Spectroscopic studies on the formation and thermal stability of DNA triplexes with a benzoannulated δ-carboline-oligonucleotide conjugate

Article abstract of DOI:10.1016/j.bmc.2008.09.026

A benzoannulated δ-carboline with a phenyl substituent has been covalently tethered to the 3′-end of a triplex-forming oligonucleotide and its ability to bind and stabilize DNA triple helices has been examined by various spectroscopic methods. UV thermal melting experiments were conducted with different hairpin duplexes and with a complementary single-stranded oligonucleotide as targets for the conjugate. The δ-carboline ligand preferentially binds triplexes over duplexes and leads to a temperature increase of the triplex-to-duplex transition by up to 23 °C. The results obtained from UV, CD and fluorescence measurements suggest that the δ-carboline ligand exhibits specific interactions with a triplex and favors binding by intercalation at the triplex-duplex junction.

Full text of DOI:10.1016/j.bmc.2008.09.026

Bioorganic & Medicinal Chemistry 16 (2008) 9106–9112
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Spectroscopic studies on the formation and thermal stability of DNA
triplexes with a benzoannulated d-carboline–oligonucleotide conjugate
Andrea Eick a, Zhou Xiao a, Peter Langer b, Klaus Weisz a,
*
a Institut für Biochemie, Ernst-Moritz-Arndt-Universität Greifswald, Felix-Hausdorff-Str. 4, D-17487 Greifswald, Germany
b Institut für Chemie, Universität Rostock, Albert-Einstein-Str. 3a, 18059 Rostock, Germany
a r t i c l e i n f o
a b s t r a c t
A benzoannulated d-carboline with a phenyl substituent has been covalently tethered to the 30-end of a
triplex-forming oligonucleotide and its ability to bind and stabilize DNA triple helices has been examined
by various spectroscopic methods. UV thermal melting experiments were conducted with different hair-
pin duplexes and with a complementary single-stranded oligonucleotide as targets for the conjugate. The
d-carboline ligand preferentially binds triplexes over duplexes and leads to a temperature increase of the
triplex-to-duplex transition by up to 23 °C. The results obtained from UV, CD and fluorescence measure-
ments suggest that the d-carboline ligand exhibits specific interactions with a triplex and favors binding
by intercalation at the triplex–duplex junction.
Article history:
Received 11 June 2008
Revised 2 September 2008
Accepted 10 September 2008
Available online 13 September 2008
Keywords:
DNA triplex
d-Carboline
Oligonucleotide conjugate
Spectroscopy
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
helical complexes, especially under physiological conditions. Strat-
egies to enhance triplex stability have thus emerged as one of the
Triple helix formation through the sequence-specific recognition
of a double-helical DNA target by a third strand oligonucleotide
(TFO) has long been recognized as a potentially powerful tool for
numerous applications in medicine and molecular biology. These in-
clude the control of gene expression,1,2 site-directed mutagenesis3,4
or mapping of genomic DNA.5,6 The TFO in such triplexes binds in the
major groove of double-helical DNA forming specific hydrogen
bondstopurinebasesononeof thetwostrands. Ingeneral, twotypes
of triple helices have been characterized depending on the orienta-
tion of the third strand. In the pyrimidine motif, a homopyrimidine
strand binds to a homopurine sequence of the duplex DNA in a par-
allel orientation with the formation of TꢀAT and C+ꢀGC triplets. Be-
cause cytosine bases in the third strand have to be protonated in
order to form two Hoogsteen hydrogen bonds to a guanine base, a
low pH value is generally required for stable triplex formation. In a
second triplex motif, a purine third strand binds in an antiparallel
orientation with respect to the oligopurine tract of the underlying
duplex forming GꢀGC and AꢀAT triplets through reverse Hoogsteen
hydrogen bonds. In addition, triplex forming oligonucleotides com-
posed of T and G may also bind to a homopurine stretch of double-
helical DNA with an orientation that depends upon the particular
base sequence [for a review, see Ref. 7].
major challenges towards more general future applications [for a
review, see Ref. 8]. In addition to various chemical modifications
of backbone, bases and sugar residues, additional stacking interac-
tions through molecular capping of nucleic acid bases by aromatic
and polycyclic compounds, that are either covalently linked to the
30- or 50-terminus of the TFO,9–11 attached at an internucleotide po-
sition12–14 or being part of a linker region in an intramolecular con-
struct,15–17 have been exploited for triplex stabilization. Some of
these moieties may also serve as probes for triplex formation, for
example, through changes of their fluorescence properties upon
binding of the conjugate to the target duplex. Whereas a large
number of natural and synthetic drugs have been shown to bind
and stabilize nucleic acid structures, only a limited number of
compounds have been found to selectively bind triplexes. The
latter include annelated quinolizinium salts,18,19 benzopyridoin-
dole (BPI) and benzopyridoquinoxaline (BPQ) derivatives,20,21 dib-
enzophenanthrolines,22 2-naphthyl substituted quinolines23 and
substituted anthraquinones.24 We have recently shown in preli-
minary studies, that a novel phenyl-substituted benzoannulated
d-carboline, 5-phenyl-6H-indolo[3,2-b]quinoline, when tethered
to a third strand oligonucleotide is effective in stabilizing DNA tri-
plexes formed with a cognate duplex.25 In the present work, we
have continued studies on the ligand binding and thermal stability
of triple helices formed by a d-carboline–TFO conjugate and vari-
ous double-helical DNA targets through UV, CD and fluorescence
measurements. Also, full experimental details of the ligand-oligo-
nucleotide coupling reaction, that provides the conjugates in good
yields, are reported.
Presently, a more widespread use of triplex formation through
the recognition of double-stranded nucleic acids by oligonucleo-
tides is often hampered by the low stability of the formed triple-
* Corresponding author. Tel.: +49 (0)3834 864426; fax: +49 (0)3834 864427.
E-mail address: weisz@uni-greifswald.de (K. Weisz).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.09.026

A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
9107
2. Results and discussion
0.80
2.1. UV melting experiments
0.76
0.72
0.68
0.64
0.60
Conjugation of the phenyl-substituted benzoannulated d-carbo-
line to the 30-terminus of an oligonucleotide was accomplished
through the coupling of the NHS-activated ligand with the 30-ami-
noalkyl-modified oligonucleotide synthesized by standard proto-
cols from FMOC-protected C7 amino link solid support.25 After a
reaction time of 5 days at elevated temperatures in a two-phase
DMF-phosphate buffer system, the coupled product was isolated
and subjected to a final HPLC purification. To promote dissolution
of the NHS ester in the solvent system, lithium chloride was used
as an additive for the coupling reaction giving a final yield of
P50% for the purified product.
0
20
40
60
80
100
Temperature-dependent UV measurements were performed on
various triplexes formed by the d-carboline–oligonucleotide conju-
gate and hairpin duplexes with 10 base pairs each (see Fig. 1). Tri-
plex formation with an intramolecularly folded and thus
entropically stabilized hairpin duplex ensures two well-separated
transitions for double-helix denaturation at high temperatures
and third-strand dissociation at much lower temperatures associ-
ated with a hyperchromicity at 260 nm. As shown by the UV melt-
ing curves in Figure 2, triplex Trip1 showed a significant increase
temperature (ºC)
Figure 2. UV melting curves measured at 260 nm for the non-conjugated reference
(solid line) and for the conjugate triplex Trip1 (broken line). Tm1 values as
determined from the maxima of the first derivative plots are indicated by the
vertical arrows.
temperature, a single cooperative transition at the triplex–duplex
melting temperature Tm1 with an associated hypochromism of
33% upon triplex formation is found and suggests, that the dissoci-
ation of the conjugated third strand from the duplex occurs by a
one-step process (not shown).
in the triplex-to-duplex melting temperature
DTm1 of 22 °C when
compared to the unmodified triplex reference, indicating strong
interactions of the attached polycyclic d-carboline with the
triple-helical structure. Additionally, when the corresponding
d-carboline absorbance at 350 nm is recorded as a function of
In contrast, a smaller ligand-induced stabilization is observed
for a duplex Dup1 formed by the hybridization of the TFO with a
5‘- C T C C T T T
X
5‘- C T C C T T T
X
5‘- G A G G A A A T C G
3‘- C T C C T T T A G C
5‘- G A G G A A A G C G
3‘- C T C C T T T C G C
T4
T4
T4
T3
T4
Trip1
Trip 2
5‘- C T C C T T T
X
5‘- C T C C T T T
X
5‘- G A G G A A A G T G
3‘- C T C C T T T C A C
5‘- G A G G A A A C C G
3‘- C T C C T T T G G C
T4
Trip3
Trip4
5‘- C T C C T T T
X
5‘- C T C C T T T
X
5‘- G A G G A A G G C G
3‘- C T C C T T C C G C
5‘- G A G T A A A G C G
3‘- C T C A T T T C G C
T4
Trip5
Trip6
5‘- C T C C T T T
5‘- G C G G A G G A A A
X
5‘- T T T C C T C C G C
3‘- A A A G G A G G C G
T4
3‘- C G C C T C C T T T
X
T T T C C T C -5‘
Trip7
Trip8
HO
HN
NH
O
O
5‘- C T C C T T T
5‘- G A G G A A A C G C
X
O
O
=
X
P
N
Dup1
Figure 1. Triplex constructs with the TFO conjugate used in this study. The conjugate with the 30-linked ligand X runs parallel to the purine stretch of the target hairpin
duplex. Single base pair substitutions in triplexes Trip2-6 are marked by bold letters.

9108
A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
single-stranded 10 mer target possessing a complementary purine
sequence (see Fig. 1). Under the same buffer conditions and pH 5,
ligand binding within the double-helical structure results in a
much smaller increase of the duplex melting temperature by
11 °C consistent with a noticeably weaker affinity of the ligand to-
wards the duplex (Table 1). Interestingly, the ligand-induced ther-
mal duplex stabilization was found to be strongly affected by the
the purine sequence (AG)ꢀ(CT) in the duplex target at the triplex–
duplex junction can be derived. Note, however, that a noticeable
decrease of the triplex melting temperature associated with a
(AG)ꢀ(CT) to (AC)ꢀ(GT) substitution at the junction of Trip3 is also
observed for the corresponding reference sample and points to a
structural change that may affect the adjacent triplex region and
leads to a general weakening of the TFO binding. To position the li-
gand away from the triplex–duplex junction, Trip7 and Trip8 were
constructed with the d-carboline facing the triplex blunt end.
pH and
DTm for the duplex melting is further reduced to only
+4 °C at pH 7. It must be stressed, that relative changes in melting
temperatures do not necessarily translate into relative binding
constants due to the temperature dependence of association. How-
ever, because melting of both double- and triple-helical systems
formed upon hybridization with the conjugate occurs within a suf-
ficiently narrow temperature range, these temperature dependen-
cies can mostly be neglected and differences in thermal
stabilizations can be used as a reasonable measure of relative bind-
ing affinities in the present case.
As seen from Table 1, base pair substitutions in the duplex over-
hang of Trip1 to form constructs Trip2–4 only result in minor
changes of the ligand-induced triplex stabilization. Whereas a base
pair replacement at the center of the duplex overhang does not af-
fect the ligand binding (Trip4), base pair substitutions at the tri-
plex–duplex junction are associated with a small decrease in
triplex stabilization (see Trip2 and Trip3). Based on the larger
Although a considerable increase in
DTm1 indicates significant tri-
plex stabilization by the d-carboline, the stabilizing effect is about
equally reduced for both constructs by 6–8 °C compared to the par-
ent triplex Trip1 with the ligand facing the triplex–duplex junc-
tion. The tight T3 miniloop in Trip8 does not favor specific
interactions with the ligand and prevents its external stacking on
the terminal base triplet.26,27 On the other hand, the flexible linker
of the conjugate allows the ligand to easily fold back into the 50-
direction and consequently intercalation between two base triplets
is expected to be the major binding mode for both Trip7 and Trip8.
A single mismatch within the center of the triplex stem is not
tolerated and no triplex formation with the unmodified TFO or
the TFO conjugate could be detected for Trip5. These results show
that the binding affinity of the ligand does not significantly com-
promise the sequence specificity of the TFO towards its cognate se-
quence and one central mismatch prevents binding of the 7mer
conjugate to its target. In contrast, a mismatch at the terminal base
triad as introduced into construct Trip6 is less destabilizing allow-
ing the formation of a triplex even with the unmodified ligand-free
D
Tm1 for Trip1 and Trip4, a modest preference of the ligand for
Table 1
UV melting temperatures of triple- and double-helical constructsa
Triplex/
duplex
Tm1 (°C) triplex–
duplex
Tm2 (°C) duplex–single
strand
DTm1 (°C) triplex–
duplex
TFO. Stabilization of the conjugate triplex Trip6 amounts to a
DTm1
of 16 °C. The smaller stabilization compared to triplex Trip1 indi-
cates that the d-carboline binding is weakened due to this change
in the base sequence. This again suggests direct interactions of the
d-carboline with the terminal base triplet at the triplex–duplex
junction that might be hampered by the distorted geometry of
the mismatched base triad.
The interaction of the d-carboline within a triple helix was
also monitored by spectrophotometric titrations of a conjugate
buffer solution with the target duplex to form Trip1. As shown
in Figure 3, the oligonucleotide-linked ligand exhibits two in-
tense absorption bands at 350 nm and at 377 nm. The titration
of double-stranded DNA induces significant perturbations in
the two spectral bands. In contrast to the absorption maximum
at 350 nm, the long-wavelength absorption band at 377 nm
Trip1
Trip2
Trip3
Trip4
Trip5
Trip6
Trip7
Trip8
Dup1
Dup1d
44 (22)
41 (21)
38 (18)
45 (22)
— (—)b
31 (15)
41 (25)
45 (31)
—
79 (81)
77 (76)
80 (81)
74 (74)
75 (75)
82 (81)
80 (82)
80 (78)
25 (14)
18 (14)
22
20
20
23
—
16
16
14
11c
4c
—
Values for underivatized reference samples are given in parentheses.
a
Uncertainty in Tm 6 1 °C.
b
Not observed.
c
D
Tm2 (°C) for duplex–single strand transition.
d
20 mM phosphate, 100 mM NaCl, pH 7.
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
260
300
340
380
420
460
500
wavelength (nm)
Figure 3. Absorption spectra of a 3.6 lM solution of oligonucleotide conjugate and increasing amounts (0–7.2 lM) of complementary double-stranded oligonucleotide at
20 °C; arrows indicate the development of the absorption bands upon DNA titration.

A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
9109
exhibits an increase in intensity together with a bathochromic
2.3. Fluorescence measurements
shift of 5 nm relative to the free conjugate upon triplex forma-
tion with the target duplex. Moreover, several isosbestic points
within the absorption region of the ligand clearly support the
presence of only two spectrally distinct species of the d-carbo-
line chromophore upon triplex formation and provide evidence
that one binding mode is adopted almost exclusively within
the triple-helical complex.
In static fluorescence measurements, the maxima in excitation
and emission wavelengths of the d-carboline in the unbound con-
jugate and in a triple-helical complex were determined above
300 nm to exclude any masking effect by the nucleic acid bases.
Whereas the excitation wavelength of the d-carboline did not
change upon triplex formation with the duplex target
(kex = 350 nm), the emission wavelength shifted from kem = 466 nm
for the free conjugate to kem = 463 nm for the triple-helical com-
plex. Such a hypsochromic shift in the fluorescence emission is
typical for a change to a less polar environment and is likely to re-
flect the ligand binding associated with protection from the aque-
ous environment.
To examine the influence of triplex formation on the fluores-
cence intensity of the d-carboline, temperature-dependent mea-
surements of a 1:1 mixture of the conjugate and the target
duplex were performed. The sample was excited at 350 nm and
the emission was monitored between 360 and 600 nm in a temper-
ature range between 20 °C and 80 °C. As shown in Figure 5, the
fluorescence intensity increases with increasing temperature and
reaches the same value at higher temperatures that is observed
for the free conjugate at corresponding temperatures. In addition,
a shift in the emission wavelength from 463 nm at 20 °C to
466 nm at 50 °C is again observed indicating, that the conjugate
with the d-carboline is dissociated from the target duplex above
50 °C. Plotting the fluorescence intensity at 466 nm against the
temperature gives a transition temperature of 44 °C in agreement
with the UV melting data, again indicating a one-step dissociation
of the d-carboline–TFO conjugate from the duplex (see inset Fig. 5).
Consequently, the observed hypochromic effect with significant
quenching of ligand fluorescence below 50 °C must be attributed
to triplex formation and the binding of the d-carboline to the
DNA. At 20 °C the fluorescence intensity of the bound d-carboline
is approximately 13 times lower compared to the fluorescence
emission of the conjugate pointing to strong electronic interactions
between the ligand and DNA bases. Note that the effect of increas-
ing temperature on the fluorescence intensity of the free conjugate
2.2. Circular dichroism
Temperature-dependent CD experiments were employed on
Trip1 to independently probe structural changes for the nucleic
acid and the d-carboline upon triplex formation. Figure 4 shows
corresponding CD spectra of the triple-helical complex in a tem-
perature range from 20 °C to 60 °C. The high-temperature spectra
with a positive CD band at 280 nm and a negative CD band near
250 nm equal summed spectra of the conjugate and the target du-
plex in line with a dissociated third strand oligonucleotide and a
free B-type hairpin DNA (not shown). In contrast, formation of a
triple-helical complex at 20 °C is indicated by a more negative
CD band at around 250 nm and in particular by an intense negative
short-wavelength band at around 210 nm.28 Another weak nega-
tive CD band, centered at the long-wavelength absorbance maxi-
mum of the d-carboline at 380 nm and fully reproduced by an
independent series of temperature-dependent experiments, is only
observed for the triple-helical complex at lower temperatures and
disappears at about 50 °C after triplex melting (see inset Fig. 4). It
demonstrates that the ligand is bound to an asymmetric environ-
ment and coupling between the transition dipoles of the ligand
chromophore and DNA results in an induced CD (ICD) for the
DNA-bound d-carboline moiety. The sign of the ICD depends upon
the specific ligand orientation within the complex through its tran-
sition dipole moment for the electronic transition and in line with
its structural resemblance to other typical intercalators, a weak
negative ICD of the d-carboline–TFO conjugate when adding a du-
plex target may be taken as evidence for an intercalative mode of
binding.29
2.0
1.5
1.0
0.5
0.0
-0.5
20
15
10
5
300
340
380
420
460
wavelength (nm)
0
-5
-10
-15
200
250
300
350
400
450
wavelength (nm)
Figure 4. Circular dichroism spectra for the conjugate triplex Trip1 at 20 °C, 30 °C, 40 °C, 50 °C and 60 °C; arrows indicate the spectral changes with increasing temperature.
The inset shows an expansion of the spectral region comprising the maximum of ligand absorbance at 380 nm.

9110
A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
240
200
160
120
80
8
6
4
2
0
250
200
150
100
50
0
20 30 40 50 60 70 80
temperature (ºC)
40
0
350
400
450
500
550
600
wavelength (nm)
Figure 5. Fluorescence emission spectra for the conjugate triplex Trip1 at 20 °C, 30 °C, 40 °C, 50 °C, 60 °C, 70 °C and 80 °C; the arrow indicates the spectral changes with
increasing temperature. The inset shows the fluorescence melting curve at 466 nm together with its first derivative.
is small in comparison to the effect observed for the ligand binding
to the triple-helical complex (not shown).
2.4. Ligand binding
N
N
N
N
H
N
H
Differences in the ligand-mediated thermal stabilization as ob-
served for triplexes Trip1–4 with the d-carboline positioned at the
triplex–duplex junction and for triplexes Trip7 and Trip8 with the
ligand positioned at the triplex blunt end may be attributed to d-
carboline–DNA interactions of different strength and point to the
triplex–duplex junction as being the favorable intercalation site
for the ligand. In fact, triplex–duplex junctions have been shown
in the past to be strong binding sites for various intercalators
and this has been ascribed to some structural distortions inherent
to the junction.30,31 It has to be noted, that due to the short triple
helix with only seven base triplets comprising pH-dependent
C+ꢀGC triads, triplexes had to be stabilized by a low pH and by
the addition of spermine for reliable measurements in the accessi-
ble temperature range. As suggested by the pH dependent thermal
stabilization of the duplex in the presence of the d-carboline, the
pyrido ring nitrogen of our ligand may partially undergo proton-
ation/deprotonation between pH 5 and 7. Because a positive charge
promotes interactions with a highly negatively charged triplex
DNA, cationic polycyclic compounds constitute the vast majority
of known triplex-selective intercalators. On the other hand, posi-
tive charges at the intercalating moiety compromise binding to
CG-rich sequences due to electrostatic repulsion with adjacent
C+ꢀGC base triplets and consequently result in only weak stabiliza-
tion of GC-rich targets in general.32 However, third strand cyto-
sines at the end of a triplex have been found to exhibit an
apparent pKa of about 5.5 and hence are expected to also be partly
deprotonated at neutral pH at the triplex–duplex junction site.33
Optimized stacking interactions with base triplets are pro-
posed to be major contributors to an intercalative mode of bind-
ing. To maximize those interactions, the overall shape of a
triplex-selective polycyclic intercalator should closely match the
crescent-shaped geometry of the three bases in a canonical base
triad for an extensive overlap. Thus, the crescent-shaped B[e]PI
and B[g]PI ligands are efficient triplex-stabilizing intercalators.34
It is interesting to note, that the geometry of the four fused ring
systems of our benzoannulated d-carboline chromophore matches
the shape of the B[f]PI ligand, that has previously been found to
not significantly affect triplex stabilities at pH 6.2 with its more
linear shape that does not match the shape of the base triplets
(Fig. 6). Likewise, as expected from these simple geometric con-
B[g]PI
B[e]PI
N
N
H
N
H
B[f]PI
6H-indolo[3,2-b]quinoline
Figure 6. Chemical structures of benzopyridoindoles B[e]PI, B[g]PI and B[f]PI and
the benzoannulated d-carboline 6H-indolo[3,2-b]quinoline.
siderations, an indolo[3,2-b]-quinoline-11-carboxylic acid lacking
a phenyl substituent and tethered through its carboxylic func-
tionality to a third strand oligonucleotide binds duplex DNA with
high affinity but failed to bind triplexes over duplexes with
noticeable selectivity and affinity at pH 6.35 Disregarding addi-
tional pH dependent effects, the considerable triplex stabilization
of our d-carboline ligand at low pH may therefore be attributed to
the phenyl substituent at the tetracyclic ring system forming
stacking interactions with the DNA bases. As an added benefit,
the more flexible phenyl substituent might be better suited in
maximizing stacking interactions with propeller-twisted base tri-
ads deviating from perfect planarity. Such a flexibility is often
found to provide for an increased triplex stabilization of polycy-
clic compounds and has been exploited for the design of highly
efficient triplex-selective intercalators.13,23,36
3. Conclusions
UV, CD and fluorescence data support well-defined interactions
for the phenyl-substituted d-carboline covalently attached to the
TFO, resulting in a significant triplex thermal stabilization of up
DTm = 23 °C upon binding a duplex target. Given its efficient
coupling to third strand oligonucleotides associated with the easy
to

A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
9111
availability of corresponding oligonucleotide conjugates, the d-
carboline moiety presented here seems to be a promising triplex
binding ligand for a variety of potential applications. As an added
advantage, the synthesis of this type of d-carboline allows for a
straightforward introduction of various substituents at different
positions of the polycyclic compound and thus make the phenyl-
substituted indoloquinoline a promising lead compound for the
further development of this new class of triplex-selective interca-
lating agents. Clearly, for a better rational design more data on
the pH dependence and sequence selectivity of d-carbolines with
appropriate triple-helical systems have to be collected in the
future.
J = 7.1 Hz, 1 H, ArH), 7.28 (t, J = 7.7 Hz, 1 H, ArH), 3.01 (s, 3 H,
CH3), 2.93 (s, 4 H, 2ꢂ CH2), 2.51 (s, 3 H, CH3).
4.1.2. Oligonucleotide-d–carboline conjugate (3)
The 30-amino-modified oligonucleotide (6.7 OD, 0.12
l
mol) was
L) and 0.1 M sodium phosphate buffer
L) containing 50 mM lithium chloride was added.
The NHS ester 2 (10 mg/ml, 2 mol) dissolved in DMF (100 L)
dissolved in water (150
(pH 8.2, 65
l
l
l
l
was added and a yellow biphasic mixture was obtained. The solu-
tion was stirred for 5 days at 50 °C until the mixture became clear
and checked by TLC (propanol/H2O 5:3). The product mixture was
dried under vacuo and the residue dissolved in water. Excessive
NHS ester 2 was extracted three times by dichloromethane. The
conjugate 3 was finally purified by HPLC and desalted by using
Sep-Pak C18-Cartridges (Waters Corporation, Milford, USA). Yield
4. Experimental
4.1. Materials
5.3 OD units (53%). MALDI-TOF-MS: m/z calcd for C98N30O49P7H121
:
2579.88. Found: 2578.51.
Unmodified and 30-amino modified oligonucleotides were pur-
chased from TIB MOLBIOL (Berlin, Germany). The concentration
was calculated using molar extinction coefficients at 260 nm
derived from a nearest-neighbour model. The d-carboline deriva-
tive 1,7-dimethyl-5-(4-carboxyphenyl)-6H-indolo[3,2-b]quinoline
1 was prepared according to previously published procedures.37
The concentration of d-carboline-modified oligonucleotide was
approximated by using the determined molar extinction coeffi-
cient of the free d-carboline carboxy derivative 1 at 260 nm
4.2. UV–vis spectroscopy
All UV measurements were performed on a Cary 100 spectro-
photometer equipped with a temperature control unit (Varian Deu-
tschland, Darmstadt). Melting curves were recorded with 1 data
point/°C at 260 nm in a temperature interval from 1 to 90 °C. To
prevent water condensation on the cuvettes (1 cm path length)
at temperatures below 25 °C, the sample chamber was constantly
flushed with nitrogen gas. The lyophilized oligonucleotides
e
260 = 26,818 Mꢁ1 cmꢁ1
Solvents were distilled using a rotary evaporator prior to use.
.
(3.6 lM) were dissolved in cacodylate buffer pH 5.0 (0.02 M so-
dium cacodylate, 0.1 M NaCl, 0.001 M spermine) and annealed
prior to each melting experiment by heating to 90 °C followed by
slow cooling to room temperature. The third strand oligonucleo-
tide or the conjugate was added to the duplex or complementary
single strand in a ratio of about 1:1. Measurements included one
heating ramp (1 °C/min), followed by one cooling (1 °C/min) and
a second heating period (1 °C/min). Heating and cooling curves
showed small hysteresis effects due to slow kinetics of 0.5–3 °C
depending on the particular triplex. The melting temperature
was determined by the maximum of the first derivative plot of
the second heating curve after appropriate smoothing.
Titration experiments were performed by sequentially adding 10
aliquots of an oligonucleotide duplex solution to a 3.6 lM solution of
theoligonucleotide conjugate to reach a final molarratioof 2:1 with-
out an appreciable change in total volume. Simultaneously, equal
amounts of oligonucleotide were added to the reference cell to avoid
large increases of DNA absorption during titration.
Buffer solutions were prepared using deionized water (Millipore
Simplicity 185). All reactions were monitored by TLC using Merck
silica gel plates 60F254. Silicagel (0.063–0.2 mm, Mallinckrodt Baker,
The Netherlands) was used for normal column chromatography.
For purifications through flash chromatography, a Yamazen AI-
580 system (Yamazen, Osaka, Japan) equipped with a UV-detector
at a fixed wavelength (254 nm) and with prepacked flash chroma-
tography columns from Yamazen (Hi-Flash size L, Inject size M)
was used. A mixture of methylene chloride and methanol in a ratio
of 95:5 was employed for elution. HPLC separation was carried out
with a Hitachi/Knauer system. The purification was performed on a
reverse column (Kromasil 100 Å C18 5
lm, 4.6 ꢂ 250 mm, Wicom
GmbH, Heppenheim, Germany). A gradient was applied using buf-
fer A (0.1 M triethylammonium acetate/CH3CN = 1:1, pH 6.8) and
buffer B (0.1 M triethylammonium acetate/CH3CN = 98:2, pH 6.8)
as eluent. A flow rate of 1.2 mL/min was used and the fractions de-
tected simultaneously at 260 nm, 281 nm, 347 nm and 406 nm.
NMR spectra were recorded on a Bruker AVANCE 600 MHz spec-
trometer at room temperature.
4.3. Circular dichroism
4.1.1. 1,7-Dimethyl-5-(4-succinimidocarboxyphenyl)-6H-
indolo[3,2-b]quinoline (2)
CD spectra were recorded with a Jasco J-810 spectropolarimeter
(Jasco, Tokyo, Japan). Wavelength scans were acquired with 20
accumulations from 205 to 450 nm, a scanning speed of 50 nm/
min, a response time of 8 s and a bandwidth of 1 nm. All CD exper-
iments were carried out in cacodylate buffer pH 5.0 (0.02 M
sodium cacodylate, 0.1 M NaCl, 0.001 M spermine) with a concen-
1,7-Dimethyl-5-(4-carboxyphenyl)-6H-indolo[3,2-b]quinoline
1 (112.7 mg, 0.3 mmol) was dissolved in DMF (4 mL). N-Hydroxy-
succinimide (NHS, 95 mg) and N-(3-dimethylaminopropyl)-N0 -eth-
ylcarbodiimide hydrochloride (DCI, 157 mg) were added and the
mixture was stirred for 2 days while monitored by TLC (CH2Cl2/
CH3OH 9:1). After completion, the solvent was removed under va-
cuo and the solid yellow residue dissolved in methylene chloride. It
was washed three times with water and another three times with a
concentrated solution of sodium chloride until the water phase
was nearly colourless. The organic phase was dried using sodium
sulfate and the solvent was removed by rotary evaporation. The
product was finally purified by column chromatography and flash
chromatography. Yield 130 mg (93%). 1H NMR (600 MHz, CD2Cl2):
d(ppm) = 8.39 (d, J = 8.0 Hz, 2 H, ArH), 8.35 (d, J = 7.7 Hz, 1 H, ArH),
7.81 (d, J = 8.0 Hz, 2 H, ArH), 7.63 (d, J = 7.9 Hz, 1 H, ArH), 7.56 (d,
J = 7.1 Hz, 1 H, ArH), 7.42 (d, J = 7.1 Hz, 1 H, ArH), 7.39 (t,
tration of 8.2 lM of conjugate. The TFO conjugate was added to the
duplex in a ratio of about 1:1.
4.4. Fluorescence measurements
Fluorescence measurements were performed with a Jasco FP-
6500 spectrofluorometer (Jasco, Tokyo, Japan). Using an excitation
wavelength of 350 nm, emission spectra from 360 to 600 nm were
acquired with a scanning speed of 50 nm/min, an emission and exci-
tation bandwidth of 5 nm and a response time of 0.1 s. To obtain a
better S/N ratio five spectra were accumulated each. All fluorescence
measurements were carried out in cacodylate buffer pH 5.0 (0.02 M

9112
A. Eick et al. / Bioorg. Med. Chem. 16 (2008) 9106–9112
17. Bevers, S.; Schutte, S.; McLaughlin, L. W. J. Am. Chem. Soc. 2000, 122, 5905–
5915.
18. Lee, J. S.; Latimer, L. J. P.; Hampel, K. J. Biochemistry 1993, 32, 5591–5597.
19. Ihmels, H.; Otto, D.; Dall’Acqua, F.; Faccio, A.; Moro, S.; Viola, G. J. Org. Chem.
2006, 71, 8401–8411.
sodium cacodylate, 0.1 M NaCl, 0.001 M spermine) and a concentra-
tion of 2.2 M of conjugate. The third strand oligonucleotide or the
conjugate was added to the duplex in a ratio of about 1:1.
l
20. Mergny, J.-L.; Duval-Valentin, G.; Nguyen, C. H.; Perrouault, L.; Faucon, B.;
Rougée, M.; Montenay-Garestier, T.; Bisagni, E.; Hélène, C. Science 1992, 256,
1681–1684.
Acknowledgment
21. Marchand, C.; Bailly, C.; Nguyen, C. H.; Bisagni, E.; Garestier, T.; Hélène, C.;
Waring, M. J. Biochemistry 1996, 35, 5022–5032.
22. Baudoin, O.; Marchand, C.; Teulade-Fichou, M.-P.; Vigneron, J.-P.; Sun, J.-S.;
Garestier, T.; Hélène, C.; Lehn, J.-M. Chem. Eur. J. 1998, 4, 1504–1508.
23. Wilson, W. D.; Tanious, F. A.; Mizan, S.; Yao, S.; Kiselyov, A. S.; Zon, G.;
Strekowski, L. Biochemistry 1993, 32, 10614–10621.
We thank Mrs. R. Schröder for her help in the synthesis of the d-
carboline derivatives.
References and notes
24. Fox, K. R.; Polucci, P.; Jenkins, T. C.; Neidle, S. Proc. Natl. Acad. Sci. U.S.A. 1995,
1. Hélène, C.; Thuong, N. T.; Harel-Bellan, A. Ann. N.Y. Acad. Sci. 1992, 660, 27–36.
2. Kim, H. G.; Miller, D. M. Biochemistry 1998, 37, 2666–2672.
3. Havre, P. A.; Gunther, E. J.; Gasparro, F. P.; Glazer, P. M. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 7879–7883.
4. Povsic, T. J.; Strobel, S. A.; Dervan, P. B. J. Am. Chem. Soc. 1992, 114, 5934–
5941.
92, 7887–7891.
25. Todorovic
´, N.; Thi Bich Phuong, N.; Langer, P.; Weisz, K. Bioorg. Med. Chem. Lett.
2006, 16, 1647–1650.
26. Boulard, Y.; Gabarro-Arpa, J.; Cognet, J. A. H.; Le Bret, M.; Guy, A.;
Téoule, R.; Guschlbauer, W.; Fazakerley, G. V. Nucleic Acids Res. 1991, 19,
5159–5167.
5. Perrouault, L.; Asseline, U.; Rivalle, C.; Thuong, N. T.; Bisagni, E.; Giovannangeli,
C.; Le Doan, T.; Hélène, C. Nature 1990, 344, 358–360.
6. Strobel, S. A.; Doucette-Stamm, L. A.; Riba, L.; Housman, D. E.; Dervan, P. B.
Science 1991, 254, 1639–1642.
7. Thuong, N. T.; Hélène, C. Angew. Chem. Int. Ed. Engl. 1993, 32, 666–690.
8. Fox, K. R. Curr. Med. Chem. 2000, 7, 17–37.
9. Sun, J.-S.; Francois, J.-C.; Montenay-Garestier, T.; Saison-Behmoaras, T.; Roig,
V.; Thuong, N. T.; Hélène, C. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 9198–9202.
10. Silver, G. C.; Nguyen, C. H.; Boutorine, A. S.; Bisagni, E.; Garestier, T.; Hélène, C.
Bioconjugate Chem. 1997, 8, 15–22.
27. Chou, S.-H.; Tseng, Y.-Y.; Chu, B.-Y. J. Biomol. NMR 2000, 17, 1–16.
28. Manzini, G.; Xodo, L. E.; Gasparotto, D.; Quadrifoglio, F.; van der Marel, G. A.;
van Boom, J. H. J. Mol. Biol. 1990, 213, 833–843.
29. Tuite, E.; Nordén, B. J. Am. Chem. Soc. 1994, 116, 7548–7556.
30. Bates, P. J.; Macaulay, V. M.; McLean, M. J.; Jenkins, T. C.; Reszka, A. P.;
Laughton, C. A.; Neidle, S. Nucleic Acids Res. 1995, 23, 4283–4289.
31. Collier, D. A.; Mergny, J.-L.; Thuong, N. T.; Hélène, C. Nucleic Acids Res. 1991, 19,
4219–4224.
32. Keppler, M. D.; James, P. L.; Neidle, S.; Brown, T.; Fox, K. R. Eur. J. Biochem. 2003,
270, 4982–4992.
11. Gianolio, D. A.; Segismundo, J. M.; McLaughlin, L. W. Nucleic Acids Res. 2000, 28,
2128–2134.
33. Leitner, D.; Schröder, W.; Weisz, K. J. Am. Chem. Soc. 1998, 120, 7123–7124.
34. Escudé, C.; Nguyen, C. H.; Mergny, J.-L.; Sun, J.-S.; Bisagni, E.; Garestier, T.;
Hélène, C. J. Am. Chem. Soc. 1995, 117, 10212–10219.
12. Jessen, C. H.; Pedersen, E. B. Helv. Chim. Acta 2004, 87, 2465–2471.
13. Filichev, V. V.; Pedersen, E. B. J. Am. Chem. Soc. 2005, 127, 14849–14858.
14. Géci, I.; Filichev, V. V.; Pedersen, E. B. Bioconjugate Chem. 2006, 17, 950–957.
15. Salunkhe, M.; Wu, T.; Letsinger, R. L. J. Am. Chem. Soc. 1992, 114, 8768–8772.
16. Letsinger, R. L.; Wu, T. J. Am. Chem. Soc. 1995, 117, 7323–7328.
35. Eritja, R. Chem. Biodivers 2004, 1, 289–295.
36. Gianolio, D. A.; McLaughlin, L. W. Bioorg. Med. Chem. 2001, 9, 2329–2334.
37. Langer, P.; Anders, J. T.; Weisz, K.; Jähnchen, J. Chem. Eur. J. 2003, 9,
3951–3964.