Synthesis and Pharmacokinetic Evaluation of Siderophore Biosynthesis Inhibitors for Mycobacterium tuberculosis

Article abstract of DOI:10.1021/acs.jmedchem.5b00391

MbtA catalyzes the first committed biosynthetic step of the mycobactins, which are important virulence factors associated with iron acquisition in Mycobacterium tuberculosis. MbtA is a validated therapeutic target for antitubercular drug development. 5′-O-[N-(Salicyl)sulfamoyl]adenosine (1) is a bisubstrate inhibitor of MbtA and exhibits exceptionally potent biochemical and antitubercular activity. However, 1 suffers from suboptimal drug disposition properties resulting in a short half-life (t_1/2), low exposure (AUC), and low bioavailability (F). Four strategies were pursued to address these liabilities including the synthesis of prodrugs, increasing the pK_a of the acyl-sulfonyl moiety, modulation of the lipophilicity, and strategic introduction of fluorine into 1. Complete pharmacokinetic (PK) analysis of all compounds was performed. The most successful modifications involved fluorination of the nucleoside that provided substantial improvements in t_1/2 and AUC. Increasing the pK_a of the acyl-sulfonyl linker yielded incremental enhancements, while modulation of the lipophilicity and prodrug approaches led to substantially poorer PK parameters.

Full text of DOI:10.1021/acs.jmedchem.5b00391

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Synthesis and Pharmacokinetic Evaluation of Siderophore
Biosynthesis Inhibitors for Mycobacterium tuberculosis
Kathryn M. Nelson, Kishore Viswanathan, Surendra Dawadi, Benjamin P
Duckworth, Helena I. Boshoff, Clifton E. Barry, and Courtney C. Aldrich
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00391 • Publication Date (Web): 25 Jun 2015
Downloaded from http://pubs.acs.org on July 1, 2015
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 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and Pharmacokinetic Evaluation of Siderophore
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
Biosynthesis Inhibitors for Mycobacterium tuberculosis
a#
b#
b
a
Kathryn M. Nelson, Kishore Viswanathan, Surendra Dawadi, Benjamin P. Duckworth,
c
rd c
a,b
Helena I. Boshoff, Clifton E. Barry 3 , Courtney C. Aldrich *
a
Center for Drug Design, Academic Health Center, University of Minnesota, Minneapolis, MN
b
5
5455, USA; Department of Medicinal Chemistry, University of Minnesota, Minneapolis,
c
Minnesota, MN 55455, USA; Tuberculosis Research Section, National Institute of Allergy and
Infectious Diseases, Bethesda, MD 20892, USA
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract. MbtA catalyzes the first committed biosynthetic step of the mycobactins, which are
important virulence factors associated with iron acquisition in Mycobacterium tuberculosis.
MbtA is a validated therapeutic target for antitubercular drug development. 5′ꢀOꢀ[Nꢀ
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
(salicyl)sulfamoyl]adenosine (1) is a bisubstrate inhibitor of MbtA and exhibits exceptionally
potent biochemical and antitubercular activity. However, 1 suffers from subꢀoptimal drug
disposition properties resulting in a short halfꢀlife (t ), low exposure (AUC), and low
1
/2
bioavailability (F). Four strategies were pursued to address these liabilities including the
synthesis of prodrugs, increasing the pK of the acylꢀsulfonyl moiety, modulation of the
a
lipophilicity, and strategic introduction of fluorine into 1. Complete pharmacokinetic (PK)
analysis of all compounds was performed. The most successful modifications involved
fluorination of the nucleoside that provided substantial improvements in t1/2 and AUC.
Increasing the pK of the acylꢀsulfonyl linker yielded incremental enhancements while
a
modulation of the lipophilicity and prodrug approaches led to substantially poorer PK
parameters.
2
ACS Paragon Plus Environment

Page 3 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Mycobacterium tuberculosis (Mtb), often considered the most successful human pathogen,
infects more than two billion people and remains a leading source of infectious disease mortality
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
and morbidity. Tuberculosis (TB), the disease caused by Mtb or one of seven other closely
2
related Mycobacterium species, has received renewed attention since the World Health
Organization (WHO) declared TB a global health emergency due the alarming rise of drug
1
resistance, pandemic of HIVꢀTB coꢀinfection, and lack of new therapeutics. Multidrug resistant
TB (MDRꢀTB) is defined as resistance to isoniazid and rifampin, which are the two most
effective TB drugs. Approximately, 9% of MDRꢀTB cases are additionally resistant to one of
the injectable antibiotics used for TB (aminoglycosides or capreomycins) and any
1
fluoroquinoline; these TB strains are classified as extensively drug resistant (XDRꢀTB). Mtb is
extraordinarily difficult to treat, requiring prolonged therapy with multiple antibiotics for 6
3
months for simple drug susceptible infections and up to 20 months for drugꢀresistant infections,
4
with poor treatment outcomes in the latter cases. Eradication of the global TB pandemic, a
longꢀterm goal of the WHO, will require improved diagnostics, vaccines, and new drugs.
Excellent progress has been made over the last decade and, for the first time in over forty years,
two new TB drugs (Bedaquiline and Delamanid) were approved for MDRꢀTB by US and
5
, 6
European regulatory agencies; although initial enthusiasm has been marred by safety concerns.
While a specific target candidate profile has not yet been established in the TB field, ideal
criteria for new antitubercular agents as advocated by the TB Alliance
(http://www.tballiance.org) include: the potential to shorten and simplify the treatment regimen,
effectiveness against multidrug resistant strains, and compatibility with HIVꢀTB coꢀinfection.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 59
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
In conceiving of new therapeutic strategies for TB we have focused on mycobacterial iron
acquisition, an essential Mtb process with a long, storied history. In the midꢀ19th century,
decades before Robert Koch identified Mtb as the etiological agent of TB, the famous French
physician Armand Trousseau noticed some anemic patients rapidly succumbed to TB
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
7
reactivation when given iron supplements. This observation is now mechanistically understood
since Mtb, as well as most microorganisms, requires iron as a redox cofactor for numerous
essential cellular processes. However, normal physiological conditions restrict the free ferric
ꢀ24
concentration to an estimated 10 M, far below the concentration the bacteria needs for
8
, 9
proliferation.
The process of withholding an essential nutrient is referred to as nutritional
immunity, and represents an extremely effective strategy to prevent bacterial growth. To combat
iron deficiency, bacteria produce small molecule iron chelators known as siderophores, often at
great metabolic expense, that are secreted into the surrounding environment and abstract iron
1
0, 11
from host proteins.
The siderophoreꢀferric complexes are then reimported into the bacteria
cytosol, usually by specific siderophore transporters, where iron is released through hydrolytic
3
+
2+
destruction of the siderophore or through reduction of Fe to Fe . The tripodal mycobactins
produced by Mycobacterium species were the first class of siderophores characterized in 1912
1
2ꢀ14
and found to be an essential substance for mycobacterial growth.
Siderophores of diverse
chemical structures are now known to be produced by virtually all microorganisms, and even
15, 16
some plants.
Following the publication of the Mtb genome in 1998, the Walsh group identified a locus of
10 genes (mbtAꢀJ) encoding for a mixed nonribosomal peptide synthetaseꢀpolyketide synthase
(NRPSꢀPKS) pathway responsible for biosynthesis of the mycobactin peptidic core. This was
subsequently verified through comprehensive mutational analysis and functional
1
7ꢀ19
characterization.
Gokhale and coꢀworkers later identified a second locus of four genes
4
ACS Paragon Plus Environment

Page 5 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
0
(
mbtKꢀN) required for synthesis of the lipid sideꢀchain found in the mycobactins. Mtb mutants
deficient in mycobactin production or transport have been shown to have severe growth defects
21ꢀ23
under iron limiting conditions, and are incapable of establishing an infection in vivo.
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
Collectively, these results provide compelling evidence for targeting iron acquisition in Mtb with
small molecules. Indeed, two diametrically opposed approaches have been pursued through
24, 25
synthesis of mycobactin analogues and inhibitors of mycobactin biosynthesis.
In spectacular
application of the first approach, Miller and coꢀworkers synthesized a mycobactinꢀartemisinin
conjugate that exploits the mycobactin uptake system for delivery of artemisinin, which
2
+
generates toxic reactive oxygen species upon intracellular reductive release of Fe from
2
6
mycobactin.
Inhibition of mycobactin biosynthesis has also been fruitful, and inhibitors of
25, 27, 28
29
30
31
enzymes within the pathway have been described for MbtA,
MbtI, MbtM, and PptT.
MbtA, responsible for incorporation of salicylic acid into the mycobactin core structure, has
2
8, 32ꢀ36
been the most extensively studied for inhibitor development.
The bisubstrate inhibitor 5′ꢀ
Oꢀ[Nꢀ(salicyl)sulfamoyl]adenosine (1, Figure 1C) was independently described by three groups
and found to exhibit exceptionally potent enzyme inhibition of MbtA, with an apparent K of ~6
i
nM under physiologically relevant superꢀsaturating concentrations of its substrates ATP and
2
5, 27, 28
salicylic acid.
This compound possesses potent antitubercular activity, with a minimum
inhibitory concentration of 0.39 µM under ironꢀlimiting conditions in vitro, and was also
effective in an acute murine TB infection model when dosed intraperitoneally, thereby
3
7
chemically validating MbtA as an antitubercular target.
Unfortunately, 1 possesses subꢀ
optimal pharmacokinetic (PK) parameters including poor oral bioavailability, low oral exposure,
37
and rapid clearance. We report herein four approaches to address the poor PK parameters of 1
including: the synthesis of prodrugs (2–3), analogs that increase the sulfamate pK (4–5),
a
modulation of the lipophiliciy of the highly polar nucleoside (6–7), and strategic introduction of
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 59
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
fluorine into the nucleoside (8–10) as shown in Figure 1C. Pharmacokinetic analysis of all
compounds was performed in conjunction with selective in vitro studies to understand the PK
behavior.
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 1. Mycobactins, biosynthesis, and inhibitors. A. Structures of the lipid soluble
mycobactins and water soluble mycobactins (refered to as carboxymycobactins) produced by
Mtb. The mycobactins and carboxymycobactins are synthesized as a suite of compounds that
differ in the chain length of the lipid tail on the central lysine residue. The most abundant chain
length is depicted. B. Biosynthesis of all mycobactins is initiated with the ATPꢀdependent
activation of salicylic acid by MbtA to form an intermediate acylꢀadenylate. In a second halfꢀ
reaction, MbtA ligates salicylic acid onto the Nꢀterminal aryl acid carrier protein domain of
MbtB. MbtB, in combination with MbtE, MbtC, MbtD, and MbtF (not shown), sequentially
builds the mycobactin peptidic core scaffold. C. 5′ꢀOꢀ[Nꢀ(salicyl)sulfamoyl]adenosine (1) was
the first described inhibitor of MbtA and is considered a bisubstrate inhibitor that mimics the
intermediate acylꢀadenylate. This modular inhibitor scaffold contains aryl (blue), linker (black),
glycosyl (pink), and nucleobase (green) domains. Prior SAR studies have examined the
importance of each atom of this scaffold and provided a comprehensive understanding of the
structureꢀactivity relationships (SAR) that govern enzyme inhibition and wholeꢀcell activity.
This information has been used to design prodrugs 2–3 and analogs 4–10 in order to explore the
SAR that governs pharmacokinetic behavior.
6
ACS Paragon Plus Environment

Page 7 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Results and Discussion
Chemistry. The simple diꢀacetate prodrug 2, wherein the polar 2′ and 3′ alcohols are
masked, as well as triꢀacetate 3, in which the phenol is additionally acylated, were initially
explored to improve the bioavailability and exposure of 1. Acetates are quickly hydrolyzed by
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
38
39
serum esterases in vivo. The diꢀacetate 2 was obtained by acylation of sulfamate 11 with Nꢀ
2
8
hydroxysuccinimidyl 2ꢀbenzyloxybenzoate to provide 12 (Scheme 1A). Trifluoracetic acid
TFA) mediated deprotection of the acetonide followed by acetylation of the resultant free 2ʹ and
ʹ alcohols afforded 13. Catalytic hydrogenolysis and reverseꢀphase high performance liquid
(
3
chromatography (HPLC) purification furnished diꢀacetate product 2. The triꢀacetate derivative 3
was prepared by direct acetylation of 1 employing acetic anhydride and pyridine in
dimethylformamide (DMF) (Scheme 1B). These conditions were notable since competitive
6
acetylation at the N position of adenosine was not observed. Compound 3 was first purified on
normal phase silica and isolated as an approximate 5:1 mixture of triꢀand diꢀacetates 3 and 2.
Further purification by reverseꢀphase HPLC was required to obtain pure 3.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 59
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 1 . Synthesis of prodrugs 2 and 3.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reaction conditions: (a) Nꢀhydroxysuccinimidyl 2ꢀbenzyloxybenzoate, Cs CO , DMF, 0 °C to rt, 21
h, 62%; (b) 80% aqueous TFA, 0 °C to rt, 4 h; (c) Ac O, pyridine, DMF, rt, 16 h, 63% (2 steps); (d) H ,
Pd/C, MeOH, rt, 2 h, 42%; (e) Ac O, pyridine, DMF, rt, 20 h, 46%.
2
3
2
2
2
In an effort to further increase the pK of the acylsulfonyl linker’s acidic nitrogen, the next
a
analogues in this set replaced the sulfamate linker with a sulfamide and added a paraꢀamino
substitution on the salicyl moiety to provide 4 and 5, respectively (Scheme 2). We developed a
new synthesis of 4 (i.e. 14→15→16→4) resulting in a 10ꢀfold improved overall yield relative to
2
8
our initially disclosed route.
Our revised synthesis features a Mitsunobu reaction of tertꢀ
40
6
6
39
butoxycarbonyl (Boc) sulfamide (29) with N ,N ꢀbis(Boc)ꢀisopropylideneadenosine (14) to
afford 15 in an impressive 96% yield for direct installation of the sulfamide group at Cꢀ5′ of the
6
nucleoside. Here the N ꢀbisꢀBoc protecting group in 14 is essential to subdue cyclonucleoside
formation that occurs from cyclization of Nꢀ3 onto Cꢀ5' during the Mitsunobu reaction by
4
1, 42
attenuating the nucleophilicity at Nꢀ3.
Coupling 15 to the Nꢀhydroxysuccinimide (NHS)
32
ester of 2ꢀOꢀmethoxymethylsalicyate (31) employing Cs CO in DMF provided 16 in 60%
2
3
yield. The adjacent Boc group on the sulfamide moiety is responsible for the reduced yields of
this acylation reaction relative to that observed with simple sulfamate analogs, which typically
8
ACS Paragon Plus Environment

Page 9 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
proceed in 80–90% yields. Global deprotection with 80% aqueous TFA furnished sulfamide 4
that was purified by reverseꢀphase HPLC and isolated in 65% yield as the triethylammonium
salt. Compound 5 was prepared in an analogous fashion (15→17→18→5) except sulfamide 15
4
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
was acylated with benzyloxycarbonyl (Cbz) protected pꢀaminosalicylate (32) employing in situ
25, 33
activation with carbonyl diimidazole (CDI)
to afford 17. Sequential hydrogenolysis of the
Cbz group with Pd/C in methanol to 18 followed by acidic deprotection furnished 5.
6
6
N ꢀCyclopropyl 6 and N ꢀcyclopropylꢀ2ꢀphenyl 7 containing the optimal sulfamide linkages
were prepared in an effort to increase the lipophilicity (Scheme 2). These individual
modifications on a sulfamate template were previously shown to improve potency and
3
4
selectivity.
For the synthesis of 6 we explored a complimentary synthetic route
(
19→20→21→22→6) employing sulfamoylation with the novel dimethylaminopyridiniumꢀ
4
4
6
activated reagent 30 as reported by Tan and coꢀworkers.
N ꢀCyclopropylꢀ2′,3′ꢀOꢀ
3
4
isopropylideneadenosine 19 was elaborated to 5′ꢀamino derivative 20 through sequential
azidation and hydrogenation. Sulfamoylation of aminonucleoside 20 was achieved with
44
dimethylaminopyridiniumꢀactivated 30 followed by deprotection of the Cbz group to give the
free sulfamide 21 that was converted to 6 through sequential salicylation and global deprotection
6
using the standard conditions developed for 4. N ꢀcyclopropylꢀ2ꢀphenyl 7 was prepared from 23
in analogy to 4 (23→24→25→7).
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 59
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 . Synthesis of 4–8.
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
a
Reaction conditions: (a) 29, PPh , DIAD, THF, 0 °C to rt, 16 h, 15=96%, 24=91%, 27=89%; (b) i)
3
30, CH Cl , rt, 14 h; ii) then H , Pd/C, MeOH–THF, rt, 22 h, 21=62%; (c) 31, Cs CO , DMF, 0 °C to rt,
2
2
2
2
3
16 h, 16=60%, 22=43%; (d) i) 32, CDI, THF, 60 °C, 2 h; ii) then 15, Cs CO , THF, rt, 16 h, 17=55%; (e)
H , Pd/C, MeOH, rt, 2 h; (f) 80% aqueous TFA, 0 °C to rt, 4 h, 4=65%, 5=34% (2 steps from 17), 6=61%,
=18% (2 steps from 24), 8=38 % (2 steps from 27); (g) DPPA, NaN , 15ꢀcrownꢀ5, THF, 0 °C to rt, 4.5 h;
h) H , Pd/C, MeOH, rt, 20=44%, (2 steps from 19). Nonstandard abbreviations: IPK = isopropylidene
2
3
2
7
3
(
2
ketal.
To complete our narrative, we explored fluorinated analogs 8–10, since fluorine is known to
4
5
affect absorptionꢀdistributionꢀmetabolismꢀexcretion (ADME) properties.
The enhanced PK
parameters of 8 prompted the synthesis of 9–10 (vide infra). Fluorinated analog 8 was prepared
from 26 in analogy to 4 (26→27→28→8) as presented in Scheme 2. The synthesis of 2′ꢀfluoroꢀ
2ꢀphenyl analog 9 is illustrated in Scheme 3 starting from commercially available 2′ꢀdeoxyꢀ2′ꢀ
fluoroguanosine 33. Acetylation of the 3' and 5' alcohols provided 34, which was transformed to
3
5 by sequential chlorination employing POCl in dimethylaniline (DMA) promoted by Et NCl
3
4
4
6, 47
followed by Sandmeyerꢀlike reaction with isoamyl nitrite and diiodomethane.
Ammonolysis
yielded adenosine derivative 36 that was immediately tertꢀbutyldimethylsilyl (TBS) protected to
provide 37. The requisite Cꢀ2 phenyl group was incorporated by BuchwaldꢀHartwig coupling
48
with phenylboronic acid to give 38. Regioselective monodesilylation of the 5′ TBS group with
1
0
ACS Paragon Plus Environment

Page 11 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
AcOH–THF–H O (9:1:1) afforded the key nucleoside intermediate 39. For installation of the
2
sulfamide, we turned to new chemistry developed in our lab, which involves treatment of a 5′ꢀ
aminonucleoside with sulfamide (NH SO NH ), inspired by a sideꢀreaction reported by Takai
2
2
2
4
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and coꢀworkers. In the present synthesis this was achieved by conversion of the 5′ꢀalcohol in
9 to azide 40. The azide was reduced by catalytic hydrogenation to provide a crude amine that
was not isolated, but treated directly with sulfamide in dioxane at 90 °C to cleanly afford 41 in
7% yield. This new sulfamide synthesis is high yielding, experimentally simpler than our
3
8
Mitsunobu protocol since it avoids contamination with triphenylphosphine oxide, and obviates
6
N ꢀprotection on the nucleoside. The completion of the synthesis was accomplished using our
standard salicylation method with 31 followed by TBS deprotection to furnish 9.
a
Scheme 3 . Synthesis of 9.
a
Reaction conditions: (a) Ac O, pyridine, DMF, rt, 12 h, 79%; (b) POCl , DMA, Et NCl, reflux, 20
2
3
4
min, 86%; (c) CH I , I , isoamylnitrite, 80 °C, 1 h, 95%; (d) NH , MeOH, 70 °C, 24 h, 56%; (e) TBSCl,
2
2
2
3
imidazole, DMAP, DMF, rt, 24 h, 97%; (f) Pd(OAc) , CyJohnPhos, K PO , Ph(BOH) , 100 °C, 12 h,
2
3
4
2
6
8
0
7%; (g) AcOH–THF–H O (9:1:1), THF, 23 °C, 24 h, 70%; (h) DPPA, DBU, NaN , 15ꢀcrownꢀ5, 24 h,
2
3
5%; (i) Pd/C , MeOH, rt, 1 h; (j) NH SO NH , dioxane, 90 °C, 3 h, 87% (2 steps); (k) 31, Cs CO , DMF,
2
2
2
2
3
°C to rt, 16 h; (l) 80% aq TFA, 0 °C, 3 h, 45% (2 steps from 41). Nonstandard abbreviations:
CyJohnPhos = (2ꢀbiphenyl)dicyclohexylphosphine.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 59
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
Based on the improved PK properties of 2′ꢀfluoro analog 8 (vide infra), we also pursued the
synthesis of 2′,3′ꢀdideoxyꢀ2′ꢀfluoro derivative 10. Synthesis commenced from 2′ꢀdeoxyꢀ2′ꢀ
50
fluoroadenosine 42 that was regioselectively TBS protected at the Cꢀ5' alcohol to furnish 43.
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
BartonꢀMcCombie deoxygenation was accomplished in two steps through conversion to the
corresponding thionocarbonate 44 followed by treatment with Bu SnH in refluxing toluene to
3
5
0
yield 45.
Deprotection of the TBS group with TFA provided alcohol 46, which was
5
1
transformed to azide 47 using methodology reported by Liu and Austin. The sulfamide moiety
at Cꢀ5' in 48 was then installed using our new protocol described in the previous section through
azide reduction of 47 and in situ sulfamoylation of the crude aminonucleoside. The standard
salicylation and TFA mediated deprotection sequence yielded the final analog 10.
a
Scheme 4 . Synthesis of 10.
a
Reaction conditions: (a) TBSCl, imidazole, DMF, 23 °C, 7 h, 90%; (b) Oꢀphenyl
chlorothionoformate, DMAP, MeCN, 23 °C, 1 h, 82%; (c) Bu SnH, AIBN, toluene, 85 °C, 3 h,79%; (d)
3
TFA, MeOH, 0 °C, 12 h, 92%; (e) i) DPPA, DBU, dioxane, 23 °C, 12 h; ii) NaN , 15ꢀcrownꢀ5, dioxane,
3
9
0 °C, 12 h, 66%; (f) Pd/C, MeOH, 23 °C, 1 h; (g) NH SO NH , dioxane, 90 °C, 3 h, 60% (2 steps); (h)
2 2 2
3
1, Cs CO , DMF, 0 °C to rt, 16 h; (i) 80% aq TFA, 0 °C, 1 h, 34% (2 steps from 48)
2 3
12
ACS Paragon Plus Environment

Page 13 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Enzyme Inhibition and Antitubercular Activity. The new analogs 5–10, excluding the
prodrugs 2 and 3, were evaluated for enzyme inhibition against recombinant MbtA under initial
32
velocity conditions as previously described (see Experimental Section).
The apparent
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
inhibition constants (appK ) were determined by fitting the concentration–response plots to the
i
Morrison equation (eq 1, see Experimental Section) since all compounds exhibited tightꢀbinding
behavior (i.e. appK ≤10 × [MbtA]). The appK values for all analogs were potent, ranging from
i
i
0.78–8.9 nM (Table 1). These results are largely in agreement with prior findings and indicate
that our targeted modifications to 1 did not adversely impact potency.
Next, we evaluated the new analogs against wholeꢀcell M. tuberculosis H37Rv under ironꢀ
2
8
deficient conditions as previously described. The minimum inhibitor concentrations (MIC) that
resulted in complete inhibition of observable growth are shown in Table 1. Notably, all new
analogs 5–10 retained fairly potent antitubercular activity with MIC values ranging from 0.098–
6
.25 µM. Good correlation between appK and MIC values was observed for 1, 4, 6 and 8 with
i
the MIC/appK ratio (both units in µM) varying only slightly with the median ratio of 72 ± 16 for
i
these four compounds. However, the relative wholeꢀcell activity for 3′ꢀdeoxyꢀ2′ꢀfluoro 10, pꢀ
aminosalicyl 5, and 2′ꢀfluoroꢀ2ꢀphenyl 9 was eroded, with the MIC increasing to 0.78–3.125 µM
despite potent enzyme inhibition. This is reflected in the MIC/appK ratios of 260, 351 and 975,
i
6
respectively. The most deleterious modification was found in the diꢀfunctionalized N ꢀ
cyclopropylꢀ2ꢀphenyl derivative 7, whose MIC of 6.25 µM is nearly 10,000ꢀfold greater than the
appKi.
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 59
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 1. Enzyme inhibition and antimycobacterial activity of 1 analogs.
appK
MIC, µM
(µg/mL)
c
CC50d
(µM)
i,
Compound
a
b
MIC/appKi
(nM)
1
4
6.6 ± 1.5 0.39 (0.20)
3.7 ± 0.6 0.19 (0.11)
8.9 ± 0.7 3.125 (1.81)
1.5 ± 0.2 0.098 (0.06)
0.76 ± 0.40 6.25 (4.27)
1.7 ± 0.1 0.19 (0.11)
59
51
>200
>200
>200
–
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
351
65
6
7
8224
112
975
260
–
8
>200
>200
–
9
1.6 ± 0.1
1.56 (1.0)
10
3.0 ± 0.2 0.78 (0.43)
a
b
Assay performed with 7 nM MbtA, 10 mM ATP, 250 µM salicylic acid, 1 mM PPi. Grown in glycerolꢀ
c
d
alanine salts (GAS) medium without ferric ammonium citrate at pH 6.6. both in units of ꢁM. Cell
cytotoxicity of HepG2 cells resulting in loss of 50% viability.
Pharmacokinetic Analysis. All compounds were evaluated in a crossover experiment
employing doubly cannulated female SpragueꢀDawley rats to determine pharmacokinetic
parameters. While mice are normally used for these studies, we elected to use rats based on the
ability to serially sample a single animal and perform oral and intravenous crossover studies,
which greatly minimized interꢀanimal variability. Each compound was orally dosed (p.o.) at 25
mg/kg in three rats, and sampling was conducted at the indicated time points (see Experimental
Section). After a threeꢀday washout period, each rat was then dosed intravenously (i.v.) at 2.5
mg/kg as a single bolus with more intensive sampling at early time points (see Experimental
Section). Pharmacokinetic parameters shown in Table 2 were calculated from concentrationꢀ
time profiles by nonꢀcompartmental analysis. From these studies, the area under the
concentrationꢀtime curve (AUC) for each dose and absolute oral bioavailability (F) were
determined. Additional relevant parameters that will be discussed are the maximum serum
concentration (Cmax) for the p.o. dose and the terminal elimination halfꢀlife (t ) and clearance
1/2
(CL) for the i.v. dose. We began our studies by administering a known antitubercular agent,
14
ACS Paragon Plus Environment

Page 15 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
isoniazid (INH). The obtained PK parameters were consistent with several published studies of
5
2, 53
INH oral dosing and detection in rat plasma.
INH was rapidly eliminated with a t1/2 of 19
min and achieved a Cmax of 13.5 µg/mL, which translated to a 65% oral bioavailability. Our
minimal target PK candidate profile would provide an oral bioavailability of 20%, Cmax at least
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
0ꢀfold greater than the MIC , and a t1/2 greater than 1 hour.
PK results of 1 and prodrugs. Following an oral dose, 1 reached a maximum serum
concentration (Cmax) of 0.4 µg/mL at 60 minutes with an AUC value of 66 min·µg/mL (Table 2).
To determine the absolute bioavailability (F), 1 was then dosed i.v. to provide an AUC of 519
min·µg/mL that yielded an oral bioavailability of 1.3%. The terminal elimination halfꢀlife and
clearance were 11 min and 4.9 mL/min·kg, respectively, indicating rapid elimination. While
these results were discouraging, we note the maximum serum concentration (Cmax) achieved is
nearly two times the MIC value. The PK parameters obtained herein are nearly identical to those
obtained in BALB/c mice (Cmax = 1.2 µg/mL and AUC = 58.6 min·µg/mL) validating the
SpragueꢀDawley rat PK model as a practical surrogate for the murine PK model.
Microsomal stability studies with mouse, rat, and human microsomes demonstrated 1 is
extremely stable to oxidative metabolism (t1/2 > 30 min) and is not glucuronidated (0%
glucuronide detected at 60 min). The intraꢀmolecular hydrogenꢀbonding of the phenol with the
sulfamate nitrogen could account for the lack of observed glucuronidation. Moreover, 1 is
highly water soluble (>50 mg/mL) and stabile under both acidic (pH 1) and alkaline (pH 12)
conditions showing less than 5% degradation after 24 h at 37 °C. Collectively, these results
suggest renal clearance of this highly polar nucleoside as the likely elimination mechanism. We
also assessed the membrane permeability using the Cacoꢀ2 monolayer model to provide a
benchmark for other analogs. These studies indicate moderate permeability with a flux (Papp(A to
ꢀ6
B)) of 1.2 × 10 cm/s, and a favorable efflux ratio of 0.78 (Table 3).
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The acetate prodrugs 2 and 3 were dosed orally, but neither of the parent compounds was
detected in the serum samples and only a negligible amount of 1 could be detected with 3.
Serum stability studies of triꢀacetate 3 (see Figure S8) indicated the phenol acetate was rapidly
cleaved to liberate diꢀacetate 2, which was slowly hydrolyzed to 1. Cacoꢀ2 studies with 2 showed
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
ꢀ7
a reduced transport rate (Papp(A to B)) of 6.2 × 10 cm/s, which is approximately 50% of 1, as well
as an efflux ratio of 1.55 that is twice that of 1 (Table 3).
Modulation of the Sulfamate pK . Our next attempt to enhance oral exposure involved
a
modulation of the pK of the charged linker in 1, whose estimated pK is ~3 based on the
a
a
2
8
experimentally determined value for a related analogue.
This acidic functional group is
expected to be ionized under physiological conditions, which may impede membrane transport,
thus analogs with pK values closer to 7 would be desirable. Replacement of the sulfamate
a
linkage with the sulfamide linkage in 4 is predicted to raise the pK to ~5 while installation of the
a
28
electronꢀdonating paraꢀamino group in 5 is expected to further raise the pK to ~6. Sulfamide
a
4
achieved a Cmax of 0.7 µg/mL or 1.4 µM, which is 7ꢀfold higher than the MIC. The oral
exposure (AUC ) was increased 50% to 100 min·µg/mL that led to a nearly 3ꢀfold improved
p.o.
bioavailability of 3.5%. Cacoꢀ2 studies of 4 did not indicate enhanced permeability since the
Papp(A to B) relative to 1 was unchanged, while the efflux ratio was nearly identical (Table 3).
Based on this modest improvement all subsequent analogs were prepared with the sulfamide
linkage.
The paraꢀamino analog 5 was then evaluated. Unfortunately, this compound had
significantly lower oral exposure with an AUCp.o. of only 17 min·µg/mL, and reached a Cmax of
0.09 µg/mL, which is far below the MIC value. The i.v. exposure was also diminished, and can
be attributed to the substantially faster clearance of 46.4 mL/min·kg, which is nearly 10ꢀfold
16
ACS Paragon Plus Environment

Page 17 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
greater than 1, and the highest value of any analog in this series. The enhanced clearance of 5 is
likely caused by the increased polarity of this molecule due to introduction of the paraꢀamino
group.
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
Impact of Lipophilicity. To evaluate the impact of lipophilicity on absorption, we evaluated
6
6
N ꢀcyclopropyl 6 and N ꢀcyclopropylꢀ2ꢀphenyl 7. These modifications increased the clogP
values from ꢀ0.89 for compound 1 to 0.5 and 2.9, respectively. The Cmax for both analogs ranged
from 0.1 to 0.14 µg/mL while the AUCp.o. values ranged from 4 to 19 min·µg/mL, indicating low
oral exposure. The halfꢀlife of both compounds marginally improved to 15 min. The oral
bioavailability of 6 was unchanged relative to 1 while the oral bioavailability of 7 was a dismal
0
.3%. The increased lipophilicity of 7 manifested as poor aqueous solubility and several
excipients (carboxymethylcellulose, 25% hydroxypropylꢀβꢀcyclodextrin, and 50% PEG400–5%
tweenꢀ80) were evaluated to potentially improve oral absorption without success. These results
suggest that enhancing the lipophilicity through addition of nonpolar groups is detrimental to
improving PK. In retrospect, the high molecular weight (MW) of 1 leaves little room to
maneuver, and both 6 and 7 exceed Lipinski's threshold MW value of 500. The permeability of
7
as measured using Cacoꢀ2 cell studies corroborate the low bioavailability and indicate low
ꢀ7
permeability with a flux of 5.2 × 10 cm/s, about twoꢀfold less than either 1 or 4, and an efflux
ratio of 3.45.
Fluorinated Analogs. Fluorine is commonly used to improve drug properties and can
increase oral exposure through reduction of cytochrome P450ꢀmediated metabolic clearance.
Fluorination at the Cꢀ2' position of nucleosides is also known to enhance anomeric stability,
which can lead to an improved PK profile if degradation by cleavage of the anomeric linkage
5
4
occurs at acidic pH. Introduction of a fluorine at the Cꢀ2' position in 8 led to a nearly 3ꢀfold
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 59
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
increase in oral exposure relative to 1, with an AUCp.o. of 188 min·µg/mL that is concordant with
a 3ꢀfold increase in halfꢀlife to 32.5 min (relative to 11 min for 1). Since the AUCi.v. is also
dramatically increased to 1043 min·µg/mL due to the increased halfꢀlife, the oral bioavailability
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
remains low at 2%. Cacoꢀ2 studies of 8 demonstrated enhanced permeability with a flux of 4.2 ×
ꢀ6
1
0 cm/s, or nearly 4ꢀfold greater than 1 and 4, and a favorable efflux ratio of 0.32 (Table 3).
Unfortunately, this did not translate to improved bioavailability, as noted above, and suggests a
limitation of the Cacoꢀ2 in vitro model for predicting oral bioavailability for this compound
class. The 2′ꢀfluoroꢀ2ꢀphenyl 9 showed a dramatic improvement in halfꢀlife to 121 min, which is
11ꢀfold greater than 1, and resulted in an astonishing i.v. exposure of 1981 min·µg/mL.
Unfortunately, the oral exposure (AUC ) was diminished relative to 8, but still 2ꢀfold greater
p.o.
than 1. The oral bioavailability therefore remained low at 0.8%. 2′,3′ꢀDideoxyꢀ2′ꢀfluoro 10 was
the last compound evaluated with the anticipation that removal of the polar 3′ꢀhydroxyl group
may enhance membrane permeability. Compound 10 indeed realized the highest oral exposure
with an AUCp.o. of 198 min·µg/mL and a Cmax of 0.56 µg/mL. Additionally, the halfꢀlife was
improved to 62 min, but the oral bioavailability remained unchanged relative to 8. Overall, these
results demonstrate that replacement of the 2′ꢀhydroxyl group with a fluorine atom results in a
dramatically enhanced halfꢀlife, which translates to improved oral exposure.
18
ACS Paragon Plus Environment

Page 19 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Pharmacokinetic parameters of 1 analogs.
a
b
i.v. PK parameters
p.o. PK parameters
Compound
AUC0ꢀinf
CL
t1/2
(min)
Cmax
(µg·mL )
AUC0ꢀinf
(µg·min·mL )
F
(%)
–
1
–1
–1
–1
–1
(min·µg·mL )
(mL·min ·kg )
1
2
3
4
5
6
7
8
9
519 ± 115
nd
4.9 ± 0.9
nd
11 ± 2
0.4 ± 0.2
66 ± 41
1.3 ± 0.7
na
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
nd
<LD
<LD
nd
nd
nd
0.03 ± 0.01
0.70 ± 0.07
0.09 ± 0.01
0.14 ± 0.01
0.10 ± 0.01
0.46 ± 0.30
0.50 ± 0.16
0.56 ± 0.12
2.9 ± 0.8
99.8 ± 13.6
16.8 ± 4.2
18.1 ± 2.1
4.0 ± 0.9
188 ± 110
140 ± 46
198 ± 49
na
304 ± 60
63 ± 47
209 ± 76
184 ± 30
1043 ± 326
1981 ± 200
959 ± 135
8.4 ± 1.9
46.3 ± 4.5
9.6 ± 0.1
13.6 ± 2
2.5 ± 0.9
1.3 ± 0.1
2.6 ± 0.5
10.6 ± 0.7
10.6 ± 2.1
15.1 ± 1.3
15.5 ± 10.6
32.5 ± 7.4
121 ± 18
62 ± 6.4
3.5 ± 0.2
3.7 ± 0.3
1.0 ± 0.4
0.3 ± 0.1
1.8 ± 0.9
0.80 ± 0.15
2.0 ± 0.08
1
0
a
b
i.v. dose (D ) = 2.5 mg/kg. p.o. dose (Dp.o.) = 25 mg/kg. Values are reported as the mean ± SD. AUC0ꢀinf: area under the
i.v.
plasma concentrationꢀtime curve from time 0 to infinity, CL: clearance, t1/2: terminal elimination halfꢀlife, Cmax: maximum
plasma concentration, F: relative oral bioavailability calculated as follows F = 100 × [(AUCp.o. × D )/(AUC . × Dp.o.)], nd: not
determined, na: not applicable, LD: below the limit of detection.
i.v.
i.v
Table 3. Cacoꢀ2 permeability studies of select compounds.
Papp, A to B
Papp, B to A
10 cm/s
Compound
ꢀ6
ꢀ6
efflux ratio
1
0 cm/s
1
2
4
7
8
1.23
0.96
0.97
0.86
1.8
0.78
1.55
0.69
3.45
0.32
0.62
1.23
0.52
4.20
1.34
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 59
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. Mean plasma concentration vs. time curves of single dosage of compound 10 in 3 rats
by both p.o. and i.v. administration. Error bars represent the standard deviation of the mean
(n=3).
Conclusion
Four complimentary strategies were examined to improve the drug disposition and
pharmacokinetic properties of a new class of antibiotics that inhibit siderophore biosynthesis in
M. tuberculosis. Our first approach involved preparation of simple diꢀ and triꢀacetate prodrugs
(2 and 3) to mask the polar hydroxyl groups of the prototypical inhibitor 1. Unfortunately,
neither prodrug was orally bioavailable. Plasma stability studies suggested rapid cleavage of 3
to 2 that underwent slow cleavage to afford 1. Alternate proꢀmoieties with faster cleavage
kinetics could be explored to mask the 2′ and 3′ hydroxyls. The complete lack of oral exposure
with 2 and 3 suggested this would not be a viable strategy and further studies along this line were
not pursued.
We hypothesized that the ionized acylꢀsulfonyl linker of the adenylation inhibitors described
herein may impede membrane permeability. The negative charge is required for potent binding
due to critical electrostatic interactions in the active site of MbtA and any attempts to remove this
20
ACS Paragon Plus Environment

Page 21 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
8
charge have resulted in complete loss of activity. Thus, we designed analogs to increase the
pK of the acylꢀsulfonyl nitrogen linker. The sulfamide analog 4 possessed 3ꢀfold improved
a
bioavailability relative to 1, as well as 50% increased oral exposure and a more than 3ꢀfold
higher Cmax/MIC ratio. Incorporation of a paraꢀamino group on the salicyl moiety in 5 led to a
significant loss in oral exposure. This is attributed to dramatically enhanced clearance caused by
the additional polar amino group. Given the modest improvements realized with this approach,
coupled with the strict requirements for an acylꢀsulfonyl linker moiety from our prior extensive
SAR studies, further modifications to the salicyl moiety were not pursued. Moreover, we later
recognized these salicylꢀnucleoside analogs can engage in intramolecular hydrogen bonding
between the sulfonyl nitrogen and the orthoꢀphenol of the salicyl moiety. This could partially
shield the polarity through delocalization of the charge and mitigate our initial concerns about
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
5
the ionized acylꢀsulfonyl linker.
The third approach to enhance ADME and PK properties involved modulation of the
lipophilicity. While it is recognized that many antibacterials have unique physicochemical
properties relative to other drug classes, we nevertheless sought to increase the lipophilicity of
5
6
the highly polar nucleoside derivative 1.
This was achieved through addition of nonꢀpolar
groups onto the nucleoside at sites previously shown to enhance biochemical potency and
3
4
6
6
antibacterial activity. The calculated logP of N ꢀcyclopropyl 6 and N ꢀcyclopropylꢀ2ꢀphenyl 7
was increased to +0.5 and +2.9, respectively. However, both compounds realized dramatically
lower oral exposures and failed to even reach a Cmax exceeding their MIC values. A more
fruitful approach to increasing lipophilicity may be through strategic deletion of polar functional
groups and atoms that are dispensable for activity, as this would additionally result in a reduction
of the molecular weight.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 59
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
Incorporation of a fluorine atom at Cꢀ2′ of the nucleoside in analogs 8–10 led to the most
pronounced improvements in oral exposure, primarily through an increase in the terminal
elimination halfꢀlife (t ) following i.v. administration from 11 minutes for 1 to 32–121 minutes
1
/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
for 8ꢀ10. The apparent terminal elimination halfꢀlives following oral dosing for all of the
compounds described herein are much longer, presumably due to slow absorption, and do not
reflect the rate of elimination. To calibrate these values, the intrinsic halfꢀlife of the firstꢀline
antitubercular agent isoniazid in a rat is only 18 minutes. Fluorine is commonly introduced to
45
enhance t1/2 through blocking P450ꢀmediated metabolism. In the case of nucleosides, a Cꢀ2′
fluoro group improves the stability of the anomeric linkage, which can be important for some
54
nucleosides that degrade by anomeric cleavage under the acidic conditions of gastric fluids. We
have shown 1 and its derivatives to be extremely stable under strongly acidic conditions, while
microsomal stability studies revealed little metabolism of 1. Thus, we hypothesize that the Cꢀ2′
fluoro group likely decreases renal clearance.
In order to provide an in vitro marker for oral bioavailability, we also measured select
analogs using the standard Cacoꢀ2 model system for intestinal epithelial permeability. These
experiments showed the Cacoꢀ2 model may be qualitatively useful, but is not quantitatively
ꢀ6
predictive of oral bioavailability. We identified a minimum threshold Papp(A to B) of 1 × 10 cm/s
that can now be used to eliminate candidates prior to PK evaluation.
Through logical modification of 1 we explored multiple strategies to enhance
pharmacokinetic parameters, and successfully increased the intrinsic halfꢀlife and oral exposure.
While were unable to reach our minimal target PK candidate profile, we believe these studies
have provided a foundation to more effectively address the PK limitations of 1. Our findings
22
ACS Paragon Plus Environment

Page 23 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
may also be applicable to other adenylation inhibitors, which have been described for dozens of
5
7, 58
other functionally related adenylating enzymes involved in diverse disease processes.
EXPERIMENTALS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General materials and methods. Chemicals and solvents were purchased from Acros
Organics, Alfa Aesar, SigmaꢀAldrich, and TCI America and were used as received. The
nucleosides 2′ꢀdeoxyꢀ2′ꢀfluoroguanosine 33 and 2′ꢀdeoxyꢀ2′ꢀfluoroadenosine 42 were obtained
2
8
39
39
34
35
59
59
from Metkinen Chemistry (Kuopio, Finland). Compounds 1, 11, 14, 19, 23, 26, 27,
40
44
32
43
28
2
9, 30, 31, 32, and Nꢀhydroxysuccinimidyl 2ꢀbenzyloxybenzoate were prepared as
described. All final compounds (1–10) were purified by preparative reverseꢀphase HPLC using
the indicated method and purity (≥95%) of the lyophilized powders was confirmed by analytical
reverseꢀphase HPLC. An anhydrous solvent dispensing system using two packed columns of
neutral alumina was used for drying THF and CH Cl , while two packed columns of molecular
2
2
sieves were used to dry DMF, and the solvents were dispensed under argon gas (Ar). Anhydrous
grade MeOH, MeCN, pyridine, and DMA were purchased from Aldrich. EtOAc and hexanes
were purchased from Fisher Scientific. All reactions were performed under an inert atmosphere
of dry argon gas (Ar) in ovenꢀdried (180 °C) glassware. TLC analyses were performed on TLC
silica gel 60F254 plates from EMD Chemical Inc. and were visualized with UV light.
Purification by flash chromatography was performed using a mediumꢀpressure flash
chromatography system equipped with flash column silica cartridges with the indicated solvent
system. Reversedꢀphase HPLC (RPꢀHPLC) purification was performed on Phenomenex Gemini
10 ꢁm C18 columns (250 × 21.2 or 10 or 4.6 mm) with detection at 254 nm employing the
1
13
indicated solvent system. H and C spectra were recorded on a 400 or 600 MHz Varian, or 700
MHz Bruker NMR spectrometer. Proton chemical shifts are reported in ppm from an internal
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 59
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
standard of residual chloroform (7.26), methanol (3.31), dimethyl sulfoxide (2.50), or monoꢀ
deuterated water (HDO, 4.79); carbon chemical shifts are reported in ppm from an internal
standard of residual chloroform (77.0), methanol (49.1), or dimethyl sulfoxide (39.5). Proton
chemical data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, dt =
doublet of triplets, t = triplet, q = quartet, pentet = pent, m = multiplet, ap = apparent, br = broad,
ovlp = overlapping), coupling constant(s), integration. Highꢀresolution mass spectra were
obtained on an LTQ Orbitrap Velos (Thermo Scientific, Waltham, MA).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
General Procedure for Mitsunobu Reaction. To a solution of nucleoside (1.0 equiv), 29
(
3.0 equiv) and PPh (1.05 equiv) in THF (0.05 M for limiting reagent) at 0 °C was added neat
3
DIAD (1.05 equiv) dropwise over 15 min. The solution was stirred at 0 °C for 45 min, then
gradually warmed to 23 °C over 3 h and stirred for 12 h at 23 °C. The reaction was concentrated
under reduced pressure to a dark yellow oil and immediately purified by flash chromatography
(linear gradient 0–100% EtOAc–hexanes) to afford the title compound.
General Procedure for Salicylation. To a solution the sulfamide nucleoside (1.0 equiv) and
Cs CO (2.5 equiv) in DMF (0.5 M for limiting reagent) at 0 °C was added 31 (1.5 equiv). The
2
3
reaction mixture was then stirred for 24 h at 23 °C during which time all starting material was
consumed as monitored by electrospray mass spectrometry in the negative mode. DMF was
removed by rotary evaporation under high vacuum (P = 0.01 torr) to afford crude salicylated
nucleoside. The crude product was used in the next global deprotection reaction directly without
further purification.
General Procedure for TFA Global Deprotection. To a solution of the crude salicylated
nucleoside in H O (10 mL/mmol) at 0 °C was added TFA (40 mL/mmol) dropwise (final
2
concentration 0.02 M). The reaction mixture was stirred for 8 h at 23 °C during which time all
starting material was consumed as monitored by electrospray mass spectrometry in the negative
24
ACS Paragon Plus Environment

Page 25 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mode. The solvent was removed under vacuum. The crude material was reꢀdissolved in 1:1
MeCN–50 mM TEAB (10–20 mg/mL) and filtered to remove insoluble solids. The resulting
solution was purified by reverse phase HPLC employing the indicated method and column using
50 mM aqueous triethylammonium bicarbonate (TEAB) at pH 7.5 (solvent A) and acetonitrile
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
(
solvent B). The appropriate fractions were pooled and lyophilized to afford the final compound
as the triethylammonium salt as a white foam.
′ꢀOꢀ[Nꢀ(2ꢀBenzyloxybenzoyl)sulfamoyl]ꢀN ,N ꢀbis(tertꢀbutyloxycarbonyl)ꢀ2′,3′ꢀOꢀ
6
6
5
isopropylideneadenosine triethylammonium salt (12). To a stirring solution of 11 (0.54 g,
0.92 mmol, 1.0 equiv) in DMF (9.2 mL) cooled to 0 °C was added Nꢀhydroxysuccinimidyl 2ꢀ
benzyloxybenzoate (0.45 g, 1.38 mmol, 1.5 equiv) followed by Cs CO (0.60 g, 1.84 mmol, 2.0
2
3
equiv) and the mixture was stirred at 23 °C for 21 h. The crude reaction mixture was
concentrated under reduced pressure to a yellow oil that was taken up in EtOAc (200 mL) and
filtered. The precipitate was washed with additional EtOAc (200 mL), and the combined filtrate
was concentrated under reduced pressure and purified by flash chromatography (linear gradient
0–5% MeOH/EtOAc with 1% Et N) to afford the title compound (0.51 g, 62%) as a white foamy
3
1
oil: R = 0.37 (1:9 MeOH–EtOAc with 1% Et N); H NMR (600 MHz, CD OD) δ 1.13 (t, J = 7.2
f
3
3
Hz, 9H), 1.35 (s, 3H), 1.40 (s, 18H), 1.61 (s, 3H), 2.83 (q, J = 7.2 Hz, 6H), 4.19–4.20 (m, 2H),
.41 (dd, J = 6.0, 3.6 Hz, 1H), 5.12–5.13 (m, 3H), 5.38 (dd, J = 6.0, 3.6 Hz, 1H), 6.37 (d, J = 3.0
Hz, 1H), 6.95 (t, J = 7.8 Hz, 1H), 7.05 (d, J = 8.4 Hz, 1H), 7.25 (t, J = 7.8 Hz, 1H), 7.31–7.34
4
1
3
(m, 3H), 7.46 (d, J = 7.8 Hz, 1H), 7.52 (d, J = 7.2 Hz, 2H), 8.84 (s, 1H), 8.89 (s, 1H); C NMR
(150 MHz, CD OD) δ 10.2, 25.7, 27.8, 28.2, 47.4, 69.8, 71.6, 83.1, 85.4, 86.0, 86.2, 92.5, 115.3,
3
1
21.3, 121.9, 128.9, 129.2, 129.3, 129.6, 129.9, 130.3, 131.7, 131.9, 138.7, 146.8, 151.2, 151.7,
–
153.3, 154.6, 157.2, 176.9; HRMS (ESI–) calcd for C H N O S [M – H] 795.2665, found
3
7
43
6
12
7
95.2659 (0.8 ppm error).
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 59
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
2′,3′ꢀOꢀdiꢀAcetylꢀ5′ꢀOꢀ[Nꢀ(2ꢀhydroxybenzoyl)sulfamoyl]adenosine (2). To
a
flask
containing 12 (0.51 g, 0.56 mmol, 1.0 equiv) at 0 °C was added 80% aqueous TFA (6 mL) that
had also been cooled to 0 °C and the mixture stirred at 0 °C for 1 h then at 23 °C for 3 h. The
reaction mixture was concentrated under reduced pressure to a yellow oil and dried under high
vacuum to remove all traces of TFA, then dissolved in DMF (6 mL) to which was added pyridine
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
(
0.17 mL, 2.1 mmol, 3.8 equiv) followed by acetic anhydride (0.33 mL, 3.5 mmol, 6.4 equiv)
and the mixture stirred at 23 °C for 16 h. The reaction was concentrated under reduced pressure
to remove all traces of DMF and give 13 as a crude yellow oil that was used directly in the next
step without purification.
To a roundꢀbottomed flask containing the crude product 13 prepared above (0.25 g, 0.33
mmol, 1.0 equiv) purged with argon was added anhydrous MeOH (3.4 mL) followed by Pd/C
(0.030 g, 10% by weight) and the flask evacuated by house vacuum and backꢀfilled with an H2
balloon 3 times before being allowed to stir vigorously under an H balloon at 23 °C for 2 h. The
2
crude reaction mixture was filtered over a pad of Celite and concentrated under reduced pressure
to a dark oil. Purification by semiꢀpreparative reverseꢀphase HPLC on a Phenomenex Gemini 10
ꢁ
m C18 (250 × 10.0 mm) column at a flow rate of 5 mL/min with isocratic elution of 35%
MeCN–50 mM aqueous triethylammonium bicarbonate for 10 min, followed by 70% MeCN for
min. The retention time of the product was 5.4 minutes (k′ = 1.2) and the appropriate fractions
5
were pooled and lyophilized to afford the title compound (90.0 mg, 42%) as a white solid: R =
f
1
0.18 (1:9 MeOH–CH Cl with 1% Et N); H NMR (600 MHz, CD OD) δ 1.23 (t, J = 7.2 Hz,
2
2
3
3
9
H), 2.00 (s, 3H), 2.10 (s, 3H), 3.03 (q, J = 7.2 Hz, 6H), 4.44 (dd, J = 12.0, 3.0 Hz, 1H), 4.47
(
dd, J = 11.4, 3.6 Hz, 1H), 4.53 (q, J = 3.6 Hz, 1H), 5.70 (dd, J = 5.4, 3.6 Hz, 1H), 5.89 (t, J =
6.0 Hz, 1H), 6.27 (d, J = 6.6 Hz, 1H), 6.75–6.79 (m, 2H), 7.28 (t, J = 7.2 Hz, 1H), 7.93 (d, J =
1
7
.8 Hz, 1H), 8.18 (s, 1H), 8.56 (s, 1H); H NMR (600 MHz, DMSOꢀd ) δ 1.17 (t, J = 7.2 Hz,
6
26
ACS Paragon Plus Environment

Page 27 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9
H), 1.99 (s, 3H), 2.10 (s, 3H), 3.09 (q, J = 7.2 Hz, 6H), 4.25 (dd, J = 11.4, 4.8 Hz, 1H), 4.33
(dd, J = 11.4, 4.2 Hz, 1H), 4.44 (q, J = 4.2 Hz, 1H), 5.56 (dd, J = 5.4, 4.2 Hz, 1H), 5.90 (t, J =
6.0 Hz, 1H), 6.20 (d, J = 6.0 Hz, 1H), 6.72–6.75 (m, 2H), 7.26 (t, J = 7.2 Hz, 1H), 7.36 (br s,
1
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
2H), 7.81 (d, J = 7.8 Hz, 1H), 8.14 (s, 1H), 8.46 (s, 1H), 13.5 (s, 1H); C NMR (150 MHz,
CD OD) δ 9.9, 20.3, 20.6, 47.8, 69.1, 73.0, 75.3, 82.7, 87.1, 118.0, 119.4, 120.2, 120.7, 131.6,
3
134.5, 141.3, 150.8, 154.2, 157.5, 162.3, 171.0, 171.5, 175.3; HRMS (ESI–) calcd for
–
C H N O S [M – H] 549.1045, found 549.1048 (0.5 ppm error).
21
21
6
10
5
′ꢀOꢀ[Nꢀ(2ꢀAcetoxybenzoyl)sulfamoyl]ꢀ2′,3′ꢀOꢀdiꢀacetyladenosine (3). To a stirring
solution of 1 (0.10 g, 0.18 mmol, 1.0 equiv) in DMF was added pyridine (0.05 mL, 0.68 mmol,
.8 equiv) followed by acetic anhydride (0.11 mL, 1.13 mmol, 6.4 equiv) and the mixture stirred
3
at 23 °C for 21 h. The crude reaction mixture was concentrated under reduced pressure to a
yellow oil and purified by flash chromatography (linear gradient 0–5% MeOH/DCM with 1%
Et N) to afford an approximately 5:1 mixture of 3 and 2. The desired compound was purified by
3
semiꢀpreparative reverseꢀphase HPLC on a Phenomenex Gemini 10 ꢁm C18 (250 × 10.0 mm)
column at a flow rate of 5 mL/min with a linear gradient 20–70% MeCN–50 mM aqueous
triethylammonium bicarbonate for 10 min, followed by 70% MeCN for 5 min. The retention
time of the product was 6.6 minutes (k′ = 1.6) and the appropriate fractions were pooled and
lyophilized to afford the title compound (57.8 mg, 46%) as a white solid: R = 0.25 (1:9 MeOH–
f
1
CH Cl with 1% Et N); H NMR (600 MHz, CD OD) δ 1.27 (q, J = 7.2 Hz, 9H), 2.00 (s, 3H),
2
2
3
3
2.13 (s, 3H), 2.26 (s, 3H), 3.17 (t, J = 6.0 Hz, 6H), 4.40 (dd, J = 14.4, 3.6 Hz, 1H), 4.44 (dd, J =
.6, 3.0 Hz, 1H), 4.54 (d, J = 3.0 Hz, 1H), 5.70–5.71 (m, 1H), 5.89 (t, J = 6.0 Hz, 1H), 6.29 (d, J
6.0 Hz, 1H), 7.02 (d, J = 7.8 Hz, 1H), 7.06 (s, 1H), 7.23 (t, J = 7.8 Hz, 1H), 7.41 (t, J = 7.2 Hz,
9
=
1
3
1H), 7.86 (d, J = 7.8 Hz, 1H), 8.19 (s, 1H), 8.59 (s, 1H); C NMR (150 MHz, CD OD) δ 9.4,
3
20.3, 20.6, 21.4, 48.0, 69.1, 73.2, 75.3, 82.9, 87.0, 120.2, 124.1, 126.7, 131.9, 132.2, 133.1,
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
41.3, 150.8, 150.9, 154.2, 157.2, 171.0, 171.5, 172.0, 173.8; HRMS (ESI–) calcd for
–
C H N O S [M – H] 591.1151, found 591.1170 (3.2 ppm error).
23
23
6
11
6
6
N ,N ꢀbis(tertꢀButoxycarbonyl)ꢀ5′ꢀNꢀ[Nꢀ(tertꢀbutoxycarbonyl)ꢀNꢀsulfamoyl]aminoꢀ5′ꢀ
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
deoxyꢀ2′,3′ꢀOꢀisopropylideneadenosine (15). The title compound was prepared from 14 (0.60
g, 1.2 mmol) using the general procedure for the Mitsunobu reaction and isolated as a foamy oil
1
(0.789 g, 96%): R = 0.25 (1:1 EtOAc–hexanes); H NMR (600 MHz, CDCl ) δ 1.40 (s, 21H),
f
3
1.46 (s, 9H), 1.49 (s, 3H), 1.61 (s, 3H), 3.94 (dd, J = 15.0, 7.8 Hz, 1H), 3.99 (dd, J = 15.6, 5.4
Hz, 1H), 4.51 (p, J = 4.8 Hz, 1H), 4.90 (p, J = 6.0 Hz, 1H), 5.24 (dd, J = 6.0, 3.6 Hz, 1H), 5.57
1
3
(dd, J = 6.0, 1.8 Hz, 1H), 6.38 (d, 1H, 1.8 Hz), 8.64 (s, 1H), 8.88 (s, 1H); C NMR (150 MHz,
CDCl ) δ 27.6, 28.2, 28.4, 28.5, 49.9, 83.6, 85.1, 85.4, 85.6, 87.5, 91.6, 115.6, 130.7, 147.6,
3
–
1
51.1, 151.5, 153.1, 153.2, 154.2; HRMS (ESI–) calcd for C H N O S [M – H] 684.2668,
28 42
7
11
found 684.2685 (2.5 ppm error).
6
6
N ,N ꢀbis(tertꢀButoxycarbonyl)ꢀ5′ꢀNꢀ(NꢀtertꢀbutoxycarbonylꢀNꢀ{[(2ꢀOꢀ
methoxymethoxy)benzoyl]sulfamoyl})aminoꢀ5′ꢀdeoxyꢀ2′,3′ꢀOꢀisopropylideneadenosine
triethylammonium salt (16). The title compound was prepared from 15 (0.52 g, 0.75 mmol, 1.0
equiv) using the general procedure for salicylation and isolated as an offꢀwhite amorphous solid
1
(0.38 g, 53%): R = 0.37 (1:9 MeOH–CH Cl with 1% Et N); H NMR (600 MHz, CDCl ) δ 1.26
f
2
2
3
3
(t, J = 7.2 Hz, 9H), 1.37 (s, 3H), 1.39 (s, 18H), 1.45 (s, 9H), 1.62 (s, 3H), 3.11 (q, J = 7.2 Hz,
6H), 3.45 (s, 2H), 4.06 (dd, J = 15.0, 6.0 Hz, 1H), 4.64 (dd, J = 18.0, 7.2, Hz, 1H), 4.60–4.63 (m,
1
H), 5.14 (s, 2H), 5.28 (dd, J = 6.0, 3.0 Hz, 1H), 5.51 (dd, J = 6.6, 3.0 Hz, 1H), 6.32 (d, 2.4 Hz,
H), 6.96 (t, J = 7.2 Hz, 1H), 7.11 (d, J = 8.4 Hz, 1H), 7.27 (t, J = 7.8 Hz, 1H), 7.51 (d, J = 7.8
1
1
3
Hz, 1H), 8.74 (s, 1H), 8.89 (s, 1H); C NMR (150 MHz, CDCl ) δ 7.9, 27.7, 28.0, 46.1, 52.8,
3
5
6.3, 63.3, 81.4, 82.3, 83.7, 84.5, 85.0, 90.3, 96.5, 114.3, 118.0, 121.9, 129.0, 129.8, 131.3,
44.0, 150.0, 150.3, 151.6, 152.2, 152.5, 153.0, 154.5; HRMS (ESI–) calcd for C H N O S [M
1
37 50
7
14
28
ACS Paragon Plus Environment

Page 29 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
–
–
H] 848.3142, found 848.3146 (0.5 ppm error).
5
′ꢀN-[Nꢀ(2ꢀHydroxybenzoyl)sulfamoyl]aminoꢀ5′ꢀdeoxyadenosine
triethylammonium
salt (4). The title compound was prepared (150 mg, 65%) from 16 (380 mg, 0.39 mmol) 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
the general procedure for TFA deprotection and purified by preparativeꢀreverse phase HPLC as
1
13
described. The analytical data ( H NMR, C NMR, HRMS) agreed with previously reported
2
8
values for 4 obtained by an alternate synthetic route.
5
′ꢀNꢀ[Nꢀ(4ꢀBenzyloxycarbonylaminoꢀ2ꢀhydroxybenzoyl)ꢀNꢀ(tertꢀ
6
6
butoxycarbonyl)sulfamoyl]aminoꢀN ,N ꢀbis(tertꢀbutoxycarbonyl)ꢀ5ʹꢀdeoxyꢀ2′,3′ꢀOꢀ
isopropylideneadenosine triethylammonium salt (17). To a stirring solution of 32 (0.75 g, 2.6
mmol, 3.0 equiv) in DMF (4.4 mL) was added CDI (0.51 g, 3.2 mmol, 3.6 equiv) and the
mixture heated at 60 °C for 2 h. The reaction mixture was cooled to 23 °C, and 15 (0.60 g, 0.87
mmol, 1.0 equiv) was added as a solution in DMF (4.4 mL) followed by solid Cs CO (0.43 g,
2
3
1.31 mmol, 1.5 equiv), and stirred for 15 h. The crude reaction mixture was concentrated under
reduced pressure to a dark oil and purified by flash chromatography (linear gradient 0–5%
MeOH–CH Cl with 1% Et N) to afford the title compound (0.46 g, 50%) as a waxy light yellow
2
2
3
1
solid: R = 0.61 (1:9 MeOH–CH Cl with 1% Et N); H NMR (600 MHz, CD OD) δ 1.09 (t, J =
f
2
2
3
3
7
.2 Hz, 9H), 1.36 (s, 9H), 1.38 (s, 18H), 1.40 (s, 3H), 1.61 (s, 3H), 2.67 (q, J = 7.2 Hz, 6H), 4.05
dd, J = 15.0, 6.6 Hz, 1H), 4.18 (dd, J = 15.0, 7.2 Hz, 1H), 4.63 (td, J = 7.2, 3.0 Hz, 1H), 5.17 (s,
2H), 5.27 (dd, J = 5.4, 3.0 Hz, 1H), 5.57 (dd, J = 6.0, 3.0 Hz, 1H), 6.33 (d, J = 2.4 Hz, 1H), 6.82
(
(d, J = 9.0 Hz, 1H), 7.06 (s, 1H), 7.30 (d, J = 7.2 Hz, 1H), 7.36 (t, J = 7.8 Hz, 1H), 7.41 (d, J =
13
7
.8 Hz, 1H), 7.75 (d, J = 8.4 Hz, 1H), 8.74 (s, 1H), 8.89 (s, 1H); C NMR (150 MHz, CD OD) δ
3
10.9, 25.7, 27.7, 28.2, 28.5, 47.3, 53.8, 67.7, 83.8, 84.0, 85.4, 87.3, 92.1, 106.8, 109.9, 115.5,
16.0, 129.1, 129.3, 129.7, 130.7, 131.9, 134.6, 138.1, 145.0, 147.3, 151.2, 151.6, 153.3, 154.4,
1
29
ACS Paragon Plus Environment