Inhibition of the HIV-1 Rev-RRE complex formation by unfused aromatic cations

Article abstract of DOI:10.1016/S0968-0896(00)00344-8

RNA viruses cause a wide range of human diseases. Development of new agents to target such viruses is an active area of research. Towards this goal, a series of diphenylfuran cations as potential inhibitors of the Rev-RRE complex have been designed and synthesized. Analysis of the interaction of the diphenylfurans with RRE and TAR RNA model systems by gel shift assays indicates that they exhibit both sequence and structure-dependent binding modes. Our results show a strong interaction between the diphenylfuran ring system and RRE bases, while the TAR interactions are much weaker with the compounds that are the best inhibitors of Rev-RRE.

Full text of DOI:10.1016/S0968-0896(00)00344-8

Bioorganic & Medicinal Chemistry 9 (2001) 1097±1113
Inhibition of the HIV-1 Rev±RRE Complex Formation
by Unfused AromaticCations
Ge Xiao, Arvind Kumar, Ke Li, C. Ted Rigl, Miroslav Bajic, Tina M. Davis,
David W. Boykin*and W. David Wilson*
Department of Chemistry, Georgia State University, Atlanta, GA 30303, USA
Received 4 October 2000; accepted 21 November 2000
AbstractÐRNA viruses cause a wide range of human diseases. Development of new agents to target such viruses is an active area of
research. Towards this goal, a series of diphenylfuran cations as potential inhibitors of the Rev±RRE complex have been designed
and synthesized. Analysis of the interaction of the diphenylfurans with RRE and TAR RNA model systems by gel shift assays
indicates that they exhibit both sequence and structure-dependent binding modes. Our results show a strong interaction between the
diphenylfuran ring system and RRE bases, while the TAR interactions are much weaker with the compounds that are the best
inhibitors of Rev±RRE. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
experiments with other ribosome targeting antibiotics,
such as thiostrepton, also indicate that RNA binds
small molecules with high speci®city and can be a highly
speci®c therapeutic target.24,25 These observations suggest
that RNA is an underutilized therapeutic target and
e€orts are underway in several labs to design drugs that
are selective for RNA±protein complexes.
Biological RNAs fold into locally base paired secondary
regions and macromolecular tertiary structures that are
essential for the biological functions of the RNA.1À4
The DNA double helix is recognized primarily through
its linear array of base pairs whereas proteins are
recognized through their shape and three dimensional
arrangement of functional groups. The single strands of
most biological RNAs can assume folded structures that
have features similar to both double-helical DNA as
well as to the folded conformations of proteins. The
highly-folded structures obtained from high-resolution
structural investigations on RNA suggest that selected
sites on such folded RNAs should be amenable to small-
molecule drug-design strategies.5À14 This approach
would be similar to the highly successful application of
such strategies to design compounds to target the active
sites of proteins.
The relative inhibitory e€ects of aminoglycosides
against the formation of RRE±Rev complexes was
determined by Zapp and co-workers23 with a competi-
tive gel shift assay that used an RRE IIB hairpin model
system and a recombinant Rev protein. The relative
activity of aminoglycosides to inhibit formation of the
Rev±RRE complex was found to be: neomycin B (0.1
mM)>tobramycin (1.0 mM)> gentamicin (10 mM)>
kanamycin A (ꢀ100 mM)>streptomycin (ꢀ100 mM),
and these results suggest a structure-dependent recogni-
tion of the RRE RNA by neomycin and closely related
aminoglycosides.23 Although aminoglycosides are not
ideal antivirals for reasons such as transport and toxicity,
they do provide exciting new leads for development.
Recent high-resolution NMR studies with aminoglycoside±
RNA complexes have de®ned the recognition principles for
the antimicrobial activity of aminoglycosides.2,15,16 Other
studies have shown that aminoglycosides can inhibit RNA
splicing,17 ribozyme activity,18À20 and can selectively
inhibit both the Rev and Tat proteins from binding to
their cognate RRE and TAR RNA binding sites in the
genomic RNA of HIV-1.21À23 Results from a range of
E€orts in our laboratory to design compounds that
speci®cally recognize RNA and exert desired biological
e€ects initially focused on identifying molecular structures
that can recognize di€erent RNA conformations.14,26À28
Through this search an unfused aromatic diphenylfuran
tetracation (DB182) that bound strongly to the RRE
RNA of HIV-1 and inhibited complex formation with
the Rev protein was found in a gel band shift assay simi-
lar to that described above for the aminoglycosides.26,27
*Corresponding authors. Tel.: +1-404-651-3798; fax: +1-404-651-1416;
e-mail: chedwb@panther.gsu.ed (D.W. Boykin). Tel.: +1-404-651-3903;
fax: +1-404-651-2751; e-mail: chewdw@panther.gsu.edu (W.D. Wilson)
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00344-8

1098
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
More detailed studies of the DB182±RRE complex sug-
gested that the compound interacts strongly with the bases
of RRE, requires the G±G and G±A base pair mis-
matches and internal loop region of RRE (Fig. 1(A)),
and causes a conformational change in RRE on com-
plex formation.26
Results
Gel band shift analysis of the RRE duplex±Rev model system
To evaluate the ability of the compounds to inhibit the
Rev±RRE complex, we have used the RRE duplex,
de®ned by Gait and co-workers29 and the Rev-17 peptide
of Frankel and co-workers30 in our assay for compound
activity. The RRE hairpin de®ned by Peterson and co-
workers31 and Rev-17 peptide model system were also
used here for some selected compounds for validation
comparison. Complex formation between the Rev-17
peptide and RRE duplex model and hairpin (Fig. 1(A))
was evaluated with a gel band shift assay (Fig. 1(B)) as
previously described.32,33 Under the conditions used in
these experiments, the RNA duplex and the hairpin
shift from the free to the complex species over a sharp
range of Rev peptide concentrations (Fig. 1). The RRE
sample in the gel shift study migrates as an intact duplex
or hairpin and forms a speci®c complex with Rev peptide.
The interaction is quite speci®c and addition of the Tat
peptide to the RRE RNA showed no complex forma-
tion over the same concentration range (not shown).
The exciting ®ndings on unfused aromatic cation±RNA
interactions prompted us to initiate a range of synthetic
and biophysical studies to obtain additional informa-
tion on the interaction of unfused aromatic tetracations
with RRE. A key goal of our e€orts to design RNA
selective drugs is to increase both the anity and selec-
tivity of the compounds for the RRE structure. In order
to evaluate the importance of each structural component
of DB182 to the inhibition of the Rev±RRE complex, a
number of new derivatives in which the furan and
phenyl groups and cationic substituents of the compound
have been systematically varied have been prepared
(Figs 2(b), 4(a) and (b), and 5(a)±(c)). Inhibition of Rev±
RRE complex formation by the compounds has been
evaluated through gel band shift experiments. The new
diphenylfuran derivatives can be divided into four
groups with modi®cations in the (1) cationic sub-
stituents, (2) phenyl rings, (3) central furan heterocycle,
and (4) position of the cationic side chain (meta versus
para) relative to the parent compound. One of the new
compounds has six positive charges, several have two
positive charges, and the majority have four positive
charges. The objectives of the studies reported here are
to determine how the components and structure of the
diphenylfuran derivatives a€ect the activity and speci®-
city for inhibition of the Rev±RRE complex, and in the
long term, to use this information to design new com-
pounds with enhanced RRE RNA interactions and
inhibition selectivity for the Rev±RRE complex.
Aminoglycosides. In Figure 2(a); a competitive gel band
shift assay illustrates the potential of aminoglycosides to
inhibit the RRE duplex±Rev peptide complex under the
experimental conditions of this work. IC50 values, the
concentration of the compound required to give 50%
inhibition of Rev binding to RRE, are given for each
compound. As Zapp et al.23 showed with an RRE±Rev
model system, neomycin B demonstrates a strong inhi-
bitory e€ect against the formation of the RRE duplex±
Rev complex and was followed by tobramycin and
gentamicin. Kanamycin A and streptomycin have no
signi®cant inhibitory e€ect under these conditions (Fig.
2(a)). With the duplex RNA model system and higher
Figure 1. (A) RRE IIB duplex, hairpin, and TAR duplex models used in thermal denaturation experiments and gel shift assays. Sequences of the
Rev-17 peptide from HIV-1 Rev protein and Tat-24 peptide from HIV-1 Tat protein that bind RRE and TAR RNAs, respectively, used in gel shift
assays. (B) Gel shift assays of (a) RRE duplex titrated with Rev-17 peptide; (b) RRE hairpin titrated with Rev-17 peptide; (c) TAR duplex titrated
with Tat-24 peptide.

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1099
temperature used for the gel band shift analysis in the
present experiments, a higher concentration of neomy-
cin is required to inhibit the Rev±RRE complex than in
previous studies. The higher temperature is preferred for
these experiments since we have observed signi®cantly
less smearing of RNA bands that occurs in the presence
of increasing concentrations of some of the cationic
compounds at low temperatures. Tm increases caused by
addition of the aminoglycosides to the RRE duplex are
also shown in Figure 2 for evaluation of the relative
binding anity of these compounds to the RRE RNA.
The e€ect of the ratio of neomycin to RRE duplex on
the observed Tm is illustrated in Figure 3. Both the Tm
increases and the ability to inhibit the Rev±RRE com-
plex increase for the aminoglycosides as their charge
increases (Fig. 2(a)).
Unfused aromaticcations. In our search for novel inhi-
bitors of the RRE±Rev complex, it was previously
shown that the diphenylfuran tetracation DB182
strongly inhibited binding of both the Rev protein and
peptide model system to RRE.26 Results for DB182 are
shown in Figure 2(b) for reference under the gel conditions
of this paper along with results for DB60, which was
Figure 2. Competitive gel shift assays of the inhibition of the RRE±Rev interaction and IC50 values, the compound concentration required to give
50% bound RNA, and the increase in RRE Tm are shown. In each gel experiment, the ®rst lane is RRE alone and subsequent lanes are RRE plus
Rev-17 peptide and increasing concentration of drug (indicated in mM). The notations b and f indicate bound and free RRE, respectively. (a) RRE
titrated with aminoglycosides; (b) RRE titrated with the diphenylfuran analogues that previously demonstrated inhibition of the Rev±RRE inter-
action.26 (continued on next page)

1100
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
Figure 2. (continued)
previously found to be a signi®cant but weaker inhibitor
than DB182, and for DB75, DB103, and DB198, which
were found to be very poor inhibitors. As can be seen,
the agreement with previous work is excellent for all
compounds, and DB182 is a signi®cantly better inhi-
bitor of the Rev±RRE interaction than neomycin.
The Tm increases for RRE caused by addition of the
diphenylfuran derivatives are also shown in Fig. 2b for
comparison with results for the aminoglycosides and
with results for the new compounds reported here. Plots
of the Tm increase versus ratio for several of the furan
derivatives are compared to results for neomycin in
Figure 3. The largest Tm increase occurs between the
ratios of zero and one compound per duplex and there
is little increase in Tm above a ratio of 2.
Figure 3. Melting temperature of the RRE±drug complex at varying
ratios of drug to RRE.

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1101
To dissect the contributions that individual molecular
units of DB182 make to the inhibition of Rev±RRE
complex formation, four di€erent series of new deriva-
tives were designed, synthesized and are reported below.
The compounds of Figure 4(a) have modi®ed cationic
substituents while those of Figure 4(b) have modi®ed
central furan heterocycles. The compounds of Figure
5(a) have modi®cations at one or both phenyl rings
while the compounds of Figure 5(b) have the side chains
moved from the para to the meta position.
Modi®cation of the cationic substituents. As shown in
Figure 4(a), compound DB339, in which the two amidine
groups were changed to amide groups (a dication),
showed only very slight inhibition at 30 mM, a sig-
ni®cant decrease in activity compared to results for
DB182 (Fig. 2(b)). DB240 and DB403, dications that
retain the alkylamidine substituents with the removal of
the terminal amino groups, also show no inhibition up
to 30 mM. Compounds DB378 and DB379 with one or
three more methylene groups in the cationic alkyl linker
Figure 4. Competitive gel shift assays for inhibition of the RRE duplex±Rev interaction and the increase in RRE melting temperature by dipheny-
furan analogues. Conditions and labels are the same as in Fig. 2. (a) Diphenylfuran analogues with modi®ed cationic substituents; (b) diphenylfuran
analogues with modi®ed central furan heterocycles. (continued on next page)

1102
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
Figure 4. (continued)
than DB182, showed activity similar to DB182. When
the terminal dimethylamino group of DB182 was
replaced by a diethylamino group, DB228, the inhibi-
tory activity decreased slightly. When the diethyl groups
were fused into a ®ve member ring, DB380, however,
there was no inhibition of the Rev±RRE complex up to
30 mM. Similarly, DB394 with a terminal piperazyl ring
is a very poor inhibitor. Clearly, inhibition of formation
of the protein±RNA complex is acutely sensitive to the
nature of the alkylamine substituent on the diphenyl-
furan aromatic system.
Compounds of Figure 4(a) that are very poor inhibitors
have low ÁTm values for their RRE complexes (less
than 1 ꢁC). The better inhibitors have ÁTm values of 3±
5 ꢁC, but ÁTm values in general are only qualitative
predictors of IC50 results for Rev±RRE inhibition.
Modi®cation or removal of the central furan heterocycle.
Compound DB400, in which the 3-position of furan is
substituted with a methyl group has similar activity to
the parent compound DB182, while a compound with a
3-tolyloxy substituent, DB414, is inactive (Fig. 4(b)). In

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1103
Figure 5. Competitive gel shift assays for inhibition of the RRE±Rev interaction and the increase in RRE melting temperature by diphenyfuran
analogues. Conditions and labels are the same as in Fig. 2. (a) Inhibition of the RRE duplex±Rev interaction by diphenylfuran analogues with
modi®ed phenyl rings; (b) inhibition of the RRE duplex±Rev interaction by diphenylfuran analogues with the position of the cationic substituent
modi®ed; (c) inhibition of the RRE hairpin±Rev interaction by selected diphenylfuran analogues. (continued on next page)

1104
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
Figure 5. (continued)

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1105
agreement with previous results,27 the corresponding
pyrrole compound, DB247, showed signi®cantly
decreased activity with an IC50 greater than 30 mM.
Relative to the furan DB182, the corresponding thio-
phene, DB316, and oxazole, DB449, compounds have
improved activity and the pyrimidine, DB436, has similar
activity. The oxazole compound DB449 was the most
active compound in the series of derivatives in Figure 4,
and it is clear that the central heterocyclic ring has a
strong in¯uence on the inhibitory activity. Several carbo-
cyclic tetracations that are closely related to DB182 are
completely inactive in the assay (not shown).
not greatly a€ect inhibition while a derivative with a
longer side chain and six total charges, DB420, resulted
in an inactive compound. Although pyrroles DB382 and
DB383 are inactive, the pyrrole analogue of DB182 has
higher activity when the cationic substituents are meta,
DB381, instead of para, DB247 (Fig. 2(b)), as observed
with the corresponding meta and para furans. The conclu-
sion from these results is that for compounds with a
central ®ve-membered ring heterocycle substituted at
the 2- and 5-positions with phenyl groups, derivatives
with meta cationic substituents are much better Rev±RRE
inhibitors than the corresponding para substituents. As
with the compounds described above, there is a general
correlation of ÁTm and inhibition ability for the com-
pounds of Figure 5.
As observed with the compounds of Figure 4(a), there is
a general correlation between ÁTm and inhibition, but
the correlation is not quantitative. Tm increases for the
RRE RNA complex with these compounds are generally
larger for the better inhibitors than for the poor inhibitors,
suggesting that binding strength for the RRE RNA is a
primary factor in e€ectiveness of the compounds as
Rev±RRE inhibitors. Although all active compounds
have signi®cant ÁTm values, some inactive compounds
have ÁTm values close to the values for the active com-
pounds. These results can be explained by the di€erence
between speci®c and nonspeci®c binding. The Tm
increase is determined by the increased stability of the
entire RNA and is sensitive to both speci®c and non-
speci®c binding while inhibition is much more dependent
on speci®c interactions that prevent binding of the Rev
peptide. Thus compounds such as DB379 and DB394
have very similar ÁTms but quite di€erent IC50 values.
Gel band shift analysis of the RRE hairpin±Rev model
system
In order to test the reliability of the RRE duplex model
system, an RRE hairpin model system was used in the
competitive gel shift assays. The conditions used are
described in Experimental. Ten representative compounds
were initially assayed (Fig. 5(c)) for their ability to inhibit
Rev binding. From these results we can see that several of
these aromatic heterocyclic compounds have signi®cant
inhibitory potential. In particular, the tetracationic com-
pound DB340 was a potent inhibitor of Rev binding
with an IC50 of 0.5 mM in the hairpin model that is
within experimental error of the 0.3 mM value observed
in the duplex model system. For the other compounds, the
measured IC50 in the hairpin model system were generally
somewhat larger than the corresponding results in the
duplex model system. The order of the relative activity
of these compounds in both of the model systems is the
same, however: DB340>>DB355>DB182>DB60>
DB247>DB515. Four additional dicationic derivatives
of DB340 (DB270, DB271, DB293, and DB302) were
also assayed on this system (Fig. 5(c)). Among them,
DB302 has the best inhibitory activity with IC50 of 22
mM. The others were not active up to 30 mM.
Modi®cation of the phenyl rings. The compounds shown
in Figure 5(a) have one or more of the phenyl rings of
the DB182 diphenylfuran system replaced by a benzimi-
dazole ring. The conversion of DB182 to DB340 has a
dramatic e€ect and produced the most potent compound
of the series with strong inhibition of Rev±RRE complex
formation at an IC50 of less than 1 mM. The benzimida-
zole DB340 binds strongly to RRE and causes a large
increase in the Tm of RRE on complex formation. Like
the amidine dication analogue of DB182 (DB75 in Fig.
2(b)), the amidine dication analogue of DB340, DB293, is
inactive. The imidazoline dication analogue of DB340,
DB302, is less active than the imidazoline dication analo-
gue of DB182 (DB60 in Fig. 2(a)). DB302 reproducibly
gave some inhibition at 3 mM but less at 30 mM as in
Figure 5(a). This may be due to a cooperative self-asso-
ciation of DB302 at the higher concentrations that lim-
its its speci®c binding to RRE.
Inhibition speci®city
In order to investigate the speci®city of the inhibition of
the Rev±RRE complex by the unfused aromatic tetra-
cations, e€ects of representative compounds on a Tat
peptide±TAR RNA duplex (Fig. 1(a)) were investigated
(Fig. 6). The Tat-24 peptide used in this work is the
same one characterized by Long and Crothers34 while the
TAR duplex (Fig. 1(a)) was shown to form a high anity
site for the Tat protein by Gait and co-workers.29 The
titration of labeled TAR RNA by the peptide is shown
in Figure 1(b), and, as with the RRE titration by Rev,
complex formation occurs over a narrow concentration
range. Compound DB182 inhibits the Tat±TAR complex
with IC50 of 3.6 mM. Pyrrole DB247, a poor inhibitor in
the Rev±RRE system, has improved activity against
Tat±TAR. Compounds DB340 and DB355, the most
active in the Rev±RRE assay, show signi®cantly reduced
activity against the Tat±TAR complex with IC50 values
greater than 15 mM. The conclusion from these results is
that the most active compounds of this series in the
The position of the cationic substituent on the unfused
aromatic ring system also has a strong e€ect on Rev±
RRE inhibition. Compound DB355 (Fig. 5(b)), in which
the para cationic substituents of DB182 are moved to
the meta position, has a signi®cantly improved activity
(IC50=1.6 mM) relative to DB182 (IC50 of 5 mM).
Interestingly, however, when the imidazole side chains
of the dication DB60 (Fig. 2(b)) are moved to the meta
position, DB361, a dramatic decrease in inhibition is
observed, and there is no activity up to 30 mM. As with
the para derivatives, addition of another methylene
group to the side chain in the meta series, DB434, did

1106
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
Figure 6. Competitive gel shift assays for inhibition of TAR±Tat interaction and the increase in TAR melting temperature by neomycin B and
representative diphenylfuran analogues. The ®rst lane is TAR alone and subsequent lanes are TAR plus Tat-24 and increasing concentration of drug
(indicated in mM). The notations b and f indicate bound and free TAR, respectively.
Rev±RRE assay are also the compounds which have the
most signi®cant speci®city for the Rev±RRE system. No
correlation of ÁTm and the ability to inhibit the Tat±
TAR complex is apparent with these compounds.
Signi®cant inhibition of complex formation is obtained
with neomycin while other related aminoglycosides have
reduced inhibition ability.
Our previous results for the compounds in Fig. 2(b)
with the Rev protein and RRE hairpin complex indi-
cated that DB182 is a strong inhibitor of complex for-
mation, the dication DB60 is a moderate inhibitor while
DB75, DB103 and DB198 are all poor inhibitors.26,27
The results in Figure 2(b) with the duplex±peptide
model are in good agreement with the previous results.
For design of the compounds reported in this paper, we
have treated DB182 as a lead compound that is com-
posed of three components that should be modi®ed in a
structure±inhibition assay: a central heterocycle, phenyl
rings that are symmetrically substituted on the hetero-
cycle, and cationic substituents at the para positions of
the phenyl rings. In the work reported here, we have
individually varied each component as well as the posi-
tion of the cationic substituent to determine how each
molecular unit contributes to the observed Rev±RRE
inhibition. The modi®cations have resulted in new
compounds with signi®cantly improved inhibitory
activity.
Discussion
The RRE duplex RNA system used in the assays
reported here provides a convenient synthetic model
system that is readily amenable to modi®cation of the
base sequence or backbone chemistry on each
strand.29,33 The duplex forms a speci®c complex with
the Rev-17 peptide (Fig. 1(a) and (b)) while addition of
the Tat peptide does not yield a complex under the same
conditions (not shown). Similar results are obtained
with inhibition of Rev binding with the RRE hairpin.
The RRE±Rev complex and free RRE are resolved in
nondenaturing gel electrophoresis experiments (Fig. 1(B)),
and this provides a method for analysis of inhibition of
complex formation by added agents. Control experiments
with aminoglycosides and the duplex±peptide model
system (Fig. 2(a)) are in agreement with previous studies
with the Rev protein and RRE hairpin model system.23,26,27

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1107
Conversion of the two dicationic substituents into
monocationic groups of similar structure yielded com-
pletely inactive compounds (Fig. 4(a)). Maintaining the
charge while lengthening the alkyl chain between the
amidine and amine or modifying the terminal amine
either did not improve the activity or yielded completely
inactive compounds. The IC50 values of Figure 4(a)
show that inhibition of complex formation by the furan
derivatives is quite sensitive to change in both the ami-
dine and amine cations. It appears that two dicationic
groups are required for speci®c binding and are critical
for e€ective inhibition of the Rev±RRE complex. A small
substitutent on the furan ring does not signi®cantly
change inhibition while a large substituent abolishes
activity (Fig. 4(b)). The RNA interactions necessary for
Rev inhibition are not available when the furan has a
bulky substituent at the 3-position. It should be noted
that the bulky O-toluyl substituent is twisted out of the
diphenylfuran aromatic plane and could cause a sig-
ni®cant perturbation in RNA interactions of the system.
The best inhibitor of all compounds in this series is the
benzimidazole DB340 (Fig. 5(a)) which inhibits the com-
plex at less than 1 mM concentration even under the rela-
tively stringent conditions of the assay used in this study
(e.g., relative to a 7 mM IC50 for neomycin).
The dicationic analogues of DB340 show poor activity
and the imidazoline derivative is less active than the
dication diphenylfuran compounds DB60 and DB75
(Fig. 2(a)). The imidazoline compound, DB302, is the
only one in the series to show less activity at 30 mM than
at 3 mM concentration and some secondary interactions
may limit its speci®c binding to RRE duplex.
The most active compounds with the greatest speci®city
in this series are DB340 and DB355. The benzimidazole
tetracation is also the most speci®c compound for the
Rev±RRE system in this series with a ratio of Rev±RRE
IC50 to that for the Tat±TAR complex greater than 50.
If the reasonable assumption is made that DB340 and
DB355 recognize RRE through a similar conformation, it
becomes useful to establish conformational similarities
between these two compounds as well as how they di€er
from derivatives with lower activity. One obvious dif-
ference between the highly active DB340 and DB355
relative to the parent para compound DB182 is that
with DB182, rotation about the phenyl±furan bond
does not yield a new conformation while in the meta
compound it does. With the benzimidazole derivative
DB340, rotating about the phenyl±furan bond does not
yield a new conformation but rotation about the benzi-
midazole±furan bond does generate a new conforma-
tional state. It thus seems possible that asymmetric
comformations of the active compounds account for the
higher activity, perhaps conformations with the two
cationic groups pointed in approximately opposite
directions with respect to the aromatic ring system.
Experiments are in progress to test this hypothesis.
The oxazole, thiophene and the six-membered pyri-
midine heterocyclic compounds all have activity that is
similar or slightly better than the parent furan (Fig. 4(b)).
With a central ®ve-atom ring, the oxazole and the thio-
phene derivatives are good inhibitors while the pyrrole
is a very poor inhibitor (Fig. 4(b)). It appears that the
RNA target site can accommodate either a central ®ve
or six-membered ring system, but the ring must have an
H-bond accepting heteroatom at the 1-position. The
pyrrole can bind to RRE and raise the global Tm value of
the RNA, but although it has the correct shape, it appar-
ently does not provide the correct type of interaction for
binding strongly at the speci®c site required for Rev inhi-
bition. IC50 values for DB340 and DB182 obtained
through a solid-phase assay35 agree with the IC50 values
reported here obtained through gel shift assays. The IC50
value for the pyrrole DB247 obtained through the solid-
phase assay is much lower than the value reported here.
The reason for this discrepancy is currently unknown. It
appears that IC50 values for strong binders such as
DB340 and DB182 are in much better agreement than
IC50 values for weak binders such as DB247.
The results presented here demonstrate that formation
of the drug±RRE complex leading to inhibition of Rev
binding is quite sensitive to the structure of the com-
pound. A general model for inhibition of Rev±RRE
complex formation has not yet been de®ned. However,
footprinting and preliminary NMR results on the
DB340±RRE complex suggest that DB340 binds near
the internal loop region of RRE and competes with
binding of the Rev peptide. Detailed binding experi-
ments for other strong binders such as DB60, DB288,
and DB355 should yield additional insight into the
mechanism of Rev inhibition and aid in the identifying
speci®c molecular motifs with high anity for RNA.
When the cationic substituents are moved from the para
to the meta positions, the furan derivative IC50 values
decrease considerably (compare isomeric derivatives in
Figs 4(b) and 5(b)). The pyrrole meta derivative also has
some activity in contrast to the inactive para compound.
Clearly the heterocycle requirement is relaxed in moving
from the para to the meta substituted compounds. The
meta imidazoline dication, however, completely loses
activity (compare DB361 in Fig. 5(b) to DB60 in Fig.
2(b)) in the para to meta switch. All meta tetracations
evaluated are more active that the corresponding para
compounds, and equally as important, the meta furan
derivative DB355 appears to be more speci®c for the Rev±
RRE complex that the para derivative. The para compound
actually has a slightly lower IC50 for the Tat±TAR complex
than for the Rev±RRE system (compare DB182 in Figs
2(b) and 6) while the meta compound has a considerably
lower IC50 for the Rev±RRE complex (compare DB355 in
Figs 5(b) and 6).
Experimental
Compound synthesis. The compounds used in these
experiments are shown in Figures 2, 4, and 5. All amino-
glycosides were purchased from Sigma Chemical Com-
pany. The synthesis of compounds DB75, DB228, DB60,
DB103, DB182, DB198, and DB240, DB403 have been
published previously26 and synthesis of all new com-

1108
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
pounds are given below. The compounds were made
into stock solutions with DEPC treated, autoclaved
deionized water. Melting points were recorded using a
Thomas Hoover (Uni-Melt) capillary melting point
apparatus or a Fisher-Johns apparatus and are uncor-
was triturated with cold dry ether and ®ltered to yield 0.51 g
(77%) pale yellow solid, mp 168±170 ꢁC dec. H NMR
1
(DMSO-d6/45 ꢁC) 7.77 (d, 4H, J=8.4), 7.67 (d, 4H,
J=8.4), 7.54 (s, 2H), 3.18 (t, 4H, J=6.8), 2.33 (t, 4H,
J=6.8), 2.16 (s, 12H), 1.12 (q, 4H, J=6.8). M m/e 490
(M+). An ethanolic solution of free base (0.3 g, 0.8
mmol) was saturated with dry HCl gas, stirred at room
temperature for 3 h, the excess ethanol was removed
under reduced pressure and the gummy yellow mass was
triturated with dry ether, the solvent was removed under
reduced pressure and the residue dried in vacuum at
50 ꢁC for 12 h, yellow solid 0.41 g (80%); mp 294±
296 ꢁC dec. 1H NMR (DMSO-d6/D2O/TSP) 7.87 (d,
4H, J=8.4), 7.82 (d, 4H, J=8.4), 7.70 (s, 2H), 3.53 (t,
4H, J=7.2), 3.11 (t, 4H, J=7.2), 2.73 (s, 12H), 2.07ꢁ(q,
4H, J=7.2). 13C NMR (DMSO-d6/D2O.TSP/45 C)
163.1, 142.9, 138.2, 129.5, 127.8, 127.6, 125.9, 54.8, 42.9,
40.3, 23.0. Anal. calcd for: C28H38N6Sꢂ 4HClꢂ 0.5H2O C,
52.09; H, 6.71; N, 13.02. Found: C, 52.08; H, 6.73; N,
12.92.
1
rected. H NMR and 13C NMR spectra were recorded
employing a Varian GX400 spectrometer and chemical
shifts (d) are in ppm relative to trimethylsilane (TMS) and
coupling constants (J) are reported in Hertz. Mass
spectra were recorded on a VG Instruments 70-SE
spectrometer (Georgia Institute of Technology, Atlanta,
GA). IR spectra were recorded using a Michelson 100
(Bomem, Inc.) instrument. Elemental analysis were
obtained from Atlantic Microlab Inc. (Norcross, GA)
and are withinÆ 0.4 of the theoretical values. All che-
micals and solvents were purchased from Aldrich
Chemical Co. or Fisher Scienti®c.
2-[5-(2-Imidazolinyl)-2-benzimidazolyl]-5-[4-(2-imidazolinyl)
phenyl] furan hydrochloride DB302. A suspension of
2-(5-cyano-2-benzimidazolyl)-5-(4-cyanophenyl)furan (31)
(0.5 g, 1.5 mmol) in 50 mL of absolute ethanol was
cooled to 0±5 ꢁC and was saturated with HCl (g). The
¯ask was stoppered and contents was stirred at room
temperature until IR spectra indicated the dis-
appearance of the nitrile peak (ca. 5 days). Anhydrous
ether was added to the suspension and solid was col-
lected by ®ltration and washed with dry ether. The
resultant di-imidate ester was suspended in absolute
ethanol (50 mL) and ethylenediamine (0.23 g, 3.8 mmol)
was added. The mixture was heated at re¯ux for 24 h and
after cooling ether was added. The solid was collected by
®ltration and dried in vacuum at 90ꢁC for 48 h. The free
base was suspended in absolute ethanol saturated with
HCl and heated at re¯ux for 2 h. After cooling, dry ether
was added and solid collected by ®ltration, washed with
acetone and dry ether, and dried in vacuum at 90 ꢁC for
48 h. Yield 0.58 g (75%) of yellow solid, mp >290 ꢁC.
MS (FAB): m/z 397 (M++1); HRMS: calcd mass (free
2,5-Bis[4-{3-(N,N-dimethylaminopropyl)carbamoyl}phenyl]
furan DB339. To a suspension of 2,5-bis[4-carboxy-
phenyl]furan diacid chloride (30) (0.69 g, 2 mmol) in 75
mL dry CH2Cl2 was added 3-(N,N-dimethylamino)propyl-
amine (2 mmol) and the mixture was stirred at room
temperature for 6 h. The solvent was removed by dis-
tillation and the solid residue was treated with ice-water,
the pH was adjusted to 10 with 2 M NaOH and the
solid which resulted was ®ltered, washed with cold
water and crystallized from ether: CHCl3 to yield a pale
yellow solid (75% yield) mp 189±190 ꢁC. The free base
was converted to its HCl salt by treating with saturated
ethanolic-HCl in an 85% yield of yellow hygroscopic
solid mp 205±208 ꢁC dec. H NMR (DMSO-d6) 10.91
1
(b, s, 2H), 8.80 (b, s, 2H), 8.03 (d, 4H, J=8.4), 7.92 (d,
4H, J=8.4), 7.23 (s, 2H), 3.40 (q, 4H, J=6), 3.15
(quintet, 4H, J=5.2), 2.76 (s, 6H), 2.75 (s, 6H), 2.02
(quintet, 4H, J=6.8). 13C NMR (DMSO-d6/D2O)
165.6, 152.4, 132.8, 131.9, 127.7, 122.9, 109.6, 54.3, 41.7,
36.2, 23.8 MS m/e 476 (M+). Anal. calcd for:
1
base): 397.1777 (M++1); observed mass: 397.1813. H
NMR (DMSO-d6/D2O, TMS): 8.28 (s, 1H), 8.17 (d, 2H,
J=8.4), 8.05 (d, 2H, J=8.2), 7.85 (d, 1H, J=8.2)
7.81(d, 1H, J=8.8), 7.52 (d, 1H, J=3.8), 7.47 (d, 1H,
J=3.8), 4.05 (s, 4H), 4.02(s, 4H). 13C NMR (D2O/
DMSO-d6): 165.6; 165.2; 155.2; 145.1; 142.4; 140.4;
137.1; 134.5; 129.7; 125.7; 125.6; 124.5; 121.3; 118.5;
117.2; 116.5; 113.1; 45.8. Anal. calcd for: C23H20-
N6Oꢂ 3HCl: C, 54.61; H, 4.58; N, 16.62; Cl, 21.03.
Found: C, 54.67; H, 4.65; N, 16.63; Cl, 20.93.
ꢂ
C28H36N4O3 2HClꢂ 0.5H2O C, 60.21; H, 7.03; N, 10.03.
Found: C, 60.51; H, 7.13; N, 9.95.
2-{4-[N-(3-N,N-Dimethyaminopropyl)amidino]-2-benzimid-
azolyl}-5-[4-(3-N,N-dimethyaminopropyl)amidino]phenyl]
furan-pentahydrochloride DB340. 2-(5-cyano-2-benzimid-
azolyl)-5-(4-cyanophenyl)furan (31) was converted into
its imidate ester, subsequently allowed to react with
3-N,N-dimethyaminopropylamine, and ®nally converted
into the hydrochloride salt using the procedure above
for DB302 to yield a yellow solid (88% yield); mp 198±
200 ꢁC dec. 1H NMR (DMSO-d6/D2O) 8.13 (d, 2H,
J=8), 8.09 (s, 1H), 7.89 (d, 2H, J=8), 7.79 (d, 1H,
J=8.4), 7.64 (d, 1H, J=8.4), 7.51 (d, 1H, J=2.8), 7.43
(d, 1H, J=2.8), 3.58±3.45 (m, 4H), 3.22±3.15 (m, 4H),
2.79 (s, 12H), 2.18±2.12 (m, 4H). 13C NMR (DMSO-d6/
D2O) 163.1, 162.0, 153.7, 144.6, 143.9, 142.2, 137.3,
133.1, 128.9, 127.7, 123.9, 123.0, 122.8, 115.8, 115.4,
114.3, 110.7, 53.8, 41.8, 40.1, 22.3, 22.3. MS m/e 515
(M+). Anal. calcd for: C29H38N8OHClꢂ 2.5H2O C,
46.89; H, 6.52; N, 15.10. Found: C, 46.82; H, 6.77; N,
15.35.
2,5-Bis-4-N-(3-dimethylaminopropyl)amidinophenylthio-
phene DB316. A solution of freshly distilled 3-dimethyl-
aminopropylamine (0.37 g, 3.6 mmol) in 2 mL absolute
ethanol was added to a suspension of bis-imidate ester
dihydrochloride (0.78 g, 1.7 mmol), prepared from 2,5-
bis[4-cyanophenyl]thiophene,36 in 30 mL ethanol (under
nitrogen). The mixture was stirred at room temperature
for 12 h and the excess solvent was distilled. The residue
was washed with dry ether and dissolved in water, ®l-
tered and basi®ed, while cooling and stirring, to yield a
gummy mass. The solid was separated from the aqueous
phase and dissolved in 60 mL of chloroform, dried over
anhydrous Na2SO4, ®ltered and concentrated. The residue

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1109
2,5-Bis[3-(2-imidazolinyl)phenyl]furan
dihydrochloride
A mixture of 2,5-bis(3-cyanophenyl)furan (2.5 g, 9.2
mmol) in 35 mL dry ethanol was saturated while stirring
and cooling (0±5 ꢁC) with dry HCl gas. The mixture was
stirred at room temperature for 6 days (aliquots of the
mixture were monitored by IR for the disappearance of
the nitrile function), diluted with 75 mL dry ether and
the resultant solid was ®ltered under nitrogen, washed
with ether to yield the imidate ester hydrochloride 3.72 g
(92%) after drying under vacuum at 30 ꢁC for 5 h. A
mixture of imidate ester HCl (0.87 g, 2 mmol) and dry
ethylenediamine (0.13 g, 2.2 mmol) in 15 mL absolute
ethanol was heated at re¯ux for 12 h. The solvent was
removed under vacuum and the residue was treated with
water and basi®ed with 2 M NaOH. The precipitated solid
was ®ltered, washed with water, dried and crystallized
from chloroform/ether, to yield an o€-white solid 0.57 g
(80%) mp 222±224 ꢁC. The free base (0.45 g, 1.3 mmol)
was converted into the salt with ethanolic HCl to yield
DB361. Divinyl sulfone (5.9 g, 5 mmol) in 20 mL dry
ethanol was added to a boiling mixture of m-bromobenz-
aldehyde (18.5 g, 1 mmol), anhydrous sodium acetate
(2.46 g, 0.03 mol) and 3-benzyl-5(2-hydroxyethyl)thiaz-
olium chloride (2.69 g, 0.01 mol) in 120 mL dry ethanol.
The mixture was allowed to re¯ux for 16 h, cooled, ®ltered,
washed with ether, water and dried. The solid was extracted
with boiling chloroform (4Â200 mL), the combined
extracts were washed with a NaHCO3 solution (20%)
and dried over anhydrous sodium sulfate. The solution
was passed through a silica gel column to yield after
removing solvent a white crystalline solid 7.90 g (40%)
mp 170±172 ꢁC dec. 1H NMR (DMSO-d6) 8.09 (m, 2H),
7.89 (m, 2H), 7.82 (m, 2H), 7.5 (m, 2H), 3.39 (s, 4H).
13C NMR (DMSO-d6) 197.1, 138.4, 135.2, 130.4, 129.9,
126.4, 121.7, 32.3. MS m/e 396 (M+). The 1,4-di(3-bromo-
phenyl)butanedione thus obtained was used directly,
without further characterization, in the next step.
ꢁ
1
0.51 g (88%) of a solid, mp 258±260 C dec. H NMR
(DMSO-d6) 8.65 (s, 2H), 8.16 (d, 2H, J=8), 7.91 (d, 2H,
J=8), 7.67 (m, 2H), 7.22 (s, 2H), 4.03 (s, 8H). 13C NMR
(DMSO-d6) 164.5, 151.9, 130.8, 129.9, 129.4, 127.4,
123.6, 122.8, 110.1, 44.4. MS (FAB) m/e 357 (M++1).
Anal. calcd. for: C22H20N4Oꢂ 2HClꢂ H2O. C, 59.06; H,
5.40; N, 12.52. Found: C, 59.29; H, 5.07; N, 12.54.
To a solution of 1,4-di(3-bromophenyl)butanedione
(7.92 g, 0.02 mol) in 80 mL boiling acetic anhydride, ®ve
drops of concd H2SO4 was added and the mixture was
allowed to re¯ux for 5 min. The solution was cooled,
poured on ice-water, stirred, and the light brown solid
which formed was ®ltered. The solid was washed with
water, dissolved in chloroform (200 mL), washed with
saturated NaHCO3, water, and dried over Na2SO4. The
solution was passed through a bed of silica gel and on
evaporation of the solvent yielded 5.7 g (75%) white
2,5-Bis[3-(N-(3-dimethylaminopropyl)amidino)]phenylfuran
tetrahydrochloride DB355. 2,5-Di(3-cyanophenyl)furan,
from above, was converted into its imidate ester, subse-
quently allowed to react with 3-N,N-dimethyaminopropyl-
amine, and ®nally converted into the hydrochloride salt
using our standard procedure described for DB302 to
yield an of hygroscopic o€-white solid which was dried
crystalline solid, mp 118±119 ꢁC. H NMR (DMSO-d6)
1
7.98 (m, 2H), 7.79 (d, 2H, J=8), 7.47 (d, 2H, J=8.0),
7.37 (m, 2H), 7.14 (s, 2H). 13C NMR (DMSO-d6) 151.3,
131.8, 130.6, 129.9, 125.7, 122.3, 122.0, 109.3. MS m/e
378 (M+). The 2,5-bis(3-bromophenyl)furan thus
obtained was used directly, without further character-
ization, in the next step.
in vacuum at 70 ꢁC for 24 h; mp 190±191 ꢁC dec. H
1
NMR (DMSO-d6) 8.25 (b, s, 2H), 8.13 (d, 2H, J=7.6),
7.73±7.63 (m, 4H), 7.23 (s, 2H), 3.56 (t, 4H, J=7.6),
3.21 (t, 4H, J=7.6), 2.79 (s, 12H), 2.12 (m, 4H). 13C
NMR (DMSO-d6) 163.1, 152.2, 130.7, 129.9, 129.6,
128.3, 127.3, 123.4, 110.2, 54.4, 42.6, 40.1, 22.6. MS m/e
475 (M++1). Anal. calcd for: C28H38N6Oꢂ 4HCl C,
52.66; H, 6.94; N, 13.16. Found: C, 52.63; H, 6.96; N,
13.10.
A mixture of 2,5-bis(3-bromophenyl)furan (5.0 g, 13
mmol) and CuCN (4.15 g, 46 mmol) in 10 mL dry
N-methyl-2-pyrrolidone was allowed to re¯ux under N2
for 2 h (reaction progress followed by TLC, silica gel,
Kodak plate, CHCl3). The mixture was cooled, stirred
with 150 mL of 10% aqueous NaCN for 5 h, ®ltered,
washed with water and dried. The solid was dissolved in
warm chloroform and passed through a neutral Al2O3
column and on evaporation of the solvent yielded 2,5-
bis(3-cyanophenyl)furan as white crystalline needles
2,5-Bis[4{N-(4-dimethylaminobutyl)amidino}phenyl]furan
tetrahydrochloride DB378. 2,5-Di(4-cyanophenyl)furan37
was converted into its imidate ester, subsequently
allowed to react with 4-dimethylaminobutylamine, and
®nally converted into the hydrochloride salt using our
standard procedure described for DB302 to yield a yellow
ꢁ
1
2.65 g (75.5%), mp 164±165 C. H NMR (DMSO-d6)
8.32 (s, 2H), 8.14 (d, 2H, J=8), 7.72 (d, 2H, J=7.6),
7.63 (d, 1H, J=8), 7.61 (d, 1H, J=7.6). 13C NMR
(DMSO-d6) 151.2, 130.7, 130.6, 129.8, 127.7, 126.8,
118.2, 112.1, 109.9. MS m/e 270 (M+). Anal. calcd for:
C18H10N2O C, 79.98; H, 3.73; N, 10.36. Found: C,
79.45; H, 3.81; N, 10.40. Alternatively, 2,5-bis(3-cyano-
phenyl)furan was prepared by cylodehydration of 1,4-
di(3-cyanophenyl)butanedione, mp 201±202ꢁC, 1H NMR
(DMSO-d6) 8.38 (s, 2H), 8.27 (d, 2H, J=7.6), 8.05 (d,
2H, J=7.6), 7.74 (m, 2H), 3.47 (s, 4H). 13C NMR
(DMSO-d6) 197.2, 137.3, 135.9, 131.9, 131.3, 129.9,
127.7, 117.7, 111.9, 32.5. MS m/e 288 (M+) obtained
from 3-cyanobenzaldehyde using Stetter methodology
(cf. ref 7).
solid (86%); mp 205±208 ꢁC dec. H NMR (DMSO-d6)
1
8.0 (d, 4H, J=7.6), 7.82 (d, 4H, J=7.6), 7.32 (s, 2H),
3.44 (t, 4H, J=6.4), 3.08 (t, 4H, J=7.2), 2.75 (s, 12H),
1.82±1.62 (m, 8H). 13C NMR (DMSO-d6) 162.9, 152.9,
134.4, 129.6, 127.8, 124.4, 112.0, 56.7, 42.7, 42.5, 24.7,
21.7. MS (FAB) m/e 503 (M++1). Anal. calcd for:
C30H42N6Oꢂ 4HClꢂ 0.75 H2O. C, 54.42; H, 7.23; N,
12.69. Found: C, 54.46; H, 7.06; N, 12.46.
2,5-Bis[4-[N-(6-dimethylaminohexyl)amidino]phenyl]furan
tetrahydrochloride DB379. 2,5-Di(4-cyanophenyl)furan37
was converted into its imidate ester, subsequently
allowed to react with 6-dimethylaminohexylamine, and
®nally converted into the hydrochloride salt using our

1110
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
standard procedure described for DB302 to yield a yellow
solid (86%); mp 195±200 ꢁC dec. H NMR (DMSO-d6)
2,5-Bis[3(2-imidazolinyl)phenyl]pyrrole dihydrochloride
DB382. 2,5-Di(3-cyanophenyl)pyrrole was converted
into its imidate ester, subsequently allowed to react with
ethylenediamine, and ®nally converted into the hydro-
chloride salt ꢁusing our standard procedure (81% yieꢁld);
1
8.0 (d, 4H, J=8), 7.84 (d, 4H, J=8), 7.31 (s, 2H), 3.43
(t, 4H, J=6.4), 3.02 (t, 4H, J=8), 2.72 (s, 12H), 1.72±
1.61 (m, 8H), 1.44±1.32 (m, 8H). 13C NMR (DMSO-d6)
162.4, 152.6, 134.1, 129.2, 127.5, 124.1, 111.6, 56.8, 42.8,
42.3, 27.1, 25.7, 25.6, 23.7. MS (FAB) m/e 559 (M++1).
Anal. calcd for: C34H50N6Oꢂ 4HClꢂ 0.75H2O. C, 56.87; H,
7.79, N, 11.70. Found: C, 56.88; H, 7.61; N, 11.63.
1
mp 364±365 C dec. H NMR (DMSO-d6/D2O/50 C)
8.54 (s, 2H), 8.13 (d, 2H, J=8), 7.72 (d, 2H, J=7.6),
7.58 (d, 2H, J=8). 7.57 (d, 2H, 7.6), 6.78 (s, 2H), 4.0 (s,
8H). 13C NMR (DMSO-d6/D2O/50 ꢁC) 165.3, 133.2, 132.4,
130.1, 129.8, 125.8, 124.3, 122.6, 109.6, 44.6. MS (FAB) m/e
+
ꢂ
2,5-Bis[4-[N-(3-(N-pyrrolidinopropyl)amidino]phenyl]furan
tetrahydrochloride DB380. 2,5-Di(4-cyanophenyl)furan37
was converted into its imidate ester, subsequently
allowed to react with 3-(N-pyrrolidino)propylamine,
and ®nally converted into the hydrochloride salt using
our standard procedure described for DB302 to yield a
yellow solid (85%); mp 200±204 ꢁC dec. 1H NMR
(DMSO-d6) 7.99 (d, 4H, J=8), 7.31 (s, 2H), 7.90 (d, 4H,
J=8), 7.31 (s, 2H), 3.60 (t, 4H, J=6.4), 3.28 (t, 4H,
J=7.2), 3.05 (b, m, 4H), 2.52±2.46 (m, 4H), 2.18±2.10
(m, 4H), 2.06±1.96 (m, 4H), 1.94±1.86 (m, 4H). 13C
NMR (DMSO-d6) 162.5, 152.5, 134.0, 129.1, 127.2,
123.8, 111.4, 53.2, 51.4, 23.9, 22.8. MS (FAB) m/e 526
(M++1). Anal. calcd for: C32H42N6Oꢂ 4HClꢂ 1.5H2O.
C, 54.93; H, 7.05; N, 12.01. Found: C, 55.03; H, 7.02; N,
11.93.
356(M +1). Anal. calcd for: C22H21N5 2HClꢂ 0.5H2O: C,
60.41; H, 5.64; N, 16.01. Found: C, 60.58; H, 5.50; N,
16.07.
2,5-Bis[3-([3-N,N-dimethylaminoethyl]amidino)phenyl]-
pyrrole dihydrochloride DB383. 2,5-Di(3-cyanophenyl)
pyrrole was converted into its imidate ester, subsequently
allowed to react with 2-N,N-dimethylaminoethylamine,
and ®nally converted into the hydrochloride salt using
our standard procedure to yield a brownish-yellow solid
(90% yield); mp 297±299ꢁC dec. H NMR (DMSO-d6/
1
D2O) 8.39 (s, 2H), 8.10 (d, 2H, J=7.2), 7.63 (d, 2H,
J=7.2), 7.57 (d, 2H, J=7.2), 7.55 (d, 2H, J=7.2), 6.78
(s, 2H), 3.52 (b m, 4H), 2.89 (s, 12H), 2.18±2.04 (m, 4H).
13C NMR (DMSO-d6/D2O/50 ꢁC) 164.0, 133.2, 132.7,
129.9, 129.3, 129.1, 125.8, 124.1, 109.6, 54.4, 43.2,38.3.
MS (FAB) m/e 446(M++1). Anal. calcd for: C26H35-
ꢂ
2,5-Bis[3-([3-N,N-dimethylaminopropyl]amidino)phenyl]-
pyrrole dihydrochloride DB381. A mixture of 1,4-di(3-
bromophenyl)butanedione (3.96 g, 1 mmol) (see DB355)
and anhydrous ammonium acetate (7.7 g, 0.1 mol) in 20
mL glacial acetic acid was heated at re¯ux, under
nitrogen, for 5 h. The mixture was cooled and the solid
which formed was ®ltered, washed with hexane and
dried under vacuum at 60 ꢁC for 24 h. The compound
was recrystallized from CHCl3/hexane to yield 2.98 g
N7 4HCl: C, 52.80; H, 6.64; N, 16.58. Found: C, 52.63;
H, 6.70; N, 16.43.
2,5-Bis[4-(N-(3-(4-N-methylpiperazino)propyl)amidino]-
phenyl]furan hexahydrochloride DB394. 2,5-Di(4-cyano-
phenyl)furan37 was converted into its imidate ester,
subsequently allowed to react with 3-(N-methylpiper-
azino)propylamine, and ®nally converted into the
hydrochloride salt using our standard procedure described
for DB302 to yield a yellow solid (82%); mp 240±245 ꢁC
dec. 1H NMR (DMSO-d6/D2O) 8.01 (d, 4H, J=8), 7.88
(d, 4H, J=8), 7.31 (s, 2H), 3.66 (b, 16H), 3.29 (t, 4H,
J=8), 2.86 (s, 6H), 2.48 (b, 4H), 2.19±2.10 (m, 4H). 13C
NMR (DMSO-d6) 162.7, 152.6, 134.1, 129.2, 127.3,
123.9, 111.5, 53.1, 49.6, 48.0, 47.8, 42.1, 22.3. MS (FAB)
m/e 585 (M++1). Anal. calcd for: C34H48N8Oꢂ
6HClꢂ1.25H2O C, 49.43; H, 6.89; N, 13.56. Found: C,
49.33; H, 6.91; N, 13.45.
(79%), mp 143±145. H NMR (DMSO-d6/50 ꢁC) 11.33
1
(s, 1H), 8.02 (s, 2H), 7.73 (d, 2H, J=7.6), 7.3ꢁ8±7.27 (m,
4H), 6.68(s, 2H). 13C NMR (DMSO-d6/50 C) 134.4,
131.8, 130.3, 128.1, 125.9, 122.8, 122.0, 108.7. MS m/e
377 (M+). 2,5-di(3-Bromophenyl)pyrrole (2.9 g, 7.7
mmol) was used directly without further characteriza-
tion by heating with CuCN in N-methyl pyrrolidone in
a standard procedure1 to yield 2,5-di(3-cyanophenyl)
pyrrole; 1.42 g (68%) of an o€ꢁ-white solid mp 210±
ꢁ
1
212 C. H NMR (DMSO-d6/50 C) 11.44 (s, 1H), 8.24
(s, 2H), 8.06 (d, 2H, J=8), 7.61 (d, 2H, J=8), 7.56 (m,
2H), 6.80 (s, 2H). 13C NMR (DMSO-d6/50 ꢁC) 133.1,
131.7, 129.6, 128.9, 128.2, 126.8, 118.6, 111.7., 109.4.
Anal. calcd for: C18H11N3: C, 80.29; H, 4.12; N, 15.60.
Found: C, 80.29; H, 4.19; N, 15.51. The bis-nitrile was
converted into its imidate ester, subsequently allowed to
react with 3-N,N-dimethylaminopropylamine, and ®nally
converted into the hydrochloride salt using our standard
procedure to yield a brownish-yellow solid (92% yield);
2,5-Bis-4{N-(4-dimethylaminopropyl)amidino}phenyl-3-
methylfuran tetrahydrochloride DB400. 2,5-di(4-cyano-
phenyl)-3-methylfuran36 was converted into its imidate
ester, subsequently allowed to react with 3-N,N-dimethy-
aminopropylamine, and ®nally converted into the
hydrochloride salt using our standard procedure to
yield a yellow solid (75% yield); mp 215±217 ꢁC dec. 1H
NMR (DMSO-d6/D2O) 7.96 (d, 2H, J=8.4), 7.92 (d,
2H, J=8.8), 7.87 (d, 2H, J=8.8), 7.84 (d, 2H, J=8.4),
7.21 (s, 1H, J=8.4), 3.54±3.48 (m, 4H), 3.21±3.15 (m,
4H), 2.78 (s, 6H), 2.77 (s, 6H), 2.35 ( s, 3H), 2.13±1.89
(m, 4H). 13C NMR (DMSO-d6/D2O) 163.4, 151.2,
147.9, 135.2, 134.5, 133.8, 129.5, 129.3, 127.6, 127.0,
125.5, 124.4, 123.2 115.2, 54.9, 43.0, 40.4, 22.9, 12.3. MS
m/e 489 (M+). Anal. calcd for: C29H40N6Oꢂ 4HClꢂ1H2O
C, 53.37; H, 7.10; N, 12.88. Found: C, 53.54; H, 7.12; N,
12.83.
mp 197±199 ꢁC dec. H NMR (DMSO-d6/D2O) 8.22 (s,
1
2H), 8.10±8.0 (m, 2H), 7.62±7.51 (m, 4H), 6.75(s, 2H),
3.58±3.50 (b, m, 4H), 3.24±3.18 (b, m, 4H), 2.79 (s, 12ꢁH),
2.18±2.04 (m, 4H). 13C NMR (DMSO-d6/D2O/50 C)
163.9, 133.6, 132.9, 130.3, 129.5, 129.4, 125.8, 123.9, 109.8,
54.9, 45.1,40.4,23.0. MS (FAB) m/e 474 (M++1). Anal.
ꢂ
calcd for: C28H39N7 4HClꢂ 2.5H2O: C, 50.60; H, 7.28;
N, 14.75. Found: C,50.61; H, 7.31; N, 14.72.

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1111
2,5-Bis{4-[N-(3-N,N-dimethyaminopropyl)amidino]phenyl}-
3-(p-tolyloxy) furan tetrahydrochloride DB 414. 2,5-di(4-
cyanophenyl)-3-tolyloxyfuran38 was converted into its
imidate ester, subsequently allowed to react with 3-N,N-
dimethyaminopropylamine, and ®nally converted into
the hydrochloride salt using our standard procedure to
yield a yellow solid (90% yield); mp 205±206 ꢁC dec. 1H
NMR (DMSO-d6/D2O) 7.99 (d, 2H, J=8.4), 7.97 (d,
2H, J=8.4), 7.80 (d, 4H, J=8.4), 7.19 (d, 2H, J=8.4),
7.12 (s, 2H), 7.03 (d, 2H, J=8.4), 3.52±3.46 (m, 4H),
3.18±3.14 (m, 4H), 2.78 (d, 6H, J=6.4), 2.77 (d, 6H,
J=6.4), 2.24 (s, 3H), 2.11±2.08 (m, 4H). 13C NMR
(DMSO-d6/D2O) 163.6, 154.7, 151.1, 145.2, 139.6,
134.2, 134.0, 133.8, 131.1, 129.5, 128.4, 127.2, 124.7,
124.0, 117.5, 106.0, 55.0, 43.2, 40.4, 23.0, 20.6. MS m/e
54.5, 42.7, 40.2, 22.7. MS m/e (FAB) 487 (M++1).
ꢂ
Anal. calcd for: C28H38N8 5HClꢂ 2H2O C, 47.70; H,
6.72; N, 15.89. Found: C, 47.99; H, 6.59; N, 15.67.
2,5-Bis-4-(N-(3-dimethylaminopropyl)amidino)phenyloxa-
zole tetrahydrochloride DB449. 2,5-di(4-cyanophenyl)
oxazole40 was converted into its imidate ester, subsequently
allowed to react with 4-dimethylaminopropylamine, and
®nally converted into the hydrochloride salt using our
standard procedure described for DB302 to yield a pale
yellow solid 85%, mp 205±208 ꢁC dec. 1H NMR (DMSO-
d6/D2O) 8.28 (d, 2H, J=8.4), 8.05 (d, 2H, J=8.4), 8.06
(s, 1H), 7.95 (d, 2H, J=8.4), 7.91 (d, 2H, J=8.4), 3.55±
3.48 (m, 4H), 3.25±3.15 (m, 4H), 2.79 (s, 12H), 2.16±
2.04 (m, 4H). 13C NMR (DMSO-d6/D2O) 162.8, 160.2,
150.6, 143.4, 131.6, 130.6, 129.4, 129.37, 128.5, 127.1,
126.7, 124.7, 54.4, 42.5, 40.2, 22.6. MS (FAB) m/e 476
(M++1). Anal. calcd for: C22H37N7Oꢂ 4HClꢂ H2O C,
50.70; H, 6.77; N, 15.33. Found: C, 50.46; H, 6.76; N,
15.07.
+
581 (M ). Anal. calcd for: C35H44N6O2 4HClꢂ 1.5H2O
C, 55.78 ; H, 6.82; N, 11.15. Found: C, 55.93; H, 6.84;
N, 11.05.
ꢂ
2,5-Bis[3-([3-N-methyl-N-(3-aminopropyl)aminopropyl]
amidino)phenyl] furan hexahydrochloride DB420. 2,5-
Di(3-cyanophenyl)furan from above, was converted into
its imidate ester, subsequently allowed to react with 3,30-
diamino-N-methyldipropylamine, and ®nally converted
into the hydrochloride salt using our standard procedure
described for DB302 to yield (72%) of a very hygro-
scopic white solid which was dried in vacuum at 70 ꢁC
for 24 h; mp 250±253 ꢁC dec. 13C NMR (DMSO-d6)/
D2O) 163.1, 152.3, 130.8, 129.9, 129.6, 128.4, 127.4,
123.6, 110.2, 53.1, 52.7, 40.3, 36.2, 22.4, 21.8. MS (FAB)
m/e 561 (M++1). Anal. calcd for: C32H48N8Oꢂ
6HClꢂ 3H2O C, 46.10; H, 7.25; N, 13.44. Found: C,
46.21; H, 7.51; N, 13.66.
2,5-Bis-3-[(N-(4-N,N-dimethylaminohexyl)amidino)]phenyl-
thiophene tetrahydrochloride DB492. 1,4-Bis-(3-bromo-
phenyl)butanedione, from above, (3.96, 0.01 mol) and
Lawesson's reagent (5 mmol, 2.02 g) in dry benzene was
heated at re¯ux for 6 h (TLC monitored). The solvent
was distilled under reduced pressure. The residue was
dissolved in CHCl3 and passed through a silica column
using ether/hexane mixture to yield an o€ white solid
2.5 g (65%) mp 135±136 ꢁC. 1H NMR (CDCl3) 7.75 (m,
2H), 7.51 (m, 2H), 7.40 (m, 2H), 7.26 (s, 2H), 7.23 (m,
2H). 13C NMR (CDCl3) 142.7, 136.2, 130.6, 130.4,
128.6, 124.9, 124.3, 123.1. MS m/e 394 (M+). The 2,5-bis
(3-bromophenyl) thiophene was used directly without fur-
ther characterization to prepare 2,5-bis-(3-cyanophenyl)
thiophene (60% yield, mp 178±180 ꢁC) by the action of
CuCN in N-methylpyrrolidone in a standard procedure.36
1H NMR (CDCl3) 7.87 (m, 2H), 7.82 (m, 2H), 7.57 (m,
2H,), 7.50 (m, 2H), 7.35 (s, 2H). 13C NMR (CDCl3)
142.3, 136.1, 131.0,. 129.9, 129.7, 129.0, 125.6, 118.3,
113.5. MS m/e 276 (M+). 2,5-Di-(3-cyanophenyl) thio-
phene, without further characterization, was converted
into its imidate ester, subsequently allowed to react with
4-dimethylaminohexyllamine, and ®nally converted into
the hydrochloride salt using our standard procedure
described for DB302 to yield (85%) a very hygroscopic
2,5-Bis-3-[(N-(4-N,N-dimethylaminobutyl)amidino)]phenyl-
furan tetrahydrochloride DB434. 2,5-di(3-cyanophenyl)
furan (see above) was converted into its imidate ester,
subsequently allowed to react with 4-N,N-dimethylamino-
butylamine, and ®nally converted into the hydrochloride
salt using our standard procedure to yield a very hygro-
scopic pale yellow solid (71% yield); mp 175±177 ꢁC
1
dec. H NMR (DMSO/D2O) 8.19±8.12 (m, 4H), 7.68±
7.63 (m, 4H), 7.23 (s, 2H), 3.47 (t, 4H, J=6.8), 3.09 (m,
4H), 2.75 (s, 12H), 1.82±1.67 (m, 8H). 13C NMR
(DMSO-d6) 163.6, 152.6, 131.1, 130.4, 130.1, 128.6,
127.5, 123.6, 110.5, 56.9, 42.7, 42.6, 24.7, 21.7. MS
(FAB) m/e 503 (M++1). Anal. calcd for:
C30H42N6Oꢂ 4HClꢂ 2H2O C, 52.63; H, 7.36; N, 12.27.
Found: C, 52.48; H, 6.92; N, 11.83.
solid; mp 147±150 ꢁC dec. H NMR (DMSO-d6/D2O)
1
8.11 (b, 2H), 7.95 (d, 2H, J=8), 7.75 (s, 2H), 7.73 (d,
2H, J=8), 7.64 (t, 2H, J=8), 3.49 (t, 4H, J=7.2), 3.02
(t, 4H, J=7.2), 2.72 (s, 12H), 1.73±1.65 (m, 8H), 1.43±
1.35 (m, 8H). 13C NMR (DMSO-d6/D2O) 162.3, 141.9,
133.9, 129.7, 129.6, 129.5, 127.2, 126.5, 124.9, 56.5, 42.5,
41.9, 26.8, 25.4, 25.3, 23.2. MS (FAB) m/e 575
(M++1). Anal. calcd for: C34H50N6Sꢂ 4HClꢂ H2O C,
55.43; H, 7.39; N, 111.40. Found: C, 55.41; H, 7.71; N,
11.40.
2,4-Bis-4-(N-(3-dimethylaminopropyl)amidino)phenylpyri-
midine pentahydrochloride DB436. 2,4-di(4-cyanophenyl)
pyrimidine39 was converted into its imidate ester, sub-
sequently allowed to react with 4-dimethylaminopropyl-
amine, and ®nally converted into the hydrochloride salt
using our standard procedure described for DB302 to
yield a very hygroscopic white solid (69%); mp 105±
109 ꢁC dec. 1H NMR (DMSO/D2O) 9.05 (d, 1H,
J=5.2), 8.61 (d, 2H, J=8), 8.48 (d, 2H, J=8), 8.17 (d,
1H, J=5.2), 8.10 (d, 2H, J=8), 8.07 (d, 2H, J=8), 3.60 (t,
4H, J=7.2), 3.25±3.19 (m, 4H), 2.79 (s, 12H), 2.18±2.15
(m, 4H). 13C NMR (DMSO/D2O) 163.2, 163.1, 162.7,
162.3, 141.7, 140.7, 129.3, 129.0, 128.5, 127.9, 116.9,
2,5-Bis-3-[(N-(4-N,N-dimethylaminobutyl)amidino)]phenyl-
thiophene tetrahydrochloride DB493. 2,5-Di(3-cyano-
phenyl)thiophene was converted into its imidate ester,
subsequently allowed to react with 4-dimethylamino-
butylamine, and ®nally converted into the hydrochloride
salt using our standard procedure described for DB302

1112
G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
to yield (90%) a pale yellow hygroscopic solid (90%);
mp 170±172 ꢁC dec. H NMR (DMSO-d6/D2O) 8.08 (b,
room temperature. For all studies, the RNA hairpin
and duplexes were reannealed by heating to 90 ꢁC and
cooled down in room temperature prior to use.
1
2H), 7.96 (d, 2H, J=7.6), 7.72 (d, 2H, J=7.6), 7.71 (s,
2H), 7.63 (t, 2H, J=7.6), 3.52 (t, 4H, J=6.8), 3.10 (t,
4H, J=7.2), 2.75 (s, 12H), 1.84±1.72 (m, 8H). 13C NMR
(DMSO-d6/D2O) 162.7, 142.1, 134.1, 129.9, 129.8,
129.7, 127.3, 126.6, 124.9, 56.3, 42.3, 42.1, 24.4, 21.2.
MS (FAB) 519 (M++1). Anal. calcd for:
C30H42N6Sꢂ 4HClꢂ 0.75H2O (676.11). C, 53.29; H, 6.78;
N, 12.43. Found: C, 53.26; H, 7.00; N, 12.25.
The Rev-17 peptide (Fig. 1(a)) with the sequence de®ned
by Tan et al.30 was purchased from Chiron. The peptide
is succinylated at the amino termini and amidated at the
carboxy termini, modi®cations that have been shown to
increase alpha-helicity of the peptide and its binding to
RRE.30 The Tat-24 peptide (Fig. 1(a)) with the sequence
de®ned by Long et al.34 was prepared as previous descri-
bed.33
2,5-Bis-4-(N-(3-dimethylaminopropyl)amidino)phenylpyrrole
tetrahydrochloride DB522. 2,5-di(4-cyanophenyl)pyrrole
(32) was converted into its imidate ester, subsequently
allowed to react with 4-dimethylaminopropylamine, and
®nally converted into the hydrochloride salt using our
standard procedure described for DB302 to yield a yellow
Thermal denaturation experiments
The RRE duplex melting studies were conducted in Rev
dilution bu€er (10 mM phosphate, 1 mM EDTA, 0.1 M
NaCl, 1 mM MgCl, 1 mM DTT). The thermal dissociation
studies were performed with a Cary 3 spectrophotometer
by following the absorbance change at 260 nm as a
function of temperature. The temperature was con-
trolled by Varian software with a Cary temperature
controller that was programmed to raise the tempera-
ture at a rate at 0.5 ꢁC/min. A thermister ®xed into a
reference cuvette was used to monitor the temperature.
Tm values were determined from the ®rst derivative
plots.41 Compounds are compared by the increase in Tm
(ÁTm=Tm of complexesÀTm of free nucleic acids) that
they produce at each concentration.
solid (88%); mp 273±275 ꢁC dec. H NMR (DMSO-d6/
1
D2O) 7.99 (d, 4H, J=8.4), 7.79 (d, 4H, J=8.4), 6.86 (s,
2H), 3.57±3.52 (m, 4H), 3.22±3.16 (m, 4H), 2.79 (s,
12H), 2.14±2.05 (m, 4H). 13C NMR (DMSO-d6/D2O)
162.7, 152.6, 134.1, 129.2, 127.3, 123.9, 111.5, 53.1, 49.6,
42.1, 22.3. MS (FAB) m/e 474 (M++1). Anal. calcd
ꢂ
for: C28H39N7 4HClꢂ 2H2O C, 51.30; H, 7.22; N, 14.96.
Found: C, 51.19; H, 7.41; N, 14.79.
RNA and peptides. The RRE hairpin and RRE and
TAR duplex strands (Fig. 1(a)) were purchased from
Cruachem with HPLC puri®cation. Oligonucleotide
concentrations were determined by absorbance at 260
nm. Single-strands of RNAs were end-labeled using
[g-32P]ATP and T4 polynucleotide kinase under stan-
dard conditions.33 For ecient labeling, it is very
important that the RNA is free of any impurities that
would be preferentially labeled. The HPLC puri®ed
RRE hairpin was labeled very poorly and for this rea-
son the RRE hairpin was puri®ed by electrophoresis
through a denaturing gel (20% acrylamide, 50% urea).
Individual bands detected by UV shadowing were
excised, eluted and Centriconed to remove ions. After
radiolabeled as previous described,33 the sample was
suspended in denaturing loading bu€er and puri®ed on
a 20% denaturing polyacrylamide gel. After auto-
radiography, the band corresponding to the 30 mer
RRE hairpin was excised from the gel and eluted in 500
mL elution bu€er (0.5 M NH4OAc, 10 mM Mg(OAc)2,
1.0 mM EDTA and 0.1% SDS) at 0 ꢁC overnight. The
RRE oligomer was precipitated from the supernatant
by addition of 80% cold ethanol and was centrifuged for
30 min. The resulting pellet was washed twice with cold
80% ethanol. The concentrations of labeled RNA were
calculated from the speci®c activity of 32P incorpora-
tion. The concentrations of unlabeled RNA solutions
were quantitated by spectrophotometry. The RRE hairpin
model stock solutions were prepared by heating the
unlabeled or 32P-50-end labeled RNA strand at micro-
molar concentrations to 90 ꢁC for 2 min in sodium phos-
phate bu€er (10 mM phosphate, 1 mM EDTA, 0.1 M
NaCl, adjusted to pH 7.0 with NaOH) with slow cooling to
room temperature. The duplex model stock solutions were
prepared by heating the mixtures of 32P-50-end labeled
complementary strands at micromolar concentrations to
90 ꢁC for 2 min in the same bu€er with slow cooling to
Gel shift assay for RRE/Rev and TAR/Tat complexes.
RNA 32P-labeled duplexes were formed as described
above, and the stock solution was diluted and added to
a sodium phosphate bu€er. RNA solutions were ali-
quoted (8 mL), and increasing concentrations of Rev-17
peptide or Tat-24 peptide were added (2 mL) to achieve
®nal peptide concentration ranges. The ®nal duplex
concentration was 5 nM. The ®nal reaction conditions
in a 12 mL total volume were 10 mM phosphate, 1 mM
EDTA, 0.1 M NaCl, 1 mM MgCl2, 1 mM DTT, 25 mg/
mL yeast tRNA (Sigma), and 10% glycerol. The reaction
mixtures were incubated for at least 15 min, and samples
(3 mL) were then applied to a 12% native polyacrylamide
gel (20Â20 cm) cast in 1ÂTBE (90 mM Tris/borate and
2 mM EDTA at pH 7.7) and prequilibrated overnight at
20 ꢁC. The samples were electrophoresed for 4 h at 220
V at 20 ꢁC. The gels were ®xed, dried in vacuo at 80 ꢁC,
and autoradiographed (Biomax, Kodak). Autoradio-
graphs were scanned on a densitometer (PDI, Inc.) and
quantitated using software supplied by PDI.
Competitive gel shift assay for inhibitors of RRE/Rev
TAR/Tat complexation. As described above, labeled
RRE duplex or hairpin solution (2 mL) was incubated
with 6 mL binding reaction bu€er (above). To this solution
was added 2 mL of competitive inhibitor at a range of
concentrations and either the Rev-17 or Tat-24 peptide
solution (2 mL). Labeled RNA duplex and competitive
inhibitor were pre-mixed in reaction bu€er to prevent
premature interaction with Rev peptide as described by
Daly et al.42 Incubation, electrophoresis and imaging were
the same as described above. The IC50 was calculated by
®tting the densitometer data to the following equation:

G. Xiao et al. / Bioorg. Med. Chem. 9(2001) 1097±1113
1113
19. Rogers, J.; Chang, A. H.; Ahsen, U. v.; Schroeder, R.
J. Mol. Biol. 1996, 36, 916.
20. Ahsen, U. v.; Davies, J.; Schroeder, R. Nature 1991, 353,
368.
‰drugŠ
% Inhibition ˆ
‰drugŠ ‡ IC50
21. Mei, H.-Y.; Galan, A. A.; Halim, N. J.; Mack, D. P.;
Moreland, J. W.; Sanders, K. B.; Truong, H. N.; Czarnik, A. J.
Bioorg. Med. Chem. Lett. 1995, 5, 2755.
22. Wang, S.; Huber, P. W.; Cui, M.; Czarnik, A. W.; Mei,
H.-Y. Biochemistry 1998, 37, 5549.
23. Zapp, M. L.; Stern, S.; Green, M. R. Cell 1993, 74, 969.
24. Rogers, M. J.; Bukhman, Y. V.; McCutchan, R. F.; Dra-
per, D. E. RNA 1997, 3, 815.
where % Inhibition is the percent of intensity of the free
RNA band; [drug] is the compound concentration after
each addition; IC50 is the calculated concentration of
the compound when the intensity of the upper band is
equal to the lower band.
25. Conn, G. L.; Gutell, R. R.; Draper, D. E. Biochemistry
1998, 37, 11980.
Acknowledgements
26. Ratmeyer, L.; Zapp, M. L.; Green, M. R.; Vinayak, R.;
Kumar, A.; Boykin, D. W.; Wilson, W. D. Biochemistry 1996,
35, 13689.
This research was supported by National Institutes of
Health Grants GM-54896 and GM-61587.
27. Zapp, M. L.; Young, D. W.; Kumar, A.; Singh, R.; Boy-
kin, D. W.; Wilson, W. D.; Green, M. R. Bioorg. Med. Chem.
1997, 5, 1149.
28. Li, K.; Xiao, G.; Rigl, T.; Kumar, A.; Boykin, D. W.;
Wilson, W. D. In Structure, Motion, Interaction and Expres-
sion of Biological Macromolecules; Sarma, R. H., Sarma, M.
H., Eds.; Adenine: 1998; 137±145.
29. Pritchard, C. E.; Grasby, J. A.; Hamy, F.; Zacharek, A. M.;
Singh, M.; Karn, J.; Gait, M. J. Nucl. Acids Res. 1994, 22, 2592.
30. Tan, R.; Chen, L.; Buettner, D. H.; Frankel, A. D. Cell
1997, 73, 1031.
References
1. Tinoco, I. J.; Bustamanta, C. J. Mol. Biol. 1999, 293, 271.
2. Gesteland, R. F.; Atkins, J. F., Eds. The RNA World II;
Cold Spring Harbor Laboratory: Plainview, NY, 1993; 1±630.
3. Niedle, S., Eds. Oxford Handbook of Nucleic Acid Struc-
ture; Oxford University Press, Oxford, 1999; p 608.
4. Draper, D. E. J. Mol. Biol. 1999, 293, 255.
5. Wong, C.-H.; Hendrix, M.; Manning, D. D.; Rosenbohm,
C.; Greenberg, W. A. J. Am. Chem. Soc. 1998, 120, 8319.
6. Hendrix, M.; Priestley, E. S.; Joyce, G. F.; Wong, C.-H.
J. Am. Chem. Soc. 1997, 119, 3641.
7. Sucheck, S. J.; Wong, A. L.; Koeller, K. M.; Boehr, D. D.;
Draker, K.-A.; Sears, P.; Wright, G. J.; Wong, C.-H. J. Am.
Chem. Soc. 2000, 122, 5230.
31. Peterson, R. D.; Bartel, D. P.; Szostak, J. W.; Horvath,
S. J.; Feigon, J. Biochemistry 1994, 33, 5357.
32. Li, K.; Fernandex-Saiz, M.; Rigl, C. T.; Kumar, A.; Ragu-
nathan, K. G.; McConnaughie, A. W.; Boykin, D. W.; Schnei-
der, H.-J.; Wilson, W. D. Bioorg. Med. Chem. 1997, 5, 1157.
33. Rigl, C. T.; Lioyd, D. H.; Tsou, D. S.; Gryaznov, S. M.;
Wilson, W. D. Biochemistry 1997, 36, 650.
8. Sucheck, S. J.; Greenberg, W. A.; Tolbert, T. J.; Wong, C.-H.
Angew. Chem., Int. Ed. 2000, 39, 1080.
9. Chow, C. S.; Bogdan, F. M. Chem. Rev. 1997, 97, 1489.
10. Pearson, N. D.; Prescott, C. D. Chem. Biol. 1997, 4, 409.
11. Hermann, T.; Westhof, E. Curr. Opin. Biotech. 1998, 9, 66.
12. Tor, Y. Angew. Chem. 1999, 38, 1579.
34. Long, K. S.; Crothers, D. M. Biochemistry 1995, 34, 8885.
35. Luedtke, N. W.; Tor, Y. Angew. Chem. Int. Ed. 2000, 39,
1788.
36. Das, B. P.; Boykin, D. W. J. Med. Chem. 1977, 20, 1219.
37. Bajic, M.; Kumar, A.; Boykin, D. W. Heterocyclic Comm.
1996, 2, 135.
13. Hermann, T.; Westhof, E. J. Med. Chem. 1999, 42, 1250.
14. Wilson, W. D.; Li, K. Curr. Med. Chem. 2000, 7, 73.
15. Puglisi, E. V.; Puglisi, J. D. In RNA Structure and
Function; Simons, R. W. Grunberg-Manago, M., Eds, Cold
Spring Harbor Press, Cold Spring Harbor 1998, p 117.
16. Lynch, S. R.; Recht, M. I.; Puglisi, J. D. Meth. Enz. 1999,
317, 240.
38. Kumar, A.; Boykin, W. D. Syn. Commun. 1995, 25, 2071.
39. Kumar, A.; Boykin, D. W.; Wilson, W. D. Eur. J. Med.
Chem. 1996, 31, 767.
40. Das, B. P.; Wallace, R. A.; Boykin, D. W. J. Med. Chem.
1980, 23, 578.
41. Wilson, W. D.; Tanious, F. A.; Fernandez-Saiz, M.; Rigl,
C. T. Methods Mol. Med. 1997, 90, 219.
17. Schroeder, R.; nov Ahsen, U. Nucl. Acids Mol. Biol. 1996,
10, 53.
18. Hermann, T.; Westhof, E. J. Mol. Biol. 1998, 276, 903.
42. Daly, T. J.; Doten, R. C.; Rennert, P.; Auer, M.; Jaksche,
H.; Donner, A.; Fisk, G.; Rusche, J. R. Biochemistry 1993, 32,
10497.