Journal of Photochemistry and Photobiology B: Biology 101 (2010) 340–347
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology B: Biology journal homepage: www.elsevier.com/locate/jphotobiol
Photophysical evaluation of mTHPC-loaded HSA nanoparticles as novel PDT delivery systems Kuan Chen a, Matthias Wacker b, Steffen Hackbarth a, Carmen Ludwig a, Klaus Langer c, Beate Röder a,⇑ a
Institute of Physics, Humboldt Universität zu Berlin, Newtonstraße 15, 12489 Berlin, Germany Institute of Pharmaceutical Technology, Goethe-Universität, Max-von-Laue-Str. 9 60438, Frankfurt am Main, Germany c Institute of Pharmaceutical Technology and Biopharmacy, Westfälische Wilhelms-Universität Münster, Corrensstr. 1, 48149 Münster, Germany b
a r t i c l e
i n f o
Article history: Received 17 December 2009 Received in revised form 18 July 2010 Accepted 4 August 2010 Available online 17 August 2010 Keywords: Photodynamic Therapy (PDT) Nanoparticles mTHPC HSA Singlet oxygen
a b s t r a c t Controlled drug release is one of the main goals of recent developments in drug carrier systems. In this work human serum albumin (HSA) nanoparticles as carriers for 5-, 10-, 15-, 20-Tetrakis (3-hydroxyphenyl)-chlorin (mTHPC) were investigated. The photophysical properties of mTHPC–HSA nanoparticles in dependence of loading ratio and level of HSA cross-linking were determined. Further the drug release after uptake by Jurkat cells and in vitro singlet oxygen kinetics were examined. The loading ratio of the mTHPC–HSA nanoparticles turned out to be of major importance for the PDT relevant electronic parameters in solution. Therefore, only HSA nanoparticles with low mTHPC-loading ratio generate singlet oxygen in D2O. However, after cellular uptake all mTHPC–HSA samples generate singlet oxygen in Jurkat cells, but the decomposition rate depends on the level of HSA cross-linking. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction Photodynamic Therapy (PDT) is an effective treatment for a number of malignant and non-malignant tumours [1]. PDT treatment is based on the presence of a drug with photosensitizing properties combined with far red or near infrared light and oxygen. In darkness photosensitizers exhibit no or low toxicity. After absorbing light the tumour-localized sensitizer molecules generate singlet oxygen, which causes the death of tumour cells or of the tumour vasculature resulting in local hypoxia and indirect cell death [2,3]. Within a few hours after successful PDT tumour tissue exhibits extensive regions of necrosis and apoptosis. These cell death mechanisms as well as inflammatory and immune responses are all induced by PDT [4]. Nanoparticles are regarded as potential PDT drug carrier system because of their high stability, high carrier capacity, feasibility to carry both hydrophobic and hydrophilic agents [5], as well as no restrictions in administration methods. Besides these advantages, nanoparticles can be designed as controllable drug release carriers. Especially, nanoparticles with suitable size accumulate in tumour tissue because of the enhanced permeability and retention (EPR) effect [6]. Moreover, by using antibody-modified nanoparticles selective targeting of tumour cells may be possible [7].
⇑ Corresponding author. Tel.: +49 30 2093 7625; fax: +49 30 2093 7666. E-mail address:
[email protected] (B. Röder). 1011-1344/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jphotobiol.2010.08.006
Human serum albumin (HSA) is a common protein in blood plasma [8,9]. As a potential candidate for drug delivery, it is biodegradable and non-antigenic. HSA can interact with many kinds of organic and inorganic molecules [10,11]. The HSA-based nanoparticles prepared as described in [12] have a defined particle diameter between 150 and 200 nm with a narrow size distribution [13,14]. As drug carrier for photosensitizers, the photophysical properties of pheophorbide a-HSA nanoparticles as well as their photocytotoxicity in Jurkat cells were evaluated in a former work [15]. For the preparation of HSA nanoparticles, glutaraldehyde was used as cross-linking agent. The amount of added glutaraldehyde mainly influences drug release and the stability of HSA nanoparticles during long-term storage [14]. 5-, 10-, 15-, 20-Tetrakis (3-hydroxyphenyl)-chlorin (mTHPC) is an efficient second-generation photosensitizer used in PDT [2,16]. The commercial mTHPC or FoscanÒ has already been approved by the European Medicines Agency (EMA) for the palliative treatment of advanced head and neck squamous cell carcinoma [17,18]. However, the tumour selectivity of mTHPC needs to improve. Moreover, the phototoxicity of mTHPC during the drug administration should be minimized to protect patients against unexpected light exposure. In this paper, HSA nanoparticles were used as mTHPC carrier. The influence of mTHPC-loading ratio on photophysical properties of mTHPC–HSA nanoparticles was evaluated. The optimized loading ratio was screened according to multiple standards: photophysical performance, reproducibility, and carrier efficiency. Furthermore, intracellular singlet oxygen
341
K. Chen et al. / Journal of Photochemistry and Photobiology B: Biology 101 (2010) 340–347
generation by mTHPC–HSA nanoparticles in Jurkat cells was investigated in order to evaluate the mTHPC release from HSA nanoparticles. The aim of this work was to investigate the influence of different mTHPC–HSA nanoparticle preparation parameters on drug release as well as intracellular singlet oxygen generation. These results will be helpful to further optimize photosensitizerloaded drug carrier systems.
Table 1a Sample description: mTHPC–HSA nanoparticles with different loading ratios.
NPL1 NPL2 NPL3 NPL4 NPL5
HSA cross-linking (%)
Loading ratio mTHPC/HSA (lg/mg)
100 100 100 100 100
4.5 8.9 17.8 44.5 89
2. Material and methods 2.1. Preparation of mTHPC–HSA nanoparticles Human serum albumin (HSA) nanoparticles were prepared by a desolvation method as described previously [12] In principle, an amount of 100 mg HSA was dissolved in 1 mL 10 mM sodium chloride solution. The pH was adjusted to 8.0 and the solution was filtered through a 0.22 lm filtration unit (Schleicher und Schüll, Dassel, Germany). Nanoparticles were formed by continuous addition of 4.0 mL ethanol under permanent stirring (380 rpm) at room temperature. A pumping device (Ismatec IPN, Glattbrugg, Switzerland) allowed a defined rate of ethanol addition (1 mL/min). After protein desolvation different amounts (23.1 lL, 57.8 lL, and 115.6 lL, respectively) of 8% aqueous glutaraldehyde solution were added to stabilize the resulting protein nanoparticles by chemical cross-linking. The common amount of glutaraldehyde used for particle stabilization is between 40% and 200% of the amount necessary for stoichiometric cross-linking of the 60 amino groups in the HSA molecule [9]. The glutaraldehyde concentrations used correspond to 40%, 100%, and 200% stoichiometric level of crosslinking of the amino groups in 100 mg HSA. Particles were stirred for 1 h and were purified by three cycles of centrifugation (20,817g, 10 min) and redispersion in 1.0 mL water in an ultrasonic bath (5 min). mTHPC molecules (Fig. 1) were loaded on HSA nanoparticles (7.656 mg/mL) by non-covalent adsorption. After mTHPC addition (1 mM), the mixtures were shaken with a minishaker (IKA-MS1, Milian AG) for 5 min and suspended 10 min by ultrasonication (Elma Hans Schmidbauer GmbH & Co. KG). In this paper, abbreviation NPL1-5 is used for samples with different mTHPC/HSA loading ratios (loading parameters see Table 1a). Abbreviations NPC40, NPC100 and NPC200 stand for mTHPC– HSA nanoparticles prepared with different level of cross-linking between the HSA molecules in the nanoparticle matrix, loading
Fig. 1. Structural formula of 5-, 10-, 15-, 20-Tetrakis (3-hydroxyphenyl)-chlorin (mTHPC).
Table 1b Sample description: mTHPC–HSA nanoparticles with different degrees of crosslinking.
NPC40 NPC100 NPC200
HSA cross-linking (%)
Loading ratio mTHPC/HSA (lg/mg)
40 100 200
19.05 20.71 20.59
parameters see Table 1b. Water was used as dispersion medium for all mTHPC–HSA nanoparticles. 2.2. Spectroscopic methods If not otherwise mentioned, the concentration of all samples was adjusted to a mTHPC content of 3 lM. mTHPC was dissolved in ethanol. mTHPC–HSA nanoparticles were diluted with distilled water. UV–Vis absorption spectra were measured with a commercial spectrometer Shimadzu UV-160A at room temperature. Fluorescence spectra were taken at room temperature using the set-up described in [19] using a Xenon lamp (XBO 150, OSRAM) with a monochromator (LOT-Oriel) at 506 nm for excitation and a polychromator with a cooled CCD matrix for detection (LOT-Oriel, Instaspec IV). Fluorescence lifetime measurements were carried out using time-correlated single photon counting (TCSPC) [20], comprising a thermoelectrical cooled microchannel plate (R3809-01, Hamamatsu, computer controlled monochromator (77200, LOT-Oriel) and a PCI TCSPC controller card (SPC630, Becker & Hickl): Excitation wavelength was kexc = 532 nm, fluorescence was recorded at 650 nm under a polarization angle of 54.7° (magic angle). Optical density (OD) of the samples was adjusted to 0.2 at the lowest energy absorption band. Data were analyzed by a home made program using the least square fit and deconvoluted by the instrument response function (IRF) including a time shift. Detailed description of the TCSPC set-up is given in [20]. Singlet oxygen luminescence at 1268 nm was detected with a liquid nitrogen-cooled Ge-diode detector Model EO-817P (North Coast, Inc., Santa Rosa, CA). All samples were excited with a Nd:YAG pumped OPO, NT 342/1 (Ekspla, Vilnius, Lithuania) at 652 nm. TPPS (5-, 10-, 15-, 20-Tetraphenyl-21H, 23H-porphine-p, p0 , p00 , p00 0 -tetrasulfonic acid, tetrasodium salt hydrate) (UD = 0.70) was taken as reference. For better signal-to-noise ratio (SNR) water was replaced by D2O. Laser flash photolysis was used to investigate the triplet lifetime of the photosensitizer in the samples. For excitation a ns-Nd:YAG laser pumped OPO (NT 342/1 (Ekspla, Vilnius, Lithuania) was operated at kexc = 652 nm. Transient triplet–triplet absorption was measured using the light of an LED (kmax = 470 nm), a Si-diode with integrated fast pre-amplifier (Elektronik Manufaktur Mahlsdorf, Germany) and a fast digital oscilloscope (HP5415). The triplet lifetime of photosensitizer can be calculated by analysing the transient absorption curve. Comparison of triplet
342
K. Chen et al. / Journal of Photochemistry and Photobiology B: Biology 101 (2010) 340–347
decay time of air saturated and oxygen-depleted sample allows an indirect estimation of singlet oxygen quantum yield. Samples were flushed with N2 for 40 min to obtain oxygen-depleted samples. The value of singlet oxygen quantum yield (UD) can be determined by the following equation:
UD ¼ UT
sTN2 sTair SD sTN2
ð1Þ
UN is the triplet quantum yield, sTN2 is the triplet lifetime in oxygen-depleted sample and sTair is the triplet lifetime in air-saturated sample, SD is the fraction of all oxygen deactivation processes, in which singlet oxygen is generated. SD is very close to 1 in case of tetrapyrroles such as mTHPC [21,22]. The singlet oxygen luminescence of photosensitizers located inside Jurkat cells was determined with time-correlated multi-photon counting (TCMPC). In this set-up, a custom dye laser (DCM) is pumped by a frequency doubled Q-switching Nd: YAG laser (Serie 600, B.M. Industries). The average power of the laser is monitored by a powermeter (FieldMaster GS, COHERENT). The fluorescence and scattering light under 1270 nm are blocked by an interference filter (bk-1270-70-B, bk-interferenzoptik, Germany). The luminescence was focused on a NIR-photomultiplier (H-10330-45, HAMAMATSU). The signal is sent to a computer equipped with a custom-made TCMPC plug-in card. For more details about intracellular singlet oxygen luminescence determination see [23]. The temporal shape of singlet oxygen luminescence in a homogeneous environment for pulsed excitation can be determined using the following equation [23]: t t exp ID ¼ N UD exp
sD
sT
sD sD sT
ð2Þ
where ID is the singlet oxygen luminescence intensity. N is the initial population of triplet state photosensitizer molecules. sD is the singlet oxygen lifetime and sT is the triplet lifetime of the photosensitizer. The obtained singlet oxygen luminescence decay curves were fitted with the two-exponential model function (Eq. (2)). As can be seen, the shape of singlet oxygen luminescence is determined by two competitive lifetimes: the triplet lifetime of photosensitizer sT and the singlet oxygen lifetime sD. The rising of singlet oxygen luminescence is determined by the faster one. The slower one decides the signal decay. The triplet lifetime of photosensitizer in solution is usually shorter than singlet oxygen lifetime. Therefore, the rising signal presents triplet lifetime sT. The decay signal is caused by singlet oxygen lifetime sD. In contrast to singlet oxygen in solution, intracellular singlet oxygen is always quenched by endogenous chemical quenchers in cells. The singlet oxygen lifetime is usually shorter than the triplet lifetime of photosensitizer. In this case, the rising signal is caused by singlet oxygen decay and the decay of the curve is determined by the triplet lifetime.
NPC200 in darkness, respectively. Before illumination cells were washed with PBS twice and re-suspended in PBS to reach a concentration of 3 105/mL. The cells were illuminated at the wavelength of 652 nm for 20 s for each sample with an average laser power of 6 mW, which gives an efficient light dose of 32 nJ/cell (efficient light dose is determined by the product of illumination intensity, time and geometric cross-section of the cell). 3. Results and discussions 3.1. Photophysical properties of mTHPC–HSA nanoparticles with different mTHPC-loading ratios 3.1.1. Absorption spectra The most obvious effect in absorption is the dependency of both the strength and spectral position of the Soret band from the loading ratio (Figs. 2a and b). The Soret band maximum of mTHPC in ethanol locates at 416.6 nm and it exhibits a large bathochromic shift of about 10 nm in aqueous solution. Adsorbed on HSA nanoparticles the Soret band maximum of NPL1 locates at 423.4 nm. The position changes slightly bathochromic with higher loading up to 424.8 nm for NPL5. For mTHPC, the extinction coefficient at Soret band maximum decreases from 2.07 105 M1 cm1 in ethanol to 1.49 105 M1 cm1 in water (Table 2). Accompanied by reducing extinction, the full width at half maximum (FWHM) of the Soret band broadens to 54.2 nm compared to 32.7 nm of mTHPC in ethanol. For the loaded HSA nanoparticles the extinction of the Soret band maximum continuously decreases from 2.52 105 M1 cm1 for NPL1 to 1.88 105 M1 cm1 for NPL5 (Fig. 3). The FWHM increases about 5.5 nm with increasing loading ratio. As can be seen from Fig. 3, the extinction coefficient of the mTHPC depends on the loading ratio. While for low loading ratio the extinction coefficient of mTHPC on HSA particles is higher compared to mTHPC in ethanol, it is continuously decreasing with higher loading ratio. This observation suggests that besides the interaction of mTHPC molecules with the HSA nanoparticles, there is another type of interaction occurring mainly for higher loading ratios. Because of geometric considerations, the interaction of mTHPC with the HSA nanoparticle itself should be nearly indepen-
2.3. Cell culture and incubation Jurkat cells (clone E 6-1 human acute T-cell leukaemia) were cultured in RPMI 1640 medium with L-glutamine (GIBCO) supplemented with 10% fetal calf serum (FCS), 100 U/mL of penicillin and 100 lg/mL of streptomycin without phenol red [24]. Cells were cultivated at 37 °C in 100% humidity and 5% CO2 and were seeded in new medium every 2–3 days. For the investigation of the level of cross-linking effect of mTHPC–HSA nanoparticles, Jurkat cells were incubated for 1 h, 3 h, 5 h, and 24 h with 3 lM mTHPC and NPC40, NPC100, and
Fig. 2a. UV–Vis absorption spectra of mTHPC in ethanol, water and of mTHPC–HSA nanoparticles in aqueous solution, mTHPC conc.: 3 lM (scattering corrected). Curves description: 1. mTHPC in ethanol; 2. mTHPC in H2O; 3–7 from peak to bottom: NPL1-5, respectively.
343
K. Chen et al. / Journal of Photochemistry and Photobiology B: Biology 101 (2010) 340–347
Fig. 2b. UV–Vis absorption spectra (Soret band region) of mTHPC in ethanol and aqueous solution, and of mTHPC–HSA nanoparticles in aqueous solution (scattering corrected). Curves description: 1. mTHPC in ethanol; 2. mTHPC in H2O; 3–7 from peak to bottom: NPL1-5, respectively.
Table 2 Soret band maxima, extinction coefficient, and full width at half maxima (FWHM) of mTHPC and mTHPC–HSA nanoparticles.
mTHPC in ethanol NPL1 NPL2 NPL3 NPL4 NPL5 mTHPC in H2O
kmax (nm)
e (M1 cm1)
FWHM (nm)
416.6 423.4 423.4 423.7 424.0 424.8 427.2
2.07 105 2.52 105 2.39 105 2.32 105 2.22 105 1.88 105 1.49 105
32.7 32.2 32.4 33.3 35.4 37.7 54.2
Fig. 4. Fluorescence spectra of mTHPC in ethanol and of mTHPC–HSA nanoparticles with different mTHPC-loading ratios (mTHPC conc.: 3 lM; kexc: 506 nm). Curves description: 1. mTHPC in ethanol; 2–6 from peak to bottom: NPL1-5, respectively.
Table 3 Fluorescence emission of mTHPC in ethanol and mTHPC–HSA nanoparticles in aqueous solution. Sample
Loading ratio mTHPC/ HSA (lg/mg)
kmax (nm) of S1,0 ? S0,0
kmax (nm) of S1,0 ? S0,1
UFl (%)a
NPL1 NPL2 NPL3 NPL4 NPL5
4.5 8.9 17.8 44.5 89.0
656.3 656.7 657.0 657.5 658.7
722.3 722.5 722.7 721.4 723.3
80.0 65.9 46.2 19.5 8.6
kmax: Location of fluorescence maximum; UFL: Fluorescence quantum yield. a Standard: mTHPC in ethanol.
mainly caused by the interaction between photosensitizer and HSA nanoparticles. For highly loaded samples, mTHPC–HSA interaction plays a minor role in comparison with interaction between neighbouring mTHPC molecules. Since the interactions are visible in ground state it is highly probable, that the mTHPC molecules undergo strong excitonic interactions. Such interactions have been reported for other multi-chromophoric chlorin systems [25,26].
Fig. 3. Extinction coefficient changing with mTHPC/HSA loading ratio, mTHPC concentration: 3 lM.
dent from the loading ratio. In contrast, at high loading ratio it seems highly probable, that closer positioning of the individual mTHPC molecules in each nanoparticle causes additional interactions between the dye molecules. As a result the intensity of this interaction does not simply depend on the number, but on the distance between neighbouring molecules on/in the nanoparticle. At low drug loading ratio (
