Discovery of Novel Small Molecule Dual Inhibitors Targeting Toll-Like Receptors 7 and 8

Article abstract of DOI:10.1021/acs.jmedchem.9b01201

Endosomal toll-like receptors (TLRs) 7 and 8 recognize viral single-stranded RNAs, a class of imidazoquinoline compounds, 8-oxo-adenosines, 8-aminobenzodiazepines, pyrimidines, and guanosine analogues. Substantial evidence is present linking chronic inflammation mediated specifically by TLR7 to the progression of autoimmunity. We identified a new TLR7/8 dual inhibitor (1) and a TLR8-specific inhibitor (2) based on our previous screen targeting TLR8. Compound 1, bearing a benzanilide scaffold, was found to inhibit TLR7 and TLR8 at low micromolar concentrations. We envisioned making modifications on the benzanilide scaffold of 1 resulting in a class of highly specific TLR7 inhibitors. Our efforts led to the discovery of a new TLR8 inhibitor (CU-115) and identification of a TLR7/8 dual inhibitor (CU-72), bearing a distinct diphenyl ether skeleton, with potential for TLR7 selectivity optimization. Given the role of TLR8 in autoimmunity, we also optimized the potency of 2 and developed a new TLR8 inhibitor bearing a 1,3,4-oxadiazole motif.

Full text of DOI:10.1021/acs.jmedchem.9b01201

Subscriber access provided by University of Massachusetts Amherst Libraries
Article
Discovery of Novel Small Molecule Dual
Inhibitors Targeting Toll-like Receptor 7 and 8
Rosaura Padilla-Salinas, Rachel Anderson, Kentaro Sakaniwa, Shuting Zhang,
Patrick Nordeen, Chuanjun Lu, Toshiyuki Shimizu, and Hang Hubert Yin
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01201 • Publication Date (Web): 05 Nov 2019
Downloaded from pubs.acs.org on November 5, 2019
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 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Novel Small Molecule Dual Inhibitors Targeting
Toll-like Receptor 7 and 8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Rosaura Padilla-Salinas†,‡, Rachel Anderson†,‡, Kentaro Sakaniwa§, Shuting Zhang‡, Patrick Nordeen†, Chuanjun
Lu†, Toshiyuki Shimizu§,* and Hang Yin*,†,‡
† Department of Chemistry and Biochemistry and BioFrontiers Institute, University of Colorado Boulder, Boulder, CO 80309, USA
‡ School of Pharmaceutical Sciences, Tsinghua University-Peking University Joint Center of Life Science, Tsinghua University, Beijing
100082, China
§ Graduate School of Pharmaceuticals Sciences, The University of Tokyo, Tokyo, 113-0033, Japan
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
ABSTRACT: Endosomal Toll-like receptors (TLRs) 7 and 8 recognize viral ssRNAs, a class of imidazoquinoline compounds, 8-oxo-adeno-
sines, 8-aminobenzodiazepines, pyrimidines, and guanosine analogs. Substantial evidence has accrued linking chronic inflammation mediated
specifically by TLR7 to the progression of autoimmunity. We identified a new TLR7/8 dual- (1) and a TLR8 specific-inhibitor (2) based on
our previous screen targeting TLR8. Compound 1, bearing a benzanilide scaffold, was found to inhibit TLR7 and TLR8 at low micromolar
concentrations. We envisioned making modifications on the benzanilide scaffold of 1 resulting in a class of highly specific TLR7 inhibitors. Our
efforts led to the discovery of a new TLR8 inhibitor (CU-115) and identification of a TLR7/8 dual inhibitor (CU-72), bearing a distinct di-
phenyl ether skeleton, with potential for TLR7 selectivity optimization. Given the role of TLR8 in autoimmunity, we also optimized the potency
of 2 and developed a new TLR8 inhibitor bearing a 1,3,4-oxadiazole motif.
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
2
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
INTRODUCTION
1
2
3
4
5
6
Nucleic acids (NA) sensors are part of the innate immune system and recognize multiple forms of intracellular self- and non-self RNA and
DNA macromolecules. Activation of NA-sensors initiates signal transduction pathways for the upregulation of type I interferon genes (IFNs).
7
8
9
,
1 2
The endosomal Toll-like receptors (TLRs) represent a major class of NA sensors that regulate inflammation. In humans, four endosomal
TLRs (TLR 3, 7, 8 and 9) have been identified. The endosomal TLRs detect viral and endogenous double-stranded RNA (dsRNA; TLR3),
single-stranded RNA (ssRNA; TLR7/8), or unmethylated CpG sequences in DNA (TLR9).1 Of the four-known endosomal TLRs, TLR7 and
TLR8 are most similar in sequence and function.4b Both TLR7 and TLR8 mount an immune response by detecting and responding to degraded
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 4
ssRNA products from bacteria and viruses. The endosomal TLR NA-sensing pathways modulate nuclear factor kappa-light-chain-enhancer
of activated B cells (NF-kB) and IFNs, which leads to the transcription of genes encoding pro-inflammatory cytokines.1
Systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and inflammatory bowel disease (IBD) are common autoimmune diseases.
Substantial evidence has accrued linking chronic inflammation mediated by signaling malfunction of TLR7 to the onset and progression of
5
autoimmune diseases. For example, studies have identified a genetic linkage and observed polymorphisms in TLRs (TLR7, TLR3, and TLR9)
6
associated with SLE. Activation of TLR7 caused by recognition of self-RNAs (e.g. microRNA let-7) and self-ssRNA (released from disrupted
, ,
7 8 9
cells) has been observed.
TLR7 signaling malfunction is also associated as a contributor to/towards increased tumor necrosis factor alpha
10
(TNF-a) production in autoimmune and autoinflammatory disorders such as SLE. A direct genetic link for TLR8 and autoimmunity remains
to be fully established, however, studies have already implicated TLR8 as a contributor in increased TNF-a production of RA.10,
Over-
, ,
11 12 13
activation of endosomal TLRs, specifically TLR7, leads to increased levels of proinflammatory cytokines linked to inflammatory disorders,
which provides a compelling rationale for targeting TLR7. Despite the considerable evidence implicating TLR7 signaling dysregulation in au-
toimmune diseases, only a few selective small molecule TLR7 inhibitors have been published, which has hindered our understanding of TLR7
14
signaling and its role in autoimmunity. Few TLR7 inhibitors have entered preclinical development; however, there are currently no clinically
available selective TLR7 inhibitors.10
Achieving TLR7 inhibition with a high degree of specificity without affecting TLR8 remains a major obstacle for small molecule drug devel-
opment. This is in part due to the fact that TLR7 and TLR8 are structurally similar and both recognize ssRNA, guanosine analogs, and non-
selective imidazoquinoline agonists and inhibitors. Structural studies have aided in the identification of two distinct ligand binding sites for
degraded ssRNA products for both TLR7 and TLR8.3, 4 The ligand binding sites for uridine, guanosine, and chemical ligands is conserved
between TLR7 and TLR8. Thus, many efforts for the rational design of TLR7 inhibitors have relied on known TLR8 agonists with high binding
specificity for TLR8 exclusively. Limitations of side effects in published TLR7 inhibitors include poor TLR selectivity and cellular potency
(Figure 1). For example, the small molecule inhibitor hydroxychloroquine targets both TLR7 and TLR9 indirectly by accumulating in the
endosome and binding NAs, preventing TLR recognition.1 The recently reported 3H-imidazoquinolines inhibitors have IC50 values in the 25
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
and 100 µM range.15, 16 Single-stranded phosphorothioate oligonucleotides have also been reported as TLR7 inhibitors.17 Other efforts have
focused on targeting downstream signaling proteins of endosomal TLRs, such as Interleukin-1 receptor-associated kinase 4 (IRAK4).1 A draw-
back of this approach is that directly targeting the downstream kinase IRAK4 could block the other nine TLR signaling pathways, causing off-
target effects.
1
2
3
4
5
6
7
8
9
R2
N
Me
OH
NH2
H2N
N
N
N
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
O
N
N
HN
N
Et
N
3
N
N
R1
N
NH2
N
N
N
Bn
Bn
Cl
Me
Me
R1 = -C9H19
R1 = Ph
OH
Hydroxychloroquine
TLR7/9 Antagonist
CU-CPT8m
CU-CPT9a R2 = OMe
CU-CPT9b R2 = OH
Selective TLR8 Antagonists
Dual TLR7/TLR8 Antagonists
Figure 1. Examples of reported TLR7, TLR7/8, and TLR8 antagonists.
Recently our group overcame TLR7/8 specificity challenges, ultimately developing a new generation of TLR8-specific inhibitors with high
18
selectivity and potency, CU-CPT8m (Figure 1). From the screen leading to the identification of CU-CPT8m, we serendipitously discovered
a small molecule displaying TLR7/8 inhibitory activity and were inspired to use medicinal chemistry and structure-activity relationship studies
(SAR) to override the TLR7/8 selectivity to favor TLR7 activity, ultimately enabling the discovery of new selective TLR7 inhibitors through
optimization of the TLR7 activity.4b Herein, we describe our efforts to develop TLR7, TLR8, and TLR7/8 inhibitors based on two distinct
scaffold families and their biophysical characterization and biological evaluation in cellular assays. Our efforts culminated in the discovery of
two new TLR8 inhibitors displaying high TLR 8 selectivity at low concentrations and the development of new TLR7/8 inhibitors with potential
optimization for TLR7 activity.
RESULTS AND DISCUSSION
High Throughput Screening (HTS) results and validations studies. Hits 1 and 2 were manually identified from an HTS targeting TLR8,
see Error! Reference source not found.a for chemical structures. The Maybridge Hit Finder library compounds (14, 400 small molecules)
follow the Lipinski guidelines for "drug-likeness" and have properties that include: no more than 5 hydrogen bond donors, no more than 10
hydrogen acceptors, a molecular weight less than 500, and an octanol-water partition coefficient (log P) less than 5. The HTS protocol and
results using the Maybridge Hit Finder library have been published for the development of the first selective TLR8 inhibitor.18
For this project, hit 1 and 2 were purchased from Maybridge and first re-tested for potency validation, cell toxicity, and TLR specificity. Both
compounds inhibited TLR7 and TLR8 signaling in a dose-dependent manner to down regulate TLR7- and TLR8-induced secreted embryonic
alkaline phosphatase (SEAP) levels with IC50 values of 2.88 ±0.25 µM (TLR7) and 1.64 ±0.40 µM (TLR8) for 1 and IC50= 1.40 ±0.06 µM
(TLR8) for 2, (see Figure S1 and Figure S2), and were nontoxic, (see Figure S4). In humans, there are a total of 10 TLRs and some share
4
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
adapter proteins and downstream signaling pathways. As such, a high degree of TLR specificity is necessary and important for therapeutic
applications. The specificity tests for 1 and 2 using stably transfected human embryonic kidney (HEK 293-Blue) cells individually overexpress-
ing each human TLR indicated no obvious inhibition of other TLRs, (Error! Reference source not found.b), except for TLR9 pathway at 1.5
µM for 1. Interestingly, compound 1 was the only inhibitor that down-regulated TLR7-induced SEAP levels at 1.5 µM. Although 1 lacked
1
2
3
4
5
6
7
8
9
specificity at 3 and 5 µM, we continued to pursue SAR since 1 displayed more activity at 1.5 µM in TLR7 cell line compared to TLR8 and
TLR9. Compounds 1 and 2 both contained a diaryl ether linkage and the aromatic tricyclic ring core scaffold, indicating those motifs are im-
portant for target selectivity. Differences such as the amide functional group and the 1,3,5-oxadiazole motif suggested those motifs could be
modified to modulate TLR7/8 selectivity for SAR. Compounds 1 and 2 represent two different scaffold families that are still capable of main-
taining high inhibitory activity and selectivity toward TLR8, which presents the opportunity to discover a new binding mode or identify inhib-
itors with altered and/or more suitable pharmacokinetic or pharmacodynamic properties for drug development. Because of the clinical rele-
vance of TLR7 and TLR8 in autoimmunity, we conducted SAR studies for both 1 and 2 and primarily focused on 1 owing to the significant
potential for discovering a much needed TLR7-selective inhibitor.
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. (a) Chemical structures of 1 and 2. (b) HEK 293 cells expressing human toll-like receptor (hTLR) gene and an inducible secreted
embryonic alkaline phosphatase (SEAP) reporter gene were incubated with 1 or 2 for 16 h. At 1.5, 3 and 5 µM 1 does not inhibit not modulate
the NF-kB inhibition induced by LPS, R848, Poly(I:C), in HEK-293 TLR4, TLR3, TLR7, and TLR9 cells. Compound 1 modulates the TLR9
pathway at ~ 3–5 µM and only exhibits partially selectivity at low concentrations. At 1.5, 3 and 5 µM 2 does not modulate the NF-kB inhibition
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
induced by LPS, Pam2CSK4, Pam3CSK4, Flic, R848, and Poly(I:C) in HEK-293 TLR1/2, TLR2/6, TLR3, TLR4, and TLR7 cells and 2 dis-
plays selectivity for TLR8. The data were normalized as [(raw data – untreated cells)/(ligand + solvent control – untreated cells)]. Ligand +
solvent is 100% activation, and untreated cells are 0% activation. The result of one representative biological replicate for three independent days
is plotted with the error bars representing the standard deviation of three technical replicates for one independent biological replicate.
1
2
3
4
5
6
7
8
9
Preliminary SAR studies for 1. Initially, five analogs were designed to identify key structural motifs contributing to TLR7/8 dual inhibitory
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
activity, see (Figure 3). Benzanilide and 1,3,4-oxadiazole motifs are common structural elements found in biologically active molecules and
19
both 1,3,4- and 1,2,4-oxadiazoles are known amide bioisosteres. The presence of the -NH, an H-bond donor, between the B- and A-ring is a
notable difference between 1 and 2 thus revealing some SAR information for modulating TLR7/8 activity.
F3C
O
F
F
F
F
O
H
N
H
F3C
N
N
N
H
CF3
HN
O
Cl
O
O
Me
Cl
Cl
Cl
Cl
F
CF3
3
4
5
6
7
Removed aryl ether
0% TLR7 inhibition
0% TLR8 inhibition
2º Alkyl amine
IC50 > 50 µM TLR7
IC50 > 50 µMTLR8
3º Methylated amide
0% TLR7 inhibition
Reverse amide
0% TLR7 inhibition
TLR8 IC50 = 1.4 (±0.1) µM TLR8 IC50= 7.10 (±1.78) µM
1,3-disubstituted B ring
0% TLR7 inhibition
TLR8 IC50
=
2.9 (±0.8) µM
Figure 3. Preliminary SAR results for 1.
We synthesized the tertiary benzanilide (3), the corresponding secondary alkyl amine (4) and reverse amide (5), (Figure 3). The tertiary
amide 3 was synthesized by alkylation of 1 with methyl iodide, and reductive amination of 2-chloro-6-fluorobenzaldehyde with 4-[3,5-bis(tri-
fluoromethyl)phenoxy aniline provided 4 in 88% yield. The synthesis of 5 began with SNAr of methyl 4-fluoro-3-nitrobenzoate, 8, to afford
intermediate 9. Reduction of the nitro group gave compound 10, which was then converted to the corresponding benzoyl chloride, 4-[3,5-
bis(trifluoromethyl)phenoxy] benzoyl chloride. Subsequent amide formation with 2-chloro-6-fluoroaniline yielded 5 in 57%, (see Scheme 1).
Compounds 6 and 7 were synthesized using the same route as 1 except 3-[3,5-bis(trifluoromethyl)phenoxy]aniline and commercially available
3,5-bis(trifluoromethyl)aniline were used as the aniline substrates.
Scheme 1. Synthesis of 5.a
F
F3C
O
R
a
O
OMe
O2N
OMe
CF3
O
8
9, R = NO2
b
10, R = H
F
F3C
O
F
H2N
H
N
c, d
Cl
CF3
O
5
Cl
6
ACS Paragon Plus Environment

Page 7 of 47
1
Journal of Medicinal Chemistry
a Reaction conditions: (a) 3,5-bis(trifluoromethyl)phenol, Cs2CO3, DMF 110 °C (79%); (b) 10 mol% Pd/C, H2, MeOH, then t-BuONO, DMF, 50
°C (75% 2-steps); (c) i. NaOH, MeOH, 50 °C; ii. Oxalyl chloride, cat. DMF, CH2Cl2; (d) 2-chloro-6-fluoroaniline, Et3N, CH2Cl2 (57% 2-steps).
2
3
4
5
6
7
8
9
Methylation of amide (3) completely abolished TLR7 activity and retained TLR8 activity of 2.9 ( 0.8) μM, which then led us to examine
the importance of the carbonyl moiety. Poor inhibition was observed for both TLR7 and TLR8 with the secondary alkyl amine, 4, indicating
the presence of a carbonyl H-bond acceptor is important and a rigid planar benzanilide is superior. The corresponding reverse benzanilide 5
was synthesized to explore different orientations of the NH amide and was determined to be inactive towards TLR7 and active towards TLR8.
A notable difference compared to 1 is that compound 5 has the amide NH adjacent to the fluorine atom. Investigators have speculated that a
fluorine proximal to an amide N–H can mask the H-bond through an electrostatic interaction; thus, we speculate TLR7 inactivity may be due
to H-bond masking caused by the adjacent fluorine atom.19
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
Cumulatively, compounds 3-5 show the presence of the -NH or H-bond donating atom is necessary for TLR7 activity. By comparing 4 and
5 we also observed that H-bond acceptors are more important for TLR8 activity. We then synthesized a 1,3-disubstituted derivative, 6, to
explore additional substitution patterns of the B-ring. Compound 6 did not inhibit TLR7, which indicates the 1,4-disubstitution pattern is
required for TLR7 activity, and 6 was slightly less active (IC50 = 7.10 ± 1.78 µM) towards TLR8 compared to 1, suggesting the TLR8 binding
pocket may tolerate additional substitutions. Lastly, the distance between the C- and A-ring was shown to be important since removing the B-
ring resulted in 0% inhibition (7), and this motivated us to focus on the optimization of the A- and C-ring for 1 scaffold.
SAR studies and chemical synthesis for 1 analog. Based on our observations for compounds 3–7 and ease of synthesis, we selected the 4-
phenoxybenzanilide (11) core as the candidate scaffold for modulation of the A-ring, (Figure 4). With the goal of rapidly introducing diversity
into the A-ring, we generated a library of 1 analogs using standard amide bond forming reactions of 13 with benzoyl or benzoic acid derivatives
to synthesize compounds 14a–14n and 15a–15g, (Scheme 2). While many benzoyl or benzoic acid derivatives are commercially available, 6-
fluorobenzoyl chloride or carboxylic acid derivates with 4- and 6-substituents typically contain halogens or have limited diversity in the alkyl,
and electron withdrawing (EWG) groups.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
O
O
1
2
C
B
N
H
A
3
4
5
11
Key
benzanilide
substructure
Hydrophobic
diphenyl ether
skeleton
6
7
8
9
R2
O
O
O
O
Y
Ar
R1
R2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ar1
Y
Ar2
Figure 4. Design of novel TLR7/8 and TLR8 inhibitors bearing a benzanilide substructure.
Scheme 1. Synthesis of 14a–14m and 15a–15g.
F3C
O
F3C
O
O
F
c
R3
N
NH2
H
CF3
F
CF3
14
13
b
R1
R2
F3C
O
14a (R1 = H)
14b (R1 = F)
14c (R1 = Cl)
14e (R1 = Br)
14f (R1 = I)
14i (R1 = CF3)
14j (R1 = OH)
14l (R1 = NH2)
14n (R1 = OCF3)
14d (R2 = Cl)
14g (R2 = I)
NO2
F
CF3
12
a
R1
14h (R3 = I)
F3C
OH
14k (R3 = OH)
14m (R3 = NH2)
F
CF3
+
F
X
F
F
F
NO2
15a
15b
X = F, Br
F
Y
H
N
R4
N
X
15c
15d
15f, X = C
15g, X = N
(Y = O, R4 = H)
15e
(Y = S, R4 = Cl)
a Reagents and conditions: (a) 3,5-bis(trifluoromethyl)phenol and 1-fluoro-4-nitrobenzene or 1-bromo-4-nitrobenzene, Cs2CO3 or K2CO3, toluene,
100 °C (61–91%); (b) Method A: SnCl2, EtOH, 70 °C (19–88%) or Method D: 10 mol% Pd/C, H2, MeOH, rt, overnight; (c) Method A: HATU,
carboxylic acid, DMF, i-Pr2NH or Et3N, aniline, 0 °C–rt , overnight or Method B: aniline, acid chloride, Et3N, CH2Cl2, 0 °C–rt.
8
ACS Paragon Plus Environment

Page 9 of 47
1
2
3
4
5
6
7
Journal of Medicinal Chemistry
We prepared carboxylic acids that were not commercially available, (Scheme 3) and designed functionalized benzoic acid derivatives ame-
nable to late-stage functionalization for access to a variety of analogs to expand our SAR exploration. The compounds were evaluated in the
SEAP assay for both TLR7 and TLR8 inhibition to determine general TLR7/8 selectivity. Initial SAR efforts for the A-ring began by exploring
the effects of other halogens at the 6-position, (Table 1). Replacing -Cl with -H, 14a, decreased the potency for TLR7 inhibition (>50 µM) and
had minimal impact on TLR8 selectivity. Encouragingly, we modified the TLR7/8 selectivity to favor TLR8 selectivity by replacing the -Cl with
an -I atom. The TLR7 activity for 14f (CU-115) was significantly reduced compared to its TLR8 activity, supporting our hypothesis that TLR
selectivity could be optimized for the dual inhibitor. In general, we observed that increasing the size of the halogen atom (6-F, 6-Br, and 6-I)
increased TLR8 activity, while TLR7 inhibition decreased. To explore the importance of the 2,6-disubstitution pattern on the A-ring, we pre-
pared the 2,5- (14d and 14g) and the 2,4-disubstituted (14h) iodo- derivatives. All the compounds displayed reduced TLR7 and TLR8 activity
relative to 1. Cumulatively, these results suggest the TLR7 binding pocket may be smaller and is less tolerant of changes in substitution patterns
of the A-ring.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis of carboxylic acid substrates 16–19.a
F
F
O
a
b
OH
OCF3
O
OCF3
16
F
F
O
OH
OH
H2N
HN
Tf
17
F
F
O
c
OH
NH2
O
NO2
18
F
O
F
d
OH
OH
BnO
HO
19
a Reaction conditions: (a) LDA, THF, -78 °C then CO2 16, 17 56%) (B) Tf2O, CH2Cl2 (86%) (C) Oxone, NaOH, NaHCO3, EDTA, H2O, Acetone
(95%) (d) i. BnBr, K2CO3, DMF, 70 °C then LiOH, THF, H2O (52% 2-steps).
Table 1. SAR Modifications to A-ring 1 scaffold.
F3C
O
F3C
O
O
F
O
R1
N
H
N
H
R1
A
CF3
CF3
TLR7
IC50 (µM)a
TLR8
IC50 (µM)a
~3
TLR7
IC50 (µM)a
TLR8
IC50 (µM)a
ND
Compound
14ad
R1
6-H
Compound
14l
R1
>50
6-NH2
ND
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
14b
6-F
6-Cl
5-Cl
NI
NI
14m
14n
14o
4-NH2
>50
>50
NI
>50
>50
NI
1
2
3
4
5
6
7
14c (1)
14d
2.88
(±0.25)
ND
1.64
(±0.04)
6-OCF3
O
F
1.11
(±0.12)
8
9
14ed
6-Br
6-I
~1
~6
15a
15b
NI
NI
NI
NI
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
F
F
14f
> 50
1.04
(±0.16)
F
F
(CU-115)
F
O
F
14g
5-I
NI
ND
15c
>50
NI
N
O
14h
14i
4-I
>50
>50
ND
15d
15e
NI
NI
NI
O
NDd
O
6-CF3
3.2 (±0.21)
S
Cl
O
14j
6-OH
4-OH
ND
NI
ND
15f
>50
>50
ND
ND
H
N
O
14k
>50 uM
15g
H
N
N
a
IC50 derived from concentration-response curve in HEK 293-Blue TLR7 and TLR8 cell lines. Data was normalized to a 1 µg/mL R848 control
(data are mean ± s.d.; n = 3 independent experiments). b NI: 0% inhibition. c ND: Not determined. IC50 was not determined for compounds displaying
20-30% inhibition at 50 µM. c Unstable compound. d Compound was insoluble at higher concentration, preventing accurate determination of IC50. An
approximate value is provided.
We hypothesized that differences in activity for compounds 1, 14b, and 14f (CU-115)could arise from electrostatic effects. The iodine can
20
engage in halogen bonding, an electrostatically driven interaction. To further explore this, we examined the -OCF3 (14n) and the ethynyl
21
group (14o). The ethynyl group is a nonclassical bioisostere that has a polarized –CH moiety, and it is a weak hydrogen bond donor. The
trifluoromethyl (-OCF3) group has been used as a bioisostere for larger halogens but cannot participate in halogen bonding like -I. For our
inhibitors, replacement of -I with the -OCF3 and the ethynyl moiety (14o) did not improve the potency, which indicates that halogen bonding
may not be the dominating factor. The s-hole effect also enables the -I atom to act as an H-bond acceptor; thus, we installed a -OH and -NH2
group to explore this possibility at both the 4- and 6-positions. Installation of H-bond donors at the 6-positions leads to loss of TLR7/8 activity
for 14k, 14l, 14j, and 14m further supporting the notion that halogen bonding is not an important factor. Lastly, complete fluorination of the
A-ring (15b) and installation of a pyridine N- at the 6-position (15c) abolished TLR7 and TLR8 activity. All compounds bearing a heterocyclic
A-ring 15-15g, except for 15e, were inactive or displayed poor activity for TLR7 and TLR8 inhibition or were toxic to HEK 293-Blue TLR cells
10
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
lines. Interestingly, thiophene 15e inhibited TLR8 signaling with an IC50 value of 3.2 (±0.21 µM), while 0% inhibition was observed for TLR7
suggesting 5-membered heterocycles may be promising. Although there is precedence for the synthesis of halothiophenes, many approaches
require the use of protecting groups or lengthy synthetic routes; thus, the thiophene analogs were not pursued.
1
2
3
4
5
6
7
8
9
Heterocyclic fused-ring derivatives were investigated to reduce the flexibility by fusing the A-ring with the B-ring and ultimately decrease the
entropic cost for binding. We pursued benzimidazole instead of the benzoxazole derivative because it contains an NH and the SAR results
indicated the amide NH was critical for retaining TLR7 activity. SNAr of 5-fluoro-2-nitroaniline with 3,5-bis(trifluoromethyl)phenol followed
by reduction of the nitro group afforded 23 in 71%. Benzoylation with 2-chloro-6-fluorobenzoyl chloride and then acid catalyzed cyclization
yielded 28 in 25% yield (2-steps), see (Scheme 4b). Compound 28 did not display inhibitory activity thus revealing a flexible tricyclic core is
necessary. Based on these results, heterocyclic cores were not pursued further, and replacement of the amide with a sulfonamide was ruled out
since no inhibition was observed for both TLR7 and TLR8, (Figure 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(a)
F3C
O
F
O
O
S
R
N
H
CF3
27a, R = Cl
27b, R = H
0% TLR7 Inhibition
0% TLR8 Inhibition
F
I
F3C
O
N
N
H
CF3
28
0% TLR7 Inhibition
0% TLR8 Inhibition
(b)
F
H
N
F3C
O
O
X
OH
CF3
29, X = Cl: 0% TLR7 inhibition
IC50 = 4.02 (±0.21) TLR8
30, X = I: IC50 = 4.73 (±0.20) TLR7
IC50 = 2.63 (±0.16) TLR8
Figure 5. SAR B-ring 1 scaffold.
We then installed a -OH group ortho- to the amide for derivatives with different halogen substituents on the A-ring to explore the effect of
two H-bond donors between the A- and B-rings. Palladium-catalyzed hydroxylation of 26, which was synthesized utilizing the general synthetic
route described in (Scheme 1) afforded 29 in 91% yield, see (Scheme 4c). Compounds 29 and 30 were comparable in activity to 1; however,
the chlorinated derivative 29 lost TLR7 activity while the iodine derivative 30 regained TLR7 activity with an IC50 value of 4.72 (±0.20) µM,
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
see (Figure 5). We also attempted to replace the B-ring with a 2,4-disubstituted pyridine. However, all attempts to synthesize the pyridine
derivative using several amide bond forming reactions yielded mixtures of the imide byproduct, which were inseparable.
1
2
3
4
5
6
7
8
9
To complete our SAR studies, we introduced diversity into the C-ring since all the targets contained a symmetrical 3,5-bis(trifluorome-
thyl)phenoxy]phenyl motif. We designed intermediate 32 as a versatile building block for synthesizing 1 analogs with two different substituents
at the 3- and 5-positions of the phenoxy C-ring which would allow us to assess the importance of the -CF3 group and unsymmetrical versus
symmetrical 3,5-disubstitution patter (Scheme 5). We envisioned using the Sandmeyer reaction to convert the aniline functional group (32)
into a variety of halogens or H-bond donating groups such as -OH, see (Table 2) for summarized SAR results.
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 3. Synthesis of 35a–35f.a
F3C
O
F3C
NO2
F3C
OH
F
F3C
O
a
b
c
+
NO2
NO2
NO2
R4
NH2
OH
NH2
31
F3C
key intermediate
32
33
O
F3C
O
O
F
d
e
N
H
NH2
R4
R4
Cl
35
34
35a (R4 = H)
35b (R4 = F)
35d (R4 = Br)
35e (R4 = I)
F3C
35c (R4 = Cl) 35f (R4 = OH)
N
35g
a Reaction conditions: (a) 10 mol% H2 Pd/C, MeOH, rt (b) K2CO3, DMF, 90 °C (99% 2-steps) (c) various conditions (d) Fe powder, AcOH, EtOH
(e) 2-chloro-6-fluorobenzoyl chloride, Et3N, CH2Cl2, (40–44% 2-steps).
Table 2. SAR modification to C-ring scaffold.
F3C
O
O
F
C
N
H
R1
Cl
Cmpd
R1
TLR7
TLR8
IC50 (µM)a
IC50 (µM)a
35a
35b
35c
35d
35e
35f
H
F
2.65 (±0.84)
2.41 (±0.27)
1.59 (±0.11)
ND
Cl
Br
I
ND
ND
2.04 (±0.23)
2.32 (±0.57)
>5 uM
ND
-OH
NI
NI
F3C
35g (CU-72)
5.10 ± 0.84
2.87± 0.21
N
12
ACS Paragon Plus Environment

Page 13 of 47
1
2
3
Journal of Medicinal Chemistry
a IC50 derived from concentration-response curve in Hek 293-Blue TLR8 cell line. Data was normalized to a 1 µg/mL R848 control (data are mean
± s.d.; n = 3 independent experiments). b NI: 0% inhibition. c ND: Not determined. IC50 was not determined for compounds displaying 20-30% inhibi-
tion at 50 µM.
4
5
6
7
8
Replacing one -CF3 group with a -H (35a) did not change the potency or selectivity indicating that one substituent is sufficient for activity,
and replacing -CF3 group with halogens atom retained TLR8 activity and abolished TLR7 activity, suggesting p-stacking interactions may be
9
22
playing a role. We observed a decrease in TLR8 activity with decreasing electronegativity for 35b–35e and lost TLR7 activity. This motivated
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
us to install a hydroxyl group because it is an electronegative group that can also act as an H-bond donor or acceptor. Installation of -OH
eliminated both TLR7 and TLR8 activity indicating that an H-bond interaction is unfavorable. We then made a derivative containing a 3-
phenoxy-5-(trifluoromethyl)pyridine motif because incorporation of a pyridine nitrogen, compound 35g (CU-72), would make the aryl ring
more electron deficient and also introduce an H-bond acceptor. Interestingly, compound 35g displayed inhibitory activity toward both TLR7
and TLR8 suggesting a favorable H-bond interaction in the binding pocket of TLR7. This exciting result affords a new site to modulate TLR7/8
selectivity of the 1 scaffold. Overall, the SAR studies yielded a wealth of information for optimization of 1 and sufficient evidence to advance
CU-115 and CU-72, the lead compounds for scaffold 1, for biological studies.
Overview of key SAR results and potency optimization for 2. 1,3,4-Oxadiazoles are common heterocyclic compounds that exhibit a broad
,
23 24
spectrum of biological activities.
Our initial SAR efforts focused on determining if the diaryl ether linker between the B- and C-rings was
necessary and examining the importance of the 1,3,4-oxadiazole motif by replacing it with a 2-substituted oxazole (36). The benzyl derivative
37 inhibited TLR8 with an IC50 = 2.52 ±0.12 µM while the oxazole (36) and ester 38 derivatives were inactive, (Figure 6). These SAR studies
demonstrated that 1,3,4-oxadiazole motif and the diaryl ether linkage are essential for the bioactivity of 2. A variety of 2 analogs were synthesized
in excellent yield using the SNAr reaction between aryl fluorides and aryl tert-butyl dimethyl silyl (TBDMS) or trimethylsilyl (TMS) ethers
25
promoted by Verkade super bases (Scheme S3b). The potency of 2 was improved by replacing -CN with a -NO2 (39b) group (IC50 = 0.77
±0.09 µM ), see (Figure 6). Electron donating groups (EDG) that can act as H-bond donor or acceptor did not inhibit TLR8 signaling or were
less active (39c and 39d). We slightly improved the potency of 39b by replacing the Cl- with -I to afford 39f (CU-68) with an IC50 value of 0.3
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
±0.20 µM. CU-68 was identified as the lead hit for scaffold 2 and advanced for biological evaluation in parallel with the lead hits CU-115 and
CU-72 benzanilide inhibitors.
1
2
3
4
5
6
CN
Cl
O
N
N
N
O
7
8
9
CN
Cl
O
N
36, 0% inhibtion
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
37, IC50 = 2.52 (±0.12) µM
O
R1
N
N
N
Me
O
CN
O
O
Cl
O
38, 0% inhiibtion
TLR8
(
TLR7
(
R3
R2
R3
H
IC50
µM)
IC50
µM)
Cmpd
R2
O
39a
39b
39c
39d
39e
H
7.26 (±0.41)
NDa
NDa
NDa
NDa
NDa
> 20
C
B
N
Cl -NO2 0.77 (±0.09)
Cl -NH2 5.29 (±0.81)
N
A
O
Cl -NHAc
Br -CN 0.88 (±0.20)
-NO2 0.30 (±0.02)
NI
39f (CU-68)
I
a ND, not determined.
Figure 6. Overview of key SAR results and potency optimization for 2.
Endosomal and non-endosomal TLR specificity studies. The specificity for the new inhibitors was first tested using stably transfected
HEK 293-Blue cells individually overexpressing each human TLR. CU-68 displayed good selectivity toward TLR8 over all TLRs at 0.5 and 20
µM, see (Figure S3.). CU-115 and CU-72 display activity for TLR7 and TLR8 only at low concentrations ( 0.5 µM) but not at 5 and 20 µM.
We speculate that at higher concentrations (5 and 20 µM) the inhibitors could be binding to the nucleic acid ligand (ODN 2006) and this
26
interaction induces a chemical change in the ligand that prevents their binding for TLR9 activation. The inhibitors displayed sufficient selec-
tivity at concentrations < 5 µM and inhibitory activity at low concentrations against TLR7 and TLR8 to advance biological studies in other
immune cells. The compounds were nontoxic at low concentration (0.5 and 20 μM) and toxic at 100 μM in Hek 293 TLR7 and TLR8 cells,
see (Figure S5a). Of course, nucleic acid binding studies are needed to determine if dsDNA or ssRNA intercalation is a potential inhibitory
mechanism.
Secondary cellular assays were used to further confirm that CU-68 and CU-115 are inhibiting TLR8. First, we tested inhibition of the hTLR7
pathway using stably transfected HEK 293-Blue cells individually overexpressing each human hTLR7 and determined that CU-68 and CU-
115 do not inhibit hTLR7 at 0.5, 1, 5, and 20 µM, see (Figure 7a and Table 1). We also used RAW.264.7 murine macrophages cells to evaluate
27
TLR7 inhibition since murine TLR8 has been shown to not respond to R848. For several years, it has been speculated that murine TLR8 is
not functional; however, controversial studies have shown TLR8 to be active in murine macrophages, and a general consensus remains to be
14
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
28
established in the field. The RAW.264 cells were stimulated with R848 for activation of nitric oxide (NO) in macrophage cells. We measured
the NO levels as an indicator of R848 induced TLR7 activation to evaluate the inhibitory activity. The results showed the inhibitors do not
inhibit signaling mediated by TLR7 in RAW.264.7 cells, see (Figure 7b) at < 5 µM for CU-68 and for CU-115 only at 1.25 µM and at concen-
1
2
3
4
5
6
7
8
9
trations > 5 µM CU-68 and CU-115 displayed inhibitory activity. Further studies are needed to clarify the observed difference in activity be-
tween species. We also conducted cellular toxicity studies by treating Hek-293 TLR7, Hek-293 TLR8, THP-Dual, Raw 264.7, and RAW Dual
cell lines with sample compounds. These studies showed that CU-115, CU-72, and CU-68 inhibitors were nontoxic at low concentrations (0.5
and 20 μM) for all cell lines used in the cell-based assays, see Figure S5. Toxicity to Hek 293 TLR7 and TLR8 cell lines was observed for CU-
115 and CU-72 at 100 µM and CU-68 displayed partial toxicity at 100 μM in THP-Dual cells, see (Figure S7b).
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. CU-68 and CU-115 do not inhibit hTLR7 in RAW 264.7 and Hek 293 hTLR7 cells: (a) Effect of CU-68 and CU-115 on NF-kB inhibition
mediated by endosomal TLR7 in RAW 264.7 murine macrophages. RAW 264.7 macrophage cells were incubated with CU-68 or CU-115 for 16 h.
Activation of TLR7 mediated by R848 results in the activation of NO synthase and the production of NO in RAW 264.7 cells. The NO level was
monitored as an indicator of R848-induced TLR7 activation to evaluate the compound inhibitory activity. At 1.25 µM CU-68 and CU-115 do not
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
inhibit nitric oxide production mediated by TLR7 except at 1.25, 5, 10, and 20µM (39–51% inhibition). Inhibition caused by cellular toxicity was ruled
1
out, see Figure S5c. (b) Effect of CU-68, CU-115, and CU-72 on NF-kB inhibition mediated by ssRNA stimulated HEK 293-TLR7 cells. Human
embryonic kidney (HEK) 293 cells expressing human toll-like receptor (hTLR) 7 gene and an inducible secreted embryonic alkaline phosphatase
(SEAP) reporter gene were incubated with sample compounds for 16 h. Ligand-induced TLR activation was determined by measuring absorbance at
620 nm for the SEAP protein and comparing wells treated with sample to 50% DMSO treated and untreated cells. (c) For both studies, the data were
normalized as [(raw data – untreated cells)/(ligand + solvent control – untreated cells)]. Ligand + solvent is 100% activation, and untreated cells are
0% activation. The data are mean s.d.; n = 3 independent experiments.
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
Therapeutic potential and biophysical characterization of CU-68, CU-115, and CU-72. The inhibition of the biologically relevant
ssRNA ligands, which mimic physiological detection of viruses or siRNA molecules, with CU-68, CU-115, and CU-72 was examined to explore
their therapeutic potential. In the presence of ssRNA ligands, CU-68 inhibited TLR8 signaling completely at 5 µM while CU-115 and CU-72
displayed weaker activity, see (Figure 8a), indicating the benzanilide inhibitors are less potent than CU-68. In HEK 293 TLR8 cells stimulated
with ssRNA agonist, CU-68 has an IC50 value of 0.62 (±0.04) µM (Figure 8b), which is less potent compared to R848 stimulated HeK 293
TLR8 cells (IC50 = 0.30 (±0.02) µM). The activation mechanisms of TLR8 with distinct agonists such as R848 and ssRNA can be mechanisti-
cally different. R848 activates TLR8 by binding to the dimer interface of TLR8 near leucine-rich repeat (LRR) 11–24 and LRR16–18.4a The
conformational change induced by R848 brings the two C termini in closer proximity ultimately activating the cytoplasmic Toll-interleukin-1
(TIR) receptor.4a In contrast, ssRNA inducted TLR8 activation is achieved through synergistic binding of two ssRNA degradation products,
uridine and short oligonucleotides.4c The uridine molecule binds to the same site on the dimerization interface as R848 while the short oligo-
nucleotides binds to the concave surface of the TLR8 horseshoe structure and binding of both molecules is necessary for full activation of TLR8
by ssRNA.4c These differences in the activation mechanisms have often times led to discrepancies of IC50 values for TLR8 signaling inhibition
induced by chemical and ssRNA ligands.18a A possible explanation for the differences in activity for CU-68 as well as CU-115 and CU-72 is
that a conformational change induced by ssRNA bound to site 2 destabilizes the ligand affinity (i.e. CU-68) at site 1, resulting in decreased
inhibitory activity. Of course, x-ray crystal structures of CU-68 bound alone and in complex with ssRNA are needed for further clarification of
specific binding mode and the inhibitory mechanism.
16
ACS Paragon Plus Environment

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
Figure 8. Therapeutic potential of CU-115, CU-72, and CU-68 and biophysical characterization of CU-68. (a) Effect of CU-68, CU-115, and CU-
51
52
53
54
55
56
57
58
59
60
72 on NF-kB inhibition mediated by ssRNA stimulated human embryonic kidney (HEK) 293-TLR8 cells. HEK 293 cells hTLR8 gene and an inducible
secreted embryonic alkaline phosphatase (SEAP) reporter gene were incubated with sample compounds for 16 h. The data were normalized as [(raw
data – untreated cells)/(ligand + solvent control – untreated cells)]. Ligand + solvent is 100% activation, and untreated cells are 0% activation. At 5
µM, CU-68 suppresses the NF-kB inhibition induced by ssRNA-Lyo-40 in HEK 293 TLR8 cells while CU-115 and CU-72 displayed weaker activity
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
(data are mean s.d.; n = 3 independent experiments). (b) CU-68 dose-response-dependent NF-kB inhibition mediated by ssRNA stimulated HEK
293-TLR8 cells (data are mean s.d.; n = 3 independent experiments). (c) ITC thermogram of CU-68 titrated into hTLR8 to determine binding
affinity. The titration of R848 into hTLR8 premixed with CU-68 (left), and the titration of R848 into hTLR8 alone (right), suggesting that CU-68
prevents R848 from binding to TLR8 (representative of one independent experiment). The raw data are presented on top and the integrated peak areas
are shown and fitted below.
1
2
3
4
5
6
7
8
9
Our ongoing efforts have not yielded co-crystal structures of CU-68, CU-115, and CU-72 in complex with TLR7 and TLR8 of sufficient
resolution for structure elucidation. Thus, the direct binding of CU-68 to the ectodomain of human TLR8 was confirmed with isothermal
titration calorimetry (ITC). Unfortunately, the titration of CU-68 into TLR8 did not yield sufficient heat signals to determine the dissociation
constant (Kd) between CU-68 and TLR8. Alternatively, we assessed the inhibitory effect of CU-68 for R848, a previously established nonse-
lective TLR7/8 agonist, 27 binding to TLR8, The heat signals during the titration of R848 into TLR8 was clearly suppressed in the presence of
CU-68, suggesting that CU-68 prevented R848 from binding and subsequent structural change of TLR8. Unfortunately, attempts to conduct
ITC analyses for CU-72 did not yield data of sufficient quality to confirm its direct binding to TLR8. Because CU-72 and CU-115 both exhib-
ited low solubilities, they could not be used for ITC experiments.
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
CU-115, CU-72, and CU-68 inhibit endosomal ssRNA sensing pathways. To investigate the NA-sensing pathway selectivity, we tested
the effect of CU-115, CU-72A, and CU-68 on inhibition of the IFN immune response mediated by human retinoic-acid-inducible protein 1
receptors (RIG-I, also known as Ddx58), and cyclic AMP-GMP synthase (cGAS), which are cytosolic NA sensing receptors. We used human
THP1-Dual monocyte gene assays, containing both the Lucia luciferase and the SEAP genes and measured expression of luciferase as an indi-
cator of ligand-induced activation. The THP1-Dual cells were stimulated with 3p-hpRNA, a known RIG-I agonist in the presence or absence
of the inhibitors, and to activate the cGAS pathway we used Y-form dsDNA (G3-YSD), a known cGAS agonist. Overall, CU-68, CU-115, and
CU-72 were unable to suppress activation of the RIG-I and cGAS signaling pathways by biologically relevant immunogenic stimuli at 0.5, 1,
and 5 µM concentrations for the THP1-Dual cell line (Figure 9a). CU 68 inhibited both pathways at 20 µM and for CU-72 we observed
inhibition at 5 and 20 µM in THP1-cells, which could possibly be caused from the inhibitors binding to the RNA or DNA ligands preventing
activation. The specificity results are consistent for the RAW-Dual cell line at 0.5 and 1 µM concentrations (Figure S6) and also support our
hypothesis the inhibitors may bind to NAs at high concentrations. Of course, further clarification of the inhibitory mechanism is necessary.
18
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Figure 9. CU-68, CU-115, and CU-72 inhibit ssRNA NA sensing pathways: (a) Effects of CU-68, CU-115, and CU-72 treatment on IFN-I inhibition
mediated by human cGAS and RIG-I cytosolic sensors. The human THP1-Dual monocyte gene assays, which contains both the Lucia luciferase and
the SEAP genes was used. The THP1-Dual cells were stimulated with 3p-hpRNA, a known RIG-I agonist in the presence or absence of the inhibitors
and to activate the cGAS pathway we used Y-form dsDNA (G3-YSD), a known cGAS agonist. Quantification of luciferase was used to monitor the
expression of luciferase via detection of Lucia luciferase reporter protein secreted by cells. The compounds were considered active if they decreased
luciferase levels as indicated by a decrease in luminescence relative light units. At 0.5 and 1 µM CU-115, CU-72, and CU-68 do not modulate type-1
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
IFN transcription mediated by cGAS-STING or the RIG-I pathways. CU-72 and CU-68 display off-target effects at > 5 µM (data are mean s.d.; n =
3 independent experiments). (b) Effect of CU-68 and CU-115 on NF-kB inhibition mediated by hTNF-a in HEK 293 TLR8 cells. Hek 293 cells
expressing hTLR8 gene and an inducible secreted embryonic alkaline phosphatase (SEAP) reporter gene were incubated with sample compounds for
16 h. Ligand-induced TLR activation with hTNF-a was determined by measuring absorbance at 620 nm for the SEAP protein and comparing wells
treated with sample to 50% DMSO treated and untreated cells. CU-68, CU-115, and CU-72 do not inhibit the immune response mediated by TNF-a
which rules out inhibition via NF-kB.
In addition, we investigated the release of pro-inflammatory cytokines using recombinant TNF-a ligand. TNF-a is an inflammatory cytokine
produced in several types of immune cells. Signaling induced by TNF-a ultimately leads to activation of NF-kB and mitogen-activated protein
kinase (MAPK) signaling pathways. We measured the SEAP levels as an indicator of TNF-a induced immune response to evaluate the inhibi-
tory activity. The results showed that CU-68, CU-115, and CU-72 do not inhibit the immune response mediated by TNF-a (Figure 9b),
which rules out inhibition via NF-kB. Lastly, enzyme-linked immunosorbent assay (ELISA) was performed to measure upregulation/inhibition
of TNF-a in human THP-1 cells (hTHP-1). The results showed that all three inhibitors abolished the TNF-a production activated by R848
in hTHP-1 (Figure 10a). We also investigated the production of interleukin-1 beta (IL-1b) in hTHP-1 cells. The compounds repressed the
expression of IL-1b (Figure 10b). These results suggest the inhibitors suppress TLR8 and TLR7 signaling pathways.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 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 10. TNF-a and IL-1b in THP-1 cells in the presence and absence of CU-68, CU-115, or CU-72. R848 (a) Dose-dependent response of CU-
68, CU-115, and CU-72 TLR8-mediated TNF-a production in THP-1 cells with indicated concentration of inhibitor. Hek hTLR8 cells were stimu-
lated with 1 µg/mL R848. Data are mean s.d.; n = 3 independent experiments. (d) Dose-dependent response of CU-68, CU-115, and CU-72 TLR8-
mediated IL-1b production in THP-1 cells with indicated concentration of inhibitor. All inhibitors down-regulated TNF-a and IL-1b production ac-
tivated by R848 in hTHP-1 cells.
Scheme 4. Synthesis of 5.a
CONCLUSIONS
In conclusion, a new benzanilide compound was identified as an interesting hit compound displaying inhibitory activity toward TLR7 and
TLR8 signaling pathways. We also identified an additional hit bearing a 1,3,4-oxadiazole motif inhibiting TLR8 exclusively. Our extensive SAR
studies led to the development of two distinct scaffold classes of TLR8 inhibitors and a new TLR7/8 dual inhibitor with potential TLR7 activity
optimization. The SAR results obtained indicate the importance of the benzanilide substructure and the hydrophobic diphenyl ether skeleton
for conserving TLR7 activity. Biological evaluation of CU-115, CU-72, and CU-68 inhibitors using human monocyte THP-1, RAW264.7 and
Hek 293-Blue TLR cells confirmed they are all active for inhibiting ssRNA-sensing pathways at low concentrations and do not inhibit other
non-endosomal TLR and cytosolic NA-sensing pathways (< 5 µM). Our ITC analyses with purified TLR8 protein provided strongly suggestive
evidence of its direct binding to the ectodomain of TLR8. Crystallographic studies are needed to identify the specific binding pocket for CU-
68 and additional SAR focused on solubility optimization for characterization studies is warranted for CU-72 and CU-115. In general, the new
1 dual inhibitor represents an important class of inhibitors with potential development for a much needed TLR7 inhibitor. We anticipate the
SAR results will be instructive for optimization studies based on the CU-72 scaffold.
EXPERIMENTAL SECTION
Protein preparation and isothermal titration calorimetry (ITC). The extracellular domain of human Toll-like receptor 8 (hTLR8, residues 27–
827) was prepared as described previously4.
20
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
ITC experiments were done in a buffer composed of 25 mM MES pH 5.5, 0.20 M NaCl, and 5% dimethyl sulfoxide (DMSO) at 298 K using a MicroCal iTC200
(GE Healthcare). The titration sequence included a single 0.4 µL injection followed by 18 injections, 2 µL each, with a spacing of 120 s between the injections.
The titration conditions were as follows: 100 µM R848 into 10 µM hTLR8 and 50 µM CU-68; 100 µM R848 into 10 µM hTLR8. OriginLab software (GE
Healthcare) was used to analyze the raw ITC data.
1
2
3
4
5
6
Chemistry. Materials and Methods. Unless otherwise noted, all non-aqueous reactions were run under an atmosphere of dried nitrogen (N2) in dried
glassware. All reagents were reagent grade and used without further purification. Moisture sensitive reagents were added via syringe. All chemicals were obtained
from Sigma-Aldrich, Acros, or Strem unless otherwise noted. Flash column chromatography was performed using EM Reagents Silica Gel 60 (230-400).
Analytical thin-layer chromatography (TLC) was performed using EM Reagents 0.25 mm silica gel 254-F plates. Visualization was accomplished with UV light,
p-anisaldehyde stain, and/or iodine. 1H NMR and 13C NMR spectra were recorded on a Brüker-III NMR spectrometer operating at 400 MHz. Chemical shifts
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
1
are reported relative to the solvent resonance peak δ 7.27 (CDCl3) for H and δ 77.23 (CDCl3) for 13C NMR. Data are reported as follows: chemical shift,
multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, bs = broad singlet, m = multiplet), coupling constants, and number of protons. High-resolution mass
spectra were obtained using a VG Autospec using an ionization mode of either ESI or CI. Infrared spectra are reported in cm-1 and recorded using an Agilent
Technologies Cary 630 FTIR. Unless otherwise noted, all yields refer to isolated yields. A MPA 160 melting point apparatus was used to measure the melting
point.
High performance liquid chromatography (HPLC) Purity Method: The purity of tested compounds (inhibitor samples > 95% purity) was
evaluated using An Agilent Technologies 1200 series HPLC instrument with an Agilent Eclipse XDB-C18 column (reverse phase) Mobile phase (A) 100%
HPLC grade acetonitrile (MeCN) with 0.1% HPLC grade trifluoroacetic acid (TFA) and mobile phase (B) 100% HPLC grade water (H2O) 0.1% HPLC grade
TFA. The gradient was 80% A at 0 min then to 80–60 to 5 min then 60% A with 5 min hold (1.0 mL/min). The mobile phase was then changed to 100% B to
10 with a 5 min hold then 60:40 (A:B) with 5 min hold then 80:20 for 7 min. A flow rate of 1.0 mL/min was used for the entire method, with a total assay time
of 22 min.
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2-chloro-6-fluoro-N-methylbenzamide (3): To a solution of N-{4-[3,5-bis(trifluorome-
thyl)phenoxy]phenyl}-2-chloro-6-fluorobenzamide (1) (0.13 mmol, 60 mg) in DMF (0.27 M) was added sodium hydride 60% in dispersion mineral oil (0.25
mmol, 6 mg) at 0 °C. The mixture was warmed to rt and stirred for 30 min. Methyl iodide (0.19 mmol, 11.7 µL) was then added dropwise at 0 °C and stirred at
0 °C. After 1 h, then more methyl iodide (0.36 mmol, 22 µL) was added and the mixture was stirred at rt overnight. The reaction was then quenched with
saturated aqueous solution of NH4Cl. The mixture was extracted with EtOAc (3x 1 mL). The organic mixture was then washed with 5% LiCl aq (5x 1 mL),
dried with Na2SO4, and concentrated by rotary evaporation to afford a white solid. Following purification by column chromatography (20% EtOAc in hexane),
the title compound was obtained in 84% yield (52 mg, 98% pure): mp 74–75 °C; 1H NMR (DMSO-d6, 400 MHz) (mixture of rotamers) δ ; 7.58 (br s, 1H),
7.36 (d, J = 8.8 Hz, 2 H), 7.35 – 7.30 (m, 1 H), 7.20 – 7.10 (m, 1 H), 7.15-7.10 (m, 1 H), 7.14–7.11 (m, 2 H), 7.06 (ddt, J = 8.1, 3.7, 0.9 Hz, 2 H), 6.92 (d, J =
Hz 1 H), 6.84 (tt, J = 8.4, 0.9 Hz, 1 H), 3.55 (s, 3 H, minor rotamer), 3.49 (s, 3 H); 13C NMR (DMSO-d6, 101 MHz) (mixture of rotamers) δ 163.5, 163.3,
158.3 (d, 1JCF = 251.0 Hz), 158.3 (d, 1JCF = 250.3 Hz), 158.6, 154.4, 141.7, 139.7, 132.4, 130.9 (d, 3JCF = 9.1 HZ), 130.9 (d, 3JCF = 9.0 HZ), 128.1, 128.0 (d, 2JCF3
= 36.9Hz, 125.2 (d, 2JCF = 10.7 Hz), 125.3 (d, 2JCF = 10.8 Hz), 122.9 (q, 1JCF3 = 274.0 Hz), 121.6, 120.7, 117.9 (q, 4JCF3 = 3.1 Hz), 116 (m, 3JCF3), 114.0 (d, 2JCF
=
21.7 Hz), 113.8 (d, 2JCF = 21.6 Hz), 37.1, 37.0; 19F NMR (365 MHz, CDCl3) δ -63.0, -111.2, 111.5; IR (film) 3070, 2925, 1659, 1614, 1584, 1462, 1365, 1279,
1234, 1171, 1130, 1108, 1007, 955, 895, 854, 823, 702, 683 cm-1; HRMS (ESI+) calcd for C13H16ClF7N6O2 [M+H]+, m/z = 492.0601, found 492.0601.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
4-(3,5-Bis(trifluoromethyl)phenoxy)-N-(2-chloro-6-fluorobenzyl)aniline (4)29: A vial containing acetic acid (0.1 mL, 1.8 mmol), 12 (0.32 g, 1
mmol), 2-chloro-6-fluorobenzaldehyde (0.24 g, 1.5 mmol), and MeOH (5 mL) was stirred at rt for 30 minutes, then cooled to 0 °C. Sodium cyanoborohydride
(0.08g, 1.2 mmol) was added in one portion, and the reaction warmed to rt while stirred for 18 hours. The resulting solution was diluted with 20 mL EtOAc,
then washed 2x with 10 mL H2O, once with 10 mL brine, dried over Na2SO4, and concentrated to dryness. Following purification by column chromatography
1
2
3
4
5
6
7
1
(0-100% CH2Cl2 in hexane), the title compound was obtained in 88% yield (0.41 g, 99% pure): mp = 76–77 °C; H NMR (400 MHz, CDCl3) (mixture of
8
9
rotamers) δ 7.49 (s, 1H), 7.30 (s, 2 H), 7.25 – 7.20 (m, 2 H), 7.06 – 6.99 (m, 1 H), 6.94 – 6.89 (m, 2 H), 6.81 – 6.75 (m, 2 H), 4.17 (s, 3 H); 13C NMR (101
MHz, CDCl3) δ 161.8 (d, 1JCF = 249.7 Hz), 160.3, 146.2, 145.4, 135.7 (d, 3JCF = 5.6 Hz), 133.0 (q, 2JCF3 = 33.6 Hz), 129.8 (d, 3JCF = 9.8 Hz), 125.8 (d, 4JCF = 3.3
Hz), 124.6 (d, 2JCF = 25.7 Hz), 123.2 (q, 1JCF3 = 279.6 Hz), 121.7, 116.8 (d, 4JCF3 = 4.1 Hz), 115.3 (sep, 3JCF3 = 3.9 Hz),114.8, 114.5 (d, 2JCF = 23.3 Hz), 40.0 (d,
3.4 Hz); 19F NMR (365 MHz, CDCl3) δ -63.0, -113.7; IR (film) 3417, 3085, 1610, 1581, 1514, 1454, 1383, 1320, 1286, 1223, 955, 881, 862, 836, 687 cm-1;
HRMS (ESI+) calcd for C21H13F7ClNO [M+H]+, m/z = 464.0652, found 464.0652.
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
General Procedures for Compounds 3, 6, 7, 13a-13o, and 14-14g:
General Procedure A: Amide Coupling via HATU30: To a flame-dried vial was added HATU (0.6 mmol), carboxylic acid (0.56 mmol), and DMF
(1 mL). The vial was cooled to 0 °C and diisopropylethylamine (0.6 mmol) was added, then the vial was removed from ice bath. After the vial was stirred for at
least 15 minutes, the appropriate aniline (0.5 mmol) dissolved in DMF (1 mL) was added, and the reaction mixture was stirred overnight. After completion, the
solvent was removed in vacuo and the crude product was partitioned between 5 mL each water and ethyl acetate. The aqueous layer was extracted 3x with 5 mL
ethyl acetate, after which the combined organic layers were washed with 5 mL brine, dried over Na2SO4, and concentrated to dryness. Purification varied by
compound, typically flash chromatography or trituration in cold hexane.
General procedure B: amide coupling via acyl chloride: To a flame-dried flask containing substituted aniline (0.6 mmol) in CH2Cl2 (2 mL) was
added acid chloride (0.72 mmol) dissolved in CH2Cl2 (2 mL). The flask was placed under nitrogen and cooled to 0 °C, after which triethylamine (0.1 mL, 0.72
mmol) was added dropwise. The reaction was allowed to return to rt while stirring overnight. The solvent was removed in vacuo and the resulting residue was
partitioned between 10 mL ethyl acetate and water. The aqueous layer was extracted 3x with 5 mL ethyl acetate, and the combined organic layers were sequen-
tially washed with 5 mL 1M NaOH and 5 mL brine, dried over Na2SO4, and concentrated to dryness. Purification varied by compound.
N-(3,5-Bis(trifluoromethyl)phenyl)-2-chloro-6-fluorobenzamide (5): Prepared by general method B using 3,5-bis-(trifluoromethyl)aniline 0.52
mmol, 0.52 mmol 2-chloro-6-fluorobenzoyl chloride, 0.78 mmol Et3N and 2.6 mL CH2Cl2. The solid was purified by flash column chromatography (20% EtOAc
in hexanes) to afford the title compound as a white solid in 28% yield (56 mg, 99% pure): 1H NMR (CDCl3, 400 MHz) δ 8.13 (d, J = 1.6 Hz, 2 H), 7.95 (s, 1
H), 7.68 (tt, J = 1.5, 0.8 Hz, 1 H), 7.39 (td, J = 8.3, 6.0 Hz, 1 H), 7.29 – 7.26 (m, 1 H), 7.10 (td, J = 8.6, 1.0 Hz, 1 H). ; 13C NMR (CDCl3, 101 MHz) δ 160.8,
2
4
1
2
4
158.3, 138.6, 132.6 (q, JCF3 = 33.8 Hz), 132.2, 132.1, 126.0 (d, JCF = 3.5 Hz), 123.0 (q, JCF3 = 274.0 Hz), 124.2 (d, JCF = 20.5 Hz), 119.8 (q, JCF3 = 3.0 Hz),
118.5 (q, 3JCF3 = 3.8 Hz); 19F NMR (376 MHz, CDCl3) δ -63.0, -112.0; IR (film) 3253, 3093, 1722, 1670, 1558, 1502, 1473, 1454, 1380, 1305, 1279, 1231, 1171,
1134, 1063, 951, 903, 884, 795, 773, 683, 527 cm-1; HRMS (ESI+) calcd for C15H7ClF7NO [M+H]+, m/z = 386.0177, found 386.0180.
4-[3,5-Bis(trifluoromethyl)phenoxy]-N-(2-chloro-6-fluorophenyl)benzamide (6): A vial containing methyl 4-(3,5-bis(trifluoromethyl)phe-
noxy)benzoate (0.15 g, 0.41 mmol), 2 M NaOH (2 mL, 4 mmol), and MeOH (4 mL) was sealed and heated at 50 °C for 2.5 hours, then partitioned between
10 mL each 3 M HCl and EtOAc. The aqueous layer was extracted 3x with 10 mL EtOAc, and the combined organic layers washed with 10 mL brine, dried over
Na2SO4, and concentrated to dryness. The resulting white solid was added to a vial containing CH2Cl2 (5 mL) and DMF (1 drop), then placed under N2 and
cooled to 0 °C. Oxalyl chloride (0.086 mL, 1 mmol) was added and the mixture gradually warmed to rt while stirring for 30 minutes. Volatiles were removed in
22
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
vacuo, then 2-fluoro-6-chloroaniline (0.09 g, 0.62) dissolved in CH2Cl2 (5 mL) was added, followed by Et3N (0.28 mL, 2 mmol). This mixture was stirred for 3
h, then concentrated. The residue was partitioned between 10 mL each EtOAc and H2O, and the aqueous layer extracted 3x with 10 mL EtOAc. Combined
organic layers were washed with brine (10 mL), dried over Na2SO4 and concentrated. The residue was triturated with hexane and then purified by column
1
2
3
4
5
6
7
1
chromatography (0-100% CH2Cl2 in hexane) to afford the title compound in 57% yield (0.111g, 99% pure) as a tan solid : mp = 178–180 °C; H NMR (400
MHz, Acetone-d6) δ 9.40 (s, 1 H), 8.21 – 8.16 (m, 2 H), 7.87 (s, 1 H), 7.78 – 7.75 (s, 2 H), 7.46 – 7.38 (m, 2 H), 7.37 – 7.30 (m, 2 H), 7.30 – 7.23 (m, 1 H); 13
C
8
9
1
4
2
NMR (101 MHz, Acetone-d6) (rotamers) δ 165.12, 165.1, 159.9 (d, JCF = 248.8 Hz), 159.4, 158.6, 134.1 (d, JCF = 2.6 Hz),133.7 (q, JCF3 = 33.7 Hz), 131.0,
130.8 (d, 4JCF = 3.0 Hz), 129.5 (dd, 3JCF = 2.2, 5.8, Hz), 125.9 (d, 4JCF = 3.5 Hz), 123.8 (q, 1JCF3 = 273.2 Hz), 120.2 (q, 4JCF3 = 4.0 Hz), 117.8 (sep, 3JCF3 = 4.0 Hz),
119.8, 115.4 (d, 2JCF = 21.3 Hz); 19F NMR (376 MHz, CDCl3) δ -63.0, -114.1; IR (film) 3245, 1666, 1502, 1465, 1450, 1376, 1283, 1246, 1167, 1126, 951, 873,
780, 706, 687, 668, 653 cm-1; HRMS (ESI+) calcd for C21H11ClF7NO2 [M+H]+, m/z = 478.0445, found 478.0443.
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-{3-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-chloro-6-fluorobenzamide(7): Prepared by general method B using with 0.15 mmol 3-[3,5-
bis(trifluoromethyl)phenoxy]aniline, 0.18 mmol 2-chloro-6-fluorobenzoyl chloride, 0.3 mmol Et3N, and 5 mL CH2Cl2. Purified by column chromatography
(0-60% CH2Cl2 in hexane) to afford the title compound as a white solid in 70% yield (51 mg, 99% pure): mp = 100–104 °C; 1H NMR (400 MHz, CDCl3) δ 7.69
(s, 1 H), 7.60 – 7.58 (m, 1 H), 7.58 – 7.55 (m, 1 H), 7.43 (s, 2 H), 7.41 – 7.33 (m, 3 H), 7.26 – 7.23 (m, 1 H), 7.08 (td, J = 8.5, 1.0 Hz, 1 H), 6.88 – 6.83 (m, 1
H); 13C NMR (101 MHz, CDCl3) δ 160.9, 160.6, 157.2 (d, 1JCF = 273.6 Hz), 158.4, 139.3, 133.4 (q, 2JCF3 = 33.8 Hz), 132.6 (d, 4JCF = 4.9 Hz), 131.9 (d, 3JCF = 9.2
Hz), 130.9, 126.0 (d, 4JCF = 3.6 Hz), 125.0 (d, 2JCF = 21.3 Hz), 123.9 (q, 1JCF3= 274.0 Hz), 118.3 (d, 4JCF3 = 4.0 Hz), 116.7 (sep, 3JCF3 = 3.9 Hz), 116.6, 116.2, 114.8
(d, JCF = 21.7 Hz), 112.0; 19F NMR (376 MHz, CDCl3) δ -62.9, -112.2; IR (film) 3253, 3201, 3137, 3085, 1674, 1659, 1607, 1577, 1555, 1488, 1454, 1372,
2
1331, 1305, 1279, 1223, 1175, 1126, 985, 940, 907, 881, 851, 795, 773, 735, 698, 571, 519 cm-1; HRMS (ESI+) calcd for C21H11ClF7NO2 [M+H]+, m/z =
478.0445, found 478.0443.
Methyl 4-(3,5-bis(trifluoromethyl)phenoxy)-3-nitrobenzoate (9): Prepared by general method C with 4.8 mmol 1-Bromo-3-(4-nitrophenoxy)-5-
(trifluoromethyl)benzene 5.8 mmol 3,5-bis(trifluoromethyl)phenol, 14.4 mmol Cs2CO3, and 20 mL DMF. Purified by column chromatography (0-50% CH2Cl2
in hexane) to afford the title compound as a pale yellow oil in 79% yield (1.55 g): 1H NMR (400 MHz, CDCl3) δ 8.70 (d, J = 2.1 Hz, 1H), 8.28 (dd, J = 8.7, 2.1
Hz, 1H), 7.74 (s, 1H), 7.49 (s, 2H), 7.14 (d, J = 8.6 Hz, 1H), 3.99 (s, 3H); 13C NMR (101 MHz, CDCl3) δ 164.2, 156.3, 151.8, 141.4, 135.5, 133.9 (q, 2JCF3
=
34.3 Hz), 127.8, 127.4, 122.6 (q, JCF3 = 274.1 Hz), 120.9, 121.9, 119.1 (q, JCF3 = 3.8 Hz), 118.5 (q, JCF3 = 4.0 Hz); 19F NMR (376 MHz, CDCl3) δ -63.0; IR
(film) 3324, 3093, 1733, 1648, 1607, 1543, 1510, 1462, 1443, 1279, 1242, 1171, 1126, 951, 761, 706, 687 cm-1; HRMS (ESI-) calcd for C16H9F6NO5 [M+Li]+,
m/z = 416.0545, found 416.0548.
1
4
3
General procedure C: Fe reduction of nitroarenes:
A flask containing the nitroarene (1 mmol), iron powder (4 mmol), 4 mL glacial AcOH, and 4mL EtOH was heated at reflux for 70 minutes. The mixture was
filtered through a pad of Celite, concentrated in vacuo, then partitioned between 10 mL each EtOAc and H2O. Residual acid was quenched with the addition of
saturated NaHCO3, and the aqueous layer extracted 3x with 10 mL EtOAc. The combined organic layers were washed with 10 mL brine, dried over Na2SO4,
and concentrated to dryness.
General Procedures D: Pd reduction of nitroarenes for Compounds 10, 31, 14k, 14l, 35b, and 35f:31
.
A flask was charged with nitroarene (9.6 mmol) and 10% Pd/C (2.9 mmol), then placed under vacuum and backfilled with three times with N2. MeOH (80
mL) degassed by sparging with N2 was added, after which H2 was introduced via a balloon. The reaction was monitored by TLC; at completion, the mixture was
filtered through a pad of Celite and concentrated to dryness. Purification method varied by compound
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
Methyl 4-(3,5-bis(trifluoromethyl)phenoxy)benzoate (10) From the nitroarene, the aniline was obtained by general procedure D with 2.1 mmol
7, 0.6 mmol 10% Pd/C, and 25 mL MeOH. The resulting white solid was dissolved in DMF (5 mL) and added to a solution of tert-butyl nitrite (0.36 mL, 3
mmol) in DMF (5 mL) at 50 °C. The mixture was stirred for 1 hour, then partitioned between 20 mL EtOAc and 20 mL H2O. The aqueous layer was extracted
3x with 10 mL EtOAc, and the combined organic layers washed 3x with 10 mL 5% LiCl, 1x with 10 mL brine, dried over Na2SO4 and concentrated to dryness.
1
2
3
4
5
6
7
1
Following purification by column chromatography (0-50% CH2Cl2 in hexane), the title compound was obtained as a yellow solid in 75% yield (0.56 g): H
8
9
NMR (400 MHz, Chloroform-d) δ 8.14 – 8.07 (m, 2H), 7.65 (s, 1H), 7.45 (s, 2H), 7.10 – 7.04 (m, 2H), 3.93 (s, 3H); 13C NMR (101 MHz, CDCl3) δ 166.3,
159.5, 157.5, 133.6 (q, 2JCF3 = 33.9 Hz), 132.3, 127.0, 121.6 (q, 1JCF3 = 272.3 Hz), 119.3 (4JCF3 = 3.7 Hz), 118.8, 117.5 (3JCF3 = 3.7 Hz), 52.4; 19F NMR (376 MHz,
CDCl3) δ -63.0 (Note: 3 was fully characterized).
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-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-fluorobenzamide (14a): Prepared by general method B using 0.36 mmol 2-fluorobenzoyl
chloride, 0.3 mmol 13, 0.9 mmol pyridine, and 2 mL CH2Cl2. The crude product was purified by column chromatography (0-45% CH2Cl2 in hexane), then
1
recrystallized from boiling methanol to afford the title compound as a white solid in 81% yield (0.11 g, 99% pure): mp = 192–193 °C; H NMR (400 MHz,
CDCl3) δ 8.52 (d, J = 16.0 Hz, 1H), 8.20 (td, J = 8.0, 1.9 Hz, 1 H), 7.77 – 7.72 (m, 2 H), 7.59 – 7.52 (m, 2 H), 7.38 (s, 2 H), 7.35 (td, J = 7.6, 1.1 Hz, 1 H), 7.24
– 7.17 (m, 1 H), 7.12 – 7.08 (m, 2 H); 13C NMR (CDCl3, 101 MHz) δ 160.6 (d, 1JCF = 247.1 Hz), 161.3 (d, 4JCF = 3.6 H), 151.4, 135.0, 134.0 (d, 3JCF = 9.6 Hz),
133.2 (q, 2JCF3 = 33.8 Hz), 132.3 (d, 4JCF = 1.9 Hz), 125.2 (d, 3JCF = 3.3 Hz), 122.9 (q, 1JCF3 = 273.9 Hz), 122.6, 121.0 (d, 2JCF = 11.3 Hz), 120.7, 117.6 (q, 4JCF3
=
2
3
4.1 Hz), 116.2 (d, JCF = 25.3 Hz), 116.3, 116.14 (sep, JCF3 = 3.7 Hz), 116.06; 19F NMR (365 MHz, CDCl3) δ -62.97, -113.14; IR (film) 3376, 3119, 3096,
1663, 1610, 1540, 1510, 1488, 1462, 1372, 1190, 1167, 1130, 951, 866, 843, 758, 683, 624 cm-1; HRMS (ESI+) calcd for C21H13F7NO2 [M+H]+, m/z = 444.0830,
found 444.0832.
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2,6-difluorobenzamide (14b): Prepared by general method B using 0.78 mmol 2,6-difluoro-
benzoyl chloride, 0.78 mmol 4-[3,5-bis(trifluoromethyl)phenoxy]aniline, 1.6 mmol Et3N, and 3 mL CH2Cl2. Purified by column chromatography (50% CH2Cl2
in hexane to 100% CH2Cl2) to afford product as a white solid in 54% yield (0.192 g, 99% pure): mp = 160–162 °C; 1H NMR (CDCl3, 400 MHz) δ 7.76 (s, 1 H),
7.74 – 7.67 (m, 2 H), 7.57 (tq, J = 1.5, 0.7 Hz, 1 H), 7.51 – 7.39 (m, 1 H), 7.42 – 7.36 (m, 2 H), 7.13 – 7.05 (m, 2 H), 7.05 – 6.96 (m, 2 H); 13C NMR (CDCl3,
3
3
3
1
101 MHz) δ 161.3 (d, JCF = 6.0 Hz), 158.9, 158.8 (d, JCF = 7.0 Hz), 158.4, 151.6, 134.5, 133.1 (2JCF3 = 32.8 Hz), 132.2 (dd, JCF = 11.6 Hz), 121.5 (q, JCF3
=
273.5 Hz), 122.2, 120.7, 117.6 4JCF3 = 3.6 Hz), 116.2 (sep, 3JCF3 = 3.6 Hz), 112.4 (d, 2JCF = 25.3 Hz); 19F NMR (365 MHz, CDCl3) δ -63.0, -111.7; IR (film) 3316,
3253, 3190, 3119, 3063, 1670, 1625, 1607, 1536, 1514, 1465, 1376, 1279, 1234, 1190, 1171, 1130, 1108, 1074, 1011, 955, 877, 843, 825, 795, 735, 717, 702,
683, 646, 594, 579 cm-1; HRMS (ESI+) calcd for C21H12F8NO2 [M+H]+, m/z = 462.0735, found 462.0736.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-chloro-6-fluorobenzamide (14c, 1):Prepared by general B with 0.6 mmol 12, 0.72 mmol
2-chloro- 6-fluorobenzoyl chloride, 0.72 mmol Et3N, and 4 mL CH2Cl2. Purified by column chromatography (0-80% CH2Cl2 in hexane) to yield title product
as a white solid in 96% yield (0.69 g, 99% pure): 1H NMR (400 MHz, CDCl3) δ 7.76 – 7.67 (m, 2 H), 7.58 (d, J = 1.6 Hz, 1 H), 7.49 (s, 1 H), 7.45 – 7.34 (m, 3
H), 7.29 (dt, J = 8.1, 0.9 Hz, 1 H, 7.17 – 7.06 (m, 3 H); 19F NMR (376 MHz, CDCl3) δ -62.97, -112.27; IR (film) 3260, 1666, 1610, 1532, 1510, 1454, 1376,
1279, 1175, 1126, 955, 881, 784, 683, 568 cm-1;
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-5-chloro-2-fluorobenzamide (14d): Prepared by general method A using 0.55 mmol 13, 0.5
mmol 5-chloro-2-fluorobenzoic acid, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. Purification by column chromatography (0-100% CH2Cl2 in hexane)
afforded the title compound as a white powder in 75% yield (0.18 g, 99% pure): mp = 122–123 °C; 1H NMR (400 MHz, CDCl3) δ 8.45 (d, J = 15.6 Hz, 1H),
8.17 (dd, J = 6.7, 2.8 Hz, 1 H), 7.78 – 7.69 (m, 2 H), 7.57 (s, 1 H), 7.50 (ddd, J = 8.8, 4.4, 2.8 Hz, 1 H), 7.38 (s, 2 H), 7.17 (dd, J = 11.4, 8.8 Hz, 1 H), 7.14 – 7.06
24
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
(m, 2 H); 13C NMR (101 MHz, Acetone-d6) δ 161.9 (dd, 3JCF = 8.2, 1.7 Hz), 159.3 (d, 2JCF3 = 249.3 Hz), 160.5, 152.0, 137.1 (d, 3JCF = 9.1 Hz), 133.7 (q, 2JCF3
=
1
2
3
4
5
6
7
33.5 Hz), 133.6 (d, 3JCF = 9.0 Hz), 130.9 (d, 4JCF = 3.1 Hz), 130.2 (d, 4JCF = 3.3 Hz), 126.8 (dd, 3JCF = 8.2, 4.5 Hz), 124.2 (q, 1JCF3 = 272.1 Hz), 122.9 (d, 3JCF = 9.2
Hz), 121.7, 119.1 (d, 2JCF = 25.0 Hz), 118.7 (d, 4JCF = 3.0 Hz), 116.7 (sep, 3JCF3 = 3.9 Hz); 19F NMR (376 MHz, CDCl3) δ -62.97, -116.14; IR (film) 3342, 1655,
1610, 1543, 1510, 1480, 1462, 1372, 1283, 1190, 1167, 1130, 1108, 951, 873, 854, 821, 676 cm-1; HRMS (ESI+) calcd for C21H11F7ClNO2 [M+Li]+, m/z =
484.0527, found 484.0533.
8
9
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-bromo-6-fluorobenzamide(14e): As general procedure A, with 0.5 mmol 13, 0.56 mmol 2-
bromo-6-fluorobenzoic acid, 0.6 mmol HATU, 0.6 mmol DIPEA, and 2 mL CH2Cl2. Purification by flash chromatography (0-60% CH2Cl2 in hexane) followed
by trituration in cold hexane yielded a white-tan solid in 76% yield (0.20 g, 99% pure): mp = 180–182 °C; 1H NMR (400 MHz, CDCl3) δ 7.74 – 7.69 (m, 2 H),
7.58 (t, J = 1.4, 0.7 Hz, 1 H), 7.52 (s, 1 H), 7.45 (dt, J = 8.0, 0.9 Hz, 1 H), 7.39 (dt, J = 1.7, 0.6 Hz, 2 H), 7.32 (td, J = 8.3, 5.9 Hz, 1 H), 7.15 (td, J = 8.5, 1.0 Hz,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1 H), 7.12–7.08 (m, 2 H); 13C NMR (101 MHz, CDCl3) δ 161.5, 159.6 (d 1JCF = 252.8 Hz), 159.0, 151.9, 134.5, 133.4 (d, 2 CF3 = 33.7 Hz), 132.2 (d, 3JCF = 8.7
J
Hz), 129.1 (d, 4JCF = 3.5 Hz), 127.2 (d, 2JCF = 21.1 Hz), 123.1 (1JCF3 = 272.8 Hz), 122.5, 120.92, 120.85 (d, 2JCF = 4.23 Hz), 117.8 (d, 4JCF = 4.5), 116.4 (sep, 3JCF3
= 3.9 Hz), 115.4 (d, 3JCF = 21.6 Hz);19F NMR (376 MHz, CDCl3) δ -62.97, -111.60; IR (film) 3264, 1666, 1610, 1532, 1510, 1462, 1279, 1171, 1126, 955, 877,
870, 843, 780, 706, 683 cm-1; HRMS (ES-) [M-H]- for C21H10BrF7NO2 , m/z 519.9783, found 519.9781.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-fluoro-6-iodobenzamide (14f, CU-115): Prepared by general method B using 0.36 mmol
2-fluoro-6-iodobenzoyl chloride, 0.3 mmol 13, 0.9 mmol pyridine, and 2 mL CH2Cl2. Purified by column chromatography (0-75% CH2Cl2 in hexane) to afford
product as a white solid in 94% yield (0.16 g, 99% pure): mp = 191–193 °C; 1H NMR (400 MHz, CDCl3) δ 7.75 – 7.69 (m, 3 H), 7.58 (s, 1 H), 7.40 (s, 2 H),
7.21 – 7.14 (m, 2 H), 7.14 – 7.09 (m, 2 H). 19F NMR (376 MHz, CDCl3) δ -62.94, -110.81. 13C NMR (101 MHz, CDCl3) δ 163.2, 158.8 (d, 1JCF = 253.4 Hz),
159.0, 151.9, 135.5 (d, 4JCF = 3.48 Hz), 134.5, 133.4 (q, 2JCF3 = 34.3 Hz), 132.6 (d, 3JCF = 8.4 Hz), 130.9 (d, 2JCF = 20.2 Hz), 123.0 (q, 1JCF3 = 272.9 Hz), 122.5,
117.8 (d, 4JCF = 3.2 Hz), 120.9, 116.4 (sep, 3JCF3 = 3.9), 116.1 (d, 2JCF = 21.7 Hz), 93.6 (d, 3JCF = 2.8 Hz); 19F NMR (376 MHz, CDCl3) δ -62.9, -110.8; IR (film)
3290, 1655, 1514, 1279, 1175, 884 cm-1; HRMS (ESI+) calcd for C16H10ClN3O2 [M+H]+, m/z = 312.0540, found 312.0541.
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2-fluoro-5-iodobenzamide (14g): Prepared by general procedure B using 0.5 mmol 12, 0.55
mmol 2-fluoro-5-iodobenzoyl chloride, 1 mmol Et3N, and 5 mL CH2Cl2. Purified by column chromatography (0-100% CH2Cl2 in hexane) to afford the title
compound as a white solid in 89% yield (0.25 g, 99% pure): mp = 140–142 °C; 1H NMR (400 MHz, CDCl3) δ 8.48 (dd, J = 7.3, 2.4 Hz, 1 H), 8.41 (d, J = 15.4
Hz, 1 H), 7.83 (ddd, J = 8.6, 4.8, 2.4 Hz, 1 H), 7.75 – 7.68 (m, 2 H), 7.57 (s, 1 H), 7.38 (s, 2 H), 7.13 – 7.06 (m, 2 H), 6.98 (dd, J = 11.8, 8.6 Hz, 1 H); 13C NMR
(101 MHz, CDCl3) δ 160.2 (d, 1JCF = 248.3 Hz), 159.8 (d, 4JCF = 3.6 Hz), 158.9, 151.6, 142.6 (d, 3JCF = 9.4 Hz), 140.9, 134.6, 133.2 (q, 2JCF3 = 33.8 Hz), 127.0,
122.9 (q, 1JCF3 = 273.9 Hz), 123.0 (d, 2JCF = 12.4 Hz), 122.6, 120.7, 118.3 (d, 2JCF = 26.3 Hz), 117.6 (d, 4JCF3 = 4.0 Hz), 116.2 (q, 3JCF3 = 3.9 Hz); 19F NMR (376
MHz, CDCl3) δ -63.0, -114.9; IR (film) 3324, 3096, 3070, 1543, 1506, 1465, 1376, 1283, 1164, 1126, 955, 877, 854, 832, 817, 687, 609, 512 cm-1; HRMS
(ESI+) calcd for C21H11F7INO2 [M+H]+, m/z = 569.9801, found 569.9790.
N-{4-[3,5-bis(trifluoromethyl)phenoxy]phenyl}-2-fluoro-4-iodobenzamide (14h): Prepared by general method A using 0.5 mmol 2-
fluoro-4-iodobenzoic acid, 0.55 mmol 13, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. The crude product was purified by column chromatog-
raphy (0-100% CH2Cl2 in hexane) followed by recrystallization from hot hexane to afford title compound as white crystals in 19% yield (53 mg, 98%
pure): mp = 172–173 °C; 1H NMR (400 MHz, CDCl3) δ 8.41 (d, J = 15.6 Hz, 1 H), 7.90 (t, J = 8.4 Hz, 1 H), 7.76 – 7.68 (m, 3 H), 7.61 (dd, J = 11.4,
1.6 Hz, 1 H), 7.57 (s, 1 H), 7.38 (s, 2 H), 7.15 – 7.06 (m, 2 H);13C NMR (101 MHz, CDCl3) 160.63 (d, 4JCF = 3.0 Hz), 150.59 (d, 4JCF = 3.7 Hz), 158.9,
158.2, 151.6. 134.8 (d, 3JCF = 3.3 Hz), 134.7, 133.4, 133.1 (q, 2JCF3 = 31.3 Hz), 125.4 (2JCF = 26.5 Hz), 121.5 (1JCF3 = 273.4 Hz), 122.6, 120.7, 117.6 (3JCF3
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
= 3.4 Hz), 116.2 (sep, 3JCF3 = 3.8 Hz), 98.7 (3JCF = 9.0 Hz) ; 19F NMR (376 MHz, CDCl3) δ -62.97, -111.49; IR 3387, 3067, 1670, 1603, 1547, 1506,
1465, 1368, 1279, 1190, 1167, 1141, 1119, 951, 899, 854, 832, 706, 687, 635, 590, 512 (film) cm-1; HRMS (ESI+) calcd for C21H11F7INO2 [M+H]+,
m/z = 569.9801, found 569.9792.
1
2
3
4
5
6
7
8
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2-fluoro-6-(trifluoromethyl)benzamide (14i): Prepared by general procedure B using 0.5
mmol 13, 0.55 mmol 2-fluoro-6-(trifluoromethyl)benzoyl chloride, 1 mmol Et3N, and 5 mL CH2Cl2. Purified by column chromatography (0-100% CH2Cl2 in
hexane) to afford the title compound as a tan solid in 75% yield (0.19 g, 99% pure): mp = 204–206 °C; 1H NMR (400 MHz, Acetone-d6) δ 10.01 (s, 1 H), 7.92
– 7.84 (m, 2 H), 7.81 – 7.74 (m, 2 H), 7.69 (d, J = 7.9 Hz, 1 H), 7.65 – 7.59 (m, 3 H), 7.29 – 7.23 (m, 2 H); 13C NMR (101 MHz, Acetone-d6) 159.3, 158.4,
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
2
3
1
3
158.2 (d, 1JCF = 248.0 Hz), 150.3, 135.05, 134.95, 131.9 (q, JCF3 = 33.6 Hz), 130.9 (d, JCF = 8.8 Hz), 120.9, 122.2 (q, JCF3 = 273.2 Hz), 121.3 (sep, JCF3 = 4.7
Hz), 120.6, 120.5, 119.7, 119.1 (d, 2JCF = 22.4 Hz), 117.0 (d, 4JCF3 = 3.6 Hz), 115.0 (sep, 3JCF3 = 4.0 Hz); 19F NMR (376 MHz, Acetone-d6) δ -59.9, -63.5, -115.9;
IR (film) 3271, 3074, 1670, 1536, 1514, 1465, 1380, 1324, 1283, 1164, 1126, 955, 918, 884, 847, 683, 512 cm-1; HRMS (ESI+) calcd for C22H11F10NO2 [M+H]+,
m/z = 512.0709, found 512.0709.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-fluoro-6-hydroxybenzamide (14j)32 : To a flame-dried flask was added 13 (0.48g, 1.5
mmol), 2-fluoro-6-hydroxybenzoic acid (0.23 g, 1.5 mmol), phosphorous trichloride (0.10 mL, 1.13 mmol), and chlorobenzene (7.5 mL), after which the mix-
ture was heated at reflux for 4 hours. After cooling, the crude product was purified by column chromatography (0-75% CH2Cl2 in hexane) to afford the title
compound as a white solid in 49% yield (0.34 g, 99% pure): mp = 142–144 °C; 1H NMR (400 MHz, CDCl3) δ 12.90 (s, 1 H), 8.78 (d, J = 23.1 Hz, 1 H), 7.70 –
7.64 (m, 2H), 7.59 (s, 1 H), 7.42 – 7.34 (m, 3 H), 7.14 – 7.09 (m, 2 H), 6.87 (dt, J = 8.5, 1.0 Hz, 1 H), 6.68 (ddd, J = 13.2, 8.2, 1.1 Hz, 1 H); 13C NMR (DMSO-
d6, 101 MHz) δ; 166.4 (d, J = 3.8 Hz), 164.0 (d, J = 4.2 Hz), 161.1 (d, 1JCF = 246.2 Hz), 158.7, 152.2, 134.3 (d, 2JCF = 13.9 Hz), 133.6, 133.3 (q, 2JCF3 = 33.7 Hz),
122.9 (q, 1JCF3 = 273.9 Hz), 123.6, 120.6, 117.8 (d, 4JCF3 = 3.7 Hz), 116.4 (sep, 3JCF3 = 3.9 Hz), 115.2 (d, 4JCF = 2.8 Hz), 105.8 (d, 2JCF = 26.4 Hz), 103.6 (d, 3JCF
=
11.8 Hz); 19F NMR (376 MHz, CDCl3) δ -62.98, -111.43; IR (film) 3446, 1659, 1607, 1555, 1506, 1376, 1279, 1234, 1015, 955, 881, 866, 843, 806, 620, 601,
519 cm-1; HRMS (ESI+) calcd for C21H12F7NO3 [M+H]+, m/z = 460.0784, found 460.0789.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-fluoro-4-hydroxybenzamide (14k): The benzyl-protected intermediate was prepared by
general method A with 19 (0.53 g, 1.7 mmol), 13 (0.37 g, 1.5 mmol), HATU (0.63 g, 1.7 mmol), DIPEA (0.5 mL, 2 mmol), and DMF (4 mL), then recrystallized
from hot ethanol to afford pure intermediate in 63% yield (0.52 g, 99% pure). A portion of the intermediate (0.30 g, 0.55 mmol) was added to a flask with 10%
Pd/C (0.016g, 0.15 mmol) and backfilled with nitrogen, after which MeOH (15 mL) and a hydrogen balloon were added. The crude mixture stirred overnight,
then filtered first through Celite, then through a small silica plug to afford the title compound as a tan solid in 99% yield (0.25 g, ): mp = 197–199 °C; 1H NMR
(400 MHz, CDCl3) δ 8.43 (d, J = 16.8 Hz, 1H), 8.11 (t, 1H), 7.76 – 7.67 (m, 2H), 7.56 (s, 1 H), 7.38 (s, 2 H), 7.14 – 7.06 (m, 2 H), 6.78 (dd, J = 8.7, 2.4 Hz, 1
H), 6.69 (dd, J = 13.7, 2.4 Hz, 1 H), 5.45 (s, 1 H); 13C NMR (101 MHz, Acetone-d6) δ 162.2 (d, 1JCF = 248.7 Hz), 162.94, 162.92, 162.8, 162.7, 160.6, 137.6,
133.7 (q, 2JCF3 = 33.6 Hz), 133.3 (d, 4JCF = 4.4 Hz), 124.2 (q, 1JCF3 = 272.0 Hz), 122.9, 118.6 (d, 4JCF = 3.0 Hz), 116.7 (sep, 3JCF3 = 4.0 Hz), 115.5 (d, 2JCF = 13.4
Hz), 113.0 (d, 4JCF = 2.6 H), 103.8 (d, 2JCF = 26.4 Hz); IR (film) 3443, 3178, 1648, 1610, 1543, 1506, 1465, 1409, 1376, 1279, 951, 884, 842, 706, 701, 683, 635,
609, 533, 512 cm-1; HRMS (ESI-) calcd for C21H12F7NO3 [M]-, m/z = 458.0267, found 458.0613.
2-Amino-N-(4-(3,5-bis(trifluoromethyl)phenoxy)phenyl)-6-fluorobenzamide (14l): A vial containing N-{4-[3,5-bis(trifluorome-
thyl)phenoxy]phenyl}-2-fluoro-6-nitrobenzamide (62 mg, 0.13 mmol) and 10% Pd/C (10 mg, 0.1 mmol) was backfilled with nitrogen, after which
MeOH (3 mL) and a hydrogen balloon were added. After stirring overnight, the reaction mixture was filtered through Celite, then concentrated in
vacuo. Following purification by column chromatography (0-100% CH2Cl2 in hexane), the title compound was afforded in 95% yield as a tan soft solid
26
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
(57 mg, 96% pure): 1H NMR (400 MHz, CDCl3) δ 8.40 (d, J = 17.5 Hz, 1 H), 7.73 – 7.63 (m, 2 H), 7.56 (s, 1 H), 7.38 (s, 2 H), 7.16 (td, J = 8.2, 6.5
Hz, 1 H), 7.12 – 7.04 (m, 2 H), 6.51 (dt, J = 8.4, 0.9 Hz, 1H), 6.43 (ddd, J = 13.2, 8.1, 1.1 Hz, 1H), 6.00 (s, 2H); 13C NMR (101 MHz, CDCl3) δ 164.52,
164.50, 162.1 (d, 1JCF = 243.7 Hz), 159.1, 152.1, 152.0, 151.4, 135.0, 133.3 (q, 2JCF3 = 33.8 Hz), 132.9 (d, 3JCF = 9.6 Hz), 123.1, 123.06, (q, 1JCF3 = 273.6
Hz), 120.8, 117.7 (d, 4JCF = 3.9 Hz), 116.2 (sep, 3JCF3 = 3.9 Hz), 113.5 (d, 4JCF = 2.4 Hz), 103.3 (d, 2JCF = 26.6 Hz); 19F NMR (376 MHz, CDCl3) δ -62.9,
-111.7; HRMS (ESI+) calcd for C21H13F7N2O2 [M+H]+, m/z = 459.0944 , found 459.0947.
1
2
3
4
5
6
7
8
9
4-Amino-N-{4-[3,5-bis(trifluoromethyl)phenoxy]phenyl}-2-fluorobenzamide (14m): Prepared by general method A with 0.5 mmol 17, 0.55
mmol 13, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF to obtain trifluoroacetamide protected compound. After removing solvents in vacuo, the crude
trifluoroacetamide was added to K2CO3 (0.14 g, 1 mmol) and MeOH (10 mL) and then heated at reflux for 4 hours. After cooling, the solvent was removed in
vacuo and the resulting residue was partitioned between 5 mL each H2O and EtOAc. The aqueous layer was 3x with 5 mL EtOAc, then combined organic layers
were washed 5 mL H2O, 5 mL brine, and dried over Na2SO4. Crude product was purified by column chromatography (0-30% EtOAc in hexane) to afford title
compound as a tan solid in 39% yield over 2 steps (32 mg, 98% pure). Note: compound is oxidized rapidly on standing. mp = 147–148 °C; 1H NMR (400 MHz,
CDCl3) δ 8.41 (d, J = 17.5 Hz, 1 H), 7.99 (t, J = 9.2, 8.6 Hz, 1 H), 7.77 – 7.67 (m, 2 H), 7.55 (s, 1 H), 7.37 (s, 2 H), 7.11 – 7.02 (m, 2 H), 6.55 (dd, J = 8.6, 2.2
Hz, 1 H), 6.39 (dd, J = 14.7, 2.2 Hz, 1 H), 4.20 (s, 2H); 13C NMR (101 MHz, CDCl3) δ 162.2 (d, 1JCF = 244.1Hz), 161.8 (d, 4JCF = 4.0 Hz), 159.3, 152.1 (d, 3JCF
= 13.0 Hz), 151.0, 135.6, 133.9 (d, 4JCF = 4.0 Hz), 133.3 (q, 2JCF3 = 33.7 Hz), 123.1 (q, 2JCF3 = 272.9 Hz), 122.5, 120.9, 117.6 (d, 4JCF3 = 3.5 Hz), 116.1 (sep, 3JCF3
= 3.8 Hz), 111.4 (d, 4JCF = 1.7 Hz), 110.2 (d, 3JCF = 11.33 Hz), 101.9 (d, 3JCF = 29.2 Hz); 19F NMR (376 MHz, CDCl3) δ -62.97, -111.81; IR (film) 3480, 3458,
3353, 3242, 3089, 3063, 1674, 1640, 1618, 1540, 1514, 1462, 1376, 1279, 1182, 1130, 1108, 951, 884, 858, 843, 706, 687, 590, 568, 534 cm-1; HRMS (ESI+)
calcd for C21H13F7N2O2 [M+H]+, m/z = 459.0944, found 459.0944.
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-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-2-fluoro-6-(trifluoromethoxy)benzamide (14n): Prepared by general procedure A using
0.4 mmol 16, 0.44 mol, 13, 0.44 mmol HATU, 0.8 mmol DIPEA, and 3 mL DMF. Purified by column chromatography (0-60% CH2Cl2 in hexane), then recrys-
tallized from hot MeOH to afford the title compound as white crystals in 21% yield (45 mg, 97% pure): mp = 179–180 °C; 1H NMR (400 MHz, CDCl3) δ 7.71
– 7.66 (m, 2 H), 7.58 (s, 1 H), 7.55 – 7.48 (m, 2 H), 7.40 (s, 2 H), 7.23 – 7.15 (m, 2 H), 7.13 – 7.07 (m, 2 H). 19F NMR (376 MHz, CDCl3) δ -57.41, -62.96, -
111.47. 13C NMR (101 MHz, CDCl3) δ 159.8 (d, 1JCF = 253.3 Hz), 158.7 (d, 2JCF = 16.0 Hz), 151.8, 146.6 (d, 3JCF = 6.2 Hz), 134.2, 133.3 (q, 2JCF3 = 33.9 Hz),
132.0 (d, 3JCF = 9.7 Hz), 122.9 (q, 1JCF3 = 273.9 Hz), 122.4, 121.6, 120.7, 119.5 (d, 2JCF = 20.8 Hz), 118.9 (d, 3JCF = 16.8 Hz), 117.7 (q, 4JCF = 4.1 Hz), 116.9 (d,
4JCF3 = 1.8 Hz), 116.3 (sep, 4JCF3 = 3.8 Hz), 114.9 (d, 2JCF = 21.7 Hz);IR (film) 3264, 3063, 1666, 1622, 1536, 1510, 1469, 1380, 1380, 1264, 1233, 1219, 1167,
1126, 955, 887, 884, 866, 847, 706 cm-1; HRMS (ESI+) calcd for C22H11F10NO3 [M+H]+, m/z = 528.0657, found 528.0663.
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2-ethynyl-6-fluorobenzamide (14o): To a dry THF solution of N-{4-[3,5-bis(trifluorome-
thyl)phenoxy]phenyl}-2-fluoro-6-[2-(trimethylsilyl)ethynyl]benzamide (80 mg, 0.15 mmol) was added a solution of tetrabutylammonium fluoride (TBAF)
(0.20 mL 1 M TBAF in THF, 0.20 mmol) dropwise at 25 °C under N2. The mixture was stirred at rt and monitored by TLC. After 18 h, the mixture was
concentrated by rotary evaporation and dissolved with CH2Cl2 and washed with H2O. The organic mixture was dried with Na2SO4, filtered, and concentrated
by rotary evaporation to afford a white solid. The solid was purified by flash column chromatography (15% EtOAc:Hexanes) to afford a colorless oil (3 mg, 97%
pure) in 4% yield: 1H NMR (CDCl3, 400 MHz) δ ; 7.60 – 7.53 (m, 2 H), 7.59–7.57 (m, 1 H), 7.50 (m, 2), 7.44 (d, J = 9.0 Hz, 2 H), 7.25–7.22 (m, 1 H), 7.19
(d, J = 8.9 Hz, 2 H), 4.89 (d, J = 1.7 Hz, 1 H); 13C NMR (CDCl3, 101 MHz) δ 158.8 (d, 1JCF = 263.0, 157.9, 155.0; 142.3 (d, 4JCF = 2.0 Hz), 138.7 (d, 3JCF = 3.2
Hz), 134.4 (d, 3JCF = 7.7 Hz), 133.5 (q, 2JCF3 = 34.1 Hz) 130.7, 130.1, 120.1, 121.5, 120.7, 118.8 (d, 4JCF3 = 3.8 Hz), 117.7 (d, 2JCF = 19.1 Hz), 117.01, (sep 3JCF3
=
2.9 Hz)116.1 (d, 3JCF = 4.1 Hz), 91.5 (unresolved quartet); 19F NMR (376 MHz, CDCl3) δ -63.0, -116.6; 19F NMR (376 MHz, CDCl3) δ -63.0 ; IR (film) 3335,
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
3275, 3115, 3078, 1696, 1648, 1614, 1599, 1514, 1491, 1465, 1451, 1286, 1242, 1175, 1130, 1026, 858, 814, 784, 706, 687 cm-1; HRMS (ESI+) calcd for
23H12F7NO2 [M+H]+, m/z = 468.0835, found 468.0831.
1
2
C
3
4
5
6
7
8
9
N-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)cyclohexanecarboxamide (15a): To a solution of 4-[3,5-bis(trifluoromethyl)phenoxy]ani-
line (0.31 mmol) and Et3N (0.37 mmol) in CH2Cl2 (0.3 M) was added cyclohexanecarbonyl chloride (0.37 mmol) at 0 °C .The reaction was warmed up to and
stirred overnight at rt. A solid precipitated from the solution. The reaction mixture was diluted with 5.0 mL of CH2Cl2 and washed with 10 mL H2O. The mixture
was extracted CH2Cl2 3x with 5.0 mL. The organic mixture was then dried with Na2SO4, filtered and concentrated by rotary evaporation to afford a white solid.
The solid was purified by flash column chromatography (15% EtOAc:Hexanes) to afford a white solid (53 mg, 99% pure) in 33% yield: mp = 178–179 °C; 1H
NMR (CDCl3, 400 MHz) δ 7.59 (d, J = 8.9 Hz, 2 H), 7.56 – 7.50 (m, 1 H), 7.39 – 7.30 (m, 2 H), 7.19 (s, 1 H), 7.02 (d, J = 9.3 Hz, 2 H), 2.25 (tt, J = 11.8, 3.5
Hz, 1 H), 1.96 (d, J = 3.5 Hz, 2 H), 1.86 (dd, J = 10.5, 4.9 Hz, 2 H), 1.79 – 1.67 (m, 1 H), 1.64 – 1.49 (m, 4 H), 1.45 – 1.17 (m, 4 H); 13C NMR (CDCl3, 101
MHz) δ 174.4, 159.1, 150.7, 135.4, 133.1 (q, 2JCF3 = 33.8 Hz), 122.9 (q, 4JCF3 = 274.2 Hz), 121.7 120.8, 117.4 (d, 4JCF3 = 3.2 Hz), 116.0 (sep, 3JCF3 = 4.0 Hz), 46.5,
29.7, 25.6; 19F NMR (376 MHz, CDCl3) δ -63.0; IR (film) 3294, 2936, 1666, 1607, 1529, 1510, 1462, 1376, 1286, 1238, 1171, 1134, 1108, 955, 881, 862, 843,
706, 683, 512 cm-1; HRMS (ESI+) calcd for C20H18F6N2O2 [M+H]+, m/z = 433.1346, found 433.1350.
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-(4-(3,5-Bis(trifluoromethyl)phenoxy)phenyl)-2,3,4,5,6-pentafluorobenzamide (15b): Prepared by general procedure B using 0.39 mmol
13, 1.2 mmol 2-fluoro-6-(trifluoromethyl)benzoyl chloride, 1.2 mmol Et3N, and 4 mL CH2Cl2. Purified by column chromatography (5% EtOAc in hexane) to
1
afford the title compound as a white solid in quantitative yield (0.200 g, 99% pure): mp = 86–88 °C; H NMR (CDCl3, 400 MHz) δ 7.66 (tt, J = 1.5, 0.8 Hz,
1H), 7.37 – 7.36 (m, 2H), 7.35 (d, J = 8.8 Hz, 2 H), 7.07 (d, J = 8.9 Hz, 2 H); 13C NMR (CDCl3, 101 MHz) δ 159.9, 157.1, 156.9, 144.3 (m, JCF), 141.7 (m, JCF),
138.9 (m, JCF), 136.3 (m JCF);133.7 (q, 2JCF3 = 34.0 Hz), 131.7, 130.2, 122.7 (1, 1JCF3 = 278.9 Hz), 120.3, 119.0 (q, 3JCF3 = 3.8 Hz), 117.7 (sep, 3JCF3 = 3.9 Hz) 19
F
3
NMR (376 MHz, CDCl3) δ -63.19, -140.02 (m), -148.28 (t, JFF-o = 21.4 Hz), -159.17– -159.29 (m); 19F NMR (376 MHz, CDCl3) δ -63.19, -140.02 (m), -
148.28 (t, 3JFF-o = 21.4 Hz), -159.17– -159.29 (m); IR (film) 1722, 1659, 1599, 1506, 1372, 1327, 1283, 1234, 1175, 1160, 1141, 1115, 1000, 951, 895, 881, 773,
706, 687, 523 cm-1; HRMS (ESI+) calcd for C21H9F11NO2 [M+H]+, m/z = 516.0453 , found 516.0441.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-3-fluoropyridine-2-carboxamide (15c) Prepared by general procedure A with 0.55 mmol
13, 0.5 mmol 3-fluoropyridine-2-carboxylic acid, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. Purified by column chromatography (0-100% CH2Cl2 in
hexane) followed by trituration in cold hexane to afford the title compound as a white-tan solid in 82% yield (0.18 g, 99% pure): mp = 150–151 °C; 1H NMR
(400 MHz, CDCl3) δ 9.93 (s, 1H), 8.47 (s, 1 H), 7.87 – 7.82 (m, 2 H), 7.66 – 7.59 (m, 1 H), 7.59 – 7.54 (m, 2 H), 7.38 (s, 2 H), 7.13 – 7.08 (m, 2 H); 13C NMR
(101 MHz, CDCl3) δ 160.1, 160.0, 159.2, 151.2, 143.9 (d, 3JCF = 5.6 Hz),137.3, 135.1, 133.3 (q, 2JCF3 = 33.8 Hz), 128.7 (d, 3JCF = 5.4 Hz), 127.0 (d, 2JCF = 19.7
4
3
Hz), 121.7, 120.9, 117.7 (d, JCF = 2.9 Hz), 116.2 (sep, JCF3 = 3.8 Hz) (unresolved C-CF3 quartet);19F NMR (376 MHz, CDCl3) δ -62.99, -118.07; IR (film)
3335, 3078, 1700, 1618, 1532, 1506, 1458, 1376, 1279, 1223, 1175, 1149, 1130, 1108, 1093, 1015, 951, 903, 884, 858, 843, 825, 802, 780, 739, 702, 624, 594,
571 cm-1; HRMS (ES-) [M-H]- for C20H10F7N2O2 , m/z 443.0630, found 443.0633.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}furan-2-carboxamide (15d): Prepared by general method B using 4-[3,5-bis(trifluorome-
thyl)phenoxy]aniline 0.31 mmol , 0.37 mmol 2-furyl chloride, 0.62 mmol Et3N and 1.2 mL CH2Cl2. The solid was purified by flash column chromatography
(15% EtOAc:Hexanes) to afford the title compound as a white solid in 38% yield (25 mg, 99% pure): mp = 136–137 °C, 1H NMR (400 MHz, CDCl3) δ 8.14
(s, 1H), 7.77 – 7.69 (m, 2 H), 7.60 – 7.50 (m, 2 H), 7.38 (dd, J = 1.5, 0.8 Hz, 2 H), 7.30 – 7.24 (m, 1 H), 7.12 – 7.04 (m, 2 H), 6.58 (dd, J = 3.5, 1.8 Hz, 1 H);
2
1
4
13C NMR (101 MHz, CDCl3) δ 159.0, 156.1, 151.2, 147.6, 144.3, 134.7, 133.0 (q, JCF3 = 34.1 Hz), 121.6 (q, JCF3 = 273.9 Hz), 121.9, 120.7, 117.5 JCF3 =4.1
28
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
z=Hz), 11;6.1 (4JCF3 = 4.3 Hz), 115.5, 112.7; 19F NMR (376 MHz, CDCl3) δ -63.0; (film) 3342, 1685, 1659, 1610, 1536, 1510, 1465, 1376, 1279, 1231, 1178,
1134, 1015, 955, 884, 858, 758, 706, 687 cm-1; HRMS (ES+) [M-H]- for C19H11F6NO3, m/z 438.0541, found 438.0532.
1
2
3
4
5
6
7
8
9
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-3-chlorothiophene-2-carboxamide (15e): Prepared by general procedure A with 0.5 mmol
13, 0.5 mmol 3-chlorothiphene-2-carboxylic acid, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. Following trituration in ice-cold hexane the title com-
pound was isolated as a pink solid in 87% yield (0.20 g, 96% pure): mp = 130–132 °C; 1H NMR (400 MHz, CDCl3) δ 8.81 (s, 1H), 7.74 – 7.69 (m, 2H), 7.56
(d, J = 5.3 Hz, 2H), 7.38 (s, 2H), 7.12 – 7.07 (m, 2H), 7.06 (d, J = 5.3 Hz, 1H); 13C NMR (101 MHz, CDCl3) δ 159.1, 158.4, 151.6, 134.6, 133.3, 133.0 (q, 2JCF3
= 42.5 Hz), 130.5, 129.7, 123.2, 123.0 (q, 1JCF3 = 272.9 Hz), 122.5, 120.9, 117.7 (d, 3JCF3 = 3.0 Hz), 116.3 (sep, 3JCF3 = 3.9 Hz); 19F NMR (376 MHz, CDCl3) δ -
62.99; IR (film) 3342, 1640, 1607, 1536, 1510, 1465, 1428, 1372, 1283, 1190, 1167, 1130, 955, 918, 884, 858, 843, 825, 724, 706, 683, 594, 516, 516 cm-1;
HRMS (ES-) [M-H]- calcd for C19H9ClF6NO2S , m/z 463.9947, found 463.9953.
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-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-1H-indole-5-carboxamide (15f): Prepared by general procedure A using 0.55 mmol 13, 0.5
mmol indole-5-carboxylic acid, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. Trituration in ice-cold hexane followed by column chromatography (0-
100% EtOAc in hexane) afforded the title compound as a white solid in 67% yield (0.17 g, 97% pure): mp = 171–173 °C; 1H NMR (400 MHz, Methanol-d4) δ
8.27 (dd, J = 1.8, 0.7 Hz, 1 H), 7.85 – 7.80 (m, 2 H), 7.75 (dd, J = 8.6, 1.8 Hz, 1 H), 7.66 (s, 1H), 7.52 – 7.46 (m, 3 H), 7.35 (d, J = 3.2 Hz, 1H), 7.19 – 7.13 (m,
2H), 6.60 (dd, J = 3.2, 0.9 Hz, 1 H); 13C NMR (101 MHz, MeOD) δ 170.4, 161.1, 152.3, 139.7, 137.9, 134.3 (q, 2JCF3 = 33.6 Hz), 129.0, 127.4, 126.6, 124.48,
124.45 (1JCF3 = 273.1 Hz), 121.9, 121.8, 121.7, 118.6 (d, 4JCF3 = 3.9 Hz), 116.8 (sep, 3JCF3 = 4.0 Hz), 121.1, 103.7; 19F NMR (376 MHz, Methanol-d4) δ -64.57;
IR (film) 2478, 2452, 2419, 2393, 1618, 1592, 1570, 1510, 1469, 1432, 1402, 1372, 1324, 1286, 1253, 1193, 1164, 1126, 1018, 1003, 951, 895, 858, 825, 784,
754, 732, 706, 687, 586, 545, 523 cm-1; HRMS (ESI-) calcd for C23H14F6N2O2 [M+H]+, m/z = 463.0881, found 463.0885.
N-{4-[3,5-Bis(trifluoromethyl)phenoxy]phenyl}-1H-benzimidazole-5-carboxamide (15g): As general procedure A with 0.55 mmol
13, 0.5 mmol 5-benzimidizole carboxylic acid, 0.55 mmol HATU, 1 mmol DIPEA, and 4 mL DMF. Purified by column chromatography (0-100%
EtOAc in CH2Cl2) followed by recrystallization from MeOH to give an iridescent white solid in 28% yield (0.066g, 92% pure): mp = 113–115 °C; 1H
NMR (400 MHz, DMSO-d6) δ 12.77 (s, 1 H), 10.37 (s, 1H), 8.39 (s, 1 H), 8.26 (s, 1H), 7.93 (d, J = 8.9 Hz, 2 H), 7.85 (d, J = 13.2 Hz, 2 H), 7.69 (s, 1
H), 7.59 (s, 2 H), 7.26 – 7.20 (m, 2 H); 13C NMR (101 MHz, Acetone-d6) δ 165.0, 158.8, 142.9, 142.2, 131.8 (q, 2JCF3 = 33.5 Hz), 127.8, 122.3 (q, 1JCF3
= 267.6 Hz), 121.6, 121.0, 121.0, 119.8, 119.6, 118.1, 116.6 (q, 4JCF3 = 4.0 Hz), 114.7 (m, 3JCF3 = 3.7 Hz), 110.8, 110.4; 19F NMR (376 MHz, DMSO-
d6) δ -61.47; IR 1648, 1607, 1532, 1506, 1231, 1175, 1126, 955, 884, 862, 836, 706, 683 (film) cm-1 HRMS (ESI+) calcd for C22H13F6N3O2 [M+H]+,
m/z = 466.0990, found 466.1022.
2-Fluoro-6-(trifluoromethoxy)benzoic acid (16)33: To a flame-dried flask containing THF (5 mL) was added lithium diisopropylamide (2M in THF;
0.6 mL, 1.2 mmol) under N2. After cooling to -78 °C, 3-(trifluoromethyl)fluorobenzene (0.18 g, 1. mmol) was added, and the reaction stirred 2 hours at -78 °C.
CO2 from a flask containing dry ice was bubbled through the solution as it was allowed to warm to rt over 1 hour, then quenched by addition of H2O. The
mixture was concentrated in vacuo, then dissolved in 10 mL 1M NaOH and washed twice with 10 mL EtOAc. Following acidification with 3M HCl, the aqueous
layer was extracted 3x with 10 mL EtOAc and the combined organic extracts were dried over Na2SO4 and concentrated to dryness to afford the title compound
as a white solid (0.126 g, 56%) which was used without further purification: mp = 92–95 °C; 1H NMR (400 MHz, DMSO-d6) δ 14.14 (s, 1 H), 7.66 (td, J = 8.5,
6.5 Hz, 1 H), 7.43 (ddd, J = 9.3, 8.5, 0.9 Hz, 1 H), 7.39 – 7.34 (m, 1 H); 13C NMR (101 MHz, Acetone-d6) δ 162.7, 161.8, 159.3, 147.1(dd, 4JCF = 2.2, 3.3 Hz),
133.5 (d, 3JCF = 9.9 Hz), 121.3 (q, 1JCF3 = 257.7 Hz), 118.4 (d, 2JCF = 20.5 Hz), 118.0, (dd, 4JCF = 1.1, 1.4 Hz), 117.4, 116.0 (d, 3JCF = 21.6 Hz);’ 19F NMR (376
29
ACS Paragon Plus Environment