Design, synthesis, and evaluation of novel biarylpyrimidines: A new class of ligand for unusual nucleic acid structures

Article abstract of DOI:10.1021/jm060315a

Biarylpyrimidines are characterized as selective ligands for higher-order nucleic acid structures. A concise and efficient synthesis has been devised incorporating Suzuki biaryl cross-coupling of dihalopyrimidines. Two ligand series are described based on

Full text of DOI:10.1021/jm060315a

J. Med. Chem. 2006, 49, 5187-5198
5187
Design, Synthesis, and Evaluation of Novel Biarylpyrimidines: A New Class of Ligand for
Unusual Nucleic Acid Structures
Richard T. Wheelhouse,*,† Sharon A. Jennings,† Victoria A. Phillips,† Dimitrios Pletsas,† Peter M. Murphy,†
Nichola C. Garbett,‡ Jonathan B. Chaires,‡ and Terence C. Jenkins§
School of Pharmacy, UniVersity of Bradford, Bradford, West Yorkshire BD7 1DP, U.K., James Graham Brown Cancer Center,
UniVersity of LouisVille, LouisVille, Kentucky 40202, and MorVus Technology Limited, Building 115, Porton Down Science Park,
Salisbury, Wiltshire SP4 0JQ, U.K.
ReceiVed March 20, 2006
Biarylpyrimidines are characterized as selective ligands for higher-order nucleic acid structures. A concise
and efficient synthesis has been devised incorporating Suzuki biaryl cross-coupling of dihalopyrimidines.
Two ligand series are described based on the parent thioether 4,6-bis[4-[[2-(dimethylamino)ethyl]mercapto]-
phenyl]pyrimidine (1a) and amide 4,6-bis(4[(2-(dimethylamino)ethyl)carboxamido]phenyl)pyrimidine (2a)
compounds. In UV thermal denaturation studies with the poly(dA)‚[poly(dT)]2 triplex structure, thioethers
showed stabilization of the triplex form (∆Tm e 20 °C). In contrast, amides showed duplex stabilization
(∆Tm e 15 °C) and either negligible stabilization or specific destabilization (∆Tm ) -5 °C) of the triplex
structure. Full spectra of nucleic acid binding preferences were determined by competition dialysis. The
strongest interacting thioether bound preferentially to the poly(dA)‚[poly(dT)]2 triplex, Kapp ) 1.6 × 105
M-1 (40 × Kapp for CT DNA duplex). In contrast, the strongest binding amide selected the (T2G20T2)4
quadruplex structure, Kapp ) 0.31 × 105 M-1 (6.5 × Kapp for CT DNA duplex).
Introduction
Results and Discussion
Ligand Design. 4,6-Biarylpyrimidine 1a is planar in its
crystal form and known to behave, in part, as a modest duplex
DNA intercalator.19 However, steric crowding inhibits adoption
of equivalent planar conformations in the analogous triphenyl
ring system. Energy profiles were determined, using semi-
empirical molecular modeling, for rotation about one inter-
annular bond of 1,3-diphenylbenzene (“triphenyl”) and a range
of substituted biarylpyrimidines (see Supporting Information).
In triphenyl, energy maxima occur when the molecule ap-
proaches a planar conformation, i.e., torsion angle θ ) 0°, 180°,
etc. These energy barriers are lowered by 1.1 kcal mol-1 in the
analogous biarylpyrimidine system because of relief of steric
interactions around the interannular bond by introduction of the
two nitrogen atoms in the pyrimidine ring. Such stabilization
is largely unaffected by the nature of the para substituent on
each of the phenyl rings. Thus, under the influence of an external
stacking constraint (e.g., crystal packing or DNA intercalation)
the near-planar conformation may become optimal for the
unfused tricyclic system.19
Examination of the crystal structure of biarylpyrimidine 1a
shows the long dimension of the putatively planar, intercalating
core portion of the molecule (13.2 Å) to be a closer match to
the larger threading distances presented by quadruplex (10.8-
13.0 Å)20 or triplex (7.0-12.4 Å)21,22 nucleic acid structures
rather than to DNA duplexes that offer distances of 5.2-7.3 Å,
depending on sequence (Figure 1). It was also envisaged that
the potentially planar chromophore could be extended by
substitution of amide groups for the thioethers (i.e., 1 f 2) and
to thereby further direct nucleic acid-binding preferences toward
higher-order structures. On this basis two distinct series of
ligands based on thioether and amide linkages (1 and 2,
respectively) bearing ω-aminoalkyl side chains were identified
for synthesis and further investigation.
Following the completion of the human genome project and
its conclusion that a relatively small number (30 000) of genes
“encode” a human being, it now appears that the definition and
functional control of the whole organism are determined not
by the simple on/off expression of genes but by the dynamic
interplay of gene expression over time.1 This offers the prospect
of a fundamentally new medicine based on regulating gene
activity rather than altering the behavior of proteins. The targets
for drug design will be nucleic acids: not only B-form DNA
sequences but also specific and unusual structures, such as the
high-order DNA triplexes and quadruplexes (tetraplexes),2-4
DNA‚RNA hybrids,5 and specific RNA secondary structures.6
A tantalizing glimpse of this new chemotherapy is seen from
evidence that (i) the c-myc oncogene promoter sequence
includes a G-rich tract that is able to adopt a quadruplex
structure7 and (ii) treatment of tumor cells with a quadruplex
DNA-binding ligand alters the expression of both this oncogene
and other genes under its control.8 Moreover, putative quadru-
plex-forming sequences appear to be common in the promoter
regions of growth control genes.9-16
Ligands with precisely defined selectivity for nucleic acid
structure therefore have a range of potential therapeutic ap-
plications, while the need to avoid undesirable cytotoxic effects
dictates that competing affinity for duplex structures must be
eliminated in any design program.17 Herein, we report fully18
the design and synthesis of biarylpyrimidines and describe
investigations of their intriguing and selective binding prefer-
ences toward higher-order DNA secondary structures. Structural
factors that either direct binding selectivity toward quadruplex
and triplex DNA or induce specific triplex destabilization have
been identified.
* To whom correspondence should be addressed. Phone: +44 (0) 1274
234710. Fax: +44 (0) 1274 235600. E-mail: r.t.wheelhouse@brad.ac.uk.
† University of Bradford.
Synthesis. To generate a structurally diverse library of
compounds, a concise, high-yielding synthetic route was
required with the potential for adaptation to solid-phase methods.
‡ University of Louisville.
§ Morvus Technology Limited.
10.1021/jm060315a CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/21/2006

5188 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
Figure 1. Comparison of the dimensions of thioether 1a (top, left) and amide 2a (top, right) with the cross-sections of base plane arrangements
in duplex, triplex, and quadruplex DNA structures.
Scheme 1a
a Reagents and conditions: (i) Pd(Ph3P)4, K2CO3, PhMe/EtOH 9:1, ∆, N2; (ii) NaOH, ∆, then HCl; (iii) NaSEt, DMF, 80 °C, 48 h, 62%; (iv) NaSEt,
DMF, ∆, e56%; (v) NaSEt, DMF, room temp, 22h; (vi) NaSBut, DMF, 70 °C, 18 h; (vii) RSCl, AcOH, 50 °C, 24 h; (viii) NaBH4, EtOH/H2O 100:15, pH
8-9, 30 min, 72%; (ix) Z(CH2)2Cl, NaOH, EtOH, e87%, then HBr; (x) Z(CH2)2Cl, K2CO3, DMF, room temp, e76%, then HBr; (xi) Z(CH2)2NH2, PyBOP,
Et3N, CH2Cl2, room temp, then HBr, e70% yield; (xii) Z(CH2)2NH2, ∆, e88%, then HBr.
Biarylpyrimidines have hitherto been prepared either by con-
densation of 1,3-diketones with formamide23,24 or by sequential
aryllithium additions to pyrimidine.19 Previous success with
heterocyclic examples of the Suzuki biaryl cross-coupling
reaction25,26 suggested a superior assembly of the biarylpyrim-
idine core by Suzuki arylation of a dihalopyrimidine,18,27-32 with
side chain diversity introduced in subsequent steps. The flexible
synthetic protocol is outlined in the Scheme 1. Briefly, 4,6-
diiodopyrimidine33 was reacted with arylboronic acids in the
presence of Pd(Ph3P)4 to afford substituted biarylpyrimidines
3a-d directly in up to 90% yield. Since reaction of bromo- or
iodoarenes is known to give superior yields in carbocyclic
examples of Suzuki reactions, commercially available 4,6-
dichloropyrimidine was converted to 4,6-diiodopyrimidine
before arylation. However, subsequent experiments showed that
in the heterocyclic cases, a chlorine atom positioned ortho to
the ring nitrogen was sufficiently activating to give good yields
so that conversion to the iodide was not always essential for
efficient synthesis.31,34
Access to the thioether-linked ligands 1 was achieved through
nucleophilic substitution of appropriate 2-chloroethylamine
derivatives by thiophenol 3j. The 4-halobiarylpyrimidines 3b

Ligands for Unusual Nucleic Acids
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5189
Table 1. ∆Tm Values (°C)a Determined for CT DNA and
Poly(dA)‚[Poly(dT)]2 Triplex DNA in the Presence of Biarylpyrimidine
Ligands (All Data Are Mean Values ( 0.1 °C)
a 50 µM DNA (in base pairs), 20 µM ligand, 10 mM Na2HPO4, NaH2PO4,
1 mM Na2EDTA, pH 7.00 ( 0.01; Tm for CT DNA ) 68.3 °C. b 50 µM
triplex (in base triads), 25 µM ligand, 10 mM sodium cacodylate, 300 mM
NaCl, 0.1 mM Na2EDTA, pH 6.00 ( 0.01; Tm1 ) 54.2 °C (triplex f duplex
2
+ single strand) and Tm ) 76.8 °C (melt to single strands). c ∆Tm
)
Tm(DNA + ligand) - Tm(DNA). d ∆Tm20/∆Tm80 ) ratio of ∆Tm at 20% to
∆Tm at 80% of the normalized absorbance change during melting. e Ligand
destabilizes the DNA triplex.
and 3d were good substrates for nucleophilic aryl substitution
reactions, the electron-withdrawing substituent effect of the
pyrimidine on the phenyl rings being equivalent to that of a
nitro group; the observed chemical shifts of the ring protons
(δH ) 8.02, 7.67 ppm for 3b) were thus similar to those of
4-bromonitrobenzene (8.02, 7.66 ppm).35 Although thiophenol
3j could be obtained by nucleophilic displacement of the
bromoarene 3b by treatment with excess NaSEt,18,36,37 reaction
conditions for cleavage of the intermediate thioether 3i were
harsh and led to variable isolated yields. Switching to the
fluoroarene 3d gave the anticipated rate enhancements in the
nucleophilic substitution reaction (complete consumption by
NaSEt in DMF in 22 h at room temperature compared with 48
h at 80 °C). However, little reaction was evident on treatment
of 3d with a wide range of sulfur, oxygen, or nitrogen
nucleophiles, even in the presence of DBU; good yields were
obtained only with sodium alkanethiolates. Significantly, use
of the 4-fluoro substrate 3d facilitated introduction of the tert-
butylthioether 3e, for which alternative methods are available
to effect cleavage to the thiophenol. Disappointingly, reaction
with Br2 in the presence of acetyl bromide yielded an intractable
disulfide polymer rather than the thioacetate,38 even upon
addition of DMAP. Cleavage with sulfinyl chlorides efficiently
converted the tert-butylthioether to the mixed disulfides 3f and
3g (78% and 91% yield, respectively).39-41 These stable
crystalline derivatives were reduced to the free thiophenol 3j
immediately prior to use.
Figure 2. Concentration dependence of melting behaviors of the
poly(dA)‚[poly(dT)]2 triplex in the absence and presence of example
compounds from the (A) thioether and (B) amide ligand series: (solid
1
lines) ∆Tm (triplex melt f duplex + single strand); (dashed lines)
∆Tm2 (duplex melt f single strands). The [ligand]/[DNA] ratio refers
to the initial concentrations of each component at low temperatures,
using a fixed DNA triplex concentration (50 µM in base triads).
all proved unsuccessful, even with addition of HOBt and NHS
to the carbodiimide reactions. High yields were achieved using
the peptide-coupling agent PyBOP, where the almost-pure amide
precipitate could often be obtained by simple filtration of the
reaction mixture. Two series of analogues, thioethers 1a-f and
amides 2a-g, were prepared. All compounds bearing basic side
chains were isolated as their hydrobromide acid addition salts.
UV Spectrophotometric Thermal Melting Studies. Three
nucleic acid systems were used for studies of thermal denatur-
ation (melting):42,43 calf thymus (CT) DNA, poly(dA)‚poly(dT)
duplex, and the poly(dA)‚[poly(dT)]2 triplex. Collated ∆Tm data
are presented in Figures 2 and 3 and Table 1. In the absence of
bound ligands, CT DNA showed a main transition at Tm ) 68.3
°C. A concentration dependence study indicated that a mole
ratio of 50 µM (base pair) CT DNA/20 µM ligand gave ∼80%
maximal stabilization with the parent ligands 1a and 2a. All of
the cationic ligands examined effected thermal stabilization of
The amide-linked derivatives 2 could be obtained by direct
aminolysis of ester 3a, but greatly improved yields and easier
purifications were achieved following hydrolysis to the free acid
3h and subsequent amide formation. The utility of various
activated carboxylic acid equivalents was investigated. Coupling
reactions using acid chlorides, TFP esters, DCC, EDC, and CDI

5190 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
Figure 3. UV melting curves for poly(dA)‚[poly(dT)]2 triplex DNA (fixed at 50 µM in base triplets) in the absence and presence of (A) thioether
ligands 1 or (B) amide ligands 2. Concentrations: DNA, 50 µM in base triplets; ligand, 25 µM. A fixed ligand concentration of 25 µM was used
with optical monitoring at 260 nm.
this DNA duplex, with induced ∆Tm of 10-15 °C. The
stabilization was independent of the functional group (thioether
or amide) linking the core aryl system to the side chain.
Furthermore, the shifted curves were not parallel to that obtained
for DNA alone and the effect upon the melting profile provides
an indicator of sequence binding preference of the ligand. For
the normalized A260 (i.e., 0-100%) vs temperature curve, the
shape of the early part of the melting curve is controlled by
A-T base-pair melting; G-C base pairs melt in the higher-
temperature range.44 A semiquantitative impression of the excess
of A-T over G-C stabilization can be obtained by comparing
the ∆Tm values at 20% and 80% of the normalized absorbance
change during melting (denoted ∆Tm20 and ∆Tm80, respectively).
On the basis of concentration dependence studies, a fixed
ratio of [ligand]/[base triplets (bt)] of (25 µM):(50 µM) was
used for comparative analysis of the two ligand series (see
Figure 2 and Table 1). Triplex stabilization was common to all
thioether ligands bearing ω-amino side chains (1a,b,f), dem-
onstrating that the charged ammonium group in the side chain
is essential for triplex stabilization. In contrast, the hydroxyethyl
(1c) and methoxyethyl (1d) derivatives showed little evidence
for interaction with either triplex- or duplex-form DNA, despite
the possibilities for greater hydrogen-bonded DNA-ligand
interactions upon binding. The corresponding amides (2c,d)
showed similarly negligible effects. The extent of triplex
stabilization is sensitive to the nature of the basic side chain,
with the bulkier and more hydrophobic groups being more
80
This factor is represented by the ∆Tm20/∆Tm ratio (Table 1).
1
For ligands in both compound series where significant duplex
effective (e.g., piperidinyl 1f, ∆Tm ) 17.4 °C; diethylamino
80
stabilization was shown (∆Tm > 10 °C), the ∆Tm20/∆Tm
1b, ∆Tm1 ) 20.2 °C) although there is no apparent correlation
with the pKa of the basic amino group (NMe2 1a < morpholino
1e < piperidino 1f < NEt2 1b).
parameter was greater than 1.0, indicating preferential stabiliza-
tion of A-T rich sequences in the CT duplex. This differential
behavior was significantly more marked for the amide com-
Minor structural alteration to the amides 2 resulted in an
entirely different pattern of ∆Tm values (Figure 3B). In general,
the simple dialkylamino compounds (e.g., 2a,b) induced a
moderate thermal stabilization of the poly(dA)‚poly(dT) duplex
80
pounds 2a,b,f,g (Tm20/∆Tm ) 1.21-1.45) than for the thio-
80
ethers 1a,b,f (Tm20/∆Tm ) 1.10-1.24).
The melting curve for poly(dA)‚[poly(dT)]2 triplex DNA in
2
1
(∆Tm e 8.1 °C) with little effect on triplex melting (∆Tm
e
the absence of ligand shows two sequential thermal transi-
1
1 °C, except for 2b). Remarkable structure-selective differences
are evident for amide ligands bearing heterocyclic side chains
(2e-g). Ligands 2f and 2g showed closely similar effects, with
tions: Tm ) 54.3 °C (triplex melt f duplex + single strand)
and Tm2 ) 77.8 °C (duplex melt f all single strands).42,43 The
effects of examined ligands on the melting behavior of the poly-
(dA)‚[poly(dT)]2 triplex are shown in Figure 3. The concentra-
tion-dependent effects of selected ligands from each series (1b,
2a) on the two melting transitions are shown in Figure 2.
Evidence for selective interactions with high-order structures
is apparent, with demonstration that the selectivity can be
directed by modifying the ligand structure from thioether to
amide. Thus, thioether 1b effected a dramatic and concentration-
dependent stabilization of the triplex melting transition up to a
2
stabilization of the poly(dA)‚poly(dT) duplex (∆Tm ≈ 14 °C)
1
and a specific destabilization of the triplex structure (∆Tm
≈
-5.3 °C) that preserves the shape of the biphasic melting curve
(Figure 3B). This behavior indicates that the ligands show
sufficiently high affinity for duplex-form poly(dA)‚poly(dT) to
competitively displace the third strand from its corresponding
DNA triplex structure.
A spectrophotometric titration study showed marked hypo-
chromicity and sharp isosbestic points as poly(dA)‚poly(dT)
duplex was titrated into a solution containing either thioether
1a (Figure 4A) or amide 2a (Figure 4B). The results are
consistent with the binding-induced ∆Tm and curve shifts
observed with CT DNA. However, a dramatic difference was
seen upon addition of poly(dA)‚[poly(dT)]2 triplex to either
1
maximum ∆Tm ≈ 24 °C, whereas the duplex melting event
was little affected (∆Tm2 e 3 °C). In the case of amide 2a, the
effects on the transitions are reversed and the magnitude of the
effects is also much smaller: the triplex to duplex transition is
1
unaffected (∆Tm < 2 °C), while the duplex to single strands
2
melt was increased with a maximum ∆Tm ≈ 9 °C.

Ligands for Unusual Nucleic Acids
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5191
Figure 4. UV-visible absorbance spectra for the titration addition of poly(dA)‚poly(dT) duplex (A, B) and poly(dA)‚[poly(dT)]2 triplex (C, D)
into ligand solutions (fixed at 25 µM) containing either thioether 1a (A, C) or amide 2a (B, D). Spectral changes are indicated for equilibrated
DNA-drug mixtures containing 0, 1, 2, 5, 10, 15, 20, 25, and 50 µM (bp) duplex or 0, 1, 2, 5, 7.5, 10, 12.5, 15, 20, and 25 µM (bt) triplex.
thioether 1a (Figure 4C) or amide 2a (Figure 4D), where the
thioether again produced sharp isosbestic points and hypochro-
micity indicative of interaction between the ligand and DNA
triplex in a discrete binding mode. In contrast, no such
interactions were apparent for amide 2a with this triplex, again
mirroring the effects seen in denaturation experiments with CT
DNA. The structural basis for these differential behaviors and
for the selective destabilization of triplex DNA is uncertain and
the subject of continuing investigations.
Competition Equilibrium Dialysis: Nucleic Acid Structure
Binding Preferences. This competitive assay is an important
tool in discerning ligand-nucleic acid interactions and binding
preferences for nucleic acid structures.45-48 Briefly, a common
pool of test ligand is dialyzed against an array of competing
nucleic acid structures in a single experiment. At equilibrium,
the free ligand concentration will be the same for each dialysis
compartment and the amount bound to each structure indicates
the binding affinity. Thus, binding constants and energies can
be obtained in a direct and quantitative manner. Data obtained
for a range of biarylpyrimidines from the thioether (1) and amide
(2) series dialyzed against a panel of 18 nucleic acid structures
are presented in Figure 5; derived thermodynamic parameters
are collected in Table 2.49
These results broadly confirm that neither ligand class shows
a strong affinity for any nucleic acid duplex or single-stranded
structure examined in the assay, irrespective of either sequence
or backbone involved. As anticipated, preferential interactions
were instead evident for high-order nucleic acid structures. The
simple thioethers 1a,b showed a strong preference for triplex
DNA (27-fold greater than for the natural duplex, calf thymus
(CT) DNA). Significantly, less ligand is bound to triplex RNA,
indicating that at least part of the bound ligand (probably the
pendent side chains) must align in the grooves of the DNA
triplex, an accommodation thwarted by the 2′-hydroxyl groups
in the all-RNA structure.21,22 In contrast, the amides 2 bind
preferentially to quadruplex DNA structures but are less
discriminating in their overall recognition profiles. Neutral
compounds (Z ) OH or OMe) showed reduced affinity for
nucleic acids, but the structure preferences generally resemble
those of their dimethylamino counterparts. In both series,
variation of the simple dialkylamino groups only influenced
binding affinity (e.g., Z ) NMe2 > NEt2) while the overall
selectivity was unaffected. However, compounds bearing satu-
rated heterocyclic side chains show different patterns of binding
preferences that are also reflected in their anomalous thermal
melting behavior.
Results from selected quantitative analysis of the competition
equilibrium dialysis data using established methods50 are
presented in Table 2. Apparent binding constants (Kapp) and free
energies of binding (∆G) are presented for thioethers with their
preferred host poly(dA)‚[poly(dT)]2 triplex nucleic acid structure
and for amides with the [d(T2G20T2)]4 parallel-stranded DNA
quadruplex. Data for the natural CT DNA duplex are included
for comparison. Parameter errors have previously been estimated
as 10-15%.42 Two further analytical descriptors can be used
to quantitate binding selectivity: (i) the specificity sum (SS)
given by summation of the normalized amounts bound to each
DNA structure and (ii) the ratio Cmax/SS. In the present assay
system, SS would have a value between 1 (i.e., perfect selectivity
for a single structure) and 18, where the latter value requires
that equal amounts are bound to all of the 18 competing
structures (i.e., a complete absence of selectivity). The Cmax
/
SS parameter provides a simple indicator for both binding
affinity and inherent selectivity, where Cmax is the largest amount
of ligand bound to any structure in the assayed nucleic acid
panel. From the experimental results of competition equilibrium
dialysis with 126 compounds from a diverse range of chemical
classes and an alternative array of 13 nucleic acid structures,

5192 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
Figure 5. Competition equilibrium dialysis data for compounds 1a-f (left) and 2a-c (right). The vertical columns indicate the relative binding
of each candidate ligand to the nucleic acid system. For a description of the nucleic acid structures used, see Table S1 (Supporting Information).
Table 2. Thermodynamic Parameters Derived from Competition Equilibrium Dialysis Data for Test Ligands against a Panel of 18 Nucleic Acid
Structures
Cb
µM
triplexa
duplexb
∆Gtriplex
∆GCTDNA
∆∆Gtriplex,CT
∆∆Gtriplex,1a
ligand
Kapp, 105 M-1
Kapp, 105 M-1
kcal mol-1
kcal mol-1
SS
Cmax/SS
kcal mol-1
kcal mol-1
1a
1b
1c
1d
1e
1f
10.4
6.4
2.9
0.8
7.4
5.8
4.0
1.60
0.93
0.41
0.11
1.09
0.84
0.57
0.04
0.03
0.08
0.07
0.24
0.06
0.19
-5.66
-5.34
-4.86
-4.10
-5.43
-5.28
-5.05
-3.54
-3.32
-3.91
-3.82
-4.55
-3.77
-4.42
3.35
3.21
3.85
4.88
5.46
4.80
8.73
3.09
1.99
0.76
0.51
1.35
1.21
0.46
-2.12
-2.02
-0.95
-0.29
-0.89
-1.51
-0.64
0.00
0.32
0.80
1.55
0.22
0.38
0.61
3j
ligand
Cb
µM
quadruplexc
duplex
∆Gquadruplex
∆GCTDNA
SS
Cmax/SS
∆∆Gquadruplex,CT
∆∆Gquadruplex,2a
Kapp, 105 M-1
K
app, 105 M-1
kcal mol-1
kcal mol-1
kcal mol-1
kcal mol-1
2a
2b
2c
10.0
12.5
3.0
1.55
2.00
0.42
0.39
0.31
0.11
-5.64
-5.79
-4.87
-4.83
-4.70
-4.10
5.81
4.05
5.12
1.73
3.09
0.59
-0.8
-1.1
-0.8
0.00
-0.15
0.76
a Poly(dA)‚[poly(dT)]2. b CT DNA. c [d(T2G20T2)]4.
C
max/SS was found42 to range from 0.06 to 9.8 with a mean
between the two ligands is more pronounced when the data are
compared for binding affinity and specificity, with Cmax/SS )
3.09 and 1.99 for 1a and 1b, respectively. The marked difference
in behavior for the diethylamine 1b and piperidinyl 1c (SS )
4.80, Cmax/SS ) 1.21) derivatives highlights the effect of side
chain flexibility on binding such that the more rigid piperidinyl
compound has both reduced affinity and discrimination in its
binding interactions.
In contrast to the thioethers, the preferred nucleic acid targets
for the amide ligands are the three quadruplex DNA forms:
the folded human telomeric intramolecular d[AG3(T2AG3)3]
quadruplex, the intermolecular G-wire [d(G10T4G10)]n structure,
and the parallel-stranded [d(T2G20T2)]4 arrangement. While
value of 2.4 ( 2.2.
The highest-affinity triplex-binding ligand (thioether 1a) has
Kapp ) 1.60 × 105 M-1 for the triplex compared with Kapp
)
/
0.04 × 105 M-1 for the CT DNA duplex. This Kapp(triplex)
Kapp(CT DNA) ) 40 ratio corresponds to ∆∆G ≈ -2 kcal mol-1
between the two systems. This energy difference indicates a
significant discrimination between the host DNA structures
shown by this ligand. However, the overall specificity for ligand
1a (SS ) 3.35) is less impressive, indicating that other
structures, principally those rich in dA‚dT base pairs, are also
favored for binding. The diethyl congener 1b was marginally
more selective in its interactions (SS ) 3.21). The difference

Ligands for Unusual Nucleic Acids
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5193
available TRAP assay53 (see Supporting Information). As
expected for quadruplex-binding ligands, moderate inhibition
of human telomerase function was found, where the lowest
telIC50 ≈ 10 µM activity compares favorably with that for
porphyrin and perylene (telIC50 g 10 µM)54-56 inhibitors and
the internal reference inhibitor used in this assay (BSU-1021,
telIC50 ) 4.9 µM)57,58 (see Supporting Information). The 2b >
2a ∼ 1f > 1a > 1d ranking of inhibitory activities determined
for the functionalized biarylpyrimidine compounds broadly
reflects their relative DNA-binding strengths with the human
telomeric quadruplex DNA structure d(AG3[T2AG3]3). Stabiliza-
tion of the folded quadruplex (or G-quadruplex) structure is
proposed to prevent access of the telomerase enzyme to the
unfolded DNA substrate required for processivity in telomeric
3′ extension.
Table 3. Chemosensitivity Data from the NCI Three-Cell-Line
Prescreen
Conclusion
Functionalized 4,6-biarylpyrimidines present a versatile and
intriguing platform for the development of ligands for the
recognition of high-order (triplex and quadruplex) nucleic acid
structures. A new, flexible, and high-yielding synthesis of these
agents has been developed that incorporates a Suzuki cross-
coupling reaction to generate the biarylpyrimidine core structure.
Two distinct classes of candidate ligand were synthesized where
the side chains are linked to the aryl core structure via thioether
(1) or amide (2) functional groups.
Spectrophotometric and competition equilibrium dialysis
assays have revealed interesting patterns of nucleic acid binding
behavior in terms of structural selectivity for the host system.
Further, relatively minor alterations in ligand structure effect a
dramatic switch in binding preference. Significantly, all mem-
bers of both classes of ligand were indifferent binders toward
duplex nucleic acid structures; binding constants Kapp < 4 ×
104 M-1 with the mammalian CT DNA reference were obtained
by the competition equilibrium dialysis method. The cationic
pendent side chains are critically important to both classes of
ligand in conferring a strong binding affinity. Thus, even
substituent groups with a combination of H-bond donors and
acceptors were unable to compensate for the absence of a formal
positive charge.
UV spectrophotometry gave the first indications of discrimi-
nation in the interactions of the test ligands between duplex
and triplex structures. Cationic thioethers showed appreciable
thermal stabilization of the poly(dA)‚[poly(dT)]2 triplex (∆Tm
e 20 °C) while having a negligible effect on the poly(dA)‚
poly(dT) duplex. The converse behavior was evident for the
cationic amides where binding-induced stabilization was greater
for duplex-form DNA rather than for triplex-form DNA. This
structural preference was confirmed by the absence of isosbestic
points and hypochromicity in titration experiments involving
poly(dA)‚[poly(dT)]2 triplex and the cationic amide 2a ligand.
A detailed picture of the nucleic acid binding properties was
obtained by the competition equilibrium dialysis method. This
assay reveals that neither class of ligand binds in a preferential
manner to any single- or double-stranded DNA structure.
Instead, in both cases, preferred binding is shown toward
multistranded triplex or quadruplex structures. The overall
structural preference is dictated by the functional group linking
the side chain, with a switch between three-stranded (thioether,
1) and four-stranded (amide 2) hosts. Within each class, the
overall pattern of binding preferences is similar, with uncharged
substituent moieties showing reduced binding affinity and less
selectivity (i.e., smaller Cmax/SS).
qualitatively similar binding to the three quadruplexes was
evident, in certain cases (2a,b), a modest selectivity (e2-fold)
is seen for the discrete four-stranded [d(T2G20T2)]4 structure used
for comparative analysis in Table 2. For compounds 2a and
2b, Kapp ) 1-2 × 105 M-1 are seen. However, binding
discrimination toward this structure rather than a standard
“random” sequence duplex (CT DNA) is an order of magnitude
lower, with the Kapp(quadruplex)/Kapp(CT DNA) ≈ 4.0 ratio representing
a differential ∆∆G ≈ -1 kcal mol-1. Overall, the diethylamino-
terminated ligand 2b gave the best combination of affinity and
selectivity in the amide series, with Cmax/SS ) 3.09. Signifi-
cantly, for both series, the poorest values of Cmax/SS < 1 are
characteristic of compounds with uncharged side chains.
The primary SAR determinant discerned from these data is
that the fundamental binding preference for high-order DNA
structures is largely directed by the biarylpyrimidine core
structure. The chalcogen-linked compounds, thioether or ether
(data not shown), favor triplex binding, while the amide linkage
switches this preference to quadruplex. Within each class, the
cationic side chains make a significant contribution to binding
affinity while having little effect on the overall patterns of
discrimination observed across the panel of 18 target structures.
The ∆∆HvH enthalpies calculated for the DNA-ligand
complexes relative to the free DNA show clear differences in
thermal stabilization of nucleic acid targets between the thioether
1 and amide 2 series (determined by partial qualitative van’t
Hoff analysis of the UV thermal melting curves; see Supporting
Information) and support the switch in binding preference. Thus,
the rank order for triplex Tm1 stabilization is given by 1a . 2a,
whereas the order is reversed for duplex Tm2 stabilization (i.e.,
2a > 1a). This behavior is reflected in the ∆Tm values (Table
1). The 1a > 2a ranking for stabilization of duplex form of CT
DNA again reflects the Tm stabilization afforded by the ligands,
although the altered AT/GC binding preference here prevents
a full analysis with this “pseudorandom” host DNA duplex
system.
Chemosensitivity and Telomerase Inhibition. Selected
compounds were assessed for in vitro chemosensitivity in the
NCI three-cell-line prescreen (Table 3). In agreement with the
low binding affinities determined for duplex DNA (Kapp e 4 ×
104 M-1), no compound examined showed appreciable cyto-
toxicity toward a range of human tumor cell lines (IC50 g 100
µM). Ligands selective for quadruplex rather than duplex DNA
structures have been identified as inhibitors of telomerase.2,17,51,52
Telomerase inhibition was assessed using the commercially

5194 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
UV DNA melting curves were determined using a Varian-Cary
400 Bio UV/vis spectrophotometer equipped with a Peltier tem-
perature controller. Heating runs were typically performed between
40 and 95 °C, heating at a rate of 1 °C min-1, while the absorbance
at 260 nm was continuously monitored. All melts were performed
in 1 cm path length quartz cells. Solutions of DNA and the candidate
ligand were prepared by direct mixing of a solution containing either
CT DNA or triplex with aliquots from the ligand stock solution.
Melting temperatures were determined from the primary data using
the method of Wilson et al.44
Spectrophotometric Titrations. The experimental procedure
used for equilibrium binding titration experiments has been
published.59 Concentrated stock solutions of ligands 1a and 2a (∼3
mM) were freshly prepared in DMSO and diluted to the required
concentration with either aqueous phosphate or cacodylate buffer
(see above), as required. Stock aqueous solutions containing either
poly(dA)‚poly(dT) or poly(dA)‚[poly(dT)]2 at known high concen-
tration (∼5 mM in bp or bt for the duplex and triplex, respectively)
were similarly prepared in the appropriate buffer system. UV-
visible absorbance spectra were recorded at 25 °C for aqueous
solutions containing the ligand (500 µL of 25 µM) in a quartz
cuvette using a Varian-Cary 400BIO instrument. Serial aliquot
portions of DNA stock solution were then added to generate
working solutions containing the reactants at several defined molar
ratios (see Figure 4). The spectrum was rerecorded at each titer
step following equilibrium stirring at 25 °C for 15 min. Spectra
were similarly recorded for mixtures containing a large known molar
excess of the DNA to determine isosbestic behavior and the
saturation approach to binding equilibrium.
The more rigid saturated heterocyclic side chains (1e,f and
2e-g) have distinct effects upon the overall recognition
properties for the ligands compared with acyclic groups of
similar size (1b, 2b). Thus, in the thioether series, a marked
increase in SS is accompanied by a lower binding affinity. For
the amides (2e-g) a specific and unexpected destabilization of
the triplex structure was discovered.
The molecular basis for a switch in structural binding
preference following alteration in the side chain linkage is not
fully understood. Molecular modeling exercises support the
design rationale that superior π-stacking and drug fits for stacked
G-tetrads (quadruplex) are obtained with the extended amide
linkage rather than the inherently more flexible thioether.
Pendent cationic charges promote this structural preference. In
contrast, for triplex binding by the thioethers, the equilibrium
dialysis data indicate a groove binding contribution to the
interaction. Again, modeling studies support the hypothesis, but
diagnostic biophysical evidence has proved elusive; stepwise
NMR titration of ligand 1a with poly(dA)‚[poly(dT)]2 triplex
(in either direction) resulted in apparently quantitative precipita-
tion of the complex rather than observable chemical shift
perturbations. Detailed investigation of the key nucleic acid
binding interactions and the molecular basis for the observed
effects is a continuing effort, focusing upon structural studies
and calorimetric and thermodynamic analyses with optimized
ligand molecules to dissect the component energetic terms for
the thioether and amide classes.
Competition Dialysis. Competition dialysis experiments were
performed exactly as previously described,45,47,49 albeit with a new
array of nucleic acid structures. The nucleic acids structures used
in the new test array are listed in Table S2 of Supporting
Information. The preparation and tests for the quality of these
nucleic acids are fully described in refs 45 and 49. Competition
dialysis studies were performed in BPES buffer, consisting of 6
mM Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, 185 mM NaCl,
pH 7.0. Test ligand molecules (1 µM concentration, 200 mL
volume) were dialyzed against the nucleic acid array. Each nucleic
acid structure was at 75 µM in a volume of 200 µL. Following 24
h of dialysis, samples were removed and ligand was dissociated
by the addition of SDS to a final concentration of 1% (v/v). The
total ligand concentration present in each nucleic acid sample was
measured spectrophotometrically using microcuvettes at an ab-
sorbance maximum appropriate for each ligand and corrected for
the slight dilution resulting from the addition of the SDS solution.
The free ligand concentration, which typically was not appreciably
different from the initial concentration of 1 µM, was determined
spectrophotometrically from an aliquot of the dialysate solution,
and the concentration of ligand bound to each nucleic acid structure
was determined by difference (Cb ) Ct - Cf).
A subset of compounds was assayed as inhibitors of human
telomerase and a moderate telIC50 ≈ 10 µM was found,
consistent with modest, unoptimized binding (Kapp ) 0.85 ×
105 M-1) to the human telomeric sequence AG3(T2AG3)3.
Significantly, both weak duplex DNA binding and concomitant
low cytotoxicity (IC50 > 100 µM) were determined. These
findings will be important for future development of such agents
as selective modulators of gene expression through interaction
with high-order DNA structures. In summary, the 4,6-biaryl-
pyrimidine ligands are an attractive class of compounds for
diversification and optimization of their nucleic acid recognition
properties. These agents are synthetically accessible, nontoxic,
and not complicated by the photosensitivity problems that
hamper other ligands with extended aromatic chromophores
(e.g., porphyrins, quinones, coralyne) reported for use in
targeting high-order nucleic acid systems.
Experimental Section
UV Thermal Melting Studies. Calf thymus (CT) DNA and
poly[dA]‚poly[dT] and poly[dT] DNA were purchased from Sigma
as the sodium salts and used without further purification. An
aqueous buffer containing 10 mM Na2HPO4/NaH2PO4, 1 mM
Na2EDTA, pH 7.00 ( 0.01, was used for CT DNA melting
experiments. An aqueous buffer containing 10 mM sodium ca-
codylate, 300 mM NaCl, 0.1 mM Na2EDTA, pH 6.00 ( 0.01, was
used for triplex DNA melting experiments. Concentrations of all
DNA solutions were determined by UV spectrophotometry using
the following extinction coefficients: CT DNA, ꢀ260 ) 12 824 (M
(base pairs))-1 cm-1; poly[dA]‚poly[dT], ꢀ260 ) 12 000 (M (base
Synthesis. Boronic acids were purchased from Frontier Scientific,
Europe Ltd., Carnforth, Lancashire, U.K. Other reagents were
purchased from Sigma-Aldrich, Gillingham, U.K., or Lancaster,
Morecambe, U.K. Solvents were purchased from Riedel-de Hae¨n
and, where stated, purified according to Perrin, Perrin, and
Armarego.60 TLC was performed on highly purified silica gel plates
with UV indicator (silica gel 60 F254), manufactured by Merck and
visualized under UV light (254 or 366 nm) or stained with iodine.
Flash chromatography was performed using silica gel (40-63 µm)
purchased from Merck and used following the methodology
described by Still et al.61 Melting points were determined using an
Electrothermal IA9200 digital melting point apparatus. Infrared data
were obtained on a Perkin-Elmer (Paragon 1000) FT-IR spectro-
photometer. Routine nuclear magnetic resonance spectra were
pairs))-1 cm-1; poly[dT], ꢀ265 ) 8700 (M (nucleotides))-1 cm-1
.
The poly[dA]‚poly[dT]2 triplex was formed by mixing equimolar
(i.e., base pairs to nucleotides) solutions of the poly[dA]‚poly[dT]
duplex and the poly[dT] single strand, then annealing by heating
to 95 °C followed by slow cooling overnight.
1
acquired on a JEOL GX270Delta spectrometer (observing H at
Ligands were dissolved in buffer or DMSO, ensuring that the
ultimate concentration of DMSO was e1-2% v/v where used.
DMSO calibration curves were used to correct the Tm values. Ligand
stock solutions were stored in the refrigerator (0-5 °C) when not
in use.
270.17 MHz and 13C at 67.80 MHz). Alternatively, where stated,
1
a JEOL ECA600 spectrometer (observing H at 600.17 MHz and
13C at 150.91 MHz) was used. 13C assignments were made with
the aid of the DEPT135 experiment. Values of J and δ of aryl AB
systems were calculated using the NMRcalc program (T. C. Jenkins,

Ligands for Unusual Nucleic Acids
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5195
1
2001). Mass spectra were obtained from the EPSRC National Mass
Spectrometry Service Centre, University of Wales, Swansea.
Elemental analysis was performed by the Advanced Chemical and
Materials Analysis Unit at the University of Newcastle upon Tyne,
U.K.
13C{1H} NMR (CDCl3, 150.91 MHz) δ 164.8 (d, JCF ) 252.9
Hz, C-4′), 163.7 (C-4, C-6), 159.3 (C-2), 133.1 (d, 4JCF ) 4.3 Hz,
C-1′), 129.3 (d, 3JCF ) 10.1 Hz, C-2′, C-6′), 116.2 (d, 2JCF ) 21.7
Hz, C-3′, C-5′), 112.1 (C5); IR (CH2Cl2) 1600s, 1573s, 1510s,
1450s, 1275s, 850s cm-1. MS (ES+) (%) 269 (100) (M + H)•+,
Anal. (C16H10F2N2) C, H, N.
General Method 1: Suzuki Coupling Reaction. 4,6-Bis(4-
methoxyphenyl)pyrimidine‚0.25H2O, 3c. A solution of 4-meth-
oxyphenylboronic acid (0.9 g, 5.9 mmol), 4,6-diiodopyrimidine,
(0.98 g, 2.9 mmol), palladium(0)tetrakis(triphenylphosphine) (45.6
mg, 0.04 mmol), and K2CO3 (0.58 g, 5.9 mmol) in toluene/EtOH
(9:1 v/v, 200 mL) was thoroughly degassed using a stream of dry
N2, then stirred vigorously under reflux. After 8 h the solution was
concentrated by evaporation under vacuum and the residue parti-
tioned between H2O and ether (50 + 50 mL) and extracted with
ether (3 × 100 mL). The combined organic extracts were dried
over MgSO4 and evaporated. Recrystallization from EtOH afforded
biarylpyrimidine 3c as a white solid (0.52 g, 61%): mp 148-149
4,6-Bis(4-tert-butylmercaptophenyl)pyrimidine, 3e. A slurry
of 4,6-bis-(4-fluorophenyl)pyrimidine 3d (1.26 g, 4.70 mmol) and
sodium tert-butylthiolate (1.16 g, 10.34 mmol) in DMF (25 mL)
was heated at 70 °C for 18 h. Water (50 mL) was added to the
reaction mixture, and the solution was extracted with ether (1 ×
150 mL, 2 × 50 mL). The combined organic extracts were dried
over MgSO4, filtered, and evaporated. Recrystallization from ethanol
gave fine, colorless needles (1.60 g, 84%): mp 151.5-152.5 °C;
1H NMR (CDCl3, 600.17 MHz) δ 9.29 (d, J ) 1.2 Hz, 1H, 2-H),
8.09 (d, J ) 1.2 Hz, 1H, 5-H), 8.08 (1/2AB, J ) 8.4 Hz, 4H, 2′-H,
6′-H), 7.66 (1/2AB, J ) 8.4 Hz, 4H, 3′-H, 5′-H), 1.31 (s, 18H,
C(CH3)3); 13C NMR (CDCl3, 150.91 MHz) δ 163.9 (C-4, C-6),
159.3 (C-2), 137.8 (C-2′, C-6′), 137.2 and 136.6 (C-1′ and C-4′),
127.2 (C-3′, C-5′), 112.9 (C5), 46.6 (CMe3), 31.2 (CH3); IR
(CH2Cl2) 2950m, 1575s, 1505s, 1450s, 1270s cm-1; MS (ES+)
(%) 409 (100) (M + H)•+. Anal. (C24H28N2S2) C, H, N.
4,6-Bis(4-trichloromethyldisulfanylphenyl)pyrimidine, 3f. A
solution of 4,6-bis(4-tert-butylmercaptophenyl)pyrimidine 3e (0.078
g, 0.2 mmol) and CCl3SCl (0.402 g, 2.2 mmol) in acetic acid (8
mL) was heated at 50 °C for 24 h. The mixture was partitioned
between ethyl acetate (100 mL) and water (100 mL). The organic
layer was separated, dried over MgSO4, filtered, and evaporated.
The pale-yellow solid was recrystallized from ethanol (0.088 g,
78%): mp 126.1-127.3 °C; 1H NMR (270.17 MHz, CDCl3) δ 9.31
(d, J ) 1.0 Hz, 1H, 2-H), 8.14 (1/2AB, J ) 8.5 Hz, 4H, 2′-H, 6′-
H), 8.07 (d, J ) 1.0 Hz, 1H, 5-H), 7.80 (1/2AB, J ) 8.5 Hz, 4H,
3′-H, 5′-H), 3.70 and 1.23 (q and t, EtOH); 13C NMR δ 163.8 (C-
4, C-6), 159.2 (C-2), 138.7 and 136.6 (C-1′ and C-4′), 129.2 (C-3′,
C-5′ or C-2′, C-6′), 128.1 (C-2′, C-6′ or C-3′, C-5′), 112.7 (C-5),
SCCl3 not observed; IR (CH2Cl2) 1575s, 1510s, 1450m. 1290m,
1010w, 825s cm-1; MS (ES+) (%) 474 (100) (M + H)•+. Anal.
(C18H10Cl6N2S4‚0.25EtOH) C, H, N.
1
°C; H NMR (CDCl3, 600.17 MHz) δ 9.20 (d, J ) 1.3 Hz, 1H,
2-H), 8.11 (1/2AB, J ) 8.9 Hz, 4H, 2′-H, 6′-H), 7.97 (d, J ) 1.3
Hz, 1H, 5-H), 7.03 (1/2AB, J ) 8.9 Hz, 4H, 3′-H, 5′-H), 3.89 (s,
6H, OCH3); 13C NMR (CDCl3) δ 165.1 (C-4, C-6), 163.4 (C-4′),
159.0 (C-2), 128.7 (1′), 123.1 (C-2′, C-6′), 114.4 (C-3′, C-5′), 111.0
(C-5), 55.4 (CH3); IR (CH2Cl2) 3450w, 1600s, 1590s, 1575s, 1510s,
1405s cm-1; MS (FAB) (%) 293 ([M + H]•+, 100), 152 (17), 107
([C6H4OCH3]•+, 36); UV-vis (10 mM sodium phosphate buffer,
pH 7.00, 1 mM EDTA, 300 mM NaCl) λmax (nm) [ꢀ (M-1 cm-1)]
332 [6026]. Anal. (C18H16N2O2‚0.25H2O) C, H, N.
4,6-Bis(4-methoxycarbonylphenyl)pyrimidine, 3a. 3a was
prepared by general method 1 from 4-methoxycarbonylphenyl-
boronic acid (1.74 g, 9.64 mmol). Solvents were removed by
evaporation, and the product was recrystallized from ethyl acetate
1
to yield 1.53 g (91%); mp 223.5-225.0 °C; H NMR (CDCl3,
600.17 MHz) δ 9.38 (d, J ) 1.4 Hz, 1H, 2-H), 8.21 (AB, 8H, 2′-
H, 3′-H, 5′-H, 6′-H), 8.19 (d, J ) 1.4 Hz, 1H, 5-H), 3.97 (s, 6H,
OCH3); 13C NMR (DMSO) δ 166.1 (CdO), 163.2 (C-4, C-6), 159.4
(C-2), 140.6 (C-1′), 132.0 (C-4′), 129.9 (C-3′, C-5′ or C-2′, C-6′),
127.9 (C-2′, C-6′ or C-3′, C-5′), 114.2 (C-5), 52.9 (OCH3); IR
(CH2Cl2) 1725s (CdO), 1580s, 1460m, 1105m, 1020w cm-1; MS
(EI) (%) 348 (M•+, 87), 317 ([M - OCH3]•+, 100), 289 (82), 203
(26), 101 (49). Anal. (C19H16N2O4) C, H, N.
4,6-Bis[4-(2-nitrophenyldisulfanyl)phenyl]pyrimidine, 3g. A
solution of 4,6-bis(4-tert-butylmercaptophenyl)pyrimidine 3e (3.680
g, 9.0 mmol) and 2-nitrobenzenesulfenyl chloride (6.83 g, 36.0
mmol) in acetic acid (130 mL) was heated at 65 °C for 2 h. The
mixture was cooled to -20 °C, filtered, and washed with ethanol.
The yellow solid was triturated with hot ethanol and filtered hot
4,6-Bis(4-bromophenyl)pyrimidine‚0.25H2O, 3b. Treatment of
diiodopyrimidine (1.92 g, 6.03 mmol) with 4-bromophenylboronic
acid (2.42 g, 12.05 mmol), according to general method 1, afforded
biarylpyrimidine 3b as white crystals from ethanol (1.6 g, 68%):
mp 196.5-197.5 °C; 1H NMR (CDCl3, 600.17 MHz) δ 9.29 (d, J
) 1.1 Hz, 1H, 2-H), 8.04 (d, J ) 1.1 Hz, 1H, 5-H), 8.02 (1/2AB,
J ) 8.6 Hz, 4H, 2′-H, 6′-H), 7.67 (1/2AB, J ) 8.6 Hz, 4H, 3′-H,
5′-H); 13C NMR (CDCl3) δ 164.5 (C-4, C-6), 160.0 (C-2), 137.0
(C-1′), 133.0 (C-2′, C-6′), 129.4 (C-3′, C-5′), 126.6 (C-4′), 112.8
1
(4.35 g, 86%): mp 220-222 °C; H NMR (270.17 MHz, CDCl3)
δ 9.26 (d, J ) 1.0 Hz, 1H, 2-H), 8.58 (d, J ) 1.0 Hz, 1H, 5-H),
8.37 (d, J ) 8.5 Hz, 2H, 3′′-H), 8.34 (1/2AB, J ) 8.5 Hz, 4H, 2′-H,
6′-H), 8.08 (d, J ) 8.5 Hz, 2H, 6′′-H), 7.86 (t, J ) 8.5 Hz, 2H,
4′′-H), 7.72 (1/2AB, J ) 8.5 Hz, 4H, 3′-H, 5′-H), 7.57 (t, J ) 8.5
Hz, 2H, 5′′-H); 13C NMR (CDCl3, 150.91 MHz) δ 163.4 (C-4, C-6),
159.5 (C-2), 145.8 (C-2′′), 138.4 (C-1′), 136.1 (C-4′), 136.0 (C-
3′′), 135.1 (C-4′), 128.9 (C-2′, C-6′), 127.9 (C-3′, C-5′), 128.3, 127.4
and 127.1 (C-4′′, C-5′′, and C-6′′), 112.9 (C-5); IR (KBr) 1585s,
1510s, 1335s, 1300m, 725m cm-1; MS (ES+) (%) 603 (50) (M +
H)•+. Anal. (C28H18N4O4S4‚H2O) C, H, N.
(C-5); IR (CH2Cl2) 1580s, 1520m, 1460m, 1260w, 1075w cm-1
;
MS (CI) (%) 388 + 390 + 392 (M•+ 54%, 1:2:1 isotopic
distribution for 81Br/81Br, 81Br/79Br, 79Br/79Br), 311 + 309 ([M -
Br]•+, 32), 284 + 282 (36), 229 ([M - 2Br]•+, 22), 203 (50), 101
(100). Anal. (C16H10Br2N2‚0.25H2O) C, H, N.
4,6-Bis(4-fluorophenyl)pyrimidine, 3d. By General Method
1. 4-Fluorophenylboronic acid (3.97 g, 27.51 mmol), 4,6-dichloro-
pyrimidine (2.06 g, 13.42 mmol), and finely powdered K2CO3 (5.56
g) in methanol/toluene (1:9, 300 mL) were mixed with strong
agitation and thorough degassing with N2 for 30 min at 60 °C.
Pd(Ph3P)4 (0.3 g) was added, and the mixture was heated to reflux
with strong agitation for 6 h. The solvent was evaporated and the
residue partitioned between chloroform (3 × 150 mL) and water
(200 mL). The organic layers were combined, dried over MgSO4,
filtered, and evaporated to dryness. The residue was recrystallized
from methanol and water (2.93 g, 82%): mp 102.6-103.6 °C (lit.24
88.9-90.6 °C). All NMR data obtained are consistent with the
assigned structure but different from those quoted in the literature.24
4,6-Bis(4-carboxyphenyl)pyrimidine‚0.5H2O, 3h. 4,6-Bis[4-
methoxycarbonylphenyl]pyrimidine 3a (0.67 g, 1.92 mmol) was
added to 2 M NaOH solution (50 mL) and heated under reflux for
12 h. The solution was left to cool and was then acidified (pH 6)
using concentrated HCl. The resultant precipitate was collected by
filtration, washed with water, ethanol, and ether, and then dried to
yield 0.61 g of a white solid (100%): mp > 300 °C (dec); 1H NMR
(DMSO) δ 13.06 (br s, 1H, OH), 9.39 (s, 1H, 2-H), 8.79 (s, 1H,
5-H), 8.51 (1/2AB, J ) 8.4 Hz, 4H, 2′-H, 6′-H), 8.12 (1/2AB, J )
8.4 Hz, 4H, 3′-H, 5′-H); 13C NMR (DMSO) δ 167.3 (CdO), 163.4
(C-4, C-6), 159.3 (C-2), 139.9 (C-1′), 134.1 (C-4′), 130.0 (C-3′,
C-5′), 127.6 (C-2′, C-6′), 113.9 (C-5); IR (KBr) 3020w (OH),
1690m (CO2H) 1100s cm-1; MS (ES) (%) 321 (M + H, 100). Anal.
(C8H12N2O4‚0.5H2O) C, H, N.
1
19F NMR (CDCl3, 564.73 MHz, CFCl3 reference) δ -109.2; H
NMR (CDCl3, 600.17 MHz) δ 9.25 (d, J ) 1.2 Hz, 1H, 2-H), 8.13
3
4
(dd, JHH ) 8.7, JHF ) 5.3 Hz, 4H, 2′-H, 6′-H), 7.98 (d, J ) 1.2
4,6-Bis(4-ethylmercaptophenyl)pyrimidine, 3i. A slurry of 4,6-
bis(4-bromophenyl)pyrimidine 3b (4.84 g, 11.5 mmol) and NaSEt
3
3
Hz, 1H, 5-H), 7.21 (t, JHH ≈ JHF ) 8.7 Hz, 4H, 3′-H, 5′-H);

5196 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
(2.42 g, 28.75 mmol) in DMF (100 mL) was heated under nitrogen
at 100 °C for 48 h. The residue obtained on evaporation of DMF
was boiled with water (75 mL), and the solids were collected by
filtration. Recrystallization from ethanol gave two crops of pale-
yellow, fluffy needles (2.51 g, 62%): mp 109.2-111.2 °C; 1H NMR
(CDCl3, 600.17 MHz) δ 9.25 (d, J ) 1.2 Hz, 1H, 2-H), 8.06
(1/2AB, J ) 8.6 Hz, 4H, 2′-H, 6′-H), 8.02 (d, J ) 1.2 Hz, 1H,
5-H), 7.41 (1/2AB, J ) 8.6 Hz, 3′-H, 5′-H), 1.38 (t, J ) 7.4 Hz,
6H, CH3), 3.04 (q, J ) 7.4 Hz, 4H, SCH2); 13C NMR (CDCl3,
150.91 MHz) δ 164.0 (C-4, C-6), 159.3 (C-2), 141.4 (C-1′), 133.8
(C-4′), 127.8 (C-3′, C-5′ or C-2′, C-6′), 127.5 (C-2′, C-6′ or C-3′,
C-5′), 111.7 (C-5), 26.7 (SCH2), 14.2 (CH3); IR (CH2Cl2) 1600m,
1575s, 1505m, 1450m, 1350w, 1100s, 850m cm-1; MS (ES+) (%)
353 (100) (M + H)•+, 324. 43. Anal. (C20H20N2S2) C, H, N.
4,6-Bis(4-mercaptophenyl)pyrimidine, 3j. NaBH4 (2.0 g, 53.1
mmol) was dissolved in a mixture of ethanol (100 mL) and water
(15 mL), pH 8-9. Disulfide 3g (3.2 g, 5.31 mmol) was then added
over 30 min. The mixture was stirred at room temperature for 2 h.
Solids were removed by filtration through an elite pad, and the
filtrate volume was reduced to a third. The red solution was
partitioned between chloroform (4 × 250 mL) and water (200 mL),
the pH being adjusted slowly with 4 M HCl to pH 4. The organic
layers were combined, dried over MgSO4, filtered, and reduced to
dryness. The resulting residue was triturated with hot hexane and
filtered hot to leave a yellow solid (1.25 g, 79%): mp 183-185
°C; 1H NMR (CDCl3) δ 9.24 (s, 1H, 2-H), 8.01 (s, 1H, 5-H), 7.98
(1/2AB, J ) 8.6 Hz, 4H, 2′-H, 6′-H), 7.39 (1/2AB, J ) 8.6 Hz, 4H,
3′-H, 5′-H), 3.60 (s, 2H, SH); 13C NMR (CDCl3) δ 164.5 (C-4,
C-6), 159.8 (C-2), 136.2 and134.7 (C-4′ and C-1′), 129.8 (C-2′,
C-6′ or C-3′, C-5′), 128.4 (C-3′, C-5′ or C-2′, C-6′), 112.3 (C5);
5′-H), 6.95 (br t, exch. D2O, 2H, CONH), 3.51 (q, J ) 5.8 Hz, 4H,
CONHCH2), 2.53 (t, J ) 5.8 Hz, 4H, CH2NMe2), 2.26 (s, 12H,
CH3); 13C NMR 167.3 (CdO), 164.6 (C-4, C-6), 160.2 (C-2), 140.1
(C-4′), 137.5 (C-1′), 128.4 (C-2′, C-6′ or C3′, C-5′), 128.0 (C3′,
C-5′ or C-2′, C-6′), 113.9 (C-5), 58.3 (CH2NMe2), 45.8 (CH3), 37.9
(CH2NHCO); HRMS (CI) 461.2665 ([M + H]•+), C26H33N6O2
requires 461.2660; MS (CI) 462 ([M + 2H]•+, 32), 461 ([M +
H]•+, 100), 447 ([M - CH3]•+, 6), 416 (9), 390 [M - CH2CH2-
NMe2]•+, 3), 193(8), 60 (50); IR (CH2Cl2) 3683w, 3410w, 3054m,
2951m, 1660s, 1586s, 1524s, 1497s, 1462s, 1290m cm-1. Hydro-
bromide salt: mp 219-220 °C, UV-vis (10 mM sodium phosphate
buffer, pH 7.00, 1 mM EDTA, 300 mM NaCl) λmax (nm) [ꢀ (M-1
cm-1)] 255 [21 670], 289 [18 338]. Anal. (C26H32N6O2‚3HBr‚
1.25H2O) C, H, N.
General Method 3: Preparation of Amides. 4,6-Bis(4[(2-
(diethylamino)ethyl)carboxamido]phenyl)pyrimidine‚2HBr‚
2H2O, 2b. 4,6-Bis(4-carboxyphenyl)pyrimidine (0.30 g, 0.94
mmol), PyBOP (1.22 g, 2.34 mmol), and triethylamine (0.33 mL,
2.34 mmol) were stirred together in dry DCM (50 mL) under N2
for 30 min. N,N-Diethylethylenediamine (0.22 mL, 1.87 mmol) was
then added, and the mixture was stirred for a further 24 h at room
temperature. Either the precipitated crude product was collected
by filtration or solvent was removed by evaporation leaving crude
product, which was taken up in 48% HBr (5 mL) and evaporated
to dryness. Recrystallization from ethanol gave the hydrobromide
salt as a white solid (0.41 g, 61%): mp 193.0-193.5 °C; 1H NMR
(DMSO) δ 9.39 (d, J ) 1.2 Hz, 1H, 2-H), 9.18 (br s, 2H, N+H),
8.92 (br t, J ) 5.9 Hz, 2H, CONH), 8.53 (1/2AB, J ) 8.5 Hz, 4H,
2′-H, 6′-H), 8.11 (1/2AB, J ) 8.5 Hz, 4H, 3′-H, 5′-H), 3.66 (m,
4H, CONHCH2), 3.27 (br m, 12H, N+CH2), 1.24 (t, J ) 7.2 Hz,
12H, CH2CH3); 13C NMR (DMSO) 175.5 (CdO), 173.0 (C-4, C-6),
166.8 (C-2), 142.0 (C-1′), 137.6 (C-4′), 130.5 (C-3′, C-5′), 130.4
(C-2′, C-6′), 118.3 (C-5), 53.2 (CONHCH2), 50.6 (N+CH2CH3),
37.6 (CONHCH2CH2), 10.8 (CH3); IR (KBr) 3300m (NH), 1650s
and 1540s (CONH), 1600s (Ar-CH), 1550s cm-1; MS(CI) (free
base) 517 (10%) (M + H)•+, 221 ([PhCONH(CH2)2NEt2]•+, 15),
100 ([(CH2)2NEt2]•+, 40), 86 ([CH2NEt2]•+, 60), 74 (100), 44 (70).
Hydrobromide salt: UV-vis (10 mM sodium phosphate buffer,
pH 7.00, 1 mM EDTA, 300 mM NaCl) λmax (nm) [ꢀ (M-1 cm-1)]
255 [19 209], 291 [16 192]. Anal. (C30H40N6O2‚2HBr‚2H2O) C, H,
N.
4,6-Bis(4-[(2-(hydroxy)ethyl)carboxamido]phenyl)pyrimidine‚
0.5H2O, 2c. Ester 3a (0.5 g, 1.56 mmol) and ethanolamine (9 mL)
were heated at 120 °C overnight under N2. Evaporation of the excess
ethanolamine left a pale-yellow residue, which was triturated twice
with boiling chloroform, collected by filtration, and dried to give
a white solid (0.46 g, 73%): mp 255-256 °C; 1H NMR (DMSO)
δ 9.36 (br s, 1H, 2-H), 8.77 (br s, 1H, 5-H), 8.67 (t, J ) 5.0 Hz,
2H, exch D2O, CONH), 8.49 (1/2AB, J ) 8.0 Hz, 4H, 2′-H, 6′-H),
8.07 (1/2AB, J ) 8.0 Hz, 4H, 3′-H, 5′-H), 4.82 (t, J ) 5.0 Hz, 2H,
OH), 3.54 (q, J ) 5.0 Hz, 4H, CH2O), 3.37 (q, J ) 5.0 Hz, 4H,
NCH2); 13C NMR (DMSO) δ 165.7(CdO), 163.1 (C-4, C-6), 159.0
(C-2), 138.4 (C-4′), 136.7 (C-1′),127.8 (C-2′, C-6′ or C-3′, C-5′),
127.2 (C-3′, C-5′ or C-2′, C-6′), 113.6 (C-5), 59.7 (OCH2), 42.3
(NCH2); IR (KBr) 3300s (OH, NH), 1650s and 1540s (CONH),
1600m (Ar-CH), 1260m, 1050m, 1025m cm-1; HRMS (CI) found
407.1715 ([M + H]•+), C22H23N4O2 requires 407.1715; MS (CI)
(%) 407 ([M + H]•+, 43), 389 ([M - OH]•+, 7), 69 (100); UV-
vis (10 mM sodium phosphate buffer, pH 7.00, 1 mM EDTA, 300
mM NaCl) λmax (nm) [ꢀ (M-1 cm-1)] 256 [16 454], 297 [17 162].
Anal. (C22H22N4O4‚0.5H2O) C, H, N.
IR (KBr) 3400m, 1620s, 1580s, 1500m, 1375m, 1100s cm-1
;
HRMS (CI) 297.0512 ([M + H]•+), C16H13N2O2 requires 297.0515;
MS (EI) (%) 296 (M•+, 70), 263 ([M - S]•+ (23), 236 (14), 134
(29), 105 (48), 44 (50), 49 (47), 43 (100).
General Method 2: Preparation of Thioethers. 4,6-Bis[4′-
[[2′′-(dimethylamino)ethyl]mercapto]phenyl]pyrimidine‚22/3HBr‚
0.25H2O, 1a. A mixture of bisthiophenol 3j (0.25 g, 0.85 mmol),
2-dimethylaminoethyl chloride hydrochloride (0.37 g, 2.55 mmol),
and K2CO3 (0.29 g, 2.12 mmol) was stirred in dry DMF (5 mL) at
room temperature for 18 h. The reaction mixture was partitioned
between water and ether (25 + 25 mL), and the organic layer was
washed twice with water, dried over MgSO4, filtered, and evapo-
rated to dryness. The crude residue was suspended in dilute HBr
(20 mL, 15% w/v) and stirred for 30 min. The solution was
evaporated to dryness, and the crude solid was recrystallized from
ethanol. Free base: 1H NMR (CDCl3) δ 9.21 (d, J ) 1.1 Hz, 1H,
H-2), 8.03 (1/2AB, J ) 8.4 Hz, 4H, 2′-H, 6′-H), 7.97 (d, J ) 1.1
Hz, 1H, 5-H), 7.39 (1/2AB, J ) 8.4 Hz, 4H, 3′-H, 5′-H), 3.10 (t, J
) 7.0 Hz, 4H, SCH2), 2.59 (t, J ) 7.0 Hz, 4H, CH2N), 2.27 (s,
12H, NCH3); 13C NMR (CDCl3) δ 164.6 (C-4, C-6), 159.9 (C-2),
141.7 (C-4′), 134.6 (C-1′), 128.5 (C-2′, C-6′ or C-3′, C-5′), 128.2
(C-3′, C-5′ or C-2′, C-6′), 12.4 (C-5), 58.9 (SCH2), 46.0 (NCH3),
30.3 (NCH2); IR (CH2Cl2) 3684w, 2948m, 1582s, 1511m, 1497m,
1460s, 1372m cm-1; MS (CI) (%) 439 ([M + H]•+, 100), 396 ([M
- NMe2]•+, 8), 368 (54), 336 ([M - SCH2CH2NMe2]•+, 72).
Hydrobromide salt: mp 258-259 °C; UV-vis (10 mM aqueous
phosphate buffer, pH 7.00, 1 mM EDTA and 300 mM NaCl) λmax
(nm) [ꢀ (M-1 cm-1)] 236 [10 636], 329 [7076]. Anal. (C24H30N4S2‚22/
3HBr‚0.25H2O) C, H, N.
4,6-Bis(4[(2-(dimethylamino)ethyl)carboxamido]phenyl)-
pyrimidine‚3HBr‚1.25H2O, 2a. A solution of ester 3a (0.45 g,
1.28 mmol) in N,N-dimethylethylenediamine (20 mL) was heated
under reflux for 8 h. Excess N,N-dimethylethylenediamine was
removed by evaporation in vacuo, and the residue was purified by
flash column chromatography (hexane/dichloromethane/triethyl-
amine, 45:45:10) to give diamide 2a (0.52 g, 88%) as a brown oil.
Treatment with 48% HBr (1 mL), evaporation to dryness, and
recrystallization from methanol-2-propanol gave the hydrobromide
Acknowledgment. This work was supported by the EPSRC
(Grant GR/N 37605 to R.T.W. and T.C.J.), Yorkshire Cancer
Research (program grant to T.C.J.), National Cancer Institute
(Grant CA35635 to J.B.C.), and an EPSRC CASE studentship
with Enact Pharma PLC.
1
salt as a white solid: mp 219-220 °C; H NMR (CDCl3) δ 9.29
Supporting Information Available: Figures and experimental
information for rotational energy barrier determination; synthesis
procedure for compounds 1b-f and 2d-g; table of calculated van’t
(d, J ) 1.3 Hz, 1H, 2-H), 8.16 (1/2AB, J ) 8.7 Hz, 4H, 2′-H, 6′-
H), 8.09 (d, J ) 1.3 Hz, 5-H), 7.91 (1/2AB, J ) 8.7 Hz, 4H, 3′-H,

Ligands for Unusual Nucleic Acids
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5197
(25) Shi, D.-F.; Wheelhouse, R. T. A Novel, High-Yielding Synthesis of
meso-Substituted Porphyrins. Tetrahedron Lett. 2002, 43, 9341-
9342.
(26) Shi, D.-F.; Wheelhouse, R. T.; Sun, D.; Hurley, L. H. Quadruplex-
Interactive Agents as Telomerase Inhibitors: Synthesis of Porphyrins
and Structure-Activity Relationship for the Inhibition of Telomerase.
J. Med. Chem. 2001, 44, 4509-4523.
(27) Murphy, P. M.; Jenkins, T. C.; Wheelhouse, R. T. Synthesis of
Molecular Twist Compounds and Evaluation of Their Interactions
with Higher-Order DNA Structures. Br. J. Cancer 2001, 85, 92-
92.
(28) Hughes, G.; Wang, C. S.; Batsanov, A. S.; Fern, M.; Frank, S.; Bryce,
M. R.; Perepichka, I. F.; Monkman, A. P.; Lyons, B. P. New
Pyrimidine- and Fluorene-Containing Oligo(arylene)s: Synthesis,
Crystal Structures, Optoelectronic Properties and a Theoretical Study.
Org. Biomol. Chem. 2003, 1, 3069-3077.
(29) Saygili, N.; Batsanov, A. S.; Bryce, M. R. 5-Pyrimidylboronic acid
and 2-Methoxy-5-pyrimidylboronic Acid: New Heteroarylpyrimidine
Derivatives via Suzuki Cross-Coupling Reactions. Org. Biomol.
Chem. 2004, 2, 852-857.
(30) Wu, C. C.; Lin, Y. T.; Chiang, H. H.; Cho, T. Y.; Chen, C. W.;
Wong, K. T.; Liao, Y. L.; Lee, G. H.; Peng, S. M. Highly Bright
Blue Organic Light-Emitting Devices Using Spirobifluorene-Cored
Conjugated Compounds. Appl. Phys. Lett. 2002, 81, 577-579.
(31) Schomaker, J. M.; Delia, T. J. Arylation of Halogenated Pyrimidines
via a Suzuki Coupling Reaction. J. Org. Chem. 2001, 66, 7125-
7128.
(32) Wong, K. T.; Hung, T. S.; Lin, Y. T.; Wu, C. C.; Lee, G. H.; Peng,
S. M.; Chou, C. H.; Su, Y. L. O. Suzuki Coupling Approach for the
Synthesis of Phenylene-Pyrimidine Alternating Oligomers for Blue
Light-Emitting Material. Org. Lett. 2002, 4, 513-516.
(33) Boucher, E.; Simard, M.; Wuest, J. D. Use of Hydrogen Bonds To
Control Molecular Aggregation. Behavior of Dipyridones and
Pyridone-Pyrimidones Designed To Form Cyclic Triplexes. J. Org.
Chem. 1995, 60, 1408-1412.
(34) Schroter, S.; Stock, C.; Bach, T. Regioselective Cross-Coupling
Reactions of Multiple Halogenated Nitrogen-, Oxygen-, and Sulfur-
Containing Heterocycles. Tetrahedron 2005, 61, 2245-2267.
(35) Aldrich product data, www.sial.com.
Hoff enthalpies; telomerase assay data; list of nucleic acid structures
used in the competition dialysis assay; table of competition dialysis
data; elemental analysis data for all compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Levine, M.; Tjian, R. Transcription Regulation and Animal Diversity.
Nature 2003, 147-151.
(2) Jenkins, T. C. Targeting Multi-Stranded DNA Structures. Curr. Med.
Chem. 2000, 7, 99-115.
(3) Arthanari, H.; Bolton, P. H. Functional and Dysfunctional Roles of
Quadruplex DNA in Cells. Chem. Biol. 2001, 8, 221-230.
(4) Mergny, J.-L.; He´le`ne, C. G-Quadruplex DNA: A Target for Drug
Design. Nat. Med. 1998, 4, 1366-1367.
(5) Ren, J.; Qu, X.; Dattagupta, N.; Chaires, J. B. Molecular Recognition
of a RNA:DNA Hybrid Structure. J. Am. Chem. Soc. 2001, 123,
6742-6743.
(6) Zaman, G. J.; Michiels, P. J.; van Boeckel, C. A. Targeting RNA:
New Opportunities To Address Drugless Targets. Drug DiscoVery
Today 2003, 8, 297-306.
(7) Siddiqui-Jain, A.; Grand, C. L.; Bearss, D. J.; Hurley, L. H. Direct
Evidence for a G-Quadruplex in a Promoter Region and Its Targeting
with a Small Molecule To Repress c-MYC Transcription. Proc. Nat.
Acad. Sci. U.S.A. 2002, 99, 11593-11598.
(8) Grand, C. L.; Han, H.; Munoz, R. M.; Weitman, S.; Von Hoff, D.
D.; Hurley, L. H.; Bearss, D. J. The Cationic Porphyrin TMPyP4
Down-Regulates c-MYC and Human Telomerase Reverse Tran-
scriptase Expression and Inhibits Tumor Growth in Vivo. Mol.
Cancer Ther. 2002, 1, 565-573.
(9) Catasti, P.; Chen, X.; Moyzis, R. K.; Bradbury, E. M.; Gupta, G.
Structure-Function Correlations of the Insulin-Linked Polymorphic
Region. J. Mol. Biol. 1996, 264, 534-545.
(10) Dai, J. X.; Dexheimer, T. S.; Chen, D.; Carver, M.; Ambrus, A.;
Jones, R. A.; Yang, D. Z. An Intramolecular G-Quadruplex Structure
with Mixed Parallel/Antiparallel G-Strands Formed in the Human
BCL-2 Promoter Region in Solution. J. Am. Chem. Soc. 2006, 128,
1096-1098.
(11) De Armond, R.; Wood, S.; Sun, D. Y.; Hurley, L. H.; Ebbinghaus,
S. W. Evidence for the Presence of a Guanine Quadruplex Forming
Region within a Polypurine Tract of the Hypoxia Inducible Factor 1
Alpha Promoter. Biochemistry 2005, 44, 16341-16350.
(12) Sun, D. Y.; Guo, K. X.; Rusche, J. J.; Hurley, L. H. Facilitation of
a Structural Transition in the Polypurine/Polypyrimidine Tract within
the Proximal Promoter Region of the Human VEGF Gene by the
Presence of Potassium and G-Quadruplex-Interactive Agents. Nucleic
Acids Res. 2005, 33, 6070-6080.
(13) Rankin, S.; Reszka, A. P.; Huppert, J.; Zloh, M.; Parkinson, G. N.;
Todd, A. K.; Ladame, S.; Balasubramanian, S.; Neidle, S. Putative
DNA Quadruplex Formation within the Human c-kit Oncogene. J.
Am. Chem. Soc. 2005, 127, 10584-10589.
(14) Todd, A. K.; Johnston, M.; Neidle, S. Highly Prevalent Putative
Quadruplex Sequence Motifs in Human DNA. Nucleic Acids Res.
2005, 33, 2901-2907.
(15) Huppert, J. L.; Balasubramanian, S. Prevalence of Quadruplexes in
the Human Genome. Nucleic Acids Res. 2005, 33, 2908-2916.
(16) Keniry, M. A. Quadruplex Structures in Nucleic Acids. Biopolymers
2000, 56, 123-146.
(36) Testaferri, L.; Tingoli, M.; Tiecco, M. A Convenient Synthesis of
Aromatic Thiols from Unactivated Aryl Halides. Tetrahedron Lett.
1980, 21, 3099-3100.
(37) Testaferri, L.; Tiecco, M.; Tingoli, M.; Chianelli, D. Simple Synthesis
of Aryl Alkyl Thioethers and of Aromatic Thiols from Unactivated
Aryl Halides and Efficient Methods for Selective Dealkylation of
Aryl Alkyl Ethers and Thioethers. Synth. Commun. 1983, 751-755.
(38) Blaszczyk, A.; Elbing, M.; Mayor, M. Bromine Catalyzed Conversion
of S-tert-Butyl Groups into Versatile and, for Self-Assembly
Processes, Accessible, Acetyl-Protected Thiols. Org. Biomol. Chem.
2004, 2, 2722-2724.
(39) Moore, C. G.; Porter, M. The Reaction of 2,4-Dinitrobenzenesul-
phenyl Chloride with Organic Monosulphides. Tetrahedron 1960,
9, 58-64.
(40) Pastuszak, J. J.; Chimiak, A. tert-Butyl Group as Thiol Protection in
Peptide-Synthesis. J. Org. Chem. 1981, 46, 1868-1873.
(41) Quintela, J. M.; Peinador, C. A Ready One-Pot Preparation for 7-Oxa-
(or thia)-3,4,6-triazabenz[d,e]anthracene and 7-Oxa-3,4,6,9-Tetraza-
benz[d,e]anthracene Derivatives. Tetrahedron 1996, 52, 10497-
10506.
(17) Perry, P. J.; Jenkins, T. C. DNA Tetraplex-Binding Drugs: Structure-
Selective Targeting Is Critical for Antitumour Telomerase Inhibition.
Mini-ReV. Med. Chem. 2001, 1, 31-41.
(18) Murphy, P. M.; Phillips, V. A.; Jennings, S. A.; Garbett, N. C.;
Chaires, J. B.; Jenkins, T. C.; Wheelhouse, R. T. Biarylpyrimidines:
A New Class of Ligand for High-Order DNA Recognition. Chem.
Commun. 2003, 1160-1161.
(19) Wilson, W. D.; Strekowski, L.; Tanious, F. A.; Watson, R. A.;
Mokrosz, J. L.; Strekowska, A.; Webster, G. D.; Neidle, S. Binding
of Unfused Aromatic Cations to DNA. The Influence of Molecular
Twist on Intercalation. J. Am. Chem. Soc. 1988, 110, 8292-8299.
(20) Wang, Y.; Patel, D. J. Solution Structure of the Human Telomeric
Repeat d[AG3(T2AG3)3] G-Tetraplex. Structure 1993, 1, 263-282.
(21) Lane, A. N.; Jenkins, T. C. Structures and Properties of Multi-
Stranded Nucleic Acids. Curr. Org. Chem. 2001, 5, 845-870.
(22) Soyfer, V. N.; Potaman, V. N. Triple-Helical Nucleic Acids; Springer-
Verlag: New York, 1996.
(42) McConnaughie, A. W.; Jenkins, T. C. Novel Acridine-Triazenes as
Prototype Combilexins: Synthesis, DNA Binding, and Biological
Activity. J. Med. Chem. 1995, 38, 3488-3501.
(43) Haq, I.; Ladbury, J. E.; Chowdhry, B. Z.; Jenkins, T. C. Molecular
Anchoring of Duplex and Triplex DNA by Disubstituted Anthracene-
9,10-diones: Calorimetric, UV Melting, and Competition Dialysis
Studies. J. Am. Chem. Soc. 1996, 118, 10693-10701.
(44) Wilson, W. D.; Tanious, F. A.; Fernandez-Saiz, M.; Rigl, C. T.
Evaluation of Drug-Nucleic Acid Interactions by Thermal Melting
Curves. Drug-DNA Interaction Protocols; Humana Press Inc.:
Totowa, NJ, 1997; pp 219-240.
(45) Ren, J. S.; Chaires, J. B. Sequence and Structural Selectivity of
Nucleic Acid Binding Ligands. Biochemistry 1999, 38, 16067-
16075.
(46) Ren, J. S.; Chaires, J. B. Preferential Binding of 3,3′-Diethyloxadi-
carbocyanine to Triplex DNA. J. Am. Chem. Soc. 2000, 122, 424-
425.
(23) Bredereck, H.; Gompper, R.; Morlock, G. Formamid-Reactionen,
VIII. Eine Neue Pyrimidin-Synthese (Formamide Reactions, VIII.
A New Pyrimidine Synthesis). Chem. Ber. 1957, 90, 942-951.
(24) Herbert, C. G.; Bass, R. G.; Watson, K. A.; Connell, J. W. Preparation
of Poly(arylene ether pyrimidine)s by Aromatic Nucleophilic Sub-
stitution Reactions. Macromol. 1996, 29, 7709-7716.
(47) Ren, J. S.; Chaires, J. B. Rapid Screening of Structurally Selective
Ligand Binding to Nucleic Acids. Methods Enzymol. 2001, 340, 99-
108.
(48) Chaires, J. B. Competition Dialysis: An Assay To Measure the
Structural Selectivity of Drug-Nucleic Acid Interactions. Curr. Med.
Chem.: Anti-Cancer Agents 2005, 5, 339-352.

5198 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Wheelhouse et al.
(49) Chaires, J. B.; Ren, J.; Hamelberg, D.; Kumar, A.; Pandya, V.;
Boykin, D. W.; Wilson, W. D. Structural Selectivity of Aromatic
Diamidines. J. Med. Chem. 2004, 47, 5729-5742.
(50) Chaires, J. B.; Ren, J.; Henary, M.; Zegrocka, O.; Bishop, G. R.;
Strekowski, L. Triplex Selective 2-(2-Naphthyl)quinoline Com-
pounds: Origins of Affinity and New Design Principles. J. Am. Chem.
Soc. 2003, 125, 7272-7283.
(51) Hurley, L. H.; Wheelhouse, R. T.; Sun, D.; Kerwin, S. M.; Salazar,
M.; Fedoroff, O. Y.; Han, F. X.; Han, H.; Izbicka, E.; Von Hoff, D.
D. G-Quadruplexes as Targets for Drug Design. Pharmacol. Ther.
2000, 85, 141-158.
(52) Read, M.; Harrison, R. J.; Romagnoli, B.; Tanious, F. A.; Gowan,
S. H.; Reszka, A. P.; Wilson, W. D.; Kelland, L. R.; Neidle, S.
Structure-Based Design of Selective and Potent G Quadruplex-
Mediated Telomerase Inhibitors. Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 4844-4849.
(53) Kim, N. W.; Piatyszek, A. M.; Prowse, K. R.; Harley, C. B.;
West, M. D.; Ho, P. L. C.; Coviello, G. M.; Wright, W. E.;
Weinrich, S. L.; Shay, J. W. Specific Association of Human
Telomerase Activity with Immortal Cells and Cancer. Science 1994,
266, 2011-2015.
(54) Wheelhouse, R. T.; Sun, D.; Han, H.; Han, F. X.; Hurley, L. H.
Cationic Porphyrins as Telomerase Inhibitors: The Interaction of
Tetra(N-methyl-4-pyridyl)porphine with Quadruplex DNA. J. Am.
Chem. Soc. 1998, 120, 3261-3262.
(55) Izbicka, E.; Wheelhouse, R. T.; Raymond, E.; Davidson, K. L.;
Lawrence, R. A.; Sun, D.; Windle, B. E.; Hurley, L. H.; Von Hoff,
D. D. Effects of Cationic Porphyrins as G-Quadruplex Interactive
Agents in Human Tumour Cells. Cancer Res. 1999, 59, 639-644.
(56) Fedorov, O. Y.; Salazar, M.; Han, H.; Chemeris, V. V.; Kerwin, S.
M.; Hurley, L. H. NMR-Based Model of a Telomerase-Inhibiting
Compound Bound to G-Quadruplex DNA. Biochemistry 1998, 37,
12367-12374.
(57) Perry, P. J.; Jenkins, T. C. Recent Advances in the Development of
Telomerase Inhibitors for the Treatment of Cancer. Expert Opin.
InVest. Drugs 1999, 8, 1981-2008.
(58) Perry, P. J.; Gowan, S. M.; Reszka, A. P.; Polucci, P.; Jenkins, T.
C.; Kelland, L. R.; Neidle, S. 1,4- and 2,6-Disubstituted Amido-
anthracene-9,10-dione Derivatives as Inhibitors of Human Telom-
erase. J. Med. Chem. 1998, 41, 3253-3260.
(59) Jenkins, T. C. Optical Absorbance and Fluorescence Techniques for
Measuring DNA-Drug Interactions. Drug-DNA Interaction Pro-
tocols; Humana Press Inc.: Totowa, NJ, 1997; pp 195-218.
(60) Perrin, D. D.; Perrin, D. R.; Armarego, W. L. F. Purification of
Laboratory Chemicals, 3rd ed.; Pergamon Press: Oxford, U.K., 1988.
(61) Still, W. C.; Khan, M.; Mitra, A. Rapid Chromatographic Technique
for Preparative Separations with Moderate Resolution. J. Org. Chem.
1978, 43, 2923-2925.
JM060315A