Phenotypic Optimization of Urea-Thiophene Carboxamides to Yield Potent, Well Tolerated, and Orally Active Protective Agents against Aminoglycoside-Induced Hearing Loss

Article abstract of DOI:10.1021/acs.jmedchem.7b00932

Hearing loss is a major public health concern with no pharmaceutical intervention for hearing protection or restoration. Using zebrafish neuromast hair cells, a robust model for mammalian auditory and vestibular hair cells, we identified a urea-thiophene carboxamide, 1 (ORC-001), as protective against aminoglycoside antibiotic (AGA)-induced hair cell death. The 50% protection (HC₅₀) concentration conferred by 1 is 3.2 μM with protection against 200 μM neomycin approaching 100%. Compound 1 was sufficiently safe and drug-like to validate otoprotection in an in vivo rat hearing loss model. We explored the structure-activity relationship (SAR) of this compound series to improve otoprotective potency, improve pharmacokinetic properties and eliminate off-target activity. We present the optimization of 1 to yield 90 (ORC-13661). Compound 90 protects mechanosensory hair cells with HC₅₀ of 120 nM and demonstrates 100% protection in the zebrafish assay and superior physiochemical, pharmacokinetic, and toxicologic properties, as well as complete in vivo protection in rats.

Full text of DOI:10.1021/acs.jmedchem.7b00932

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Article
Phenotypic Optimization of Urea-Thiophene Carboxamides
to Yield Potent, Well Tolerated and Orally Active Protective
Agents Against Aminoglycoside-Induced Hearing Loss
Sarwat Chowdhury, Kelly N. Owens, R. Jason Herr, Qin Jiang, Xinchao Chen, Graham
Johnson, Vincent E. Groppi, David W. Raible, Edwin W Rubel, and Julian A. Simon
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00932 • Publication Date (Web): 09 Oct 2017
Downloaded from http://pubs.acs.org on October 9, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Phenotypic Optimization of UreaꢀThiophene Carboxamides
to Yield Potent, Well Tolerated and Orally Active
Protective Agents Against AminoglycosideꢀInduced
Hearing Loss
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
2
3
3
3
Sarwat Chowdhury, Kelly N. Owens, R. Jason Herr, Qin Jiang, Xinchao Chen, Graham
4
,5
6
2,7
2
*,1,8,9
Johnson, Vincent E. Groppi, David W. Raible, Edwin W Rubel and Julian A. Simon
1
8
9
Clinical Research, Human Biology and, Public Health Sciences Divisions, Fred Hutchinson
Cancer Research Center, Seattle, Washington 98109, United States
2
Virginia Merrill Bloedel Hearing Research Center, University of Washington, Seattle,
Washington 98195, United States
3
AMRI, Albany, New York 12203, United States
4
NuPharmAdvise LLC, Sanbornton, New Hampshire 03269 United States
5
Oricula Therapeutics, Seattle Washington, 98154, United States
6
Center for Discovery of New Molecules, University of Michigan, Ann Arbor, Michigan 48109,
United States
7
Department of Biological Structure, University of Washington, Seattle, Washington 98195,
United States
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
Hearing loss is a major public health concern with no pharmaceutical intervention for hearing
protection or restoration. Using zebrafish neuromast hair cells—a robust model for mammalian
auditory and vestibular hair cells— we identified a ureaꢀthiophene carboxamide, 1 (ORCꢀ001) as
protective against aminoglycoside antibiotic (AGA)ꢀinduced hair cell death. The 50% protection
(HC ) concentration conferred by 1 is 3.2 µM with protection against 200µM neomycin
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
approaching 100%. 1 was sufficiently safe and drugꢀlike to validate otoprotection in an in vivo
rat hearing loss model. We explored the structure activity relationship (SAR) of this compound
series to improve otoptrotective potency, improve pharmacokinetic properties and eliminate offꢀ
target activity. We present the optimization of 1 to yield 90 (ORCꢀ13661). 90 protects
mechanosensory hair cells with HC of 120 nM and demonstrates 100% protection in the
50
zebrafish assay, superior physiochemical and pharmacokinetic and toxicologic properties as well
as complete in vivo protection in rats.
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Acquired hearing loss can be caused by a variety of environmental and organic factors ranging
from noise exposure and pharmaceutical agents to infections, late onset genetic disorders and
1, 2
aging. Hair cells of the inner ear are the mechanosensors that transduce a physical stimulus ꢀ
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
vibrations transmitted by the fluidꢀfilled compartment of the cochlea and vestibule ꢀ into nerve
impulses that are converted in the central nervous system into the perception of sound and
vestibular reflexes. Most forms of hearing loss are sensorineural in nature, meaning they result
from death or injury of the cochlear hair cells or other cell types in the organ of Corti of the
3
cochlea or the neurons connecting the hair cells to the brain. Unlike aquatic vertebrates and
4, 5
birds, mammals lack the ability to regenerate hair cells making hearing loss irreversible. The
mechanisms by which hair cells are killed are complex and overlapping. Elements of
programmed cell death, or apoptosis, as well as necrotic signatures have been identified in
5
ꢀ8
damaged hair cells in vitro and in vivo. Antioxidants including aspirin, Nꢀacetyl cysteine
9
(NAC) and selenium compounds have been reported to offer some degree of protection in model
3
, 10
11ꢀ13
organisms,
but have yielded disappointing results in human clinical trials.
Mechanistically, stress pathways, including the cꢀJun Nꢀterminal kinase (JNK), mixed lineage
kinase (MLK) and p38 kinase pathways have been implicated in drugꢀinduced ototoxicity and
8
, 14ꢀ17
small molecule inhibitors of these pathways have been proposed as protective agents.
However, clinical trials have again been disappointing, possibly because the protection is not
1
8
robust and falls off at higher doses of the ototoxic agent. These clinical findings indicate a need
for better therapeutic agents effective at higher doses of ototoxic agent.
Aminoglycoside antibiotics (AGAs) constitute a class of natural product antibiotics that are
1
1
effective against a variety of clinically important gramꢀnegative bacteria. However, because of
side effects that include irreversible hearing and balance problems (ototoxicity) and kidney
damage (nephrotoxicity), their use is limited in North America and Western Europe to combating
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
9ꢀ21
lifeꢀthreatening infections.
Exceptions are the use of systemic AGAs for controlling flareꢀups
of chronic Pseudomonas aeruginosa infections in cystic fibrosis patients and multiꢀdrug resistant
22ꢀ25
tuberculosis.
AGAs exert their antimicrobial effect partly by binding to ribosomal RNA and
26ꢀ29
altering or inhibiting ribosomal protein synthesis.
A more controversial model of toxicity in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bacterial and mammalian cells through direct induction of reactive oxygen species (ROS) has
3
0ꢀ32
been proposed but remains hotly contested.
Additionally, proteins that bind AGAs have been
3
3
identified and proposed as relevant mediators of in vivo ototoxicity.
In an unbiased approach, we used the lateral line neuromast hair cells in freeꢀswimming larval
zebrafish (Danio rerio) to screen for compounds that protect these mechanosensory hair cells
34
against AGAꢀinduced cell death. Like mammalian auditory and vestibular hair cells, lateral line
hair cells transduce a mechanical stimulus of changing liquid pressure to nerve impulses and like
their auditory counterparts, are exquisitely sensitive to damage or death due to AGA exposure.
Neuromast hair cell toxicity can be used as a highly reliable and versatile screening assay to
3
4
identify modifiers, either enhancers or suppressors, of ototoxic agents. The assay is equally
effective in identifying genetic and pharmacological modifiers. In a screen of structurally
diverse, small molecule, drugꢀlike compounds, we identified a thiopheneꢀureaꢀcarboxamide 1
20
(
ORCꢀ001, PROTOꢀ1, Figure 1), as an agent that provided doseꢀdependent protective activity
(
otoprotection) against a range of AGA concentrations. Further experiments revealed that 1 was
sufficiently drugꢀlike to allow proofꢀofꢀconcept testing in a rat model of AGAꢀinduced hearing
loss (see below). However, 1 exhibited pharmacokinetic and toxicological liabilities including
only adequate solubility, moderate oral bioavailability, short in vivo halfꢀlife and moderate
inhibition of the human EtherꢀàꢀgoꢀgoꢀRelated Gene (hERG, aka K 11.1) potassium ion channel.
V
Inhibition of hERG by a diverse array of pharmaceuticals has been linked to cardiac arrhythmias
35ꢀ38
through QT interval prolongation.
Because of this, hERG inhibition is generally viewed as a
liability in drug development. Finally, as a hit in a phenotypic screen, 1 was uncharacterized in
ACS Paragon Plus Environment

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
9, 40
terms of its molecular target or mechanism of action.
To overcome these limitations, we
carried out a systematic evaluation of analogs using a phenotypic hair cell protection assay in
zebrafish to generate a structure activity relationship for compounds that protect against
aminoglycoside antibioticꢀinduced hearing loss. The results of this evaluation and validation of
hearing protection in mammals by one of the final lead compounds are the subjects of the current
manuscript.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(Figure 1, here)
Results
Synthesis of 1 and 3-aryl urea derivatives. In a phenotypic screen using zebrafish neuromast
hair cells we identified 1 as highly protective against aminoglycoside antibioticꢀinduced hair cell
death. Early ADME characterization revealed several promising properties as a lead molecule
(adequate aqueous solubility and membrane permeability, reasonable plasma protein binding, no
issues with CYP stability or inhibition and an acceptable clearance profile), although we
identified several pharmacokinetic limitations and a hERG liability (hERG IC 3.26 µM).
5
0
However, 1 was sufficiently safe and drugꢀlike to be used to validate otoprotection in an in vivo
rat model of aminoglycosideꢀinduced hearing loss as a measure of proof of concept (vide infra).
We then explored the structure activity relationships of this compound series with three goals in
mind: improve otoprotective potency, improve pharmacokinetic properties and
eliminate/minimize offꢀtarget activity.
41, 42
The preparation of 1 and N3’ꢀaryl derivatives began with the Gewald condensation
of 4ꢀ
ethylꢀ4ꢀpiperidone with cyanoacetamide in the presence of elemental sulfur and morpholine to
produce 2ꢀaminoꢀ4,5,6,7ꢀtetrahydrothieno[2,3ꢀc]pyridineꢀ3ꢀcarboxamide 28 as shown in Scheme
1. Subsequent reaction of 28 with aryl isocyanates produced the urea derivatives 1-15 in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
moderate to good yields. Formation of HCl salts stabilized compounds containing a basic
nitrogen in the tetrahydropyridine ring and related structures, which otherwise slowly underwent
aereal oxidation to generate fluorescent conjugated products that interfered in the primary
bioassay. In all, over 400 analogues including amines and quaternary ammonium salts were
prepared and evaluated in the zebrafish neuromast hair cell protection assay.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
Scheme 1, here)
In vivo screen of 1 analogs in zebrafish. New compounds were tested in the zebrafish assay as
34, 43
previously described.
In summary, newly hatched freeꢀswimming AB zebrafish larvae were
o
raised at 28.5 C in petri dishes and transferred to cell culture baskets placed in 6ꢀwell culture
plates in groups of ten fish per basket. Fish were preꢀtreated with test compound for 1 hour at
five threeꢀfold dilutions followed by treatment with 200 µM neomycin. Following neomycin
exposure, fish were rinsed briefly with a staining agent (DASPEI), anesthetized and mounted on
the stage of an epifluorescence dissecting microscope. Hair cell staining of ten neuromasts on
one side of each animal was evaluated visually each neuromast was scored for presence of a
normal compliment of hair cells, with reduced or absent DASPEI staining indicating a reduction
in hair cell number. Composite scores were calculated for animals in each treatment group,
normalized to the control group and expressed as % hair cell survival. If hair cell survival was at
least 50%, the HC was calculated. These data are shown in Tables 1ꢀ8 and representative dose
50
response relationship experiments for several analogues in the zebrafish hair cell protection assay
are shown in Figures 2 and 3. All zebrafish protocols were approved by the University of
Washington IACUC.
(Figures 2 and 3, here)
ACS Paragon Plus Environment

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure Activity Relationship (SAR). For the purposes of the determination of structure
activity relationships, 1 was conceptualized as composed of six structural units: 1) N3ꢀaryl
substituents on the urea; 2) the 1,3ꢀdiꢀsubstituted urea; 3) thiophene core; 4) primary carboxamide
C2 substitution; 5) tetrahydropyridine core, and; 6) Nꢀethyl group. The goal of the SARꢀ
determining studies described here was threeꢀfold: first, to determine the contribution of each
structural element described above to the observed biological activity; second, to improve the
protective activity of the compound series, and; third, to improve the pharmacokinetic and
toxicological properties of the lead compounds. A systematic variation of each structural element
and its effect on protective activity is described.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The examination of the effects of N3ꢀsubstitution to replace the pꢀchlorophenyl group relied on
the reaction of readily synthesized isocyanates or their equivalents with 2ꢀaminoꢀthiopheneꢀ3ꢀ
carboxamide 28. A variety of aromatic and nonꢀaromatic isocyanates gave rise to a panel of
substituted ureas as shown in Table 1. The pattern of activity was indicative of a narrow range of
allowable substituents with pꢀchlorophenyl being close to optimal. Several aryl ureas had
comparable protective activity to 1 (HC = 3.12 µM), such as the classic isosteric replacements
5
0
at the C4 position (2, 3, 7 and 8, HC = 1.6ꢀ3.4 µM range). Interestingly, the unsubstituted
50
analog 14 was inactive, and the bioisosteric 3,4ꢀdisubstituted pair 5 and 6 differed in potency by
0ꢀfold (HC = 1.78 for 6 versus 17.4 µM for 5), reinforcing the narrow requirement for
1
50
allowable substitution. The ‘extended’ benzyl analogues represented by 11 were inactive, as
were Nꢀalkyl derivatives, represented by 12 and 13, clearly lying out of the tolerable substitution
range. Heterocyclic derivative 15 was similarly inactive, as were the majority of the 2ꢀsubstituted
aromatic analogues (data not shown).
(Table 1, here)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As shown in Table 2, replacement of the primary carboxamide with a carboxylic acid 16, methyl
ester 17, Nꢀmethyl amide 18, carbonitrile 19 and tertiary amide 20 greatly reduces the activity,
suggesting that both protons on the primary amide are required for activity. Similarly,
carboxamide bioisosteric compounds 1,2,5ꢀoxadiazole 21 and 1,3,4ꢀoxadiazole 22 are inactive.
Only the 2ꢀhydroxyethyl carboxamide 24 and 2ꢀ(Nꢀmorpholino)ethyl carboxamide 25 retain
modest activity (both HC = 9.4 µM), suggesting the possibility that an internal hydrogen bond
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
with the proximal urea proton may provide a preꢀorganization of molecular structure which
contributes to compound potency.
(
Table 2, here)
As shown in Table 3, efforts to replace the diꢀsubstituted urea moiety were largely unsuccessful,
although the diꢀsubstituted thiourea 26 did provide modest potency, albeit twoꢀfold less (HC =
50
6.57 versus. 3.12 µM). Replacement of urea with several types of amide structures resulted in no
protective activity in 27, 29ꢀ30 or 32ꢀ33, and unsurprisingly the core compound 28 without a diꢀ
substituted urea functional group was inactive. Cyclopropyl malonamide 31 displayed a lack of
potency, and likewise substitution at either the urea N1 or N3 position was also not tolerated (data
not shown). Replacement of the diꢀsubstituted urea with a peptide bond or as a squarate diꢀamide
also completely abolished protective activity (data not shown).
(
Table 3, here)
The thiophene portion of the core fused bicyclic ring is also important for activity (Table 4).
Several 5ꢀ and 6ꢀmembered ring thiophene variants were prepared from commercial building
blocks to give 34-37, none of which provided protective activity. The bioisosteric
tetrahydroisoquinoline 38 was also prepared, but likewise displayed no potency in the bioassay.
ACS Paragon Plus Environment

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(Table 4, here)
The variation of the parent tetrahydrothieno[2,3ꢀc]pyridine ring proved feasible without
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
eliminating biological activity, with protective activity remaining in the 1ꢀ5 µM range (Table 5).
Movement of the nitrogen to produce tetrahydrothieno[3,2ꢀc]pyridine 41 was tolerated (HC =
50
4.61 µM), as was ringꢀcontraction to the dihydrothieno[2,3ꢀc]pyrrole 42 (HC = 4.34 µM) and
5
0
ringꢀexpansion to the tetrahydrothienoazepines 43 and 44 (HC = 3.98 and 5.80 µM,
5
0
respectively). However, replacement of the tetrahydropyridine ring with the carbocyclic
cyclohexanes 46ꢀ48 completely abolished protective activity. Breaking up the fused bicyclic core
to provide the tertiary aminomethyl tetraꢀsubstituted thiophene 45, provided a compound of
moderate but measurable activity (HC = 6.92 µM).
50
(
Table 5, here)
Interestingly, removal of the Nꢀethyl substituent as in 40 had only a modest effect on activity
(
HC = 3.26 µM for 40 versus 3.12 µM for 1), leading to an exploration of the effect of Nꢀ
50
substitution on potency, as shown in Table 6. A series of Nꢀalkyl hydrocarbon analogues 49ꢀ58
were prepared to provide a simple exploration of changes in length and steric bulk. While the
increase in bulk at the terminal end of the linear case reduced activity (Nꢀethyl 1, HC = 3.12 µM,
5
0
to Nꢀisobutyl 50, HC = 4.51 µM, to Nꢀneopentyl 51, HC = 5.56 µM, to Nꢀisopropyl 52, HC
50
50
50
=
1.49 µM), a few alkyl substituents bearing terminal cyclic motifs led to improved additional
improvements in protective activity [Nꢀ(cyclopropyl)ethyl 53, HC = 1.22 µM, to Nꢀ
5
0
(
cyclobutyl)ethyl 54, HC = 1.36 µM, to Nꢀ(cyclopentyl)ethyl 55, HC = 1.83 µM, to Nꢀ
50 50
(cyclohexyl)methyl 56, HC = 1.89 µM, versus 3.12 µM for 1]. Unfortunately, with along with
50
increases in hearing protection potency, a number of compounds in this same set also exhibited
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
increases in hERG activity as determined by the patch clamp assay (hERG IC = 2.67 µM for 53
50
and 0.73 µM for 54 versus 3.26 µM for 1). To test the effect amine basicity on potency,
carbamate 59 and aminoethyl ester 60 were prepared, with the result that protective activity was
completely suppressed. Interestingly, as the hydrocarbon chain connecting the trialkylamine to
the ester functionality was increased to two carbons to give 61, potency was increased (HC =
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
1
.10 µM for 61 versus 3.12 µM for 1). When the length was increased between three to five
methylene units in 62-64, the potency was improved to less than a micromolar activity (HC =
50
0.6ꢀ0.7 µM range). Unlike the hydrocarbon substituent series, however, the increases in potency
were not necessarily paralleled by their hERG inhibition, (hERG IC = 3.84 µM for 62 versus
5
0
3
.26 µM for 1). Compounds incorporating an amide functionality distal to the tetrahydropyridine
nitrogen as in 65-67 also retained improved protective activity, providing some of the most potent
compounds prepared in the Nꢀalkyl SAR series (HC = 0.1ꢀ0.5 µM range).
5
0
(Table 6, here)
In order to more fully explore the requirement for a basic nitrogen in the tetrahydropyridine ring,
we synthesized a panel of N,Nꢀdialkyl quaternary ammonium salts (Table 7). This provides a
permanent positive charge within the ring, whereas the previous compounds were assumed to be
protonated at physiological pH. The initial Nꢀmethyl analogue 68 was found to be slightly more
potent than the parent compound (HC = 1.96 µM for 68 versus 3.12 µM for 1), but interestingly
5
0
it exhibited the same level of hERG inhibition (hERG IC = 3.38 µM for 68 versus 3.26 µM for
5
0
1
). We were initially pleased to find that increasing steric bulk around the nitrogen yielded
compounds with increasing activity (Nꢀethyl 68, HC = 1.96 µM, to Nꢀisobutyl 71, HC = 0.38
5
0
50
µM, to Nꢀneopentyl 72, HC = 0.29 µM), although a limit was reached with Nꢀisopropyl
50
compound 70 (HC = 0.96 µM). Similarly, increasing the size of a cyclic moiety on the terminal
50
end of the Nꢀsubstituent led to increasingly more potent compounds [Nꢀ(cyclopropyl)ethyl 73,
ACS Paragon Plus Environment

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HC = 0.36 µM, to Nꢀ(cyclobutyl)ethyl 74, HC = 0.21 µM, to Nꢀ(cyclopentyl)ethyl 75, HC =
50
50
50
0.20 µM, to Nꢀ(cyclohexyl)methyl 76, HC = 0.10 µM]. It was therefore a pleasant discovery
5
0
that the hERG activity decreased inversely from that observed in the Nꢀalkyl series. For instance,
interaction with the hERG channel decreased as the zebrafish potency increased for the size
increase from the Nꢀethyl compound 68 (HC = 1.96 µM, hERG IC = 3.38 µM) to Nꢀisobutyl
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
5
0
50
compound 71 (HC = 0.38 µM, hERG IC = 9.75 µM) to Nꢀ(cyclohexyl)methyl compound 76
50
50
(
HC = 0.10 µM, hERG IC > 30 µM). As with the basic Nꢀsubstituted variants in Table 6,
50 50
quaternary ammonium salts containing methyl ether and ester functional groups tethered by
increasing length aliphatic linkers yielded highly active compounds 78ꢀ81 shown in Table 7.
(
Table 7, here)
A number of other quaternary ammonium congeners were prepared as more significant scaffold
modifications (Table 7). Spirocyclic pyrrolidineꢀdihydrothieno[2,3ꢀc]pyridinium compound 82
and spirocyclic morpholineꢀ dihydrothieno[2,3ꢀc]pyridinium compound 83 were both more potent
than the NꢀethylꢀNꢀmethyl parent compound (HC = 0.41 µM for 82 and HC = 1.21 µM for 83
5
0
50
versus HC = 1.96 µM for 68), but the former displayed the same level of hERG inhibition (IC
50
5
0
=
3.51 µM for 82 versus IC = 3.38 µM for 68), whereas the latter compound displayed only
50
moderate hERG activity (IC = 12.4 µM for 83). The quaternary dihydrothieno[2,3ꢀc]pyrrole
50
analogue 84 was similar to the parent tetrahydrothieno[2,3ꢀc]pyridine 68 both in terms of hearing
protection (HC = 3.10 µM for 84) and hERG inhibition (IC = 3.29 µM for 84). Two
50
50
compounds in which the quaternary ammonium functional group was outside of the fused
bicyclic ring system were prepared, but both were less active (HC = 3.92 µM for 85 and HC =
5
0
50
5
.87 µM for 86).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The steric environment around the basic tetrahydropyridine nitrogen was further explored by the
synthesis of the 2,2,6,6ꢀtetramethyl tetrahydropyridine derivative 87 (Table 8), which was
prepared from tetramethyl piperidone through the route depicted in Scheme 1. This compound
was more active than the parent compound (HC = 1.24 µM for 87 versus HC = 3.12 µM for 1),
5
0
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
but more importantly it was devoid of hERG liability (IC > 30 µM for 87 versus IC = 3.26 µM
5
0
50
for 1). In vitro ADME profiling of 87 demonstrated that it had excellent metabolic stability, no
issues with CYP inhibition, but suffered from poor aqueous solubility (13 µM) and modestly
increased plasma protein binding (96.5% in human and 95.8% in rat plasma).
(
Table 8, here)
In an effort to evaluate the importance of conformational flexibility of the tetrahydropyridine unit,
we set out to synthesize the bicyclic, tropaneꢀlike compound 91 as a racemate (Figure 5). The
standard Gewald sequence using commercially available tropinone to react with cyanoacetamide
and elemental sulfur in the presence of base yielded a mixture with two predominant products,
1
each having the correct molecular weight of the desired product as well as H NMR spectra
consistent with the expected product structure. An admixture of the two yielded distinct but
partially overlapping NMR spectra suggesting that the two did not interconvert under these
conditions. The 2ꢀaminothiophene materials were difficult to purify and fully characterize and
instead were reacted as a mixture with pꢀchlorophenyl isocyanate to give a mixture of racemic
1
regioisomers 88 and 91 (Figure 5). Again, both compounds exhibited mass and HꢀNMR spectra
consistent with the expected product. Both compounds had good protective activity with 88
slightly more active than 91 [HC = 0.36 µM for 88 versus HC = 0.83 µM for 91]. The
50
50
expected product contains two asymmetric centers at the bridgehead carbon atoms and would be
produced as a racemate. The more active of the two racemic compounds 88 was resolved by
4
4
chiral HPLC chromatography to yield the individual enantiomers as 89 and 90 (ORCꢀ13661).
ACS Paragon Plus Environment

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The structural ambiguity of the products derived from the original tropinone ꢀ Gewald reaction
was eventually resolved by determining the single crystal Xꢀray diffraction structure of 90 HCl
salt as shown in Figure 4 (Supporting Information, Figure S2, S3 and Table S2). Surprisingly,
the isolation of 90 resulted from a previously undescribed side reaction pathway in the Gewald
condensation, resulting in a reversed regiochemistry of the thiophene relative to the bicyclic
tropinone starting material.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(Figure 4, here)
The crystal structure unambiguously proves the absolute structure of 90 as (1R,8S)ꢀ4ꢀ{[(4ꢀ
6
chlorophenyl)carbamoyl]amino}ꢀ11ꢀmethylꢀ5ꢀthiaꢀ11ꢀazatricyclo[6.2.1.0², ]undecaꢀ2(6),3ꢀdieneꢀ
3
ꢀcarboxamide hydrochloride. Based on this the crystal structure, we assigned the structures of
8
9, 90, 92 and 93 as shown in Figure 5, with 91, 92 and 93 representing the “expected” Gewald
products. The mechanism of the “unexpected” Gewald reaction to yield 88, 89 and 90 is under
investigation. An alternate synthetic approach starting from chiral (+) 2ꢀtropinone was developed
1
subsequently and provided 90 that was identical by mass spectrometry, HꢀNMR spectroscopy
and chiral HPLC as the material derived from achiral 3ꢀtropinone. The asymmetric synthesis of
9
0 will be the subject of a subsequent publication.
(Figure 5, here)
Protection assays in zebrafish revealed that 90 is the more potent enantiomer (HC = 0.12 µM)
50
while 89 is considerably less effective (HC > 2.25 µM). The eudismic ratio (ER) of 18.8
5
0
suggests that 90 interacts with a chiral target. The less potent racemate 91 was also resolved by
chiral HPLC to provide the individual enantiomers as 92 and 93. The levorotatory enantiomer 93
is more potent (HC = 0.27 µM) while the dextrorotatory enantiomer 92 is considerably less
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
active (HC = 7.21 µM, ER 26.7). Reaction of the NꢀBOC tropinone under standard Gewald
50
conditions followed by pꢀchlorophenylꢀurea formation yielded a single product 94 (Figure 5),
albeit in poor yield, that was carried through NꢀBOC removal and chiral resolution to yield the Nꢀ
1
H compounds 95 and 96. Based on H NMR chemical shift of the diagnostic Cꢀ6 (Xꢀray structure
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
numbering) methine proton, we assigned the structures of 94, 95 and 96 as shown (Figure 5). As
with the Nꢀmethyl compounds 92 and 93, the levorotatory NꢀH enantiomer 96 was more potent
(
HC = 0.27 µM) while the dextrorotatory enantiomer 95 was less active (HC = 3.92 µM, ER
50 50
14.7), lending additional support to our structural assignments. Completing the NꢀH and Nꢀ
methyl regiochemical tropaneꢀfused thiophene sets, we reacted racemic NꢀBOC 2ꢀtropinone
under the Gewald conditions to afford racemic 97. Chiral HPLC separation of 97 could only be
accomplished prior to NꢀBOC group deprotection. Subsequent removal of the NꢀBOC group
yielded enantiomers 98 and 99.
Examination of the structural relationship among the active enantiomers of the two thiophene
orientation Gewald products for the Nꢀmethyl 90 and 93, and NꢀH compounds 96 and 99 reveals
a similar relationship to that seen with the regioꢀisomeric tetrahydropyridines 1 and 41 (Table 5).
The altering the position of the basic nitrogen relative to the thiophene ring while maintaining the
overall scaffold geometry yields compounds with similar biological activity. The tropinone
compounds 90, 93, 96 and 99 exhibit superior protective activity with HC in the 33 ꢀ 270 nM
5
0
range while their corresponding enantiomers 89, 92, 95 and 98 have significantly reduced activity.
Based on activity and accessibility, indoleꢀcontaining Nꢀalkyl derivative 67 (Table 6), 2,2,6,6ꢀ
tetramethyl tetrahydropyridine derivative 87 (Table 8) and 90 were selected for early preꢀclinical
characterization along with 1 (1).
Pre-clinical characterization of 1, 67, 87 and 90. Although the hitꢀtoꢀlead phase of the
campaign resulted in several very potent and efficacious derivatives of 1 1, the triage of
compounds using the early ADME results was the determining factor in deciding which
ACS Paragon Plus Environment

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analogues to take forward for early pharmacokinetic evaluation. As a number of the quaternary
ammonium salt compounds were not orally bioavailable and also found to inhibit the hERG
channel, we elected to focus on the tertiary amine derivatives. As a result, we chose to profile
three compounds, in addition to 1 (1, HC = 3.12 µM): the terminal indoleꢀcontaining Nꢀalkyl
5
0
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
compound 67 (Table 6, HC = 0.10 µM), the 2,2,6,6ꢀtetramethyl tetrahydropyridine derivative 87
50
(
Table 8, HC = 1.24 µM) and the bicyclic tropaneꢀlike compound 90, Table 8, HC = 0.12 µM).
50 50
Therefore, the pharmacokinetic, metabolic and inꢀvitro toxicological properties for these
candidates were determined. The compounds have moderate to good aqueous solubility [85 µM
for 1, 86 µM for 67, 83 µM for 87, and 76 µM for 90 for the free base forms], acceptable plasma
protein binding profiles [95.9% (human) and 94.1% (rat) 1, 99.9% (human) and 99.9% (rat) for
6
7, 96.5% (human) and 95.8% (rat) for 87, and 90.7% (human) and 90.1% (rat) for 90], and
elimination kinetics with halfꢀlives of 2.1 hours for 1, 7.1 hours for 67, 6.2 hours for 87 and 7.8
hours for 90. Compounds 1, 87 and 90 exhibited no in vitro toxicity against the HepG2 cell line
(IC >300 µM), whereas compound 67 inhibited growth of HepG2 cells is IC 17µM. All four
5
0
50
compounds were well tolerated in an acute toxicity assay in rats at 50 mg/kg i.v. As noted, hit
compound 1 inhibited the hERG ion channel (hERG IC = 3.26 µM) giving a ratio of 1.2
5
0
(
hERG/HC ). On the other hand, compounds 67, 87 and 90 were less potent hERG inhibitors
50
with IC values of 20.1 µM, >30 µM and 5.6 µM, respectively, and with hERG/HC ratios of
50
50
2
01, >24 and 47, respectively. In addition to hERG inhibition, 1 was found to gradually undergo
partial air oxidation to yield an inactive, highly fluorescent byproduct. Compound 67 was shown
to have marginal oral bioavailability (>1%), while 1, 87 and 90 had acceptable to good oral
bioavailability (46%, 21% and >50%). Compound 87 and 90 are resistant to the oxidation
pathway observed for 1 (as well as 67) degradation by virtue of geminal dimethyl substitution in
8
7 and the bicyclic tropane framework of 90.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compounds 1 and 90 were tested for in vitro and in vivo interference with aminoglycoside
antimicrobial activity. There was no interference by either drug with neomycin minimal
inhibitory concentration (MIC) or minimal bactericidal concentration (MBC) against E. coli
ATCC25922. Additionally, 90 exhibited no interference in the MIC for tobramycin, amikacin,
kanamycin and neomycin against five strains of P. aeruginosa or amikacin MIC against M.
tuberculosis. Finally, 90 showed no interference in an in vivo mouse pulmonary tuberculosis
model with amikacin, kanamycin and moxifloxacin (see Supporting Information, Table S4ꢀ1 and
S4ꢀ2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In vivo testing of ORC compounds in rats. On the basis of early preꢀclinical evaluation,
compounds 1 and 90 were tested in an in vivo rat model of aminoglycosideꢀinduced hearing loss.
The auditory brain stem response (ABR) assay is a widely used, nonꢀinvasive method for
quantification of auditory function in vertebrates, including humans. Male Sprague Dawley rats
5ꢀ7 weeks old were purchased from Harlan Laboratories. Briefly, rats are anesthetized, placed in
a soundꢀattenuated enclosure on a heating pad to maintain body temperature at 37.5 ºC. Electrical
responses from the ear and CNS are detected with subcutaneous needle electrodes. Positive and
negative electrodes are inserted at the left temporal bone above the pinna and at the vertex of the
skull, with the ground electrode in the thigh. Freeꢀfield pureꢀtone stimuli are generated by
custom software, and presented as short (5 ms) acoustic pulses with 1ꢀms rise/fall times, at a
repetition rate of 19/s. In addition, broadband clicks are presented at the beginning and end of
each session to assess any changes in the animal’s condition. Neural responses are amplified,
bandpass filtered (100–3000 Hz), digitized using custom software and displayed onꢀline as an
average response to the repeated acoustic stimulus. A sequence of pure tone stimuli of varying
frequencies (2, 4, 8,16 and 32 kHz) and intensities (usually 15 – 90 dB, SPL) is presented and the
resulting waveforms are recorded. All individual responses and averaged responses were saved.
Examination of the averaged soundꢀinduced waveforms by a trained experimenter yields the
ACS Paragon Plus Environment

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hearing thresholds—minimum sound level required to induce a reliable waveform at each
frequency tested.
(
Figure 6, here)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In preliminary experiments, we showed that 1 injected intravenously (i.v.) was nonꢀtoxic to rats at
doses up to 25 mg/kg. This dose also yielded a plasma halfꢀlife of 1.7 hours. Based on literature
precedent, we carried out an in vivo hearing protection experiments using kanamycin (500
mg/kg/day s.c.). The results of these experiments are shown in Figure 6. After initial ABR
testing, male SpragueꢀDawley rats 50ꢀ60 days old were treated with kanamycin 500 mg/kg/day
(
s.c.) for 14 days and then allowed to recover without treatment for an additional two weeks
before final ABR testing. A second group was subjected to the same treatment with the addition
of exposure to 1 at 25 mg/kg/day by intraperitoneal injection. Control groups included animals
receiving 1 alone, the vehicle for 1 (Cremaphor EL:DMSO:EtOH:PBS) alone or no treatment.
ABR testing (Figure 6) reveals a highly significant hearing loss at 8ꢀ32 kHz, that is greatest at the
higher frequencies in animals treated with kanamycin alone. This hearing loss was completely
prevented by 1 at 8 and 16 kHz and mostly resolved at the highest frequency tested (32 kHz). All
rat protocols were approved by the University of Washington IACUC.
(Figure 7, here)
We selected compound 90 as our lead compound for full preꢀclinical evaluation based on its
superior protective activity in the zebrafish assay, decreased hERGꢀinhibitory activity and
superior pharmacokinetic properties (Figure 7 and Supporting Information, Table S1) and
efficiency of large scale synthesis. Compound 90 was tested for in vivo protection using
amikacin, a more widely employed aminoglycoside antibiotic than kanamycin in the clinic. For
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
testing the in vivo mammalian efficacy of 90, we used Fischer 344 rats 7ꢀ9 weeks old at the time
of testing were purchased from Taconic. Rats were exposed to amikacin alone at 320 mg/kg/day
(
s.c.) for 12 days or the same amikacin exposure plus 90 (5 mg/kg/day, p.o.) for the same period.
Additional rats were treated with 90 alone (5 mg/kg/day [n=3], or 25 mg/kg/day [n=3], p.o.) for
2 days. In all experiements 90 was delivered by oral gavage. We find that exposure to amikacin
compared to kanamycin causes significantly more reproducible hearing loss in animals treated
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
with AGA alone. For oral gavage the vehicle for 90 was PEG300:DMA:EtOH:H 0. ABR
2
thresholds were determined for each rat prior to drug treatment. Rats were then treated with
amikacin alone or amikacin plus 90 daily for 12 days and then allowed to recover for 2 weeks
before final ABR testing. As shown in Figure 8, Fischer 344 rats treated with amikacin (320
mg/kg/day s.c.) alone for 12 days exhibited moderate to severe hearing loss at all frequencies
tested. Coꢀadministration of amikacin (320 mg/kg/day s.c.) and 90 (5 mg/kg/day p.o.) for 12 days
resulted in highly significant hearing protection across all frequencies tested. Furthermore, the
resulting average hearing thresholds of this group are well within the normal range, and 6 of the 8
rats tested in this group showed complete protection from hearing loss due to the aminoglycoside
exposure. None of the rats treated with 90 alone showed reliable changes in hearing thresholds
(
data not shown). In support of the audiological findings, immunohistochemical examination of
the inner and outer hair cells of the organ of Corti from experimental animals showed significant
disruption of normal tissue architecture (Figure 9, panel A) in Fischer 344 rats treated with
amikacin alone (Figure 9, panel B). The amikacinꢀinduced hair cell death in the organ of Corti
was suppressed by coꢀtreatment with 90.
(
Figures 8 and 9, here)
ACS Paragon Plus Environment

Page 19 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discussion
Protection of hearing through pharmacological intervention against the myriad insults that are
45
capable of causing hearing loss has been a long soughtꢀafter goal. We describe the development
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
and preclinical characterization of a family of ureaꢀthiophene carboxamides culminating in 90, an
orally active and well tolerated agent with excellent pharmacokinetic and metabolic properties.
Compound 90 protects rat hearing against amikacinꢀinduced ototoxicity while showing no effect
on in vitro or in vivo antibacterial activity of several clinically relevant AGAs. Our structure
activity relationship (SAR) for this compound series is based on over 300 analogs interrogating
34
all elements of the core structure of the screening hit 1. Our studies identified structural
elements that are absolutely required for protective activity as well as those that can be
substantially altered without disrupting the biological activity. The pꢀchlorophenyl urea
substituent appears to be optimal as a wide variety of substituted aryl urea compounds exhibit
comparable activity at best. Significant changes of the aryl substituent nature or position lead to
loss of activity. The bisꢀmonoꢀsubstituted urea is absolutely required for activity as even modest
change such as replacement by an amide, cyclopropyl malonamide or Nꢀmethylation abolishes
activity. Likewise, the thiopheneꢀ3ꢀcarboxamide is critical for protective activity as replacement
of either thiophene or carboxamide with even conservative structural variants leads to dramatic
loss of activity.
The Nꢀethyl tetrahydropyridine moiety of 1 provided the richest ground for structural variation.
A basic nitrogen or quaternary ammonium salt is needed for activity, presumably existing in a
mostly, or completely cationic state at physiological pH for favorable electrostatic interactions
with its target. The positive charge can however reside at the tetrahydropyridine 2ꢀposition (as in
1) or 3ꢀposition (as in 41, Table 5), in a 5ꢀmembered ring (as in 42, Table 5) or 7ꢀmembered ring
(as in 43 and 44, respectively, Table 5) or as part of a rigid bicycloꢀ[3.2.1]ꢀazaꢀoctane (tropane)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ring system (as in 90 and 93, Table 8). Extending the Nꢀalkyl substituent to 2ꢀnaphthylꢀ or
indoleꢀcontaining amides (as in 66 and 67, Table 6) yielded the most active compounds in the
zebrafish assays but at the cost of physicochemical properties. Compound 67 has the high
potency value (HC = 70 nM), but lacks solubility and oral bioavailability making it unsuitable
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
for in vivo testing in the rat. Based on the high potency of aryl amide analogs like 67, we
investigated the possibility of appending functional groups that would be useful for mechanistic
and target identification studies. We discovered that a wide range of substituents are tolerated
when connected to the tetrahydropyridine nitrogen through a 7ꢀ to 17ꢀatom linker. Analogs
containing a fluorescent probe (i.e. BODIPY) as well as an affinity tag (i.e. biotin) have been
synthesized based on the existing SAR and are currently being used in mechanistic and target
identification efforts. Among the most active ORC compounds in this series are 90, HC = 120
5
0
nM), 93 (HC = 270 nM), 96 HC = 270 nM) and 99 HC = 33 nM), which vary only in Nꢀ
50
50
50
substitution on the 8ꢀazabicyclo[3.2.1] ring system and position of the 2ꢀcarbon bridge relative to
the thiophene ring (Figure 5). Because of it superior potency in the zebrafish hair cell protection
assay and improved pharmacokinetic properties 90 was advanced through INDꢀenabling studies
under the auspices of Oricula Therapeutics.
Conclusion
In the absence of a molecular target, elaboration of our original screening hit 1, into a full SAR
program was made possible using the phenotypic zebrafish neuromast hair cell viability assay to
predict in vivo activity in a mammalian model. This effort yielded 90 as a promising clinical
candidate for the prevention of aminoglycoside induced hearing loss in patients. Clinical
adoption of such an agent would reinvigorate the safe use of this underused broad class of
effective antibiotics and thus expand their utility in the treatment of serious and lifeꢀthreatening
bacterial infections.
ACS Paragon Plus Environment

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Experimental Section
Zebrafish neuromast hair cell protection assay.
Adult AB zebrafish were bred and newly fertilized embryos were collected the week prior to
o
testing and raised at 28.5 C in petri dishes containing embryo medium (1 mM MgSO , 0.15 mM
4
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
KH PO , 50 µM Na HPO , 1 mM CaCl , 500 µM KCl, 115 mM NaCl, 700 µM NaHCO ). Newly
2
4
2
4
2
3
hatched freeꢀswimming larvae were fed paramecium at 4 days post fertilization (dpf). All
procedures were approved by the University of Washington Animal Care and Use Committee.
For initial efficacy testing of each new experimental compound, fish 5ꢀ7 dpf were transferred to
cell culture baskets placed in 6ꢀwell culture plates containing 7 mL of embryo medium.
Typically, tests are done with ten fish per basket but work well with up to 50 fish. All treatment
and wash volumes are 7 mL. Fish were preꢀtreated with test compound for 1 hour at five initial
concentrations (1.56, 3.125, 6.25, 12.5, and 25 µM). Following this exposure to the test
compound alone, fish were moved to a new well containing the test compound plus 200 µM
neomycin (neomycin sulfate, Sigma, St. Louis, MO, catalog # N1142) for 1 hr. If significant hair
cell protection was observed at the lowest concentration the dosages were reduced by half until
hair cell protection was eliminated. All experimental compounds were tested alongside the
following controls groups: fish treated with neomycin (200 µM) alone; fish treated with test
compound (25 µM) alone); fish treated with 1 at the same concentration range as test compound;
and untreated fish using the same procedures.
Following neomycin exposure, fish were rinsed briefly 4 times in embryo medium and 700 µl of
0
.05% DASPEI (2ꢀ{4ꢀ(dimethylamino)styryl}ꢀNꢀethylpyridinium iodide, Molecular Probes,
Eugene, OR) was added and allowed to stain for 15 minutes. DASPEI brightly labels the hair
4
6
cells of each neuromast due to their robust packing in mitochondria. Following this labelling
and two rinses in embryo medium, the fish were anesthetized with 350 µl MS222 (0.55 µg/mL
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
final concentration, 3ꢀaminobenzoic acid ethyl ester, methansulfonate salt, Sigma, St. Louis,
MO).
For assessment of the doseꢀresponse protective functions, the fish in each experimental and
control group were transferred to separate wide depression slides in embryo medium with
MS222. Slides were mounted on the stage of an epifluorescence dissecting microscope equipped
with a DAPSEI filter set (excitation 450ꢀ490 nM and barrier 515 nM, Chroma Technologies,
Brattleboro, VT). The hair cell staining of ten neuromasts (SO1, SO2, IO1, IO2, IO3, IO4, M2,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
7
MI1, MI2 and O2) on one side of each animal was visually evaluated. Each neuromast was
scored for presence of a normal compliment of hair cells (score=2), reduced DASPEI staining
indicating a reduction in hair cell number (score=1), or absence of DASPEI staining (score=0).
The total score for each animal was tabulated, to give a composite score that ranged from 0 to 20
for each fish. Average scores and standard deviations were calculated for animals in each
treatment group. Scores were normalized to the control group (vehicle only, no drug, no
neomycin) and expressed as % hair cell survival. If hair cell survival was at least 50%, the 50%
effective concentration (HC ) was calculated as a linear extrapolation from the nearest
50
concentrations of protective drug that produced hair cell survival below and above 50%.
Rat Auditory Brain Stem Response (ABR) assay for hearing protection.
Male Fischer 344 rats were purchased from Harlan Laboratories (Livermore CA) at 40ꢀ50 days
postnatal. Auditory brainstem responses (ABRs) were measured from each rat twice, once 1ꢀ2
weeks prior to drug treatments and again at 2 weeks after the termination of drug treatment. Rats
were anesthetized with isoflurane, placed on a heating pad to maintain body temperature near
3
7°C, and placed in a soundꢀattenuating chamber. ABRs responses were recorded using standard
subcutaneous needle electrodes with the positive and negative electrodes at the left temporal bone
above the pinna and the vertex of the skull, and the ground electrode in the thigh. Freeꢀfield pure
ACS Paragon Plus Environment

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tone stimuli were generated, and ABR recordings were digitized using custom software. Tone
pips were 5 ms in duration with 1ꢀms rise/fall times, presented at a repetition rate of 19/s. In
addition, broadband clicks were presented at the beginning and end of each session to assess for
any changes in the animal’s condition. All stimuli were calibrated onꢀline at the beginning of each
experiment with a calibrated probe microphone. Neural responses were preꢀamplified (100x; AꢀM
Systems amplifier Model 3000), sent through an MA3 amplifier with an additional 20 dB postꢀ
preamp gain (Tucker Davis Technologies), bandpass filtered (100–3000 Hz; KrohnꢀHite filter
model 3550), and digitized at 24.4 kHz. We sampled responses in a 15ꢀms window (with a 5ms
stimulus onset delay). The threshold was defined as the lowest sound pressure level (SPL) in
which a recognizable waveform was present and repeatable. Thresholds were determined at 2, 4,
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
8
, 16, and 32 kHz and for a broadꢀband click. Stimuli were presented 500 times from 80 to 20 dB
SPL in steps of 10 and then 1000 repetitions in steps of 5 dB SPL when approaching threshold.
Near threshold each series was repeated to determine the reliability of the waveform at the
estimated threshold, 5 dB above and 10 dB below the estimated threshold. When animals
appeared to be deaf at a particular frequency, the stimulus was presented at the maximum
intensity generated by our system (90–100 dB SPL at least twice at 1000 repetitions to be assured
of a complete hearing loss. The threshold was then arbitrarily set at 100 dB SPL. All averaged
responses were examined on line and thresholds were estimated. Following the experimental
session all responses were printed and rescored twice, once by the experimenter and once by an
ABR expert who was blinded with regard to the experimental condition. Threshold estimates
were within 5 dB in over 97 % of the response traces, and the few that differed by >10dB were
reꢀexamined by the two evaluators to come to a consensus. Data were analyzed as raw thresholds
at each test frequency and as “Threshold Shift”, comparing preꢀtest threshold with postꢀ drug
administration threshold, where a positive number indicates a hearing loss (in dB) due to the drug
treatment.
Chemistry
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
All reactions were performed under a dry atmosphere of nitrogen unless otherwise specified.
Indicated reaction temperatures refer to the reaction bath, while room temperature (rt) is noted as
2
5 ºC. Commercial grade reagents and anhydrous solvents were used as received from vendors
and no attempts were made to purify or dry these components further. Removal of solvents under
reduced pressure was accomplished with a Buchi rotary evaporator at approximately 28 mm Hg
pressure using a Teflonꢀlinked KNF vacuum pump. Thin layer chromatography was performed
using either 1” x 3” AnalTech No. 02521 or Merck 60 F254 silica gel plates with fluorescent
indicator using appropriate solvent mixtures. Visualization of TLC plates was made by
observation with either short wave UV light (254 nm lamp), 10% phosphomolybdic acid in
ethanol or in iodine vapors. Medium pressure Flash column chromatography was carried out
using either a Teledyne Isco CombiFlash Companion Unit with RediSep Rf silica gel columns or
a Biotage Isolera with SiliCycle HP cartridges. Proton NMR spectra were obtained either on 300
MHz Bruker Nuclear Magnetic Resonance Spectrometer or 500 MHz Bruker Nuclear Magnetic
Resonance Spectrometer and chemical shifts (δ) are reported in parts per million (ppm) and
coupling constant (J) values are given in Hz, with the following spectral pattern designations: s,
singlet; d, doublet; t, triplet, q, quartet; dd, doublet of doublets; m, multiplet; br, broad singlet;
sym, symmetrical. Tetramethylsilane (TMS) was used as an internal reference. Melting points
are uncorrected and were obtained using a MELꢀTEMP Electrothermal melting point apparatus.
Mass spectroscopic analyses were performed either using positive mode electron spray ionization
(ESI) on a Varian ProStar LCꢀMS with a 1200L quadrupole mass spectrometer or using positive
mode atmospheric pressure chemical ionization (APCI) on a Shimadzu LCꢀMS system. High
performance liquid chromatography (HPLC) purity analysis was conducted using a Varian Pro
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Star HPLC system with a binary solvent system A and B using a gradient elution [A, H O with
2
0
.1% trifluoroacetic acid (TFA); B, CH CN with 0.1% TFA] and flow rate = 1 mL/min, with UV
3
detection at 254 nm. All final compounds were purified to ≥95% purity and these purity levels
were measured by a Varian Pro Star HPLC system. Three different Varian Pro Star HPLC
ACS Paragon Plus Environment

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
methods were used to establish compound purity. HPLC Method A: Phenomenex Luna C18(2)
column (4.6 × 250 mm); mobile phase, A = H O with 0.1% TFA and B = CH CN with 0.1%
2
3
TFA; gradient 10–95% B (0.0–10 min; hold for 6 min); UV detection at 254 nm. HPLC Method
B: SunFire C18 column (4.6 × 250 mm); mobile phase, A = H O with 0.1% TFA and B = CH CN
2
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
with 0.1% TFA; gradient: 10–100% B (0.0–20 min; hold for 5 min); UV detection at 254 nm.
HPLC Method C: SunFire C18 column (4.6 × 250 mm); mobile phase, A = H O with 0.1% TFA
2
and B = CH CN with 0.1% TFA; gradient: 0–100% B (0.0–15 min; hold for 5 min)); UV
3
detection at 254 nm.
2
ꢀAminoꢀ6ꢀethylꢀ4,5,6,7ꢀtetrahydrothieno[2,3ꢀc]pyridineꢀ3ꢀcarboxamide. A stirred
mixture of Nꢀethylꢀ4ꢀpyrrolidinone (22.7 g, 178 mmol), 2ꢀcyanoacetamide (16.5 g, 197 mmol),
sulphur (6.87 g, 215 mmol) and morpholine (31.5 g, 362 mmol) in ethanol (350 mL) was heated
to reflux under nitrogen for 5 h. After this time, the reaction mixture was cooled to room
temperature and concentrated under reduced pressure. The residue was mixed with saturated
aqueous sodium bicarbonate (200 mL), water (200 mL). The aqueous mixture was extracted with
methylene chloride (5 × 200 mL). The combined organic extracts were dried over anhydrous
sodium sulfate, filtered, and concentrated under reduced pressure. The resulting residue was
triturated with cold methanol (30 mL) and filtered. The filter cake was washed with cold
methanol (2 × 10 mL) and then dried under reduced pressure to provide the product as a yellow
1
solid (21.7 g, 54%): H NMR (300 MHz, DMSOꢀd ) δ 6.98 (s, 2H), 6.52 (bs, 2H), 3.29 (s, 2H),
6
2
.66 (d, J = 4.8 Hz, 2H), 2.60 (d, J = 4.8 Hz, 2H), 2.46 (q, J = 7.2 Hz, 2H), 1.04 (t, J = 7.2 Hz,
+
3
H); LRMS m/z 226 [M+H] .
2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ6ꢀethylꢀ4,5,6,7ꢀtetrahydrothieno[2,3ꢀc]pyridineꢀ3ꢀ
carboxamide (1). To the stirred solution of 2ꢀaminoꢀ6ꢀethylꢀ4,5,6,7ꢀtetrahydrothieno[2,3ꢀ
c]pyridineꢀ3ꢀcarboxamide (450 mg, 2.00 mmol) in anhydrous tetrahydrofuran (10 mL) at room
temperature under nitrogen was added a solution of 4ꢀchlorophenyl isocyanate (400 mg, 2.40
mmol) in anhydrous methylene chloride (6 mL) dropwise over 3 min. Then the reaction mixture
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was stirred overnight at room temperature. The reaction mixture was filtered. The filter cake was
washed with methylene chloride (5 mL) and then dried under reduced pressure to provide the
1
product 1 (free base) as a white solid (301 mg, 38%): H NMR (300 MHz, DMSOꢀd ) δ 10.94 (s,
6
1
H), 10.12 (s, 1H), 7.52–7.45 (m, 4H), 7.34 (d, J = 9.0 Hz, 2H), 3.45 (s, 2H), 2.77 (d, J = 4.8 Hz,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
H), 2.65 (d, J = 4.8 Hz, 2H), 2.55–2.45 (m, 2H), 1.07 (t, J = 6.6 Hz, 3H); LRMS m/z 379
+
[
M+H] .
2
ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ6ꢀethylꢀ4,5,6,7ꢀtetrahydrothieno[2,3ꢀc]pyridineꢀ3ꢀ
carboxamide Hydrochloride (1 HCl). To the stirred mixture of compound 1 (free base) (157 mg,
.400 mmol) in methylene chloride (50 mL) at room temperature was added hydrochloric acid (2
0
M in diethyl ether, 0.300 mL, 0.600 mmol). After addition, the mixture was concentrated under
reduced pressure. The resulting solid was triturated with methylene chloride and filtered to afford
1
compound (1 HCl) as pale yellow solid: H NMR (500 MHz, DMSOꢀd ) δ 10.94 (s, 1H), 10.84
6
(
bs, 1H), 10.22 (s, 1H), 7.75–6.90 (m, 6H), 4.49 (d, J = 14.5 Hz, 1H), 4.22–4.16 (m, 1H), 3.65–
3.62 (m, 1H), 3.38–3.05 (m, 5H), 1.32 (t, J = 7.2 Hz, 3H); HRMS (ESI) m/z calculated for
+
C H ClN O S [M+H] 379.09899, found 379.10085.
17
19
4
2
(
+/ꢀ)ꢀ2ꢀAminoꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀ epiminocyclohepta[b]ꢀthiopheneꢀ3ꢀ
carboxamide. The stirred mixture of tropinone (3.00 g, 21.6 mmol), 2ꢀcyanoacetamide (1.99 g,
3.7 mmol), sulphur (830 mg, 25.9 mmol), and morpholine (3.75 g, 43.0 mmol) in ethanol (80
2
mL) was heated to reflux under nitrogen for 4 h. After this time, the reaction mixture was cooled
to room temperature and concentrated under reduced pressure. The resulting residue was diluted
with saturated aqueous sodium bicarbonate (60 mL) and extracted with methylene chloride (3 ×
1
50 mL). The combined organic extracts were dried over anhydrous sodium sulfate, filtered, and
concentrated under reduced pressure. The resulting residue was purified by flash column
chromatography on silica gel eluting with methanol/methylene chloride (1:9) to provide
compound (+/ꢀ)ꢀ2ꢀAminoꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀepiminocyclohepta[b]ꢀthiopheneꢀ
+
3ꢀcarboxamide as a brown solid (396 mg, 8%): LRMS m/z 238 [M+H] .
ACS Paragon Plus Environment

Page 27 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
+/ꢀ)ꢀ2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀ
epiminocyclohepta[b]thiopheneꢀ3ꢀcarboxamide (88). To the stirred mixture of 2ꢀAminoꢀ9ꢀ
methylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀepiminocyclohepta[b]ꢀthiopheneꢀ3ꢀcarboxamide (375 mg, 1.58
mmol) in methylene chloride (10 mL) at room temperature under nitrogen was added a solution
of 4ꢀchlorophenyl isocyanate (255 mg, 1.66 mmol) in methylene chloride (10 mL). The reaction
mixture was stirred overnight and concentrated under reduced pressure. The resulting residue
was purified by flash column chromatography on silica gel eluting with methanol/methylene
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
chloride (15:85) to provide compound 88 as an offꢀwhite solid (369 mg, 60%): H NMR (300
MHz, DMSOꢀd ) δ 10.70 (bs, 1H), 10.05 (bs, 1H), 7.90–6.50 (m, 6H), 4.11 (bs, 1H), 2.99 (d, J =
6
+
1
4.1 Hz, 1H), 2.35–1.95 (m, 7H), 1.79 (bs, 1H), 1.48 (bs, 1H); LRMS m/z 391 [M+H] .
(ꢀ)ꢀ2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀ
epiminocyclohepta[b]thiopheneꢀ3ꢀcarboxamide (89) and (+)ꢀ2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ9ꢀ
methylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀepiminocyclohepta[b]thiopheneꢀ3ꢀcarboxamide (90)
88 (720 mg) was separated by chiral preparative HPLC (20 µm CHIRALCEL OD, 5 cm
×
50 cm, 100 mL/min flow rate, 120 mg/injection) eluting with 0.1% diethylamine in 20%
+
ethanol/heptane to provide 89 (268 mg, 37%)as a white solid, LRMS m/z 391 [M+H] , followed
+
by 90 (250 mg, 35%) as a white solid. LRMS m/z 391 [M+H] . Free base optical rotations. 89: ꢀ
1
9.5º (c = 0.2, MeOH), T=25 ºC. 90: +18.5º (c = 0.2, MeOH), T=25 ºC.
ꢀ)ꢀ2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀ
(
epiminocyclohepta[b]thiopheneꢀ3ꢀcarboxamide Hydrochloride 89 and (+)ꢀ2ꢀ[3ꢀ(4ꢀ
Chlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀepiminocyclohepta[b]thiopheneꢀ3ꢀ
carboxamide Hydrochloride (89). To the stirred mixture of compound 89 (free base) (240 mg,
0
.610 mmol) in methanol (30 mL) at room temperature was added 1 M hydrochloric acid (1.25
mL, 1.25 mmol) dropwise. After addition, the mixture was stirred for 10 min. After this time,
the mixture was diluted with water (10 mL) and lyophilized to provide compound 89 as a white
1
solid (255 mg, 97%): H NMR (300 MHz, DMSOꢀd ) δ 11.24 (bs, 0.40 H), 10.34 and 10.32 (2 s,
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
H), 10.19–10.16 (m, 1.60H), 7.51–7.48 (m, 4H), 7.35 (d, J = 9.0 Hz, 2H), 4.95–4.82 (m, 1H),
4.19–4.08 (m, 1H), 3.43–3.17 (m, 1H), 2.88–2.68 (m, 4H), 2.45–2.09 (m, 3H), 1.89–1.80 (m,
1
H); optical rotation [α] = ꢀ5.5º (c = 0.2, MeOH); HRMS (ESI) m/z calculated for
D
+
C H ClN O S [M+H] 391.09899, found 391.09906.
18
19
4
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(+)ꢀ2ꢀ[3ꢀ(4ꢀChlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀ
epiminocyclohepta[b]thiopheneꢀ3ꢀcarboxamide Hydrochloride 89 and (+)ꢀ2ꢀ[3ꢀ(4ꢀ
Chlorophenyl)ureido]ꢀ9ꢀmethylꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ5,8ꢀepiminocyclohepta[b]thiopheneꢀ3ꢀ
carboxamide Hydrochloride (90). To the stirred mixture of compound 90 (free base)(220 mg,
0
.60 mmol) in methanol (30 mL) at room temperature was added 1 M hydrochloric acid (1.25
mL, 1.25 mmol) dropwise. After addition, the mixture was stirred for 10 min. After this time,
the mixture was diluted with water (10 mL) and lyophilized to provide compound 90 as a white
1
solid (243 mg, 96%): 90: H NMR (300 MHz, DMSOꢀd ) δ 11.20 (bs, 0.34H), 10.34 and 10.31 (2
6
s, 1H), 10.18 (s, 1H), 10.16 (bs, 0.66H), 7.54–7.33 (m, 6H), 4.94–4.90 (m, 1H), 4.20–4.06 (m,
1H), 3.42–3.16 (m, 1H), 2.88–2.65 (m, 4H), 2.49–2.28 (m, 2H), 2.21–2.09 (m, 1H), 1.89–1.80
(
m, 1H); optical rotation [α] = +3.5º (c = 0.2, MeOH); HRMS (ESI) m/z calculated for
D
+
C H ClN O S [M+H] 391.09899, found 391.09907.
18
19
4
2
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS Publications Website at DOI:
(
to be inserted by publisher).
Figure S1: Reproducibility of Zebrafish Lateral Line Hair Cells Viability Assay;
Table S1: Pharmacokinetic and Toxicological Properties of 90; Figure S2 & S3,
Figures S2, S3 and Table S2: Compound 90 Crystal Structure Determination
ACS Paragon Plus Environment

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Tables S4 and S5: Nonꢀinterference in Antimicrobial Activity of Aminoglycoside
Antibiotics; Full Experimental Procedures and Characterization of Compounds 1-
99 (PDF)
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
SMILES strings for reported compounds (CSV)
Accession code
Xꢀray crystallographic structure of 90 HCl has been deposited at The Cambridge Crystallographic
Database (CSD) with the accession code CCDCꢀ1557948 (http://ccdc.cam.ac.uk).
AUTHOR INFORMATION
Corresponding Author
*
phone: 206ꢀ667ꢀ6241. fax: 206ꢀ667ꢀ5255. Email: jsimon@fredhutch.org.
ORCID
Julian A. Simon: 0000ꢀ0001ꢀ9235ꢀ7565
Notes
The authors declare no competing financial interest.
Acknowledgements
We thank AMRI chemistry colleagues Randall Davis and Feryan Ahmed for synthetic
efforts in support of this program. We also thank Dr. Brandon Mercado of the Yale
University Chemical and Biophysical Instrumentation Center for Xꢀray crystal structure
determination of 90 HCl. We are also grateful for the assistance of Kelly Owens, Allison
Hixon, Divya Mehta, Heather Kinnear and Patricia Wu for carrying out the zebrafish
ACS Paragon Plus Environment