Pc 4 Conjugated Iron Oxide Nanoparticles as Photodynamic Therapeutic Carriers
Photodynamic therapy (PDT) is getting more and more popular because of its low invasiveness and low toxicity. It is used in treatment of e.g. acne, malignant cancer or arteriosclerosis by inducing cell apoptosis or necrosis in the target tissue. PDT utilizes the property of certain photosensitizers to produce highly reactive singlet oxygen through illumination at non-toxic wavelengths, which renders the final therapeutic drug. Phthalocyanines are such photosensitizers that absorb in the far red spectrum of visible light and therefore are a promising tool in PDT[2,3]. These fluorescent dyes can also be used for imaging. The silicon based phthalocyanine Pc 4 is a second-generation photosensitizer for PDT with a maximum absorption at 675 nm and has been successfully applied in PDT[4,5].
Also very popular is the loading of carrier particles with drugs for their delivery. Beside many other carriers, dendrimers, polymeric micelles or inorganic nanoparticles have been designed for this purpose. Inorganic magnetic nanoparticles provide both the additional possibility of magnetic resonance imaging and the possibility of manipulation by external magnetic fields as long as their magnetic properties are strong enough. The requirements of magnetic particle properties that are needed for in vivo applications in biomedicine are manifold: i) the magnetic material must be non-toxic; ii) the particles must be superparamagnetic (ferromagnetic particles would aggregate immediately); iii) but should exhibit saturation magnetizations that allow for the manipulation by ferromagnetic materials, e.g. to remove them almost quantitatively from the body; iv) and they should show a narrow size-distribution to avoid undesired interparticle interaction; v) furthermore, the colloids have to be soluble in physiological solutions of high salt concentration (colloid stability); vi) the particles surface coating must offer functional groups that allow for further manipulation in terms of coupling them to specific compounds of interest, e.g. PEG, proteins, antibodies, phthalocyanines, etc., by well-known conjugation techniques (e.g. 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimid- (EDC) or disuccinimidyl carbonate (DSC)-coupling). The only magnetic materials that have been published to be non-toxic to our knowledge are iron oxides (Fe3O4, ϒ-Fe2O3) and manganese ferrite (MnFe2O4). The most promising syntheses concerning monodispersity, crystallinity, and non-aggregation are done in high boiling hydrophobic solvents by thermal decomposition of fatty acid salts of iron in the presence of excess fatty acids as stabilizing and size controlling surfactants[9-11]. As those particles are hydrophobic and not soluble in water, their surface- groups have to be further replaced by hydrophilic molecules to provide hydrophilicity.
In this study we report the preparation of aqueous based Pc 4-conjugated 15 nm magnetite particles in four steps: 1: synthesis of hydrophobic particles; 2: ligand exchange, i.e. replacing oleic acid by aspartic acid (ASA); 3: derivatization of Pc 4; 4: conjugation of derivatized Pc 4 to ASA-coated magnetite nanoparticles. Furthermore, cell-killing studies under LED illumination at 670 nm showed high efficacy without toxic effects observed in the dark. A sketch of the overall procedure is shown in the graphical abstract. For an enlarged TOC image see S6.
The synthesis of hydrophobic, monodisperse magnetite nanoparticles of controlled size, morphology, and crystallinity has been up-scaled and optimized to a gram-scale (yield after work-up: 3.6 g, ~70%). The particles are superparamagnetic with typical values of saturation magnetization as continuously aobserved in numerous studies[12,13]. TEM-images of these particles are presented in Figure 1. High monodispersity and crystallinity is obviously. No aggregation is seen. It has been published that this synthesis route produces particles containing mainly magnetite and only a small fraction of maghemite (noted as “FeOx”).
Figure 1. TEM-images of as-prepared, hydrophobic, monodisperse particles at different magnifications. The main particle diameter is 15 nm. The high crystallinity can be seen on the HRTEM image with inserted electron diffraction pattern.
The water-transfer of these particles was successfully done by ligand exchange with aspartic acid (ASA), providing amine groups at the outer particle surface that allow for the successful conjugation of the phthalocyanine Pc 4. It is not easy to derivatize Pc 4. The only idea was to activate the Pc-Si-OH group by DSC as is published for the activation of R-C-OH groups followed by conjugation to amino groups . Thus, we needed a strategy to make the particles water soluble and place amino groups around in the same time. As we had good results with exchanging oleic acid by dimercaptosuccinic acid we tried to do the same with aspartic acid. The particles were soluble in water after some modification in presence of ASA while they were not before. Moreover, the aqueous solution had a glance (total missing of eye-visible scattering) that we have never seen before in water based iron oxide colloids and TEM investigation (Figure 2) showed no aggregation. There are also no visible changes in size and morphology compared to Figure 1.
Phthalocyanine-conjugation: prior to conjugation optical, spectra of Pc 4 dissolved in DMSO and water had been taken. Figure 3 (top) provides a typical result. A spectrum of Pc 4 conjugated to FeOx in water is also shown. No further blue-shift after conjugation is seen indicating non-aggregation onto the surface of the nanoparticles. The absorbance of Pc 4 in water shows only a slight red-shift to 677 nm but remarkable quenching.
Figure 3. UV-Vis spectra of (top, from top to bottom) pure 10 μM Pc 4 in DMSO, immediately after dissolving in water (10 μM), and 0.3 μM conjugated to FeOx in water. Bottom: singlet-oxygen measurement of pure 2.5 μM Pc 4 in water.
Aqueous Pc 4 solutions have been prepared in different concentrations and volumes. These solutions contain a very smallamount of DMSO and are only stable for several hours (see experimental part for details).
Additionally, singlet oxygen phosphorescence of its transition between the 1⧠+g and 1⧠g configuration at 1270 nm formedthrough illumination of Pc 4 dissolved in oxygen saturated water was investigated (the kinetics of this transition is dependenton oxygen concentration), showing singlet oxygen formation (Figure 3, bottom). 4,4’-benzophenone in acetonitrile and erythrosine blue in water were measured as positive control. We expect that the observed blue-shift of Pc 4 dissolved in DMSO (Figure 3 top) compared to Pc 4 dissolved in water is not a result of aggregation, but related to differences in permittivity of the solvents.
ASA-coated particles have been designed for the attachment of Pc 4. The OH-group of Pc4 can be activated by disuccinimidylcarbonate (DSC) in presence of diisopropylethyl amine (DIPEA) leading to NHS-ester activated Pc 4, Figure 4, that can be conjugated to amine groups. The activated Pc-Si-OH structure is an estimation based on the published structure of DSCactivated R-C-OH groups. This reaction needs absolute water-free conditions. Simply drying of DMSO by molecular sieves is not enough and DSC only reacts with residual water. This problem could be solved by adding DSC in excess. An image of as prepared activated Pc 4- and diluted particle-dye conjugate solutions is provided in the SI (Figure S1).
Figure 4. Left: structure of Pc 4. Right: expected structure of DSC activated Pc 4.
Finally, as proof of principle, the ability of Pc 4-conjugated magnetite nanoparticles to induce cell killing was studied in a RAW macrophage cell line. Figure 5 shows results of cell-killing studies after 2 h (A) and 24 h (B) of nanoparticle incubation. The IC50-values are 0.03 ng/μl Pc 4, i.e. 1.9 ng/μl FeOx and 0.23 ng/μl Pc 4, i.e. 14 ng/μl FeOx, respectively. No dark toxicity was observed with either time points (Figures 5A and 5B). The reason for the decreased efficacy after an increase in the incubation time could be, e.g. particle agglomeration with time or intracellular processing of nanoparticles. More experiments will be needed to clarify this observation.
A protocol for the preparation of aqueous based Pc 4-conjugated magnetite nanoparticles has been developed as a tool for the photodynamic therapy of arteriosclerotic heart disease. The major challenges of our work were the ligand exchange with ASA, the derivatization of the Pc 4 dye and its conjugation while avoiding dye-aggregation on top of the nanoparticles surface and preserving water-solubility of the final colloidal solution. Moreover, the type of activation of Pc 4 is unprecedented for silicon based phthalocyanines, and solves a difficult synthetic challenge in this particular case, i.e. it might open a way to the derivatization of these important Pc derivatives.The solutions remain stable for at least some weeks up to 1.5 years, the procedure could be reproduced 6 times, and cell-killing experiments with the final conjugates under illumination at 670 nm showed very high cell killing efficacy, while no toxic effect could be observed in the dark.
Figure 5. The relative viability of RAW cells after 2h (A) and 24 h (B) incubation.
Further experiments to analyze the real structure of activated Pc 4 are planned for the future and will be published in a second article.
All chemicals were from Sigma-Aldrich. Dimethyl sulfoxide, pyridine and dichloromethane were dried where required. All others have been used as received.
Magnetite particles: the synthesis procedure started with dissolving 60 mmol iron(III)acetylacetonate and 180 mmol oleic acid in 400 ml octadecene in a 1000 ml three-necked flask. The second stage consisted of heating up to 300°C while nitrogen is bubbled through the solution first, with a distillation-condenser to evaporate side-products and second, continued with further heating up to 320°C (bp) with a reflux-condenser while keeping under reflux for 60 min. After cooling down below 100°C the particles were separated by a permanent magnet after the addition of three times excess 2-butanol as the non-solvent, dissolved in toluene by ultrasonification, separated again magnetically after addition of the non-solvent and finally dissolved in 50 ml toluene or chloroform for storage, i.e. 65 mg/ml. The particle concentration has been calculated by totally drying out 1 ml solution after three washing steps with acetone and assuming a monolayer of oleic acid with a space requirement of 0.4 nm2 per one oleic acid molecule.
Transfer to water: 150 μl triethylamine, 3 ml water and 10 mg aminosuccinic acid (ASA) dissolved in 2 ml dimethylsulfoxide (DMSO) were mixed with 10 mg particles in 5 ml chloroform and sonicated for 4.5 h. The water was exchanged every 45 min to avoid temperatures above 40°C. The final magnetically separated particles were dissolved either in 2 ml DMSO or 2 ml deionized water. This exchange is not very efficient. Only 60% of the particles could be collected as a well dispersed non-aggregated fraction after 30 min ultrasonification.
Pc 4-synthesis: The synthesis of Pc 4 was carried out adapting procedures from the literature. In particular, the reaction between silicon phthalocyanine (SiPc) dihydroxide (SiPc(OH)2) and 3-(methoxydimethylsilyl)-N,N-dimethylpropan- 1-amine in distilling pyridine yields a SiPc derivative with two dimethylaminopropylsiloxy axial substituents. The selective displacement of one of these substituents with trichloroacetate, and the subsequent substitution of the latter with an OH ligand, allows obtaining Pc 4 in an overall yield of 70% (see SI for details, Scheme S1, Figures S2-S4).
Pc 4-conjugation: briefly, Pc 4, DSC and DIPEA were dissolved in 5 ml DMSO in a molar ratio of 1:3:6 and incubated overnight resulting in activated Pc 4. Pc 4-activation was done following a protocol describing R-C-OH activation of meso-dimercaptosuccinic acid coated ZnS nanoparticles. Without any workup this solution was mixed with an aqueous particle solution in a molar ratio of 120 Pc 4 molecules per amine functionalized magnetite particles (calculated by the FeOx surface area divided by the Pc 4 size plus a small distance between the dyes) and incubated overnight resulting in Pc 4 conjugated magnetic nanoparticles. (Potentially remaining excess DSC is immediately reacting with water in this case.) It was found that up to 140 molecules could be quantitatively attached to one particle without observing any Pc 4-aggregation. The conjugated particles were cleaned by magnetic separation 3 times and finally dissolved in much less volumes of pure water in concentrations up to 4.88 mg/ml.
Cell-killing studies: Mouse macrophage RAW264.7 (ATCC; TIB- 71) cell lines were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10 % FBS, 1% penicillin- streptomycin. The efficacy of Pc 4-particle conjugates was analyzed using MTS-assay (Promega). In this assay, cells are first split into a 96-well plate. One day later, conjugates are applied to the cells, and allowed to incubate on the cells for 1-24h. After selected time, cells were illuminated with a custom-made LED device (Figure S5) emitting light at 670 nm wavelength for 10 minutes, with LED light intensity of 3.04 mW/cm2. 24 hours later MTS-reagent is applied to the plate, and measured one hour later by spectrophotometry (Multiskan Ascent, MTX Labsystems). Cells without conjugates and/or without light activation were always used as a control, as well as the control solutions in which conjugates were made.
TEM: The particles have been investigated on a Philips CM12 at 120 kV. Samples in organic solutions have been prepared by simply dropping 10 μl from 1 cm above onto a carbon coated copper grid. Aqueous based samples were prepared by carefully placing a 10 μl drop onto a carbon coated copper grid for 15 min particle adsorption followed by drying with filter paper.
UV/VIS: Spectra have been taken with standard single beam spectrometer in 1 cm cuvettes and 2 mm slit. Samples were prepared from a 0,1 M stock solution in DMSO by dilution either in DMSO or pure water between 100 and 400 times. Singlet oxygen: Samples have been excited with a high pressure xenon lamp at 683 nm and detected at 1274 nm with PicoQuant FluoTime 300 and oxygen detector Hamamatsu NIR PMT. Erythrosin blue in water and 4,4’-benzophenone in acetonitrile were measured additionally as a positive control.
The authours greatfully acknowledge funding from FP7- NMP “CosmoPHOS-nano” under grant agreement no.310337. M.G. and J.D.R. also acknowledge financial support by UMO- 2016/23/B/NZ7/01288 grant from National Science Centre (Poland)
1. Chen James, Keltner Llew, Christophersen Julenea, Zheng Frank, Krouse Michaelc, Singhal Anil et al. New Technology for Deep Light Distribution in Tissue for Phototherapy. Cancer J. 2002, 8(2): 154-163.
3. S. I. Ogura, K. Tabata, K. Fukushima, T. Kamachi, I. Okura, J. Porphyr. Phthalocya 2006, 10, 1116-1124.
4. JD Miller, ED Baron, H Scull, A Hsia, JC Berlin, T McCormick et al. Photodynamic therapy with the phthalocyanine photosensitizer Pc 4: the case experience with preclinical mechanistic and early clinical-translational studies. Appl. Pharm. 2007, 224(3): 290-299.
5. Baron ED, Malbasa CL, Santo-Domingo D, Fu P, Miller JD, Hanneman KK et al. Silicon phthalocyanine (Pc 4) photodynamic therapy is a safe modality for cutaneous neoplasms: results of a phase 1 clinical trial. Laser Surg. Med. 2010, 42, 888-895.
8. UI Tromsdorf, NC Bigall, MG Kaul, OT Bruns, MS Nikolic, B Mollwitz et al. Size and Surface Effects on the MRI Relaxivity of Manganese Ferrite Nanoparticle Contrast Agents. Nano Lett. 2007, 7(8): 2422-2427.
13. C Martínez-Boubeta, K Simeonidis, M Angelakeris, N Pazos-Pérez, M Giersig, A Delimitis et al. Critical radius for exchange bias in naturally oxidized Fe nanoparticles. Phys Rev B. 2006, 74, 054430(10).
b) United States Patent 1998, Nº: 5763602.