HIV-1 in strand transfer chelating inhibitors: A focus on metal binding

Article abstract of DOI:10.1021/mp100343x

Most active and selective strand transfer HIV-1 integrase (IN) inhibitors contain chelating functional groups that are crucial feature for the inhibition of the catalytic activities of the enzyme. In particular, diketo acids and their derivatives can coordinate one or two metal ions within the catalytic core of the enzyme. The present work is intended as a contribution to elucidate the mechanism of action of the HIV-IN inhibitors by studying the coordinative features of H₂L¹ (L-708,906), an important member of the diketo acids family of inhibitors, and H₂L², a model for S-1360, another potent IN inhibitor. Magnesium(II) and manganese(II) complexes of H₂L¹ and H₂L² were isolated and fully characterized in solution and in the solid state. The crystal structures of the manganese complex [Mn(HL²)₂(CH₃OH) ₂]·2CH₃OH were solved by X-ray diffraction analysis. Moreover, the speciation models for H₂L² with magnesium(II) and manganese(II) ions were performed and the formation constants of the complexes were measured. M(HL²)₂ (M = Mg ²⁺, Mn²⁺) was the most abundant species in solution at physiological pH. All the synthesized compounds were tested for their anti-IN activity, showing good results both for the ligand and the corresponding complexes. From analysis of the speciation models and of the biological data we can conclude that coordination of both metal cofactors could not be strictly necessary and that inhibitors can act as complexes and not only as free ligands.

Full text of DOI:10.1021/mp100343x

ARTICLE
pubs.acs.org/molecularpharmaceutics
HIV-1 IN Strand Transfer Chelating Inhibitors: A Focus on Metal
Binding
Alessia Bacchi,† Mauro Carcelli,† Carlotta Compari,‡ Emilia Fisicaro,‡ Nicolino Pala,§ Gabriele Rispoli,†
Dominga Rogolino,*,† Tino W. Sanchez,|| Mario Sechi,§ and Nouri Neamati||
†Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica and ‡Dipartimento di Scienze Farmacologiche,
Biologiche e Chimiche Applicate, Universitꢀa di Parma, Parco Area delle Scienze, 17/A, 43124 Parma, Italy
§Dipartimento Farmaco Chimico Tossicologico, Universitꢀa di Sassari, Via Muroni 23/A, 07100 Sassari, Italy
Department of Pharmaceutical Sciences, School of Pharmacy, University of Southern California, 1985 Zonal Avenue, PSC 304,
Los Angeles, California 90089, United States
S Supporting Information
b
ABSTRACT: Most active and selective strand transfer HIV-1
integrase (IN) inhibitors contain chelating functional groups that
are crucial feature for the inhibition of the catalytic activities of the
enzyme. In particular, diketo acids and their derivatives can
coordinate one or two metal ions within the catalytic core of the
enzyme. The present work is intended as a contribution to
elucidate the mechanism of action of the HIV-IN inhibitors by
studying the coordinative features of H2L1 (L-708,906), an im-
portant member of the diketo acids family of inhibitors, and H2L2, a
model for S-1360, another potent IN inhibitor. Magnesium(II) and
manganese(II) complexes of H2L1 and H2L2 were isolated and
fully characterized in solution and in the solid state. The crystal
structures of the manganese complex [Mn(HL2)2(CH3OH)2]
3
2CH3OH were solved by X-ray diffraction analysis. Moreover, the
speciation models for H2L2 with magnesium(II) and manganese-
(II) ions were performed and the formation constants of the complexes were measured. M(HL2)2 (M = Mg2þ, Mn2þ) was the most
abundant species in solution at physiological pH. All the synthesized compounds were tested for their anti-IN activity, showing good
results both for the ligand and the corresponding complexes. From analysis of the speciation models and of the biological data we can
conclude that coordination of both metal cofactors could not be strictly necessary and that inhibitors can act as complexes and not
only as free ligands.
KEYWORDS: HIV-1 integrase, strand transfer selective inhibitors, S-1360, metal complexes, coordinating pharmacophores
’ INTRODUCTION
highly conserved among polynucleotidyl transferases.10 These
three conserved amino acid residues coordinate one or two
divalent metal ions. IN catalyzes the insertion of the viral DNA
into the host cell genome through two different steps, 30-
processing and strand transfer. During 30-processing IN selec-
tively cleaves the last two nucleotides (GT) of viral DNA, to
generate two CA-30-hydroxyl recessed ends, which are the
reactive intermediates required for the next step. The enzyme,
still bound to the 30-processed viral DNA, translocates to the
nucleus of the infected cell as a part of the preintegration
complex, wherein the terminal 30-OH of the viral DNA attacks
the host DNA in the strand transfer step.
The antiviral therapy currently in use against the human
immunodeficiency virus type 1 (HIV-1) provides good results,
coming close to stopping viral evolution. However, it does not
completely suppress viral replication in cells, and several factors
like drug toxicity, resistant strains and problems with patient
adherence clearly indicate the necessity to develop new and more
potent drugs. In the last ten years, HIV-1 integrase (IN), which
catalyzes the integration of proviral cDNA into the host cell
genome,1-7 has emerged as a promising target for drug design.
The efforts in this direction resulted in the recent approval by the
US FDA of the first IN inhibitor, raltegravir (brand name
Isentress, Merck & Co.).8 Currently, there are several other IN
inhibitors under clinical trial (Figure 1).9
Received: October 8, 2010
Accepted: February 16, 2011
IN is a 32 kDa protein that folds into three domains: the
C-terminal domain, the N-terminal domain and the catalytic core
domain. The core domain contains the “D,D(35)E” motif that is
Revised:
February 9, 2011
Published: February 16, 2011
r
2011 American Chemical Society
507
dx.doi.org/10.1021/mp100343x Mol. Pharmaceutics 2011, 8, 507–519
|

Molecular Pharmaceutics
ARTICLE
chelating groups. Modification of the hydrophobic scaffold can in
fact produce a dramatic variation of activity. This is probably due
to the fact that this moiety provides a series of interactions within
the active site, as precisely observed in the solved crystal structure
of the intasome.22
It is worth noting that magnesium is the metal cofactor in
many processes involving formation and modification of phos-
phate chains that are fundamental for the action of polymerases,
exonucleases, ribonucleases, transposases and integrases. Effec-
tively, metal-chelating compounds have been developed as antiviral
agents, and most of them have been derived exactly from DKAs.27 It
is therefore of great interest to gain insight into their mechanism
of action, in order to design more effective drugs not only against
HIV IN but also against many other viral enzymes, such as
Hepatitis C Virus polymerase and Influenza endonuclease. More-
over, some DKA IN inhibitors have shown anti-RNase H
acivity,28-30 and recently some inhibitors of both enzymes have
been identified.30,31
Figure 1. HIV-1 IN inhibitors raltegravir, elvitegravir and 5CITEP.
The “two-metal ions binding model” has been suspected for a
long time and it became an important strategy for the develop-
ment of new and potent IN inhibitors.23,32-35 As mentioned
earlier, DKAs are able to coordinate one or two metal ions (in
particular Mg2þ or Mn2þ), since they present two binding sites:
the keto-enol and the carboxylate functions. Complexes with a
2:2 ligand:metal stoichiometry were predicted by quantum
chemistry calculations36 and isolated in our previous studies.37,38
Also the investigation of the coordinative ability of members of
the polyhydroxylated styrylquinoline series by X-ray crystallo-
graphic analysis39 evidenced the formation of bimetallic com-
plexes of Mg2þ and Cu2þ, with a Mg-Mg distance of 3.221(1)
Å, that is notably in agreement with the metal-metal distance of
3.9 Å encountered in the crystal structure of Escherichia coli DNA
polymerase as well as of IN-DNA intasome (ranging from 3.32 to
4.46 Å). The chelating ability of ligands containing the quinolone
pharmacophoric fragment was elucidated by crystallographic
studies on magnesium(II) complexes40,41 and by a series of other
studies using several computational tools.42,43 Nevertheless, the
data regarding the metal/inhibitors equilibria or regarding the
biological activity of isolated metal complexes are surprisingly
scarce. Moreover, it is not definitively stated if it is strictly necessary
for the ligand to coordinate both metal ions or it is sufficient to
block only one cation to inhibit strand transfer. In fact, the recent
development of novel quinolone HIV-1 IN strand-transfer inhib-
itors18,19,44 (for example elvitegravir,45 Figure 1, that is currently
undergoing phase III clinical trials) pointed out that a monoketo
acid fragment can be an alternative to the keto-enol acid motif.
Therefore, an in-depth understanding of their coordinating
ability is of paramount importance, since it will allow a more
efficient drug design of potent inhibitors.
Figure 2. (a) Chemotype of DKA β-diketo-based inhibitors; (b) H2L1;
(c) the triazolic bioisostere S-1360 with its model H2L2 used in the
present work.
Several classes of IN inhibitors have been identified, dating
back to the studies on diketo acids (DKAs, Figure 2a) and their
bioisosteres, such as the keto-enol tetrazole 5CITEP (Figure 1)
and the keto-enol triazole S-1360 (Figure 2c),1,11-14 until
the more recent examples, carboxamide15-17 and quinolone
derivatives.18-21 The keto-enolic acid motif can be replaced by
several bioisosteres that can mimic a ketone, an enol and a
carboxyl moiety. The presence of this common scaffold is
thought to be fundamental for inhibition, since it allows the
coordination of the divalent metal ions, i.e. the possibility to
block the active site by competing with the viral DNA substrate
(Figure 3). These considerations have been recently supported
by the resolution of the X-ray structure of the full-length enzyme
in complex with the viral DNA, in the absence and in the
presence of inhibitors.22 The crystal structure of the intasome
(the complex between viral DNA and IN) with raltegravir and
elvitegravir shows that they are located within the active site with
their chelating fragments oriented toward the metal cofactors,
while the benzyl groups fit within a tight pocket of the enzyme.
Several studies23-26 have pointed out the critical importance of
the hydrophobic substituent adjacent to the pharmacophoric
In this study we report the synthesis and characterization of
two ligands, the (2Z)-2-hydroxy-4-oxo-4-(3,5-benzyloxy)phenyl-
but-2-enoic acid (H2L1, L-708,906, Figure 2b),11 which exemplifies
the DKA family of inhibitors, and the (Z)-3-hydroxy-1-phenyl-
3-(1H-1,2,4-triazol-3-yl)prop-2-en-1-one (H2L2), designed as
a model of S-1360 (Figure 2c).46 Since the primary aim of this
study is to investigate the coordinating ability of IN strand
transfer inhibitors, we used a simplified model of S-1360 to
avoid long synthetic procedures, yet maintaining the pharma-
cophoric chelating scaffold. The magnesium(II) and man-
ganese(II) metal complexes were synthesized and character-
ized, together with the X-ray crystal structure of ligand
H2L2 and of [Mn(HL2)2(CH3OH)2] 2CH3OH. A series of
3
508
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Figure 3. The two-metal inhibition mechanism. (A) Strand transfer reaction and (B) the inhibitor chelates the two metal ions blocking the host DNA
binding.
Scheme 1. Preparation of Ligand H2L1 a
a Reagents and conditions: (i) Dimethyloxalate, anhydrous DMF, NaH, 0-5 °C, then 65-70 °C for 5 h; (ii) 2 N NaOH, methanol, room temperature
for 5 h, then 1 N HCl.
Scheme 2. Preparation of Ligand H2L2 a
a Reagents and conditions: (i) 65 °C, 1 h; (ii) diglyme, reflux, 1.5 h; (iii) chlorotriphenylmethane, triethylamine, anhydrous N,N-dimethylacetamide,
rt, 15 min; (iv) acetophenone, 1 M LiHMDS in THF, anhydrous THF, -70 °C to rt, 1.5 h; (v) 3 N HCl, dioxane, 80 °C, 30 min.
potentiometric measurements were also carried out with
H2L2. Finally, the anti-HIV-1 IN activity of the free ligands
and the related metal complexes have been evaluated in
enzymatic assays, and a mechanistic hypothesis has been
detailed.
’ RESULTS AND DISCUSSION
Chemistry. Ligand H2L1 (Scheme 1) was synthesized by using
a modified literature procedure.25,47 Singlets for CH2 protons
were present at 4.58-4.60 ppm in the spectra recorded in
509
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Scheme 3. Synthesis of Complexes 9 and 10 with Ligand H2L1 (a) and 11 and 12 with Ligand H2L2 (b)
DMSO-d6, indicating the coexistence of both the enolic and
ketonic tautomers (98:2 and 97:3, for H2L1 and 2, respectively).
Moreover, a mixture of the two geometric Z and E stereoisomers
was detected in a 6:1 ratio (for E and Z, respectively).47
The ligand H2L2 was prepared by following the same synthetic
approach we previously used for S-136048 (Scheme 2). It was
characterized by means of IR, 1H NMR and mass spectrometry.
In the 1H NMR spectrum, recorded in DMSO-d6, both the keto
and the enolic forms are visible: the C-H proton is at 7.28
ppm (enolic form), while a singlet for the CH2 protons of the
keto tautomer is at 4.80 ppm (17%). Finally, H2L2 was crystal-
lized from a MeOH/HCl solution as H2L2 1/2HCl and
3
characterized by X-ray diffraction analysis (see Supporting
Information).
Scheme 3a details the synthesis of complexes 9 and 10. The
ligand is bideprotonated, giving rise to dimeric complexes of type
M2L1 nH2O. Mn(CH3COO)2 was sufficient to deprotonate
2
3
Figure 4. Crystal structure of [Mn(HL2)2(CH3OH)2] 2CH3OH. Mn
H2L1, while in the case of magnesium, the hydroxide had to be
3
occupies a center of symmetry, the labeling is reported only for the
independent unit. Hydrogen bonds to the pair of uncoordinated
methanol molecules are dotted. Thermal ellipsoids are at the 50%
probability level.
used. The necessity to carefully choose the proper salt has emerged
also during the studies on hydroxyisoquinoline dione dual inhib-
itors30 so that deprotonation and therefore complexation could
occur only with a sufficiently basic anion. The complete depro-
tonation of the ligand and the coordination of the diketo acid
moiety to the metal can be inferred through the IR spectra of the
complexes: the OH absorption disappeared (2850-3100 cm-1
in the free ligand), and the CdO bands shifted from 1709
(ligand) to 1679 and 1575 cm-1 for 9 and 10, respectively. For
complex 9, a sharp intense band at 3695 cm-1 was attributable to
the presence of coordinated water molecules, since it disappeared
after heating the compound at 120 °C. The 1H NMR spectra of 9
confirmed the absence of acidic protons. We observed a 0.15-
0.20 ppm downfield shift of all the signals vs the free ligand.
Elemental analysis confirmed the proposed 1:1 stoichiometry.
The DKA ligands can coordinate in the hydroxy-carboxylate49 or
in the acetyl-acetonate form.50 In the IR spectra of 9 and 10, the
band for a free CdO was absent, which allowed exclusion of the
possibility that the ligand is in the hydroxy-carboxylate form.
Our37 and others'36 previous studies suggest the presence of
M2L2 species, where both the binding modes of the ligand are
used to give a dimer where the acetyl, the hydroxy and the
carboxylate groups coordinate to the metal; it can be assumed that,
also for 9 and 10, dimeric species are formed (Scheme 3a, inset).
The reactivity of H2L2 was slightly different. Complexes of
general formula M(HL2)2 4H2O (M = Mg2þ 11, M = Mn2þ
3
12), can be obtained by using the chloride or the acetate of the
metal in the presence of a base to ensure deprotonation of the
ligand (pH = 8, Scheme 3b). In these conditions, only the enolic
proton is lost, while the triazolic moiety is still protonated.
Monodeprotonation of the ligand and coordination were in-
ferred by spectroscopic tools (IR, 1H NMR) and mass spectro-
metry and were further confirmed by X-ray diffraction analysis on
the manganese complex.
Under harsh conditions (pH = 10), also the triazole ring of
H2L2 was deprotonated, with formation of Mg2L2 5H2O (13)
2
3
in analogy with the DKA parent compound.37 Therefore, a
510
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Figure 5. Frontal (left) and side-on (right) views of the supramolecular ribbons generated by hydrogen bonds (dotted) in the crystal structure of
[Mn(HL2)2(CH3OH)2] 2CH3OH.
3
Figure 6. Distribution diagrams for the systems under investigation (ligand:M = 2:1, [H2L2] = 2 mM).
Table 1. Logarithms of Formation Constants (βpqr = [MpLq-
Hr]/[M]p[L]q[H]r) in Methanol/Water = 9:1 v/v, I=0.1MKCl
at 25 °C for the Ligand under Study with Mn(II) and Mg(II)a
Table 2. Inhibition of HIV-1 IN Catalytic Activities of
Ligands and Complexesa
IC50 (μM)
p
q
r
M = Mn(II)
M = Mg(II)
compds
Mb
M:Lc
30-processing
strand transfer
SId
1
1
1
1
1
1
2
2
1
1
1
1
2
16.88 (0.11)
25.32 (0.09)
33.04 (0.16)
-2.01 (0.09)
9.53 (0.08)
16.60 (0.14)
23.92 (0.08)
33.02 (0.10)
-3.67 (0.07)
6.45 (0.32)
H2L1
9
9 ( 2
70 ( 4
4 ( 1
>333
0.5 ( 0.3
1.6 ( 0.4
0.3 ( 0.1
69 ( 0.4
43 ( 7
18
44
13
>5
>2
2.4
5
Mg2þ
Mn2þ
2:2
2:2
10
-1
H2Le
H2L2
11
0
>100
p
q
r
Mg2þ
Mn2þ
1:2
1:2
100
42 ( 16
19 ( 4
12
S-1360f
87 ( 7
11 ( 2
0
0
1
1
1
2
L þ H h LH
L þ 2H h LH2
11.431 (0.007)
19.657 (0.008)
0.6 ( 1
18
a MnCl2 was used in the reaction buffer. b M, metal ion. M:L, metal:
c
a Standard deviations are given in parentheses. Charges are omitted for
simplicity.
ligand ratio. d SI: selectivity index. e Data from ref 37. f Data from ref 54.
coordination compound of the type M(HL2)2, with the ligand
monodeprotonated, was the most abundant species at physiolo-
gical conditions, but at higher pH the ligand is completely
While for H2L1 a classical two-metal chelation model has been
found, two preferred binding modes have been disclosed for the
triazolic bioisostere. Both models can be regarded as two-metal
binding, but in one case the Mg2þ ions show close interactions
only with the keto-enolic fragment. Other recent computational
studies seem to suggest a two-metal binding also for the diketo
triazole pharmacophore.52
deprotonated and dimeric Mg2L2 species was also formed
2
(see potentiometric data below).
It is interesting to note that these data are in agreement with
induced-fit docking studies performed on H2L1 and S-1360.51
511
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Figure 7. A representative gel showing inhibition of purified IN by selected compounds H2L2 and complexes 9-12. (a) A 21-mer blunt-end
oligonucleotide corresponding to the U5 end of the HIV-1 LTR, 50 end-labeled with 32P, is reacted with purified IN. The initial step involves nucleolytic
cleavage of two bases from the 30-end, resulting in a 19-mer oligonucleotide. Subsequently, 30 ends are covalently joined at several sites to another
identical oligonucleotide that serves as the target DNA. This reaction is referred to as strand transfer, and the products formed migrate more slowly than
the original substrate (shown in the Figure 6b as STP, for strand transfer products). (b) Line named as DNA indicates DNA alone; line termed INT
indicates IN and DNA with no drug; other lines: IN, DNA and selected drug concentrations (μM) as indicated in each line. STP indicates strand transfer
products.
X-ray Crystallography of [Mn(HL2)2(CH3OH)2] 2CH3OH.
than the adjacent C7-C8. The terminal rings are not coplanar
with the chelation system (C8-C9-C10-N1 = 19°; C8-C7-
C1-C6 = -36°) and form a dihedral angle of 45° between them.
The NH groups on the triazole rings point to the internal side of
the complex, in cis position with respect to the Mn-O coordina-
tion bonds. The complex is supramolecularly associated to a pair
of uncoordinated solvation methanol molecules, which are
docked by NH O and CH O interactions with the NH
3
Slow evaporation of a methanolic solution of 12 afforded crystals
of [Mn(HL2)2(CH3OH)2] 2CH3OH suitable for X-ray diffrac-
3
tion. Crystals were not stable in air, probably due to the loss of
crystallization solvent molecules. [Mn(HL2)2(CH3OH)2] con-
sists of a centrosymmetric molecular complex where the Mn2þ
center is octahedrally coordinated by two monodeprotonated
ligands occupying the equatorial plane in a bidentate fashion, and
two apical methanol molecules. Two uncoordinated solvation
methanol molecules complete the cell contents. Figure 4 shows
the molecular structure and labeling of [Mn(HL2)2(CH3-
3 3 3
3 3 3
of the triazole rings and the O2of the CO groups (N3-H O4=
3 3 3
2.742(4)Å, 169(3)°; C13-H O2(i) = 3.421(3)Å, 144(1)°;
3 3 3
i = -x, 1 - y, 1 - z) (Figure 5).
OH)2] 2CH3OH. Deprotonation of the ligand occurs at the
It is interesting to note that, in the X-ray structure of 5CITEP
cocrystallized with the enzyme,12 the inhibitor is located in the
center of the active site without coordinating directly the mag-
nesium ion, but all four nitrogen atoms of the tetrazole ring are
hydrogen-bonded to the protein. The involvement of the tria-
zolic ring in extensive hydrogen bonding in the crystal structure
3
keto-enolic system, whose chelation on the metal originates a
pentatomic ring slightly puckered in an envelope shape (Cremer
and Pople53 puckering coordinates: QT = 0.15, θ2 = 115°). The
two C-O bonds of the keto enolic system are perfectly equiv-
alent in length, while a certain asymmetry is observed on the
carbon backbone, since the C8-C9 bond is significantly shorter
of [Mn(HL2)2(CH3OH)2] 2CH3OH confirms the role this
3
512
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
moiety has in molecular recognition, even when it does not
directly coordinate the metal ions. Note, also, that the OH methanol
coordinated to magnesium was recently used to simulate the
processed viral 30-DNA end, ready to attack a host DNA phos-
phodiester bond in a theoretical study on magnesium chelation
of IN inhibitors.52
Potentiometric Measurements and Calculations. In order
to obtain more information about the behavior of H2L2 in
solution, a potentiometric study was carried out with Mg2þ and
Mn2þ. To reduce solubility problems, the titrations were per-
formed in methanol/water = 9/1 v/v, where all the species are
soluble in the examined pH range. First, the protonation con-
stants of the ligand were determined as pKa1 = 8.226(0.008) and
pKa2 = 11.431(0.008); the former is attributed to the enol OH,
the latter to the proton bound to the nitrogen in the heterocycle.
The metal ion coordination lowers the pKa1 so that complexes
with the monodeprotonated ligand are formed also in an acidic
environment (see Figure 6). Afterward, the stoichiometry of the
complexes with Mn2þ and Mg2þ were determined. 1:1 and 1:2
metal to ligand ratios have been considered, and the correspond-
ing diagrams are shown (Figure 6 and Figure S1 in the Support-
ing Information). The best fit of the experimental titration curves
was obtained by the set of species shown in Table 1. The spe-
cies found in solution for Mn(II) ([M(HL2)]þ, M(HL2)2,
[M(HL2)L2]-, ML2 and [ML2(OH)]-) are the same for Mg-
(II), with similar formation constants for the two systems. This
model is corroborated by very good statistical parameters and by
the overlapping between experimental and computed titration
curves. Dimeric models of type M2(ligand)2 were also taken into
account, but gave rise to a worse fitting or were rejected by
calculation. However, the observation of a 1:1 species in solution
is not in contrast with the isolation of dimeric complexes in
solid state.
Figure 8. Comparison of inhibition of HIV-1 IN catalytic activities of
complexes 9 and 10 (a) and 11 and 12 (b) using either MnCl2 or MgCl2
in the reaction buffer.
It is worth noting that Mn(II) is more efficient than Mg(II) in
fully deprotonating the ligand: [Mn(HL2)L2]- and MnL2 start
forming at pH ∼ 6 whereas [Mg(HL2)L2]- starts forming at pH
around 7.5, and MgL2 is present in very small amount in alkaline
environment. Besides, the [ML2(OH)]- species is formed in a
small amount only above pH 10. As shown by the distribution
diagrams in Figure 6, at a physiological pH, the species [MHL2]þ
and M(HL2)2 are the most abundant for Mg(II), whereas all the
species, excluding [MnL2(OH)]-, are found for Mn(II).
In conclusion, H2L2 behaves in a different way with respect to
the previously studied DKA ligand.37,38 In fact, as can be seen in
the distribution diagram, complexes with a 1:1 metal to ligand
ratio are less stable with H2L2 than with the parent DKA com-
pound H2L1. Moreover, the formation constants of M(HL2)2 are
much greater than for the other species in solution.
(IC50, 30-processing > 333; strand transfer = 69 ( 4 μM).37 These
data further confirm the fundamental role of the hydrophobic
substituent of the chelating moiety, and also that the substitution
of the diketo acid functionality with a triazolic ring does not affect
the activity.
The complexes 9-12 retained an activity profile analogous to
the corresponding free ligands (Table 2 and Figure 7), with IC50
values for inhibition of strand transfer ranging from 0.15 to 1.6
μM for 9 and 10, and from 19 to 49 μM for 12 and 11,
respectively (for representative gels, see Figure 7b). Interestingly,
with IC50 values of 0.15 ( 0.03 and 3 ( 0.01 μM (strand transfer
and 30-processing, respectively) the manganese complex 10
proved to be the most potent compound, with 6-fold greater
potency than its magnesium congener 9 (IC50s = 0.9 ( 0.5
and 52 ( 18 μM, for strand transfer and 30-processing). A
similar “metal dependent”37,55 behavior was also shown by 11
and 12.
Complexes 9 and 10, as well as 11 and 12, inhibit IN in similar
concentration range, when tested in the presence of either Mg or Mn
ions in the reaction buffer (Figure 8). When tested in the presence of
Mg ions in the reaction buffer, 9 and 10 show an increase in potency,
whereas a slight decrease in inhibition activity for 11 and 12 was
observed (Figure 8). These results could be related to the selec-
tivity of the free ligands vs the metal, resulting in different equi-
libria in the medium. The concentration of Mg2þ in vitro and
in vivo therefore makes plausible the hypothesis that the free
ligands could also act as complexes in their active form, as they
Inhibition of HIV-1 IN. The ligands H2L1 and H2L2 and their
complexes 9-12 were tested for their ability to inhibit 30-
processing and strand transfer catalytic activities by oligonucleo-
tide-based assays (Table 2 and Figure 7). Inhibition of strand
transfer activities in in vitro assays for complexes 9-12 were also
evaluated by using either Mn2þ and Mg2þ. All tested compounds
showed anti-IN activity in the nanomolar/micromolar range. In
particular, as far as the inhibitory activities of the ligands are
concerned, H2L1 (IC50, strand transfer = 0.5 ( 0.3 μM; 30-
processing = 9 ( 2 μM) was 100-fold more potent than H2L2
(IC50, strand transfer = 43 ( 7 μM; 30-processing > 100 μM).
H2L2 was about 70-fold less potent than its parent compound
S-1360 (IC50 = 0.6 ( 1 μM against strand transfer);54 it has
an activity similar to the previously studied H2L (Figure 2a)
513
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Figure 9. Metal complex ML2 and related proposed structures highlighting the hypothetical metal binding model into the active site of the IN-DNA
intasome (depicted in cartoon). Magenta spheres indicate the Mg2þ ions of the complex inhibitors, while the Mg2þ ions in the catalytic core are shown as
orange spheres. Panels a-f show graphical representations of putative interactions of complexes/ligand in the IN active site. The complex ML2 (a) or
ML (c) bind directly to IN. ML2 (b) or ML (d) interacts with IN and chelates a second ion on the active site. A complex with one (e) or two (f) metal ions
in the site may be formed upon binding of ML2 or ML to IN and subsequent dynamic exchange of the metal ions. The structure pdb.3L2Q was selected
and optimized to be used as a model target. These images were prepared using the PyMOL program.
could coordinate ions in solution before interaction at their
putative binding site.
terminal 30-OH functionality may be interacting with one mag-
nesium ion. The biological profile of complex 12, with respect to
that of its free ligand H2L2, seems to support this assumption.
Finally, as expected, selective strand transfer inhibition was confir-
med for all tested compounds, as evidenced by their selectivity indices.
One- or Two-Metal Binding Inhibition Mechanism?. A
simple molecular recognition study of the binding mode of
ligand/Mg2þ complexes in 1:1 and 2:1 (L:M) stoichiometric
ratio on the active site of IN was performed by means of a
molecular modeling docking approach. On the basis of the X-ray
crystal structure of IN-DNA intasome (pdb.3L2Q), a model
active site with a set of magnesium species in the IN active site
was built by manual docking. Energy minimization and equili-
bration and visual inspection were performed by the Molecular
Operating Environment (MOE) program. Trajectory and poses
were sampled and visually examined by using the PyMOL
program. The triazolic nitrogen of the ligand was taken as neutral,
as suggested from the pKa, even if it is known that the pKa of
several ligands can vary by about 1-1.5 units upon coordination
or within enzyme active site.
If we take into consideration H2L2 and its complexes, it is clear
from potentiometric and synthetic data that a direct involvement
of the triazole in the coordination to a second metal ion would be
difficult. The local conditions inside the active site could allow
the deprotonation of the triazolic moiety and the coordination of
a second metal ion. Otherwise, the triazolic group can establish
hydrogen bonding with the protein side chain. This fact, in
synergy with the complexation of one metal ion, can block the
interaction between IN and the DNA substrate, in analogy with
the situation found in the crystal structure of 5CITEP in complex
with the enzyme.12 In other words, the reactive species could be
the monometallic complex that in the intasome interacts with the
second metal ion and the protein. Again, as recently observed by
theoretical investigation,52 when a methanol molecule, which
simulates the terminal 30-OH of viral DNA, replaces a coordi-
nated water molecule in the complex, as in [Mn(HL2)2-
(CH3OH)2] 2CH3OH, the metal chelation is still well estab-
3
lished. This suggests that, in the real binding site of IN, the
514
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
A graphical representation of putative interactions of the metal
complex MgL2 and related proposed structures into the active
site of the IN-DNA intasome is depicted in Figure 9 (panels a-f).
Several hypotheses may be invoked. The complex ML2 (a)
or ML (c) can bind directly to IN, interacting as a complex
according to an “interfacial mechanism”: the complex could act as
a static entity, preventing the interaction between the enzyme
and the DNA substrate, thus blocking the integration at the
strand transfer stage. It is also possible that ML2 (b) or ML (d)
binds to IN and chelates a second ion that is present on the active
site, giving rise to a dimer within the active site. Furthermore, a
complex with one (e) or two (f) metal ions in the site may be
formed upon binding of ML2 or ML to IN and subsequent
dynamic exchange of the metal ions (Figure 9).
’ EXPERIMENTAL SECTION
Materials and Methods. Anhydrous solvents and all reagents
were purchased from Sigma-Aldrich, Merck or Carlo Erba and
used without further purification. All reactions involving air- or
moisture-sensitive compounds were performed under a nitrogen
atmosphere using oven-dried glassware and syringes to transfer
solutions. Melting points (mp) were determined using an Elec-
trothermal melting point or a K€ofler apparatus and are uncor-
rected. Analytical thin-layer chromatography (TLC) was carried
out on Merck silica gel F-254 plates. Flash chromatography
purifications were performed on Merck silica gel 60 (230-400
mesh ASTM) as the stationary phase.
NMR spectra were recorded at 27 °C on a Bruker Avance
300 FT and Varian XL-200 spectrophotometers by using SiMe4
as internal standard; the assignment of exchangeable protons
(OH and NH) was confirmed by the addition of D2O. IR spectra
were obtained with a Nicolet 5PCFT-IR spectrophotometer in
the 4000-400 cm-1 range, in reflectance mode on the powder.
Elemental analyses were performed by using a Carlo Erba Model
EA 1108 apparatus. The ESI(þ)-MS spectra were collected by
using a quadrupole-time-of-flight micro mass spectrometer (Micro-
mass, Manchester, U.K.) equipped with a pneumatically assisted ESI
interface. The system was controlled by Masslynx software version
4.0 (Micromass). The nebulizing gas (nitrogen, 99.999% purity)
and the desolvation gas (nitrogen, 99.998% purity) were delivered at
a flow-rate of 10 and 600 L/h respectively. Continuum mode full-
scan mass spectra were acquired using an acquisition time of 1 s and
an interscan delay of 0.1 s. QqTOF external calibration was
performed using a 0.1% phosphoric acid solution and a fifth-order
nonlinear calibration curve was usually adopted.
’ CONCLUSIONS AND PERSPECTIVES
The present study focuses on H2L1 and H2L2, two DKA HIV-1
IN inhibitors, and, in particular, on their Mg2þ and Mn2þ
complexes, with the aim to elucidate at the molecular level some
important mechanistic aspects that could be extended to the
chelating inhibitors in general. By comparing the spectro-
scopic, potentiometric and structural data, it is clear that the
diketo acid functionality chelates divalent metal ions, forming
complexes with metals in different stoichiometric ratios.
Moreover, analysis of the biological results suggests that these
compounds can also act as complexes in their active form.
Moreover, the electronic properties of the aromatic frame-
work influence the metal-chelating ability of the pharmaco-
phore and, consequently, the activity. Therefore, the diffe-
rence in activities may be related to the complexes they
preferentially form in solution.
Intermediates 5-7 were prepared using the starting material 3
and 4 as previously described.48
Another relevant point that emerges from this study is that
H2L2 preferentially chelates only one metal ion through the
keto-enol fragment, while the triazolic moiety is excluded by
coordination at least at normal physiological conditions. This is
confirmed by spectroscopic and potentiometric measurements
Methyl 4-(3,5-Bis(benzyloxy)phenyl)-2-hydroxy-4-oxo-
but-2-enoate (2). To a solution of 3,5-dibenzyloxy-acetophe-
none (1, 8.3 mmol) and dimethyl oxalate (10.0 mmol) in DMF
(10 mL), NaH (60% oil dispersion, 10.0 mmol) was added at
5 °C. The reaction mixture was stirred at rt for 3.5 h and then
heated at 50 °C for 1 h. After cooling in an ice-bath, the reaction
was quenched with water and acidified with 3 N HCl. The
precipitate was filtered and washed several times with water to
afford a white powder corresponding to a mixture of the Z/E
isomers (6:1). Yield: 73%. Mp: 91-92 °C [lit. 121-123 °C, for
and by X-ray diffraction analysis on [Mn(HL2)2(CH3OH)2]
3
2CH3OH. However, the triazolic moiety is involved in a complex
web of hydrogen bonds both in the structure of the ligand and
in the structure of the Mn2þ complex; in analogy, it could be
involved in extensive interactions with IN and/or DNA.
Therefore, at least for this ligand, a two-metal binding model
could be reformulated: inhibition of IN by DKA-like inhibitors
strictly requires the chelation of at least one metal ion within
the catalytic core. It can be thought that the selectivity dis-
played by DKAs for strand transfer could not derive by the fact
that inhibitors need both metal ions to bind strongly to the
enzyme. It is plausible that chelation involves, in the first
instance, only one metal and that binding of the inhibitor to
the protein is possible only after the conformational changes
occurring with 30-processing, which would allow the accom-
modation of the ligand within the catalytic core. This would
also explain the importance of the aromatic substituents of the
inhibitors, since the hydrophobic portion is fundamental in
orientating the drug within the protein active site.
E-isomer].47 IR (cm-1): νCdOester 1725, νCdOketo 1625. H
1
NMR (DMSO-d6) for the Z-isomer: δ 7.48-7.34 (m, 10H,
Ar-H), 7.27 (s, 2H, Ar-H), 7.14 (s, 1H, Ar-H), 7.01 (s, 1H,
CH), 5.19 (s, 4H, CH2), 3.86 (s, 3H, OCH3). ESI/MS (þ, m/z):
418 [Mþ]. Anal. Calcd for C25H22O6: C 71.75; H 5.30. Found: C
71.68; H 5.61.
(Z)-4-(3,5-Bis(benzyloxy)phenyl)-2-hydroxy-4-oxobut-2-
enoic Acid (H2L1). To a solution of 2 (1.0 mmol) in methanol
(15 mL), 2 N NaOH (4.0 mmol) was added. The resulting
mixture was vigorously stirred for 5 h. After dilution with water,
the reaction mixture was acidified with 1 N HCl. The white-beige
precipitate was filtered off, washed with water, and recrystallized
from hexane/ethyl acetate to give a pale yellow powder. Yield:
85%. Mp: 171 °C [lit. 170-172 °C].25 IR (cm-1): νCdOacid
1
We expect that the structural insights obtained from this study
can be useful to look at the “two metal chelation” model in a
larger context, where, together with coordination, hydrogen
bonding and hydrophobic interactions should be considered as
a whole.
1707, νCdOketo 1623. H NMR (DMSO-d6): δ 7.49-7.30
(m, 10H, Ar-H), 7.26 (d, 2H, Ar-H), 7.09 (s, 1H, Ar-H),
7.00 (s, 1H, CH), 5.19 (s, 4H, CH2). ESI/MS (þ, m/z): 404
[Mþ]. Anal. Calcd for C24H20O6: C 71.26; H 4.99. Found: C
71.00; H 5.07.
515
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
Synthesis of (Z)-3-Hydroxy-1-phenyl-3-(1-trityl-1H-1,2,4-
triazol-5-yl)prop-2-en-1-one (8). A 1 M solution of LiHMDS in
THF (3.87 mmol) was added at -70 °C to a solution of
acetophenone (2.98 mmol) in THF (9 mL) under a nitrogen
atmosphere. The reaction mixture was slowly warmed to -10 °C
and cooled to -70 °C. 7 (3.87 mmol) in THF (15 mL) was
added, and the reaction mixture was slowly warmed to rt and
stirred at the same temperature for 1.5. h. The reaction was
quenched with a 10% aqueous solution of NH4Cl and was
acidified with 3 N HCl. The solution was extracted with ethyl
acetate and the combined organic layers were washed with brine
and dried over sodium sulfate, filtered and the solvent removed in
vacuo to give an orange solid that was triturated from diethyl
ether to afford a pale orange solid. Yield: 89%. Mp: 169-171 °C
for C22H16MgN6O4 4H2O: C 50.36, H 4.61, N 16.00. Found: C
3
50.48; H 4.35; N 15.78. ESI/MS (þ, m/z): 453 [MHþ].
Mn(HL2)2 4H2O (12). The same as Mg(HL2)2, by using
3
Mn(CH3COO)2. A yellow precipitated was formed, which was
filtered off and washed with water/diethyl ether. IR (cm-1): νNH
=
3397; νCdO = 1598, 1569, 1417. Anal. Calcd for C22H16Mn-
N6O4 4H2O: C 47.61; H 4.35; N 15.13. Found: C 47.92; H 4.17;
3
N 15.09. ESI/MS (þ, m/z): 484 [MHþ].
Mg2L2 5H2O (13). Ligand H L2 (100 mg, 0.47 mmol) was
2
2
3
dissolved in 15 mL of methanol, and KOH 1 M was added until
pH 10 was reached. Mg(CH3COO)2 (1 equiv) dissolved in 2 mL
of water was added at room temperature. After a few minutes, a
white precipitated was formed, which was stirred at room
temperature for 2 h. It was then filtered off and washed with
1
water. IR (cm-1): νOH = 3420; νCdO = 1605, 1571, 1527. H
1
(dec). IR (cm-1): νCdOcarbonyl 1615. H NMR (DMSO-d6):
NMR (DMSO-d6): δ 8.18-7.93 (br, 3H, CH þ Harom); 7.54
δ 14.60-14.20 (s, br, 1H, OH), 8.31 (s, 1H, CH), 8.02 (d, 2H,
Ar-H), 7.70-7.25 (m, 12H, Ar-H), 7.22 (s, 1H, COCH),
7.20-6.92 (m, 7H, Ar-H). ESI/MS (þ, m/z): 457 [Mþ].
Synthesis of (Z)-3-Hydroxy-1-phenyl-3-(1H-1,2,4-triazol-
3-yl)prop-2-en-1-one (H2L2). 3 N HCl (7.8 mL) was added
to a solution of compound 8 (2.34 mmol) in 1,4-dioxane
(30 mL), and the mixture was stirred at 80 °C for 30 min. Then,
the reaction was cooled to rt and the solvent was evaporated
in vacuo to give a residue, which was dissolved in diethyl ether and
extracted with 1 N NaOH. The combined aqueous layers were
acidified with 3 N HCl, and the precipitate formed was filtered
and washed with water and ethyl acetate. The solid residue was
crystallized from ethyl acetate to afford a yellow solid. Yield: 85%.
Mp: 197-198 °C (dec). IR (cm-1): νCdO 3580, νCdOcarbonyl
1610, νCdC 1573. 1H NMR (DMSO-d6): δ 14.80-14.60 (s, br,
1H, OH), 8.73 (s, br, 1H, CH), 8.07-7.97 (m, 2H, Ar-H),
7.71-7.51 (m, 3H, Ar-H), 7.28 (s, 1H, COCH), 4.80 (CH2,
keto isomer). ESI/MS (þ, m/z): 215 [Mþ]. Anal. Calcd for
(br, 3H, Harom); 6.99 (s, 1H, CHenolic). Anal. Calcd for C22H14-
Mg2N6O4 5H2O: C 46.82; H 4.28; N 14.88. Found: C 47.08; H
3
4.10; N 14.57. ESI/MS (þ, m/z): 475 [MHþ].
X-ray Crystallography. Crystals of H2L2 (1/2)HCl suitable
3
for X-ray diffraction were obtained by slow evaporation of a
methanol/HCl solution of the ligand. Data were processed by
Lorentz and polarization corrections for H2L2 (1/2)HCl, and by
3
the SAINT package,56 corrected for absorption effects by the
SADABS57 procedure (Tmax/Tmin = 1.000/0.699), data reduc-
tion performed up to d = 0.90 Å, for [Mn(HL)2(CH3-
OH)2] 2CH3OH. The phase problem was solved by direct
3
methods58 and refined by full matrix least-squares on all F2.
Anisotropic displacement parameters59 were refined for all non-
hydrogen atoms, while hydrogen atoms were located from
difference Fourier maps; methanol hydrogens were introduced
in idealized positions. The final maps were featureless. For the
discussion use was made of the Cambridge Structural database
facilities.60
C11H9N3O2 0.25H2O: C 60.13; H 4.35; N 19.11. Found: C
3
Potentiometric Measurements. The metal ions stock solu-
59.86; H 4.04; N 18.85.
tions were prepared from MnCl2 4H2O (Janssen) and Mg-
Mg2L2 5H2O (9). Ligand H2L1 (100 mg, 0.2 mmol) was
3
1
3
Cl2 6H2O (Aldrich). Their concentrations were determined by
3
dissolved in 20 mL of warm methanol and the pH adjusted to 7
using EDTA as a titrant. The sodium salt of Eriochrome black T
in the presence of triethanolamine and hydroxylamine chloride
for Mn(II) and Mg(II) was used as an indicator. Equilibrium
constants for protonation and complexation reactions were
determined by means of potentiometric measurements, carried
out in methanol/water = 9:1 v/v solution at ionic strength 0.1 M
KCl and 25 ( 0.1 °C, in the pH range 2.5-11 under N2.
Temperature was controlled to (0.1 °C by using a thermostatic
circulating water bath (ISCO GTR 2000 IIx). Appropriate
aliquots of ligand solution, prepared by weight, were titrated
with standard KOH (solvent: methanol/water = 9:1 v/v, I =
0.1 M KCl) with and without metal ions, applying constant-
speed magnetic stirring. Freshly boiled methanol and bidistilled
water, kept under N2, were used throughout. The experimental
procedure in order to reach very high accuracy in the determina-
tion of the equilibrium constants in this mixed solvent has been
described in detail elsewhere.61 The protonation constants of the
ligands were obtained by titrating 20-50 mL samples of each
ligand (1 ꢀ 10-3 to 3 ꢀ 10-3 M). For the complex formation
constants, the titrations were performed with different ligand/
metal ratios (1 up to 4). At least two measurements (about 60
experimental points each) were performed for each system.
Potentiometric titrations were carried out by a fully automated
apparatus equipped with a CRISON GLP 21-22 digital voltmeter
(resolution 0.1 mV) and a 5 mL Metrohm Dosimat 655 autoburet,
by using NaOH 1 M. Mg(OH)2 (14 mg, 0.2 mmol) was added,
and the reaction mixture was stirred for 24 h at room tempera-
ture. On concentrating the solution, a light yellow precipitate
appeared, which was filtered off and washed with water. IR (cm-1):
νOH = 3695; νCdO = 1594, 1373. 1H NMR (DMSO-d6):
δ 7.47-7.31 (m, 11H, Ar-H); 7.17 (s, 2H, Ar-H), 6.96 (s,
1H, CHenolic); 5.16 (s, 4H, OCH2). Anal. Calcd for C48H36-
Mg2O12 5H2O: C 61.43; H 4.94. Found: C 61.37; H 5.51.
3
Mn2L2 H2O (10). Ligand H2L1 (100 mg, 0.2 mmol) was
1
3
dissolved in 20 mL of warm methanol and added to an aqueous
solution of Mn(CH3COO)2 (60 mg, 0.2 mmol). Immediately, a
yellow-brown precipitated was formed, which was stirred for
additional 4 h, filtered off and washed with water. IR (cm-1):
νCdO = 1575, 1341. Anal. Calcd for C48H36Mn2O12 H2O: C
3
61.81; H 4.10. Found: C 61.79; H 4.10.
Mg(HL2)2 4H2O (11). Ligand H2L2 (150 mg, 0.7 mmol) was
3
dissolved in 20 mL of methanol and the pH adjusted to 8 by using
1 M KOH. An aqueous solution of MgCl2 (0.35 mmol) was
added, and a white precipitate appeared. The reaction mixture
was stirred at room temperature for 2 h, and the precipitate was
filtered off and washed with water. IR (cm-1): νOH = 3640; νNH
=
3422; νCdO = 1597, 1540, 1426. 1H NMR (DMSO-d6):
δ 14.30 (s, br, 1H, NH), 8.21 (s, br, 1H, CH); 7.91 (d, 2H,
Harom); 7.44 (m, 3H, Harom), 6.94 (s, 1H, CHenolic). Anal. Calcd
516
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
both controlled by a homemade software in BASIC, working on
an IBM computer. The electrodic chain (Crison 5250 glass elec-
trode and KCl 0.1 M in methanol/water = 9:1 v/v calomel
electrode, Radiometer 401) was calibrated in terms of [Hþ] by
means of a strong acid-strong base titration, by Gran’s method,62
allowing the determination of the standard potential, E° (372.5 (
0.4 mV), and of the ionic product of water, Kw (pKw = 14.40 (
0.05) in the experimental conditions used. The software HYPER-
QUAD63 was used to evaluate the protonation and complexation
constants from emf data.
Biological Materials, Chemicals, and Enzymes. All com-
pounds were dissolved in DMSO, and the stock solutions were
stored at -20 °C. The γ[32P]-ATP was purchased from Perkin-
Elmer. The expression system for wild-type IN was a generous
gift of Dr. Robert Craigie, Laboratory of Molecular Biology,
NIDDK, NIH, Bethesda, MD.
Preparation of Oligonucleotide Substrates. The oligonu-
cleotides 21top, 50-GTGTGGAAAATCTCTAGCAGT-30, and
21bot, 50-ACTGCTAGAGATTTTCCACAC-30, were purchased
from Norris Cancer Center Core Facility (University of South-
ern California) and purified by UV shadowing on polyacrylamide
gel. To analyze the extent of 30-processing and strand transfer
using 50-end labeled substrates, 21top was 50-end labeled using
T4 polynucleotide kinase (Epicenter, Madison, WI) and γ[32P]-
ATP (Amersham Biosciences or ICN). The kinase was heat-
inactivated, and 21bot was added in 1.5 molar excess. The mixture
was heated at 95 °C, allowed to cool slowly to room temperature,
and run through a spin 25 minicolumn (USA Scientific) to
separate annealed double-stranded oligonucleotide from unin-
corporated material.
1:1. X-ray diffraction analysis for H2L2 (1/2)HCl and complex
3
[Mn(HL2)2(CH3OH)2] 2CH3OH with crystal data, structure
3
refinement, and most relevant bond lengths and angles. This
material is available free of charge via the Internet at http://pubs.
acs.org. Crystallographic data (excluding structure factors) for
H2L2 (1/2)HCl and [Mn(HL2)2(CH3OH)2] 2CH3OH have
3
3
been deposited with the Cambridge Crystallographic Data
Centre as supplementary publications nos. CCDC 791917-
791918. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.
(fax: (þ44) 1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
’ AUTHOR INFORMATION
Corresponding Author
*University of Parma, Dipartimento di Chimica Generale ed
Inorganica, Chimica Analitica, Chimica Fisica, Viale delle Scienze
17/A, 43124 Parma, Italy. Phone: þ39 0521-905-424. Fax: þ39
0521-905-557. E-mail: dominga.rogolino@unipr.it.
’ ACKNOWLEDGMENT
The authors thank the “Centro Interfacoltꢀa Misure Giuseppe
Casnati” and the “Laboratorio di Strutturistica Mario Nardelli” of
the University of Parma for facilities and Giuseppe Foroni for
technical assistance. M.S. is grateful to Fondazione Banco di
Sardegna for its partial financial support and to Dr. Andrea Brancale
for the use of the MOE program. The work in N.N.’s laboratory was
supported by funds from the Campbell Foundation.
Integrase Assays. To determine the extent of 30-processing
and strand transfer, wild-type IN was preincubated at a final
concentration of 200 nM with the inhibitor in reaction buffer
(50 mM NaCl, 1 mM HEPES, pH 7.5, 50 μM EDTA, 50 μM
dithiothreitol, 10% glycerol (w/v), 7.5 mM MnCl2, 0.1 mg/mL
bovine serum albumin, 10 mM 2-mercaptoethanol, 10% DMSO,
and 25 mM MOPS, pH 7.2) at 30 °C for 30 min. Then, 20 nM of
the 50-end 32P-labeled linear oligonucleotide substrate was added,
and incubation was continued for an additional 1 h. Reactions
were quenched by the addition of an equal volume (16 μL) of
loading dye (98% deionized formamide, 10 mM EDTA, 0.025%
xylene cyanol and 0.025% bromophenol blue). An aliquot (5 μL)
was electrophoresed on a denaturing 20% polyacrylamide gel (0.09
M tris-borate pH 8.3, 2 mM EDTA, 20% acrylamide, 8 M urea).
Gels were dried, exposed in a PhosphorImager cassette,
analyzed using a Typhoon 8610 Variable Mode Imager (Amer-
sham Biosciences) and quantitated using ImageQuant 5.2.
Percent inhibition (% I) was calculated using the following
equation:
’ ABBREVIATIONS USED
HIV-1, human immunodeficiency virus type 1; IN, integrase; ST,
strand transfer; DKA, diketo acid; S-1360, (2Z)-1-[5-(4-fluoro-
benzyl)-2-furyl]-3-hydroxy-3-(1H-1,2,4-triazol-5-yl)prop-2-en-
1-one; CSD, Cambridge Structural Database; LiHMDS, lithium
bis(trimethylsilyl)amide.
’ REFERENCES
(1) Pommier, Y.; Johnson, A. A.; Marchand, C. Integrase inhibitors
to treat HIV/AIDS. Nat. Rev. Drug Discovery 2005, 4, 236–248.
(2) Neamati, N.; Marchand, C.; Pommier, Y. HIV-1 integrase
inhibitors: past, present, and future. Adv. Pharmacol. 2000, 49, 147–165.
(3) D’Angelo, J.; Mouscadet, J. F.; Desmaele, D.; Zouhiri, F.; Leh, H.
HIV-1 integrase: the next target for AIDS therapy?. Pathol. Biol. 2001,
49, 237–246.
(4) Neamati, N. Structure-based HIV-1 inhibitor design: a future
perspective. Expert Opin. Invest. Drugs 2001, 10, 281–296.
(5) Anthony, N. J. HIV-1 integrase: a target for new AIDS che-
motherapeutics. Curr. Top. Med. Chem. 2004, 4, 979–990.
(6) Nair, V.; Chi, G. HIV integrase inhibitors as therapeutic agents in
AIDS. Rev. Med. Virol. 2007, 17, 277–295.
(7) Dayam, R.; Al-Mawsawi, L. Q.; Neamati, N. HIV-1 Integrase
Inhibitors: An Emerging Clinical Reality. Drugs R&D 2007, 8, 155–168.
(8) MK-0518 meeting transcript. Department of Health and Human
Services; Food and Drug Administration; Center for Drug Evaluation and
Research; Antiviral Drugs Advisory Committee; 2007 September 5. Avail-
able from www.fda.gov/ohrms/dockets/ac/cder07.htm#AntiviralDrugs.
(9) Al-Mawsawi, L. Q.; Al-Safi, R. I.; Neamati, N. Expert. Opin.
Emerging Drugs 2008, 13, 213–225.
% I ¼ 100 ꢀ ½1 - ðD - CÞ=ðN - CÞꢁ
where C, N, and D are the fractions of 21-mer substrate converted
to 19-mer (30-processing product) or ST products for DNA
alone, DNA plus IN, and IN plus drug, respectively. The IC50
values were determined by plotting the logarithm of drug con-
centration versus percent inhibition to obtain concentration that
produced 50% inhibition.
’ ASSOCIATED CONTENT
(10) Kulkosky, J.; Jones, K. S.; Katz, R. A.; Mack, J. P.; Skalka, A. M.
Residues critical for retroviral integrative recombination in a region that
is highly conserved among retroviral/retrotransposon integrases and
S
Supporting Information. Distribution diagrams for mag-
b
nesium and manganese complexes with H2L2 and ligand:M ratio
517
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
bacterial insertion sequence transposases. Mol. Cell. Biol. 1992, 12,
2331–2338.
(11) Hazuda, D. J.; Felock, P.; Witmer, M.; Wolfe, A.; Stillmock, K.;
Grobler, J. A.; Espeseth, A.; Gabryelski, L.; Schleif, W.; Blau, C.; Miller,
M. D. Inhibitors of strand transfer that prevent integration and inhibit
HIV-1 replication in cells. Science 2000, 287, 646–50.
(12) Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.; Yoshinaga,
T.; Fujishita, T.; Sugimoto, H.; Endo, T.; Murai, H.; Davies, D. R.
Structure of the HIV-1 integrase catalytic domain complexed with an
inhibitor: a platform for antiviral drug design. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 13040–13043.
(13) Pais, G. C. G.; Burke, T. R. Novel aryl diketo-containing
inhibitors of HIV-1 integrase. Drugs Future 2002, 27, 1101–1111.
(14) Egbertson, M. S. HIV integrase inhibitors: from diketoacids to
heterocyclic templates: a history of HIV integrase medicinal chemistry at
Merck West Point and Merck Rome (IRBM). Curr. Top. Med. Chem.
2007, 7, 1251–1272.
Pommier, Y.; Burke, T. R., Jr. Structure Activity of 3-Aryl-1,3-diketo-
containing compounds as HIV-1 Integrase inhibitors. J. Med. Chem.
2002, 45, 3184–3194.
(26) Zhao, X. Z.; Maddali, K.; Vu, B. C.; Marchand, C.; Hughes, S. H.;
Pommier, Y.; Burke, T. R., Jr. Examination of halogen substituent effects on
HIV-1 integrase inhibitors derived from 2,3-dihydro-6,7-dihydroxy-1H-
isoindol-1-ones and 4,5-dihydroxy-1H-isoindole-1,3(2H)-diones. Bioorg.
Med. Chem. Lett. 2009, 19, 2714–2717.
(27) Kirschberg, T.; Parrish, J. Metal chelators as antiviral agents.
Curr. Opin. Drug Discovery 2007, 10, 460–472.
(28) Shaw-Reid, C. A.; Munshi, V.; Graham, P.; Wolfe, A.; Witmer,
M.; Danzeisen, R.; Olsen, D. B.; Carroll, S. S.; Embrey, M.; Wai, J. S.;
Miller, M. D.; Cole, J. L.; Hazuda, D. J. Inhibition of HIV-1 Ribonuclease
H by a Novel Diketo Acid, 4-[5-(Benzoylamino)thien-2-yl]-2,4-dioxo-
butanoic Acid. J. Biol. Chem. 2003, 278, 2777–2780.
(29) Tramontano, E.; Esposito, F.; Badas, R.; Di Santo, R.; Costi, R.;
La Colla, P. 6-[1-(4-Fluorophenyl)methyl-1H-pyrrol-2-yl]-2,4-dioxo-5-
hexenoic acid ethyl ester a novel diketo acid derivative which selectively
inhibits the HIV-1 viral replication in cell culture and the ribonuclease H
activity in vitro. Antiviral Res. 2005, 65, 117–124.
(30) Billamboz, M.; Bailly, F.; Barreca, M. L.; De Luca, L.; Mousca-
det, J.-F.; Calmels, C.; Andreola, M.-L.; Witvrouw, M.; Christ, F.;
Debyser, Z.; Cotelle, P. Design, synthesis, and biological evaluation of
a series of 2-hydroxyisoquinoline-1,3(2H,4H)-diones as dual inhibitors
of human immunodeficiency virus type 1 integrase and the reverse
transcriptase RNase H domain. J. Med. Chem. 2008, 51, 7717–7730.
(31) Wang, Z.; Tang, J.; Salomon, C. E.; Dreis, C. D.; Vince, R.
Pharmacophore and structure-activity relationships of integrase inhibi-
tion within a dual inhibitor scaffold of HIV reverse transcriptase and
Integrase. Bioorg. Med. Chem. 2010, 18, 4202–4211.
(32) Kijama, R.; Kawasuji, T. PCT Int. Appl. 2001, WO-01/95905.
(33) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara, T.;
Kiyama, R. A platform for designing HIV integrase inhibitors. Part 2: A
two-metal binding model as a potential mechanism of HIV integrase
inhibitors. Bioorg. Med. Chem. 2006, 14, 8420–8429.
(15) Wang, Y.; Serradell, N.; Bolos, J.; Rosa, E. MK-0518, HIV
integrase inhibitor. Drugs Future 2007, 32, 118–122.
(16) Rowley, M. The Discovery of Raltegravir, An Integrase Inhi-
bitor for the Treatment of HIV Infection. Prog. Med. Chem. 2008,
46, 1–28.
(17) Summa, V.; Petrocchi, A.; Bonelli, F.; Crescenzi, B.; Donghi,
M.; Ferrara, M.; Fiore, F.; Gardelli, C.; Gonzalez, P. O.; Hazuda, D. J.;
Jones, P.; Kinzel, O.; Laufer, R.; Monteagudo, E.; Muraglia, E.; Nizi, E.;
Orvieto, F.; Pace, P.; Pescatore, G.; Scarpelli, R.; Stillmock, K.; Witmer,
M. V.; Rowley, M. Discovery of raltegravir, a potent, selective orally
bioavailable HIV-integrase inhibitor for the treatment of HIV-AIDS
infection. J. Med. Chem. 2008, 51, 5843–5855.
(18) Sato, M.; Motomura, T.; H. Aramaki, H.; Matsuda, T.; Yamashita,
M.; Ito, Y.; Kawakami, H.; Matsuzaki, Y.; Watanabe, W.; Yamataka, K.;
Ikeda, S.; Kodama, E.; Matsuoka, M.; Shinkai, H. Novel HIV-1 Integrase
Inhibitors Derived from Quinolone Antibiotics. J. Med. Chem. 2006,
49, 1506–1508.
(19) Dayam, R.; Al-Mawsawi, L. Q.; Zawahir, Z.; Witvrouw, M.;
Debyser, Z.; Neamati, N. Quinolone 3-Carboxylic Acid Pharmacophore:
Design of Second Generation HIV-1 Integrase Inhibitors. J. Med. Chem.
2008, 51, 1136–1144.
(34) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara, T.;
Kiyama, R. 3-Hydroxy-1,5-dihydro-pyrrol-2-one derivatives as advanced
inhibitors of HIV integrase. Bioorg. Med. Chem. 2007, 15, 5487–5492.
(35) Grobler, J. A.; Stillmock, K.; Hu, B.; Witmer, M.; Felock, P.;
Espeseth, A. S.; Wolfe, A.; Egbertson, M.; Bourgeois, M.; Melamed, J.;
Wai, J. S.; Young, S.; Vacca, J.; Hazuda, D. J. Diketo acid inhibitor
mechanism and HIV-1 integrase: implications for metal binding in the
active site of phosphotransferase enzymes. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 6661–6666.
(20) Di Santo, R.; Costi, R.; Roux, A.; Artico, M.; Lavecchia, A.;
Marinelli, L.; Novellino, E.; Palmisano, L.; Andreotti, M.; Amici, R.;
Galluzzo, C. M.; Nencioni, L.; Palamara, A. T.; Pommier, Y.; Marchand,
C. Novel Bifunctional Quinolonyl Diketo Acid Derivatives as HIV-1
Integrase Inhibitors: Design, Synthesis, Biological Activities, and Me-
chanism of Action. J. Med. Chem. 2006, 49, 1939–1945.
(21) Pasquini, S.; Mugnaini, C.; Tintori, C.; Botta, M.; Trejos, A.;
Arvela, R. K.; Larhed, M.; Witvrouw, M.; Michiels, M.; Christ, F.;
Debyser, Z.; Corelli, F. Investigations on the 4-Quinolone-3-carboxylic
Acid Motif. 1. Synthesis and Structure-Activity Relationship of a Class of
Human Immunodeficiency Virus type 1 Integrase Inhibitors. J. Med.
Chem. 2008, 51, 5125–5129.
(22) Hare, S.; Gupta, S. S.; Valkov, E.; Engelman, A.; Cherepanov, P.
Retroviral intasome assembly and inhibition of DNA strand transfer.
Nature 2010, 464, 232–236.
(23) Hazuda, D. J.; Anthony, N. J.; Gomez, R. P.; Jolly, S. M.; Wai,
J. S.; Zhuang, L.; Fisher, T. E.; Embrey, M.; Guare, J. P., Jr.; Egbertson,
M. S.; Vacca, J. P.; Huff, J. R.; Felock, P. J.; Witmer, M. V.; Stillmock,
K. A.; Danovich, R.; Grobler, J.; Miller, M. D.; Espeseth, A. S.; Jin, L.;
Chen, I.-W.; Lin, J. H.; Kassahun, K.; Ellis, J. D.; Wong, B. K.; Xu, W.;
Pearson, P. G.; Schleif, W. A.; Cortese, R.; Emini, E.; Summa, V.;
Holloway, M. K.; Young, S. D. A naphthyridine carboxamide provides
evidence for discordant resistance between mechanistically identical
inhibitors of HIV-1 integrase. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 11233–11238.
(36) Maurin, C.; Bailly, F.; Buisine, E.; Vezin, H.; Mbemba, G.;
Mouscadet, J. F.; Cotelle, P. Spectroscopic Studies of Diketoacids-Metal
Interactions. A Probing Tool for the Pharmacophoric Intermetallic
Distance in the HIV-1 Integrase Active Site. J. Med. Chem. 2004,
47, 5583–5596.
(37) Sechi, M.; Bacchi, A.; Carcelli, M.; Compari, C.; Duce, E.;
Fisicaro, E.; Rogolino, D.; Gates, P.; Derudas, M.; Al-Mawsawi, L. Q.;
Neamati, N. From Ligand to Complexes: Inhibition of Human Im-
munodeficiency Virus Type 1 Integrase by β-Diketo Acid Metal Com-
plexes. J. Med. Chem. 2006, 49, 4248–4260.
(38) Bacchi, A.; Biemmi, M.; Carcelli, M.; Carta, F.; Compari, C.;
Fisicaro, E.; Rogolino, D.; Sechi, M.; Sippel, M.; Sotriffer, C.; Sanchez,
T. W.; Neamati, N. From Ligand to Complexes. Part 2. Remarks on
Human Immunodeficiency Virus Type 1 Integrase inhibition by
β-Diketo Acid Metal Complexes. J. Med. Chem. 2008, 51, 7253–7264.
(39) Courcot, B.; Firley, D.; Fraisse, B.; Becker, P.; Gillet, J. M.;
Pattison, P.; Chernyshov, D.; Sghaier, M.; Zouhiri, F.; Desmaele, D.;
d’Angelo, J.; Bonhomme, F.; Geiger, S.; Ghermani, N. E. Crystal and
electronic structures of magnesium(II), copper(II), and mixed magne-
sium(II)-copper(II) complexes of the quinoline half of styrylquinoline-
type HIV-1 integrase inhibitors. J. Phys. Chem. B 2007, 111, 6042–6050.
(40) Drevensek, P.; Koꢁsmrlj, J.; Giester, G.; Skauge, T.; Sletten, E.;
Sepꢁciꢂc, K; Turel, I. X-Ray crystallographic, NMR and antimicrobial
activity studiesof magnesium complexes of fluoroquinolones - racemic
(24) Dayam, R.; Neamati, N. Active site binding modes of the beta-
diketoacids: a multi-active site approach in HIV-1 integrase inhibitor
design. Bioorg. Med. Chem. 2004, 12, 6371–6381.
(25) Pais, G. C. G.; Zhang, X.; Marchand, C.; Neamati, N.; Cow-
ansage, K.; Svarovskaia, E. S.; Pathak, V. K.; Tang, Y.; Nicklaus, M.;
518
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519

Molecular Pharmaceutics
ARTICLE
ofloxacin and its S-form, levofloxacin. J. Inorg. Biochem. 2006,
100, 1755–1763.
(41) Turel, I.; Zivec, P.; Pevec, A.; Tempelaar, S; Psomas, G.
Compounds of antibacterial agent Ciprofloxacin and Magnesium -
Crystal structures and molecular modeling calculations. Eur. J. Inorg.
Chem. 2008, 3718–3727.
(42) Chen, I. J.; Neamati, N.; MacKerell, A. D., Jr. Structure-based
inhibitor design targeting HIV-1 integrase. Curr. Drug Targets: Infect.
Disord. 2002, 2, 217–234.
(61) Fisicaro, E.; Braibanti, A. Potentiometric titrations in metha-
nol/water medium: Intertitration variabilitꢀa. Talanta 1988, 10, 769.
(62) Gran, G. Determination of the equivalence point in potentio-
metric titrations. Part II. Analyst 1952, 77, 661.
(63) Gans, P.; Sabatini, A.; Vacca, A. Investigation of equilibria in
solution. Determination of equilibrium constants with the HYPER-
QUAD suite of programs. Talanta 1996, 43, 1739.
ꢁ
(43) Maurin, C.; Bailly, F.; Cotelle, P. Structure-activity relationship
of HIV-1 integrase inhibitor-enzyme-ligand interaction. Curr. Med.
Chem. 2003, 10, 1795–1810.
(44) Sato, M.; Motomura, T.; Aramaki, H.; Matsuda, T.; Yamashita,
M.; Ito, Y.; Matsuzaki, Y.; Yamataka, K.; Ikeda, S.; Shinkai, H. Quinolone
carboxylic acids as a novel monoketo acid class of Human Immunode-
ficiency Virus Type1 Integrase inhibitors. J. Med. Chem. 2009,
52, 4869–4882.
(45) DeJesus, E.; Berger, D.; Markowitz, M.; Cohen, C.; Hawkins,
T.; Ruane, P.; Elion, R.; Farthing, C.; Zhong, L.; Chen, A. K.; McColl, D.;
Kearney, B. P. Antiviral Activity, Pharmacokinetics, and Dose Response
of the HIV-1 Integrase Inhibitor GS-9137 (JTK-303) in Treatment-
Naive and Treatment-Experienced Patients. J. Acquired Immune Defic.
Syndr. 2006, 43, 1–5.
(46) Johnson, A. A.; Marchand, C.; Pommier, Y. HIV-1 integrase
inhibitors: a decade of research and two drugs in clinical trial. Curr. Top.
Med. Chem. 2004, 4, 671–686.
(47) Maurin, C.; Bailly, F.; Cotelle, P. Improved preparation and
structural investigation of 4-aryl-4-oxo-2-hydroxy-2-butenoic acids and
methyl esters. Tetrahedron 2004, 60, 6479–6486.
(48) Sechi, M.; Carta, F.; Sannia, L.; Dallocchio, R.; Dess, A.;
Neamati, N. Design, synthesis, molecular modeling, and anti-HIV-1
integrase activity of a series of photoactivatable diketo acid-containing
inhibitors as affinity probes. Antiviral Res. 2009, 81, 267–276.
(49) Gatehouse, B. M.; O’Connor, M. J. The crystal and molecular
structure of racemic [bis(ethylenediamine)acetylpyruvato]cobalt(III)
iodide monohydrate. Acta Crystallogr. 1976, B32, 3145–3148.
(50) Kawai, H.; Kitano, Y.; Mutoh, M.; Hata, G. Synthesis, structure
and antitumor activity of a new water-soluble platinum complex, (1R,
2R-cyclohexanediamine-N,N0)[2-hydroxy-4-oxo-2-pentenoato(2-)-O2]-
platinum(II). Chem. Pharm. Bull. 1993, 41, 357–361.
(51) Barreca, M. L.; Iraci, N.; De Luca, L.; Chimirri, A. Induced-Fit
docking approach provides insight into the binding mode and mechanism of
action of HIV-1 Integrase inhibitors. ChemMedChem 2009, 4, 1446–1456.
(52) Liao, C.; Nicklaus, M. C. Tautomerism and Magnesium chela-
tion of HIV-1 Integrase inhibitors: a theoretical study. ChemMedChem
2010, 5, 1053–1066.
(53) Cremer, D.; Pople, J. A. General definition of ring puckering
coordinates. J. Am. Chem. Soc. 1975, 97, 1354.
(54) Al-Mawsawi, L. Q.; Sechi, M.; Neamati, N. Single amino acid
substitution in HIV-1 integrase catalytic core causes a dramatic shift in
inhibitor selectivity. FEBS Lett. 2007, 581, 1151–1156.
(55) Marchand, C.; Johnson, A. A.; Karki, R. G.; Pais, G. C. G.;
Zhang, X.; Cowansage, K.; Patel, T. A.; Nicklaus, M. C.; Burke, T. R.,
Jr.; Pommier, Y. Metal-Dependent Inhibition of HIV-1 Integrase by
β-Diketo Acids and Resistance of the Soluble Double-Mutant (F185K/
C280S). Mol. Pharmacol. 2003, 64, 600–609.
(56) SAINT: SAX, Area Detector Integration; Siemens Analytical
instruments Inc.: Madison, Wisconsin, USA.
(57) SADABS: Area-Detector Absorption Correction; Siemens Indus-
trial Automation, Inc.: Madison, Wisconsin, USA, 1996.
(58) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. SIR92—a program for automatic solution
of crystal structures by direct methods. J. Appl. Crystallogr. 1994, 27, 435.
(59) SHELXL97: Sheldrick, G. M. SHELXL97. Program for structure
refinement; University of G€ottingen: G€ottingen, Germany, 1997.
(60) Allen, F. H.; Kennard, O.; Taylor, R. Systematic analysis of
structural data as a research technique in organic chemistry. Acc. Chem.
Res. 1983, 16, 146–153.
519
dx.doi.org/10.1021/mp100343x |Mol. Pharmaceutics 2011, 8, 507–519