Novel HIV-1 capsid-targeting small molecules of the PF74 binding site

Article abstract of DOI:10.1016/j.ejmech.2020.112626

The PF74 binding site in HIV-1 capsid protein (CA) is a compelling antiviral drug target. Although PF74 confers mechanistically distinct antiviral phenotypes by competing against host factors for CA binding, it suffers from prohibitively low metabolic sta

Full text of DOI:10.1016/j.ejmech.2020.112626

Journal Pre-proof
Novel HIV-1 Capsid-Targeting Small Molecules of the PF74 Binding Site
Lei Wang, Mary C. Casey, Sanjeev Kumar V. Vernekar, Rajkumar Lalji Sahani,
Jayakanth Kankanala, Karen A. Kirby, Haijuan Du, Atsuko Hachiya, Huanchun
Zhang, Philip R. Tedbury, Jiashu Xie, Stefan G. Sarafianos, Zhengqiang Wang
PII:
S0223-5234(20)30598-5
DOI:
https://doi.org/10.1016/j.ejmech.2020.112626
EJMECH 112626
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 5 May 2020
Revised Date: 22 June 2020
Accepted Date: 25 June 2020
Please cite this article as: L. Wang, M.C. Casey, S.K.V. Vernekar, R.L. Sahani, J. Kankanala, K.A. Kirby,
H. Du, A. Hachiya, H. Zhang, P.R. Tedbury, J. Xie, S.G. Sarafianos, Z. Wang, Novel HIV-1 Capsid-
Targeting Small Molecules of the PF74 Binding Site, European Journal of Medicinal Chemistry, https://
doi.org/10.1016/j.ejmech.2020.112626.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2020 Elsevier Masson SAS. All rights reserved.

Graphic abstract

Graphic abstract
Highlights
•
•
•
•
Design of novel analogs and sub-chemotypes of HIV-1 CA-targeting antiviral PF74.
5-Hydroxyindole analogs (8,9 and 12) showed much improved potency over PF74
2-Indolone analogs (16-24) decreased the T of CA hexamers
m
The potencies of α- and β-naphthyl analogs (33 and 27) were comparable to PF74

Novel HIV-1 Capsid-Targeting Small Molecules of the PF74
Binding Site
a, §
b
a
a
Lei Wang , Mary C. Casey , Sanjeev Kumar V. Vernekar , Rajkumar Lalji Sahani ,
a
c
c
d
Jayakanth Kankanala , Karen A. Kirby , Haijuan Du , Atsuko Hachiya , Huanchun
c
c
a
c
* , a
Zhang , Philip R. Tedbury , Jiashu Xie , Stefan G. Sarafianos , and Zhengqiang Wang
a Center for Drug Design, College of Pharmacy, University of Minnesota, Minneapolis, MN 55455, USA
b Department of Molecular Microbiology and Immunology, University of Missouri School of Medicine, Christopher S. Bond Life
Sciences Center, Columbia, MO 65211, USA
c Laboratory of Biochemical Pharmacology, Department of Pediatrics, Emory University School of Medicine, Atlanta, GA
3
0322, USA
d Clinical Research Center, Nagoya Medical Center, National Hospital Organization, Nagoya, Aichi 460-0001, Japan
*
Corresponding Author
Zhengqiang Wang, Ph.D. Email: wangx472@umn.edu; Phone: +1 (612) 626-7025.
§
Present address: Department of Pharmacy, School of Chemical Engineering, Dalian
University of Technology, Dalian 116024, China
Abstract: The PF74 binding site in HIV-1 capsid protein (CA) is a compelling antiviral
drug target. Although PF74 confers mechanistically distinct antiviral phenotypes by

competing against host factors for CA binding, it suffers from prohibitively low metabolic
stability. Therefore, there has been increasing interest in designing novel sub-chemotypes of
PF74 with similar binding mode and improved metabolic stability. We report herein our
efforts to explore the inter-domain interacting indole moiety for designing novel
CA-targeting small molecules. Our design includes simple substitution on the indole ring,
and more importantly, novel sub-chemotypes with the indole moiety replaced with a few
less electron-rich rings. All 56 novel analogs were synthesized and evaluated for antiviral
activity, cytotoxicity, and impact on CA hexamer stability. Selected analogs were tested for
metabolic stability in liver microsomes. Molecular modeling was performed to verify
compound binding to the PF74 site. In the end, 5-hydroxyindole analogs (8,9 and 12)
showed improved potency (up to 20-fold) over PF74. Of the novel sub-chemotypes, α- and
β-naphthyl analogs (33 and 27) exhibited sub micromolar antiviral potencies comparable to
that of PF74. Interestingly, although only moderately inhibiting HIV-1 (single-digit
micromolar EC s), analogs of the 2-indolone sub-chemotype consistently lowered the
5
0
melting point (T ) of CA hexamers, some with improved metabolic stability over PF74.
m
Keywords: HIV-1; capsid-targeting antivirals; PF74; metabolic stability
1. Introduction
Although dozens of antivirals and fixed dose combinations [1] have been approved to
treat the infection of human immunodeficiency virus type 1 (HIV-1), these drugs are not
curative and HIV-1 remains a global healthcare challenge. Managing HIV-1 infection
requires lifelong treatment, and hence, the virus will eventually develop resistance to current
drug classes. Combating resistant viruses entails antivirals with novel molecular targets. The
multifunctional HIV-1 capsid protein (CA) [2-4], which is a key component of the HIV-1 gag
polyprotein [5], represents an emerging and highly attractive target in HIV-1 drug discovery

[
6-8]. CA is the building block of the mature HIV-1 capsid core [9, 10], and core stability
critically depends on CA-CA interactions. Disrupting these interactions in the early stages of
viral replication can perturb core stability to result in premature uncoating, impaired reverse
transcription, and loss of infection [11, 12]. In addition, CA also interacts with multiple
cellular factors [13] including TRIM5α [14, 15], cleavage and polyadenylation specific
factor 6 (CPSF6) [16, 17], nucleoporins 153 [18-20] and 358 [21, 22] (NUP153, NUP358),
MxB [23, 24], and Cyclophilin A (CypA) [25-27]. These CA-host interactions regulate
multiple post entry events, such as uncoating, cytoplasmic trafficking, reverse transcription,
nuclear transport, integration site distribution, and the evasion of innate immunity [28].
During the late stage of viral replication, CA-CA interactions also drive the assembly and
maturation of new infectious viral particles [29]. Therefore, CA-targeting small molecules
could confer both early and late stage antiviral phenotypes.

Figure 1. Compound binding sites and chemotypes targeting HIV-1 CA. (A) Ligand binding
sites. Three distinct small molecule binding sites at CANTD are known for C-4, PF74 and
BM-4, respectively. Host factor cyclophilin A binds to the loop on top. Polypeptide CAI
binds to the CACTD. Binding modes were reproduced in Maestro based on PDB 4XFZ [30]
with ligands C-4 & BM-4 (PDB: 4E92 [31]) and CAI (PDB: 2BUO [32]) aligned. The
picture was created in PyMOL while shadows of binding sites were rendered in PowerPoint;
(B) major chemotypes binding to each site: BI-2 and GS-CA1 bind to the PF74 site; BD-3
and CAP-1 bind to the BM-4 site; arylquinazoline (AQ) and ebselen bind to CACTD
.
Efforts targeting HIV-1 CA have identified a few chemotypes with distinct binding sites
Figure 1) [6, 7, 33]. CA is highly helical consisting of an N terminal domain (CANTD) and a
(
C terminal domain (CACTD) with a flexible linker in-between [30, 34, 35]. Although
CACTD-targeting compounds, such as polypeptide CAI [32], arylquinazoline (AQ) [36], and
the covalently binding ebselen [37], have been reported, considerably more efforts have been
directed toward targeting CANTD. At the base of the CANTD is the binding site for three
chemical classes represented by BM-4 [31], BD-3 [31] and CAP-1 [38]. Interestingly, the
backbone of BM-4 is a benzoimidazole core flanked by a pyrazole ring and a phenyl ring
(
Figure 1B). However, when the flanking moieties are reversed, the resulting compound C-4
39] binds to a completely different site near the top of the CANTD and around the base of the
[
cyclophilin A binding loop [40] (Figure 1A). In general, compounds binding to these two
sites perturbed CA assembly in vitro and moderately inhibited HIV-1 in cell culture [6].
There was no evidence that they act in the early steps of viral replication, and no direct
evidence linking the antiviral activity to the targeting of CA. By contrast, by far the most
interesting CA-targeting compounds are the three chemical classes that bind to the PF74

binding site, which also include the BI compounds [41] and the GS-CA compounds [42]
Figure 1B). These compounds all inhibited HIV-1 at both the early and the late stages of
(
viral replication. Not surprisingly, there appears to be a positive correlation between the
antiviral potency and the structural complexity: the simplest BI compounds exhibited
moderate (low micromolar) antiviral activity and the structurally highly complex GS-CA
compounds demonstrated sub-nanomolar antiviral activity, whereas the structurally
elaborate, yet highly synthetically accessible peptidomimetic PF74 inhibited HIV-1 in the
sub-micromolar range. The binding mode [30] and the activity profile [43, 44] of PF74 have
been particularly well-characterized, revealing a dual antiviral mechanism of action: at low
concentrations PF74 competes against host factors for capsid binding; at high concentrations
it induces premature uncoating, and consequently, impairs reverse transcription [11].
Figure 2. Binding mode of PF74 based on PDB 4XFZ [30] and the design of novel analogs.
(A) Detailed molecular interaction network of PF74 (magenta sticks) with residues (lines) in
both the CANTD (transparent cyan surface) and the adjacent CACTD (blue dot). The indole

moiety of PF74 (boxed) interacts with Q63, K70 of CANTD and Y169, R173, and K182 of the
adjacent CACTD; (B) Novel analogs designed to explore the two-domain interactions of the
indole moiety.
The PF74 binding pocket is lined with residues in H3 and H4 of the CANTD (cyan) and H8
and H9 of the adjacent CACTD (blue), where an extensive network of molecular interactions
defines the PF74 binding mode (Figure 2A) [30]. Structurally, PF74 features a phenylalanine
core, connected by an aniline moiety at the carboxylate end and an indole-3-acetic acid at the
amino end (Figure 2B). Although all three components provide key molecular interactions
for CA binding, the indole ring (boxed, Figure 2A) uniquely interacts with both the CANTD
(via H-bond with Q63 and π-cation interaction with K70) and the adjacent CACTD (via Y169,
R173 and K182) [30]. These interactions are lacking with the BI compounds, which do not
have a structural equivalent to the indole moiety of PF74, likely accounting for their weaker
antiviral potency. The core and the aniline moiety of PF74 engage extensively with the
CANTD [30] (Figure 2A): 1) the phenylalanine core forms two H-bonds with N57, and an
additional one with K70; 2) the aniline moiety makes contact with N53 (via the N-methyl
group) and A105, T107, and Y130 (via the phenyl ring); and 3) the phenyl ring of the
phenylalanine core forms hydrophobic interactions with residues M66 and L69. Since the
indole moiety provides the unique inter-domain binding interactions, there have been
reported efforts replacing the indole moiety with a simpler and synthetically more accessible
1,2,3-triazole ring, though the replacement resulted in substantially decreased potency (by
>10-fold) [45, 46]. However, a recent report showed that replacing the indole ring with
a
piperazinone moiety yielded compounds with better antiviral activity (up to 6-fold ) than
PF-74 [47]. We have previously conducted a comprehensive SAR on PF74 with the
synthesis of a large number of analogs [48], and have identified a structurally novel and
metabolically stable CA-targeting small molecule [49]. In the current report, we describe our

own efforts targeting the two-domain interactions (Figure 2B). We first synthesized a series
of PF74 analogs with a substituent on the indole ring, and then designed and synthesized
analogs of a few novel sub-chemotypes, each featuring a distinct ring system to replace the
indole. In the end, 56 analogs were synthesized and tested, many of which demonstrated
significant anti-HIV-1 activity and interesting effects on CA hexamer stability, and some
exhibited improved metabolic stability over PF74.
2. Results and discussion
2.1 Chemistry
Briefly, commercially available (tert-butoxycarbonyl)-L-phenylalanine (58) was
treated with various amines under a well-established method using T P or HATU as the
3
coupling agent in the presence of DIPEA as base to afford 59. After removal of Boc
protecting group using TFA, amine salts 60 were obtained which were further reacted with
commercially available acids 61 to produce analogs 1-57. Details about the synthesis of 59
and 60 were described in the Supporting Information.
Scheme 1. Synthesis of novel PF74 analogs.

Reagents and conditions: (a) amine, HATU (or T P), DIPEA, DMF, rt, 12 h; (b) TFA; DCM,
3
rt, 4-6 h; (c) HATU, DIPEA, DMF, rt, 12 h.
2.2 Biological assays and SAR
All analogs were first evaluated in a thermal shift assay, where the effect of a compound on
protein stability was measured by the change in protein melting point compared to DMSO
control (ꢀTm). A positive value in ꢀTm indicates a stabilizing effect and a negative value a
destabilizing effect on the protein. Of note, the CA protein used in the thermal shift assay is in
a covalently crosslinked hexameric state. Thus, the ꢀTm values likely reflect local changes
that may affect stabilization and exclude inter-hexamer effects, which are important
correlates of overall capsid core stability. To simplify presentation of the data, we refer to the
effects of these compounds as stabilization or destabilization of “CA hexamer.” All
compounds were also screened at 20 µM in a cell-based antiviral assay against HIV-1.
Compounds demonstrating significant inhibition were then tested at 2 µM. Promising
compounds were further assessed in dose response fashion for antiviral EC values. All
50
compounds were also tested for cytotoxicity either by screening at 100 or 50 µM, or
determination of CC values. For some compounds, % inhibition at two concentrations (2
50
µM and 20 µM) was reported instead of the EC . PF74 was resynthesized and tested in these
5
0
assays (1, ꢀTm = 7.4 °C, EC = 0.61 µM, CC = 76 µM).
50
50
1
2
3
2
.2.1. Substitutions at the indole (R and R ) and aniline ring (R )
1
This series features PF74 analogs with the indole ring substituted (R ) with an OMe (3-7) or
2
2
an OH (8-15). It is noteworthy that all these analogs lack the methyl group at R (R = H).
2
This is because when compared to PF74 (R = Me, ꢀTm = 7.4 °C, EC = 0.61 µM) analog 2
50
2
(
R = H, ꢀTm = 6.1 °C, EC = 0.46 µM) showed a largely comparable activity profile.
50
Hence, the SAR within this series concerns mainly the effect of the H-bond enabling and

1
electron donating groups at R , in combination with the substitution effects on the aniline
3
1
ring (R ). Notably, an OMe substitution at R by itself did not improve the activity profile (3
3
vs 2), though slight improvement was observed when R is a para-Cl. This chlorine effect
3
was significant as revealed by the direct comparison between analogs 7 (R = 4-Cl, ꢀTm =
3
8
.0 °C, EC = 0.31 µM) and 3 (R = H, ꢀTm = 4.9 °C, EC = 0.56 µM), possibly due to a
5
0
50
halogen bond (see molecular modeling for similar analogs). Among the compounds, 8-15,
1
substituted at R with OH, were found the most potent ones of the series. Particularly,
analogs 8 (EC = 0.053 ꢁM), 9 (EC = 0.035 ꢁM) and 12 (EC = 0.032 ꢁM) conferred up to
5
0
50
50
~
20-fold higher potency than PF74 (EC = 0.61 ꢁM) and analog 2 (EC = 0.46 ꢁM).
50 50
Furthermore, 9 (CC > 100 ꢁM) and 12 (CC > 100 ꢁM) showed lower toxicity than PF74
5
0
50
1
(
CC50 = 76 ꢁM). These results suggest that an R substituent capable of both H-bond
donating and H-bond accepting (e.g. OH) confers better antiviral potency than does an
H-bond accepting group (e.g. OMe). However, the significantly improved antiviral potency
with 8, 9 and 12 over PF74 is not correlated with results from the thermal shift assay, where
o
o
very similar ꢀTm (6.3-6.8 C for 8, 9 , 12 vs 7.4 C for PF74) was observed. It is possible that
the afore-mentioned OH group could affect capsid assembly via unknown post-binding
molecular mechanisms. In addition, PF74 is known to inhibit both the early and the late
stages of HIV-1 replication cycle. Mechanistic studies are currently underway to determine if
the most potent compounds (e.g. 8, 9, 12) from this series display similar antiviral profile.
Table 1. Anti-HIV-1 activity, cytotoxicity, and CA hexamer stability profiles of PF74
1
2
3
analogs (modifications at R , R , R ).

c
CC50
TSA
1
2
3
a
Compound
R
R
R
EC50 (µM)
b
(µ
M)
ꢀTm (°C)
7.4
PF74 (1)
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.61 ± 0.2
0.46 ± 0.1
76 ± 9
> 100
48 ± 5
> 50
2
3
4
5
6
7
8
9
H
6.1
OMe
OMe
OMe
OMe
OMe
OH
H
0.56 ± 0.3
4.9
4-Me
4-F
3-F
4-Cl
H
0.28 ± 0.02
0.57 ± 0.02
0.99 ± 0.002
0.31 ± 0.02
0.053 ± 0.02
0.035 ± 0.004
0.46 ± 0.03
0.32 ± 0.02
0.032 ± 0.01
0.31 ± 0.01
0.38 ± 0.1
5.4
> 50
6.6
> 50
4.2
> 50
8.0
55 ± 2
> 100
> 50
6.8
OH
4-Me
4-F
3-F
4-Cl
3-Cl
3-Br
6.3
10
11
12
13
14
15
OH
3.1
OH
47 ± 6
> 100
> 100
> 100
65 ± 18
4.4
OH
6.6
OH
6.9
OH
5.8
OH
3-CF
0.48 ± 0.02
6.8
3
a
Concentration of compound inhibiting HIV-1 replication by 50%, expressed as the
mean ± standard deviation from at least two independent experiments.
b
Concentration of compound causing 50% cell death, expressed as the mean ± standard
deviation from at least two independent experiments.

c
TSA: thermal shift assay. ꢀTm: change of CA hexamer melting point in presence of
compound compared to DMSO control.
5
2
.2.2. Novel sub-chemotypes with indole bioisosteres (R ) and modifications at the aniline
3
4
ring (R and R )
Table 2 summarizes our SAR studies on a few novel sub-chemotypes featuring different ring
structures as the indole replacement. The first new sub-chemotype includes nine analogs
(16-24) in which the indole moiety is replaced by an indolin-2-one ring. The first four
analogs (16-19) bear a 7-methyl group and the next five (20-24) are 4,7-dimethyl substituted.
Two prominent SAR trends were observed: the additional 4-methyl group (20-24) led to
improved antiviral potency (20 vs 16, 22 vs 19, 23 vs 17); and both 3-Cl and 4-Cl on the
3
aniline moiety (R ) also enhanced potency (18 vs 16, 24 / 23 vs 20). A beneficial chlorine
3
effect conferred by both a 3-Cl and a 4-Cl of the aniline moiety (R ) was observed with this
sub-chemotype only (see molecular modeling). With PF74 and all other sub-chemotypes,
only the 4-Cl is properly positioned for halogen bonding. Overall, analogs of this series
exhibited low micromolar potencies, and more intriguingly, seem to destabilize the CA
hexamer. The next series consists of twelve analogs (25-36) with the indole moiety replaced
by a naphthyl ring, either via a β substitution (25-30) or an α substitution (31-36). The
substitution site (β vs α) did not appear to impact antiviral activity (32 vs 26, 33 vs 27) as
3
prominently as the R group, where a 4-Me (26) or 4-Cl (27) conferred submicromolar
3
activities, though in the α series (31-36) two additional compounds (31 and 36, R = H) also
demonstrated low micromolar antiviral activity. Next we synthesized seven analogs (37-43)
5
where R is a 1,3-dimethyl-purine-2,6-dione ring system. Unfortunately, none of them
inhibited HIV-1 significantly at 2 µM. Discernible impact on CA hexamer stability was not
observed either. The last two sub-chemotypes (44-57) synthesized both feature a [6,5] spiro
ring system. Overall, analogs of these two sub-chemotypes did not significantly impact CA

hexamer stability. As for antiviral potency, while many of the bicyclic analogs (44-53)
inhibited HIV-1, such as compounds 44, 46, 47, 50, and 53, the four tricylic analogs (54-57)
did not show any antiviral activity.
Table 2. Anti-HIV-1 activity, cytotoxicity, and CA hexamer stability profiles of novel
3
4
5
analogs (R , R , R ).
c
inhibition
TSA
a
b
Compd
Code
EC50
CC50
5
4
3
R
R
R
% at 2 µM
ꢀTm
(µM)
(µM)
/
20 µM
(°C)
7
6 ±
9
Me
H
91/98
0.61 ± 0.2
7.4
PF74 (1)
Me
Me
Me
Me
H
0/91
ND
ND
> 50
~ 50
-1.2
-0.4
-1.7
-0.9
16
17
18
19
4-Cl
3-Cl
3-F
17/96
d
14/85
0/93
3.3 ± 0.3 > 100
ND
> 50
d
Me
H
0/72
4.9 ± 0.3 > 100
-1.8
2
0

d
d
Me
Me
Me
4-F
3-F
0/70
0/70
5.4 ± 0.5 > 100
5.2 ± 0.4 60 ± 6
-1.7
-1.8
-1.3
-2.4
2
2
2
2
1
2
3
4
6
3 ±
d
4-Cl
10/98
27/99
2.8 ± 0.1
1
3
d
Me
Me
3-Cl
H
2.1 ± 0.2 43 ± 2
40/98
80/99
ND
< 50
< 50
4.8
6.4
2
5
6
0
.99 ±
Me 4-Me
2
0
.005
0
.63 ±
Me
4-Cl
95/100
< 50
7.0
2
7
0
.02
Me
Me
Et
3-Cl
3-F
H
0/94
0/97
ND
ND
ND
< 50
< 50
< 50
4.3
5.2
4.3
28
29
30
46/97
Me
H
66/100
96/100
1.1 ± 0.04 < 50
6.0
5.7
3
1
2
Me 4-Me
1.0 ± 0.2
< 50
3

0
.83 ±
.03
Me
4-Cl
98/100
< 50
7.2
3
3
0
Me
Me
Et
3-Cl
3-F
H
53/96
44/97
53/98
ND
ND
< 50
< 50
4.4
5.4
4.9
34
35
36
1.8 ± 0.09 < 50
Me
H
10/94
18/98
14/85
1/67
ND
ND
ND
ND
ND
ND
> 50
> 50
> 50
> 50
> 50
> 50
0.9
1.6
0
37
38
39
40
41
42
Me 4-Me
Me
Me
Me
Me
4-F
3-F
0.5
1.5
0
4-Cl
3-Cl
6/96
0/71

Et
H
0/94
ND
> 50
0.8
4
3
Me
H
H
H
-/89
-/0
8.9 ± 1.5
>20
>100
>100
>100
>100
> 50
> 50
>100
< 50
> 50
>100
1.3
0
44
45
46
47
48
49
50
51
52
53
Me 4-OMe
Me 4-Me
-/93
-/92
9/99
24/98
-/96
9/98
-/36
-/92
5.1 ± 2
1.9 ± 0.5
ND
1.8
2.7
3.1
1.3
2.2
3.3
1.5
2.0
Me
Me
Me
Me
Me
Et
4-F
3-F
4-Cl
3-Cl
3-Br
H
ND
2.5 ± 0.5
ND
> 20
7.6 ± 0.9
Me
H
-/32
> 20
> 100
0
5
4

H
H
-/22
-/20
-/41
> 20
> 20
> 20
> 100
> 100
> 100
0.6
0.8
0.8
55
56
57
Me 4-OMe
Me 4-Me
a
Concentration of compound inhibiting HIV-1 replication by 50%, expressed as the mean ±
standard deviation from at least two independent experiments.
b
Concentration of compound causing 50% cell death, expressed as the mean ± standard
deviation from at least two independent experiments.
c
TSA: thermal shift assay. ꢀTm: change of CA hexamer melting point in presence of
compound compared to DMSO control.
d
For these compounds, % inhibition is reported at 1 µM / 10 µM.
2.3. Metabolic stability in liver microsomes
Despite its potent antiviral activity, well-characterized mode of binding and easy
synthetic accessibility, PF74 is seriously flawed as an antiviral lead due to its
prohibitively poor metabolic stability. In human liver microsomes (HLMs), the half-life
(
t ) of PF74 was less than 1 min [46, 50]. We have previously tested around 20 PF74
1/2
analogs in HLMs and they all showed very poor metabolic stability (t = 0.6-2.1 min)
1
/2
[
48]. However, an analog featuring a quinazoline-2,4-dione as the indole replacement
was found to be metabolically stable in HLMs (t1/2 = 31 min) [49]. To gauge the
metabolic stability of the sub-chemotypes described herein, we tested selected
compounds in HLMs and mouse liver microsomes (MLMs) and the results are

summarized in Table 3. Consistent with literature reports, the t1/2 of PF74 in our
metabolic assays was 0.7 min in HLMs and 0.6 min in MLMs. Peptidomimetics, a
compound class to which PF74 belongs, are particularly liable toward phase I
metabolism, presumably because they are good substrates for liver metabolizing enzyme
subfamily cytochrome P450 3A (CYP3A) [51]. This extensive liver metabolism
constitutes a major pharmacokinetic (PK) barrier for many FDA-approved HIV-1
protease inhibitors [52]. For example, darunavir was reported to have a half-life (t ) of
1
/2
3.6 min in a human liver microsomal incubation system with less CYP protein content
compared to ours (0.1 mg/mL vs 0.5 mg/mL) [53]. The main mitigating strategy has
been to co-administer the antiviral drugs with a CYP3A inhibitor, such as ritonavir [54]
or Cobicistat (Cobi) [55], as a PK enhancer [52]. Therefore, our metabolic stability
assays were performed under two distinct sets of conditions: with PK enhancer Cobi or
without Cobi (Table 3). Without Cobi, PF74 analogs (3 and 7) exhibited very short
half-life (1.2 min and 1.8 min, respectively). The metabolic stability remained poor for
analogs of novel sub-chemotypes, though a few of them, such as 16, 24, 37 and 46, did
show substantially (7 to 10-fold) improved half-life over PF74 in HLMs. For the five
selected compounds tested in the presence of PK enhancer Cobi, drastically improved
metabolic stability was observed with all (Table 3). Collectively, these results
demonstrated that replacement of the indole moiety with a less electron-rich ring could
lead to metabolically more stable compounds, and that inhibiting CYP3A could mitigate
the phase I metabolic liability of this type of compounds.
Table 3. Phase I metabolic stability in liver microsomes t (min).
1
/2
a
HLM (+Cobi
b
b
c
Compound
a
MLM
MLM (+Cobi )
HLM
c
)

c
PF74
0.7
1.2
1.8
7.0
1.5
6.7
1.8
2.7
0.6
1.1
4.7
5.2
3.3
1.1
2.5
91
--
0.6
0.5
0.6
1
34
--
3
7
--
--
1
2
2
2
2
3
3
3
4
4
4
5
6
0
4
5
7
1
3
7
6
7
8
0
--
--
--
0.6
1.4
0.5
1.4
0.5
0.6
1.2
0.7
1.2
0.5
1.1
--
107
--
34
--
>120
--
18
--
102
--
7.4
--
--
--
--
--
--
--
>120
20
a
b
c
HLM: human liver microsome
MLM: mouse liver microsome
Microsomal stability measured in the presence of CYP3A inhibitor Cobi.
2.4 Molecular modeling
To evaluate the binding mode of the new analogs, we performed molecular modeling of key
compounds based on the co-crystal structure of native HIV-1 capsid protein bound to PF74
(PDB code: 4XFZ [30]). Analog 12 is found to orient in a similar fashion as the parent
compound PF74 (Figure 3A), and share common key interactions on CANTD domain
including (i) hydrogen bonding with N57 via the carbonyl and NH of phenylalanine core,
with Q63 via indole N-H, and with K70 via the other carbonyl moiety; (ii) Cation-π

interaction with K70 via the indole moiety (Figure 3A). However, unlike PF74, the two
additional substituents of 12, 4-Cl on the aniline moiety and 5-OH of the indole ring, confer
a halogen bond (pink dashed line) and another hydrogen bond with K182 on the adjacent
CACTD domain (black dotted line), respectively (Figure 3A). These additional interactions
are consistent with improved target binding affinity (Glide score: -7.6 kcal/mol for 12 vs
-
5.8 kcal/mol for PF74) and largely increased potency (EC : 0.032 µM for 12 vs 0.61 µM
50
for PF74). Importantly, while the halogen bond conferred by the 4-Cl of the aniline moiety
appeared to be less important for potency than the hydrogen bond by the 5-OH (2 vs 8 vs
1
2), moving the Cl from 4 position to 3 (compound 13) led to a 10-fold potency drop, likely
due to the potential steric clash posed by the 3-Cl. Similar effects were observed with 3-Br
14) and 3-CF (15). The potency difference between 3-Cl and 4-Cl aniline analog was
(
3
prominent throughout SAR studies (27 vs 28, 33 vs 34, and 50 vs 51). The only exception
is with the 4,7-dimethylindolin-2-one sub-chemotype where a beneficial halogen effect was
observed for both the 4-Cl analog 23 (Figure 3B) and the 3-Cl analog 24 (Figure 3C). In
both cases, the loss of the key H-bonding with Q63 and the cation-π interactions with K70,
was compensated by a halogen-bond (pink dashed line) with N74 (4-Cl) and N53 (3-Cl),
respectively. Similarly, compound 27 having the 2-naphthyl ring in place of indole ring
exhibited an identical potency and protein binding affinity to PF74, despite the loss of a
key hydrogen-bonding with Q63 (Figure 3D). Possible orientation of the aniline ring of the
compound 27 as shown in Figure 3D could improve hydrophobic contacts with N53, G106,
and Y130, as a result of halogen bond between G106 and 4-Cl aniline, and could explain
the observed potency for this and another similar compound 33.

Figure 3. Docking poses of key compounds based on native HIV-1 capsid protein bound to
PF74 (Glide score: -5.8) (PDB code: 4XFZ [30]). (A) Predicted binding mode of
compound 12 (Glide score: -7.6 kcal/mol). (B) Predicted binding mode of 23 (Glide score:
-6.4 kcal/mol). (C) Predicted binding mode of 24 (Glide score: -6.8 kcal/mol). (D)
Predicted binding mode of 27 (Glide score: -6.1 kcal/mol). Hydrogen-bonding,
halogen-bonding and cation-π interactions are depicted as black dotted lines, pink dashed
lines, and double headed arrows, respectively. CA with key residues around the binding
site, and ligands 12, 23, 24, and 27 are colored grey and violet, respectively. The nitrogen,
oxygen, and chlorine atoms are colored blue, red, and green, respectively.
3. Conclusions

In this work, we have designed, synthesized and evaluated 56 compounds consisting of PF74
analogs and a few novel PF74-like chemotypes featuring rings less electron-rich than the
indole ring. In the end, 5-hydroxyindole analogs (8, 9, 12) were up to 20-fold more potent
than the prototype PF74. Analogs with low to sub micromolar antiviral activities were also
identified from all sub-chemotypes, except for the 1,3-dimethyl-purine-2,6-dione
sub-chemotype. Of all sub-chemotypes studied herein, the naphthalene ring conferred the
best potency as both the α- and β-naphthyl analogs (33 and 27) exhibited antiviral potencies
comparable to that of PF74. Interestingly, the 2-indolone sub-chemotype lowered the T of
m
CA hexamers, indicating a unique HIV-1 CA-destabilizing mechanism. The best 2-indolone
analog, 24, also showed 10-fold longer half-life in HLMs than PF74. These findings
necessitate future medicinal chemistry to further optimize these novel sub-chemotypes,
particularly the CA-destabilizing 2-indolone chemotype.
4. Experimental section
4.1 Chemistry
General Procedures. All commercial chemicals were used as supplied unless otherwise
indicated. Flash chromatography was performed on a Teledyne Combiflash RF-200 with
1
13
RediSep columns (silica) and indicated mobile phase. H and C NMR spectra were recorded
on a Varian 600 MHz or Bruker 400 spectrometer. Diastereomeric ratio (dr) was determined
1
by H NMR analysis. Mass data were acquired using an Agilent 6230 TOF LC/MS
spectrometer. All NMR and mass spectrometers are located in the shared instrument rooms at
the Center for Drug Design, University of Minnesota.
4.1.1. Synthesis of 59 and 60
Synthesis of intermediates 59 and 60 are described in Supporting Information.

4.1.2.
Synthesis
of
(S)-N-Methyl-2-(2-(2-methyl-1H-indol-3-yl)acetamido)-N,3-diphenylpropanamide (1)
To a solution of commercially available 2-(2-methyl-1H-indol-3-yl)acetic acid (100 mg, 0.53
mmol, 1 equiv.) in DMF (3 mL), HATU (402 mg, 1.06 mmol, 2 equiv.) and DIPEA (205 mg,
1.59 mmol, 3 equiv.) were added and the mixture was stirred at room temperature for 20 min
before (S)-2-amino-N-methyl-N,3-diphenylpropanamide (TFA salt, 235 mg, 0.64 mmol, 1.2
equiv.) was added. The mixture was further stirred overnight at room temperature. Upon
completion, H O was added and the reaction mixture was extracted with EtOAc (3x30 mL).
2
The organic phases were combined and washed with brine, dried over anhydrous MgSO4,
filtered and concentrated. The product was purified by Combiflash on silica gel using
1
EtOAc/hexane (1:2 to 6:1) as eluent. Yield 70%. H NMR (600 MHz, DMSO-d ) δ 10.68 (s,
6
1
7
2
1
1
1
4
H), 8.27 (d, J = 7.0 Hz, 1H), 7.38 (dd, J = 17.5, 6.5 Hz, 3H), 7.27 (dd, J = 18.7, 6.7 Hz, 3H),
.18 (d, J = 7.6 Hz, 1H), 7.12 (s, 3H), 6.98 – 6.89 (m, 1H), 6.84 (d, J = 6.9 Hz, 1H), 6.79 (s,
H), 4.41 (s, 1H), 3.51 – 3.31 (m, 2H), 3.15 (s, 3H), 2.82 (d, J = 11.5 Hz, 1H), 2.67 (d, J =
1
3
1.5 Hz, 1H), 2.23 (s, 3H); C NMR (150 MHz, DMSO-d ) δ 171.5, 171.0, 143.3, 138.0,
6
35.4, 133.3, 130.0, 129.2, 128.8, 128.4, 128.2, 128.0, 126.7, 120.3, 118.4, 118.4, 110.5,
−
05.3, 52.2, 37.6, 37.4, 31.3, 11.7; HRMS (ESI) m/z calcd for C H N O [M − H]
2
7
26
3
2
24.2031, found 424.2029.
(S)-2-(2-(1H-Indol-3-yl)acetamido)-N-methyl-N,3-diphenylpropanamide (2). The synthetic
method was similar to that of compound 1 except that 2-(1H-indol-3-yl)acetic acid (100
mg, 0.57 mmol, 1 equiv.) and (S)-2-amino-N-methyl-N,3-diphenylpropanamide (TFA salt,
1
2
52 mg, 0.68 mmol, 1.2 equiv.) were used as starting materials. Yield 65%. H NMR (600
MHz, CD OD) δ 7.42 (d, J = 7.9 Hz, 1H), 7.37 – 7.31 (m, 4H), 7.18 – 7.05 (m, 5H), 7.03 –
3
6.94 (m, 3H), 6.75 (d, J = 7.3 Hz, 2H), 4.66 (t, J = 7.2 Hz, 1H), 3.61 (dd, J = 22.4, 15.8 Hz,

13
2
H), 3.16 (s, 3H), 2.87 (dd, J = 13.3, 6.6 Hz, 1H), 2.64 (dd, J = 13.3, 6.6 Hz, 1H); C NMR
(
150 MHz, CD OD) δ 172.7, 171.7, 142.5, 136.7, 136.5, 129.4, 128.7, 128.0, 127.9, 127.2,
3
127.0, 126.4, 123.5, 121.1, 118.6, 118.0, 110.9, 107.7, 51.9, 37.7, 36.6, 32.2; HRMS (ESI)
−
(
−) m/z calcd for C H N O [M − H] 410.1874, found 410.1878.
2
6
24
3
2
(S)-2-(2-(5-Methoxy-1H-indol-3-yl)acetamido)-N-methyl-N,3-diphenylpropanamide
(3).
The synthetic method was similar to that of compound
1
except that
2-(5-methoxy-1H-indol-3-yl)acetic acid (100 mg, 0.49 mmol, 1 equiv.) and
(S)-2-amino-N-methyl-N,3-diphenylpropanamide (TFA salt, 217 mg, 0.59 mmol, 1.2 equiv.)
1
were used as starting materials. Yield 42%. H NMR (600 MHz, CDCl ) δ 8.34 (s, 1H), 7.36
3
−
7.31 (m, 3H), 7.26 − 7.24 (m, 1H), 7.13 (t, J = 7.4 Hz, 1H), 7.07 − 7.04 (m, 2H), 6.97 − 6.88
m, 5H), 6.66 (d, J = 7.4 Hz, 2H), 6.30 (d, J = 8.0 Hz, 1H), 4.81 – 4.77 (m, 1H), 3.79 (s, 3H),
.67 – 3.60 (m, 2H), 3.18 (s, 3H), 2.76 (dd, J = 13.3, 7.0 Hz, 1H), 2.56 (dd, J = 13.3, 7.1 Hz,
(
3
1
1
3
1
3
H); C NMR (150 MHz, CDCl ) δ 171.3, 171.0, 154.4, 142.4, 136.0, 131.4, 129.7, 129.1,
3
28.3, 128.2, 127.4, 127.3, 126.7, 124.4, 113.0, 112.1, 108.4, 100.1, 55.8, 51.3, 38.8, 37.6,
−
3.3; HRMS (ESI) m/z calcd for C H N O [M − H] 440.1980, found 440.1983.
2
7
26
3
3
(S)-2-(2-(5-Methoxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenyl-N-(p-tolyl)propanamide
(4). The synthetic method was similar to that of compound 1 except that
2-(5-methoxy-1H-indol-3-yl)acetic acid (100 mg, 0.49 mmol, 1 equiv.) and
(S)-2-amino-N-methyl-3-phenyl-N-(p-tolyl)propanamide (TFA salt, 225 mg, 0.59 mmol, 1.2
1
equiv.) were used as starting materials. Yield 63%. H NMR (600 MHz, CDCl ) δ 8.60 (s,
3
1H), 7.22 (d, J = 8.7 Hz, 1H), 7.13 – 7.12 (m, 3H), 7.08 − 7.05 (m, 2H), 6.88 − 6.69 (m, 7H),

6.33 (d, J = 8.2 Hz, 1H), 4.82 − 4.79 (m, 1H), 3.79 (s, 3H), 3.66 − 3.59 (m, 2H), 3.15 (s, 3H),
1
3
2
.77 (dd, J = 13.3, 7.0 Hz, 1H), 2.58 (dd, J = 13.3, 6.9 Hz, 1H), 2.35 (s, 3H); C NMR (100
MHz, CDCl ) δ 171.5, 170.9, 154.3, 139.9, 138.1, 136.1, 131.5, 130.3, 129.2, 128.3, 127.4,
3
127.0, 126.7, 124.5, 112.8, 112.2, 108.2, 100.1, 55.8, 51.2, 38.8, 37.7, 33.4, 21.1; HRMS
−
(
ESI) m/z calcd for C H N O [M − H] 454.2136, found 454.2139.
2
8
28
3
3
(S)-N-(4-Fluorophenyl)-2-(2-(5-methoxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenylprop
anamide (5). The synthetic method was similar to that of compound 1 except that
-(5-methoxy-1H-indol-3-yl)acetic acid (100 mg, 0.49 mmol, 1 equiv.) and
2
(S)-2-amino-N-(4-fluorophenyl)-N-methyl-3-phenylpropanamide (TFA salt, 228 mg, 0.59
1
mmol, 1.2 equiv.) were used as starting materials. Yield 62%. H NMR (600 MHz, CD OD)
3
δ 7.24 (d, J = 8.8 Hz, 1H), 7.17 − 7.12 (m, 3H), 7.06 (s, 1H), 7.01 − 6.98 (m, 3H), 6.86 − 6.76
(m, 5H), 4.57 (t, J = 7.4 Hz, 1H), 3.79 (s, 3H), 3.57 (s, 2H), 3.11 (s, 3H), 2.89 (dd, J = 13.3,
1
3
7
1
1
1
4
.5 Hz, 1H), 2.67 (dd, J = 13.3, 7.3 Hz, 1H); C NMR (150 MHz, CD OD) δ 174.15, 173.20,
3
63.35 (d, J = 246.9 Hz), 155.20, 139.96, 137.82, 133.26, 130.67 (d, J = 7.8 Hz), 130.19,
CF
CF
29.50, 128.75, 127.92, 125.67, 117.37 (d, J = 23.0 Hz), 113.01 (d, J = 9.3 Hz), 108.97,
CF
CF
−
01.28, 56.27, 53.23, 39.26, 38.07, 33.66; HRMS (ESI) m/z calcd for C H FN O [M − H]
2
7
25
3
3
58.1885, found 458.1886.
(
S)-N-(3-Fluorophenyl)-2-(2-(5-methoxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenylprop
anamide (6). The synthetic method was similar to that of compound 1 except that
-(5-methoxy-1H-indol-3-yl)acetic acid (100 mg, 0.49 mmol, 1 equiv.) and
S)-2-amino-N-(3-fluorophenyl)-N-methyl-3-phenylpropanamide (TFA salt, 228 mg, 0.59
2
(

1
mmol, 1.2 equiv.) were used as starting materials. Yield 59%. H NMR (600 MHz, CDCl )
3
δ 8.36 (s, 1H), 7.31 − 7.26 (m, 2H), 7.18 (t, J = 7.3 Hz, 1H), 7.12 − 7.10 (m, 2H), 7.03 − 7.00
(m, 2H), 6.90 − 6.87 (m, 2H), 6.77 − 6.73 (m, 3H), 6.28 (d, J = 7.9 Hz, 1H), 4.79 – 4.75 (m,
1H), 3.81 (s, 3H), 3.69 − 3.62 (m, 2H), 3.14 (s, 3H), 2.75 (dd, J = 13.1, 8.0 Hz, 1H), 2.61 (dd,
1
3
J = 13.1, 6.5 Hz, 1H); C NMR (150 MHz, CDCl ) δ 171.3, 170.9, 162.8 (d, J = 249.0 Hz),
3
CF
1
1
3
4
54.4, 143.8, 135.8, 131.5, 130.7 (d, J = 9.0 Hz), 129.2, 128.4, 127.4, 126.9, 124.4, 123.3,
CF
15.2 (d, JCF = 21.0 Hz), 114.7 (d, J = 22.8 Hz), 113.0, 112.2, 108.3, 100.1, 55.8, 51.4,
CF
−
9.2, 37.5, 33.4; HRMS (ESI) m/z calcd for C H FN O [M − H] 458.1885, found
27
25
3
3
58.1887.
(
S)-N-(4-Chlorophenyl)-2-(2-(5-methoxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenylprop
anamide (7). The synthetic method was similar to that of compound 1 except that
-(5-methoxy-1H-indol-3-yl)acetic acid (100 mg, 0.49 mmol, 1 equiv.) and
2
(S)-2-amino-N-(4-chlorophenyl)-N-methyl-3-phenylpropanamide (TFA salt, 237 mg, 0.59
1
mmol, 1.2 equiv.) were used as starting materials. Yield 80%. H NMR (600 MHz, CD OD)
3
δ 7.25 − 7.22 (m, 3H), 7.17 − 7.12 (m, 3H), 7.06 (s, 1H), 6.99 (s, 1H), 6.82 − 6.75 (m, 5H),
4
1
1
1
4
.57 (t, J = 7.4 Hz, 1H), 3.78 (s, 3H), 3.57 (s, 2H), 3.09 (s, 3H), 2.88 (dd, J = 13.2, 7.7 Hz,
1
3
H), 2.67 (dd, J = 13.2, 7.1 Hz, 1H); C NMR (100 MHz, CD OD) δ 174.1, 173.0, 155.3,
3
42.5, 137.8, 135.0, 133.3, 130.8, 130.2, 129.5, 128.8, 127.9, 125.7, 113.0, 113.0, 109.0,
−
01.4, 56.3, 53.3, 39.4, 37.9, 33.7
74.1590, found 474.1595.
; HRMS (ESI) m/z calcd for C H ClN O [M − H]
27 25 3 3

(S)-2-(2-(5-Hydroxy-1H-indol-3-yl)acetamido)-N-methyl-N,3-diphenylpropanamide
(8).
The synthetic method was similar to that of compound
1
except that
2-(5-hydroxy-1H-indol-3-yl)acetic acid (100 mg, 0.52 mmol, 1 equiv.) and
(S)-2-amino-N-methyl-N,3-diphenylpropanamide (TFA salt, 232 mg, 0.63 mmol, 1.2
1
equiv.) were used as starting materials. Yield 49%. H NMR (600 MHz, CDCl ) δ 8.42 (s,
3
1H), 7.30 − 7.27 (m, 3H), 7.13 − 7.11 (m, 2H), 7.08 − 7.06 (m, 2H), 6.90 (s, 1H), 6.86 − 6.77
(m, 4H), 6.68 (d, J = 7.3 Hz, 2H), 6.60 (d, J = 8.1 Hz, 1H), 4.79 (q, J = 7.3 Hz, 1H), 3.55 (s,
13
2
H), 3.17 (s, 3H), 2.79 (dd, J = 13.4, 7.0 Hz, 1H), 2.59 (dd, J = 13.4, 7.3 Hz, 1H); C NMR
(
100 MHz, CDCl ) δ 171.7, 171.6, 150.4, 142.3, 136.0, 131.4, 129.8, 129.2, 128.3, 128.2,
3
127.8, 127.3, 126.7, 124.7, 112.5, 112.0, 107.6, 103.0, 51.4, 38.7, 37.8, 33; HRMS (ESI) m/z
−
calcd for C H N O [M − H] 426.1823, found 426.1826.
2
6
24
3
3
(S)-2-(2-(5-Hydroxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenyl-N-(p-tolyl)propanamide
(9). The synthetic method was similar to that of compound 1 except that
2-(5-hydroxy-1H-indol-3-yl)acetic acid (100 mg, 0.52 mmol, 1 equiv.) and
(S)-2-amino-N-methyl-3-phenyl-N-(p-tolyl)propanamide (TFA salt, 241 mg, 0.63 mmol, 1.2
1
equiv.) were used as starting materials. Yield 80%. H NMR (600 MHz, CDCl ) δ 8.27 (s,
3
1
2
2
H), 7.26 (s, 1H), 7.16 − 7.08 (m, 6H), 6.91 (s, 1H), 6.79 − 6.78 (m, 3H), 6.72 (d, J = 7.1 Hz,
H), 6.53 (d, J = 8.2 Hz, 1H), 6.27 (s, 1H), 4.83 − 4.79 (m, 1H), 3.57 (s, 2H), 3.15 (s, 3H),
1
3
.80 (dd, J = 13.4, 6.9 Hz, 1H), 2.60 (dd, J = 13.4, 7.2 Hz, 1H), 2.33 (s, 3H); C NMR (150
MHz, CDCl ) δ 171.7, 171.3, 150.2, 139.7, 138.2, 136.1, 131.4, 130.3, 129.2, 128.3, 127.8,
3
127.0, 126.7, 124.6, 112.5, 112.0, 107.8, 103.1, 51.2, 38.7, 37.8, 33.2, 21.1; HRMS (ESI) m/z
−
calcd for C H N O [M − H] 440.1980, found 440.1984.
2
7
26
3
3

(
S)-N-(4-Fluorophenyl)-2-(2-(5-hydroxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenylprop
anamide (10). The synthetic method was similar to that of compound 1 except that
-(5-hydroxy-1H-indol-3-yl)acetic acid (100 mg, 0.52 mmol, 1 equiv.) and
2
(S)-2-amino-N-(4-fluorophenyl)-N-methyl-3-phenylpropanamide (TFA salt, 241 mg, 0.63
1
mmol, 1.2 equiv.) were used as starting materials. Yield 49%. H NMR (600 MHz, CD OD)
3
δ 7.19 − 7.14 (m, 5H), 7.03 − 7.00 (m, 3H), 6.86 − 6.82 (m, 4H), 6.69 (d, J = 8.6 Hz, 1H), 4.56
(t, J = 7.4 Hz, 1H), 3.54 (s, 2H), 3.11 (s, 3H), 2.88 (dd, J = 13.2, 7.7 Hz, 1H), 2.68 (dd, J =
1
3
1
1
1
3.2, 7.1 Hz, 1H); C NMR (150 MHz, CD OD) δ 174.2, 173.2, 163.4 (d, J = 246.9 Hz),
3 CF
51.5, 140.96, 139.94, 137.8, 133.0, 130.7 (d, JCF = 8.8 Hz), 130.3, 129.5, 129.2, 127.9,
25.8, 117.4 (d, J = 23.0 Hz), 112.8 (d, J = 9.7 Hz), 108.3, 103.6, 53.2, 39.4, 38.1, 33.7;
CF
CF
−
HRMS (ESI) m/z calcd for C H FN O [M − H] 444.1729, found 444.1732.
2
6
23
3
3
(
S)-N-(3-Fluorophenyl)-2-(2-(5-hydroxy-1H-indol-3-yl)acetamido)-N-methyl-3-phenylprop
anamide (11). The synthetic method was similar to that of compound 1 except that
-(5-hydroxy-1H-indol-3-yl)acetic acid (100 mg, 0.52 mmol, 1 equiv.) and
2
(S)-2-amino-N-(3-fluorophenyl)-N-methyl-3-phenylpropanamide (TFA salt, 241 mg, 0.63
1
mmol, 1.2 equiv.) were used as starting materials. Yield 39%. H NMR (600 MHz, CDCl )
3
δ 8.30 (s, 1H), 7.27 − 7.11 (m, 5H), 6.99 − 6.94 (m, 2H), 6.85 − 6.75 (m, 5H), 6.53 (d, J = 8.0
Hz, 1H), 6.47 (s, 1H), 6.32 (s, 1H), 4.78 – 4.74 (m, 1H), 3.59 (s, 2H), 3.12 (s, 3H), 2.78 (dd, J
1
3
=
13.2, 8.1 Hz, 1H), 2.63 (dd, J = 13.2, 6.6 Hz, 1H); C NMR (150 MHz, CDCl ) δ 171.6,
3
171.5, 162.7 (d, J = 249.4 Hz), 150.4, 143.6, 135.8, 131.4, 130.8 (d, J = 9.5 Hz), 129.2,
CF CF
1
28.4, 127.7, 126.9, 124.6, 123.3, 115.3 (d, J = 21.2 Hz), 114.6 (d, J = 20.9 Hz), 112.6,
CF
CF