Repurposing the antihistamine terfenadine for antimicrobial activity against staphylococcus aureus

Article abstract of DOI:10.1021/jm5010682

Staphylococcus aureus is a rapidly growing health threat in the U.S., with resistance to several commonly prescribed treatments. A high-throughput screen identified the antihistamine terfenadine to possess, previously unreported, antimicrobial activity against S. aureus and other Gram-positive bacteria. In an effort to repurpose this drug, structure-activity relationship studies yielded 84 terfenadine-based analogues with several modifications providing increased activity versus S. aureus and other bacterial pathogens, including Mycobacterium tuberculosis. Mechanism of action studies revealed these compounds to exert their antibacterial effects, at least in part, through inhibition of the bacterial type II topoisomerases. This scaffold suffers from hERG liabilities which were not remedied through this round of optimization; however, given the overall improvement in activity of the set, terfenadine-based analogues provide a novel structural class of antimicrobial compounds with potential for further characterization as part of the continuing process to meet the current need for new antibiotics.

Full text of DOI:10.1021/jm5010682

Subscriber access provided by TULANE UNIVERSITY
Article
Repurposing the antihistamine terfenadine for
antimicrobial activity against Staphylococcus aureus
Jessamyn I. Perlmutter, Lauren T. Forbes, Damian J. Krysan, Katherine Ebsworth-Mojica,
Jennifer M Colquhoun, Jenna L. Wang, Paul M. Dunman, and Daniel Patrick Flaherty
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm5010682 • Publication Date (Web): 19 Sep 2014
Downloaded from http://pubs.acs.org on September 27, 2014
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 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Repurposing the antihistamine terfenadine for
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
antimicrobial activity against Staphylococcus aureus
†
†
‡
‡,†
Jessamyn I. Perlmutter, Lauren T. Forbes, Damian J. Krysan, Katherine Ebsworth-Mojica,
†
§
†
,§
Jennifer M. Colquhoun, Jenna L. Wang, Paul M. Dunman, Daniel P. Flaherty*
†
University of Rochester Medical Center, Department of Microbiology and Immunology
6
01 Elmwood Ave., Rochester, NY 14642
‡
University of Rochester Medical Center, Department of Pediatrics
6
01 Elmwood Ave., Rochester, NY 14642
§
University of Kansas, Specialized Chemistry Center, Delbert M. Shankel Structural Biology
Center, 2034 Becker Dr., Lawrence, KS 66047, United States
Keywords: terfenadine, S. aureus, bacterial type II topoisomerase inhibitor, DNA Gyrase
inhibitor
ABSTRACT
Staphylococcus aureus is a rapidly growing health threat in the U.S. with resistance to several
commonly prescribed treatments. A high-throughput screen identified the antihistamine
terfenadine to possess, previously unreported, antimicrobial activity against S. aureus and other
Gram-positive bacteria. In an effort to repurpose this drug, structure-activity relationship studies
yielded 84 terfenadine-based analogs with several modifications providing increased activity
versus S. aureus and other bacterial pathogens, including Mycobacterium tuberculosis.
Mechanism of action studies revealed these compounds to exert their antibacterial effects, at
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 83
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
least in part, through inhibition of the bacterial type II topoisomerases. This scaffold suffers from
hERG liabilities which were not remedied through this round of optimization; however, given
the overall improvement in activity of the set, terfenadine-based analogs provide a novel
structural class of antimicrobial compounds with potential for further characterization as part of
the continuing process to meet the current need for new antibiotics.
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
INTRODUCTION
The ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella
pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) are
1
responsible for 51% of all U.S. hospital-acquired infections (HAIs). Among these, S. aureus
accounts for 16% of HAIs and is responsible for more deaths in the U.S. annually than
1
-3
HIV/AIDS.
The organism’s high morbidity and mortality rates are, in part, attributable to the
4
fact that S. aureus has developed resistance to currently available antibiotics. The quinolone
5
class of antibiotics was once a predominant treatment option for S. aureus infections; however,
6
,7
due to increasing quinolone resistance, these drugs continue to have diminishing efficacy.
The antimicrobial activity of the quinolones and fluoroquinolones, such as ciprofloxacin
(Figure 1) and levofloxacin, is thought to be mediated by their ability to inhibit the DNA re-
ligation activity of the bacterial type II topoisomerases, DNA gyrase and topoisomerase IV.
Resistance can arise from decreased access to these cellular targets or by mutations within the
8
,9
type II topoisomerases.
Despite the rise in resistance to quinolones, their previous success
validates the type II topoisomerases as valuable targets in searching for novel antimicrobial
scaffolds. Indeed, academic and industrial laboratories have devoted much effort toward
developing novel bacterial type II topoisomerase inhibitors (NBTIs) featuring compound
10-15
scaffolds chemically distinct from those of the quinolone class of antibiotics,
including the
ACS Paragon Plus Environment
2

Page 3 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N-linked piperidine 6-(methyloxy)-4-(2-{4-[([1,3]oxathiolo[5,4-c]pyridin-6-ylmethyl)amino]-1-
1
6
piperidinyl}ethyl)-3-quinolinecarbonitrile dihydrochloride 43 (GSK299423). However, despite
NBTI development efforts, pharmaceutical pipelines are still severely lacking quality antibiotic
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
17
candidates and the loss of several antibiotic groups in the pharmaceutical industry only
1
8,19
exacerbates the problem.
This, coupled with increasing prevalence of quinolone-resistant S.
aureus, makes it paramount that novel small molecule inhibitors are discovered for type II
topoisomerases for the therapeutic intervention of staphylococcal disease and possibly other
bacterial pathogens.
Figure 1. Structures of ciprofloxacin, NBTI 43 and terfenadine (1a)
The practice of drug repurposing, in which a drug previously developed to treat one
disease is then identified for possibly treating a second disease, has emerged as an attractive
20,21
alternative for drug discovery research.
The repurposed drug most likely has already been
2
2
optimized for physicochemical and pharmacokinetic properties providing a more attractive
starting point as far as these factors are concerned. Recently the National Center for Advancing
Translational Science (NCATS) has invested in this area with the formation of the NIH Chemical
Genomic Center Pharmaceutical Collection as both an informatics and screening resource for
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 83
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
23
drug repurposing research, lending credence to the potential and popularity of drug
repurposing.
Given the need for novel antimicrobials and the attractive features of drug repurposing,
we previously performed a high-throughput screen (HTS) to identify FDA approved drugs with
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
4
bactericidal activity toward the ESKAPE pathogens. Results identified a set of compounds that
were developed for other indications but displayed previously unreported antimicrobial activity.
One particular hit, the antihistamine terfenadine (Figure 1), was found to possess antimicrobial
activity versus the planktonic, biofilm, and small-colony variant forms of S. aureus. Given the
activity and relatively convenient synthetic route to analogs, terfenadine provided an attractive
starting point for studying the structure-activity relationship (SAR) of S. aureus antimicrobial
activity. However, terfenadine is not without its flaws. The clinical use of the drug was
discontinued in favor of its active metabolite fexofenadine (Allegra®) because a segment of the
2
5,26
patient population exhibited cardiac arrhythmia, attributed to prolonged QT interval,
due to
2
7
inhibition of the human ether-á-go-go related gene (hERG) potassium channel. Nonetheless, it
13
has been shown previously that it is possible to reduce hERG liabilities via an SAR strategy
and given the encouraging results from the HTS, we decided it would be beneficial to embark on
an SAR-optimization study of terfenadine (1a) and its analogs for inhibition of S. aureus and
those results are reported herein.
RESULTS AND DISCUSSION
Chemistry
A total of 84 terfenadine-based analogs were synthesized for optimization of
14
antimicrobial activity against S. aureus strain UAMS-1, a well-studied osteomyelitis clinical
2
8
isolate, by standard CLSI methods. The majority of analogs were synthesized by one of two
ACS Paragon Plus Environment
4

Page 5 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
routes while several required alternate routes or further modification. The first route employs a
substitution reaction with diphenyl(piperidin-4-yl)methanol (7) and corresponding substituted
chloro-phenylbutanones (8) followed by subsequent reduction of the ketone intermediate (9)
yielding analogs 1a-h and 1j-l (Scheme 1). An alternate pathway was used to synthesize analog
1i in which the methyl 4-(4-chlorobutanoyl)benzoate 8i was prepared according to a previously
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
9
reported procedure, reduced, and subjected to a Finkelstein reaction with 7 to yield the desired
analog (Scheme 2). This ester was then hydrolyzed to the corresponding carboxylic acid 1m.
Compound 1n was synthesized by Suzuki-Miyaura coupling using a procedure adapted from
30
Moseley et al (Scheme 3A). The final analog in this set, the known metabolite of terfenadine
3
1
(
1p also known as fexofenadine), was generated according to a previously published
3
2
procedure (Scheme S2 in supporting information).
Scheme 1. General synthetic route for terfenadine (1a) and analogs series 1.
Reagents and conditions: (a) NaHCO , 2-butanone/water, 85 °C, 16 h, 23-95%; (b) NaBH ,
3
4
MeOH, rt, 3 h, 52-95%.
Scheme 2. Synthetic route for analogs 1i and 1m.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 83
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
Reagents and conditions: (a) 1,3-propanedithiol, CH Cl , rt, 1.5 h then BF ·OEt , 0 °C – rt, 18 h,
2
2
3
2
8
6%; (b) NaHMDS, THF, -78 °C then 1-chloro-3-iodopropane, rt, 18 h, 31%; (c)
bis(trifluoroacetoxy)iodobenzene, CH CN/water, rt, 1 h, 69%; (d) NaBH , MeOH, rt, 3 h, 87%;
3
4
(e) NaHCO , NaI, CH CN, reflux, 18 h, 37%; (g) LiOH, THF/water, rt, 3 h, 47%.
3
3
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 3. Synthetic routes for analogs 1n (A), 3l-3n (B).
Reagents and conditions: (a) R-B(OH) , K CO , CH CN/water, 60 °C, 18 h, 30 – 88%; (b)
2
2
3
3
NaBH , MeOH, rt, 3 h, 94%;
4
The second route to a majority of analogs was via nucleophilic substitution in which 7
was coupled with various substituted phenyl alkyl halides or tosylates (10) yielding analogs 2a-
d, 3a-i, 4a-r, 4t, 4w and 4y-bb (Scheme 4). Benzyl bromides were not available for four desired
analogs, thus reductive amination was employed for analogs 4s, 4u, 4v and 4x with the
corresponding aldehydes 11a-d (Scheme 4). A number of analogs required further modification
such as reduction of the 4-nitro group of 3i providing the 4-amino derivative 3j followed
subsequent di-methylation yielding the 4-dimethylamino analog 3k (Scheme 5). Analogs 3l-3n
were synthesized from 3e using the aforementioned Suzuki-Miyaura cross-coupling procedure
(Scheme 3B). Saponification of methyl esters 4y-4aa resulted in carboxylic acids 4cc-4ee. A set
of 5-membered heterocycles at the 4-position were synthesized in which the 4-iodo derivative
4bb was subjected to Suzuki-Miyaura cross coupling with the corresponding boronic acids or
ACS Paragon Plus Environment
6

Page 7 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
potassium trifluoroborate salts to provide 4ff-4ii while a copper mediated Ullman type coupling
3
3
of pyrrole in acetonitrile afforded the pyrrole derivative 4jj (Scheme 6).
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 4. General synthetic route for analog series 2, 3 and 4.
Reagents and conditions: (a) 7, K CO or Et N, acetonitrile, reflux,18 h, 20-94%; (b) For analogs
2
3
3
4
s, 4u, 4v and 4x: 7, Na(OAc) BH, THF, rt to 65 °C, 3 h, 16-29%.
3
Scheme 5. Synthetic route for analogs 3j and 3k.
Reagents and conditions: (a) Raney nickel, NaBH , CH Cl /MeOH, 0 °C – rt, 28 h, 61%; (b)
4
2
2
paraformaldehyde, NaBH CN, AcOH, rt, 20 h, 84%.
3
Scheme 6. Synthetic route for analogs 4ff-4jj
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 83
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
Reagents and conditions: For 4ff and 4hh (a) R-B(OH) , K CO , CH CN/water, 60 °C, 18 h, 85
2
2
3
3
–
89%; For 4gg and 4ii (b) R-BF K, Pd(OAc) , RuPhos, NaHCO , 100 °C, W, 60 min, 68 –
3
2
3
69%; For 4jj (c) Cu, pyrrole, Cs CO , CH CN, 80 °C, 21 h, 47%.
2
3
3
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
Compounds 2e and 2f were prepared by alkylation of 7 with 3-bromopropan-1-ol
providing the alcohol intermediate 12 which was subsequently tosylated and subjected to
substitution conditions with 4-tert-butylaniline or 4-tert-butylphenol respectively (Scheme 7).
Analog 2g was synthesized via reductive amination of 9a to arrive at the benzylic amine. Each
enantiomer for terfenadine, 2h and 2i, were synthesized via a previously reported procedure
3
4
(supporting information).
Scheme 7. Synthetic route to analogs 2e, 2f and 6.
Reagents and conditions: 3-bromopropanol, Et N, acetonitrile, reflux, 3 h, 65%; (b) For 2e and
3
2
f: TsCl, Et N, CH Cl , rt, 20 h, 33%; (c) For 2e, 4-tert-butylaniline; for 2f, 4-tert-butylphenol,
3 2 2
Et N, acetonitrile, reflux, 18 h, 15-45%; (d) For 6: 4-phenylphenol, triphenylphosphine, DIAD,
3
THF, rt, 18 h, 77%.
ACS Paragon Plus Environment
8

Page 9 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
35
A slightly more polar tert-butyl isostere, methyl oxetane, was prepared using an adapted
36
procedure from Wuitschik et al. The key step being a rhodium catalyzed coupling of 4-(2-
hydroxyethyl)phenylboronic acid with the Michael acceptor intermediate. Further modification
resulted in analog 3o (Scheme 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
Scheme 8. Synthetic route for analog 3o.
Reagents and conditions: (a) BuLi, THF, 0 °C, 45 min then diethyl chlorophosphate, 0 °C, 30
min then cool to -78 °C, oxetan-3-one, 2 h, 75%; (b) [Rh(cod)Cl] , KOH, 4-(2-hydroxyethyl)
2
phenylboronic acid, 1,4-dioxane, 100 °C, W, 30 min, 76%; (c) Et N, CH Cl , TsCl, rt, 20 h,
3
2
2
2
7%; (d) 7, Et N, CH CN, reflux, 18 h, 49%; (e) Mg, MeOH, 50 °C, 4.5 h, 25%.
3 3
Analogs 5a and 5b (Figure 2) were created utilizing the same substitution and subsequent
reduction conditions described for previous series. The diphenyl piperazine intermediate was
3
7
synthesized using a previously reported procedure followed by substitution with 8a and
reduction to arrive at 5c. Finally, 5d and 5e were created by reducing the alcohol of 7 in either
3
8
TFA alone or in the presence of sodium borohydride. The reduced intermediates were then
subjected to the same substitution/reduction procedure as described for previous series
(
supporting information).
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 83
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
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
Figure 2. Analogs 5a-5e displaying modifications to diphenyl piperidine region.
The final analog combined favorable changes to the linker and pendant phenyl in an
attempt to further enhance potency. This compound was synthesized via Mitsunobu chemistry in
3
9
which the intermediate 12 was coupled with 4-phenylphenol providing 6 (Scheme 7).
Design and SAR for Terfenadine Analogs
The SAR for all terfenadine analogs was studied using antimicrobial potency toward
planktonic S. aureus UAMS-1 cells, as measured by minimum inhibitory concentration testing
(MIC). Data from these assays helped drive iterative analog design and synthesis. Mechanism
of action studies on terfenadine were performed in parallel. The most promising analogs were
carried into MIC assays testing across a spectrum of Gram-positive and Gram-negative
pathogens as well as M. tuberculosis. Furthermore, a common strategy for reducing hERG
4
0,41
activity is to lower the logP of the compound
thus attempts at incorporating polar functional
groups while optimizing for potency were undertaken with this goal in mind.
Series 1 compounds were designed to gain initial SAR information at the 4-position of
the pendant phenyl (red region in Figure 1) of 1a while preserving the linker (Table 1). A scan
of lipophilic steric bulk showed that anti-S. aureus activity correlated with the size of the
lipophilic groups (4-tert-butyl (1a) > 4-iso-propyl (1b) > 4-methyl (1c) > 4-H (1d)). A halogen
ACS Paragon Plus Environment
1
0

Page 11 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scan at the para-position also displayed a similar trend (4-bromo (1g) > 4-chloro (1f) > 4-fluoro
(1e)). The 4-phenyl analog 1n further supports this observation as it displayed a modest increase
in potency (MIC to 8 g/mL). Unfortunately the introduction of polarity at this position resulted
in loss of antimicrobial activity as observed for the 4-methoxy (1h), 4-methyl carboxylate and its
corresponding carboxylic acid (1i and 1m) as well as for the primary metabolite of 1a,
carboxylic acid 1p and the methyl ester precursor 1o (Table 1). These results suggest
lipophilicity and steric bulk at the 4-position of the pendant phenyl are optimal for anti-
staphylococcal activity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. MIC values for Series 1 and 2.
S. aureus
Compound
n
X
R
MIC
g/mL)
16
(
1
1
a
b
3
3
3
3
3
3
3
3
3
4
2
1
3
3
3
3
3
2
1
0
3
3
3
3
3
3
2
1
0
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
CH(OH)
tert-butyl
iso-propyl
CH
H
32
1c
1d
3
128
>256
128
64
1
e
F
Cl
Br
1
f
1
g
32
1h
OMe
>256
>256
16
8
16
>256
8
>256
>256
8
8
8
8
8
8
16
16
16
1
i
CO
2
Me
1
j
tert-butyl
tert-butyl
tert-butyl
1
k
1
l
1m
CO
phenyl
C(CH CO
2
C(CH CO H
2
H
1n
1o
1p
2a
3
)
2
2
Me
3
)
2
CH
CH
CH
CH
NH
O
2
2
2
2
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
tert-butyl
2b
2
c
2
d
2
e
2f
2
2
g
h
CH(NH
CH(OH) (S)
CH(OH) (R)
2
)
2
i
9
a
CO
CO
CO
CO
8
32
>256
>256
9k
9
l
9
m
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 83
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 second set of analogs was designed to study modifications to the linker (green region
in Figure 1) while maintaining the 4-tert-butyl group on the pendant phenyl. Compounds 1j-1l
scanned linker length while maintaining the benzylic alcohol and showed that changes to the
length were tolerated and did not reduce activity (MICs = 8 and 16 g/mL). A study of linker
length while altering the oxidation state on the benzylic carbon provided similar results. Analogs
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
a-2d all contained fully reduced benzylic carbons, scanning from four to one carbon linkers
respectively, and provided MICs = 8 g/mL. The ketone precursor to the hit (9a) maintained
activity (MIC = 8 g/mL) while the three-carbon derivative (9l) showed a slight reduction in
activity compared to 1a (MIC = 32 g/mL). The ketone containing two-carbon and amide
containing linkers resulted in a loss of activity possibly due to restriction of movement for the
phenyl group, at least in terms of amide linker. Further exploration of the necessity of the
alcohol led to a benzylic amino derivative (2g) and the S-OH (2h) and R-OH (2i) enantiomers,
all of which exhibited no change in activity compared to the hit. The final two analogs in this set
replaced the benzylic carbon with nitrogen (2e) and oxygen (2f) resulting in slight increases in
activity to MICs = 8 g/mL for each. The SAR of the linker region suggests the secondary
alcohol is not necessary for S. aureus antimicrobial activity while shortening the linker in some
cases led to a small, yet consistent, improvement in activity.
Given the modest success of the one and two carbon linkers from the previous set the
next two series further studied SAR on the pendant phenyl while utilizing shorter, more
accessible linkers. The two carbon linker set provided mixed results as far as potency but
maintained the trend of favoring lipophilic bulk at the 4-position (Table 2). Moving the tert-
butyl group around the pendent phenyl (3a and 3b, MIC = 16 g/mL for each) was tolerated and
did not alter activity significantly compared to 2c. The same was observed for bulkier
derivatives as the 4-bromo (3e), 4-trifluoromethyl (3f) displayed MICs = 16 g/mL for each, and
ACS Paragon Plus Environment
1
2

Page 13 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
-phenyl analog 3l had an MIC = 8 g/mL. Smaller and more polar substitutions in this set led
to varying ranges of reduction in activity or a complete loss.
These trends continued in the one-carbon linker series as the 4-phenyl analog 4g was the
most active with an MIC = 4 g/mL, a 4-fold increase in potency compared to the hit (Table 2).
Unfortunately, all polar modifications at the 2-, 3-, or 4-positions led to a significant reduction or
complete loss of S. aureus antimicrobial activity. A 4 to 8-fold reduction in activity was
observed for heterocycles at the para-position as indicated by the 3- and 2-thiophene analogs 4ff
and 4gg, 3- and 2-furan analogs 4hh and 4ii and the N-linked pyrrole 4jj, compared to 4g
suggesting the slight increase in polarity for these common phenyl isosteres affect their activity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. MIC values for Series 3 and 4.
Compound
n
R
S. aureus MIC
Compound
n
R
S. aureus MIC
(g/mL)
>256
>256
>256
>256
>256
>256
128
(g/mL)
3
a
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
3-tert-butyl
2-tert-butyl
4-F
4-Cl
4-Br
4-CF
4-OMe
16
16
128
32
16
16
>256
64
4k
4l
4m
4n
4o
4p
4q
4r
4s
4t
4u
4v
4w
4x
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3-CN
2-CN
4-CF
3-CF
2-CF
4-F
3-F
2-F
4-OMe
3-OMe
2-OMe
4-OH
3-OH
2-OH
3b
3
c
3
3
d
3
3
e
3
3
f
3
3g
a
3
h
3-pyridyl-4-tert-butyl
>256
128
128
128
128
3
i
4-NO
4-NH
4-N(CH
4-phenyl
4-(4-pyridyl)
4-(3-pyridyl)
2
64
3
j
2
>256
>256
8
32
32
>256
32
>256
32
128
128
128
4
3
k
3
)
2
3l
3
m
>256
64
3
3
n
o
4-C(CH
2
OCH
2
)CH
3
4y
4z
4-CO
3-CO
2-CO
2
2
2
Me
Me
Me
>256
>256
>256
>256
>256
>256
16
32
32
32
32
4a
3-tert-butyl
2-tert-butyl
4-iso-propyl
4-Me
3-Me
2-Me
4-phenyl
3-phenyl
2-phenyl
4-CN
4
b
4aa
4cc
4dd
4ee
4ff
4gg
4hh
4ii
4
c
4-CO
3-CO
2-CO
2
2
2
H
H
H
4
d
4
e
4f
4-(3-thiophene)
4-(2-thiophene)
4-(3-furan)
4-(2-furan)
4-(1-pyrrole)
4
4
g
h
64
>256
>256
4
i
4
j
4jj
a
Analog 3h employs a pendant 3-pyridine in place of the pendant phenyl
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 83
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
A small set of analogs were designed to explore more global changes to the piperidinyl
diphenylmethanol (blue region in Figure 1) side of 1a (Figure 2). Analogs 5a-5d showed a
complete loss of activity toward S. aureus; however, the removal of the alcohol while
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
3
maintaining sp character of the carbon (5e) was tolerated (Table 3). This provides some
evidence, however small, that future work could establish SAR trends on this side of the
molecule.
Table 3. MIC values for series 5
Compound S. aureus MIC (g/mL)
5
5
5
5d
5e
a
b
c
>256
>256
>256
>256
8
In summary, there appears to be a trend in which lipophilicity may be driving the potency
of this scaffold, at least in general. A scatter plot of logP vs MIC (Chart S1 in supporting
information) displays this trend with the average logP generally decreasing as potency decreases.
However, there are a few analogs that lie outside of the trend, such as 4g, which has a lower
4
2,43
calculated logP (estimated by ALOGPS)
than 1a but displays a 4-fold increase in potency.
Therefore, while the trend is general there are a few instances in which functional group identity
or placement still might play a role in the potency. A strong relationship was noticed at the 4-
position of the pendant phenyl with lipophilic bulk being beneficial for potency against S.
aureus. Downsizing the tert-butyl group of the hit 1a resulted in a reduction of potency, however
slight, on all occasions while the addition of a phenyl group in this position was the lone
substitution leading to a modest increase of potency.
The addition of hetercyclic isosteres (pyridinyl, thiophenyl and furyl moieties) are
comparable in size to the phenyl, however, contain slightly larger polar surface areas. The
ACS Paragon Plus Environment
1
4

Page 15 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
benefit of steric bulk in these cases may have been mostly negated due to increased polarity,
however slight, ultimately resulting in an overall reduction of potency but not a complete loss.
Modifications, either polar or non-polar, to other positions of the pendant phenyl were
unfavorable leading to reduction or loss of potency. Therefore SAR on the pendant phenyl,
while relatively steep for the ortho- and meta-positions, did offer some room for improvement at
the para-position as far as potency against S. aureus is concerned. However, further optimization
to the piperidinyl diphenylmethanol side of 1a may be required for reduction of logP while
maintaining or improving potency.
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
Several modifications to the linker region provided a slight increase in potency. The
benzylic alcohol was not necessary for activity as indicated by the removal of the hydroxy group
leading to moderately increased antimicrobial potency across the four different linker lengths
(2a-2d). It was also shown that substituting an amine for the alcohol or resolving the
enantiomers had no effect on activity. Furthermore, substituting the benzylic carbon for an
amine or ether linkage resulted in a 2-fold increase in activity compared to the hit. Given these
data, it was evident that modifications to the linker could lead to an analog with enhanced
activity.
A final analog was designed to combine the favorable ether linker modification from 2f
and the 4-phenyl substitution on the pendant group to yield 6. The combination of these
modifications resulted in a synergistic boost in potency with an MIC = 1 g/mL against S.
aureus comparable to an MIC = 0.5 g/mL for the widely prescribed fluoroquinoline,
ciprofloxacin, in the same assay (Figure 3). The success of the combined modifications offers
optimism that more potent analogs may be within reach with future SAR studies. However,
compound properties such as solubility and logP remain an issue and will be addressed with
future work.
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 83
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
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
Figure 3. MIC values for ciprofloxacin, 4g and 6.
Mechanism of Action
Transcription profiling was used to compare the cellular response of S. aureus strain
UAMS-1 following challenge with a sub-inhibitory concentration (0.5 x MIC) of terfenadine,
with that of cells treated with 0.5 x MIC of the known topoisomerase II inhibitor, ciprofloxacin,
or antibiotics with independent mechanisms of action.
Results (Figure S1 in supporting
information) revealed that terfenadine-treated cells elicit a cellular response most similar to that
of ciprofloxacin and cells treated with the DNA damaging agent mitomycin C in comparison to
cells treated with sub-inhibitory concentrations of cell wall active agents, RNA synthesis
inhibitors, or general stress responses such as cold-shock conditions and metal depletion. For
instance, 221 (43%) of the 509 genes that were differentially expressed within ciprofloxacin
treated cells were also differentially expressed within terfenadine-treated cells. A total of 149
(67%) of these 221 genes were also differentially expressed within mitomycin treated cells, but
none of the other stress conditions evaluated. Of direct relevance to this work, while the
majority of these genes (55%) are annotated as hypothetical proteins two key components of the
organism’s SOS response, lexA and umuC, as well as twelve phage-associated replication
ACS Paragon Plus Environment
1
6

Page 17 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
proteins were identified and known to be responsive to the topoisomerase inhibitor,
4
4,45
ciprofloxacin.
The similarities between the cellular response of terfenadine-, ciprofloxacin-,
and mitomycin-challenged cells, provided an initial indication that the scaffold’s antimicrobial
activity may be mediated via topoisomerase II inactivation.
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
As a direct test of whether the antimicrobial properties of terfenadine correlate with type II
topoisomerase inhibition, S. aureus DNA gyrase and topoisomerase IV inhibition assays were
used to measure each enzyme’s activity in the absence and presence of various concentrations of
terfenadine. As shown in Figure 4A (lanes 1 and 2), the addition of 0.5 units (U) of S. aureus
DNA gyrase catalyzed supercoiling of relaxed circular plasmid pBR322 DNA, resulting in
increased substrate mobility in an agarose gel, confirming that the assay conditions were
appropriate to measure gyrase activity. Similar assays in the presence of increasing
concentrations of terfenadine (62.5, 125, 250 and 500 µM), demonstrated that the compound was
a relatively mild inhibitor of S. aureus gyrase with an apparent IC value of 190 µM. Similarly,
50
terfenadine appeared to exhibit moderate topoisomerase IV inhibitory activity (Figure 5A).
Figure 4. DNA gyrase supercoiling inhibition assays (A) dose-response gel for 1a; (B) gel
showing inhibition of DNA gyrase with analogs 1a, 4g and 6 with the positive control
ciprofloxacin at 100 M.
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 83
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
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
Figure 5. Topoisomerase IV inhibition assay (A) dose-response gel for 1a; (B) gel showing
inhibition of topoisomerase IV with analogs 1a, 4g and 6 with the positive control ciprofloxacin
at 100 M.
While these values are admittedly modest they are comparable to the topoisomerase II
inhibitor, ciprofloxacin, in these assays, which displayed IC values of 110 and 4 µM for S.
5
0
aureus DNA gyrase and topoisomerase IV, respectively (data not shown). This, combined with
the fact that 1a did not appear to affect the activity of other enzymes assessed (S. aureus RNase
J2 and RnpA; data not shown), supports the hypothesis that the compound’s antimicrobial
activity may, in part, correspond to its topoisomerase II inhibitory activity.
Accordingly, IC values were measured for the first 48 analogs synthesized (Table 4) to
5
0
determine whether improvements to the compound’s antimicrobial activity tracked with
enhanced enzyme inhibition. Results of those studies revealed two overarching features of the
analog series. First, most compounds (11 of 12 in total) displaying modest improvement in
ACS Paragon Plus Environment
1
8

Page 19 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antimicrobial activity (MIC ≤ 8 g/mL), in comparison to terfenadine, also exhibited at least
modest improvement in potency toward S. aureus DNA gyrase and/or topoisomerase IV.
Second, most compounds exhibiting reduced antimicrobial activity (MIC ≥ 32 g/mL) displayed
decreased potency toward the enzymes. Compounds 4g and 6, which exhibited the most
improved antimicrobial properties toward S. aureus (MIC values of 4 g/mL and 1 g/mL,
respectively) also displayed modest improvements toward S. aureus DNA gyrase inhibition
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
(
IC s = 130 and 50 M, respectively) when compared to 1a. Indeed, as shown in Figure 4B
5
0
when tested at 100 M, 4g, 6 and ciprofloxacin, all displayed improved inhibition of S. aureus
DNA gyrase supercoiling activity, in comparison to the parent molecule, terfenadine. This
suggests that even if the gains in potency for 4g and 6 over terfenadine are modest, they do
appear to display a real improvement in DNA gyrase inhibition. Moreover, both derivatives
maintained S. aureus topoisomerase IV inhibition activity that approximated the inhibitory
activity of that of the parent molecule, 1a (Figure 5B).
Even though the terfenadine scaffold seems to be inhibiting the type II topoisomerases, it
appears there may be other mechanisms contributing to the overall bactericidal effect. It is well
4
6
noted that receptor promiscuity is correlated with lipophilicity. Therefore, it is not out of the
question that these compounds could be binding to multiple targets and the combined effect of
these targets may be exerting the antibacterial effect. Analogs 2g, 2i, 3d, 3e and 3n all are
shown to have activity versus S. aureus, with MICs ranging from 16 – 32 g/mL. However,
these analogs possess reduced or no activity in the DNA gyrase and topoisomerase IV assays.
Therefore, they must be exerting an antimicrobial effect via other means, supporting the multiple
target hypothesis. Moreover, attempts to isolate resistant strains of S. aureus provided no stable
resistant organisms also supporting a multiple target hypothesis. It should be noted that given the
above correlations, expense and laborious nature of the enzyme assays, and fact that no
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 83
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
improvements in antimicrobial activity (≥ 32 g/mL) were observed for the remaining
compounds synthesized, their IC values were not measured.
5
0
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
Table 4. DNA Gyrase and Topoisomerase IV IC Measurements
5
0
Compound
DNA Gyrase IC50
Topoisomerase
IV IC50 (M)
4
Compound
DNA Gyrase IC50
Topoisomerase
IV IC50 (M)
100
(
M)
110
190
127
(M)
127
410
93
90
73
Ciprofloxacin
a (terfenadine)
2h
2i
9a
9k
9l
9m
3a
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
4g
5a
5b
6
1
206
273
110
110
133
247
>333
>333
>333
267
1
b
1c
>500
>500
>500
133
>500
>500
440
100
133
73
>500
13
>500
>500
93
>333
>333
>333
>333
>333
>333
>333
133
1
d
1
e
93
93
1
f
1g
1h
>333
260
247
17
>333
>333
263
93
>333
80
93
250
>333
125
>333
>333
50
1
i
260
1
j
>333
>333
>333
>333
>333
>333
133
1
k
333
100
1l
1
m
>333
100
>333
>333
320
100
147
213
250
1
1
1
n
o
p
2a
227
280
2
b
93
93
90
193
100
>500
2
c
>333
160
>333
>333
160
2
d
2e
2f
>333
>333
2
g
DNA Gyrase Modeling Studies
As was shown in the previous section, the terfenadine-based scaffolds are likely to, at
least in part, exert their antibacterial effect via inhibition of the type II topoisomerases. Recently
1
6
NBTIs, such as 43 have been reported. These NBTIs also feature a piperidine with aromatic
regions linked to the N- and 4-position. Therefore, we hypothesized that the terfenadine-based
analogs may be binding in the same region as 43. A docking study was carried out using the
software Surflex module of SYBYL from Certara. The receptor protein (PDB ID 2XCS) was
prepared by removing the ligand, 43, and taking the corresponding ligand binding pocket defined
as a 20 Ǻ sphere around the GSK ligand. Out of 30 poses the best pose was selected on the basis
ACS Paragon Plus Environment
2
0

Page 21 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of Combined CScore (ChemScore, G_Score, D_Score and PMF_Score) and the scores were 8.85
for 43, 7.65 for 6, and 5.99 for 1a, respectively. The scores generally correlated with the
experimentally determined DNA gyrase activity (43 = 14 nM, 6 = 50 M and 1a = 190 M
respectively).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To assess the potential of the terfenadine analogs for binding to other known sites on
DNA gyrase, the same docking study was performed with a solved structure of ciprofloxacin
1
6
(
PDB ID 2XCT) and the docking score of the best poses for 43 and 6 were 2.22 and 0.72
respectively. The same modeling was performed at the ATP site on the GyrB subunit, in which a
pyrrolamide ligand was extracted from the ATP-binding site in a structure published by Eakin et
4
7
al (PDB ID 3U2K). Both 43 and 6 were docked and provided similar docking scores (5.05 and
.23, respectively). These in silico results suggest a higher likelihood of the terfenadine-based
5
analogs interacting in the same region as 43 than in the other two known inhibitor binding sites.
Information from the above modeling of binding site suggests that the aromatic character
of the pendant bi-phenyl moiety intercalates between DNA basepairs similar to 43 (Figure 6A).
However, due to the length of the linker, the diphenyl methanol portion may be able to interact
with a section of the protein surface adjacent to the pocket in which the oxathiolo-pyridine group
of 43 binds. The best pose of 6 suggests that hydrogen bonds may form between the alcohol
with an acceptor phosphate group in the DNA backbone, and a donor Arg1122, respectively,
with additional potential hydrophobic π-π stacking interactions (Figure 6B). While the modeling
suggests a similar binding site, further work is needed to validate these results. Future plans
include designing analogs to interrogate these possible interactions, competition assays with
known inhibitors and solving a terfenadine-based ligand bound DNA gyrase structure aimed at
providing more definitive evidence as to the binding site for this scaffold.
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 83
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
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
Figure 6. Docking studies utilizing the ligand-bound DNA gyrase structure PDB ID 2XCS (A)
Overlay of solved ligand, 43 (purple ligand), and best docked pose for 6 (green ligand) within
binding site. Protein surface shaded yellow and DNA surface shaded blue. (B) Potential
interactions from the best docked pose of 6 (light blue ligand). Docking studies in two known
alternative binding sites, the ATP and ciprofloxacin binding sites, were also performed using
solved ligand-bound structures (PDB IDs 3U2K and 2XCT, respectively) and the binding score
were inferior compared to the 43 binding site.
Antimicrobial Spectrum of Activity of Terfenadine, 4g and 6
As stated above, terfenadine (1a) was found to have anti-staphylococcal activity in a HTS
campaign using the S. aureus strain UAMS-1, a methicillin-susceptible strain. In order to
determine the spectrum of antimicrobial activity for 1a, the activity of the scaffold toward
genetically diverse S. aureus strains, other Gram-positive and Gram-negative bacterial species of
immediate healthcare concern, and Mycobacterium tuberculosis was determined. As shown in
ACS Paragon Plus Environment
2
2

Page 23 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 5, each compound’s antimicrobial activity was conserved across all Gram-positive strains
evaluated. More specifically, the MIC values of 1a, 4g, and 6 were 16, 4, and approximately 2
g/mL, respectively toward methicillin-resistant (MRSA), vancomycin-intermediate (VISA),
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
vancomycin-resistant (VRSA) S. aureus, as well as flouroquinolone resistant strains. Likewise,
they demonstrated corresponding antimicrobial activity toward Enterococcus faecium,
Enterococcus faecalis, and M. tuberculosis. Neither 1a nor its analogs were active against wild-
type Gram-negative species tested thus far. The MIC is >256 g/mL for both K. pneumoniae
CKP4 and E. coli 8314. The MIC value for A. baumannii 98-37-09 is 256 g/mL. However,
when tested versus a membrane-compromised, efflux pump-deficient strain of E. coli (tolC,
imp), all three compounds did show activity. Therefore, the lack of efficacy versus Gram-
negative species may be due to an inability of these compounds to traverse the outer membrane
to gain entry to the cellular target and/or a result of expulsion via efflux pumps. Nonetheless,
these results indicate that a terfenadine scaffold could potentially be optimized for use as an
antimicrobial against Gram-positive organisms and potentially Gram-negative organisms if
treated with an efflux pump inhibitor.
Table 5. Antimicrobial Spectrum for terfenadine, 4g and 6.
Terfenadine (1a)
4g
6
Strain (Relevant Resistance)
MIC (μg/mL)
MIC(μg/mL)
MIC (μg/mL)
S. aureus U1
16
16
16
16
16
16
16
16
4
4
4
4
8
8
8
1
2
2
2
2
2
2
S. aureus CRC61(Ciprofloxacin)
S. aureus CRC118 (Ciprofloxacin)
S. aureus USA300 NRS-384 (MRSA)
S. aureus USA 300-0114 (MRSA)
S. aureus Mu50 (VISA)
S. aureus VRSA-1 (VRSA)
E. faecium 824-05
8
2
E. faecalis OG1RF
M. tuberculosis mc 6020
16
16
4
4
2
1
2
A. baumannii 983709
K. pneumoniae CKP4
E. coli 8314
256
>256
>256
8
>256
>256
>256
2
>128
>128
>128
4
E. coli (tolC, imp-)
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 83
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
hERG Activity of Terfenadine Analogs
Compounds 4g and 6 emerged as the most promising analogs with MICs = 4 and 1
g/mL, respectively, against S. aureus. Unfortunately, the SAR did not allow for addition of

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
polar functionality and the predicted logP for each compound remains relatively high when
compared to the hit. The predicted logP values for 1a, 4g and 6 are 5.89, 5.73 and 6.30,
respectively (estimated by ALOGPS). Therefore, it appears no improvement was able to be
gained in logP while optimizing for potency in this set of analogs. This observation carries over
to the measured hERG activity for each compound. Terfenadine (1a) displayed a hERG IC =
5
0
1
30 nM while analogs 4g and 6 display hERG IC s = 210 and 140 nM respectively. While
50
hERG activity was not able to be reduced during this campaign, future SAR work on the
piperdinyl diphenylmethanol side of the molecule may still provide opportunities for addition of
polarity or reduction of the pK of the molecule, thus reducing logP and potentially reducing
a
hERG activity.
Conclusions
The project commenced with a phenotypic whole-cell HTS against the ESKAPE
pathogens using a library of off-patent FDA-approved drugs. The HTS identified the
antihistamine, terfenadine (1a), as a hit possessing previously unreported antimicrobial activity
versus S. aureus (MIC = 16 g/mL). In an effort to repurpose 1a, a total of 84 terfenadine-based
analogs were designed and synthesized for optimized antimicrobial activity versus S. aureus.
Analogs 4g and 6 displayed improved activity (4 and 1 g/mL respectively) compared to 1a and
promising activity toward other bacterial pathogens of immediate healthcare concern. The SAR
study revealed bulky lipophilic substituents at the 4-position of the pendant phenyl, shortening
the linker and/or replacing the benzylic carbon with oxygen enhance activity.
ACS Paragon Plus Environment
2
4

Page 25 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Mechanism of action studies suggest 1a, 4g and 6 are likely acting as bacterial type II
topoisomerase inhibitors targeting both DNA gyrase and topoisomerase IV. However, there is
evidence to suggest this scaffold may have multiple targets. Docking studies show these analogs
may bind within the same site as the NBTI 43. While antimicrobial activity versus S. aureus
strains tested, including flouroquinolone, methicillin and vancomycin resistant isolates, and M.
tuberculosis was improved, the hERG liability and physicochemical properties for this class of
compounds still remains an issue. We believe 6, and possibly 4g, provide good starting points
for further optimization to the piperidinyl diphenylmethanol side of the molecule. Furthemore, a
strategy has been devised to improve antimicrobial activity while reducing hERG activity. In
future work we also plan to confirm the possible DNA gyrase binding site for these analogs via
competition assays and crystallography studies.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry General Experimental
Purity of all final compounds was confirmed by HPLC/MS analysis and all compounds
1
13
have a final purity of ≥ 95% purity. H and C NMR spectra were recorded on a Bruker AM 400
spectrometer (operating at 400 and 101 MHz respectively) or a Bruker AVIII spectrometer
(
operating at 500 and 126 MHz respectively) in CDCl with 0.03% TMS as an internal standard
3
or DMSO-d . The chemical shifts () reported are given in parts per million (ppm) and the
6
coupling constants (J) are in Hertz (Hz). The spin multiplicities are reported as s = singlet, bs =
broad singlet, d = doublet, t = triplet, q = quartet, dd = doublet of doublet and m = multiplet. The
LCMS analysis was performed on an Agilent 1200 RRL chromatograph with photodiode array
UV detection and an Agilent 6224 TOF mass spectrometer. The chromatographic method
utilized the following parameters: a Waters Acquity BEH C-18 2.1 x 50 mm, 1.7 m column;
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 83
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
UV detection wavelength = 214 nm; flow rate = 0.4 mL/min; gradient = 5 - 100% acetonitrile
over 3 minutes with a hold of 0.8 minutes at 100% acetonitrile; the aqueous mobile phase
contained 0.15% ammonium hydroxide (v/v). The mass spectrometer utilized the following
parameters: an Agilent multimode source which simultaneously acquires ESI+/APCI+; a
reference mass solution consisting of purine and hexakis(1H, 1H, 3H-tetrafluoropropoxy)
phosphazine; and a make-up solvent of 90:10:0.1 MeOH:Water:Formic Acid which was
introduced to the LC flow prior to the source to assist ionization.
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
General Method A:
1
-(4-(Tert-butyl)phenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one (9a). To
a vial was added diphenyl(piperidin-4-yl)methanol 7 (1.19 g, 4.45 mmol), 1-(4-(tert-
butyl)phenyl)-4-chlorobutan-1-one 8a (1.01 g, 4.24 mmol), sodium bicarbonate (0.42 g, 5.09
°
mmol) with water and 2-butanone (18 mL, 1:5). The reaction stirred at 85 C for 16 h and was
then allowed to cooled to rt and diluted with water (50 mL). The reaction was diluted with water
(20 mL) and extracted with EtOAc (3 x 50 mL). The organic layers were combined then dried
with MgSO , filtered, concentrated and purified by MPLC (0 - 10% MeOH:CH Cl ) to produce
2
2
4
1
pure 9a (1.37 g, 2.92 mmol, 69% yield) as a colorless oil. H NMR (500 MHz, CDCl ): δ 7.90
3
(d, J = 8.4 Hz, 2H), 7.49 – 7.45 (m, 6H), 7.31 – 7.25 (m, 4H), 7.20 – 7.14 (m, 2H), 2.97 – 2.92
(m, 4H), 2.45 – 2.36 (m, 3H), 2.09 (br s, 1H), 2.00 – 1.87 (m, 4H), 1.49 – 1.32 (m, 4H), 1.34 (s,
13
9
1
H). C NMR (125 MHz, CDCl ): δ 199.7, 156.5, 146.0, 134.5, 128.1, 128.0, 126.4, 125.7,
3
25.4, 79.4, 57.9, 43.9, 43.4, 44.1, 36.2, 35.0, 31.0, 26.2, 21.9. LCMS Retention time: 4.207 min.
+
LCMS purity 99.5%. HRMS (ESI): m/z calcd for C H NO [M+H] 470.3053, found 470.3076.
32
39
2
General Method B:
-(4-(Tert-butyl)phenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-ol (1a). To a
1
vial was added 9a (0.20 g, 0.422 mmol) and MeOH (2 mL). The sodium borohydride (0.032 g,
ACS Paragon Plus Environment
2
6

Page 27 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
0
.844 mmol) was then added and the reaction stirred at rt for 3 h. The reaction was concentrated,
water (5 mL) was added and a white precipitate formed. The precipitate was filtered out and
then dissolved in CH Cl (10 mL), dried with MgSO , filtered and concentrated to produce pure
2
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
1
1
a (0.15 g, 0.320 mmol, 76% yield) as on oil. H NMR (400 MHz, CDCl ): δ 7.52 – 7.46 (m,
3
4H), 7.33 – 7.25 (m, 8H), 7.21 – 7.15 (m, 2H), 4.61 – 4.56 (m, 1H), 3.16 – 3.11 (br m, 1H), 3.00
–
6
1
2
2.94 (m, 1H), 2.51 – 2.34 (m, 4H), 2.10 – 1.88 (m, 4H), 1.83 – 1.75 (m, 1H), 1.70 – 1.45 (m,
1
3
H), 1.30 (s, 9H). C NMR (125 MHz, CDCl ): δ 149.4, 146.1, 146.0, 142.7, 128.2, 128.1,
3
26.4, 126.3, 125.7, 125.6, 125.3, 125.0, 79.2, 73.4, 58.9, 54.7, 53.3, 44.2, 39.7, 34.4, 31.4, 26.0,
5.9, 24.1. LCMS Retention time: 4.137 min. LCMS purity 97.5%. HRMS (ESI): m/z calcd for
+
C H NO [M+H] 472.3209, found 472.3219.
32
41
2
General Method C:
1-([1,1'-Biphenyl]-4-ylmethyl)piperidin-4-yl)diphenylmethanol (4g). To a vial was added the
(
4-phenylbenzyl bromide 10t (0.10 g, 0.40 mmol), 7 (0.098 g, 0.37 mmol) and acetonitrile (3
mL). The triethylamine (0.077 mL, 0.55 mmol) was added and the reaction stirred at 70 °C for 4
h. The reaction was cooled to rt and diluted with water then extracted with EtOAc. The EtOAc
layer was concentrated and the crude product was purified by reverse-phase MPLC (10 - 100%
1
MeCN:water) to produce the pure 4g (0.12 g, 0.27 mmol, 72% yield). H NMR (400 MHz,
CDCl ): δ 7.61 – 7.55 (m, 2H), 7.55 – 7.51 (m, 2H), 7.51 – 7.46 (m, 4H), 7.46 – 7.39 (m, 2H),
3
7
2
.39 – 7.32 (m, 3H), 7.32 – 7.26 (m, 4H), 7.22 – 7.12 (m, 2H), 3.55 (s, 2H), 3.04 – 2.80 (m, 2H),
13
.50 – 2.36 (m, 1H), 2.11 – 1.88 (m, 3H), 1.56 – 1.43 (m, 4H). C NMR (125 MHz, CDCl ): δ
3
145.9, 140.9, 139.9, 137.1, 129.6, 128.7, 128.1, 127.1, 127.0, 126.8, 126.4, 125.7, 79.5, 62.8,
5
3.9, 44.1, 26.4. LCMS Retention time: 4.101 min. LCMS purity 98.5%. HRMS (ESI): m/z
+
calcd for C H NO [M+H] 434.2478, found 434.2485.
3
1
31
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 83
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
-(4-(Iso-propyl)phenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one
(9b).
Method A: 7 (0.62 g, 2.32 mmol), 4-chloro-1-(4-iso-propylphenyl)butan-1-one (8b) (0.50 g, 2.21
mmol), sodium bicarbonate (0.22 g, 2.65 mmol) with water and 2-butanone (15 mL, 1:5) to
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
produce pure 9b (0.48 g, 1.06 mmol, 48% yield) as a colorless oil. H NMR (500 MHz, CDCl ):
3
δ 7.89 (d, J = 8.3 Hz, 2H), 7.51 – 7.42 (m, 4H), 7.33 – 7.26 (m, 6H), 7.20 – 7.13 (m, 2H), 2.99 –
2
.90 (m, 4H), 2.46 – 2.33 (m, 3H), 2.08 (s, 1H), 2.00 – 1.85 (m, 4H), 1.61 (s, 1H), 1.50 – 1.33
m, 4H), 1.27 (d, J = 6.8 Hz, 6H).
1-(4-(Methyl)phenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one
(
(9c).
Method A: 7 (0.50 g, 1.87 mmol), 4-chloro-1-(4-methylphenyl)butan-1-one (8c) (0.35 g, 1.78
mmol), sodium bicarbonate (0.18 g, 2.14 mmol) with water and 2-butanone (15 mL, 1:5) to
1
produce pure 9c (0.27 g, 0.64 mmol, 36% yield) as a colorless oil. H NMR (500 MHz, CDCl ):
3

7.86 (d, J = 8.2 Hz, 2H), 7.51 - 7.43 (m, 4H), 7.32 - 7.26 (m, 4H), 7.24 (d, J = 8.0 Hz, 2H),
7.18 (tt, 2H), 3.02 - 2.83 (m, 4H), 2.46 - 2.32 (m, 6H), 2.21 (s, 1H), 2.05 - 1.81 (m, 4H), 1.57 -
1
.34 (m, 4H).
1
-Phenyl-4-(4-(Hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one (9d). Method A: 7 (0.52
g, 1.96 mmol), 4-chloro-1-phenylbutan-1-one 8d (0.30 mL, 1.87 mmol), sodium bicarbonate
(0.19 g, 2.24 mmol) with water:2-butanone (18 mL, 1:5) to produce pure 9d (0.18 g, 0.43 mmol,
1
2
3% yield) as a colorless oil. H NMR (500 MHz, CDCl ): δ 8.00 – 7.89 (m, 2H), 7.59 – 7.51
3
(m, 1H), 7.50 – 7.39 (m, 6H), 7.34 – 7.25 (m, 4H), 7.21 – 7.14 (m, 2H), 3.03 – 2.81 (m, 4H),
2
.46 – 2.32 (m, 3H), 2.09 (br s, 1H), 2.01 – 1.86 (m, 4H), 1.53 – 1.29 (m, 4H).
1-(4-Fluorophenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one (9e). Method
A: 7 (0.39 g, 1.44 mmol), 4-chloro-1-(4-fluorophenyl)butan-1-one 8e (0.28 g, 1.37 mmol),
sodium bicarbonate (0.14 g, 1.64 mmol) with water (3 mL) and 2-butanone (15 mL) 9e (0.17 g,
1
0.40 mmol, 29% yield) as a colorless oil. H NMR (400 MHz, CDCl ): δ 8.00 – 7.96 (m, 2H),
3
ACS Paragon Plus Environment
2
8

Page 29 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7
.48 – 7.44 (m, 4H), 7.31 – 7.26 (m, 4H), 7.20 – 7.15 (m, 2H), 7.14 – 7.08 (m, 2H), 2.97 – 2.94
(m, 4H), 2.46 – 2.36 (m, 3H), 2.15 (br s, 1H), 2.01 – 1.88 (m, 4H), 1.52 – 1.35 (m, 4H).
1
-(4-Chlorophenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one (9f). Method
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: 7 (0.63 g, 2.34 mmol), 4-chloro-1-(4-chlorophenyl)butan-1-one 8f (0.48 g, 2.23 mmol),
sodium bicarbonate (0.23 g, 2.68 mmol) with water:2-butanone (18 mL, 1:5) to produce pure 9f
1
(
0.39 g, 0.88 mmol, 39% yield) as a colorless oil. H NMR (400 MHz, CDCl ): δ 7.90 (d, J =
3
8
.4 Hz, 2H), 7.48 – 7.44 (m, 4H), 7.42 (d, J = 8.6 Hz, 2H), 7.31 – 7.26 (m, 4H), 7.20 – 7.15 (m,
2H), 3.96 – 2.89 (m, 4H), 2.45 – 2.34 (m, 3H), 2.08 (br s, 1H), 1.99 – 1.87 (m, 4H), 1.50 – 1.30
m, 4H).
-(4-Bromophenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one (9g). Method
(
1
A: 7 (0.44 g, 1.64 mmol), 1-(4-bromophenyl)-4-chlorobutan-1-one (8g) (0.41 g, 1.56 mmol),
sodium bicarbonate (0.16 g, 1.88 mmol) with water (3 mL) and 2-butanone (15 mL) to produce
1
pure 9g (0.25 g, 0.50 mmol, 32% yield) as a colorless oil. H NMR (400 MHz, CDCl ): δ 7.83 –
3
7
2
.79 (m, 2H), 7.60 – 7.56 (m, 2H), 7.48 – 7.45 (m, 4H), 7.31 – 7.26 (m, 4H), 7.20 – 7.15 (m,
H), 2.95 – 2.90 (m, 4H), 2.45 – 2.35 (m, 3H), 2.17 (br s, 1H), 2.00 – 1.88 (m, 4H), 1.51 – 1.33
(m, 4H).
-(4-Methoxyphenyl)-4-(4-(hydroxydiphenylmethyl)piperidin-1-yl)butan-1-one
1
(9h).
Method A: 7 (0.49 g, 1.83 mmol), 4-chloro-1-(4-methoxyphenyl)butan-1-one (8h) (0.37 g, 1.74
mmol), sodium bicarbonate (0.18 g, 2.09 mmol) with water (3 mL) and 2-butanone (15 mL) to
1
produce pure 9h (0.28 g, 0.64 mmol, 36% yield) as a colorless oil. H NMR (400 MHz, CDCl ):
3
δ 7.94 (d, J = 8.9 Hz, 2H), 7.51 – 7.44 (m, 4H), 7.32 – 7.26 (m, 4H), 7.21 – 7.12 (m, 2H), 6.92
(d, J = 8.9 Hz, 2H), 3.87 (s, 3H), 3.05 – 2.82 (m, 4H), 2.40 (q, J = 7.5, 6.6 Hz, 3H), 2.02 – 1.86
(m, 4H), 1.50 – 1.24 (m, 5H).
ACS Paragon Plus Environment
2
9