Protoporphyrin IX Nanoparticle Carrier - American Chemical Society

12534

Langmuir 2008, 24, 12534-12538

Protoporphyrin IX Nanoparticle Carrier: Preparation, Optical Properties, and Singlet Oxygen Generation Liane M. Rossi,* Paulo R. Silva, Lucas L. R. Vono, Adjaci U. Fernandes, Dayane B. Tada, and Maurı´cio S. Baptista Institute of Chemistry, UniVersity of Sa˜o Paulo, USP, Sa˜o Paulo 05508-000, SP, Brazil ReceiVed March 17, 2008. ReVised Manuscript ReceiVed August 14, 2008 The present study is focused on developing a nanoparticle carrier for the photosensitizer protoporphyrin IX for use in photodynamic therapy. The entrapment of protoporphyrin IX (Pp IX) in silica spheres was achieved by modification of Pp IX molecules with an organosilane reagent. The immobilized drug preserved its optical properties and the capacity to generate singlet oxygen, which was detected by a direct method from its characteristic phosphorescence decay curve at near-infrared and by a chemical method using 1,3-diphenylisobenzofuran to trap singlet oxygen. The lifetime of singlet oxygen when a suspension of Pp IX-loaded particles in acetonitrile was excited at 532 nm was determined as 52 µs, which is in good agreement with the value determined for methylene blue in acetonitrile solution under the same conditions. The Pp IX-loaded silica particles have an efficiency of singlet oxygen generation (η∆) higher than the quantum yield of free porphyrins. This high efficiency of singlet oxygen generation was attributed to changes on the monomer-dimer equilibrium after photosentisizer immobilization.

* Corresponding author. Phone: +55 11 30912181. Fax: +55 11 38155579. E-mail: [email protected].

porphyrinic ring into a chlorin ring. This process produces BPDM (Visudyne), another derivative of Pp IX which is found on the market for use in PDT. Some of these drugs are currently approved by the Food and Drug Administration for the treatment of superficial cancers. However, there are still many problems for the clinical application of existing photosensitizers, such as low purity, because they are a complex mixture of several partially unidentified porphyrins and have poor selectivity for tumor tissue, low extinction coefficients that require the administration of relatively large amounts of drug to obtain a satisfactory phototherapeutic response, and finally, high accumulation rate in skin, which induces a prolonged cutaneous light sensitivity lasting for up to weeks after PDT treatment, obliging patients to stay out of the sunlight to avoid a severe sunburn reaction. Thus, the formulation of an injectable Pp IX product with satisfactory stability in aqueous solutions and which is easily targeted to the desired site of action is of great interest. In recent years, an increasing number of researchers have considered the possibility of using nanoparticles in photodynamic therapy (PDT).7-10 Successful loading of a chlorin derivative in silica nanoparticles has been reported to give nanoparticle platforms that only deliver singlet oxygen for PDT while conserving the photosensitizer effect.11 It was shown that nanoparticles can accumulate at the tumor site due to enhanced endocytotic activity and leaky vasculature in the tumors.12 Also, nanoparticles can be modified superficially using special targeting moieties (such as antibodies) for site-specific action. Encapsulation of drugs in silica nanoparticles is very promising because silica surfaces are easy to functionalize, nontoxic, chemically inert, and optically transparent. Besides protecting the drug from external interferences, the silica matrix should improve the drug

(1) (a) Bonnett, R. Chem. Soc. ReV 1995, 24, 19. (b) Dougherty, T. J. Photochem. Photobiol. 1987, 45, 879. (2) Snyder, J. W.; Skovsen, E.; Lambert, J. D. C.; Ogilby, P. R. J. Am. Chem. Soc. 2005, 127, 14558. (3) Xiao, Z.; Tamimi, Y.; Brown, K.; Tulip, J.; Moore, R. Urol. Oncol. 2002, 7, 125. (4) Lottner, C.; Bart, K. C.; Bernhardt, G.; Brunner, H. J. Med. Chem. 2002, 45, 2064. (5) Bonnett, R.; Gabriel, M. Tetrahedron 2001, 57, 9513. (6) Savellano, M. D.; Hasan, T. Photochem. Photobiol. 2003, 77, 431.

(7) DeRosa, M. C.; Crutchley, R. J. Coord. Chem. ReV. 2002, 351, 233–234. (8) Zeitouni, N. C.; Oseroff, A. R.; Shieh, S. Mol. Immunol. 2003, 39, 1133. (9) Wang, S.; Gao, R.; Zhao, F.; Selke, M. J. Mater. Chem. 2004, 14, 487. (10) Tada, D. B.; Vono, L. L. R.; Duarte, E. L.; Itri, R.; Kiyohara, P. K.; Baptista, M. S.; Rossi, L. M. Langmuir 2007, 23, 8194. (11) Yan, F.; Kopelman, R. Photochem. Photobiol. 2003, 78, 587. (12) (a) Konan, Y. N.; Gurny, R.; Alle´mann, E. J. Photochem. Photobiol., B: Biol. 2002, 66, 89. (b) Roby, A.; Erdogan, S.; Torchilin, V. P. Eur. J. Pharm. Biopharm. 2006, 62, 235.

Introduction Protoporphyrin IX (Pp IX) is a naturally occurring porphyrin constituent of hemoglobin, cytochrome c and other biologically relevant molecules. It is used as a drug in photodynamic therapy (PDT) but its direct application is limited due to its aggregation and low solubility in a physiological medium. PDT is based on the concept that photosensitizer molecules generate singlet oxygen upon irradiation, which acts as the primary cytotoxic agent responsible for irreversible damage of the treated tissues.1 In contrast with other conventional medical treatments, in PDT it is unnecessary to release the loaded drug, but the diffusion of surrounding oxygen molecules and the release of the produced active oxygen species must be sufficient for a therapeutic effect.2 The indirect application of Pp IX is made possible by δ-aminolevulinic acid in chloridric acid solution (ALA · HCl). ALA is a pro-drug, a precursor for all in vivo porphyrins, which is cycled by biosynthesis generating tissue Pp IX.3 Modifications to the Pp IX molecule have been made to increase the solubility of this potential drug in a physiological medium and stimulate its use in PDT. An important strategy is to explore the reactivity of vinyl groups to functionalize Pp IX with polar groups. To this end hydroxyl groups have been inserted in the vinylic positions 31 and 81 to produce hematoporphyrin IX. The hematoporphyrin IX, in both acid and basic media, polymerizes to form a mixture of chromophores named Photofrin, which is the most used drug in PDT. Further functionalization of carboxylic groups has also been used for the preparation of several derivatives.4,5 Another strategy is to functionalize the vinylic group by means of a Diels-Alder reaction,6 transforming the

10.1021/la800840k CCC: $40.75  2008 American Chemical Society Published on Web 10/04/2008

Protoporphyrin IX Nanoparticle Carrier

selectivity toward cancer cells while reducing its toxicity toward normal tissues, as has been observed in the hyperpermeability of tumor microvessels to large molecules.13 The pore cutoff size of several tumor models has been reported to range between 300 and 700 nm.14 The preparation of silica particles loaded with organic molecules can be difficult due to the highly hydrophobic nature of organic molecules compared to the hydrophilic surface of silica. One strategy used to prepare rhodamine and fluoresceindoped silica spheres is to make dye molecules water-soluble through linking with a hydrophilic dextran molecule.15 During tetraethoxysilane (TEOS) polymerization the dye molecules are doped inside the silica matrix. However, dye leakage from the silica matrix to aqueous solution has been observed. Another strategy is to modify the organic molecules with organosilane reagent in a strategy that assures chemical attachment of the drug to the inorganic framework during polymerization with TEOS.16 The main focus of the present study is the synthesis and purification of an organosilane porphyrin to develop a nanoparticulate carrier for Pp IX prepared through a modified sol-gel approach with the silyl-modified porphyrin. The entrapment of the drug in a nanoparticle-based system is investigated as an approach to the use of Pp IX as a PDT photosensitizer. The photophysical and photochemical properties of the prepared nanoparticles were also investigated.

Materials and Methods Materials. All organic chemicals and solvents were of reagent grade. Dimethylformamide, (3-aminopropyl)triethoxysilane (APTES), and oxalyl chloride were distilled prior to use. Preparation of Silyl-Functionalized Pp IX. First, 100 mg of Pp IX (1.65 × 10-4 mol) were refluxed under a dry nitrogen atmosphere with an excess of freshly distilled oxalyl chloride (2 mL). The deep purple solution obtained was refluxed with stirring for 30 min, followed by evaporation of the excess of oxalyl chloride to give Pp IX dichloride as a purple film.17 The purple residue was stirred with an excess of freshly distilled APTES (1 mL) for 2 h, under a dry nitrogen atmosphere to give dimethyl-8,13-divinyl-3,7,12,17-tetramethyl-21H,23H-porphine-2,18-dipropyl-amidepropyltriethoxysilane. The product was purified by distillation of the excess APTES and characterized by IR and NMR. IR: CdO 1632; 1H NMR (500 MHz CDCl3): δH 9.44; 9.25; 9.00 and 8.76 (4s, 4H, CH; 5, 10, 15, and 20), 7.71 and 7.59 (2m, 2H, CH; 31 and 81), 5.96-5.86 and 5.73-5.60 (2m, 4H, CH2, 32 and 82), 3.67-3.66 (m, 12H, SiOCH2CH3), 3.49 (4s,12H, CH3: 21, 71, 121, and 181), 4.17-4.12 (4H; CH2; 131 and 171), 2.76 (NCH2CH2CH2SI), 1.65 (NCH2CH2CH2SI), 1.06 (m, 18H, SiOCH2CH3), 3.36-3.31 (4H, CH2; 132 and 172), 0.56 (NCH2CH2CH2SI), -5.53 (s, 2H, H-pyrol). 13C NMR (500 MHz CDCl ): 173.02 (CON); 130.17 and 129.97 3 (CH, 31 and 81), 120.40 and 120.05 (CH2, 32 and 82), 96.71; 96.54; 96.34 and 96.30 (CH; 5, 10, 15, and 20), 58.39 (6C, SiOCH2CH3), 43.91 and 42.93 (NCH2CH2CH2SI), 35.30 and 35.13 (CH2; 132 and 172), 25.18 and 25.14 (NCH2CH2CH2SI), 22.46 and 22.05 (2C, CH2; 131 and 171), 18.26 (6C, SiOCH2CH3), 12.32; 12.16; 11.09 and 9.49 (CH3: 21, 71, 121, and 181), 8.65 and 7.55 (NCH2CH2CH2SI). Preparation of Pp IX-Loaded Silica Nanoparticles. Silica particles were prepared by adding 200 µL of silyl-functionalized Pp IX to an ethanol solution (100 mL) containing ammonium hydroxide (6 mL) and tetraethylorthosilicate (TEOS) (4.2 mL) under stirring. The mixture was stirred overnight. The nanospheres were isolated (13) Hobbs, S. K.; Monsky, W. L.; Yuan, F.; Roberts, W. G.; Griffith, L.; Torchilin, V. P.; Jain, R. K. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 4607. (14) Brigger, I.; Dubernet, C.; Couvreur, P. AdV. Drug DeliVery ReV. 2002, 54, 631. (15) Zhao, X.; Bagwe, R. P.; Tan, W. AdV. Mater. 2004, 16, 173. (16) Collinson, M. M. Trends Anal. Chem. 2002, 21, 30. (17) Taima, H.; Okubo, A.; Yoshioka, N.; Hidenari Inoue, H. Tetrahedron Lett. 2005, 46, 4161.

Langmuir, Vol. 24, No. 21, 2008 12535 by centrifugation (7000 rpm, 10 min) and washed three times with ethanol, two times with water, and again with ethanol. Instrumentation and Methods. Transmission Electron Microscopy (TEM) Analysis. The morphology of the nanoparticles obtained was characterized on a Philips CM 200 microscope operating at an accelerating voltage of 200 kV. The samples for TEM were prepared by dispersion of the nanoparticles in aqueous solution at room temperature and then collected on a carbon-coated copper grid. Direct Determination of Singlet Oxygen. The singlet oxygen lifetime was determined from phosphorescence decay curves at nearinfrared (NIR). Data were recorded with a time-resolved NIR fluorimeter (Edinburgh Analytical Instruments) equipped with Nd:YAG laser (Continum Surelite III) for sample excitation at 532 nm. The emitted light was passed through a silicon and an interference filter and a monochromator before being detected with a NIRPhotomultiplier (Hamamatsu Co. R5509), according to a methodology reported elsewhere.18 Singlet oxygen lifetime was determined by applying first-order exponential fitting to the curve of the phosphorescence decay at 1270 nm using Origin 7.0 software. Indirect Determination of Singlet Oxygen. 1,3-Diphenylisobenzofuran (DBPF), obtained from Acros Organics, was used to determine the release of singlet oxygen to the solution by the particles at 16 °C in acetonitrile. The samples were immediately prepared before use by transferring 40 µL of DBPF stock solution (8 mM) to 2 mL of a suspension of Pp IX-loaded particles or methylene blue (MB) free solution (used as standard), with the same absorption at 532 nm, in a quartz cuvette in the dark. MB was chosen as the standard because of the high solublility in acetonitrile, whereas the porphyrins are much less soluble and can thus aggregate and consequently decrease quantun yield in solution. The absorption of both solutions (MB and Pp IX-loaded particles in acetonitrile) were normalized to the same values at 532 nm, assuming that we can calculate the real absorption of the photosensitizer immobilized in the particles by measuring the absorption spectrum of a suspension of Pp IX particles and subtracting the baseline scattering by multiple point level baseline correction. The experiments were carried out by irradiating samples with a 532 nm laser beam provided by Morgotron whereas absorption spectra were obtained after certain time intervals with a Shimadzu UV-2401PC spectrophotometer. The absorption intensities at different wavelengths near the DPBF maximum absorption were plotted against irradiation time, and the time for the DBPF absorption decrease was calculated by applying first-order exponential fitting to the curve obtained using Origin 7.0 software. The time for the decrease in absorption of DPBF (t, in seconds) is inversely proportional to its reaction rate with singlet oxygen (k, in seconds-1), and the relationship expressed in eq 1 provides a measure of its singlet oxygen delivery efficiency (η∆particle). A MB acetonitrile solution was used as the standard and the relation given in eq 1 was used to calculate the release of singlet oxygen by the particles,

η∆particle ) φMB

tMB tparticle

(1)

where tMB is the time for the decrease in absorption of DPBF in the presence of MB-free in acetonitrile solution adjusted to a first-order exponential decay, tparticle is the time for the decrease in absorption of DPBF in the presence of Pp IX-loaded particles in acetonitrile adjusted to a first-order exponential decay, and φMB is the singlet oxygen quantum yield of MB-free in acetonitrile solution. Determination of Singlet Oxygen Quantum Yield of MB (ΦMB). The absolute value of the singlet oxygen quantum yield in acetonitrile was determined by the photochemical DPBF method reported by Spiller et al.19 The samples were prepared in dry acetonitrile (18) (a) Junqueira, H. C.; Severino, D.; Dias, L. G.; Gugliotti, M.; Baptista, M. S. Phys. Chem. Chem. Phys. 2002, 4, 2320. (b) Severino, D.; Junqueira, H. C.; Gabrielli, D. S.; Gugliotti, M.; Baptista, M. S. Photochem. Photobiol. 2003, 77, 459. (19) Spiller, W.; Kliesch, H.; Wohrele, D.; Hackbarth, S.; Roder, B.; Schnurpfeil, G. J. Porphyrin Phthalocyanines 1998, 2, 145.

12536 Langmuir, Vol. 24, No. 21, 2008

Rossi et al.

immediately before use by transferring 40 µL of DBPF stock solution (3 mM) to 2 mL of a MB solution in a quartz cuvette in the dark. The experiments were carried out by irradiating the samples, cooling (18 °C), and constantly stirring, with a 532 nm laser beam adjusted to light intensity of 5.4 × 1016 photons · cm-2 · s-1, whereas absorption spectra were obtained after each irradiation cycle (2 s) with a Shimadzu UV-2401PC spectrophotometer. The absorption intensities at 411 nm were used to calculate the concentration of DPBF after each cycle (ε ) 22500 M-1 · cm-1 in acetonitrile20). The quantum yield, ΦDPBF, for each irradiation cycle was calculated using equations given in ref 19. The value 1/ΦMB is the intercept obtained from the Stern-Volmer plot (1/ΦDPBF versus 1/[DPBF]). The experiment was repeated five times to give the value of ΦMB ) 0.6 ( 0.04 in acetonitrile.

Results and Discussion The entrapment of protoporphyrin IX in silica nanospheres was achieved by modification of Pp IX molecules with an organosilane reagent as shown in Scheme 1. The first step was the formation of the more reactive oxacyl chloride followed by condensation with (3-aminopropyl)triethoxysilane. The silylfunctionalized Pp IX was characterized by IR, 1H NMR, and 13C NMR. The IR spectrum showed an additional band at 1632 cm-1, which corresponds to the axial deformation band of CdO, characteristic of Pp IX amide derivatives. The 1H NMR spectrum showed the characteristic peaks of this family of compounds, in good agreement with other previously reported porphyrin compounds, and a series of peaks characteristic of the new group inserted. The peaks at δ 2.76, 1.65, and 0.56 ppm are coupled (according to COSY spectrum) and the shifts correspond to the δ, β, and γ positions of the inserted groups with respect to the amide group. The peaks at δ 3.67-3.66 and 1.06 ppm correspond to the CH2 and CH3 groups of the six ethoxy groups bound to Si, respectively. The endocyclic 1H of the porphyrin ring, usually found between δ -3 and -4 ppm, shifted to -5.53 ppm after the porphyrin modification. The 13C NMR spectrum also changed according to the additional groups. The peaks at δ 43.91 and 43.93 ppm correspond to two C atoms of the inserted group in R position with respect to the amides. The peaks at δ 25.18 and 25.14 correspond to the β position and at δ 8.65 and 7.55 to the γ position. The carbon atoms of the ethoxy groups appeared at δ 58.39 and 18.26 corresponding to CH2 and CH3, respectively. The chromophore was also characterized by UV-vis and fluorescence showing the same pattern characteristic of other known porphyrin molecules with the absorption maxima corresponding to the Soret band near 400 nm and the four Q bands between 500 and 620 nm. As shown in Figure 1, the maximum absorption of hematoporphyrin IX (HpIX) appears at 395 nm, Scheme 1. Synthesis of Dimethyl-8,13divinyl-3,7,12,17-tetramethyl-21H,23H-porphine-2,18-dipropylamidepropyltriethoxysilane (silyl-Pp IX)

Figure 1. Electronic absorption spectra of dichloromethane solutions of (a) hematoprophyrin IX, (b) protoporphyrin IX, and (c) silylfunctionalized protoporphyrin IX.

and the maximum corresponding to Pp IX at 398 nm was shifted to 405 nm upon silyl functionalization. The position of the Soret band is very sensitive to the aggregation state of the porphyrins. The red shift is in agreement with the decrease in aggregation of the silyl-Pp IX compared with that of hematoporphyrin and Pp IX in dichloromethane.21,22 The silyl-functionalized Pp IX molecule was used for the synthesis of monodisperse silica spheres through a modification of the process developed by Sto¨ber et al.23 and recently reported by us,24 which resulted in organo-SiO2 spheres with the drug distributed homogeneously in the inorganic matrix. When Pp IX molecules without modification were used to synthesize silica spheres using the same sol-gel Sto¨ber method, most Pp IX molecules were found to be in the aqueous phase, resulting in silica spheres with undetectable amounts of drug in the UV-vis and fluorescence measurements. Transmission electron microscopy (TEM) images of the Pp IX-loaded silica revealed nearly spherical particles (Figure 2), with the presence of some aggregates. Analysis of the TEM micrographs, by measuring the diameter of around 200 randomly selected particles in enlarged TEM images, allowed the calculation of the mean particle diameter of 77 ( 12 nm. The photophysical and photochemical properties of the Pp IX-loaded silica spheres were investigated. The absorption and emission spectra of a suspension of Pp IX-loaded silica spheres and of silyl-Pp IX free in dichloromethane solution are shown in Figure 3. The absorption maxima corresponding to the Soret band (ca. 400 nm) and the Q bands (in the range of 500-650 nm) observed in the porphyrin absorption spectrum (Figure 3a) were also observed in a suspension of Pp IX-loaded silica spheres (Figure 3b), although the bands are overlapped by the light scattering of the solid particles, which is responsible for spectral broadening. Similar behavior was found in aqueous solution as shown in the inset of Figure 3. The emission bands of the Pp IX-loaded silica particles in aqueous solution (λem. ) 615 and 676 nm) were slightly shifted to the blue region compared with the emission spectrum of silyl-Pp IX free in aqueous solution (λem. ) 620 and 679 nm) (20) Engelmann, F. M.; Mayer, I.; Araki, K.; Toma, H. E.; Baptista, M. S.; Maeda, H.; Osuka, A.; Furuta, H. J. Photochem. Photobiol., A: Chem. 2004, 163, 403. (21) Hutener, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525. (22) Borissevitch, I. E.; Tominaga, T. T.; Imasato, H.; Tabak, M. J. Lumin. 1996, 69, 65. (23) Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62. (24) Rossi, L. M.; Shi, L.; Quina, F. H.; Rosenzweig, Z. Langmuir 2005, 21, 4277.

Protoporphyrin IX Nanoparticle Carrier

Figure 2. Transmission electron microscopy image of Pp IX-loaded silica spheres.

Figure 3. Absorption spectra of silyl-Pp IX in dichloromethane solution (a) and of a suspension of Pp IX-loaded silica spheres in dichloromethane (b). Inset: Absorption spectra of silyl-Pp IX in aqueous solution (a′) and of a suspension of Pp IX-loaded silica spheres in aqueous solution (b′).

Langmuir, Vol. 24, No. 21, 2008 12537

oxygen, which diffuses through the pores, leading to the formation of singlet oxygen, which can then diffuse out of the porous matrix to produce a cytotoxic effect in tumor cells. Singlet oxygen generation was evidenced by the characteristic phosphorescence spectra with a maximum at 1270 nm obtained by monitoring the intensity of phosphorescence at several wavelengths after photoexcitation of a suspension of Pp IXloaded silica spheres (Figure 5 inset). The singlet oxygen lifetime of 52 µs was determined from the curve of phosphorescence decay at 1270 nm (Figure 5). This value was found to be similar to the lifetime of singlet oxygen generated irradiating a solution containing methylene blue, a very well-studied photosensitizer, prepared with the same solvent and under the same conditions of the photosensitizer under study. Both values seems to be lower than the well-established lifetime of singlet oxygen in acetonitrile around 70-80 µs.25 However, it is known that methylene blue does not suppress 1O2,18 so the fact that both solutions gave the same value may suggest that this decrease has no relation to the presence of the silica matrix in Pp IX-loaded silica spheres, but the presence of trace amounts of water in the solvent should be the reason for the lower value of the lifetime of singlet oxygen measured in acetonitrile. To better understand if singlet oxygen quenching by the silica matrix is an issue in our system, we ran three experiments to detect the emission spectra of singlet oxygen generated by hematoporphyrin IX in acetonitrile solution, followed by the addition of two portions of bare silica spheres. The results showed no changes in the emission intensity at 1270 nm for Hp IX acetonitrile solution before (I1270nm ) 569 ( 4) and after the addition of the first and second portion of silica to the solution containing the photosensitizer (I1270nm ) 568 ( 5 and I1270nm ) 574 ( 8, respectively). The delivery of 1O2 to the solution has been estimated indirectly using 1,3-diphenylisobenzofuran (DPBF) as a singlet oxygen chemical probe in acetonitrile.19 DPBF reacts irreversibly with 1O generated by photoexcitation of a sensitizer and the reaction 2 can be easily followed spectrophotometrically by recording the decrease in the intensity of the DPBF absorption at around 400 nm. The changes in the absorption spectra of DPBF with time upon irradiation in the presence of Pp IX-loaded silica spheres can be seen in Figure 6. Note the absence of direct reactions between DPBF and Pp IX since the Pp IX absorption remains unchanged during the experiment. To make sure that the absorption decrease of DPBF is induced by singlet oxygen, control

Figure 4. Emission spectra of silyl-Pp IX in aqueous solution (a) and of Pp IX-loaded silica spheres in aqueous solution (b) for the samples excited at 500 nm.

(Figure 4). The latter was very difficult to measure because of the instability of the silyl-Pp IX in water and subsequent precipitation. We expected the silica matrixes to be porous enough to allow the entrapped photosensitizing drug to interact with molecular

Figure 5. Emission transient at 1270 nm obtained after photoexcitation at 532 nm of a suspension of Pp IX-loaded silica spheres in acetonitrile. Inset: Emission spectrum of singlet oxygen generated by a suspension of Pp IX-loaded silica spheres in acetonitrile.

12538 Langmuir, Vol. 24, No. 21, 2008

Figure 6. Absorption spectra of DPBF in the presence of Pp IX-loaded silica spheres in acetonitrile after different times of irradiation with a 532 nm laser beam. Inset: (a, b) Decay curve of the absorption of DPBF at 424 nm in the presence of Pp IX-loaded silica spheres as a function of irradiation time. (b, 0) Control 1: Absorption of DPBF at 424 nm as a function of time of irradiation at 532 nm. (c, /) Control 2: Absorption of silyl-Pp IX at 424 nm as a function of time of irradiation at 532 nm.

tests under exactly the same experimental conditions were done and the results are shown in the inset of Figure 6. Irradiation at 532 nm over 80 s causes no change either in the absorption of DPBF in the absence of Pp IX molecules or in the absorption of silyl-Pp IX free in solution. Therefore, DPBF is stable under 532 nm irradiation and Pp IX is not oxidized by the singlet oxygen generated upon irradiation (inset of Figure 6, curves b and c) and the observed decay curve (inset of Figure 6, curve a) could be caused only by the simultaneous presence of photosensitizer, DPBF, and the irradiation. The decay curves of the absorption of DPBF at 424 nm as a function of time upon irradiation in the presence of Pp IX-loaded silica spheres, and a MB acetonitrile solution adjusted to the same absorption at 532 nm (used as standard), were fitted to give the decay time of absorption of DPBF caused by the photoexcitation of a suspension of Pp IX particles (64 s) and by the photoexcitation of MB (101 s). One can argue that it is possible to compare our MB standard solution (no absorption in 400 nm) with a photosensitizerloaded silica suspension (with absorption in this region caused by light scattering) without causing error in the DPBF decay time measurements. The decay times of the absorption of DPBF at different wavelengths (380, 390, 410, 424, and 440 nm) in the MB solution were obtained with an error of 0.3% when calculated in those different wavelengths, and in the Pp IX-loaded silica spheres suspension the error was 2.6%. The higher error found in the Pp IX -loaded silica spheres suspension is probably due to the pronounced light scattering (about 1 unit of absorbance difference was found from 380 to 440 nm), but still very accept-

Rossi et al.

able under our experimental conditions (the DPBF decay time calculated was 64.4 s at 380 nm and 63 s at 440 nm). The efficiency of singlet oxygen delivery (η∆) of Pp IX-loaded silica particles was calculated using eq 1 and the absolute value of the single oxygen quantum yield of MB in acetonitrile determined by the photochemical DPBF methodology (experimental session), Φ∆ ) 0.6 ( 0.04, to give a η∆ value of 0.9 ( 0.06 for the porphyrin immobilized in silica spheres. We trust that this value is accurate because we did not observe considerable effect of scattering in the absorption measurments of DPBF decay time and we did not observe any effect of scattering in the NIR emission measurments as described above. This value is higher than the quantum yield of free protoporphyrin solution (Φ∆ ) 0.77 in CCl4; Φ∆ ) 0.60 in aqueous/TX100 solution),26 which indicates that the singlet oxygen generated by the photosensitizer immobilized in the silica matrix diffuses efficiently through the solution. The increase in the quantum efficiency may be a consequence of a decrease in the monomer-dimer equilibrium because the drug is firmly attached to the silica matrix, while in solution such equilibrium is possible and should cause a decrease in the 1O2 quantum yield. This result is in line with the results shown in Figure 1 which suggest a decrease in aggregation of the silyl-functionalized Pp IX compared with that of Pp IX free in solution. It is also known that the aggregation reduces the quantum yield and the lifetime of the excited triplet state of porphyrins and consequently should reduce the 1O2 generation yield.27,28 We have successfully prepared Pp IX nanoparticle carriers using a methodology that permits the photosensitizer to be firmly attached to the silica matrix through previous silyl modification of the porphyrin molecules. Although the photosensitizer is entrapped in the silica matrix, it can be excited by irradiation to generate 1O2 that diffuses to the solution with an efficiency of singlet oxygen delivery higher than the quantum yield of the free photosensitizer. This means that the immobilization of the Pp IX molecules increased the potential of the photosensitizer to perform PDT. Moreover, silica is a very versatile support material, structurally more stable, biocompatible, and chemically more resistant than the organic materials, offering ease of surface modification and biotargeting for site-specific delivery. The methodology here reported is very general and can be applied to several other drugs for the preparation of nanoparticles as drug carriers. Acknowledgment. The authors are grateful to the Brazilian agencies FAPESP and CNPq for financial support. We also thank Prof. Pedro K. Kiyohara (IF-USP) for TEM images. LA800840K (25) Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1995, 24, 663. (26) Redmond, R. W.; Gamlin, J. N. Photochem. Photobiol. 1999, 70, 391. (27) Aggarwal, L. P. F.; Baptista, M. S.; Borissevitch, I. E. J. Photochem. Photobiol., A: Chem. 2007, 186, 187. (28) (a) Borissevitch, I. E.; Tominaga, T. T.; Schmitt, C. C. J. Photochem. Photobiol., A: Chem. 1998, 114, 201. (b) Borissevitch, I. E.; Gandini, S. C. M. J. Photochem. Photobiol., B: Biol. 1998, 43, 112.