Integrated strategies for identifying leads that target the ns3 helicase of the hepatitis C virus

Article abstract of DOI:10.1021/jm401432c

Future treatments for individuals infected by the hepatitis C virus (HCV) will likely involve combinations of compounds that inhibit multiple viral targets. The helicase of HCV is an attractive target with no known drug candidates in clinical trials. Here

Full text of DOI:10.1021/jm401432c

Subscriber access provided by DUESSELDORF LIBRARIES
Article
Integrated Strategies for Identifying Leads that
Target the NS3 Helicase of the Hepatitis C Virus
Steven R. LaPlante, Anil Padyana, Asitha Abeywardnae, Pierre R Bonneau,
Mireille Cartier, Rene Colombe, Araz Jakalian, Jessi Jones, Xiang Li, Shuang
Liang, Ginette McKercher, Peter W White, Qiang Zhang, and Steven J. Taylor
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 07 Jan 2014
Downloaded from http://pubs.acs.org on January 8, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Integrated Strategies for Identifying Leads that Target the
NS3 Helicase of the Hepatitis C Virus
1
2
2
1
Steven R. LaPlante*, Anil K. Padyana, Asitha Abeywardane, Pierre Bonneau, Mireille
1
1
1
3
2
Cartier, René Coulombe, Araz Jakalian, Jessi WildesonꢀJones, Xiang Li, Shuang
2
1
1
2
2
Liang, Ginette McKercher, Peter White, Qiang Zhang, Steven J. Taylor
.
1
Departments of Chemistry and Biological Sciences, Boehringer Ingelheim (Canada) Ltd, R&D,
100 Cunard Street, Laval, Québec, H7S 2G5, CANADA.
2
2
3
Department of Medicinal Chemistry, Department of Biotherapeutics, 900 Ridgebury Road,
Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT 06877 USA.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 56
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: Future treatments for individuals infected by the hepatitis C virus (HCV) will
likely involve combinations of compounds that inhibit multiple viral targets. The helicase of
HCV is an attractive target with no known drug candidates in clinical trials. Herein we describe
an integrated strategy for identifying fragment inhibitors using structural and biophysical
techniques. Based on an Xꢀray structure of apo HCV helicase, in silico and bioinformatic
analyses of HCV variants, we identified that one site in particular (labeled 3+4) was the most
conserved and attractive pocket to target for a drug discovery campaign. Compounds from
multiple sources were screened to identify inhibitors/binders to this site, and enzymatic and
biophysical assays (NMR and/or SPR) were used to triage the most promising ligands for
exploration by Xꢀray crystallography. Medicinal chemistry and biophysical evaluations focused
on exploring the most promising lead series. The strategies employed here can have general
utility in drug discovery.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ACS Paragon Plus Environment

Page 3 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Chronic hepatitis C virus (HCV) infection is a serious health problem affecting approximately 2ꢀ
1
3% of the world population. Left untreated, HCV infection can lead to extensive liver damage
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
1b
such as cirrhosis and hepatocarcinomas, ultimately leading to liver failure and/or cancer.
Intense research into treatments for HCV has been ongoing since the discovery of the virus in
2
1989. Until recently, the only therapies available were combinations of nonꢀspecific immune
modulatory agents (pegylated interferon and ribavirin), which have limited efficacy and severe
3
sideꢀeffects particularly against major genotype 1. The first two HCVꢀspecific antivirals,
nd
inhibitors of the viral serine protease NS3, were approved in 2011, and 2 generation protease
4
inhibitors are in late stage development or recently approved. However, to minimize the
emergence of resistance, the protease inhibitors are still coꢀadministered with interferon (IFN)
5
and ribavirin. Results from recent trials suggest that combinations of multiple antivirals
targeting complementary HCV targets, with or without ribavirin, will be needed to remove
6
interferon from standard treatment regimens, particularly for genotype 1 (GTꢀ1) . Recently,
sofosbuvir, which is an NS5B polymerase inhibitor, was recently approved for IFNꢀfree GTꢀ2
and GTꢀ3.
HCV has a small genome encoding five nonstructural proteins, including four distinct
7
enzymatic activities that have been the focus of extensive research for development of therapies.
In addition to the NS3 protease, several classes of inhibitors for the viral RNAꢀdependent RNA
8
polymerase are in lateꢀstage clinical trials. The other two enzymatic activities are
autoproteolysis of the NS2ꢀNS3 junction by a cysteine protease domain of NS2, and ATPꢀ
dependent RNA unwinding activity encoded by the Cꢀterminal twoꢀthirds of the biꢀfunctional
NS3 helicase protein. There are few reports of inhibitors for either of these activities, and to our
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 56
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
knowledge none have advanced to clinical development. NS2ꢀNS3 activity is inherently difficult
9
,10
to assay, and it is not surprising that little progress has been reported with this potential target.
In comparison, helicase activity is relatively straightꢀforward to evaluate, and this target has been
widely screened. However, very few viralꢀspecific inhibitors have been reported, and most have
been found to act via undesirable mechanisms (e.g. nonspecific binding to the nucleic acid
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
1,12, 13f
substrate instead of the protein, or interfering with the assay signal).
Despite this limited
13,14
success, there is an abundance of structural and mechanistic information on this helicase,
no obvious reason why helicase would not be druggable.
and
In this work, we describe our structure based drug discovery efforts to target this essential
HCV enzyme. We started by evaluating the druggability of this protein, and determined that
efforts were to focus exclusively at the most conserved site (labeled as site 3+4) near amino acid
residue W501. Compound screening and subsequent triaging of the hits employed strategic
biophysical methods and specificity counterscreens, which led to nine series of compounds.
Prioritization processes then afforded the indole series which provided potent and attractive leads
based on encouraging structure activity relationship (SAR) trends, stoichiometry of binding,
mechanism of inhibition and favorable solution property behavior.
RESULTS AND DISCUSSION
HCV Helicase Druggability and Analysis of Potential Ligand Binding Pockets. HCV
helicase is known to unwind doubleꢀstranded RNA in a NTPꢀdependent fashion, and this activity
can be stimulated in vitro by the addition of single stranded nucleic acid. Most RNA helicases
are grouped into two major superfamilies based on the occurrence of seven to eight conserved
4
ACS Paragon Plus Environment

Page 5 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
structural motifs which are essential to enzymatic function. Residues in those motifs contribute
to either NTP or RNA binding or couple the energy of NTP hydrolysis to dsRNA unwinding
activity. These conserved motifs are localized within the two Nꢀterminal domains (see Domains
I and II on the Xꢀray structure of HCV NS3 helicase complexed with a singleꢀstranded DNA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
2
octamer in Figure 1).
Figure 1. Solid surface view of an Xꢀray structure of HCV helicase bound to a DNA octamer.
Shown on the left is the full view of the Xꢀray structure of HCV helicase with in silico derived
annotated PASS sites (colored) which predict potential pockets and binding sites. The NTP
binding site corresponds to sites 1 and 2. The zoomed view on the right displays site 3+4 with
the singleꢀstranded DNA octamer (green atoms) superposed onto the structure for visual
32
purposes, PDB code 1A1V.
Based on this Xꢀray structure of the HCV helicase with the DNA substrate removed, we
performed an in silico pocket finding study using the program PASS (Putative Active Site with
33
Spheres) in order to identify potential binding sites. Our in silico PASS analyses identified
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 56
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
eight potential ligand binding sites on the three domains. Figure 1 shows these eight sites which
were mostly located in regions of the conserved helicase motifs and the substrate binding regions.
Having identified sites that could theoretically accommodate small molecule inhibitors,
we proceeded to assess the druggability of these sites, and the homology of these pockets across
genotypes. To evaluate the potential for these issues, two additional dimensions of data were
acquired for each of the residues that lined these sites. An amino acid conservation analysis was
performed by comparing the sequences of 361 HCV genome sequences, and residue flexibility
analysis was performed based on the crystallographic Bꢀfactors. Data for both dimensions are
provided in Figure 2 for the helicase residues located within PASS sites 3 and 4 (vide infra).
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
PASS Site 3
RNA binding site on domain III)
PASS Site 4
(RNA binding on domain I)
(
Residue
All
Genotype 1 Flexibility
only
Residue
All
Genotype 1 Flexibility
Genotypes
Genotypes
only
E493
D496
A497
W501
E555
N556
F557
P558
360/361
360/361
346/361
361/361
149/361
356/361
296/361
308/361
215/216
215/216
214/216
216/216
51/216
+
-
G255
T269
G271
K272
A275
Y502
361/361
359/361
357/361
360/361
361/361
359/361
216/216
215/216
215/216
216/216
216/216
215/216
-
+
-
-
-
+
-
++
+
-
214/216
196/216
216/216
-
-
Figure 2. Residue conservation and flexibility analyses of amino acids that line the PASS ligand
binding sites 3 and 4. The numbers shown indicate the same residues found versus the number
of viral sequences analyzed. Flexibility was based on Xꢀray crystallographic Bꢀfactors.
Residues with higher Bꢀfactors were denoted by a “+ or ++” and suggested higher flexibility,
whereas lower Bꢀfactors were denoted by a “ꢀ“ and suggested lower flexibility.
6
ACS Paragon Plus Environment

Page 7 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Analysis of the data shows that residues in site 4 are highly conserved across genotype
sequences; the same residue is noted for >350 genotypes; out of 361 studied (see the column
labelled All Genotypes in Figure 2). This attractive feature suggested that this site had the
potential for targeting panꢀgenotype activity. More variability was noted for site 3, where the
bioinformatics analysis showed that residue E555 was conserved in only 149 of 361 genotypes
conferred, and exhibited spatial flexibility (see ++ under the Flexibility column, Figure 2) based
on its crystallographic Bꢀfactor. However, other residues in site 3 were highly conserved such as
W501 which was observed in 361/361 of the genome sequences. Similar analyses on other
PASS sites showed significantly more sequence heterogeneity and flexibility (see Supporting
Information). Based on the above analysis, we rationalized that the most favorable opportunities
for identifying a useful small molecule drug was to target exclusively sites 3 and/or 4 (3+4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our campaign then turned to traditional high throughput screening (HTS) as well as
novel fragment based methods as lead finding approaches to target this site. As a first line of
6
approach, we screened our corporate compound collection (>10 compounds) using RNAꢀ
unwinding assays with the NS3 helicase domain and the fullꢀlength NS3/NS4A heterodimer
monitoring helicase activity. These campaigns identified compounds which inhibited the assay
signal, but all were shown to inhibit either by substrate intercalation, interference with the
readout of the assay or have some other undesirable mechanism. Ultimately, no compounds
were identified that had sufficient activity, selectivity, and which also inhibitored by the desired
mechanism. Without any suitable compounds identified from HTS, we transitioned to a
fragmentꢀbased drug discovery (FBDD) campaign.
Fragment Screens – Various Lead Sources and Screening Methods. We then
implemented an integrated FBDD strategy which involved screening compounds from various
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 56
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
sources to maximize opportunities to identify hit candidates to funnel through triaging filters
then subsequent optimization (Figure 3). To maximize the chances of finding leads in FBDD
campaign, we screened fragments from multiple sources. One source of fragments came from
revisiting fragmentꢀlike hits from the earlier HTS campaigns that were rejected due to poor
potency. Other sources included fragments from inꢀhouse screening collections tested in a RNAꢀ
unwinding functional assay (FA) at high compound concentrations (up to 500 ꢀM) to allow the
detection of weakly active binders. Libraries of fragments from commercial sources were also
screened in a surface plasma resonance assay (SPR). In parallel, additional compounds were
identified from virtual screening of inꢀhouse and external vendor compound collections,
followed by functional assays (FA), then verifications by biophysical filtering methods (NMR or
SPR).
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
Our workflow employed stringent triaging and prioritization strategies (using SPR and a
toolbox of NMR methods) to identify quality binders and specific inhibitors of HCV helicase.
An early objective was to prioritize quality ligands that could be funneled to Xꢀray
crystallography studies to acquire atomicꢀlevel views of complexes with NS3 helicase. This and
NMR evidence served to validate ligands that bound to site 3+4 which then triggered and
enabled structureꢀbased drug design, idea generation and analogue verification.
It was noteworthy that, in practice, this strategy was heavily burdened by falseꢀpositive
hits especially given the high concentrations of compounds required for FA testing and for
biophysical binding studies. At these higher concentrations, unusual physical properties and
binding phenomenon were encountered (solubility, aggregation, promiscuity, vide infra) that
adversely affected some biophysical analyses and interpretations. Therefore, specific followꢀup
29
triaging methods were implemented to eliminate compounds with such adverse properties.
8
ACS Paragon Plus Environment

Page 9 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Overall, the functional assays, counter screens and the NMR assays proved most useful for
evaluating and promoting quality hits for maximizing success in crystallization trials.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Screening methods and compound sources used to identify weak binding fragments for
Xꢀray crystallography screens then medicinal chemistry.
Hit and Chemotype Classification: Binders/inhibitors identified from the workflow
described above were then classified according to scaffold commonalities, and related analogues
were sought (inꢀhouse and commercial) then tested. The more promising series were clustered
into nine chemotypes based on the similarities of the core scaffolds as shown in Figure 4.
Compounds from these series were then subjected to soaking or coꢀcrystallization for
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 56
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
confirmation of binding to the desired 3+4 site. No evidence of binding to helicase was noted for
series S6, S7 and S9 in crystallization trials, and therefore these were deprioritized. Compounds
in series S5 were confirmed as binders to helicase, but ligand electron densities were found at
sites other than site 3+4. Five chemotypes (S1ꢀS4 and S8) were identified that were confirmed
to bind to site 3+4, and thirteen examples are shown overlaid in Figure 5. The most notable
feature of this superposition is that the compounds are centralized within a wide groove above
W501. Some compounds hydrogenꢀbond to the backbone of G255 and others appear to partially
enter a lipophilic pocket or interact with polar residues located in the bottom right of the pocket.
Overall, the wide groove and small lipophilic pocket suggests that targeting this site and
optimizing for potency could be challenging. From this point of view, chemotypes S2ꢀS4 were
considered the most promising. However, series S3 and S4 were subsequently deprioritized (see
Supporting Information) and efforts were focused on the S2 indole series as described in detail
below.
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
0
ACS Paragon Plus Environment

Page 11 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Shown are the chemotype series (S1ꢀS9) and inhibitory potencies of fragment hits
classified by common scaffold features.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Shown is an overlay of twelve Xꢀray crystal structures of ligandꢀhelicase complexes
involving site 3+4. The ligands are derived from the five chemotype series (S1ꢀS4, S8).
Series S2 (Indole) - Detection and Explorations. Series S2 was identified using STD
NMR as a primary screen of our inꢀhouse fragment library. The library comprised ~1000
carefullyꢀselected, small fragments (MW < 300) that are highly soluble and triaged to avoid selfꢀ
aggregating compounds. The screen consisted of the preparation of samples that each comprised
a mixture of three fragments at 300 ꢀM each and HCV helicase at 3.5 ꢀM. NMR experiments
1
were then run for each sample where a 1D H NMR spectrum was acquired along with a 1D
1
STDꢀNMR spectrum. The 1D H NMR spectrum provided detection of potential solubility
issues and identification of diagnostic proton resonances for each fragment. The presence of
signals in the STD experiment indicated that one or more of the fragments experienced binding
to helicase. The samples that gave rise to STD signals were then subjected to followedꢀup
deconvolution experiments where single compound studies identified which of the three
1
2
ACS Paragon Plus Environment

Page 13 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
fragments from hit sets bound to helicase. Figure 7B shows the 1D H reference spectrum (red)
for fragment 1, and Figure 7C displays the signals produced in the STD spectrum which
confirms that fragment 1 indeed binds to helicase. Fragment 1 was analogous to similar
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
13c
compounds known in the literature.
1
Figure 7. Identification of indole fragment by STDꢀNMR (A) Fragment binder 1. (B) 1D H
NMR reference spectrum of fragment 1. (C) Fragment 1 was discovered by STDꢀNMR signals
bottom spectrum in blue). The samples for STD experiments consisted of a total volume of 500
L in deuterated buffer containing 20 mM deuterated BisꢀTris, pH 6.5, 30 mM NaCl, and 1 mM
deuterated DTT.
(
ꢀ
NMR binding and competition experiments were then employed to assess binding of
compound 1 specifically to site 3+4. Figure 8A shows that the sharp and intense peak of free 1
1) broadened and shifted upon addition and binding to helicase (2). Addition of the literature
(
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 56
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
13c
compound 2 known to bind to site 3+4 (see Figure 8),
resulted in the displacement of
compound 1 from the binding pocket and resonances resembling those of the original free state.
This was then corroborated by the crystal structure of 1 which showed that it indeed
bound to site 3+4 (Figure 8B). It also revealed that the binding mode of the fragment was
similar to that of 2 where the carboxylic acid hydrogen bonded to G255 and T269 residues. A
comparison of Figure 8B and 8C suggested that a substituent growth vector from the nitrogen
position toward the lipophilic pocket could provide additional interactions and potency. To this
end, multiple analogues of 2 were made but without significant improvements in potency. Before
further evaluating this indole series of compounds, additional NMR studies were performed to
better understand the solution binding mode for control purposes.
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
Figure 8. Characterization of Indole fragments by NMR and Xꢀray crystallography. (A) H
NMR binding and competition experiments ꢀ (1) 1H NMR of a resonance of free compound 1; (2)
Compound 1 in the presence of helicase; Compound 1 in the presence of helicase and compound
2
. (B) Xꢀray structure of compound 2 bound to site 3+4 of helicase. (C) Compound 1 bound to
site 3+4 of helicase.
1
4
ACS Paragon Plus Environment

Page 15 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Direct binding of 2 to helicase was verified by the methylꢀshift NMR assay as shown in
15
Figure 9B. Methyl resonances changed positions upon the addition 2. Also, N TROSY
experiments (Figure 9D) showed that many resonance peaks of apo helicase shifted positions
upon the binding of compound 2 (blue versus red peaks of the complex) and formed a stable
complex. In comparison, the Supporting Information shows an example of an unstable complex
involving compound 3 from the S4 chemotype series. Thus, both NMR experiments clearly
confirmed that compound 2 bound to helicase in solution.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. Characterization of tool compound 2. (A) The primary structure and activity of
1
compound 2. (B) H NMR methyl region of apo helicase (blue) and complex with compound 2
1
9
(red). (C) F NMR spectra, ꢀ (1) resonance of free compound 2. (2 and 3) resonance of
compound 2 after adding helicase. (4 and 5) Resonance of compound 2 after adding helicase
1
15
then d(U)24. (D) Hꢀ N TROSY of apo HCV helicase (blue) and complex with 2 (red, 1:5 ratio
of helicase to ligand).
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 56
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
9
The presence of fluorine atoms on 2 also allowed for F NMR experiments to determine
if 2 was competitive with binding of a substrate, as anticipated for inhibitors of site 3+4 (see
bound nucleic acid octamer in Figure 1). These experiments are shown in Figure 9C where an
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
intense and sharp F NMR resonance was noted for free compound 2 (blue, (1)), which then
broadened and shifted upfield (left) when helicase was incrementally added, (red resonances (2)
and (3)). The shift was reversible, although not completely, when the substrate mimic d(U)24
was subsequently added (see the black resonances (4) and (5)). Thus, these studies were
consistent with the desired binding properties, which improved the attractiveness of the indole
series for further medicinal chemistry efforts.
Exploratory medicinal chemistry efforts focused on the initial hit 1 as the starting
scaffold. Changes to this scaffold by removing the bromine atom from position 5 of compound 1
resulted in an equipotent compound (Table 1, 4, entry 1), and it also fortuitously brought about a
reꢀorientation of the indole. This subsequently allowed Nꢀsubstitutions with better access to the
lipophilic pocket, while also maintaining the carboxylic acid interactions with G255 and T269
amino acid residues (Figure 10A). Addition of a benzyl substituent (Table 1, 5, entry 2) gave a
compound with 2ꢀfold improved potency. This slight improvement could be the result from
favorable πꢀπ stacking with Trp501 or general van der walls interactions. Further substitution at
position 3 of the phenyl ring was then used to target the lipophilic pocket adjacent to Trp501.
The resultant SAR suggested that large lipophilic groups were preferred such that CH < Cl < I <
3
alkynyl (Table 1, entries 3ꢀ6, 6-9), consistent with the lipophilic nature of the pocket. Indeed, the
crystal structure complexes of 7 and 9 showed favorable access to the pocket with increased
1
6
ACS Paragon Plus Environment

Page 17 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
VDW contact with residues T298, L301, C494 and Y501, lining the pocket wall (Figure 10A and
B). Xꢀray coꢀstructures also revealed the flexibility of the polar residue T298 that appears to act
as a gatekeeper at the lip of the pocket (Figure 10). However, substitutions with hydrogen bond
donor/acceptors targeting T298, as in hydroxyl or amine groups, led to losses in potency (Table 1,
10-11, entry 7ꢀ8). Interestingly, incorporation of an electron withdrawing nitro group (Table 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
12-13, entry 9ꢀ10) resulted in improved potency. In the coꢀcrystal structure of 13 (Figure 10C),
neither the methyl ether nor the nitro groups accessed the lipophilic pocket, suggesting that the
increase in potency may result from electronic effects associated with a shift of the phenyl ring
and a more favorable piꢀpi stacking of the inhibitor above W501.
Table 1. In vitro inhibition of HCV Helicase by indole fragment hit and analogs designed to
access the lipophilic pocket in the 3+4 site
O
OH
N
R1
Entry
1
Compound
R1
H
Helicase IC (ꢀM)
LE
50
4
500
0
.25
.26
2
3
5
6
240
85
0
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 56
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
Cl
0
0
.27
.28
4
5
7
8
71
I
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
52
0.28
6
9
31
OH
0
.25
7
8
10
11
150
NH2
NA
>500
O
+
N
O
0.26
9
12
13
46
O
+
N
O
0
.26
10
14
O
a
IC50 values represented as ꢀM, and averaged from a minimum of 2 determinations. Ligand Efficiency (LE)
calculated by ꢀRTln(IC ) / #heavy atoms.
50
1
8
ACS Paragon Plus Environment

Page 19 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Xꢀray coꢀcrystal structures of indole inhibitors targeting lipophilic pocket in the 3+4
site of HCV helicase. (A) Superposition of compound 1 (pink) and 7 (yellow) showing the
rotation of the indole core. (B) Coꢀstructure of compound 9.(C) Coꢀstructure of compound 13
showing the repositioning of the nitrophenyl ring and modulation of the lipohilic pocket size by
conformational change in the gateꢀkeeper residue T298 .
Based on fragment 7, further optimization was then pursued at the 6ꢀposition of the
indole scaffold (Table 2). Halogen substitution such as chlorine and bromine was tolerated
(Table 2, 14ꢀ15, entries 1ꢀ2), and incorporation of phenyl or 5ꢀnitro phenylamine led to a twoꢀ
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 56
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
fold loss in potency (Table 2, 16-17, entries 3ꢀ4). Substituting 3,5ꢀdiamino phenyl targeting
1
3c
W501 and also described in the literature,
led to a significant improvement in potency (Table
2, 18, entry 5). Combining the rightꢀhand side of 13 with 18 led to hybrid molecule 19 (Table 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
entry 6), the most potent compound in this series. The coꢀcrystal structure of 19 (Figure 8) shows
a good overlay of the core and the phenyl diamino group is projected into solvent with one of the
amines donating a hydrogen bond to the backbone carbonyl of Trp501. Interestingly, the 2ꢀ
methoxyꢀ5ꢀnitro benzene moiety of 19 binds closer to the lipophilic pocket compared to 13,
while maintaining the πꢀπ interaction with W501 (Figure 11B). However, the potency gain came
at the expense of installing larger groups as reflected by a lower LE. Nonetheless, it was
encouraging that this substitution rendered compounds 19 to be more specific. The activity
window between the HCV helicase assay and the unrelated helicase A assay was 36ꢀfold.
However, the aniline substituent present on compound 19 would need to be replaced given
mutagenic considerations. In conclusion, these studies suggested that the S2 indole series could
serve as a starting point for further optimization. As a representative of this series, compound 19
was subjected to further biophysical characterization as discussed in the next section.
Table 2. In vitro inhibition of HCV Helicase by elaborated indole analogs.
Helicase IC50
ꢀM)
LE
Entry
Compound Structure
(
O
OH
9
4
0.25
1
14
N
Cl
Cl
2
0
ACS Paragon Plus Environment

Page 21 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
OH
9
4
0.25
2
15
N
Cl
Br
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
OH
1
40
0.19
0.17
3
4
16
17
N
Cl
O
OH
H N
2
140
N
Cl
+
N
O
O
O
OH
3
5
0.21
0.23
5
6
18
19
H N
2
N
Cl
NH2
O
OH
O
N
3
+
H N
2
N
O
NH2
O
a
IC50 values represented as ꢀM, and averaged from a minimum of 2 determinations. Ligand Efficiency (LE)
calculated by ꢀRTln(IC ) / #heavy atoms.
50
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 56
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 11. Binding mode of series S2 inhibitors in site 3+4 of HCV helicase. (A) The electronꢀ
density map of compound 19 bound to HCV helicase shown in stick representation (green), with
molecular interactions depicted as dotted lines. The 2Fo - Fc electronꢀdensity map calculated
using the ligand fit model and is represented as a mesh colored in black at a contour level of 1.0σ.
(B) Superposition of coꢀstructures of compound 13 (magenta) and 19 (green).
2
2
ACS Paragon Plus Environment

Page 23 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Biophysical Characterization and Binding Kinetics for Compound 19 - Kd
Determinations. To determine whether the IC50 values from the unwinding assays were
representative of true binding affinities of inhibitors to HCV helicase, we sought to quantify the
binding of a representative compound, 19, by one or more biophysical methods that did not
require immobilization of the protein. For this purpose, NMR was used to analyze the methyl
region of the helicase. Figure 12B shows two helicase resonances in the methyl signature region
that changed following incremental additions of inhibitor 19. This was plotted in Figure 9A
where the shift changes (Hz) versus compound concentration were fit using Graphpad (oneꢀsite
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
binding), and K ’s of 32 and 28 ꢀM (+/ꢀ 8) were determined. These values differ by ~10ꢀfold
d
from the IC values determined in the enzymatic assay (Table 2). However, NMR spectra of
5
0
helicase in two different buffers (at pH 7.5 and 6.5) showed significant differences in resonance
positions suggesting that structural or dynamic differences existed (Supporting Information).
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 56
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
1
Figure 12. K by NMR for compound 19 binding to unlabeled HCV helicase. The H NMR
d
methyl region of HCV helicase is shown as a function of compound concentration. The
solutions consisted of 23 ꢀM HCV helicase GTꢀ1a, 25 mM TrisꢀHCl (deuterated), pH 7.5, 5 mM
MgCl2, 0.5 mM DTT (deuterated), 3% glycerol (deuterated), 1 mM EDTA (deuterated), 300
mM NaCl, 0.01% IGEPAL, 5% DMSO (deuterated).
Independently, an isothermal titration calorimetry (ITC) experiment also showed binding
of compound 19 to HCV helicase, though with a K value (33 ꢁM) that was also tenꢀfold higher
d
than IC value from the unwinding assay (Figure 13). To achieve sufficient signal in ITC, the
5
0
protein needed to be at very high concentrations, and in this case buffer components and pH had
to be adjusted to achieve the required concentration. We therefore sought to identify a different
biophysical technique which would require lower protein concentrations in two different buffers.
2
4
ACS Paragon Plus Environment

Page 25 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 13. Thermodynamic and structural characterization of inhibitor 19. (A) Determination of
the affinity of inhibitor 19 for the HCV helicase domain by calorimetry. Top: raw data obtained
at 25°C for injections of inhibitor x (300 ꢁM) into the sample cell containing HCV helicase (35
ꢁ
M). Bottom: calculated enthalpies (squares) after subtraction of the heat of dilution as
determined by the later injections. Data were fit to a simple one binding site model to give the
parameters shown in the inset. The association constant K is equivalent to a K value of 33 ꢁM.
d
(
B and C) CD titration of compound 19 into helicase domain. Spectra obtained for solutions of
the HCV helicase (8.6 ꢁM) and 0 (dark blue), 5 ꢁM (maroon), 10 ꢁM (green), 20 ꢁM (purple),
0 ꢁM (turquoise), and 80 ꢁM (orange) compound 19. (B) Represents data obtained in pH 7.5
4
Tris buffer (similar to that used for ITC), (C) Represents data obtained in pH 6.5 PIPES, similar
to that used for unwinding assays. (D) Plot of CD signal change vs. inhibitor concentration data
at pH 7.5 (diamonds) and 6.5 (squares). Each data set was fit to a one site binding model by
nonlinear regression giving K estimates of 41 ± 8 ꢁM (max signal change 3.2 ± 0.3 mdeg) and
d
3.5 ± 0.3 ꢁM (max signal change 4.6 ± 0.1) in pH 7.5 and 6.5 buffers, respectively. As a control,
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 56
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
it was noted that compound 19 had no significant effect on the CD spectrum of BSA (data not
shown).
Unlike ITC’s requirements for high protein concentrations, circular dichroism (CD) (Figure
13B and 13C) does not have such limitations and was used to further characterize the binding of
compound 19 to HCV helicase. The CD spectrum of a protein in the nearꢀUV region 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
1
9
determined by its tertiary structure in the region of aromatic amino acids. It is sensitive to
subtle perturbations in structure, which may be sensitive to ligand binding. Increasing
concentrations of compound 19 mixed with HCV helicase caused a progressive change in the
nearꢀUV CD spectrum as shown in Figures 16B and 16C, for pH 7.5 (ITC/NMR) and pH 6.5
(
assay) buffers, respectively. In fact, the K obtained by this method in the ITC/NMR buffer (41
d
±
8 ꢁM) at pH 7.5 was very similar to that obtained by ITC and NMR, and the K obtained in pH
d
6.5 assay buffer (3.5 ± 0.3 ꢁM) was very close to that obtained in the unwinding assay. Taken
together, the NMR differences noted for helicase in different buffers (Supporting Information)
are consistent with the differences in determined K values by CD. These results help to
d
demonstrated that the IC values measured in enzymatic assays accurately reflected the affinity
5
0
of compounds for the HCV helicase domain.
CONCLUSIONS
Despite the limited success reported so far in identifying specific lead inhibitors of HCV
helicase, our results suggest that this essential antiviral target may be druggable. In this work,
we screened fragments and compounds from multiple sources and evaluated hits by stringent
biophysical and crystallographic methods. Indole compounds from series S2 can be considered
as starting points suitable for further potency improvements and cellꢀculture activity in replicon
2
6
ACS Paragon Plus Environment

Page 27 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
assays. This series clearly binds at the more conserved 3+4 site, but it may be noteworthy that
the pocket is wide and grooveꢀlike and there is an inherent flexibility of the protein in this region.
To determine if the latter influences the intrinsic ligandꢀbinding energy, one may consider
acquiring a better understanding of the complementarity of the conformation and dynamics of
the ligands with the helicase receptor in the free and bound states. It is conceivable that the
indole series is one of the best chemotypes for this complementarity given that numerous
pharmaceutical companies have screened their collections, and indoleꢀbased compounds have
received the most attention.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
1
3c
Chemistry. Compounds 2 and 3 were prepared as previously described. Compounds 1,
4 and 6, were purchased from commercial vendors and used directly without further purification.
Preparations of fully elaborated compounds 5ꢀ7, and 10ꢀ12 are shown in Scheme 1. The indole
acetic acid was alkylated with the corresponding benzyl bromides under basic conditions to
afford elaborated indoles 5ꢀ7 and 13ꢀ12. The synthesis of analogues 8 and 9 are shown in
Scheme 2. Indole ester was alkylated with 2ꢀiodobenzylbrominde to afford the intermediate
iodide. This iodide was saponified to afford the corresponding acid 8. Alternatively prior to
saponification the iodide was subjected to Sonogasheria coupling conditions (cat. Pd, Base), and
TMSꢀacetylene to afford the coupled acetylide 9, that was treated with stoichiometric amounts of
fluoride followed by LiOH hydrolysis to afford acid 9.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 56
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 synthesis of analogs 14 and 15 are shown in Scheme 3. 5ꢀBromor or chloro indole was
coupled with 3ꢀchloro benzylbromide to afford intermediates 12ꢀ13 that were subsequently
saponified to the corresponding acids 14ꢀ15 under basic conditions.
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
The synthesis of analogs 16ꢀ19 are shown in Scheme 3. Intermediate bromide 13 was coupled
with either phenylboronic acid or 3amino,5 nitrophenylboronic acid under Suzuki conditions to
afford intermediated 16 and 17. The resulting esters were saponified under mild basic conditions
to afford 18 and 19. Nitroaryl 19 was further reduced with zinc to afford bisꢀaniline 20.
The synthesis of analog 19 is shown in Scheme 3. Bromoindole was treated with 2ꢀmethoxy,
5ꢀnitrobenzyl bromide under mild basic conditions to afford intermediate 21 that was coupled
under Suzuki conditions with 3,5 diaminophneylboronic acid. The resulting biaryl ester was
saponified under middle basic conditions to afford analog 23.
Chemistry General Remarks. Starting materials were obtained from commercial
1
suppliers and used without further purification unless otherwise stated. H NMR spectra were
recorded on a Bruker UltraShield spectrometer operating at 400 MHz in solvents, as noted.
Proton coupling constants (J) are rounded to the nearest Hz. All coupling constants are reported
in hertz (Hz), and multiplicities are labeled s (singlet), bs, (broad singlet), d (doublet), t (triplet),
q (quartet), dd (doublet of doublets), dt (doublet of triplets), and m (multiplet). All NMR spectra
were referenced to tetramethylsilane (TMS δH 0, δC 0). All solvents were HPLC grade or higher.
The reactions were followed by TLC on precoated Uniplate silica gel plates purchased from
Analtech. The developed plates were visualized using 254 nm UV illumination or by PMA stain.
Flash column chromatography on silica gel was performed on Redi Sep prepacked disposable
silica gel columns using an Isco Combiflash, Biotage SP1 or on traditional gravity columns.
Reactions were carried out under an atmosphere of argon at room temperature, unless otherwise
2
8
ACS Paragon Plus Environment

Page 29 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
noted. Mass spectrometry data were obtained using the Micromass Platform LCZ (flow
injection). Purity for all final compounds was >95%, and purity was evaluated by the following:
System 1: analytical HPLC using a Varian Dynamax SDꢀ200 pump coupled to a Varian
Dynamax UVꢀ1 detector. Solvents: (A) water + 0.05% TFA and (B) acetonitrile + 0.05% TFA,
flow 1.2 mL/min. Column: Vydac RPꢀ18, 5 m, 250 mm × 4.6 mm. Photodiode array detector at
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
220 nm. Gradient: from 95% to 20% solvent A over 25 min.
System 2: HP 1110 Agilent LCMS using a quaternary G1311A pump coupled to a
Micromass Platform LCZ detector. Solvents: (A) water + 0.1% formic acid and (B) acetonitrile +
0.1% formic acid, flow 1.5 mL/min. Photodiode array detector at 190 or 400 nm. (a) Agilent
Zorbax Eclipse XDBꢀC8 5 ꢁm, 4.6 mm × 150 mm column, 4.6 mm × 30 mm, 3.5 ꢁm, from 99%
to 5% solvent A over 10 min or (b) Column Agilent Zorbax C18 SB 3.5 ꢁm, 4.6 mm × 30 mm
cartridge, from 95% to 5% solvent A over 2.5 min.
Scheme 1.
O
1. R1: H
2. R1: Me
3. R1: Cl
O
Br
OH
OH
4. R1: NO2
+
5. R1: OBn
N
R1
6. R1: OH
N
H
R1
7. R1: NH2
(1-Benzyl-1H-indol-3-yl)-acetic acid (1). Sodium hydride (55 mg, 1.4 mmol) was
0
suspended in DMF (4 mL) and to cooled to 0 C in an iceꢀwater bath. Indoleꢀ3ꢀacetic acid (80
0
mg. 0.45 mmol) was added and the resulting slurry stirred at 0 C for 30 minutes. Benzyl
bromide (93 mg, 0.55 mmol) was added and the reaction warmed to room temperature for 1 h.
The reaction mixture was quenched with water, and the solvent evaporated in vacuo. The crude
mixture purified by reverse phase HPLC to afford 1 (25 mg, 21 %).LCMS (ES+) m/z found 266;
2
9
ACS Paragon Plus Environment