Accepted Manuscript Research paper Low level LED photodynamic therapy using curcumin loaded tetraether liposomes Lili Duse, Shashank Reddy Pinnapireddy, Boris Strehlow, Jarmila Jedelská, Udo Bakowsky PII: DOI: Reference:

S0939-6411(17)30816-0 https://doi.org/10.1016/j.ejpb.2017.10.005 EJPB 12610

To appear in:

European Journal of Pharmaceutics and Biopharmaceutics

Received Date: Revised Date: Accepted Date:

11 July 2017 6 October 2017 6 October 2017

Please cite this article as: L. Duse, S. Reddy Pinnapireddy, B. Strehlow, J. Jedelská, U. Bakowsky, Low level LED photodynamic therapy using curcumin loaded tetraether liposomes, European Journal of Pharmaceutics and Biopharmaceutics (2017), doi: https://doi.org/10.1016/j.ejpb.2017.10.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Low level LED photodynamic therapy using curcumin loaded tetraether liposomes Lili Duse1, Shashank Reddy Pinnapireddy1, Boris Strehlow1, Jarmila Jedelská1, Udo Bakowsky1*

1

Department of Pharmaceutics and Biopharmaceutics, University of Marburg, 35037 Marburg, Germany

*Corresponding author:

Prof. Dr. Udo Bakowsky Department of Pharmaceutics and Biopharmaceutics Robert Koch Str. 4 35037 Marburg Tel: +49 6421 28 884 Fax: +49 6421 27016 E-mail: [email protected]

Dedicated to the 70th birthday of Prof. Dr. Ulrich Rothe for his contributions to the field of tetraether lipid technology.

Abstract Oncological use of photodynamic therapy is an evolving field in cancer therapeutics. Photosensitisers are prone to accumulation inside healthy tissues causing undesirable effects. To avoid this, we have developed tetraether lipid liposomal formulations containing curcumin which is a naturally occurring anti-cancer substance and deemed to be safe towards healthy cells. Upon excitation with light at a specific wavelength, curcumin produces reactive oxygen species (ROS) in presence of oxygen, thereby exhibiting a cytotoxic effect towards the surrounding tissues, giving a total control on the onset of therapy. In our study, we examined two different liposomal formulations wherein curcumin is encapsulated within the hydrophobic milieu with the intent to increase its bioavailability. Hydrodynamic diameter, surface charge, stability, morphology and haemocompatibility of the liposomes were studied. The results confirmed the formation of stable nanometre range liposomal vesicles (200 - 220 nm) containing curcumin which were haemocompatible with coagulation time less than 50 s and a haemolytic potential below 40%. Increased ROS generation post irradiation (>50% compared to un-irradiated samples) was confirmed using fluorescence spectroscopy. The efficiency and selectivity of the PDT was demonstrated by assessing their viability post irradiation and by qualitative analysis using confocal microscopy showing nuclear perforation induced by PDT. Photo-destructive effects of PDT on the microvasculature were studied in vivo using chick chorioallantoic membrane model (CAM). Considerable phototoxicity could be observed in the irradiated area of the CAM 30 min post irradiation. Phototoxic effects in vitro (in SK-OV-3 and PCS-100-020™) and in vivo (in chorioallantoic membrane model) in combination with a novel custom manufactured LED irradiating device showed a formulation dependant selective photodynamic effect of the curcumin liposomes.

Keywords:

Photodynamic therapy,

LED,

chorioallantoic membrane, serum stability

curcumin,

tetraether

liposomes,

cancer,

1. Introduction In the field of drug delivery, tumour targeting has always been of a great interest and has often faced many obstacles. One of the major challenges is the efficient delivery of the chemotherapeutic drug to the affected tissues. Many therapies rely on the fact that rapidly proliferating cells are more prone to the cytotoxic effects of the chemotherapeutic agent [1]. Unfortunately, this approach apart from being cytotoxic to the surrounding tissues, leads to many other undesirable effects [2]. Modern liposomal and nanoparticle based tumour therapies employ natural or synthetic vehicles to efficiently deliver the drug into the tissues. Liposomes in particular, are a highly efficient and biocompatible system for targeted delivery of drugs, nucleic acids and other biomolecules to tumour sites. Due to their size, ease of formulation, biocompatibility, and lipid specific cellular uptake mechanisms, liposomes can be tailor made to suit the therapy. Amongst non-invasive tumour therapies available today, photodynamic therapy (PDT) is one of the least complicated without the requirement of complex equipment and set-up. This simple and effective therapy requires a combination of a drug capable of being excited at a specific wavelength i.e. a photosensitising agent (photosensitiser) and light. Upon successful administration of the photosensitiser and its subsequent internalisation in the target tissue, light is used to induce selective damage to the tumour tissue. This process of photosensitisation hugely relies upon the presence of molecular oxygen and forms the basis for a successful PDT [3, 4]. After absorbing energy from the light source, the photosensitiser interacts with molecular oxygen via energy transfer process or electron transfer process and generates singlet oxygen or forms free radicals respectively [5]. These reactive oxygen species (ROS) oxidise cellular and subcellular organelles inducing either apoptosis or necrosis leading to cell death [6]. PDT has an added advantage of selectively targeting tumour tissues with almost no or minimal effect towards healthy tissues [7]. The most common disadvantage of photosensitisers is their non-specificity, which often leads to its accumulation in healthy surrounding tissues causing undesirable effects. This could be avoided by optimising the dosage of the photosensitiser used, intensity of irradiated light, exposure time and the choice of photosensitiser itself. We employed curcumin, a naturally occurring substance as a photosensitiser of choice in our study due to its proven anticancer properties and safe toxicity profile at therapeutic concentrations [8]. Being multifaceted, curcumin has been used since centuries for treating different kinds of ailments, has been used successfully in photodynamic therapy in the oral cavity, and has been employed for the treatment of different cancers [9]. The mechanism of photo-destruction induced by curcumin in tumour cells has been reported to be primarily due to apoptosis caused by redox signalling, caspase-3 activation and inhibition of cyclooxygenase-2 apart from other nonapoptotic mechanisms such as chromatin degradation, DNA fragmentation and inhibition of protein phosphatases [10-13]. Depending upon the solvent used, curcumin can be excited at different wavelengths, facilitating the use of various light sources [14]. Furthermore, curcumin accumulates selectively in tumour tissues leading to selective photocytotoxicity of tumour tissues [15].

Combining liposomal therapies with other therapeutic regimens is a field which is garnering much interest lately due to their synergistic effects [16]. In the present research, we focus on utilising the beneficial properties of liposomes along with an already popular and well-documented PDT [17]. Among other delivery systems, liposomes are a popular choice for delivery of drugs and biomolecules due to their close resemblance towards biological membranes and their ease of optimisation. Biocompatibility, bioavailability and circulation time of poorly soluble drugs can be enhanced using liposomes [18]. In this study, we used new liposomal formulations comprising tetraether lipids. Tetraether lipids have larger lipophilic chains and deemed superior in terms of stability and less prone to oxidation due to the presence of ether bonds [19, 20]. This property of tetraether lipids enables a wider range of applications and a broader choice of delivery routes. As with other therapies, it is difficult to compare the outcome of PDT in vitro to that of in vivo. To understand the effect of liposomal curcumin based PDT in vivo, we have made use of chick chorioallantoic membrane model (CAM) to get a better insight of the synergistic effects of this novel combination. It has been reported previously that PDT causes destruction of tumour microcirculation, thereby, restricting blood flow of cancerous tissues [21, 22]. CAM is perfectly suitable for studying this effect due to the presence of the microvasculature at the surface facilitating local irradiation without the need of any complex invasive procedures [23, 24]. Apart from increasing the bioavailability and cellular uptake of photosensitiser, the other rationale of our study was to make PDT cost effective, efficient and safe. Hence, we have utilised a tailor-made prototype low power light emitting diode (LED) array for the irradiation of the photosensitiser. LED’s have an edge over traditional lasers wherein they consume less energy thereby facilitating the use of batteries to power them up which makes them extremely portable. LED’s dissipate less heat thus minimising energy loss, and can be spatially arranged to suit the area or organ of interest which needs to be irradiated. Due to the advancements in the field of semiconductor technology, LED’s have remarkable durability and are relatively inexpensive [25]. Moreover, as opposed to lasers, they emit light over a larger surface area with homogenous energy dissipation enabling treatment of larger lesions and tumours in fewer sittings. Due to their broad spectrum, they compensate for the spectral shift of the photosensitiser caused by the addition of liposomes [26]. To better comprehend and to optimise this therapy, the liposomal formulations have been characterised for their size and surface charge using dynamic light scattering and laser Doppler anemometry respectively. The structural morphology was analysed using transmission electron microscopy. The photo-destructive effects of this unique combinatorial therapy have been evaluated in vitro by cytotoxicity assays and fluorescence microscopy; and in vivo by optical observation of microvasculature destruction in the chorioallantoic membrane model. Haemocompatibility of the liposomal formulations has been determined using haemolysis and activated partial thromboplastin time (aPTT) tests. Serum stability test was performed to determine the stability of liposomes in serum. Irradiation experiments were carried out both on tumour cells and normal primary human cells to demonstrate the selectivity of this therapy.

2. Materials and methods 2.1 Materials Curcumin was obtained from Sigma Aldrich (Taufkirchen, Germany). A polar lipid fraction containing tetraether lipids caldarchaeol (GDGT) and calditoglycerocaldarchaeol (GDNT) was extracted from the biomass of Sulfolobus acidocaldarius (SIT Rosenhof GmbH, Heiligenstadt, Germany) as previously mentioned [27]. The ends of GDNT (~90% of the extract) contain phosphatidylmyoinositol and β-glucose respectively while GDGT (~10% of the extract) contains phosphatidylmyoinositol and β-D-galactosyl-D-glucose on either ends respectively [28]. 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) and 1,2-distearoyl-snglycero-3-phospho-(1'-rac-glycerol) (DSPG) were a gift from Lipoid GmbH (Ludwigshafen, Germany). 2’,7’-dichlorofluorescin diacetate (DCFDA) and tert-Butyl hydroperoxide (TBHP) were obtained from Sigma Aldrich. All other chemicals used were of analytical grade, all buffers used in this study were prepared in the laboratory unless otherwise mentioned. Double distilled water, which was autoclaved and filter sterilised using 0.2 µm polyethersulphone filters (Whatman GmbH, Dassel, Germany) was used for all experiments.

2.2 LED Device A prototype low power LED device consisting of an array of light emitting diodes designed to fit multiwell plates was custom manufactured by Lumundus GmbH (Eisenach, Germany). The device is equipped with two different matrices of LED’s capable of emitting light at wavelengths of 457 and 620 nm respectively. Irradiation time, current (interchangeable between 20, 40, 60, 80 and 100 mA) and wavelength could be adjusted using dedicated buttons. Radiation fluence was calculated based on the current and irradiation time.

2.3 Cell lines and cell culture Wild type human ovarian adenocarcinoma cells (SK-OV-3) and primary human coronary artery endothelial cells PCS-100-020™ were procured from American Type Culture Collection (ATCC, Manassas, USA). SK-OV-3 cell line was cultivated at 37 ºC and 7% CO2 under humid conditions in IMDM medium (Capricorn Scientific, Ebsdorfergrund, Germany) supplemented with 10% foetal bovine serum (Sigma Aldrich). PCS-100-020™ cells were maintained at 37 ºC and 5% CO2 under humid conditions in MCDB 153 basal medium (Biochrom GmbH, Berlin, Germany) supplemented with PCS-100-041™ vascular endothelial cell growth kit (ATCC). Cells were grown as monolayers and passaged upon reaching 80% confluency.

2.4 Preparation of curcumin loaded liposomes Liposomes were prepared by mixing appropriate molar ratios (as described in Table 1) of lipids (dissolved in 2:1 (v/v) Chloroform ethanol mixture) and curcumin (dissolved in methanol) together in a 10 mL round bottom flask. The curcumin to lipid ratio was always kept constant at 1:30. Batches containing 0.3 mg curcumin and 10 mg lipids were prepared. Using a Laborota 4000 rotary evaporator (Heidolph Instruments, Schwabach, Germany) equipped with a vacuum pump (KNF Neuberger GmbH, Freiburg, Germany), the curcuminlipid mixture was evaporated to obtain a thin film. The film was hydrated using 20 mM HEPES buffered saline (pH 7.4) and sonicated in a bath sonicator (Elmasonic P30 H, Elma Hans Schmidbauer, Singen, Germany) at 56 °C to obtain a uniform dispersion of liposomes.

2.5 Dynamic light scattering and laser Doppler velocimetry The hydrodynamic diameter of the liposomes was analysed by dynamic light scattering (DLS) using a Zetasizer Nano ZS (Malvern Instruments, Herrenberg, Germany). The liposomes were diluted to a ratio of 1:100 with 20 mM HEPES (pH 7.4) prior to the measurements. For analysis of the data, viscosity (0.88 mPa.s) and refractive index (1.33) of water at 25 ºC were considered. The instrument is equipped with a 10 mW HeNe laser and the measurements were performed at a wavelength of 633 nm and a detection angle of 173º backscatter [29]. Measurement position and laser attenuation were automatically adjusted by the instrument. The instrument performs 15 size runs per measurement with each lasting 10 s. Zeta potential measurements were performed by laser Doppler velocimetry (LDV) using the Zetasizer Nano ZS in a clear disposable folded capillary cell (DTS1060, Malvern Instruments). The liposomes were diluted as described above prior to the measurement. Depending upon the sample, the instrument automatically performs 15-100 runs per measurement. Three independent formulations were measured for both DLS (hydrodynamic diameter; intensity measurement ) and LDV (zeta potential) and the results are expressed as averages of these measurements.

2.6 Transmission Electron Microscopy Transmission electron microscopy (TEM) was performed using a JEM-3010 UHR Transmission electron microscope (Jeol Ltd., Tokyo, Japan), equipped with a retractable highresolution slow scan CCD-Camera (Gatan Inc., Pleasanton, USA). Liposomes were diluted to 1:10 ratio with 10 mM HEPES buffer (pH 7.4) prior to negative staining using 2% uranyl acetate. 300 mesh Formvar coated copper grids (Plano GmbH, Wetzlar, Germany) were used for mounting the sample [30]. Equal parts of the sample and stain were mixed together and the grid was incubated for 5 min in this solution following which it was examined at an accelerating voltage of 300 kV and 110 µA emission current with current densities between 50-60 pA/cm2.

2.7 Encapsulation efficiency The encapsulation efficiency (EE) was determined by extracting curcumin from the liposomes. 300 µl of both liposomal formulations were centrifuged for 90 min at 2000 x g in an Eppendorf Centrifuge 5418 (Eppendorf GmbH, Wesseling-Berzdorf, Germany). Briefly, the pellet was diluted in 200 µl ethanol and 200 µl HEPES (pH 7.4). 200 µl of supernatant was diluted with 200 µl ethanol and curcumin from both samples was quantified spectrophotometrically at 430 nm using Multiskan GO plate reader (Thermo Fischer Scientific GmbH, Dreieich, Germany). All non-dissolved liposomes were removed during different purification steps by centrifugation. Calibration curve was recorded with known curcumin concentrations in the same solvent composition. Then EE (% w/w) was calculated as follows:

2.8 Stability of curcumin containing liposomes To simulate physiological conditions, 2 ml serum was diluted with 20 mM HEPES (pH 7.4) to 60% and was mixed with 0.4 ml curcumin containing liposomes in a ratio of 5:1 (v/v). Similarly, 20 mM HEPES (pH 7.4) was mixed with 0.4 ml curcumin containing liposomes in a ratio of 5:1 (v/v). Both the mixtures were incubated at 37 °C in a shaking incubator at 100 rpm for 24 h. In control experiments only serum with HEPES was incubated with the same parameters. All samples were diluted to a ratio of 1:20 with 20 mM HEPES (pH 7.4) prior to the measurements at different time intervals. Three independent measurements were performed using DLS and LDV [31].

2.9 Cell culture and irradiation experiments Cells were seeded onto 96-well transparent microtiter plates (Nunclon Delta, Thermo Fisher Scientific GmbH, Dreieich, Germany) at a seeding density of 10000 cells/0.35 cm2 (per well). After 24 h, the cells were incubated with various concentrations of curcumin loaded liposomes and curcumin (dissolved in DMSO) in triplicates for 1, 2 and 4 h periods. DMSO in concentration similar to the concentration present in curcumin solution was used as a solvent control. Subsequently, the cells were irradiated at 457 nm at various fluence levels and were incubated overnight. An unirradiated microtiter plate (dark) was used as a control. On the following day, the medium was removed and replaced with medium containing MTT dye and incubated for 4 h. The resulting formazan crystals were dissolved in DMSO and the absorbance was determined at 570 nm using a FluoStar Optima plate reader (BMG Labtech, Offenburg, Germany). Viability of untreated cells (blank) was considered as 100%.

2.10 Activated partial thromboplastin time (aPTT) test aPTT test was performed to determine the effect of the formulations on blood coagulation. Fresh blood was drawn into citrate tubes followed by centrifugation at 1500 x g for 10 min to separate the plasma fraction. The aPTT test was performed in a Coatron M1 coagulation analyser (Teco Gmbh, Neufahrn, Germany) using the TEClot aPTT-S Kit as per the manufacturer’s protocol with slight modifications [29]. Briefly, 25 µL of plasma was mixed with 25 µL of sample. 25 µL of aPTT reagent was added to the sample to activate coagulation factors followed by the addition of prewarmed 0.025 M calcium chloride. Coagulation was confirmed spectrophotometrically and the time was recorded in seconds.

2.11 Ex vivo Haemolysis Assay To determine the effect of the formulations in blood, human erythrocytes were isolated from fresh blood as described previously [32]. Briefly, fresh blood was drawn into tubes containing EDTA followed by centrifugation of the whole blood. The obtained red blood cell pellet was washed thrice with PBS buffer (pH 7.4) and diluted to 1:50 with PBS. The erythrocytes were incubated together with the formulations for 1 h at 37 ºC in V-bottom microtiter plates in an orbital shaker KS4000 IC (IKA Werke, Staufen, Germany). The plates were then centrifuged and the absorbance of the collected supernatant was determined at 540 nm in a FluoStar Optima plate reader. PBS buffer (pH 7.4) and 1% Triton X-100® were used as controls and the absorbance values of Triton X-100® was considered as 100% haemolysis.

2.12 Determination of Reactive Oxygen Species Reactive oxygen species (ROS) were determined using cell permeant DCFDA according to the cellular reactive oxygen species protocol from Abcam (Cambridge, UK) with slight modifications. Briefly, SK-OV-3 cells were seeded onto 96-well microtiter plates as mentioned above. On the following day, the cells were incubated with liposomes for 1 h, 50 µM TBHP was used as positive control for the ROS experiments. After the incubation time, cells were washed using PBS containing Ca2+ and Mg2+ (pH 7.4) and the medium was replaced with fresh medium, following which, they were irradiated at 457 nm at a radiation fluence of 8.6 J/cm2. The cells were then washed again and incubated with medium (IMDM without phenol red) containing 25 µM DCFDA for 1 h. The cells were subsequently washed and lysed using cell culture lysis reagent (Promega, Mannheim, Germany) and fluorescence was recorded using a FluoStar Optima plate reader (λex 480 nm / λem 520 nm).

2.13 Intracellular visualisation of PDT For the visualisation experiments, 90000 cells per well were seeded onto 12-well plates (Nunclon Delta) containing cover slips ( 15 mm). The plates were incubated for 24 h before being used for the experiments. Liposomes containing curcumin were added drop wise to each well; the plates were gently swirled and incubated. After 1 h, the supernatant was removed and replaced with fresh medium and the plates were irradiated at a radiation fluence of 8.6 J/cm2. The cells were then washed twice with PBS containing Ca 2+ and Mg2+ (pH 7.4) and fixed with 4% formaldehyde solution for 20 min after which the cell nucleus was counterstained with 0.1 µg/mL 4',6-diamidino-2-phenylindole (DAPI) for 20 min. Finally, the cells were washed with PBS (pH 7.4) and the cover slips were mounted onto slides and sealed using FluorSave ™ (Calbiochem Corp, La Jolla, USA). The cells were examined under a Confocal Laser Scanning Microscope (Zeiss Axiovert 100M, Carl Zeiss Microscopy GmbH, Jena, Germany). An Argon Ion laser (Coherent Enterprise, Coherent Inc., California, USA) with 364 and 488 nm wavelengths for observing nuclear counterstaining and curcumin respectively was used. Detector equipped with long pass filter of 505 nm for curcumin and band pass filter of 385-470 nm for DAPI was used for recording the micrographs. All micrographs were recorded at a similar detector gain and pinhole size.

2.14 CAM Experiments Specific pathogen free fertilised chicken eggs weighing 50-60 g were obtained from VALO BioMedia GmbH (Osterholz-Scharmbeck, Germany) and were incubated at 37 °C in an Ehret KMB 6 hatching incubator (Dipl. Ing. W. Ehret GmbH, Emmendingen, Germany) equipped with an automatic turner in a humidified atmosphere (>60% RH). For the CAM experiments a previously established protocol was utilised with modifications[33]. On egg development day (EDD) 3, a 3 mm  hole was bored into the basal part of the shell and approximately 3 ml of albumin was drawn out of the egg using a syringe to prevent the adhesion of the CAM to the apical part of shell. The apical part of the shell was cut open (Ø 30 mm) to expose the CAM. The bottom end was sealed using a cellophane tape and the apical part using a paraffin film. The eggs were further incubated on a frame designed to hold them upright to facilitate the development of CAM. On EDD 11, 0.1 ml of the liposomal formulations (700 µM) was applied intravenously into the veins and the eggs were incubated for 5 min. 5 mm Ø polypropylene rings cut out of plastic straws were used to localise the application area [34]. The area localised with the ring was irradiated at a radiation fluence of 3.6 J/cm2 (λex 405 nm, 4 Hz frequency) using a Weber Needle Endolaser (Weber Medical GmbH, Lauenförde, Germany) equipped with a PGL-V1 WM Blue Diode Laser. Images of the microvasculature were acquired prior to and post irradiation using a Stemi 2000-C stereo microscope (Carl Zeiss GmbH, Jena, Germany) attached with a Moticam 5 CMOS camera (Motic Deutschland GmbH, Wetzlar, Germany).

2.15 Statistical analysis All experiments were performed in triplicates and the values are presented as mean ± standard deviation, unless otherwise stated. Two-tailed Student’s t-test was performed to identify statistical significance differences. Probability values of p < 0.05 were considered significant. Statistical differences are denoted as “*” p < 0.05, “**” p < 0.01 and “***” p < 0.001.

3. Results and Discussion 3.1 Curcumin loaded liposomes Upon rehydration of the lipid film, curcumin binds to the hydrophobic tails present in the liposomal bilayer thereby, greatly increasing its bioavailability in aqueous media [18]. We used two different liposomal formulations to study the effect of liposomal encapsulation on bioavailability, cellular uptake, stability and photocytotoxicity of curcumin in vitro. The DS liposomes comprised DSPC and DSPG lipids, while T10 comprised 10% of TELs in addition to DSPC. Although there was no drastic increase in the hydrodynamic diameter of the liposomes, the zeta potential decreased consistently with increasing amount of TELs (Table 1). This is due to presence of positively charged head groups in the TEL lipid mixture. The polydispersity index (PDI) of the liposomal formulations suggests that the liposomes prepared showed a polydisperse size distribution. Tetraether lipids are known to stabilise the liposomes by enhancing their membrane integrity and increase their thermal stability, a feature which could be exploited for oral delivery of liposomal formulations [28, 35]. Upon incorporation of TELs in the liposomal formulation, a moderate increase in size was observed which however did not affect the internalisation of liposomes into the cells. Table 1: Physicochemical properties of curcumin loaded liposomes (0.3 mg curcumin per 10mg lipids); hydrodynamic diameter is expressed as a measure of particle size distribution by intensity, using three independent formulations of each (n=3). Liposomes DS T10

Composition [mol%] DSPC:DSPG 80:20 DSPC:TELs 90:10

Diameter [nm] ± SD 211.30 ± 3.65

Zeta Potential [mV] ± SD -22.63 ± 0.21

PDI ± SD 0.29 ± 0.05

208.70 ± 2.41

-5.89 ± 1.66

0.17 ± 0.05

3.2 Haemocompatibility To study the effect of our liposomal formulations on blood, we have carried out haemolysis test and aPTT test which determine the effect on erythrocytes and change in coagulation time respectively upon addition of liposomes (Fig. 1). Haemolysis assay determines the amount of haemoglobin released from the erythrocytes upon exposure to liposomes (Fig. 1B) The haemoglobin released from the erythrocytes reacts with atmospheric oxygen to form oxyhaemoglobin which could be determined spectrophotometrically [36]. The haemolytic potential of the liposomes used in the study appeared to be minimal. Similarly, aPTT analysis revealed that the liposomal curcumin increased the coagulation time by only 17.1 s for DS liposomes and 9.7 s for T10 liposomes, whereas curcumin alone increased the time by 112.4 s. An aPTT above 70 s denotes continuous bleeding leaving the patients with the risk of haemorrhage [37].

Haemolysis assay could help understand the correlation between in vitro and in vivo results and is also useful in determination of safe concentration to be administered in an in vivo set up.

Fig. 1. A. aPTT test and B. Haemolysis assay of the DS (DSPC:DSPG) and T10 (DSPC:TEL) liposomes. All curcumin containing liposomes have a curcumin concentration of 0.3 mg per 10 mg lipids. Blood plasma was used a control in aPTT experiments; Triton™ X-100 lyses all the cells and was hence used as positive control in haemolysis assay. Samples were measured in triplicates and results are expressed as the mean ± SD (n=3). Probability values of p < 0.05 were considered significant. Statistical differences are denoted as “*” p < 0.05, “**” p < 0.01 and “***” p < 0.001. For statistical analysis, the results were compared against the results of fresh blood plasma.

3.3 TEM Analysis TEM observation revealed circular multilamellar bilayer vesicles (Fig. 2). The diameter determined from the micrographs corresponded to that of the diameter measurements obtained by DLS. TEM enabled imaging of liposomes without laborious sample preparation methods that could otherwise lead to a change in the structural and morphological characteristics of the liposomes. It is worthwhile to mention that the liposomes were imaged in their native form i.e. in the buffer with which they were rehydrated; this was enabled by negative staining of the samples, which improved the contrast of the images without necessitating the use of staining agents to stain the liposomal membrane.

Fig. 2. A. TEM micrographs of DS (DSPC:DSPG) liposomes and B. T10 (DSPC:TEL) liposomes. Curcumin concentration is 0.3 mg per 10 mg lipids. Liposomes were negatively stained with uranyl acetate. The liposomal bilayer is shown in the magnified micrograph of the liposome (inset).

3.4 Encapsulation efficiency The results from the encapsulation studies (Table 2) show that the more than 80% of curcumin was encapsulated inside the DSPC:DSPG liposomes. The encapsulation efficiency increases to 90% upon addition of TELs to the liposomal formulation. This could be due to the fact that TEL liposomes readily encapsulate hydrophobic substances but at the same time do no easily release the same [24]. Table 2: Theoretical load of encapsulation efficiency of curcumin loaded liposomes (0.3 mg curcumin per 10 mg lipids), n=3. Formulation DS T10

Theoretical load [µg/ml] ± SD 333.0 333.0

Practical load [µg/ml] ± SD 274.3 ± 19.8 279.2 ± 7.3

%EE 83.5 ± 6.1 91.4 ± 2.4

3.5 Stability of curcumin containing liposomes To investigate the effect of serum on the stability of our liposomes, dynamic light scattering und laser Doppler velocimetry analyses were performed after incubating liposomes in serum. To simulate the physiological conditions, the experiments were carried at 37 °C in a shaking incubator. The results indicate that the liposomes shrank in size after being incubated with serum and 20 mM HEPES (Table 3 and Table 4). This was true for both, DS and T10 liposomes. The PDI indicates that the T10 liposomes were more stable than the DS liposomes, which is evident from the increase in PDI of the latter. This effect was however pronounced after 24 h. The increase in the PDI of the liposomes could be attributed to the decrease in homogeneity in the presence of serum. This is due to the formation of a protein corona around liposomes [38]. Similar to the size, the zeta potential also decreased upon incubation with serum. This decrease in size and zeta potential was independent of the formulation and size of liposomes. The decrease in the liposomal size seems to be due to the osmotic forces, as reported previously [39]. Table 3: Stability of curcumin loaded liposomes (0.3 mg curcumin per 10 mg lipids) incubated for 24 h in serum in a volume ratio of 5:1. Hydrodynamic diameter is expressed as a measure of particle size distribution by intensity; n=3. Formulation DS

T10

Time [h]

Diameter [nm] ± SD

Zeta Potential [mV] ± SD

PDI ± SD

0 1 24 0 1 24

211.30 ± 3.65 171.85 ± 1.99 154.10± 2.05 208.70 ± 2.41 180.70 ± 1.14 173.57 ± 2.15

-22.63 ± 0.21 -14.85 ± 0.21 -13.95 ± 0.64 -8.89 ± 1.66 -5.91 ± 0.87 -3.31 ± 0.15

0.29 ± 0.05 0.37 ± 0.05 0.46 ± 0.03 0.17 ± 0.05 0.19 ± 0.01 0.19 ± 0.05

Table 4: Stability of curcumin loaded liposomes (0.3 mg curcumin per 10 mg lipids) incubated for 24 h in HEPES (pH 7.4) in a volume ratio of 5:1. Hydrodynamic diameter is expressed as a measure of particle size distribution by intensity, n=3. Formulation DS

T10

Time [h]

Diameter [nm] ± SD

Zeta Potential [mV] ± SD

PDI ± SD

0 1 24 0 1 24

211.30 ± 3.65 215.63 ± 1.95 210.60 ± 1.08 208.70 ± 2.41 203.70 ± 3.89 201.80 ± 3.17

-22.63 ± 0.21 -21.56 ± 0.18 -19.62 ± 0.25 -8.89 ± 1.66 -5.87 ± 1.89 -3.59 ± 0.87

0.29 ± 0.05 0.33 ± 0.03 0.31 ± 0.05 0.17 ± 0.05 0.17 ± 0.05 0.21 ± 0.02

3.6 Photocytotoxicity studies Photo-destructive effect of PDT on tumour cells was evident in the SK-OV-3 ovarian carcinoma cells. The DS liposomes have shown the highest photocytotoxicity followed by T10 (Table 5), suggesting that increasing amount of the TELs might have made the liposomal membrane too stable leading to incomplete release of curcumin [28]. This effect could

however be exploited for controlled release and oral delivery of drugs. 1 h incubation time was found to be optimal for all the liposomal formulations. Additional incubation up to 4 h showed marginal increase in the photocytotoxicity. This suggests that much of the internalisation occurred within 1 h of incubation. After a series of optimisation experiments with a radiation fluence of 1.4, 4.3, 8.6 and 13.2 J/cm2 using DS liposomes, maximum photocytotoxicity was observed with a radiation fluence of 13.2 J/cm2 for the LED induced PDT. Since the difference between 8.6 and 13.2 J/cm2 was minimal (Fig. 3), the former was used for further irradiation experiments. It is worthwhile to mention that regardless of the dose, beyond 50 µM, the effect of PDT on the cells was almost the same. Tumour internalisation and localisation of the photosensitiser has a substantial effect on the outcome of the therapy. The results in Fig.3C show a high photocytotoxicity effect of free curcumin at maximum curcumin concentration, dissolved in DMSO. DMSO solvent control did not show an substantial toxicity at the concentrations used in the experiments (data not shown). This indicates that the phototoxicity caused by free curcumin dissolved in DMSO was solely due to curcumin itself. However similar effect could be achieved with much lower amounts of liposomal curcumin. The selectivity of our novel PDT strategy was demonstrated by comparing the results with a primary human cell line PCS-100-020™. No significant difference was observed between the dark and the irradiated line PCS-100-020™ cells indicating that photo-destruction was indeed confined to tumour cells (Fig. 4). Table 5: IC50 values of the phototoxicity induced by curcumin loaded DSPG:DSPC (DS) liposomes, DSPC:TEL (T10) liposomes and free curcumin (dissolved in DMSO) in SK-OV3 cells. The IC50 values were calculated from the graphs and tabulated for respective radiation doses (fluence). Formulation DS

T10

Curcumin (dissolved in DMSO)

Fluence (J/cm2) 3.2 6.1 8.6 10.9 13.2 3.2 6.1 8.6 10.9 13.2 3.2 6.1 8.6 10.9 13.2

IC50 (µM) 16.1 14.7 13.5 12.2 9.8 28.5 14.2 11.6 10.4 8.7 18.5 10.3 10.0 9.1 4.9

Fig. 3. A. Maximum effective radiation fluence optimisation using DS (DSPC:DSPG), B. T10 (DSPC:TEL) liposomes and C. Free curcumin dissolved in DMSO in SK-OV-3 tumour cells. All liposomal formulations contain 0.3 mg curcumin per 10 mg lipids. Cells were incubated for 1 h at 37 °C and were irradiated at different intensities. Dark was used as negative control and represents the cells without irradiation. Data are expressed as the mean ± SD (n=3).

Fig. 4. A. Photocytotoxic effects of DS (DSPC:DSPG), T10 (DSPC:TEL) and free curcumin (dissolved in DMSO) on SK-OV-3 ovarian carcinoma cells and B. PCS-100-020™ normal human primary coronary artery cells. All samples were incubated for 1h and were irradiated at a radiation fluence of 3.2 J/cm2. Dark represents cells without any irradiation. Data are expressed as the mean ± SD (n=3).

3.7 ROS generation ROS studies indicated the generation of cytotoxic ROS after PDT. The amount of ROS generated was analysed by measuring the fluorescence of the lysed cells. DCFDA is

deacetylated intracellularly to an intermediate compound which in turn is oxidised by ROS to form fluorescent DCF [40]. As expected, DS liposomes exhibited the highest ROS generation followed by T10, corresponding to the results of the PDT (Fig. 5). This indicates that DS liposomes caused maximum damage to the tumour cells among the other formulations. ROS generation is the backbone of PDT and is a determining factor in the efficacy of the therapy.

Fig. 5. ROS generation mediated by PDT in SK-OV-3 cells using different liposomal

formulations. The amount of ROS generated was analysed by measuring the fluorescence of the lysed cells. After 45 min incubation with 25 µM DCFDA the cells were incubated for 1h with different curcumin containing liposomes and curcumin (50 µM), TBHP was used as a positive control (50 µM) and were irradiated at 457 nm at a radiation fluence of 8.6 J/cm2. Blank represents untreated cells. Data are expressed as the mean ± SD (n=3).

3.8 CLSM studies To visualise the effect of PDT on tumour cells, qualitative fluorescence microscopic analysis was performed. Substantial intracellular localisation of the curcumin was observed in both dark and irradiated samples which were incubated 1 h with the liposomal formulations suggesting a rapid uptake of liposomal curcumin (Fig. 6). The micrographs confirm the photodestruction induced by PDT wherein nuclear perforation of the irradiated cells is quite evident. This could be attributed to the fact that curcumin upon photo activation induces chromatin condensation and DNA fragmentations which are the main modes of action of curcumin induced PDT [5]. The results obtained were similar across all the liposomes used

Fig. 6. A. CLSM micrographs of SK-OV-3 cells incubated with DS liposomes and B.) T10 liposomes. The curcumin concentration in liposomes was 50 µM and the radiation fluence used was 8.6 J/cm2.

3.9 Photo-destruction of microvasculature PDT induced destruction of microvasculature was evident from the in vivo experiments in the CAM model. A set of controls viz. unloaded liposomes, PBS buffer (pH 7.4) and irradiation alone were used to prove that the photo-destruction of microvasculature was exclusively due to the synergistic effects of curcumin loaded liposomes and PDT. After a delay of around 30 min, destruction of microvasculature and an anti-angiogenesis effect in the CAM was observed in the eggs (Fig. 7). Since the photo-destructive effects exhibited by the different liposomal formulations were similar, we chose DS liposomes for the interpretation of the results. Due to the transparency of the CAM, uptake and localisation of the photosensitiser could be clearly observed more over it enables effective irradiation without necessitating any invasive procedures. A complete destruction of the blood vessels was observed at the site of irradiation signifying the efficacy of this therapy. PDT is known to cause damage to the tumour microvasculature thereby, resulting in cessation of blood flow to the tumours. Besides it is known to cause necrosis of the tumour tissues resulting in anoxia, as reported previously [41].

Fig. 7. A. Photo-destruction of CAM microvasculature before (above) and after (below) PDT with curcumin encapsulated (0.3 mg curcumin per 10 mg lipids) DS (DSPC:DSPG) liposomes, B. T10 (DSPC:TEL) liposomes and C. free curcumin dissolved in DMSO. The area localised with the ring was irradiated with a radiation fluence of 3.6 J/cm2 (λex 405 nm, 4 Hz frequency) using a Weber Needle Endolaser. Scale bar 500 µm.

4. Conclusion The use of novel liposomal formulations in this study enabled the use of higher amounts of curcumin, which would otherwise be impossible due to its poor solubility in aqueous media. Based upon the results obtained from our study, curcumin, which is elsewise safe for normal cells, has apoptotic effect on tumour cells upon irradiation therefore selectively inhibiting tumour growth. Since production of cytotoxic ROS occurs only after irradiation of the intracellular photosensitiser, tumour tissues can be targeted with high precision, thus making this therapy site specific. Together with liposomes, curcumin’s salubrious properties could be further exploited to take this comprehensive therapy to a new horizon thereby, making it a promising contender in the field of PDT. Serum stability test and haemocompatibility studies prove the stability and applicability of the curcumin loaded liposomes intravenously. This could help realise the use of these liposomes for photodynamic therapy in a clinical setup. The LED device used in our study has all the benefits of a conventional laser set up used for PDT, apart from having an added advantage of being portable and economical, meaning the therapy could be realised even in areas where a fullfledged laser set up is otherwise not feasible. Transforming this novel strategy into a safe and effective therapy would be of our prime research interest in future.

Acknowledgements The authors would like to thank Prof. Cornelia M. Keck for reviewing the manuscript, Mrs. Eva Maria Mohr for her help and technical assistance, Ms. Sandra Ditzler for assistance with the haemocompatibility studies and Dr. Gihan Mahmoud for her assistance. The authors are also grateful to Lumundus GmbH for developing the prototype LED device used in this study and to Lipoid GmbH for providing the lipids.

References [1] S. Jaracz, J. Chen, L.V. Kuznetsova, I. Ojima, Recent advances in tumor-targeting anticancer drug conjugates, Bioorg Med Chem, 13 (2005) 5043-5054. [2] L. Brannon-Peppas, J.O. Blanchette, Nanoparticle and targeted systems for cancer therapy, Adv Drug Deliv Rev, 56 (2004) 1649-1659. [3] S.B. Brown, E.A. Brown, I. Walker, The present and future role of photodynamic therapy in cancer treatment, Lancet Oncol, 5 (2004) 497-508. [4] E. Paszko, C. Ehrhardt, M.O. Senge, D.P. Kelleher, J.V. Reynolds, Nanodrug applications in photodynamic therapy, Photodiagnosis Photodyn Ther, 8 (2011) 14-29. [5] J.C. Ahn, J.W. Kang, J.I. Shin, P.S. Chung, Combination treatment with photodynamic therapy and curcumin induces mitochondria-dependent apoptosis in AMC-HN3 cells, Int J Oncol, 41 (2012) 2184-2190. [6] J. Atchison, S. Kamila, C. McEwan, H. Nesbitt, J. Davis, C. Fowley, B. Callan, A.P. McHale, J.F. Callan, Modulation of ROS production in photodynamic therapy using a pH controlled photoinduced electron transfer (PET) based sensitiser, (2015). [7] S. Banerjee, A. Dixit, A.A. Karande, A.R. Chakravarty, Remarkable Selectivity and Photo-Cytotoxicity of an Oxidovanadium(IV) Complex of Curcumin in Visible Light, European Journal of Inorganic Chemistry, 2015 (2015) 447-457. [8] R. Wilken, M.S. Veena, M.B. Wang, E.S. Srivatsan, Curcumin: A review of anti-cancer properties and therapeutic activity in head and neck squamous cell carcinoma, in: Mol Cancer, 2011, pp. 12. [9] S.C. Gupta, S. Patchva, B.B. Aggarwal, Therapeutic Roles of Curcumin: Lessons Learned from Clinical Trials, in: AAPS J, 2013, pp. 195-218. [10] N. Okada, E. Muraoka, S. Fujisawa, M. Machino, Effects of curcumin and capsaicin irradiated with visible light on murine oral mucosa, In Vivo, 26 (2012) 759-764. [11] T. Atsumi, K. Tonosaki, S. Fujisawa, Comparative cytotoxicity and ROS generation by curcumin and tetrahydrocurcumin following visible-light irradiation or treatment with horseradish peroxidase, Anticancer Res, 27 (2007) 363-371. [12] K. Piwocka, E. Jaruga, J. Skierski, I. Gradzka, E. Sikora, Effect of glutathione depletion on caspase-3 independent apoptosis pathway induced by curcumin in Jurkat cells, Free Radic Biol Med, 31 (2001) 670-678. [13] X. Han, B. Xu, C.S. Beevers, Y. Odaka, L. Chen, L. Liu, Y. Luo, H. Zhou, W. Chen, T. Shen, S. Huang, Curcumin inhibits protein phosphatases 2A and 5, leading to activation of mitogen-activated protein kinases and death in tumor cells, (2012). [14] C.F. Chignell, P. Bilski, K.J. Reszka, A.G. Motten, R.H. Sik, T.A. Dahl, Spectral and photochemical properties of curcumin, Photochem Photobiol, 59 (1994) 295-302. [15] J. Ravindran, S. Prasad, B.B. Aggarwal, Curcumin and cancer cells: how many ways can curry kill tumor cells selectively?, Aaps j, 11 (2009) 495-510. [16] C.M. Hu, S. Aryal, L. Zhang, Nanoparticle-assisted combination therapies for effective cancer treatment, Ther Deliv, 1 (2010) 323-334. [17] E. Paszko, G.M. Vaz, C. Ehrhardt, M.O. Senge, Transferrin conjugation does not increase the efficiency of liposomal Foscan during in vitro photodynamic therapy of oesophageal cancer, Eur J Pharm Sci, 48 (2013) 202-210. [18] A. Kunwar, A. Barik, R. Pandey, K.I. Priyadarsini, Transport of liposomal and albumin loaded curcumin to living cells: an absorption and fluorescence spectroscopic study, Biochim Biophys Acta, 1760 (2006) 1513-1520. [19] A. Ozcetin, S. Mutlu, U. Bakowsky, Archaebacterial Tetraetherlipid Liposomes, in: V. Weissig (Ed.) Liposomes : Methods and Protocols, Volume 1: Pharmaceutical Nanocarriers, 2009, pp. 87-96.

[20] U. Bakowsky, U. Rothe, E. Antonopoulos, T. Martini, L. Henkel, H.J. Freisleben, Monomolecular organization of the main tetraether lipid from Thermoplasma acidophilum at the water-air interface, Chem Phys Lipids, 105 (2000) 31-42. [21] C. Abels, R.M. Szeimies, P. Steinbach, C. Richert, A.E. Goetz, Targeting of the tumor microcirculation by photodynamic therapy with a synthetic porphycene, J Photochem Photobiol B, 40 (1997) 305-312. [22] M.W. Reed, F.N. Miller, T.J. Wieman, M.T. Tseng, C.G. Pietsch, The effect of photodynamic therapy on the microcirculation, J Surg Res, 45 (1988) 452-459. [23] P. Nowak-Sliwinska, T. Segura, M.L. Iruela-Arispe, The chicken chorioallantoic membrane model in biology, medicine and bioengineering, Angiogenesis, 17 (2014) 779-804. [24] G. Mahmoud, J. Jedelska, B. Strehlow, U. Bakowsky, Bipolar tetraether lipids derived from thermoacidophilic archaeon Sulfolobus acidocaldarius for membrane stabilization of chlorin e6 based liposomes for photodynamic therapy, Eur J Pharm Biopharm, 95 (2015) 8898. [25] B.-C. Lin, K.-J. Chen, H.-V. Han, Y.-P. Lan, C.-H. Chiu, C.-C. Lin, M.-H. Shih, P.-T. Lee, H.-C. Kuo, Advantages of Blue LEDs With Graded-Composition AlGaN/GaN Superlattice EBL, IEEE Photonics Technology Letters, 25 (2016) 2062-2065. [26] A. Kunwar, A. Barik, K. Indira Priyadarsini, R. Pandey, Absorption and fluorescence studies of curcumin bound to liposomes and lymphocytes effect of γ- irradiation, Board of Research in Nuclear Sciences, India, 2006. [27] K.H. Engelhardt, S.R. Pinnapireddy, E. Baghdan, J. Jedelska, U. Bakowsky, Transfection Studies with Colloidal Systems Containing Highly Purified Bipolar Tetraether Lipids from Sulfolobus acidocaldarius, Archaea, 2017 (2017) 8047149. [28] T.K. Khan, P.L. Chong, Studies of archaebacterial bipolar tetraether liposomes by perylene fluorescence, Biophys J, 78 (2000) 1390-1399. [29] S.R. Pinnapireddy, L. Duse, B. Strehlow, J. Schäfer, U. Bakowsky, Composite liposomePEI/nucleic acid lipopolyplexes for safe and efficient gene delivery and gene knockdown, Colloids and Surfaces B: Biointerfaces, 158 (2017) 93-101. [30] S.R. Pinnapireddy, L. Duse, D. Akbari, U. Bakowsky, Photo-Enhanced Delivery of Genetic Material Using Curcumin Loaded Composite Nanocarriers, Clin Oncol, 2 (2017). [31] J. Senior, G. Gregoriadis, Stability of small unilamellar liposomes in serum and clearance from the circulation: the effect of the phospholipid and cholesterol components, Life Sci, 30 (1982) 2123-2136. [32] B.C. Evans, C.E. Nelson, S.S. Yu, K.R. Beavers, A.J. Kim, H. Li, H.M. Nelson, T.D. Giorgio, C.L. Duvall, Ex Vivo Red Blood Cell Hemolysis Assay for the Evaluation of pHresponsive Endosomolytic Agents for Cytosolic Delivery of Biomacromolecular Drugs, in: J Vis Exp, 2013. [33] A. Ozcetin, A. Aigner, U. Bakowsky, A chorioallantoic membrane model for the determination of anti-angiogenic effects of imatinib, Eur J Pharm Biopharm, 85 (2013) 711715. [34] R. Manjunathan, M. Ragunathan, Chicken chorioallantoic membrane as a reliable model to evaluate osteosarcoma—an experimental approach using SaOS2 cell line, Biol Proced Online, 17 (2015). [35] E.L. Chang, Unusual thermal stability of liposomes made from bipolar tetraether lipids, Biochem Biophys Res Commun, 202 (1994) 673-679. [36] M.A. Dobrovolskaia, S.E. McNeil, Handbook of Immunological Properties of Engineered Nanomaterials (Frontiers in Nanobiomedical Research), World Scientific Publishing Company, 2013. [37] M. Wintrobe, Wintrobe's Clinical Hematology, 9 ed., Lea & Febiger, Philadelphia, 1993. [38] P.d. Pino, B. Pelaz, Q. Zhang, P. Maffre, G.U. Nienhaus, W.J. Parak, Protein corona formation around nanoparticles – from the past to the future, (2013).

[39] J. Wolfram, K. Suri, Y. Yang, J. Shen, C. Celia, M. Fresta, Y. Zhao, H. Shen, M. Ferrari, Shrinkage of pegylated and non-pegylated liposomes in serum, Colloids Surf B Biointerfaces, 114 (2014) 294-300. [40] H.R. McLennan, M. Degli Esposti, The contribution of mitochondrial respiratory complexes to the production of reactive oxygen species, J Bioenerg Biomembr, 32 (2000) 153-162. [41] M.W. Reed, T.J. Wieman, D.A. Schuschke, M.T. Tseng, F.N. Miller, A comparison of the effects of photodynamic therapy on normal and tumor blood vessels in the rat microcirculation, Radiat Res, 119 (1989) 542-552.

Graphical abstract