Design, synthesis, and characterization of novel tetrahydropyran-based bacterial topoisomerase inhibitors with potent anti-gram-positive activity

Article abstract of DOI:10.1021/jm400963y

There is an urgent need for new antibacterial drugs that are effective against infections caused by multidrug-resistant pathogens. Novel nonfluoroquinolone inhibitors of bacterial type II topoisomerases (DNA gyrase and topoisomerase IV) have the potential

Full text of DOI:10.1021/jm400963y

Subscriber access provided by Georgetown University | Lauinger and Blommer Libraries
Article
Design, Synthesis and Characterization of Novel Tetrahydropyran-based
Bacterial Topoisomerase Inhibitors with Potent Anti-Gram Positive Activity.
Jean-Philippe Surivet, Cornelia Zumbrunn, Georg Rueedi, Christian Hubschwerlen,
Daniel Bur, Thierry Bruyère, Hans Locher, Daniel Ritz, Wolfgang Keck, Peter Seiler,
Christopher Kohl, Jean-Christophe Gauvin, Azely Mirre, Verena Kaegi, Marina Dos
Santos, Mika Gaertner, Jonathan Delers, Michel Enderlin-Paput, and Maria Boehme
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm400963y • Publication Date (Web): 22 Aug 2013
Downloaded from http://pubs.acs.org on August 23, 2013
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 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis and Characterization of Novel Tetrahydropyranꢀbased Bacterial
Topoisomerase Inhibitors with Potent AntiꢀGram Positive Activity.
*
+
Authors: JeanꢀPhilippe Surivet, Cornelia Zumbrunn, Georg Rueedi, Christian Hubschwerlen, Daniel
+
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
Bur, Thierry Bruyère, Hans Locher, Daniel Ritz, Wolfgang Keck, Peter Seiler, Christopher Kohl,
JeanꢀChristophe Gauvin, Azely Mirre, Verena Kaegi, Marina Dos Santos, Mika Gaertner, Jonathan
Delers, Michel EnderlinꢀPaput, Maria Boehme.
Actelion Pharmaceuticals Ltd, Gewerbestrasse 16, CHꢀ4123 Allschwil, Switzerland
ABSTRACT
There is an urgent need for new antibacterial drugs that are effective against infections caused by
multiꢀdrug resistant pathogens. Novel nonꢀfluoroquinolone inhibitors of bacterial type II
topoisomerases (DNA gyrase and topoisomerase IV) have the potential to become such drugs as they
display potent antibacterial activity and exhibit no targetꢀmediated crossꢀresistance with
fluoroquinolones. Bacterial topoisomerase inhibitors that are built on a tetrahydropyran ring linked to
a bicyclic aromatic moiety through a synꢀdiol linker, show potent anti Gramꢀpositive activity, covering
isolates with clinically relevant resistance phenotypes. For instance, analog 49c was found to be a dual
DNA gyraseꢀtopoisomerase IV inhibitor, with broad antibacterial activity and low propensity for
+
spontaneous resistance development, but suffered from high hERG K channel block. On the other
+
hand, analog 49e displayed lower hERG K channel block, while retaining potent in vitro antibacterial
activity and acceptable frequency for resistance development. Furthermore, analog 49e showed
moderate clearance in rat and promising in vivo efficacy against Staphylococcus aureus in a murine
infection model.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
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
INTRODUCTION
The selective pressure exerted by the use of antibiotics has inevitably resulted in the dissemination of
drug resistant bacteria, among which the ESKAPE pathogens (Enterococcus faecium, Staphylococcus
aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacter
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
species) are of utmost concern. These pathogens cause not only the major part of hospitalꢀacquired
infections (leading to a death toll of approximately 100,000/year in the US) but also constitute a
2
ꢀ4
financial burden to healthcare systems all over the world . The development of novel small molecule
st
antibacterial drugs is therefore urgently needed. At the dawn of the 21 century, it was common belief
that genomic and associated new technologies would deliver a plethora of new antibacterial drug
5
,6
targets and thus new chemical entities lacking crossꢀresistance with clinically used antibacterial
drugs. Unfortunately, the efforts have not yet borne fruit and a significant research effort is still
7
8
dedicated to the ‘old’ but clinically validated targets to fill the antibiotic pipeline .
Bacterial type II topoisomerases (DNA gyrase and topoisomerase IV (Topo IV)) are heterotetramers of
2 2 2 2 2 2
GyrA GyrB and ParC ParE (E. coli) or GrlA GrlB (S. aureus) respectively. They have been
validated as targets for antibacterial therapy ever since it was recognized that the fluoroquinolone class
of drugs can inhibit the topoisomerase function. The antibacterial spectrum of activity was expanded
9
from nalidixic acid to newest fluoroquinolones , but crossꢀresistance still emerged. Among several
1
0
resistance mechanisms , single amino acid changes in the quinolone resistanceꢀdetermining region
1
1,12
(
QRDR) of GyrA (Ser84Trp) or GrlA (Ser80Phe)
of S. aureus reduces dramatically
1
3,14
fluoroquinolones affinity binding and therefore their inhibitory effect . However, a new avenue was
opened in the field of bacterial topoisomerase inhibition when researchers at GlaxoSmithKline (GSK)
and Aventis independently disclosed closely related novel nonꢀfluoroquinolone based topoisomerase
15,16
inhibitors (so called NBTIs) , devoid of cross resistance with fluoroquinolones. The Aventis
program, which was continued at Novexel, capitalized on the initial hit scaffold and its attractive
17
properties, leading to 1 (NXLꢀ101) (Figure 1). Unfortunately, its development had to be discontinued
18
after detection of unacceptable QT interval prolongation . The GSK program took a different
19
20
21
direction resulting in successive generations of NBTIs 2 , 3 and 6 and could even profit at a later
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
stage from the resolution of the Xꢀray structure of 4 (Figure 1) bound to DNA and S. aureus DNA
2
2
gyrase . The crystallographic studies provided the basis to rationalize the lack of cross resistance
between NBTIs and fluoroquinolones and also led to the design of novel inhibitors (Figure 1).
+
Recently, a publication by AstraZeneca showed that 2 and 3 block the hERG K channel significantly
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<10ꢀM) which may lead to unacceptable cardiac safety liabilities due to QT interval
prolongation. Using compound 3 as lead compound, the AstraZeneca team has successfully improved
2
3
the cardiac safety profile of their newly designed NBTIs . However, the resulting lead compound 5
suffered from low exposure in mice and no data concerning the stability toward resistance
24
25
development were presented. Researchers at Johnson and Johnson and Pfizer have also reported
their efforts on NBTIs.
Figure 1: Representative novel bacterial topoisomerase inhibitors (NBTIs)
Here we report the path to design and synthesize novel dual inhibitors of bacterial DNA gyrase and
Topo IV from lead molecules 1 and 2. The StructureꢀActivityꢀRelationships (SARs) of bacterial type
II topoisomerase inhibitors were elaborated and used in the course of our optimization program that
26
led to compound 49e , whose characterization will be presented.
WORKING HYPOTHESIS
Biochemical evaluation of compound 1 indicated that it owns antibacterial activity against Gramꢀ
positive bacteria (including methicillinꢀ and fluoroquinoloneꢀresistant strains) primarily through potent
S. aureus DNA gyrase inhibition (Table 1, supercoiling assay). By contrast, in Gramꢀnegative
pathogens, the primary target of 1 is likely Topo IV, as indicated by its better inhibitory effect on E.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
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
coli Topo IV (Table 1, relaxation assay). DNA gyrase and Topo IV differ in their relative sensitivities
to 1, in opposition to the mode of action of fluoroquinolones which target preferentially Topo IV in
2
7
Gramꢀpositive and DNA gyrase in Gramꢀnegative pathogens . Compound 2 is not only more potent
against Gramꢀpositive bacteria, but also displays a broader antibacterial spectrum including E. coli as
well as other Gramꢀnegative pathogens (Table 2). Compound 2 is a true dual inhibitor of GyraseꢀTopo
IV in both S. aureus and E. coli (Table 1). A S. aureus singleꢀstep mutant strain, obtained from wildꢀ
type S. aureus grown in the presence of 1, contained a single mutation in the gyrA gene corresponding
to a D83N amino acid exchange. Notably, while compound 2 showed a marginal inhibitory activity
against this mutated DNA gyrase, it still inhibits growth of this strain (MIC 0.25 mg/L) through Topo
IV inhibition. We therefore concluded that a balanced dual inhibition profile would be highly
advantageous in preventing rapid emergence of spontaneous resistance.
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 1: Inhibitory activities of compounds 1, 2, 5 and 6 against bacterial topoisomerases
Topo IV (IC , ꢀM)
Gyrase (IC , ꢀM)
50
50
Relaxation Inhibitory Activity
Supercoiling Inhibitory Activity
a
Compds
S. aureus
E. coli
S. aureus
S. aureus
S. aureus
E. coli
b
c
b
d
e
c
wt
wt
wt
QR
D83N
wt
1
2
5
6
128
2
0.125
0.03
0.03
0.125
128
0.125
0.008
0.03
128
8
32
0.03
8
0.125
0.008
0.125
0.5
8
2
8
32
0.03
2
0.125
0.5
f
CIP
8
>256
>256
a
c
b
Compounds 1, 2, 5 and 6 were prepared in our laboratories as reported in literature. Isolated from Staphylococcus aureus ATCC 29213.
d
Isolated from Escherichia coli ATCC 25922. Quinoloneꢀresistant (QR) enzyme containing a S84L mutation isolated from
e
f
Staphylococcus aureus Aꢀ798. Enzyme containing the D83N mutation isolated from a laboratory strain resistant to 1. Ciprofloxacin,
commercially available.
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
8
Table 2: Antibacterial activities of compounds 1, 2, 5 and 6 (MIC in mg/L)
S. aureus
S. aureus
S. aureus
E. faecium
S. pneumoniae.
E. coli
P. aeruginosa
Compds
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
c
d
e
f
g
wt
MRSA, QR
D83N
QR
wt
wt
wt
1
2
5
6
0.125
0.015
0.015
0.015
0.25
0.25
0.015
0.015
0.015
>8
0.25
1
1
1
>8
0.25
8
>8
8
0.063
0.5
0.063
0.25
0.015
>8
>8
0.25
1
0.06
1
h
CIP
>8
0.25
>8
1
0.016
a
b
c
d
Staphylococcus aureus ATCC 29213. Staphylococcus aureus Aꢀ798. Laboratory strain. Enterococcus faecium Aꢀ949 vancomycinꢀ
e
f
g
resistant. Streptococcus pneumoniae ATCC 49619. Escherichia coli ATCC 25922. Pseudomonas aeruginosa ATCC 27853.
h
Ciprofloxacin (fluoroquinolone).
At the start of our investigations, no crystallographic data for GyrA complexed to a NBTI was
available to support rational drug design of dual GyrA/ParC inhibitors. Enzymatic data recorded for
compound 2, showed crossꢀresistance with 1, suggesting that both entities share a common binding
site in the proximity of Asp83. NBTIs chemical structure can be divided in three parts: a methoxy
substituted bicyclic aromatic leftꢀhand side (LHS), a monoꢀ or bicyclic lipophilic rightꢀhand side
(
RHS) and a variable linker joining the two units. Superposition of compounds 1 and 2 (Figure 2)
displayed a good overlay for LHS and RHS thereby defining the length of the linker as well as
directionalities of the respective exit vectors. The linker, which can itself be divided in three parts (i.e.
a central ring spacer and two flanked connecting units reaching the LHS and RHS respectively),
differs significantly in compounds 1 and 2, sharing only a saturated 1,4ꢀdisubstituted sixꢀmembered
ring as common motif. The amide functionality in 2 is coꢀplanar with its LHS, leading to a straighter
linker if compared with the hydroxyꢀpropylidene (ꢀCHOHꢀ(CH
2 2
) ꢀ) functionality in 1. Both inhibitors
comprise a basic nitrogen in close vicinity to the RHS.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
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: a. Modeled superimposition of inhibitors 1 (blue bonds) and 2 (green bonds). Red: oxygen
atoms; blue: nitrogen atoms; white: carbon atoms; yellow: sulfur atoms; green: fluorine atoms.
The more balanced GyrA/ParC activity of 2 led us to design NBTIs containing a transꢀ3,6ꢀdisubtituted
tetrahydropyran (THP) ring and a two atom (ꢀAꢀBꢀ) unit to connect this ring to the LHS (Scheme 1).
We hypothesized that such a linker should be suitable to establish geometrically restricted compounds
with favorable physicochemical and pharmacological profiles. Therefore, benzylic alcohols or diols,
respectively (ꢀAꢀBꢀ = ꢀCHOHꢀCH
solubility if compared to amide (ꢀAꢀBꢀ = ꢀHNꢀCOꢀ) or ether (ꢀAꢀBꢀ = ꢀOꢀCH
products on the way to the desired diol compounds, the (E)ꢀalkene (ꢀAꢀBꢀ = ꢀCH=CHꢀ) and its
saturated analog (ꢀAꢀBꢀ = ꢀCH ꢀCH ꢀ) could be obtained. As we could not predict the effect of the
2
ꢀ or ꢀCHOHꢀCHOHꢀ), were assumed to produce superior aqueous
2
ꢀ) units. As side
2
2
THPꢀring stereochemistry, we decided to explore both enantiomeric series. In case the –AꢀBꢀ motif is
a diol, we have chosen to limit our investigations to synꢀdiols for synthetic reasons. Finally, known
RHSꢀunits were connected to the THPꢀcore through the aminomethyl side chain.
CHEMISTRY
As illustrated in Scheme 1, the exploration of the SAR of the newly designed topoisomerase inhibitors
was facilitated by using a modular synthetic approach based on the following three key building
blocks: i) a bicyclic aromatic LHS, ii) a central THP core, and iii) a bicyclic aromatic RHS. The
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
impact of the stereochemistry of the central THP core can be rapidly explored by using building
blocks 7 and its enantiomer entꢀ7.
Scheme 1
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
In contrast to the wellꢀdocumented synthesis of the alcohol (3R,6S)ꢀ7 from triꢀOꢀAcꢀDꢀglucal , the
synthetic access to the enantiomer (3S,6R)ꢀentꢀ7 had to be developed first. According to Scheme 2, the
3
0
known
ωꢀalkenoate 8 was reduced to the corresponding alcohol with lithium borohydride in
tetrahydrofuran, and a treatment with meta-chloroperbenzoic acid under buffered conditions yielded
epoxide 9 as an equimolar mixture of diastereomers. The following acidꢀpromoted cyclization, using
(+/ꢀ)-10ꢀcamphorsulfonic acid (CSA), afforded a separable mixture of enantiopure entꢀ7 (arising from
of the specific attack of the alcohol on the (S)ꢀconfigurated epoxide) and the pyrrolidine derivative 10
(resulting from the cyclization of –NHꢀtertꢀbutoxycarbonyl on the (R)ꢀconfigurated epoxide).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
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
Scheme 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
According to Scheme 1, the introduction of the AꢀB linker required the preparation of appropriately
substituted central building blocks featuring a proper reaction handle FG . The following
2
modifications of alcohol (3R,6S)ꢀ7 leading to central building blocks 12, 15 and 18 are summarized in
Scheme 3. Amide 12 was obtained in two steps starting with the oxidation of alcohol 7 to acid 11 by
29
sodium periodate in presence of a catalytic amount of ruthenium trichloride . A subsequent treatment
of the acid 11 with Nꢀhydroxysuccinimide (NHS) and N,Nꢀdicyclohexylcarbodiimide provided its
activated form which upon treatment with gaseous ammonia yielded the targeted amide 12, with no
3
1
loss of stereochemical information (Scheme 3).
a
Scheme 3
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
2
3
The oxidation of alcohol 7 using pyridineꢀsulfur trioxide complex (py.SO ) and dimethylsulfoxide in
3
3
dichloromethane afforded aldehyde 13. The Wittig olefination leading to alkene 14 was performed
by the reaction of methylenetriphenylphosphorane (generated from methyltriphenylphosphonium
bromide and potassium tertꢀbutoxide) with aldehyde 13. The hydroboration of alkene 14 with 9ꢀ
borabicyclo[3.3.1]nonane in tetrahydrofuran led after oxidative workꢀup to the intermediate primary
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
35
alcohol that was further oxidized to aldehyde 15 using DessꢀMartin periodinane (Scheme 3). As
observed in the case of amide 12, it is noteworthy to state that no epimerization occurred by passing
from alcohol 7 to homologated aldehyde 15. Phenyltetrazole sulfone 18 was obtained by oxidation of
sulfide 17 with hydrogen peroxide in presence of ammonium molybdate. The formation of 17 was
initially achieved by direct coupling of alcohol 7 with phenyltetrazole thiol (PTꢀSH) under Mitsunobuꢀ
36
3
type conditions (combining triphenylphosphine (PPh ) and diisopropyl azodicarboxylate (DIAD)) .
However, nucleophilic substitution of iodide 16 with PTꢀSH in presence of potassium hydroxide in
refluxing ethanol turned to be more effective. The intermediate iodide 16 was obtained through
tosylation followed by displacement with sodium iodide in refluxing acetone (Scheme 3).
2
The bromides 19a and 22aꢀ25a (R = Br) were prepared by refluxing the corresponding hydroxyl
2
37
derivatives (R = OH) in neat phosphorous tribromide. As illustrated in Scheme 4, the preparation of
aldehydes 19bꢀ22b consisted in trapping a lithio anion that was generated either from bromides 19a or
2
2a (treatment with nꢀbutyl lithium or lithium diisopropylamide at ꢀ78°C), or by fluorineꢀdirected
orthoꢀlithiation of compounds 20a or 21a with dimethylformamide. Dechlorination of aldehyde 21c
was done by hydrogenolysis using palladium on carbon and triethylamine under hydrogen atmosphere
2
to yield aldehyde 21b. The syntheses of aldehydes 23bꢀ26b started from bromides 23aꢀ25a (R = Br)
2
38
39
or triflate 26a (R = OTf) via Suzuki coupling with transꢀphenylvinyl boronic acid giving rise to the
respective phenyl alkenes 23cꢀ26c which were converted to the corresponding intermediate diols 23dꢀ
2
6d. The desired aldehydes 23bꢀ26b were then subsequently formed under standard periodic cleavage
conditions. On the other hand, aldehyde 27b was obtained by hydrolysis of dibromide 27c obtained
37
from methyl quinoxaline 27a by treatment with an excess of Nꢀbromosuccinimide in presence of
azobisisobutyronitrile (Scheme 4).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
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
Scheme 4
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
According to Scheme 5, the general synthetic access to final NBTI described herein involved the
coupling of the central THP core to the bicyclic aromatic LHS, followed by a reductive amination
using readily accessible RHS aldehydes. The ether linkage of derivative 28a was constructed using a
26
Mitsunobu coupling reaction (DIAD, PPh
3
) between alcohol 7 and hydroxyꢀnaphthyridine 23e . In
40
contrast, amide 12 was coupled to bromide 23a via a Buchwaldꢀtype reaction to form the amideꢀ
containing derivative 29a. Treatment of bromide 23a with nꢀbutyl lithium at ꢀ78°C generated the lithio
anion that reacted with aldehyde 15 to yield benzylic alcohol 30a as an equimolar mixture of
4
1
42
diastereomers . JuliaꢀKocieński coupling of PTꢀsulfone 18 and aldehyde 23b using potassium
bis(trimethylsilyl)amide as a base provided (E)ꢀalkene 31a which was hydrogenated over palladium on
carbon to afford the corresponding saturated derivative 34a. On the other hand, Sharpless asymmetric
4
3
41
dihydroxylation of alkene 31a gave rise to synꢀdiols with an excellent diastereoselectivity (>95%) .
The use of commercial ADꢀmixꢀ or ADꢀmixꢀ gave access to both diastereomeric series as
demonstrated with the formation of intermediates (1S,2R)ꢀ35a and (1R,2S)ꢀ36a The corresponding
α
β
.
free amines 28bꢀ31b and 34bꢀ36b were obtained by treatment with trifluoroacetic acid (TFA), either
neat or diluted in dichloromethane. From these amines, variable RHS moieties were introduced via
4
4
45
46
reductive amination reaction with aldehydes 32 or 33 , giving rise to NBTIs 28cꢀ31c, 28dꢀ31d,
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
4cꢀ36c, 34d and 35d. A set of THPꢀbased topoisomerase inhibitors belonging to the enantiomeric
series could be prepared by the same synthetic pathway but using sulfone entꢀ10 as a partner in the
JuliaꢀKocieński coupling. Likewise, asymmetric dihydroxylation of entꢀ31a using either ADꢀmixꢀ or
α
ꢀβ
proceeded smoothly to afford the corresponding synꢀdiols entꢀ35c and entꢀ36c (Scheme 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 5
The JuliaꢀKocieński coupling of aldehydes 19bꢀ22b, 24bꢀ27b and sulfone 18 provided (E)ꢀalkenes 37ꢀ
41
4
4 with (E)/(Z)ꢀselectivities bigger than 6/1 . The (1R,2S)ꢀsynꢀdiols 45aꢀ52a were prepared by using
ADꢀmixꢀβ as dihydroxylating reagent. After removal of the tertꢀbutoxycarbonyl (Boc) group with
TFA, the resulting amines 45bꢀ52b were transformed to NBTIs 45cꢀ52c using aldehyde 32 in the
reductive amination reaction. Finally, the intermediate amines 48b and 49b were coupled under
47
48
49
reductive amination conditions with aldehydes 33, 53 , 54 and 55 to access NBTIs 48dꢀf and 49dꢀ
g (Scheme 6).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
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
Scheme 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
RESULTS AND DISCUSSION
The antibacterial activity, expressed as MIC (minimal inhibitory concentration, [mg/L]) of the novel
chemical entities, was measured against a panel of bacterial strains including Staphylococcus aureus
(
S. aureus), Streptococcus pneumoniae (S. pneumoniae), Escherichia coli (E. coli) and Pseudomonas
aeruginosa (P. aeruginosa) ATCC strains as well as a S. aureus D83N DNA gyrase mutant strain,
resistant to NBTIs such as 1. All the novel chemical entities were also evaluated for inhibition of DNA
+
gyrase and topo IV isolated from wild type S. aureus and E. coli and for block of hERG K channels.
As shown in Table 3, all molecules based on a 6ꢀmethoxyꢀ1,5ꢀnaphthyridine scaffold showed potent
binding to S. aureus DNA gyrase and E. coli Topo IV, with IC50 in the nanomolar range for the best
compounds such as 30, 34 and 35. The ether linker present in 28cꢀd led to lower enzyme potency, and
accordingly, weaker antibacterial activities. The two amide derivatives 29cꢀd displayed much weaker
antibacterial activities compared to the related GSK inhibitor 2. However, precipitation of 29c due to
low solubility prevented reliable inhibition measurements. The most potent antibacterial activities
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were measured for derivatives containing an ethylene, a hydroxyꢀ or a 1,2ꢀdihydroxyꢀethylene unit
(30cꢀd, 34cꢀd and 35cꢀd respectively) whereas alkenes 31c,d showed higher MICs. Compounds
comprising a pyridothiazinone moiety as RHS displayed consistently superior target potency
compared to molecules containing the dioxinopyridine moiety, which translated into lower MICs and a
broader antibacterial spectrum, as exemplified by compounds 30c and 30d. While both molecules
displayed good antibacterial activities on S. aureus, 30c showed a 4ꢀfold lower MIC on wild type S.
pneumoniae (0.03 mg/L), and at least a 8ꢀfold lowered MIC on E. coli (1 mg/L) when compared to
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
30d. The same trend was observed for 34c and 34d as well as for 35c and 35d. The lack of
antibacterial activity of compounds 28dꢀ35d against wildꢀtype E. coli was not further investigated, but
may result from a combination of reduced enzyme potency on target and lower cytoplasmic
accumulation in E. coli compared to the related derivatives having a pyridothiazinone RHS (28cꢀ36c).
Moreover, the pyridothiazinone inhibitors exhibit good target potency against S. aureus Topo IV and
E. coli DNA gyrase respectively, whereas inhibitors comprising a dioxinopyridine failed to do so.
Similar to inhibitor 2, the most potent dual S. aureus DNA gyraseꢀTopo IV inhibitors (34c and 35c)
displayed antibacterial activities against the S. aureus D83N mutant (MIC range 0.5ꢀ1 mg/L). In
+
general, as shown in Table 3, a high level of hERG K channel block was measured for this set of
compounds with the exception of compound 35d, suggesting that designing compounds with high
+
antibacterial activity and low hERG K channel block should be possible.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
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
Table 3: SAR of twoꢀatom linker AꢀB
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
Gyrase
SCIA (IC50
Topo IV
hERG
e
Cpds
RHS
Linker
MIC (mg/L)
)
RIA (IC50
)
block
a
R b
c
d
S. a.
0.5
2
S. a.
S. p.
E. c.
S. a.
0.125
0.5
E. c.
2
S. a.
8
E. c.
28c
1
2
1
2
1
2
1
2
1
2
1
2
4
>8
>8
>8
0.5
>8
4
0.25
4
4
>8
>8
>8
1
0.125
0.5
70
19
99
71
99
66
94
88
97
82
36
27
28d
>8
>8
2
f
2
2
9c
0.25
0.125
0.03
0.03
0.125
0.25
0.03
0.06
0.03
0.5
0.25
0.03
0.125
1
0.125 0.5
0.125 >8
<0.03 0.5
0.125 >8
0.125 0.5
0.125 >8
0.03 0.12
<0.03
<0.03 0.5
0.03
0.125
0.125
<0.03
0.125
0.03
9d
>8
>8
>8
2
3
0c
0d
31c
1d
4c
4d
5c
35d
3
>8
4
3
>8
1
1
>8
1
>8
0.5
>8
0.5
8
0.125
0.03
3
0.06
0.5
0.06
0.5
3
4
8
8
0.125
0.125
0.125
3
0.5
4
1
0.06
8
8
a
b
c
d
Staphylococcus aureus ATCC 29213. Staphylococcus aureus Aꢀ234 gyrA: D83N. Streptococcus pneumoniae ATCC 49619. Escherichia
e
+
f
coli ATCC 25922, % block of hERG K channel measured at 10 ꢀM. Precipitation observed during assay measurements. All compounds
showed MIC >8 mg/L on Pseudomonas aeruginosa ATCC 27853.
Lead Optimization
The absolute stereochemistry of the synꢀdiol linker modulated the antibacterial activity to a significant
extent (Table 4). While inhibitors 35c and entꢀ36c featuring a (1S,2R)ꢀconfigurated dihydroxyꢀ
ethylene linker displayed low MICs against Gramꢀpositive S. aureus and S. pneumoniae (0.03 to 0.06
mg/L) and Gramꢀnegative E. coli (1ꢀ2 mg/L), the MICs recorded for their respective (1R,2S)ꢀ
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
configurated diastereomers 36c and entꢀ35c, were more than 10ꢀfold higher against those pathogens.
These observations were in agreement with the lower DNA topoisomerases inhibitory activities
recorded for 36c and entꢀ35c in comparison to those of 35c and entꢀ36c and 52c respectively.
Interestingly, the (3R,6S)ꢀ or (3S,6R)ꢀ configuration of the THP core (referring to the progenitor
building blocks 7 and ent-7) had no influence on antibacterial potency. All diastereomers or
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
+
enantiomers of 35c that were evaluated for hERG K channel block, exerted higher affinity for this
channel if compared to 35c and no clear trend could be observed with respect of the different
stereochemistries.
Table 4: Influence of the stereochemistry of the 1,2ꢀdihydroxyꢀethylene linker and the THP ring
Gyrase
Topo IV
hERG
e
Cpds
MIC (mg/L)
SCIA (IC50
)
RIA (IC50
)
block
a
R b
c
d
S. a.
S. a.
S. p.
E. c.
S. a.
E. c.
0.5
S. a.
E. c.
3
5c
0.03
0.5
1
0.5
0.06
0.5
1
1
0.03
0.125
0.5
0.5
0.125
36
45
93
70
entꢀ35c
36c
>8
>8
0.5
4
>8
1
8
8
>8
>8
8
0.125
0.125
0.125
ent-36c
0.03
0.03
0.03
0.5
a
b
c
d
Staphylococcus aureus ATCC 29213. Staphylococcus aureus Aꢀ234 gyrA: D83N. Streptococcus pneumoniae ATCC 49619. Escherichia
e
+
coli ATCC 25922, % block of hERG K channel measured at 10 ꢀM. All compounds showed MIC >8 mg/L on Pseudomonas aeruginosa
ATCC 27853.
As the stereochemistry of the central THPꢀcore had no influence on antibacterial potency, we chose to
pursue our investigations on the series stemming from the more readily available alcohol 7 and several
analogs of 35c featuring various bicyclic aromatic LHS moieties, were evaluated. As shown in Table
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
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
5
, its 6ꢀmethoxyquinoline analog 51c was less active. The nitrogen atom at the generic position noted
U contributes to an enhanced antibacterial potency. This was confirmed by comparing 47c with 49c
and 46c with 52c, respectively. Indeed, derivatives 45c, 48c, 49c and 52c, wherein U = N, shared a
similar antibacterial profile, with good MICs against wildꢀtype S. aureus and S. pneumoniae. As
previously observed, these dual inhibitors of S. aureus and E. coli type II topoisomerases, exhibited
antibacterial activity against D83N S. aureus mutant and wild type E. coli. In contrast, the nature of
the groups in generic positions V, W and X (N, CH or CF) did not modulate potency to a significant
extent. In general, derivatives 46c, 47c, 50c and 51c, wherein U = CH did not exhibit a dual target
inhibitor profile and lacked antibacterial activity against D83N S. aureus mutant. Nonetheless, less
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
+
block of the hERG K channel was observed for derivatives with U = CH. For compounds with U = N,
+
slight differences were observed with respect to hERG K channel block, with IC50s in the 10
ꢀM
range in case of X = CF (48c and 49c), whereas analogs for which X = CH (35c and 52c) showed
+
slightly more potent hERG K channel block.
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 5: SAR of Left Hand Sides
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
Gyrase
Topo IV
hERG
e
Cpds
LHS
MIC (mg/L)
SCIA (IC50
)
RIA (IC50
)
block
a
R b
c
d
S. a.
S.a.
S. p.
E. c.
S. a.
E. c.
S. a.
S. a.
35c
0.03
0.03
0.25
0.25
0.06
0.03
0.03
0.25
0.03
0.5
0.25
>8
>8
1
0.06
0.06
0.25
0.25
0.5
1
0.03
0.5
0.5
0.125
70
72
17
15
44
54
20
24
55
45c
2
8
4
4
1
4
4
2
0.03
0.125
0.125
0.03
0.125 0.5
0.125
0.125
0.125
0.125
0.03
46c
47c
48c
2
2
0.5
8
0.5
0.5
2
2
49c
0.25
4
0.06
0.25
0.25
0.06
0.03
0.5
8
50c
51c
52c
0.125
0.25
0.125
0.125
0.125
>8
2
8
0.5
0.03
0.5
2
a
d
b
c
Staphylococcus aureus ATCC 29213. Staphylococcus aureus Aꢀ234 gyrA: D83N. Streptococcus pneumoniae ATCC 49619.
e
+
Escherichia coli ATCC 25922, % block of hERG K channel measured at 10 ꢀM.. All compounds showed MIC >8 mg/L on
Pseudomonas aeruginosa ATCC 27853.
+
From the latter observations, the influence of the RHSꢀmoiety on hERG K channel block was
established based on the 3ꢀfluoronaphthyridine and 6ꢀfluoroquinoxaline scaffolds (Table 6). Whereas
+
both compounds 48c and 49c blocked hERG K channel in the low micromolar range, significantly
reduced block was observed for the inhibitors 48d, 48f, 49d and 49g containing either a
dioxinopyridine, a dioxinopyridazine, an oxathiinopyridine or an oxathiinopyridazine moiety.
+
However, reduced hERG K channel block tracked in parallel with lower topoisomerase inhibition.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
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
For instance, compounds 48d, 48f, 49f and 49g, featuring a dioxinopyridine or a pyridazineꢀbased
moiety, still displayed potent inhibition of S. aureus DNA gyrase, but weak S. aureus Topo IV
activity, thus resulting in limited antibacterial activity on the S. aureus D83N mutant strain (MIC
8mg/L or higher). The exchange of dioxino ring for a oxathiino ring resulted still a low S. aureus Topo
IV activity but to an increased potency against S. aureus DNA gyrase. A good compromise between
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
+
antibacterial activity and block of hERG K channel could be obtained for compound 49e.
Structural analysis of diol 49e
With the Xꢀray structure of compound 4 in complex with S. aureus DNA gyrase and a 20ꢀbp DNA
2
2
duplex in hand , inhibitor 49e was docked into the active site of the proteinꢀDNA complex in order to
rationalize the proposed SAR (Figure 3). The 6ꢀmethoxyꢀ1,5ꢀnaphthyridine moiety of 49e sits between
two stretched DNA base pairs. Saturated linkers (ethylene, hydroxylꢀ or 1,2ꢀdihydroxyꢀethylene)
allowed an unstrained fit of compounds as observed for 4, whereas amides, ethers and alkenes
diverged from a linker path as defined by 4, ultimately leading to a suboptimal positioning of a
(charged) Hꢀbond donor that can interact with Asp83. Furthermore, the two hydroxyl units of 49e
appear to be in a position to establish Hꢀbonds with neighboring base pairs. The RHS of 49e resides in
a hydrophobic pocket at the interface of the two GyrA subunits. The oxathiolopyridine ring of
compound 4 and similarly the hydrophobic RHS of 49e establish van der Waals interactions with the
surrounding hydrophobic residues, thereby featuring interactions between the sulfur atoms of the
inhibitors and methionine side chains of GyrA.
ACS Paragon Plus Environment

Page 19 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Docking pose of compounds 49e (yellow) superimposed on original inhibitor 4 (pink) in
GyrB27ꢀA56 complex Xꢀray (PDBꢀcode: 2XCS). DNA is shown in green, while gyrase Cꢀalpha
backbone is blue.
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 6: Novel bacterial topoisomerase inhibitors with reduced hERG K channel block
Y
O
Z
N
H
N
O
N
F
OH
O
O
N
F
LHS
N
N
OH
LHS1
LHS2
Gyrase
SCIA (IC )
Topo IV
hERG
Cpds
LHS
Y
Z
e
MIC (mg/L)
RIA (IC50)
block
5
0
a
R b
c
d
S. a.
S. a.
S. p.
E. c.
S. a.
E. c.
S. a.
S. a.
4
8d
48e
8f
9d
9e
49f
9g
1
1
1
2
2
2
2
CH
CH
N
O
S
≤0.015
≤0.015
≤0.015
0.03
8
0.125
0.03
8
8
0.125
8
8
0.5
0.125
2
11
33
0
2
8
2
2
8
0.03
0.125
0.03
8
8
8
8
8
8
>8
8
4
S
0.125
0.03
>8
>8
8
4
CH
CH
N
O
S
0.125
0.125
0.5
10
19
3
4
≤0.015
≤0.015
≤0.015
0.06
0.03
8
S
>8
>8
0.125
8
4
N
O
0.125
>8
0.5
0.125 >8
>8
2
7
a
b
c
d
Staphylococcus aureus ATCC 29213. Staphylococcus aureus Aꢀ234 gyrA: D83N. Streptococcus pneumoniae ATCC 49619. Escherichia
e
+
coli ATCC 25922. % block of hERG K channel measured at 10 ꢀM. All compounds showed MIC >8 mg/L on Pseudomonas aeruginosa
ATCC 27853.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
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
Antibacterial profile of NBTI 49e
As shown in Table 7, compound 49e was highly active against all S. aureus isolates tested including
5
0
MRSA and fluoroquinoloneꢀresistant strains (MIC , 0.03 mg/L) . In addition it showed good activity
90
against other clinically relevant Gramꢀpositive bacteria including S. pneumoniae and E. faecium VRE
with MIC90s of ≤ 1 mg/L. Significant activity was also observed against Moraxella catarrhalis and
Haemophilus influenzae. However, only weak or no activity was detected against other Gramꢀnegative
bacteria such as Escherichia coli or Pseudomonas aeruginosa. In summary, compound 49e has an
antibacterial spectrum suitable for the treatment of Gramꢀpositive and respiratory tract infections.
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 7. Antibacterial activities of 49e against clinical isolates (in mg/L)
Nb of
Organisms
MIC50
MIC90
Range
strains
35
Staphylococcus aureus
≤ 0.015
≤ 0.015
0.03
0.03
≤ 0.015ꢀ0.06
≤ 0.015ꢀ0.03
MRSA subset
25
Staphylococcus epidermidis
Enterococcus faecalis
16
20
20
24
14
13
11
11
0.06
0.5
0.25
0.5
0.03ꢀ0.25
0.06ꢀ1
Enterococcus faecium VRE
Streptocococcus pneumoniae
Streptococcus pyogenes
Streptococcus agalactiae
Haemophilus influenzae
Moraxella catarrhalis
0.25
0.06
0.03
0.25
0.5
1
0.03ꢀ4
0.125
0.06
0.5
0.03ꢀ0.25
0.015ꢀ0.125
0.125ꢀ0.5
0.25ꢀ8
4
0.03
0.125
≤0.015ꢀ125
Spontaneous resistance frequencies of selected compounds were measured in S. aureus. Both 49c and
9e that displayed dual target activity (gyrase and Topo IV) showed 10 to 100ꢀfold lower resistance
4
frequencies compared to compound 1 targeting only gyrase (Table 8). This finding confirmed our
working hypothesis that emergence of resistance should be slower with inhibitors acting on both
targets.
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 8. Frequencies of spontaneous resistance development in S. aureus ATCC 25923
Cpds
Resistance frequency
at 4x the MIC at 16x the MIC
a
ꢀ6
ꢀ7
1
2.7 x 10
8.8 x 10
5.0 x 10
2.1 x 10
ꢀ
8
8
ꢀ9
49c
<2.1 x 10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ
ꢀ8
49e
1.1 x 10
a
assayed as an equimolar mixture of diastereomers (at benzylic position).
In order to assess the pharmacokinetic properties, mouse, rat and human liver microsomal stabilities,
expressed as intrinsic clearances, were determined (Table 9). With decreasing lipophilicity achieved
through introduction of hydroxyl groups on the ethylene linker (drop in clogP of 1.2 unit), a lower
metabolic turnover was observed, with similar trends for the three species. A marked enhancement in
in vitro metabolic stability was further obtained for quinoxaline 52c.
Table 9: Hepatic intrinsic microsomal clearance (ꢀL/(min*mg))
a
Cpds
mouse
Rat
Human
cLogP
3
4c
9e
954
291
39
>1250
344
347
30
2.43
1.20
0.57
4
52c
92
45
a
Assayed at 1ꢀM compound concentration.
Compounds 34c and 52c, both featuring a pyridothiazinone RHS, showed a high in vivo clearance in
the rat, whereas 49e with an oxathiinopyridine RHS, was cleared much more slowly and displayed
significant bioavailability (F = 41%). Rat clearance measured for compounds 34c, 49e and 52c (Table
1
0) only partially matched the corresponding in vitro microsomal stability data in line with the notion
that in vitro CLint is by and large an unbound parameter not confounded by plasma protein binding.
Not surprisingly, the in vivo clearance of 49e and 52c reflects the in vitro metabolic stability once
5
1
u
corrected for the fraction unbound in plasma (unbound clearance, CL ) .
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
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
Table 10: Pharmacokinetic parameters in rat
a
,
b
d
e
f
g
h
CL
CLu
Vss
t1/2
AUC (iv)
AUC (po)
(ng*h/mL)
F
c
Cpds
f
u
(
mL/min//kg)
(mL/min/kg)
(L/kg)
(h)
(ng*h/mL)
(%)
i
3
4c
9e
36
8.7
53
nd
nd
3.2
1
1.2
3
463
959
313
1320
776
29
41
26
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
0.017
0.18
512
294
52c
18
5.9
817
a
Pharmacokinetic parameters following iv bolus injection of a 1 mg/kg dose (compounds 34c and 52c) or 0.5 mg/kg dose (compound
b
c
d
e
f
g
4
9e). Clearance. Fraction unbound in plasma. Unbound clearance. Volume of distribution at steady state. Halfꢀlife. AUC, area under
h
the curve. Oral bioavailability F was assessed following the administration of a 10 mg/kg dose (compounds 34c and 52c) or 1 mg/kg
i
dose(compound 49e). Not determined.
The antiꢀstaphylococcal activity of compound 49e was confirmed in vivo in a neutropenic murine
52
thigh infection model (MIC S. aureus = 0.06 mg/L). When administered subcutaneously at a dose of
4
0 mg/kg, 49e mediated a net 1ꢀlog reduction of colony forming units (CFU) in the thigh measured 6 h
post infection.
CONCLUSION
We have reported the design of novel bacterial dual DNA gyrase and Topo IV inhibitors comprising a
tetrahydropyran core and have elaborated SAR findings that led us to identify an ethyleneꢀdiol motif
as a suitable moiety connecting the central THPꢀring to the LHS. Our efforts to optimize antibacterial
+
and pharmacological profile and simultaneously decrease block of hERG K channel have resulted in
synthesis and characterization of a series of potent dual GyrA/ParC inhibitors which display strong
antibacterial activity against clinically relevant pathogens such as Staphylococci, Enterococci, and
Streptococci. Furthermore, we have demonstrated that dual inhibition of gyrase and Topo IV is
required to minimize the rate of resistance development. Although a fully balanced GyrA/ParC
inhibition profile could be achieved, our efforts to further advance such balanced inhibitors were
+
thwarted by unacceptable hERG K channel block. However, we have identified compound 49e which
shows an exquisite antibacterial profile, while having a low to moderate resistance frequency in vitro
ACS Paragon Plus Environment

Page 23 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
and significantly decreased hERG K channel block. The pharmacokinetic parameters of 49e were
sufficient to reach exposures needed for in vivo efficacy in a murine thigh infection model.
THPꢀbased bacterial topoisomerase inhibitors show promising properties and deserve further efforts to
be eventually transformed into new clinically used antibacterial agents.
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
MIC testing. Minimal inhibitory concentrations (MICs) were determined by the broth microdilution
2
8
method according to guidelines of the Clinical and Laboratory Standards Institute (CLSI) . Stock
solutions of compounds were prepared in DMSO and serially diluted in cationꢀadjusted Muellerꢀ
Hinton broth (final DMSO concentration, 2.5 % (v/v). Reference antibiotics (ciprofloxacin, linezolid)
were tested in parallel and MICs against quality control strains were within the accepted CLSI ranges.
9
10
Spontaneous resistance frequencies. Large numbers of bacteria (10 ꢀ10 ) were plated on
MuellerꢀHinton agar plates containing the antibiotic compound at 4ꢀ and 16ꢀfold the MIC
concentration. Resistance frequencies were calculated by dividing the numbers of colonies growing
after 48 h of incubation at 37°C through the total numbers of viable bacteria plated (as determined by
colony count on drugꢀfree agar).
DNA gyrase and Topo IV inhibition assays. DNA gyrase supercoiling inhibition assay was
performed with relaxed pBR322 (purchased from Inspiralis, UK) as a substrate. For E. coli, the
reaction mixture (20 ꢁL) contained 25 mM Tris HCl pH 7.5, 24 mM KCl, 4 mM MgCl
mM ATP, 1.6 mM spermidine, 6.5% glycerol, 50 ꢁg/mL Bovine Serum Albumin (BSA) and 0.1 ꢁg
relaxed plasmid. For S. aureus the reaction mixture (20 ꢁL) contained 50 mM Tris HCl pH 7.5, 20
mM KCl, 5 mM MgCl , 5 mM DTT, 3 mM ATP, 700 mM potassium glutamate, 50 ꢁg/mL BSA and
2
, 2 mM DTT,
1
2
0.1 ꢁg relaxed plasmid. The reactions were carried out at 37 °C for 1 h. DNA Topo IV relaxation
assay was performed with supercoiled pBR322 as a substrate. For E. coli, the reaction mixture (20 ꢁL)
2
contained 50 mM Tris HCl pH 7.8. 20 mM KCl, 6 mM MgCl , 10 mM DTT, 1 mM ATP, 1 mM
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
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
spermidine, 100 ꢁg/mL BSA and 0.1 ꢁg supercoiled pBR322. For S. aureus, the reaction mixture (20
2
ꢁL) contained 50 mM Tris HCl pH 7.5. 20 mM KCl, 5 mM MgCl , 5 mM DTT, 1.5 mM ATP, 5 mM
spermidine, 100 mM potassium glutamate, 5 ꢁg/mL BSA and 0.1 ꢁg supercoiled pBR322. The
reaction was carried out at 37 °C for 1 h. Both supercoiling and relaxation reactions were stopped by
adding a mixture of EDTA, bromophenol blue and glycerol. 10 ꢁL of each reaction mixture were
loaded on a 1% agarose gel in TAE buffer (40 mM Trisꢀacetate, 1 mM EDTA pH 8.0) and run at 1.0
V/cm for 14 to 16 h. Gels were stained in water with ethidium bromide and visualized under
ultraviolet light. For supercoiling inhibition assays (SCIA), 50% inhibitory concentration (IC50) was
determined visually as being the compound concentration at which the supercoiled band was reduced
by 50%. For relaxation inhibition assays (RIA), IC50 was defined as the concentration of inhibitor
necessary to inhibit the formation of 50% relaxed DNA from supercoiled. The IC50 was taken as an
average of two inhibitory concentrations if inhibition did not occur exactly within a particular
concentration range (for representative gels, see Supporting Information) . All gels were also subject
to analysis using AlphaEase Stand Alone Software (Alpha Innotech).
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
53
hERG assays . Compounds were evaluated for block of the hERG K channel using CHO cells
stably expressing the hERG gene (bSys, Witterswil, Switzerland) and the QPatch platform (Sophion,
Ballerup, Denmark). K tail currents were measured at ꢀ40 mV following a 500ꢀms depolarization to
+
2+
+
20 mV from a holding voltage of ꢀ80 mV. The external solution contained 4 mM K , 1 mM Mg and
2
+
1
.2 mM Ca . Compound effects were quantified 3 minutes after application to the cells.
Mice, in life procedures. All animal housing and experiments were performed in agreement with
the Swiss Federal Ordinance for animal protection, the animal welfare guide lines from the Cantonal
Vetenary Office Basellandschaft, and the Actelion Pharmaceuticals Ltd internal animal welfare
guidelines. Eight to twelve weekꢀold, specific pathogenꢀfree, female NMRI mice weighing 27 to 30 g
were used for all studies (Harlan, The Netherlands). Mice were rendered neutropenic (neutrophils <
3
1
00/mm ) by injecting cyclophosphamide (Sigma) intraperitoneally 4 days (150 mg/kg) and 1 day
ACS Paragon Plus Environment

Page 25 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
100 mg/kg) before thigh infection. Previous studies have shown that this regimen produces
neutropenia in this model for 5 days. Broth cultures of freshly plated S. aureus (laboratory descendent
of DSM 11823) were grown to logarithmic phase overnight to an absorbance of 0.3 to 0.9 at 580 nm.
Thigh infections with S. aureus were performed by injection of 0.05 ml of inoculum (107.0ꢀ8.0
CFU/ml) intra muscularly into both thighs of mice 2 h before administration of compound 49e
(formulated in 5% mannitol in water, administered at 10 ml/kg). Postꢀmortem, thigh samples were
homogenized (UltraꢀTurrax® tube drive ꢀ IKA), serially diluted and plated (spiral platter EddyJet®
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
(
IUL)). Results were expressed as colony forming units (CFU; CFUꢀcounter Flash & Grow® ꢀ IUL),
correlating to remaining bacterial load in the thigh at the end of the study.
Pharmacokinetics in the Rat. The pharmacokinetic profile was determined in the male Wistar rat
(
n = 3). For this purpose, compound 34c was formulated as an aqueous solution at pH 4.5 for
intravenous administration at a dose of 1 mg/kg and as aqueous suspension pH 5.6 for oral dosing at
0 mg/kg; compound 49e was formulated as a solution in 5% DMSO in pure water for intravenous
1
administration at a dose of 0.5 mg/kg, and as a microsuspension in 7.5% aqueous modified gelatin for
oral dosing at 1 mg/kg; compound 52c was formulated as a solution in 10 % PG, 10% PEG400 in
water at pH 7 for intravenous administration at a dose of 1 mg/kg and as a microsuspension in 0.5% in
methylcellulose for oral dosing at 10 mg/kg. Blood samples were taken at regular time intervals over a
period of 24 h from a preimplanted catheter, and plasma was generated by centrifugation at 4000 rpm
at 4 °C. Plasma concentrations were determined using a specific and sensitive LCMS/ MS method
with a limit of quantification of 1.5 ng/mL.
General Chemical Methods. Starting materials, reagents and solvents were obtained from
commercial sources and used as received. All reactions were carried out with continuous stirring under
1
13
atmosphere of dry nitrogen. The resonance frequency for H ( C) on a Varian Gemini 300 is 300MHz
75MHz). Chemical shifts ( ) are reported in parts per million (ppm) relative to deuterated solvent as
the internal standard ( H: CDCl 7.26ppm, DMSOꢀd6 2.50ppm), coupling constants (J) are in hertz
(
δ
δ
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
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
(
Hz). Peak multiplicities are expressed as follows: singlet (s), doublet (d), doublet of doublets (dd),
triplet (t), quartet (q), multiplet (m), broad singlet (br s). Melting points were recorded with a Büchi
520 apparatus and are uncorrected. Purification of intermediates and final products was carried out on
2
normal phase using an ISCO CombiFlash system and prepacked SiO cartridges eluted with optimized
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
gradients of either heptaneꢀethyl acetate mixture or dichloromethaneꢀmethanol (doped with 1% v/v of
aq NH OH for basic compounds). Progress of the reactions was monitored by thinꢀlayer
4
2
54
chromatography (TLC) analysis (Merck, 0.2 mm silica gel 60 F on glass plates) or by LCꢀMS
(
Methods and equipments are described in Supporting Information). Unless otherwise stated, all target
compounds had purity > 95%, established on a Waters Atlantis T3, 5 ꢂm, 4.6 mm × 30 mm, eluting
with a gradient of 5−95% of acetonitrile in water containing 0.04% of trifluoroacetic acid; or on a
Waters XBridge C18, OBD, 5 ꢂm, 4.6 mm × 50 mm (Waters, Switzerland), eluting with a gradient of
5
4
−95% of acetonitrile in water containing 13 mM of NH OH UV at 230 nm and 254nm. Purity and
identity were further confirmed by NMR spectroscopy. Optical rotations were measured on a Jasco Pꢀ
1
030 apparatus. HPLCꢀHRMS Methods and equipments are described in Supporting Information.
t
Retention times ( R) are expressed in minutes. Personal protective equipment (mask and gloves)
should be worn when handling aldehydes 19bꢀ27b (irritant). Compound 23a is commercially
available.
tertꢀbutyl ((2S)ꢀ1ꢀhydroxyꢀ4ꢀ(oxiranꢀ2ꢀyl)butanꢀ2ꢀyl)carbamate (9). To a suspension of
30
LiBH
4
(1.15 g, 53 mmol) in THF (300 mL) at rt was added a solution of 8 (12.9 g, 53 mmol) in THF
(
100 mL). The mixture was stirred at rt for 4 h, poured into water and extracted with EtOAc. The
organic layer was washed with brine, dried over MgSO and concentrated to give the alcohol (11.4 g,
4
1
9
9% yield). Colorless oil; H NMR
δ
(CDCl ): 5.75ꢀ5.65 (m, 1H), 5.00ꢀ4.90 (m, 2H), 4.5 (br, 1H,
3
OH), 3.70ꢀ3.45 (m, 3H), 2.10ꢀ2.00 (m, 2H), 1.60ꢀ1.35 (m, 2H), 1.38 (s, 9H). The latter (11.4 g, 53
mmol) was dissolved in DCE (300 mL) and water (250 mL) and 1 M phosphate buffer pH 8 (150 mL)
was added. m-CPBA (14.3 g) was added and the mixture vigorously stirred overnight. The two phases
were separated and the aq phase was extracted once more with DCM. The combined organic layers
3 4
were washed with a saturated NaHCO solution, dried over MgSO , filtered and concentrated to
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
dryness. The residue was purified by chromatography (heptaneꢀEtOAc) to give 9 (7.74 g, 63% yield,
1
mixture of diastereomers). Colorless oil; H NMR
3
δ (CDCl ): 4.85ꢀ4.90 (m, 1H), 3.50ꢀ3.75 (m, 3H),
2.90ꢀ3.00 (m, 1H), 2.51ꢀ2.80 (m, 1H), 2.6 (br, 1H, OH), 2.50ꢀ2.55 (m, 1H), 1.40ꢀ1.80 (m, 4H), 1.42 (s,
9H).
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
tertꢀbutyl ((3S,6R)ꢀ6ꢀ(hydroxymethyl)tetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (entꢀ7).
solution of the epoxide 9 (2.3 g, 10 mmol) in DCM (50 mL) was treated with CSA (0.23 g, 1 mmol).
The mixture was stirred at rt for 3 h, diluted with a saturated NaHCO solution (100 mL). The two
layers were separated and the organic layer was dried over Na SO , filtered and concentrated to
A
3
2
4
dryness. The residue was purified by chromatography (EtOAc) to give entꢀ7 (0.874 g, 37% yield).
1
Colorless solid; m.p.= 112.9 °C; [
α
] = + 7.1 (c 0.96, MeOH); H NMR
δ
(CDCl ): 4.28 (br, 1H), 4.10ꢀ
D
3
4.20ꢀ10 (m, 1H), 3.30ꢀ3.70 (m, 5H), 3.04 (t, J =10.6 Hz, 1H), 2.00ꢀ2.20 (m, 2H), 1.20ꢀ1.75 (m, 11H);
13
+
C NMR
δ (DMSOꢀd6) 155.6, 78.5, 78.3, 70.9, 65.0, 47.1, 29.8, 28.9 (3C), 27.8;; ESIꢀMS (M+H)
+
t
m/z 232.2; HR ESIꢀMS (M+Na) m/z = 254.1370 (calc. for C11
4
H21NO Na: 254.1368); R = 0.84.
tertꢀbutyl ((3R,6S)ꢀ6ꢀcarbamoyltetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (12). To a solution
29
of 11 (3 g, 12.2 mmol, prepared from 7 as described ) in EtOAc (50 mL) was added NHS (1.5 g, 13
mmol) and DCC (2.7 g, 13.1 mmol). The reaction was stirred overnight at rt. After filtration and
concentration to dryness, the residue was taken up in THF (180 mL). Gaseous NH was bubbled
3
through the solution for 10 min, and the reaction proceeded at rt for 1 h. Silica gel (20 g) was added,
and the volatiles were removed in vacuo. The residue was purified by chromatography (DCMꢀMeOH
1
1
9ꢀ1) to afford 12 (1.3 g, 43%). White solid; H NMR
δ
(DMSOꢀd6): 7.15 (br s, 1H), 7.07 (br s, 1H),
6.83 (d, J = 8.0 Hz, 1H), 3.85 (m, 1H), 3.56 (m, 1H), 3.32 (m, 1H), 3.01 (t, 10.6 Hz, 1H), 1.84ꢀ1.96
13
(
m, 2H), 1.34ꢀ1.43 (m, 2H), 1.36 (s, 9H); C NMR
δ
(DMSOꢀd6): 173.2, 155.5, 78.3, 76.9, 70.4, 46.5,
+
t
2
9.6, 28.7 (3C), 28.5; HR ESIꢀMS (M+H) m/z = 245.1502 (calc. for C11
H
21
N O
2 4
: 245.1501); R =
0
.82.
tertꢀbutyl ((3R,6S)ꢀ6ꢀformyltetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (13). To a solution of
(
7
(37.5 g, 162.13 mmol) in DCM (310 mL) cooled to ꢀ10°C was added DIPEA (84.75 mL, 495 mmol).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
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 solution of py.SO
3
complex (69.47 g, 218 mmol) in DMSO (225 mL) was slowly added. The
reaction mixture was stirred for 2 h at 0 °C .The reaction mixture was partitioned between water (150
mL) and DCM (220 mL). The two layers were separated and the aq layer was extracted twice with
2 4
DCM (2 x 150 mL). The combined organic layers were dried over Na SO , filtered and concentrated
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
to dryness. The residue was filtered through silica gel (EtOAcꢀheptane) to afford 13 (33.58 g, 90%).
1
White solid; m.p.= 108.9 °C; H NMR
3
δ (CDCl ): 9.63 (br s, 1H), 4.37 (br s, 1H), 4.16 (ddd, J = 1.7,
4
1
1
.5, 10.9 Hz, 1H), 3.72 (dd, J = 2.9, 10.7 Hz, 1H), 3.62 (br s, 1H), 3.14 (t, J = 10.4 Hz, 1H), 2.12 (m,
13
H), 1.96 (m, 1H), 1.56 (m, 1H), 1.44 (s, 9H), 1.38 (overlaid m, 1H); C NMR
δ
(CDCl
3
): 201.3,
+
+
55.1, 80.6, 79.7, 70.8, 45.9, 29.2, 28.4 (3C), 25.0; ESIꢀMS (M+H) m/z 230.0; HR ESIꢀMS (M+Na)
t
m/z = 252.1214 (calc. for C11
H19NO
4
Na: 252.1211); R = 0.78.
tertꢀbutyl ((3R,6S)ꢀ6ꢀvinyltetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (14). t-BuOK (31.74 g,
282.89 mmol) was added in one portion to a white suspension of methyltriphenylphosphonium
bromide (101.05 g, 282.89 mmol) in THF (340 mL) at rt under nitrogen. The resulting suspension was
stirred for 1 h at rt and a solution of 13 (32.43 g, 141.44 mmol) in THF (85 mL) was added. The
4
mixture was stirred 30 min. at rt. A 10% NaHSO solution (120 mL) was added and the mixture was
diluted with EtOAc (200 mL). The two layers were decanted and the aq layer was extracted once with
EtOAc (250 mL). The combined organic layers were washed with brine, dried over Na SO , filtered
2
4
and concentrated to dryness. The residue was purified by chromatography (heptaneꢀEtOAc) to afford
1
14 (28.98 g, 90%). White solid; H NMR
δ
(CDCl
3
): 5.84 (ddd, J = 5.6, 10.5, 17.3 Hz, 1H), 5.24 (dt, J
=
1.5, 17.3 Hz, 1H), 5.12 (dt, J = 1.5, 10.5 Hz, 1H), 4.26 (br. s, 1H), 4.10 (ddd, J = 2.1, 4.7, 10.8 Hz,
1H), 3.73 (m, 1H), 3.61 (m, 1H), 3.06 (t, J = 10.5 Hz, 1H), 2.10 (m, 1H), 1.79 (m, 1H), 1.25ꢀ1.60 (m,
+
2H), 1.44 (s, 9H); ESIꢀMS (M+H) m/z= 228.2.
tertꢀbutyl ((3R,6S)ꢀ6ꢀ(2ꢀoxoethyl)tetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (15). To an iceꢀ
chilled solution of 14 (1.47 g, 6.46 mmol) in THF (30 mL) was added 9ꢀBBNꢀH (2.37 g, 19.4 mmol).
The mixture was stirred overnight at rt. After cooling to 0 °C, 3 N NaOH (25 mL) and 50% aq H
12 mL) were cautiously added. The resulting mixture was stirred at rt for 2 h. The two layers were
2
O
2
(
decanted and the aq layer was extracted with EtOAc (2 x 200 mL). The combined organic layers were
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
washed with 10% aq NaHSO
3
2 4
solution (100 mL), water (50 mL) and dried over Na SO . After
filtration and concentration to dryness, the residue was purified by chromatography (heptaneꢀEtOAc)
1
to afford the resulting alcohol (1.15 g, 72%). White solid; H NMR
3
δ (CDCl ): 4.16 (br s, 1H), 4.01
(ddd, J = 2.2, 4.7, 10.7 Hz, 1H), 3.69 (t, J = 5.2 Hz, 2H), 3.52 (br s, 1H), 3.39 (m, 1H), 2.93 (t, J =
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
0.7 Hz, 1H), 2.02 (br s, 1H), 2.00 (m, 1H), 1.58ꢀ1.72 (m, 3H), 1.45 (m, 1H), 1.37 (s, 9H), 1.24 (qd, J
+
=
4.1, 12.2 Hz, 1H); ESIꢀMS (M+H) m/z 246.4. The latter alcohol (1.15 g, 4.68 mmol) was dissolved
in DCM (20 mL), and after cooling to 0 °C, a solution of DessꢀMartin periodinane (15 wt% in DCM,
8 mL) was added. After warming to rt, the reaction proceeded for 2 h. The reaction mixture was
1
concentrated to dryness and the residue purified by chromatography (heptaneꢀEtOAc) to afford 15
1
(
1.06 g, 93%). White solid; H NMR
δ
(CDCl
3
): 9.79 (t, J = 2.3 Hz, 1H), 4.29 (br s, 1H), 4.07 (ddd, J
=
2.1, 4.7, 10.7 Hz, 1H), 3.77 (m, 1H), 3.61 (br s, 1H), 3.04 (t, J = 10.7 Hz, 1H), 2.65 (ddd, J = 2.3,
7
1
4
2
.6, 16.5 Hz, 1H), 2.49 (ddd, J = 2.3, 4.8, 16.5 Hz, 1H), 2.12 (m, 1H), 1.79 (m, 1H), 1.50 (overlaid m,
1
3
H), 1.49 (s, 9H), 1.33 (qd, J = 4.1, 12.2 Hz, 1H); C NMR
3
δ (CDCl ): 200.9, 155.1, 79.5, 72.3, 71.3,
+
9.2, 46.2, 30.7, 30.3, 28.3 (3C); HR ESIꢀMS (M+Na) m/z = 266.1367 (calc. for C12
H21NO
4
Na:
t
66.1368); R = 0.81.
tertꢀbutyl ((3R,6S)ꢀ6ꢀ(iodomethyl)tetrahydroꢀ2Hꢀpyranꢀ3ꢀyl)carbamate (16). To an iceꢀ
2
9
3
chilled solution of 7 (40.9 g, 176.8 mmol) in DCM (840 mL) were added successively Et N (59.5
mL, 423.9 mmol), DMAP (3.01 g, 24.56 mmol) and TsCl (42.4 g, 222.4 mmol). The reaction
proceeded 4 h with warming to rt. A saturated NaHCO solution (350 mL) was added. The two phases
3
were separated and the organic layer was evaporated in vacuo. The residue was diluted with EtOAc
(
4
900 mL) and the organic layer was washed three times with a saturated CuSO solution (3 x 200 mL),
brine (200 mL), dried over MgSO , filtered and concentrated to dryness to afford the crude tosylate
4
1
(
75.83 g); H NMR
δ
(CDCl ): 7.78 (d, J = 8.7 Hz, 2H), 7.33 (d, J = 8.7 Hz, 2H), 4.20 (br s, 1H), 4.01
3
(ddd, J = 2.1, 4.7, 10.8 Hz, 1H), 3.96 (d, J = 5.7 Hz, 2H), 3.54 (br s, 1H), 3.48 (m, 1H), 2.93 (t, J =
10.8 Hz, 1H), 2.44 (s, 3H), 2.09 (m, 1H), 1.69 (m, 1H), 1.48ꢀ1.18 (m, 2H), 1.42 (s, 9H); ESIꢀMS
+
(M+H) m/z
=
386.2. The latter (75.83 g) was taken up in acetone (700 mL) and NaI (90.0 g, 600.4
mmol) was added. The reaction mixture was refluxed 24 h. The reaction mixture was cooled to rt and
ACS Paragon Plus Environment