DL-Buthionine-Sulfoximine

Chlorin e6 Conjugated Methoxy-Poly(Ethylene Glycol)-Poly(D,L- Lactide) Glutathione Sensitive Micelles for Photodynamic Therapy

Preeti Kumari • Milan Paul • Himanshu Bhatt • Sri Vishnu Kiran Rompicharla • Debolina Sarkar • Balaram Ghosh • Swati Biswas
1 Department of Pharmacy, Birla Institute of Technology & Science-Pilani, Hyderabad Campus, Jawahar Nagar, Medchal Hyderabad, Telangana 500078, India

ABSTRACT
Purpose
In this study, we developed a polymeric micellar system for glutathione-mediated intracellular delivery of a photosensitizer, chlorin e6 (Ce6) by synthesizing an amphi- philic polymer, methoxy-poly(ethylene glycol)-poly(D,L-lac- tide)-disulfide-Ce6 (mPEG-PLA-S-S-Ce6), which self- assembled in aqueous environment to form micelles.
Methods
The polymer-drug conjugate was characterized by NMR. The singlet oxygen (2O1) generation and in vitro release of Ce6 micelles were evaluated. Further, glutathione- mediated intracellular drug delivery was assessed in human alveolar adenocarcinoma cells (A549), mouse mammary car- cinoma cells (4 T1) and 3D A549 spheroids.
Results
The micellar system protected Ce6 from aggregation leading to improved 2O1 generation compared to free Ce6. Due to the availability of glutathione, the disulfide bonds in the micelles were cleaved resulting in rapid release of Ce6 evident from the in vitro study. The Ce6 micelles displayed quicker drug release in presence of glutathione monoester (GSH-OEt) pre-treated A549 and 4 T1 cells compared to without pre-treated cells. In vitro phototoxicity of micelles dis- played enhanced toxicity in 10 mM GSH-OEt pre-treated A549 and 4 T1 cells compared to untreated cells. As antici- pated, Ce6 micelles showed lower drug release in presence of 0.1 mM of buthionine sulfoximine (BSO) pretreated A549 and 4 T1 cells exhibiting lower phototoxicity. Further, A549 3D spheroids treated with Ce6 micelles showed significant inhibition in growth, enhanced phototoxicity, and cellular ap- optosis in comparison to free Ce6.
Conclusion
The above results showed that the developed strategy could be effective in improving the PDT efficacy of Ce6, and the developed polymeric micellar system could be utilized as a PDT regimen for cancer.

INTRODUCTION
In recent years, photodynamic therapy (PDT) has been widely utilized for the treatment of cancer. PDT involves the use of photosensitizers (PSs) that get activated by specific wavelength of light resulting in sequence of photochemical events causing phototoxicity (1). PSs absorb the photons that results in their excitation from ground singlet state to excited singlet state, which is further promoted to longer-lived excited triplet state by inter-system crossing. The excited triplet state perform reactions of two types, Type I and Type II (2,3). The reaction designated as type I occurs mainly at low concentration of oxygen or in presence of polar environment (4). However, it has been seen that Type II reaction prevails throughout the photodynamic therapy and singlet oxygen (2O1) is primarily considered to be cytotoxic causing cellular damages. Even though PSs are considered to be highly cytotoxic, however, hydrophobicity of the porphyrin moieties, restricted tissue penetration, requirement of specific wavelength of light source to trigger the activity of PSs, need for the presence of oxygen and non-specific localization of photosensitizers to non-tumor tissues are few of limitations hampering its successful usage for the treatment of cancer.
Various nanocarriers have been developed over the years for the delivery of photosensitizers such as polymeric micelles, liposomes, nanogels, dendrimers, and polymer-drug conjugates (5–8). Polymeric micelles have been widely ex- plored for drug delivery applications (9). These nanocarriers accumulate in solid tumors via enhanced permeation and re- tention (EPR) effect, which improves the therapeutic efficiency of the loaded cargo materials manifold (10). In spite of having the advantage of EPR effect, the untimely release of drug in the systemic circulation provides drug access to unwanted sites giving rise to toxicity (11,12). Further, concentration of drug reaching the target site is poor enabling to administer high doses leading to adverse side effects.
In the recent years, extensive attempts have been made in the progress of stimuli-sensitive nanocarriers that can initiate the release of drugs in response to particular cell signals (13,14). Nanocarriers bearing disulfide bonds, which can be cleaved by the intracellular glutathione (GSH) have been ex- plored for intracellular drug release (15,16). GSH, a thiol- containing reducing tripeptide cleaves the disulfide bonds in the cytoplasm. The intracellular concentration of GSH (~10 mM) is significantly elevated compared to extracellular fluids (~2 μM) (17). Consequently, this huge variation in GSH concentration provides a platform to design a stimuli-sensitive nanocarrier system to deliver drugs intracellularly and de- creasing the undesired side effects.
In this study, we synthesized an amphiphilic polymer,Ce6-conjugated methoxy-poly(ethylene glycol)- poly(D,L-lactide) (mPEG-PLA), where Ce6 is linked to mPEG-PLA by disulfide bond to form mPEG-PLA-S-S- Ce6, which self-assembles to form polymeric micelles in aqueous environment. The rationale behind this devel- oped micellar drug conjugate is to improve solubility of Ce6, its circulation stability by reducing its aggregation, selective accumulation to cancer cells via EPR effect and stimuli-sensitive drug release inside the cancer cells thus enhancing the therapeutic efficacy. Additionally in our previous work, we had developed Ce6 conjugated methoxy-poly(ethylene glycol)-poly(D,L-lactide) (mPEG- PLA) nanoparticles to improve the pharmacokinetic properties of Ce6 (18). The delivery system was effective in improving the dissolution and permeation of Ce6, de- creased its aggregation, and thereby, improved its pho- todynamic therapeutic effect. In this work, we have in- troduced stimuli-sensitivity in the nanocarrier system for faster drug release and enhanced therapeutic effect. The micelles were characterized for particle size, polydisper- sity index, zeta potential, generation of singlet oxygen, and in vitro release profile in presence of reduction stim- ulus. The GSH-mediated intracellular drug delivery was studied against mouse mammary carcinoma cells (4 T1) and human alveolar adenocarcinoma cells (A549) by pre- treatment of the cells with either buthionine sulfoximine (BSO) or, glutathione monoester (GSH-OEt) thus in- creasing or decreasing the GSH concentration in the cytoplasm, respectively. Further, growth inhibition, penetration efficiency, phototoxicity of micelles were assessed in a 3D spheroidal tumor model of A549 lung cancer cells.

MATERIALS AND METHODS
Materials
Methoxy poly(ethylene glycol) 5000, Glutathione (GSH), D,L- Lactide, Glutathione monoester (GSH-OEt), DAPI (4,6- dia- midino-2-phenylindole), Buthionine sulfoximine (BSO), 9,10- dimethylanthracene (DMA), Anhydrous dimethyl sulfoxide (DMSO) , 3 ,3 ′ – D i t hi od ip ro pi on ic a c id, N – Hydroxysuccinimide (NHS), propidium iodide (PI), N-Boc ethylene diamine, 1-Ethyl-3-(3-dimethylaminopropyl)-carbo- diimide (EDC), Trifluoroacetic acid, para-formaldehyde, and tetrahydrofuran were purchased from Sigma-Aldrich, Bangalore, India. The Ce6 was obtained from Frontier Scientific, Inc. (USA). The dialysis membranes were obtained from Spectrum Laboratories (USA). Annexin V-FITC/PI de- tection kit was procured from Invitrogen Co. (USA). Calcein Blue AM, and Singlet oxygen sensor green (SOSG) was pro- cured from Thermo Fisher Scientific (USA).
Accutase™, Fluoromount-G, Dulbecco’s Modified Eagle’s Medium (DMEM), Penicillin-Streptomycin, trypsin-EDTA, and heat-inactivated fetal bovine serum (FBS) were purchased from Himedia Labs (India).
Human alveolar adenocarcinoma cells (A549) and mouse mammary carcinoma cells (4 T1) were procured from National Center for Cell Sciences (Pune, India). All culture media were added with 1% penicillin-streptomycin solution, and 10% heat-inactivated fetal bovine serum. The A549 cells were kept in the incubator at 37°C with containing 5% CO2.
Synthesis of Disulfide Linked Methoxy-Poly(Ethylene Glycol)-Poly(D,L-Lactide)-Chlorin e6 Conjugate (mPEG-PLA-S-S-Ce6)

Synthesis of mPEG-PLA
The di-block copolymer methoxy-poly(ethylene glycol)- poly(D,L-lactide) (mPEG-PLA) were prepared through the reaction of ring opening polymerization following the previ- ous reported process (19). In brief, 2 g of mPEG, 0.8 g of D,L- lactide and 0.008% w/w of stannous octoate were dissolved in toluene (6 mL) and they were added into a tube and kept for stirring at 160°C for 6 h. The reaction mixture was evaporat- ed, and the crude product was mixed with tetrahydrofuran, further precipitated by ice-cold diethyl ether. The precipitated product was kept for drying, dissolved in water, and dialyzed against distilled water by dialysis membrane (MWCO 12–14,000 Da, Spectrum Laboratories, USA) for 24 h and then, lyophilized to obtain white fluffy powder.

Synthesis of mPEG-PLA-S-S-COOH
3,3′-Dithiodipropionic acid (8.76 mg) was reacted with N- Hydroxysuccinimide (NHS) (4.7 mg) and 1-Ethyl-3-(3-dime- thylaminopropyl)-carbodiimide (EDC) (7.8 mg) in dimethyl- formamide (DMF; 2 mL) for 1 h in dark at room temperature (RT). mPEG-PLA (500 mg) in DMF (2 mL) was added drop- wise into the above solution and kept on stirring for 12 h at RT in dark. The unreacted material was removed by dialysis against distilled water by dialysis membrane (MWCO 12– 14,000 Da). Further, lyophilized to obtain pure mPEG- PLA-S-S-COOH.

Synthesis of NH2-Terminated Ce6
Ce6 (100 mg) was reacted with EDC (64.21 mg), and NHS (38.53 mg) in DMF (2 mL) for 1 h in dark at RT. N-Boc ethylene diamine (29.53 mg) dissolved DMF (1 mL) was added dropwise into the above solution. The reaction was continued for overnight at RT in dark. The product was treated with trifluoroacetic acid: dichloromethane (TFA:DCM) in a ratio of 3:7 at RT for 2 h. After that the volatiles were removed in vacuum. The mixture was neutralized and fractioned with saturated solution of sodium bicarbonate (NaHCO3) and eth- yl acetate (EtOAc). The organic layer was dried and concen- trated under reduced pressure to get the product.

Synthesis of mPEG-PLA-S-S-Ce6
To the solution of mPEG-PLA-S-S-COOH (100 mg) in DMF (2 mL), NHS (2.2 mg) and EDC (3.7 mg) was added for 1 h. The reaction mixture was stirred for 4 h in dark at RT before addition of Ce6-NH2 (51.37 mg) dissolved in DMF dropwise. The reaction mixture was kept on stirring for 12 h at RT in dark. The reaction mixture was purified against distilled water using a dialysis membrane (MWCO 3500 Da) and then lyophilized to obtain final polymer mPEG-PLA-S-S- Ce6. The compound was characterized using 1H NMR, Bruker spectrometer (AVANCE model, Germany) working at 300 MHz at RT. The sample was dissolved in CDCl3 for NMR at concentration of 5 mg/mL.

Fourier-Transform Infrared Spectroscopy (FTIR) Analysis
The lyophilized mPEG-PLA-S-S-Ce6 was accurately weighed and mixed with KBr to make the pellet. The pellet of lyoph- ilized mPEG-PLA-S-S-Ce6 was analyzed by FTIR spectrom- eter (FT/IR-4200, Jasco, USA) individually. The spectrum was recorded in range from 4000 to 400 cm−1. The final spectrum of the mPEG-PLA-S-S-Ce6 was assessed to observe the characteristic peaks.

PREPARATION AND CHARACTERIZATION OF THE MPEG-PLA-S-S-CE6
REDOX-RESPONSIVE MICELLES
The synthesized mPEG-PLA-S-S-Ce6 conjugate, an am- phiphilic polymer has ability to self- assemble, forms mPEG-PLA-S-S-Ce6 micelles in aqueous solution. The polymer dissolved in DMSO at 10.0 mg/mL was stirred for 20 min. Then, 5 mL of distilled water was added dropwise to the preceding solution and stirred continu- ously for another 4 h at RT. Finally, the solution was transferred to dialysis membrane (MWCO 3500 Da, Spectrum Laboratories, USA) and dialyzed against dis- tilled water for 48 h to form micelles.

Critical Micelle Concentration (CMC)
Critical micelle concentration of the mPEG-PLA-S-S-Ce6 conjugate was evaluated utilizing pyrene as a hydrophobic fluorescence probe (20). Pyrene solution (50 μL; 10 mg/mL in chloroform) was mixed with micellar solution in a series of concentration ranging from 3.125 to 100 μg/mL. The mix- ture was kept on overnight stirring for encapsulation of pyrene in the core of micelles in dark. Further, the solutions were filtered and analyzed using microplate reader (Spectramax™, Molecular Devices, USA) at an excitation wavelength of 339 nm with the slit-width of both excitation and emission set at 5 nm. The CMC was determined as the inflection point. The graph was plotted using log concentra- tions of mPEG-PLA-S-S-Ce6 conjugate and the intensities of fluorescence at ratio of I373/I383 at X and Y axis, respectively.

Dynamic Light Scattering (DLS) Measurements
mPEG-PLA-S-S-Ce6 micelles were investigated for particle size, polydispersity index, and surface potential by dynamic light scattering technique (Nano ZS90, Malvern Instruments Ltd., UK). The particle size was determined using cell (dia. 10 mm) at 25°C.

Transmission Electron Microscopy
The morphology of mPEG-PLA-S-S-Ce6 micelles was exam- ined by transmission electron microscopy (TEM). Micelles were stained with 2% uranyl acetate and de-stained with dis- tilled water. The sample were kept on copper grids with films, dried for 10 min, and viewed by TEM (TEM, JEM-1200EX, JEOL, Tokyo, Japan).

Evaluation of Singlet Oxygen Generation (SOG)
Generation of singlet oxygen (1O2) by mPEG-PLA-S-S-Ce6 was determined using 9,10-dimethylanthracene (DMA). The change in fluorescence intensity of DMA upon reaction with 1O2 was determined using fluorescent spectroscopy as men- tioned previously (21). The mPEG-PLA-S-S-Ce6 (1.5 mg/ mL; water) was mixed with DMA (20 mM) and kept aside for 10 min. The samples were treated with or without 10 mM GSH. After incubation, the sample solution was illu- minated with 633 nm laser with power of 50 mW/cm2 (UniRAM Micro Raman Systems, Spectrolab Systems Limited, United Kingdom). The irradiated samples formed non-fluorescent product, 9,10 endoperoxide which caused de- crease in the fluorescence intensity of DMA (Ex. 360 nm; Em. 380–550 nm) as determined at 10 s interval using a fluores- cence spectrophotometer.
Further, the singlet oxygen generation by mPEG- PLA-S-S-Ce6 in distilled water was determined using singlet oxygen sensor green (SOSG) following reported procedure (22,23). Briefly, 2.5 mM of SOSG was mixed with mPEG-PLA-S-S-Ce6, which is treated with or without 10 mM of GSH. The mixture was irradiated using 633 nm laser source. The irradiated sample gen- erated 1O2 which reacted with SOSG producing SOSG endoperoxide (SOSG-EP). The fluorescence of SOSG- EP (Ex. 510 nm; Em. 525–536 nm) was measured using fluorescence spectrophotometer.

In Vitro Ce6 Release from mPEG-PLA-S-S-Ce6 Micelles
The release kinetics of Ce6 from mPEG-PLA-S-S-Ce6 micelles was determined using a dialysis method (24). 1 mL of mPEG-PLA-S-S-Ce6 micelles containing 30 μg/mL of Ce6 was put into a dialysis membrane (MWCO 3500 Da), dipped in the phosphate buffered saline (40 mL, PBS; pH 7.4) having 0.1% (w/v) Tween 80 for maintaining the sink condi- tion. The dissolution medium was supplied with no GSH, 2 μM GSH or 10 mM GSH. The solutions were kept on shaking at 150 rpm at 37 ± 0.5°C. At predetermined time points, 1 mL of the dissolution medium was removed and replaced with the same volume of fresh dissolution medium. The drug release of free Ce6 and mPEG-PLA-S-S-Ce6 micelles in release medium without GSH was measured as control, and the procedure was the same as described above. The cumulative amount of Ce6 released from mPEG-PLA-S- S-Ce6 micelles was determined by UV-Visible spectropho- tometer at λmax 405 nm.

Cell Experiments
Cell Culture
Human alveolar adenocarcinoma cells (A549) and mouse mammary carcinoma cells (4 T1) were cultured in DMEM medium added with 1% penicillin-streptomycin and 10% heat-activated fetal bovine serum (FBS). Cells were kept in a humidified environment with 5% CO2 at 37°C and passaged every 2–3 days.

Cellular Uptake Study of mPEG-PLA-S-S-Ce6 Micelles
The cellular uptake of mPEG-PLA-S-S-Ce6 micelles were observed by confocal microscope (Leica Microsystems, Germany) qualitatively and quantified using flow cytometer. A549 and 4 T1 cells were seeded on round cover-slips placed in 12-well plates (5 X 104 cells/well) and incubated overnight. On the following day, the cells were incubated with free Ce6 and mPEG-PLA-S-S-Ce6 micelles at 5 μg/mL of Ce6 concentration for 1 h and 4 h. After incubation, the treated cells were rinsed with cold PBS (pH 7.4). Cell nuclei were stained with DAPI for 5 min in dark. The cells were fixed with 4% para-formaldehyde for 15 min at 25°C. The slides were mounted using Fluoromount G as the mounting media and visualized under confocal microscope. DAPI and Ce6 were excited using 358 nm and He-Ne laser, and 461 and 650 nm emission filter was used to detect the fluorescence, respective- ly. The confocal images collected were evaluated using Image J software.
For flow cytomertic measurements, A549 and 4 T1 cells were plated in 6-well plates (5 X 105 cells/well) and incubated overnight. Next day, the cells were treated with free Ce6 and mPEG-PLA-S-S-Ce6 micelles at Chlorin e6 concentration of 5 μg/mL in complete medium for 1 and 4 h, respectively. After both the time points, the cells were rinsed with cold phosphate buffer saline, trypsinized, and re-suspended in PBS, pH 7.4 (200 μL). Data for 10,000 gated events were collected and analysis was performed using flow cytometer (Amnis Flowsight, United States) and IDEAS software.

Redox-Dependent Intracellular Chlorin e6 Release from mPEG-PLA-S-S-Ce6 Micelles
The redox-sensitive release of Ce6 from mPEG-PLA-S-S-Ce6 micelles was determined using confocal microscopy and flow cytometer as described above. The A549 and 4 T1 cells in 12- well plates were pre-treated with 10 mM GSH-OEt for 2 h. The cells were rinsed with PBS (pH 7.4), treated with mPEG- PLA-S-S-Ce6 for 4 h. After incubation, cells were rinsed with PBS (pH 7.4), stained with DAPI for 5 min in the dark and treated with 4% para-formaldehyde at 25°C for 15 min. The slides were mounted using Fluoromount G and observed un- der a confocal microscope.
For flow cytometer analysis, A549 and 4 T1 cells in 6-well plates were pre-treated with 10 mM GSH-OEt for 2 h. After 2 h, cells were rinsed with PBS (pH 7.4), treated with mPEG- PLA-S-S-Ce6 for additional 4 h. The cells were rinsed with PBS, harvested, and pellet was suspended again in 200 μL of phosphate buffer saline, pH 7.4 for flow cytometry analysis.

In Vitro Phototoxicity of mPEG-PLA-S-S-Ce6
The in vitro phototoxicity assay of free Ce6 and mPEG-PLA-S- S-Ce6 were performed on A549 and 4 T1 cells. In particular, 5 X 104 cells/well were plated in 96-well plates and kept overnight at 37°C for cell adhesion. Then, cells were incubat- ed with GSH-OEt (10 mM) for 2 h or BSO (0.1 mM) for 12 h. Without pre-treatment, cells were taken as control. After dis- carding the GSH-OEt or BSO, cells were washed with PBS, 100 μL of a serial concentration of free Ce6 and mPEG-PLA- S-S-Ce6 were added to each well, and kept in incubator for 12 h. After incubation interval, the cells were illuminated with 633 nm laser at power of 50 mW/cm2 for 2 min. Irradiated cells were incubated further for 12 h in dark at 37°C. Thereafter, the culture medium was discarded, and 50 μL of 5 mg/mL MTT in serum/phenol red free medium was added to each well.
After 3 h incubation period, the MTT solution was dis- carded and 150 μL of DMSO was added to each well to solubilize the purple formazan crystals formed through reduc- tion of MTT by the mitochondria of live cells. The absor- bance was determined at 570 nm and subtracted from the background absorbance taken at 630 nm. Cell viability was calculated using below equation:
Cell viability ð%Þ ¼ ABSsample=ABScontrol × 100
where, ABSsample = the absorbance of the transformed MTT in formulation treated cells and ABScontrol = the absorbance of transformed MTT in medium incubated cells (positive control).

DNA Fragmentation Assay
A549 cells were seeded into 6 well assay plates at cell density 1× 105/well and kept overnight at 37°C. Next day, the cells were treated with mPEG-PLA-S-S-Ce6 micelles and free Ce6 (Ce6 concentration. 50 μm). Following an incubation period of 12 h, the treated cells were irradiated for 5 min using 663 nm laser source at intensity of 50 mW/cm2. The cells were further incubated for 24 h. The DNA was extracted from cells using GSure® DNA genomic isolation kit and resolved on 1.5% gel containing 0.5 μg/ml ethidium bromide. The fragmentation was visualized using Fusion Pulse gel doc sys- tem (Vilber, Germany).

Nuclear Staining
Chromatin condensation is a morphological hallmark of apo- ptosis. For nuclear staining, A549 cells with cell density of 1 × 105 / well were grown in 6 well plate. The following day, the cells were treated with mPEG-PLA-S-S-Ce6 and free Ce6 at Ce6 concentration of 50 nM and incubated for 12 h. Next, the treated cells were irradiated for 5 min with 663 nm laser source at intensity 50 mW/cm2 and kept for incubation for another 12 h. The cells were washed with PBS, fixed, treated with DAPI (10 μg/ml) and acridine orange (0.1 mg/ml), and visualized using fluorescence microscope.

Avascular A549 Spheroids Experiments
Avascular A549 Spheroids Culture
Avascular A549 spheroid model was prepared by liquid over- lay method according to previous reports (25). Briefly, the 96- well plates were treated with 50 μL of agar solution (1.5% (w/ v) made in serum free Dulbecco’s Modified Eagle’s Medium. A549 cells (1 X 104/well) were seeded in 96-well plates, centri- fuged at 1500 rcf for 15 min at 25°C. The spheroids of 400– 500 μm size obtained after 4–5 days were utilized for the study.

Penetration of Micellar Ce6 in A549 Spheroids
The penetration efficiency of mPEG-PLA-S-S-Ce6 and free Ce6 in A549 spheroids was assessed following 1 and 4 h of incubation with mPEG-PLA-S-S-Ce6 and free Ce6 at Ce6 concentration of 5 μg/mL by confocal microscope. The Z- stack images of treated A549 spheroids were taken at 10X magnification at intervals of 10 μm.

Internalization of mPEG-PLA-S-S-Ce6 in Spheroids
The A549 3D spheroids were treated with mPEG-PLA-S-S- Ce6 as well as free Ce6 at 5 μg/mL of Ce6 concentration for 1 h and 4 h, respectively. The spheroids were rinsed thorough- ly with phosphate buffer saline (pH 7.4), mixed with 50 μL of Accutase™ solution and shaken moderately for 10 min at 37°C. After 10 min, the cell suspension of spheroids was col- lected in 15 mL tubes. The activity of Accutase™ was neu- tralized using FBS (500 μL). 10 spheroids suspension were pooled together for flow cytometer analysis. The fluorescence of Ce6 was captured using flow cytometer.

Growth Inhibition of Spheroids
The growth inhibition of A549 spheroids were evaluated by incubating the spheroids with mPEG-PLA-S-S-Ce6 and free Ce6 at 3 μg/mL of Ce6 concentration for 4 h. The treated spheroids were illuminated with 633 nm wavelength laser (50 mW/cm2) for 2 min. The experiment was executed in dark condition. The size of spheroids were captured every 2 days using fluo- rescence microscope (Leica Microsystems, Germany) with 10X magnification. The control spheroids were incubated in complete medium. The images represent the diameter of spheroid ± SD.

Live/Dead Cell Assay in Spheroids
The live/dead assay was carried out in spheroids using Calcein Blue AM (stains live cells blue) and propidium iodide (PI) (stains dead cells red). The spheroids were treated for 4 h with mPEG-PLA-S-S-Ce6 and free Ce6 at Ce6 concentration of 3 μg/mL. Then, the treated spheroids were kept for irradi- ation using 633 nm laser at intensity of 50 mW/cm2 for 15 min. Thereafter, the medium was removed, washed with phosphate buffer saline (pH 7.4), and kept for incubation for additional 4 h in fresh medium. After incubation, spheroids were stained with 2 μM of calcein blue AM and 4 μM propi- dium iodide, kept for 20 min at 37°C. The stained spheroids were observed using fluorescence microscope.

In Vitro Phototoxicity in Spheroids
In vitro phototoxicity was measured with Presto blue reagent as reported by the manufacturer. Briefly, the A549 spheroids were treated with a serial concentration of mPEG-PLA-S-S-Ce6 and free Ce6, and incubated for 12 h. Next, the treated cells were illuminated using 633 nm laser (50 mW/cm2) for 2 min. The irradiated cells were further incubated for 12 h in dark at 37°C. Then, the A549 spheroids were rinsed with phosphate buffer saline and incubated with Accutase™ solution (50 μL) and were shaken moderately for 10 min at 37°C. The cell suspension of the spheroids was collected in 15 mL sterile tubes. Accutase™ activity was neutral- ized by adding 500 μL of fetal bovine serum. Further, cells were kept in the centrifugation for 5 min at 1000 rpm. The resulting cell pellet was again suspended in 90 μL of culture medium and 10 μL of Presto Blue reagent, and kept for incubation for 2 h. The UV ab- sorbance was measured at 570 nm using the reference standard (600 nm). Data was produced in triplicates with each triplicate comprising of ten spheroids.

Apoptosis in the Spheroid Cells
The apoptosis study was performed employing Annexin V assay to determine the apoptotic potential of Ce6. The sphe- roids having size range of 40–500 μm were treated for 12 h with mPEG-PLA-S-S-Ce6 and free Ce6 at Ce6 concentration of 50 nM. After incubation, the treated spheroids were irra- diated for 30 min with 633 nm laser source at intensity of 50 mW/cm2 and again they were kept for 12 h. Thereafter, spheroids were trypsinized, rinsed with cold binding buffer, and mixed with Accutase™ solution (100 μL). The suspension of spheroidal cells were centrifuged, pellet was re-suspended in Annexin V buffer (100 μL), and stained with AnnexinV-FITC (5 μL) and propidium iodide (10 μL). The cell suspension was kept in dark for 15 min. Further, the cell suspension was di- luted with Annexin V binding buffer to 500 μL volume and analyzed by flow cytometer.

Statistical Analysis
The results were evaluated statistically by Student’s t test to assess the significance between all the groups. p values were determined by Graph Pad prism 5 software (GraphPad Software, California, USA). The data was represented as mean ± standard deviation, n = 3 or 4, from 3 set of experi- ments. The p < 0.05 was denoted statistically significant. *, **, *** in figures represented p values <0.05, 0.01 and 0.001, respectively. RESULT AND DISCUSSION Synthesis and Characterization of Disulfide Linked Methoxy-Poly(Ethylene Glycol)-Poly(D,L-Lactide)-Chlorin e6 Conjugate (mPEG-PLA-S-S-Ce6) Scheme 1 shows the synthesis scheme of mPEG-PLA-S-S-Ce6 conjugate. The non-ionic amphiphilic copolymer, mPEG- PLA was prepared using ring-opening polymerization reac- tion according to previous procedure (19). The mPEG-PLA copolymer was reacted with pre-activated 3,3 ′- Dithiodipropionic acid to yield mPEG-PLA-S-S-COOH. An amino group (-NH2) was added to the Ce6 using N-Boc ethylene diamine by a two-step process resulting in the forma- tion of Ce6-NH2. The Ce6-NH2 was then coupled with pre- activated carboxylic acid group of mPEG-PLA-S-S-COOH to yield the final product, mPEG-PLA-S-S-Ce6. The 1H NMR spectra of mPEG-PLA-S-S-Ce6 was shown in Fig. 1. The distinct signals at 1.68 and 5.19 ppm correspond to protons of PLA whereas, the signal at 3.65 ppm confirms the presence of PEG moiety. The peaks at 1.4–1.8, 2.9, and 4.2–4.3 ppm belong to the protons of Ce6, which indicates that the Ce6 was conjugated to mPEG-PLA successfully. Scheme 1 Synthesis procedure of mPEG-PLA-S-S-Ce6. i. EDC/ NHS/ DIPEA, Room temperature (RT); ii. Trifluoroacetic acid, RT; iii. Sn(Oct)2 / 160°C FTIR spectral analysis indicated the successful synthe- sis of mPEG-PLA-S-S-Ce6 by the reaction between the amine groups of Ce6 and the carboxylic groups of the mPEG-PLA-S-S, which resulted in C=O-N-H bond for- mation. The FTIR spectrum of mpEG-PLA-S-S-Ce6 has been represented in Fig. S1. The presence of the peak at 1601 cm−1 (C=O stretching amides) and defor- mation of the signal at 3437 cm−1 (-NH stretching amide) proved the formation of (C-NH) amide bond. The presence of the peaks at 845 cm− 1 and 1276 cm−1 proved the S-S bisulfide bond formation and presence of CSH stretching. Characterization of mPEG-PLA-S-S-Ce6 Micelles The physicochemical properties of the mPEG-PLA-S-S-Ce6 micelles such as particle size, zeta potential, morphology, redox-responsive behavior, and CMC were determined. The amphiphilic nature of mPEG-PLA-S-S-Ce6 provides a chance to form micelles via self-assembly in aqueous solution. The hydrophobicity of the mPEG-PLA-S-S-Ce6 micelles was evaluated by pyrene as a fluorescent probe (26). In the aque- ous environment, pyrene exhibits weak fluorescence intensity due to its self-quenching and poor solubility. However, pyrene intensity significantly increases due to its encapsulation inside the hydrophobic core of the micelles in aqueous environment. Pyrene is most frequently utilized as a fluorescence probe due to its effectiveness in determining the self-aggregation proper- ty of polymers or surfactants. In Fig. 2a, with increase in con- centration of conjugate, fluorescence intensity increased sig- nificantly, indicating the CMC concentration at 25 μg/mL. The low CMC concentration reveals that self-assembled micelles formed easily and maintained stability in dilute condition. The size of the micelles in the range of 20–200 nm are advantageous to maintain lower level of reticuloendothelial system (RES) internalization, slight renal excretion, and accu- mulation in solid tumor through EPR effect (27,28). Evaluation of morphology by TEM revealed that morphology of micelles were almost spherical with low polydispersity (Fig. 2c). As shown in Fig. 2b, the particle size of the mPEG-PLA-S- S-Ce6 micelles was 177.96 ± 2.64 nm with a polydispersity index of 0.259 ± 0.038. The zeta potential was found to be −12.3 ± 0.97 mV. Evaluation of Singlet Oxygen Generation (SOG) The generation of singlet oxygen (1O2) indicates the photo- toxicity, along with quenching and activation of the conjugated Ce6 in the micelles (29). In this work, we used 9,10-dimethylanthracene (DMA) as an efficient chemical trap for 1O2 (30). The mPEG-PLA-S-S-Ce6 with or without 10 mM GSH was irradiated using 633 nm laser source. The change in DMA fluorescence intensity due to generation of 1O2 was examined. As shown in Fig. 3a, mPEG-PLA-S-S-Ce6 with GSH generated higher 1O2 than mPEG-PLA-S-S-Ce6 without GSH. The free Ce6 generated less 1O2 due to its aggregation in aqueous environment. Further, 1O2 generation potential of Ce6 in mPEG-PLA- S-S-Ce6 conjugate treated with or without GSH was evaluat- ed using singlet oxygen sensor green (SOSG). The fluores- cence of SOSG in mPEG-PLA-S-S-Ce6 without treatment with GSH was lower compared to GSH treated mPEG- PLA-S-S-Ce6, suggesting the instantaneous presence of active Ce6 for the generation of 1O2, which was possible by cleavage of the disulfide bond in presence of GSH (Fig. 3b). These data reveals that the presence of disulfide bond can be easily ma- nipulated to enhance the phototoxicity, which can be utilized for the intracellular GSH-activated Ce6-mediated photody- namic therapy. The free Ce6 photo-activity was inhibited in aqueous environment owing to its aggregation triggered by pie-pie interactions between neighboring aromatic rings of free Ce6 resulted in self-quenching in the molecule. In Vitro Ce6 Release from mPEG-PLA-S-S-Ce6 Micelles The in vitro release profile of Ce6 from the mPEG-PLA-S-S- Ce6 micelles was determined in PBS (pH 7.4) at 37°C in the presence and absence of 10 mM GSH. As shown in Fig. 3c, it was observed that the free Ce6 was completely diffused through the dialysis bag within 20 h. Without GSH treatment, the cumulative release of Ce6 from mPEG-PLA-S-S-Ce6 micelles reached 9.85% in 24 h. At extracellular GSH con- centration (2 μM), the Ce6 release was almost similar com- pared to the micelles without treatment with GSH. However, in presence of 10 mM GSH, Ce6 was released rapidly from mPEG-PLA-S-S-Ce6 as about 57.52% of Ce6 was released in 24 h. These results indicate that mPEG-PLA-S-S-Ce6 micelles could efficiently retard Ce6 release in absence of GSH, whereas Ce6 could be released quickly in presence of GSH at intracellular concentration. It can be due to the cleav- age of disulfide bond which accelerated the release of drug. Cellular Internalization of mPEG-PLA-S-S-Ce6 Micelles A time-dependent (1 and 4 h) cellular uptake of mPEG-PLA- S-S-Ce6 micelles on A549 and 4 T1 cells were carried out by flow cytometry and confocal microscopy (Figs. 4 and 5). Figures 4a and 5a illustrated the confocal microscope images of A549 and 4 T1 cells treated with free Ce6 and mPEG-PLA- S-S-Ce6 micelles for 1 and 4 h. It was observed that mPEG- PLA-S-S-Ce6 micelles were internalized by A549 and 4 T1 cells as seen by green fluorescence within the cells. The geometric mean of Ce6 fluorescence in A549 cells for free Ce6, and mPEG-PLA-S-S-Ce6 micelles were 170,408.75 ± 1.05, 227,928.07 ± 2.18 for 1 h and 284,777.37 ± 3.92, 354,555.63 ± 1.48 for 4 h, respectively (Fig. 4b, c). The geometric mean fluorescence observed in 4 T1 cells treated with free Ce6, and mPEG-PLA-S-S- Ce6 micelles were, 43,876.32 ± 2.02, 72,991.93 ± 3.78 for 1 h and 80,468.81 ± 1.72, 172,937.25 ± 1.61 for 4 h, respectively (Fig. 5b, c). The results obtained by confocal microscopy corroborate with the flow cytometer analysis results that the cells incubated with mPEG-PLA-S-S-Ce6 micelles demonstrated higher fluorescence at both the time points (1 h and 4 h) by indicating of higher cellular internalization compared to free Ce6. Redox-Dependent Intracellular Ce6 Release from mPEG-PLA-S-S-Ce6 Micelles As reported earlier, polymeric micelles are capable of avoid- ing endosomes and moves into the cytoplasmic organelles (31). The redox sensitive release of Ce6 from mPEG-PLA-S-S-Ce6 micelles in the intracellular environment was determined. Most of the cancer cells shows high level of cytosolic glutathi- one (GSH) around 100-to 1000-fold higher compared to ex- tracellular fluids. GSH reduction is the common redox system in tumor cells (17). The glutathione monoester (GSH-OEt) which penetrates cell membrane have been used to enhance the intracellular GSH concentration by ethyl ester hydrolyza- tion in cytoplasm (16,32). The A549 and 4 T1 cells were pre- treated with 10 mM GSH-OEt for 2 h to elevate the concen- tration of intracellular GSH. Further, cells were washed and treated with mPEG-PLA-S-S-Ce6 for 4 h. The fluorescence intensity of Ce6 was analyzed using confocal microscopy and flow cytometer. As shown in Figs. 4 and 5, the Ce6 fluores- cence was significantly enhanced in the GSH-OEt treated cells compared to without GSH-OEt pre-treatment. In Figs. 4b and 5b, the histogram of the pre-treated cells with GSH- OEt shifted to the right compared to the cells without pre- treatment. The geometric mean of fluorescence in A549 and 4 T1 cells for 4 h was 477,152.93 ± 3.02, and 297,152.35 ± 1.77, respectively. The results suggest that the Ce6 was rapidly released owing to the cleavage of disulfide linkage in existence of GSH. The results obtained by confocal microscopy corrob- orate with the flow cytometer result revealing that cells pre- treated with GSH-OEt showed higher fluorescence in com- parison to without GSH-OEt pre-treatment at both the time points. In Vitro Phototoxicity of mPEG-PLA-S-S-Ce6 In vitro phototoxicity of mPEG-PLA-S-S-Ce6 micelles were determined by MTT assay in A549 and 4 T1 cells. The cells were pretreated with 10 mM of GSH-OEt for 2 h or 0.1 mM of BSO for 12 h and without pretreat- ment were assigned as control. As mentioned earlier, GSH-OEt, a glycine esterified form of GSH upregulates the level of intracellular GSH concentration by ethyl ester hydrolyzation in the cytoplasm. On the other hand, BSO, an ir reversible inhibitor o f g am ma- glutamylcysteine synthetase, one of the rate-limiting en- zyme in glutathione synthesis decreases the amount of GSH (33). As shown in Fig. 6, cell viability varied signif- icantly with the cells pretreated with GSH-OEt or BSO and without pre-treatment. The cell viability of A549 and 4 T1 cells pre-treated with GSH, decreased signifi- cantly with a percentage viability of 29.96 ± 3.45% and 25.54 ± 5.5% at 24 h at the highest concentration of Ce6, respectively. The results indicate the rapid release of Ce6 due to the cleavage of disulfide bond in the mPEG-PLA-S-S-Ce6 micelles in the presence of intracel- lular GSH. However, A549 and 4 T1 cells pretreated with BSO showed cell viability of 61.76 ± 1.83% and 68.51 ± 2.6% at 24 h, respectively at the highest concen- tration of Ce6 indicating the decrease in the amount of intracellular GSH suppressing the Ce6 release from micelles. In the A549 without pretreatment, free Ce6 and mPEG-PLA-S-S-Ce6 micelles demonstrated cell via- bility of 55.51 ± 1.57%, and 45.62 ± 2.67%, respectively, whereas in 4 T1 cells, the cell viability of free Ce6 and mPEG-PLA-S-S-Ce6 micelles were 63.59 ± 3.44%, 52.66 ± 5.87% at 24 h, respectively. The tumor cells pos- sess approximately 4-fold increase in the concentration of glutathione compared to the normal tissues. Therefore, the glutathione sensitive delivery system can be success- fully utilized for the tumor-targeted drug delivery. However, mPEG-PLA-S-S-Ce6 micelles were not cyto- toxic in the absence of light. DNA Fragmentation and Nuclear Staining The DNA fragmentation assay by agarose gel electrophoresis was carried out to investigate the cause of cell death. The mPEG-PLA-S-S-Ce6 treated cells displayed higher DNA fragmentation with a ladder-like pattern compared to free Ce6 which confirmed increased degradation of DNA follow- ing treatment (Fig. 7a). The control DNA was intact with no DNA sign of degradation. Further, the nuclear staining was performed using DAPI and acridine orange (AO) to assess live and dead cells (Fig. 7b). The AO permeates into both dead and live cells and emits green fluorescence, whereas DAPI stains the live cells only and emits blue fluorescence. Nuclear staining of mPEG-PLA-S-S-Ce6-treated A549 cells with acri- dine orange (AO) and DAPI showed distinctive changes in the cellular morphology compared to the untreated cells. The presence of higher extent of nuclear fragmentation (NF) and cytoplasm shrinkage (CS) in mPEG-PLA-S-S-Ce6-treated cells indicated that the cells were undergoing apoptosis and DNA damage to higher extent compared to free Ce6 treat- ment. The result supported the MTT assay findings that the mPEG-PLA S-S-Ce6 treatment exhibited higher therapeutic efficacy compared to the treatment using free Ce6. Penetration of Micellar Ce6 in Spheroids The three-dimensional (3D) spheroid culture serves as an im- portant model mimicking the in vivo tumor growth such as architecture, gradients of pH and PO2 and vascularization. The 3D model overcomes limitations of the two-dimensional (2D) monolayer cell cultures (34). Due to their 3D architec- ture, it has been utilized for study of penetration of anticancer drugs into tumor tissue (35). Previous reports suggest that the surface charge of the nanoformulations influences the pene- tration depth (36). The nanocarriers possessing negative sur- face potential probably diffuses deeper in spheroids owing to the decreased non-specific adsorption by negatively charged extracellular matrix components of spheroids such as collagen and proteoglycans (37,38). The penetration of mPEG-PLA-S- S-Ce6 micelles and free Ce6 were studied in A549 spheroids for 1 and 4 h, respectively. As shown in Fig. 8a, Ce6 fluores- cence from mPEG-PLA-S-S-Ce6 micelles was observed in the layers of the A549 spheroids evident from the Z-projections of the stacks images. The small size and negative surface charge of mPEG-PLA-S-S-Ce6 micelles helps in easy penetration into the spheroids. The mPEG-PLA-S-S-Ce6 micelles incu- bated for 4 h showed higher fluorescence intensity at every section of depth with maximum penetration of about 70 μm. However, free Ce6 accumulated at the periphery of the sphe- roids showed weak fluorescence, and was not able to penetrate towards the core, which might be due its solubility and slow cellular internalization. The spheroids incubated with the me- dia (control) demonstrated no fluorescence. Internalization of the mPEG-PLA-S-S-Ce6 in Spheroids To quantify the internalization of mPEG-PLA-S-S-Ce6 micelles in A549 spheroids compared to free Ce6, fluores- cence intensity was measured using flow cytometer. Spheroids incubated with mPEG-PLA-S-S-Ce6 micelles showed significantly higher fluorescence intensity compared to free Ce6 indicated by more right shift of the curve in the histogram obtained from flow cytometer (Fig. 8b). The mPEG-PLA-S-S-Ce6 micelles-treated spheroids showed in- crease in geometric mean of fluorescence (14,537.56 ± 2.51, 25,977.41 ± 1.83) than spheroids treated with free Ce6 (11,094.53 ± 2.95, 17,895.43 ± 4.06) at both the time points. The data acquired by flow cytometry (Fig. 8b) corroborates with the confocal microscopy results (Fig. 8a) suggesting that mPEG-PLA-S-S-Ce6 micelles are capable of penetrating in- side the spheroids owing to their small size, as discussed above. Growth Inhibition of Spheroids The growth inhibition on A549 spheroids were analyzed by assessing the diameter of spheroids using fluorescence micro- scope. The spheroids incubated with mPEG-PLA-S-S-Ce6 micelles irradiated with a 633 nm laser showed significant reduction in size, suggesting the efficacy of the PDT treat- ment. The time-dependent change in the diameter and mor- phology of irradiated spheroids are shown in Fig. 9. The mean diameter of A549 spheroids after treatment with mPEG-PLA- S-S-Ce6 micelles at day 3 was 587.5 ± 8.06 μm and reduced to 569.17 ± 12.51 μm on day 6, whereas free Ce6 treated spheroids showed 611.25 ± 34.94 μm on day 3 and reached a maximum of 633.75 ± 12.12 μm on day 6 (Fig. 9). However, control spheroids exhibited significant increase in growth dur- ing the experiment. Live/Dead Cell Assay in Spheroids The viability of A549 spheroids were evaluated qualitatively using Calcein Blue AM and propidium iodide (PI). The Calcein Blue AM, a non-fluorescent membrane permeant dye initiates fluorescence upon interacting with the intracellu- lar esterases in the live cells. The live cells retained calcein producing blue fluorescence whereas dead cells exhibited red fluorescence following PI staining, which went inside cell through damaged membrane and bound with the nucleic acids (39). The spheroids incubated with mPEG-PLA-S-S- Ce6 micelles demonstrated higher red fluorescence compare to free Ce6 revealing the increased death in spheroidal cells (Fig. 10a). However, control spheroids exhibited more blue fluorescence emitted from the metabolically-active spheroidal cells owing to the selective uptake of calcein blue AM. In Vitro Phototoxicity in Spheroids To determine the phototoxicity in spheroids, highly sensitive Presto Blue reagent was used. This reagent involves resazurin indicating the cell viability (40). In live cells, resazurinis gets reduced to resorufin in cellular respiration by taking electrons from NADPH, NADH, FMNH, FADH, and cytochromes (41). The reduction results in conversion of non-fluorescent Presto Blue to fluorescent form indicating the number of met- abolically viable cells (42). It appeared that mPEG-PLA-S-S- Ce6 micelles showed less phototoxicity to spheroids than to the monolayer cells. The cell viability (59.99 ± 3.14%) at high- est concentration of Ce6 was significantly higher in spheroids compared to monolayer cells (Fig. 10b). This could be due to the tissue depth of the spheroids which can weaken the irra- diated light resulting in generation of 2O1 at lesser extent compared to monolayer cells. Apoptosis Study in Spheroidal Cells The extent of apoptosis caused by mPEG-PLA-S-S-Ce6 micelles was determined by Annexin V-FITC/PI stain using flow cytometer. PI and Annexin V-FITC were utilized as fluo- rescent probes to differentiate apoptotic or dead cells from live cells. The histogram plots of the stained spheroidal cells shown in Fig. 11 represents the four quadrants, lower left (Q1, non- apoptotic cells), lower right (Q2, early apoptotic; Annexin V stained cells), upper right (Q3, late apoptotic; double stained) cells), upper left (Q4, necrotic; PI stained cells), respectively. As shown in Fig. 11, it was observed that insignificant difference in necrotic cell population in different samples whereas, early and late apoptotic cells of mPEG-PLA-S-S-Ce6 micelles were significantly greater in comparison to free Ce6. mPEG-PLA- S-S-Ce6 induced about 54.59% of apoptosis (6.49% early and 48.1% late) whereas free Ce6 exhibited 40.4% of apoptosis (5.2% early and 35.2% late). The results suggest that the mPEG-PLA-S-S-Ce6 micelles were effective in killing the can- cer cells by the induction of apoptosis. CONCLUSION Here, Ce6, a NIR-activatable third generation photosensitizer was utilized for development of redox-sensitive PDT. An am- phiphilic polymer conjugate, mPEG-PLA-S-S-Ce6 was syn- thesized, and characterized by 1H NMR. Further, the micelles were prepared and characterized by DL-Buthionine-Sulfoximine and TEM studies, which proved low polydispersity and spherical morphology. The mPEG-PLA-S-S-Ce6 micelles demonstrated significant increase in the generation of 1O2 in presence of GSH. In vitro release study also suggested a stimuli-sensitive release based on the GSH concentration. mPEG-PLA-S-S-Ce6 micelles displayed superior cellular internalization and photo- toxicity compared to free Ce6 in both the tested cell lines. Higher phototoxicity was observed when cells were pretreated GSH-OEt which confirms the sensitivity of the synthesized carrier to the stimulus. The mPEG-PLA-S-S-Ce6 micellar treatment exhibited DNA damage to the cancer cells to a greater extent compared to the free Ce6 as assessed by DNA fragmentation and nuclear staining experiments. Further, the studies in multicellular 3D tumor spheroids indicated a higher penetration, superior growth inhibition, phototoxicity, and induction of apoptosis, which suggests that the GSH- cleavable mPEG-PLA-S-S-Ce6 micelles were efficient in de- livering Ce6 intracellularly. Therefore, the newly synthesized mPEG-PLA-S-S-Ce6 micellar nanoconjugate holds a great potential to be further investigated for its successful applica- tion in photodynamic therapy.