Synthesis, properties, and photodynamic inactivation of Escherichia coli using a cationic and a noncharged Zn(II) pyridyloxyphthalocyanine derivatives

Article abstract of DOI:10.1016/j.bmc.2005.01.063

The photodynamic effect of a cationic Zn(II) N- methylpyridyloxyphthalocyanine (ZnPc 2) and a noncharged Zn(II) pyridyloxyphthalocyanine (ZnPc 1) has been compared in both homogeneous media bearing photooxidizable substrates and in vitro using a typical G

Full text of DOI:10.1016/j.bmc.2005.01.063

Bioorganic & Medicinal Chemistry 13 (2005) 3037–3045
Synthesis, properties, and photodynamic inactivation of
Escherichia coli using a cationic and a noncharged Zn(II)
pyridyloxyphthalocyanine derivatives
*
´
Ines Scalise and Edgardo N. Durantini
´ ´ ´
Departamento de Quımica, Universidad Nacional de Rıo Cuarto, Rıo Cuarto, Agencia Postal Nro. 3,
X5804BYA Rıo Cuarto, Cordoba, Argentina
´
´
Received 29 September 2004; accepted 28 January 2005
Abstract—The photodynamic effect of a cationic Zn(II) N-methylpyridyloxyphthalocyanine (ZnPc 2) and a noncharged Zn(II) pyr-
idyloxyphthalocyanine (ZnPc 1) has been compared in both homogeneous media bearing photooxidizable substrates and in vitro
using a typical Gram-negative bacterium Escherichia coli. Absorption and fluorescence spectroscopic studies were analyzed in dif-
ferent media. Fluorescence quantum yields (/F) of 0.23 for ZnPc 1 and 0.22 for ZnPc 2 were calculated in N,N-dimethylformamide
(DMF). The singlet molecular oxygen, O2(1Dg), production was evaluated using 9,10-dimethylanthracene (DMA) in DMF yielding
values of UD = 0.56 for ZnPc 1 and 0.59 for ZnPc 2. A faster decomposition of L-tryptophan (Trp), which was used as biological
substrate model, was obtained using ZnPc 2 as a sensitizer with respect to ZnPc 1. In biological medium, the E. coli cultures were
treated with 10 lM of sensitizer for different times at 37 ꢀC in the dark. Both ZnPcs 1 and 2 are rapidly bound to E. coli cells in 5 min
and the amount of cell-bound sensitizer is not appreciably changed incubating the cultures for longer times. The recovered ZnPc 2
after one washing step is ꢀ3 times higher than 1, reaching a value of ꢀ3 nmol/106 cells. After irradiation with visible light, a higher
photoinactivation of cells was found for ZnPc 2. Thus, a ꢀ4.5 log (99.997%) decrease of cell survival was obtained after 30 min of
irradiation. On the other hand, a very low photodamage was found for cells treated with ZnPc 1 (ꢀ0.5 log). Also, these results were
established by stopping of growth curves for E. coli. In the structure of ZnPc 2, the cationic centers are isolated from the phthalo-
cyanine ring by an ether bridge, which also provides a higher mobility of the charges facilitating the interaction with the outer mem-
brane of the Gram-negative bacteria. These studies show that cationic ZnPc 2 is an efficient phototherapeutic agent with potential
applications in photodynamic inactivation of bacteria.
ꢁ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
inactivation.3,4 The incidence of these mechanisms de-
pends mainly on the sensitizer, substrate, and the nature
of the medium.5
Photodynamic therapy (PDT) is an innovative treat-
ment for several types of cancer. This therapy is based
on the administration of a photosensitizer, which is
selectively incorporated in tumor cells. The subsequent
exposure to visible light in the presence of oxygen specifi-
cally inactivates neoplastic cells.1,2 Two photoreaction
processes are principally implicated in the photodamage
of cells, one involves the generation of free radicals (type
I) and the other produces singlet molecular oxygen,
O2(1Dg) (type II), as the main species responsible for cell
Recently, a nononcological application of PDT has been
established for photoinactivation of bacteria in an
attempt to overcome the problem of bacterial strains
resistant to current antibiotics.6,7 Thus, pathogenic
microorganisms growing in vivo as localized foci of
infection, on skin or on accessible mucous membrane,
would be candidates for photodynamic destruction. Pre-
vious investigation showed that phthalocyanine deriva-
tives can photosensitize the inactivation of various
microbial pathogens.8–11 Gram-positive bacteria are
efficiently photoinactivated by a variety of sensitizers,
whereas Gram-negative bacteria are usually resistant
to the action of negatively charged or neutral photo-
sensitizers.10,12 However, the presence of positively
Keywords: Phthalocyanine; Photodynamic effect; Escherichia coli;
Photodynamic therapy; Photoinactivation; Bacteria.
*
Corresponding author. Fax: +54 358 467 6233; e-mail:
edurantini@exa.unrc.edu.ar
0968-0896/$ - see front matter ꢁ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.01.063

3038
I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
charged functional groups in the photosensitizer
structure allows an extensive photoinduced inactivation
of Gram-positive and also of Gram-negative bacterial
cells. The positive charge on the photosensitizer mole-
cule appears to promote a tight electrostatic interaction
with negatively charged sites at the outer surface of the
bacterial cells. This association increases the efficiency of
the photoinactivation processes.8,10
2. Results and discussion
2.1. Synthesis
ZnPc 1 was synthesized by a two-step procedure.19,20
First, 4-(4-pyridyloxy)phthalonitrile was prepared by
a nucleophilic ipso-nitro substitution reaction of 4-
nitrophthalonitrile with 4-hydroxypyridine in the pres-
ence of K2CO3 (Scheme 1). Dinitrile derivative was
isolated by flash chromatography with 66% yield. The
cyclotetramerization of dinitrile with zinc(II) acetate in
the presence of organic base 1,8-diazabicyclo[5.4.0]-
undec-7-ene (DBU) was performed in n-pentanol
(Scheme 2). After reflux for 18 h, the reaction results
in the formation of the corresponding ZnPc 1 as
mixtures of constitutional isomers with 68% yield.
In previous studies, we have investigated the photodyn-
amic activity of porphyrin derivatives in different biomi-
metic media and in vitro on the Hep-2 human larynx
carcinoma cell line.13–15 In particular, the presence of a
cationic charge in the sensitizer structure produces an in-
crease in the uptake of this amphiphilic porphyrin into
Hep-2 cells with respect to a nonionic porphyrin
model.16 In this case, a higher cellular incorporation is
also accompanied by a higher photocytotoxic activity
on Hep-2 cells. Recently, meso-substituted cationic
porphyrin derivatives with asymmetric charge distribu-
tion were synthesized as efficient sensitizers to inactivate
Escherichia coli cells.17
Cationic sensitizer 2 was obtained treating the ZnPc 1
with methyl iodide for 48 h at reflux in acetone (Scheme
2). The exhaustive methylation produces ZnPc 2 with
95% yield. In the structure of ZnPc 2, the cationic cen-
ters are isolated from the phthalocyanine macrocycle
ring by ether bonds. This spacer provides a higher
mobility of the charges, which could facilitate the inter-
action with the outer membrane of the Gram-negative
bacteria.
In this paper, the photodynamic activity of a cationic
Zn(II) N-methylpyridyloxy phthalocyanine (ZnPc 2)
was compared with that of a noncharged Zn(II) pyr-
idyloxyphthalocyanine (ZnPc 1), in homogeneous med-
ium bearing a photooxidizable substrate and in vitro
using a typical Gram-negative bacterium E. coli. Phtha-
locyanines derivatives exhibit a high absorption coeffi-
cient in the visible region of the spectrum, mainly in
the phototherapeutic window (600–800 nm) and a long
lifetime of triplet excited state to produce efficiently
O2(1Dg).3 The use of phthalocyanines for sterilization
of infection contaminants in red blood cells has been
of major interest because of the intense absorption of
red light where hemoglobin absorption is minimal.7,18
The behaviors of ZnPcs 1 and 2 in solution were com-
pared with that of Zn(II) phthalocyanine (ZnPc). These
neutral and cationic phthalocyanines derivatives have
appropriate photophysical properties to be used as pho-
tosensitizer in PDT.3 In biological medium, ZnPc 1 was
used as a noncationic model. Thus, the main purpose of
this study was to evaluate the photoinactivation efficien-
cies of these photosensitizers in a Gram-negative repre-
sentative bacterium E. coli.
2.2. Spectroscopic studies and phthalocyanine properties
The absorption spectra of ZnPcs 1 and 2 in N,N-dimeth-
ylformamide (DMF) are gathered in Figure 1. The spec-
tra show the typical Soret and Q-bands, characteristic of
zinc(II) phthalocyanines (ZnPc).3,21 The values of molar
coefficient (e) in DMF are reported in Table 1. The Q-
band presents a bathochromic shift by ꢀ10 nm when
compared with that of ZnPc (670 nm). In addition, the
wavelength absorption maxima of these ZnPcs were
OH
N
K2CO3
CN
CN
CN
CN
N
+
O
DMF
80˚C, 3h
66%
O2N
Scheme 1. Synthesis of 4-(4-pyridyloxy)phthalonitrile.
+
N
N
O
O
O
+
O
N
N
N
Zn
N
N
CN
N
N
N
N
DBU
N
N
N
N
CH3I
4I-
4
N
N
N
Zn
N
N
Zn(CH3COO)2
1-pentanol
reflux
O
N
CN
reflux
72 h
95%
+
N
N
18 h
68 %
O
O
O
N
O
ZnPc 1
ZnPc 2
N
+
Scheme 2. Synthesis of ZnPcs 1 and 2.

I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
3039
Table 2. Absorption and fluorescence emission data for ZnPcs 1 and 2
in different solvents
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
A
Medium
Absorption kmax (nm)
Emission
kmax (nm)a
ZnPc 1
DMF
369
613
607
613d
611
680
688
758
746
—
Methanol
PBS
2% w/v SDSb
353
347c
367
671
—
677
—
676
683
755
ZnPc 2
DMF
374
351
340
368
612
605
629d
610
678
669
—
687
678
—
756
744
—
Methanol
PBS
2% w/v SDSb
300
400
500
600
700
800
675
683
753
wavelength (nm)
a kexc = 608 nm.
b 2% w/v in water.
c Shoulder.
B
d Broadening.
there is not aggregation of these ZnPcs in the systems.
A solvatochromic effect in pure solvents is observed on
the location of the absorption band showing a blue-shift
(ꢀ10 nm) upon solubilization in methanol. On the other
hand, these ZnPcs are aggregated in PBS, as it is typical
for many phthalocyanine derivatives.3,22 The cationic
ZnPc 2 shows a broadening band at 629 nm. However,
they are solubilized as monomers in 2% w/v SDS, which
was used as a medium to evaluate the amount of phtha-
locyanine bound to cells.
650
700
750
800
wavelength (nm)
C
The steady-state fluorescence emission spectra of ZnPcs
1 and 2 were performed in DMF (Fig. 1B). The spectra
show two bands in the red spectral region, which are
characteristic for similar Zn(II) phthalocyanines.21 By
comparison with ZnPc as a reference, the values of fluo-
rescence quantum yields (/F) were obtained in DMF.
Values of /F are shown in Table 2. Also, emission spec-
tra were obtained in different media (Table 2). As ex-
pected, no emission was observed in PBS where these
phthalocyanines are mainly aggregated. A small Stokes
shift (ꢀ7 nm) was observed, indicating that the spectro-
scopic energy is nearly identical to the relaxed energy of
the singlet state. Taking into account the energy of the
0–0 electronic transitions, the energy levels of the singlet
excited states (Es) were calculated (Table 1). These re-
sults are in agreement with those previously reported
for similar phthalocyanines in different media.22
300
400
500
600
700
wavelength (nm)
Figure 1. (A) Absorption spectra; (B) fluorescence emission spectra,
k
exc = 612 nm; and (C) fluorescence excitation spectra, kem = 760 nm,
of ZnPc 1 (- - -) and ZnPc 2 (—) in DMF.
The corrected fluorescence excitation spectra of the
ZnPcs 1 and 2 were measured in DMF (Fig. 1C), moni-
toring the emission at 760 nm. In both cases, the spectra
resemble the absorption spectra (Fig. 1A), indicating
that these sensitizers are essentially unaggregated in this
medium.
Table 1. Spectroscopic data in DMF and partition coefficients (P) of
ZnPcs 1 and 2
Phthalocyanine e (MÀ1 cmÀ1
)
/
Es (eV)
P
c
F
ZnPc 1
ZnPc 2
1.7 · 105a
0.23 0.01 1.81
0.22 0.01 1.81
1.6 0.5
0.88 0.1
1.1 · 105b
a 680 nm.
The fluorescence spectrum of ZnPc 2 was also analyzed
in biological medium. The E. coli cells were treated with
10 lM of sensitizer 2 for 30 min at 37 ꢀC in the dark.
After one washing step, a cell suspension in PBS
showed a shape close to that of the fluorescence emis-
sion in homogeneous media (Fig. 1B) with maxima at
686 and 755 nm. This result indicates that the binding
b 678 nm.
c /F(ZnPc) = 0.28.
studied in different media (Table 2). A sharp absorption
band was obtained in organic solvents indicating that

3040
I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
of s ZnPc 2 to E. coli cells leads to photosensitizer
disaggregation.16
0.5
0.4
0.3
0.2
0.1
0.0
A
On the other hand, the n-octanol/water partition coeffi-
cients (P = [phthalocyanine]o/[phthalocyanine]w) of
these phthalocyanines were evaluated at 25 ꢀC.3,15,23
The values of P obtained are reported in Table 1. As ex-
pected, the hydrophilic character increases in cationic
ZnPc 2 with respect to 1.19
2.3. Photosensitized decomposition of substrates
2.3.1. Photooxidation of 9,10-dimethylanthracene
(DMA). The aerobic irradiation of photosensitizers 1
and 2 in DMF was performed in the presence of
DMA. This substrate quenches O2(1Dg) exclusively by
chemical reaction.14 Typical semilogarithmic plots,
describing the progress of the reaction for DMA are
shown in Figure 2A. From these plots the values of
the observed rate constant (kobsDMA) were obtained
(Table 3). The quantum yields of O2(1Dg) production
(UD) were calculated comparing the slope for the
phthalocyanines with the corresponding slope obtained
for the reference, ZnPc (U = 0.56).24,25 As can be ob-
served in Table 3, a similar efficiency in the O2(1Dg) pro-
duction was found for ZnPc 1 and ZnPc in DMF, while
ZnPc 2 presents a slightly higher UD value. These are
quite reasonable values for Zn(II) phthalocyanines in
this solvent.21,24 However, the values of UD can signifi-
cantly change in a different medium, diminishing when
the sensitizer is partially aggregated.
0
50
100
150
200
250
irradiation time (s)
1.2
1.0
0.8
0.6
0.4
0.2
0.0
B
2.3.2. Decomposition of tryptopha (Trp). The amino acid
Trp represents a suitable substrate for testing the effi-
ciency of photosensitizing agents, since it is efficiently
photooxidized by both type I and type II reaction mech-
anisms.26 The photoprocess follows first-order kinetics
with respect to Trp concentration, as shown in Figure
0
100
200
300
400
500
irradiation time (s)
Figure 2. First-order plots for the photooxidation of (A) DMA
(35 lM) and (B) Try (25 lM) photosensitized by ZnPc 1 (.), ZnPc 2
(m), and ZnPc (d) in DMF; kirr = 590–800 nm. Values represent
means standard deviation of three separate experiments.
2B for [Trp] = 25 lM. From the plots in Figure 2B,
Trp
the values of the kobs
were calculated for ZnPcs 1
and 2. As can be observed in Table 2, the photooxida-
tion rate is about five times higher for cationic ZnPc 2
than that for 1 and ZnPc. Thus, these results are not
in agreement with those obtained for the DMA photo-
oxidation, indicating that the other photoreaction pro-
cess, probably an electron transfer process, is also
involved in the decomposition of Trp sensitized by ZnPc
2. In addition, the photosensitized oxidation of Trp was
found to be more than two orders of magnitude more
efficient via electron transfer than via O2(1Dg).27 Also,
electron transfer quenching of the sensitizer singlet ex-
cited state by Trp has been postulated.28 Since cationic
sensitizers can bind to these substrates, probably by elec-
trostatic attraction, an electron transfer pathway may
also be contributing together with Type II photoprocess
to Trp decomposition in DMF.5,16
2.4. Photobleaching of sensitizers
Photobleaching of ZnPcs was carried out under the
same irradiation conditions used for photoinactivation
of E. coli cultures (see Experimental). The decomposi-
tions were analyzed following the decrease in absorption
of the phthalocyanine Q-band. Figure 3A shows the
Table 3. Kinetic parameters of ZnPcs 1 and 2 in DMF
Parameter
ZnPc 1
ZnPc 2
ZnPc
DMA
kobs
a
(sÀ1
)
(1.6 0.1) · 10À2
0.56
(4.5 0.1) · 10À4
(2.9 0.2) · 10À4
(2.2 0.1) · 10À5
(1.7 0.1) · 10À2
0.59
(2.5 0.1) · 10À3
(3.6 0.2) · 10À4
(5.6 0.3) · 10À5
(1.6 0.1) · 10À2
0.56
(4.2 0.1) · 10À4
(2.6 0.2) · 10À4
(1.6 0.1) · 10À5
UD
kobs (sÀ1
)
Try
P
kobs (sÀ1
)
UP
a Ref. 25 UD(ZnPc) = 0.56.

I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
3041
changed incubating the cultures for 30 min. However,
the amount of cationic ZnPc 2 recovered after one wash-
ing step is considerably higher than noncationic ZnPc 1
and it reaches a value of ꢀ3 nmol/106 cells. Thus, the
presence of cationic N-methylpyridyl groups, on the
periphery of phthalocyanine macrocycle, considerably
increases the binding affinity of sensitizer 2 for E. coli
cells.
A
0.2
0.1
0.0
2.5.2. Photosensitized inactivation of E. coli. After the
treatment of E. coli cells with 10 lM of sensitizers 1
and 2 for 30 min at 37 ꢀC in the dark, the cultures were
irradiated with visible light after one washing step. The
survival curves at different light exposure levels are
shown in Figure 4B. As can be observed, the photoinac-
tivation activities of cationic ZnPc 2 are considerably
600
700
wavelength (nm)
B
0.25
0.20
0.15
0.10
0.05
0.00
4
A
3
2
1
0
0
200
400
600
800
irradiation time (s)
Figure 3. (A) Absorption spectra changes of ZnPc 2 irradiated with
visible light in DMF (Dt = 10 min). (B) First-order plots for the
photobleaching of ZnPc 1 (.), ZnPc 2 (m), and ZnPc (d) in DMF
irradiated with visible light. Values represent means standard
deviation of three separate experiments.
0
5
10
15
20
25
30
incubation time (min)
absorption spectra changes observed on photolysis of
ZnPc 2 in DMF. As can be observed, the diminishments
in the Q-bands are not accompanied by the formation of
a new band in this region. The photobleaching reactions
7
6
5
4
3
2
1
0
B
P
follow a first-order kinetic (Fig. 3B). The values of kobs
and the initial quantum yield of photobleaching (UP) are
shown in Table 3. The results of UP for ZnPcs 1 and 2
are similar to those obtained for other structurally com-
parable Zn(II) phthalocyanine derivatives.3,21 Under
these conditions, ZnPc 2 presents a higher efficiency of
photobleaching in DMF. However, the mechanism of
phthalocyanine photodegradation can be complex and
differs in solution and in biological media.5
2.5. Studies in vitro on E. coli cells
2.5.1. Binding of phthalocyanines to cells. The ability of
cationic ZnPcs 2 to bind to bacterial cells was compared
with that obtained for a noncationic ZnPc 1. The E. coli
cultures were incubated with phthalocyanine for differ-
ent times at 37 ꢀC in the dark. In each case, the sensitizer
associated with cells was determined by fluorescence
analysis (see Experimental). The results after one wash-
ing step are summarized in Figure 4A. Under these con-
ditions, these phthalocyanines are rapidly bound to E.
coli cells in 5 min and the binding is not appreciably
0
5
10
15
20
25
30
irradiation time (min)
Figure 4. (A) Amount of ZnPcs 1 (.) and 2 (.) recovered after one
washing step from E. coli cells treated with 10 lM of sensitizer for
different incubation times, at 37 ꢀC in the dark. (B) Survival curves of
E. coli incubated with 10 lM of ZnPcs 1 (.) and 2 (m) for 30 min at
37 ꢀC in the dark, washing once before illumination and exposed to
visible light for different irradiation times. Values represent
means standard deviation of three separate experiments.

3042
I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
higher than that found for ZnPc 1. Thus, the photody-
namic activity of ZnPc 2 produces a ꢀ4.5 log decrease
of E. coli cell survival, when the cultures are irradiated
for 30 min. This diminishing represents ꢀ99.997% of cell
inactivation. While, under these conditions a very low
diminishing in the cellular viability was found using
ZnPc 1 as a sensitizer. On the other hand, control experi-
ments showed that the viability of E. coli was unaffected
by illumination alone or by dark incubation with 10 lM
of the photosensitizer for 30 min, indicating that the cell
mortality obtained after irradiation of the cultures is
produced by the photosensitization effect of phthalo-
cyanines.
studies in vitro show that the efficiency of E. coli inactiv-
ation is appreciably higher for cationic ZnPc 2 than non-
charged ZnPc 1. An analogous behavior was previously
found using a cationic zinc pyridinium phthalocyanine
derivative in comparison with a neutral tetra-diethanol-
amine phthalocyanine and a negatively charged tetra-
sulfonated phthalocyanine.8 The insensitivity of Gram-
negative bacteria to photoinactivation was attributed
to the presence of the outer membrane, which endows
the surface of Gram-negative bacteria with a negative
charge.10,29,30 In these cases, the binding of the sensitizer
to E. coli cells appears to be necessary to exert its photo-
cytotoxic effect. Consequently, O2(1Dg) is produced in
the membrane microenvironment, where the agent is
localized. Thus, the photodynamic activity is mainly
associated with ZnPc 2, which are tighter bound to
cells.8,10 On the other hand, when the sensitizer remains
in the PBS, such as neutral or anionic agents, the photo-
dynamic effect is almost not observed. Therefore, non-
cationic phthalocyanines, such as ZnPc 1, are
unsuccessful sensitizers for Gram-negative bacteria,
unless that the outer membrane permeability is
altered.8,12,10 The results indicate that the failure of
ZnPc 1 to show any photosensitizing activity on the
bacteria is most probably due to its failure to bind to
them. These data confirm that cationic ZnPc 2 is an
efficient photosensitizer with potential applications in
bacteria inactivation by photodynamic treatment.
Also, growth delay of E. coli cultures was performed in
the medium (10% w/v TS broth) to ensure that photo-
sensitization was still possible when the cells were not
under starvation conditions or the potential damaging
effects of phosphate buffer washing.8 As can be observed
in Figure 5, under these conditions, growth was arrested
when E. coli was exposed to 10 lM of ZnPc 2. After
irradiation in the presence of sensitizer 2, the cells no
longer appeared to be growing as measured by turbidity
at 550 nm. As observed above, the effect of ZnPc 1 is
slower than the sensitizer 2. On the other hand, E. coli
cells exposed to 10 lM of sensitizers in the dark or not
treated with sensitizer and illuminated showed no
growth delay compared with controls. Therefore, the
data illustrate that the observed growth delay is due to
the photoinactivation effect of the sensitizer 2 on the
cells.8,10
3. Experimental
3.1. General
In conclusion, the spectroscopic and photodynamic
properties of both sensitizers 1 and 2 in homogeneous
medium are quite similar except for Trp, which presents
a higher photodecomposition rate when the cationic
phthalocyanine is used as a sensitizer. However, the
UV–visible and fluorescence spectra were recorded on a
Shimadzu UV-2401PC spectrometer and on a Spex Fluo-
roMax fluorometer, respectively. Proton nuclear mag-
netic resonance (1H NMR) spectra were recorded on a
Bruker ARX 300 multinuclear spectrometer at
300 MHz. Mass Spectra were taken with a Varian Matt
312 operating in EI mode at 70 eV and with a ZAB-SEQ
Micromass equipment. All the chemicals from Aldrich
(Milwaukee, WI, USA) were used without further puri-
fication. Sodium dodecyl sulfate (SDS) from Merck
(Darmstadt, Germany) and zinc phthalocyanine (ZnPc)
from Aldrich were used as received. Solvents (GR grade)
from Merck were distilled. Ultrapure water was ob-
tained from Labonco equipment model 90901-01. The
cultures were exposed for different time intervals to vis-
ible light. The light source used was a Novamat 130 AF
slide projector equipped with a 150 W lamp. The light
was filtered through a 2.5 cm glass cuvette filled with
water to absorb heat. Wavelength ranges between 350–
800 and 590–800 nm were selected by optical filters
(OG590 cut-off filter). The light intensity at the treat-
ment site was 30 mW/cm2 (k = 670 nm) (Radiometer
Laser Mate-Q, Coherent).
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
100
200
300
time (min)
Figure 5. Growth delay curves of E. coli incubated with 10 lM of
ZnPc 1 (.) and ZnPc 2 (m) and exposed to different irradiation times
with visible light in 10% w/v TS broth at 37 ꢀC. The arrow marks the
time point where the appropriate concentration of phthalocyanine was
3.2. Photosensitizers
According to the synthetic procedure, phthalocyanine
macrocycle is obtained as a mixture of the correspond-
ing regioisomers, which only one is named below.
added. Control culture untreated irradiated ( ), treated with 10 lM of
s
ZnPc 1 (,) and ZnPc 2 (n) and not irradiated. Values represent
means standard deviation of three separate experiments.

I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
3043
3.2.1. 4-(4-Pyridyloxy)phthalonitrile. A solution of 4-
nitrophthalonitrile (500 mg, 2.89 mmol) and 4-hydroxy-
pyridine (302 mg, 3.18 mmol) in 15 mL of DMF was
stirred for 10 min under argon atmosphere. Then, dry
potassium carbonate (2 g, 14.5 mmol) was added and
the mixture was heated at 80 ꢀC for 3 h. The mixture
was treated with water (50 mL) and extracted with
dichloromethane/methanol (5%) (two portions of
50 mL each). The solvent was removed under reduced
pressure. The product was purified by flash chromato-
graphy (silica gel, dichloromethane/methanol 10%) and
yielded 422 mg (66%) of the pure 4-(4-pyridyloxy)phtha-
lonitrile. TLC (silica gel) Rf (dichloromethane/methanol
taking into account the refractive index of the
solvents.32
3.4. Partition coefficient measurements
1-Octanol/water partition coefficients (P) were deter-
mined at 25 ꢀC using equal volumes of water (2 mL)
and 1-octanol (2 mL). Typically, a solution of each
phthalocyanine (ꢀ10 lM) was stirred in the thermostat
after the equilibrium was reached (ꢀ8 h). An aliquot
(100 lL) of aqueous and organic phases was dissolved
in 2 mL of DMF and the final phthalocyanine concen-
tration determined by absorption spectroscopy.15,23
1
10%) = 0.45. Melting point 126 ꢀC. H NMR (DMSO-
d6, TMS) d [ppm] 6.15 (d, 2H, J = 7.8 Hz), 8.00 (d,
2H, J = 7.8 Hz), 8.01 (dd, 1H, J = 2.2 Hz, 8.6 Hz), 8.18
(d, 1H, J = 8.6 Hz), 8.34 (d, 1H, J = 2.2). MS [m/z] 221
(M+) (221.0589 calculated for C13H7N3O). Anal. Calcd
C 70.58, H 3.19, N 19.00; found C 70.49, H 3.22, N
19.05.
3.5. Steady state photolysis
Solutions of either 9,10-dimethylanthracene (35 lM in
DMF, 2 mL) or L-tryptophan (Trp, 25 lM, 2 mL) and
photosensitizer (k = 680 nm, absorbance 0.1) in DMF
were irradiated in quartz cuvettes with 590–800 nm light.
The kinetics of photooxidation were studied by following
the decrease of the absorbance (A) at kmax = 378 nm for
DMA and the fluorescence intensity (F) at k = 340 nm for
Trp. The Trp fluorescence was excited by 290 nm light. A
control experiment showed that under these experimen-
tal conditions the fluorescence intensity is linearly corre-
lated with Trp concentration. The observed rate
constants (kobs) were obtained by a linear least-squares
fit of the semilogarithmic plot of LnA0/A versus time
or LnF0/F versus time. Photooxidation of DMA was
used to determine singlet molecular oxygen, O2(1Dg),
production by the photosensitizers.14,24 ZnPc was used
as the standard in DMF (UD = 0.56).24,25 Measurements
of the sample and reference under the same conditions
afforded UD for ZnPc 1 and 2 by direct comparison of
the slopes in the linear region of the plots. Sensitizer
photobleaching was evaluated irradiating a phthalocya-
nine solution in DMF (k = 680 nm, absorbance 0.2) with
visible light in the same conditions described below for
the treatment of E. coli cultures. Quantum yields of
photobleaching (UP) were calculated as [initial rate of
disappearance of phthalocyanine]/[initial rate of absorp-
tion of photons by the reaction mixture].33–35 All the
experiments were performed at 25.0 0.5 ꢀC. The
pooled standard deviation of the kinetic data, using dif-
ferent prepared samples, was less than 5%.
3.2.2. Zinc(II) 2,9,16,23-tetrakis(4-pyridyloxy)phthalocya-
nine 1 (ZnPc 1). A solution of 4-(4-pyridyloxy)phthalo-
nitrile (400 mg, 1.81 mmol) and zinc(II) acetate
dihydrate (110 mg, 0.50 mmol) in 15 mL of n-pentanol
was stirred for 10 min under argon atmosphere. Then,
300 lL of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU,
2.01 mmol) was added and the mixture was refluxed
for 18 h. The reaction mixture was cooled to room tem-
perature and precipitated with 50 mL of cyclohexane.
The solid was filtered and washed with water and hex-
anes. The solid was recrystallized from methanol/water
to yield 297 mg (68%) of ZnPc 1. Melting point
1
>300 ꢀC. H NMR (DMSO-d6, TMS) d [ppm] 6.2–6.8
(m, 8H), 8.0–8.3 (m, 8H), 8.1–9.4 (br, 12H). MS [m/z]
948 (M+) (948.1648 calculated for C52H28N12O4Zn).
Anal. Calcd C 65.73, H 2.97, N 17.69; found C 65.64,
H 3.02, N 17.60.
3.2.3. Zinc(II) 2,9,16,23-tetrakis[4-(N-methylpyridyloxy)]-
phthalocyanine iodide (ZnPc 2). A mixture of ZnPc 1
(240 mg, 0.25 mmol) and 20 mL of methyl iodide in
10 mL of acetone was stirred for 48 h at reflux. The sol-
vents were removed under vacuum. The solid was re-
suspended in cyclohexane and filtered. The solid was
washed with hexanes, diethylether, and acetone, to yield
1
360 mg (95%) of ZnPc 2. Melting point >300 ꢀC. H
NMR (DMSO-d6, TMS) d [ppm] 4.2–4.3 (m, 12H,
–CH3), 8.0–9.5 (m, 28H). MS [m/z] 1516 (M+)
(1515.8766 calculated for C56H40I4N12O4Zn). Anal.
Calcd C 44.31, H 2.66, N 11.07; found C 44.38, H
2.75, N 11.15.
3.6. Bacterial strain and preparation of cultures
E. coli strain (EC7) recovered from clinical urogenital
material was used. The strain was resistant to ampicillin
and sulfamethoxazol–trimethoprim. E. coli strain was
grown aerobically at 37 ꢀC in 30% w/v tryptic soy (TS)
broth overnight. Aliquots (ꢀ40 lL) of the culture were
aseptically transferred to 4 mL of fresh medium (30%
w/v TS broth) and incubated at 37 ꢀC to mid logarithmic
phase (absorbance ꢀ0.6 at 660 nm). Cells in the loga-
rithmic phase of growth were harvested by centrifuga-
tion of broth cultures (3000g for 15 min), washed once
with 10 mM of PBS and re-suspended in 4 mL of PBS.
Then, the cells were diluted 1/1000 in PBS, correspond-
ing to ꢀ106 colony forming units (CFU)/mL. In all the
experiments, 2 mL of the cell suspensions in Pyrex brand
3.3. Spectroscopic studies
Absorption spectra were recorded at 25.0 0.5 ꢀC
using 1 cm path length cells. The fluorescence quantum
yields (/F) of phthalocyanines were calculated by com-
parison of the area below the corrected emission spec-
trum with that of ZnPc as a fluorescence standard,
exciting at kex = 608 nm.31 A value of /F = 0.28 for
ZnPc in DMF was calculated by comparison with the
fluorescence spectrum in pyridine using /F = 0.30 and

3044
I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
´
culture tubes (13 · 100 mm) was used and the sensitizer
was added from a stock solution of phthalocyanine
(3.0 · 10À3 M) in DMF. Viable bacteria were monitored
and their number calculated by counting the number of
colony forming units after appropriated dilution on agar
plates.8 Bacterial cultures grown under the same condi-
tions and light exposures, but without addition of any
photosensitizer, served as controls.
tina, Fundacion Antorchas and SECYT Universidad
´
Nacional de Rıo Cuarto for financial support. E.N.D.
is a Scientific Member of CONICET.
References and notes
1. Villanueva, A.; Vidania, R.; Stockert, J. C.; Can˜ete, M.;
Juarrans, A. Photodynamic effects on cultured tumor cells.
Cytoskeleton alterations and cell death mechanisms. In
Photobiology; Nalwa, H. S., Ed.; Handbook of Photo-
chemistry and Photobiology; American Scientific, 2002;
Vol. 4, Chapter 3, pp 79–117.
2. Ackroyd, R.; Kelty, C.; Brown, N.; Reed, M. The history
of photodetection and photodynamic therapy. Photochem.
Photobiol. 2001, 74, 656–669.
3. Bonnett, R. Chemical Aspects of Photodynamic Therapy;
Gordon and Breach Science: Amsterdam, 2000.
4. DeRosa, M. C.; Crutchley, R. J. Photosensitized singlet
oxygen and its applications. Coord. Chem. Rev. 2002, 233–
234, 351–371.
5. Ochsner, M. Photophysical and photobiological processes
in photodynamic therapy of tumours. J. Photochem.
Photobiol. B: Biol. 1997, 39, 1–18.
3.7. Phthalocyanine binding to bacteria cells
Suspensions of E. coli (2 mL, ꢀ106 CFU/mL) in PBS
were incubated in the dark at 37 ꢀC with 10 lM of sen-
sitizer for different times. The cultures were centrifuged
(3000g for 15 min) and then washed with PBS. The cell
pellets obtained by centrifugation (3000g for 15 min)
were re-suspended in 2% aqueous SDS (2 mL), incu-
bated overnight at 4 ꢀC and sonicated for 30 min. The
concentration of sensitizer in the supernatant was mea-
sured by spectrofluorimetry (kexc = 670 nm, kem
=
684 nm) in solution of 2% SDS in PBS. The fluorescence
values obtained from each sample were referred to the
total number of bacteria contained in the suspension.
The concentration of the phthalocyanine in this sample
was estimated by comparison with a calibration curve
obtained with standard solutions of the sensitizer in
2% w/v SDS ([sensitizer]ꢀ0.05–0.5 lM).
6. Taylor, P. W.; Stapleton, P. D.; Luzio, J. P. New ways to
treat bacterial infections. Drug Discovery Today 2002, 7,
1086–1091.
7. Hamblin, M. R.; Hasan, T. Photodynamic therapy: a new
antimicrobial approach to infectious disease? Photochem.
Photobiol. Sci. 2004, 3, 436–450.
8. Minnock, A.; Vernon, D. I.; Schofield, J.; Griffiths, J.;
Parish, J. H.; Brown, S. B. Photoinactivation of bacteria.
Use of a cationic water-soluble zinc phthalocyanines to
photoinactivate both Gram-negative and Gram-positive
bacteria. J. Photochem. Photobiol. B: Biol. 1996, 32, 159–
164.
9. Wood, S.; Nattress, B.; Kirkham, J.; Shore, R.; Brookes,
S.; Griffiths, J.; Robinson, C. An in vitro study of the use
of photodynamic therapy for the treatment of natural oral
plaque biofilms formed in vivo. J. Photochem. Photobiol.
B: Biol. 1999, 50, 1–7.
10. Minnock, A.; Vernon, D. I.; Schofield, J.; Griffiths, J.;
Parish, J. H.; Brown, S. B. Mechanism of uptake of a
cationic water-soluble pyridinium zinc phthalocyanine
across the outer membrane of Escherichia coli. Antimicrob.
Agents Chemother. 2000, 44, 522–527.
11. Segalla, A.; Borsarelli, C. D.; Braslavsky, S. E.; Spikes, J.
D.; Roncucci, G.; Dei, D.; Chiti, G.; Jori, G.; Reddi, E.
Photophysical, photochemical and antibacterial photosen-
sitizing properties of a novel octacationic Zn (II)-phtha-
locyanine. Photochem. Photobiol. Sci. 2002, 1, 641–648.
12. Lacey, J. A.; Phillips, D. The photosensitization of
Escherichia coli using disulphonated aluminium phthalo-
cyanines. J. Photochem. Photobiol. A: Chem. 2001, 142,
145–150.
13. Milanesio, M. E.; Moran, M. S.; Yslas, E. I.; Alvarez, M.
G.; Rivarola, V.; Durantini, E. N. Synthesis and biological
evaluation of methoxyphenyl porphyrin derivatives as
potential photodynamic agents. Bioorg. Med. Chem. 2001,
9, 1943–1949.
14. Milanesio, M. E.; Alvarez, M. G.; Yslas, E. I.; Borsarelli,
C. D.; Silber, J. J.; Rivarola, V.; Durantini, E. N.
Photodynamic studies of metallo 5,10,15,20-tetrakis-
(4-methoxyphenyl) porphyrin: photochemical character-
ization and biological consequences in a human carcinoma
cell line. Photochem. Photobiol. 2001, 74, 14–21.
15. Scalise, I.; Durantini, E. N. Photodynamic effect of
metallo 5-(4-carboxyphenyl)-10,15,20-tris(4-methylphen-
yl) porphyrins in biomimetic AOT reverse micelles
3.8. Photosensitized inactivation of bacteria cells
Cell suspensions of E. coli (2 mL, ꢀ106 CFU/mL) in
PBS were incubated with 10 lM of phthalocyanine for
30 min in the dark at 37 ꢀC. After that, the cultures were
washed once with PBS and re-suspended in 2 mL of
PBS. The cultures were exposed for different time inter-
vals to visible light (350–800 nm). Control experiments
were carried out without illumination in the absence
and in the presence of sensitizer. Control and irradiated
cell suspensions were serially diluted with PBS, each
solution was plated in triplicate on TS agar and the
number of colonies formed after 18–24 h incubation at
37 ꢀC was counted. Each experiment was repeated sepa-
rately three times.
3.9. Growth delay experiment
E. coli cells were grown overnight as described above. A
portion (60 lL) of this culture was transferred to 20 mL
of fresh TS broth (10%) medium. The suspension was
homogenized and aliquots of 2 mL were incubated at
37 ꢀC. The culture grown was measured by turbidity at
550 nm using a Tuner SP-830 spectrophotometer.8
When the cultures reached the log phase (absorbance
ꢀ0.1), 10 lM of sensitizer was added. Then, the flasks
were irradiated with visible light as described above.
Samples were also taken to determine the viability of
the cells as described earlier.
Acknowledgments
The authors are grateful to Consejo Nacional de Inves-
´ ´
tigaciones Cientıficas y Tecnicas (CONICET) of Argen-

I. Scalise, E. N. Durantini / Bioorg. Med. Chem. 13 (2005) 3037–3045
3045
containing urease. J. Photochem. Photobiol. A: Chem.
2004, 162, 105–113.
25. Seotsanyana-Mokhosi, I.; Kuznetsova, N.; Nyokong, T.
Photochemical of tetra-2,3-pyridinoporphyrazines. J. Pho-
tochem. Photobiol. A: Chem. 2001, 140, 215–222.
26. Foote, C. S. Definition of type I and type II photosensi-
tized oxidation. Photochem. Photobiol. 1991, 54, 659–
660.
27. Wessels, J. M.; Foote, C. S.; Ford, W. E.; Rodgers, M. A.
J. Photooxidation of tryptophan: O2(1Dg) versus electron-
transfer pathway. Photochem. Photobiol. 1997, 65, 96–102.
16. Milanesio, M. E.; Alvarez, M. G.; Silber, J. J.; Rivarola,
V.; Durantini, E. N. Photodynamic activity of monocati-
onic and non-charged methoxyphenylporphyrin deriva-
tives in homogeneous and biological media. Photochem.
Photobiol. Sci. 2003, 2, 926–933.
17. Lazzeri, D.; Durantini, E. N. Synthesis of meso-substi-
tuted cationic porphyrins as potential photodynamic
agents. ARKIVOC (ARKAT USA) 2003, 10, 227–239.
18. Chan, W.-S.; Zuk, M.; Ben-Hur, E. Phthalocyanines. In
Photodynamic Tumor Therapy, 2nd and 3rd Generation
Photosensitizers; Moser, J. G., Ed.; Harwood Academic:
Amsterdam, 1998; Chapter 2.4, pp 63–73.
19. Wo¨hrle, D.; Iskander, N.; Graschew, G.; Sinn, H.;
Friedrich, E. A.; Maier-Borst, W.; Stern, J.; Schlag, P.
Synthesis of positively charged phthalocyanines and their
activity in the photodynamic therapy of cancer cells.
Photochem. Photobiol. 1990, 51, 351–356.
20. Ng, A. C. H.; Ng, D. K. P. Synthesis and photophysical
properties of nonaggregated phthalocyanines bearing
dendritic substituents. Macromolecules 1999, 32, 5292–
5298.
´
´
28. Daraio, M. E.; Vo¨lker, A.; Aramendıa, P. A.; San Roman,
E. Tryptophan quenching of zinc-phthalocyanine and
porphycene fluorescence in micellar CTAC. Photochem.
Photobiol. 1998, 67, 371–377.
29. Nicaido, H.; Vaara, M. Molecular basic of bacterial outer
membrane permeability. Microbiol. Rev. 1985, 49, 1–32.
30. Vaara, M. Agents that increase the permeability of outer
membrane. Microbiol. Rev. 1992, 56, 395–411.
31. Demas, J. N.; Crosby, G. A. The measurement of
photoluminescence quantum yields. J. Phys. Chem. 1971,
75, 991–1024.
32. Seybold, P. G.; Gouterman, M. Porphyrins XIII: fluores-
cence spectra and quantum yields. J. Mol. Spectrosc. 1969,
31, 1–13.
21. Ogunsipe, A.; Maree, D.; Nyokong, T. Solvent effect on
the photochemical and fluorescence properties of zinc
phthalocyanine derivatives. J. Mol. Struct. 2003, 650, 131–
140.
33. Spikes, J. D. Quantum yields and kinetics of the photo-
bleaching of hematoporphyrin, photofrin II, tetra(4-sulfo-
natophenyl)-porphine and uroporphyrin. Photochem.
Photobiol. 1992, 55, 792–808.
22. Marti, C.; Nonell, S.; Nicolau, M.; Torres, T. Photophys-
ical properties of neutral and cationic tetrapyridinopor-
phyrazines. Photochem. Photobiol. 2000, 71, 53–59.
23. Indig, G. L.; Anderson, G. S.; Nichols, M. G.; Bartlelett,
J. A.; Mellon, W. S.; Sieber, F. Effect of molecular
structure on the performance of triarylmethane dyes as
therapeutic agents for photochemical purging of autolo-
gous bone marrow grafts from residual tumor cells. J.
Pharmaceutical. Sci. 2000, 89, 88–99.
34. Bezdetnaya, L.; Zeghari, N.; Belitchenko, I.; Barberi-
Heyob, M.; Merlin, J.-L.; Potapenko, A.; Guillemin, F.
Spectroscopic and biological testing of photobleaching of
porphyrins in solutions. Photochem. Photobiol. 1996, 64,
382–386.
35. Hadjur, C.; Lange, N.; Rebstein, J.; Monnier, P.; van den
`
Bergh, H.; Wagnieres, G. J. Spectroscopic studies of
photobleaching and photoproduct formation of meta-
(tetrahydroxyphenyl)chlorin (m-THPC) used in photody-
namic therapy. The production of singlet oxygen by m-
THPC. J. Photochem. Photobiol. B: Biol. 1998, 45, 170–
178.
24. Redmond, R. W.; Gamlin, J. N. A compilation of singlet
yields from biologically relevant molecules. Photochem.
Photobiol. 1999, 70, 391–475.