Probing the HIV-1 NCp7 Nucleocapsid Protein with Site-Specific Gold(I)-Phosphine Complexes

Article abstract of DOI:10.1021/acs.inorgchem.7b01762

In this work, we examined a series of thiophilic Au(I) compounds based on [Au(L)(PR₃)] (L = Cl^-, 4-dimethylaminopyridine (dmap); R= ethyl (Et), cyclohexyl (Cy)) for chemoselective auration of the C-terminal HIV nucleocapsid protein NCp7 F2 and the full HIV NCp7 (NC, zinc finger (ZnF)) as probes of nucleocapsid topography. The choice of phosphine allowed electronic and steric effects to be considered. The use of the heterocycle leaving group allowed us to study the effect of possible π-stacking with the essential tryptophan residue of NC on the reactivity and selectivity, mimicking the naturally occurring interaction between the zinc finger and nucleic acids. We also examined for comparison the standard gold-phosphine compound auranofin, which contains an S-bound glucose coordinated to the {Au(PEt₃)} moiety. Both the nature of the phosphine and the nature of L affect the reactivity with the C-terminal NCp7 F2 and the full NC. ³¹P NMR spectroscopy showed the formation of long-lived {Au(PR₃)}-ZnF species in all cases, but in the case of NCp7 F2, a selective interaction in the presence of the dmap ligand was observed. In the case of auranofin, an unusual Au-His (rather than Au-Cys) coordination was indicated on NC. The overall results suggest that it is useful to consider three aspects of zinc finger structure in considering the profile of chemical reactivity: (i) the zinc-bound cysteines as primary nucleophiles; (ii) the zinc-bound histidine as a spectator ligand; and (iii) ancillary groups not bound to Zn but essential for ZnF function such as the essential tryptophan in NCp7 F2 and NC. Modification of fully functional NC zinc finger by the Cy₃P-containing species confirmed the inhibition of the NC-SL2 DNA interaction, as evaluated by fluorescence polarization.

Full text of DOI:10.1021/acs.inorgchem.7b01762

Article
pubs.acs.org/IC
Probing the HIV‑1 NCp7 Nucleocapsid Protein with Site-Specific
Gold(I)−Phosphine Complexes
Raphael E. F. de Paiva,†,‡ Zhifeng Du,‡ Erica J. Peterson,‡ Pedro P. Corbi,† and Nicholas P. Farrell*,‡
‡Institute of Chemistry, University of Campinas − UNICAMP, P.O. Box 6154, 13083-970 Campinas-SP, Brazil
†Department of Chemistry, Virginia Commonwealth University, 1001 West Main Street, Richmond, Virginia 23284-2006, United
States
S
* Supporting Information
ABSTRACT: In this work, we examined a series of thiophilic
Au(I) compounds based on [Au(L)(PR3)] (L = Cl−, 4-
dimethylaminopyridine (dmap); R= ethyl (Et), cyclohexyl
(Cy)) for chemoselective auration of the C-terminal HIV
nucleocapsid protein NCp7 F2 and the “full” HIV NCp7 (NC,
zinc finger (ZnF)) as probes of nucleocapsid topography. The
choice of phosphine allowed electronic and steric effects to be
considered. The use of the heterocycle “leaving group” allowed
us to study the effect of possible π-stacking with the essential
tryptophan residue of NC on the reactivity and selectivity,
mimicking the naturally occurring interaction between the zinc
finger and nucleic acids. We also examined for comparison the
“standard” gold−phosphine compound auranofin, which
contains an S-bound glucose coordinated to the {Au(PEt3)} moiety. Both the nature of the phosphine and the nature of L
affect the reactivity with the C-terminal NCp7 F2 and the “full” NC. 31P NMR spectroscopy showed the formation of long-lived
{Au(PR3)}−ZnF species in all cases, but in the case of NCp7 F2, a selective interaction in the presence of the dmap ligand was
observed. In the case of auranofin, an unusual Au−His (rather than Au−Cys) coordination was indicated on NC. The overall
results suggest that it is useful to consider three aspects of zinc finger structure in considering the profile of chemical reactivity:
(i) the zinc-bound cysteines as primary nucleophiles; (ii) the zinc-bound histidine as a “spectator” ligand; and (iii) ancillary
groups not bound to Zn but essential for ZnF function such as the essential tryptophan in NCp7 F2 and NC. Modification of
fully functional NC zinc finger by the Cy3P-containing species confirmed the inhibition of the NC−SL2 DNA interaction, as
evaluated by fluorescence polarization.
alkylating agents such as iodoacetamide, 4-vinylpyridine, and
acrylamide has also been used in mass spectrometric peptide
mapping for protein identification.11
INTRODUCTION
■
One of the most remarkable structural and functional
characteristics of the HIV-1 nucleocapsid protein (HIV
NCp7, NC) is the presence of two -Cys-X2-Cys-X4-His-X4-
Cys- (C2HC) zinc finger (ZnF) domains, typically found on
nucleic acid binding proteins.1,2 The presence of ZnF domains
is highly conserved among retroviruses, and any mutation of
the Zn-bound residues results in loss of biological function.1,3
On the infectious HIV-1, NCp7 appears to be closely and
strongly associated with RNA in the viral core.4 The two major
functions of NCp7 are RNA binding and viral encapsidation,
but evidence also suggests that NCp7 has a role in some other
processes such as RNA dimerization, Gag−Gag interactions,
membrane binding, reverse transcription, and stabilization of
the protein complex preintegration.1,4
The cysteine residues of NCp7 are some of the most
nucleophilic of all zinc-bound thiolates in proteins.5−8 As such,
they are substrates for alkylation by electrophiles such as
maleimide and iodoacetamide.9 In cellular assays, N-ethyl-
maleimide can inhibit retroviral infectivity in a concentration-
dependent manner.10 Modification of cysteine residues by
The zinc finger template is a ready source for metal ion
replacement.12 The properties of gold compounds and
[AuCl(PR3)] in particular make them an ideal system for
probing cysteine nucleophilicity in biomolecules, directly
analogous to the “organic” electrophiles. Indeed, early studies
on Au(I) compounds such as aurothiomalate suggested
possible zinc finger inactivation as contributing to their
antiarthritic activity.13 The complex [AuCl(PEt3)] has been
examined for specificity of reaction with respect to model zinc
peptides with differing zinc coordination cores.14 The complex
was also used for targeting the de novo-designed coiled coil
TRIL23C, binding to Cys residues.15 With [AuCl(PPh3)], both
{(PPh3)AuF} and {AunF} species (where F is formally the apo
peptide) are formed with the C-terminal finger of NCp7 F2 and
the F3 finger of the transcription factor Sp1-F3, where the latter
Received: July 10, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Many different methods were attempted to isolate the solid product,
but it rapidly decomposes with reduction to Au(0). In order to
characterize and use this compound, the synthetic procedure was
tends to show a higher propensity for the {AunF} species over
the {(PPh3)AuF} species.16 The coordination preferences of
linear Au(I) in the cysteine-rich environments of both these
individual zinc fingers have been probed by traveling-wave ion
mobility mass spectrometry, and conformational isomers have
been identified.17 The properties of [AuCl(PR3)] can be further
modified by varying the phosphine for both steric and
electronic effects. Phosphines and carbenes are particularly
well suited as carrier ligands when designing Au(I) complexes,
as they form stable and strong bonds with Au(I) and the
formed Au(I) complexes retain good air and moisture
stability.18,19
1
carried out in solution. For H and 31P NMR experiments, 1 equiv of
the chloride precursor (1) and 1 equiv of solid AgNO3 were mixed in
deuterated acetonitrile. The suspension was stirred for 1 h, and the
AgCl that precipitated was removed with a syringe filter. To the
remaining solution was added 1 equiv of dmap. Spectral changes
confirmed the substitution. The same procedure, but with non-
deuterated acetonitrile, was used to prepare the compound for mass
spectrometry (MS) assays.
1H NMR data for compounds 1, 2, 3, and 4 are provided in Figure
S1.
NCp7 (F2 and “Full Finger” NC) Preparation. Preparation was as
previously published.17,20,21 An adequate amount of apo-NCP7 F2 or
apo-NC was dissolved in water, and 1.2 equiv of zinc acetate was
added per zinc core. The pH was adjusted to 7.2−7.4 using a solution
of NH4OH. The solution was incubated for 2 h at 37 °C prior to any
other experiments. The ZnF formation was confirmed by circular
dichroism (CD) spectroscopy and MS, where the overall CD profile
and species distribution in MS were in agreement with data previously
reported in both cases.17,20,21
The more general structure [Au(L)(PR3)] also allows for
modification of L if we consider L as “leaving group” and PR3 as
“carrier group” for interactions with biomolecules. The two
aromatic residues in the structure of NCp7 (Phe16 and Trp37)
are responsible for π-stacking with purine and pyrimidine
residues on RNA and DNA.1,3,4 The importance of these
ancillary residues in dictating the reactivity can be explored by
introducing aromatic coligands with good π-stacking ability
such as a purine or 4-dimethylaminopyridine (dmap) in the L
position of [Au(L)(PR3)].16 In this work, we combined these
two components (phosphine ligands based on basicity and
steric hindrance and heterocycle L versus Cl−) to explore the
analogy between “organic” and Lewis acid electrophiles and to
examine the gold compounds for chemoselective auration and
as useful probes of NC topography. We also examined for
comparison the “standard” gold−phosphine compound aur-
anofin, which contains an S-linked sugar ligand coordinated to
the {Au(PEt3)} moiety. Both the nature of the phosphine and
the nature of L affect the reactivity with the C-terminal NCp7
F2 and the fully functional NC.
Ligand Scrambling Evaluation. 31P NMR Spectroscopy. A 10
mmol L−1 solution of compounds 1−4 was prepared in deuterated
acetonitrile. 31P NMR spectra were recorded on a Varian Mercury 300
MHz (7.05 T) NMR spectrometer, referenced to trimethyl phosphate,
using 256 scans at 20, 27, 34, 41, and 48 °C.
UV−Vis Spectroscopy. The [AuCl(Cy3P)] (3) (4.3 × 10−4 mol
L−1) and [Au(dmap)(Cy3P)]+ (4) (1.7 × 10−4 mol L−1) solutions
were prepared in acetonitrile. UV−vis spectra were recorded on an HP
8453 spectrophotometer equipped with a diode array detector. The
temperature was set to 20, 27, 34, 41, 48, or 55 °C using a Peltier unit.
After 55 °C was reached and the respective spectra were acquired, the
temperature was returned to 20 °C and the spectra were recorded
once again.
Interaction with Model Biomolecules. Ligand Displacement
by N-Acetyl-L-cysteine (N-Ac-Cys). NMR Spectroscopy. To a 10 mmol
L−1 solution of the Au(I)−phosphine compound in acetonitrile was
EXPERIMENTAL SECTION
■
1
added 1 equiv of N-Ac-Cys. H and 31P NMR spectra were acquired
Materials. H[AuCl4] and triethylphosphine (Et3P) were purchased
from Acros. Tricyclohexylphosphine, 4-dimethylaminopyridine, 2,2′-
thiodiethanol, and auranofin were obtained from Sigma-Aldrich.
Deuterated solvents were obtained from Cambridge Isotopes or
Sigma-Aldrich.
immediately after mixing and over time (after 1, 2, 24, and 96 h).
Electrospray Ionization Mass Spectrometry (ESI-MS). To a 14
mmol L−1 of the Au(I)−phosphine compound in acetonitrile was
added 1 equiv of N-Ac-Cys dissolved in water. ESI-MS spectra were
acquired immediately after mixing and at appropriate time points later
where necessary.
Tryptophan Fluorescence Quenching Assay. Stock solutions (7.5
mmol L−1) of the dmap-functionalized compounds (2 and 4) were
prepared. The quenchers were titrated into a cuvette containing 3.0
mL of a 5 μmol L−1 solution of N-acetyltryptophan to give N-Ac-
Trp:quencer ratios from 1:1 up to 1:100. The spectral window
monitored was 300−450 nm with λex = 280 nm. The detector voltage
was 750 V, the temperature was 20 °C, and the scan rate was 600 nm/
min. For the Stern−Volmer model, the linearized data were obtained
at λmax = 362.9 nm.
Interactions with Zinc Fingers. Circular Dichroism Spectros-
copy. Samples of NCp7 F2, apo-NCp7 F2, “full” NC, or “full” apo-NC
peptides (50 μM) were used. For time-based measurements, 1.3 equiv
of the Au(I)−phosphine compound was added, and spectra were
recorded immediately after mixing and after 1, 6, and 24 h. For
concentration-dependent measurements (titration), [Au(dmap)-
(Et3P)]+ was added in the range ri = 0.3−1.3, where ri is the Au
complex:ZnF molar ratio.
Mass Spectrometry. For MS experiments, 1 mM reaction mixtures
(1 equiv of Au complex per ZnF core) were prepared in water/
acetonitrile mixtures at pH 7.0 (adjusted using NH4OH). The reaction
solutions were incubated at room temperature for 0, 2, 6, or 24 h. The
samples were sprayed using a final concentration of ∼100 μM.
Experiments were carried out on an Orbitrap Velos from Thermo
Electron Corporation operated in positive mode. Samples (25 μL)
were diluted with methanol (225 μL) and directly infused at a flow
Synthesis and ZnF Preparation. Chloro(phosphine)gold(I)
Complexes. [AuCl(Et3P)] (1) was synthesized by an adaptation of
an already published method.14,17 In brief, H[AuCl4] (0.50 mmol, 194
mg) was dissolved in 1.0 mL of distilled water, and the resulting
solution was cooled in an ice bath. 2,2′-Thiodiethanol (1.17 mmol,
117 μL) was dissolved in 360 μL of EtOH, and this solution was added
slowly (over ∼4 h, 10 μL per addition) to the Au(III) solution.
Throughout the addition, the solution changed from yellow (Au(III))
to colorless (Au(I)), and the brown-orange byproduct was filtered off.
Et3P was dissolved in 360 μL of EtOH, and the resulting solution was
cooled in the freezer and then added to the Au(I) solution. A white
crystalline solid was obtained immediately. The stirring was carried for
a further 30 min, and the solid was isolated by filtration and purified by
recrystallization from water/EtOH (1:1). δ (31P) in CD3CN = 33.63
ppm. Anal. Calcd for C6H15AuClP (%): C, 20.56; H, 4.31. Found (%):
C, 20.58; H, 4.22.
[AuCl(Cy3P)] (3) was synthesized following the same procedure as
described for [AuCl(Et3P)]. δ(31P) in CD3CN = 55.42 ppm. Anal.
Calcd for C18H33AuClP (%): C, 42.16; H, 6.49. Found (%): C, 41.62;
H, 6.04.
[Au(dmap)(Cy3P)]+ (4) was synthesized following the procedure
previously reported by us for [Au(dmap)(Ph3P)]+.21 δ (31P) in
CD3CN = 51.82 ppm. Anal. Calcd for [Au(dmap)(Cy3P)](NO3),
C25H43AuN3O3P (%): C, 45.39; H, 6.55; N, 6.35. Found (%): C,
44.19; H, 6.02; N, 6.12.
The compound [Au(dmap)(Et3P)]+ (2) is light-sensitive in the
solid state but stable in solution with δ(31P) in CD3CN = 30.05 ppm.
B
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Designed gold(I) compounds 1−4. Auranofin was studied within the Et3P series because of structural similarities. The Tolman cone angle
(in degrees) and pKa values are also given. Triphenylphosphine (PPh3) data are also provided for comparison. The NC “full” zinc finger structure is
shown (with NCp7 F2 boxed), highlighting the Zn-bound residues.
rate of 0.7 μL/min using a source voltage of 2.30 kV. The source
temperature was maintained at 230 °C throughout.
containing 1.25 mM NaCl and 0.125 mM HEPES (pH 7.2) in a 96-
well black, low-binding microplate (Greiner). The NC concentration
was chosen from the NC−SL2 binding experiment such that 90% of
SL2 was bound. FP readings were recorded immediately.
31P NMR Spectroscopy. NCp7 F2. Aliquots (500 μL) of a 0.3 mM
stock solution of NCp7 F2 were mixed with 1.3 equiv of compounds
1−5 dissolved in acetonitrile. The reactions were monitored from t = 0
up to 4 days. Spectra were recorded on a Varian Mercury 300 MHz
(7.05 T) spectrometer, referenced to trimethyl phosphate, using 256
scans and standard phosphorus parameters.
Predicting Zinc Finger Auration Sites Using Blue Star STING.
The module Java Protein Dossier (JPD) was used to analyze and
compare the four zinc-bound ligands of NCp7 F2 (Cys36, Cys39,
His44, and Cys49).22,23 Three parameters were selected for
comparison: electrostatic potential at the last heavy atom (LHA),
accessibility, and sponge. Electrostatic potential values were calculated
using the Delphi program according to the modifications done by
NC “Full” Finger. Aliquots (375 μL) of a 0.129 mM stock solution
of NC “full” zinc finger were mixed with 2.6 equiv of compounds 1−4
and auranofin dissolved in acetonitrile. The reactions were monitored
from t = 0 up to 8 days. The final spectra, acquired after 8 days of
incubation, were recorded with standard phosphorus parameters and
256 scans as above.
NC−SL2 DNA Interaction and Inhibition. The experiments
followed our literature protocol.20,21 An NC and SL2 binding control
experiment was performed as follows: A range of concentrations of
NC (amino acids 1−55) were mixed with 100 nM 3′-fluorescein-
labeled hairpin SL2 DNA (sequence GGGGCGACTGGTGAG-
TACGCCCC) in a final volume of 50 μL of buffer containing 1.25
mM NaCl and 0.125 mM HEPES (pH 7.2) in a 96-well black, low-
binding microplate (Greiner). Fluorescence polarization (FP) readings
were recorded immediately on a Beckman Coulter DTX880 plate
reader. To examine inhibition of NC-SL2 binding, compounds 3 and 4
at various concentrations were incubated with 5 μM NC for 1 h before
addition of 100 nM SL2 DNA in a final volume of 50 μL of buffer
24,25
J
Rocchia and further adapted to PD requirements.
The numerical
values are expressed in units of kT/e. The amino acid accessibility was
calculated according to the SurfV program.26 JPD shows three values:
for the protein chain in isolation, for the protein chain in complex with
the other chain (if) present in the PDB file, and finally, a relative
accessibility (the last one given by the table of absolute solvent-
accessible areas for amino acids). Numerical values are expressed in
units of Å2. Sponge was calculated as the sum of van der Waals
volumes for all atoms encountered within a sphere of a given radius
divided by the volume of the sphere. The values reported in Table 2
were obtained for spheres centered on the LHA with radius of 5 Å.
The numerical values are dimensionless. The evaluated parameters
consider the peptide in its apo form.
C
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Summary of 31P NMR Chemical Shifts for the Reaction Products of the Au(I)−Phosphine Compounds with C-
a
Terminal NCp7 F2 and “Full” NC
δ(31P)/ppm
compound
reagent
NCp7 F2 products (Δδ)
“full” NC product (Δδ)
[AuCl(Et3P)]
33.63
30.05
39.05
55.42
51.82
43.04 (9.41); 49.62 (15.99)
46.96 (16.91)
47.28 (13.65)
47.28 (17.23)
32.40 (−6.65)
64.12 (8.70)
64.11 (12.29)
[Au(dmap)(Et3P)]+
Auranofin
43.04 (3.99); 34.67 (−4.39)
56.54 (1.12); 59.80 (4.37)
57.22 (5.40)
[AuCl(Cy3P)]
[Au(dmap)(Cy3P)]+
a
N-Ac-Cys products: [Au(dmap)(Et3P)]+, 47.24 ppm; [Au(dmap)(Cy3P)]+, 57.67 ppm.
Figure 2. (A) 31P NMR spectra of the reaction products obtained after incubation of NCp7 F2 with Au(I)−phosphine compounds for 6 h: (1)
[AuCl(Et3P)]; (2) [Au(dmap)(Et3P)]+; (3) [AuCl(Cy3P)]; (4) [Au(dmap)(Cy3P)]+. (B) 31P NMR spectra after 8 days of the reaction products
obtained upon incubation of NC with (1) [AuCl(Et3P)], (2) [Au(dmap)(Et3P)]+, (3) [AuCl(Cy3P)], and (4) [Au(dmap)(Cy3P)]+.
chemical shift change (0.457 ppm) was that for compound 4
containing the dmap ligand (Scheme S3). Overall it is possible
to affirm that ligand scrambling happens to only a small extent
in our systems.
RESULTS AND DISCUSSION
■
Ligand Scrambling Evaluation. The structures of the
compounds of general structure [Au(L)(PR3)] are shown in
Figure 1. They were chosen to vary the electronic and steric
properties of the phosphine as well as the putative leaving
group L. The use of 4-NMe2-pyridine (dmap) allowed
evaluation of the influence of a group capable of stacking
with the tryptophan of NCp7 F2, while auranofin with the S-
bound sugar moiety was expected to be kinetically inert with
respect to displacement by the similar S-donor Cys.
Interaction with Model Biomolecules. The interaction
of Au(I)−PR3 compounds with N-acetyl-L-cysteine (N-Ac-Cys)
as a model for sulfur-containing proteins was followed by ESI-
MS and 31P NMR spectroscopy. When the chloride precursors
1 and 3 were mixed with N-Ac-Cys, immediate precipitation of
a white solid occurred, explained by the possible polymeric
nature of the product {Au(I)(N-Ac-Cys)}n, which has been
characterized previously.16,30 In contrast, when the dmap-
functionalized compounds were mixed with N-Ac-Cys, the
products remained a sufficient time in solution, with 31P NMR
shifts from 30.05 to 47.24 ppm and 51.82 to 57.87 ppm for the
PEt3 and PCy3 compounds, respectively, consistent with
replacement of dmap by a sulfur donor (Figure S3).
In the ESI-MS spectra of the chloride precursors 1 and 3
(carried out at much lower concentrations than the NMR
experiments), strong peaks at m/z 792.13 and 1116.42 are
assigned to a sulfur-bridged [(R3P)Au−μ-(N-Ac-Cys)−Au-
(PR3)] species for PEt3 and PCy3, respectively (Figures S4 and
S5). The ligand scrambling products [Au(PR3)2]+ are also
observed. Simple chloride replacement by N-Ac-Cys, corre-
sponding to [(Cy3P)Au(N-Ac-Cys)], is also observed for
compound 3 at m/z 640.23 (Figure S5). A trinuclear gold
Ligand Scrambling. Ligand scrambling is an inherent
property of Au(I)−phosphine complexes (eq 1):27,28
2[(R3P)AuL]+ → [(R3P)2Au]+ + [AuL2]+
(1)
The extent of disproportionation depends on intrinsic (steric
hindrance and basicity of the phosphine as well as the nature of
the coligand) and extrinsic (polarity of the solvent, ionic
strength of the medium) factors.
The 31P NMR chemical shifts of [Au(PEt3)2]+ and
[Au(PCy3)2]+ have been reported as 44.17 and 62.12 ppm,
respectively.29 Here we evaluated only the intrinsic factors, with
the extrinsic factors kept constant (acetonitrile as the solvent
without added electrolytes). UV−vis and 31P NMR spectra of
complexes 1−4 at varying temperatures showed only modest
changes (Figures S1 and S2 and Schemes S1−S3). The largest
D
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. ESI-MS spectra (positive mode) for the reactions between NCp7 F2 (abbreviated ZnF for clarity) and (A) [AuCl(Et3P)] (1), (B)
[Au(dmap)(Et3P)]+ (2), (C) [AuCl(Cy3P)] (3), and (D) [Au(dmap)(Cy3P)]+ (4). Spectra for 1, 3, and 4 were obtained immediately after
incubation, and that for 2 was obtained after 6 h.
species is observed at m/z 1151.13 upon incubation of 1 with
N-Ac-Cys, corresponding to [Au3(N-Ac-Cys)2(Et3P)2]+, but
this species is no longer present after 24 h (Figure S4B). For
the dmap-functionalized compounds (2 and 4), the profile is
the same, but the intensities of the thiolate-substituted products
relative to [Au(R3P)2]+ are much lower than for 1 and 3
(Figures S6 and S7). [(R3P)Au]+ species corresponding to
cleavage of the Au−N bond are also observed in both cases,
along with low-intensity peaks corresponding to monodentate
thiolate binding [(R3P)Au(N-Ac-Cys)], indicative of a lower
reaction rate for compounds 2 and 4 compared with the
chloride precursors 1 and 3.
Besides the effects of electronic structure and steric demands
of the phosphine, a second relevant feature for probe design in
the general structure [Au(L)(PR3)] is how potential stacking of
the dmap-functionalized compounds 2 and 4 with the single
tryptophan entity of NCp7 F2 could affect the overall reaction
profile, considering dmap as the leaving group. The association
constants (KSV) for [Au(dmap)(Et3P)]+ (2) and [Au(dmap)-
(Cy3P)]+ (4) with N-Ac-Trp were calculated as 4.0 × 105 and
6.2 × 105 M−1, respectively, from fluorescence quenching assays
using the Stern−Volmer model (Figure S8). The straight-line
shape of the plots for both systems is indicative of pure static
quenching, as expected for π-stacking. The measured
association constants are almost an order of magnitude higher
than observed for [Au(dmap)(PPh3)]+, reflecting the stronger
basicity of the PEt3 and PCy3 ligands.16 The stacking is also
significantly favored by an order of magnitude in comparison
with previously reported [(dmap)(dien)Au(III)]3+ and
[(dmap)(dien)Pt(II)]2+ complexes.31,32 In general, Au(I) and
Au(III) compounds are too reactive for measurement of their
association constants with the zinc fingers themselves, but it is
noteworthy that the Ka found is the highest we have observed
in our studies and also comparable to those of many organic
moieties used to target NCp7.33,34
Electrophilic Attack on Zinc Fingers Studied by 31P
NMR Spectroscopy. When interacting with a biomolecule,
His versus Cys coordination can be easily distinguished by 31P
NMR spectroscopy, which is sensitive to the nature of the L
ligand in a R3P−Au−L compound. Table 1 summarizes the
data obtained by probing NCp7 F2 and the full NC, and the
spectra are shown in Figure 2.
Probing NCp7 F2. Upon reaction with NCp7 F2, all of the
compounds presented downfield shifts for the 31P signal,
consistent with Au−sulfur binding. Two phosphorus-containing
species were observed for the chloride-containing [AuCl-
(Et3P)] and [AuCl(Cy3P)], whereas the dmap-functionalized
compounds showed only a single phosphorus-containing signal
with the target protein. The pattern is the same for both
phosphines, with the signal arising from the dmap compounds
intermediate in chemical shift between the two signals resulting
from reaction of the chloride species. The results confirm the
E
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. ESI-MS spectra (positive mode) for the reactions between NC “full” ZnF and (A) [AuCl(Cy3P)] immediately after incubation, (B)
[AuCl(Cy3P)] after 6 h incubation, (C) [Au(dmap)(Cy3P)]+ immediately after incubation, and (D) [Au(dmap)(Cy3P)]+ after 6 h incubation.
persistence of {PR3Au}−protein species previously observed
for the PPh3 ligand, but all of the resonances are much sharper,
suggesting that the specific environment is less flexible in the
present case. These results further imply a greater selectivity as
a consequence of replacement of the chloride by the stronger σ
donor dmap, allied with possible contributions from the π−π
stacking (Figure 2A).
in all cases, along with multigold species attributed to the
presence of the multiple coordination sites on the zinc finger:
[Au(L)(PR3)] → [Au(PR3)−ZnF] → [Au(PR3)−apo‐F]
and
[Au(PR3)−ZnF] → [Au−ZnF] → {AuF}
ESI-MS. The formation of S-bound products was confirmed
by ESI-MS. There are many species observable in the gas phase,
but in general a limiting sequence of reactions can be discerned
The two phosphine compounds show similar profiles with
slight differences between the Cl and dmap precursors (Figure
3). For the Cl-containing precursors, in examining the 3+
F
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
charge state species, the presence of the {(PR3)Au} moiety
bound to the peptide is confirmed by the presence of peaks
attributed to Au(PR3)−ZnF and Au(Et3P)−apo-NCp7 species,
the latter presumably caused by ejection of zinc from the
former. A multinuclear Au2(Et3P)−apo-NCp7 species, equiv-
alent to that found with N-Ac-Cys, is also observed for 1
(Figure 3A). At longer reaction times (2, 6, and 24 h), the
spectra are quite similar. The presence of an interesting Au−
ZnF species is seen for both phosphine cases, and in the PEt3
case it is accompanied by a weak peak for the simple “gold
finger” AuF.
6+) upon incubation (Figure 4A). The bisphosphine
compound [Au(Cy3P)2]+ is also observed at m/z 757.43,
along with the aurated species (Cy3P)Au−Zn1F (8+ at m/z
871.77, 7+ at m/z 996.03). A Zn1F species in multiple charge
states (10+, 9+, 8+) is also observed (Figure 4A). After 6 h
(Figure 4B), the spectrum is dominated by apo peptide species
(11+ to 6+), and AuF is the only Au-containing peptide species
(8+, 7+).
Production of apo peptide was not observed for the C-
terminal finger (Figure 3C). The initial spectrum obtained for
the interaction between [Au(dmap)(Cy3P)]+ and NCp7 is
similar to that obtained for [AuCl(Cy3P)] (Figure 4C). On the
other hand, a much more complex metalation profile was
observed after 6 h incubation, with the bisphosphine compound
still appearing as the most intense species (Figure 4D). A
mixture of Zn2F (8+, 7+), Zn1F (9+, 8+, 7+), and apo peptide
(8+, 7+, 6+) was observed. In terms of aurated species, AuF
(8+, 7+, 6+) and the bimetallic species Au−Zn1F (8+, 7+) were
identified. The profile is again different from that of the C-
terminal NCp7 F2, which, combined with the different 31P
NMR spectra, suggests that the different environments can lead
to different reaction pathways and different products. The “full”
NC has a CD profile very similar to that observed for NCp7 F2,
with a valley at 200 nm and a positive absorption region with a
maximum at 218 nm. Interestingly, a quick reaction (at t = 0) is
observed for compound 4, with an increase in intensity of the
negative absorption at 200 nm. For both compounds the
positive band remains, and the spectra again indicate the
presence of ordered structuresthe spectra are similar but not
identical to those generated using Au(III) compounds20
(Figure S14).
The Case of Auranofin: Is Histidine a Spectator
Ligand? Auranofin remains the prototypical gold-based drug
and is structurally related to compounds 1 and 2. In vivo, loss
of the sugar moiety is part of the activation of the drug, but in
biophysical studies, the presence of the S-bound ligand would a
priori suggest lower reactivity toward S-rich ligands such as zinc
fingers compared with Cl− and even N-based leaving groups.
There is some discrepancy in the literature on the reactivity of
auranofin with zinc fingers, which is possibly a reflection of the
techniques and conditions employed. Auranofin reacts slowly
with a Cys2His2 model apo peptide but does not displace Zn
from the Zn peptide.36 On the other hand, monoadduct {AuF}
species were identified by mass spectrometry when auranofin
reacts with a Zn(Cys2HisCys) modeled from a PARP sequence
but not with the apo peptide.37 Interactions of auranofin with
zinc fingers corresponding to the C-terminal HIV-2 NC protein
(Cys2HisCys) from two different isolates also showed Zn
displacement using spectrophotometric and mass spectrometric
techniques.38 In the system studied here, ESI-MS (Figure S15)
and CD spectra with NCp7 ZnF2 demonstrated features similar
to those found for compounds 1 and 2. The 31P NMR spectra
are very instructive: weak signals are observed for both zinc
fingers (Figure S16) but a shielding effect is observed in both
cases. For the C-terminal finger, signals at 43.04 and 34.67 ppm
are relative to auranofin itself at 39.05 ppm. The low-field shift
is easily attributed to Cys displacement and is identical to that
observed for reaction with 1. The upfield shifts have been
observed in reactions of auranofin with bovine serum albumin
and have been attributed to [(PEt3)Au−N-donor)] species.39,40
In our case, therefore, these resonances are reasonably
attributed to His (N) rather than Cys (S) binding, comparing
all of the chemical shifts. In agreement, MS−MS data for the
Slightly different profiles are seen for the dmap derivatives
(Figure 3B,D). Especially, while the [Au(PR3)−apo-F] and
Au−ZnF species are seen, the companion [Au(PR3)−ZnF] is
not observed in either case. This may suggest that the expected
initial stacking of the dmap moiety assists in weakening the
affinity of the peptide for the structural zinc atom. For
[Au(dmap)(Et3P)]+ (2), the auration of the peptide is much
slower, and the products are not observed immediately; zinc
ejection is only observed after 6 h, but the profile otherwise
remains similar to those of the chloride products (Figure 3B).
Circular Dichroism. Circular dichroism spectroscopy can be
used to evaluate conformational changes caused by zinc
ejection. The characteristic CD profile of NCp7 (ZnF2) has
a negative signal centered around 205 nm, which is shifted to
200 nm upon zinc ejection.31,35 The most prominent change in
the CD spectrum of the apo peptide compared with NCp7
(ZnF2) is the loss of the positive absorption bands at ∼190 nm
and in the range 215−225 nm. All of the compounds cause
some degree of conformational change in the NCp7 F2
secondary structure upon interaction, with a decrease in the
intensity of the positive bands and a slight blue shift for the
negative one (Figure S9). However, the spectra over time show
neither a total decrease of the positive band nor an increase in
intensity indicative of the apo peptide as the major species. CD
spectroscopy reflects the contributions of multiple species, and
these observations overall suggest the formation of structured
gold finger products with an organized secondary structure.
Upon further analysis of the intensity of the positive band at
215−230 nm, it is possible to observe that the dmap-containing
compounds (Figure S9B,E) seem to be less reactive than the
chloride counterparts (Figure S9A,D).
Probing the NC “Full” Zinc Finger. The fully functional
two-finger NC relies on the cooperation between the two zinc
finger domains, which are linked by a flexible, highly basic
sequence RAPRKKG, for recognizing specific nucleic acid
sequences. Figure 2B shows the spectra obtained for the
reaction products, where a single, identical 31P signal was
observed for the pairs 1 and 2 and also 3 and 4. Furthermore,
this 31P signal is different from those observed for the C-
terminal NCp7 F2 (individual comparisons are shown in
Figures S10−S13). This is indicative that the phosphine ligand
appears in different chemical environments when interacting
with the two peptides (C-terminal and “full” zinc finger) and
that the cysteine environments are also not identical. The only
case where the chemical shifts of the reaction products with
both peptides are almost identical is for [Au(dmap)(Et3P)]+
(Figure S11).
ESI-MS and CD Spectroscopy. Because of the photo-
sensitivity of compound 2 and its lack of isolation, further
spectroscopic studies were carried out only for the PCy3
compounds and the “full” NC (Figure 4). For [AuCl(PCy3)],
unreacted Zn2F is observed in multiple charge states (8+, 7+,
G
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Fluorescence-polarization-based binding assays. (A) Fully functional NC−SL2 binding. (B) [AuCl(Cy3P)] (triangles) and
[Au(dmap)(Cy3P)]+ (squares) inhibition of NC−SL2 complex formation.
Scheme 1. Schematic of NC Auration Leading to Inhibition of Nucleic Acid Binding
modified C-terminal NCp7 F2 show aurated fragments
indicative of Au−His binding as well as fragments associated
with the expected Au−Cys binding. (Figure S17) In the case of
NC, the product peak at 32.40 ppm is also attributable to a
[(PEt3)Au−His] moiety, along with unreacted auranofin,
observed even after 8 days. The observation of only
(PEt3)Au−His binding in the full NC whereas the C-terminal
finger shows both [(PEt3)Au−Cys] and [(PEt3)Au−His] peaks
again suggests that the environment of the cysteines is subtly
changed compared with that in the fully functional peptide.
NC “Full” Zinc Finger−SL2 DNA Interaction Inhibition.
As shown in Figure 5A, fluorescence polarization (FP) was used
to assay complex formation between the NCp7 “full” zinc finger
and SL2 DNA.20,21 A Kd of 899 46 nM was determined using
the Michaelis−Menten/Hill model (n = 1) (Figure 5A), similar
to published results.20,21,41 Figure 5B shows the results when
[AuCl(Cy3P)] (3) and [Au(dmap)(Cy3P)]+(4) were added to
reactions containing 5 μM NC, where 90% of SL2 was bound.
Dissociation of the NCp7−SL2 complex was observed, and
inhibition constants of 28.6 and 22.0 μM were obtained for
[AuCl(Cy3P)] and [Au(dmap)(Cy3P)]+, respectively, using an
exponential regression model. FP demonstrates the final
outcome of the auration process, since the natural function of
the NC “full” zinc finger relies on its interaction with the viral
RNA. The large Au(I)−PCy3 moiety seems to be required for
the inhibition, since Et3P-containing compounds caused no
significant inhibition (data not shown). In the case of auranofin,
at least this can be attributed to its kinetic inertness. The
compounds display similar efficiency to [Au(dien)(9-
EtGua)]3+.20 The micromolar inhibition in all cases is
somewhat weaker but still within the range of some inhibitory
compounds discovered from library screening of organic
moieties.33,34 The proposed mechanism of inhibition caused
by the Cy3P series is presented in Scheme 1.
Predicting the Auration Sites in Zinc Fingers. We have
previously used traveling-wave ion mobility MS to examine the
conformational preferences of AuF from NCP7 F2, with two
major Cys-Au-Cys conformations identified through Cys36-
Cys49 and Cys39-Cys49 binding.17 One minor conformation was
also found.17 It is reasonable to consider that the origin of the
conformers may arise from loss of PR3 from the initial
{PR3Au−F} species. In this case the observation of two 31P
NMR signals could indicate the two different environments
dictated by the difference in the electrostatic potentials of the
cysteines, and it is furthermore now reasonable to suggest that
different conformations of {AuF} arise from the loss of the PR3
from distinct {PR3Au−F} species and auration of a second
cysteine. We previously used the bioinformatics software Blue
Star STING (Sequence To and withIN Graphics) (see the
Experimental Section) to evaluate potential binding site
preferences of the {Pt(en)} moiety in Sp1 transcription factor
zinc fingers.22,23,42 The analysis was broadly in agreement with
previous analyses showing Cys39 and Cys49 as most likely sites
for electrophilic attack.5,6 In the present case, Table 2
summarizes the accessibility properties as well as the electro-
static potentials of the individual Zn-bound peptide residues.
The cysteines in the zinc coordination sphere can be grouped
into two different categories: Cys36, with an electrostatic
potential at the LHA (E.P.@LHA) of −18.3, and Cys39 and
Cys49, with E.P.@LHA close to −35. Taking into account how
crowded the neighborhood of each residue is, we suggest that
H
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Java Protein Dossier (JPD) Descriptors for Zinc-
Coordinated Residues Found in the NCp7 Structure (from
PDB Entry 1MFS)
in this case through the slow displacement of the S-bound
thioglucose. We are currently pursuing further details of this
reaction. It is relevant in this context to note the work of
Meade, who has used DNA-tethered Co(III) chelates (with
known high affinity for nitrogenous ligands) to selectively
modify histidines in a model Cys2HisCys zinc finger peptide as
well as the Cys2His2 Sp1 transcription factor.43−45 The
combined results clearly show that histidine is not simply a
“spectator” ligand in zinc finger coordination spheres and opens
up interesting targeting possibilities for coordination com-
pounds based on their kinetic inertness.
a
E.P.@
LHA
description
residue
accessibility sponge
HIV-1 nucleocapsid protein, F1
C15
C18
H23
C28
C36
C39
H44
C49
−18.68
−27.16
−84.13
−10.76
−18.3
−33.35
−90.22
−35.91
0
0.99
0.53
0.53
0.70
0.73
0.53
0.69
0.66
66.84
42.85
17.19
2.42
HIV-1 nucleocapsid protein, F2
52.88
41.04
15.17
SUMMARY
■
a
See the Experimental Section for details of the parameters.22,23,42
.
In this work, we explored structural variations in the phosphine
ligand to fine-tune the reactivity of a series of Au(I) Lewis acid
electrophiles for probing zinc finger proteins. By combining MS
with 31P NMR spectroscopy, we were able to follow the
interaction products formed within a couple of seconds up to a
few days, since the PR3−Au−peptide species appear remarkably
robust. Whereas the reaction of compounds 1 and 4 with NCp7
F2 yielded phosphines in more than one chemical environment,
as observed by 31P NMR spectroscopy, one single signal was
observed for the dmap-functionalized compounds. These
features may be attributed to its enhanced π-stacking
properties, mimicking the naturally occurring interaction
between NCp7 and nucleic acids, as well as its expected slower
displacement. The π-stacking observed may also enhance Zn
ejection. Only one signal is observed for the full NC, and the
final products of the pairs 1 and 2 and also 3 and 4 suggest
some shielding of cysteines in the full peptide. The linking basic
residues and the zinc fingers have been proposed to cooperate
Cys49 should be slightly more prone to initial electrophilic
attack than Cys39 but also less accessible (Figure 6).
Histidine as the Site of Binding. To our knowledge, the
observation of histidine binding is the first for the reactions of
the HIV nucleocapsid protein with electrophiles. The vast
majority of discussions on NC reactivity, especially in relation
to drug targeting, have focused on the nucleophilicity and
availability of the cysteines.5−8 However, the bioinformatics
parameters do show that the histidines are reasonably accessible
with good electrostatic potential. Comparison of the histidine
accessibilities in the two fingers strongly suggests the C-
terminal His 44 as the potential specific site of binding for
auranofin. The observation of the histidine-bound auranofin
adduct raises the possibility that kinetic preferences may dictate
the initial binding sites followed by formation of the most
thermodynamically stable Au−Cys species, perhaps observable
Figure 6. Contact interaction map for the zinc-coordinated residues in the zinc finger nucleocapsid NC protein: (A) N-terminal finger; (B) C-
terminal finger. The protein structure was taken from PDB entry 1MFS. Cys18 and Cys39 have significantly reduced calculated contacts.
I
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
to select and package RNA in vivo.46 Probing the topography of
NCp7 using gold compounds shows that the cysteines are in
slightly different environments in the full peptide, which may
affect their nucleophilicitythis is important since these
residues have been extensively targeted for drug intervention.
The slow formation of a Au−histidine adduct on the “full” NC
by auranofin may also be interpreted in terms of selectivity
through kinetic preferences. Overall, the use of the series has
elucidated a number of steric and electronic features of the
nucleocapsid zinc cores that could eventually be used in
targeting.47,48 The analysis of ZnF interactions with the gold
complexes studied here suggests that it is generally useful to
consider three aspects of zinc finger structure in considering the
profile of chemical reactivity: (i) the zinc-bound cysteines as
primary nucleophiles; (ii) the role of histidine as a “spectator”
ligand and secondary nucleophile; and (iii) ancillary groups not
bound to Zn but essential for ZnF function, such as the
essential tryptophan in NCp7 F2 and NC. All of these features
can be seen to play a role in some of the interactions observed
in this work. The inhibition of NC−SL2 binding was closely
associated with the steric hindrance of the phosphine ligands
with the PCy3-containing compounds capable of inhibiting the
biomolecular interaction, as evaluated by fluorescence polar-
ization. Although the phosphine can be surely lost throughout
the reaction, as evidenced by MS data, it still plays an important
role in the inhibition mechanism, as it may dictate conforma-
tional preferences in the expected preferred Cys−Au−Cys
conformation.
reverse transcription and molecular mechanism. Prog. Nucleic Acid Res.
Mol. Biol. 2005, 80, 217−286.
(2) Klug, A. The discovery of zinc fingers and their applications in
gene regulation and genome manipulation. Annu. Rev. Biochem. 2010,
79, 213−221.
(3) Wu, H.; Mitra, M.; McCauley, M. J.; Thomas, J. A.; Rouzina, I.;
Musier-Forsyth, K.; Williams, M. C.; Gorelick, R. J. Aromatic residue
mutations reveal direct correlation between HIV-1 nucleocapsid
protein’s nucleic acid chaperone activity and retroviral replication.
Virus Res. 2013, 171, 263−277.
(4) Keane, S. C.; Heng, X.; Lu, K.; Kharytonchyk, S.; Ramakrishnan,
V.; Carter, G.; Barton, S.; Hosic, A.; Florwick, A.; Santos, J.; Bolden, N.
C.; McCowin, S.; Case, D. A.; Johnson, B.; Salemi, M.; Telesnitsky, A.;
Summers, M. F. Structure of the HIV-1 RNA Packaging Signal. Science
2015, 348, 917−921.
(5) Maynard, A. T.; Covell, D. G. Reactivity of Zinc Finger Cores:
Analysis of Protein Packing and Electrostatic Screening. J. Am. Chem.
Soc. 2001, 123, 1047−1058.
(6) Maynard, A. T.; Huang, M.; Rice, W. G.; Covell, D. G. Reactivity
of the HIV-1 nucleocapsid protein p7 zinc finger domains from the
perspective of density-functional theory. Proc. Natl. Acad. Sci. U. S. A.
1998, 95, 11578−11583.
(7) Lee, Y.-M.; Lim, C. Physical Basis of Structural and Catalytic Zn-
Binding Sites in Proteins. J. Mol. Biol. 2008, 379, 545−553.
(8) Lee, Y.-M.; Lim, C. Factors Controlling the Reactivity of Zinc
Finger Cores. J. Am. Chem. Soc. 2011, 133, 8691−8703.
(9) Chertova, E. N.; Kane, B. P.; McGrath, C.; Johnson, D. G.;
Sowder, R. C.; Arthur, L. O.; Henderson, L. E. Biochemistry 1998, 37,
17890−17897.
(10) Morcock, D. R.; Thomas, J. A.; Gagliardi, T. D.; Gorelick, R. J.;
Roser, J. D.; Chertova, E. N.; Bess, J. W.; Ott, D. E.; Sattentau, Q. J.;
Frank, I.; Pope, M.; Lifson, J. D.; Henderson, L. E.; Crise, B. J. J. Virol
2005, 79, 1533−1542.
(11) Sechi, S.; Chait, B. T. Modification of Cysteine Residues by
Alkylation. A Tool in Peptide Mapping and Protein Identification.
Anal. Chem. 1998, 70, 5150−5158.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b01762.
(12) Quintal, S. M.; dePaula, Q. A.; Farrell, N. P. Zinc Finger
Proteins as Templates for Metal Ion Exchange and Ligand Reactivity.
Chemical and Biological Consequences. Metallomics 2011, 3 (2), 121−
139.
Characterization of compounds and data on scrambling
1
effects; H NMR and MS spectra for reactions with N-
Ac-Cys; calculation of fluorescence quenching; CD and
individual 31P NMR spectra; ESI-MS spectrum of
auranofin−NCp7 F2 adducts (PDF)
(13) Larabee, J. L.; Hocker, J. R.; Hanas, J. S. Mechanisms of
Aurothiomalate−Cys2 His2 Zinc Finger Interactions. Chem. Res.
Toxicol. 2005, 18 (12), 1943−1954.
(14) Franzman, M. A.; Barrios, A. M. Spectroscopic Evidence for the
Formation of Goldfingers. Inorg. Chem. 2008, 47, 3928−3930.
(15) Peacock, A. F. A.; Bullen, G. A.; Gethings, L. A.; Williams, J. P.;
Kriel, F. H.; Coates, J. Gold-Phosphine Binding to de Novo Designed
Coiled Coil Peptides. J. Inorg. Biochem. 2012, 117, 298−305.
(16) Abbehausen, C.; Peterson, E. J.; De Paiva, R. E. F.; Corbi, P. P.;
Formiga, A. L. B.; Qu, Y.; Farrell, N. P. Gold(I)-Phosphine-N-
Heterocycles: Biological Activity and Specific (Ligand) Interactions on
the C-Terminal HIVNCp7 Zinc Finger. Inorg. Chem. 2013, 52 (19),
11280−11287.
AUTHOR INFORMATION
Corresponding Author
*E-mail: npfarrell@vcu.edu.
ORCID
Raphael E. F. de Paiva: 0000-0003-2549-0344
Nicholas P. Farrell: 0000-0001-7160-7182
■
Notes
(17) Du, Z.; de Paiva, R. E. F.; Nelson, K.; Farrell, N. P. Diversity in
Gold Finger Structure Elucidated by Traveling-Wave Ion Mobility
Mass Spectrometry. Angew. Chem., Int. Ed. 2017, 56 (16), 4464−4467.
(18) Wurm, T.; Asiri, A. M.; Hashmi, A. S. K. NHC−Au(I)
Complexes: Synthesis, Activation, and Application. In N-Heterocyclic
Carbenes: Effective Tools for Organometallic Synthesis; Nolan, S. P., Ed.;
Wiley-VCH: Weinheim, Germany, 2014; pp 243−270.
(19) Schwerdtfeger, P.; Hermann, H. L.; Schmidbaur, H. Stability of
the Gold(I)−Phosphine Bond. A Comparison with Other Group 11
Elements. Inorg. Chem. 2003, 42 (4), 1334−1342.
(20) Spell, S. R.; Mangrum, J. B.; Peterson, E. J.; Fabris, D.; Ptak, R.;
Farrell, N. P. Au(III) compounds as HIV nucleocapsid protein
(NCp7)−nucleic acid antagonists. Chem. Commun. 2017, 53, 91−94.
(21) Tsotsoros, S. D.; Lutz, P. B.; Daniel, A. G.; Peterson, E. J.; de
Paiva, R. E. F.; Rivera, E.; Qu, Y.; Bayse, C. A.; Farrell, N. P.
Enhancement of the physicochemical properties of [Pt(dien)-
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was funded by the National Science Foundation
(NSF) Grant CHE-1413189. We also thank Cien
■
̂
cia Sem
Fronteiras CAPES PVES 154/2012, the Brazilian Council of
Technological and Scientific Development (CNPq) 442123/
2014-0, and the Brazilian Coordination Agency for the
Improvement of High-Level Personnel (CAPES) PVES
0580/2013 for support.
REFERENCES
■
(1) Levin, J. G.; Guo, J.; Rouzina, I.; Musier-Forsyth, K. Nucleic acid
chaperone activity of HIV-1 nucleocapsid protein: critical role in
J
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(nucleobase)]2+ for HIVNCp7 targeting. Chem. Sci. 2017, 8, 1269−
1281.
(39) Coffer, M. T.; Shaw, C. F., III; Eidsness, M. K.; Watkins, J. W.,
II; Elder, R. C. Reactions of Auranofin and Et3PAuCl with Bovine
Serum Albumin. Inorg. Chem. 1986, 25, 333−339.
(22) Neshich, G.; Mancini, A. L.; Yamagishi, M. E.; Kuser, P. R.;
Fileto, R.; Pinto, I. P.; Palandrani, J. F.; Krauchenco, J. N.; Baudet, C.;
Montagner, A. J.; Higa, R. H. STING Report: convenient web-based
application for graphic and tabular presentations of protein sequence,
structure and function descriptors from the STING database. Nucleic
Acids Res. 2005, 33, D269−D274.
(40) Kinsch, E. M.; Stephan, D. W. 31P nuclear magnetic resonance
and fluorescence study of the interaction of an anti-arthritic gold
phosphine drug with albumin. Inorg. Chim. Acta 1984, 91, 263−297.
(41) Breuer, S.; Chang, M. W.; Yuan, J.; Torbett, B. E. identification
of HIV-1 inhibitors targeting the nucleocapsid protein. J. Med. Chem.
2012, 55, 4968−4977.
(23) Neshich, G.; Rocchia, W.; Mancini, A. L.; Yamagishi, M. E. B.;
Kuser, P. R.; Fileto, R.; Baudet, C.; Pinto, I. P.; Montagner, A. J.;
Palandrani, J. F.; Krauchenco, J. N.; Torres, R. C.; Souza, S.; Togawa,
R. C.; Higa, R. H. JavaProtein Dossier: a novel web-based data
visualization tool for comprehensive analysis of protein structure.
Nucleic Acids Res. 2004, 32, W595−W601.
(42) Du, Z.; de Paiva, R. E. F.; Qu, Y.; Farrell, N. P. Tuning the
reactivity of Sp1 zinc fingers with platinum complexes. Dalton Trans.
2016, 45, 8712−8716.
(43) Louie, A. Y.; Meade, T.J. A. Cobalt complex that selectively
disrupts the structure and function of zinc fingers. Proc. Natl. Acad. Sci.
U. S. A. 1998, 95, 6663−6668.
(44) Heffern, M. C.; Kurutz, J. W.; Meade, T. J. Spectrosocpic
characterization of the inhibitory mechanism of Cys2His2 zinc finger
transcription factors by cobalt(III) Schiff bases. Chem. - Eur. J. 2013,
19, 17043−17053.
(24) Honig, B.; Nicholls, A. Classical Electrostatics in Biology and
Chemistry. Science 1995, 268 (5214), 1144−1149.
(25) Rocchia, W.; Sridharan, S.; Nicholls, A.; Alexov, E.; Chiabrera,
A.; Honig, B. Rapid Grid-Based Construction of the Molecular Surface
and the Use of Induced Surface Charge to Calculate Reaction Field
Energies: Applications to the Molecular Systems and Geometric
Objects. J. Comput. Chem. 2002, 23, 128−137.
(45) Vistain, L. F.; Yamamoto, N.; Rathore, R.; Cha, P.; Meade, T. J.
Targeted Inhibition of Snail Activity in Breast Cancer Cells by Using a
CoIII-Ebox Conjugate. ChemBioChem 2015, 16, 2065−2072.
(26) Sridharan, S.; Nicholls, A.; Honig, B. A New Vertex Algorithm
To Calculate Solvent Accessible Surface Areas. Biophys. J. 1992, 61,
A174.
́
(46) De Rocquigny, H.; Gabus, C.; Vincent, A.; Fournie-Zaluski, M.-
C.; Roques, B.; Darlix, J.-L. Viral RNA annealing activities of human
immunodeficiency virus type 1 nucleopcapsid protein require only
peptide domains outside the zinc fingers. Proc. Natl. Acad. Sci. U. S. A.
1992, 89, 6472−6476.
(27) Shaw, C. F., III Gold-Based Therapeutic Agents. Chem. Rev.
1999, 99, 2589−2600.
(28) Ahmad, S. The chemistry of cyano complexes of gold(I) with
emphasis on the ligand scrambling reactions. Coord. Chem. Rev. 2004,
248, 231−243.
(47) Mori, M.; Kovalenko, L.; Lyonnais, S.; Antaki, D.; Torbett, B. E.;
́
Botta, M.; Mirambeau, G.; Mely, Y. Nucleocapsid Protein: A Desirable
Target for Future Therapies Against HIV-1. Curr. Top. Microbiol.
Immunol. 2015, 389, 53−92.
(29) Isab, A. A.; Hussain, M. S.; Akhtar, M. N.; Wazeer, M. I. M.; Al-
Arfaj, A. R. 13C, 15N and 31P NMR study of the disproportionation of
cyanogold(I) complexes: [R3PAu13C15N]. Polyhedron 1999, 18, 1401−
1409.
(48) Garg, D.; Torbett, B. E. Advances in targeting nucleocapsid-
nucleic acid interactions in HIV-1 Therapy. Virus Res. 2014, 193, 135−
143.
(30) Corbi, P. P.; Quintao, F. a.; Ferraresi, D. K. D.; Lustri, W. R.;
̃
Amaral, a. C.; Massabni, a. C. Chemical, Spectroscopic Character-
ization, and in Vitro Antibacterial Studies of a New gold(I) Complex
with N -Acetyl- L -Cysteine. J. Coord. Chem. 2010, 63 (8), 1390−1397.
(31) Spell, R. S.; Farrell, N. P. Reactivity with Biomolecules and Zinc
Finger Peptides. Inorg. Inorg. Chem. 2015, 54, 79−86.
(32) Tsotsoros, S. D.; Bate, A. B.; Dows, M. G.; Spell, S. R.; Bayse, C.
A.; Farrell, N. P. Modulation of the stacking interaction of MN4 (M=
Pt, Pd, Au) complexes with tryptophan through N-heterocyclic
ligands. J. Inorg. Biochem. 2014, 132, 2−5. PMID 24206773
(33) Stephen, A. G.; Worthy, K. M.; Towler, E.; Mikovits, J. A.; Sei,
S.; Roberts, P.; Yang, Q.; Akee, R. K.; Klausmeyer, P.; McCloud, T. G.;
Henderson, L.; Rein, A.; Covell, D. G.; Currens, M.; Shoemaker, R. H.;
Fisher, R. J. Identification of HIV-1 nucleocapsid protein: nucleic acid
antagonists with cellular anti-HIV activity. Biochem. Biophys. Res.
Commun. 2002, 296, 1228−1237.
(34) Shvadchak, V.; Sanglier, S.; Rocle, S.; Villa, P.; Haiech, J.; Hibert,
́
M.; Van Dorsselaer, A.; Mely, Y.; de Rocquigny, H. Identification by
high throughput screening of small compounds inhibiting the nucleic
acid destabilization activity of the HIV-1 nucleocapsid protein.
Biochimie 2009, 91, 916−923.
(35) de Paula, Q. A.; Mangrum, J. B.; Farrell, N. P. Zinc finger
proteins as templates for metal ion exchange: Substitution effects on
the C-finger of HIV nucleocapsid NCp7 using M(chelate) species (M
= Pt, Pd, Au). J. Inorg. Biochem. 2009, 103, 1347−1354.
(36) Laskay, U. A.; Garino, C.; Tsybin, Y. O.; Salassa, L.; Casini, A.
Gold Finger Formation Studied by High-Resolution Mass Spectrom-
etry and in Silico Methods. Chem. Commun. 2015, 51, 1612−1615.
(37) Castiglione Morelli, M. A.; Ostuni, A.; Matassi, G.; Minichino,
C.; Flagiello, A.; Pucci, P.; Bavoso, A. Spectroscopic Investigation of
Auranofin Binding to Zinc Finger HIV-2 Nucleocapsid Peptides.
Inorganica Chim. Inorg. Chim. Acta 2016, 453, 330−338.
(38) Mendes, F.; Groessl, M.; Nazarov, A. A.; Tsybin, Y. O.; Sava, G.;
Santos, I.; Dyson, P. J.; Casini, A. Metal-based inhibition of poly(ADP-
ribose) polymerase−the guardian angel of DNA. J. Med. Chem. 2011,
54, 2196−2206.
K
DOI: 10.1021/acs.inorgchem.7b01762
Inorg. Chem. XXXX, XXX, XXX−XXX