Picomolar inhibitors of HIV-1 reverse transcriptase: Design and crystallography of naphthyl phenyl ethers

Article abstract of DOI:10.1021/ml5003713

Catechol diethers that incorporate a 6-cyano-1-naphthyl substituent have been explored as non-nucleoside inhibitors of HIV-1 reverse transcriptase (NNRTIs). Promising compounds are reported that show midpicomolar activity against the wild-type virus and sub-20 nM activity against viral variants bearing Tyr181Cys and Lys103Asn mutations in HIV-RT. An X-ray crystal structure at 2.49 ? resolution is also reported for the key compound 6e with HIV-RT.

Full text of DOI:10.1021/ml5003713

Letter
pubs.acs.org/acsmedchemlett
Picomolar Inhibitors of HIV‑1 Reverse Transcriptase: Design and
Crystallography of Naphthyl Phenyl Ethers
Won-Gil Lee,† Kathleen M. Frey,‡ Ricardo Gallardo-Macias,† Krasimir A. Spasov,‡ Mariela Bollini,†
Karen S. Anderson,*,‡ and William L. Jorgensen*,†
†Department of Chemistry, Yale University, New Haven, Connecticut 06520-8107, United States
‡Department of Pharmacology, Yale University School of Medicine, New Haven, Connecticut 06520-8066, United States
S
* Supporting Information
ABSTRACT: Catechol diethers that incorporate a 6-cyano-1-naphthyl
substituent have been explored as non-nucleoside inhibitors of HIV-1
reverse transcriptase (NNRTIs). Promising compounds are reported that
show midpicomolar activity against the wild-type virus and sub-20 nM
activity against viral variants bearing Tyr181Cys and Lys103Asn
mutations in HIV-RT. An X-ray crystal structure at 2.49 Å resolution
is also reported for the key compound 6e with HIV-RT.
KEYWORDS: Anti-HIV agents, NNRTIs, protein crystallography
on-nucleoside inhibitors of HIV-1 reverse transcriptase
(NNRTIs) are a central component of anti-HIV
compounds.5 Other issues are also in play including activity
toward viral variants containing clinically common mutations in
HIV-RT, general cytotoxicity, and pharmacological properties,
especially solubility. The two most commonly occurring
mutations are Lys103Asn and Tyr181Cys.6 Indolizine 5a (X
= Z = H; Y = CN) is impressive in multiple regards; it has EC50
values of 0.380 nM toward the WT virus and 11 nM toward
HIV-1 bearing the challenging K103N/Y181C double muta-
N
chemotherapy.1 In view of the continuing severity of the
HIV/AIDS pandemic,2 great effort is directed at seeking new
compounds with improvements in activity toward variant
strains of HIV-1 and in reducing dosages and undesirable side-
effects.3 There are five FDA-approved NNRTIs including
evafirenz and rilpivirine, which have particular importance since
they are components of the one-a-day triple-combination
therapies Atripla and Complera.1 Our research has led to
discovery of some extraordinarily potent NNRTIs in the
catechol diether class.4 In particular, the dichloro (X, Z = Cl)
and difluoro analogues of 1 have EC50 values of 0.055 and 0.320
nM in an assay using human MT-2 cells infected with wild-type
(WT) HIV-1, while efavirenz and rilpivirine have EC50s of 2.0
and 0.670 nM.4
tion, it shows no cytotoxicity toward human T-cells (CC50
>
100 μM), and it has good aqueous solubility, 38 μg/mL. By
comparison, rilpivirine is somewhat less potent toward the WT
virus (0.670 nM), somewhat more potent toward the double
mutant (2.0 nM), significantly more cytotoxic (CC50 = 8 μM),
and far less soluble (ca. 0.1 μg/mL). However, oddly, in our
assay with virus containing the single Y181C mutation, 5a has
an EC50 of 310 nM, while rilpivirine yields 0.65 nM. Though we
have also reported crystal structures of 5a and four analogues of
1 and 2 with WT HIV-RT,5,7,8 clear explanation of the lower
activity of 5a toward the Y181C-bearing variant and the
significant improvement toward the double mutant is elusive.
For NNRTIs in general, it is almost always the case that activity
is much reduced in going from the single to the double variant.
Under the circumstances, we have continued to explore
additional derivatives of the catechol diethers. Herein, the
focus is on 1-naphthyl analogues 6. These were not investigated
However, 1 like rilpivirine (2) bears a cyanovinylphenyl
group, which may evoke concern as a potential Michael
acceptor leading to off-target effects. In order to circumvent
this, computational analyses led us to explore bicyclic
replacements among which 3−5 provided very potent
Received: September 10, 2014
Accepted: October 9, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ml5003713 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
earlier since it was unclear that their larger size would be
accommodated in the NNRTI binding site and because among
the variety of bicyclic heterocycles some could be expected to
yield more optimal aryl−aryl interactions in the binding site.5
Table 1. Inhibitory Activity (EC50, μM) for HIV-1 and
Cytotoxicity (CC50, μM) in MT-2 Cell Assays
K103N
a
a
a
b
cmpd
X
Y
Z
WT
Y181C
/Y181C
CC50
1a
1b
1c
3a
3b
3c
4a
4b
4c
4d
5a
5b
5c
5d
5e
5f
Cl
F
Cl
F
0.000055
0.00032
0.00031
0.056
0.049
0.016
0.046
38
0.220
0.085
0.024
5.5
10
45
18
72
1.2
81
H
H
H
H
H
H
F
Cl
H
Me
Cl
H
Cl
H
H
H
Cl
F
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
H
0.010
NA
0.80
0.34
27
2.7
0.018
NA
0.42
>100
>100
>100
7.0
0.019
1.9
0.26
0.040
3.5
4.0
Cl
H
H
H
F
0.0070
0.00038
0.0037
0.00070
0.0027
0.00040
0.0025
0.060
NA
NA
0.31
0.22
0.17
0.60
0.25
0.16
5.7
0.011
0.022
0.068
0.22
>100
39
40
H
F
100
50
F
0.010
0.11
F
Cl
H
H
H
Cl
H
Cl
Cl
Cl
H
H
F
15
6a
6b
6c
6d
6e
6f
H
H
H
H
H
H
F
5
80
Me
Cl
0.25
12
15
40
As summarized in Scheme 1, synthesis proceeded in a
manner similar to that for 4 and 5.5 Substituted 1-
hydroxynaphthalenes underwent Cu(I)-catalyzed addition to
2-bromoanisoles to yield naphthyl phenyl ethers. The methoxy
group was unmasked with BBr3 to give the phenols, which were
alkylated with N-Bz-protected 1-bromoethyluracil and then
deprotected. Full details are provided in the Supporting
Information. The identity of assayed compounds was confirmed
0.014
1.0
0.71
>100
45
Cl
0.038
0.43
0.019
0.013
0.0053
0.049
0.010
0.048
0.008
0.0056
0.032
NA
0.52
CN
CN
CN
CN
CN
CN
CN
CN
CN
0.00053
0.0022
0.0045
0.011
0.015
0.0075
0.046
0.250
0.048
0.150
0.006
0.021
0.150
NA
>100
24
6g
6h
6i
16
Cl
F
20
0.00049
0.0015
0.0011
0.0019
0.005
>100
>100
>100
>100
18
6j
Cl
H
F
1
by H and 13C NMR and high-resolution mass spectrometry;
6k
6l
HPLC analyses established purity as >95%. Aqueous solubilities
were measured using a shake-flask procedure.5,9 Saturated
solutions were made by stirring excesses of the compounds in
Britton−Robinson buffer for 48 h at 25 °C. The pH of the
buffer solutions is 6.5 as measured by a Corning General
Purpose pH Combination probe (4136L21). The supernatant
was collected using a Pall Life Sciences Acrodisc syringe filter
with a 0.2 μm pore size and analyzed by UV−vis
spectrophotometry (Agilent 8453). Piroxicam is used as a
control; we consistently obtain values of 5−7 μg/mL, while the
literature value is 5.9 μg/mL.9
F
6m
Cl
F
c
nev
0.11
>100
15
c
efv
0.002
0.010
0.008
0.030
0.005
0.002
c
etv
0.001
11
c
rpv
0.00067
0.00065
8
a
For 50% protection in MT-2 cells; NA for EC50 > CC50 or >100 μM.
For 50% inhibition of MT-2 cell growth. Nevirapine (nev); efavirenz
b
c
(efv); etravirine (etv); and rilpivirine (rpv).
The procedures for the human MT-2 T-cell assays have been
described in detail.4,5,10,11 Briefly, activities against the IIIB and
variant strains of HIV-1 were measured; EC50 values are
obtained as the dose required to achieve 50% protection of the
infected MT-2 cells by the MTT colorimetric method. CC50
values for inhibition of MT-2 cell growth by 50% are obtained
as well. The antiviral and toxicity curves used triplicate samples
at each concentration.
a
Scheme 1. Synthesis of 1-Naphthyl Phenyl Ethers 6
The activity results for the naphthyl ethers are summarized in
Table 1 along with corresponding data for key compounds in
the 1 and 3−5 series and for four FDA-approved NNRTIs.
This is the initial report for 1c and 6a−m. Among the naphthyl
ethers, 6e, 6i, and 6k emerged as extremely potent NNRTIs.
Before discussing the structure−activity data in more detail, it is
helpful to present the crystallographic results for the complex of
6e with HIV-RT. As detailed in the Supporting Information,
crystals were obtained using the recombinant RT52A enzyme
that diffracted to 2.49 Å on the X29A beamline at the
Brookhaven NSLS. A rendering of the NNRTI binding site
from the structure is provided in Figure 1. Consistent with the
crystal structures for 1a, 1b, and 5a,5,7 key interactions for 6e
include aryl−aryl contacts with Tyr181, Tyr188, and Trp229
a
Reagents and conditions: (a) CuI, Cs2CO3, 2,2,6,6-tetramethyl-3,5-
heptanedione, dioxane, 100 °C, 48 h; (b) BBr3, DCM, −78 °C → 0
°C, 3 h; (c) K2CO3, DMF, 60 °C, 3 h; (d) NH4OH, DCM, 16 h.
B
dx.doi.org/10.1021/ml5003713 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Table 2. Experimental Aqueous Solubility at pH 6.5 (S in
μg/mL) and Computed ClogP
cmpd
S
ClogP
cmpd
S
ClogP
a
1b
1c
5a
5e
6e
10.8
510
3.09
3.38
2.70
3.14
3.30
nev
efv
etv
rpv
rpv
167
2.65
4.67
5.22
5.75
5.75
68.0
b
37.9
43.8
≪1
0.02
0.24
c
d
4.3
a
b
c
d
Reference 12. Reference 13. Reference 14; pH 7. Reference 15;
pH 7.4.
of an angstrom closer to the tyrosine ring than for the left edge
of the naphthalene ring system in Figure 1. Thus, the
interaction between the indolizinyl fragment and Tyr181 may
be more attractive than for naphthyl, and loss of the
interactions in going to Cys181 is more damaging for 5 than
6. Further crystallographic and computational analyses are
being pursued.
Figure 1. Rendering from the crystal structure of 6e with wild-type
HIV-1 reverse transcriptase. Carbon atoms of 6e are colored yellow.
Some residues in front of the ligand have been removed for clarity.
The PDB code is 4WE1.
The activity data for 6e, 6f, 6k, and 6l are particularly
notable. Compounds 6k and 6l are more potent than efavirenz
for all three viral strains, and 6k is essentially identical to
etravirine in potency. The naphthyl ethers are also less
cytotoxic than efavirenz, etravirine, and rilpivirine with excellent
results (CC50 > 100 μM) for the very potent 6e, 6i, 6k, and 6l.
The EC50 results for etravirine and rilpivirine are impressive,
though they are accompanied by extremely low solubility. The
solubility data are summarized in Table 2, which includes
previously unpublished results for 1c and 6e. As noted before,
the normal range for solubility of oral drugs is 4−4000 μg/mL
corresponding to an S of 10−2−10−5 M for a compound with a
molecular weight of 400.16,17 Though 6e barely makes it into
this range, its solubility represents a ca. 40-fold improvement
over etravirine and rilpivirine. The indolizines are an additional
10-fold more soluble, similar to efavirenz. For the cyanovinyl-
phenyl series, a curious result is that, while the difluoro 1b has a
solubility of 11 μg/mL, duplicate measurements for the chloro
analogue 1c confirmed its remarkably high solubility, 510 μg/
mL. The ClogP values for the listed catechol diethers are all
near 3, which is again in the normal range of 0−5 for oral
drugs,18 while etravirine and rilpivirine fall in the 5−6 region.
In summary, catechol diethers that incorporate a 6-cyano-1-
naphthyl substituent have been explored as NNRTIs.
Compounds 6e and 6i are very potent (0.5 nM) inhibitors of
wild-type HIV-1, while 6k has EC50 values below 10 nM for all
three viral strains. In comparison to heterobicyclic alternatives,
the naphthyl ethers show improved performance against the
variant strain of the virus incorporating the Tyr181Cys
mutation in HIV-RT. Compound 6e also shows much
enhanced solubility in comparison to the most recently FDA-
approved NNRTIs, etravirine and rilpivirine, though its
solubility is 10-fold less than for the corresponding indolizine
5a. A crystal structure for the complex of 6e with WT HIV-RT
confirmed the expected binding mode. Compounds such as 6e,
6f, and 6k, which feature sub-20 nM potency toward all three
viral strains and very low cytotoxicity, are promising for further
study.
and the naphthyl fragment, a glancing edge-to-face interaction
between Tyr181 and the catechol ring, and a hydrogen bond
between the C2 carbonyl group of the uracil ring and the
backbone NH of Lys103. The cyano group on C6 of the
naphthalene projects into the channel leading to the RT
polymerase active site; computer simulations using standard
methods4,5 indicate that the cyano nitrogen is hydrogen-
bonded to a water molecule.
The importance of the cyano group is apparent in the WT
activities for 6a, 6b, 6c, and 6e; the unsubstituted 6a is 100-fold
less active than 6e. The difference in the activities of the core
structures is reflected in the data for 1c (0.32 nM), 3c (340), 4b
(19), 5b (3.7), and 6f (2.2), which are identically substituted
with the cyano group and a chlorine. The activity for 1c likely
benefits from the added torsional degree of freedom for the
cyanovinyl side chain, which allows some optimization of the
interaction with Trp229. The torsion angle between the phenyl
ring and vinyl group is 173° and 135° in the crystal structures
for 1a and 1b,7 while in Monte Carlo simulations for the
complex of 1a and HIV-RT it ranges from 130−180° with an
average of 155°.4 The naphthyl compound is then the next
most active, followed by the indolizine. Among the
cyanonaphthalenes, 6e−6m, substitutions with fluorine or
chlorine are not beneficial for WT activity except for the X =
F analogue 6i, which has similar potency to 6e (0.5 nM).
Additional options for substitution are limited owing to the
tight packing in the binding site; aza analogues of the
naphthalene group are also unlikely to be productive as there
are no opportunities for hydrogen bonding with the aza
nitrogens. In the indolizine series, the most potent compound
is 5a, the parent cyano analogue.
Turning to the results for the viral variants, significant gains
have been made with the naphthyl ethers. There are multiple
compounds with EC50 values below 20 nM for the Y181C or
K103N/Y181C variants. The reason for the more than 10-fold
improvements toward the Y181C variant over the results for
identically substituted indolizines appears to be subtle. When
the crystal structures for 5a and 6e are overlaid, there is little
difference except the indolizinyl fragment is tipped up toward
Trp229 more than the naphthyl group. This causes the edge of
the indolizine nearest Tyr181 (C6 and C7) to be a few tenths
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic procedures, NMR and HRMS spectral data for
compounds 1c and 6a−m, and crystallographic details. The
crystal structure data for the complex of 6e and HIV-RT have
C
dx.doi.org/10.1021/ml5003713 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
atitis B Virus and Human Immunodeficiency Virus Type 1 in Vitro.
Biochem. Pharmacol. 1994, 47, 171−174.
(11) Ray, A. S.; Yang, Z.; Chu, C. K.; Anderson, K. S. Novel Use of a
Guanosine Prodrug Approach to Convert 2′,3′-Didehydro-2′,3′-
dideoxyguanosine into a Viable Antiviral Agent. Antimicrob. Agents
Chemother. 2002, 46, 887−891.
(12) Morelock, M. M.; Choi, L. L.; Bell, G. L.; Wright, J. L.
Estimation and Correlation of Drug Water Solubility with Pharmaco-
logical Parameters Required for Biological Activity. J. Pharm. Sci. 1994,
83, 948−952.
(13) Weuts, I.; Van Dycke, F.; Voorspoels, J.; de Cort, S.;
Stokbroekx, S.; Leemans, R.; Brewster, M. E.; Xu, D.; Segmuller, B.;
Turner, Y. T. A.; Roberts, C. J.; Davies, M. C.; Qi, S.; Craig, D. Q. M.;
Reading, M. Physicochemical Properties of the Amorphous Drug, Cast
Films, and Spray Dried Powders to Predict Formulation Probability of
Success for Solid Dispersions: Etravirine. J. Pharm. Sci. 2011, 100,
260−274.
(14) Janssen, P. A. J.; Lewi, P. J.; Arnold, E.; Daeyaert, F.; de Jonge,
M.; Heeres, J.; Koymans, L.; Vinkers, M.; Guillemont, J.; Pasquier, E.;
Kukla, M.; Ludovici, D.; Andries, K.; de Bethune, M.-P.; Pauwels, R.;
́
been deposited in the RCSB Protein Data Bank with the PDB
code 4WE1. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*(K.S.A.) E-mail: karen.anderson@yale.edu.
*(W.L.J.) E-mail: william.jorgensen@yale.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Gratitude is expressed to the National Institutes of Health
(AI44616, GM32136, GM49551) for research support and for
a fellowship to KMF (AI104334). We also thank the National
Synchrotron Light Source at Brookhaven National Laboratory
for beam time and assistance on X29A.
Das, K.; Clark, A. D., Jr.; Frenkel, Y. V.; Hughes, S. H.; Medaer, B.; De
Knaep, F.; Bohets, H.; De Clerck, F.; Lampo, A.; Williams, P.; Stoffels,
P. In Search of a Novel Anti-HIV Drug: Multidisciplinary
Coordination in the Discovery of 4-[[4-[[4-[(1E)-2-Cyanoethenyl]-
2,6-dimethylphenyl]amino]-2-pyrimidinyl]amino]benzonitrile
(R278474, Rilpivirine). J. Med. Chem. 2005, 48, 1901−1919.
(15) Sun, L.-Q.; Qin, B.; Huang, L.; Qian, K.; Chen, C.-H.; Lee, K.-
H.; Xie, L. Optimization of 2,4-Diarylanilines As Non-Nucleoside
HIV-1 Reverse Transcriptase Inhibitors. Bioorg. Med. Chem. Lett. 2012,
22, 2376−2379.
ABBREVIATIONS
■
HIV, human immunodeficiency virus; HIV-RT, HIV reverse
transcriptase; NNRTI, non-nucleoside inhibitor of HIV-RT; Bz,
benzoyl; DCM, dichloromethane; DMF, dimethylformamide;
HPLC, high-performance liquid chromatography
REFERENCES
■
(1) De Clercq, E. The Nucleoside Reverse Transcriptase Inhibitors,
Nonnucleoside Reverse Transcriptase Inhibitors, and Protease
Inhibitors in the Treatment of HIV Infections (AIDS). Adv. Pharmacol.
2013, 67, 317−358.
(2) Reynolds, C.; de Koning, C. B.; Pelly, S. C.; van Otterlo, W. A. L.;
Bode, M. L. In Search of a Treatment for HIV: Current Therapies and
the Role of Non-Nucleoside Reverse Transcriptase Inhibitors
(NNRTIs). Chem. Soc. Rev. 2012, 41, 4657−4670.
(16) Jorgensen, W. L.; Duffy, E. M. Prediction of Drug Solubility
from Structure. Adv. Drug Delivery Rev. 2002, 54, 355−366.
(17) Bollini, M.; Cisneros, J. A.; Spasov, K. A.; Anderson, K. S.;
Jorgensen, W. L. Optimization of Diarylazines As Anti-HIV Agents
with Dramatically Enhanced Solubility. Bioorg. Med. Chem. Lett. 2013,
23, 5213−5216.
(18) Jorgensen, W. L. Efficient Drug Lead Discovery and
Optimization. Acc. Chem. Res. 2009, 42, 724−733.
(3) Zhan, P.; Chen, X.; Li, D.; Fang, Z.; De Clercq, E.; Liu, X. HIV-1
NNRTIs: Structural Diversity, Pharmacophore Similarity, and
Implications for Drug Design. Med. Res. Rev. 2013, 33, E1−E72.
(4) Bollini, M.; Domaoal, R. A.; Thakur, V. V.; Gallardo-Macias, R.;
Spasov, K. A.; Anderson, K. S.; Jorgensen, W. L. Computationally-
Guided Optimization of a Docking Hit to Yield Catechol Diethers as
Potent Ani-HIV Agents. J. Med. Chem. 2011, 54, 8582−8591.
(5) Lee, W.-G.; Gallardo-Macias, R.; Frey, K. M.; Spasov, K. A.;
Bollini, M.; Anderson, K. S.; Jorgensen, W. L. Picomolar Inhibitors of
HIV Reverse Transcriptase Featuring Bicyclic Replacement of a
Cyanovinylphenyl Group. J. Am. Chem. Soc. 2013, 135, 16705−16713.
́
(6) de Bethune, M.-P. Non-nucleoside Reverse Transcriptase
Inhibitors (NNRTIs), Their Discovery, Development, and Use in
the Treatment of HIV-1 Infection: A Review of the Last 20 Years
(1989−2009). Antiviral Res. 2010, 85, 75−90.
(7) Frey, K. M.; Bollini, M.; Mislak, A. C.; Cisneros, J. A.; Gallardo-
Macias, R.; Jorgensen, W. L.; Anderson, K. S. Crystal Structures of
HIV-1 Reverse Transcriptase with Picomolar Inhibitors Reveal Key
Interactions for Drug Design. J. Am. Chem. Soc. 2012, 134, 19501−
19503.
(8) Frey, K. M.; Gray, W. T.; Spasov, K. A.; Bollini, M.; Gallardo-
Macias, R.; Jorgensen, W. L.; Anderson, K. S. Structure-Based
Evaluation of C5 Derivatives in the Catechol Diether Series Targeting
HIV-I Reverse Transcriptase. Chem. Biol. Drug Des. 2014, 83, 541−
549.
́ ́
(9) Baka, E.; Comer, J. E. A; Takacs-Novak, K. Study of Equilibrium
Solubility Measurement by Saturation Shake-Flask Method Using
Hydrochlorothiaide as Model Compound. J. Pharm. Biomed. Anal.
2008, 46, 335−341.
(10) Lin, T. S.; Luo, M. Z.; Liu, M. C.; Pai, S. B.; Dutschman, G. E.;
Cheng, Y. C. Antiviral Activity of 2′,3′-Dideoxy-β-L-5-fluorocytidine
(β-L-EddC) and 2′,3′-Dideoxy-β-L-cytidine (β-L-ddC) against Hep-
D
dx.doi.org/10.1021/ml5003713 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX