Efficient photocleavage of DNA by cationic porphyrin-acridine hybrids with the effective length of diamino alkyl linkage

Article abstract of DOI:10.1248/cpb.49.287

Positively charged porphyrins bearing an acridine with various lengths of diamino alkyl linkage, 5-[4-[(6-chloro-2- methoxy-9-acridyl)aminoalkylaminocarbonyl]phenyl]-10,15,20-tris (4-N-methylpyridiniumyl)porphine triiodide, alkyl=ethyl, butyl, hexyl, of o

Full text of DOI:10.1248/cpb.49.287

March 2001
Chem. Pharm. Bull. 49(3) 287—293 (2001)
287
Efficient Photocleavage of DNA by Cationic Porphyrin–Acridine Hybrids
with the Effective Length of Diamino Alkyl Linkage
Yoshinobu ISHIKAWA, Aya YAMASHITA, and Tadayuki UNO*
Faculty of Pharmaceutical Sciences, Kumamoto University, 5–1 Oe-honmachi, Kumamoto 862–0973, Japan.
Received August 30, 2000; accepted November 16, 2000
Positively charged porphyrins bearing an acridine with various lengths of diamino alkyl linkage, 5-[4-[(6-
chloro-2-methoxy-9-acridyl)aminoalkylaminocarbonyl]phenyl]-10,15,20-tris(4-N-methylpyridiniumyl)porphine
triiodide, alkyl؍ethyl, butyl, hexyl, or octyl, were synthesized. They exhibited more enhanced photocleavage ac-
tivity of pUC18 plasmid DNA than TMPyP, meso-tetrakis(4-N-methylpyridiniumyl)porphine, which is well
known to bind to DNA tightly and to cleave DNA effectively; the hybrid linked with the hexamethylene chain
showed particularly high activity. An equilibrium dialysis experiment demonstrated that the binding ability of
the hybrids to calf thymus (CT) DNA correlated quantitatively with the photocleavage activity. The lack of the
substantial red-shift of the Soret maxima of the hybrids through the titration with CTDNA denied the intercala-
tive binding of the porphyrin part. In their circular dichroism (CD) spectral change on binding to CTDNA, two
negative peaks appeared at 275 nm and at 285—290 nm in the UV range. The latter negative peak was observed
for hybrids, but not for TMPyP, and thus we assigned it to induced CD (ICD) derived from intercalation of acri-
dine chromophore. In the visible range, the hybrids showed only a positive peak around their Soret maxima, and
this feature suggested the porphyrin moiety lay in the DNA groove. In addition, the length of the linker markedly
influenced the ellipticity of their visible ICD, suggesting that the proximity of the porphyrin moiety to DNA was
greatly affected by the linker.
Key words porphyrin; DNA; acridine; circular dichroism; photocleavage; equilibrium dialysis
The development of compounds which modify nucleic The synthetic procedures for the positively charged por-
acid structure is a current research area and can make a sig- phyrin–acridine hybrids 1—4 are illustrated in Chart 1. One
nificant contribution to anti-tumor and -viral chemother- equivalent of terephthalaldehydic acid methyl ester, 3 eq of
1
,2)
3—5)
apy. For example, drugs like bleomycin (BLM)
capable pyridine-4-aldehyde, and 4 eq of pyrrole were reacted in pro-
of binding to and cleaving nucleic acids are expected to sup- pionic acid for 1.5 h. The resultant mixture including six por-
press the proliferation of malignant cells by blocking the phyrin isomers was separated on silica gel. The fifth eluted
flow of genetic information. From this point of view, cationic isomer, 5-(4-methoxycarbonylphenyl)-10,15,20-tris(4-pyridyl)-
meso-tetrakis(4-N-methylpyridiniumyl)porphine (TMPyP) has porphine, 9, obtained in 5.9% yield was then hydrolyzed with
been carefully studied and noted for its bifunctionality: i) the lithium hydroxide to give 5-(4-carboxyphenyl)-10,15,20-
tight interaction with DNA through the three types of binding tris(4-pyridyl)porphine, 10, almost quantitatively. Condensa-
mode including intercalation, outside binding in the groove, tion of 1 eq of 10 with 5—7 eq of acridine-diaminoalkane de-
2
0)
and outside binding with self-stacking along the DNA sur- rivatives, 5—8, using 5—7 eq of carbonyldiimidazole
6
,7)
face,
ity.
and ii) the marked photochemical nuclease activ- (CDI) in an absolute 50/50 (v/v) DMF–DMSO solution af-
8
—10)
11,12)
Such a positively charged photosensitizer
would forded corresponding porphyrin–acridine hybrids, 11—14, in
1
3,14)
also be useful in the photodynamic therapy of cancer.
22—46% yield. The products were then methylated with ex-
Some hybrid molecules composed of a non-charged por- cess methyl iodide in DMF for 3 h to give the final products,
phyrin and acridines have been synthesized as a functional 1—4, quantitatively. Characterization of the compounds was
1
5—17)
18,19)
1
BLM model.
Acridine is known as an intercalator,
accomplished using H-NMR spectroscopy, matrix assisted
and hence the metal complexes of the hybrids bind to DNA laser desorption ionization time-of-flight mass spectrometry
via the acridine moiety, and then cause strand breaks by gen- (MALDI-TOF-MS), and elemental analysis.
erating reactive oxygen species in the presence of reducing
or oxidizing agents.
Photocleavage of DNA To estimate the potency of these
novel hybrids, the photo-nuclease activity of 1—4 was exam-
To achieve higher and more efficient chemical modifica-
tion of negatively charged DNA, we have for the first time
synthesized positively charged porphyrin–acridine hybrids,
1
—4, which are linked with various lengths of a diamino
alkyl chain. In this paper we report their photo-nuclease ac-
tivity and interaction with DNA. Our data show that they ex-
hibit more enhanced photo-induced nuclease activity than
TMPyP, and suggest that acridine and porphyrin moiety in
the hybrids interact with DNA by the intercalative and out-
side groove binding, respectively.
Results and Discussion
Synthesis of the Cationic Porphyrin–Acridine Hybrids
∗
To whom correspondence should be addressed. e-mail: unot@gpo.kumamoto-u.ac.jp
© 2001 Pharmaceutical Society of Japan

2
88
Vol. 49, No. 3
Fig. 1. Agarose Gel (0.8%) Electrophoresis of Plasmid DNA Treated with
TMPyP and 1—4 in the Presence of Light
Cleavage conditions: 57.9 mM pUC18 plasmid DNA, 10 mM sodium phosphate buffer
pH 7.0), irradiation at 526 nm for 30 min at 25 °C. a) Lane I, DNA alone; lane II to VI,
(
1
.0 mM TMPyP, 1, 2, 3, and 4. Form I and form II denote the supercoiled and nicked cir-
cular forms of plasmid DNA, respectively; b) a bar graph shows the densitometric data
for form II (%) of each lane, and a line graph shows the amount of TMPyP and 1—4
bound to CTDNA. The dialysis results were obtained after equilibration for 24 h, using
1
tube.
.0 mM TMPyP and 1—4 in each dialysate solution and 58.0 mM CTDNA in the dialysis
a dialysate buffer solution (10 mM sodium phosphate, pH 7.0)
containing a hybrid (1.0 mM). After equilibration at 25 °C for
2
4 h, the amount of the hybrid bound to CTDNA (C ) was
b
2
1)
evaluated by a visible absorbance measurement. Results
are summarized in Fig. 1b as a line graph. Among the hy-
brids, 3 (C ϭ9.6 mM) was bound in the highest amount, and
b
the amount of other hybrids bound to CTDNA decreased in
the order of 2 (8.1 mM), 4 (7.1 mM), and 1 (5.7 mM). These
equilibrium dialysis results of the hybrids closely correlated
with their photo-nuclease activity, in that their quantitative
Chart 1
ined using supercoiled double-stranded pUC18 plasmid relation for the amount of the hybrid bound to CTDNA cor-
DNA. A mixture of the hybrid (1.0 mM) and the plasmid responded to their relative capability of photochemical modi-
2
2)
DNA (57.9 mM) in a buffer (10 m M sodium phosphate, pH fication of the plasmid DNA.
7
.0) was irradiated at 526 nm of the Q band region for 30 min
Absorption Spectra Complexation between a ligand
at 25 °C. Acridine moiety in the hybrids is not excited at this molecule and DNA leads to absorption spectral changes that
wavelength. After irradiation, the conversion of the super- can be used to monitor the binding process. All hybrid solu-
coiled DNA (form I) to nicked circular DNA (form II) was tions (5.0 mM) were titrated with a stock solution of CTDNA
visualized by agarose gel electrophoresis with ethidium bro- in a buffer (100 mM NaCl and 10 mM sodium phosphate, pH
mide staining. Figure 1 shows the results of the gel elec- 7.0). The intensity of the Soret bands for all hybrids de-
trophoresis (a), and the densitometric data of the form II dis- creased at the initial stage, and then increased with further
tribution of each lane as a bar graph (b). In this reaction con- DNA additions. The spectral change for titration of 1 with
dition, 1, having an acridine with dimethylene linkage, and CTDNA is shown in Fig. 2 as a representative of all hybrids,
TMPyP, having no acridine, showed almost the same degrees and Fig. 3 shows its absorbance change at the Soret maxi-
(
44 and 43%, respectively) of the conversion from form I to mum. The spectral behavior demonstrated that at least two
form II. However, the hybrids 2 (73%), 3 (97%) and 4 (63%) binding steps exist. While the large decrease in absorbance
with longer linkage exhibited higher degrees of the conver- (hypochromicity) was observed in common among 1, 2, 3,
sion than TMPyP. Based on these results, it is clear that the and 4 (33, 38, 38, and 34%, respectively), the substantial red-
positively charged porphyrin–acridine hybrid 3 with hexam- shift of the Soret bands was not observed through the titra-
ethylene linkage has the most efficient photo-nuclease activ- tion. The absence of the red-shift indicates that there is no
23)
ity of all the porphyrins tested, and that the effective length sign of intercalation of TMPyP moiety in the hybrids.
of the diamino alkyl linkage in the hybrids is responsible for
this high nuclease activity.
At the initial step below the DNA concentration of 6—8
mM the intensity of the Soret maximum decreased monoto-
Equilibrium Dialysis An equilibrium dialysis experi- nously for each hybrid. Since isosbestic points were observed
ment to know how many hybrids can bind to DNA was per- for this step, as shown in Fig. 4 for 1, the optical contribu-
formed. Five-tenths milliliters of calf thymus DNA tions should come from two distinct species. We thus
(
CTDNA) solution (58.0 mM) was dialyzed against 100 ml of performed Scatchard analysis for the initial binding

March 2001
289
Fig. 4. Absorption Spectra of 1 (5.0 mM) with Addition of CTDNA at
RϾ0.75 in 10 mM Sodium Phosphate, 0.1 M NaCl (pH 7.0)
Fig. 2. Absorption Spectra of 1 (5.0 mM) with Addition of Various
CTDNA Concentrations in 10 mM Sodium Phosphate, 0.1 M NaCl (pH 7.0)
Fig. 5. Scatchard Plot for the Binding of 1 to CTDNA
Fig. 3. Absorbance Change of 1 at 429 nm with Addition of CTDNA
Table 1. Binding Parameters of Hybrids to CTDNA at RϾ0.6
2
4)
equilibrium. The Scatchard plot for the binding of 1 to
CTDNA is shown in Fig. 5, and the values of K, the binding
constant, and n, the number of base pairs occupied by the
bound hybrid, are summarized in Table 1. The values of n
evaluated for all hybrids suggest that one hybrid molecule
occupies one base pair site of DNA at the initial binding step,
1
2
3
4
Ϫ1
K/mM
n
58
0.97
23
0.94
8.9
0.96
8.1
0.98
though it is not clear whether the orientation of hybrids is or- 285—290 nm was observed for all hybrids, but not for
dered (e.g., outside stacking along the DNA surface) or not. TMPyP, and thus we assign it to induced CD derived from in-
Since the values of K and n were obtained at R (input ratio of tercalation of acridine chromophore in the hybrids.
[
hybrid]/[base pairs])Ͼ0.6, care should be taken that these
Induced CD (ICD) Spectra ICD is very helpful for
are not associated with other experiments performed at analysis of the interaction of chiral DNA with achiral por-
6
,7,23)
RՅ0.1.
phyrin.
The ICD spectra can clarify the binding mode of
Circular Dichroism (CD) Spectra CD is a powerful the porphyrin chromophore. Figure 7 shows the ICD spectra
spectroscopic technique that has been widely used to detect for TMPyP and hybrids bound to CTDNA in the 350—500
conformational changes of DNA and to study the effects of nm region at Rϭ0.10 (top), 0.043 (middle), and 0.010 (bot-
2
5,26)
interaction with external ligands.
The CD spectra gener- tom) in a buffer (100 mM NaCl and 10 mM sodium phosphate,
ally consist of positive and/or negative bands and are there- pH 7.0). For TMPyP both positive and negative peaks show-
fore more sensitive and informative than absorption spec- ing almost equivalent ellipticity were observed in the Soret
2
7)
tra. In particular, the CD spectra in the UV range is sensi- region. However, the ICD for all the hybrids showed only
tive to the conformational change of the duplex, and hence large positive peaks and decrease of the ellipticity in the
CTDNA was titrated with a stock solution of TMPyP or hy- order of 1, 2, 3, and 4.
brids in a buffer (100 mM NaCl and 10 mM sodium phos-
To examine the influence of ionic strength on the binding
phate, pH 7.0). Figure 6 shows the CD spectral changes on to CTDNA, the ICD were recorded at various salt (NaCl)
addition of TMPyP (top), 1 (middle), and 3 (bottom) to concentrations. Figure 8 shows the ICD spectra at Rϭ0.017
CTDNA. In the case of TMPyP, the decrease of positive CD in a buffer containing 100 mM (top), 20 mM (middle), or 0 mM
2
8)
at 270 nm was observed through the titration. The appear- (bottom) of NaCl. The positive peak of TMPyP decreased
ance of this negative peak is thought to reflect the conforma- with decrease in the salt concentration. The ICD spectral
tional perturbation of DNA caused by interaction with change would indicate alteration of the binding mode. On the
TMPyP. On the other hand, in the case of hybrids another other hand, for hybrids the presence of the positive peaks and
negative peak at 285—290 nm appeared together with a neg- the order of their ellipticity remained unchanged. These
ative peak at around 275 nm. This large negative peak at spectral data under the conditions employed at various R or

2
90
Vol. 49, No. 3
Fig. 6. CD Spectral Changes of CTDNA with Addition of TMPyP (Top), Fig. 7. ICD Spectra of Hybrids and TMPyP in the Presence of CTDNA in
1
7
(Middle), and 3 (Bottom) in 10 mM Sodium Phosphate, 0.1 M NaCl (pH
.0)
10 mM Sodium Phosphate, 0.1 M NaCl (pH 7.0)
The concentrations of the porphyrins and CTDNA were as follows: top, 5.0 mM and
5
0 mM; middle, 5.0 mM and 116 mM; bottom, 4.9 mM and 500 mM.
The CTDNA concentrations were as follows: top, 31.4 mM; middle, 31.0 mM; bottom,
8.6 mM.
2
cient photo-nuclease activity than TMPyP. The introduction
salt concentrations suggest that the porphyrin part in all the of acridine intercalator to cationic porphyrin chromophore
hybrids always lay in the DNA groove. Furthermore, it was brought about outside groove binding selectivity to the por-
deduced that the interaction of the porphyrin chromophore in phyrin, and the various lengths of their diamino alkyl linkage
the hybrid with DNA reduced by degrees, as the diamino markedly influenced the ellipticity of their visible ICD. Since
alkyl linkage became longer. It was concluded that the intro- the photo-nuclease activity of hybrids varied depending on
duction of an acridine intercalator to the lead compound the length of their diamino alkyl linkage, a definitive factor
TMPyP could result in outside groove binding selectivity to for the activity must be a suitable distance between the out-
the cationic porphyrin.
side groove surface of DNA and the positively charged por-
phyrin moiety in the hybrids. These hybrids are the first
cationic porphyrins showing more excellent photocleavage of
DNA than TMPyP.
Experimental
1
Instrumentation The H-NMR spectra were recorded on a JEOL GX-
4
00 or JNM-A-500 spectrometer. The MALDI TOF mass spectra were mea-
TM
sured on a Bruker REFLEX . The UV–visible absorption measurement
was performed on a Beckman DU650 spectrophotometer. The CD spectra
were recorded on a JASCO J-720 spectropolarimeter.
Materials and Methods CTDNA was purchased from Sigma Chemical
Co. CTDNA solutions were quantitated spectrophotometrically using e260ϭ
Ϫ1
Ϫ1
1
3,200 M (base pairs) cm . The tosylate salt of TMPyP was purchased
from Dojin Chemical Co.
-(2-Aminoethyl)amino-6-chloro-2-methoxyacridine (5) 6,9-Dichloro-
-methoxyacridine (1.20 g, 4.30 mmol) was dissolved in ethylenediamine
20 ml) and stirred at 70 °C for 1 h. After filtration, water (100 ml) was added
to the filtrate. The yellow powder was separated, collected on a filter, and
9
2
(
Conclusion We have synthesized new DNA-interactive,
cationic porphyrin–acridine hybrids linked with various
lengths of a diamino alkyl chain. They exhibited more effi- dried. Yield 1.0 g (81%).

March 2001
291
(
e): 420 (7070), 362 (3070), 345 (2500), 328 (1530), 285 (42700). EI-MS
ϩ
m/z: 358 (M , Calcd for C H N O Cl : 357.89). Anal. Calcd for
2
0
24
3
1
1
C H N O Cl ·1 (H O): C, 63.91; H, 6.97; N, 11.18. Found: C, 63.84; H,
2
0
24
3
1
1
2
6
.59; N, 10.91.
-(2-Aminooctyl)amino-6-chloro-2-methoxyacridine (8) To 1,8-di-
9
aminooctane (7.7 g) melted at 80 °C, 6,9-dichloro-2-methoxyacridine (0.77
g, 2.82 mmol) was added and stirred for 1 h. Water (300 ml) was added to
the reaction mixture, and then it was cooled in a refrigerator. The oily prod-
uct was dissolved in hydrochloric acid (15 ml) and water (100 ml) by warm-
ing. After filtration, orange powder precipitated, then was filtered off,
washed with tetrahydrofuran (THF) and dried. Yield 854 mg (62%).
1
H-NMR (CD OD) d: 1.41 (8H, m), 1.66 (2H, m), 2.00 (2H, m), 2.91
3
(
2H, t, Jϭ7.7 Hz), 4.01 (3H, s), 4.14 (2H, t, Jϭ7.7 Hz), 7.48 (1H, dd, Jϭ9.2,
and 1.8 Hz), 7.63 (1H, dd, Jϭ9.5, 2.2 Hz), 7.75 (1H, d, Jϭ9.5 Hz), 7.79 (1H,
d, Jϭ2.2 Hz), 7.83 (1H, d, Jϭ1.8 Hz), 8.45 (1H, d, Jϭ9.2 Hz). UV l
max
(
DMSO) nm (e): 425 (8170), 345 (3310), 329 (2970), 284 (44900). EI-MS
ϩ
m/z: 386 (M , Calcd for C H N O Cl : 385.94). Anal. Calcd for
2
2
28
3
1
1
C H ClN O·2HCl·1.5H O: C, 54.38; H, 6.85; N, 8.65. Found: C, 54.41;
2
2
28
3
2
H, 6.53; N, 8.29.
-(4-Methoxycarbonylphenyl)-10,15,20-tris(4-pyridyl)porphine (9)
5
A
mixture of terephthalaldehydic acid methyl ester (1.6 g, 0.01 mol) and pyri-
dine-4-aldehyde (2.7 ml, 0.03 mol) was dissolved in propionic acid (160 ml),
and was refluxed with stirring. To this solution was slowly added pyrrole
(
2.6 ml, 0.04 mol). The mixture was successively refluxed for 1.5 h. After
propionic acid was removed under reduced pressure, chloroform (100 ml)
was added to the residue, which was then neutralized with saturated
NaHCO solution. The organic phase was dried with anhydrous Na SO , and
3
2
4
then the crude material including six porphyrin isomers was chro-
matographed on silica gel (2% methanol/CHCl ). Addition of heptane to the
3
eluent and slow evaporation of the solvent afforded 9 as purple crystals.
Yield: 400 mg (5.9%)
1
H-NMR (CDCl ) d: Ϫ2.89 (2H, s), 4.12 (3H, s), 8.16 (6H, m), 8.30 (2H,
3
d, Jϭ8.1 Hz), 8.47 (2H, d, Jϭ8.1 Hz), 8.86 (8H, m), 9.06 (6H, m). UV l
max
(
(
DMSO) nm (e): 642 (2010), 587 (3420), 546 (3720), 513 (10100), 418
ϩ
191000). FAB-MS m/z: 676 (M , Calcd for C H N O : 675.76). Anal.
4
3
29
7
2
Calcd for C H N O ·0.5CHCl : C, 71.04; H, 4.04; N, 13.33. Found: C,
4
3
29
7
2
3
7
0.80; H, 4.04; N, 13.72.
-(4-Carboxyphenyl)-10,15,20-tris(4-pyridyl)porphine (10) To a so-
5
lution of 9 (400 mg, 0.592 mmol) in 40 ml of DMF was added lithium hy-
droxide monohydrate (1.20 g, 28.6 mmol) and water (5 ml), and the mixture
was stirred at room temperature for 10 h. The pH of the reaction mixture was
adjusted to 4 with 1 M HCl, and the product was extracted from the aqueous
layer with chloroform. The organic layer was washed with water twice and
dried with anhydrous Na SO . Addition of heptane and slow evaporation
Fig. 8. ICD Spectra of Hybrids and TMPyP in the Presence of CTDNA in
1
0 mM Sodium Phosphate and Various Amounts of NaCl (pH 7.0)
The concentrations of the porphyrins, CTDNA, and NaCl were as follows: top, 4.9
mM, 298 mM, and 0.1 M; middle, 5.0 mM, 298 mM, and 0.02 M; bottom, 5.0 mM, 298 mM,
and 0 M.
2
4
gave purple crystals. Yield: 340 mg (84%).
1
H-NMR (CDCl ) d: Ϫ3.02 (2H, s), 8.27 (6H, m), 8.34 (2H, d, Jϭ8.1
3
1
H-NMR (CD OD) d: 3.01 (2H, t, Jϭ6.6 Hz), 3.85 (2H, t, Jϭ6.6 Hz), Hz), 8.40 (2H, d, Jϭ8.1 Hz), 8.90 (8H, m), 9.05 (6H, m). UV lmax (DMSO)
3
3
7
8
3
.98 (3H, s), 7.31 (1H, dd, Jϭ9.2, 2.2 Hz), 7.42 (1H, dd, Jϭ9.5, 2.6 Hz),
.53 (1H, d, Jϭ2.6 Hz), 7.83 (1H, d, Jϭ9.5 Hz), 7.86 (1H, d, Jϭ2.2 Hz), FAB-MS m/z: 662 (M , Calcd for C42
.27 (1H, d, Jϭ9.2 Hz). UV lmax (DMSO) nm (e): 420 (6420), 362 (2900),
45 (2270), 329 (1300), 285 (39800). EI-MS m/z: 301 (M , Calcd for
nm (e): 642 (1790), 587 (3390), 546 (3610), 513 (11300), 418 (232000).
ϩ
H
N
O
: 661.73). Anal. Calcd for
C H N O ·H O: C, 74.21; H, 4.30; N, 14.42. Found: C, 74.02; H, 3.99; N,
42 27 7 2 2
27
7
2
ϩ
14.18.
C H ClN O: 301.78). Anal. Calcd for C H ClN O·H O: C, 60.09; H,
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminoethylaminocarbonyl]phenyl]-
10,15,20-tris(4-pyridyl)porphine (11) Compound 10 (48 mg, 73 mmol)
was dissolved in 20 ml of an absolute 50/50 (v/v) DMF–DMSO solution
under argon at 0 °C. To this solution 8 ml of an absolute 50/50 (v/v) DMF–
DMSO solution of CDI (35 mg, 0.22 mmol) was slowly added, and the reac-
16
16
3
16 16
3
2
5
.67; N, 13.14. Found: C, 60.09; H, 5.68; N, 13.19.
-(2-Aminobutyl)amino-6-chloro-2-methoxyacridine (6) 6,9-Dichloro-
-methoxyacridine (0.64 g, 2.31 mmol) was dissolved in 1,4-diaminobutane
9
2
(
8 ml) and stirred at 70 °C for 1 h. After filtration, water (100 ml) was added
to the filtrate. The yellow powder was separated, collected on a filter, and tion mixture was stirred for 6 h at 0 °C. To this reaction mixture was added
dried. Yield 472 mg (58%).
10 ml of an absolute 50/50 (v/v) DMF–DMSO solution of 5 (109 mg, 0.36
1
H-NMR (CD OD) d: 1.52 (2H, m), 1.79 (2H, m), 2.62 (2H, t, Jϭ7.3 Hz), mmol) in 20 min at 0 °C. The solution was stirred 20 h at room temperature.
3
3
.79 (2H, t, Jϭ7.3 Hz), 3.95 (3H, s), 7.25 (1H, dd, Jϭ9.4, 2.0 Hz), 7.39 (1H, Chloroform (100 ml) was poured into the solution, which was then washed
dd, Jϭ9.5, 2.6 Hz), 7.47 (1H, d, Jϭ2.6 Hz), 7.80 (1H, d, Jϭ9.5 Hz), 7.83 with saturated NaHCO
solution three times and water once, and dried with
. The solvent was removed and hybrid 11 was purified on
7740), 363 (3330), 345 (2740), 329 (1620), 285 (44700). EI-MS m/z: 329 a silica gel chromatograph (5 to 20% methanol/CHCl ). Addition of heptane
M , Calcd for C H ClN O: 329.83). Anal. Calcd for C H ClN O·H O: to the eluent and slow evaporation gave a brown powder, which was col-
3
(
(
(
1H, d, Jϭ2.2 Hz), 8.21 (1H, d, Jϭ9.4 Hz). UV lmax (DMSO) nm (e): 421 anhydrous Na SO
2 4
3
ϩ
18
20
3
18 20
3
2
C, 62.15; H, 6.38; N, 12.08. Found: C, 62.11; H, 6.05; N, 11.76.
-(2-Aminohexyl)amino-6-chloro-2-methoxyacridine (7) 6,9-Dichloro-
-methoxyacridine (0.63 g, 2.28 mmol) was dissolved in 1,6-diaminohexane 7.48 (1H, d, Jϭ8.8 Hz), 7.65 (1H, d, Jϭ9.2 Hz), 7.89 (1H, d, Jϭ9.2 Hz),
lected by centrifugation and dried. Yield: 15 mg (22%).
1
9
H-NMR (DMSO-d ) d: 3.92 (2H, q, br), 4.02 (3H, s), 4.36 (2H, m, br),
6
2
(
8 ml) and stirred at 80 °C for 1 h. After filtration, water (200 ml) was added
7.98 (1H, s), 8.09 (1H, s), 8.25 (4H, s, br), 8.68 (6H, d, Jϭ5.7 Hz), 8.70 (1H,
to the filtrate. The yellow powder was separated, collected on a filter, and
d, Jϭ8.8 Hz), 9.01 (8H, m), 9.24 (6H, d, Jϭ5.7 Hz). UV lmax (DMSO) nm
dried. Yield 611 mg (74 %).
(e): 642 (2870), 587 (5480), 546 (5750), 512 (17600), 418 (371000), 285
1
ϩ
H-NMR (CD OD) d: 1.38 (6H, m), 1.76 (2H, m), 2.55 (2H, t, Jϭ7.3 (70500). MALDI-TOF-MS m/z: 946.19 (M , Calcd for C58
H
41ClN10
O
:
3
2
Hz), 3.79 (2H, t, Jϭ7.3 Hz), 3.95 (3H, s), 7.24 (1H, dd, Jϭ9.2, 2.2 Hz), 7.38
945.49). Anal. Calcd for C58
H N O Cl ·0.5CHCl ·1.5H O: C, 68.07; H,
41 10 2 1 3 2
(
1H, dd, Jϭ9.5, 2.6 Hz), 7.47 (1H, d, Jϭ2.6 Hz), 7.80 (1H, d, Jϭ9.5 Hz), 4.35; N, 13.57. Found: C, 67.82; H, 4.37; N, 13.38.
7
.84 (1H, d, Jϭ2.2 Hz), 8.20 (1H, d, Jϭ9.2 Hz). UV lmax (DMSO) nm
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminobutylaminocarbonyl]phenyl]-

2
92
Vol. 49, No. 3
1
0,15,20-tris(4-pyridyl)porphine (12) Compound 10 (65 mg, 98 mmol) This reaction is quantitative.
was dissolved in 2 ml of an absolute 50/50 (v/v) DMF–DMSO solution
under argon at 0 °C. To this solution 2 ml of an absolute 50/50 (v/v) DMF– 10,15,20-tris(4-N-methylpyridiniumyl)porphine Triiodide (1): H-NMR
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminoethylaminocarbonyl]phenyl]-
1
DMSO solution of CDI (80 mg, 0.49 mmol) was slowly added, and the reac-
tion mixture was stirred for 6 h at 0 °C. To this reaction mixture was slowly br), 4.72 (6H, s), 4.73 (3H, s), 7.55 (1H, d, Jϭ9.2 Hz), 7.68 (1H, d, Jϭ9.9
added 2 ml of an absolute 50/50 (v/v) DMF–DMSO solution of 6 (239 mg, Hz), 7.86 (1H, d, Jϭ9.9 Hz), 7.91 (1H, s), 7.95 (1H, s), 8.24 (2H, d, Jϭ8.3
(DMSO-d ) d: Ϫ3.03 (2H, s), 3.97 (2H, q, br), 4.09 (3H, s), 4.38 (2H, m,
6
0
.687 mmol) at 0 °C. The solution was stirred 37 h at room temperature. Hz), 8.31 (2H, d, Jϭ8.3 Hz), 8.68 (1H, d, Jϭ9.2 Hz), 8.97 (6H, d, Jϭ6.6
Chloroform (70 ml) was poured into the solution, which was then washed Hz), 9.17 (8H, m), 9.43 (6H, d, Jϭ6.6 Hz). UV lmax (DMSO) nm (e): 644
with saturated NaHCO solution three times and water once, and dried with (2320), 589 (6020), 551 (6150), 516 (16900), 423 (261000), 286 (65900).
3
ϩ
anhydrous Na SO . The solvent was removed and hybrid 12 was purified on MALDI-TOF-MS m/z: 990.33 (M , Calcd for C H ClN O : 990.60).
2
4
61 50
10
2
a silica gel chromatograph (5 to 20% methanol/CHCl ). Addition of heptane Anal. Calcd for C H ClN O I ·4H O·CH I: C, 46.97; H, 3.88; N, 8.84.
3
61 50
10
2
3
2
3
to the eluent and slow evaporation gave a brown powder, which was col- Found: C, 47.07; H, 3.88; N, 9.10.
lected by centrifugation and dried. Yield: 39 mg (38%).
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminobutylaminocarbonyl]phenyl]-
1
1
H-NMR (DMSO-d ) d: Ϫ3.03 (2H, s), 1.75 (2H, m, br), 1.93 (2H, m, 10,15,20-tris(4-N-methylpyridiniumyl)porphine Triiodide (2): H-NMR
6
br), 3.45 (2H, q, br), 3.98 (3H, s), 4.00 (2H, m, br), 7.40 (1H, d, Jϭ8.4 Hz), (DMSO-d ) d: Ϫ3.03 (2H, s), 1.82 (2H, m, br), 2.08 (2H, m, br), 3.50 (2H,
.50 (1H, d, Jϭ9.0 Hz), 7.77 (1H, s), 7.81 (1H, d, Jϭ9.0 Hz), 7.86 (1H, s), q, Jϭ6.1 Hz), 4.02 (3H, s), 4.25 (2H, m, br), 4.72 (6H, s), 4.73 (3H, s), 7.60
.24 (2H, d, Jϭ8.4 Hz), 8.27 (6H, d, Jϭ5.7 Hz), 8.29 (2H, d, Jϭ8.4 Hz), (1H, d, Jϭ9.2 Hz), 7.75 (1H, d, Jϭ11.7 Hz), 7.84 (1H, d, Jϭ11.7 Hz), 7.86
6
7
8
8
.44 (1H, d, Jϭ8.4 Hz), 8.88 (1H, t, br), 8.90 (8H, s, br), 9.05 (6H, d, Jϭ5.7
(1H, s), 7.95 (1H, s), 8.32 (4H, s, br), 8.62 (1H, d, Jϭ9.5 Hz), 9.01 (6H, d,
Hz). UV l
(DMSO) nm (e): 642 (3430), 587 (6680), 546 (7070), 513 Jϭ6.6 Hz), 9.10 (8H, m), 9.49 (6H, d, Jϭ6.6 Hz). UV lmax (DMSO) nm (e):
max
ϩ
(
22500), 418 (601000), 285 (75900). MALDI-TOF-MS m/z: 974.13 (M ,
644 (2730), 589 (7510), 551 (7590), 517 (21600), 425 (313000), 286
ϩ
Calcd for C H ClN O : 973.55). Anal. Calcd for C H N O Cl · (67500). MALDI-TOF-MS m/z: 1018.82 (M , Calcd for C H ClN O :
6
0
45
10
2
60 45 10
2
1
63 54
10
2
CH OH·2H O: C, 70.34; H, 5.13; N, 13.45. Found: C, 70.36; H, 4.96; N,
1018.65). Anal. Calcd for C H ClN O I ·8H O·CH I·DMF: C, 45.76; H,
63 54 10 2 3 2 3
3
2
1
3.05.
4.59; N, 8.76. Found: C, 45.85; H, 4.27; N, 8.84.
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminohexylaminocarbonyl]phenyl]-
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminohexylaminocarbonyl]phenyl]-
1
1
0,15,20-tris(4-pyridyl)porphine (13) Compound 10 (89 mg, 0.134 10,15,20-tris(4-N-methylpyridiniumyl)porphine Triiodide (3): H-NMR
mmol) was dissolved in 10 ml of an absolute 50/50 (v/v) DMF–DMSO solu- (DMSO-d ) d: Ϫ3.02 (2H, s), 1.53 (4H, m, br), 1.70 (2H, m, br), 1.99 (2H,
6
tion under argon at 0 °C. To this solution CDI (153 mg, 0.941 mmol) was
added, and the reaction mixture was stirred for 2 h at 0 °C. To this reaction 4.74 (3H, s), 7.60 (1H, d, Jϭ9.2 Hz), 7.73 (1H, d, Jϭ11.4 Hz), 7.84 (1H, d,
mixture was added 7 (341 mg, 0.941 mmol) at 0 °C. The solution was stirred
Jϭ11.4 Hz), 7.85 (1H, s), 7.95 (1H, s), 8.33 (4H, s, br), 8.57 (1H, d, Jϭ9.2
m, br), 3.45 (2H, q, Jϭ6.2 Hz), 4.01 (3H, s), 4.14 (2H, m, br), 4.73 (6H, s),
4
7 h at room temperature. Chloroform (100 ml) was poured into the solution, Hz), 9.01 (6H, d, Jϭ6.6 Hz), 9.19 (8H, m), 9.49 (6H, d, Jϭ6.6 Hz) UV lmax
which was then washed with diluted HCl solution three times in order to re- (DMSO) nm (e): 644 (2480), 589 (6710), 551 (6810), 517 (18900), 425
move unreacted 7, and with water three times, and dried with anhydrous (298000), 285 (60800). MALDI-TOF-MS m/z: 1046.60 (M , Calcd for
ϩ
Na SO . The solvent was removed and hybrid 13 was purified on a silica gel
C H N O Cl : 1046.71). Anal. Calcd for C H ClN O I ·8H O: C,
2
4
65 58 10 2 1 65 58 10 2 3 2
chromatograph (5 to 20% methanol/CHCl ). The brown powder was precipi- 49.68; H, 4.75; N, 8.91. Found: C, 49.79; H, 4.47; N, 9.13.
3
tated by addition of diethyl ether, collected on a filter and dried. Yield: 64
5-[4-[(6-Chloro-2-methoxy-9-acridyl)aminooctylaminocarbonyl]phenyl]-
10,15,20-tris(4-N-methylpyridiniumyl)porphine Triiodide (4): H-NMR
1
mg (46%).
1
H-NMR (DMSO-d ) d: Ϫ3.03 (2H, s), 1.46 (4H, m, br), 1.64 (2H, m, (DMSO-d ) d: Ϫ3.02 (2H, s), 1.41 (8H, m, br), 1.67 (2H, m, br), 1.93 (2H,
6
6
br), 1.85 (2H, m, br), 3.64 (2H, q, br), 3.89 (2H, m, br), 3.96 (3H, s), 7.40 m, br), 3.43 (2H, q, Jϭ6.6 Hz), 3.99 (3H, s), 4.12 (2H, m, br), 4.73 (6H, s),
(
1H, d, Jϭ9.2 Hz), 7.49 (1H, d, Jϭ8.7 Hz), 7.75 (1H, s), 7.80 (1H, d, Jϭ8.7
4.74 (3H, s), 7.58 (1H, d, Jϭ9.2 Hz), 7.72 (1H, d, Jϭ9.2 Hz), 7.82 (1H, d,
Jϭ9.5 Hz), 7.84 (1H, s), 7.93 (1H, s), 8.34 (4H, s, br), 8.54 (1H, d, Jϭ9.2
Hz), 9.01 (6H, d, Jϭ6.6 Hz), 9.09 (8H, m), 9.25 (6H, d, Jϭ6.6 Hz). UV lmax
Hz), 7.85 (1H, s), 8.26 (6H, d, Jϭ5.9 Hz), 8.27 (2H, d, Jϭ8.4 Hz), 8.30 (2H,
d, Jϭ8.4 Hz), 8.40 (1H, d, Jϭ9.2 Hz), 8.82 (1H, t, br), 8.90 (8H, m, br), 9.04
(
(
1
6H, d, Jϭ5.9 Hz). UV lmax (DMSO) nm (e): 642 (3540), 587 (6770), 546 (DMSO) nm (e): 644 (2450), 589 (6580), 551 (6700), 517 (18800), 425
ϩ
7140), 513 (22100), 418 (486000), 284 (66700). MALDI-TOF-MS m/z: (287000), 285 (62600). MALDI-TOF-MS m/z: 1074.19 (M , Calcd for
ϩ
001.7 (M , Calcd for C H ClN O : 1001.60). Anal. Calcd for C H ClN O : 1074.76). Anal. Calcd for C H ClN O I ·7H O: C, 50.88;
6
2
49
10
2
67 62
10
2
67 62
10
2
3
2
C H N O Cl ·2H O: C, 71.77; H, 5.15; N, 13.50. Found: C, 71.55; H, H, 4.84; N, 8.86. Found: C, 50.88; H, 4.99; N, 9.08.
62
49 10
2
1
2
4
.87; N, 13.53.
DNA Photocleavage Assay Photoirradiation was performed at 25 °C
5
-[4-[(6-Chloro-2-methoxy-9-acridyl)aminooctylaminocarbonyl]phenyl]- using a Hitachi 650—660 fluorescence spectrophotometer. A sample con-
1
0,15,20-tris(4-pyridyl)porphine (14) Compound 10 (88 mg, 0.133 mmol)
taining supercoiled pUC18 plasmid DNA (57.9 mM) and a hybrid (1.0 mM)
was irradiated in a buffer (10 mM sodium phosphate, pH 7.0) at 526 nm for
was dissolved in 10 ml of an absolute 50/50 (v/v) DMF–DMSO solution
under argon at 0 °C. To this solution CDI (151 mg, 0.931 mmol) was added, 30 min. After irradiation, DNA was analyzed by 0.8% agarose gel elec-
and the reaction mixture was stirred for 2 h at 0 °C. To this reaction mixture trophoresis at 100 V for 45 min. The gel was incubated in a solution of ethid-
was added 8 (453 mg, 0.931 mmol) and 4-ethylmorpholine (355 ml, 2.79
mmol) at 0 °C. The solution was stirred 40 h at room temperature. Chloro-
form (100 ml) was poured into the solution, which was then washed with di-
luted HCl solution three times and water three times, and dried with anhy-
drous Na SO . The solvent was removed and hybrid 14 was purified on a sil-
ium bromide for 5 min and the DNA bands were detected by fluorescence
under a UV lamp. The densitometric data of the bands were obtained with
an ATTO Densitograph version 4.0 for Macintosh.
Equilibrium Dialysis Assay A volume of 100 ml of the dialysate buffer
solution (10 mM sodium phosphate, pH 7.0) containing 1.0 mM hybrid was
2
4
ica gel chromatograph (5 to 20% methanol/CHCl ). The brown powder was placed in a beaker. Extinction coefficient of the Soret band of the hybrid was
3
precipitated by addition of diethyl ether, collected by centrifugation and then spectrophotometrically determined. A 0.5 ml buffer solution (10 mM
dried. Yield: 33 mg (24%).
sodium phosphate, pH 7.0) containing CTDNA (58.0 mM) was pipetted into
1
H-NMR (DMSO-d ) d: Ϫ3.02 (2H, s), 1.33 (8H, m, br), 1.61 (2H, m, a 0.5 ml Spectro/Por DispoDialyzer tube, which was then soaked in the
6
br), 1.78 (2H, m, br), 3.39 (2H, q, br), 3.81 (2H, m, br), 3.93 (3H, s), 7.35 dialysate buffer solution. The beaker was covered with Parafilm and
(
1H, d, Jϭ9.2 Hz), 7.43 (1H, d, Jϭ9.2 Hz), 7.68 (1H, s), 7.78 (1H, d, Jϭ9.2
wrapped in foil, and its content was allowed to equilibrate with stirring for
Hz), 7.83 (1H, s), 8.26 (6H, d, Jϭ5.9 Hz), 8.27 (2H, d, Jϭ8.6 Hz), 8.30 (2H, 24 h at 25 °C. After the completion of equilibration, the DNA sample was
d, 8.6 Hz), 8.35 (1H, d, Jϭ9.2 Hz), 8.81 (1H, t, br), 8.90 (8H, m, br), 9.04 transferred to a tube, and was taken to a final concentration of 1% sodium
(
(
1
6H, d, Jϭ5.9 Hz). UV lmax (DMSO) nm (e): 642 (3730), 587 (7170), 546 dodecyl sulfate by the addition of its 10% stock solution. The total concen-
7580), 513 (23600), 418 (519000), 285 (67800). MALDI-TOF-MS m/z: tration of the hybrid (C ) within the dialysis tube was determined spec-
t
ϩ
029.8 (M , Calcd for C H ClN O : 1029.65). Anal. Calcd for trophotometrically using wavelength and extinction coefficient of the hybrid.
6
4
53
10
2
C H ClN O ·1.5H O: C, 72.75; H, 5.34; N, 13.26. Found: C, 72.83; H,
The free hybrid concentration (C ) was determined spectrophotometrically
64
53
10
2
2
f
5
.13; N, 13.25.
using an aliquot of the dialysate solution. The amount of bound hybrid (Cb)
General Procedure for Methylation of Hybrids Hybrids 11—14 were was determined by the difference, C ϭC ϪC .
b
t
f
methylated in 5 ml of DMF with methyl iodide (0.8 ml) for 3 h at room tem-
Spectral Measurements Aliquots of a CTDNA solution were added to
perature. The solvent and methyl iodide were removed under vacuum. The the solution of a hybrid (5.0 mM). The spectral measurement was performed
residue was taken up in DMF and precipitated with diethyl ether. The brown
at 25 °C in a buffer containing 10 mM sodium phosphate and 100 mM sodium
powder was collected by centrifugation, washed with diethyl ether and dried. chloride (pH 7.0). All spectral data were corrected for the dilution effect.

March 2001
293
The binding constant (K) and the number of base pairs occupied by the 13) Boyle R. W., Dolphin D., Photochem. Photobiol., 64, 469—485
bound drug (n) were estimated from the absorbance change of the Soret
(1996).
maximum, using the Scatchard equation:
14) Alex His. R., Rosenthal D. I., Glatstein E., Drugs, 57, 725—734
(
1999).
r/cϭK (nϪr)
1
5) Lown J. W., Sondhi S. M., Ong Chi-Wi, Biochemistry, 25, 5111—5117
where r is the binding ratio and c is the concentration of unbound hybrid.
(1986).
Values for r and c were calculated according to the treatment of Peacocke 16) Uno T., Sowa N., Shimabayashi S., Chem. Pharm. Bull., 42, 988—990
2
9)
and Skerrett. The linear fitting of r vs. r/c was carried out using a Kaleida-
Graph version 3.0. To a buffered aqueous solution (10 mM sodium phosphate
and 100 mM sodium chloride, pH 7.0) of CTDNA (about 30 mM) was added
(1994).
17) Drexler C., Hossieini M. W., Pratviel G., Meunier B., Chem. Com-
mun., 1998, 1343—1344.
aliquots of the TMPyP or hybrid solution, and CD spectra in the UV range 18) Denny W. A., Baguley B. C., “Molecular Aspects of Anticancer Drug-
were recorded.
DNA Interactions,” Vol. 2, ed. by Neidle S., Waring M., CRC Press
Inc., Boca Raton, 1994, pp. 270—311.
Acknowledgments This research was supported in part by a Grant-in- 19) Todd A. K., Adams A., Thorpe J. H., Denny W. A., Wakelin L. P. G.,
Aid to T. U. for Scientific Research from the Ministry of Education, Science,
Sports and Culture of Japan (grant no. 11470476).
Cardin C. J., J. Med. Chem., 42, 536—540 (1999).
20) Lown J. W., Joshua A. V., J. Chem. Soc., Chem. Commun., 1982,
1
298—1300.
References and Notes
21) Ren J., Chaires J. B., Biochemistry, 38, 16067—16075 (1999).
1
)
Meunier B. (ed.), “DNA and RNA Cleavers and Chemotherapy of 22) The reason that TMPyP, showing the highest quantity among the por-
Cancer and Viral Diseases,” NATO ASI Series, Kluwer Academic
Publishers, London, 1996.
Waring M. J., Ann. Rev. Biochem., 50, 159—192 (1981).
Natrajan A., Hecht S. M., “Molecular Aspects of Anticancer Drug-
DNA Interactions,” Vol. 2, ed. by Neidle S., Waring M., CRC Press
Inc., Boca Raton, 1994, pp. 197—242.
phyrins, exhibited lower efficiency of the conversion than the hybrids
is not yet understood; TMPyP can bind to DNA by both intercalative
and groove binding modes, thus there may be a possibility that interca-
lation was an inert mode of photo-nuclease activity of TMPyP in our
experiment.
2
3
)
)
23) Pasternack R. F., Gibbs E. J., Villafranca J. J., Biochemistry, 22,
2406—2414 (1983).
4
5
6
7
)
)
)
)
Burger R. M., Chem. Rev., 98, 1153—1170 (1998).
Claussen C. A., Long E. C., Chem. Rev., 99, 2797—2816 (1999).
Fiel R. J., J. Biomol. Struc. Dyn., 6, 1259—1274 (1989).
Pasternack R. F., Gibbs E. J., Met. Ions Biol. Syst., 33, 367—397
24) Scatchard G., N.Y. Acad. Sci., 51, 660—672 (1949).
25) Rodger A., Norden B., “Circular Dichroism and Linear Dichroism,”
Oxford Chemistry Masters, Oxford University Press, Oxford, New
York, Tokyo, 1997
26) Zimmer C., Luck G., “Advances in DNA Sequence Specific Agents,”
Vol. 1, ed. by Hurley L. H., JAI Press Inc., Greenwich, 1992, pp. 51—
88.
(
1996).
Fiel R. J., Datta-Gupta N., Mark E. H., Howard J. C., Cancer Res., 41,
543—3545 (1981).
8
)
)
3
9
Praseuth D., Gaudemer A., Verlhac J.-B., Kraljic I., Sissoeff I., Guille
E., Photochem. Photobiol., 44, 717—724 (1986).
0) Munson B. R., Fiel R. J., Nucleic Acids Res., 20, 1315—1319 (1992).
1) Marzilli L. G., New. J. Chem., 14, 409—420 (1990).
27) Uno T., Hamasaki K., Tanigawa M., Shimabayashi S., Inorg. Chem.,
36, 1676—1683 (1997).
28) Carvlin M. J., Fiel R. J., Nucleic Acids Res., 11, 6121—6139 (1983).
1
1
1
2) Villanueva A., Juarranz A., Diaz V., Gomez J., Canete M., Anti-Cancer 29) Peacocke A. R., Skerrett J. N., J. Chem. Soc., Faraday Trans., 52,
Drug Des., 7, 297—303 (1992). 261—279 (1956).