Synthesis and photochemotherapeutic activity of thiopyrano[2,3-e]indol-2- ones

Article abstract of DOI:10.1016/j.bmcl.2005.03.016

A series of derivatives of the new ring system thiopyrano[2,3-e]indol-2-one was prepared with the aim of obtaining new photochemotherapeutic drugs. Biological screenings were performed on this new class of photoactivable drugs and a strong antiproliferative effect was observed upon irradiation with UVA light. The compound bearing a methyl substituent at the pyrrole nitrogen resulted as the most interesting showing IC₅₀ in the nanomolar range.

Full text of DOI:10.1016/j.bmcl.2005.03.016

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2291–2294
Synthesis and photochemotherapeutic activity of
thiopyrano[2,3-e]indol-2-ones
Paola Barraja,a Laura Sciabica,a Patrizia Diana,a Antonino Lauria,a
Alessandra Montalbano,a Anna Maria Almerico,a Gaetano Dattolo,a
b
b
Girolamo Cirrincione,a, Silvia Disaro, Giuseppe Basso,
*
`
Giampietro Violac and Francesco DallÕAcquac
a
`
Dipartimento Farmacochimico, Tossicologico e Biologico, Universita degli Studi di Palermo Via Archirafi 32, 90123 Palermo, Italy
b
`
Dipartimento di Pediatria, Universita degli Studi di Padova, Via Giustiniani 3, 35131 Padova, Italy
`
Dipartimento di Scienze Farmaceutiche, Universita degli Studi di Padova, Via Marzolo 5, 35131 Padova, Italy
c
Received 28 October 2004; revised 12 January 2005; accepted 3 March 2005
Abstract—A series of derivatives of the new ring system thiopyrano[2,3-e]indol-2-one was prepared with the aim of obtaining new
photochemotherapeutic drugs. Biological screenings were performed on this new class of photoactivable drugs and a strong anti-
proliferative effect was observed upon irradiation with UVA light. The compound bearing a methyl substituent at the pyrrole nitro-
gen resulted as the most interesting showing IC50 in the nanomolar range.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Linear furocoumarins, such as psoralen 1, belong to an
important class of photoactivable drugs which are cur-
rently used in PUVA (psoralen + UVA) therapy for
the treatment of various skin diseases including psoria-
sis, vitiligo and tumours such as cutaneous T-cell lym-
phoma (CTCL).1 This is due to their capacity to
intercalate into DNA and photobind with it yielding
either monoadducts or interstrand cross-links by one
or two subsequent photocycloadditions.2–4 However
several short term (erythema) and long term (mutations
and skin cancer) side effects are associated with this ther-
apy. On the contrary angelicin 2, an angular psoralen,
proved to form only monoadducts with DNA. In fact
its geometry avoids the simultaneous alignment of the
two reactive sites.5 We have recently reported the syn-
thesis and biological activity of the new ring system pyr-
rolo[2,3-h]quinolin-2-one 3, an angelicin hetero-
analogue, in which nitrogen atoms replace both oxygens
on the furan and the pyrone ring.6
O
O
R1
H
N
O
Me
O
O
O
N
R
O
1
2
3
All the tested compounds were phototoxic on HT-1080
(human fibrosarcoma) and LoVo (human intestinal ade-
nocarcinoma) cell lines with IC50 in the micromolar or
submicromolar range (0.4–16 lM) with remarkable
UVA dose-dependence. Interestingly, some compounds
of this series show even higher cytotoxicity than 8-meth-
oxypsoralen (8-MOP), 5-methoxypsoralen (5-MOP) and
angelicin used as reference drugs. However, studies of
linear dichroism (LD) strongly suggest that the new
derivatives do not interact efficaciously with DNA, thus
indicating a different mechanism in respect to that of
furocoumarins.
It has been reported that the introduction of a sulfur on
the pyrone moiety of psoralen improves the interaction
with DNA both in the dark and under UVA light.7 In
the search for new angelicin analogues with better anti-
proliferative activity and lower toxicity, and considering
Keywords: Thiopyrano-indoles; Photochemotherapeutic activity;
Apoptosis induction.
*
Corresponding author. Fax: +39 091 6169999; e-mail: gcirrinc@
unipa.it
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.03.016

2292
P. Barraja et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2291–2294
O
C
Cl
O
C
SEt
O
O
C
SEt
H
H
H
O
ii
i
iii
Me
N
Me
N
R
Me
N
R
N
R
Me
R
O
4a-c
S
5a-c
6a-c
7a-c
iv
iv
O
X
HO
SEt
X
EtO
SEt
vi
v
X
X
X
N
R
Me
N
Me
X
N
Me
R
R
12a,b X-X= CH=CH
13a-c X-X= CH2-CH2
10a,b X-X= CH=CH
11a-c X-X= CH2-CH2
8a,b X-X= CH=CH
9a-c X-X= CH2-CH2
a R= Ph b R= Me c R=Bn
Scheme 1. Synthesis of compounds 12a,b (X–X = CH@CH2) and 13a–c (X–X = CH2–CH2). Reagents and conditions: (i) DMF/POCl3 in
dichloromethane, 0 ꢁC then reflux; (ii) HSEt/K2CO3 in DMF, rt; (iii) DDQ in benzene, reflux; (iv) (EtO)2P(O)CH2CO2Et/t-BuOK in
dichloromethane, rt; (v) KOH, ethanol 50%, reflux; (vi) PPA, D.
our previous satisfactory results on pyrroloquinoli-
nones, we decided to investigate the synthesis of the
new ring system thiopyrano[2,3-e]indol-2-ones and to
study the biological activity of its derivatives. The syn-
thetic approach to such compounds outlined in Scheme
1, started from ketones 4a–c prepared by known proce-
dures, by reaction of freshly prepared 2-acetonyl-1,3-
cyclohexanedione with the proper amines in acetic acid.6
Chloroformylation was performed in dichloromethane
with an excess of the Vilsmeier–Haack reagent (DMF/
POCl3). Strict temperature control (0 ꢁC), during the
addition of the formylating mixture, and a short reflux-
ing time (5 min), are crucial to achieve the best yields
(70–75%). The chloroformylated derivatives 5a–c are
unstable at room temperature and require storage at
À20 ꢁC. Nucleophilic substitution of the chlorine atom
by the ethanethiolate anion yields the thioethers 6a–c
(94–96%).
indol-2-ones 12a,b and 13a–c (30–40%).8 The phototox-
icity of the title compounds was investigated on two
cultured cell lines: HL-60 and LoVo. Table 1 shows
the extent of cell survival expressed as IC50 which is
the concentration, expressed in micromolars, that in-
duces 50% inhibition of cell growth, after irradiation
at different UVA doses.9 Control experiments with
UVA light or drugs alone were carried out without sig-
nificant cytotoxic effects, except that, it was not possible
to use the highest UVA dose (6.5 J cmÀ2) for HL-60 cells
because it causes in this cell line a significant cytotoxicity
(data not shown). It can be noted that the compounds
exhibit different values of IC50 depending on the substi-
tution pattern, and a remarkable UVA dose-depen-
dence. In particular compound 12b showed the highest
cytotoxicity, especially in the solid tumour when com-
pared to 8-MOP, and angelicin used as reference com-
pounds. It is interesting to note that the most active
compound bears a methyl group on the pyrrole nitro-
gen, whereas a phenyl or a benzyl group strongly re-
duces the activity.
Oxidation of the dihydro derivatives with 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone (DDQ) effected conver-
sion of 6a–c into the stable aldehydes 7a–c (80–84%).
Transformation into the vinylogous esters 8a,b and
9a–c was achieved by reaction with the appropriate Wit-
tig–Horner reagent either on the dihydro aldehydes (70–
75%) or the aromatic ones (72–76%). Hydrolysis to the
corresponding acids 10a,b (86–88%) and 11a–c (82–
85%) and cyclization with PPA yielded thiopyrano[2,3-e]-
In parallel to the cytotoxic evaluation, we used flow
cytometry to study cell cycle variations upon irradia-
tion.11 In Figure 1, the effect in the leukaemic cell line,
of the most active compound 12b is shown after 24, 48
and 72 h from the irradiation. The results indicate that
treatment with 12b in combination with UVA induces
Table 1. IC50 (lM) values obtained for the test compounds at different UVA doses in two human tumour cell linesa
Cytotoxicity (IC50, lM)a
Compounds
HL-60b
LoVo
2.5c
3.2
2.5
3.2
6.5
2.9 1.1
12a
13a
6.2 0.6
2.6 0.3
0.6 0.06
4.8 0.9
1.4 0.2
1.2 0.1
2.7 0.3
1.3 0.2
0.3 0.1
2.7 0.3
1.2 0.4
0.9 0.2
>20
12.3 1.7
4.2 1.4
0.8 0.2
7.2 1.2
0.7 0.1
0.9 0.1
5.1 1.2
1.0 0.1
>20
3.9 0.4
0.07 0.02
1.8 0.6
0.4 0.1
0.8 0.1
12b
13c
8-MOP
Angelicin
1.1 0.4
1.6 0.2
a Values are means SEM of three independent experiments.
b HL-60, human promyelocytic leukaemia; LoVo intestinal human adenocarcinoma.
c UVA dose expressed in joule per centimetre square as measured at 365 nm by a Cole Parmer radiometer.

P. Barraja et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2291–2294
2293
Figure 1. DNA flow cytometric analysis. HL-60 cells were irradiated with compound 12b at concentration of 5 lM and a UVA dose of 2.5 J cmÀ2
The cells were harvested for designated times as shown in the figure and described in Ref. 11. The data shown here are from a representative
experiment repeated two times with similar results. Upper panels: irradiated controls; lower panels 12b treated cells.
.
a remarkable reduction of the S phase at 24 h from irra-
diation. This is followed by massive induction of apop-
tosis, as observed by the appearance of a sub G1 peak
(apoptotic peak) that refers to cells with DNA content
lower than G1. In fact, apoptosis induces the activation
of endogenous nucleases which are responsible for nu-
cleic acid degradation.12
Acknowledgments
This work was financed by MIUR (Ministero dellÕIst-
`
ruzione, dellÕUniversita e della Ricerca). G.B. thanks
`
also for financial support the Fondazione ÔCitta della
speranzaÕ and Progetto finalizzato MIUR-CNR:
ÔCaratterizzazione biologica e strategie terapeutiche
`
innovative nelle malattie sistemiche dellÕadulto e nellÕeta
pediatricaÕ.
Considering that DNA is the major target of psoralen,
we have investigated the binding between the drugs
and the macromolecule. UV–vis and fluorimetric titra-
tion with DNA have shown neither variation of the
absorption spectra nor the presence of isosbestic points
indicative of the formation of a complex between the
test compounds and the DNA. Furthermore, a poor
fluorescence quenching of the emission of the free drug
in the presence of DNA, suggests that the title com-
pounds are loosely bound to DNA. These data are also
confirmed by LD measurements on DNA–drug com-
plexes.13,14 The obtained results (data not shown) indi-
References and notes
1. DallÕAcqua, F.; Viola, G.; Vedaldi, D. Molecular Basis of
Psoralen Photochemotherapy Chapter 142. In CRC
Handbook of Organic Photochemistry and Photobiology;
Hoorspool, W. M., Lenci, F., Eds.; CRC: Boca Raton
USA, 2004; p 1.
2. Tessman, J. W.; Isaacs, S. T.; Hearst, J. E. Annu. Rev.
Biochem. 1985, 54, 1151.
3. Musajo, L.; Rodighiero, G. In Photophysiology; Giese, A.
C., Ed.; Accademic: New York, 1972; Vol. 7, p 115.
4. DallÕAcqua, F.; Viola, G. In Photobiology for the 21st
Century; Coohil, T. P., Valenzeno, D. P., Eds.; Valdenmar
Publishing company: Overland park Kansas, 2001; p 325.
5. Bordin, F.; DallÕAcqua, F.; Guiotto, A. Pharmacol. Ther.
1991, 52, 331.
cate
a complete absence of any signal in the
chromophore absorption region of the thiopyrano deriva-
tives. This fact strongly suggests that the new derivatives
do not significantly interact with the macromolecule as
already observed for other series of angelicin
heteroanalogues.6
6. Barraja, P.; Diana, P.; Lauria, A.; Montalbano, A.;
Almerico, A. M.; Dattolo, G.; Cirrincione, G.; Viola,
G.; DallÕAcqua, F. Bioorg. Med. Chem. Lett. 2003, 13,
2809, and references herein cited.
According to these preliminary results, we can conclude
that the new thiopyrano derivatives seem to be interest-
ing as potential drugs for PUVA photochemotherapy.
7. (a) Jacobs, A. E.; Piette, J. J. Photochem. Photobiol. B
1994, 22, 9; (b) Jacobs, A. E.; Piette, J. J. Med. Chem.
1995, 38, 869.
However, further studies are necessary to identify the
biological targets involved in their mechanism of action
as well as the downstream events involved in the activa-
tion of the apoptosis. Identification of these specific tar-
gets may have significant implications for the
therapeutic application of the new derivatives.
1
8. All final products were characterized by IR, H and 13C
NMR and elemental analysis. A representative example,
7,8-dimethyl-6,7-dihydrothiopyrano[2,3-e]indol-2(5H)-
one 13b: 40% mp 172–173 ꢁC; IR 1620 (CO) cmÀ1
;
1H NMR (DMSO-d6): 2.22 (3H, s, CH3), 2.77–2.79. (4H,

2294
P. Barraja et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2291–2294
m, H-5 and H-6), 3.43 (3H, s, CH3), 5.99 (1H, s, H-9), 6.16
30 min in an atmosphere of 5% CO2 at 37 ꢁC, the control
plate was placed in the dark and then irradiated with two
HPW 125 Philips lamps, principally emitting at 365 nm.
After irradiation, the solution was replaced by the
complete medium by centrifugation and the plates were
incubated for 72 h. Cell viability was assayed by the MTT
(1H, d, J = 10.2 Hz, H-3), 7.22 (1H, d, J = 10.2 Hz, H-4);
13C NMR (DMSO-d6): 12.1 (q), 20.8 (t), 29.4 (t), 30.3 (q),
101.5 (d), 108.2 (s), 116.5 (s), 117.4 (d), 120.2 (s), 130.8 (s),
131.7 (s)144.6 (d), 184.4 (s, CO). Anal. Calcd for
C13H13NOS: C, 67.50; H, 5.66; N, 6.06. Found: C,
67.80; H, 5.69; N, 6.40.
[(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazolium
9. Exponentially growing HL-60 leukaemia cells were resus-
pended at a density of 5 · 104 cells/mL in a complete
medium (RPMI containing 10% foetal bovine serum,
100 UI/mL penicillin G and 100 lg/mL streptomycin) and
seeded in 96 well culture plates. LoVo, intestinal human
adenocarcinoma cells were resuspended at a density of
5 · 104 cells/mL in a complete medium (HamÕF12 con-
taining 10% foetal bovine serum, 100 UI/mL penicillin G
and 100 lg/mL streptomycin) and seeded in a 96 well
culture plates which were allowed to adhere for 18 h to
culture plates before addition of the drugs. After the
medium was removed, by centrifugation, 100 lL of the
drug solution, dissolved in DMSO and diluted with
HankÕs balanced salt solution (HBSS pH = 7.2), was
added to each well. The plate was then incubated for
bromide)] method.10 Cell growth at each drug concentra-
tion was expressed as percentage of untreated controls and
the concentration resulting in 50% (IC50) growth inhibi-
tion was determined by linear regression analysis.
10. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
11. For cell cycle experiments 5 · 105 HL-60 cells were
treated, as described for phototoxicity experiments, fixed
in ethanol 70% after 24, 48 and 72 h, respectively, from the
treatment and stained with propidium iodide. Samples
were analyzed on flow cytometer (Epics XL-MCL, Coul-
ter), using the Multicycleꢂ software (Phoenix, CA).
12. Compton, M. M. Cancer Metast. Rev. 1992, 11, 105.
´
13. Norden, B.; Kubista, M.; Kurucsev, T. Q. Rev. Biophys.
1992, 25, 51.
´
14. Norden, B.; Kurucsev, T. J. Mol. Recognit. 1994, 7, 141.