Structure-Based Design of a Small Molecule CD4-Antagonist with Broad Spectrum Anti-HIV-1 Activity

Article abstract of DOI:10.1021/acs.jmedchem.5b00709

Earlier we reported the discovery and design of NBD-556 and their analogs which demonstrated their potential as HIV-1 entry inhibitors. However, progress in developing these inhibitors has been stymied by their CD4-agonist properties, an unfavorable trait for use as drug. Here, we demonstrate the successful conversion of a full CD4-agonist (NBD-556) through a partial CD4-agonist (NBD-09027), to a full CD4-antagonist (NBD-11021) by structure-based modification of the critical oxalamide midregion, previously thought to be intolerant of modification. NBD-11021 showed unprecedented neutralization breath for this class of inhibitors, with pan-neutralization against a panel of 56 Env-pseudotyped HIV-1 representing diverse subtypes of clinical isolates (IC₅₀ as low as 270 nM). The cocrystal structure of NBD-11021 complexed to a monomeric HIV-1 gp120 core revealed its detail binding characteristics. The study is expected to provide a framework for further development of NBD series as HIV-1 entry inhibitors for clinical application against AIDS.

Full text of DOI:10.1021/acs.jmedchem.5b00709

Subscriber access provided by RUTGERS UNIVERSITY
Article
Structure-based design of a small molecule CD4-
antagonist with broad spectrum anti-HIV-1 activity
Asim Kumar Debnath, Francesca Curreli, Young Do Kwon, Hongtao Zhang, Daniel Scacalossi, Dmitry
S Belov, Artur A Tikhonov, Ivan A Andreev, Andrea Altieri, Alexander V. Kurkin, and Peter D Kwong
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00709 • Publication Date (Web): 24 Aug 2015
Downloaded from http://pubs.acs.org on August 25, 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 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structureꢀbased design of a small molecule CD4ꢀantagonist with broad spectrum antiꢀ
HIVꢀ1 activity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a¶
b¶
a
a
c
Francesca Curreli , Young Do Kwon , Hongtao Zhang , Daniel Scacalossi , Dmitry S. Belov ,
c
c
c
c
b
Artur A. Tikhonov , Ivan A. Andreev , Andrea Altieri , Alexander V. Kurkin , Peter D. Kwong &
a*
Asim K. Debnath
a
b
c
Laboratory of Molecular Modeling and Drug Design, Lindsey F. Kimball Research Institute, New
York Blood Center, New York, NY 10065, USA.
Vaccine Research Center, National Institute of Allergy and Infectious Diseases, National
Institutes of Health, Bethesda, MD 20892, USA.
EDASA Scientific, Scientific Park, Moscow State University, Leninskie Gory, Bld.75, 77 –
01b; 119992 Moscow ꢀ Russia
1
¶ These authors contributed equally.
Address for communication: Dr. Asim K Debnath (adebnath@nybloodcenter.org)
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 71
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
Earlier we reported the discovery and design of NBDꢀ556 and their analogs which
demonstrated their potential as HIVꢀ1 entry inhibitors. However, progress in developing these
inhibitors has been stymied by their CD4ꢀagonist properties, an unfavorable trait for use as
drug. Here, we demonstrate the successful conversion of a full CD4ꢀagonist (NBDꢀ556) through
a partial CD4ꢀagonist (NBDꢀ09027), to a full CD4ꢀantagonist (NBDꢀ11021) by structureꢀbased
modification of the critical oxalamide midꢀregion, previously thought to be intolerant of
modification. NBDꢀ11021 showed unprecedented neutralization breath for this class of
inhibitors, with panꢀneutralization against a panel of 56 Envꢀpseudotyped HIVꢀ1 representing
diverse subtypes of clinical isolates (IC as low as 270 nM). The coꢀcrystal structure of NBDꢀ
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
50
11021 complexed to a monomeric HIVꢀ1 gp120 core revealed its detail binding characteristics.
The study is expected to provide a framework for further development of NBD series as HIVꢀ1
entry inhibitors for clinical application against AIDS.
2
ACS Paragon Plus Environment

Page 3 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Entry of human immunodeficiency virus type 1 (HIVꢀ1) to host cells consists of highly
1
orchestrated events and is essential for maintaining the virus lifeꢀcycle . HIVꢀ1 entry process
2ꢀ
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
starts when its surface envelope glycoprotein gp120 bind to the host cell primary receptor CD4
4
. CD4 binding triggers conformational changes in gp120 which facilitate its binding to the host
5ꢀ7
cell coꢀreceptor (secondary) CCR5 or CXCR4 . Coꢀreceptor binding initiates conformational
changes in gp41 which trigger the insertion of its fusion protein and formation of a stable
hexahelical bundle which promotes fusion of virus membrane with cell membrane and virusꢀcell
entry. Each of these steps of the entry pathway has been recognized as targets for developing
8ꢀ11
drugs for prevention and therapy of this deadly disease . Approval of two drugs by the FDA
that target HIVꢀ1 entry has validated entry inhibition to host cells is an effective strategy for
developing drugs. One of the clinically applied drugs, maraviroc (Selzentry) works against
CCR5 and the other drug enfuvirtide (Fuzeon) targets the HIVꢀ1 envelope glycoprotein gp41.
There are no licensed drugs currently available that target HIVꢀ1 gp120.
In the quest to discover effective drugs that target gp120, we identified two inhibitors –
NBDꢀ556 and NBDꢀ557 – in 2005 by targeted screening of small molecules from commercial
12
libraries . Subsequently, the report that these compounds, despite their small molecular
13
weights (337 and 382 Da), mimic CD4 remarkably well sparked interest in multiple research
13ꢀ26
groups to target this site to develop NBDꢀ556 related inhibitors
. In 2011, we reported the
27
structure of NBDꢀ556 bound to HIVꢀ1 gp120 by Xꢀray crystallography . Several other crystal
15,16,28
structures of NBDꢀ556 analogs bound to HIVꢀ1 gp120 have also been reported
. These
structures reveal that NBDꢀ556 and its analogs bind to a cavity (termed the Phe43 cavity,
because of the peripheral contact of the cavity by the side chain phenyl group of residue 43 of
CD4); however, the aromatic chlorophenyl group of NBDꢀ556 and its analogs penetrate deep
into the Phe43 cavity. The structures confirm that NBDꢀ556 does not retain the critical Hꢀ
bond/salt bridge interaction with Asp368gp120 as was observed with Arg59CD4. Unfortunately,
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 71
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
ꢀ
+
NBDꢀ556 and its analogs behave as CD4ꢀagonist and enhance HIVꢀ1 infectivity in CD4 CCR5
cells. We and others attempted to modify region I, II and III of NBDꢀ556 (Figure 1a) and
concluded that region I could be minimally modified and that region III is amenable for
modification. However, it was generally concluded that modifications to region II were
detrimental to binding and antiviral activity, because the oxalamide moiety contributes to the
binding by forming two hydrogen bonds with gp120 residues. We therefore focused on making
modifications to region I and especially to region III. To enhance binding affinity and antiviral
potency, we reasoned that it might be essential to gain the critical Hꢀbond/salt bridge interaction
of the basic moieties in region III with the conserved Asp368gp120. Towards this goal we explored
different scaffolds with basic groups. We synthesized a series of compounds with a piperidine
and a thiazole ring attached with a flexible linker, tested those molecules in a large set of Envꢀ
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
9
pseudotyped HIVꢀ1, and observed measurable enhancement of antiviral activity . However,
both functional and biophysical assays confirmed that this class of compounds retains agonist
property similar to NBDꢀ556. Interestingly, one of these new compounds, NBDꢀ09027, showed
1
9
reduced agonist property compared to NBDꢀ556 both in functional and biophysical studies .
These observations motivated us to determine the Xꢀray structure of NBDꢀ09027 with HIVꢀ1
19
gp120 in order to investigate the interactions . The structure revealed that although the 4ꢀ
chlorophenyl oxalamide group was superimposable with the NBDꢀ556 in the Xꢀray structure, the
basic “nitrogen” of the piperidine ring of the new scaffold was within 4.4 Å of the Asp368gp120
,
though not close enough to form an Hꢀbond or a salt bridge indicating that additional
modifications would be needed to obtain this critical interaction.
Here we report a systematic structureꢀbased approach to modifying region II by
preserving a key Hꢀbond interaction of one of the most active agonists, NBDꢀ09027, with gp120
but to alter positioning of the piperidine moiety used in NBDꢀ09027 to form a second key Hꢀbond
with Asp368gp120. This report demonstrates that modifications in region II are feasible and may
be critical for enhancing antiviral activity of the NBD series entry inhibitors. Most notably, we
4
ACS Paragon Plus Environment

Page 5 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
successfully convert of a full CDꢀagonist (NBDꢀ556) to a full CD4ꢀantagonist (NBDꢀ11021), and
achieve the desired broad spectrum antiꢀHIVꢀ1 activity which has not been previously
accomplished for this class of entry inhibitors.
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
RESULTS
Rational design of NBDꢀ11021
It has been recently reported that VRC01 and NIH45ꢀ46, two highly potent broadly
neutralizing antibodies (bNAbs), induce conformational changes of monomeric gp120 in the
bridging sheet similar to that induced by CD4; thereby showing enhanced binding to mAb 17b.
Therefore, VRC01 and NIH45ꢀ46 induce exposure of the coꢀreceptor binding site, suggesting
29
that they are CD4ꢀagonists . On the contrary, VRC03 and VRCꢀPG04, two other highly potent
bNAbs, induce monomeric gp120 conformation, especially within residues 428ꢀ431gp120 in
strands β20/21, which is distinctly different, thus showing no enhancement in binding to mAb
17b. VRC03 and VRCꢀPG04 thus do not induce exposure of the coꢀreceptor binding site in
gp120, suggesting that they are CD4ꢀantagonists, not agonists. In addition, they also observed
that a critical residue, Met434 on β21 in gp120 in both VRCꢀPG04ꢀ and VRC03ꢀbound
structures, adopts a conformation that clashes with CDRH2 of mAb 17b, whereas this residue
adopts a nonꢀclashing conformation when bound to VRC01 and NIH45ꢀ46, explaining the
29
differences in recognition by the mAb 17b . Interestingly, irrespective of their agonist or
antagonist characteristics they all have a positively charged Arg (Arg79) interacting with
3
0
Asp368gp120. Mutation of Asp368gp120 abrogates the neutralizing capacity of these bNAbs .
Therefore, this interaction is essential for binding of these highly potent bNAbs to gp120 with
high affinity as well as to elicit potent neutralization efficiency. We reasoned that in order to first
enhance the potency of these new series of NBD inhibitors we also need to introduce this
interaction.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 71
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
To enhance antiviral activity and to overcome the agonist property of NBDꢀ09027, we
decided to modify Region II by incorporating a pyrrole ring replacing part of the oxalamide ring
towards the phenyl group. We kept the remaining structure identical. We reasoned that
introduction of the pyrrole ring would help in preserving the Hꢀbond interaction of the NH of
pyrrole with Asn425, but its rigid geometry might alter the conformation of the scaffold in Region
III. This conformational change may facilitate movement of the piperidine ring towards
Asp368gp120 for the desired Hꢀbond/salt bridge interaction.
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
Synthesis of NBDꢀ11021
As illustrated in Schemeꢀ1 the target compound was envisioned to arise from coupling
between pyrroleꢀ2ꢀcarboxylic acid 2’ and amine 1’, which in turn could be prepared by the
addition of the thiazolyl anion to a suitably protected aldehyde or imine 3’.
Thiazole core was prepared by the reaction between thioformamide and ethyl 2ꢀ
chloroacetoacetate (Schemeꢀ2). The reduction of the ester group with LiAlH and the
4
subsequent protection of the alcohol group with TBSCl furnished the first of the three key
building blocks 7. The synthesis of the second building block 11 commenced with the oxime
31
preparation from 4′ꢀchloroacetophenone followed by the Trofimov reaction . The Friedel–Crafts
acylation of the resulting 2ꢀarylpyrrole 9 with trifluoroacetic acid anhydride followed by the
haloform reaction provided the required acid 11 in 78% yield over two steps.
The synthesis of the third part of the target compound and an assembly of the fragments
began with the protection of the pipecolic acid with allyl chloroformate. The formation of the
Weinreb amide 2, followed by the reduction with LiAlH provided the aldehyde 3. The sulfinimide
4
32
4
was prepared from the racemic tertꢀbutanesulfinamide and racemic aldehyde 3 as a mixture
of stereoisomers. The addition of thiazolyl lithium derived from 7 to the imine 4 furnished a
chromatographically separable mixture of isomers in a combined 85% yield. The first fraction
6
ACS Paragon Plus Environment

Page 7 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
contained one pure racemic diastereomer 12a (according to LCMS) and the second fraction
contained presumably a mixture of 2 diastereomers (differing in the configuration of the sulfur
atom) assigned together as 12b (Schemeꢀ2). The isolated isomers 12a and 12b were used in
the subsequent steps separately.
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 synthetic intermediates derived from 12a have been labelled with an “a” letter in
their numbering system (13a, 14a, 15a= NBDꢀ11021A),
The synthetic intermediates derived from 12b have been labelled with a “b” letter in the
numbering system (13b, 14b, 15b= NBDꢀ11021B).
After sulfinyl and TBS groups were cleaved with anhydrous methanolic HCl solution two
pure racemic diastereomers 13a and 13b were produced separately from 12a and 12b. No
stereochemical assignment was performed at this point and both amines were transformed
independently into the target compounds. Only inseparable mixtures of isomers were observed
in the addition reaction when Boc or Cbz protecting groups were used. It is also important to
note that separation could not be achieved in the next steps because of the low differences in
mobility of the compounds on silica gel. Finally, amines 13a and 13b were coupled with acid 11,
33
and Alloc group was removed according to the literature procedure to afford the title
compounds 15a (NBDꢀ11021A) and 15b (NBDꢀ11021B) in racemic form.
1
13
The final compounds did show very similar H and C NMR spectra. To prove once
1
13
again that these compounds are in fact isomers we measured H and C NMR spectra from a
mixture of 15a (NBDꢀ11021A) and 15b (NBDꢀ11021B). The spectral data showed a double set
of signals clearly indicating that 15a (NBDꢀ11021A) and 15b (NBDꢀ11021B) are in fact
stereoisomers.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 71
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
Separation of NBDꢀ11021 diastereomer and assignment of absolute configuration
Initial synthesis of NBDꢀ11021 yielded a diastereomeric mixture (4 isomers), which we
designated “NBDꢀ11021” and used in different assays. However, we were able to synthesize
NDBꢀ11021 as two separate isomeric mixtures (comprising of a mixture of two enantiomers of
each single racemate) and designated these as NBDꢀ11021A and NBDꢀ11021B. We
subsequently used chiral separation fluid chromatography (SFC) to separate these mixtures into
individual enantiomers, which we named NBDꢀ11021A1, NBDꢀ11021A2 and NBDꢀ11021B1,
NBDꢀ11021B2.
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
Vibrational circular dichroism (VCD) was used to delineate the absolute configuration of
these four isomers. They were assigned (Figure 1a) as NBDꢀ11021A1 (16S,25R), NBDꢀ
11021A2 (16R,25S), NBDꢀ11021B1 (16S,25S) and NBDꢀ11021B2 (16R,25R).
NBDꢀ11021 exhibits antiꢀHIVꢀ1 activity in multiꢀcycle and singleꢀcycle assay
Following purification of the two isomers A and B of NBDꢀ11021 and subsequent
separation of the fractions to NBDꢀ11021A1, NBDꢀ11021A2, NBDꢀ11021B1 and NBDꢀ11021B2,
we evaluated the antiꢀHIVꢀ1 activity and the cytotoxicity of these compounds in different cell
lines using both multiꢀcycle and single cycle assays (Table 1) to select the best isomers for
subsequent studies in larger panel of ENV pseudotyped reference viruses. NBDꢀ11021A2 and
NBDꢀ11021B2 showed improvement in antiviral activity compared to the diastereomeric mixture
of NBDꢀ11021 exhibiting an IC as low as 0.8 ꢁM in MTꢀ2 cells against HIVꢀ1 . NBDꢀ11021B1
50
IIIB
showed a significant loss in antiviral activity with respect to the initial diastereomeric NBDꢀ11021
and the isomers NBDꢀ11021A and NBDꢀ11021B in multiꢀcycle assay.
We detected similar level of cytotoxicity in all the isomers with respect to the original
diastereomeric mixture of NBDꢀ11021 with minimal variations considering different sensitivity of
the diverse cell lines we tested. While the antiꢀviral neutralization appeared specific, we note
that neutralization of AꢀMLV was observed (see below).
8
ACS Paragon Plus Environment

Page 9 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The NBDꢀ11021 shows similar antiviral activity against CXCR4ꢀ and CCR5ꢀtropic HIVꢀ1
Interaction of gp120 with the cellular receptor CD4 and coꢀreceptors CCR5 or CXCR4
3
4,35
results in the formation of fusion pores and release of the HIV genome into target cells
.
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
CCR5ꢀusing viruses are responsible for both interꢀindividual transmission and for sustaining the
pandemics whereas CXCR4ꢀusing viruses, usually dualꢀtropic R5X4, emerge in about 50% of
36
individuals in the late, immunologically suppressed stage of the disease . Therefore, it was
pertinent to verify whether NBDꢀ11021 shows activity against both CCR5ꢀtropic and CXCR4ꢀ
tropic HIV viruses. To this end, U87ꢀCD4ꢀCCR5 and U87ꢀCD4ꢀCXCR4 cells were infected with
pseudoviruses HIVꢀ1ADA (CCR5ꢀtropic) and HIVꢀ1HXB2 (CXCR4ꢀtropic) and treated with varying
doses of NBDꢀ11021 and its isomers. Our results indicate that this compound inhibited both
CCR5ꢀ and CXCR4ꢀtropic HIVꢀ1 with similar efficiency with an IC50 of ~1.7ꢀ2.4 ꢁM (Table 1),
which was about 4ꢀ5ꢀfold improvement in activity with respect to NBDꢀ556 and NBDꢀ09027.
NBDꢀ11021 blocks gp120ꢀCD4 interaction
Previously we reported that several NBDꢀcompounds, such as NBDꢀ556, NBDꢀ557,
NBDꢀ09027 and NBDꢀ10007, bind to the Phe43 cavity and inhibit interaction between CD4 and
12,20,27,28
gp120
. Since, in this study, we made major modification in region II, we investigated
how this change might affect the interaction with both CD4ꢀpositive and CD4ꢀnegative cells. To
accomplish this, we evaluated the effect of NBDꢀ11021 on the infection of a CD4ꢀdependent
virus HIVꢀ1ADA in Cf2Th/CD4ꢀCCR5 target cells that express CD4 and CCR5 and a CD4ꢀ
independent mutant HIVꢀ1ADAN197S in Cf2ThꢀCCR5 target cells that express CCR5, but not
12,17
CD4. As was observed with NBDꢀ556 and NBDꢀ09027
, NBDꢀ11021 inhibited the CD4ꢀ
dependent virus in a doseꢀdependent manner with an IC of 2.4±0.2 ꢁM (Figure 1b) suggesting
50
that this compound inhibits interaction between gp120 and CD4. However, while NBDꢀ556 and
NBDꢀ09027 did not inhibit infection of the target cells by the CD4ꢀindependent virus HIVꢀ
1ADAN197S, NBDꢀ11021 inhibited the CD4ꢀindependent virus with similar activity (IC =2.1±0.15
50
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 71
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
37
ꢁ
M). A similar inhibition was reported for entry inhibitor BMSꢀ378806 . Moreover, we and
others have recently shown that NBDꢀ556 can also act as CD4ꢀagonist promoting CCR5
13,19
binding and enhancing HIVꢀ1 entry into CD4ꢀnegative cells expressing CCR5
while the
1
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
2
3
4
5
6
7
8
9
0
enhancement of infection by NBDꢀ09027 was about twoꢀfold lower . To further evaluate the
activity of NBDꢀ11021, we infected CD4ꢀnegative Cf2ThꢀCCR5 cells with recombinant CD4ꢀ
dependent HIVꢀ1ADA in the presence of various concentrations of NBDꢀ11021 (Figure 1c). NBDꢀ
17,19
5
56, NBDꢀ09027 and BMSꢀ378806 were used as control. As reported previously
, NBDꢀ556
enhanced HIVꢀ1 entry into CD4ꢀnegative cells and NBDꢀ09027 showed moderate enhancement
with respect to NBDꢀ556; however, NBDꢀ11021, similar to BMSꢀ378806, did not enhance HIVꢀ1
ꢀ
+
infectivity in CD4 CCR5 cells suggesting that the structural modification of NBDꢀ09027
successfully converted it to a CD4ꢀantagonist.
To verify this finding biophysically, we used surface plasmon resonance (SPR) to
determine if NBDꢀ11021 induced the antibody 17bꢀrecognized conformation in gp120. sCD4,
NBDꢀ556, and NBDꢀ09027 were used as control. In the presence of a 15ꢀfold excess of sCD4,
27
the YU2 core gp120 demonstrated enhanced binding to antibody 17b compared to the YU2
m
corem gp120 alone. Among the NBD compounds tested NBDꢀ556 showed the highest
enhancement of antibody 17b binding to YU2 core gp120 whereas NBDꢀ09027 showed much
m
reduced enhancement. Remarkably, no enhancement in binding of mAb 17b to YU2 corem
gp120 was observed in the presence of NBDꢀ11021 (Figure 1d) suggesting that NBDꢀ11021
functions as an antagonist.
NBDꢀ11021 binds to gp120
We used competitive SPR to validate and to characterize the target of NBDꢀ11021 on
gp120. As shown in Figure 1e NBDꢀ11021 competes with CD4 to bind to gp120 in a doseꢀ
dependent manner. The recent structure of antibody VRCꢀPG04 (also called antibody VRCꢀ
38
PG04) with a trimeric form of HIVꢀ1 Env (BG505 SOSIP.664) demonstrated that this antibody
10
ACS Paragon Plus Environment

Page 11 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
recognizes the prefusion mature closed conformation of HIVꢀ1 Env. We used competitiveꢀSPR
to verify whether NBDꢀ11021 shared the same binding site as VRCꢀPG04. The data in Figure 1f
suggest that at higher ratios of NBDꢀ11021 versus VRCꢀPG04, NBDꢀ11021 effectively
competed with VRCꢀPG04 to inhibit its binding to gp120 in a doseꢀdependent manner. The data
suggest that NBDꢀ11021 binding site on gp120 overlapped with that of VRCꢀPG04, which binds
to the Phe43 cavity in gp120.
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
NBDꢀ11021 shows broad spectrum inhibition against a large diverse panel of ENV
pseudotyped reference viruses
To further evaluate the breadth of antiꢀHIVꢀ1 activity we tested NBDꢀ11021 against 56
HIVꢀ1 Envꢀpseudotyped reference viruses of diverse clades including primary and
39
transmitted/founder HIVꢀ1 isolates . These ENVꢀpseudoviruses were derived from patient
serum and exhibit a neutralization phenotype that is typical of most primary isolates. We also
used the two purified isomers NBDꢀ11021A2 and NBDꢀ11021B2 which showed the best antiꢀ
HIVꢀ1 activity in our preliminary assays, and we compared their activity with compounds NBD–
556 and NBDꢀ09027. Six representative doseꢀdependent inhibition plots of the neutralization
assay of selected pseudoviruses (#11314, #11316, #11911, #11526, NIH #11888, and
KNH1144) are shown in Figure S1. The results in Table 2 indicate that NBDꢀ11021, even as
mixture of isomers, showed improved antiviral activity against all viral isolates tested including
subtype C with an overall range of activity of 0.6 ꢁM to 5.3 ꢂM of which 82% of the viruses
showed sensitivity at ≤ 3 ꢂM and 39 % at ≤ 2 ꢂM. By contrast, NBDꢀ556 showed relatively
much lower antiviral activity against most of the isolates tested with an overall range in activity
of 1.9 – 27.8 ꢂM of which only 7% of the viruses showed sensitivity at ≤ 3 ꢂM and only one at ≤
2 ꢂM. Additionally, the two most active isomers NBDꢀ11021A2 and NBDꢀ11021B2 separated
from the Isomers A and B respectively, exhibited improved activity over the diastereomeric
mixtures of NBDꢀ11021 (mean IC : 2.38 ꢂM) (Figure S2). For example, NBDꢀ11021A2 isomer
50
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 71
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
had an IC in the range of 0.32ꢀ3.7 ꢁM (mean IC : 1.71 ꢂM) with 68% of the viruses showed
50
50
sensitivity ≤ 2 ꢂM. Similarly, NBDꢀ11021B2 isomer showed antiviral activity in the range of
.27ꢀ2.7 ꢁM (mean IC : 1.57 ꢂM) with 71% of the viruses showed sensitivity ≤ 2 ꢂM indicating
0
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
that the separation of isomers resulted in improved antiꢀHIV activity. Surprisingly, NBDꢀ11021
(isomer mixtures and pure isomers) also showed inhibition against HIVꢀ1 pseudotyped with Aꢀ
40
MLV envelope glycoprotein. Recently, AꢀMLV is shown to enter cells by macropinocytosis .
41,42
HIVꢀ1 entry has also been shown to occur by endocytic pathway
including macropinocytosis
43
in macrophages . By contrast, NBDꢀ11021 and its two isomers showed poor activity against
HIVꢀ1 pseudotyped with VSVꢀG envelope. Furthermore, NBDꢀ11021 had also poor activity
against SHIVSF162P3, SIVmac186 and HIVꢀ2 isolates (Table S2).
Antiviral activity on NBDꢀ11021 against a diverse set of laboratoryꢀadapted and primary
HIVꢀ1 isolates
As a next step, we decided to test the inhibitory activity of NBDꢀ11021 against a diverse
set of seven laboratoryꢀadapted HIVꢀ1 strains and eight primary HIVꢀ1 isolates representing the
diverse subtypes (B, C, E and G) and coꢀreceptor usage (R5, X4 and R5X4). In addition, we
have tested this compound against one RT inhibitorꢀresistant, one protease inhibitorꢀresistant
and one fusion inhibitorꢀresistant isolates. The inhibitory activity of NBDꢀ11021 was compared
to the activity of earlier compounds NBDꢀ556 and NBDꢀ09027. The data indicated substantial
improvement of antiviral activity of NBDꢀ11021 from the initially identified compounds (Table 3).
In fact, NBDꢀ556 inhibited HIVꢀ1 labꢀadapted isolates with an IC in the range of 4.8 ꢀ ≥ 118 ꢁM
50
and the IC of NBDꢀ09027 was in the range of 4 ꢀ 35.8 ꢁM while the IC of NBDꢀ11021 was in
50
50
the range of 1.2 ꢀ 7.3 ꢁM. Similarly, for all the drug resistant and primary isolates HIVꢀ1 strains
NBDꢀ11021 was the only compound that showed a consistent inhibitory activity. Remarkably,
NBDꢀ11021 not only had a low micromolar activity against all the primary HIVꢀ1 isolates we
tested in PBMC but it was also the only compound which exhibited activity against the HIVꢀ1
12
ACS Paragon Plus Environment

Page 13 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
isolates of subtype C we tested in this study, HIVꢀ193IN101 and HIVꢀ193MW959 while NBDꢀ556 and
NBDꢀ09027 showed poor to no activity against these HIVꢀ1 strains. Also, the HIVꢀ1BCF02 of
group O, showed practically no sensitivity to NBDꢀ09027 but it was sensitive to NBDꢀ11021.
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
NBDꢀ11021 prevents cellꢀtoꢀcell fusion
To investigate whether NBDꢀ11021 and its two isomers NBDꢀ11021A2 and NBDꢀ
11021B2 prevent cellꢀtoꢀcell fusion we coꢀcultured MAGIꢀCCR5 cells and HL 2/3 cells in a virusꢀ
free culture in the presence of different concentrations of compounds. We used BMSꢀ378806
12,44
and NBDꢀ556 as controls because they were previously shown to inhibit cellꢀcell fusion
. As
expected both controls BMSꢀ378806 and NBDꢀ556 inhibited fusion between MAGIꢀCCR5 and
HL 2/3 with an IC of 14 nM and 6.5 ꢁM, respectively (Table 4). Moreover, NBDꢀ11021 and its
5
0
two isomers NBDꢀ11021A2 and NBDꢀ11021B2 also inhibited cellꢀtoꢀcell fusion with similar
activity ranges by other NBD compounds.
NBDꢀ11021 also targets HIVꢀ1 Reverse Transcriptase (RT) but shows no activity against
integrase
The activity of NBDꢀ11201 against both MLV and VSVꢀG clearly indicated that
there may be additional mechanism of action of this molecule. Since both MLV and VSVꢀG has
different entry pathway we hypothesized that the other target of this compound could be at a
later step in the life cycle of the virus. This led us to test this molecule against HIVꢀ1 RT and
HIVꢀ1 integrase. We did not select HIVꢀ protease (PR) because these series of compounds
showed potent activity in a single cycle assay. For the RT assay we tested NBDꢀ11021A2 and
NBDꢀ556 and used Nevirapine and AZTꢀTP as controls. NBDꢀ11021 showed HIVꢀ1 RT
inhibition with an IC of 43.4 ꢁM; however, NBDꢀ556 did not show any activity even at 200 ꢁM.
5
0
Nevirapine and AZTꢀTP showed potent activity against HIVꢀ RT with IC of 0.20 and 0.008 ꢁM,
50
respectively (Table 4).
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 71
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
However, neither NBDꢀ11021A2 nor NBDꢀ556 showed any activity against HIVꢀ1
integrase whereas the control integrase inhibitor Raltegravir showed inhibition with an IC50 of
0.21 ꢁM (Table 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
NBDꢀ11021 blocks cellꢀtoꢀcell transmission of HIVꢀ1
HIVꢀ1 infection occurs via both, cellꢀfree and cellꢀassociated transmission; however, the
relative importance of the later transmission pathway is not clear yet. Recently, Abela et al.
elegantly demonstrated that some of the CD4BS directed antibodies and CD4M47, a peptideꢀ
45
based CD4ꢀmimetic entry inhibitor, lost efficiency in inhibiting cellꢀcell transmission of HIVꢀ1 .
However, in a more recent report by Agosto, et al. it was shown that several antiretroviral drugs
46
including some entry inhibitors effectively inhibited cellꢀcell transmission . In view of those
findings, we investigated the ability of NBDꢀ11021 to inhibit cellꢀcell transmission. To this end,
we used Ghost X4/R5 as target cells and H9/HIVꢀ1IIIB and MOLTꢀ4/HIVꢀ1ADA as effector cells.
While NBDꢀ556 and NBDꢀ09027 had poor activity against the CXCR4ꢀtropic HIVꢀ1 cellꢀcell
transmission (IC50 ~60 ꢁM) (Figure 2a), NBDꢀ11021 and its two isomers NBDꢀ11021A2 and
NBDꢀ11021B2 inhibited cellꢀcell transmission effectively. NBDꢀ11021B2 was the most active
inhibitor with an IC of 2 ꢁM. With the CCR5ꢀtropic cellꢀcell transmission assay we found that
50
NBDꢀ556 had similar IC (~60 ꢁM) as observed with CXCR4ꢀtropic virus while the NBDꢀ09027
50
had practically no inhibitory activity (Figure 2b). By contrast, NBDꢀ11021 and the two isomers
again showed inhibition though slightly higher than in the CXCR4ꢀtropic cellꢀcell transmission
system.
Crystal structure of NBDꢀ11021 in complex with gp120 core
To understand underlying mechanism(s) of improved antiviral potency of NBDꢀ11021,
we have determined the crystal structure of gp120 core in complex with NBDꢀ11021 and
compared it with NBDꢀ09027ꢀboundꢀgp120 structure. We used clade A/E 93TH057 H375S
gp120 core to determine NBDꢀ11021ꢀbound gp120 structure. Among the four isomers of NBDꢀ
e
14
ACS Paragon Plus Environment

Page 15 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1021, NBDꢀ11021A1 fits the best to the electron density map. NBDꢀ11021 bound to the Phe43
cavity in a manner similar to that observed with other NBDꢀanalogs; the chlorophenyl ring and
the linker region of the compound penetrate the cavity while the region III of the compound
resides in the vestibule of the cavity (Figure 3a,b,c). When the two structures were
superimposed, the phenyl rings were on top of each other, but the two different linker regions
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
(
oxalamide of NBDꢀ09027 and pyrrole of NBDꢀ11021) prompted the region III of the compounds
to deviate substantially from each other in the cavity (Figure 3b,c). Val255 and Thr257 make
hydrophobic interactions with the phenyl ring of Region I, Trp427 with the linker region (Region
II), and Gln428 and Gly429 make hydrophobic interactions with the 5ꢀmembered thiazole ring of
the Region III of NBDꢀ11021 (Figure 3d). Interestingly, the nitrogen atom (N13) in the pyrrole
ring of the linker makes a hydrogen bond with Asn425 (Figure 3d). Most interestingly, the new
linker of NBDꢀ11021 allowed its piperdine ring to position near the Asp368gp120 and to make a
hydrogen bond with the nitrogen atom of the piperidine ring (Figure 3d), which has not been
observed in NBDꢀ09027ꢀbound gp120 structure. This interaction may contribute significantly to
the improved affinity of NBDꢀ11021 to gp120 and its antiviral properties over NBDꢀ09027.
Similar interactions between the guanidinium group of DMJꢀIꢀ228, an NBD analog, and
15
Asp368gp120 was reported previously (PDB ID: 4DKQ) .
DISCUSSION
The highly conserved Phe43 cavity has been validated as potential target for developing
12,19,20,47,48
entry inhibitors as drugs for HIVꢀ1 therapy and prophylaxis
. The discovery of NBDꢀ556
in 2005 by our group and the subsequent finding that this class of small molecules mimics CD4
and induces conformational changes in gp120 in similar manner generated considerable
interest to pursue on designing NBDꢀ556 analogs as entry inhibitors. However, these inhibitors,
48
except one reported recently , induce gp120 conformation suitable for binding by CCR5 coꢀ
receptor; therefore, they function as CD4ꢀagonist, an undesirable property for developing these
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 71
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
inhibitors as drug candidates because they may promote HIVꢀ1 entry and enhance infection in
ꢀ
+
CD4 CCR5 target cells.
Recently, we designed NBDꢀ09027, a more active analog of NBDꢀ556, by modifying the
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
scaffold in region III. Interestingly, despite its higher potency in singleꢀcycle HIVꢀ1 infection
assay we demonstrated by both functional and biophysical studies that this molecule still
possesses the undesired CD4ꢀagonist property. However, we noted that the tendency of this
molecule to function as agonist was about 50% less than NBDꢀ556. Subsequently the Xꢀray
structure of NBDꢀ09027 bound to gp120 indicated the absence of a critical interaction between
the piperidine ring and Asp368gp120 of gp120. We reasoned that to enhance binding affinity and
antiviral activity we needed further modification to achieve this interaction. In this study we
presented the successful modification of part of the oxalamide moiety with a pyrrole ring in
48
region II, which was reported as intolerant to modification . The Xꢀray structure of NBDꢀ11021ꢀ
gp120 indicated that the introduction of a more rigid 5ꢀmembered ring in region II forced the
molecule to adopt a conformation different from NBDꢀ09027 and bind to the cavity in a different
orientation than NBDꢀ09027. The structure confirmed that this modification helped in acquiring
an Hꢀbond interaction between the piperidine ring and Asp368gp120. However, this interaction
alone cannot be attributed to the CD4ꢀantagonist property of NBDꢀ11021 because the Xꢀray
structural information reported here is limited due to the deletion of the V3 loop in the gp120
structure. Therefore, the effect of this compound on the V3 loop conformation is not available.
34,49,50
V3 loop has been implicated as the major binding site for the CCR5 coꢀreceptor binding
51,52
and mutations in the V3 loop affects CCR5 binding
. Therefore, it is most likely that a
cumulative effect of the binding of NBDꢀ11021 in the Phe43 cavity in gp120 which is in the
proximity of the V3 loop may have altered the conformation of gp120 that is no longer suitable
for CCR5 binding. Consequently, NBDꢀ11021 behaves as CD4ꢀantagonist.
Since NBDꢀ11021A2 and NBDꢀ11021B2 showed somewhat better antiviral activity, we
made an attempt to fit these enantiomers to the electron density map of the gp120 structure in
16
ACS Paragon Plus Environment

Page 17 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
complex with NBDꢀ11021 mixture. We found that both these isomers could be modeled in such
a way that the thiazole ring is now facing to the Asp368 and the nitrogen atom in the thiazole
ring is ~4 Å apart from Asp368 to form Hꢀbond (Figure S10). Although the fit of these isomers is
not as ideal as NBDꢀ11021A1, these putative binding modes of isomers may provide additional
guidance to optimize this lead compound.
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 singleꢀcycle infectivity assay with a large number of Envꢀpseudotyped HIVꢀ1
demonstrated substantial improvement in antiviral potency of NBDꢀ11021A2 and NBDꢀ11021B2
compared to NBDꢀ556. Considerable improvement in antiviral potency of NBDꢀ11021 compared
to NBDꢀ556 and NBDꢀ09027 was also observed against replicating laboratory strains and
ꢀ
primary isolates in multiꢀcycle assay. Most significantly, the cellꢀbased functional assay in CD4
+
CCR5 cells demonstrated that NBDꢀ11021 did not enhance HIVꢀ1 infection in these cells.
Furthermore, the SPR (BIAcore) based assessment of CCR5 antibody surrogate mAb 17b
binding to the NBDꢀ11021ꢀgp120 core complex indicated that this inhibitor did not enhance mAb
17b binding. The data confirm that NBDꢀ11021 is a CD4ꢀantagonist. When we used AꢀMLV
envelope pseudotyped with HIVꢀ1 backbone to test the specificity of the NBDꢀ11021, we
surprisingly found that NBDꢀ11021 also inhibits this virus which belongs to the same retrovirus
family. The AꢀMLV pseudovirus has HIVꢀ1 backbone. Therefore, NBDꢀ11021 may also inhibit a
4
0
post entry step. AꢀMLV has been shown recently to enter cells by macropinicytosis .
Furthermore, HIVꢀ1 protease (PR) inhibitors have been shown to also work against MLV PR
despite the fact that these two PR show substantial differences in specificity towards substrates
and their molecular models of MLV PR suggested that the residues forming the subsites for
53
substrate recognition in both PRs are different . Therefore, it is not unusual to find antiviral
activity of the same compounds in different viruses with distinctly different structural features. In
order to unravel other possible target of NBDꢀ11021 we tested one of its isomers, NBDꢀ
11021A2, against HIVꢀ1 RT and integrase and indeed we found that it inhibits HIVꢀ1 RT but at
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 71
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
relatively higher dose (IC of 43.4 ꢁM). NBDꢀ11021A2 did not show any activity against HIVꢀ1
50
integrase. Recently, a pyrazoloꢀpiperidine based HIVꢀ1 entry inhibitor that targets coꢀreceptors
CCR5/CXCR4 has been shown to also target HIVꢀ1 RT with an IC of 9.0 ꢁM. This compound
50
54
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
was ~5ꢀfold potent against HIVꢀ1 RT than NBDꢀ11021 .
CONCLUSION
In this study we report successful conversion of a CD4ꢀagonist NBDꢀ556 to a CD4ꢀ
antagonist NBDꢀ11021 by systematic structureꢀbased design, synthesis and functional and
biophysical assays. The antagonist showed the level of breadth in antiviral activity against a
panel of diverse HIVꢀ1 Envꢀpseudotyped viruses in singleꢀcycle assay never reported before for
small molecule inhibitors that target the Phe43 cavity. Furthermore, our study indicates that
NBDꢀ11021 also efficiently inhibits cellꢀtoꢀcell fusion and cellꢀassociated virus infection. Besides
inhibiting viral entry we also uncovered that NBDꢀ11021 also inhibits HIVꢀ1 RT but at much
higher concentration. This discovery is expected to pave the way to further optimize NBDꢀ11021
to a clinical candidate for HIVꢀ1 prevention and therapy.
EPERIMENTAL SECTION
Cells, virus and plasmids
MTꢀ2 cells (Human Tꢀcell leukemia cells were obtained through the NIH AIDS Reagent
Program (ARP) from Dr. D. Richman), H9/HTLVꢀ/HIVꢀ1IIIB cells (obtained through the NIH ARP
from Dr. R. Gallo) and Human PBMC (isolated from buffy coats of healthy HIVꢀ1 negative donor
obtained from the New York Blood Center, NY) were grown in RPMI 1640 medium
supplemented with 10% fetal bovine serum (FBS), penicillin and streptomycin (100 U/ꢂl each).
TZMꢀbl cells (a HeLa cell line that expresses CD4, CXCR4 and CCR5 and expresses
luciferase and ßꢀgalactosidase under control of the HIVꢀ1 promoter, obtained from Dr. J. C.
18
ACS Paragon Plus Environment

Page 19 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
55 56
Kappes, Dr. X. Wu and Tranzyme Inc. through the NIH ARP)
HEK 293Tcells (ATCC),
57
Cf2Th/CD4ꢀCCR5 and CD4ꢀCf2ThꢀCCR5 cells (kindly provided by Dr. J. G. Sodroski ) were
grown in DMEM medium supplemented with 10% FBS, penicillin and streptomycin.
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
U87ꢀCD4ꢀCCR5 and U87ꢀCD4ꢀCXCR4 cells (obtained through the ARP from Drs. H.
58
Deng and D. Littman) were grown in DMEM supplemented with 15% FBS, 1 ꢁg/ml puromycin,
300 ꢁg/ml G418, penicillin and streptomycin.
GHOST (3) X4/R5 cells (obtained through the NIH ARP from Drs. V. N. Kewalramani
59
and D. Littman) were grown in DMEM supplemented with 10% FBS, 500 ꢂg/ml G418, 100
ꢂg/ml hygromycin, 1 ꢂg/ml puromycin, penicillin and streptomycin.
MOLTꢀ4/CCR5 cells (obtained through the NIH ARP from Drs. M. Baba, H. Miyake and
Y. Iizawa) were grown in RPMI 1640 medium supplemented with 10% FBS, 1mg/ml G418
penicillin and streptomycin.
MAGIꢀCCR5 cells (obtained from Dr. J. Overbaugh through the NIH ARP), and HL 2/3
cells (obtained from Drs. B.K. Felber and G.N. Pavlakis through the NIH ARP), were grown in
DMEM medium supplemented with 10% FBS, penicillin and streptomycin.
HIVꢀ1 labꢀadapted and primary isolates, SIV, SHIV and HIVꢀ2 strains were obtained
through the NIH ARP.
Env expression vector pSVIIIenvꢀADA and mutant pSVIIIenvꢀADAꢀN197S DNA were
57
kindly provided by Dr. J. G. Sodroski . HIVꢀ1 Env molecular clone expression vector pHXB2ꢀ
60
env (X4) DNA was obtained through the ARP from Drs. K. Page and D. Littman . HIVꢀ1 Env
molecular clones of gp160 genes for HIVꢀ1 Env pseudovirus production were obtained as
follows: clones representing the standard panel A, A/D, A2/D, D and panel C
(
QB099.391M.ENV.B1, QB099.391M.ENV.C8 and QC406.70M.ENV.F3) were obtained through
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 71
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
61,62
the NIH ARP from Dr. J. Overbaugh
. The HIVꢀ1 Env molecular clones panel of subtype A/G
and CRF01_AE clone 269 were obtained through the NIH ARP from Drs. D. Ellenberger, B. Li,
63
M. Callahan, and S. Butera . The AE clones AA058 and CM244 were kindly provided by Drs.
Robert J. McLinden and AgnèsꢀLaurence Chenine from US Military HIV Program, Henry M.
Jackson Foundation (Silver Spring, MD). The HIVꢀ1 Env panel of standard reference subtype B
Env clones were obtained through the NIH ARP from Drs. D. Montefiori, F. Gao and M. Li (PVO,
clone 4 (SVPB11); TRO, Clone 11 (SVPB12); AC10.0, clone 29 (SVPB13); QH0692, clone 42
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
(
(
SVPB6); SC422661, clone B (SVPB8)); from Drs. B. H. Hahn and J. F. SalazarꢀGonzalez
pREJO4541, clone 67 (SVPB16); pRHPA4259, clone 7 (SVPB14); pWITO4160 clone 33
(SVPB18)); from Drs. B.H. Hahn and D.L. Kothe (pTHRO4156 clone 18 (SVPB15), pCAAN5342
56,64,65
clone A2 (SVPB19))
. The subtype B pWEAUd15.410.5017 and p1058_11.B11.1550 were
3
9
obtained through the NIH ARP from Drs. B. H. Hahn, B. F. Keele and G. M. Shaw . The
subtype C HIVꢀ1 reference panel of Env clones were also obtained through the NIH ARP from
Drs. D. Montefiori, F. Gao, S. A. Karim and G. Ramjee (Du 156.12; Du172.17); from Drs. D.
Montefiori, F. Gao, C. Williamson and S. A. Karim (Du422.1), from Drs. B. H. Hahn, Y. Li and
J.F. SalazarꢀGonzalez (ZM197M.PB7; ZM233M.PB6; ZM249M.PL1; ZM214M.PL15); from Drs.
E. Hunter and C. Derdeyn (ZM53M.PB12; ZM135M.PL10a; ZM109F.PB4); from Drs. L. Morris,
66ꢀ68
K. Mlisana and D. Montefiori, (CAP45.2.00.G3; CAP210.2.00.E8)
. The paired envelope
variants subtype D/A Maternal (MF535.W0M.ENV.C1) and Infant (BF535.W6M.ENV.A1)
69
molecular clones were obtained through the NIH ARP from Dr. J. Overbaugh .
The ENV pseudotyped genes of BG505.T332N, KNH1144 and B41 were kindly provided
by Dr. J. P. Moore of the Weil Cornell Medical College, NY. SVꢀAꢀMLVꢀenv from Drs. N. Landau
and D. Littman and pHEFꢀVSVG from Dr. L. Chang were obtained through the NIH ARP.
The ENVꢀdeleted proꢀviral backbone plasmids, pNL4ꢀ3.Luc.Rꢀ.Eꢀ DNA (Dr. N.
70,71
env
56,65
Landau)
and pSG3∆ DNA (Drs. J. C. Kappes and X. Wu)
were obtained through the
20
ACS Paragon Plus Environment

Page 21 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NIH ARP. The Envꢀdeleted proviral backbone plasmid pNL4ꢀ3KFS DNA was kindly provided by
Dr. E. Freed from NIH/NCI. The DNA fragment encoding revꢀgp160 was subcloned from the
72
YU2 strain. The expression vector was constructed as previously described .
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Soluble CD4 (sCD4) was purchased (ImmunoDiagnostics, Woburn, MA).17b hybridoma
was kindly provided by Dr. James Robinson.
Chemistry
General. Commercial reagents and solvents were used without further purification. All reactions
were performed in the air atmosphere unless otherwise stated. Reactions were monitored by
thin layer chromatography (TLC) carried out on Merck TLC Silica gel plates (60 F254), using UV
light for visualization and basic aqueous potassium permanganate or iodine fumes as a
1
13
developing agent. NMR H and C spectra were recorded on Bruker Avance 400 instrument
with operating frequency of 400 and 100 MHz respectively and calibrated using residual
undeuterated chloroform (δH = 7.28 ppm) and CDCl (δC = 77.16 ppm), or undeuterated DMSO
3
(
δH = 2.50 ppm) and DMSOꢀd (δC = 39.51 ppm) as internal references. The following
6
abbreviations are used to set multiplicities: s = singlet, d = doublet, t = triplet, q =quartet, m =
multiplet, br. = broad.
1ꢀ((allyloxy)carbonyl)piperidineꢀ2ꢀcarboxylic acid (1)
DLꢀPipecolic acid (18.20 g, 0.141 mol, 1.5 equiv) was dissolved in water (70 mL), containing
NaOH (5.60 g, 0.140 mol). To this solution THF was added (70 mL) and the reaction mixture
was cooled to 0 °C. At this temperature from two separate funnels were added allyl
chloroformate (10 mL, 94.1 mmol in 20 mL of THF) and NaOH solutions (5.60 g, 0.140 mol in 70
mL of water). The reaction mixture was allowed to warm to an ambient temperature, acidified
with 24 mL of conc. aqueous hydrochloric acid and extracted with dichloromethane (3x100mL).
The combined organic phases were dried over anhydrous Na SO , and evaporated to give a
2
4
white solid which was used without further purification. M=18.00 g, Yield=90%.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
H NMR (CDCl , 400 MHz): δ = 1.26 ꢀ 1.54 (m, 2 H), 1.63 ꢀ 1.78 (m, 3 H), 2.21 ꢀ 2.34 (m, 1 H),
3
2
.92 ꢀ 3.14 (m, 1 H), 4.09 (dd, J=1.0 Hz, 1 H), 4.63 (s, 2 H), 4.96 (d, J=27.5, 13.0 Hz, 1 H), 5.16
5.29 (m, 1 H), 5.33 (d, J=16.6 Hz, 1 H), 5.86 ꢀ 6.02 (m, 1 H), 9.94 (br. s., 1 H).
ꢀ
13
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
C NMR (CDCl , 100 MHz): δ = (20.75, 20.82), (24.6, 24.8), (26.7, 26.8), (41.8, 41.9), (54.2,
3
54.4), (66.5, 66.6), 117.6, 132.9, (155.9, 156.6), 177.4.
Allyl 2ꢀ(methoxy(methyl)carbamoyl)piperidineꢀ1ꢀcarboxylate (2)
CDI (15.20 g, 93.8 mmol, 1.1 equiv,) was added portionwise to a solution of 1 (18.00 g, 84.5
mmol) in MeCN (90 mL) (Caution! CO evolution!). The resulting solution was stirred for 5
2
minutes and N,Oꢀdimethylhydroxylamine hydrochloride (10.00 g, 109.3 mmol, 1.2 equiv) was
added in one portion. The reaction mixture was stirred overnight and evaporated to half of a
volume. The residue was dissolved in 500 mL of CH Cl and washed with 5ꢀ10% aqueous
2
2
hydrochloric acid and then with 5% K CO solution. The organic phase was dried over
2
3
anhydrous Na SO , and evaporated to give the pure compound 2 as a nonꢀviscous liquid.
2
4
M=18.91 g, Yield=87%.
1
H NMR (CDCl , 400 MHz): δ = 1.35 ꢀ 1.58 (m, 2 H), 1.59 ꢀ 1.67 (m, 1 H), 1.76 (m, 2 H), 1.93 ꢀ
3
2.07 (m, 1 H), 3.18 (s, 3 H), 3.43 ꢀ 3.65 (m, 1 H), 3.71 (s, 1 H), 3.81 (s, 2 H), 4.03 (m, 1 H), 4.53
ꢀ 4.68 (m, 2 H), 5.02 ꢀ 5.14 (m, 1 H), 5.16 ꢀ 5.37 (m, 2 H), 5.94 (m, 1 H).
13
C NMR (CDCl , 100 MHz): δ = 19.5, (24.75, 24.83), 26.5, (31.9, 32.1) (br.), (42.1, 42.2), (51.3,
3
5
1.5), 61.34, 66.10, (117.2, 117.7), 133.1, 156.9, 173.2.
Allyl 2ꢀformylpiperidineꢀ1ꢀcarboxylate (3)
LiAlH (2.65 g, 69.7 mmol) was added in small portions to the solution of 2 (17.91 g, 70.0 mmol)
4
in THF (70 mL) at 0 °C. The resulting reaction mixture was stirred at 0 °C until the complete
consumption of the starting material (~1 hour, TLC control, 1:1 hexanes:EtOAc). After cooling to
ꢀ
20 °C the excess of LiAlH was quenched by a dropwise addition of water (~5 mL). After that
4
1
0% aqueous HCl (100 mL) was added dropwise and the resulting cold solution was
immediately extracted with CH Cl (3x100 mL). The combined organic phases were dried over
2
2
22
ACS Paragon Plus Environment

Page 23 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
anhydrous Na SO , and evaporated to give an oil which was used without further purification.
2
4
M=9.16 g, Yield=66%.
1
H NMR (CDCl , 400 MHz): δ = 1.16 ꢀ 1.36 (m, 1 H), 1.46 (q, J=12.7 Hz, 1 H), 1.55 ꢀ 1.78 (m, 3
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
H), 2.16 ꢀ 2.32 (m, 1 H), 2.80 ꢀ 3.08 (m, 1 H), 3.96 ꢀ 4.26 (m, 1 H), 4.57 ꢀ 4.79 (m, 3 H), 5.13 ꢀ
5.40 (m, 2 H), 5.85 ꢀ 6.08 (m, 1 H), 9.62 (s, 1 H).
13
C NMR (CDCl , 100 MHz): δ = 20.9, (23.5, 23.7), (24.6, 24.8), (42.4, 42.8), 61.1, 66.5, 117.7,
3
132.8, 156.4, (201.0, 201.2).
Allyl 2ꢀ(((tertꢀbutylsulfinyl)imino)methyl)piperidineꢀ1ꢀcarboxylate (4)
To a solution of 3 (9.16 g, 46.5 mmol) in THF (46 mL), tertꢀbutylsulfonamide (7.62 g, 63.0 mmol,
1.35 equiv) was added in one portion followed by the addition of Ti(OiPr) (28 mL, 94.6 mmol, 2
4
equiv). The resulting solution was stirred overnight and loaded directly on silica.
Chromatography with 1:1 hexanes:EtOAc and then sole EtOAc as eluent gave fractions which
contained the title compound and some Ti(OiPr) as admixture (cloudy fractions). Ti(OiPr) was
4
4
quenched by the addition of 1 mL of water, TiO *nH O was filtered through paper and the filtrate
2
2
was evaporated and dried in vacuo. M=10.61 g, Yield=76%.
The title compound exists as a mixture of two diastereomers. Diastereomeric ratio was not
determined. The configuration of the double bond is also unknown; however it should probably
be Eꢀconfiguration.
1
H NMR (CDCl , 400 MHz): δ = 1.18 (s, 4 H), 1.22 (s, 5H), 1.25 ꢀ 1.57 (m, 3 H), 1.59 ꢀ 1.91 (m,
3
3
H), 2.17 (d, J=11.5 Hz, 1 H), 2.84 ꢀ 3.10 (m, 1 H), 3.97 ꢀ 4.18 (m, 1 H), 4.43 ꢀ 4.67 (m, 1 H),
.99 ꢀ 5.35 (m, 3 H), 5.83 ꢀ 5.99 (m, 1 H), 7.93 (dd, J=5.8, 1.7 Hz, 1 H).
4
13
C NMR (CDCl , 100 MHz): δ = (20.2, 20.4), (22.5, 22.7) (3C), (24.9, 25.0) (br.), 26.5 (br.), 41.6
3
(br.), 55.9 (br.), 57.0, 66.3, (117.5, 117.9 (br.)), 133.0, 155.9, (168.0, 168.5 (br.)).
Ethyl 4ꢀmethylthiazoleꢀ5ꢀcarboxylate (5)
To a solution of formamide (14.5 mL, 0.364 mol, 1.5 equiv) in 1,4ꢀdioxane (100 mL) P S (17.30
2
5
g, 77.7 mmol, 0.3 equiv) was added in small portions in the course of 5 hours under vigorous
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 71
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
stirring. The reaction mixture was stirred for 2 hours and ethyl 2ꢀchloroacetoacetate (35.4 mL,
0.243 mol) was added dropwise. The reaction mixture was heated at ~60 °C for 8 hours, poured
into cold water (300 mL) and extracted with CH Cl (3x100 mL). Combined organic phases were
2
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
washed with 5% K CO , dried over anhydrous Na SO , evaporated and distilled at reduced
2 3 2 4
pressure. M= 25.0 g, Yield= 60%.
1
H NMR (CDCl , 400 MHz): δ = 1.39 (t, J=7.2 Hz, 3 H), 2.79 (s, 3 H), 4.36 (q, J=7.1 Hz, 2 H),
3
8.84 (s, 1 H).
13
C NMR (CDCl , 100 MHz): δ = 14.2, 17.2, 61.2, 122.2, 155.2, 160.4, 161.9.
3
(
4ꢀmethylthiazolꢀ5ꢀyl)methanol (6)
A solution of 5 (25.0 g, 0.146 mol) in THF (150 mL) was added dropwise to a suspension of
LiAlH (5.55 g, 0.146 mmol) in THF (150 mL). The reaction mixture was stirred for 1 hour and
4
quenched by successive addition of water (6 mL), 10 % NaOH (6 mL) solution and water (12
mL). The precipitate was filtered and washed several times with THF. The filtrate was
evaporated to give 6, which was used without purification. M= 17.90 g, Yield= 95%.
1
H NMR (CDCl , 400 MHz): δ = 2.44 (s, 3 H), 4.83 (s+br.s, 2+1 H), 8.65 (s., 1 H).
3
13
C NMR (CDCl , 100 MHz): δ = 14.5, 55.9, 132.2, 148.7, 151.1.
3
5ꢀ(((tertꢀbutyldimethylsilyl)oxy)methyl)ꢀ4ꢀmethylthiazole (7)
(36.0 g, 0.279 mol) was dissolved in DMF (280 mL), imidazole (28.5 g, 0.419 mol, 1.5 equiv)
6
was added in one portion followed by portionwise addition of TBSCl (50.2 g, 0.335 mol, 1.2
equiv). The reaction mixture was stirred overnight at 50ꢀ60 °C, cooled to room temperature,
diluted with water (0.5 L) and extracted with hexane (3x100 mL). The combined organic phases
were dried over anhydrous Na SO , and evaporated to give an oil which was purified by
2
4
distillation at reduced pressure.
M=61.20 g, Yield = 90%.
1
H NMR (CDCl , 400 MHz): δ = 0.12 (s, 6 H), 0.94 (s, 9 H), 2.42 (s, 3 H), 4.85 (s, 2 H), 8.64 (s,
3
1
H).
24
ACS Paragon Plus Environment

Page 25 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13
C NMR (CDCl , 100 MHz): δ = ꢀ5.22, 15.2, 18.4, 25.9, 58.0, 132.5, 148.0, 150.6.
3
1ꢀ(4ꢀchlorophenyl)ꢀNꢀhydroxyethanimine (8)
Hydroxylamine hydrochloride (50.11 g, 0.72 mol) was added to a solution of 4’ꢀ
chloroacetophenone (85.74 g 0.55 mol) in 140 mL of ethanol. Then with stirring the solution of
NaOH (28.84 g, 0.72 mol) in 140 mL of water (exothermic reaction!) was added in one portion.
The biphasic reaction mixture was refluxed for 3 hours and poured into 1000 mL of ice water.
The precipitate was filtered off and dried thoroughly on a rotary evaporator to afford 91.00 g
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
(
97%) of an offꢀwhite to lightꢀyellow solid (the color depends on a quality of the starting
acetophenone) with m.p. = 98ꢀ99 °C.
1
H NMR: (DMSO, 400 MHz) δ = 2.11 (s, 3H), 7.25 (d, J = 8.6 Hz, 2H), 7.54 (d, J = 8.6 Hz, 2H),
10.93 (br. s, 1H).
2ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrrole (9)
KOH (74.02 g, 1.32 mol) was added to the solution of 1ꢀ(4ꢀchlorophenyl)ꢀNꢀhydroxyethanimine
80.00 g, 0,47 mol) in DMSO (230 mL). Further scale up lead to the significant decrease in yield.
(
The heating was turned on and KOH was fused with a heatgun (T > 100 °C). The reaction
mixture was stirred at 130ꢀ140 °C and the solution of dibromoethane (81.2 mL, 0.94 mol) in
DMSO (80 mL) was slowly added dropwise during the course of 6ꢀ7 hours. Every hour a 30 g
portion of KOH (totally 5 portions) was added and fused with a heatgun to achieve a fine
suspension (otherwise the reaction mixture could not be stirred properly, especially at the last 2ꢀ
3 hours of the dibromoethane addition).
After the addition was finished the reaction mixture was cooled to RT and poured into ~1500 g
of ice water. The resulting dark solution was extracted with benzene (5×300 mL). The first two
extractions were difficult to execute because the phase separation level was hardly noticeable.
The combined organic phases were dried over anhydrous Na SO and the solvent evaporated
2
4
on a rotary evaporator. The residue was distilled under high vacuum and the highꢀboiling
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 71
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
fractions collected. Fractions with b.p. > 130 °C were collected. The product was crystallized in
a wide condenser. The product was melted with a heat gun and washed off in a beaker with
light petroleum ether. The pyrrole was filtered off, washed with petroleum ether and dried on air
to afford 10.60 g (13%) of offꢀyellow solid with strong odor. (m.p. = 140 °C). The product is
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
readily oxidized on air. Thus it should be used in the next step promptly.
1
H NMR: (CDCl , 400 MHz) δ = 6.30 (s, 1Н), 6.51 (s, 1H), 6.86 (s, 1H); 7.33 (d, J = 8.4 Hz, 2H),
3
7.39 (d, J = 8.6 Hz, 2H), 8.38 (br. s, 1H).
13
C NMR (CDCl , 100 MHz): δ = 106.6, 110.5, 119.4, 125.1 (2C), 129.2 (2C), 131.1, 131.4,
3
131.8.
1ꢀ(5ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrrolꢀ2ꢀyl)ꢀ2,2,2ꢀtrifluoroethanone (10)
To the solution containing 2ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrrole 9 (18.20 g, 0.102 mol) and pyridine
(
11.0 ml, 0.13 mol) in 140 ml of diethyl ether trifluoroacetic anhydride (18.9 ml, 0.14 mol) was
added dropwise with an external cold water bath cooling. It should be noted that shortly after the
beginning of the addition the precipitate starts to form. After the addition was complete the
mixture was stirred for 3h and poured into 500 ml of cold water. The resulting precipitate was
filtered off, washed with water and dried on a rotary evaporator to afford 25.23 g (90%) of an offꢀ
white solid with m.p. = 189ꢀ191 °C.
1
H NMR: (DMSO, 400 MHz) δ = 6.92ꢀ6.96 (m, 1H), 7.25ꢀ7.29 (m, 1H), 7.51 (d, J = 6.9 Hz, 2H),
8.00 (d, J = 8.8 Hz, 2H), 12.93 (br. s, 1H).
13
C NMR: (DMSO, 100 MHz) δ = 111.1, 117.1 (q, J=289.8 Hz), 123.3 (q, J=2.9 Hz), 126.2,
128.1 (2C), 128.7, 128.9 (2C), 133.7, 142.3, 167.8 (q, J=35.0 Hz).
5ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrroleꢀ2ꢀcarboxylic acid (11)
The solution of NaOH (2.86 g, 72 mmol) in 25 mL of water was added to the solution of 1ꢀ(5ꢀ(4ꢀ
chlorophenyl)ꢀ1Hꢀpyrrolꢀ2ꢀyl)ꢀ2,2,2ꢀtrifluoroethanone 10 (9.79 g, 36 mmol) in 25 mL of ethanol.
The resulting reaction mixture was refluxed for 6 hours and cooled to RT. Ethanol was removed
on a rotary evaporator. The resulting preꢀcooled solution was carefully acidified to pH 3 with an
26
ACS Paragon Plus Environment

Page 27 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
equivalent quantity of concentrated HCl solution (~ 3 ml). The resulting precipitate was filtered
off, washed successively with cold water and diethyl ether and dried on a rotary evaporator to
afford 6.90 g (87%) of grey solid with m.p. = 171ꢀ172 °C.
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
H NMR: (DMSO, 400 MHz) δ = 6.60 (d, J = 3.5 Hz, 1H), 6.71 (d, J = 3.7 Hz, 1H), 7.39 (d, J =
8.6 Hz, 2H), 7.86 (d, J = 8.6 Hz, 2H), 11.87 (br. s, 1H). CO H missing due to broadening.
2
13
C NMR: (DMSO, 100 MHz) δ = 107.8, 115.3, 126.5 (2С), 126.9, 128.6 (2С), 130.8, 131.1,
134.0, 162.8.
Allyl 2ꢀ((5ꢀ(((tertꢀbutyldimethylsilyl)oxy)methyl)ꢀ4ꢀmethylthiazolꢀ2ꢀyl)(1,1ꢀ
dimethylethylsulfinamido)methyl)piperidineꢀ1ꢀcarboxylate (12)
7 (12.90 g 53.1 mmol, 1.5 equiv) was dissolved in THF (33 mL) and cooled to ꢀ78 °C. Please
note that large excess of 7 is not necessary. It was used in excess because of the limited
quantity of 4. At this temperature nꢀBuLi (2.5 M, 22 mL, 55 mmol, 1.5 equiv) was added
dropwise under the nitrogen atmosphere. The reaction mixture was stirred for 30 minutes at ꢀ78
°
C, and 4 (10.64 g, 35.5 mmol) was added dropwise as a solution in THF (35 mL). The reaction
mixture was slowly (~1 hour) warmed to 0 °C, acidified with acetic acid (3.18 mL, 53 mmol) and
quenched with water (200 mL). The biphasic mixture was extracted with CH Cl (3x100 mL).
2
2
The combined organic phases were dried over anhydrous Na SO , and evaporated to give a
2
4
mixture of compounds 7, 12a and 12b (brown oil) which was separated by a column
chromatography. Eluent: hexanes:EtOAc (1:1) then pure EtOAc.
First fraction: starting material 7 Rf~0.6ꢀ0.7 (EtOAc)
Second fraction (12a): R =0.56 (EtOAc)
f
Third fraction (12b): R =0.39 (EtOAc)
f
Collected: pure 12a 1.98 g, mixed fraction (12a+12b) 8.52 g, pure 12b 5.98 g (85% overall).
Isomers 12a and 12b were used in subsequent steps separately.
N1 (12a):
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
H NMR (CDCl , 400 MHz): δ = 0.10 (s, 6 H), 0.92 (s, 9 H), 1.19 (s, 8 H), 1.25 (s, 1 H), 1.41 ꢀ
3
1
.55 (m, 2 H), 1.58 ꢀ 1.87 (m, 5 H), 2.33 (s, 3 H), 3.01 ꢀ 3.22 (m, 1 H), 4.14 (m, 1 H), 4.42 ꢀ 4.53
m, 1 H), 4.58 ꢀ 4.70 (m, 2 H), 4.79 (s, 2 H), 5.07 (d, J=10.0 Hz, 1 H), 5.24 (d, J=10.4 Hz, 1 H),
5.35 (d, J=16.6 Hz, 1 H), 5.98 (ddt, J=16.9, 11.0, 5.4, 5.4 Hz, 1 H).
(
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
13
C NMR (CDCl , 100 MHz): δ = ꢀ5.2, 14.3, 15.3, 18.3, 19.2, 22.6 (3C), 24.9 (br.), 25.4, 25.9
3
(
3C), 40.2 (br.), 55.9, 58.1, 60.4, 66.7, 117.7, 132.9, 133.5, 147.1. (3 signals at the downfield
regions are missing).
N2 (12b):
1
H NMR (CDCl , 400 MHz): δ = 0.08 (s, 6 H), 0.91 (s, 9 H), 1.23 (s, 6 H), 1.27 (s, 3 H), 1.38 ꢀ
3
1.84 (m, 6 H), 2.09 ꢀ 2.27 (m, 1 H), 2.30 (s, 3H), 2.62 ꢀ 3.05 (m, 1 H), 3.84 ꢀ 4.07 (m, 1 H), 4.45
(br. s., 3 H), 4.73 ꢀ 4.79 (m, 2 H), 4.82ꢀ5.03 (m, 1 H), 5.12 ꢀ 5.25 (m, 2 H), 5.63 ꢀ 5.90 (m, 1 H).
13
C NMR (CDCl , 100 MHz): δ = ꢀ5.3, 15.0, 18.6, 22.5 (3C), 22.8, 24.1, 24.7 (br.), 25.7, 25.75
3
(
3C), 25.8, 40.0, (56.4, 56.6), 56.5, 57.8, 66.1, 117.1, 117.7, 133.0, 147.0, 166.0, 168.2.
Allyl 2ꢀ(amino(5ꢀ(hydroxymethyl)ꢀ4ꢀmethylthiazolꢀ2ꢀyl)methyl)piperidineꢀ1ꢀcarboxylate
13a and 13b)
(
The 1M HClꢀMeOH solution was prepared by dropwise addition of AcCl (2.10 mL, 29.5 mmol) to
methanol (30 mL). The resulting solution was cooled to an ambient temperature and added to
the flask containing 12a (1.9752 g, 3.64 mmol). After the dissolution of 12a, the reaction mixture
was stirred for 1 hour, evaporated (without heating), dissolved in MeOH (50 mL) and solid KOH
(
~1g) was added to the solution. The mixture was stirred for 10 minutes and loaded on silica
gel. Eluting with CH Cl :MeOH (50:1→20:1→5:1) provided the title amine 13a. M=0.9052 g,
2
2
Yield 76%.
The second isomer (13b) was prepared according to the same procedure from 12b. M=0.8544
g, Yield 73%.
R =0.42 CHCl :MeOH (for both amines).
f
3
N1 (13a):
28
ACS Paragon Plus Environment

Page 29 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
H NMR (CDCl , 400 MHz): δ = 1.35 ꢀ 1.54 (m, 3 H), 1.54 ꢀ 1.74 (m, 3 H), , 2.33 (s, 3 H), 2.4ꢀ3.4
3
(
br. s., 3 H) 2.77ꢀ2.98 (br. s., 1 H), 4.05ꢀ4.31 (br. s, 1 H), 4.31ꢀ4.50 (br. s., 1 H), 4.48 ꢀ 4.53 (m, 1
H), 4.62 (d, J=4.4 Hz, 2 H), 4.73 (s, 2 H), 5.22 (dd, J=10.5, 1.2 Hz, 1 H), 5.32 (br. d., J=18.2 Hz,
1 H), 5.96 (ddd, J=22.5, 10.7, 5.4 Hz, 1 H).
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
3
C NMR (CDCl , 100 MHz): δ = 15.0, 19.4, 23.2, (25.2, 25.7) (br.), (39.8, 40.0) (br.), (52.4,
3
5
2.8) (br.), 56.5, (56.7, 57.0) (br.), 66.3, 117.5, 131.7 (br.), 133.1, 148.5 (br.), 156.4, (171.7,
72.2) (br.).
1
N2 (13b):
1
H NMR (CDCl , 400 MHz): δ = 1.40 ꢀ 1.54 (m, 1 H), 1.55 ꢀ 1.71 (m, 4 H), 2.11 ꢀ 2.20 (m, 1 H),
3
2.30 (s, 3 H), 2.34 (br. s., 3 H), 3.03 (t, J=12.8 Hz, 1 H), 3.98 ꢀ 4.07 (m, 1 H), 4.21 ꢀ 4.31 (m, 2
H), 4.32 ꢀ 4.40 (m, 1 H), 4.49 (d, J=10.0 Hz, 1 H), 4.67 (s, 2 H), 5.12 (dd, J=10.5, 1.3 Hz, 1 H),
5
.16 (dd, J=17.3, 1.5 Hz, 1 H), 5.77 (ddd, J=22.6, 10.8, 5.5 Hz 1 H).
13
C NMR (CDCl , 100 MHz): δ = 15.0, 19.1, 25.1, 25.4, 40.4, 52.4, 56.4, 56.8, 66.0, 117.1,
3
1
31.2, 133.2, 148.3, 155.4, 171.9.
Allyl 2ꢀ((5ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrroleꢀ2ꢀcarboxamido)(5ꢀ(hydroxymethyl)ꢀ4ꢀ
methylthiazolꢀ2ꢀyl)methyl)piperidineꢀ1ꢀcarboxylate (14a and 14b)
To a suspension of 5ꢀ(4ꢀchlorophenyl)ꢀ1Hꢀpyrroleꢀ2ꢀcarboxylic acid (556 mg, 2.51 mmol) in
DMF (4.2 mL) DIPEA (0.43 mL, 2.47 mmol) was added followed by HBTU (951 mg, 2.50 mmol).
The resulting solution was stirred for 10 minutes and a solution of 13a (905.2 mg, 2.78 mmol) in
DMF (4.2 mL) was added in one portion. The reaction mixture was stirred for 2 hours and
loaded on silica gel. Eluting with CH Cl :MeOH (50:1) gave 14a. M=1.4751 g, Yield=100%. The
2
2
title compounds contained DMF, DIPEA, (Me N) CO, CH Cl and MeOH as admixtures.
2
2
2
2
14a:
1
H NMR (CDCl , 400 MHz): δ = 1.20 ꢀ 1.60 (m, 7 H), 1.65 (d, J=12.3 Hz, 1 H), 2.29 (s, 3 H), 3.0ꢀ
3
4
.0 (br. s. 1H), 3.97 ꢀ 4.10 (m, 1 H), 4.59 (br. s., 2 H), 4.75 (s+m, 2+1 H), 4.99 ꢀ 5.23 (m, 2 H),
29
ACS Paragon Plus Environment