Development of a Dual-Acting Antibacterial Agent (TNP-2092) for the Treatment of Persistent Bacterial Infections

Article abstract of DOI:10.1021/acs.jmedchem.6b00485

The clinical management of prosthetic joint infections and other persistent bacterial infections represents a major unmet medical need. The rifamycins are one of the most potent antibiotic classes against persistent bacterial infections, but bacteria can

Full text of DOI:10.1021/acs.jmedchem.6b00485

Subscriber access provided by University of Sussex Library
Drug Annotation
Development of a Dual-Acting Antibacterial Agent (TNP-2092)
for the Treatment of Persistent Bacterial Infections
Zhenkun Ma, and Anthony Simon Lynch
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00485 • Publication Date (Web): 23 Jun 2016
Downloaded from http://pubs.acs.org on June 24, 2016
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 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
Development of a Dual-Acting Antibacterial Agent (TNP-2092) for the Treatment of
Persistent Bacterial Infections
7
8
9
Zhenkun Ma1* and Anthony Simon Lynch2
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
1TenNor Therapeutics Ltd., 218 Xinghu St. Suzhou Industrial Park, China 215123
2Janssen Research & Development LLC., 1400 McKean Road, Spring House, PA 18940, USA.
*Corresponding Author
ABSTRACT
The clinical management of prosthetic joint infections and other persistent bacterial infections
represents a major unmet medical need. The rifamycins are one of the most potent antibiotic
classes against persistent bacterial infections, but bacteria can develop resistance to rifamycins
rapidly and the clinical utility of the rifamycin class is typically limited to antibiotic combinations
to minimize the development of resistance. To develop a better therapy against persistent
bacterial infections, a series of rifamycin based biꢀfunctional molecules were designed,
synthesized and evaluated with the goal to identify a dualꢀacting drug that maintains the potent
activity of rifamycins against persistent pathogens and at the same time minimize the
development of rifamycin resistance. TNPꢀ2092 was identified as a drug candidate and is
currently in an earlyꢀstage of clinical development for the treatment of prosthetic joint infections.
Keywords: rifamycin, dualꢀaction, bacterial persistent infections, bacterial biofilms, prosthetic
joint infections, antibiotic resistance.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 41
1
2
3
4
INTRODUCTION
5
6
7
8
9
Prosthetic joint infections (PJIs) are difficult to treat and are often associated with significant
morbidity and substantial health care expenditures.1 The incidence of prosthesis implantations
and as a consequence, the burden of PJIs is expected to continue to rise rapidly. In the United
States, there were about 1 million total hip and knee replacement surgeries performed in 2010
and this number is projected to reach 4 million by 2030. However, the implementation of
enhanced infection prevention measures does not appear to have resulted in a significant
reduction in the PJI incident rate in the United States. In fact, the incidence rate has increased
slightly from 1.99 to 2.18% for hip and from 2.05 to 2.18% for knee replacement from 2001 to
2009, respectively.2 The management of PJIs is difficult, requiring surgical intervention
accompanied by prolonged antibiotic therapy. For earlyꢀstage infections (less than 30 days postꢀ
implantation or less than 3 weeks after the onset of infectious symptoms) with a wellꢀfixed
prosthesis without a sinus tract, a debridement and prosthesis retention strategy is advised. All
other patients should proceed to oneꢀstage or twoꢀstage prosthesis exchange (revision) surgery.
For certain patients, exchange surgery is not possible and limb amputation becomes the last
resort 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
Staphylococcus aureus and coagulaseꢀnegative staphylococci (CoNS) are responsible for the
majority of PJIs and are associated with higher treatment failure rates than observed with other
causative pathogens. The treatment strategy for staphylococcal PJIs involves a prolonged
rifampinꢀbased combination therapy, including 2ꢀ6 weeks of intravenous antibiotic therapy in
combination with oral rifampin, followed by rifampin plus a companion antibiotic for a total of 3ꢀ6
months.3 The formation of bacterial biofilms on the surface of prosthesis appears to play an
important role in the pathogenesis of PJI and is believed to be the underlying mechanism for
persistence. It is a wellꢀknown phenomenon that pathogens living in bacterial biofilms can
effectively evade host immune responses and can become tolerant to antibiotic treatment.4
ACS Paragon Plus Environment

Page 3 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PJIs are devastating to patients and have significant economic impact to the healthcare system.
Patient suffering and treatment costs increase dramatically with more invasive surgical
intervention strategies. Early diagnosis and early treatment are critical to reduce the need for
exchange surgery, permanent resection and amputation. However, the least invasive PJI
treatment strategy – debridement and retention procedure can only be applied to patients who
have a wellꢀfixed prosthesis without sinus tract involvement and within the first month of
prosthesis implantation or within 3 weeks of the onset of infection symptoms. Despite the
stringent inclusion criteria and prolonged antibiotic therapy, this treatment strategy can still be
associated with low and unreliable treatment outcomes with the overall success rate reported as
14%ꢀ100%.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
Overall, there is an urgent need for better therapies against persistent bacterial infections
associated with bacterial biofilm formation. A new antibacterial agent that is more efficacious in
eradicating pathogens living in bacterial biofilms, particularly for those formed by S. aureus and
CoNS, could represent a major advancement in PJI treatment. Such an agent could potentially
lead to improved cure rate, shorter treatment duration and simplified therapy under the current
treatment guidelines. More importantly, such an agent could potentially change the PJI
treatment paradigm, allowing more patients to utilize the debridement and prosthesis retention
treatment strategy and reducing the need for more aggressive surgeries. A more efficacious
therapy against bacterial biofilms accompanied by an early diagnostic biomarker for PJI could
potentially make a surgeryꢀfree PJI treatment strategy a reality.
The rifamycin family of antibiotics is one of the most potent classes known against pathogens
living in bacterial biofilms.5 Rifampin is a member of the rifamycin family and plays an essential
role in current staphylococcal PJI management.6 However, pathogens develop resistance to
rifamycins rapidly as single point mutations on bacterial RNA polymerase typically confer high
levels of rifamycin resistance. Hence, in clinical practice, rifamycins need always to be used in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 41
1
2
3
4
5
6
7
8
9
combination with other antibiotics in order to minimize resistance development. To take
advantage of the potent activity of the rifamycin class against pathogens living in biofilms and at
the same time minimizing the emergence of resistance, a new strategy is currently in
development by attaching another antibiotic functionality to the rifamycin core to form hybrid
antimicrobials with a dualꢀmechanism of action. In this article we summarize the efforts in
discovering and developing dualꢀaction molecule 1 (TNPꢀ2092, CBRꢀ2092) based on rifamycins.
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
RIFAMYCINS AS LEADS FOR NEW AGENTS TO TREAT PERSISTENT BACTERIAL
INFECTIONS
The rifamycin class of antibiotics was first discovered in 1957 as natural products produced by
Streptomyces mediterranei.7 However, the naturally occurring rifamycins are unsuitable as
therapeutic agents due to relatively poor potency and undesirable pharmacokinetics. Synthetic
modifications of the natural products have led to significant improvements in potency and
pharmacokinetic properties. Currently, there are four semiꢀsynthetic compounds, rifampin,
rifapentine, rifabutin and rifaximin in clinical use. Rifampin is the cornerstone agent in a fourꢀ
drug combination regimen for the treatment of tuberculosis.8 Rifapentine possesses a longer
halfꢀlife than rifampin and can be used as intermittent therapy for the treatment of tuberculosis.
Rifabutin exhibits a reduced cytochrome P450 induction potential and has advantage in treating
tuberculosis and human immunodeficiency virus coꢀinfected patients. Rifaximin is unique in its
pharmacokinetics; it is not orally absorbable and is used for the treatment of symptoms
associated with gastrointestinal tract infections, including travelers’ diarrhea, hepatic
encephalopathy and irritable bowel syndrome with diarrhea.9
The mechanism of action of the rifamycin class has been extensively studied. Rifamycins are
DNAꢀdependent RNA polymerase inhibitors. They bind to the β subunit of the bacterial RNA
polymerase in the RNA exit channel and block RNA elongation during the transcription
ACS Paragon Plus Environment

Page 5 of 41
Journal of Medicinal Chemistry
1
2
process.10 Pathogens develop resistance to rifamycins rapidly mainly due to point mutations in
the rifamycinꢀbinding region of the β subunit of the bacterial RNA polymerase. A single point
mutation of one of the key residues in the rifamycin binding pocket can lead to high levels of
rifamycin resistance.11
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
One of the key characteristics of the rifamycin class is its activity against persistent bacterial
infections. Unlike the majority of other antibiotic classes that require active bacterial growth to
exert their antibacterial activity, rifamycins are active against pathogens in slow growing,
stationary and nonꢀreplicating metabolic states, owing to its ability to interfere with the critical
bacterial gene transcription process and its ability to distribute to deepꢀseated infection sites.
These unique attributes have important clinical applications and have made rifamycin class
agents the drugꢀofꢀchoice for many persistent bacterial infections. Rifampin is the most
important member of the rifamycin class from a clinical perspective and is a key component of
the fourꢀdrug regimen for the treatment of tuberculosis. Rifampin is mainly responsible for
shortening the duration of tuberculosis therapy from 12 months to the current 6 months.8, 12
Rifampin is also being utilized or studied for the treatment of other persistent and difficultꢀtoꢀ
treat bacterial infections including infective endocarditis, osteomyelitis and PJIs. 5, 13
Rifampin has demonstrated good activity against pathogens living in bacterial biofilms in vitro
and in animal models of biofilmꢀassociated, persistent infections5, 14 and is recommended as a
key component in the treatment of PJIs according to various clinical practice guidelines.3, 15 The
contribution of rifampin in the management of PJIs has been demonstrated in many clinical
trials.6b, 16 It has been well documented that rifampin should always be used in combinations
with other antibiotics to minimize the development of resistance.17 In addition, the companion
antibiotic can help to kill planktonic bacterial and prevent bacteremia. Various antibiotics have
been explored as companion agents for rifampin in the management of the early stage of
staphylococcal PJIs. The clinical practice guidelines published by the Infectious Diseases
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 41
1
2
3
4
5
6
7
8
9
Society of America (IDSA) recommends 2ꢀ6 weeks of a pathogenꢀspecific intravenous
antimicrobial therapy in combination with rifampin 300ꢀ450 mg given orally twice daily followed
by rifampin plus a companion oral drug for a total of 3 months for a total hip arthroplasty
infection and 6 months for a total knee arthroplasty infection.3 Recommended oral companion
drugs for rifampin are two fluoroquinolones, ciprofloxacin and levofloxacin, based on in vitro and
in vivo animal studies in combination with clinical evidence.16b
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
Currently, the rifampinꢀcontaining drug combinations are perhaps the best option available for
the treatment of staphylococcal PJIs and associated with improved cure rate. However, the
current treatment strategy is clearly inadequate as it requires parallel surgical intervention to be
effective and despite this, the success rate is still low and unreliable and the consequence of
treatment failure is devastating.
DESIGN, SYNTHESIS AND EVALUATION OF DUAL-ACTING RIFAMYCINS
One potential approach to improve PJI treatment is through the development of rifamycin based
dualꢀaction molecules. This approach takes advantage of the proven efficacy of the rifamycin
combinations against PJIs and in the same time to alleviate the problems associated with the
use of drug combinations.
There are a number of problems associated with rifampin based drug combinations in the
treatment of PJI. First, rifampin and its companion drug may have different pharmacokinetic
profiles. For example, they may have different serum halfꢀlives and tissue distributions. This can
lead to unbalanced drug exposure at the infection site, leaving rifampin “unprotected” from a
resistanceꢀdevelopment perspective at certain points in the dosing cycle and therein allowing for
the development of rifampin resistance. Second, the antibacterial spectrum of the companion
antibiotic (e.g., ciprofloxacin) can be very different from rifampin, and represents another cause
for concern for the development of resistance by organisms not fully covered by both antibiotics.
ACS Paragon Plus Environment

Page 7 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Third, there can be significant additive toxicity associated with the use of drug combinations or
other undesired drugꢀdrug interactions that impact the pharmacokinetics of companion agents.
Fourth, patients have to take multiple different antibiotics; particularly in the initial phase of the
PJI treatment where rifampin is given orally and other drugs are given intravenously. Finally,
another specific issue associated with the recommended PJI regimen rifampin plus a
fluoroquinolone (ciprofloxacin or levofloxacin) combination is the antagonism between these two
agents. The antagonistic effect between rifampin and the fluoroquinolone has been observed in
vitro and in vivo with different members of the fluoroquinolone class and against different
pathogens.18 A dualꢀacting molecule that contains a rifamycin and a fluoroquinolone structural
motif can theoretically address the above problems associated with rifampin and
fluoroquinolone drug combinations.
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 general strategy to minimize antibiotic resistance development, dualꢀacting antibacterial
molecules that can attack bacteria with two different modes of action have been explored for
many years. This area has been reviewed several times in the past few years, including a
comprehensive chemistry review published in 2015.19 One of the early examples of the dualꢀ
acting approach is 2 (Ro 23ꢀ9424), a hybrid compound comprised of a cephalosporin and a
fluoroquinolone moiety.20 Compound 2 itself behaves like a cephalosporin and does not
possess the mode of action of a fluoroquinolone. However, 2 can be cleaved either chemically
or metabolically to release the fleroxacin moiety and this degradation product can then exert its
antibacterial activity via the mode of action of a fluoroquinolone (Figure 1). Compound 2 is
therefore not a true dualꢀacting antibacterial agent and is inꢀeffect a cephalosporin acting as a
fluoroquinolone proꢀdrug. However, 2 failed to demonstrate any advantage over a cephalosporin
or a fluoroquinolone in preventing resistance development.21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1: Proposed enzymatic activation process for cephalosporinꢀfluoroquinolone hybrid
compound 2.
More recent efforts have focused on the development of nonꢀcleavable dualꢀaction molecules.
Antibiotic classes involved include fluoroquinolones, macrolides, βꢀlactams, oxazolidinones,
tetracyclines, aminoglycosides, rifamycins, glycopeptides and some less known antibiotic
classes.19 One of the nonꢀcleavable antibiotic hybrids is 3 (cadazolid), formed by a
fluoroquinolone and an oxazolidinone core via a stable linker (Figure 2). Compound 3 is
currently in Phase III clinical trials for the treatment of Clostridium difficile infections. Mode of
action studies conducted with C. difficile indicated that 3 retains the mode of action of an
oxazolidinone but failed to demonstrate a substantial contribution from the fluoroquinolone
function.22 In effect, 3 behaves like a more potent linezolid with a low systemic exposure and a
high local concentration in the gastrointestinal tract.
ACS Paragon Plus Environment

Page 9 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: Structure of oxazolidinoneꢀfluoroquinolone hybrid compound 3 (Cadazolid).
One of the key components for a successful dualꢀaction antibacterial program is to implement a
robust compound evaluation strategy. An experienced medicinal chemist with knowledge of the
structureꢀactivity relationships for each of the parent antibiotic classes can design and
synthesize certain hybrid compounds containing two antibiotic motifs. However, the resulting
hybrid compounds may or may not have the desired dual modes of action. Critical data need to
be generated rapidly to guide the lead optimization and candidate selection process. A robust
compound evaluation strategy should include assays to progressively assess compounds in
enzyme, wholeꢀcell and animal levels. Such an evaluation scheme is shown in Figure 3 which
includes (i) validated biochemical assays to assess the activity and potency of the hybrid
compounds against target proteins/enzymes of both parent drugs, (ii) whole cell assays on
isogenic bacterial panels with defined antibiotic resistance mutations on drug targets associated
with both parent drugs, and (iii) validated animal models of the target disease indication(s).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: Biological evaluation strategy for a dualꢀaction antimicrobial program.
Rifamycins are a wellꢀstudied class of antibiotics with an extensive knowledge of structureꢀ
activity relationships pertaining to in vitro and in vivo activity accumulated over the past 50 years.
Recently, the details of the interaction between rifampin and its drug target, bacterial RNA
polymerase, have been elucidated by highꢀresolution crystallization studies of the Thermus
aquaticus core enzyme in complex with rifampin.10 Both crystal structural interactions between
rifampin and RNA polymerase and structureꢀactivity relationships suggest that the groups
attached to the Cꢀ3 and Cꢀ25 positions of rifampin are pointing to open space and can therefore
be explored as sites for attaching a secondary antibiotic group [Figure 4].
ACS Paragon Plus Environment

Page 11 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Structural interactions between rifampin and RNA polymerase (4a) and structureꢀ
activity relationships (4b) suggest that the groups attached to the Cꢀ3 and Cꢀ25 positions of the
rifamycin scaffold are pointing to open space and can be explored for attaching a secondary
antibiotic motif. (Image 4a reprinted and modified from Cell, Vol. 104, Campbell EA, Korzheva N,
Mustaev A, Murakami K, Nair S, Goldfarb A, Darst SA, Structural mechanism for rifampicin
inhibition of bacterial RNA polymerase, Page 901ꢀ912, Copyright 2001, with permission from
Elsevier).
The majority of rifamycinꢀbased hybrid antibiotic compounds prepared to date utilize the Cꢀ3
position as the linking point. In addition to accessing an open space from the rifamycin scaffold
docked into its binding site on RNA polymerase, the Cꢀ3 position is chemically versatile which
can be easily functionalized to introduce various groups serving as linkers.23 The Cꢀ3 position
together with its adjacent Cꢀ4 position can also form a new heterocycle that in turn can serve as
the linking point for a secondary antibiotic.24 The carbonyl group at the Cꢀ11 position can be
functionalized via an oxime group and serve as the linking point for a secondary antibiotic to
access the same open space as that of the Cꢀ3 position.25 The acetyl group at the Cꢀ25 position
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 41
1
2
can be replaced with a carbamate group and serve as a linking point for further derivatization.26
The positions on rifamycin that have been explored as linking points and the chemistry
strategies utilized to link a secondary antibiotic functionality are summarized in Figure 5.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: Positions on the rifamycin scaffold and strategies that have been utilized for linking a
secondary antibiotic group.
The selection of the secondary antibiotic is important. From a mechanistic standpoint, the
secondary antibiotic should be active against the target pathogen(s) that the resulting dualꢀ
acting compound is intended to treat and it should not have crossꢀresistance with the primary
antibiotic. In addition, it should possess a wellꢀdefined structureꢀactivity relationship and/or high
resolution coꢀcrystal structural information allowing for the rapid assembly of hybrid molecules
with dual mechanisms of action. Several antibiotic classes have been explored as matching
partners for developing dualꢀacting rifamycins to date. They include quinolones (including nonꢀ
fluoroquinolones, fluoroquinolones and their bioisostere quinolizinones)23, 25b, 25c, 27, macrolides
(including 14ꢀ, 15ꢀ and 16ꢀmembered ring macrolides)28, oxazolidinones28 and
nitroimidazoles.24c In vitro minimum inhibitory concentration (MIC) data published in the cited
ACS Paragon Plus Environment

Page 13 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
patent literature indicate that these hybrid series are active against rifamycinꢀsusceptible and
resistant bacterial strains – an indicator of genuine dualꢀaction.
7
8
9
OPTIMIZATION AND DEVELOPMENT OF RIFAMYCIN-QUINOLONE DUAL-ACTING
ANTIMICROBIALS
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
Rifamycinꢀquinolone hybrid compounds have been studied the most extensively among those
rifamycinꢀbased hybrid series. A variety of quinolone core structures as depicted in Figure 6
have been explored as the matching partners for rifamycin.23, 25a, 25c, 27 These quinolone core
structures are associated with various fluoroquinolone, nonꢀfluoroquinolone and quinolizinone
subꢀclasses. Quinolizinones (also known as 2ꢀpyridones) are close analogs of quinolones where
the Nꢀ1 nitrogen atom is switched with the Cꢀ5 carbon atom to produce a family of highly active
bioisosteres.29 Quinolizinones, exemplified by 4 (ABTꢀ719), are potent and balanced inhibitors
of bacterial DNA gyrase and topoisomerase IV and exhibit potent antibacterial activities against
both Gramꢀpositive and Gramꢀnegative pathogens.29ꢀ30 However, no quinolizinone class of
antibacterial agents has been commercialized to date. The positions on the rifamycin scaffold
that have been utilized for attaching the quinolone structures are Cꢀ3, Cꢀ3/Cꢀ4, Cꢀ11 and Cꢀ25
using the strategy discussed previously (Figure 5). The structures for the linker group vary
widely based on published patent literature and appear to have significant impact on the
antibacterial activity of the hybrid compounds. However, the detailed structureꢀactivity
relationships around the rifamycinꢀquinolone series have yet to be reported.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: Quinolone core structures that have been explored as matching partners for
rifamycins.
Compound 1 is the leading molecule from the rifamycinꢀquinolone hybrid series currently in
clinical development.31 The compound is formed by a rifamycin and a quinolizinone core
connected via the Cꢀ3 position from the rifamycin side and the Cꢀ8 position from the
quinolizinone side. The linker group resembles the structural characteristics of the side chains of
rifampin and 4 joined together covalently (Figure 7).
ACS Paragon Plus Environment

Page 15 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7: Structure of rifampin, quinolizinone 4 and dualꢀacting molecule 1.
Compound 1 was synthesized by connecting the rifamycin core (14), quinolizinone core (11)
and linker group (10) together (Scheme 1).23 Starting from commercially available (S)ꢀ1ꢀbenzylꢀ
3ꢀhydroxypyrrolidine, linker group 10 was prepared in 5 steps. 8ꢀChloroꢀ1ꢀcyclopropylꢀ7ꢀfluoroꢀ
9ꢀmethylꢀ4ꢀoxoꢀ4Hꢀquinolizinoneꢀ3ꢀcarboxylate (11) was prepared according to published
procedures.29 Coupling of 10 and 11 provided compound 12. Hydrolysis of the ethyl ester and
deprotection of the amine protecting group, provided compound 13. Finally, 13 was attached to
the rifamycin core by first converting the piperidine amino group to a hydrazine, followed by
coupling with 3ꢀformalrifamycin (14). Compound 1 has been manufactured in kilogram quantities
in a similar fashion for multiple times. The synthesis of 1 is illustrated in the experimental
session.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 41
1
2
3
4
HO
NC
H2N
N
5
N
N
7
5
6
7
8
i,ii
iii
6
O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
N
8
iv
O
O
O
O
N
O
N
v
N
N
N
NH
10
9
O
O
F
N
OEt
11
vi
Cl
O
O
O
F
O
N
N
OEt
HN
vii,viii
N
N
O
O
F
N
OH
12
N
N
HO
13
OH
O
OH OH
O
O
ix,x
MeO
14
NH
O
CHO
O
OH
O
Compound 1
Scheme 1: Synthesis of dualꢀacting molecule 1.
Reagents and conditions: (i) MsCl, Et3N, toluene, 0oC, 2h; (ii) Bu4N+CNꢀ, CH3CN, 65oC, 16h; (iii)
Ti(OꢀiPr)4, EtMgBr, Et2O/THF, ꢀ78oC, 29 min; warm to rt, BF3•Et2O, 2h; (iv) 8, THF/AcOH, rt, 1 h,
then NaBH(OAc)3, rt, overnight; (v) H2, Pd/C, AcOH, 60 psi, rt, 18 h; (vi) 11, NaHCO3, CH3CN,
reflux, 5h; (vii) LiOH, EtOH, 60oC, 1h; (viii) CF3CO2H, CH2Cl2, 0oC ꢀ rt, 1h; (ix) NaOH, H2Nꢀ
OSO3H, 0oC, 1h; (x) 3ꢀformal rifamycin (14), MeOH/THF, rt, 30 min.
ACS Paragon Plus Environment

Page 17 of 41
Journal of Medicinal Chemistry
1
2
3
4
PRECLINICAL CHARACTERIZATION
5
6
7
8
9
Preclinical studies of the in vitro and in vivo activities of 1 were described in a series of posters
presented at the 47th (2007) Interscience Conference on Antimicrobial Agents and
Chemotherapy32 and in two subsequent publications.31
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 is shown in Table 1, biochemical studies using S. aureus enzymes indicate that 1 retains
rifampinꢀlike potency as an inhibitor of RNA Polymerase (RNAP) and exhibits balanced
(equipotent) activity against DNA gyrase and DNA topoisomerase IV (Topo IV) while retaining
activity against a common quinoloneꢀresistant (FQꢀR) variant (ParCS80F) of DNA topoisomerase
IV.32a No inhibition of human topoisomerase IIꢀalpha (Top2a) by 1 was observed. As is shown
in Table 2, minimum inhibitory concentration (MIC) values determined for otherwise isogenic
strains of S. aureus ATCC 29213 bearing rifamycin (rpoBH481Y) and/or quinolone (parCS80F and
gyrAS84L) resistance alleles confirm that 1 retains rifampinꢀlike potency against quinoloneꢀ
resistant strains and gatifloxacinꢀlike potency against rifampinꢀresistant strains. Again, these
data support the notion that 1 exhibits antimicrobial activity via combined contributions from both
the rifamycin and quinolizinone functions and are consistent with triple targeting of RNA
polymerase, DNA gyrase and DNA topoisomerase IV. A serial stepwise passage resistance
study conducted with 1 over a period of 26 days yielded a terminal isolate with broth and agar
MICs of >16 and 4 ꢁg/mL, respectively, that possessed five target mutations (rpoBR484H
,
gyrA∆L520, S84L, parC<R236>, H103Y) but was unaltered in efflux properties. Data from timeꢀkill and
metabolic labeling experiments also support the notion that the primary activity of 1 in S. aureus
is driven by its rifamycin function and that activity against rifamycinꢀresistant strains is mediated
by the quinolizinone function. In metabolic labeling experiments, 1 has a primary effect on de
novo RNA synthesis with delayed, secondary effects on DNA, protein and cell wall synthesis.
The more pronounced effect of 1 on DNA synthesis than that observed with rifampin
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 41
1
2
3
4
5
6
presumably reflects the secondary mode of action conferred by the quinolizinone part of the
molecule.
7
8
9
Table 1. Biochemical characterization of 1 and comparators.
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
IC50 (ꢀM)1
CC50 (ꢀM)2
RNAP
RNAP
Topo IV
Gyrase
Topo IV
(FQꢀR,
Gyrase
Compound
Human
Topo2a
(Wild
(RifꢀR,
(Wild
(Wild
(FQꢀR,
GyrAS84L
)
Type)
RpoBH481Y
)
Type)
Type)
ParCS80F
)
Compound 1
Rifampin
0.034
>25
1.7
ꢀ
1.5
ꢀ
2.7
ꢀ
>150
ꢀ
>150
ꢀ
0.015
>25
Ciprofloxacin
Gatifloxacin
ꢀ
ꢀ
ꢀ
ꢀ
0.4
0.2
5
31
13
>150
18
>150
>150
0.7
1The minimum concentration necessary to inhibit 50% (IC50) of the RNA product formed in the
absence of test agents; 2Minimum concentration necessary to induce 50% cleavage of a
covalently closed DNA substrate (CC50).
Table 2. Antibacterial activity of 1 and comparator agents against isogenic Staphylococcus
aureus strains bearing rifamycin and/or quinolone resistance mutations.
Staphylococcus aureus isogenic strain panel
MIC (ꢀg/mL)
rpoBH481Y
gyrAS84L
parCS80F
Compound
gyrAS84L
parCS80F
rpoBH481Y rpoBH481Y
Wild
gyrAS84L parCS80F
rpoBH481Y
gyrAS84L
parCS80F
Type
1 (agar)
0.008
0.008
0.008
0.008
0.06
0.12
0.06
0.25
ACS Paragon Plus Environment

Page 19 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1 (broth)
0.015
0.008
0.015
0.008
0.25
0.015
0.008
2
0.015
0.008
16
0.12
> 250
0.25
0.25
0.06
0.50
> 250
0.25
0.25
0.12
0.25
> 250
2
2
> 250
16
Rifampin
Ciprofloxacin 0.25
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
Levofloxacin
Gatifloxacin
0.25
0.06
0.25
1
8
1
8
0.12
0.25
8
0.25
8
Table 3 lists the in vitro antimicrobial activity of 1 against 175 staphylococci and 125
streptococci compared to those determined in parallel for rifampin and ciprofloxacin. The broth
microdilution MICs at which 90% of methicillinꢀsusceptible or methicillinꢀresistant S. aureus
(MSSA, MRSA) or S. epidermidis (MSSE, MRSE) isolates are inhibited (MIC90) by 1 are 0.015,
0.015, 0.008, and 0.5 ꢁg/mL, respectively. The activity of 1 against MSSA, MRSA and MSSE
was equivalent to rifampin and improved over ciprofloxacin, whereas 1 was improved over both
rifampin and ciprofloxacin against MRSE. MIC90 values for 1 against S. pyogenes, S. agalactiae
and S. pneumoniae were 0.12, 0.25, and 0.12 ꢁg/mL, respectively and again, equivalent to
rifampin and improved over ciprofloxacin.31b, 32c
Table 3. Antibacterial activities of 1 against clinical isolates of gramꢀpositive bacteria.
MIC (ꢁg/mL)1
Organism (n)
Compound
Range
50%
90%
S. aureus
MSSA (51)2
1
Rifampin
Ciprofloxacin
1
≤0.004 ꢀ 0.03
≤0.008 ꢀ 0.06
0.06 ꢀ > 4
0.008
0.015
0.25
0.015
0.015
1
MRSA (54)2
≤ 0.004 ꢀ 2
0.015
0.015
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 41
1
2
3
4
5
6
Rifampin
≤ 0.008 ꢀ > 4
0.12 ꢀ > 4
0.015
> 4
0.015
> 4
Ciprofloxacin
7
8
9
S. epidermidis
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
MSSE (35)2
1
≤ 0.004 ꢀ 0.015
≤ 0.004 ꢀ 0.06
0.12 ꢀ > 4
≤ 0.004 ꢀ 1
≤ 0.004 ꢀ > 4
0.12 ꢀ > 4
0.008 ꢀ 0.12
≤ 0.008 ꢀ 0.25
0.12 ꢀ 2
≤ 0.008
0.015
0.25
0.008
0.015
> 4
0.008
0.015
> 4
Rifampin
Ciprofloxacin
1
MRSE (35)2
S. pyogenes (35)3
S. agalactiae (35)3
S. pneumoniae (55)3
0.5
Rifampin
Ciprofloxacin
1
> 4
> 4
0.03
0.06
0.5
0.12
0.12
1
Rifampin
Ciprofloxacin
1
0.06 ꢀ 2
0.12
0.12
0.5
0.25
0.12
1
Rifampin
Ciprofloxacin
1
0.03 ꢀ 2
0.25 ꢀ 1
0.06 ꢀ 0.25
0.12 ꢀ1
0.12
0.25
2
0.12
0.25
> 4
Rifampin
Ciprofloxacin
0.5 ꢀ > 4
1Broth microdilution method; 2Tested in MHII broth with 0.002 % (vol:vol) polysorbateꢀ80;
3Tested in MHII broth with 5% (vol:vol) LHB and 0.002 % (vol:vol) polysorbateꢀ80.
ACS Paragon Plus Environment

Page 21 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 1 also exhibited prolonged postꢀantibiotic and subꢀMIC effects in S. aureus and are
indicative of efficient intracellular uptake and retention. In timeꢀkill studies, 1 exhibited
bactericidal activity against both staphylococci and streptococci with rapid, quinoloneꢀlike killing
of rifampinꢀresistant strains observed. In spontaneous resistance studies, 1 exhibited activity
consistent with balanced contributions from its composite parent antibiotics with a mutant
prevention concentration (MPC) of 0.12 ꢁg/mL and a resistance frequency of <10ꢀ12 determined
at 1ꢁg/mL in agar for S. aureus ATCC 29213. Finally, in studies of the killing of S. aureus
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
residing intracellularly in murine macrophage cells, 1 exhibited prolonged bactericidal activity.31,
32c, 32d
The comparative in vitro activity of 1 and rifampin+quinolone combinations was assessed using
a series of checkerboard assays with fractional inhibitory concentration (FIC) indices determined
by the broth microdilution method, timeꢀkill studies, biofilm killing assays and through
determination of MPCs.32b Antagonism of the rapid killing of S. aureus by quinolones in the
presence of rifampin was apparent in checkerboard assays via minimal bactericidal
concentration (MBC) endpoints and in high density timeꢀkill assays. In the latter, a
rifampin+ciprofloxacin combination (1+0.25 ꢁg/mL) is bacteriostatic (< 99.9% kill at 24h) against
S. aureus ATCC 29213 and resulted in the outgrowth of rifampinꢀresistant isolates. In contrast,
1 was observed to be bactericidal (≥ 99.9% kill at 24h at 0.25 ꢁg/mL) without the emergence of
rifampin resistance. Also in contrast to tested rifampin+quinolone cocktails, 1 promoted
eradication of colony biofilms formed by S. aureus MSSA or MRSA isolates and exhibited
bactericidal activity (>99% kill at 24h at 4 ꢁg/mL) that is greater than that of
rifampin+moxifloxacin (≤ 99 % kill at 4+4 ꢁg/mL) combination in killing S. aureus cells within
murine macrophages.
In further in vitro studies, 1 was found to exhibit antiꢀbiofilm activity in a number of alternate
assays and retains potent bactericidal activity against staphylococcal cells in the biofilm state.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 41
1
2
32d
3
4
In contrast to existing rifamycin class agents, resistance development in biofilm populations
5
6
was not observed with 1. These studies hold promise for the potential of 1 in the treatment of
7
8
biofilmꢀassociated infections.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Pharmacokinetic (PK) properties of 1 were determined following intravenous (IV) administration
to mice, rats, rabbits and dogs and indicated a plasma T1/2 of 0.4 to 4.1 h in mice, rats and
rabbits and 1.6 to 11.7 h in dogs.32f The calculated volume of distribution of 1 ranged from 0.08
to 0.52 and 0.09 to 0.17 L/kg in rodents/rabbits and dogs, respectively. Plasma clearance
ranged from 0.07 to 0.36 and 0.01 to 0.08 L/h/kg in rodents/rabbits and dogs, respectively.
Calculated AUC and Cmax values were dose proportional for all species. Data from the
neutropenic thigh infection model and the drip flow biofilm reactor identified the 24h AUC/MIC
ratio as the PD index best correlated with efficacy. 32f
The in vivo efficacy of 1 was determined in various rodent models of staphylococcal and
streptococcal infections including acute systemic infections and chronic biofilmꢀassociated
infections of indwelling medical devices.32e In a mouse septicemia model, ED50 values for 1
ranged from 1.4 to 3.8 mg/kg for MSSA, MRSA and FQꢀresistant S. aureus and 4.8 to 10.7
mg/kg for S. epidermidis and S. pyogenes infections. In a mouse skin abscess model, 1
exhibited efficacy comparable to rifampin or comparators used to treat complicated skin and
skin structure infections (cSSSIs) with log10 CFU reductions of >3.1 to 5.3 observed for MRSA,
S. epidermidis or S. pyogenes. In the rat central venous catheter (CVC) infection model, 1
exhibited equivalent efficacy to rifampin, but significantly improved over cSSSI comparator
agents with log10 CFU reductions of 2.8 to 3.4 observed for MSSA and FQꢀresistant S. aureus.
Compound 1 showed effective penetration into granuloma pouch fluid and reduced S. aureus
counts by 1 to 3.4 log10 CFU after single IV doses of 5 to 25 mg/kg. In the mouse subcutaneous
catheter implant model, 1 was more efficacious than rifampin and cSSSI comparators achieving
log10 CFU reductions of 5.03, 3.98 and 2.51 against susceptible, FQꢀresistant and rifampinꢀ
ACS Paragon Plus Environment

Page 23 of 41
Journal of Medicinal Chemistry
1
2
3
4
resistant S. aureus isolates, respectively. In addition, 1 exhibited continued suppression of the
5
6
catheterꢀlocalized biofilm infections after therapy was withdrawn.
7
8
9
The in vivo efficacy of 1 was further investigated in a rabbit model of infective endocarditis (IE)
mediated by MRSA.32g, 32h In these studies, rabbit IE was induced with MRSA strain 67ꢀ0,
following transcarotidꢀtransaortic valve indwelling catheterization. At 24h postꢀinfection, animals
received either (i) no therapy (control), (ii) 1 at 10, 25 or 40 mg/kg, iv, bid or (iii) vancomycin at
15 mg/kg, iv, bid. All treatments were for 3 days and 24h after the last dose, target tissues were
quantitatively cultured; relapse of infection was also investigated 3d after treatment was
discontinued. MICs/MBCs of 1 and vancomycin for MRSA 67ꢀ0 were 0.008/0.5 mg/mL, and 1/1
mg/mL, respectively. In these studies, all regimens significantly decreased MRSA densities vs.
controls with 1 at 25 and 40 mg/kg exhibiting the greatest efficacy and also prevented relapse.
In contrast to vancomycin, sterile tissue cultures were observed in vegetations, kidneys and
spleens for all 1 treatment groups following the relapse period with no resistance development
to 1 observed.
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 comparative efficacy of 1 and rifampin+quinolone combinations was also determined in the
rabbit IE model including studies of a quinoloneꢀresistant variant of MRSA 67ꢀ0 (QMRSA)
bearing both norAup (efflux) and DNA topoisomerase IV (parCS80F) mutations.32h At 24h postꢀ
infection, animals received either (i) no therapy (control), (ii) 1 40 mg/kg, iv, bid, ciprofloxacin 20
mg/kg, iv, tid + rifampin 10 mg/kg, im, bid or (iii) levofloxacin 20 mg/kg, iv, bid + rifampin 10
mg/kg, im, bid, all for 3 days. At 24h postꢀdose, target tissues were cultured. Relapse of
infection was investigated 3d after treatment was discontinued. MICs/MBCs of 1, ciprofloxacin,
levofloxacin and rifampin were 0.008/0.5, 16/16, 2/2 and 0.008/2 mg/mL, respectively.
Compound 1 significantly reduced QMRSA densities in all target tissues at the end of therapy
and relapse. Levofloxacin+rifampin significantly reduced densities but to a lower extent than
observed with 1 and ciprofloxacin+rifampin exhibited the lowest overall efficacy.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 41
1
2
3
4
SUMMARY AND FUTURE PERSPECTIVES
5
6
7
8
9
The development of rifamycin based dualꢀacting antibacterial agents represents a promising
strategy for the treatment of persistent bacterial infections such as PJIs. The dualꢀaction
approach could potentially address several major issues associated with the use of rifamycin
drug combinations and therefore provide a better therapeutic solution in treating persistent
bacterial infections where the standardꢀofꢀcare requires a rifamycin combination therapy.
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 dualꢀaction molecule 1 has demonstrated advantages over rifampin and fluoroquinolone
combinations in several aspects. First, it possesses potent and balanced activity against three
cellular targets RNA polymerase, DNA gyrase and topoisomerase IV in S. aureus and CoNS.31
Second, it is not a substrate of fluoroquinolone efflux systems and has a low propensity for
resistance development. Third, 1 exhibits potent activity against drug resistant pathogens
including those resistant to one or both parent drugs and demonstrates better efficacy and
resistance development profile than rifamycin+fluoroquinolone combinations in both in vitro and
in vivo biofilm models. Fourth, 1 appears able to alleviate certain safety concerns associated
with its parent antibiotic classes such as CYP3A4 induction and hERG inhibition. In completed
human Phase I single and multiple ascending dose clinical trials of an intravenous formulation, 1
demonstrates good pharmacokinetic and safety profiles supporting further clinical development
for the treatment of PJIs and other persistent bacterial infections.
One of the unexpected observations during the development of rifamycin based dualꢀaction
compounds such as 1 is that such dualꢀaction molecules exhibit synergistic effects in vitro and
in vivo improved over the simple parent drug combinations. Such synergies are apparent with
several dualꢀaction classes and may represent a general benefit of the dualꢀaction approach.
The source and mechanism for such synergy is unclear at this time and represents a direction
ACS Paragon Plus Environment

Page 25 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
of ongoing research for rifamycinꢀquinolone hybrid agents. Understanding the mechanisms of
this synergistic effect may help to design better dualꢀacting molecules in the future.
7
8
9
The spectrum of the antibacterial activity of a dualꢀacting molecule can be very different from its
parental antibiotics. RNA polymerase and DNA gyrase/topoisomerase IV, the respective drug
targets for rifamycins and fluoroquinolones, are essential enzymes in all pathogens. The
spectrums of activity for rifamycins and fluoroquinolones are largely a reflection of their ability to
reach their intracellular targets. Rifamycins are mainly active against Gramꢀpositive pathogens
with difficulties to penetrate the outer cell membrane of Gramꢀnegative organisms, while
fluoroquinolones are broad spectrum and active against both Gramꢀpositive and Gramꢀnegative
pathogens. The dualꢀaction hybrid molecule 1 is mainly active against Gramꢀpositive pathogens
with a spectrum similar to rifampin. However, 1 also demonstrates a better activity than rifampin
against certain Gramꢀnegative bacteria such as Escherichia coli and Acinetobacter baumannii,
indicating that 1 may have improved membrane permeability and/or be able to avoid the native
efflux systems of these organisms.
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 dualꢀaction molecule generally possesses very different absorption, distribution, metabolism
and excretion (ADME) properties from its parental antibiotics and its clinical application should
be optimized based on its ADME characteristics. Indeed, dualꢀaction molecules such as 1
typically have high molecular weights and exhibit low oral bioavailability and therefore may not
be suitable for the development of oral dosage formulations for systemic delivery. However, this
low oral absorption could become an advantage for the treatment of diseases associated with
gastrointestinal tract infections and this strategy is currently being actively pursued in clinical
trials for indications associated with gastrointestinal tract infections.
EXPERIMENTAL SECTION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 41
1
2
3
4
5
6
7
8
9
Compound 1 has been synthesized many times from milligram to kilogram quantities. The
following procedure reproduced from a patent publication is a representation of the process
used for large quantity manufacturing of 1.23 All starting materials used are either purchased
from commercial sources or prepared according to published procedures. Operations involving
moisture and/or oxygen sensitive materials are conducted under an atmosphere of nitrogen.
Flash chromatography is performed using silica gel 60 as normal phase adsorbent or C18 silica
gel as reverse phase adsorbent. Thin layer chromatography (TLC) and preparative thin layer
chromatography (PTLC) are performed using preꢀcoated plates purchased from E. Merck and
spots are visualized with ultraviolet light followed by an appropriate staining reagent. Nuclear
magnetic resonance (NMR) spectra are recorded on a Varian 400 MHz magnetic resonance
spectrometer. 1H NMR chemical shift are given in partsꢀper million (δ) downfield from TMS using
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
the residual solvent signal (CHCl3 =δ 7.27, CH3OH =δ 3.31) as internal standard. H NMR
information is tabulated in the following format: number of protons, multiplicity (s, singlet; d,
doublet; t, triplet; q, quartet; m, multiplet; td, triplet of doublet; dt, doublet of triplet), coupling
constant (s) (J) in hertz. The prefix app is occasionally applied in cases where the true signal
multiplicity is unresolved and prefix br indicates a broad signal. Electroꢀspray ionization mass
spectra are recorded on a Finnegan LCQ advantage spectrometer. Highꢀperformance liquid
chromatography (HPLC) analysis for the final compound is performed on Agilent 1100
instrument using a Waters Xterra RP18 column (5 mcꢀm, 4.6 x 250 mm) and gradient elution
(solvent A: 20 mM NaH2PO4/acetonitrile, 60:40 v/v; solvent B: acetonitrile). HPLC purities for the
final compound are ≥95%.
Synthesis of (S)-Methanesulfonic acid 1-benzyl-pyrrolidin-3-yl ester (A)
(S)ꢀ1ꢀBenzylꢀ3ꢀhydroxypyrrolidine (5, 20.2 g, 114 mmol) was dissolved in toluene (200 mL). To
this stirred solution, triethylamine (20 mL, 142 mmol) was added followed by methanesulfonyl
chloride (10.5 mL, 136 mmol) during a period of 40 min at 0ºC. The resultant slurry was allowed
ACS Paragon Plus Environment

Page 27 of 41
Journal of Medicinal Chemistry
1
2
to stir for two hours at 0oC and 7% sodium bicarbonate (200mL) was then added to the reaction
mixture. The organic layer was then separated and the aqueous layer extracted with toluene
(3x100 mL). The combined organic layers were washed with saturated sodium bicarbonate
(3x200 mL), dried over Na2SO4 and concentrated under reduced pressure to give (S)ꢀ
methanesulfonic acid 1ꢀbenzylꢀpyrrolidinꢀ3ꢀyl ester as a yellow oil (28 g, 96%). The material
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
was used directly for the next step without purification. H NMR (400 MHz, CDCl3) δ 7.34ꢀ7.17
(m, 5H), 5.24ꢀ5.18 (m, 1H), 3.71 (d, JAB = 13.0 Hz, 1H), 3.64 (d, JAB = 13.0 Hz, 1H), 3.01 (s, 3H),
2.89ꢀ2.80 (m, 3H), 2.56ꢀ2.50 (m, 1H), 2.38ꢀ2.31 (m, 1H), 2.14 ꢀ2.07 (m, 1H).
Synthesis of (R)-1-benzyl-3-cyanopyrrolidine (6)
(S)ꢀMethanesulfonic acid 1ꢀbenzylꢀpyrrolidinꢀ3ꢀyl ester (28 g, 110 mmol) was dissolved in
anhydrous acetonitrile (60 mL) and solid tetrabutylammonium cyanide (59 g, 220 mmol) was
added at once at room temperature. The resulting mixture was heated at 65ºC for 16 h and
cooled to room temperature. Saturated NaHCO3 (100 mL) solution was added. The organic
layer was then separated and the aqueous layer extracted with toluene (3x100 mL). The
combined organic layers were washed with water (3x100 mL), brine, and dried over Na2SO4
and concentrated under reduced pressure to give (R)ꢀ1ꢀbenzylꢀ3ꢀcyanopyrrolidine as brown oil
(21 g), which was vacuum distilled at 150oC/5 mmHg to give a colorless liquid (19 g, 85%): [α]23
1
ꢀ 22.0º (c = 2.5, MeOH); H NMR (400 MHz, CDCl3) δ 7.36ꢀ7.17 (m, 5H), 3.66 (app s, 2H),
3.05ꢀ2.99 (m, 1H), 2.93 (app t, J = 8.8 Hz, 1H), 2.72ꢀ2.60 (m, 3H), 2.31ꢀ2.22 (m, 1H), 2.17ꢀ2.11
(m, 1H).
Synthesis of (R)-1-(1-Benzyl-pyrrolidin-3-yl)-cyclopropylamine (7)
To a solution of (R)ꢀ1ꢀbenzylꢀ3ꢀcyanopyrrolidine (6, 10 g, 53.4 mmol) in anhydrous ether (200
mL) was added Ti(OiPr)4 (17.2 mL, 58.8 mmol) and the resulting solution was cooled to ꢀ78ºC.
EtMgBr (3.0 M in Et2O, 35 mL, 105 mmol, Aldrich) was added dropꢀwise during a period of 40
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 41
1
2
min at ꢀ78oC. Anhydrous THF (50 mL) was added to the resultant yellow suspension to facilitate
stirring and the solution was stirred for 20 min at ꢀ78ºC. The reaction mixture was allowed slowly
to warm up to room temperature, then BF3—Et2O (13.4 mL, 107 mmol) was added and the
resultant dark brown suspension was stirred at room temperature for 2 hrs. Both 1N HCl (100
mL) and Et2O (200 mL) were introduced subsequently to the reaction mixture and stirring was
kept for 20 min until both organic and aqueous phases became clear. It was then basified using
aqueous 20% NaOH (100 mL) and stirred for 30 min. The organic phase was separated and the
aqueous phase was extracted with Et2O (2x100 mL). The combined organic phase was dried
over Na2SO4 and concentrated under reduced pressure to give (R)ꢀ1ꢀ(1ꢀbenzylꢀpyrrolidinꢀ3ꢀyl)ꢀ
cyclopropylamine as brown oil (10.5 g, 90%), which was vacuum distilled to give a colorless
liquid (9 g, 77%). The product is sufficiently pure to use in next step. For characterization, a
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
small amount of sample was purified by PTLC using solvent NH4OH/MeOH/CH2Cl2 (1/60/340):
1
[α]23 ꢀ 10.0º (c = 2.5, MeOH); H NMR (400 MHz, CDCl3) δ 7.33ꢀ7.23 (m, 5H), 3.62 (d, JAB
=
12.8 Hz, 1H), 3.56 (d, JAB = 12.8 Hz, 1H), 2.66ꢀ2.57 (m, 3H), 2.40ꢀ2.37 (m, 1H), 1.96ꢀ1.91 (m,
2H), 1.69ꢀ1.62 (m, 1H), 0.55ꢀ0.48 (m, 2H), 0.41ꢀ0.36 (m, 2H).
Synthesis of (R)-4-{[1-(1-Benzyl-pyrrolidin-3-yl)-cyclopropyl]-methyl-amino}-piperidine-1-
carboxylic acid tert-butyl ester (9)
To a solution of (R)ꢀ1ꢀ(1ꢀbenzylꢀpyrrolidinꢀ3ꢀyl)ꢀcyclopropylamine (7, 18.6 g, 87 mmol) and 4ꢀ
oxoꢀpiperidineꢀ1ꢀcarboxylic acid tertꢀbutyl ester (8, 22.7 g, 113 mmol) in THF (500 mL) was
added glacial HOAc (31 mL). The solution was stirred at room temperature for one hour and
then NaBH(OAc)3 (65 g, 306 mmol) was added. The resulted suspension was stirred at room
temperature overnight before 37% aqueous solution of formaldehyde (30 mL, 262 mmol) was
added. The reaction mixture was allowed to stir for 3 hrs and completed, then added aqueous
20% NaOH (300 mL) at 0oC. The mixture was allowed to stir for 2 h, and layers were separated.
The aqueous layer was extracted with EtOAc (3x300 mL). The combined organic phase was
ACS Paragon Plus Environment

Page 29 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
dried over sodium sulfate, concentrated in vacuo. The residue was purified by flash
chromatography (gradient elution, 1% to 10% methanol in dichloromethane) to give the title
compound as a pale yellow oil (27 g): ESI MS m/z 414.3 (M + H+). 1H NMR (400 MHz, CDCl3) δ
7.33ꢀ7.23 (m, 5H), 4.24ꢀ4.02 (m, 2H), 3.88ꢀ3.80 (m, 1H), 3.60 (d, JAB = 12.8 Hz, 1H), 3.53 (d, JAB
= 12.8 Hz, 1H), 2.82ꢀ2.57 (m, 5H), 2.31 (s, 3H), 2.28ꢀ2.22 (m, 1H), 1.94ꢀ1.80 (m, 4H), 1.44 (s,
9H), 1.42ꢀ1.28 (m, 3H), 0.55ꢀ0.53 (m, 4H). The chiral purity of the product was analyzed by
ChiralꢀCel OD column (0.46 cm × 25 cm, Daicel Chemical industries, Ltd, column catalog no.
OD00CEꢀE1031) with acetonitrile in the presence of 0.1% diethylamine as mobile phase (flow
rate 1mL/min; retention time 5.23 min desired, 5.93 isomer) to be 95% ee.
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
Synthesis of (R)-4-[Methyl-(1-pyrrolidin-3-yl-cyclopropyl)-amino]-piperidine-1-carboxylic
acid tert-butyl ester (10)
(R)ꢀ4ꢀ{[1ꢀ(1ꢀbenzylꢀpyrrolidinꢀ3ꢀyl)ꢀcyclopropyl]ꢀmethylꢀamino}ꢀpiperidineꢀ1ꢀcarboxylic acid tertꢀ
butyl ester (9, 14 g) was dissolved in 150 mL glacial acetic acid at room temperature in a
hydrogenation flask, air was evacuated, and filled with nitrogen. To this was added 30% Pd/C (5
g), flask was mounted to Parr shaker, nitrogen was evacuated and filled with hydrogen, and the
mixture was shaken under hydrogen atmosphere at 60 psi in a Parr for 18 h. Reaction mixture
was diluted with toluene (300 mL), filtered through a pad of celite. The filtrate was concentrated
in vacuo, residue was digested in cold 30%NaOH at 0oC and the product was extracted with
ethyl acetate. The combined ethyl acetate solution was charged with 50 g of charcoal (Darco
Gꢀ60), allowed to stir for 2 h, filtered, concentrated to give pale yellow oil, which can be vacuum
distilled to give a clear oil (9.6 g). 1H NMR (400 MHz, CD3OD) δ 4.04 (br s, 1H), 3.01 (dd, J =
10.4 Hz and 7.6 Hz, 1H), 2.84 (dd, J = 8.0 Hz and 5.2 Hz, 2H), 2.64ꢀ2.55 (m, 3H), 2.48ꢀ2.43 (m,
2H), 2.30 (s, 3H), 2.28ꢀ2.23 (m, 1H), 2.23ꢀ1.74 (m, 3H), 1.40 (s, 9H), 1.35ꢀ1.28 (m, 2H), 1.18ꢀ
1.12 (m, 2H), 0.53ꢀ0.45 (m, 4H).
ACS Paragon Plus Environment