3-Alkyl- and 3-aryl-7H-furo[3,2-g]-1-benzopyran-7-ones: synthesis, photoreactivity, and fluorescence properties.

Article abstract of DOI:10.1002/1521-4184(200205)335:5<187::AID-ARDP187>3.0.CO;2-E

A series of 3-alkyl- and 3-aryl-7H-furo[3, 2-g]-1-benzopyran-7-ones, known as linear furocoumarins, was synthesized and evaluated for their dark- and photobinding (crosslink formation) with DNA as well as for their spectrophotometric and fluorescent properties, lipophilicity, and ability to photobleach N, N-dimethyl-p-nitrosoaniline (RNO) after irradiation with UVA light. 8-Methoxypsoralen (8-MOP, 9-methoxy-7H-furo[3, 2-g]-1-benzopyran-7-one) and 4, 5', 8-trimethylpsoralen (TMP, 2, 5, 9-trimethyl-7H-furo[3, 2-g]-1-benzopyran-7-one) were used as reference compounds in all tests. The investigations support the formation of a molecular complex between the furocoumarins and DNA.Crosslink formation with DNA after irradiation with UVA light was detectable for compounds with a methyl or phenyl substituent in position 3, but not for those bearing either a tert-butyl, a 4-methoxyphenyl, or a 2, 5-dimethoxyphenyl group. All furocoumarins exhibited sufficient absorption in the UVA wavelength range and are fluorescent. All compounds showed a higher lipophilicity than 8-MOP. Generally the 3-alkyl substituted furocoumarins had a capacity to photobleach RNOwhich was higher than that of the 3-aryl substituted ones. Some of the 3-aryl substituted furocoumarins displayed a photobleaching ability which was similar to or lower than that of 8-MOP.

Full text of DOI:10.1002/1521-4184(200205)335:5<187::AID-ARDP187>3.0.CO;2-E

Arch. Pharm. Pharm. Med. Chem. 2002, 5, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 187
Paul Körner
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzopyran-
7-ones:
Pharmazeutisches Institut,
Universität Kiel, Kiel, Germany
Synthesis, Photoreactivity, and Fluorescence
Properties
A series of 3-alkyl- and 3-aryl-7H-furo[3,2-g]-1-benzopyran-7-ones, known as linear
furocoumarins, was synthesized and evaluated for their dark- and photobinding
(crosslink formation) with DNA as well as for their spectrophotometric and fluores-
cent properties, lipophilicity, and ability to photobleach N,N-dimethyl-p-ni-
trosoaniline (RNO) after irradiation with UVA light. 8-Methoxypsoralen (8-MOP, 9-
methoxy-7H-furo[3,2-g]-1-benzopyran-7-one) and 4,5Ј,8-trimethylpsoralen (TMP,
2,5,9-trimethyl-7H-furo[3,2-g]-1-benzopyran-7-one) were used as reference com-
pounds in all tests.The investigations support the formation of a molecular complex
between the furocoumarins and DNA.Crosslink formation with DNA after irradiation
with UVA light was detectable for compounds with a methyl or phenyl substituent in
position 3, but not for those bearing either a tert-butyl, a 4-methoxyphenyl, or a 2,5-
dimethoxyphenyl group.All furocoumarins exhibited sufficient absorption in the UVA
wavelength range and are fluorescent. All compounds showed a higher lipophilicity
than 8-MOP. Generally the 3-alkyl substituted furocoumarins had a capacity to pho-
tobleach RNO which was higher than that of the 3-aryl substituted ones.Some of the
3-aryl substituted furocoumarins displayed a photobleaching ability which was simi-
lar to or lower than that of 8-MOP.
Keywords: Furocoumarins; DNA intercalation; Fluorescence; Lipophilicity; Photo-
reactivity
Received: August 8, 2001 [FP619]
tive effects but also for the undesired side-effects such
as carcinogenicity, genotoxicity, and the formation of skin
Introduction
Furocoumarins (psoralens, 7H-furo[3,2-g]-1-benzo-
pyran-7-ones) are widely used in PUVA therapy (psor-
alen plus UVA irradiation) as photoreactive drugs in the
treatment of various skin diseases such as psoriasis,
mycosis fungoides, vitiligo [1–3], and in photopheresis,
an extracorporal form of photochemotherapy [4, 5].The
derivatives, which are usually employed for both PUVA
and photopheresis, are 8-methoxypsoralen (8-MOP),
5-methoxypsoralen (5-MOP), and 4,5Ј,8-trimethylpsor-
alen (TMP) [6]. The biological activity of these furocou-
marins has been correlated to their ability to photoreact
with DNA [2].The planar compounds are able to interca-
late between two base pairs of a DNA double helix in the
dark.Upon UVA irradiation, mono- and diadducts – inter-
strand crosslinks – can be formed.This DNA photodam-
age is responsible for some of the desired antiprolifera-
erythemas [2].Both mono- and diadducts are effective in
inhibiting DNA synthesis, although decreased photoin-
duced mutagenicity has been found in the case of mono-
functional as opposed to bifunctional furocoumarins [7].
For this reason monofunctional furocoumarins have
been prepared and investigated by several groups [2,
8–12] with the aim of reducing these undesired side ef-
fects. It has also been shown that the reactivity of the
double bonds can be modified by substitution with meth-
yl groups, thus influencing charge donation, steric ef-
fects and the lipophilic nature of the furocoumarins [13].
A further field of interest is the responsible mechanism of
PUVA. Furocoumarins also react with proteins and
membrane lipids, either by direct photoaddition or by
generation of singlet oxygen, which suggests that paral-
lel to the DNA reaction other cellular targets should be
taken into account when attempting to explain furocou-
marin photosensitization [14].Because of the diversity of
furocoumarin reactivity, the development of new deriva-
tives with specific target oriented properties is still an im-
portant task [2].
Correspondence: Paul Körner, Pharmazeutisches Institut,
Universität Kiel, Gutenbergstr. 76, D-24118 Kiel, Germany.
Phone:+49 (0)431 880 1137, Fax:+49 (0)431 8801352, e-mail:
pKoerner@pharmazie.uni-kiel.de.
© WILEY-VCH Verlag GmbH, 69451 Weinheim, 2002
0365-6233/05/0187 $ 17.50+.50/0

188 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
The author was interested in the preparation and investi-
gation of furocoumarins which have hydrophobic and
bulky substituents on the furan moiety, in order to check
the influence of the structural changes in position 3 of the
furocoumarin skeleton on the interaction with DNA as
well as on the absorption and fluorescence spectra, the
lipophilicity, and the photobleaching effect of N,N-di-
methyl-p-nitrosoaniline. This method has been used by
several groups for measuring the generation of singlet
oxygen [6, 10, 15–19].
The present paper describes the preparation of five se-
ries of furocoumarins.They consisted of two series of 3-
alkyl derivatives (3-methyl (3 a–c) and 3-tert-butyl (3 d–
f)) and three series of 3-aryl derivatives (3-phenyl
(3 g–i), 3-(4-methoxyphenyl) (3 j–l), and 3-(2,5-dimeth-
oxyphenyl) (3 m–o)).
Linear furocoumarins with a 3-methyl (3 a–c) or 3-phenyl
(3 g–h) group at the furan moiety have also been pre-
pared by several groups in the past [20–22]. In this
present work, in a comparative study, these compounds
are investigated together with newly synthesised deriva-
tives containing a 3-tert-butyl-, a 3-(4-methoxyphenyl)-
or a 3-(2,5-dimethoxyphenyl) group. 8-MOP and TMP
were used as reference compounds in all tests.
Chemistry
The furocoumarins (3 a–o) were obtained from the cor-
responding 7-hydroxy-2H-1-benzopyran-2-ones (1 a–
c), as indicated in Scheme 1. 7-Hydroxy-2H-1-benzo-
pyran-2-one (1 a) and 4-methyl-7-hydroxy-2H-1-benzo-
pyran-2-one (1 b) are commercially available com-
pounds. 4,8-Dimethyl-7-hydroxy-2H-1-benzopyran-2-
one (1 c) [23] was prepared via the Pechmann reaction.
The synthesis of furocoumarins was then achieved in
two steps by modifying the method of MacLeod [21].The
7-hydroxy-2H-1-benzopyran-2-ones (1 a–c) were con-
verted into β-ketoethers (2 a–o) by alkylation with the ap-
propriate haloketone under phase-transfer catalysis
conditions.Base-catalyzed intramolecular condensation
in 1 N aqueous KOH under nitrogen for 8 h and subse-
quent acidification gave the corresponding linear furo-
coumarins (3 a–o). Following this synthetic route, sever-
al compounds, bearing different substituents R3 (R3 =
methyl, tert-butyl, phenyl, 4-methoxyphenyl, 2,5-dimeth-
Table 1. Substitution pattern of furocoumarins (3).
Compound
R1
R2
R3
R4
3 a
3 b
3 c
3 d
3 e
H
H
Me
H
H
Me
H
H
Me
H
H
Me
H
H
Me
Me
Me
tert-Bu
tert-Bu
tert-Bu
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Me
Me
H
Me
Me
H
Me
Me
H
Me
Me
H
3 f
3 g
3 h
3 i
3 j
3 k
Ph
Ph
4-MeOPh
4-MeOPh
4-MeOPh
2,5-(MeO)2Ph
2,5-(MeO)2Ph
2,5-(MeO)2Ph
H
3 l
3 m
3 n
3 o
8-MOP
TMP
H
Me
Me
H
Me
MeO
Me
Scheme 1. a) acetone, K2CO3 , 18-crown-6, reflux, R3-
CO-CH2-X, X = Cl, Br; b) 1. 1 N KOH, reflux, 2.30 %
H3PO4 .
Me
H

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 189
Table 2. Some characteristics of compounds 2 d–f, 2 i, 2 k–o.
Comp.
No.
Yield
(%)
Mpa
(°C)
Formula
MW
Anal.
Calcd.:
Found:
2 d
2 e
2 f
45
40
98
116
123
198
158
200
176
182
205
C15H16O4
C16H18O4
C17H20O4
C19H16O4
C19H16O5
C20H18O5
C19H16O6
C20H18O6
C21H20O6
260.29
274.31
288.34
308.33
324.33
338.36
340.33
354.36
368.38
C, 69.22; H, 6.20
C, 68.90; H, 6.13
C, 70.06; H, 6.61
C, 69.85; H, 6.66
40
C, 70.81; H, 6.99
C, 70.86; H, 7.17
2 i
62.5
82
C, 74.01; H, 5.23
C, 73.63; H, 5.13
2 k
2 l
C, 70.36; H, 4.97
C, 70.29; H, 5.06
77
C, 71.00; H, 5.36
C, 70.50; H, 5.34
2 m
2 n
2 o
a
83
C, 67.06; H, 4.74
C, 66.81; H, 4.90
92
C, 67.79; H, 5.12
C, 67.73; H, 5.32
75
C, 68.47; H, 5.47
C, 68.38; H, 5.46
Recrystallization solvent: ethanol.
Table 3. 1H-NMR spectroscopic data and MS data of compounds 2 d–f, 2 i, 2 k–o.
Comp.
No.
1H-NMR; δ (ppm)
MS (70 eV): m/z (%)
1
3
3
2 d
2 e
2 f
7.98 (d, H, J = 9.1 Hz, 4-H), 7.38 (d, 1 H, J = 9.3 Hz, 5-H),
260 (21, M+), 176 (24), 147
(20), 89 (12), 63 (8), 57 (100),
41 (25)
3
6.93 (m, 2 H, 8-H, 6-H), 6.3 (d, 1 H, J = 9.5 Hz, 3-H), 5.52 (s,
2 H, –OCH2CO), 1.19 (s, 9 H, –COC(CH3)3)a
3
7.65 (d, 1 H, J = 9.1 Hz, 5-H), 6.92 (m, 2 H, 6-H, 8-H), 6.64
274 (37, M+), 189 (47), 161
(28), 103 (9), 85 (8), 57 (100),
41 (24)
4
(q, 1 H, J = 1.2 Hz, 3-H), 5.51 (s, 2 H, –OCH2CO), 2.39 (d,
4
a
3 H, J = 1.2 Hz, 4-CH3), 1.22 (s, 9 H, –COC(CH3)3
3
3
7.38 (d, 1 H, J = 8.8 Hz, 5-H), 6.58 (d, 1 H, J = 8.8 Hz, 6-H),
5.0 (s, 2 H, –OCH2CO), 2.39 (s, 3 H, 4-CH3), 2.38 (s, 3 H, 8-
CH3), 1.28 (s, 9 H, –COC(CH3)3)b
288 (40, M+), 204 (51), 189
(22), 175 (8), 145 (9), 115 (8),
91 (5), 77 (4), 57 (100), 51 (3),
41 (21)
3
3
2 i
8.0 (d, 2 H, J = 7.6 Hz, C6-2Ј,6Ј-H), 7.64 (dd, 1 H, J = 7.5 Hz,
308 (19, M+), 203 (5), 189 (10),
115 (4), 105 (100), 91 (10), 77
(29), 65 (6), 51 (7), 43 (1)
3J = 7.5 Hz, C6-4Ј-H), 7.52 (dd, 2 H, J = 7.6 Hz, J = 7.6 Hz,
C6-3Ј,5Ј-H), 7.37 (d, 1 H, J = 8.8 Hz, 5-H), 6.71 (d, 1 H, J =
8.8 Hz, 6-H), 6.15 (s, 1 H, 3-H), 5.38 (s, 2 H, –OCH2CO), 2.38
(s, 6 H, 4-CH3, 8-CH3)b
3
3
3
3

190 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
Table 3 (Continued)
Comp.
No.
1H-NMR; δ (ppm)
MS (70 eV): m/z (%)
3
3
2 k
7.98 (d, 2 H, J = 8.8 Hz, 4-CH3OC6-2Ј,6Ј-H), 7.51 (d, 1 H, J =
324 (7, M+), 135 (100), 121 (5),
107 (7), 92 (6), 77 (17), 64 (4),
51 (4), 44 (6)
3
8.6 Hz, 5-H), 6.99 (d, 2 H, J = 8.8 Hz, 4-CH3OC6-3Ј,5Ј-H ),
3
3
3
6.94 (dd, 1 H, J = 8.9 Hz, J = 2.4 Hz, 6-H), 6.79 (d, 1 H, J =
2.2 Hz, 8-H), 6.13 (s, 1 H, 3-H), 5.32 (s, 2 H, –OCH2CO), 3.90
(s, 3 H, CH3O), 2.38 (s, 3 H, 4-CH3)
3
3
2 l
8.00 (d, 2 H, J = 8.9 Hz, 4-CH3OC6-2Ј,6Ј-H ), 7.55 (d, 1 H, J =
338 (5, M+), 190 (2), 162 (3),
150 (5), 135 (100), 121 (9),
107 (9), 92 (9), 77 (25), 64 (6),
51 (6), 44 (61)
3
8.9 Hz, 5-H), 7.10 (d, 2 H, J = 8.9 Hz, 4-CH3OC6-3Ј,5Ј-H ),
3
4
6.98 (d, 1 H, J = 8.9 Hz , 6-H), 6.21 (d, 1 H, J = 1.1 Hz, 3-H),
5.70 (s, 2 H, –OCH2CO), 3.87 (s, 3 H, CH3O), 2.38 (d, 3 H,
4-CH3, J = 1.1 Hz), 2.28 (s, 3 H, 8-CH3)b
4
3
4
2 m
7.64 (d, 1 H, J = 9.5 Hz, 4-H), 7.98 (d, 1 H, J = 3.1 Hz, 2,5-
340 (6, M+), 165 (100), 150 (4),
135 (2), 122 (6), 107 (12), 89
(6), 77 (14), 63 (6), 51 (6), 44
(11)
3
(CH3O)2C6-6Ј-H ), 7.39 (d, 1 H, J = 8.6 Hz, 5-H), 7.15 (dd, 1 H,
3
3
4J = 3.2 Hz, J = 9.0 Hz, 2,5-(CH3O)2C6-4Ј-H), 6.99 (d, 1 H, J
3
4
= 9.0 Hz, 2,5-(CH3O)2C6-3Ј-H), 6.93 (dd, 1 H, J = 8.6 Hz , J =
4
3
2.2 Hz, 6-H), 6.76 (d, 1 H, J = 2.0 Hz, 8-H), 6.25 (d, 1 H, J =
9.5 Hz, 3-H), 5.33 (s, 2 H, –OCH2CO), 3.97 (s, 3 H, CH3O),
3.81 (s, 3 H, CH3O)b
3
4
2 n
7.54 (d, 1 H, J = 8.8 Hz, 5-H), 7.49 (d, 1 H, J = 3.2 Hz, 2,5-
354 (6, M+),165 (100), 150 (4),
135 (2), 122 (6), 107 (11), 91
(5), 77 (15), 65 (6), 51 (7), 44
(39)
4
3
(CH3O)2C6-6Ј-H) , 7.15 (dd, 1 H, J = 3.2 Hz, J = 9.1 Hz,
2,5-(CH3O)2C6-4Ј-H), 7.01 (d, 1 H, J = 9.1 Hz, 2,5-(CH3O)2C6-
3Ј-H), 6.96 (dd, 1 H, J = 8.8 Hz , J = 2.5 Hz, 6-H), 6.76 (d,
1 H, J = 2.5 Hz, 8-H), 6.15 (d, 1 H, J = 1.0 Hz, 3-H), 5.33 (s,
2 H, –OCH2CO), 3.97 (s, 3 H, CH3O), 3.81 (s, 3 H, CH3O), 2.39
(d, 3 H, 4-CH3)b
3
3
4
4
4
4
3
2 o
7.45 (d, 1 H, J = 2.4 Hz, 2,5-(CH3O)2C6-6Ј-H), 7.35 (d, 1 H, J
368 (6, M+), 190 (3), 165 (100),
150 (5), 122 (8), 107 (14), 91
(10), 77 (22), 65 (9), 51 (9), 44
(50)
4
3
= 8.7 Hz, 5-H), 7.13 (dd, 1 H, J = 2.6 Hz, J = 9.0 Hz, 2,5-
(CH3O)2C6-4Ј-H), 6.98 (d, 1 H, J = 9.00 Hz, 2,5-(CH3O)2C6-3Ј-
H), 6.64 (d, 1 H, J = 8.8 Hz, 6-H), 6.13 (s, 1 H, 3-H), 5.37 (s,
3
3
2 H, –OCH2CO), 3.94 (s, 3 H, CH3O), 3.80 (s, 3 H, CH3O), 2.39
(s, 3 H, 4-CH3), 2.36 (s, 3 H, 8-CH3)b
a
DMSO-d6; b CDCl3.
Table 4. Some characteristics of compounds 3 d–f, 3 i–o.
Comp.
No.
Yield
(%)
Mpa
(°C)
Formula
MW
Anal. or HRMS
Calcd.:
Found:
3 d
3 e
3 f
56
51
60
144
152
193
C15H14O3
C16H16O3
C17H18O3
242.27
256.30
270.33
242.09430
242.09415
256.10995
256.10983
270.12558
270.12550

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 191
Table 4. (Continued)
Comp.
No.
Yield
(%)
Mpa
(°C)
Formula
MW
Anal. or HRMS
Calcd.:
Found:
3 i
66
53
75
72
77
75
64
191
199
192
231
197
204
179
C19H14O3
C18H12O4
C19H14O4
C20H16O4
C19H14O5
C20H16O5
C21H18O5
290.31
292.29
306.32
320.34
322.31
336.34
350.37
C, 78.61; H, 4.86
C, 78.23; H, 4.86
3 j
292.07355
292.07360
3 k
3 l
C, 74.50; H, 4.61
C, 74.29; H, 4.61
C, 74.99; H, 5.03
C, 74.73; H, 5.00
3 m
3 n
3 o
C, 70.80; H, 4.38
C, 70.45; H, 4.37
C, 71.42; H, 4.79
C, 71.12; H, 4.81
350.11542
350.11520
a
Recrystallization solvent: ethanol
Table 5. 1H-NMR spectroscopic data, and MS data of compounds 3 d–f, 3 i–o.
Comp.
No.
1H-NMR; δ (ppm)
MS (70 eV): m/z (%)
3
3 d
7.82 (d, 1 H, 5-H, J = 9.6 Hz), 7.77 (s, 1 H, 4-H), 7.42 (s, 1 H,
9-H), 7.40 (s, 1 H, 2-H), 6.36 (d, 1 H, 6-H, 3J = 9.6 Hz), 1.45
(s, 9 H, 3-C(CH3)3)a
242 (45, M+), 227 (100), 199
(17), 128 (7), 115 (6), 99 (4),
85 (19), 77 (3), 63 (5), 57 (5),
41(5)
3 e
3 f
3 i
7.86 (s, 1 H, 4-H), 7.39 (s, 1 H, 2-H), 7.37 (s, 1 H, 9-H), 6.25
(s, 1 H, 6-H), 2.52 (d, 3 H, 5-CH3), 1.46 (s, 9 H, 3-C(CH3)3)a
256 (45, M+), 241 (100), 213
(17), 185 (7), 141 (6), 115 (4),
92 (19), 67 (3), 51 (5), 41 (5)
7.73 (s, 1 H, 4-H), 7.40 (s, 1 H, 2-H), 6.23 (s, 1 H, 6-H), 2.55
(s, 3 H, 9-CH3), 2.52 (s, 3 H, 5-CH3), 1.46 (s, 9 H, 3-C(CH3)3)a
270 (M+), 255 (100), 227 (14),
128 (7), 115 (6), 99 (26), 77
(6), 67 (11), 63 (5), 43 (7)
7.85 (s, 1 H, 4 H), 7.83 (s, 1 H, 2 H), 7.65–7.40 (m, 5 H, –C6H5),
6.29 (s, 1 H, 6-H), 2.65 (s, 3 H, 9-CH3), 2.51 (s, 3 H, 5-CH3)a
290 (100, M+), 262 (48), 233
(6), 219 (5), 202 (7), 189 (12),
178 (4), 165 (6), 131 (8), 101
(7), 89 (8), 76 (6)
3 j
7.85 (s, 1 H, 2-H), 7.81 (d, 1 H, 3J = 9.6 Hz, 5-H), 7.78 (s, 1 H,
292 (100, M+), 277 (29), 249
(26), 165 (15), 132 (9), 88 (9),
44 (15)
3
4-H), 7.55 (d, 2 H, J = 7.0 Hz, 4-CH3OC6-2Ј,6Ј-H), 7.50 (s,
3
1 H, 9-H), 7.05 (d, 2 H, J = 7.0 Hz, 4-CH3OC6-3Ј,5Ј-H), 6.4 (d,
3
1 H, J = 9.6 Hz, 6-H), 3.88 (s, 3 H, OCH3)a

192 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
Table 5. (Continued)
Comp.
No.
1H-NMR; δ (ppm)
MS (70 eV): m/z (%)
3
3 k
7.93 (s, 1 H, 4-H), 7.77 (s, 1 H, 2-H), 7.55 (d, 2 H, J = 8.6 Hz,
306 (100, M+), 291 (14), 278
(17), 263 (20), 235 (6), 178 (8),
139 (9), 117 (10)
3
4-CH3OC6-2Ј,6Ј-H), 7.49 (s, 1 H, 9-H), 7.06 (d, 2 H, J =
8.6 Hz, 4-CH3OC6-3Ј,5Ј-H), 6.28 (s, 1 H, 6-H), 3.88 (s, 3 H,
CH3O), 2.51 (s, 3 H, 5-CH3)a
3
3 l
7.78 (s, 2 H, 2-H, 4-H), 7.55 (d, 2 H, J = 8.5 Hz, 4-CH3OC6-
320 (100, M+), 305 (12), 292
(17), 277 (15), 178 (5), 146 (7),
124 (8)
3
2Ј,6Ј-H), 7.05 (d, 2 H, J = 8.5 Hz, 4-CH3OC6-3Ј,5Ј-H), 6.27 (s,
1 H, 6-H), 3.88 (s, 3 H, CH3O), 2.63 (s, 3 H, 9-CH3), 2.50 (s,
3 H, 5-CH3)a
3
3 m
7.98 (s, 1 H, 2-H), 7.81 (d, 1 H, J = 9.6 Hz, 5-H), 7.80 (s, 1 H,
322 (100, M+), 307 (32), 279
(26), 208 (14), 161 (15), 148
(20), 136 (16), 125 (24), 76
(14), 63 (14)
4
4-H), 7.50 (s, 1 H, 9-H), 7.17 (d, 1 H, J = 3.0 Hz, 2,5-
(CH3O)2C6-6Ј-H), 6.99 (d, 1 H, J = 8.9 Hz, 2,5-(CH3O)2C6-3Ј-
H), 6.91 (dd, 1 H, J = 3.0 Hz, J = 8.9 Hz, 2,5-(CH3O)2C6-4Ј-
H), 6.38 (d, 1 H, J = 9.6 Hz, 6-H) 3.85 (s, 3 H, CH3O), 3.83 (s,
3
4
3
3
3 H, CH3O)a
3 n
3 o
7.94 (s, 1 H, 2-H), 7.90 (s, 1 H, 4-H), 7.49 (s, 1 H, 9-H), 7.15
336 (100, M+), 321 (26), 293
(22), 265 (10), 250 (9), 222 (9),
200 (10), 165 (10), 148 (19),
136 (10), 82 (6), 67 (7)
4
3
(d, 1 H, J = 2.9 Hz, 2,5-(CH3O)2C6-6Ј-H), 7.15 (d, 1 H, J =
4
3
8.9 Hz, 2,5-(CH3O)2C6-3Ј-H), 6.93 (dd, 1 H, J = 2.9 Hz, J =
8.9 Hz, 2,5-(CH3O)2C6-4Ј-H), 6.27 (s, 1 H, 6-H), 3.85 (s, 3 H,
CH3O), 3.83 (s, 3 H, CH3O), 2.49 (s, 3 H, 5-CH3)a
4
7.96 (s, 1 H, 2-H), 7.75 (s, 1 H, 4-H), 7.15 (d, 1 H, J = 2.4 Hz,
3
2,5-(CH3O)2C6-6Ј-H), 7.0 (d, 1 H, J = 8.9 Hz, 2,5-(CH3O)2C6-
350 (100, M+), 335 (29), 307
(19), 214 (21), 175 (15), 161
(16), 148 (44), 139 (21), 89
(13), 76 (16)
4
3
3Ј-H), 6.92 (dd, 1 H, J = 2.4 Hz, J = 8.9 Hz, 2,5-(CH3O)2C6-
4Ј-H), 6.26 (s, 1 H, 6-H), 3.84 (s, 3 H, CH3O), 3.82 (s, 3 H,
CH3O), 2.64 (s, 3 H, 9-CH3), 2.48 (s, 3 H, 5-CH3)a
a
CDCl3
oxyphenyl) in position 3 and a different number of methyl
groups in position 5 and 9 of the furocoumarin skeleton,
were prepared. The synthesis of 1 c [23], 2 a [20], 2 b
[21], 2 c [24], 2 g [21], 2 h [21], 2 j [25], 3 a [20], 3 b [21],
3 c [24], 3 g [21], and 3 h [21] has been described in the
literature. All synthesized compounds have been char-
acterized by H-NMR- and mass spectroscopy and the
new ones were further characterized by elemental anal-
ysis or high resolution mass spectroscopy.
The maxima (λmax (nm)) of the absorption spectra in the
ground state of the synthesised furocoumarins (20–
60 µM) dissolved in methanol are reported in Table 6.
The large band shown in the spectra by 3 a (292 nm) and
3 d (296 nm) is shifted to longer wavelength if there is an
aryl group in 3-position, e.g. 3 g (298 nm), 3 j (302 nm),
and 3 m (302 nm), instead of an alkyl group. In addition
methyl groups in 5- and 9-position lead to a further
bathochromic effect in the absorption spectra (3 c, 3 f,
3 i, 3 l, 3 o). All compounds exhibit sufficient absorption
in the UVA wavelength range of 320–400 nm and in par-
ticular 350 nm, the maximum of the UVA lamp emission.
1
Results and discussion
Absorption and fluorescence spectra
The maxima of the fluorescence excitation and emission
spectra of the furocoumarins (4 µM) in methanol are re-
ported in Table 6. The fluorescence excitation spectra
show bands with maxima at around 300 or 330 nm. For
compounds 3 a–i the maxima in the fluorescence excita-
The absorption and fluorescence spectra of the potential
photoreactive compounds were measured with the aim
of characterizing their spectrophotometric- and fluores-
cence properties.

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 193
tion spectra agree with the absorption maxima, in the
case of compounds 3 j–o the spectra do not coincide.
Former findings of Lysenko et al. [26, 27], measuring
furocoumarins in ethanol, indicate that there exist mono-
meric and different aggregated forms of furocoumarins
in solution. They suggested that only aggregated forms
of furocoumarins possess the ability to fluoresce at a
high degree, which may be responsible for their ob-
served maxima discrepancy comparing the absorption
and excitation spectra of different furocoumarins, which
is similar to the observations in this work.
gated by fluorescence spectroscopy [28] and by meas-
uring the thermal transitions of the DNA-macromolecule
(melting experiments) [32].
It is well known that a preliminary complex in the ground
state occurs between furocoumarins and DNA, in which
the planar moiety of the furocoumarin undergoes inter-
calation inside DNA [2]. A furocoumarin in an interacting
state should have a fluorescence maxima (λmax) and a
fluorescence intensity (If), which is different from that of
the free state [33, 34].The formation of a molecular com-
plex between the compounds and DNA is supported by
fluorescence quenching [28, 34, 35] and a shift of λmax of
the fluorescence excitation spectra of the small ligands.
The recording of fluorescence excitation spectra instead
of UV/VIS spectra, which have been used by other
groups [13, 36, 37], has the advantage of lower furocou-
marin concentrations in the aqueous medium.The fluo-
rescence measurement data for the interaction of the
synthesised furocoumarins and the reference com-
pounds 8-MOP and TMP (4 µM) with calf-thymus DNA
(600 µM) are shown in Table 6.The DNA titration experi-
ments showed that the fluorescence intensity of the
3-methyl (3 a–c), and 3-tert-butyl derivatives (3 d–e) was
quenched to a different extent with increasing DNA con-
centration, which is in accordance with previously re-
ported furocoumarins [28, 34]. 8-MOP and TMP, which
are known as DNA-intercalating compounds, showed a
quenching of fluorescence intensity and a red shift of
9 nm in the fluorescence excitation spectra, when the
compounds were dissolved in DNA-buffer solution in-
stead of buffer solution alone.The 3-alkyl derivatives al-
so showed a red shift, which is similar to that of the refer-
ence compounds, suggesting that the 3-methyl and 3-
tert-butyl substituent may intercalate into DNA. The
3-aryl derivatives displayed a more different result. The
3-phenyl furocoumarins (3 g–i) showed a blue shift of
4–6 nm and a fluorescence quenching and the 3-(4-
methoxyphenyl) compounds (3 j, 3 k) showed red shifts
and fluorescence quenching when the compounds were
dissolved in DNA-buffer solution instead of buffer solu-
tion alone, suggesting that these compounds may inter-
calate in DNA. In the case of 3 l and 3 o no shift and
quenching were observable, apparently due to only very
weak or missing interactions between the compounds
and DNA.
Lipophilicity
In a study on the biological action of furocoumarins, a re-
lationship between the octanol/water partition coeffi-
cient (log P) and the structural changes resulting in in-
creased affinities towards DNA was reported [28]. The
variation of hydrophobicity seems to be a factor that af-
fects the affinity of furocoumarins towards the internal li-
pophilic parts of duplex DNA [28–29].For this reason the
log P values were determined in order to characterize
the lipophilicity of the synthesized furocoumarins. It is
well established that reversed phase HPLC can provide
quantitative information about the hydrophobicity of sol-
ute molecules, especially for highly hydrophobic com-
pounds [30].The log P values were obtained by a HPLC
procedure [31], using an alkyl-bonded stationary phase
and a methanol-water mixture as mobile phase. The
compounds exhibited a wide range of log P values
(2.23–4.69, Table 7). All furocoumarins showed a higher
lipophilicity than 8-MOP. Comparing the furocoumarins
in the series with each other, it can be observed that the
introduction of a methyl group in 5-position (3 b, 3 e, 3 h,
3 k, 3 n) and in addition in 9-position (3 c, 3 f, 3 i, 3 l, 3 o)
increases the lipophilicity.In the present studies no clear
relationship between lipophilicity and DNA interaction
could be found.
Interaction of furocoumarins with DNA
The photoreactivity of the 3-substituted compounds with
nucleic acids was studied in order to investigate the influ-
ence of bulky and lipophilic substituents at the furan
group on the ability to form crosslinks with DNA. Be-
cause the photoreactions are governed by preceding
dark reactions, the interaction of furocoumarins with
DNA in the dark was also examined.
It has also been demonstrated that DNA shows helix sta-
bilization when complexed with furocoumarins, leading
to an increase in thermal helix to coil transition tempera-
ture (Tm increase) [29, 32].The Tm (50 % of total thermal
transition) increase could be measured from the melting
profiles of poly(dA-dT)-poly(dA-dT) in the presence and
absence of furocoumarin.The thermal denaturation data
for interaction of the synthesized furocoumarins (3 a–o)
and the reference compounds 8-MOP and TMP are
As already described, three steps are generally as-
sumed in the formation of furocoumarin-DNA photoad-
ducts [2].The first is a non-covalent interaction of the in-
tercalation type. The non-covalent binding of the linear
furocoumarins (3 a–o, 8-MOP, TMP) to DNA was investi-

194 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
Table 6. Absorption and fluorescence properties of furocoumarins 3 shown in methanol, and in buffer pH 7.4 without or
in presence of DNA.
Compound
λmax (nm);
λmax (nm)
λmax (nm)
λmax (nm)
Ib/Ife
∆
λmax shift
(nm)f
ex
log ε MeOHa
MeOHb
bufferc
buffer, DNAd
ex/em
ex/em
ex/em
3 a
3 b
3 c
3 d
3 e
292; 4.00
292; 4.00
298; 4.05
296; 4.05
292; 4.08
298; 3.91
298; 4.10
296; 4.08
300; 4.03
302; 4.08
302; 3.97
304; 4.12
302; 4.17
302; 3.97
306; 4.10
302; 4.02
296; 4.01
292/450
292/436
298/445
296/445
292/430
298/439
298/468
296/456
300/456
338/450
333/454
333/467
342/442
334/442
334/467
302/476
296/442
299/478
298/460
304/466
302/475
299/458
308/432
341/454
333/450
324/417
331/456
331/442
323/442
n.d.
310/474
310/460
311/466
310/475
309/458
314/432
337/454
327/450
318/450
341/456
337/442
323/442
n.d.
0.9
0.6
0.4
0.8
0.7
1.0
0.9
0.9
0.5
0.8
0.8
1.0
n.d.
n.d.
1.1
0.9
0.5
11g
12g
7g
8g
10g
6g
3 f
3 g
3 h
3 i
3 j
3 k
4h
6h
6h
10g
6g
3 l
0g
3 m
3 n
3 o
8-MOP
TMP
n.d.
n.d.
0g
9g
9g
n.d.
n.d.
329/447
302/508
302/464
329/442
311/503
311/464
a
Absorption wavelength maxima (λmax) in MeOH (scanned wavelength range:from 280 to 400 nm) and log ε (logarithm
of molar extinction).
b
c
d
e
Fluorescence excitation and emission wavelength maxima (λmax) in MeOH; slits 15 and 20 nm.
Excitation and emission wavelength maxima (λmax) in buffer pH 7.4; slits 15 and 20 nm.
Excitation and emission wavelength maxima (λmax) in DNA-buffer solution pH 7.4 (600 µM); slits 15 and 20 nm.
Quotient of fluorescence intensities of furocoumarins in DNA-buffer solution pH 7.4 (600 µM; Ib) and free in buffer
pH 7.4 (If).
Shift of maximum wavelength of fluorescence excitation spectrum in DNA-buffer solution pH 7.4 (600 µM) in relation
to buffer pH 7.4.
Red shift. h Blue shift. N.d. means not determined.
f
g
shown inTable 7.The results show that for some furocou-
After intercalation of the furocoumarins in the dark the
second step of the formation of furocoumarin-DNA pho-
toadducts is the formation of photomonoadducts upon
UVA irradiation of the non-covalent complex.The forma-
tion usually involves the 5,6 double bond of a pyrimidine
and the double bond of either the furan-side or the py-
rone-side of the furocoumarin [1, 2]. In the last step the
furocoumarin linked by the furan-side can absorb a sec-
ond photon and, if the geometrical arrangement is favo-
rable, an interstrand crosslink with a pyrimidine belong-
ing to the opposite DNA strand can be formed (Figure 1).
marins (3 a–c, 3 f–j, 8-MOP, and TMP) an increase in Tm
is observed, although only to a small extend in compari-
son with the shifts noticed for some compounds under
UVA irradiation conditions. The furocoumarins (3 d–e,
3 k–o) show no marked increase in Tm values.The data
of the fluorescence measurement and the melting exper-
iments support the formation of molecular complexes
between most of the furocoumarins and DNA and indi-
cate the possible intercalation of these ligands inside du-
plex DNA. In the case of the 3-aryl substituted com-
pounds (3 l, 3 o) the results indicate no DNA-interaction.
Evidence for intercalation of these compounds can be
obtained from DNA melting studies under irradiation.
Crosslinking requires that the furocoumarin penetrates
into the interior of the double helix [1, 2].
In order to investigate the potential of compounds for in-
terstrand crosslink formation, melting studies under
60 min irradiation conditions at a 10 : 1 molar ratio of
calf-thymus DNA and furocoumarin and under 30 min

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 195
pounds are able to generate crosslinks similar to the ref-
erence compounds TMP and 8-MOP. Compounds con-
taining a 3-tert-butyl (3 d–f), or 3-(4-methoxyphenyl)
(3 j–l), or 3-(2,5-dimethoxyphenyl) (3 m–o) substituent
exhibited no marked effect on Tm, suggesting that these
compounds do not produce any photoadducts with
poly(dA-dT)-poly(dA-dT) or calf thymus DNA.
The photoreaction of furocoumarins with DNA in general
is influenced by several factors [38–40], e.g.the complex
formation and geometrical arrangement of a furocou-
marin after intercalation. Further investigations might be
necessary to fully explain the present results, neverthe-
less some suggestion can be made: In comparison with
the 3-methyl derivatives (3 a–c) a possible reason for the
decreased (3 g–i) or missing (3 d–f and 3 j–o) capacity
to generate crosslinks with DNA might be a steric hin-
drance of the bulky 3-position, which may influence the
geometrical arrangement of the furocoumarin in the
DNA helix before irradiation with UVA light. In addition,
specific furocoumarin excited states properties within
the non-covalent complex might be important comparing
the 3-phenyl derivatives (3 g–i) with the methoxyphenyl
substituted compounds (3 j–o).
Figure 1. Possible photochemical formation of 2,3- and
5,6-monoadducts and 2,3,5,6-interstrand crosslink ad-
ducts (according to Song et al. [35] with modifications).
Photobleaching of N,N-dimethyl-p-
irradiation conditions at a 1 : 1 molar ratio of poly(dA-dT)-
poly(dA-dT) and furocoumarin were made. The thermal
denaturation data for interaction of the synthesized furo-
coumarins (3 a–o) and the reference compounds
8-MOP and TMP under UVA irradiation conditions are
shown in Table 7.
nitrosoaniline (RNO)
Since PUVA is responsible for cell damage and the reac-
tion with DNA, and proteins and lipids may also be influ-
enced by the production of singlet oxygen [14], the ca-
pacity of synthesised compounds to photobleach RNO
was evaluated in an in vitro test system [41], too.
In the calf-thymus DNA studies the greatest increase in
Tm was induced by compound 3 c. For TMP and the fol-
lowing furocoumarins it was gradually lower (3 c > TMP
ӷ 3 b > 3 i > 3 h > 3 a ~ 8-MOP > 3 g). The other com-
pounds (3 d–f, 3 j–o) show no marked increase inTm val-
ues.The amount of compound covalently bound to DNA
generally increases with the A-T content [38]. For this
reason it was expected, that there would be an increase
in Tm values, when poly(dA-dT)-poly(dA-dT) was used
instead of calf-thymus DNA. In this study compounds
3 a–c and 3 g–j also induce stabilization of the duplex
DNA (3 b ~ 3 c ~ TMP ӷ 3 a ~ 3 i > 8-MOP ~ 3 g > 3 h ӷ
3 j).Compounds 3 d–f and 3 k–o exhibited no marked ef-
fect on Tm through UVA irradiation, which suggests, in
addition to the results of the calf-thymus-DNA studies,
that these compounds do not produce photoadducts or
cross-links with DNA. Comparing the five series of furo-
coumarins with each other, it could be observed that the
3-methyl (3 a–c) and 3-phenyl (3 g–i) derivatives in-
duced intense stabilization of poly(dA-dT)-poly(dA-dT)
and the calf-thymus DNA helix and consequently
marked photobinding, suggesting that these com-
Furocoumarins and UVA irradiation act through various
mechanisms: a type I (radical formation), a type II (ener-
gy transfer mechanism, with formation of singlet oxygen)
and the anoxic type III (C4-cycloaddition reactions) [32].
The photobleaching capacity of the furocoumarins was
studied according to the method of Krajic and El Moshni
[41] which is based on the bleaching of N,N-dimethyl-p-
nitrosoaniline by singlet oxygen in the presence of sensi-
tizer and histidine (for more details see the experimental
part). The results, reported in Table 7, show that all the
furocoumarins are able to photobleach RNO, although
with quite different yields.Generally the 3-alkyl substitut-
ed compounds (3 a–f) show a photobleaching capacity
which is higher than that of the 3-aryl substituted ones
(3 g–o). Some 3-aryl substituted furocoumarins have an
ability to photobleach RNO which is similar to or lower
than that of 8-MOP (3 g, 3 j, 3 k, 3 m, 3 o).
A consequence of the weak photobleaching capacity of
the 3-aryl substituted compounds 3 g, 3 j, 3 k, 3 m, and
3 o may be a decreased singlet oxygen generation ca-
pacity, similar to that of 8-MOP, which is known to be a

196 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
Table 7. Tm increase shown by DNA in presence of furocoumarins, photobleaching of RNO and log P values as meas-
ure of lipophilicity.
Compound
Induced
∆Tm shifta (°C)
CT–DNA
Induced
∆Tm shiftb (°C)
poly–(dA–dT)
Photo–bleaching
of RNOc
log P d
(%)
UV–A
dark
UV-A
3 a
3 b
3 c
3 d
3 e
3 f
3 g
3 h
3 i
3 j
3 k
3 l
3 m
3 n
3 o
2.0 ± 0.5e
4.5 ± 0.5e
8.5 ± 0.5e
–0.3 ± 0.3
0.2 ± 0.3
0.1 ± 0.3
1.1 ± 0.3e
2.3 ± 0.5e
2.9 ± 0.5e
0.1 ± 0.3
–0.3 ± 0.3
0.0 ± 0.3
–0.4 ± 0.4
0.1 ± 0.4
–0.4 ± 0.4
1.8 ± 0.5e
7.4 ± 0.5e
n.d.
0.5 ± 0.2e
1.8 ± 0.2e
1.3 ± 0.2e
–0.1 ± 0.1
0.2 ± 0.2
0.7 ± 0.2e
0.9 ± 0.2e
0.5 ± 0.2e
0.8 ± 0.3e
1.0 ± 0.1e
0.3 ± 0.3
0.1 ± 0.1
0.3 ± 0.2
0.4 ± 0.3
0.3 ± 0.2
0.5 ± 0.2e
0.8 ± 0.1e
n.d.
5.1 ± 0.4e
>15.0e
24 ± 1.0
32 ± 1.0
34 ± 1.6
28 ± 1.0
37 ± 1.0
20 ± 1.0
4 ± 1.0
10 ± 1.0
13 ± 1.0
3 ± 0.7
7 ± 0.7
9 ± 0.9
6 ± 1.5
12 ± 1.4
5 ± 1.0
2.23 ± 0.09
2.69 ± 0.08
3.36 ± 0.06
3.42 ± 0.06
3.84 ± 0.05
4.57 ± 0.05
3.26 ± 0.06
3.84 ± 0.05
4.57 ± 0.05
3.51 ± 0.06
3.93 ± 0.05
4.63 ± 0.05
3.51 ± 0.06
3.93 ± 0.05
4.69 ± 0.04
1.69 ± 0.08
3.36 ± 0.06
n.d.
>15.0e
–0.5 ± 0.2
–0.2 ± 0.3
–0.4 ± 0.3
4.2 ± 0.2e
3.0 ± 0.2e
5.6 ± 0.5e
1.0 ± 0.4e
0.3 ± 0.2
0.4 ± 0.2
0.3 ± 0.3
0.0 ± 0.1
–0.1 ± 0.1
4.0 ± 0.2e
>15.0e
8–MOP
TMP
Methylene blue
7 ± 0.6
14 ± 1.0
26 ± 1.0
n.d.
All values are mean ± s.e.m.
a
Mean value from two determinations for a compound: calf-thymus (CT) DNA ratio of 1 : 10.
Mean value from two to three determinations for a compound: poly (dA–dT) ratio of 1 : 1.
Decrease in absorbance of N,N-dimethyl-p-nitrosoaniline (RNO) at 440 nm after 3 h of UVA irradiation. d Log P
values as measure of lipophilicity determined by HPLC.
b
c
e
Induced ∆Tm shift was significantly different with respect to control ( p < 0.05).
poor singlet oxygen producer [15, 16], possibly resulting
in a lower skin phototoxicity in comparison with the
3-alkyl substituted furocoumarins (3 a–f). On the other
hand, the 3-alkyl substituted furocoumarins (3 a–f) may
be the more photoreactive compounds, which may be in-
teresting for developing new furocoumarins with specific
target orientation.
interaction could be found, suggesting that steric effects,
such as a steric hindrance generated by the bulky sub-
stituents at the 3-position, are responsible for this action.
All novel derivatives are more lipophilic than 8-MOP and
for this reason it may be expected that these compounds
also associate with the relatively hydrophobic environ-
ment of lipid membranes or the hydrophobic pockets of
proteins as further possible targets.The absorption and
fluorescence spectra in methanol show that the investi-
gated furocoumarins absorb UVA light and have fluores-
cent properties similar to those of other known furocou-
marins. In addition, the different capacity of compounds
to photobleach RNO, which might be correlated with the
generation capacity of singlet oxygen, also influencing
the reaction with DNA, proteins, and lipids, demonstrate
that these furocoumarins may be interesting photoreact-
ing agents, which should be investigated for further prop-
erties and other possible targets.
Conclusions
The development of furocoumarins based on the intro-
duction of bulky tert-butyl (3 d–f), 4-methoxyphenyl (3 j–
l), and 2,5-dimethoxyphenyl (3 m–o) substituents in the
furan moiety of the planar compounds leads to novel de-
rivatives which have, in comparison with 8-MOP and
TMP, no crosslinking ability. This might be of interest in
view of the side effects of furocoumarins in PUVA thera-
py. No clear relationship between lipophilicity and DNA

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 197
4,8-Dimethyl-7-(2-oxo-2-phenylethoxy)-2H-1-benzopyran-
2-one (2 i) was prepared from 1 c (0.95 g, 5 mmol) and 2-chlo-
ro-acetophenone (0.93 g, 6 mmol).
Acknowledgements
The author thanks Mrs Elke Stoll and Mrs Susan Häuer
for technical assistance.
7-(2-(4-Methoxyphenyl)-2-oxoethoxy)-4-methyl-2H-1-benzo-
pyran-2-one (2 k) was prepared from 1 b (0.88 g, 5 mmol) and
2-bromo-4Ј-methoxyacetophenone (1.37 g, 6 mmol).
Experimental part
4,8-Dimethyl-7-(2-(4-methoxyphenyl)-2-oxoethoxy)-2H-1-
benzopyran-2-one (2 l) was prepared from 1 c (0.95 g, 5 mmol)
and 2-bromo-4Ј-methoxyacetophenone (1.37 g, 6 mmol).
Chemistry
Melting points were determined on a ReichertThermovar melt-
ing point apparatus and are uncorrected. Nuclear magnetic
resonance (NMR) spectra were recorded on a Bruker ARX
300 MHz FT NMR spectrometer, and chemical shifts (ppm) are
reported relative to tetramethylsilane as internal standard in
CDCl3 or in DMSO-d6 or the solvents peak (CDCl3 7.24 ppm,
DMSO-d6 2.5 ppm). Signals are designated as follows: s (sin-
glet), d (doublet), dd (doublet of doublets), t (triplet), q (quadru-
plet), m (multiplet), ex (exchangeable with D2O). Mass spectra
were recorded on a Hewlett-Packard 5989A spectrometer (EI,
70 eV).High resolution mass spectra were recorded on a Finni-
gan MAT 8230 MS (70 eV), Institute for Organic Chemistry, Uni-
versity of Kiel. The 1H-NMR and MS spectra of known com-
pounds were in agreement with their chemical structure. Ele-
mental analyses were performed by a Hewlett-Packard CHN
autoanalyzer, Institute for Inorganic Chemistry, University of
Kiel.Yields were not optimized. HPLC (Waters 590 pump, Wa-
ters 486 absorbance detector set at 210 nm or 310 nm) was
performed on a LiChrospher 100 RP-18 column (5 µm parti-
cles; Merck).The absorbance spectra were recorded with a HP
8845A diode-array spectrophotometer fitted with a Julabo
F20-C heating controller. The fluorescence spectra were ob-
tained on a Perkin Elmer LS 50B luminescence spectrometer
fitted with a Julabo F20-C heating controller and fused quartz
cuvettes of 1 cm path length.
7-(2-(2,5-Dimethoxyphenyl)-2-oxoethoxy)-2H-1-benzopyran-
2-one (2 m) was prepared from 1 a (0.81 g, 5 mmol) and
2-bromo-2Ј,5Ј-dimethoxyacetophenone (1.55 g, 6 mmol).
7-(2-(2,5-Dimethoxyphenyl)-2-oxoethoxy)-4-methyl-2H-1-
benzopyran-2-one (2 n) was prepared from 1 b (0.88 g,
5 mmol) and 2-bromo-2Ј,5Ј-dimethoxy-acetophenone (1.55 g,
6 mmol).
7-(2-(2,5-Dimethoxyphenyl)-2-oxoethoxy)-4,8-dimethyl-2H-1-
benzopyran-2-one (2 o) was prepared from 1 c (0.95 g, 5 mmol)
and 2-bromo-2Ј,5Ј-dimethoxyacetophenone (1.55 g, 6 mmol).
General procedure for preparation of 7H-furo[3,2-g]-1-benzo-
pyran-7-ones [21]
The required 7-(2-alkyl-2-oxoethoxy)-2H-1-benzopyran-2-
ones or 7-(2-aryl-2-oxoethoxy)-2H-1-benzopyran-2-ones
(3 mmol) were heated in potassium hydroxide (1 N) solution
(300 mL) under reflux in an atmosphere of nitrogen for 8 h.The
solution was cooled, acidified with 30 % phosphoric acid, and
kept at 4 °C overnight. The precipitate was collected by filtra-
tion, washed with water, dried, and recrystallized from ethanol
(modified after MacLeod [21]).
The following compounds were prepared using the same gen-
eral procedure for preparation of 7H-furo[3,2-g]-1-benzopyran-
7-ones [21].
Compound 1 c [23] was prepared according to the general pro-
cedure for preparation of coumarins [42]. Compounds 2 a [20],
2 b [21], 2 c [24], 2 g [21], 2 h [21], and 2 j [25] were prepared
according to general procedure for preparation of β-ketoethers
[43].Compounds 3 a [20], 3 b [21], 3 c [24], 3 g [21], and 3 h [21]
were prepared according to general procedure for preparation
of 7H-furo[3,2-g]-1-benzopyran-7-ones [21].
3-(1,1-Dimethylethyl)-7H-furo[3,2-g]-1-benzopyran-7-one
(3 d) was prepared from 2 d (0.78 g, 3 mmol).
3-(1,1-Dimethylethyl)-5-methyl-7H-furo[3,2-g]-1-benzopyran-
7-one (3 e) was prepared from 2 e (0.82 g, 3 mmol).
General procedure for preparation of β-ketoethers [43]
5,9-Dimethyl-3-(1,1-dimethylethyl)-7H-furo[3,2-g]-1-benzo-
pyran-7-one (3 f) was prepared from 2 f (0.87 g, 3 mmol).
The appropriate haloketone (6 mmol), potassium carbonate
(2.8 g, 20 mmol) and 18-crown-6 (1 mL of stock solution (70 mg
of 18-crown-6 in 100 mL of acetone)) were added to a solution
of the required 7-hydroxy-2H-1-benzopyran-2-one 1 (5 mmol)
in acetone (100 mL). The mixture was refluxed for 5 h (moni-
tored by TLC). After cooling, the solution was filtered and evap-
orated under reduced pressure. The solid residue was recrys-
tallized from ethanol.
5,9-Dimethyl-3-phenyl-7H-furo[3,2-g]-1-benzopyran-7-one
(3 i) was prepared from 2 i (0.93 g, 3 mmol).
3-(4-Methoxyphenyl)-7H-furo[3,2-g]-1-benzopyran-7-one
(3 j) was prepared from 2 j (0.93 g, 3 mmol).
3-(4-Methoxyphenyl)-5-methyl-7H-furo[3,2-g]-1-benzopyran-
7-one (3 k) was prepared from 2 k (0.97 g, 3 mmol).
The following compounds were prepared using the same gen-
eral procedure for preparation of β-ketoethers [43].
5,9-Dimethyl-3-(4-methoxyphenyl)-7H-furo[3,2-g]-1-benzo-
pyran-7-one (3 l) was prepared from 2 l (1.02 g, 3 mmol).
7-(3,3-Dimethyl-2-oxobutoxy)-2H-1-benzopyran-2-one (2 d)
was prepared from 1 a (0.81 g, 5 mmol) and 1-bromo-3,3-
dimethylbutan-2-one (1.07 g, 6 mmol).
3-(2,5-Dimethoxyphenyl)-7H-furo[3,2-g]-1-benzopyran-7-one
(3 m) was prepared from 2 m (1.02 g, 3 mmol).
4-Methyl-7-(3,3-dimethyl-2-oxobutoxy)-2H-1-benzopyran-2-
one (2 e) was prepared from 1 b (0.88 g, 5 mmol) and 1-bromo-
3,3-dimethylbutan-2-one (1.07 g, 6 mmol).
3-(2,5-Dimethoxyphenyl)-5-methyl-7H-furo[3,2-g]-1-benzo-
pyran-7-one (3 n) was prepared from 2 n (1.06 g, 3 mmol).
4,8-Dimethyl-7-(3,3-dimethyl-2-oxo-butoxy)-2H-1-benzo-
pyran-2-one (2 f) was prepared from 1 c (0.95 g, 5 mmol) and
1-bromo-3,3-dimethylbutan-2-one (1.07 g, 6 mmol).
3-(2,5-Dimethoxyphenyl)-5,9-dimethyl-7H-furo[3,2-g]-1-ben-
zopyran-7-one (3 o) was prepared from 2 o (1.11 g, 3 mmol).

198 Körner et al.
Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
DNA-furocoumarin solutions were prepared by addition of the
compounds from a stock solution in DMSO (1 mM) to give a fi-
nal drug concentration of 10 µM.The solutions were irradiated
for 60 min (output measured at 360 nm at a distance of 26 cm:
0.9 mW/cm2).The addition of 20 µL of DMSO and UVA irradia-
tion for 60 min had no effect on the transition temperature of the
DNA in the absence of 3.
Biological assay methods
Calf-thymus DNA, DMSO, NaCl, histidine, methylbenzene,
ethylbenzene, propylbenzene, butylbenzene, pentylbenzene,
and methanol Lichrosolv were purchased from Merck. Tris(hy-
droxymethyl)aminomethane (Tris), 2-(N-morpholino)pro-
panesulfonic acid (MOPS), poly(dA-dT)-poly(dA-dT), N,N-
dimethyl-p-nitrosoaniline, methylene blue, and 4,5Ј,8-trimeth-
ylpsoralen were purchased from Sigma. 8-Methoxypsoralen
was purchased from Aldrich.
Poly(dA-dT)-poly(dA-dT)-DNA (according to a modified meth-
od of Wulff et al.[44]):an aqueous DNA-stock solution was pre-
pared by dissolving 25 units of the poly(dA-dT)-poly(dA-dT)-
DNA in 2.5 mL of MOPS-buffer (0.1 M, pH 7.0); working solu-
tions contained 10 µM DNA (20 µL stock solution). DNA-furo-
coumarin solutions were prepared by addition of the com-
pounds from a stock solution in DMSO (10–3 M) to give a final
drug concentration of 10 µM.The solutions were either incubat-
ed for 30 min in the dark or irradiated for 30 min. (output meas-
ured at 360 nm at a distance of 26 cm: 0.9 mW/cm2).The addi-
tion of 20 µL of DMSO and UVA irradiation for 30 min had no ef-
fect on the transition temperature of the polynucleotide in the
absence of 3.
UV/VIS and fluorescence excitation and emission spectra of
furocoumarins
The absorbance spectra were recorded using working solu-
tions containing 20 to 60 µM of the furocoumarin in methanol
(20 °C).
The fluorescence excitation and emission spectra of furocou-
marins were studied using working solutions containing 4 µM of
the furocoumarin in methanol (20 °C).The excitation and emis-
sion monochromators were set at the excitation or emission
maxima of the furocoumarin solutions in methanol. The slit for
the excitation monochromator was set at 15 nm and the slit for
the emission monochromator was set at 20 nm.
Determination of photobleaching of N,N-dimethyl-p-nitroso-
aniline (RNO)
Intrinsic fluorescence and dark binding of furocoumarins to
DNA
Photobleaching capacity was determined in aqueous solutions
according to a modified method of Kraljic and El Moshni [41].
Solutions were prepared in phosphate buffer (0.07 M, pH 7.4)
containing the furocoumarin (30 µM), N,N-dimethyl-p-ni-
trosoaniline (40 µM) and histidine (10 mM).The solutions were
irradiated for increasing periods of time and N,N-dimethyl-p-ni-
trosoaniline bleaching was determined by absorbance at
440 nm.Results are given as the mean of three determinations
after 3 h of UVA irradiation (output measured at 360 nm at a
distance of 13 cm:2.5 mW/cm2).Control experiments were per-
formed in the absence of histidine and showed that the tested
furocoumarins were not able to bleach RNO directly, which is
similar to the observations of Joshi and Pathak [15] who investi-
gated 8-MOP and other furocoumarins.They carried out singlet
oxygen quenching studies with azide ions and 1,4-diazabi-
cyclo[2,2,2]octane to obtain additional evidence for singlet oxy-
gen, and they also showed, that superoxide dismutase, which
is a superoxide radical anion quencher, has no effect on the
RNO bleaching test, suggesting that superoxide radical anion,
which is produced by furocoumarins, too, does not influence
the values of RNO bleaching [15, 16].
An aqueous DNA stock solution was prepared by dissolving
calf-thymus DNA (1.2 mM) in Tris-buffer (0.05 M; pH 7.4, 25 °C)
containing 5 mM NaCl;working solutions contained 10–600 µM
DNA. Drug and DNA-drug solutions were prepared by addition
of the compounds from a stock solution in DMSO (400 µM) to
give a final drug concentration of 4 µM.The solutions were incu-
bated for 15 min in the dark. The maxima and the isosbestic
points for furocoumarin and DNA-furocoumarin complex, and
fluorescence excitation and emission spectra were deter-
mined.The interaction of furocoumarins with DNA was investi-
gated by monitoring their intrinsic fluorescence at the isosbes-
tic points in the free and bound (DNA-furocoumarin complex)
states and in addition by recording the fluorescence excitation
and emission spectra.The slit for the excitation monochromator
was set at 15 nm and the slit for the emission monochromator
was set at 20 nm.
UVA irradiation
Irradiation was performed by means of a 30-W Philips TL 36
P25/07 N lamp, wavelength range 320–420 nm, maximal emis-
sion at 350 nm; output measured at 360 nm with a Holtkamp
UVA meter 360, was 2.5 and 0.9 mW/cm2 at a distance of 13
and 26 cm, respectively.
Log P determination
There is a linear relationship between the capacity factor log kЈ
of compounds and their log P values [31]. The log kЈ values of
five compounds (methylbenzene (0.04), ethylbenzene (0.19),
propylbenzene (0.37), butylbenzene (0.56), and pentylben-
zene (0.74)) with known log P values (2.69, 3.15, 3.68, 4.26,
4.82) [45] were determined as reference values. A plot of log kЈ
versus log P generated from this calibration mixture was used
for the calculation of log P values of furocoumarins, which log kЈ
values were determined by the same procedure as for the refer-
ence compounds. Chromatographic retention data are ex-
pressed by the logarithm of the capacity factor log kЈ, defined
as log (tr – t0)/t0, where tr and t0 are the retention times of the
furocoumarin and of a non-retained compound (methanol).So-
lutions of the furocoumarins in methanol were injected into the
chromatograph. Methanol/water/phosphoric acid (77/23/0.1)
pH 3.1 was used as eluant (flow rate 1 mL/min). Because this
reversed-phase HPLC procedure was also used as standard
method for the determination of log P values of carbonic acids
(data not published), a pH of 3.1 was set.
Thermal transition studies
Absorbance-temperature profiles were determined. The tem-
perature was directly measured in the cuvettes with a digital
thermometer. Heating was applied at a rate of 1° min–1 with ab-
sorbance (260 nm) and temperature data sampling at 1 min in-
tervals. DNA helix to coil transition temperatures (Tm) were de-
termined at the midpoint of the melting profiles. Furocoumarin
induced alterations in the DNA melting behavior are given by:
∆Tm = Tm (DNA + furocoumarin) – Tm (DNA); results are given
as the mean of two to three determinations. Calf-thymus DNA
and poly(dA-dT)-poly(dA-dT)-DNA were used as DNA com-
pounds.
Calf-thymus DNA: an aqueous DNA-stock solution was pre-
pared by dissolving calf-thymus DNA (1.2 mM) in MOPS-buffer
(0.1 M, pH 7.0); working solutions contained 100 µM DNA.

Arch. Pharm. Pharm. Med. Chem. 2002, 335, 187–199
3-Alkyl- and 3-Aryl-7H-furo[3,2-g]-1-benzo. . . 199
[24] D. R. Bender, J. E. Hearst, H. Rapoport, J. Org. Chem.
1979, 44, 2176–2180.
References
[1] J. A. Parrish, T. B. Fitzpatrick, L. Tannenbaum, M. A.
[25] Y.-L. Chen, T.-C. Wang, S.-C. Liang, C.-M. Teng, C.-C.
Pathak, N. Engl. J. Med. 1974, 291, 1207–1211.
Tzeng, Chem. Pharm. Bull. 1996, 44, 1591–1595.
[2] D. Averbeck, Photochem. Photobiol. 1989, 50, 859–882.
[3] F. P. Gasparro, Photochem. Photobiol. 1996, 63, 553–557.
[4] R. L. Edelson, Ya l e J. B i o l . M e d . 1989, 62, 565–577.
[26] E. P. Lysenko, A. Y. Potapenko, V. L. Sukhorukov, Bio-
physics. 1988, 33, 804–807.
[27] E. P. Lysenko, V. L. Sukhorukov, Z.Vunderlikh, L. I. Shago-
va, G. A. Kuznetsova, L. V. Kuz’mina, A. Y. Potapenko,
Russ. J. Phys. Chem. 1989, 63, 1518–1521.
[5] H. P. van Iperen, G. M. J. Beijersbergen van Henegouwen,
J. Photochem.Photobiol. B 1997, 39, 99–101.
[28] F. Dall’Acqua, D. Vedaldi. F. Bordin, F. Baccichetti, F. Car-
lassare, M.Tamaro, P. Rodighiero, G. Pastorini, A. Guiotto,
G. Recchia, M. Cristofolini, J. Med. Chem. 1983, 26, 870–
876.
[6] O.Gia, A.Anselmo, M.T.Conconi, C.Antonello, E.Uriarte,
S. Caffieri, J. Med. Chem. 1996, 39, 4489–4496.
[7] D. Averbeck, E. Moustacchi, Photochem. Photobiol. 1980,
31, 475–478.
[29] F.P.Gasparro, R.Dall’Amico, D.Goldminz, E.Simmons, D.
Weingold, Yale J. Biol. Med. 1989, 62, 579–593.
[8] F.Dall’Acqua, D.Vidaldi, S.Caffierii, A.Guiotto, P.Rodighi-
ero, F. Baccichetti, F. Carlassare, F. Bordin, J. Med. Chem.
1981, 24, 178–184.
[30] P. Vicini, E. Fisicaro, M. T. Lugari, Arch. Pharm. Pharm.
Med. Chem. 2000, 333, 135–144.
[9] P. Rodighiero, M. Palumbo, S. Marciani Magno, P. Manzini,
O. Gia, R. Piro, A. Guiotto, J. Heterocycl. Chem. 1986, 23,
1405–1410.
[31] S. H. Unger, J. R. Cook, J. S. Hollenberg, J. Pharm. Sci.
1978, 67, 1364–1367.
[32] D.Vedaldi, S.Caffieri, S.Frank, F.Dall’Acqua, A.Jakobs, J.
[10] F. Bordin, F. Carlassare, M.T. Conconi, A. Capozzi, F. Ma-
jone, A. Guiotto, F. Baccichetti, Photochem. Photobiol.
1992, 55, 221–229.
Piette, Farmaco 1995, 50, 527–536.
[33] M. Sasaki, I. Nakasato, H. Sugiura, H. Fujita, T. Sakata,
Photochem. Photobiol. 1987, 46, 551–555.
[11] A.Chilin, C.Marzano, A.Guiotto, P.Manzini, F.Baccichetti,
F. Carlasse, F. Bordin, J. Med. Chem. 1999, 42, 2936–
2945.
[34] M. Sasaki, F. Meguro, E. Kumazawa, H. Fujita, H. Kakishi-
ma, T. Sakata, Mutat. Res. 1988, 197, 51–58.
[12] J. Moron, C. H. Nguyen, E. Bisagni, J. Chem. Soc. Perkin
Trans. I, 1983, 225–229.
[35] P.-S. Song, K. J. Tapley Jr., Photochem. Photobiol. 1979,
29, 1177–1197.
[13] C. Antonello, G. Zagato, S. Bobilio, C. Marzano, O. Gia, E.
[36] V. Adam, M. Berger, J. Cadet, F. DallЈAcqua, B. Epe, S.
Frank, D. Ramaiah, S. Raoul, C. Saha-Möller, D. Vedaldi,
Photochem. Photobiol. 1996, 63, 768–778.
Uriarte, Farmaco 1994, 49, 277–280.
[14] Z. Zarebska, J. Photochem. Photobiol. B. 1994, 23, 101–
109.
[37] M. Gupta, R. Ali, J. Biochem. 1984, 95, 1253–1257.
[15] P. C. Joshi, M. A. Pathak, Biochem. Biophys. Res.
Commun. 1983, 112, 638–646.
[38] A. Oroskar, G. Olack, M.J. Peak, F. P. Gasparro, Photo-
chem. Photobiol. 1994, 60, 567–573.
[16] M. A. Pathak, P. C. Joshi, Biochim. Biophys. Acta 1984,
798, 115–126.
[39] J. P. Demaret, S. Brunie, J. P. Ballini, P.Vigny, Photochem.
Photobiol. 1989, 50, 7–21.
[17] D.Vedaldi, S. Frank, A. Pozzan, G. Miolo, F. Dall’Acqua, S.
Caffieri, P. Rodighiero, A. Guiotto, Farmaco 1995, 50,
615–623.
[40] F. Dall’Acqua, S. Marciani, G. Rodighiero, Z. Naturforsch.
1971, 26b, 561–569.
[41] I.Kraljic, S.A.El Moshni, Photochem.Photobiol.1978, 28,
[18] M. Ifster, P. Fossa, G. Menozzi, L. Mosti, F. Baccichetti, C.
577–581.
Marzano, M. Simonato, Farmaco 1995, 50, 669–678.
[42] O. Dann, G. Illing, Liebigs Ann. Chem. 1957, 605, 158–
[19] F.Bordin, F.Baccichetti, C.Marzano, F.Carlassare, G.Mio-
lo, A. Chilin, A. Guiotto, Photochem. Photobiol. 2000, 71,
254–262.
167.
[43] S. Lam, E. Grushka, J. Chassin, J. Anal. Chem. 1978, 50,
1398–1399.
[20] J. N. Ray, S. S. Silooja, V. R. Vaid, J. Chem. Soc. 1935,
[44] H. Wulff, H. Rauer, T. Düring, C. Hanselmann, K. Ruff, A.
Wrisch, S. Grissmer, W. Hänsel, J. Med. Chem. 1998, 41,
4542–4549.
813–816.
[21] J. K. MacLeod, B. R. Worth, R. J. Wells, Aust. J. Chem.
1978, 31, 1533–1544.
[45] H. Kaup, Ph. D. Thesis, University of Kiel, 1996.
[22] V. N. Gupta, T. R. Seshadri, J. Sci. Ind. Res. 1935, 16B,
257–260.
[23] S. Ragaswami, T. R. Seshadri, Proc. Indian Acad. Sci.
Sect. A 1937, 5, 112–114.