Novel pyrone side tetracyclic psoralen derivatives: Synthesis and photobiological evaluation

Article abstract of DOI:10.1021/jm0210919

This study reports the synthesis of tetrahydrobenzo- (4-6) and benzopsoralen (7-9) derivatives obtained by condensing the fourth ring to the pyrone side of the tricyclic psoralen moiety. The new compounds are characterized by having a methoxy, a hydroxy, or a dimethylaminopropoxy side chain inserted at position 8 of the psoralen chromophore. The evaluation of the photoantiproliferative activity on human tumor cell lines along with skin phototoxicity on guinea pigs revealed an interesting photobiological pattern for the dimethylaminopropoxy derivatives 6 and 9: they are in fact able to exert an antiproliferative effect up to 1 order of magnitude higher than that of the well-known drug 8-MOP, but they are devoid of skin phototoxicity. The ability of both 6 and 9 to photoadd to DNA is demonstrated by the isolation and characterization of the 4′,5′-monoadducts. AM1 calculations were also performed to gain further insight into the molecular basis of their photobiological behavior.

Full text of DOI:10.1021/jm0210919

3800
J . Med. Chem. 2003, 46, 3800-3810
Novel P yr on e Sid e Tetr a cyclic P sor a len Der iva tives: Syn th esis a n d
P h otobiologica l Eva lu a tion
Lisa Dalla Via,*,† Eugenio Uriarte,‡ Elias Quezada,‡ Alessandro Dolmella,† Maria Grazia Ferlin,† and
Ornella Gia†
Department of Pharmaceutical Sciences, University of Padova, Via Marzolo 5, 35131 Padova, Italy, and
Department of Organic Chemistry, University of Santiago de Compostela, Santiago de Compostela, Spain
Received November 8, 2002
This study reports the synthesis of tetrahydrobenzo- (4-6) and benzopsoralen (7-9) derivatives
obtained by condensing the fourth ring to the pyrone side of the tricyclic psoralen moiety. The
new compounds are characterized by having a methoxy, a hydroxy, or a dimethylaminopropoxy
side chain inserted at position 8 of the psoralen chromophore. The evaluation of the
photoantiproliferative activity on human tumor cell lines along with skin phototoxicity on guinea
pigs revealed an interesting photobiological pattern for the dimethylaminopropoxy derivatives
6 and 9: they are in fact able to exert an antiproliferative effect up to 1 order of magnitude
higher than that of the well-known drug 8-MOP, but they are devoid of skin phototoxicity.
The ability of both 6 and 9 to photoadd to DNA is demonstrated by the isolation and
characterization of the 4′,5′-monoadducts. AM1 calculations were also performed to gain further
insight into the molecular basis of their photobiological behavior.
In tr od u ction
to obtain new tetracyclic structures that showed a
photobinding ability toward DNA comparable to that
of 8-MOP, notwithstanding their monofunctional be-
havior.10 Moreover, some methylpsoralen and hydroxy-
methylpsoralen derivatives, also characterized by the
presence of a fourth ring (benzenic or cyclohexenylic)
fused to the 4′,5′-position, stimulated a further interest
in these tetracyclic structures. Indeed, they appeared
to be almost devoid of skin phototoxicity and character-
ized by a better ability to interact with DNA versus
8-MOP.12 Nevertheless, the biological potentiality of this
new moiety appeared to be limited by a low aqueous
solubility. In this connection, a subsequent line of
research proposed the insertion in the tetracyclic chro-
mophore of a basic side chain protonable at physiological
pH. The results obtained by inserting a dimethylami-
nopropoxy side chain in position 11 or 5, in analogy with
the well-known drugs 8-MOP and 5-MOP, appeared
very promising.13,14 In particular, for the new tetracyclic
water-soluble structures thus obtained, an effective
interaction with DNA along with a remarkable anti-
proliferative activity was achieved. These characteristics
appeared to be more pronounced for the analogues of
8-MOP than for those of 5-MOP. In addition, noteworthy
was the decrease of erythema induction for the tetrahy-
drobenzo derivatives and the lack of any skin photo-
sensitizing effects for the benzo derivatives, in contrast
to the parent drugs.14
Photochemotherapy constitutes an intriguing ap-
proach for the treatment of hyperproliferative diseases.1
PUVA therapy (psoralen plus UVA light) plays a
significant role in this field. Its usefulness in the
treatment of skin hyperproliferative diseases (such as
psoriasis)2-4 or for the cutaneous T-cell lymphoma5,6 has
been well-established. However, PUVA treatment is
limited by both short-term (erythema, hyperpigmenta-
tion) and long-term (benign keratoses, premalignant
keratoses, skin cancers) side effects. Of the psoralen
structures, 8-methoxypsoralen (8-MOP) is the drug most
widely employed in PUVA therapy, but also its congener
5-methoxypsoralen (5-MOP) is successfully employed,
even thought to a lesser extent.3,7 Their cellular target-
(s) and their molecular mechanism of action have been
already clarified. In particular, psoralens are able to
absorb a photon from UVA light (365 nm) to give rise
to a covalent adduct with the pyrimidine bases of DNA,
mainly thymine, by means of a photoaddition reaction
involving the 4′,5′-furan-side double bond and/or the 3,4-
pyrone-side double bond of the tricyclic moiety and the
5,6-double bond of the base. The ability of the furan side
monoadduct to photoreact further, causing the 3,4-
double bond to form a diadduct, has been demonstrated
for psoralens.8
With the aim of developing drugs endowed with a
better photochemotherapeutic pattern, i.e., high anti-
proliferative ability along with lower skin photosensi-
tization, a large number of new derivatives have been
synthesized and studied.9-11 In particular, a synthetic
approach led to the condensation of a benzenic ring at
the photoreactive 4′,5′-double bond, making it possible
On the basis of such promising results, it appeared
to be of interest to pursue research into the tetracyclic
psoralen derivatives. In this connection, the condensa-
tion of the fourth nucleus (benzenic or cyclohexenylic)
at the pyrone side of the psoralen molecule,15-17 instead
of the furan one, can allow further in-depth investiga-
tion into the photobiological role of the tetracyclic
psoralen moieties.
* To whom correspondence should be addressed. Phone: +39 049
8275712. Fax: +39 049 827 5366. E-mail: lisa.dallavia@unipd.it.
† University of Padova.
In this paper the synthesis of six new tetracyclic
derivatives, characterized by having the fourth aromatic
‡ University of Santiago de Compostela.
10.1021/jm0210919 CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/06/2003

Pyrone Side Tetracyclic Psoralen Derivatives
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18 3801
Sch em e 1a
or cyclohexenyl ring condensed at the 3,4-double bond
of the pyrone belonging to the psoralen moiety, is
reported. Both tetrahydrobenzo- (4-6) and benzo de-
rivatives (7-9) carrying a methoxy, a hydroxy, and a
dimethylaminopropoxy substituent at position 11 were
studied. In particular, the antiproliferative activity on
human tumor cell lines, skin phototoxicity, and inter-
action with DNA were investigated. Finally, the isola-
tion and characterization of the furan photoadduct
between both the tetrahydrobenzo- (6) and benzodim-
ethylaminopropoxy derivatives (9) with thymine are
reported.
Furthermore, we have tried to devise a very simple
theoretical model of the binding of 6, 9, and the related
compounds having the fourth ring (cyclohexenylic or
benzenic) fused at the furan side to DNA, to gain further
possible hints about the photobiological role of these
moieties.
Resu lts a n d Discu ssion
Ch em istr y. The compounds studied (4-9) were
obtained starting from 2-methoxyresorcinol (1), as
shown in Scheme 1. Pechmann condensation by treat-
ment of compound 1 with ethyl 2-oxocyclohexanecar-
boxylate afforded the corresponding hydroxycoumarin
2, which through a Williamson reaction with chloro-
acetone (refluxing acetone/K2CO3, 24 h) gave the oxo
ether 3 in 71% yield. Cyclization of 3 by heating in
strongly alkaline solution afforded the tetrahydroben-
zofurocoumarin 4 in 82% yield. The unsubstituted
terminal ring of 4 was aromatized by heating with DDQ
(2,3-dichloro-5,6-dicyano-1,4-benzoquinone) in refluxing
toluene, which afforded the benzofurocoumarin 7 in 60%
yield.
The methoxypsoralen derivatives 4 and 7 were trans-
formed to 5 and 8 in 87% and 78% yields, respectively,
by hydrolysis of their methoxy groups to hydroxyl
groups with aluminum trichloride in refluxing methyl-
ene chloride. Finally, treatment of compounds 5 and 8
with 3-chloro-N,N-dimethylpropylamine and NaH in the
presence of NaI in refluxing dimethylformamide af-
forded compounds 6 and 9 in 62% and 67% yields,
respectively.
P h otobiologica l Activity. The antiproliferative ac-
tivity of 4-9 was evaluated on two human tumor cell
lines: HeLa (human cervix adenocarcinoma cells) and
HL-60 (human promyelocytic leukemic cells). The well-
known photochemotherapeutic drug 8-MOP was tested
under the same experimental conditions and considered
as reference compound. The results have been expressed
as IC50 values, i.e., the concentration of compound able
to induce the death of 50% of the cells with respect to a
control culture. Table 1 reports the data relative to all
new tetracyclic compounds and 8-MOP, for both cell
lines taken into account.
a
Reagents: (a) ethyl 2-oxocyclohexanecarboxylate, H2SO4; (b)
chloroacetone, K2CO3, acetone; (c) NaOH; (d) DDQ, toluene; (e)
AlCl3, CH2Cl2; (f) 3-chloro-N,N-dimethylpropylamine, NaH, NaI,
DMF.
Ta ble 1. Cell Growth Inhibition and Skin Phototoxicity in
Guinea Pigs in the Presence of Examined Compounds and
8-MOP as Reference Drug
IC50 (µM ) of cell lines
HeLa
UVA
HL-60
erythema
intensitya
compd dark
dark
>20
>20
UVA
4
5
6
7
8
9
>20 15.7 ( 1.3
>20 17.4 ( 0.2
>20 0.68 ( 0.08
>20 >20
13.0 ( 0.8
17 ( 1
0.95 ( 0.13
11.8 ( 0.3
>20
-
-
16 ( 1
-
>20
>20
-
-
>20 >20
>20 0.64 ( 0.16
9.9 ( 0.7 0.67 ( 0.11
5.4 ( 0.7
-
8-MOP >20
10 ( 3
>20
+ (with
edema)
a
+, strong; -, absent.
After exposure to UVA light (0.793 J cm-2 at 365 nm)
the compounds characterized by having the dimethyl-
aminopropoxy side chain, both tetrahydrobenzo- (6) and
benzopsoralen (9), are able to exert a noticeable anti-
proliferative ability. In detail, the measured IC50 values
appear to be from 5 to 15 times lower with respect to
those obtained for the reference compound.
to that detected for the drug and always significantly
lower if compared with that exerted by the dimethyl-
aminopropoxy derivatives, in both the examined cell
lines. With respect to the hydroxy derivatives, it can be
stated that they are the less active compounds. Par-
ticularly, 8 is inactive toward both cell lines, while for
5 a very low level of activity can be detected.
As regards the methoxy derivatives (4 and 7), they
show an antiproliferative activity weaker with respect

3802 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18
Dalla Via et al.
The absence of cytotoxic effects in the dark, as usually
noticed with the psoralen moiety, is also confirmed for
the new compounds, except that in the case of the most
active, 6 and 9, in the HL-60 cells. Nevertheless, the
measured values are more than 1 order of magnitude
higher with the respect to those obtained upon UVA
irradiation.
The skin phototoxicity of the new tetracyclic deriva-
tives was determined by evaluating the appearance of
erythema, a marker of cutaneous sensitization, on the
depilated skin of guinea pigs according to the procedure
described in the Experimental Section. The results,
reported in Table 1, clearly indicate that all the new
derivatives, unlike 8-MOP, are nonphototoxic, even
when tested at concentrations higher than that of the
reference drug.
With regard to the benzopsoralen derivatives 7-9, the
lack of skin phototoxicity does not constitute an unusual
effect. Indeed, the same result has already been ob-
tained also for the benzopsoralens characterized by the
condensation of the fourth ring at the 4′,5′-double bond
and carrying the same chromophoric structure as 7-9,
as previously published.14 In contrast, some interesting
features emerge if the two tetracyclic tetrahydroben-
zopsoralen moieties, derived from the condensation of
the fourth ring at the 4′,5′- (see ref 14) or 3,4-double
bond (compound 4-6), are compared. In this case, the
condensation at the pyrone side photoreactive double
bond appears to be clearly more promising with respect
to that at the furan side. Indeed, the dimethylamino-
propoxy derivative carrying the cyclohexenyl ring fused
at 4′,5′ showed a certain phototoxic ability,14 while the
correspondent compound 6 appears to be devoid of
erythematous effect, even though both are able to exert
a noticeable and comparable cytotoxic activity on both
HeLa and HL-60 cells.
F igu r e 1. Linear flow dichroism spectra for compounds 6 (A)
and 9 (B) at different [drug]/[DNA] ratios: a ) 0; b ) 0.02; c
) 0.04; d ) 0.08. [DNA] ) 1.6 × 10-3 M.
Non cova len t Bin d in g to DNA. The ability of the
tricyclic psoralen structure in the ground state to form
a complex with DNA via an intercalation mode of
binding is well-established. The formation of a molecular
complex between the tetracyclic derivatives 6 and 9 and
the macromolecule has been investigated by means of
flow linear dichroism experiments. The spectra of DNA
solutions in the presence of increasing amounts of the
considered compounds are reported in Figure 1A (com-
pound 6) and B (compound 9).
that are polarized in the plane of the chromophore,18
the noted negative signal must be ascribable to a
parallel orientation to the plane of DNA bases. This is
in agreement with an intercalative mode of binding.
Flow linear dichroism experiments have been per-
formed also for compounds 4, 5, 7, and 8. Nevertheless,
for these derivatives, no significant dichroic signal
appeared at high wavelengths, thus indicating a limited
ability to interact with the macromolecule.
P h otobin d in g to DNA. It is well-known that upon
UVA irradiation, the 4′,5′-and 3,4-double bonds of the
psoralen chromophore are able to give rise to a C4-
cycloaddition involving the 5,6-double bond of pyrimi-
dine bases, mainly thymine. Previous studies have
demonstrated that the presence of a fourth ring con-
densed to the furan side can variously influence its
capacity to photoreact. In detail, the condensation of a
cyclohexenyl ring did not compromise its ability to form
a photoadduct, while in the presence of a benzene ring
this capacity appeared completely abolished.12,14 In
particular, the electronic delocalization that takes place
in this latter case in the tetracyclic system may be
considered responsible for the inability of the involved
double bond to photoadd to DNA bases. On the contrary,
the steric hindrance caused by the lack of aromaticity
All dichroic spectra show a strong negative signal at
260 nm, typical of the macromolecule; nevertheless, in
the presence of the tested compounds (traces b-d), a
further dichroic signal appears at higher wavelengths
(300-360 nm). Due to the fact that no contribution from
DNA base pairs can be detected in this spectral range,
this signal has to be attributed to the added compounds
which, on the contrary, absorb in this region. Since
small molecules, such as 6 and 9, cannot themselves
become oriented in the flow field, the presence of this
signal proves that these compounds have formed a
molecular complex with DNA that allows them to
undergo an orientation. The negative LD signal of 6 and
9 at wavelengths different from those of DNA is of the
same sign as the strong LD band at 260 nm derived
from the purine and pyrimidine base pairs. Assuming
that also for benzo- and tetrahydrobenzopsoralens, like
in psoralens, all strong absorptions are πfπ* transitions

Pyrone Side Tetracyclic Psoralen Derivatives
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18 3803
spectrum (Figure 4), a doublet attributed to HC-5′ at δ
5.13 with a 3J ) 5.25 Hz due to its coupling with the
thymine HC-6 appears, which is evidence for the pho-
toadduct formation between tetrahydrobenzopsoralen
derivative 6 and the pyrimidine. The HC-6 signal occurs
at δ 3.88. Furthermore, the shift of a furan methyl
singlet from δ 2.23 in the spectrum of 6 to δ 1.42 and
that of a thymine methyl singlet from δ 1.63 in the
spectrum of thymine to δ 1.46, that is, toward the
aliphatic region, represents a further confirmation of
the occurrence of photoaddition, since it shows that the
furan ring added thymine by its 4′,5′-double bond,
yielding a no more aromatic system. Moreover, to assign
the 13C spectrum, obtain bond connectivities, and es-
tablish the multiple bond proton-carbon connectivities,
1H-13C HMQC and HMBC spectra were obtained. In
particular, the presence of four saturated carbon signals
(δ 44.9, 53.3, 54.3, and 85.1) that are assignable to the
carbons 5 and 6 of the thymine and 4′ and 5′ of the furan
ring is consistent with the occurrence of a C4-cyclo-
adduct between the 4′,5′-double bond of 6 and the 5,6-
double bond of thymine. Finally, the confirmation of the
photoadduct structure derives from HMBC analysis (see
the inset in Figure 4), which also allows us to assign to
the isolated thymine-6 photoadduct, in agreement with
previous isolated furan side adducts, a cis configura-
tion.20,21 This means that, upon intercalation, 6 photo-
adds to thymine in such a way that the psoralen moiety
and the DNA base lie on the same side of the cyclo-
butane plane.
F igu r e 2. UV absorption spectra of an ethanol solution of
furan cycloadduct obtained from salmon testes DNA and 6
before (line a) and after irradiation at 254 nm (10, 30, 60, 120
min, lines b-e, respectively).
in the hydrogenated fourth ring does not counteract the
capacity of the involved double bond to photoreact.
To evaluate the covalent photobinding process of 6
and 9, it appeared interesting to analyze the 4′,5′-
photoadduct formation.
The irradiation (365 nm) of an aqueous solution of
DNA in the presence of the two considered compounds,
followed by precipitation and acid hydrolysis, as re-
ported in the Experimental Section, allows us to isolate
a product characterized by a strong violet fluorescence.
This property is usually consistent with the molecular
structure of the furan side monoadduct.19 Figure 2
shows the UV spectrum of an ethanol solution of the
fluorescent compound obtained by the photoreaction
between 6 and DNA (line a). It can be observed that its
trend is similar to that already observed for other C4-
cycloadducts between the furan double bond of the
furocoumarins and the 5,6-double bond of pyrimidine:
The fluorescent photoproduct isolated from the pho-
toreaction between DNA and 9 was analyzed by mass
spectrometry, and the resulted spectrum is reported in
Figure 5, where the major peak, which appears at m/z
478, corresponds to the thymine-9 C4-cycloadduct.
Cr oss-Lin k in g. A peculiar property of the furan side
monoadduct is the possibility to absorb a second photon
(365 nm) to photoreact also at the 3,4-double bond, thus
giving rise to an interstrand cross-link with a pyrimi-
dine belonging to the opposite DNA strand. As already
stated, the condensation of a cyclohexenyl ring to the
4′,5′-photoreactive furan side allowed the photoaddition
to the same double bond. Consequently, they have been
demonstrated to behave as bifunctional compounds. On
the contrary, for the analogous tetracyclic benzoderiva-
tives the aromatic fourth ring at the level of the furan
double bond prevents the photoaddition, rendering these
compounds monofunctional molecules.12,14
20
an evident absorption at 330 nm appears, while the
peculiar furocoumarin band around 300 nm disappears.
To provide additional evidence, photoreversion experi-
ments were performed. Indeed, it is known that, when
irradiated at 254 nm, the C4-cycloadducts undergo
breakage, yielding the parent compounds, i.e., the
furocoumarin, and the DNA base. Figure 2 also reports
the UV absorption spectra of the photoproduct after
increasing periods of irradiation at 254 nm (lines b-e).
After 2 h, the recorded spectrum shows the disappear-
ance of the peak at 330 nm and the occurrence of the
absorption band at 301 nm, typical of 6 (line e). Similar
results have been obtained also for 9 (data not shown).
Nevertheless, confirmation of the above assumption
was obtained after the characterization of the isolated
photoproducts.
Also for the new tetracyclic benzopsoralens 7-9, the
inability to give rise to diadducts with the macromol-
ecule has been confirmed (data not shown). In this case
the fourth ring at the pyrone side did not allow the furan
monoadduct to photoreact further. Regarding the tet-
rahydrobenzo congeners 4-6, Figure 6 shows the results
obtained by denaturation-renaturation experiments, in
comparison with 8-MOP. For all the new tetrahy-
drobenzopsoralens, a certain ability to form cross-links
has been detected that decreases in the following
order: 6 > 4 > 5. Nevertheless, this ability appears to
be lower with respect to that obtained for the reference
drug.
As regards the fluorescent compound isolated from
the photoreaction between 6 and DNA, Figure 3 reports
the mass spectrum where the peak at m/z 482 is
consistent with a thymine-6 photoadduct. To charac-
terize the structure of the photoproduct, 1H and 13C
NMR spectra were performed. With regard to the 1H
Considering 6 and 9, the comparison between the
above data and the photoantiproliferative activity re-
ported in Table 1, suggests that for these moieties the

3804 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18
Dalla Via et al.
F igu r e 3. Mass spectrum of the 6-thymine photoadduct.
ability to induce bifunctional damage to the DNA does
not seem to be determinant for the cytotoxic effect,
probably also in connection with the very moderate
amount of cross-links induced by 6.
In a previous study, for a tetrahydrobenzopsoralen
analogue to 6, which differs from the latter by the
condensation of the fourth ring at the 4′,5′-double bond
rather than at the 3,4-double bond, we demonstrated a
higher capacity to form cross-links versus 8-MOP.14
On the basis of the overall obtained results, it is
possible to affirm that the condensation of the cyclo-
hexenyl ring at the 3,4-pyrone side seems to inhibit the
ability of the tetracyclic moiety to form cross-links with
respect to the condensation at the 4′,5′-furan side.
Th eor etica l Ca lcu la tion s. To investigate more in
depth the different photo-cross-linking ability of 6 with
respect to its congener characterized by having the
fourth cyclohexenyl ring condensed at the 4′,5′-photo-
reactive double bond, molecular modeling studies were
carried out. For the sake of completeness, in these
studies 9 and its congener with the benzene ring
condensed at the 4′,5′-double bond14 were also included.
As a first step, a molecular geometry optimization at
the AM1 level on all compounds was performed (Figure

Pyrone Side Tetracyclic Psoralen Derivatives
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18 3805
F igu r e 4. 1H NMR spectrum of the 6-thymine furan photoadduct. In the inset relevant HMBC spectrum regions are reported.
7). All molecules show the tetracyclic skeleton to be flat
except for the two cyclohexene carbons farthest from the
rest of the system, which can lie alternatively above and
below the chromophore mean plane, thus originating
two possible conformations. For 6, such conformations
(I, II) have practically equal energies of about -93 kcal/
mol. The two corresponding conformations of its conge-
ner (III, IV) behave similarly and have also equal
energies of about -91 kcal/mol (that is, 2 kcal/mol
higher in energy). The displacement of the cyclohexene
carbons in the two molecules from the plane of the
tetracycle was within 0.42 Å in I and II and within 0.37
Å in III and IV. With respect to the benzo derivatives,
the optimized geometry of 9 (V) has an energy of -59.4
kcal/mol, while its congener (VI) has an energy of -56.2
kcal/mol.
It is known that the occurrence of a covalent photo-
addition to the double helix requires, in the first place,
an effective intercalation complex between two base
pairs. To compare the ability of the molecules taken into
account to fit into a model of their binding site, a very
simple representation of the DNA intercalation site was
prepared by modeling in vacuo a tetranucleotide, d(TA)2,
where the two base pairs were positioned as described
in previous studies.22 Also the position of the compounds
(I-VI) was chosen in such a way as to reproduce the
same arrangement of the intercalating compound de-
scribed in such a study.22 Accordingly, the dimethylami-
nopropyl tail of all the molecules ends up in the major
groove. The intercalation site was also geometrically
optimized at the AM1 level. Figure 8 shows a represen-
tation of the interaction of 6 with the d(TA)2 tetranucle-
otide.
Finally, the energies of the interaction of I-VI with
the d(TA)2 tetranucleotide were investigated. The com-
plexes showed relative energies of 2.98 (I), 6.51 (II), 0.00
(III), 3.51 (IV), 30.7 (V), and 30.7 (VI) kcal/mol, respec-
tively. The values obtained for the tetrahydrobenzo
derivatives (I-IV) suggest that the condensation of the
fourth ring at the 4′,5′-double bond renders the psoralen
moiety better suited to fit into the intercalation pocket
with respect to the condensation at the 3,4-site. Thus,
the more unfavorable interaction energy of 6 could
account for the resulting decrease in its ability to form
cross-links when compared with the congener.
With respect to V and VI, the energy of the complexes
are the same, suggesting that no significant difference
exists between the two benzo derivatives. The energies
of the interaction of V and VI with the tetranucleotide
are much higher than those shown by I-IV. Neverthe-
less, it has to be underlined that they appear to be
mainly determined by the energies of the two benzo
derivative moieties alone.

3806 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18
Dalla Via et al.
F igu r e 5. Mass spectrum of the 9-thymine photoadduct.
Con clu sion s
both series of tetracyclic derivatives, a methoxy, a
hydroxy, or a dimethylaminopropoxy side chain was
inserted at position 11 of the chromophore.
New tetracyclic psoralen derivatives were synthesized
and their photobiological behavior studied. These com-
pounds are characterized by the condensation of a
benzenic or a cyclohexenyl ring at the photoreactive 3,4-
double bond of the psoralen moiety. Furthermore, in
The evaluation of the antiproliferative activity on
human tumor cell lines after UVA irradiation evidenced
a remarkable effect for the new compounds bearing the

Pyrone Side Tetracyclic Psoralen Derivatives
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18 3807
ability has been found, a different pattern appears in
the case of the tetrahydrobenzopsoralen congeners. In
more detail, the skin phototoxicity was decreased with
respect to the reference drug by the condensation of the
fourth cyclohexenyl ring at the 4′,5′-double bond of
psoralen moiety, but it is completely abolished if the
condensation takes place at the pyrone side. Further-
more, it is worth underlining the fact that for both
tetrahydrobenzopsoralen congeners, the antiprolifera-
tive activity is comparable and significantly higher with
respect to 8-MOP.
The study of the interaction of both 6 and 9 with DNA
in the ground state and after UVA irradiation indicated
that the behavior already demonstrated for the psor-
alens occurred. In particular, they form a complex with
DNA via an intercalative mode of binding and, after
irradiation, are able to photoadd to the macromolecule.
Specifically, the C4-cycloadducts involving the 4′,5′-
double bond of the psoralen chromophore and the 5,6-
double bond of thymine were isolated and characterized
by spectroscopic techniques.
F igu r e 6. Cross-linking of compounds 4-6 and 8-MOP to
double-stranded DNA from salmon testes (nucleotide/drug
ratio ) 75) as a function of irradiation time.
For tetrahydrobenzopsoralen derivatives 4-6, the
ability to form interstrand cross-links has been dem-
onstrated, even though it is lower with respect to that
found for 8-MOP. The comparison with the analogue of
6 endowed with the fourth cyclohexenyl ring at the 4′,5′-
double bond highlighted the differing capacities of the
two congeners to induce cross-links in the macromol-
ecule. In detail, the amount induced by 6 appears to be
significantly lower with respect to that obtained for its
analogue. The energy profile of the two compounds,
derived from theoretical calculations, accounted for the
observed behaviors. In fact a more unfavorable inter-
action energy was detected for 6, and this seems to
account for the more difficult subsequent cycloaddition
reaction.
F igu r e 7. A representation of 6 (top left), 9 (top, right), and
their furan side congeners.
In conclusion, the photobiological properties of com-
pounds 6 and 9 allow us to consider the new benzo- and
tetrahydrobenzotetracyclic moieties as an interesting
model in the development of new photochemotherapeu-
tic drugs. Moreover, a further reason for the interest in
6 lies in the possibility that it allows correlation between
chemical structures with the aim of better delineating
the structure-activity relationships that regulate the
photobiological effects of PUVA therapy.
F igu r e 8. A representation of the interaction of 6 with the
d(TA)2 tetranucleotide.
dimethylaminopropoxy side chain, both tetrahydrobenzo
(compound 6) and benzo (compound 9). Indeed, they are
able to exert a cytotoxic ability from 5 to 15 times higher
with respect to that obtained for the reference drug
8-MOP. Otherwise, the activity scored for the synthe-
sized compounds carrying the hydroxy or the methoxy
group appeared to be lower than that of 8-MOP or, in
some cases, even absent at the tested concentrations.
The evaluation of the appearance of erythema, a marker
of cutaneous photosensitization, increased the interest
in derivatives 6 and 9, due to the fact that both appear
to be clearly nonphototoxic. The insertion of a fourth
ring can be considered responsible for having conferred
this helpful feature. In this regard, taking into consid-
eration a previous study14 in which analogues to 6 and
9 characterized by the fourth ring condensed at the 4′,5′-
double bond were extensively studied, some consider-
ations can be made. It can be noted that, while for both
benzopsoralen derivatives a lack of skin photosensitizing
Exp er im en ta l Section
Melting points are uncorrected and were determined with
a Reichert Kofler thermopan or in capillary tubes in a Bu¨chi
510 apparatus. IR spectra were recorded with a Perkin-Elmer
1640FT spectrometer (KBr disks, υ in cm-1). 1H NMR (300
MHz) and 13C NMR (75.4 MHz) spectra were recorded with a
Bruker AMX spectrometer, using TMS as internal standard
(chemical shifts in δ values, J in hertz). Mass spectrometry
was carried out with a Kratos MS-50 or a Varian MAT-711
spectrometer. Elemental analyses were performed by a Perkin-
Elmer 240B microanalyzer and were within (0.4% of calcu-
lated values in all cases. Flash chromatography (FC) was
performed on silica gel (Merck 60, 230-400 mesh); analytical
TLC was performed on precoated silica gel plates (Merck 60
F
254, 0.25 mm).
3,4-Cycloh exen e-7-h yd r oxy-8-m et h oxycou m a r in (2).
Ethyl 2-oxocyclohexanecarboxylate (16.39 mL, 17.49 g, 102.76
mmol) was added to 2-methoxyresorcinol (1, 12.0 g, 85.63
mmol) in concentrated H2SO4 (120 mL) at 0 °C. The mixture
was stirred for 10 h at room temperature, poured into ice-
water (500 mL), and left overnight. A yellow precipitate was

3808 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18
Dalla Via et al.
recovered by filtration, washed with water, and purified by
FC with 1:1 hexane-ethyl acetate as eluent, giving pure 2
(19.55 g, 93%). Mp: 182-183 °C. IR: 3363, 2941, 1689, 1577,
1315, 1092, 798. 1H NMR (DMSO-d6): 1.72 (m, 4H, CH2(CH2)2-
CH2), 2.38 (m, 2H, CH2 in 4), 2.70 (m, 2H, CH2 in 3), 3.81 (s,
3H, CH3O), 6.85 (d, J ) 8.80, 1H, H-6), 7.26 (d, J ) 8.80, 1H,
H-5), 10.16 (bs, 1H, HO). 13C NMR (DMSO-d6): 20.77 (CH2),
21.11 (CH2), 23.39 (CH2), 24.61 (CH2), 60.46 (CH3), 112.76
(CH), 118.25 (C), 118.43 (C), 118.68(CH), 133.89 (C), 145.88
(C), 147.83 (C), 152.23 (C), 160.51(C). MS m/z (%): 247 ([M +
1]+, 67), 246 (M+, 100), 230 (44), 218 (42), 190 (68), 157 (14).
Anal. C14H14O4: C, H.
then treated with ion-exchange resin IRA-4000, giving pure 6
(185 mg, 62%). Mp: 112-113 °C. IR: 3418, 2938, 1705, 1591,
1400, 1115, 753.1H NMR (DMSO-d6): 1.78 (m, 4H, CH2(CH2)2-
CH2), 1.85 (m, 2H, CH2CH2CH2), 2.12 (s, 6H, (CH3)2N), 2.23
(d, J ) 1.30, 3H, CH3 in 4′), 2.43 (t, J ) 7.00, 2H, CH2N), 2.43
(m, 2H, CH2 in 4), 2.88 (m, 2H, CH2 in 3), 4.40 (t, J ) 6.40,
2H, CH2O), 7.56 (s, 1H, H-5), 7.85 (d, J ) 1.30, 1H, H-5′). 13C
NMR (DMSO-d6): 7.29 (CH3), 20.76 (CH2), 20.98 (CH2), 23.58
(CH2), 25.02 (CH2), 27.65 (CH2), 45.10 (2 CH3), 55.35 (CH2),
71.50 (CH2), 107.08 (CH), 115.61 (C), 116.69 (C), 120.43 (C),
126.56 (C), 130.37 (C), 140.99 (C), 143.27 (CH), 146.20 (C),
147.60 (C), 160.21 (C). MS m/z (%): 356 ([M + 1]+, 10), 355
(M+, 44), 310 (4), 270 (16), 214 (12), 58 (100). Anal. C21H25
NO4: C, H, N.
-
7-Aceton yloxy-3,4-cycloh exen e-8-m eth oxycou m ar in (3).
To a solution of 2 (19.27 g, 78.25 mmol) in dry acetone (370
mL) were added chloroacetone (14.48 g, 156.50 mmol) and K2-
CO3 (21.63 g, 156.50 mmol), and the reaction mixture was
refluxed for 24 h. The precipitate was filtered out and the
filtrate was concentrated under vacuum to a solid residue that
was purified by FC with 6:4 hexane-ethyl acetate as eluent,
giving pure 3 (16.85 g, 71%). Mp: 141-143 °C. IR: 2941, 1734,
3,4-Ben zo-4′-m eth yl-8-m eth oxyfu r o[3,2-g]cou m a r in (7).
A solution of tetrahydrobenzopsoralen 4 (1.0 g, 3.517 mmol)
and DDQ (0.798 g, 3.517 mmol) in toluene (80 mL) was
refluxed for 6 h. The mixture was cooled, the precipitate
collected, the solvent evaporated under reduced pressure, and
the resulting residue purified by FC with CH2Cl2 as eluent,
giving pure 7 (591 mg, 60%). Mp: 193-194 °C. IR: 2946, 1713,
1600, 1386, 1245, 1115, 840. 1H NMR (CDCl3): 2.27 (d, J )
1.30, 3H, CH3), 4.27 (s, 3H, CH3O), 7.40 (bs, 1H, H-5′), 7.49
(m, 1H, CH-CH-C3), 7.68 (s, 1H, H-5), 7.77 (m, 1H, CH-
CH-C4), 8.09 (m, 1H, CH-C4), 8.31 (m, 1H, CH-C3). 13C
NMR (CDCl3): 7.85 (CH3), 61.22 (CH3), 105.57 (CH), 115.03
(CH), 115.83 (C), 120.19 (C), 121.87 (CH), 127.70 (C), 128.25
(CH), 130.60 (CH), 133.16 (C), 134.74 (CH), 134.87 (CH),
135.61 (C), 142.72 (CH), 146.62 (C), 160.86 (C). MS m/z (%):
281 ([M + 1]+, 18), 280 (M+, 100), 237 (70), 209 (7), 181 (10),
152 (32). Anal. C17H12O4: C, H.
1
1697, 1609, 1302, 1115, 795. H NMR (CDCl3): 1.82 (m, 4H,
CH2(CH2)2CH2), 2.31 (s, 3H, CH3CO), 2.55 (m, 2H, CH2 in 4),
2.72 (m, 2H, CH2 in 3), 4.01 (s, 3H, CH3O), 4.70 (s, 2H, CH2O),
6.73 (d, J ) 8.90, 1H, H-6), 7.22 (d, J ) 8.90, 1H, H-5). 13C
NMR (CDCl3): 21.90 (CH2), 22.18 (CH2), 24.53 (CH2), 25.86
(CH2), 27.12 (CH3), 62.23 (CH3), 74.67 (CH2), 110.50 (CH),
116.66 (C), 118.65 (CH), 122.22 (C), 137.19 (C), 147.03 (C),
147.70 (C), 152.83 (C), 161.85 (C), 205.22 (C). MS m/z (%): 302
(M+, 100), 259 (26), 245 (55), 214 (17), 128 (14), 115 (17). Anal.
C17H18O5: C, H.
3,4-Cycloh exen e-4′-m et h yl-8-m et h oxyfu r o[3,2-g]cou -
m a r in (4). A mixture of 3 (16.16 g, 53.45 mmol) and 1 M
NaOH (280 mL) was refluxed for 24 h, cooled, and acidified
with 3 M HCl. The precipitate was recovered by filtration,
washed with water, and purified by FC with 9:1 hexane-ethyl
acetate as eluent, yielding pure 4 (12.45 g, 82%). Mp: 204-
205 °C. IR: 3112, 2945, 1711, 1623, 1590, 1343, 1158, 827. 1H
NMR (CDCl3): 1.84 (m, 4H, CH2(CH2)2CH2), 2.25 (d, J ) 1.25,
3H, CH3), 2.57 (m, 2H, CH2 in 4), 2.83 (m, 2H, CH2 in 3), 4.25
(s, 3H, CH3O), 7.25 (s, 1H, H-5), 7.42 (d, J ) 1.25, 1H, H-5′).
13C NMR (CDCl3): 7.74 (CH3), 21.38 (CH2), 21.53 (CH2), 23.96
(CH2), 25.67 (CH2), 61.14 (CH3), 106.08 (CH), 115.84 (C),
117.33 (C), 121.60 (C), 127.10 (C), 132.16 (C), 141.28 (C),
142.43 (CH), 146.53 (C), 147.45 (C), 161.38 (C). MS m/z (%):
285 ([M + 1]+, 18), 284 (M+, 100), 256 (19), 228 (52), 191 (5),
128 (8). Anal. C17H16O4: C, H.
3,4-Ben zo-8-h yd r oxy-4′-m eth ylfu r o[3,2-g]cou m a r in (8).
This compound was prepared from 7 (712 mg) in the same way
as 5 from 4. The crude product was purified by FC with 99:1
methylene chloride-methanol as eluent, yielding pure 8 (525
mg, 78%). Mp: 288 °C (dec). IR: 3385, 2923, 1713, 1605, 1302,
1
1121, 766. H NMR (DMSO-d6): 2.28 (d, J ) 1.10, 3H, CH3),
7.63 (m, 1H, CH-CH-C3), 7.83 (d, J ) 1.10, 1H, H-5′), 7.93
(m, 1H, CH-CH-C4), 8.02 (s, 1H, H-5), 8.24 (dd, J ) 0.80
and 7.84, 1H, CH-C4), 8.50 (m, 1H, CH-C3), 10.51 (bs, 1H,
HO). 13C NMR (DMSO-d6): 7.52 (CH3), 103.34 (CH), 114.43
(C), 115.69 (C), 119.31 (C), 122.80 (CH), 126.64 (C), 128.32
(CH), 129.56 (CH), 130.32 (C), 135.06 (CH), 135.52 (C), 137.54
(C), 143.15 (CH), 144.36 (C), 160.17 (C). MS m/z (%): 267 ([M
+ 1]+, 17), 266 (M+, 100), 237 (8), 210 (7), 181 (16), 152 (12).
Anal. C16H10O4: C, H.
3,4-Cycloh exen e-8-h yd r oxy-4′-m et h ylfu r o[3,2-g]cou -
m a r in (5). A mixture of AlCl3 (844 mg, 6.331 mmol) and dry
CH2Cl2 (25 mL) was stirred for 2 h at room temperature. A
solution of compound 4 (600 mg, 2.11 mmol) in dry CH2Cl2
(15 mL) was added, and stirring was continued for another 3
h. The reaction mixture was then acidified with HCl and
extracted with CH2Cl2 (4 × 100 mL), the extract was dried
(Na2SO4), and the solvent was evaporated under reduced
pressure, leaving a residue that upon purification by FC with
98:2 CH2Cl2/MeOH as eluent yielded pure 5 (497 mg, 87%).
Mp: 312-314 °C. IR: 3306, 2956, 1696, 1592, 1290, 1112, 824.
1H NMR (DMSO-d6): 1.77 (m, 4H, CH2(CH2)2CH2), 2.22 (d, J
) 1.30, 3H, CH3), 2.43 (m, 2H, CH2 in 4), 2.86 (m, 2H, CH2 in
3), 7.33 (s, 1H, H-5), 7.80 (d, J ) 1.30, 1H, H-5′), 10.41 (bs,
1H, HO). 13C NMR (DMSO-d6): 7.45 (CH3), 20.84 (CH2), 21.08
(CH2), 23.60 (CH2), 25.09 (CH2), 103.50 (CH), 115.63 (C),
116.64 (C), 120.19 (C), 126.13 (C), 129.48 (C), 138.13 (C),
143.04 (CH), 144.30 (C), 148.05 (C), 160.53 (C). MS m/z (%):
271 ([M + 1]+, 18), 270 (M+, 100), 242 (53), 227 (28), 214 (74),
201 (20), 177 (19), 115 (17). Anal. C16H14O4: C, H.
3,4-Ben zo-8-(3-d im eth yla m in op r op oxy)-4′-m eth ylfu r o-
[3,2-g]cou m a r in (9). This compound was prepared from 8
(150.0 mg) in the same way as 6 from 5. The crude product
was purified by FC with 98:2 methylene chloride-methanol
as eluent, giving pure 9‚HCl (146 mg, 67%). Mp: 237 °C. IR:
2923, 2700, 1720, 1475, 1298, 1115, 769. 1H NMR (DMSO-
d6): 2.14 (m, 2H, CH2CH2CH2), 2.30 (d, J ) 1.15, 3H, CH3 in
4′), 2.87 (s, 6H, (CH3)2N), 3.41 (m, 2H, CH2N), 4.48 (t, J )
5.75, 2H, CH2O), 7.70 (m, 1H, CH-CH-C3), 7.91 (d, 1H, J )
1.15, H-5′), 7.97 (dt, 1H, J ) 7.80 and 1.20, CH-CH-C4), 8.25
(dd, 1H, J ) 7.80 and 1.10, CH-C4), 8.33 (s, 1H, H-5), 8.57
(m, 1H, CH-C3). 13C NMR (DMSO-d6): 7.47 (CH3), 24.73
(CH2), 42.52 (2 CH3), 54.37 (CH2), 70.49 (CH2), 107.98 (CH),
114.76 (C), 115.85 (C), 119.37 (C), 122.93 (CH), 127.22 (C),
128.69 (CH), 129.55 (CH), 130.54 (C), 135.09 (C), 135.23 (CH),
140.75 (C), 143.60 (CH), 146.30 (C), 160.01 (C). MS m/z (%):
352 ([M + 1]+, 9), 351 (M+, 40), 278 (5), 266 (7), 250 (7), 237
(21), 181 (14), 152 (45), 58 (100). Anal. C21H22ClNO4: C, H,
Cl, N.
P h otobiologica l Meth od s. Cell Cu ltu r es. HL-60 and
HeLa cells were grown in RPMI 1640 (Sigma Chemical Co.)
supplemented with 15% heat-inactivated fetal calf serum
(Seromed) and Nutrient Mixture F-12 [HAM] (Sigma Chemical
Co.) supplemented with 10% heat-inactivated fetal calf serum
(Seromed), respectively. Penicillin (100 U/mL), streptomycin
(100 µg/mL), and amphotericin B (0.25 µg/mL) (Sigma Chemi-
cal Co.) were added to both media. The cells were cultured at
37 °C in a moist atmosphere of 5% carbon dioxide in air.
3,4-Cycloh exen e-8-(3-d im eth yla m in op r op oxy)-4′-m eth -
ylfu r o[3,2-g]cou m a r in (6). A mixture of hydroxytetrahy-
drobenzofurocoumarin 5 (229.0 mg, 0.847 mmol), 3-chloro-N,N-
dimethylpropylamine hydrochloride (160 mg, 1.016 mmol),
60% NaH (61 mg, 2.541 mmol), NaI (152 mg, 1.016 mmol),
and dimethylformamide (25 mL) was heated for 4 h at 100 °C
and then concentrated under reduced pressure. The residue
was purified by FC with 98:2 CH2Cl2-methanol as eluent and

Pyrone Side Tetracyclic Psoralen Derivatives
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18 3809
Ir r a d ia tion P r oced u r e. Irradiations were performed by
means of Philips HPW 125 lamps equipped with a Philips filter
emitting over 90% at 365 nm. Irradiation intensity was
checked on a UV-X radiometer (Ultraviolet Products Inc.,
Cambridge, UK) for each experimental procedure.
9:0.8:0.2. UV spectra were recorded on a Perkin-Elmer model
Lambda 5 spectrophotometer. The 1H and 13C assignment was
obtained by utilizing HMQC and HMBC spectra on a Bruker
Avance DMX600 instrument. Mass spectrometry measure-
ments were performed on a ElectroSpray Ionization (ESI) time-
of-flight (ToF) instrument (model Mariner, Perseptive-Biosys-
tem) by dissolving the samples in water/acetonitrile/formic acid
(50:49:1) solution.
P h otor ever sa l of Ad d u cts. An ethanol solution of the
adduct (ca. 20 µg/mL) was irradiated in quartz cuvettes with
a mineral lamp (254 nm). The photosplitting reaction was
followed spectrophotometrically.
Eva lu a tion of In ter str a n d Cr oss-Lin k s in Vitr o. Evalu-
ation of cross-links was carried out by measuring the rena-
turation capacity of cross-linked DNA after thermal denatur-
ation. Aliquots of a solution of DNA and examined compound
[DNA]/[drug] ) 75 were introduced into calibrated glass tubes,
immersed in a thermostatically controlled bath, and then
irradiated for various periods of time. After irradiation the
samples were thermally denatured (95 °C for 15 min) and
quickly cooled in ice. The renaturation capacity of DNA, due
to cross-link formation, was investigated by recording absor-
bance at 260 nm. Data were expressed in terms of log % of
nonrenaturated fraction of irradiated compound-DNA com-
plex relative to irradiated DNA, as suggested by Blais et al.25
In h ibition Gr ow th Assa ys. HeLa cells (105) were seeded
into each well of a 24-well cell culture plate. After incubation
for 24 h, the medium was replaced with an equal volume of
Dulbecco’s modified Eagle medium (DMEM, Sigma Chemical
Co.) without phenol red, and various concentrations of the test
agent were added. One hour later the cells were irradiated
with a UVA dose of 0.793 J cm-2. After irradiation, the medium
containing the compounds was removed, and the cells were
incubated in complete F-12 medium for 24 h. In the case of
the experiments carried out in the dark, the replacement of
the incubation medium was omitted and the cells were
incubated in complete F-12 medium in the presence of the test
compound for 24 h.
HL-60 cells (105) were seeded into each well of a 24-well
cell culture plate. After incubation for 24 h, various concentra-
tions of the test agents were added in complete medium. The
cells were kept in the dark for 1 h, irradiated with a UVA dose
of 0.793 J cm-2, and then incubated for a further 24 h.
A trypan blue assay was performed to determine cell
viability. Cytotoxicity data were expressed as IC50 values, i.e.,
the concentrations of the test agent inducing 50% reduction
in cell numbers compared with control cultures.
Sk in P h ototoxicity. Skin phototoxicity was tested on
depilated albino guinea pigs (outbred Dunkin-Hartley strain),
as previously reported.20 An ethanol solution of each new
compound was applied topically to the skin up to 50 µg/cm2.
For 8-MOP, the concentration used was 10 µg/cm2. The
animals were then kept in the dark for 45 min and the treated
skin was irradiated with 20 kJ m-2 of UVA; erythema was
scored after 48 h.
Th eor etica l Ca lcu la tion s. Theoretical optimization of the
molecular geometries, followed by manual docking into a DNA
binding site model was calculated at the AM1 semiempirical
quantum mechanical level,26 using the AM1 module of the PC
Spartan Pro program27 run on an SGI 320 workstation.
Ack n ow led gm en t. We express our thanks to Prof.
Stefano Mammi for NMR analysis and for the valuable
discussion.
Nu cleic Acid . Salmon testes DNA was purchased from
Sigma Chemical Co. (Cat. D-1626). Its hypochromicity, deter-
mined according to the method of Marmur and Doty,23 was
over 35%. The DNA concentration was determined using
extinction coefficient 6600 M-1cm-1 at 260 nm.
Lin ea r F low Dich r oism . Linear dichroism (LD) measure-
ments were performed on a J asco J 500A circular dichroism
spectropolarimeter converted for LD and equipped with an
IBM PC and a J asco J interface.
Refer en ces
(1) Dalla Via, L.; Marciani Magno, S. Photochemotherapy in the
treatment of cancer. Curr. Med. Chem. 2001, 8, 1405-1418.
(2) Parrish, J . A.; Fitzpatrick, T. B.; Tannenbaum, L.; Pathak, M.
A. Photochemotherapy of psoriasis with oral methoxsalen and
longwave ultraviolet light. N. Engl. J . Med. 1974, 291, 1207-
1211.
(3) Lowe, N. J .; Chizhevsky, V.; Gabriel, H. Photo(chemo)therapy:
General principles. Clin. Dermatol. 1997, 15, 745-752.
(4) Bethea, D.; Fullmer, B.; Syed, S.; Seltzer, G.; Tiano, J .; Rischko,
C.; Gillespie, L.; Brown, D.; Gasparro, F. P. Psoralen photo-
biology and photochemotherapy: 50 years of science and medi-
cine. J . Dermatol. Sci. 1999, 19, 78-88.
Linear dichroism is defined as
LD(λ) ) A|(λ) - A
(λ)
(5) Edelson, R. L.; Berger, C. L.; Gasparro, F. P.; J egasothy, B.;
Heald, P.; Wintroub, B.; Vonderheid, E.; Knobler, R.; Wolff, K.;
Plewig, G.; et al. Treatment of cutaneous T-cell lymphoma by
extracorporeal photochemotherapy. N. Engl. J . Med. 1987, 316,
297-303.
where A| and A correspond to the absorbances of the sample
when polarized light is oriented parallel or perpendicular to
the flow direction, respectively. The orientation is produced
by a device designed by Wada and Kozawa24 at a shear
gradient of 500-700 rpm and each spectrum was accumulated
four times.
(6) Van Iperen, H. P.; Beijersbergen van Henegouwen G. M. J .
Clinical and mechanistic aspects of photopheresis. J . Photochem.
Photobiol B: Biol. 1997, 39, 99-109.
(7) McNeely, W.; Goa, K. L. 5-Methoxypsoralen. A review of its
effects in psoriasis and vitiligo. Drugs 1998, 56, 667-690.
(8) Song, P.-S.; Tapley, K. J ., J r. Photochemistry and photobiology
of psoralens. Photochem. Photobiol. 1979, 29, 1177-1197.
(9) Rodighiero, G.; Dall’Acqua, F.; Averbeck, D. New psoralen and
angelicin derivatives. In Psoralen DNA photobiology; Gasparro,
F. P., Ed.; CRC Press: Boca Raton, FL, 1988; pp 37-114.
(10) Palumbo, M.; Rodighiero, P.; Gia, O.; Guiotto, A.; Marciani
Magno, S. Benzofurocoumarins: new monofunctional DNA-
photobinding agents. Photochem. Photobiol. 1986, 44, 1-4.
(11) Guiotto, A.; Chilin, A.; Manzini, P.; Dall’Acqua, F.; Bordin, F.;
Rodighiero, P. Synthesis and antiproliferative activity of furo-
coumarin isosters. Farmaco 1995, 50, 479-488.
(12) Gia, O.; Mobilio, S.; Palumbo, M.; Pathak, M. A. Benzo- and
tetrahydrobenzo-psoralen congeners: DNA binding and photo-
biological properties. Photochem. Photobiol. 1993, 57, 497-503.
(13) Gia, O.; Conconi, M. T.; Uriarte, E. Benzo- and tetrahydrobenzo-
furocoumarins: New potential photochemotherapeutic drugs.
Farmaco 1997, 5, 289-293.
A solution of salmon testes DNA (1.5 × 10-3M) in ETN
buffer (containing 10 mM TRIS, 10 mM NaCl, and 1 mM
EDTA, pH ) 7) was used. Spectra were recorded at 25 °C at
different [DNA]/[drug] ratios.
P r ep a r a tion of Ad d u cts. Volumes of concentrated solu-
tions of the examined compound were added to salmon testes
DNA in ETN solution (1.5 × 10-3M) to achieve a DNA/
compound ratio of about 80. The mixture was irradiated in a
glass dish with four Philips HPW 125 lamps, arranged two
above and two below the dish, at a distance of 7 cm, for 120
min at room temperature. After irradiation the DNA was
precipitated with NaCl (up to 1 M concentration) and cool
ethanol (2 volumes), and the precipitated DNA was collected,
washed with 80% ethanol, dried, and then dissolved in a
measured volume of buffer. The final solution was made 0.5
N with HCl, heated at 100 °C for 2 h, neutralized, and
extracted exhaustively with CHCl3. After this procedure the
organic layers were collected, dried under high vacuum, and
dissolved in ethanol, and the adduct was separated on TLC
plates and eluted with CHCl3/ethanol/ammonium hydroxide
(14) Dalla Via, L.; Gia, O.; Marciani Magno, S.; Santana, L.; Teijeira
M.; Uriarte, E. New tetracyclic analogues of photochemothera-
peutic drugs 5-MOP and 8-MOP: Synthesis, DNA interaction,
and antiproliferative activity. J . Med. Chem. 1999, 42, 4405-
4413.

3810 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 18
Dalla Via et al.
(15) Parekh, M. G.; Trivedi, K. N. Synthesis of cyclohexenocoumarins.
III. Reactions of 3-hydroxy-6-oxo-7,8,9,10-tetrahydro-6H-diben-
zo[b,d]pyran. Aust. J . Chem. 1970, 23, 407-412.
monoaddition products from 8-methoxypsoralen and 4,5′,8-
trimethylpsoralen. Biochemistry 1982, 21, 861-871.
(22) Gia, O.; Anselmo, A.; Pozzan, A.; Antonello, C.; Marciani Magno,
S.; Uriarte, E. Some new methyl-8-methoxypsoralens: Synthesis,
photobinding to DNA, photobiological properties and molecular
modelling. Farmaco 1997, 52, 389-397.
(23) Marmur, J .; Doty, P. Determination of the base composition of
the deoxyribonucleic acid from its thermal denaturation tem-
perature. J . Mol. Biol. 1962, 5, 109.
(16) Confalone, P. N.; Confalone, D. L. The design and synthesis of
monofunctional psoralens structurally related to methoxsalen
and trioxsalen. Tetrahedron 1983, 39, 1265-1271.
(17) Soman, S. S.; Trivedi, K. N. Synthesis of furocoumarins. Part
33. Synthesis of 6/7-methoxy-8/9-methylcycloalkeno[c]psoralens.
Indian J . Chem. B 1992, 31B, 30-34.
(18) Tjerneld, F.; Norden, B.; Ljunggren B. Interaction between DNA
and 8-methoxypsoralen studied by linear dichroism. Photochem.
Photobiol. 1979, 29, 1115-1118.
(19) Caffieri, S.; Rodighiero, P.; Vedaldi, D.; Dall’Acqua, F. Methy-
lallopsoralen-thymine 3,4- and 4′,5′-monoadducts formed in the
photoreaction with DNA. Photochem. Photobiol. 1985, 42, 361-
366.
(20) Gia, O.; Anselmo, A.; Conconi, M. T.; Antonello, C.; Uriarte, E.;
Caffieri, S. 4′-methyl derivatives of 5-MOP and 5-MOA: Syn-
thesis, photoreactivity, and photobiological activity. J . Med.
Chem. 1996, 39, 4489-4496.
(21) Kanne, D.; Straub, K.; Rapoport, H.; Hearst, J . E. Psoralen-
deoxyribonucleic acid photoreaction. Characterization of the
(24) Wada, A.; Kozawa, S. Instrument for the studies of differential
flow dichroism of polymer solutions. J . Polym. Sci., Part A 1964,
2, 853-864.
(25) Blais, J .; Vigny, P.; Moron, J .; Bisagni, E. Spectroscopic proper-
ties and photoreactivity with DNA of new monofunctional
pyridopsoralen. Photochem. Photobiol. 1984, 39, 145-156.
(26) Dewar, M. J . S.; Zoebish, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: a new general purpose quantum mechanical molecular
model. J . Am. Chem. Soc. 1985, 107, 3902-3909.
(27) PC SPARTAN Pro. Version 1.0.5. Wavefunction, Inc., Irvine, CA,
2000.
J M0210919