopathological evaluation by H&E staining. All animal
handling and mouse experiments were conducted in accordance with national and international
guidelines and upon approval from the institutional ethics committee. All recommendations enlisted in
the ACS Ethical Guidelines were followed while performing the mouse experiments.
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
RNA isolation, cDNA preparation and Quantitative PCR. Lesional skin tissue derived from
6 control C57BL/6 mice and 6 compound 23 (15 mg/kg) treated mice, following 7 consecutive days of
oral administration (according to a protocol identical to that described in in-vivo efficacy study), was
minced into tiny pieces, frozen in liquid nitrogen, crushed using a mortar & pestle in TRIzol Reagent
(Invitrogen) and RNA was isolated according to the manufacturer's protocol. cDNA was produced using
the High- Capacity cDNA Reverse Transcription Kit (Applied Bisosystems). qPCR was performed to
study the expression of the specified genes using SyBR Green Master Mix (BioRad), (normalization
was done with respect to the expression of β-actin- housekeeping gene).
Primer sequences:
Mu-β actin
Mu-TNFα
Mu-MX1
Mu-MX2
F: 5’-GAGGTATCCTGACCCTGAAGTA-3’
R: 5’-GCTCGAAGTCTAGAGCAACATAG-3’
F: 5’-AATGGCCTCCCTCTCATCAGTT-3’
R: 5’-CCACTTGGTGGTTTGCTACGA-3’
F: 5’-TTCAAGGATCACTCATACTTCAGC-3’
R: 5’-GGGAGGTGAGCTCCTCAGT-3’
F: 5’-TTCCAGCATCTGAATGCCTAC-3’
29
ACS Paragon Plus Environment