Photodynamic effects of porphyrin and chlorin photosensitizers in human colon adenocarcinoma cells

Article abstract of DOI:10.1016/j.bmc.2004.07.011

A series of tetraaryl porphyrins and chlorins were synthesized and tested for photodynamic activity in human colon adenocarcinoma cells. Photodynamic therapy (PDT) is a cancer treatment involving systemic administration of a tumor-localizing photosensitiz

Full text of DOI:10.1016/j.bmc.2004.07.011

Bioorganic & Medicinal Chemistry 12 (2004) 4853–4860
Photodynamic effects of porphyrin and chlorin photosensitizers
in human colon adenocarcinoma cells
Stefano Banfi,a,* Enrico Caruso,a Stefania Caprioli,a Lugi Mazzagatti,a Gianfranco Canti,b
Raffaella Ravizza,a Marzia Gariboldia and Elena Montia
aDBSF, University of Insubria, Via A. da Giussano 10, 21052 Busto Arsizio (VA), Varese, Italy
bDepartment of Pharmacology, University of Milan, Milan, Italy
Received 7 April 2004; revised 1 July 2004; accepted 6 July 2004
Available online 3 August 2004
Abstract—Photodynamic therapy (PDT) is a cancer treatment involving systemic administration of a tumor-localizing photosensi-
tizer; this, when activated by the appropriate wavelength of light, interacts with molecular oxygen to form a toxic, short-lived species
known as singlet oxygen, which is thought to mediate cellular death. Photofrinꢀ, a complex mixture of porphyrin oligomers has
recently received FDAapproval for the photodynamic treatment of esophageal and endobronchial carcinoma, but its photodynamic
and toxicity profiles are far from ideal. In the present study we evaluated a series of porphyrin-based PSs, some of which newly
synthesized by our group, with the aim to identify agents with more favorable characteristics. For the most effective compounds
in the porphyrin series, chlorin analogs were also synthesized; for comparison, the screening also included Photofrinꢀ. Cytotoxicity
studies were performed by the MTT assay on a cultured human colon adenocarcinoma cell line (HCT116); the results indicate that
the 3,4,5-trimethoxyphenyl, 3OH- and 4OH-phenyl, and the sulfonamidophenyl derivatives are significantly more potent than Pho-
tofrinꢀ. Flow cytometric studies and fluorescence microscopy indicate that in PDT-treated HCT116 cells death occurs mainly by
apoptosis.
In summary, novel PSs described in the present study, belonging both to the porphyrin and chlorin series, have proven more
effective than Photofrinꢀ in killing colon cancer cells in vitro; extending these observation to in vivo models, particularly regarding
the deeper reaching chlorin derivatives, might lead to significant advances in the development of tumor PDT.
ꢁ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
damage concur to the therapeutic effects of PSs;2 photo-
damage at specific subcellular sites, most notably mito-
chondria, ultimately leads to cell death by apoptosis.3
Photodynamic therapy (PDT) provides a viable thera-
peutic option for the treatment of a number of solid
malignancies, as well as nonmalignant diseases. Sys-
temic administration of a photosensitizing agent (PS),
exogenously administrated or endogenously generated,
is followed by irradiation with visible light of appropri-
ate wavelength (i.e., compatible with absorption spec-
trum of the PS) and dose. Upon energy absorption,
the PS is driven into an excited triplet state, which inter-
acts with ground state molecular oxygen to produce sin-
glet oxygen, a highly cytotoxic species with short lifetime
(<0.04ls) and radius of action (<0.02lm).1 It appears
that both direct cytotoxic activity and microvascular
The application of PDT to cancer treatment is particu-
larly attractive, as it holds the potential for minimal side
toxicities, due both to the preferential distribution and
longer retention times of PSs in tumors as compared
with normal tissues, and to the fact that light irradiation
is limited to the tumor region. Furthermore, the use of
PDT is not precluded by prior radiotherapy, chemother-
apy or surgery.4
Porfimer sodium (Photofrinꢀ), a complex of porphyrin
oligomers, was the first PS to win approval by regula-
tory agencies in several countries and is now licensed
for treatment of cancers in the esophagus, lung, stom-
ach, cervix, and bladder. However, a number of prob-
lems related with the use of Photofrinꢀ, such as
extended skin photosensitivity and poor absorption of
tissue-penetrating red light, have led to the development
Keywords: PDT; HCT116; Tetraaryl porphyrins; Chlorins; Photofrinꢀ.
*
Corresponding author. Address: DBSF, University of Insubria, Via
JH Dunant, 3, 21100 Varese, Italy. Tel.: +39-0332-421550; fax: +39-
0332-421554; e-mail: stefano.banfi@uninsubria.it
0968-0896/$ - see front matter ꢁ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.07.011

4854
S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
Table 1. IC50 values ( SE) obtained from 4–6 independent experi-
ments. Statistically significant differences were assessed by the analysis
of variance, followed by Dunnetꢁs test, comparing PSs 1–12 versus
Photofrinꢀ
of novel photosensitizers with more favorable character-
istics. These so-called second-generation PSs have
shorter periods of photosensitization, longer activation
wavelengths (and, therefore, are activated deeper within
tissues), higher yields of singlet oxygen, and tumor selec-
tivity. Alarge number of polyunsaturated compounds
with different structures have been actively investigated,
including chlorins, texaphyrins, purpurins, and phthalo-
cyanines; however, only one new agent, the m-tetra(hy-
PS
IC50 (ng/mL) SE
73.67 8.04
IC50 (nM) SE
n.d
Photofrinꢀ
1
>10,000*
4428 125*
10.74 1.82*
23.20 3.46*
5.13 0.98*
3.07 0.37*
221.81 36.42*
24.8 1.28*
28.69 4.12*
6.51 1.32*
3358.00 96*
17.74 1.26*
>13,608
6026 170
11 1.86
2
3
droxyphenyl)chlorin
(m-THPC
or
temoporfin,
marketed as Foscanꢀ), has recently been approved by
the European Medical Agency (EMEA) for use in head
and neck squamous cell carcinomas.
4
18 2.68
7.5 1.4
5
6
4.52 0.6
274 44.99
36.4 1.88
42 6.05
7
8
The present study is a preliminary screen of the photo-
dynamic activity of a small number of compounds ob-
tained by modification of the tetraaryl porphyrin/
chlorin ring (Fig. 1). The choice of the core structure
was based on a number of considerations: (i) temopor-
fin, one of the most active PSs currently in clinical trials,
is a tetraarylchlorin derivative; (ii) a variety of substitu-
ents, featuring different polarity and length, can be
introduced on the meso-phenyls, either by using different
substituted aromatic aldehydes for the synthesis or by
chemical modification of the tetraphenyl-porphyrin
skeleton, and this could yield novel PS with a better pro-
9
10
11
12
6.66 1.35
4178 120
13.8 0.98
* p < 0.05 versus Photofrinꢀ.
file than the agents in current clinical use. Asecondary
goal in this study was to establish whether the presence
of a tetraaryl porphyrin or tetraaryl chlorin ring signif-
icantly affects the photodynamic activity and cell pene-
tration characteristics of these compounds. In vivo
applications of porphyrin derivatives are hampered by
their spectroscopic characteristics, which do not feature
absorption in the red region (they exhibit an intense So-
ret absorption band around 420nm and a lower band at
510nm) and, therefore, are only activated by wave-
lengths with limited tissue penetration; however, the de-
sign and synthesis of tetraaryl porphyrins, as well as
their in vitro use in tumor cells, are far less cumbersome
as compared to the corresponding tetraaryl chlorin
derivatives, and this makes them eminently suitable for
in vitro screening of large panels of compounds. The
most active compounds in the screen could then be con-
verted to the corresponding chlorins (which exhibit an
absorption band at 650nm), in view of subsequent in
vivo applications.
5
3
5
3
4
4
4
4
3
5
3
5
5
3
5
3
N
N
H
NH
HN
NH
HN
N
H
N
Porphyrins
Chlorins
4
4
4
4
3
5
3
5
3
4
5
Porphyrins
Twelve tetraaryl derivatives, including five porphyrin/
chlorin pairs, were tested for photodynamic cytotoxicity
on a human colon adenocarcinoma cell line (HCT116)
and the results were compared with those obtained with
Photofrinꢀ. The series included a number of derivatives
newly synthesized by our group and/or for the first time
tested (Table 1, 3,4, 7, 10, 11, 12), as well as temoporfin
(8) and its tetraaryl porphyrin analog (5).
H
H
H
1
OCH3
H
OCH3
OCH3
2
3
4
OCH3
OCH3
SO2AEE *
H
OH
H
H
5
6
7
H
H
H
OH
OH
OH
OH
5
3
4
2. Results and discussion
Chlorins
9
H
OH
H
The ꢀmethoxyꢁ porphyrins 1–3 were synthesized by con-
densation of aromatic aldehyde and pyrrole, following
the general method described by Lindsey and co-work-
ers, in 41%, 22%, and 33% yields, respectively, after col-
umn chromatography.5 Compound 4 was isolated as
previously reported,6 while the ꢀhydroxyꢁ porphyrins 5–
7, were obtained in good yields from the parent methoxy
through reaction with BBr3 in CH2Cl2.
10
OCH3
OCH3
OCH3
11
12
OH
OH
H
OH
H
SO2AEE *
* SO2AEE = SO2NHCH2CH2OCH2CH2OH
Figure 1. Photosensitizer structures.

S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
4855
The choice of the aryl substituents of the porphyrin
rings was dictated by the necessity to modulate the
hydrophilic/lipophilic character of the molecules, the
starting point being tetraphenylporphyrin (H2-TPP),
which is the most lipophilic and is known as a poorly
efficient photosensitizer. Although the exact mechanism
of sensitizer uptake by tumor cells is still unknown, there
is evidence that porphyrins are bound by plasma low
density lipoproteins (LDLs) and internalized following
interaction of LDLs with their receptors on cell mem-
branes; binding to LDLs is probably related to the am-
phiphilic character of the molecule.7 Aminor change in
lipophilicity can be obtained by the introduction of 1–3
methoxy appendix for each phenyl. In this series, com-
pound 3, featuring three methoxy groups on each phenyl
can be considered as the analogous of the so-called ꢀpeg-
ylatedꢁ porphyrins, which have been studied as an alter-
native to the hydroxy substituted ones, and have shown
an increased tumor selectivity and a prolonged activity
in vivo.8
recovered in about 50%, 68%, and 92% yields, respec-
tively. In some experiments the formation of bacterio-
chlorins (BCs) could not be avoided, as evidenced by
the presence of an absorption band at 730nm in UV–
vis spectra; the product of porphyrin over-reduction
was eliminated by carefully treating the mixture with a
diluted toluene solution of chloranyl. Chlorin 11 could
not be obtained from the corresponding 3,4,5-trihydroxy
phenyl porphyrin; however, it was isolated from chlorin
10 after demethylation with BBr3.
The formation of BCs as side products of chlorin syn-
thesis deserves a comment. The reduction of porphyrins
with tosylhydrazide and base is quite trivial and it is gen-
erally carried out with a large excess of reducing agent;
chlorin concentration is spectroscopically evaluated by
measuring the appearance of an absorption band at
650nm. The ratio between the intensity of the Soret
absorption band at 420nm and the absorption band at
650nm was assumed as an index of chlorin formation.
In order to achieve the desired value (420/650 = 6) the
reaction should be prolonged for many hours; therefore,
these harsh conditions end in the unavoidable formation
of some BCs, which is identified by the 730nm absorp-
tion band. Actually, based on their spectroscopic char-
acteristics, BCs could be even more interesting than
chlorins for in vivo applications, but the advantage of
activation by highly tissue-penetrating wavelengths is
overthrown by their greater thermal and photochemical
instability. In vivo experiments comparing m-THPC
with the corresponding BC m-THPBC, indicate that
the photobleaching rate is 20 times greater for the latter
than for the chlorin derivative.11
The methoxy groups have the advantage to easily under-
go further transformation to the corresponding hydroxy
derivatives by reaction with BBr3 at 0ꢂC. The presence
of one or more hydroxyl groups increases the degree
of hydrophilicity of the porphyrins; in this respect,
methoxyphenyl-porphyrins are considered the precursor
of the hydroxyphenyl-containing tetrapyrrole macrocy-
cles, of which the m-THCP is the most valuable
example.
Chlorosulfonation of phenyl rings is a well known pro-
cedure, which allows the anchorage of side arms to the
tetraaryl porphyrin structures; on the H2-TPP it occurs
on the four para-positions, yielding a symmetrically sub-
stituted chlorosulfonyl porphyrin.9 The chlorosulfonic
group can then be reacted with oxygen or nitrogen
nucleophiles to give sulfonic ester or sulfonamide; we
chose to synthesize the latter compound because of its
higher stability versus hydrolysis. The aminoethoxyetha-
nol (AEE) chain was then inserted with the aim to mod-
ify the lipophilic character of the tetraphenylporphyrin.
2.1. Cytotoxicity studies
Figure 2 shows the dose/response curves obtained in
HCT116 cells following exposure to the different por-
phyrin (panel A) and chlorin (panel B) derivatives for
24h and irradiation with visible light for 2h, and com-
pares them with the curve obtained for Photofrinꢀ
under identical experimental conditions. The intrinsic
cytotoxicity of tertraaryl derivatives was assessed by
omitting the irradiation step from the treatment proto-
col, and was found to be negligible in all cases (data
not shown). Thus, photoactivation is an absolute
requirement for the cytotoxic activity of these
Chlorins 8–10 and 12 were synthesized from the corre-
sponding porphyrin by reaction with toluene-4-sul-
fonylhydrazide, anhydrous K2CO3, and pyridine, using
the method described by Whitlock et al.,10 and were
A - porphyrin
B - chlorin
1.00
1.00
0.75
0.50
0.25
0.00
0.75
0.50
0.25
0.00
1
10
100
1000
10000
1
10
100
1000
10000
dose (ng/ml)
dose (ng/ml)
Figure 2. Representative dose–response curves obtained in HCT116 cells following exposure to the different porphyrin (panel A) and chlorin (panel
B) derivatives for 24h and irradiation with visible light for 2h. Curves obtained with PSs 1-12 are compared with the curve obtained for Photofrin
(j) under identical experimental conditions. Panel A:
Ã
1, s 2, r 3, Ã 4, . 5, m 6, d 7; panel B: . 8, m9, r 10, d 11, h 12.

4856
S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
Figure 3. Visualization of apoptotic nuclei by fluorescent microscopy in HCT116 cells following exposure to PSs: panel A––control; panel B––
photosensitizer 6 (4-hydroxy porphyrin) for 24h (3ng/mL) and irradiation with visible light for 2h.
compounds. All the compounds tested proved to be sig-
nificantly more cytotoxic than Photofrinꢀ, except the
trihydroxy-substituted derivatives (both porphyrin and
chlorin) and the 4-methoxy- and 3-methoxy-porphyrin
analogs, which were practically devoid of effect. This is
confirmed by the IC50 values calculated for each curve
and reported in Table 1. Interestingly, cell death induced
by tetraaryl derivatives occurs mainly by apoptosis,
in agreement with observations reported by other
authors.12,13 Apoptotic nuclei following photodynamic
treatment of HCT116 cells with 4-OH-tetraaryl porphy-
rin (the most effective compound in the series tested) at a
concentration corresponding to its IC50 value are shown
in Figure 3, in which the chromatin condensation typical
of the late phases of apoptosis is clearly visible.
phyrin and 17.74 1.26ng/mL for the chlorin;
mean SE), their molecules contain a polyfunctional-
ized side arm, to which a number of moieties could be
conveniently attached to improve pharmacokinetics, cel-
lular uptake, subcellular localization and/or tumor tar-
geting. The branched arms can be modified at the
sulfonyl chloride level, reacting nitrogen nucleophiles
with different structures or, in a subsequent step, exploit-
ing the presence of an hydroxyl moiety. As an example,
a-aminoacid methylester residues have recently been
grafted on the sulfonated tetraphenyl-porphyrin and
the phototoxic activity was tested on HT29 adenocarci-
noma cells.16
Both trimethoxy derivatives (3 and 10, IC50 values
10.74 1.82 and 6.51 1.32ng/mL, respectively;
mean SE) were found to be significantly more cyto-
toxic than either temoporfin (compound 8) or Photo-
frinꢀ. However, these derivatives contain 3,4,5-substi-
tuted phenyl rings and distinctions among the reactivi-
ties of the three positions are no longer possible;
therefore, their molecular structures will not bear much
further alterations, thus limiting future developments
for these compounds. To the best of our knowledge,
compounds 3 and 10 have not yet been tested on HCT
116 cells; about trimethoxy-phenyl porphyrin, Dolphin
has recently reported the activity of dihydroxychlorin
derivative of compound 3 on L1210 cells.17
Notably, the cytotoxic activity exerted by 3, 4, 5, 6, and
the newly synthesized derivatives (10, 12) was greater
than that exhibited by temoporfin (8), while chlorin 9
showed comparable cytotoxic activity. Temoporfin has
been approved in 2001 by the European Medical Agency
(EMEA) for use in head and neck squamous cell carci-
nomas, and is the latest addition to the small group of
compounds currently used for PDT. Its photodynamic
and toxicity profiles compare favorably with those of
Photofrinꢀ: in fact, in the present study it was found
to be significantly more cytotoxic than the latter (IC50
values 24.8 1.28ng/mL vs. 73.67 8.04, p < 0.05;
mean SE), and this results confirm previous in vitro
and clinical observations.14,15 In addition, the main-
adverse effects of temoporfin have been related to local
tumor necrosis, whereas extended photosensitivity, rep-
resenting the major drawback to the clinical use of Pho-
tofrinꢀ, can easily be prevented with the adoption of
appropriate measures. Among the novel PSs, both tri-
hydroxy derivatives (compounds 7 and 11, IC50 values
221.81 36.42 and 3358.00 96.00ng/mL, respectively;
mean SE) display a significantly lower cytotoxicity
than either temoporfin (compound 8) or Photofrinꢀ
(IC50 values 24.8 1.28 and 73.67 8.04ng/mL, respec-
tively; mean SE); therefore, it is hard to foresee any
further developments for these derivatives. The two ami-
noethoxyethanol derivatives (compounds 4 and 12), for
the first time used as photosensitizers, seem far more
promising as lead compounds: in fact, while their cyto-
toxic effect does not significantly differ from that of
temoporfin (IC50 values 23.20 3.46ng/mL for the por-
The relative cytotoxicities of compounds containing a
porphyrin or a chlorin rings deserve some comments.
To the best of our knowledge, direct comparisons of
the phototoxicity of porphyrins and the corresponding
chlorins have never been reported in the literature, as
most of the studies address differences in the absorption
profiles between the two families rather than biological
effects. The results obtained in the present study indicate
that the presence of a fully unsaturated tetrapyrrole ring
or a mono hydrogenated ring in the tetraaryl derivatives
may significantly affect the cytotoxic properties of the
compounds. Whereas no significant differences were ob-
served for the trimethoxy (3 and 10) and aminoethoxy-
ethanol (4 and 12) pairs, the 3-hydroxy and 4-hydroxy
chlorin derivatives (8 and 9, IC50 values 24.8 1.28
and 28.69 4.12ng/mL, respectively; mean SE) were
significantly less cytotoxic than the corresponding por-
phyrins (5and 6, IC50 values 5.13 0.98 and

S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
4857
3.07 0.37ng/mL, respectively; mean SE). Prelimi-
nary uptake data obtained for some of the tested PSs
indicate that photodynamic activity is loosely correlated
to the intracellular PS levels and that, in the case of the
4-hydroxy porphyrin/chlorin pair, the chlorin derivative
is less cell-permeant than the corresponding porphyrin,
as assessed by fluorometric measures on cell extracts
(data not shown). This observation can hardly be related
to structural differences between porphyrins and chlo-
rines; in fact, while it is obvious that the presence of
one extra double bond can influence the spectroscopic
behavior of the polyunsaturated structure, the whole
conformation of the molecule is not affected by the addi-
tion of two hydrogen atoms. In conclusion, the observed
difference of phototoxicity between porphyrins 5, 6 and
the corresponding chlorins 8, 9 cannot be accounted for
by their general structural features.
corded on a Bruker 400MHz spectrometer in CDCl3 or
[d6] DMSO; chemical shifts are expressed in ppm rela-
tive to chloroform (7.26). Mass spectrometric measure-
ments were performed on
a Finnigan LCQ-MS
instrument fitted with an electrospray ionization (ESI)
or atmospheric pressure chemical ionization (APCI)
ion sources, equipped with an ion trap mass analyzer.
HPLC analyses were conducted with a Thermo Separa-
tion Products (TSP) instrument coupled with a Finnigan
LCQ-MS. The instrument was fitted with a 0.21 · 15cm
column (Supelco, Discovery) packed with C-18 re-
versed-phase particles (5lm) and operated with an iso-
cratic elution with A:B/30:70 ratio (A = H2O, 0.1%
CH3COOH; B = CH3CN:CH3OH/80:20) at 0.2mL/min.
Elemental analyses were performed on a ThermoQuest
NA2100, C, H, N analyzer, equipped with an electronic
mass flow control and a thermal conductivity detector.
Another peculiar observation inferred from the results
of the present work is that, under the conditions
adopted for cytotoxicity assays, no substantial differ-
ences in activity were observed between 3- and 4-hy-
droxyphenyl substituted tetrapyrroles, either in the
porphyrin or chlorin series. About chlorins, recent and
old papers report an activity 1–2 order of magnitude
higher for the meta isomer with respect to the para, in
both in vitro and in vivo experiments.17,18 These results
cannot be accounted for by structural evidence, and, as
far as we know, no explanation has been offered for the
observed difference in activity. In a HPLC/MS study
currently ongoing in our laboratory 3-hydroxy deriva-
tives do exhibit a higher affinity for the stationary phase
(RP-C18) than the 4-OH substituted agents; this obser-
vation indicates that 3-hydroxy-derivatives are more
lipophilic, which could account for their greater photo-
toxicity. However, differences in cytotoxicity were not
apparent in HCT116 cells, suggesting that cell type-spe-
cific factors may also play a role in the response to PDT.
Analytical thin-layer chromatography (TLC) was per-
formed using Merck 60 F254 silica gel (precoated sheets,
0.2mm thick) or on Macherey–Nagel F254 silica gel C18-
100 (precoated sheets, 0.25mm thick). Silica gel 60 (70–
230 mesh, Merck) was used for column chromatography.
Aromatic aldehydes, 5,10,15,20-tetraphenyl-21H,23H-
porphyrin [H2-TPP] and 5,10,15,20-tetra-(4-hydroxy-
phenyl)-21H,23H-porphyrin (6), were commercial prod-
ucts (Sigma–Aldrich) and used as received. Pyrrole and
BF3ÆEt20 were freshly distilled prior to use. Dichloro-
methane used for porphyrin synthesis was distilled from
CaCl2 directly into the reaction flask.
4. Synthesis of free base tetraarylporphyrins
5,10,15,20-Tetra(3-methoxyphenyl)-21H,23H-porphyrin
[H2-T(m-OMeP)P, 1], 5,10,15,20-tetra(4-methoxyphen-
yl)-21H,23H-porphyrin [H2-T(p-OMeP)P, 2], and
5,10,15,20-tetra(4,5,6-trimethoxyphenyl)-21H,23H-por-
phyrin [H2-T(OMe3P)P, 3] were synthesized via conden-
sation of the corresponding aromatic aldehydes and
pyrrole under mixed-acid catalysis, as recently reported
by Lindsey and co-workers.5 The chlorosulfonyl deriva-
tive of H2-TPP was obtained by reacting free-base por-
phyrins with chlorosulfonic acid, following the method
described by Rocha Gonsalves et al.9 5,10,15,20-tetra-
(4-N-ethoxyethanolsulfonamidophenyl)-21H,23H-porph-
yrin [H2-T(p-SO2 AEE)P-P, 4] was synthesized as previ-
ously described.6
In summary, we can conclude that trimethoxyphenyl
derivatives (3, 10) and hydroxyphenyl compounds (5, 6,
8, 9) are the most active among the tested compounds.
Substituting an aminoethoxyethanol moiety in the phenyl
rings of porphyrins and chlorins yields compounds (4, 12)
with cytotoxic potencies very similar to those exhibited
by currently used PS that lend themselves to further
chemical modifications aimed at improving their antitu-
mor efficacy and/or their pharmacokinetics. Subsequent
in vitro assessments of the photodynamic activity of mod-
ified aminoethoxyethanol derivatives will have to be per-
formed directly on chlorin derivatives, as they have better
light absorbing characteristics than the corresponding
porphyrins (and are, therefore, better candidates for in
vivo applications) and may differ considerably from the
latter in both cell permeability and cytotoxicity.
The general procedure for porphyrin syntheses is fully
described for the first compound, the others being pre-
pared under similar conditions.
4.1. 5,10,15,20-Tetra(3-methoxyphenyl)-21H,23H-
porphyrin [H2-T(m-OMe)P-P, 1]
3. Experimental
3.1. General
BF3ÆEt2O (6lL, 6 · 10À3 mmol) and 0.35mL (4.5mmol)
of trifluoroacetic acid (TFA) were added to a solution of
0.6mL (0.681g, 5mmol) of m-anisaldehyde and 0.35mL
(0.345g, 5mmol) of freshly distilled pyrrole in 500mL of
CH2Cl2; the mixture was kept at rt for 2h. When all the
UV–vis absorption spectra were measured on a Perkin–
Elmer Lambda 10 instrument. 1H NMR spectra were re-

4858
S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
aldehyde was reacted (TLC:SiO2; hexane/CH2Cl2 6/4)
1.01g (4mmol) of chloranyl were added and the mixture
was kept under reflux for 4h. The solvent was evapo-
rated and the raw material purified by column chroma-
tography (SiO2; CH2Cl2). The fractions were tested by
TLC, those containing only one single spot were col-
lected and the recovered product was further purified
by refluxing with MeOH (20mL). The solid matter
was filtered and washed on the filter with a small
amount of CH2Cl2 affording 374mg (41%) as a purple
after this period, 30mL of CH2Cl2 was added to the
reaction mixture together with the required amount of
1M aqueous NaOH to neutralize the mixture. The lay-
ers were separated and the organic phase was further
washed with water, dried (Na2 SO4) and concentrated
to dryness yielding a solid product (31mg 33%). 1H
NMR (CDCl3) d: À2.97 (s, 2H, NH); 7.24 (d, 4H);
7.60 (dd, 12H); 8.89 (s, 8H); 9.89 (s, 4H, OH). HPLC:
retention
time = 2.57min.
UV–vis(water)
:
420nm
(KBr):
(e = 102,000); 516nm (e = 11,000). IR
solid; UV–vis(dichloromethane)
:
418nm (e = 189,000);
3402cmÀ1 (m OH); 1605cmÀ1 (m C@C); 1259cmÀ1 (m
CO). MS-ESI+: m/z 679.7 (100%); 680.6 (44%); MS-
ESIÀ: m/z 677.4 (100%); 678.4 (40%). Molecular weight
calculated for C44H30N4O4 = 678.7.
514nm (e = 12,000). 1H NMR (CDCl3) d: À2.77 (s,
2H); 4.00 (s, 12H); 7.36 (dd, 4H); 7.67 (t, 4H); 7.83 (t,
8H); 8.91 (s, 8H). IR (KBr): 2930cmÀ1 (m CH);
1152cmÀ1 (m CO). HPLC: retention time = 4.49min.
MS-APCI+: m/z 735.6 (100%); 736.5 (45%). Molecular
weight calculated for C48H38N4O4 = 734.8.
5.2. 5,10,15,20-Tetra(3,4,5-trihydroxyphenyl)-21H,23H-
porphyrin [H2-T(OH)3P-P, 7]
4.2. 5,10,15,20-Tetra(4-methoxyphenyl)-21H,23H-
porphyrin [H2-T(p-OMe)P-P, 2]
Asolution of 142mg (0.146mmol) of porphyrin in 15mL
of CH2Cl2 was treated with 0.6mL (0.6mmol) of BBr3
solution as described above. The insoluble product was
recovered by filtration and purified by gel permeation
column chromatography on Bio-Beads resins, with
CH3CN as eluant, to give 73mg (62%) of the pure prod-
Compound 2 was synthesized as described above;
the isolated pure product was 315mg, corresponding
to 21.8% yields. UV–vis(dichloromethane)
:
422nm
(e = 253,000); 518nm (e = 18,000). 1H NMR (CDCl3)
d: À2.73 (s, 2H, NH); 4.12 (s, 12H); 7.30 (d, 8H); 8.14
(d, 8H); 8.88 (s, 8H). IR (KBr): 2927cmÀ1 (m CH);
1147cmÀ1 (m CO). HPLC: retention time = 6.45min.
MS-APCI+: m/z 735.5 (100%); 736.6 (45%); MS-APCIÀ:
m/z 734.4 (100%). Molecular weight calculated for
C48H38N4O4 = 734.8.
uct. UV–vis(water)
:
424nm (e = 101,000); 516nm
(e = 8500). IR (KBr): 3512cmÀ1 (m OH). 1H NMR
(DMSO) d: 7.62 (s, 8H); 8.60 (s, 8H); 9.80 (m, 12H,
OH). MS-ESI+ (flow injection) 807.5 highest peak of
molecular cluster; MS-ESIÀ 805.3. Molecular weight cal-
culated for C44H30N4O12 = 806.7. Anal. Calcd: C, 65.50;
H, 3.74; N, 6.94. Found; C, 66.18; H, 3.77; N, 6.82.
4.3. 5,10,15,20-Tetra(3,4,5-trimethoxyphenyl)-21H,23H-
porphyrin [H2-T(OMe)3P-P, 3]
6. Synthesis of chlorins from porphyrins
3,4,5-Trimethoxybenzaldehyde (981mg, 5mmol) and
0.35mL (5mmol) of pyrrole were reacted as described
for of compound 1. The raw material was purified by col-
umn chromatography (SiO2, CH2Cl2/MeOH 98/2) and the
recovered product was crystallized dissolving the solid into
the minimum amount of dichloromethane then settling as
a precipitate by careful addition of hexane (CH2Cl2/hex-
ane 1/2; v/v) to give 400mg (33%) of the pure product.
6.1. 5,10,15,20-Tetra(3-hydroxyphenyl)-2,3-dihydro-
21H,23H-chlorin [H2-T(m-OH)P-CL, 8]
Toluene-4-sulfonylhydrazide (87.8mg, 0.472mmol) and
163mg (1.18mmol) of K2CO3 were added to a solution
of 80mg (0.118mmol) of porphyrins 5 in 10mL of pyr-
idine; the mixture was refluxed for 2h; the reaction pro-
gress was monitored by means of UV–vis spectroscopy,
based on the occurrence of the chlorin band at 650nm.
The same initial amount of hydrazide and K2CO3 were
added every hour until the reaction was completed
(determined evaluating the ratio of the intensity between
the band at 420nm and that one at 650nm; this ratio
must be equal or lower than 6). In some cases, the for-
mation of an absorption band at 730nm was detected,
which is characteristic of the product of over-reduction
(bacteriochlorin). At the end of the reaction, 35mL of
AcOEt and 18mL of H2O were added and the mixture
was kept at 70ꢂC for 1h. The organic layer was isolated,
washed a few times with HCl 10%, then with H2O, dried
(Na2SO4) and concentrated to yield a solid product
(38mg 47.4%). The absorption spectra of the isolated
compound was again controlled and, in the case of the
presence of the peak at 730nm, the solid was dissolved
in CH2Cl2 and treated with a diluted toluene solution
of chloranyl. This procedure must be carried out with
particular attention to avoid oxidation of the chlorin
back to the initial porphyrin.
UV–vis(water): 430nm (e = 133,000); 518nm (e = 22,000).
IR (KBr): 2924cmÀ1 (m CH); 1122cmÀ1 (m CO). 1H
NMR (CDCl3) d: À2.75 (s, 2H), 3.99 (s, 24H, OCH3);
4.21 (s, 12H, OCH3); 7.49 (s, 8H); 8.98 (s, 8H). HPLC:
retention time = 2.64min. MS-APCI+: m/z 975.5
(100%); 976.5 (50%) 978 (5%). Molecular weight calcu-
lated for C56H54N4O12 = 975.0.
5. Synthesis of hydroxy substituted porphyrins from
methoxy porphyrins
5.1. 5,10,15,20-Tetra(3-hydroxyphenyl)-21H,23H-
porphyrin [H2-T(m-OH)P-P, 5]
Asolution of 100mg (0.136mmol) of porphyrin 1 in
14mL of CH2Cl2 was stirred at 0ꢂC for 0.5h, then
2.72mL (2.72mmol) of 1M BBr3 solution was added.
The mixture was kept at 0ꢂC for 1h and at rt for 18h;

S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
4859
As the final step of the reaction, the solution was con-
6.4. 5,10,15,20-Tetra(3,4,5-trihydroxyphenyl)-2,3-di-
hydro-21H,23H-chlorin [H2-T(OH)3P-CL, 11]
centrated and purified by column chromatography
(SiO2; CH2Cl2/MeOH 9/1) to give 30mg (37%) of the
pure product.
Asolution of 20mg (20mmol) of chlorin 10 in 8mL of
CH2Cl2 was treated with 0.6mL (0.6mmol) of BBr3
solution as described above. The insoluble product
was recovered by filtration and purified by gel permea-
tion chromatography on Bio-Beads resins, with
CH3CN as eluant, yielding 10mg (62.5%) of the desired
product.
UV–vis(water): 420nm (e = 74,700); 650nm (e = 18,000).
IR (KBr): 3455cmÀ1 (m OH); 1604cmÀ1 (C@C);
1
1258cmÀ1 (m CO). H NMR (CDCl3) d:À1.65 (s, 2H,
NH); 4.16 (s, 4H) 7.09 (m, 8H); 7.52 (m, 8H); 8.24 (d,
2H); 8.36 (s, 2H); 8.63 (d, 2H); 9.73 (s, 2H, OH); 9.79
(s, 2H, OH). MS-ESI+ (flow injection): m/z 681.6
(100%); 682.5 (45%); MS-ESIÀ (flow injection): m/z
679.4 (100%). Molecular weight calculated for
C44H32N4O4 = 680.8.
UV–vis(water): 420nm (e = 77,000); 638nm (e = 12,000).
IR (KBr): 3490cmÀ1 (m OH); 1H NMR (DMSO):
À1.50 (s, 2H, NH); 4.06 (s, 4H); 7.50 (m, 4H); 7.80 (m,
4H); 8.40 (s, 6H) 9.60 (m, 12H, OH). MS-ESI+: m/z
809.5 highest peak of molecular cluster. Molecular
weight calculated for C44H32N4O12 = 803.7. Anal. Calcd:
C, 65.76; H, 4.01; N, 2.24. Found; C, 66.23; H, 4.01; N,
2.28.
6.2. 5,10,15,20-Tetra(4-hydroxyphenyl)-2,3-dihydro-
21H,23H-chlorin [H2-T(p-OH)P-CL, 9]
The same procedure described for compound 8 was used
for this compound, thus 160mg (0.236mmol) of porphy-
rins 6 in 10mL of pyridine was treated with 131.8mg
(0.708mmol) of toluene-4-sulfonylhydrazide and
163mg (1.18mmol) of K2CO3. The reaction was fol-
lowed as described above and, at the end, the title com-
pound was isolated as a solid product (109mg, 67.7%).
UV–vis(water): 424nm (e = 64,000); 656nm (e = 10,100).
IR (KBr): 3409cmÀ1 (m OH); 1601cmÀ1 (m C@C);
6.5. 5,10,15,20-Tetra(4-N-ethoxyethanolsulfonamidophen-
yl)-2,3-dihydro-21H,23H-chlorin [H2-T(p-SO2 AEE)P-P,
12]
The same procedure described for compound 10 was
used for this compound, thus 80mg (62mmol) of por-
phyrins 4 in 7mL of pyridine was treated with 34.6mg
(186mmol) of toluene-4-sulfonylhydrazide and 43mg
(310mmol) of K2CO3. The reaction was followed as de-
scribed above, and the title compound was isolated as a
solid product (42mg, 52%).
1
1255cmÀ1 (m CO). H NMR (CDCl3) d: À1.56 (s, 2H,
NH); 4.11 (s, 4H); 7.12 (dd, 8H); 7.67 (d, 4H); 7.86 (d,
4H); 8.20 (d, 2H); 8.35 (s, 2H); 8.60 (d, 2H); 9.74 (s,
2H, OH); 9.87 (s, 2H, OH). HPLC: retention
time = 4.54. MS-ESI+: m/z 681.4 (100%); 682.4 (40%);
MS-ESIÀ m/z 679.3 (100%). Molecular weight calcu-
lated for C44H32N4O4 = 680.8.
UV–vis(water): 420nm (e = 72,000); 650nm (e = 11,000).
IR (KBr): 3430cmÀ1 (m OH); 1610cmÀ1 (m C@C). H
1
NMR (DMSO) d: À1.50 (s, 2H, NH); 3.15 (t, 8H),
3.40 (m, 8H), 3.45 (m, 20H), 4.20 (s, 4H); 4.66 (br s,
4H), 7.55 (d, 8H), 7.95 (d, 8H); 8.25 (d, 2H), 8.34 (d,
2H), 8.60 (d, 2H). MS-MALDI-TOF: m/z 1284.9 highest
peak of molecular cluster. Molecular weight calculated
for C60H68N8O16S4 = 1285.5. Anal. Calcd: C, 56.05; H,
5.33; N, 8.71. Found; C, 55.71; H, 5.40; N, 8.62.
6.3. 5,10,15,20-Tetra(3,4,5-trimethoxyphenyl)-2,3-di-
hydro-21H,23H-chlorin [H2-T(OMe)3P-CL, 10]
Asolution of 40mg (0.041mmol) of porphyrins 3 in
5mL of pyridine was treated with 30.5mg (0.164mmol)
of toluene-4-sulfonylhydrazide and 56.6mg (0.41mmol)
of K2CO3. The mixture was refluxed for 2h. After this
period the reaction was checked through UV–vis every
hour and the same amount of hydrazide and K2CO3
was added until the reaction was completed. Then
15mL of AcOEt and 8mL of H2O were added and the
mixture was kept at 70ꢂC for 1h. The organic phase
was isolated, washed (HCl 10%, then H2O), dried
(Na2SO4), and concentrated affording 38mg (92%) of
a solid product.
6.6. Cytotoxicity studies
Human adenocarcinoma HCT116 cells were obtained
from the American Type Culture Collection (Rockville,
MD, USA) and maintained in DMEM (Mascia-Brun-
elli, Milano, Italy) supplemented with 10% fetal bovine
serum (Mascia Brunelli) at 37ꢂC in a humidified 5%
CO2 atmosphere. The antiproliferative effects of the dif-
ferent PSs, including Photofrin, were assessed using the
MTT assay.19 Briefly, 5 · 104 cells/mL were seeded onto
96-well plates and allowed to grow for 48h prior to
treatment with different PS concentrations. After 24h,
the PS-containing medium was replaced by PBS, and
cells were irradiated under visible light (halogen lamp
500W) for 2h (average value between 400 and 700nm
determine with a LICOR-1800 spectroradiometer; light
irradiance 5.5 · 10À2 mWcmÀ2 nmÀ1 and a light energy
of 39.6mJcmÀ2 nmÀ1). At the end of this time cells were
incubated for 24h in drug free medium; MTT was then
added to each well (final concentration 0.4mg/mL) for
3h at 37ꢂC and formazan crystals formed through
UV–vis(water): 424nm (e = 93,000); 630nm (e = 14,500).
IR (KBr): 2931cmÀ1 (m CH); 1127cmÀ1 (CO). 1H
NMR (CDCl3): À1.47 (s, 2H, NH); 3.94 (s, 12H,
OCH3); 3.96 (s, 12H, OCH3); 4.14 (s, 6H, OCH3); 4.16
(s, 6H, OCH3); 4.28 (s, 4H); 7.10 (s, 4H); 7.39 (s, 4H);
8.11 (d, 4H). HPLC: retention time = 2.62min. MS-
APCI+: m/z 977.6 (100%); 978.6 (40%); 979.5 (10%);
MS-APCIÀ: m/z 976.5 (100%); 977.4 (55%);. Molecular
weight calculated for C56H56N4O12 = 977.1. Anal. Calcd:
C, 68.83; H, 5.77; N, 5.73. Found; C, 69.38; H, 5.75; N,
5.62.

4860
S. Banfi et al. / Bioorg. Med. Chem. 12 (2004)4853–4860
MTT metabolism by viable cells were dissolved in
DMSO. Optical densities were measured at 570nm
using a Universal Microplate Reader EL800 (Bio-
Tek Instruments). Possible intrinsic (i.e., nonphoto-
dynamic) cytotoxic effects of the PS were assessed on
control treated as described above, but omitting cell
irradiation.
References and notes
1. MacDonald, I. J.; Dougherty, T. J. J. Porphyrins Phthalo-
cyanines 2001, 5, 105.
2. Dolmans, D. E. J. G. J.; Fukumura, D.; Jain, R. K.
Nature Rev. Cancer 2003, 3, 380.
3. Oleinick, N. L.; Morris, R. L.; Belichenko, I. Photochem.
Photobiol. Sci. 2002, 1, 1.
4. Hopper, C. Lancet Oncol. 2000, 1, 212.
IC50 (dose affecting 50% of cells) values were obtained
by nonlinear regression analysis, using the GraphPad
PRISM 3.03 software (GraphPad Software Inc., San
Diego CA).
5. Chevalier, F.; Geier, G. R., III; Lindsey, J. S. J. Porphyrins
Phthalocyanines 2002, 6, 186.
6. Banfi, S.; Cassani, E.; Caruso, E.; Cazzaro, M. Bioorg.
Med. Chem. 2003, 11, 3595.
7. Osterloh, J.; Vicente, M. G. H. J. Porphyrins Phthalocya-
nines 2002, 6, 305.
8. Westermann, P.; Glanzmann, T.; Andrejevic, S.; Braich-
otte, D. R.; Forrer, M.; Wagnieres, G. A.; Monnier, P.;
van den Bergh, H.; Mach, J.-P.; Folli, S. Int. J. Cancer
1998, 76, 842.
9. Gonsalves, A. M. dꢁA. R.; Johnstone, R. A. W.; Pereira,
M. M.; de SantꢁAna, A. M. P.; Serra, A. C.; Sobral, A. J.
F. N.; Stocks, P. A. Heterocycles 1996, 43, 829.
10. Whitlock, H. W.; Hanauer, R.; Oester, M. Y.; Bower, B.
K. J. Am. Chem. Soc. 1969, 91, 7485.
11. Rovers, J. P.; deJode, M. L.; Rezzoug, H.; Grahn, M. F.
Photochem. Photobiol. 2000, 72(3), 358.
12. Marchal, S.; Bezdetnaya, L.; Guillemin, F. Biochemistry
(Mosc) 2004, 69, 45.
7. Visualization of apoptotic cells
Apoptotic cells were visualized by fluorescence micro-
scopy. HCT116 cells were seeded onto coverslips,
allowed to grow for 48h and subsequently treated with
the different PSs, at concentrations approximating their
respective IC50 values. After 24h, the PS-containing
medium was replaced by PBS, and cells were irradiated
for 2h, as previously described. At the end of this
time cells were incubated for 24h in drug-free
medium. Cells still attached to coverslips were washed
in phosphate-buffered saline (PBS) and fixed in 70%
methanol for at least 30min at À20ꢂC, washed in
PBS, and incubated in PBS containing RNAse A
(1kU/mL), propidium iodide (50mg/mL), and 0.05%
Nonidet (N)P40. After 30min on ice, coverslips were
mounted onto glass slides with Mowiol 4-88 and ob-
served with an Olympus photomicroscope equipped
for epifluorescence.
13. Bourre, L.; Rousset, N.; Thibaut, S.; Eleouet, S.; Lajat,
Y.; Patrice, T. Apoptosis 2002, 7, 221.
14. Reuther, T.; Kubler, A. C.; Zillmann, U.; Flechtenmacher,
C.; Sinn, H. Lasers Surg. Med. 2001, 29, 314.
15. Mlkvy, P.; Messmann, H.; Regula, J.; Conio, M.; Pauer,
M.; Millson, C. E.; MacRobert, A. J.; Bown, S. G.
Neoplasma 1998, 45, 157.
16. Sobral, A. J. F. N.; Eleouet, S.; Rousset, N.; Gonsalves,
´
A . M. dꢁA. R.; Le Meur, O.; Bourre, L.; Patrice, T. J.
Porphyrins Phthalocyanines 2002, 6, 456.
17. Dolphin, D.; Macalpine, J. K.; Boch, R. J. Porphyrins
Phthalocyanines 2002, 6, 146.
18. Bonnett, R.; White, D. R.; Winfield, U.-J.; Berenbaum,
M. C. Biochem. J. 1989, 261, 277.
Acknowledgements
This work was supported by a grant from the ꢀProgetto
di Eccellenza per la Ricerca di Ateneoꢁ, 2003/04, Univer-
sity of Insubria (Italy).
19. Alley, M. C.; Scudiero, D. A.; Monks, A.; Hursey, M. L.;
Czerwinski, M. J.; Fine, D. L.; Abbott, B. J.; Mayo, J. G.;
Shoemaker, R. H.; Boyd, M. R. Cancer Res. 1988, 48, 589.