Selective small molecules blocking HIV-1 tat and coactivator PCAF association

Article abstract of DOI:10.1021/ja044885g

Development of drug resistance from mutations in the targeted viral proteins leads to continuation of viral production by chronically infected cells, contributing to HIV-mediated immune dysfunction. Targeting a host cell protein essential for viral reproduction, rather than a viral protein, may minimize the viral drug resistance problem as observed with HIV protease inhibitors. We report here the development of a novel class of N1-aryl-propane-1,3-diamine compounds using a structure-based approach that selectively inhibit the activity of the bromodomain of the human transcriptional co-activator PCAF, of which association with the HIV trans-activator Tat is essential for transcription and replication of the integrated HIV provirus. Copyright

Full text of DOI:10.1021/ja044885g

Published on Web 02/04/2005
Selective Small Molecules Blocking HIV-1 Tat and Coactivator PCAF
Association
Lei Zeng,‡ Jiaming Li,† Michaela Muller,‡ Sherry Yan,‡ Shiraz Mujtaba,‡ Chongfeng Pan,†
Zhiyong Wang,*,† and Ming-Ming Zhou*,‡
Structural Biology Program, Department of Physiology and Biophysics, Mount Sinai School of Medicine,
New York UniVersity, One GustaVe L. LeVy Place, New York, New York 10029-6574, and Department of Chemistry,
UniVersity of Science and Technology of China, Hefei, Anhui 230026, P.R. China
Received August 24, 2004; E-mail: ming-ming.zhou@mssm.edu; zwang3@ustc.edu.cn
The replication cycle of the human immunodeficiency virus
(HIV) presents several viable targets for anti-HIV chemotherapy.
The current anti-HIV drugs specifically target the viral reverse
transcriptase, protease and integrase.1 However, because of the
development of viral drug resistance from mutations in the targeted
proteins, continuous viral production by chronically infected cells
contributes to HIV-mediated immune dysfunction,2 and there is still
no cure for AIDS. A rapid growing AIDS epidemic calls for new
therapeutic strategies targeting different steps in the viral life cycle.
Therapeutic intervention at the stage of HIV gene expression can
complement the existing therapy to interfere with virus production.
Transcription of the integrated HIV provirus is regulated by the
concerted action between cellular transcription factors and a unique
viral trans-activator Tat. Tat binds to a viral RNA TAR and recruits
cyclin T1 and cyclin-dependent kinase 9 that hyperphosphorylates
and enhances elongation efficiency of the RNA polymerase II.3
Tat transactivation requires acetylation of its lysine 50 and
recruitment of histone lysine acetyltransferase transcriptional co-
activators for remodeling the nucleosome that contains the integrated
proviral DNA.4 Our recent study shows that Tat coactivator
recruitment requires its acetylated lysine 50 (AcK50) binding to
the bromodomain (BRD) of the coactivator PCAF,5a and microin-
jection of anti-PCAF BRD antibody blocks Tat transactivation.5b
These data suggest that Tat/PCAF recruitment via a BRD-AcK
binding is essential for HIV transcription, and this interaction serves
as a new therapeutic target for intervening HIV replication.
To validate this possible new anti-HIV therapeutic target, we
aim to develop small-molecule inhibitors that block Tat/PCAF
binding by targeting the BRD of PCAF. Targeting a host cell protein
essential for viral reproduction, rather than a viral protein, may
minimize the problem of drug resistance due to mutations of the
viral counterpart as observed with protease inhibitors. Here we
report the development of a novel class of N1-aryl-propane-1,3-
diamine compounds using a structure-based approach that binds
the PCAF BRD selectively over other structurally similar BRDs.
The bromodomain, present in chromatin-associated proteins and
histone lysine acetyltransferases,6a is an acetyl-lysine binding
domain.6b Bromodomain/AcK binding plays an important role in
control of chromatin remodeling and gene transcription.6c BRDs
adopt the highly conserved structural fold of a left-handed four-
helix bundle (RZ, RA, RB, and RC), as first shown in the PCAF
BRD 6b (Figure 1A). The ZA and BC loops at one end of the bundle
form a hydrophobic pocket for AcK binding. The structure of the
PCAF BRD bound to a Tat-AcK50 peptide5a shows that AcK50
interacts with protein residues V752, Y802 and Y809, Y47(AcK-
3) with V763, and R53(AcK+3) and Q54(AcK+4) with E756,
Figure 1. Structural basis of ligand recognition of the PCAF BRD. (A)
Tat-AcK50 peptide recognition by the PCAF BRD. (B) Superimposition
of the BRD structure in free (gray) and bound to the compound 2 (yellow).
(C) Structure of the BRD/2 complex, showing the compound 2 binding
site. (D) GRASP view of the compound 2 binding pocket.
conferring a specific intermolecular association. The structures of
CBP BRD/p53-AcK382 and GCN5p BRD/H4-AcK16 complexes7
show that the residues in BRDs important for AcK recognition are
largely conserved, whereas sequence variations in the ZA and BC
loops enable discrimination of different binding targets. Notably,
as compared to the other parts of the protein, the ZA and BC loops
contain significant sequence variations with amino acid deletion
or insertion, supporting the notion that different sets of residues in
the ZA and/or BC loops dictate BRD ligand specificity by
interacting with residues flanking the acetyl-lysine in a target
protein.6c
To develop selective small-molecule inhibitors for blocking Tat/
PCAF association, we conducted NMR-based chemical screening
for the BRD by monitoring ligand-induced protein signal changes
in 2D 15N HSQC spectra.8 We placed an emphasis on identifying
compounds that bind selectively the BRD near but not just the AcK
binding pocket, as the former may be more selective for this BRD.
From screening of a few thousands of small-molecules from
commercial libraries, we discovered several compounds including
1 that meet this criterion. Compound 1 binds the PCAF BRD with
an affinity comparable to that of the Tat-AcK50 peptide7a (see
below). Importantly, these compounds do not bind the structurally
similar BRDs from CBP and TIF1â at millimolar concentration.
We next synthesized a series of 1 analogues to probe the
structure-activity relationship (SAR) (Table 1). We assessed their
binding to the PCAF BRD by measuring an IC50 in an ELISA assay,
in which a compound competes against a biotinylated Tat-AcK50
peptide for binding to the GST-fusion BRD immobilized to
‡ Mount Sinai School of Medicine
† University of Science and Technology of China
9
2376
J. AM. CHEM. SOC. 2005, 127, 2376-2377
10.1021/ja044885g CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Structure-Activity Relationship Data for Derivatives of 1
cmpd
R1
X
n
R2
IC50 (µM)
+
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
2-NO2
2-NO2, 4-CH3
2-NO2, 4-CH2-CH3 NH
2-NO2, 3-CH3
2-NO2, 5-CH3
2-NO2, 4-Ph
2-NO2, 4-CN
2-NO2, 5-CN
2-CH3, 5-NO2
2-COO-
NH
NH
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
2
4
3
2
2
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3+
-NH3
5.1 ( 0.1
1.6 ( 0.1
7.2 ( 0.1
5.9 ( 0.1
10.8 ( 0.1
Figure 2. Inhibition of PCAF BRD/Tat-AcK50 binding by 2. In this assay,
2 inhibits the biotinylated Tat AcK50 peptide immobilized on streptavidin-
agarose by binding to the GST-PCAF BRD, as assessed by anti-GST
Western blot. Lower panel indicates an equal amount of BRD used in each
assay.
NH
NH
NH
NH
NH
NH
NH
NH
O
O
O
O
O
>10,000
34.9 ( 0.1
63.4 ( 0.6
77.8 ( 0.4
1). The aromatic ring of 2 is sandwiched between the side chains
of Y802 and A757 on one side and Y809 and E756 on the other
side, and the propane carbon chain is surrounded by the hydrophobic
portions of the side chains of P747, E756, and V752. Finally, the
4-methyl of 2 fills a small hydrophobic cavity formed by side chains
of A757, Y802, and Y809 (Figure 1D), contributing to a 3-fold
increase over 1 in binding to the PCAF BRD. Notably, out of the
ligand-interacting residues, only Y802 is among the conserved
residues at the AcK binding site in BRDs, thus explaining the
selective binding by this class of compounds to the PCAF BRD
over the structurally similar CBP and TIF1â BRDs.
>10,000
>10,000
125.6 ( 0.6
2-COOCH3
2-NO2
2-NO2, 4-CH3
2-NO2, 4-CH3O
2-NO2, 4-Cl
2-NO2, 5-CH3
2-NO2, 3-CH3
2-NO2
2-NO2
4-NO2
4-NO2
180.0 ( 0.5
102.7 ( 0.4
215.1 ( 0.6
203.6 ( 0.6
164.8 ( 0.8
O
CH2
NH
NH
NH
NH
NH
NH
>10,000
145.9 ( 0.7
>2,000
>10,000
>2,000
3-NH2, 4-NO2
2-NO2, 4-Cl
2-Cl, 4-NO2
-COO-
-(OH)CH3 >10,000
-(OH)CH3 >10,000
In summary, we have developed a class of novel small molecules
that can effectively inhibit the PCAF BRD/Tat-AcK50 association
in vitro by selectively binding to the BRD (Figure 2). The detailed
SAR understanding of the lead compounds 1 and 2 will facilitate
our efforts to optimize their affinity and selectivity by branching
out to interact with the neighboring AcK binding pocket by the
tethering techniques.9 Such small-molecule inhibitors will help
validate the novel anti-HIV/AIDS therapeutic strategy by targeting
a cellular protein to block HIV transcription and replication.
glutathione-coated 96-well microtiter plate. The SAR study reveals
salient features of BRD recognition of 1. First, the BRD prefers a
4-methyl group on the aniline ring, which improves IC50 by 3-fold
to 1.6 µM (2 vs 1). While substitution of a 4-ethyl, 3- or 5-methyl
group on the aniline ring slightly weakens the binding (3-5 vs.
1), addition of a 4-phenyl group nearly abolishes the binding (6 vs
1). Adding a 4- or 5-cyano group weakens the binding by ∼7-
12-fold (7 and 8 vs 1). Second, a 2-nitro group on the aniline ring
is vital for the binding. Swapping of 2-nitro and 5-methyl causes
a 7-fold reduction in binding (9 vs 5). Surprisingly, substitution of
2-nitro with 2-caroxylate or 2-caroxyl ester abrogates the binding
(10 and 11 vs 1). Third, the functional importance of the 2-nitro is
further supported by the effects of changing the NH to an O linkage
in the aniline, which severely compromises the binding to the PCAF
BRD (12-17 vs 1-5). Moreover, changing to a carbon linkage
eliminates the binding (18 vs 1). Fourth, the BRD prefers an amino
three-carbon aliphatic chain in 1sa four-carbon chain reduces the
binding by 30-fold (19 vs 1) and a two-carbon chain nearly loses
the binding (20 vs 1). Alteration of 1 by two key elements, i.e.
changing to a four-carbon chain and 4-nitro, abolishes the binding
(21 vs 1). Finally, the terminal amine group is also an important
functional moiety for the BRD binding (22-24 vs 1).
To understand ligand selectivity of the PCAF BRD, we solved
the 3D structures of the protein bound to 1 and 2. The two ligands
are bound in the protein structure in nearly the same manner. For
clarity, only the 2-bound structure is reported here, which is similar
to the free structure except for the ZA and BC loops that move
closer to each other by clamping onto the ligand (Figure 1B). 2 is
engulfed by residues in the ZA and BC loops outside the AcK
binding pocket, blocking BRD binding to AcK of a target protein
(Figure 1C). The 2-nitro group of 2 possibly forms a hydrogen
bond with the phenolic -OH of Y809 and/or Y802, and the terminal
-NH3+ interacts eletrostatically with the side-chain carboxylate of
E750. The functional importance of the 2-nitro and the aniline NH
likely results from a possible six-member ring structure formed
between these two groups. However, it is not clear why substitution
of 2-nitro with 2-carboxylate abrogates the BRD binding (10 vs
Acknowledgment. We thank the National Institutes of Health
(to M.-M.Z.) for financial support of this work.
Supporting Information Available: Full experimental procedures.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Garg, R.; Gupta, S. P.; Hua, G.; Babu, M. S.; Debnath, A. K.; Hansch, C.
Chem. ReV. 1999, 99, 3525-3601.
(2) (a) Ho, D. D.; Zhang, L. Nat. Med. 2000, 6, 757-761. (b) Wei, X. et al.
Nature 1995, 373, 117-122. (c) Richman, D. D. AIDS Res. Hum.
RetroVirus 1992, 8, 1065-1071.
(3) (a) Keen, N. J.; Churcher, M. J.; Karn, J. EMBO J. 1997, 16, 5260-
5272. (b) Karn, J. J. Mol. Biol. 1999, 293, 235-254. (c) Jones, K. A.
Genes. DeV. 1997, 11, 2593-2599. (d) Kao, S. Y.; Calman, A. F.; Luciw,
P. A.; Peterlin, B. M. Nature 1987, 330, 489-493.
(4) (a) Ott, M.; Schnolzer, M.; Garnica, J.; Fischle, W.; Emiliani, S.; Rackwitz,
H. R.; Verdin, E. Curr. Biol. 1999, 9, 1489-1492. (b) Kaehlcke, K. et al.
Mol. Cell 2003, 12, 167-176.
(5) (a) Mujtaba, S.; He, Y.; Zeng, L.; Farooq, A.; Carison, J. E.; Ott, M.;
Verdin, E.; Zhou, M.-M. Mol. Cell. 2002, 9, 575-586. (b) Dorr, A.;
Kiermer, V.; Pedal, A.; Rackwitz, H. R.; Henklein, P.; Schubert, U.; Zhou,
M.-M.; Verdin, E.; Ott, M. EMBO J. 2002, 21, 2715-2733.
(6) (a) Jeanmougin, F.; Wurtz, J. M.; Douarin, B. L.; Chambon, P.; Losson,
R. Trends Biochem. Sci. 1997, 22, 151-153. (b) Dhalluin, C.; Carlson, J.
E.; Zeng, L.; He, C.; Aggarwal, A. K.; Zhou, M.-M. Nature 1999, 399,
491-496. (c) Zeng, L.; Zhou, M.-M. FEBS Lett. 2002, 513, 124-128.
(7) (a) Mujtaba, S. et al. Mol. Cell. 2004, 13, 251-263. (b) Owen, D. J.;
Ornaghi, P.; Yang, J. C.; Lowe, N.; Evans, P. R.; Ballario, P.; Neuhaus,
D.; Filetici, P.; Travers, A. A. EMBO J. 2000, 19, 6141-6149.
(8) (a) Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Q. ReV. Biophys. 1999,
32, 211-240. (b) Moore, J. M. Curr. Opin. Biotechnol. 1999, 10, 54-
58.
(9) (a) Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Science 1997, 278, 498-
9. (b) Erlanson, D. A.; Braisted, A. C.; Raphael, D. R.; Randal, M.; Stroud,
R. M.; Gordon, E. M.; Wells, J. A. Proc. Natl. Acad. Sci. U.S.A. 2000,
97, 9367-9372.
JA044885G
9
J. AM. CHEM. SOC. VOL. 127, NO. 8, 2005 2377