Novel ebselen-porphyrin conjugates: Synthesis and nucleic acid interaction study

Article abstract of DOI:10.1016/j.bmcl.2007.05.074

Novel porphyrins bearing ebselen (2-phenyl-1,2-benzisoselenazol-3[2H]-one) moiety were synthesized and characterized. Their interactions with herring sperm DNA were studied by means of UV-visible, fluorescence, circular dichroism spectroscopy, and gel electrophoresis.

Full text of DOI:10.1016/j.bmcl.2007.05.074

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4266–4270
Novel ebselen–porphyrin conjugates: Synthesis
and nucleic acid interaction study
Zhi Xue,a An-Xin Hou,a,* Daniel Wei-Jing Kwongb and Wai-Kwok Wonga,b,
*
aDepartment of Chemistry, Wuhan University, Wuhan 430072, PR China
bDepartment of Chemistry, Hong Kong Baptist University, Hong Kong, PR China
Received 4 February 2007; revised 10 April 2007; accepted 10 May 2007
Available online 26 May 2007
Abstract—Novel porphyrins bearing ebselen (2-phenyl-1,2-benzisoselenazol-3[2H]-one) moiety were synthesized and characterized.
Their interactions with herring sperm DNA were studied by means of UV–visible, fluorescence, circular dichroism spectroscopy, and
gel electrophoresis.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Investigations of porphyrin–nucleic acid interactions
have been the subject of extensive research during the
past decade and a great many of important results were
achieved.1 The study on porphyrin–nucleic acid interac-
tions was largely motivated by such potential applica-
tion as photodynamic inactivation of microorganisms
and viruses, the probing of the secondary and tertiary
structures of nucleic acids, and DNA-targeted antitu-
mor activity.2 For the porphyrin–DNA interaction,
three binding modes (intercalative binding, external
groove binding, and outside binding with self-stacking
along the DNA helix) were widely accepted.3 Previous
studies have attributed the anticancer activities of cat-
ionic porphyrins to their intercalative binding with
DNA.4 But DNA-binding porphyrins included cationic,
neutral as well as anionic porphyrins.5 Thus, it is imper-
ative to study the interactions of DNA with a wide vari-
ety of porphyrins. Results from these studies will be
valuable toward the development of porphyrin-based
compounds as novel DNA-targeted anti-tumor or anti-
biotic drugs.
1,2-benzisoselenazol-3-[2H]-one) has been studied exten-
sively.6 The combination of porphyrin with ebselen
might provide stronger anti-tumor activity due to the
fact that porphyrins have specific affinity for tumor cells,
which would enable the ebselen moiety to accumulate
efficiently inside the tumor cells. Furthermore, it is rea-
sonable to expect that the conjugate might display syn-
ergistic or additive anti-tumor effects due to the two
moieties.
The synthetic route for 5-ebselenyl-10,15,20-triphenyl-
porphyrin is shown in Scheme 1. Reduction of elemental
selenium with potassium borohydride followed by treat-
ment with diazotized anthranilic acid gave 2,20-diselen-
obis(benzoic acid), which reacted with thionyl chloride
to give 2-chloroselenobenzoyl chloride (1).7 Meanwhile,
the key of the synthesis is the preparation of 5-(4-amino-
phenyl)-10,15,20-triphenylporphyrin (2). In the classical
Adler’s condensation, neither 4-nitrobenzaldehyde nor
4-aminobenzaldehyde condensed with pyrrole to give
2. However, 2 could be synthesized in good yield via
nitration followed by reduction of tetraphenylporphyrin
(TPPH2).8 Reaction between 1 and 2 gave the targeted
conjugate 3, which bore an ebselen moiety. The progress
of the reaction was monitored by TLC. The targeted
complex had a slightly higher Rf than 2, in CHCl3.
The product 3 was air- and moisture-stable and could
be purified by flash column chromatography. The metal-
lo-ebselenylporphyrins (M = Co(II), Cu(II), Mn(II), and
Zn(II)) were obtained by metalation of 2 first, then
followed by reaction with 1. The resulting metallo-ebsel-
enylporphyrins (4–7) were readily soluble in DMF.
Herein we report the synthesis of a series of novel ebse-
len–porphyrin conjugates and the results of their inter-
actions
with
DNA,
including
their
DNA
photocleavage activities. The anti-oxidant, anti-inflam-
matory and anti-tumor activity of ebselen (2-phenyl-
Keywords: Porphyrins; Ebselen; HS DNA; Binding mode.
*
Corresponding authors. Tel.: +86 27 6875 2323; fax: +86 27 6875 4067
(A.-X.H.); tel.: +852 3411 7011; fax: +852 3411 7348 (W.-K.W.);
e-mail addresses: houanxin@yahoo.com; wkwong@hkbu.edu.hk
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.05.074

Z. Xue et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4266–4270
4267
Scheme 1. Synthetic routes for porphyrins with ebselen moiety.
Table 1. UV–visible and fluorescence spectral changes of the ebselen–
porphyrins (3–7) upon titration with HS DNA
ESI-MS analysis confirmed the metalation of the
ebselen–porphyrin complexes. Due to the paramagnetic
nature of porphyrins (4–6), their 1H NMR spectral data
were not available. The synthetic procedure and spectro-
scopic data of porphyrins 3 and 7 are given in the
reference.9
Porphyrins
UV–vis data of the Soret band
Red shift (nm) Hypochromicity (%) Intensity
enhancement
(%)
12.8
Fluorescence
3
4
5
6
7
3
9
1.5
31.7
The interaction of porphyrins 3–7 with herring sperm
DNA (HS DNA) was evaluated using UV–visible
absorption and fluorescence spectroscopy. To enhance
solubility, the porphyrins were first dissolved in DMF
and then diluted to 10 lM with deionized water. The
diluted porphyrin solutions were titrated with increasing
concentrations of HS DNA in Tris–HCl buffer and the
volume percentage of DMF was kept to less than 5%.
The concentration of the HS DNA solution, in terms
of base pairs, was determined by its absorbance at
88.9
75.1
2
5.1
ꢀ66.6
84.7
2
15
184.2
38.5
The porphyrin–DNA mixture contained 10 lM of the ebselen–por-
phyrin and different molar ratios of DNA, CDNA/Cporphyrin = 0–10. A
negative hypochromicity means a corresponding hyperchromicity. All
experiments were performed at room temperature in buffer solution
(0.05 M Tris–HCl, 0.1 M NaCl, pH = 7.4).
260 nm using the molar absorptivity, eabs
,
of
1.31 · 104 Mꢀ1 cmꢀ1. The excitation (kex) and emission
(kem) wavelengths of the porphyrins were determined
with the diluted porphyrin solutions (10 lM) in the
absence of DNA. For 7, the kex and kem are 412 and
650 nm, respectively, in 0.05 M Tris–HCl buffer (pH
7.4) and 0.1 M NaCl. In the fluorescence titration exper-
iments, after addition of DNA, the resultant solution
was stirred vigorously and maintained at room temper-
ature for 30 min before measurement was taken. The
spectral changes upon DNA addition were different
for different porphyrins, ranging from a modest 13%
increase in intensity in 3 to ca. 185% enhancement in
6. These results, together with the UV–visible spectral
changes, are summarized in Table 1. In the UV–visible
titration experiments, 3 and 5 gave no significant
changes, 4 and 7 showed a relatively large red shift
(9 nm and 15 nm, respectively) and hypochromicity
(31.7% and 84.7%, respectively), and 6 gave a small
red shift (2 nm) but a substantial hyperchromicity
(66.6%). Among the three binding modes demonstrated
for porphyrins, DNA intercalation is characterized by a
large red shift (>10 nm) and a significant hypochromic-
ity (up to 40%) in its Soret band; external groove bind-
ing by a minor (or no) spectral shift and an occasional
hyperchromicity; and outside binding with self-stacking
sharing similar characteristics with the intercalative
binding mode.2 According to this criterion, 7 appears
to intercalate into DNA and 4 binds to DNA via outside
binding with self-stacking. The remaining three ebselen–
porphyrins seem to exhibit external groove binding
mode. It is well known that the nature of the central me-
tal ions of metalloporphyrins could influence their bind-
ing characteristics with DNA.1b,10 This can partly
explain the different binding modes shown by the five

4268
Z. Xue et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4266–4270
ebselen–porphyrins. The UV–visible titration spectra of
7 are shown in Figure 1.
Intercalative binding mode was characterized by large
binding constant as well. The apparent binding constant
(Kapp) could be calculated according to Eq. (1):
ꢀ
ꢁ
½DNAꢁ
1
total
¼
½DNAꢁ
total
ðjeapp ꢀ ef jÞ
ðjeb ꢀ ef jÞ
1
È
É
þ
ð1Þ
Kappðjeb ꢀ ef jÞ
where eapp, ef and eb correspond to Aobsd/[porphyrin], the
extinction coefficient for the free porphyrin, and the
extinction coefficient for the porphyrin in the fully
Figure 2. Fluorescence spectra (kex = 412 nm) of 7 (10 lM) in the
absence (—) and the presence (- - -) of HS DNA (CDNA/Cporphyrin = 10)
in 0.05 M Tris–HCl buffer (pH 7.4) and 0.1 M NaCl.
bound form, respectively. From a plot of [DNA]total
/
(jeapp ꢀ efj) versus [DNA]total, shown in the inset of
Figure 1, Kapp is obtained by the ratio of the slope to
the intercept.11 Kapp of 7 was determined to be
7.40 · 105 Mꢀ1, which compared favorably with the
binding constant in the intercalative binding between
TMPyP (TMPyP = tetrakis(4-methylpyridiniumyl)por-
phyrin) and poly[d(G-C)2].1b
the Soret region of porphyrins can be used as a signature
for its binding mode with DNA: a positive induced CD
band is indicative of external groove binding, whereas a
negative induced CD band is indicative of intercalative
binding.1b,15 To further investigate the binding modes,
the CD spectra of these five porphyrins (3–7) upon addi-
tion of HS DNA were measured. Results showed that
only 7 exhibited a negative CD signal in the Soret band
(Fig. 3), while the other four porphyrins (3–6) exhibited
a positive CD signal. These CD results corroborate with
the results obtained from UV–vis and fluorescence titra-
tion experiments: in their binding interactions with HS
DNA, 7 binds with an intercalative mode, while 3–6
bind externally, with 4 undergoing self-stacking along
the DNA helix and 3, 5, and 6 presumably prefer groove
binding with DNA.
Fluorescence spectroscopy was also applied to study
porphyrin–DNA interactions.12 Fluorescence intensity
of the five ebselen–porphyrins was enhanced to various
extents upon addition of HS DNA. These data are given
in Table 1. While four ebselen–porphyrins, 3–6, showed
only minor spectral shifts, kem,max of 7 was substantially
blue-shifted (Dk = 24 nm) in the presence of DNA
(Fig. 2). This unique feature might be indicative of the
formation of a novel porphyrin–DNA complex. An
enhanced fluorescence was once believed to be one of
the criteria for intercalative binding.1c,13 However, sub-
sequent studies revealed that external groove binding
showed similar changes as well.14 Thus, intercalative
binding mode could not be inferred solely on the basis
of enhanced fluorescence upon DNA addition. Previous
studies on the interactions of porphyrins and metallo-
porphyrins with DNA suggested that at low concentra-
tions, the sign of the induced circular dichroism (CD) in
DNA photocleavage activities of these ebselen–porphy-
rins were also measured using the plasmid DNA relaxa-
tion assay.11 The gel image is given in Figure 4. At
concentrations up to 20 lM (limited by their water sol-
ubility), only modest DNA photocleavage activities,
ranging from 15 to 25%, were seen in the four ebse-
Figure 1. UV–visible spectra of porphyrin 7 (10 lM) in the absence (- - -)
and presence (—) of increasing molar ratios of HS DNA (CDNA
/
Cporphyrin = 0, 0.1, 0.2, 0.4, 0.8, 1.0, 2.0, 4.0, 10.0). Hypochromism at
427 nm was observed at low CDNA/Cporphyrin ratios but red-shifted to
near 440 nm at high CDNA/Cporphyrin ratio. The inset is a plot of
Figure 3. Induced circular dichroism spectrum of 7 with HS DNA (a)
in the absence of DNA, (b) in the presence of DNA) [porphy-
rin] = 5 lM, CDNA/Cporphyrin = 20.0. The spectrum was recorded in
0.05 M Tris–HCl buffer, pH = 7.4, and 0.1 M NaCl.
[DNA]total/(jeapp ꢀ efj) versus [DNA]total
.

Z. Xue et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4266–4270
4269
Acknowledgments
W.-K.W. thanks the Hong Kong Baptist University
(FRG/05-06/II-03) and Hong Kong Research Grants
Council (HKBU 2021/03P) for financial support of this
work. We thank the reviewers for their critical com-
ments on the manuscript.
References and notes
1. (a) Fiel, R. J.; Howard, J. C.; Datta, G. N. Nucleic Acids
Res. 1979, 6, 3093; (b) Pasternack, R. F.; Gibbs, E. J.;
Villafranca, J. Biochemistry 1983, 22, 2406; (c) Sehlstedt,
U.; Kim, S. K.; Carter, P.; Goodisman, J.; Vollano, J. F.;
Norden, B.; Dabrowiak, J. C. Biochemistry 1994, 33, 417;
(d) Sari, M. A.; Battioni, J. P.; Mansuy, D.; Le Pecq, J. B.
Biochem. Biophys. Res. Commun. 1986, 141, 643; (e) Ford,
K. G.; Neidle, S. Bioorg. Med. Chem. 1995, 3, 671.
2. (a) Sol, V.; Branland, P.; Chaleix, V.; Granet, R.;
Guilloton, M.; Lamarche, F.; Verneuil, B.; Krausz, P.
Bioorg. Med. Chem. Lett. 2004, 14, 4207; (b) Weisman-
Shomer, P.; Cohen, E.; Hershco, I.; Khateb, S.; Wolfovitz-
Barchad, O.; Hurley, L. H.; Fry, M. Nucleic Acids Res.
2003, 31, 3963; (c) Anton, B. G.; Neocles, B. L. Biochem-
istry 1999, 38, 15425.
3. (a) Fiel, R. J. J. Biomol. Struct. Dynam. 1989, 6, 1259; (b)
Lugo-Ponce, P.; McMillin, D. R. Coord. Chem. Rev. 2000,
208, 169.
4. (a) Villanueva, A.; Juarranz, A.; Diaz, V.; Gomez, J.;
Canete, M. Anticancer Drug Des. 1992, 7, 297; (b)
Pasternack, R. F.; Gibbs, E. J. Met. Ions Biol. Syst.
1996, 33, 367.
Figure 4. Agarose gel electrophoresis of DNA photocleavage assay of
(a) porphyrin 7 and 3; and (b) porphyrin 4 and 5. In (a) lane 1:
plasmid DNA (pBluescript); lane 2: 7 (1 lM) + DNA; lane 3: 7
(5 lM) + DNA; lane 4: 7 (10 lM) + DNA; lane 5: 7 (20 lM) + DNA;
lane 6:
(10 lM) + DNA; lane 9: 3 (20 lM) + DNA. In (b) lane 1: plasmid
DNA (pBluescript); lane 2: (1 lM) + DNA; lane 3:
(5 lM) + DNA; lane 4: 5 (10 lM) + DNA; lane 5: 5 (20 lM) + DNA;
lane 6: (1 lM) + DNA; lane 7: (5 lM) + DNA; lane 8:
3 (1 lM) + DNA; lane 7: 3 (5 lM) + DNA; lane 8: 3
5
5
4
4
4
(10 lM) + DNA; lane 9: 4 (20 lM) + DNA. Photo-irradiation was
conducted using a transilluminator at 455 nm for 45 min under
ambient temperature.
5. (a) Balaz, M.; Steinkruger, J. D.; Ellestad, G. A.; Berova,
N. Org. Lett. 2005, 7, 5613; (b) Lauceri, R.; Purrello, R.;
Shetty, S. J.; Vicente, M. G. H. J. Am. Chem. Soc. 2001,
123, 5835.
6. (a) Moussaoui, S.; Obinu, M. C.; Daniel, N.; Reibaud, M.;
Blanchard, V.; Imperato, A. Exp. Neurol. 2000, 166, 235;
(b) Parnham, M.; Sies, H. Expert Opin. Investig. Drugs
2000, 9, 607.
len–porphyrins studied, that is, porphyrin 4 (15%), 3
(20%), 7 (20%), and 5 (25%).
In summary, we have synthesized a series of metallo-
ebselenylporphyrins. Using UV–vis, fluorescence, and
circular dichroism techniques, their binding properties
with HS DNA were investigated. While the metal-free
ebselen–porphyrin and its Co(II), Cu(II), and Mn(II)
complexes demonstrate outside binding with DNA, the
Zn(II) complex showed an intercalative binding mode.
This result stands in contrast to the more extensively
studied meso-substituted cationic Zn(II)–TMPyP com-
plex, which binds externally to the DNA, presumably
because of the steric constraint from an axial aquo li-
gand.1b The reason for such distinct behavior was not
apparent at present. However, it should be pointed out
that two Zn(II) non-meso-substituted cationic porphy-
rins have recently been reported to bind to DNA via
intercalation as well.15 Despite the modest DNA
photocleavage activities observed for 3, 4, 5, 7, which
only show that they are not efficient in causing DNA
strand breaks, more definitive photodynamic assay
using appropriate cancer cells will be conducted to prop-
erly evaluate the potential of these compounds as antitu-
mor agents. With respect to some porphyrins bearing
bioactive groups which exhibited efficient antibacterial
activity,16 these compounds will be also applied to other
bioassay such as antibacterial or anti-inflammation
activity.
7. Kamigata, N.; Izuka, H.; Izuoka, A.; Kobaysahi, M. Bull.
Chem. Soc. Jpn. 1986, 59, 2179.
8. Kruper, W. J.; Chamberlin, T. A.; Kochanny, M. K.
J. Org. Chem. 1989, 54, 2753.
9. Selected procedure: Under an argon atmosphere, fuming
HNO3 (0.32 g, 5.0 mmol, fuming) was added dropwise
over a period of 20 min at 0 ꢁC to a solution of
tetraphenylporphyrin (TPPH2) (1.0 g, 1.62 mmol) in
CHCl3 (150 cm3). The reaction was completed within half
an hour. The reaction mixture was then washed thrice with
distilled water (3 · 100 cm3) to remove the excess HNO3.
The CHCl3 layer was then dried over MgSO4, filtered, and
evaporated to dryness. Then the crude product was
chromatographed (silica gel: 100 g, eluent: CHCl3/petro-
leum ether 2:1) and crystallized from methanol. The
second band was gathered to give 5-(4-nitrophenyl)-
10,15,20-triphenylporphyrin (p-NO2-TPPH2) in 50% yield.
p-NO2-TPPH2 (2.00 g, 3.03 mmol) was dissolved in 60 cm3
of concentrated hydrochloric acid under nitrogen. Tin(II)
chloride dihydrate (2.08 g, 9.24 mmol) was added to the
solution. The reaction solution was heated to 65 ꢁC for 1 h
and then poured into a beaker filled with ice for cooling.
Sodium hydroxide was added to the mixture with vigorous
stirring until the color of the solution turned red brownish.
The solution was then extracted several times with
chloroform. The combined chloroform solution was
evaporated to dryness. The residue was purified by column

4270
Z. Xue et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4266–4270
cence (H2O): kex = 412 nm, kem = 650 nm; IR (KBr, cmꢀ1
chromatography (silica gel: 200 g, eluent: CHCl3/petro-
)
leum ether 4:1). Crystallization in methanol gave the free
porphyrin NH2-TPPH2 in 60% yield. The corresponding
metalloporphyrin was synthesized by the addition of
excess M(OAc)2 in methanol to the free porphyrin in
chloroform at reflux, and purified by column chromato-
graphy. A chloroform solution containing compound 1
(30 mg, 0.12 mmol) and freshly distilled triethylamine
(28 lL, 0.2 mmol)) was added dropwise to a chloroform
solution of NH2-TPPH2 (50 mg, 0.08 mmol) at room
temperature under an argon atmosphere. The progress of
the reaction was monitored by TLC. When the starting
material had completely reacted, the reaction mixture was
washed by saturated NaHCO3 solution and deionized
water. The chloroform layer was separated and evapo-
rated to dryness under vacuum. The crude product was
purified by flash column chromatography to give 3 and
metallo products (4–7). Selected data: compound 3: UV–
vis (CH2Cl2): kmax (nm, loge) 418 (4.42), 368 (4.08);
1597 (C@O); HRMS (ESI, m/z) (calculated 873.0895,
found 873.1013); 1H NMR (DMSO-d6, 300 MHz): d 8.75–
8.91 (8H, pyrrolic), d 6.97–8.18 (23H, ArH).
10. Daryono, H. T.; Shunsuke, M.; Takehiro, A.; Naoki, Y.;
Hidenari, I. J. Inorg. Biochem. 2001, 85, 219.
11. Mettath, S.; Munson, B. R.; Pandey, R. K. Bioconjug.
Chem. 1999, 10, 94.
12. (a) Le Pecq, J. B.; Paoletti, J. B. C. J. Mol. Biol. 1967, 27,
87; (b) Pasternack, R. F.; Caccam, M.; Keogh, B.;
Stephenson, T. A.; Williams, A. P.; Gibbs, E. J. J. Am.
Chem. Soc. 1991, 113, 6835.
13. (a) Sari, M. A.; Battioni, J. P.; Duprt, D.; Mansuy, D.; Le
Pecq, J. B. Biochemistry 1990, 29, 4205; (b) Zupan, K.;
Herenyi, L.; Toth, K.; Majer, Z.; Csik, G. Biochemistry
2004, 43, 9151.
14. Hyun, K. M.; Choi, S. D.; Lee, S.; Kim, S. K. Biochim.
Biophys. Acta 1997, 1334, 312.
15. Yellappa, S.; Seetharamappa, J.; Rogers, L. M.; Chitta,
R.; Singhal, R. P.; D’Souza, F. Bioconjug. Chem. 2006, 17,
1418.
16. (a) Li, H. P. Bioorg. Med. Chem. Lett. 2006, 16, 6298; (b)
Zhang, J. Y.; Wu, X. J.; Cao, X. P.; Yang, F.; Wang, J. F.;
Zhou, X.; Zhang, X. L. Bioorg. Med. Chem. Lett. 2003, 13,
1097.
Fluorescence:
cmꢀ1
kex = 417 nm, kem = 646 nm; IR (KBr,
)
1606 (C@O); HRMS (ESI, m/z) (calculated
811.1850, found 811.1901); 1H NMR (CDCl3,
300 MHz): d 8.85–8.93 (8H, pyrrolic), d 7.56-8.30 (23H,
ArH),
d
ꢀ2.77 (inner NH). Compound 7: UV–vis
(CH2Cl2): kmax (nm, loge) 548 (4.11), 421 (5.63); Fluores-