M.Q. Mesquita et al. / Dyes and Pigments xxx (2014) 1e11
5
2.10. Photosensitization procedure
3. Results
Bacterial cultures grown overnight were ten-fold diluted in
phosphate buffered saline (PBS), pH 7.4, to reach a final concen-
tration of w107 CFU mLꢀ1. Then, the bacterial suspension was
equally distributed in 100 mL acid-washed and sterilized glass
beakers (10 mL in each beaker). In the experiments performed with
free chlorin 3, an appropriate quantity of chlorin was added to
achieve a final concentration of 20 mM. Two controls were included
in each irradiation experiment; a light control (LC) submitted to the
same irradiation protocol as the samples but without PS and the
3.1. Synthesis and characterization of photosensitizer
Briefly, 5,10,15,20-tetrakis(pentafluorophenyl)porphyrin 1 was
obtained by condensation of pentafluorobenzaldehyde with pyr-
role in a refluxing mixture of acetic acid and nitrobenzene [31].
Chlorin 2 was obtained from the reaction of 1 with para-
formaldehyde and N-methylglycine and their cationization with an
excess of iodomethane in toluene at 40 ꢁC for 24 h afforded the
cationic chlorin 3 [30,32,33] (Fig. 1).
dark control (DC) containing the PS at concentration of 20
covered with aluminium foil.
mM and
The synthetic routes to materials A and B are shown in Schemes
and 2. The grafting of chlorin 2 on the modified 3-
1
In the experiments with materials A and B, seven samples were
prepared for each material. To the bacterial suspension diluted in
PBS, the amount of insoluble material to achieve final concentra-
bromopropylsilica and Merrifield resin was carried out in 1,2-
dichlorobenzene and in toluene at reflux, respectively. After 48 h
and 72 h, TLC of each reaction mixtures showed that most of the
starting chlorin 2 was converted into new green-coloured materials
that remain in the baseline. The resulting solids were filtered and
washed with the appropriate solvent in order to remove the re-
sidual unbound chlorin. The relative amount of chlorin 2 in the two
materials A1 and B1 was ca. 1.6% and 4.3% (w/w), respectively.
Considering that an increase in the number of positive charges
could improve the phototoxicity of the obtained materials towards
Gram-negative bacterium E. coli, materials A1 and B1 were subse-
quently treated with 1-methylimidazole and pyridine affording
materials A2, A3, B2 and B3, respectively. In order to check the
toxicity and phototoxicity of the starting silica (A) and the Merri-
field resin (B), the supports (without the addition of chlorin 2) were
also treated with 1-methylimidazole and pyridine under the same
experimental conditions used to obtain A5e6/B5e6. A concise
identification of all the prepared materials is represented in Table 1.
The UVevis spectra of the chlorin 3 and materials A1eA3 and
B1eB3 show the typical features of chlorins: a Soret band at ca.
400 nm and an intense absorption peak at ca. 650 nm, as exem-
plified in Fig. 3 for chlorin 3 and materials A1 and B1. These spectral
features indicate the success of the immobilization process and the
preservation of the photophysical features of the PS.
tions of 5.0
mM, 10
m
M and 20
mM with respect to PS in the case of
material A, and final concentrations of 10
m
M, 100 M and 200 mM
m
with respect to PS in the case of material B were added. In addition
to light and dark controls, two other controls were prepared, cor-
responding to the light and dark control of materials A (A4eA6) and
B (B4eB6). Light controls of supports were prepared by adding
equivalent amounts of support A (4e6) and B (4e6), respectively to
assays performed with A1e3 and B1e3, and exposed to light. The
dark controls were prepared in the same way but were kept in the
dark. Before each assay, the immobilized PS materials were soni-
cated for 45 min to material dispersion.
The samples were incubated for 10 min in the dark under
100 rpm stirring (25 ꢁC) to promote the PS binding to E. coli cells.
Then, samples were exposed in parallel to white light (PAR radia-
tion, 13 OSRAM 20 lamps of 18 W each, 380e700 nm) with an
irradiance of 40 W mꢀ2, for 180 min, under 100 rpm stirring (me-
chanical stirring when using chlorin 3 and magnetic stirring when
immobilized PS were used). As the recombinant bioluminescent
E. coli emits light at temperatures below 30 ꢁC, the beakers were
placed in a water bath in order to keep the samples at a constant
temperature (25 ꢁC). Sample aliquots were collected at time 0 and
after 30, 60, 90, 120, 150 and 180 min of exposure and the biolu-
minescence signal was measured in triplicate in the luminometer
(TD-20/20 Luminometer, Turner Designs, Inc., USA). Three inde-
pendent assays were performed for all experiments.
In addition, the EDX analysis shows the presence of fluorine in
materials A1 and B1 that also corroborates the successful immo-
bilization of chlorin 2 on the selected supports. The silica support
appears as particles with irregular shape and dimensions between
2 and 40 mm that decreased to below 30 mm after the immobili-
zation of the chlorin 2 (Fig. 4). However, the immobilization of the
chlorin did not affect the spheroidal shape and the size of the
2.11. Reuse of the materials
Merrifield resin, which was 100 ꢃ 14
mm as seen by SEM, nor after a
After each PI assay the materials A1e3/B1e3 were removed
from the bacterial suspension by vacuum filtration through a filter
paper NM 615 (MachereyeNagel) and subjected to successive
washing with small volumes (5 ꢂ 20 mL) of different solvents:
water, CHCl3/MeOH (15%), CHCl3 and petroleum ether in order to
remove possible bacterial residues and the remaining bacteria.
Materials were then dried in the oven at 50 ꢁC overnight and
subjected to a new cycle of PDI. Light and dark controls were
included in each irradiation experiment and were similar to the
ones obtained for the photodynamic procedure. Three independent
assays were performed for all experiments.
photosensitized treatment (Fig. 5). After the treatment with 1-
methylimidazole and pyridine, the EDX spectra of the samples
A2, B2 and A3, B3 indicate the presence of nitrogen, which is in
agreement with the attachment of these bases to the silica and
Merrifield resin supports.
3.2. Singlet oxygen (1O2) generating capacity
In order to evaluate the potentiality of the new materials for the
photodynamic inactivation of bacteria, their capacity to generate
singlet oxygen was estimated qualitatively using 1,3-
diphenylisobenzofuran (DPiBF) as 1O2 indicator [36]. The yellow
DPiBF (with a maximum wavelength at 415 nm) reacts with 1O2 in a
[4 þ 2] cycloaddition affording a colourless o-dibenzoylbenzene.
The decay of DPiBF absorption band at 415 nm is proportional to the
formation of 1O2. Chlorin 3 is a high 1O2 generator, causing a total
decay of DPiBF absorption within 15 min, when irradiated with
light at an irradiance of 9.0 mW cmꢀ2 (Fig. 6). This result corrob-
orates the direct measurements of phosphorescence of 1O2 in DMF
generated by chlorin 3 (FD ¼ 0.71) [30].
2.12. Statistical analysis
Statistics were performed by using SPSS (SPSS 15.0 for Win-
dows, SPSS Inc., USA). Normal distributions were assessed by Kol-
mogoroveSmirnov test. The significance among materials on
bacterial inactivation was assessed by one-way analysis of variance
(one-way ANOVA) model with the Bonferroni post-hoc test. A value
of P < 0.05 was considered significant.
Please cite this article in press as: Mesquita MQ, et al., Pyrrolidine-fused chlorin photosensitizer immobilized on solid supports for the