A specific protein disorder catalyzer of HIV-1 Nef

Article abstract of DOI:10.1016/j.bmc.2011.10.051

The HIV-1 auxiliary protein Nef is required for the onset and progression of AIDS in HIV-1-infected persons. Here, we have deciphered the mode of action of a second-generation inhibitor of Nef, DLC27-14, presenting a competitive IC₅₀ of ～16 μM measured by MALDI-TOF experiments. Thermal protein denaturation experiments revealed a negative effect on stability of Nef in the presence of a saturating concentration of the inhibitor. The destabilizing action of DLC27-14 was confirmed by a HIV protease-based experiment, in which the protease sensitivity of DLC27-14-bound Nef was three times as high as that of apo Nef. The only compatible docking modes of action for DLC27-14 suggest that DLC27-14 promotes an opening of two α-helices that would destabilize the Nef core domain. DLC27-14 thus acts as a specific protein disorder catalyzer that destabilizes the folded conformation of the protein. Our results open novel avenues toward the development of next-generation Nef inhibitors.

Full text of DOI:10.1016/j.bmc.2011.10.051

Bioorganic & Medicinal Chemistry 19 (2011) 7401–7406
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
A specific protein disorder catalyzer of HIV-1 Nef
Adrien Lugari a, Sebastian Breuer b, Thibault Coursindel c, Sandrine Opi d,e,f, Audrey Restouin d,e,f
,
Xiaoli Shi a,g, Alexis Nazabal h, Bruce E. Torbett b, Jean Martinez c, Yves Collette d,e,f, Isabelle Parrot c,
Stefan T. Arold i, , Xavier Morelli
a,
⇑
⇑
a IMR Laboratory, CNRS-UPR 3243, Centre National de la Recherche Scientifique, Institut de Microbiologie de la Méditerranée (IMM) and Aix-Marseille Universités,
31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France
b Department of Molecular and Experimental Medicine, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA
c Institut des Biomolécules Max Mousseron (IBMM), UMR 5247 CNRS-Université Montpellier 1–Université Montpellier 2, CC17-03, Place Eugène Bataillon, 34095 Montpellier
Cedex 5, France
d INSERM U891, Centre de Recherche en Cancérologie de Marseille, Marseille F-13009, France
e Institut Paoli-Calmettes, Marseille F-13009, France
f Université de la Méditerranée, Marseille F-13007, France
g State Key Laboratory of Structural Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, 155 Yang Qiao Xi Lu, Fuzhou, Fujian 350002, China
h CovalX AG, Grabenstrasse 11A, Schlieren CH-8952, Switzerland
i Department of Biochemistry and Molecular Biology, Unit 1000, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030-4009, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The HIV-1 auxiliary protein Nef is required for the onset and progression of AIDS in HIV-1–infected per-
sons. Here, we have deciphered the mode of action of a second-generation inhibitor of Nef, DLC27-14,
Received 31 August 2011
Revised 11 October 2011
Accepted 17 October 2011
Available online 20 October 2011
presenting a competitive IC50 of ꢀ16
lM measured by MALDI-TOF experiments. Thermal protein dena-
turation experiments revealed a negative effect on stability of Nef in the presence of a saturating concen-
tration of the inhibitor. The destabilizing action of DLC27-14 was confirmed by a HIV protease-based
experiment, in which the protease sensitivity of DLC27-14–bound Nef was three times as high as that
of apo Nef. The only compatible docking modes of action for DLC27-14 suggest that DLC27-14 promotes
Keywords:
HIV-1 Nef inhibitor
Molecular modeling
Thermostability
Mass spectrometry
Protein–protein interaction inhibitor
an opening of two a-helices that would destabilize the Nef core domain. DLC27-14 thus acts as a specific
protein disorder catalyzer that destabilizes the folded conformation of the protein. Our results open novel
avenues toward the development of next-generation Nef inhibitors.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
56–206), a stable globular domain that contains a proline-rich mo-
tif, which allows the binding of Src homology 3 (SH3) kinase do-
During HIV-1 infection, HIV-1 auxiliary proteins such as Nef
interact with many host proteins. These host–pathogen protein–
protein interactions (PPIs) contribute crucially to the efficiency of
viral infection, pathogenicity, and disease progression.1–4 Numer-
ous cellular partners of Nef have been identified, and they can be
divided into two groups: proteins involved in the trafficking of cell
surface receptors and proteins involved in cell signaling pathways.5
Most of these PPIs involve binding to the Nef core domain (residues
mains.6 The biological properties of the Nef:SH3 domain complex,
as well as the structural data available for this interaction7,8, have
led Nef to be considered as a suitable model for a structure-based
drug design program.6
In this emerging field of PPI inhibition to target host–pathogen
interactions, we identified in 2007 the first PPI inhibitor targeting
the HIV-1 Nef:SH3 interaction.9 This compound, DLC27 (Supple-
mentary Fig. S1, panel a), was developed by coupling a virtual
screening method to medium-throughput biological tests, the so-
called 2P2I (protein–protein interaction inhibition) method.
According to our previous analysis9, DLC27 competes for the SH3
domain through specific binding to the region of Nef that binds
to the SH3 RT loop (Supplementary Fig. S1, panel b). This RT loop
binding region includes a mostly hydrophobic groove delimited
by Nef residue Trp113.
Abbreviations: HIV-1, human immunodeficiency virus-1; AIDS, acquired
immune deficiency syndrome; PPIs, protein–protein interactions; SH3, Src homol-
ogy 3; ITC, isothermal titration calorimetry; SPR, surface plasmon resonance;
MALDI-TOF MS, matrix-assisted laser desorption/ionization-time-of-flight mass
spectrometry; NMR, nuclear magnetic resonance; HSQC, heteronuclear single
quantum coherence spectroscopy; STD, saturation transfer difference; DSLS,
differential static light scattering.
⇑
Corresponding authors. Tel.: +33 (0)491 164 501; fax: +33 (0)491 164 540
(X.M.); tel./fax: +1 713 792 6059 (S.T.A.).
In the present work, we used numerous biological and
biophysical methods to test more than 30 DLC27 derivatives.
Among these, we identify DLC27-14 as a promising and unusual
E-mail addresses: starold@mdanderson.org (S.T. Arold), morelli@ifr88.cnrs
-mrs.fr (X. Morelli).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.10.051

7402
A. Lugari et al. / Bioorg. Med. Chem. 19 (2011) 7401–7406
second-generation inhibitor of HIV-1 Nef, because DLC27-14 acts
as a specific protein disorder catalyzer that destabilizes the folded
conformation of the Nef core domain. Thus DLC27-14 could act as
a potent antiviral compound.
the proteins (Fig. 1, panel a). Adding DLC27 thus partially inhibits
formation of the Nef:SH3-Hck complex. At the same concentration,
one derivative compound, DLC27-14, totally inhibited formation of
the Nef:SH3-Hck complex (Fig. 1, panel b). DLC27-14 therefore acts
as a more potent inhibitor of Nef:SH3-Hck complex formation than
does DLC27. To confirm these results, we performed several exper-
iments at different compound concentrations and derived an IC50
2. Results and discussion
of 16 lM for the inhibition of the Nef:SH3-Hck complex by
DLC27-14 (Fig. 1, panel c). The amount of DLC27 needed to derive
an IC50 value was significantly higher (not saturating at a concen-
2.1. Discovery of potential lead compounds from DLC27
derivatives
tration of 100 lM), indicative of an IC50 >50 lM.
DLC27 is a dual-head compound bearing a polar head with a
carboxylic acid group and an apolar head composed of a tert-butyl
in the para position of a phenyl cycle. The heads are separated by
a carbamate-type linker (Supplementary Fig. S1, panela). According
to the binding model for DLC27 derived previously9, the apolar head
of the compound inserts itself into the hydrophobic RT loop binding
region of the SH3 domain binding surface of Nef (Supplementary
Fig. S1, panel b, region within gray dashed line). This hydrophobic
groove is delimited by Nef residue Trp113 and accommodates resi-
due 96 of the SH3 RT loop (according to the sequence of the Fyn SH3
domain)7,8, which is critical for the Nef:SH3 interaction.10 Indeed,
Nef has a canonical polyproline SH3 recognition sequence, but addi-
tional contacts occur with the RT loops of the SH3 domains of the Src
family kinases Hck and Fyn, in particular with SH3 residue 96 (using
Fyn numbering). This tertiary recognition enhances significantly the
specificity and affinity of these SH3 domains for HIV-1 Nef.10–12 The
Nef binding pocket for SH3 residue 96 is therefore a hot spot for the
Nef:SH3 interaction and an attractive target for inhibitors.
We identified an additional hydrophobic pocket in the structure
of the Nef:SH3 complex, adjacent to the binding pocket for residue
96. (Supplementary Fig. S1, panel b, region within red dashed line).
We therefore decided to synthesize new DLC27 derivatives that
could interact with this new pocket by an extension of the apolar
head of DLC27. We expected this additional interaction to result
in molecules that have higher affinity and specificity for Nef than
DLC27.
DLC27-14 has a second phenyl cycle, which extends the apolar
head of DLC27, as shown in Figure 2, panel a. The chemical path-
way used to synthesize DLC27-14 is presented in Figure 2, panel b.
2.3. In silico modeling of DLC27 derivatives bound to Nef
To structurally characterize the binding mode of DLC27-14 on
Nef, we set up nuclear magnetic resonance (NMR) experiments
and crystallization assays. Those experiments, including NMR het-
eronuclear single quantum coherence spectroscopy (HSQC) and
saturation transfer difference (STD) experiments as well as soaking
and co-crystallization experiments, failed to yield exploitable data
because the presence of the compound in DMSO induced Nef pre-
cipitation in solution or melting of Nef crystals even at less than
equimolar ratios. We therefore used in silico docking methods to
obtain structural insights into the Nef:DLC27-14 complex.
Figure 3, panel a shows the binding model of DLC27 on Nef, pro-
posed by Betzi et al9 Surprisingly, using the same docking program
and protein files, we could not dock DLC27-14 onto the Nef surface.
Effectively, in our DLC27 binding model the apolar head group of
DLC27 pushes against a hydrophobic surface contributed by
Trp113 and, to a lesser extent, Phe90. DLC27-14 is too long to fit
into the RT-loop binding pocket as does DLC27. However DLC27-
14 is also incapable of reaching over the barrier established by
the side chain of Trp113 to attain the second adjacent hydrophobic
pocket. Thus, to allow fitting DLC27-14 onto the Nef surface, the
Trp113 side chain has to move away. This is plausible because
Trp113 has an above-average mobility in the crystal structure of
the apo-Nef core domain (mean Trp113 B-factor = 76.8 Å2 com-
pared with an overall B-factor = 37.5 Å2 for all atoms of the struc-
ture).7 From the structure, there are two ways for the Trp113
side chain to move away: (i) by rotating towards the second hydro-
phobic pocket. This movement is plausible because it corresponds
to only a subtle exaggeration of the movement that the Trp113 side
chain undergoes to accommodate the bulky Fyn Arg96 residues.7
(ii) By using a different rotamer, which rotates the Trp113 side
chain out of the pocket. A suitable rotamer corresponds to the sec-
ond most probable conformation of the residue in solution, accord-
ing to the SYBYL tools (Tripos). Both plausible Trp113 side chain
conformations create enough space to allow docking of DLC27-
14. In both cases the apolar extension of DLC27-14 reaches into
the adjacent second hydrophobic pocket of Nef while the polar
head of the compound stills occupies the SH3 RT loop binding site.
In doing so, DLC27-14 disrupts a hydrophobic connection between
the side chains of Trp113 and Phe90. We therefore hypothesized
that DLC27-14 (as well as DLC27, to some extent) catalyzes Nef
unfolding by destabilizing a large hydrophobic contact of its core
domain.
We synthesized 30 DLC27 derivatives, named DLC27-1 to
DLC27-30, by replacing the apolar head of DLC27 with various
longer chemical groups with hydrophobic properties. All the com-
pounds were tested in cell-based assays, as previously described by
Betzi et al9 Unfortunately, all the newly synthesized compounds
had very low solubility (which is expected according to their calcu-
lated logD values), and thus no useful data could be obtained from
these assays.
2.2. DLC27-14 is an improved inhibitor of the Nef:SH3-Hck
complex
Because of the very low solubility, we could not obtain dissoci-
ation constants (Kd) for the compounds using classical biophysical
methods such as isothermal titration calorimetry (ITC) or surface
plasmon resonance (SPR). We therefore developed a mass-spec-
troscopy (MS)–based method to assess the capacity of the most
promising DLC27 derivatives to inhibit the Nef:SH3-Hck complex.
This technique, based on MALDI-TOF MS, enabled us to assess
the potency of an inhibitor to block the formation of a protein com-
plex in the presence of a bridging agent. This method successfully
produced binding data, because it is relatively insensitive to DMSO
(which is needed to solubilize the compounds) and to protein
destabilization.
When applied to the Nef:SH3-Hck complex, we detected three
major peaks, which were assigned to Nef and SH3-Hck in mono-
meric forms and to the Nef:SH3-Hck complex (Fig. 1, panel a). In
the presence of DLC27, we observed a decrease in the peak corre-
sponding to the Nef:SH3-Hck complex, which was associated with
increases in the peaks corresponding to the monomeric forms of
2.4. Characterization of DLC27 and DLC27-14 mode of Inhibition
To decipher the mode of action of the compounds, we used a
differential static light scattering (DSLS) based thermal protein
denaturation assay on Nef in complex with DLC27 and DLC27-14
molecules. Interestingly, DLC27 in DMSO strongly decreased the

A. Lugari et al. / Bioorg. Med. Chem. 19 (2011) 7401–7406
7403
Figure 1. Inhibition of the Nef:SH3-Hck complex, assessed by high-mass MALDI-TOF. (a) High-mass MALDI-TOF spectrum of the Nef:SH3-Hck protein complex before (blue)
and after (red) 1 h of incubation with DLC27 100ꢁ. The complex was detected with MH+ = 26.254 kDa. The two subunits forming the complex were also detected, with
MH+ = 6.891 kDa and MH+ = 18.849 kDa (b) High-mass MALDI-TOF spectrum of the Nef:SH3-Hck protein complex before (blue) and after (red) 1 h of incubation with DLC27-
14 100ꢁ. The complex was detected with MH+ = 26.189 kDa. The two subunits forming the complex were also detected, with MH+ = 6.938 kDa and MH+ = 18.964 kDa. (c)
Effect of DLC27-14 concentration on the percentage of binding for the Nef:SH3-Hck protein complex. The percentage of protein complex binding obtained by high-mass
MALDI-TOF experiments was plotted as a function of inhibitor concentration (logarithmic scale) to determine the IC50 value of DLC27-14.
aggregation temperature of Nef by ꢀ10° compared with a control
with Nef and DMSO only (Fig. 4, panel a). DLC27 thus already
destabilizes Nef. As DLC27-14 had very low solubility and a pro-
pensity to precipitate Nef in solution, we could not obtain DSLS
data for this compound even at very low temperatures. Because
of its similarity to DLC27 and our binding and modeling data, we
expected that the DLC27-14 also destabilizes Nef.
compared how HIV-1 protease cleaves Nef at the Trp57/Leu58
HIV protease cleavage site in presence or absence of DLC27 and
DLC27-14 using a cleavage enzyme-cytometric bead array (CE-
CBA) experiment (Fig. 4b and c and Supplementary Fig. S2). This
substrate cleavage assay relies on native Nef protein expressed as
a fluorescent reporter fusion protein (GST-Nef-mVenus) coupled
to anti-GST antibody coated beads. Nef cleavage by the HIV prote-
ase is then followed through the loss of fluorescence on the beads
utilizing flow cytometry.13 The CE-CBA allows precise determina-
tion of alterations in protease catalytic efficiency and facilitates
the identification of small molecule inhibitors of the protease or
its substrate Nef. Structural analysis shows that the Trp57/Leu58
2.5. Proteolytic sensitivity to HIV protease
A significant destabilization of the Nef core domain is expected
to increase the sensitivity of Nef to proteolysis. We therefore

7404
A. Lugari et al. / Bioorg. Med. Chem. 19 (2011) 7401–7406
Figure 2. DLC27-14 structure and synthesis. (a) Structure of DLC27-14. The gray circle represents the additional phenyl cycle compared with DLC27. (b) Chemical synthesis
pathway of the most potent derivatives of DLC27.
cleavage site interacts with the second hydrophobic pocket in the
Nef core domain in proximity of Trp113.14,15 We therefore rea-
soned that destabilization of the Nef core by Trp113 conforma-
tional changes exposes the HIV protease cleavage site. Indeed,
Nef stability and enhanced proteolytic sensitivity. In silico molecu-
lar modeling suggested that DLC27-14 (and to a lesser extend
DLC27) destabilizes the Nef core domain by disrupting a hydropho-
bic barrier established by Trp113. Because of this solubility issue,
we did not succeed to measure a biological activity for our DLC27-
DLC27 accelerated Nef proteolysis, with EC50 = 109.1 1.9
DLC27-14 demonstrated an improved capacity to catalyze Nef pro-
teolysis, with EC50 = 34.3 3.7 M. These results agree with those
lM.
X series, except for slight cell toxicity. Indeed, up to 40
lM DLC27
l
only reduced cell viability to 86.6%, while 40 M of DLC27-14 re-
l
from our MS experiments and validate that our optimized com-
pound acts as a specific protein disorder catalyzer that destabilizes
the folded conformation of the Nef core domain and thus increases
its sensitivity to HIV protease. The destabilizing properties of these
compounds would explain why we could not obtain structural data
on their binding mode using NMR or X-ray crystallography.
duced cell viability down 73% (Supplementary Fig. S3). Thus, a ma-
jor issue towards further development of this DLC27 series will be to
increase the solubility of the compounds, without further impacting
on cell viability, but displaying anti-viral activity by targeting the
function of Nef on virion infectivity and anti-viral defense (CD4
and MHC I down regulation). Thus we demonstrated that the new
optimized compound, DLC27-14, acts as a specific protein disorder
catalyzer and thus could act as a potent antiviral compound.
3. Summary and conclusions
In the present work, we evaluated 30 chemically synthesized
DLC27 derivatives as potential second-generation inhibitors of the
Nef:SH3-Hck complex. In these compounds, the apolar head of
DLC27 was replaced with various longer chemical groups with
hydrophobic properties. As we could not evaluate the binding
capacity of these new-generation inhibitors using standard
biophysical experiments such as NMR, ITC, SPR, or X-ray crystallog-
raphy, we developed an original in vitro experiment using MALDI-
TOF MS that enabled us to evaluate IC50 values. We found the
compound DLC27-14 to be a more potent and more efficient inhib-
itor than DLC27 of the Nef:SH3-Hck domain interaction, with an IC50
4. Methods
4.1. Synthesis of Compounds
All solvents were dried and freshly distilled before use. All
reactions were carried out under an atmosphere of dry argon in
flame-dried glassware. Reactions were magnetically stirred and
monitored by thin-layer chromatography using silica gel plates
(Kieselgel 60 F254, Merck) or by analytical HPLC. For validation,
we carried out NMR experiments on the compounds. For details of
HPLC and NMR experiments, please refer to Supplementary data.
For details about the synthesis of (4-bromo-phenoxy)-acetic
acid ethyl ester,16 (4-bromo-phenoxy)-acetic acid,17 3-[2-(4-bromo
of about 16
lM. DSLS experiments and analysis of cleavage of Nef by
the HIV protease showed that DLC27 and DLC27-14 decreased the

A. Lugari et al. / Bioorg. Med. Chem. 19 (2011) 7401–7406
7405
Figure 3. DLC27 and DLC27-14 binding models on Nef. (a) Docking model of DLC27 on surface representation of Nef, colored according to lipophilic potential (PDB code:
1AVZ, chain B). The docking model was obtained using Surflex-Dock (version 2.4) (Tripos) and represented using LITHIUM software (Tripos). (b and c) Docking models of
DLC27-14 on Nef. The Nef protein presents an alternative conformation for the residue Trp113, as explained in Methods. For (b), the hydrophobic pocket was built based on
PDB code 1AVZ (chain B) and Trp113 was rotated by about 35° towards the second pocket using the ‘regularize zone’ mode in COOT.21 For (c), the alternative Trp113
conformation corresponds to the second most populated conformation for residue Trp113 according to Sybyl ‘set sidechain conformation’ tool.
-phenoxy)-acetylamino]-benzoic acid methyl ester,18 3-[2-(40-tert-
butyl-biphenyl-4-yloxy)-acetylamino]-benzoic acid methyl ester
(DLC27-10), and 3-[2-(40-tert-butyl-biphenyl-4-yloxy)-acetylami
no]-benzoic acid (DLC27-14), please refer to Supplementary data.
DLC27 or DLC27-14, diluted in 100% DMSO with a concentration
of 50 M (2ꢁ), 150 M (3ꢁ), 250 M (5ꢁ), 300
M (1ꢁ), 100
(6ꢁ), 350 M (7ꢁ), 450 M (9ꢁ), 500 M (10ꢁ), or 5 mM (100ꢁ).
The mixture was incubated for 1 h at rt. After incubation of Nef
with each inhibitor, the mixture was mixed with 5 L of SH3-
Hck (5 M). The mixture was then incubated for 1 h at rt. The final
concentrations of the proteins in the mixture were 2.5 M for Nef
and 2.5 M for SH3-Hck. After incubation, the samples were sub-
mitted to cross-linking using the K100 MALDI MS stabilization
kit (CovalX). First, 1
L of the cross-linking reagent (2 mg mLꢂ1
was added to each solution, and the mixture was incubated for
1 h. After incubation, 1 L of each mixture obtained was mixed
with 1
L of recrystallized sinapinic acid matrix (10 mg ꢂ1) in ace-
tonitrile/water (1:1, v/v), trifluoroacetyl 0.1%. After mixing, 1 L of
l
l
l
l
lM
l
l
l
l
4.2. Protein expression and purification
l
l
The HIV-1 Nef core domain (residues 56–206) and the Hck SH3
domain were expressed in E. coli BL21 DE3 as GST (glutathione S-
transferase) fusion proteins. The purification used was derived
from that described in Franken et al.19 Cells were grown at 37 °C
until OD600 = 0.6 was reached and then induced using 0.2 mM IPTG
and cultured overnight at 20 °C. Cells were pelleted by centrifuga-
tion (15 min at 8,000g), and pellets were resuspended in 20 mM
Tris (pH 8.0), 150 mM NaCl, 2 mM EDTA, and 5 mM dithiothreitol
(DTT). After French press and centrifugation (45 min at 45000g),
the supernatant was purified by GST-affinity column chromatogra-
phy (GE Healthcare Life Sciences). After incubation of peak frac-
tions with thrombin protease (20:1) at 4 °C, the cleaved GST tag
was removed with a second GST-affinity column. Protein peak frac-
tions were then loaded onto a Superdex 75 column and eluted with
50 mM PO4 (pH 7.0), 150 mM NaCl, 5 mM DTT, and 2.5 mM EDTA
for Nef and 20 mM HEPES (pH 8.0), 150 mM NaCl, and 2 mM EDTA
for SH3-Hck. Purified proteins were further concentrated and
stored at ꢂ80 °C.
l
l
)
l
l
l
each sample was spotted onto the MALDI plate. The analysis was
performed in triplicate. After crystallization at rt, the plate was
introduced into the MALDI mass spectrometer and analyzed imme-
diately. See Supplementary data for details of the analysis.
4.4. Computer files and docking experiments
Both DLC27 and DLC27-14 were designed as standard .mol2
files using the SYBYL chemical structure drawing tool (Tripos)
and further minimized. For docking experiments, the Nef structure
was taken from the crystallographic structure of the Nef protein in
complex with the Fyn SH3 domain (PDB code: 1AVZ, chain B)7 as
described by Betzi et al9 All dockings were computed using the
Surflex-Dock (version 2.4) Geom X algorithm within the SYBYL
suite (Tripos). Please refer to Supplementary data for details.
4.3. MS experiments
All the high-mass measurements were performed using a
MALDI mass spectrometer (ABI 4800 MALDI-TOF/TOF, ABSCIEX)
equipped with the HM2 high-mass detection system (CovalX).
The CovalX HM2 system is designed to optimize ion detection in
the high-mass range, up to 1500 kDa, with nanomolar sensitivity.
4.5. DSLS-based thermal protein denaturation
Thermal protein denaturation based on DSLS was performed
using a multiwell instrument (StarGazer-384, Harbinger Biotech),
Previously purified Nef (5 lM, 4.5 lL) was mixed with 0.5 lL of

7406
A. Lugari et al. / Bioorg. Med. Chem. 19 (2011) 7401–7406
and His-tagged Nef protein was produced according to the proce-
120
100
80
60
40
20
0
a
dure described in Breuer et al. 2006.20 The aggregation tempera-
ture (Tagg) was determined by increasing the temperature of
control
DLC27
10
200
l
l
M His-tagged Nef from 25 °C to 85 °C in the presence of 10–
M compound in 50 mM HEPES (pH 7.5), 750 mM NaCl, and
1% (v/v) DMSO. Tagg was calculated by fitting a Boltzmann sigmoi-
dal curve function using GraphPad Software version 5.0.
4.6. Enzymatic proteolysis
Nef cleavage mediated by HIV protease was measured using
cleavage enzyme-cytometric bead array (CE-CBA) technology as
described in Breuer et al. 2011.13 Please refer to Supplementary
data for details.
35 40 45 50 55 60 65 70 75 80
Temperature (ºC)
Acknowledgment
b
protease
The project was supported by the ANRS (Agence Nationale de
Recherche sur le SIDA et les Hépatites Virales) and the CNRS
(Centre National de la Recherche Scientifique). This research is sup-
ported in part by the National Institutes of Health through MD
Anderson’s Cancer Center Support Grant (CA016672). A.L. and
S.O. were supported by a fellowship from the ANRS. T.C. was sup-
ported by a fellowship from the MESR (Ministère de l’Enseigne-
ment Superieur et de la Recherche). X.S. was supported by a
fellowship from the French Embassy in China. S.B. was supported
by CHRP-F09-SRI-205. We are grateful to Dr. I. Engels and S. Espin-
ola for providing valuable technical expertise with the DSLS-based
thermal protein denaturation at the Genomics Institute of the Nov-
artis Research Foundation. We also would like to greatly thank
K. Muller for editorial assistance.
CAW/LEAQ
Nef
Nef
Nef
GST-
GST-
- mVenus
58-206
1-57
proteolysis
Nef
- mVenus
1-57
58-206
Supplementary data
c
120
100
80
60
40
20
0
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.10.051.
DLC27
DLC27-14
References
1. Peterlin, B. M.; Trono, D. Nat. Rev. Immunol. 2003, 3, 97.
2. Fackler, O. T.; Alcover, A.; Schwartz, O. Nat. Rev. Immunol. 2007, 7, 310.
3. Das, S. R.; Jameel, S. Indian. J. Med. Res. 2005, 121, 315.
4. Stevenson, M. Nat. Med. 2003, 9, 853.
5. Arold, S. T.; Baur, A. S. Trends Biochem. Sci. 2001, 26, 356.
6. Saksela, K. Curr. Drug Targets Immune Endocr. Metabol. Disord. 2004, 4, 315.
7. Arold, S.; Franken, P.; Strub, M. P.; Hoh, F.; Benichou, S.; Benarous, R.; Dumas, C.
Structure 1997, 5, 1361.
8. Lee, C. H.; Saksela, K.; Mirza, U. A.; Chait, B. T.; Kuriyan, J. Cell 1996, 85, 931.
9. Betzi, S.; Restouin, A.; Opi, S.; Arold, S. T.; Parrot, I.; Guerlesquin, F.; Morelli, X.;
Collette, Y. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 19256.
10. Lee, C. H.; Leung, B.; Lemmon, M. A.; Zheng, J.; Cowburn, D.; Kuriyan, J.; Saksela,
K. EMBO J 1995, 14, 5006.
0
1
2
3
log[CMPD](µM)
11. Moarefi, I.; LaFevre-Bernt, M.; Sicheri, F.; Huse, M.; Lee, C.; Kuriyan, J.; Miller,
W. Nature 1997, 385, 650.
12. Collette, Y.; Arold, S.; Picard, C.; Janvier, K.; Benichou, S.; Benarous, R.; Olive, D.;
Dumas, C. J. Biol. Chem. 2000, 275, 4171.
13. Breuer, S.; Sepulveda, H.; Chen, Y.; Trotter, J.; Torbett, B. E. Biochemistry 2011,
50, 4371.
14. Grzesiek, S.; Bax, A.; Hu, J. S.; Kaufman, J.; Palmer, I.; Stahl, S. J.; Tjandra, N.;
Wingfield, P. T. Protein Sci. 1997, 6, 1248.
15. Breuer, S.; Schievink, S. I.; Schulte, A.; Blankenfeldt, W.; Fackler, O. T.; Geyer, M.
PLoS One 2011, 6, e20033.
16. Sathisha, K. R.; Khanum, S. A.; Chandra, J. N.; Ayisha, F.; Balaji, S.; Marathe, G.
K.; Gopal, S.; Rangappa, K. S. Bioorg. Med. Chem. 2011, 19, 211.
17. Khanum, S. A.; Khanum, N. F.; Shashikanth, M. Bioorg. Med. Chem. Lett. 2008, 18,
4597.
18. Lee, K.; Lee, J. H.; Boovanahalli, S. K.; Jin, Y.; Lee, M.; Jin, X.; Kim, J. H.; Hong, Y.
S.; Lee, J. J. J. Med. Chem. 2007, 50, 1675.
19. Franken, P.; Arold, S.; Padilla, A.; Bodeus, M.; Hoh, F.; Strub, M. P.; Boyer, M.;
Jullien, M.; Benarous, R.; Dumas, C. Protein Sci. 1997, 6, 2681.
20. Breuer, S.; Gerlach, H.; Kolaric, B.; Urbanke, C.; Opitz, N.; Geyer, M. Biochemistry
2006, 45, 2339.
Figure 4. Biochemical profiling of DLC27 and DLC27-14. (a) Thermostability of Nef in
complex with DLC27. Shown is a representative thermal protein denaturation
experiment (DSLS-based thermal protein denaturation curve of Nef in the presence of
200 lM compound). Tagg was calculated as mean SEM of three independent
experiments. Tagg (Nef + DLC27) = 50.29 0.03 °C, and Tagg (Nef + DMSO) = 58.83
0.25 °C. (b) Schematic description of the cleavage enzyme—cytometric bead array
(CE-CBA). A recombinant reporter fusion protein (GST-Nef-mVenus) was immobi-
lized on cytometric beads labeled with anti-GST antibody. The HIV protease
recognizes and cleaves at the specific cleavage site of Nef, liberating Nef’s core
domain and mVenus reporter protein. Loss of mVenus decreases the fluorescence
signal on the beads, which can be quantified by flow cytometry. (c) Effects of the
compounds on Nef sensitivity to proteolysis. The effect of the compounds on the
stability of the Nef structure was analyzed by observing the proteolytic processing of
Nef by HIV protease. Shown is the percentage of cleavage as a function of compound
concentration (normalized to the bead fluorescence of the DMSO control as 0%
cleavage and the intrinsic bead fluorescence as100% cleavage). Therelative EC50 value
was calculated as described in Methods and presented as mean SEM of three
independent experiments. For DLC27, EC50 = 109.1 1.9
EC50 = 34.3 3.7 M.
lM; for DLC27-14,
l
21. Emsley, P.; Cowtan, K. Acta Crystallogr., Sect. D Biol. Crystallogr. 2004, 60, 2126.