Discovery, SAR study and ADME properties of methyl 4-amino-3-cyano-1-(2-benzyloxyphenyl)-1H-pyrazole-5-carboxylate as an HIV-1 replication inhibitor

Article abstract of DOI:10.1039/d0md00025f

Inspired by the antiviral activity of known pyrazole-based HIV inhibitors, we screened our in-house library of pyrazole-based compounds to evaluate theirin celluloactivity against HIV-1 replication. Two hits with very similar structures appeared from single and multiple-round infection assays to be non-toxic and active in a dose-dependent manner. Chemical expansion of their series allowed an in-depth and consistent structure-activity-relationship study (SAR) to be built. Further ADME evaluation led to the selection of 4-amino-3-cyano-1-(2-benzyloxyphenyl)-1H-pyrazole-5-carboxylate with an advantageous pharmacokinetic profile. Finally, examination of its mode of action revealed that this compound does not belong to the three main classes of anti-HIV drugs, a feature of prime interest in the context of viral resistance.

Full text of DOI:10.1039/d0md00025f

RSC
Medicinal Chemistry
View Article Online
View Journal
RESEARCH ARTICLE
Discovery, SAR study and ADME properties of
methyl 4-amino-3-cyano-1-(2-benzyloxyphenyl)-
Cite this: DOI: 10.1039/d0md00025f
1H-pyrazole-5-carboxylate as an HIV-1 replication
inhibitor†
a
Jeanne Fichez,
Stéphane Priet,
Guillaume Prestat,
Cathia Soulie,b Laurent Le Corre,a Sophie Sayon,b
cd
Karine Alvarez,c Olivier Delelis,e Patrick Gizzi,f
a
a
Christine Gravier-Pelletier,
Anne-Geneviève Marcelin,b
Vincent Calvezb and Patricia Busca
a
*
Inspired by the antiviral activity of known pyrazole-based HIV inhibitors, we screened our in-house library
of pyrazole-based compounds to evaluate their in cellulo activity against HIV-1 replication. Two hits with
very similar structures appeared from single and multiple-round infection assays to be non-toxic and active
in a dose-dependent manner. Chemical expansion of their series allowed an in-depth and consistent
structure–activity-relationship study (SAR) to be built. Further ADME evaluation led to the selection of
Received 23rd January 2020,
Accepted 14th March 2020
4-amino-3-cyano-1-(2-benzyloxyphenyl)-1H-pyrazole-5-carboxylate
with
an
advantageous
pharmacokinetic profile. Finally, examination of its mode of action revealed that this compound does not
belong to the three main classes of anti-HIV drugs, a feature of prime interest in the context of viral
resistance.
DOI: 10.1039/d0md00025f
rsc.li/medchem
However, HAART is still not curative,4 and this life-long
treatment suffers from several pitfalls such as inherent drug
Introduction
Human immunodeficiency virus (HIV) and the resulting
acquired immunodeficiency syndrome (AIDS) still represent a
major global public health threat. According to the Joint
United Nations Program on HIV and AIDS (UNAIDS), since
the beginning of the epidemic, more than three decades ago,
77.3 million people have become infected with HIV and 35.4
million of them have died from AIDS-related illnesses.1
Highly active anti-retroviral therapy (HAART), introduced in
1996, led to a significant decrease of AIDS-related deaths by
38% and helped to save 11.4 million lives.2 HAART has also
improved the life expectancy of HIV/AIDS patients which
nowadays is approaching that of the general population.3
toxicity,5 emergence or transmission of drug-resistant HIV-1
strains,6–8 and the decline in patient adherence.9 In response
to these concerns, a global effort is required to pursue the
development of novel and potent anti-HIV-1 drugs, which
represents
a
continuing challenge.10 In this context,
prompted by the promising antiretroviral activities of several
pyrazole-based inhibitors such as PNU-32945 I,11 Lersivirine
II12 and benzylamino derivatives III13 (Fig. 1), we decided to
screen
a
pyrazole-based chemical library, which was
previously developed in our laboratory.14,15
We report herein the discovery of new pyrazolic HIV-1
inhibitors, their synthesis, biological activities and ADMET
properties.
Interestingly,
methyl
4-amino-3-cyano-1-(2-
benzyloxyphenyl)-1H-pyrazole-5-carboxylate that was selected
a LCBPT, UMR CNRS 8601, Université de Paris, Paris, France.
E-mail: patricia.busca@parisdescartes.fr
b Laboratoire de Virologie-CERVI, UMR S 1136, Hôpital Pitié Salpêtrière –
Sorbonne Université, Paris, France
c Aix-Marseille Univ, Laboratoire AFMB, UMR CNRS 7257, Marseille, France
d UMR EPV, Université Aix-Marseille – IRD 190 – INSERM 1207 – EHESP, Marseille,
France
e LBPA, UMR CNRS 8113, ENS Paris-Saclay, Cachan, France
f PCBIS, UMS CNRS 3286, ESBS – Université de Strasbourg, Illkirch, France
† Electronic supplementary information (ESI) available: Biology: experimental
procedures, full results of the first screening, dose–response curves; chemistry:
general procedures, compound characterization, 1H and 13C NMR spectra. See
DOI: 10.1039/d0md00025f
Fig. 1 Structures of known pyrazole-based HIV inhibitors.
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
as a lead compound does not inhibit any of the viral enzymes
that are classically targeted in antiviral therapy.
Results and discussion
Discovery of the hit compounds through a double screening
Our in-house chemical library, which was originally designed
to target a kinase domain, is composed of 110 pyrazole-based
molecules that are well balanced between four
Fig. 3 Structures of the three selected hits and biological results @50
μM.
heteroaromatic
collections:
aminopyrazoles
A,
pyrazolopyridines B, pyrazolopyrimidines C and pyrazolo-
oxadiazoles D (Fig. 2). Interestingly, while compounds from
families C and D behave like kinase inhibitors, those from
series A and B are not active against this kind of target.16
These compounds were tested for their potential
inhibitory activity against HIV-1 replication in cell cultures.
This primary screening was carried out using HeLa P4 cells
containing a stably integrated LTR linked to a Lac Z reporter
gene. These cells were infected with HIV-1 Laï viruses, and
were incubated for 48 h with 50 μM of each compound, using
zidovudine as a positive control (see the ESI†). This kind of
phenotypic assay was chosen because it allows for several
steps of the HIV replication cycle to be evaluated at once. In
parallel, the cytotoxicity of each compound was evaluated
under the same culture conditions according to the MTS
standard protocol (see the ESI†). This test is also necessary to
assess the relevancy of inhibition values.
After this first screening at 50 μM, none of the 24
pyrazolopyrimidines C nor the 27 pyrazolo-oxadiazoles D
showed enough activity to be selected (the full table of results
can be found in the ESI†). However, three compounds
displayed an inhibition activity higher than 50%, and a
toxicity value lower than 15%: aminopyrazoles A.20 and A.21,
as well as pyrazolopyridine B.19 (Fig. 3). Interestingly, the
benzyloxy group present in both A.20 and A.21 is likely to be
part of the pharmacophoric requirements.
Dose-dependent assay
To further validate these hits, and to rule out the possibility
of false positives, the inhibition assay was repeated in a dose-
dependent manner using a 1.0 to 50 μM concentration range
of the inhibitors (see the ESI†).
The resulting dose–response curves allowed for the
concentration at which 50% of the replication of HIV is
inhibited to be determined (EC50 value). These experiments
revealed that A.20 and A.21 are able to inhibit the HIV
replication in a dose-dependent manner. In contrast, it was
found that B.19 is likely to be a false positive because its
apparent inhibition potency is related to its toxicity and not
to its activity, and therefore it was discarded. Encouragingly,
the EC50 values of A.20 and A.21 were found to be 19.4 μM
and 15.2 μM, respectively, which are values that are
comparable to the 13.3 μM EC50 of tenofovir (TDF, clinically
used inhibitor) which we determined under the same
conditions.17 Hence, A.20 and A.21 were validated as hits and
selected for the next step of biological evaluation.
Multiple-round infection assay
With the aim to ascertain their antiviral potency, the
validated hits were then evaluated in different cell-types, with
another source of the virus, and also incubated for a longer
amount of time. For this multiple-round infection assay,
human lymphoblastoid MT2 cells were incubated for 3 days
with Laï virus at MOI 3 and the antiviral effect of A.20 and
A.21 was monitored in a concentration range from 0 to 50
μM using zidovudine as a positive control (see the ESI†).
According to these results, A.20 and A.21 are undoubtedly
efficient in counteracting HIV-1 replication, even at this high
concentration of virus. To further develop this series of
aminopyrazoles, the synthesis of analogous compounds was
therefore undertaken.
Chemical expansion of the A series
To study and verify the impact of substituents on the phenyl
ring of pyrazoles A, new analogues were synthesized
according to the two-step strategy which we previously
developed (Table 1).14,15 Starting from commercially available
anilines I.1–28, firstly a diazotization reaction with sodium
nitrite in aqueous hydrochloric acid was performed. This was
Fig. 2 Molecular scaffolds included in the screened chemical library.
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
Table 1 Two-step synthesis of aminopyrazoles A.1–28a
Compound
–R
Overall yield (%)
A.1
A.2
A.3
H
69
38
48
16
35
56
44
35
51
38
48
32
38
44
82
83
35
68
42
42
61
18
50
73
22
31
77
44
2-NO2
3-NO2
4-NO2
2-Br
3-Br
4-Br
2-Cl
3-Cl
4-Cl
2-F
A.4
A.5b
A.6
A.7
A.8b
A.9
A.10
A.11b
A.12
A.13
A.14b
A.15b
A.16b
A.17b
A.18
A.19
A.20
A.21
A.22b
A.23b
A.24
A.25b
A.26b
A.27
A.28
3-F
4-F
2-I
3-CF3
4-CF3
2-CO2tBu
3-CO2tBu
4-CO2tBu
2-OBn
3-OBn
4-OBn
2-OMe
3-OMe
4-OMe
2-CCH
3-CCH
4-CCH
a
b
Yields are calculated over two steps. Hydrazones II.x and pyrazoles A.x are described in the ESI.†
immediately followed by condensation with malononitrile in
First of all, the activity of N-phenyl-pyrazole A.1 (without
any substituents) remains below the critical threshold of
30% inhibition. When compared to the activity of hits A.20
and A.21 (63% and 65% respectively), this low value
supports our hypothesis regarding the key role that might
be played by the benzyloxy group. Then, nine compounds
which show good inhibition all share the common feature
of bearing electron withdrawing groups such as NO2 (A.2–4),
Br (A.5–6), Cl (A.8), F (A.12), or CO2tBu (A.18–19). Besides
the pure electronic effect on the phenyl ring, these groups
could also be involved in non-covalent interactions such as
hydrogen or halogen-bonding. Finally, none of the newly
synthesized pyrazoles were found to be more efficient; with
A.20 and A.21, bearing a benzyloxy substituent, being the
best inhibitors within this series. When compared to their
methyloxy counterparts (A.23–24), these hits are at least
twice as active, suggesting that the benzyl moiety is likely to
interact with a hydrophobic pocket. These observations are
summarized in the following figure (Fig. 4). As no other
compound appeared to be as potent as A.20 and A.21, these
two hits were further evaluated in order to assess their
drugability.
basic
dicyanohydrazones II.1–28 in good to excellent yields. The
second step is Thorpe–Ziegler reaction which was
medium
to
afford
the
corresponding
a
performed with methyl bromoacetate and potassium
carbonate in dioxane under microwave irradiation. Desired
aminopyrazoles A.1–28 were obtained in moderate to good
yields ranging from 16 to 83% over two steps. Thanks to this
synthetic effort, we obtained
a homogeneous series of
aminopyrazoles with comparable substitution patterns which
are suitable for structure–activity relationship studies.
Structure–activity relationship (SAR)
To draw out the SAR pattern, the inhibition and toxicity
values of the previously and newly synthesized pyrazoles are
gathered in the following table (Table 2). For the sake of
clarity, we used a minimum threshold of 30% for inhibition,
and a maximum one of 15% for toxicity. Indeed, beyond this
value, the cell mortality distorts the value of inhibition
significantly. The application of this double filter revealed 11
interesting compounds (Table 2).
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
Table 2 In vitro evaluation of the inhibitory activity and toxicity of
aminopyrazoles A.1–28a
less lipophilic, A.20 is ten times more soluble; it displays
better membrane permeability by passive diffusion. As the
compound doesn't contain any hydrolysable group, its
plasma stability is excellent. Its low metabolic stability is
offset by a high level of protein binding, which should ensure
good availability over time. According to these results, A.20
has the best drugability potential, therefore we decided to
pursue our study with this compound.
Compound
–R
Inhibition (%)
Toxicity (%)
A.1
A.2
A.3
A.4
A.5
A.6
A.7
A.8
H
<30
42
47
43
32
42
67
30
<30
56
7.0
0
0
2-NO2
3-NO2
4-NO2
2-Br
3-Br
4-Br
2-Cl
3-Cl
4-Cl
2-F
0
4.8
9.0
>15
2.4
11.0
>15
3.6
0.0
6.0
>15
>15
14
A.9
Identification of the potential target of hit A.20
A.10
A.11
A.12
A.13
A.14
A.15
A.16
A.17
A.18
A.19
A.20
A.21
A.22
A.23
A.24
A.25
A.26
A.27
A.28
At present, 30 drugs have been officially approved for the
treatment of HIV/AIDS.18 These molecules target essential
steps or viral enzymes that are involved in the HIV replication
cycle: CCR5 co-receptor,19 viral fusion,20 reverse transcriptase
(RT),21 integrase (IN)22,23 and protease (PR).24 It is noteworthy
to say that the current WHO preferred first-line regimens
predominantly consist of RT inhibitors (RTI), in some
instances associated with PR or IN strand-transfer inhibitors
(PRI or INSTI).25,26 However, despite tremendous efforts to
develop new therapeutic agents, no novel target exploitation
is to be expected from the few inhibitors in clinical trials at
the moment.27
In this context, to explore the mechanism of action of
A.20, we tried to discover its biological target. Since A.20 was
selected through a screening assay that cannot detect PR
inhibitors (one single cycle of HIV replication does not allow
detection of post-integrative step inhibitors), no further
investigation was performed on this target. We therefore
focused our study on RT and IN strand transfer (INST)
activities.
<30
48
3-F
4-F
2-I
<30
<30
<30
<30
<30
36
47
63
65
<30
<30
<30
<30
<30
<30
Nd
3-CF3
4-CF3
2-CO2tBu
3-CO2tBu
4-CO2tBu
2-OBn
3-OBn
4-OBn
2-OMe
3-OMe
4-OMe
2-CCH
3-CCH
4-CCH
2.8
0
14.0
9.0
11.0
3.8
9.2
6.0
14.6
7.0
14.0
Nd
a
Data represent mean values of two experiments in duplicate.
To evaluate the inhibitory activity of compound A.20 on
the RT of HIV-1, its relative efficiency of incorporation was
measured using subtype B WT HIV-1 RT in an in vitro
susceptibility assay, using nevirapine and AZT as positive
controls (see the ESI†). Under these conditions, A.20
Fig. 4 General trends resulting from SAR analysis.
appeared to be completely ineffective toward RT (IC50
1000).
>
To check the propensity of A.20 to inhibit the reaction
mediated by IN, strand transfer assays were carried out with
increasing concentrations of inhibitors, from 100 nM to 333
μM, using raltegravir as a positive control (see the ESI†).
Regardless of the concentration used, no effect on the strand
transfer reaction efficiency was observed, thereby
ADME properties
A full pharmacokinetic profile has been established for A.20
and A.21. The overall results are gathered in Table 3.
Although the structures of A.20 and A.21 are very similar,
A.20 shows a better overall profile than A.21. Being slightly
Table 3 ADME properties of hits A.20 and A.21
ADME properties
A.20
A.21
PBS solubility
Octanol/water partition coefficient
Plasma protein binding
9.6 0.5 μM
log D7.4 = 3.69 0.06
95 1%
1.03 0.06 μM
log D7.4 = 3.84 0.05
100 0%
Permeability (PAMPA)
Metabolic stability – half life
In vitro stability
Maximum plasma concentration after intravenous administration in mice
Maximum plasma concentration after oral administration in mice
log Pe = −5.41 0.04
t1/2 = 3.0 0.5 min
100%
log Pe = −5.97 0.04
t1/2 = 4.6 0.5 min
100%
238 29 ng mL−1 at 10 min
18 4 ng mL−1 at 10 min
241 26 ng mL−1 at 10 min
Nd
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
demonstrating that A.20 does not impact the catalytic
properties of INST at all.
From these results, we can conclude that our lead
compound, A.20, does not belong to the three main classes
of anti-HIV drugs (RTI, PI and INSTI), and thus is likely to
inhibit an underexploited or unknown target.
Persaud, J. D. Tucker, F. Barre-Sinoussi, International
AIDS Society Towards a Cure Working Group, G. Alter, J.
Auerbach, B. Autran, D. H. Barouch, G. Behrens, M.
Cavazzana, Z. Chen, É. A. Cohen, G. M. Corbelli, S.
Eholié, N. Eyal, S. Fidler, L. Garcia, C. Grossman, G.
Henderson, T. J. Henrich, R. Jefferys, H.-P. Kiem, J.
McCune, K. Moodley, P. A. Newman, M. Nijhuis, M. S.
Nsubuga, M. Ott, S. Palmer, D. Richman, A. Saez-Cirion,
M. Sharp, J. Siliciano, G. Silvestri, J. Singh, B. Spire, J.
Taylor, M. Tolstrup, S. Valente, J. van Lunzen, R.
Walensky, I. Wilson and J. Zack, International AIDS
Society global scientific strategy: towards an HIV cure
2016, Nat. Med., 2016, 22, 839.
Conclusion
It is still of key importance to develop new inhibitors of HIV
replication in order to counteract the rapid and unavoidable
virus mutations, with a preference for those which are
orthogonal to the current drugs. In this study, we identified
two promising hits A.20 and A.21 with interesting EC50
values, similar to that of tenofovir. Further multiple round
infection assays confirmed their inhibitory activity on long
term incubation. Based on the same aminopyrazole scaffold,
a series of derivatives were synthesized to draw an SAR study
that ascertained A.20 and A.21 as hits, and the OBn
substituent as an essential group for activity. ADME
evaluations allowed us to select A.20 as the better lead
compound. The final part of our study revealed that A.20
does not behave as usual anti-HIV drugs (RTI, PI and INSTI)
and thus belongs to the sought-out category of inhibitors
which could advantageously slow down the development of
resistance.
5 A. Carr, Nat. Rev. Drug Discovery, 2003, 2, 624.
6 C.-F. Perno, G. Moyle, C. Tsoukas, W. Ratanasuwan,
J. Gatell and M. Schechter, J. Med. Virol., 2008, 80,
565.
7 S. Hué, R. J. Gifford, D. Dunn, E. Fernhill and D. Pillay,
J. Virol., 2009, 83, 2645.
8 L. Menéndez-Arias, Antiviral Res., 2013, 98, 93.
9 T. Mathes, D. Pieper, S.-L. Antoine and M. Eikermann, HIV
Med., 2013, 14, 583.
10 F. Dabis and L.-G. Bekker, Science, 2017, 357, 335.
11 M. J. Genin, C. Biles, B. J. Keiser, S. M. Poppe, S. M. Swaney,
W. G. Tarpley, Y. Yagi and D. L. Romero, J. Med. Chem.,
2000, 43, 1034–1040.
This study provided us with A.20 as a validated lead
compound, and a solid SAR basis to design and develop more
potent inhibitors. Moreover, fostered by the complete lack of
activity on the classical targets, further studies will be
dedicated to uncover its mechanism of action.
12 M. Platten and G. Fätkenheuer, Expert Opin. Invest. Drugs,
2013, 22, 1687.
13 T. Mizuhara, T. Kato, A. Hirai, H. Kurihara, Y. Shimada,
M. Taniguchi, H. Maeta, H. Togami, K. Shimura, M.
Matsuoka, S. Okazaki, T. Takeuchi, H. Ohno, S. Oishi
and N. Fujii, Bioorg. Med. Chem. Lett., 2013, 23,
4557–4561.
Conflicts of interest
14 L. L. Corre, L. Tak-Tak, A. Guillard, G. Prestat, C.
Gravier-Pelletier and P. Busca, Org. Biomol. Chem.,
2015, 13, 409.
15 L. L. Corre, M. C. D. Lang, C. Garbay, C. Gravier-Pelletier, P.
Busca, M. Ethève-Quelquejeu and E. Braud, Synthesis,
2016, 48, 4569.
16 Unpublished results.
17 Unpublished results.
18 Y. Mehellou and E. De Clercq, J. Med. Chem., 2010, 53,
521.
There are no conflicts to declare.
Acknowledgements
The authors thank the “Agence Nationale de Recherches sur le
Sida et les hépatites virales” for support (ANRS, ref ECTZ20422),
the “Ministère de l'Enseignement supérieur, de la Recherche et
de l'Innovation” for J. F. PhD grant and the UMR 8601 CNRS,
Université Paris Descartes for HRMS and NMR facilities.
19 H. R. Shah and J. K. Savjani, Eur. J. Med. Chem., 2018, 147,
115.
Notes and references
1 UNAIDS - Global HIV & AIDS statistics http://www.unaids.
org/en/resources/fact-sheet.
20 D. Eggink, B. Berkhout and R. W. Sanders, Curr. Pharm. Des.,
2010, 16, 3716.
2 World Health Organization- Key facts http://www.who.int/
mediacentre/factsheets/fs360/en/.
21 A. D. Holec, S. Mandal, P. K. Prathipati and C. J. Destache,
Curr. HIV Res., 2017, 15, 411.
3 J. L. Marcus, C. R. Chao, W. A. Leyden, L. Xu, C. P.
Quesenberry, D. B. Klein, W. J. Towner, M. A. Horberg and
M. J. Silverberg, J. Acquired Immune Defic. Syndr., 2016, 73,
39.
4 S. G. Deeks, S. R. Lewin, A. L. Ross, J. Ananworanich, M.
Benkirane, P. Cannon, N. Chomont, D. Douek, J. D.
Lifson, Y.-R. Lo, D. Kuritzkes, D. Margolis, J. Mellors, D.
22 Y. Pommier, A. A. Johnson and C. Marchand, Nat. Rev. Drug
Discovery, 2005, 4, 236.
23 V. Nair and G. Chi, Rev. Med. Virol., 2007, 17, 277.
24 A. K. Ghosh, H. L. Osswald and G. Prato, J. Med. Chem.,
2016, 59, 5172.
25 U.S. Department of Health and Human Services - Guidelines for
the Use of Antiretroviral Agents in Adults and Adolescents Living
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
with HIV, https://aidsinfo.nih.gov/contentfiles/lvguidelines/
adultandadolescentgl.pdf.
https://www.who.int/hiv/pub/guidelines/hivdr-guidelines-
2017/en/.
26 World Health Organization – The Public Health Response
to Pretreatment HIV drug resistance, WHO Guide, 2017,
27 H.-V. Tieu, B. S. Taylor, J. Jones and T. J. Wilkin, Top. Antivir.
Med., 2018, 26, 40.
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020