Metal-dependent inhibition of HIV-1 integrase

Article abstract of DOI:10.1021/jm0201417

Human immunodeficiency virus type 1 integrase (HIV-1 IN) is an essential enzyme for effective viral replication. Therefore, IN inhibitors are being sought for chemotherapy against AIDS. We had previously identified a series of salicylhydrazides as potent inhibitors of IN in vitro (Neamati, N.; et al. J. Med. Chem. 1998, 41, 3202-3209.). Herein, we report the design, synthesis, and antiviral activity of three novel mercaptosalicylhydrazide (MSH) derivatives. MSHs were effective against the IN catalytic core domain and inhibited IN binding to HIV LTR DNA. They also inhibited catalytic activities of IN in IN-DNA preassembled complexes. Site-directed mutagenesis and molecular modeling studies suggest that MSHs bind to cysteine 65 and chelate Mg²⁺ at the active site of HIV-1 IN. Contrary to salicylhydrazides, the MSHs are 300-fold less cytotoxic and exhibit antiviral activity. They are also active in Mg²⁺-based assays, while IN inhibition by salicylhydrazides is strictly Mn²⁺-dependent. Additionally, in target and cell-based assays, the MSHs have no detectable effect on other retroviral targets, including reverse transcriptase, protease, and virus attachment, and exhibit no detectable activity against human topoisomerases I and II at concentrations that effectively inhibit IN. These data suggest that MSHs are selective inhibitors of HIV-1 IN and may serve as leads for antiviral therapeutics.

Full text of DOI:10.1021/jm0201417

J . Med. Chem. 2002, 45, 5661-5670
5661
Meta l-Dep en d en t In h ibition of HIV-1 In tegr a se
Nouri Neamati,*,† Zhaiwei Lin,‡ Rajeshri G. Karki,‡ Ann Orr,§ Kiriana Cowansage,† Dirk Strumberg,§
Godwin C. G. Pais,‡ J ohannes H. Voigt,‡ Marc C. Nicklaus,‡ Heather E. Winslow,§ He Zhao,‡ J im A. Turpin,|
J izu Yi, Anna Marie Skalka, Terrence R. Burke, J r.,§ and Yves Pommier§
Department of Pharmaceutical Sciences, University of Southern California, School of Pharmacy, 1985 Zonal Avenue,
Los Angeles, California 90089, Laboratory of Medicinal Chemistry, Center for Cancer Research, NCIsFrederick,
Frederick, Maryland 21702, Laboratory of Molecular Pharmacology, Division of Basic Sciences, National Cancer Institute,
Bethesda, Maryland 20892, Southern Research Institute, 431 Aviation Way, Frederick, Maryland 21701, and Fox Chase
Cancer Center, Philadelphia, Pennsylvania
Received April 1, 2002
Human immunodeficiency virus type 1 integrase (HIV-1 IN) is an essential enzyme for effective
viral replication. Therefore, IN inhibitors are being sought for chemotherapy against AIDS.
We had previously identified a series of salicylhydrazides as potent inhibitors of IN in vitro
(Neamati, N.; et al. J . Med. Chem. 1998, 41, 3202-3209.). Herein, we report the design,
synthesis, and antiviral activity of three novel mercaptosalicylhydrazide (MSH) derivatives.
MSHs were effective against the IN catalytic core domain and inhibited IN binding to HIV
LTR DNA. They also inhibited catalytic activities of IN in IN-DNA preassembled complexes.
Site-directed mutagenesis and molecular modeling studies suggest that MSHs bind to cysteine
65 and chelate Mg2+ at the active site of HIV-1 IN. Contrary to salicylhydrazides, the MSHs
are 300-fold less cytotoxic and exhibit antiviral activity. They are also active in Mg2+-based
assays, while IN inhibition by salicylhydrazides is strictly Mn2+-dependent. Additionally, in
target and cell-based assays, the MSHs have no detectable effect on other retroviral targets,
including reverse transcriptase, protease, and virus attachment, and exhibit no detectable
activity against human topoisomerases I and II at concentrations that effectively inhibit IN.
These data suggest that MSHs are selective inhibitors of HIV-1 IN and may serve as leads for
antiviral therapeutics.
In tr od u ction
3′-processing and 3′-end joining (also referred to as
integration or strand transfer), respectively.
Successes of the multidrug cocktails using inhibitors
of protease and reverse transcriptase have altered the
natural course of AIDS by efficiently suppressing viral
replication for long periods of time. However, issues of
patient adherence, drug toxicity, the emergence of
multidrug-resistant phenotypes, and the presence of
persistent reservoirs of virus replication have high-
lighted the need to develop alternative therapeutic
approaches utilizing other targets essential in the viral
replication cycle. One such target is the integrase (IN)
enzyme. Integration of retroviral DNA into host chro-
mosomes is essential for viral replication. For HIV-1,
this process is mediated by IN, a 32 kDa virally encoded
protein, and the conserved terminal genomic sequences
in the HIV long terminal repeats (LTR) (for recent
reviews see ref 1). Following reverse transcription in the
cytoplasm of infected cells, IN cleaves two nucleotides
from each of the viral DNA ends containing the highly
conserved CA motif. After subsequent migration to the
nucleus as a part of a large nucleoprotein complex, IN
catalyzes the insertion of the resulting recessed 3′-
termini (CA-OH-3′) into a host chromosome by a direct
transesterification reaction. These two events are termed
Structural and biochemical understanding of IN has
led to the development of in vitro assays that have been
used to identify IN inhibitors.2-5 Such assays generally
utilize synthetic oligodeoxynucleotides corresponding to
the U5 end of the viral LTR, recombinant IN, and a
divalent metal cofactor (Mn2+ or Mg2+). Diverse classes
of drugs active against IN have been discovered using
these in vitro assays.6-10 Among all reported inhibitors,
the diketoacids show remarkable selectivity for strand
transfer and exhibit antiviral activity in cell culture.11
Presently, a compound designated as S-1360 discovered
by the scientists from Shionogi & Co. Ltd. is under
phase II clinical trials12 (for a recent review, see ref 9).
With the recent progress in the field of drug design and
discovery (see, for example, ref 13), new clinical candi-
dates should surface in an accelerated manner.
Previously, we reported that hydrazides are novel
noncatechol-containing inhibitors of IN.14-16 Structure-
activity relationship studies among these inhibitors
have revealed that the salicyl moiety is required for
activity.15,16 Thus, we hypothesized that these drugs
could act by chelating the divalent metal in the IN active
site.16 Herein, we report the design, synthesis, and
antiviral activity of three novel MSHs and the utiliza-
tion of several in vitro assays to define the molecular
interactions of these drugs with the IN-DNA com-
plexes. The present results indicate that MSHs inhibit
IN in Mg2+-based assays and are active against IN-
DNA preassembled complexes. We propose that MSH
* To whom correspondence should be addressed. Phone: 323-442-
2341. Fax: 323-442-1390. E-mail: neamati@usc.edu.
† University of Southern California, School of Pharmacy.
‡ NCIsFrederick.
§ National Cancer Institute.
| Southern Research Institute.
Fox Chase Cancer Center.
10.1021/jm0201417 CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/23/2002

5662 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26
Neamati et al.
F igu r e 1. Structure of salicylhydrazide 1 and the synthetic route to MSHs 2-4: (a) SOCl2/benzene, reflux; (b) thiosalicylhydrazide/
pyridine/toluene, room temp, overnight; (c) I2/DMF, room temp, overnight; (d) Et3P/THF/H2O, room temp, 1 h.
binds to cysteine 65 on the active site of IN and chelates
Mg2+. In addition, antiviral-targeted-based assays show
that the MSHs are selective against IN and exhibit
antiviral activity.
10 mM 2-mercaptoethanol, under these assay conditions
it is plausible that 4 could be reduced to 3 and
subsequently to 2, a possibility that we had hoped to
take advantage of in the design of these compounds.
Similarly, under reducing conditions found in vivo, the
disulfides could also be reduced to the corresponding
mercapto derivatives. In fact, we found that in the IN
reaction buffer, which contains mercaptoethanol, a
significant portion of 4 converted to 3 in approximately
half an hour. This was corroborated by TLC analysis,
where after 30 min, solutions of 4 in plain DMF or in
IN buffer solution clearly indicated the conversion
between 2 and 3 in the presence of mercaptoethanol
(data not shown). This is consistent with literature
reports that weak reducing agents such as triethylphos-
phine,17 hydrazine,18 hydriodic acid, and mercapto-
ethanol19 can serve to reduce the disulfide bonds to the
thiols.
Hyd r a zid es In h ibit IN Ca ta lytic Activities a n d
Exer t a Rem a r k a ble Diva len t Meta l Dep en d en cy.
The IN-catalyzed 3′-processing and DNA strand transfer
were measured in an in vitro assay employing purified
IN, a 21-mer duplex oligonucleotide corresponding to
the U5 end of the HIV LTR sequence (Figure 2A) and a
divalent metal ion (Mn2+ or Mg2+). Figure 2B shows a
representative gel illustrating inhibition of both 3′-
processing and strand transfer reactions by hydrazides.
Although in vitro assays are generally more efficient
in the presence of Mn2+ as a cofactor, it has been
proposed that the actual physiological divalent cation
is Mg2+ (see, for example, ref 20). We therefore, com-
pared the extent of 3′-processing and strand transfer
in the presence of either Mn2+ (Figure 2B) or Mg2+
(parts C and D of Figure 2, Table 1). Although 1 was
inactive in Mg2+, the mercapto-containing compounds
Resu lts
Syn t h esis of MSH s. In initial attempts to pre-
pare desired bis(thiosalicyl)hydrazide 2, salicylmono-
hydrazides bearing protected thiol groups were reacted
with thiosalicyl chloride. These reactions provided
intractable dark mixtures, underwent premature depro-
tection, or presented difficulties in removing thiol
protection. Alternatively, refluxing methyl thiosalicylate
in pure, excess hydrazine yielded only the corresponding
monohydrazide. In light of these problems, an alterna-
tive “internal protection” scheme was employed that
relied on initial formation of homodimeric disulfide
(DSA, Figure 1). In this approach, an internal disulfide
bond served as a temporary thiol-blocking group, which
could be removed by reacting the bisthiosalicyl-
hydrazide that had been formed. Therefore, starting
from the commercially available 2,2′-dithiosalicylic acid,
dimeric acid chloride (DSC) was prepared by treatment
with thionyl chloride. Upon reaction with thiosalicyl-
hydrazide (TSH), dimeric bisthiosalicylhydrazide 3
resulted. Triethylphosphine-mediated reduction of 3
gave the desired final product 2 (Figure 1). Compounds
2-4 provided analytical data consistent with their
assigned structures. (Compound 4 gave combustion
analysis results that deviated 0.62% from the theoretical
N value.) All inhibitors exhibited comparable IC50 values
(2-5 µM), indicating that mercapto groups (compounds
2) can substitute for hydroxyls (compound 1) and that
cyclization does not adversely influence potency. Since
IN assays are routinely carried out in the presence of

Metal-Dependent Inhibition of HIV-1 Integrase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26 5663
F igu r e 2. MSHs inhibit IN in the presence of Mn2+ and Mg2+. (A) A 21-mer blunt-end oligonucleotide corresponding to the U5
end of the HIV-1 LTR (5′ end-labeled with 32P) is reacted with purified IN. The initial step involves nucleolytic cleavage of two
bases from the 3′-end, resulting in a 19-mer oligonucleotide. Subsequently, 3′-ends are covalently joined at several sites to another
identical oligonucleotide that serves as the target DNA. This reaction is referred to as 3′-end joining or strand transfer, and the
products formed migrate more slowly than the original substrate (shown in the figure as STP for strand transfer products). Also
shown is the concentration-dependent inhibition of HIV-1 IN by hydrazides 1-4 using Mn2+ (B) or Mg2+ (C) as a cofactor. Drug
concentrations in micromolar units are indicated above each lane. (D) Quantitation of part C is shown.
inhibitory only in Mn2+-containing reactions (data not
Ta ble 1. Metal-Dependent Inhibition of HIV-1 Integrase by
Hydrazides 1-4
shown). This inhibitory effect of 1 in the presence of
a
IC50
,
µM
Mn2+ was independent of the order of the addition of
DNA (data not shown). This implies that inhibition of
preassemblyb
postassemblyc
strand
3′-processing transfer
IN by 1 is absolutely Mn2+-dependent. The Mn2+
-
strand
dependent activity of 1 was further established with
other assays as described below. Additionally, assays
were performed with different concentrations of the
various metals. IN can cleave its substrate DNA in the
presence of Mn2+ with concentrations as low as 3 mM
and as high as 50 mM and with Mg2+ in the range 7-25
mM (data not shown). When compounds 1 and 2 were
examined within these concentration ranges, similar
IC50 values were observed as for 1 at all the tested
concentrations of Mn2+ whereas 2 was slightly more
potent at higher concentrations of Mn2+ (data not
shown).
MSHs In h ibit P r eassem bled IN-DNA Com plexes.
IN has a high affinity for its DNA substrate, and under
physiological conditions it perhaps is always associated
with the viral LTRs. In vitro, IN also forms a tight
complex with viral DNA. To investigate whether MSHs
can effect assembled complexes, IN was allowed to bind
to DNA on ice first, before addition of drugs. Figure 3
demonstrates that the salicylhydrazides (1-4) inhibited
3′-processing and strand transfer in the presence of
Mn2+ within the same range (IC50 ) 5-35 µM). How-
ever, when similar reactions were performed in the
presence of Mg2+, compound 1 was inactive, while the
MSHs 2-4 inhibited both 3′-processing and strand
compound 3′-processing transfer
with Mn2+
1
2
3
4
2.0 ( 0.75
5.1 ( 1.2
3.2 ( 1.0
2.4 ( 0.9
0.7 ( 0.1
5.0 ( 1.0
3.7 ( 1.1
7.2 ( 2.3
12 ( 2
35 ( 5
8 ( 2
10 ( 2
12 ( 1
9 ( 3
5 ( 1
5 ( 1
with Mg2+
1
2
3
4
>1000
20; 30
15; 18
11; 8
>1000
11; 20
11; 15
11; 10
>1000
>1000
20
20
7
18
18
5
a
IC50 (50% inhibitory concentration) values with standard
deviations are mean values of three independent experiments. IC50
values from independent experiments are explicitly indicated.
b
Integrase was preincubated with metals and drugs for 30 min
followed by DNA for 1 h. c Integrase was preincubated with metals
and DNA on ice for 15 min followed by drugs for 1 h.
2-4 were active with either Mg2+ or Mn2+. Thus, by
contrast to other salicylhydrazides and more generally
to polyhydroxylated aromatics,6,21,22 the MSHs (2-4)
inhibit IN in reactions catalyzed with Mn2+ or Mg2+
.
To further examine the metal dependency of 1 in
purified IN assays, a variety of different conditions were
employed. For example, when IN was allowed to mix
without metal and in the presence of all possible
combinations of Mn2+, Mg2+, or Ca2+, compound 1 was

5664 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26
Neamati et al.
presence of Zn2+ and that this effect was more pro-
nounced when the full-length IN was used instead of
the core (data not shown).
Hyd r a zid es In h ibit IN-DNA Bin d in g. DNA-IN
cross-linking assay as described in Figure 4A was used
to assess inhibition of DNA binding.23 The salicyl-
hydrazides 1-4 inhibited the formation of IN-DNA
complexes in the presence of Mn2+ (Figure 5A). How-
ever, only the MSHs (2-4) were active in the presence
of Mg2+ (Figure 5B). Interestingly, all compounds were
less potent in this assay, which argues for the fact that
these compounds form ternary complexes with DNA and
IN and that they are not DNA binders. UV cross-linking
experiments provided similar results (data not shown).
Assays were performed with different combinations of
metals in order to investigate the Mn2+ requirement of
1. When the Schiff base assays were performed in the
presence of other divalent metals (Ca2+, Zn2+, Co2+, and
all possible combinations of these metals), 1 inhibited
DNA cross-linked product only in reactions where Mn2+
was present (data not shown). These results and the
dependency on the type of divalent metal are consistent
with the inhibition of DNA integration results and
suggest that salicylhydrazides 1-4 inhibit IN activity
by blocking IN binding to its substrate DNA.
MSH 2 Bin d s to Cystein e 65 a t th e Active Site
of IN a n d Ch ela tes Mg2+. We predicted that com-
pounds 2 and 3 could potentially form disulfide bonds
with Cys 65 of IN. To investigate the drug binding sites
of IN, we mutated Cys 65, which is close to the active
site of IN. Figure 6A shows that in the presence of Mn2+
,
MSH 2 inhibits the C65S mutant with similar potency
as the wild-type IN (see Figure 2B). However, in the
presence of Mg2+, MSH 2 lost its inhibitory potency
significantly (Figure 6B). Compound 1 inhibited this
mutant with an IC50 value identical to that of the wild-
type enzyme in the presence of Mn2+, and likewise, it
was inactive in the presence of Mg2+ (data not shown).
F igu r e 3. Inhibition of IN-DNA preassembled complexes.
Concentration-dependent inhibition of IN-assembled product
by hydrazides 1-4 in the presence of Mn2+ (A) or Mg2+ (B) is
shown. DNA and IN were assembled on ice for 15 min prior
to the addition of drugs. Drug concentrations in micromolar
units are indicated above each lane. (C) Quantitation of part
B is shown.
Because Cys 65 is close to the highly conserved DDE
residues and the MSHs are potential metal chelators,
we investigated the possible interactions of MSH with
this part of the enzyme by performing molecular model-
ing studies. Parts A and B of Figure 7 show the energy-
minimized structure of MSH 1 and 2, respectively. Parts
C and D of Figure 7 show the energy-minimized complex
between 1 and 2, respectively, with HCO2-Mg2+ as
calculated by ab initio studies. Though both 1 and 2
when optimized alone had lower energy values in the
trans conformation, 2 when optimized as a complex with
HCO2-Mg2+ was found to be more stable in the cis
conformation. Parts A and B of Figure 8 show the
docking model of 1 and 2 in the HIV-IN core domain.
From the docking analysis, it was observed that 1 forms
three hydrogen bonds with the enzyme. Both the hy-
droxyl groups on the aromatic ring form hydrogen
bonds, one with Asp 64 and the other with Asn 120,
while one hydrazide NH is hydrogen-bonded to Cys 65.
However, in the case of compound 2, no hydrogen bond
formation was observed with the enzyme. A possible
way that 2 forms a stable complex with the enzyme is
by formation of a disulfide bond between the sulfydryl
group and Cys 65. Figure 8C shows that 2, once allowed
to form a disulfide bond with C65, perfectly positions
transfer as efficiently as in Mn2+ (parts B and C of
Figure 3 and Table 1). Does IN assemble with its
substrate DNA on ice? To answer this question, we
examined the kinetics of IN-DNA complex formation
using a Schiff base assay as described before23 (Figure
4A). Parts B and C of Figure 4 illustrate that by 15 min
such complexes are formed and can be trapped using
either sodium borohydride and sodium cyanoboro-
hydride, respectively. To our knowledge, this is the first
report demonstrating that IN can assemble with its
DNA substrate on ice and that this assay can substitute
for Ca2+-based IN-DNA assembly.
IN is known to be capable of assembling with its DNA
substrate in the presence of Ca2+ without proceeding
to enzymatic cleavage of the DNA.22,24 Using this assay,
we found again that the MSHs (2-4) were active in the
presence of both Mg2+ and Mn2+ while the salicyl-
hydrazide 1 was selectively active in the presence of
Mn2+ (data not shown). Moreover, when compound 1
was assayed for the inhibition of disintegration in the
presence of a combination of Mn2+ and Zn2+, it was
observed that inhibition by 1 was stimulated by the

Metal-Dependent Inhibition of HIV-1 Integrase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26 5665
F igu r e 4. Kinetics of IN-DNA complex formation on ice. (A) The assay uses IN-DNA cross-linking by formation of a Schiff
base between IN and a duplex oligonucleotide containing an abasic site. IN and DNA were allowed to mix on ice at various times
in the presence of Mn2+ or Mg2+. The reactions were trapped using either sodium borohydride (B) or sodium cyanoborohydride
(C).
F igu r e 6. Metal-dependent inhibition of C65S mutant. In-
hibition of 3′-processing and strand transfer by compounds 1
and 2 in the presence of Mn2+ (A) and Mg2+ (B).
molecular dynamics calculations, which are after all
F igu r e 5. Metal-specific inhibition of IN binding to LTR DNA
based on classical molecular mechanics (MM) force field
calculations, to capture any true orbital interactions, let
alone bond formation (or bond breaking). It is therefore
not surprising that no closer approach of the two sulfur
atoms in question was observed; at closer distances, the
only effects that the classical force field will reproduce
are steric and/or electrostatic repulsion. It would require
a quantum mechanical (QM) approach, such as QM/
MM, to tackle the disulfide bond formation. These
computations are beyond the scope of this paper.
Selectivity of In h ibition by Sa licylh yd r a zid es.
To examine selectivity of hydrazides 1-4, these com-
pounds were first assayed for inhibition of other retro-
by salicylhydrazides. Shown is the effect of salicylhydrazides
1-4 on DNA binding of IN in the presence of Mn2+ (A) or Mg2+
(B) PhosphorImager picture shows the inhibition of the 39 kDa
product corresponding to the IN-DNA covalent complex in the
presence of the indicated concentrations of drug.
itself on the active site to chelate the Mg2+ ion and to
form a ternary complex with IN.
The closest distance that was observed during the MD
runs between the Cys 65 sulfur and the sulfur atom of
MSH 2 was about 4.86 Å, which has been shown to be
a typical distance at which disulfide bond formation can
be expected to become possible.25,26 It is impossible for

5666 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26
Neamati et al.
cylhydrazides induced a cleavage complex (data not
shown) or inhibited the ability of topoisomerase I to
generate camptothecin-mediated cleavage complexes at
concentrations that effectively inhibited IN (data not
shown). Interestingly, when these compounds were
examined in an assay specific for topoisomerase II, an
enzyme known to require divalent metal for activity, a
similar metal-dependency profile in activity was ob-
served as compared to those results against IN (Figure
9). For example, compound 1 inhibited topoisomerase
II in a strictly Mn2+-dependent fashion, while the MSHs
did not suppress the topoisomerase II mediated cleavage
complex within the similar range that they inhibited
IN. Suppression of topoisomerase II activity was similar
in reactions with the presence or absence of etoposide
(data not shown). These results not only show selectivity
against IN but also metal-dependent inhibition of two
different enzymes. Taken as a whole, these data suggest
that the tested compounds are more selective for IN and
that their inhibitory potency is not due to nonspecific
binding to the protein.
F igu r e 7. B3LYP/6-31G* geometry optimized structures of
(A) compound 1, (B) compound 2, (C) 1-Mg2+‚2HCO2 com-
-
plex, and (D) 2-MSH‚Mg2+‚2HCO2 complex.
-
viral targets. At concentrations that inhibit IN, none of
these agents exhibited detectable activity on HIV-1 RT,
protease, and viral attachment (Table 2). We next
examined whether salicylhydrazides affected host cell
topoisomerases. An assay specific for eukaryotic topo-
isomerase I27 demonstrated that none of the four sali-
MSHs Exh ibit An tivir a l Activity. In the standard
XTT cytoprotection assay performed in the NCI anti-
viral screening program, we observed that the MSHs
F igu r e 8. Hypothetical binding of MSHs in the active site of IN with chelation of Mg2+: (A, B) docking model of 1 and 2 in the
active site of an IN core domain, obtained using GLIDE; (C) simulated structure of 2 bound to Cys 65 of the HIV IN core domain.
Green dashed lines are for hydrogen bonds, and white dashed lines are for interactions with the Mg2+ ion.
Ta ble 2. Testing of Hydrazides 1-4 against Viral Replication and Viral Proteins
IC50 (µM)
RT
antiviral activity
compound
EC50 (µM)
CC50 (µM)
TIb
protease
attachment
RAdT
rCdG
1
2
3
4
a
0.1
c
>90
>90
>90
c
> 100
> 100
> 100
c
>100
>100
>100
c
>100
>100
>100
4.3 ( 1.3
17.8 ( 1.9
6.0 ( 4.0
14.4 ( 0.6
34.0 ( 1.5
24.9 ( 1.0
3.3
1.9
4.1
a
b
Not reached because of cytotoxicity. TI, therapeutic index ) (CC50/EC50). c Not tested.

Metal-Dependent Inhibition of HIV-1 Integrase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26 5667
tant for ligand binding.16 The 2-hydroxyphenyl, the
R-keto, and the hydrazine moiety when in a planar
conformation could interact with IN by chelation of the
metal in the IN active site, since inhibition of IN
catalytic activity and DNA binding are strictly Mn2+
-
dependent. Hydrophobic binding sites potentially par-
ticipate in complementarity of molecular shape between
ligand and receptor. The results from this latter study
indicate that only those compounds, which possess these
features in a linear orientation, are potent inhibitors of
IN.16 Unfortunately, despite their remarkable anti-IN
potency, the hydrazides examined were highly cytotoxic
(IC50 values in the nanomolar range) and none showed
antiviral activity. Therefore, the design and synthesis
of nontoxic hydrazide-based inhibitors have been elu-
sive.
Design of MSHs. Studies from our laboratory have
identified compounds containing free mercapto or sulf-
oxide groups such as diaryl sulfones28 and 2-mercapto-
benzenesulfonamides,29 which inhibit IN and exhibit
antiviral activity. The X-ray structure of one compound,
naphthalene disulfonate complexed with ASV IN, was
subsequently solved30 and shown to be located close to
the enzymatic active site. This binding altered the
conformation of a critical flexible loop and affected the
conformation of active site residues.30 This naphthalene
disulfonate inhibits IN and shows moderate antiviral
activity against HIV-1 infected CEM cells. Cumula-
tively, these data suggest that the mercapto-containing
groups are potentially novel inhibitors, which differ
significantly from polyhydroxylated aromatics with
respect to antiviral and cytotoxicity profiles. The present
study further demonstrates that MSHs are 300-fold less
cytotoxic than parent compound 1. In accord with our
previous study showing that hydrazides can diffuse into
the active site and form ternary complexes with the
enzyme, which are highly Mn2+-dependent, we sought
to take potential advantage of the differential coordina-
tion of Mn2+ and Mg2+ as exhibited in sulfur and oxygen
functionalites. In contrast to Mn2+, hard metal ions such
as Mg2+ do not coordinate to sulfur effectively, leading
to our prediction that differential sulfur and oxygen
affinities for Mn2+ versus Mg2+ would be expected to
play major roles in the biological profiles of 1 and its
mercapto derivative 2. For this reason, analogues of 1
were prepared by replacing hydroxyl groups with free
mercapto groups. However, since rapid oxidation of
thiols to the disulfides was a possible side reaction,
analogues 3 and 4 were designed so that both mercapto
and disulfide forms would produce the desired free
mercaptohydrazide 2 upon reduction in the cell or in
vitro in the presence of 2-mercaptoethanol, as commonly
required for enzymatic assays.
F igu r e 9. Suppression of etoposide-stabilized topoisomerase
II cleavage complexes by hydrazides 1-4. A 254-base-pair
DNA fragment from the first c-myc intron was prepared
between positions 3035 and 3288 by PCR and the lower DNA-
strand labeled with 32P at the 5′ terminus. The topoisomerase
II cleavage complex shows both subunits (topoisomerase II acts
as a homodimer) covalently bound to each 5′-end of the cleaved
DNA, generating a transient double-strand break with a four-
base-pair stagger. Reactions were performed in the presence
Mg2+ (B) or Mn2+ (C) containing 100 µM etoposide (VP16).
Drug concentrations in micromolar units are indicated above
each lane. Lane 1: control, no topoisomerase II, no drug treat-
ment. Lane 2: purine ladder obtained after formic acid re-
action. Lane 3: topoisomerase II only. Lane 4: topoisomerase
II plus VP16.
2-4 protected HIV-1 infected cells with 50% effective
concentration (EC50) values ranging from 4 to 18 µM
(Table 2). They were also toxic at low micromolar (50%
cytotoxic concentration (CC50) values ranging from 14
to 34 µM. In contrast, compound 1 exhibited a CC50
value of 0.1 µM without protection of the HIV-1 infected
cells. The 300-fold reduction in cytotoxicity of MSHs
relative to 1 may explain their antiviral activity. Studies
are underway to examine the antiviral activities of these
compounds against different viral strains.
On the basis of the relative affinity of hydroxyl versus
mercapto groups toward soft and hard metals, it was
envisioned that replacement of the salicylhydroxyl with
a mercapto group could result in a different activity
profile against IN. According to Pearson’s HSAB theory,
hard metal ions such as Mg2+ are Lewis acids that
interact with hard bases such as phosphates and car-
boxylate oxygens and water in preference to softer Lewis
acids such as Mn2+, which interact more strongly with
sulfur. For example, it has been shown that Mg2+
coordinates with a ratio of 31 000 oxygen to sulfur in
Discu ssion
Hydrazides are novel non-catechol-containing in-
hibitors of IN.14-16 Structure-activity relationships
among these inhibitors have revealed that the salicyl
moiety is required for activity15,16 and, furthermore, that
superimposition of the lowest energy conformations
yields three sites whose properties appear to be impor-

5668 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26
Neamati et al.
complex with ATPâS, while Mn2+ coordinates with a
ratio of 158 oxygen to sulfur.31 In a more recent study
with bacterial transposase, it was demonstrated that
enzyme selectivity and efficiency were changed from
Mg2+ to Mn2+ by replacing Asp114 in the active site with
a cysteine.32 Therefore, the unique predilection of di-
valent metal in substrate binding could be exploited as
a mechanistic tool for elucidating the role of metal
binding in the inhibition of metal-dependent enzymes
such as IN.
Selectivity Issu es. The low inhibitory potency of
MSHs against various other viral and nonviral proteins
tested attests to their selectivity for IN. Therefore, this
class of compounds is different from previously de-
scribed hydroxylated aromatics, which frequently do not
exhibit this selectivity. Although more potent MSH
analogues will aid in future in vivo studies, the study
presented here provides initial evidence for the underly-
ing hypothesis, which we have put forth previously.16
That is, selectively chelating the metal ion at the active
site of IN or in other ways selectively interfering with
the highly conserved acidic residues offers potentially
important approaches for the inhibition of IN. For MSH
2, this selectivity issue might be due to the formation
of a disulfide bond with C65 at the active site and
subsequent juxtaposition of the diketo functionality to
chelate Mg2+. Obviously, there are many enzymes with
active cysteine residues that could potentially react with
compound 2. The fact that MSHs exhibit antiviral
activity and do not inhibit other viral and nonviral
enzymes may indicate that MSHs are selective inhibi-
tors of IN.
(methanol/diethyl ether) afforded pure product as an off-white
solid (2.50 g, 78% yield). Mp: 255 °C (dec). H NMR (DMSO-
d6): δ 10.82 (d, J ) 22.9 Hz, 2H, 2NH), 10.59 (d, J ) 21.7 Hz,
2H, 2NH), 7.90-7.24 (m, 16H, 4C6H4-), 5.40 (br s, 2H, 2SH).
FABMS m/z: 607 (M + H). Anal. (C28H22N4O4S4) C, H, N.
1
N,N′-Bis(2,2′-d ith iosa licyl)h yd r a zid e 4. To a solution of
N,N′-bis(2-mercaptobenzoyl)-2,2′-dithiosalicylhydrazide 3 (0.5
g, 0.82 mmol) in DMF (3 mL) with methanol (3 mL) was added
iodine (0.21 g, 0.82 mmol), and the mixture was stirred at room
temperature (15 h). To the mixture was added H2O (10 mL),
whereupon the crude product came out of solution as a brown
solid. Recrystallization (DMF/H2O) afforded pure 4 as a white
1
powder (0.46 g, 90%). Mp: 148 °C (dec). H NMR (DMSO-d6):
δ 11.75 (br s, 4H, 2NHNH), 8.07-7.48 (m, 16H, 4C6H4).
FABMS m/z: 605 (M + H). Anal. (C28H20N4O4S4) C, H, N.
N,N′-Bis(2-m er ca p toben zoyl)h yd r a zid e 2. To a suspen-
sion of N,N′-bis(2-mercaptobenzoyl)-2,2′-dithiosalicylhydrazide
3 (0.50 g, 0.82 mmol) in tetrahydrofuran (40 mL) containing
H2O (4 mL) was added slowly triethylphosphine (0.19 g, 1.6
mmol), and the mixture was stirred under argon (0.5 h).
Solvents were removed in vacuo, yielding a brown solid, which
was purified by silica gel flash chromatography (gradient of
chloroform and acetic acid) to afford pure product as a beige
solid (0.40 g, 80%). Mp: 170 °C (dec). 1H NMR (CDCl3): δ 10.55
(s, 2H, NHNH), 7.71 (d, J ) 7.6 Hz, 2H, C6H4-), 7.53 (d, J )
7.8 Hz, 2H, C6H4-), 7.42 (t, J ) 7.8 Hz, 2H, C6H4-), 7.28 (t, J
) 7.6 Hz, 2H, C6H4-), 5.40 (br s, 2H, 2SH). FABMS m/z: 303
(M - H). Anal. (C14H12N2O2S2‚0.2H2O) C, H, N.
Ma ter ia ls, Ch em ica ls, a n d En zym es. All compounds
were dissolved in DMSO, and the stock solutions were stored
at -20 °C. Etoposide (VP16) was obtained from Bristol-Myers
Co., Wallingford, CT. Human c-myc inserted into pBR322,
restriction enzymes, T4 polynucleotide kinase, polyacrylamide/
bisacrylamide, and Taq DNA polymerase were purchased from
Lofstrand Labs (Gaithersburg, MD), Life Technologies, Inc.
(Gaithersburg, MD), New England Biolabs (Beverly, MA), or
Qiagen Inc. (Valencia, CA). [g-32P]-ATP was purchased from
DuPont NEN (Boston, MA). PCR oligonucleotide primers were
obtained from GIBCO BRL (Gaithersburg, MD). All positive
control compounds for individual assays except AZTTP were
obtained from the NCI chemical repository. The reference
reagents for the individual assays are as follows: attachment,
Farmatalia (NSC 65016)33 and dextran sulfate (NSC 620255);
reverse transcriptase inhibition, rAdT template/primer-AZTEC
(Sierra BioResearch, Tuscon, AZ), rCdG template/primer-
UC3834 (NSC 629243); protease inhibition, KNI-27235 (NSC
651714). The expression systems for the wild-type IN, soluble
mutant INF185KC280S, and the truncated mutant IN50-212 (F185K)
were generous gifts of Drs. T. J enkins and R. Craigie, Labora-
tory of Molecular Biology, NIDDK, NIH, Bethesda, MD. The
cysteine mutant enzymes were prepared as described.36,37
Exp er im en ta l Section
Melting points were determined on either a Gallenkamp or
a Mel Temp II melting point apparatus and are uncorrected.
Elemental analyses were obtained from Atlantic Microlab Inc.,
Norcross, GA, and are within 0.4% of the theoretical except
where indicated. Fast atom bombardment mass spectra
(FABMS) were acquired with a VG Analytical 7070E mass
spectrometer under the control of a VG data system. 1H NMR
data were obtained on a Bruker AC250 spectrometer (250
MHz) and are reported in ppm relative to TMS and referenced
to the solvent in which they were run. Anhydrous solvents
were obtained commercially and used without further drying.
Flash column chromatography was performed using E. Merck
silica gel 60 (particle size, 230-400 mesh). Thiosalicyl-
hydrazide (TSH) was prepared from methylthiosalicylate
(MTS), and 2,2′-dithiosalicylic acid (DSA) was purchase from
Sigma-Aldrich Co.
P r ep a r a tion of Oligon u cleotid e Su bstr a tes, In tegr a se
Assa y, a n d DNA Bin d in g Assa y Usin g Sch iff Ba se F or -
m a tion . These protocols were carried out essentially as
described.23,38
P r ep a r a tion of En d -La beled DNA F r a gm en ts by P CR.
The 254-base-pair DNA fragment from the first c-myc intron
was prepared between positions 3035 and 3288, with numbers
referring to GenBank genomic positions using the following
oligonucleotides: 5′-GTAATCCAGAACTGGATCGG-3′ for the
upper strand and 5′-ATGCGGTCCCTACTCCAAGG-3′ for the
lower strand (annealing temperature of 56 °C). Single-end
labeling was obtained by 5′-end labeling of the specific primer
oligonucleotide. A 10 pM aliquot of DNA was incubated for 60
min at 37 °C with 10 U of T4 polynucleotide kinase and 10
pM of [γ-32P]-ATP (100 µCi) in kinase buffer (70 mM Tris-HCL,
pH 7.6, 0.1 M KCL, 10 mM MgCl2, 5 mM dithiothreitol, and
0.5 mg/mL bovine serum albumin). Reactions were stopped
by heat denaturation at 70 °C for 15 min. After purification
using Sephadex-G25 columns (Boehringer Mannheim), the
labeled oligonucleotides were used for PCR. Approximately 0.1
µg of the c-myc DNA that had been restricted by XhoI and
XbaI was used as template for the PCR. A 10 pM aliquot of
each oligonucleotide primer, one of them being 5′-labeled, was
P r ep a r a tion of 2,2′-Dith iosa licyl Ch lor id e (DSC). To a
suspension of 2,2′-dithiosalicylic acid (DSA) (5.0 g, 16.3 mmol)
in benzene (100 mL) was added thionyl chloride (18.5 g, 155.6
mmol). The mixture was refluxed (3 h) and then solvent was
removed in vacuo to provide crude DSC as a brown solid in
sufficient purity for further use. Mp: 142-147 °C. 1H NMR
(CDCl3): δ 8.47 (dd, J 1 ) 8.1 Hz, J 2 ) 1.5 Hz, 2H, C6H4-),
7.84 (dd, J 1 ) 8.1 Hz, J 2 ) 1.2 Hz, 2H, C6H4-), 7.84 (td, J 1
8.1 Hz, J 2 ) 1.5 Hz, 2H, C6H4-), 7.47 (td, J 1 ) 8.1 Hz, J 2
1.2 Hz, 2H, C6H4-).
N,N′-Bis(2-m er ca p toben zoyl)-2,2′-d ith iosa licylh yd r a -
zid e 3. To a solution of thiosalicylhydrazide (1.77 g, 10.6 mmol)
in pyridine (20 mL) under argon was added a solution of 2,2′-
dithiosalicyl chloride (DSC) (2.0 g, 5.3 mmol) in toluene (40
mL) and DMF (20 mL). Then the mixture was stirred at room
temperature (36 h) and solvent was removed in vacuo to afford
a brown residue. The residue was treated with H2O (40 mL)
to yield crude 3 as a white precipitate. Recrystallization
)
)

Metal-Dependent Inhibition of HIV-1 Integrase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26 5669
used in 19 temperature cycle reactions (each cycle with 94 °C
for 1 min, annealing for 1 min, and 72 °C for 2 min). The last
extension was for 10 min. DNA was purified using PCR Select-
II columns (5Prime-3Prime, Inc. Boulder, CO).
below). Sixty QMD cycles were performed with different seeds
for the random number generator for each cycle. Each QMD
cycle consisted of 60 ps MD at 800 K in an NVT ensemble,
keeping all protein atoms fixed except the side chain of Cys
65 and MSH 2. The vdW interactions were scaled to 2%, and
the Coulombic interactions were scaled to 20%. The coordi-
nates were saved every 200 fs and subsequently minimized
by 300 steps of the Polak-Ribiere conjugated gradient algo-
rithm (CG-PR). During the minimization, all atoms except the
Mg2+ and the side chains within a radius of 8 Å around the
initial position of the Mg2+ ion were held fixed. For each of
the 60 runs, the lowest energy conformation out of the 300
minimized structures was stored. The frame with lowest
energy among the 60 stored structures is depicted in Figures
8C.
Top oisom er a se II In d u ced DNA Clea va ge Rea ction s.
DNA fragments ((5-10) × 104 dpm/reaction) were equilibrated
with or without drug in 1% DMSO, 10 mM Tris-HCl, pH 7.5,
50 mM KCl, 5 mM MgCl2, 2 mM dithiothreitol, 0.1 mM
Na2EDTA, 1 mM ATP, and 15 µg/mL bovine serum albumine
for 5 min before addition of 8 units (80 ng) of purified human
topII (kindly provided by Dr. J . L. Nitiss, Department of
Molecular Pharmacology, St. J ude Children’s Research Hos-
pital, Memphis, TN) in 10 µL of final reaction volume.
Reactions were performed at 37 °C for 30 min and thereafter
stopped by adding 1% sodium dodecyl sulfate (SDS) and 0.4
mg/mL proteinase K (final concentrations) followed by an
additional incubation at 55 °C for 30 min.
HIV-1 Cell a n d Ta r get-Ba sed Assa ys. The cell-based p24
attachment assay has been described in detail elsewhere.39
Assays for activity against HIV-1 reverse transcriptase rAdT
(template/primer) and rCdG (template/primer) using recom-
binant HIV-1 reverse transcriptase (a kind gift from S. Hughes
ABL Basic Research NCI-FCRDC, Frederick, MD) have been
previously described.40 The substrate cleavage of recombinant
HIV-1 protease in the presence of test compounds was quanti-
fied using an HPLC-based methodology with the artificial
substrate Ala-Ser-Glu-Asn-Try-Pro-Ile-Val-amide (Multiple
Peptide Systems, San Diego, CA) previously described.41,42 The
anti-HIV drug testing performed at NCI is based on a protocol
described by Weislow et al.43
Molecu la r Mod elin g. Three-dimensional structures of the
compounds 1 and 2 were constructed in Macromodel 7.0. Each
structure was energy-minimized using the Merck molecular
force field (MMFF).44 The geometry of these local energy
minima was used as the initial structure for conformational
analysis and subjected to 1000 steps of Monte Carlo (MC)
simulation. Two minimum conformations for each molecule,
one in the cis and one in the trans orientation, reached by the
MC simulations were then fully optimized in the gas phase at
the density functional theory (DFT) level of theory, using the
B3LYP/6-31G* basis set, using the ab initio program Titan
(running on a Windows 98 PC).45
Geometry optimization of compounds 1 and 2 complexed
with Mg2+‚2HCO2- was also carried out at the B3LYP/6-31G*
level in the gas phase, in Gaussian 98, revision A11.46 For this,
different possibilities in which the molecules can form the
complexes were considered and each of them were optimized
individually.
Docking studies of compounds 1 and 2 were performed on
the core domain of HIV-1 integrase using the program
GLIDE.47 The domain A of the crystal structure of the HIV
IN (PDB code 1QS4)48 provided the geometry of the core
domain. The crystallographic water molecules and the bound
ligand were removed. The residues missing in the 1QS4
structure (residues 141-144) were built using the biopolymer
module in Sybyl 6.8,49 and the conformation was adjusted by
the loop search method such that it was consistent with that
observed in the crystal structure 1BL3.pdb.50 For each com-
pound 100 poses were saved and the pose with the best GLIDE
score is depicted in parts A and B of Figure 8. The pose with
the best GLIDE score was refined further by 60 quenched
molecular dynamics (QMD) simulation as explained below. For
each QMD cycle, all atoms beyond a radius of 12 Å around
the initial position of the Mg2+ ion were held fixed.
To study the possibility of covalent bond formation between
compound 2 and Cys 65, a bond between one sulfur of MSH 2
and the sulfur of Cys 65 was added. This was subjected to a
QMD simulation using the Insight II 97.0/Discover 3.0 model-
ing package51 (running on SGI computers) utilizing the cff91
force field.52 Additional parameters for Mg2+ were determined
using its ionic radius and small-molecule crystallographic data.
A dielectric constant of 1.00 and the cell multipole method with
fine accuracy were used for all nonbonding interactions except
for the quenched molecular dynamics (MD) simulations (see
Ack n ow led gm en t. We acknowledge Elizabeth Gla-
ser Pediatric AIDS Foundation (student intern award
program) support to Kiriana Cowansage.
Refer en ces
(1) Asante-Appiah, E.; Skalka, A. M. HIV-1 integrase: structural
organization, conformational changes, and catalysis. Adv. Virus
Res. 1999, 52, 351-369.
(2) Hazuda, D. J .; Hastings, J . C.; Wolfe, A. L.; Emini, E. A. A novel
assay for the DNA strand-transfer reaction of HIV-1 integrase.
Nucleic Acids Res. 1994, 22, 1121-1122.
(3) Mazumder, A.; Neamati, N.; Sunder, S.; Owen, J .; Pommier, Y.
Retroviral Integrase: A Novel Target in Antiviral Development;
Basic In Vitro Assays with the Purified Enzyme. In Methods in
Cellular and Molecular Biology: Antiviral Evaluation; The
Humana Press: Totowa, NJ , 1999.
(4) Chow, S. A. In vitro assays for activities of retroviral integrase.
Methods 1997, 12, 306-317.
(5) Debyser, Z.; Cherepanov, P.; Pluymers, W.; De Clercq, E. Assays
for the evaluation of HIV-1 integrase inhibitors. Methods Mol.
Biol. 2001, 160, 139-155.
(6) Neamati, N.; Sunder, S.; Pommier, Y. Design and discovery of
HIV-1 integrase inhibitors. Drug Discovery Today 1997, 2, 487-
498.
(7) Neamati, N.; Marchand, C.; Pommier, Y. HIV-1 integrase
inhibitors: past, present, and future. Adv. Pharmacol. 2000, 49,
147-165.
(8) Neamati, N. Structure-based HIV-1 integrase inhibitor design:
a future perspective. Expert Opin. Invest. Drugs 2001, 10, 281-
296.
(9) Neamati, N. Patented small molecule inhibitors of HIV-1 inte-
grase: a ten-year saga. Expert Opin. Ther. Pat. 2002, 12, 709-
724.
(10) Pommier, Y.; Neamati, N. Inhibitors of human immunodeficiency
virus integrase. Adv. Virus Res. 1999, 52, 427-458.
(11) Hazuda, D. J .; Felock, P.; Witmer, M.; Wolfe, A.; Stillmock, K.;
et al. Inhibitors of strand transfer that prevent integration
and inhibit HIV-1 replication in cells. Science 2000, 287, 646-
650.
(12) Yoshinaga, T.; Sato, A.; Fujishita, T.; Fujiwara, T. In Vitro
Activity of a New HIV-1 Integrase Inhibitor in Clinical Develop-
ment. Presented at the 9th Conference on Retroviruses and
Opportunistic Infections, Seattle, WA, 2002.
(13) Neamati, N.; Barchi, J . J ., J r. New paradigms in drug design
and discovery. Curr. Top. Med. Chem. 2002, 2, 211-227.
(14) Hong, H.; Neamati, N.; Wang, S.; Nicklaus, M. C.; Mazumder,
A.; et al. Discovery of HIV-1 integrase inhibitors by pharma-
cophore searching. J . Med. Chem. 1997, 40, 930-936.
(15) Zhao, H.; Neamati, N.; Sunder, S.; Hong, H.; Wang, S.; et al.
Hydrazide-containing inhibitors of HIV-1 integrase. J . Med.
Chem. 1997, 40, 937-941.
(16) Neamati, N.; Hong, H.; Owen, J . M.; Sunder, S.; Winslow, H.
E.; et al. Salicylhydrazine-containing inhibitors of HIV-1 inte-
grase: implication for a selective chelation in the integrase active
site. J . Med. Chem. 1998, 41, 3202-3209.
(17) Urpi, F.; Vilarrasa, J . New synthetic tricks. Advantages of using
triethylphosphine in some phosphorous-based reactions. Tetra-
hedron Lett. 1986, 27, 4623-4624.
(18) Maiti, S. N.; Spevak, P.; Singh, M. p.; Micetich, R. G.; Reddy, V.
N. Reductive cleavage of symmetrical disulfides with hydrazine.
Synth. Commun. 1988, 18, 575-582.
(19) Houk, J .; Whitesides, G. M. Structure-activity relationships for
thiol-disulfide interchanges. J . Am. Chem. Soc. 1987, 109, 6825-
6836.
(20) Cowan, J . A. The biological Chemistry of Magnesium; VCH: New
York, 1995; pp 1-23.

5670 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 26
Neamati et al.
(21) Fesen, M. R.; Pommier, Y.; Leteurtre, F.; Hiroguchi, S.; Yung,
J .; et al. Inhibition of HIV-1 integrase by flavones, caffeic acid
phenethyl ester (CAPE) and related compounds. Biochem.
Pharmacol. 1994, 48, 595-608.
(38) Neamati, N.; Hong, H.; Mazumder, A.; Wang, S.; Sunder, S.; et
al. Depsides and depsidones as inhibitors of HIV-1 integrase:
discovery of novel inhibitors through 3D database searching. J .
Med. Chem. 1997, 40, 942-951.
(22) Hazuda, D. J .; Felock, P. J .; Hastings, J . C.; Pramanik, B.; Wolfe,
A. L. Differential divalent cation requirements uncouple the
assembly and catalytic reactions of human immunodeficiency
virus type 1 integrase. J . Virol. 1997, 71, 7005-7011.
(23) Mazumder, A.; Neamati, N.; Pilon, A. A.; Sunder, S.; Pommier,
Y. Chemical trapping of ternary complexes of human immuno-
deficiency virus type 1 integrase, divalent metal, and DNA
substrates containing an abasic site. Implications for the role of
lysine 136 in DNA binding. J . Biol. Chem. 1996, 271, 27330-
27338.
(24) Ellison, V.; Brown, P. O. A stable complex between integrase
and viral DNA ends mediates human immunodeficiency virus
integration in vitro. Proc. Natl. Acad. Sci. U.S.A. 1994, 91,
7316-7320.
(25) Careaga, C. L.; Falke, J . J . Thermal motions of surface alpha-
helices in the D-galactose chemosensory receptor. Detection by
disulfide trapping. J . Mol. Biol. 1992, 226, 1219-1235.
(26) Creighton, T. E. Disulfide bond formation in proteins. Methods
Enzymol. 1984, 107, 305-329.
(27) Pommier, Y.; Kohlhagen, G.; Kohn, K. W.; Leteurte, F.; Wani,
M. C.; et al. Interaction of an alkylating camptothecin derivative
with a DNA base at topoisomerase 1-DNA cleavage sites. Proc.
Natl. Acad. Sci. U.S.A. 1995, 92, 8861-8865.
(28) Neamati, N.; Mazumder, A.; Zhao, H.; Sunder, S.; Burke, T. R.,
J r.; et al. Diarylsulfones, a novel class of human immuno-
deficiency virus type 1 integrase inhibitors. Antimicrob. Agents
Chemother. 1997, 41, 385-393.
(29) Neamati, N.; Mazumder, A.; Sunder, S.; Owen, J . M.; Schultz,
R. J .; et al. 2-Mercaptobenzenesulfonamides as novel class of
human immunodeficiency type 1 (HIV-1) integrase and HIV-1
replication. Antimicrob. Agents Chemother. 1997, 8, 485-495.
(30) Lubkowski, J .; Yang, F.; Alexandratos, J .; Wlodawer, A.; Zhao,
H.; et al. Structure of the catalytic domain of avian sarcoma virus
integrase with a bound HIV-1 integrase-targeted inhibitor. Proc.
Natl. Acad. Sci. U.S.A 1998, 95, 4831-4836.
(31) Pecoraro, V. L.; Hermes, J . D.; Cleland, W. W. Stability constants
of Mg2+ and Cd2+ complexes of adenine nucleotides and thio-
nucleotides and rate constants for formation and dissociation of
MgATP and MgADP. Biochemistry 1984, 23, 5262-5271.
(32) Sarnovsky, R. J .; May, E. W.; Craig, N. L. The Tn7 transposase
is a heteromeric complex in which DNA breakage and joining
activities are distributed between different gene products. EMBO
J . 1996, 15, 6348-6361.
(33) Clanton, D. J .; Buckheit, R. W.; Terpening, S. J .; Kiser, R.;
Mongelli, N.; et al. Novel sulfonated and phosphonated analogs
of distamycin which inhibit the replication of HIV. Antiviral Res.
1995, 27, 335-354.
(39) Rice, W. G.; Supko, J . G.; Malspeis, L.; Buckheit, R. W. J .;
Clanton, D.; et al. Inhibitors of HIV nucleocapsid protein zinc
fingers as candidates for the treatment of AIDS. Science 1995,
270, 1194-1197.
(40) Turpin, J . A.; Terpening, S. J .; Schaeffer, C. A.; Yu, G.; Glover,
C. J .; et al. Inhibitors of human immunodeficiency virus type 1
zinc fingers prevent normal processing of gag precursors and
result in the release of noninfectious virus particles. J . Virol.
1996, 70, 6180-6189.
(41) Rice, W. G.; Bader, J . P. Discovery and in vitro development of
AIDS antiviral drugs as biopharmaceuticals. Adv. Pharm. 1995,
33, 389-438.
(42) Rice, W. G.; Baker, D. C.; Schaeffer, C. A.; Graham, L.; Bu, M.;
et al. Inhibition of multiple phases of human immunodeficiency
virus type 1 replication by a dithiane compound that attacks
the conserved zinc fingers of retroviral nucleocapsid proteins.
Antimicrob. Agents Chemother. 1997, 41, 419-426.
(43) Weislow, O. W.; Kiser, R.; Fine, D.; Bader, J .; Shoemaker, R.
H.; et al. New soluble-formazan assay for HIV-1 cytopathic
effects: application to high-flux screening of synthetic and
natural products for AIDS antiviral activity. J . Natl. Cancer Inst.
1989, 81, 577-586.
(44) Macromodel, version 7.0; Schrodinger Inc.: Portland, OR, 1999.
(45) TITAN, version 1; Schrodinger Inc.: Portland, OR, 1999.
(46) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery,
J . A., J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam,
J . M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;
Tomasi, J .; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J .; Petersson, G.
A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Salvador, P.; Dannen-
berg, J . J .; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Baboul, A. G.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; J ohnson, B.; Chen, W.; Wong, M. W.; Andres, J . L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J . A.
Gaussian, revision A.11; Gaussian, Inc.: Pittsburgh, PA, 2001.
(47) Schrodinger, I. Glide/ First Discovery 1.8; Schrodinger, Inc.:
Portland, OR, 2001.
(48) Goldgur, Y.; Craigie, R.; Cohen, G. H.; Fujiwara, T.; Yoshinaga,
T.; et al. Structure of the HIV-1 integrase catalytic domain
complexed with an inhibitor:
a platform for antiviral drug
(34) Bader, J . P.; McMahon, J . B.; Schultz, R. J .; Narayanan, V. L.;
Pierce, J . B.; et al. Oxathiin carboxanilide, a potent inhibitor of
human immunodeficiency virus reproduction. Proc. Natl. Acad.
Sci. U.S.A. 1991, 88, 6740-6744.
(35) Kageyama, S.; Mimoto, T.; Murakawa, Y.; Nomizu, M.; Ford,
H., J r.; et al. In vitro anti-human immunodeficiency virus (HIV)
activities of transition state mimetic HIV protease inhibitors
containing allophenylnorstatine. Antimicrob. Agents Chemother.
1993, 37, 810-817.
(36) Asante-Appiah, E.; Merkel, G.; Skalka, A. M. Purification of
untagged retroviral integrases by immobilized metal ion affinity
chromatography. Protein Expression Purif. 1998, 12, 105-110.
(37) Yi, J .; Asante-Appiah, E.; Skalka, A. M. Divalent cations
stimulate preferential recognition of a viral DNA end by HIV-1
integrase. Biochemistry 1999, 38, 8458-8468.
design. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 13040-13043.
(49) SYBYL, version 6.8; Tripos Inc.: St. Louis, MO, 2001.
(50) Maignan, S.; Guilloteau, J . P.; Zhou-Liu, Q.; Clement-Mella, C.;
Mikol, V. Crystal structures of the catalytic domain of HIV-1
integrase free and complexed with its metal cofactor: high level
of similarity of the active site with other viral integrases. J . Mol.
Biol. 1998, 282, 359-368.
(51) InsightII; Accelrys: San Diego, CA. 1997.
(52) Maple, J . R.; Dinur, U.; Hagler, A. T. Derivation of force fields
for molecular mechanics and dynamics from ab initio energy
surfaces. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5350-5354.
J M0201417