Free and Poly-Methyl-Methacrylate-Bounded BODIPYs: Photodynamic and Antimigratory Effects in 2D and 3D Cancer Models

Simple Summary Photodynamic therapy (PDT) is a minimally invasive and highly selective technique to treat solid tumors and other malignancies. To exert a significant cytotoxic effect, PDT must simultaneously gather a photosensitizer (PS), a light at a specific wavelength, and oxygen. Although several PSs have been developed so far, systems with higher selectivity and efficacy are still needed to improve PDT anticancer treatment. This work shows how BODIPYs photosensitizers, loaded onto polymethyl methacrylate nanoparticles, can effectively reduce tumor cell viability in vitro and lower their migratory ability, thus, potentially reducing the metastatic tumor potential. Abstract Several limitations, including dark toxicity, reduced tumor tissue selectivity, low photostability and poor biocompatibility hamper the clinical use of Photodynamic therapy (PDT) in cancer treatment. To overcome these limitations, new PSs have been synthetized, and often combined with drug delivery systems, to improve selectivity and reduce toxicity. In this context, BODIPYs (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene) have recently emerged as promising and easy-to-handle scaffolds for the preparation of effective PDT antitumor agents. In this study, the anticancer photodynamic effect of newly prepared negatively charged polymethyl methacrylate (nPMMA)-bounded BODIPYs (3@nPMMA and 6@nPMMA) was evaluated on a panel of 2D- and 3D-cultured cancer cell lines and compared with free BODIPYs. In particular, the effect on cell viability was evaluated, along with their ability to accumulate into the cells, induce apoptotic and/or necrotic cell death, and inhibit cellular migration. Our results indicated that 3@nPMMA and 6@nPMMA reduce cancer cell viability in 3D models of HC116 and MCF7 cells more effectively than the corresponding free compounds. Importantly, we demonstrated that MDA-MB231 and SKOV3 cell migration ability was significantly impaired by the PDT treatment mediated by 3@nPMMA and 6@nPMMA nanoparticles, likely indicating the capability of this approach to reduce metastatic tumor potential.


Introduction
Photodynamic therapy (PDT) is a therapeutic modality applied to the treatment of several diseases, such as acne, infections, and cancer [1]. In PDT, cytotoxic reactive oxygen species (ROS), mainly singlet oxygen ( 1 O 2 ), are generated locally by the combined action of a photosensitizer (PS) and light, and selectively eradicate diseased cells. One of the main advantages of PDT relies on its potential dual selectivity, e.g., the preferential PS accumulation at the diseased tissue and the focused light irradiation of the target area, making it a minimally invasive therapeutic option, especially for anticancer treatment. cells (MCF7 and MDA-MB231) and compared with free BODIPYs. In particular, the effect on cell viability was evaluated, along with their ability to accumulate into the cells, induce apoptotic and/or necrotic cell death, and inhibit cellular migration. To achieve more insight into the potential of these novel nanosystems as PDT anticancer agents, we evaluated their performance in HCT116 and MCF7 cells grown as 3D spheroids. Compared to monolayer cell lines, the 3D model better recapitulates the in vivo avascular tumor characteristics, such as normoxic/hypoxic regions and cell-cell interactions, thus representing a more suitable benchmark for evaluating the PDT effect [26]. Thus, the effects on cell growth and cell viability and the ability of free and nPMMA-bounded BODIPYs to penetrate through extracellular spaces in the spheroids and accumulate in the cells were assessed.

Reagents and Chemicals
The 3-Sulfopropyl methacrylate potassium salt, methyl methacrylate (MMA), sodium dodecyl sulphate, potassium persulphate (KPS) and all chemicals required for preparation and characterization of the nanoparticles were purchased from Aldrich (Milan, Italy) and used without further purification. All reagents for cell culture and in vitro experiments were purchased from Euroclone (Milan, Italy).

BODIPYs Characterization
BODIPYs 1-octanol-water partition coefficient (P) was determined at 25 • C by means of equal volumes of pre-equilibrated milliQ water (3 mL) and 1-octanol (3 mL). Specifically, an aqueous solution of the BODIPYs (40 µM) was stirred for 8 h at 25 • C in the presence of octanol, then, 200 µL of both aqueous and organic phases were diluted with DMF up to 2 mL and the BODIPY final concentration was determined by absorption spectroscopy (Cary 50, Agilent Technologies, Milan, Italy). The relative rates of 1 O 2 produced by the BODIPYs were experimentally measured by monitoring the disappearance of the 410 nm absorbance band of 1,3-diphenylisobenzofuran in isopropanol, an 1 O 2 scavenger, and normalized with respect to values obtained using Rose Bengal as standard [5].

Nanoparticles Synthesis and Characterization
An aqueous solution (100 mL) of 3-sulfopropyl methacrylate potassium salt (SPM, 0.65 mmol, 151 mg) and sodium dodecyl sulfate (SDS, 0.01 mmol, 3 mg) was put into a 250 mL three-neck reactor under mechanical stirring and bubbled with a nitrogen flux. After 15 min, methyl methacrylate (MMA, 4 mL, 40.2 mmol) was added dropwise and heated at 80 • C. Once the temperature reached 80 • C the radical initiator, potassium persulfate (KPS, 0.25 mmol, 93 mg), was introduced into the solution. The resulting white suspension was left under heating and stirring for 4 h and then cooled at room temperature. After cooling, the milky suspension was dialyzed for 5 days, changing the outer water twice a day (dialysis Visking tubing; Φ = 28.6 mm; MWCO = 12-14 KDa).
The hydrodynamic diameter and zeta-potential were analyzed at 25 • C through dynamic light scattering (DLS) measurements on a NanoBrook Omni Particle Size Analyzer (Brookhaven Instruments Corporation, New York, NY, USA). The nanoparticles' morphology was analyzed by scanning electron microscopy (SEM) on a Philips XL30, and the sample was prepared by drying a droplet of NP solution (0.5 mg/mL or 1 mg/mL).

Nanoparticle Loading, Stability and Release Experiments
Briefly, positively charged BODIPYs were combined with nPMMA nanoparticles by electrostatic interaction: a certain amount of BODIPY (typically 500 µg) was dissolved in THF or acetone at a concentration of 0.5 mg/mL and added to an aqueous solution of negatively charged PMMA (5 mg/mL) in an Eppendorf vial, and then stirred in a Vortex apparatus for 20 s at room temperature. Each sample was then centrifuged for 30 min at 4 • C (48000 RCF) in an Allegra 64R centrifuge (Beckman Coulter, rotor F1202). Upon centrifugation the supernatant was removed and the remaining pellet was resuspended and washed with milliQ water till cleared of unbounded BODIPY, as determined by UV-Vis analysis (Cary 100, Agilent Technologies, Milan, Italy). To increase BODIPYs' loading, this procedure was iteratively repeated three times.
The loading content was evaluated by recording the absorption spectra of the filtrates by UV-Visible spectroscopy and comparing to a calibration curve of free BODIPYs obtained by plotting the concentration against their absorbance at 530 nm.
The loading efficiency and content were then calculated, based on the following formula: Loading efficiency (LE, %) = amount of BODIPY loaded onto NPs (µg)/amount of BODIPY initially used (µg) × 100 Loading content (LC, %) = amount (µg) of loaded BODIPY/amount (µg) of nanoparticles × 100 In vitro stability studies were performed over time (48 h) at 37 • C in FBS 20% in phosphate buffered saline (PBS; pH 7.4). All experiments were carried out in the dark to avoid possible disassembly contribution, due to ROS formation under natural light exposure. In a typical experiment, 10 µL of nanoparticle solution (NPs conc: 1 mg/mL; BODIPY conc: 15 µg/mL) were diluted with 1 mL of the medium, while being maintained at 37 • C. Changes in particle size distribution over time were monitored by means of DLS.
The corresponding 3f@nPMMA and 6f@nPMMA nanoparticles were obtained and characterized following the same procedures described for non-fluorescent derivatives 3 and 6. Briefly, 20 µg of 3f (or 6f) was dissolved in water and loaded, as previously reported, onto nPMMA (1 mg/mL). This was followed by three cycles of centrifugation/washing to remove unloaded BODIPYs.
The release of BODIPYs from the nanoparticles was measured by spectrophotometric analysis on 3f@nPMMA and 6f@nPMMA. An amount of 1 mL of the obtained suspension was inserted into a dialysis tube (Φ = 1 cm; MWCO = 12-14 KDa) and placed into a sealed 20 mL plastic container. Then, 6 mL of PBS were added to the container, which was kept shaking at 37 • C for a total of 48 h. At selected time points, 1 mL of the external solution was withdrawn and replaced with fresh PBS, and the amount of fluorescent BODIPY was determined against its corresponding calibration curve, obtained by plotting the concentration from 1 µg/mL to 5 µg/mL against their respective absorbance at 596 nm.

Reactive Oxygen Species (ROS) Determination
The Reactive Oxygen Species (ROS) production was evaluated for 3@nPMMA and 6@nPMMA using the chemical probe 2,7-dichlorodihydrofluorescein diacetate (H 2 DCFDA) that, in the presence of ROS, is converted to fluorescent 2,7-dichlorofluorescein (DCF). For the probe preparation, 2 mL of NaOH (0.01 M) were added to 500 µL of a methanol solution of H 2 DCFDA (1.1 mM) and stirred for 30 min at room temperature. Then, 10 mL of phosphate buffer (pH = 7.4) were added to obtain the final ROS probe solution. For ROS detection, samples were prepared as follows: a certain amount of 3@nPMMA and 6@nPMMA solutions, corresponding to a BODIPYs' final concentration of 4 µg/mL was added to a cuvette containing 500 µL of phosphate buffer and 218 µL of ROS probe, and water was added to a final volume of 1 mL. Samples were then irradiated with a white 320-700 nm emission wavelength tungsten lamp (300 W, light intensity 45 mW/cm 2 at 670 nm, Phillips, Bologna, Italy). The light was positioned at a distance of 40 cm from the cuvette, and the absorption spectra were recorded at different time points with a Cary 100 UV-Vis spectrophotometer (Agilent Technologies, Milan, Italy), reading absorbance at 500 nm.

Cell Lines and In Vitro Culture Conditions
The human colorectal cancer HCT116, breast adenocarcinoma MCF7 and MDA-MB231 and ovarian cancer SKOV3 cell lines were obtained from ATCC (American Type Culture Collection, Manassas, VA, USA). Cells were maintained under standard culture conditions (37 • C; 5% CO 2 ) in RPMI-1640 medium (MCF7, and MDA-MB231 cells) or in DMEM medium (SKOV3 and HCT116 cells) supplemented with 10% fetal calf serum, 1% glutamine and 1% antibiotics mixture. For HCT116 and SKOV3 cells, 1% sodium pyruvate and 1% non-essential amino acids were also added to the culture medium.
To produce the corresponding spheroids, HCT116 and MCF7 cells were grown as a monolayer, detached, and 2.5 × 10 3 cells/well were then seeded onto 96U plates Nunclon Sphera (Thermo, Milan, Italy) and incubated at 37 • C in a 5% CO 2 atmosphere. Spheroids were used at day 7 from seeding.
Unless otherwise indicated, cells were seeded and allowed to attach and growth for 48 h before treatment with the compounds. After 24 h of incubation at 37 • C, the drugcontaining culture medium was removed, and cells were irradiated for 1 h in drug-free PBS and then incubated in the dark for 24 h before performing the assays. The irradiation step was always performed using a green LED lamp composed of 12 × 3 W diodes placed on a 11 cm diameter disk and equipped with a heat sinker and with a maximum emission at 525 nm and a width at half maximum of 70 nm (fluence rate 3.036 × 10 −3 W/cm 2 ).

Cell Viability Assays on 2D and 3D Dimensional Models
Cell survival following exposure to the free-and nPMMA-bounded BODIPYs was evaluated using the MTT assay, as described elsewhere [28]. Cells were seeded onto 96-well plates (3 × 10 3 cells/well) and treated with a range of drug concentrations (0.1-50 nM). At the end of the previously described experimental protocol (Section 2.5.1), MTT reagent was added to each well at a final concentration of 0.4 mg/mL for 3 h at 37 • C. Cell viability was determined by measuring the absorbance at 570 nm, using an iMark Microplate Reader (BIORAD), following dissolution of formazan crystals, formed by MTT metabolism in viable cells.
Possible intrinsic (i.e., not photodynamic) effects on cell viability, due to BODIPYs (free or nPMMA-loaded) without irradiation, were assessed by exposing cells to concentrations ten-fold higher than those used in the PDT experiments, but excluding the irradiation step. The IC 50 values (i.e., the concentration affecting 50% of cell survival fraction) were obtained by nonlinear regression analysis using the GraphPad PRISM 4.03 software (GraphPad Software Inc., San Diego, CA, USA).
The phototoxic effects of 3, 3@nPMMA, 6 and 6@nPMMA on HCT116 and MCF7 spheroids were assessed, based on a dye exclusion assay. Briefly, spheroids were treated with the compounds at concentrations corresponding to the IC 50 values obtained by the MTT assays on HCT116 and MCF7 cells cultured as monolayers. After 24 h incubation, following replacement of the drugs-containing medium with PBS, spheroids were irradiated under a green LED light for 1 h (fluence rate 3.036 × 10 −3 W/cm 2 ). After irradiation, spheroids were incubated in the dark at 37 • C in a drug-free medium, and, after 24, 48 and 72 h, three to five spheroids for each treatment were independently collected, disaggregated using trypsin-EDTA solution, and live cells were counted using a Burker hemocytometer, following Trypan Blue staining. Control spheroids were treated only with culture medium and incubated/irradiated as were the treated ones.

Flow Cytometric Analysis
The percentage of apoptotic and necrotic cells was evaluated by flow cytometric analysis. For this set of experiments, cells were exposed to the BODIPYs as free or loaded onto nPMMA at their respective IC 50 values. To assess the percentage of apoptotic cells, at the end of treatment, both adherent and detached cells were harvested, washed in PBS, and fixed in 70% ethanol at −20 • C for 30 min. After a further wash in PBS, the DNA was stained with a solution of propidium iodide (PI) in PBS (50 µg/mL) in the presence of RNAse A (30 U/mL) at room temperature for 15 min before analyzing the samples. Intracellular accumulation of the BODIPYs was performed in all cell lines cultured as monolayer and exposed to the free and nPMMA-bounded fluorescent 3f and 6f (100 nM for 24 h). At the end of the exposure time, treated cells were detached by trypsinization, washed thoroughly in ice-cold PBS, resuspended in PBS and analyzed, exploiting the intrinsic fluorescence of these compounds. All samples were analyzed with a FACSCalibur flow cytometer (Becton Dickinson Mountain View, CA, USA) and data were processed using Cel-lQuestPRO software (Becton Dickinson). Fluorescent emission of PI was collected through a 575 nm band-pass filter, acquired in log mode, and the percentage of apoptotic cells in each sample was determined, based on the sub-G1 peaks detected in mono-parametric histograms. BODIPY fluorescence was collected through a 530-nm band-pass filter and BODIPY accumulation was quantitated in arbitrary units, based on the median fluorescence intensity (MFI).

Effects on Cell Migration
The effects on cell migration eventually induced by 3, 3@nPMMA, 6 and 6@nPMMA were evaluated on SKOV3 and MDA-MB231 cell lines and assessed by the Scratch Wound Healing assay. Cells were seeded onto 6-well plates (MDA-MB231: 1.3 × 10 5 /well; SKOV3: 0.7 × 10 5 /well) and allowed to grow for 48 h (approximately to confluence) followed by a 24 h treatment with subtoxic concentrations, corresponding to their respective IC 20 , of 3 (0.5 nM), 3@nPMMA (1 nM for MCF7; 0.5 nM others), 6 (0.2 nM for MDA-MB231; 0.4 nM others) and 6@nPMMA (1 nM for HCT116; 1.5 nM for MDA-MB231; 2.5 nM others). A scratch was made with a pipette tip, and the drug-containing medium was replaced by fresh PBS, and cells were irradiated with a green LED light for 1 h (fluence rate 3.036 × 10 −3 W/cm 2 ) followed by incubation in a drug-free medium at 37 • C for 24 h. Pictures of the scratch wound were taken immediately after the irradiation step (t0) and after 24 h through a camera connected to an Olympus IX81 microscope.

Western Blot Analysis
The expression of MMP2 and MMP9 in whole cell lysates following treatment with 3, 3@nPMMA, 6 and 6@nPMMA, at concentrations corresponding to their respective IC 20 , was detected by Western blot analysis of whole cell lysates. Briefly, cells were resuspended in lysis buffer (NaCl 120 mM, NaF 25 mM, EDTA 5 mM, EGTA 6 mM, sodium pyrophosphate 25 mM in tris-buffered saline TBS 20 mM pH 7.4, phenylmethanesulfonyl fluoride PMSF 2 mM, Na 3 VO 4 1 mM, phenylarsine oxide 1 mM, 1% NP-40 and 10% Protease Inhibitor Cocktail), incubated for 10 min on ice after adding SDS (final concentration 0.1%) and lysates were collected by centrifugation (12,800 rpm for 20 min). Protein concentration was determined by the BCA assay (Pierce, Italy) and 50 µg of protein per sample were loaded onto 8% polyacrylamide gels and separated under denaturing conditions. Protein bands were then transferred onto Hybond-P membranes (Amersham Biosciences, Italy) and Western blot analysis was performed by standard techniques with mouse monoclonal antibody directed against MMP2 and MMP9 (Santa Cruz Biotechnology, Inc.). Equal loading of the samples was verified by re-probing the blots with a mouse monoclonal anti-actin antibody (Santa Cruz Biotechnology, Inc.). Protein bands were visualized through the Gbox (Syngene, Chemi-XT4) using peroxidase-conjugated anti-mouse secondary antibodies (Sigma-Aldrich) and the Westar Supernova Substrate (Cyanagen). Densitometric analysis was performed by Image-J software.

Localization and Diffusion of the PSs Inside Spheroids
Spheroids were obtained as reported at Section 2.5.1 and treated with the fluorescent version of the free-and nPMMA-bounded BODIPYs, 3f, 6f, 3f@nPMMA and 6f@nPMMA, at 100 nM concentration in the dark. After 24 h of incubation, PSs fluorescence intensity and distribution into spheroids were evaluated by confocal microscopy. Spheroids were transferred from 96-well plates to 35 mm dishes, washed with PBS and directly observed with a Leica SP5 Confocal Microscope. Images from poles to equatorial plane of the spheroids were acquired. The intensity of fluorescence in the spheroids exposed to the different BODIPYs were compared using LAS LAF software. No evidence of a different diffusion rate of the BODIPYs was seen through the acquired sections, so, therefore, the diffusion rate was analyzed exclusively at the equatorial plane. To this end, 15 radial lines were randomly drawn on the image of the equatorial plane and fluorescence at each pixel was recorded.

Statistical Analysis
Statistical analysis of all biological data was performed by means of one-or twoway ANOVA, with Bonferroni's test for multiple comparisons, using GraphPad PRISM 4.03 software.

Photodynamic Characteristics of BODIPYs 3 and 6
The photodynamic features of BODIPYs 3 and 6 in terms of lipophilicity and 1 O 2 generation capability were characterized. In particular, both BODIPYs had higher molar extinction coefficient values, compared to their bromine precursors [29], most likely due to the introduction of the pyridinium group favoring dissolution in DMSO (Table 1).
Despite the presence of the pyridinium group and the positive charge, the octanol/water partition coefficients (LogP, Table 1) seemed to indicate the higher solubility of 3 and 6 in the organic phase, as compared to the aqueous one. This behavior could be explained by the large neutral part of the molecules and the presence of the two iodine atoms. As expected, compound 6, harboring a chain of eight carbon atoms, proved to be more lipophilic than 3 with its appendix of four carbon atoms.
The relative rates of 1 O 2 production of 3 and 6 were obtained by comparison with Rose Bengal (RB), known as the most powerful singlet oxygen producer [29]. Considering equal to 1 as the 1 O 2 production from RB, compounds 3 and 6 developed 0.8 and 0.6 1 O 2 , respectively, therefore accounting for good performances of the two BODIPYs. Table 1. Wavelength of maximum absorption and extinction coefficient, partition coefficient octanol/water and 1 O 2 generation for BODIPYs 3 and 6.

BODIPY
ε (M −1 cm −1 ) a LogP 1 O 2 Generation b iodine atoms. As expected, compound 6, harboring a chain of eight carbon atoms, proved to be more lipophilic than 3 with its appendix of four carbon atoms. The relative rates of 1 O2 production of 3 and 6 were obtained by comparison with Rose Bengal (RB), known as the most powerful singlet oxygen producer [29]. Considering equal to 1 as the 1 O2 production from RB, compounds 3 and 6 developed 0.8 and 0.6 1 O2, respectively, therefore accounting for good performances of the two BODIPYs.

Nanoparticle Synthesis and Characterization
Negatively charged core-shell polymethyl methacrylate nanoparticles (nPMMAs) were synthesized by an emulsion co-polymerization reaction ( Figure 1A), affording nanoparticles with an average hydrodynamic diameter of 145 ± 5 nm ( Figure 1B) and a zeta potential of -51.4 mV. The morphological analysis, performed by scanning electron microscopy (SEM), indicated that the particles were spherical in shape and regularly distributed, with an average dry diameter of 132 ± 4 nm. The difference in the NPs' radii, observed with DLS and SEM techniques, could be ascribed to the different environments in which the measurements were performed, e.g., water and air, respectively ( Figure 1C). could be explained by the large neutral part of the molecules and the presence of the two iodine atoms. As expected, compound 6, harboring a chain of eight carbon atoms, proved to be more lipophilic than 3 with its appendix of four carbon atoms. The relative rates of 1 O2 production of 3 and 6 were obtained by comparison with Rose Bengal (RB), known as the most powerful singlet oxygen producer [29]. Considering equal to 1 as the 1 O2 production from RB, compounds 3 and 6 developed 0.8 and 0.6 1 O2, respectively, therefore accounting for good performances of the two BODIPYs.

Nanoparticle Synthesis and Characterization
Negatively charged core-shell polymethyl methacrylate nanoparticles (nPMMAs) were synthesized by an emulsion co-polymerization reaction ( Figure 1A), affording nanoparticles with an average hydrodynamic diameter of 145 ± 5 nm ( Figure 1B) and a zeta potential of -51.4 mV. The morphological analysis, performed by scanning electron microscopy (SEM), indicated that the particles were spherical in shape and regularly distributed, with an average dry diameter of 132 ± 4 nm. The difference in the NPs' radii, observed with DLS and SEM techniques, could be ascribed to the different environments in which the measurements were performed, e.g., water and air, respectively ( Figure 1C).

Nanoparticle Synthesis and Characterization
Negatively charged core-shell polymethyl methacrylate nanoparticles (nPMMAs) were synthesized by an emulsion co-polymerization reaction ( Figure 1A), affording nanoparticles with an average hydrodynamic diameter of 145 ± 5 nm ( Figure 1B) and a zeta potential of −51.4 mV. The morphological analysis, performed by scanning electron microscopy (SEM), indicated that the particles were spherical in shape and regularly distributed, with an average dry diameter of 132 ± 4 nm. The difference in the NPs' radii, observed with DLS and SEM techniques, could be ascribed to the different environments in which the measurements were performed, e.g., water and air, respectively ( Figure 1C). Optimization experiments performed by fine-tuning of the NPs/BODIPYs ratio, determined the best loading conditions in terms of efficiency (96%) and loading content, which resulted in 3.75% and 4.2% for 3@nPMMA and 6@nPMMA, respectively ( Figure  2A). Under these conditions, we obtained stable and reproducible nanoparticles with an average hydrodynamic diameter of 144 ± 5 nm (PDI = 0.12) and 205 ± 11 nm (PDI = 0.2) for 3@nPMMA and 6@nPMMA, respectively ( Figure 2B). The zeta potential values of nanoparticles after loading were -25.07 and -19.02 for 3@nPMMA and 6@nPMMA, respectively, indicating that the overall negative shell was preserved, and, thus, favoring their colloidal stability. Optimization experiments performed by fine-tuning of the NPs/BODIPYs ratio, determined the best loading conditions in terms of efficiency (96%) and loading content, which resulted in 3.75% and 4.2% for 3@nPMMA and 6@nPMMA, respectively (Figure 2A). Under these conditions, we obtained stable and reproducible nanoparticles with an average hydrodynamic diameter of 144 ± 5 nm (PDI = 0.12) and 205 ± 11 nm (PDI = 0.2) for 3@nPMMA and 6@nPMMA, respectively ( Figure 2B). The zeta potential values of nanoparticles after loading were −25.07 and −19.02 for 3@nPMMA and 6@nPMMA, respectively, indicating that the overall negative shell was preserved, and, thus, favoring their colloidal stability. which resulted in 3.75% and 4.2% for 3@nPMMA and 6@nPMMA, respectively ( Figure  2A). Under these conditions, we obtained stable and reproducible nanoparticles with an average hydrodynamic diameter of 144 ± 5 nm (PDI = 0.12) and 205 ± 11 nm (PDI = 0.2) for 3@nPMMA and 6@nPMMA, respectively ( Figure 2B). The zeta potential values of nanoparticles after loading were -25.07 and -19.02 for 3@nPMMA and 6@nPMMA, respectively, indicating that the overall negative shell was preserved, and, thus, favoring their colloidal stability.  Figure 2C). Our results indicated a similar stability trend for both nanoparticles: an initial diameter increase was observed during the first 4 h of the experiment, after which, the particles' sizes gradually decreased until Indeed, the stability of BODIPY-loaded nPMMA nanoparticles was evaluated at 37 • C for 48 h in FBS 20% in PBS pH 7.4, v/v ( Figure 2C). Our results indicated a similar stability trend for both nanoparticles: an initial diameter increase was observed during the first 4 h of the experiment, after which, the particles' sizes gradually decreased until reaching the initial value, and then stayed stable for 15 days. Therefore, although the presence of 20% FBS induced an initial size increase, likely because of the interaction with serum proteins, the sizes of the nanoparticles remained within an acceptable range (<200 nm), and no visible aggregation/precipitation phenomena were observed.
Release experiments were performed by spectrophotometric analysis taking advantage of the fluorescent version of BODIPYs, namely 3f and 6f ( Figures S1-S5), and were performed by the dialysis method, using, as release medium, PBS pH 7.4, owing to the solubility of both fluorescent BODIPYs in aqueous media. The BODIPYs' release from the nanoparticles was lower than 10% for 6f@nPMMA and around 20% of the loaded amount for 3f@nPMMA, reaching a plateau after 6 h ( Figure 2D). These data indicated that the electrostatic bond between the positively charged BODIPYs and nPMMA nanoparticles was satisfactorily stable for biological application. The lower release observed for 6f, as compared to 3f, could be ascribed to the longer chain of this BODIPY, favoring its hydrophobic interaction with the NP shell, and, thus, overall increasing the loading stability.
The ROS generation of 3 and 6 loaded onto nPMMA nanoparticles was evaluated spectrophotometrically by measuring the increase in the absorption peak at 500 nm of 2,7-dichlorofluorescein (DCF). Indeed, DCF was only produced in the presence of ROS (see baseline Figure S6) upon subsequent hydrolysis and oxidation of the H 2 DCFDA probe (see Section 2.4.2) [30]. A solution of 3/6@nPMMA nanoparticles and ROS probe was irradiated with a white tungsten lamp (300 W) for different time intervals and the DCF absorbance was recorded. The 3@nPMMA and 6@nPMMA displayed very similar ROS-producing abilities ( Figure 3) and were comparable to those of free 3 and 6 ( Figure S7). The absorbance peak at 500 nm increased with the irradiation time in a light dose-dependent manner.
The ROS generation of 3 and 6 loaded onto nPMMA nanoparticles was evaluated spectrophotometrically by measuring the increase in the absorption peak at 500 nm of 2,7dichlorofluorescein (DCF). Indeed, DCF was only produced in the presence of ROS (see baseline Figure S6) upon subsequent hydrolysis and oxidation of the H2DCFDA probe (see § 2.4.2) [30]. A solution of 3/6@nPMMA nanoparticles and ROS probe was irradiated with a white tungsten lamp (300 W) for different time intervals and the DCF absorbance was recorded. The 3@nPMMA and 6@nPMMA displayed very similar ROS-producing abilities ( Figure 3) and were comparable to those of free 3 and 6 ( Figure S7). The absorbance peak at 500 nm increased with the irradiation time in a light dose-dependent manner. Figure 3. ROS production of 3@nPMMA and 6@nPMMA nanoparticles displayed as absorption spectra of DCF measured at different irradiation times (minutes).

Effect on Cell Viability
The photodynamic activity of 3, 6, 3@nPMMA and 6@nPMMA on HCT116, SKOV3, MCF7 and MDA-MB231 cells was assessed by MTT assay upon treatment with increasing BODIPY concentrations, followed by light irradiation and 24 h incubation in a drug-free medium. The IC50 values were obtained from the corresponding dose-response curves ( Table 2). All the tested compounds/formulations showed cytotoxic effects at nanomolar concentrations. In particular, in three out of four cell lines, 3@nPMMA was as potent as the corresponding unbounded derivative 3, while 6@nPMMA was always significantly less potent than 6. The intrinsic cytotoxicity of the photosensitizers was assessed by omitting the irradiation step from the treatment protocol and was found to be negligible.

Effect on Cell Viability
The photodynamic activity of 3, 6, 3@nPMMA and 6@nPMMA on HCT116, SKOV3, MCF7 and MDA-MB231 cells was assessed by MTT assay upon treatment with increasing BODIPY concentrations, followed by light irradiation and 24 h incubation in a drug-free medium. The IC 50 values were obtained from the corresponding dose-response curves ( Table 2). All the tested compounds/formulations showed cytotoxic effects at nanomolar concentrations. In particular, in three out of four cell lines, 3@nPMMA was as potent as the corresponding unbounded derivative 3, while 6@nPMMA was always significantly less potent than 6. The intrinsic cytotoxicity of the photosensitizers was assessed by omitting the irradiation step from the treatment protocol and was found to be negligible. Table 2. IC 50 values (nM) obtained in HCT116, SKOV3, MCF7 and MDA-MB231 cells by the MTT assay following 24 h incubation with the free and nPMMA-bounded BODIPYs (3 and 6), 1 h irradiation and 24 h incubation in a drug-free medium (mean ± SD of 4/5 independent experiments; * p < 0.05, ** p < 0.01 and *** p < 0.001 vs. respective free BODIPY).

Cellular Uptake
To get insight on the different cytotoxic effects exerted by 3@nPMMA and 6@nPMMA, as compared to their free forms, we studied the cellular uptake behavior of our formulations. Indeed, it is widely acknowledged that the cellular uptake/localization of PSs and their ROS generation capability are crucial elements to the successful application of PDT, and that different cellular responses could be ascribed to diverse uptake behaviors [31].
In our study, cellular internalization was assessed by taking advantage of the fluorescent analogs of 3 and 6, namely 3f and 6f, and their nanoparticles loaded version (3f@nPMMA and 6f@nPMMA). Our data showed that upon 24 h incubation in the presence of 100 nM of 3f, 6f, 3f@nPMMA and 6f@nPMMA, both 6f formulations exhibited higher intracellular accumulation, as compared to 3f formulations ( Figure 4A). The 3f and 3f@nPMMA showed a similar accumulation trend in HTC116, SKOV3 and MDA-MB231 cell lines, while in MCF7 cells 3f internalization was significantly higher, in respect to 3f@nPMMA ( Figure 4B), and in good agreement with MTT results (Table 2). Furthermore, as compared to 6f@nPMMA, free 6f exhibited greater accumulation in all cell lines ( Figure 4C), in accordance with the higher cytotoxicity of 6, in respect to 6@nPMMA.

Cellular Uptake
To get insight on the different cytotoxic effects exerted by 3@nPMMA and 6@nPMMA, as compared to their free forms, we studied the cellular uptake behavior of our formulations. Indeed, it is widely acknowledged that the cellular uptake/localization of PSs and their ROS generation capability are crucial elements to the successful application of PDT, and that different cellular responses could be ascribed to diverse uptake behaviors [31].

PDT-Induced Cell Death
It has been largely reported that PDT triggers cell death, mainly through apoptosis and necrosis [32,33]. In agreement with this evidence, our results indicated that apoptosis played a major role in tumor cell death induced by the four compounds in three out of four cell lines. Different extents of apoptosis were observed when HCT116, SKOV3 and MDA-MB231 cells were treated with equitoxic concentrations of the compounds. As shown in Figure 5A, 3 and 3@nPMMA induced the same extent of apoptotic cell death in HCT116 and MDA-MB231 cells, while in SKOV3 and MCF7 cells 3@nPMMA induced a significantly higher percentage of apoptosis than 3. Both 6 and 6@nPMMA induced similar percentages of apoptosis in MCF7 and MDA-MB231 breast cancer cells, while a different trend was observed in SKOV3 and HCT116 cells, with 6@nPMMA being more effective than 6 in SKOV3 cells.
shown in Figure 5A, 3 and 3@nPMMA induced the same extent of apoptotic cell death in HCT116 and MDA-MB231 cells, while in SKOV3 and MCF7 cells 3@nPMMA induced a significantly higher percentage of apoptosis than 3. Both 6 and 6@nPMMA induced similar percentages of apoptosis in MCF7 and MDA-MB231 breast cancer cells, while a different trend was observed in SKOV3 and HCT116 cells, with 6@nPMMA being more effective than 6 in SKOV3 cells. The analysis of necrosis induction in our model cell lines showed that all formulations induced a low necrotic cell death in HCT116, SKOV3 and MDA-MB231 cells (Figure 5B), although a significant increase in necrotic cells, over controls, was observed in HCT116 and MDA-MB231 cells treated with 6 and in SKOV3 treated with 3@PMMA. Interestingly, high levels of necrosis were observed in MCF7 cells, most probably compensating the lower extent of apoptotic cell death in this cell line. These results were in accordance with literature data, indicating that MCF7 was rather refractory to PDT-induced apoptosis, while being more likely to undergo necrosis [34,35]. These data indicated different cell type apoptotic/necrotic responses to equitoxic concentrations of free and nPMMA-bounded 3 and 6.

Effects on Cell Migration
In a previous paper, we demonstrated that HCT116 and MCF7 cells did not show intrinsic migratory capacity under conditions similar to those used in the current study. Therefore, the effects of subtoxic concentrations (IC 20 ) of 3, 3@nPMMA, 6, and 6@nPMMA on cell migration were assessed only on SKOV3 and MDA-MB231 cells, using the scratch wound healing assay. The scratch was monitored through a camera connected to a microscope and pictures were taken immediately after the treatment with the four formulations (t0) and after 24 h in the dark or upon irradiation. Importantly, using subtoxic compound concentrations avoided massive cell death, allowing evaluation of the effects of PDT on cell migration. The representative images obtained are shown in Figures S8 and S9. Furthermore, to better represent the scratch wound healing results, percentages of the open scratch wound detected at 24 h were normalized vs. the same percentage at t0 (Figure 6). In the absence of, or upon treatment with, nude nPMMA nanoparticles, both cell lines completely covered the surface during the 24 h observation time, regardless of the presence of the irradiation step (Figures S8 and S9).
to elicit the antimigratory effect of the compounds.
Following photoactivation, a significant inhibition of cell migration was observed in both cell lines for all tested compounds. Interestingly, the nPMMA-bounded BODIPYs were significantly more potent than the free molecules in reducing cell migration ( Figure  6). The balance between extracellular matrix destruction and deposition is essential for maintaining tissue architecture and functions. The degradation of the extracellular matrix, and of the basal membrane, is the first step for cancer cell migration and invasion. Matrix metalloproteinases (MMPs), and their tissue inhibitors, are known to modulate these processes, and higher expression of different MMPs facilitates tumor cell invasion and In the absence of irradiation, no significant inhibition of cell migration over control was observed in both cell lines, indicating that photoactivation was a necessary condition to elicit the antimigratory effect of the compounds.
Following photoactivation, a significant inhibition of cell migration was observed in both cell lines for all tested compounds. Interestingly, the nPMMA-bounded BODIPYs were significantly more potent than the free molecules in reducing cell migration ( Figure 6).
The balance between extracellular matrix destruction and deposition is essential for maintaining tissue architecture and functions. The degradation of the extracellular matrix, and of the basal membrane, is the first step for cancer cell migration and invasion. Matrix metalloproteinases (MMPs), and their tissue inhibitors, are known to modulate these processes, and higher expression of different MMPs facilitates tumor cell invasion and metastasis [36,37]. In human cancers, a strong correlation has been reported between high levels of MMPs and invasiveness [38]. In particular, MMP2 (gelatinase-A) and MMP9 (gelatinase-B) are deeply associated with the presence of metastatic tumors [39].
To evaluate if the antimigratory effects observed in SKOV3 and MDA-MB231 cells, following treatment with 3, 3@nPMMA, 6 and 6@nPMMA, could be related to their abilities to induce alterations in MMP2 and MMP9 expression, we analyzed their protein levels following PDT with the same subtoxic concentrations used for the scratch wound healing assay. Our results indicated that nPMMA-bounded BODIPYs could significantly reduce MMP2 and MMP9 protein levels in both cell lines, compared to free BODIPYs and controls, confirming their anti-migratory effect (Figures 7 and 8).
(gelatinase-B) are deeply associated with the presence of metastatic tumors [39].
To evaluate if the antimigratory effects observed in SKOV3 and MDA-MB231 cells, following treatment with 3, 3@nPMMA, 6 and 6@nPMMA, could be related to their abilities to induce alterations in MMP2 and MMP9 expression, we analyzed their protein levels following PDT with the same subtoxic concentrations used for the scratch wound healing assay. Our results indicated that nPMMA-bounded BODIPYs could significantly reduce MMP2 and MMP9 protein levels in both cell lines, compared to free BODIPYs and controls, confirming their anti-migratory effect (Figures 7 and 8).

Effects on Viability
Previous studies from our group performed on HCT116, SKOV3, MCF7 and MDA-MB231 cell lines, indicated that only HCT116 and MCF7 cells could form spheroids [40].

Effects on Viability
Previous studies from our group performed on HCT116, SKOV3, MCF7 and MDA-MB231 cell lines, indicated that only HCT116 and MCF7 cells could form spheroids [40]. Therefore, the photodynamic effects of 3, 3@nPMMA, 6 and 6@nPMMA on 3D cultured cells were only evaluated on spheroids obtained from these two cell lines, and Figure 9 shows the obtained growth curves.

BODIPY Distribution and Intensity
Confocal microscopy analysis of the distribution and intensity of the fluorescence of the PSs in the images of the equatorial planes obtained on HCT116 and MCF7 spheroids showed that the penetration of fluorescent 3f and 6f compounds, free or loaded onto nPMMA nanoparticles, was limited to the external cell rim of the spheroids, while no appreciable penetration was detected in the inner core ( Figure 10). The internalization extent of compound 6f was generally greater than 3f. At the same time, no significant difference was observed in the fluorescence intensity of 3f@nPMMA and 6f@nPMMA in both cell lines compared to the free BODIPYs.
In agreement with confocal microscopy results, lower amounts of 3f, compared with 6f, were found in cells obtained from both HCT116 and MCF7 spheroids, following 24 h treatment with the free and nPMMA-bounded BODIPYs (100 nM). Furthermore, no Control spheroids from both HCT116 and MCF7 cells increased their size following 24, 48 and 72 h incubation, as indicated by the increasing cell number. In particular, the HCT116 spheroids experienced faster growth than the MCF7 spheroids. Treatment with all compounds, followed by photoactivation, induced a reduction in viable cell numbers in both HCT116 and MCF7 spheroids. Interestingly, 3@nPMMA and 6@nPMMA were significantly more potent than 3 and 6. The intrinsic cytotoxicity of the BODIPYs was assessed by omitting the irradiation step from the treatment protocol and was found to be negligible in all cases.

BODIPY Distribution and Intensity
Confocal microscopy analysis of the distribution and intensity of the fluorescence of the PSs in the images of the equatorial planes obtained on HCT116 and MCF7 spheroids showed that the penetration of fluorescent 3f and 6f compounds, free or loaded onto nPMMA nanoparticles, was limited to the external cell rim of the spheroids, while no appreciable penetration was detected in the inner core ( Figure 10).
The internalization extent of compound 6f was generally greater than 3f. At the same time, no significant difference was observed in the fluorescence intensity of 3f@nPMMA and 6f@nPMMA in both cell lines compared to the free BODIPYs.
In agreement with confocal microscopy results, lower amounts of 3f, compared with 6f, were found in cells obtained from both HCT116 and MCF7 spheroids, following 24 h treatment with the free and nPMMA-bounded BODIPYs (100 nM). Furthermore, no differences in the amount of 3f@nPMMA and 6f@nPMMA, compared to free compounds, were observed in both cell lines ( Figure 11).

BODIPY Distribution and Intensity
Confocal microscopy analysis of the distribution and intensity of the fluorescence of the PSs in the images of the equatorial planes obtained on HCT116 and MCF7 spheroids showed that the penetration of fluorescent 3f and 6f compounds, free or loaded onto nPMMA nanoparticles, was limited to the external cell rim of the spheroids, while no appreciable penetration was detected in the inner core ( Figure 10). The internalization extent of compound 6f was generally greater than 3f. At the same time, no significant difference was observed in the fluorescence intensity of 3f@nPMMA and 6f@nPMMA in both cell lines compared to the free BODIPYs.
In agreement with confocal microscopy results, lower amounts of 3f, compared with 6f, were found in cells obtained from both HCT116 and MCF7 spheroids, following 24 h treatment with the free and nPMMA-bounded BODIPYs (100 nM). Furthermore, no differences in the amount of 3f@nPMMA and 6f@nPMMA, compared to free compounds, were observed in both cell lines ( Figure 11).

Discussion
Although several photosensitizers are available for PDT application, some critical drawbacks, such as dark toxicity, poor molar extinction coefficients, low photostability, selectivity, and biocompatibility, limit their translation into clinical practice [32,41]. Furthermore, most clinically approved PSs are cyclic tetrapyrroles, whose purification, synthesis, or synthetic modification is challenging and low-yielding [42]. Finally, due to the lack of tumor selectivity, the therapeutically active PS dosage is relatively high, thus posing additional issues of unwanted phototoxicity and side effects [32].
Recent studies have shown that the BODIPY (4,4-difluoro-4-bora-3a,4a-diaza-sindacene) scaffold represents an interesting chemical structure to develop PSs with improved features for clinical application, including PDT [9][10][11][12]. In addition, the use of suitably designed nanosystems for delivering PS molecules has been shown to be a promising approach in different preclinical anticancer PDT experiments [43][44][45]. In principle, loading the PS onto nanoparticles could favor its preferential accumulation at the target tissue, thanks to passive targeting, and protect the PS from degradation, thus, lowering the dosages and reducing the side effects.
In this work, we described, for the first time, the electrostatic loading of positively charged BODIPYs, namely 3 and 6, on negatively charged PMMA nanoparticles, affording reproducible nanoparticles, e.g., 3@nPMMA and 6@nPMMA, with diameters in the 140-200 nm range and with high colloidal stability. With respect to positively charged PMMA nanoparticles, previously described by our group [22,23], nPMMA displayed lower intrinsic toxicity, while maintaining suitable internalization capacity (Figures 4, 10 and 11) and ROS generation ability (Figure 3).

Discussion
Although several photosensitizers are available for PDT application, some critical drawbacks, such as dark toxicity, poor molar extinction coefficients, low photostability, selectivity, and biocompatibility, limit their translation into clinical practice [32,41]. Furthermore, most clinically approved PSs are cyclic tetrapyrroles, whose purification, synthesis, or synthetic modification is challenging and low-yielding [42]. Finally, due to the lack of tumor selectivity, the therapeutically active PS dosage is relatively high, thus posing additional issues of unwanted phototoxicity and side effects [32].
Recent studies have shown that the BODIPY (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene) scaffold represents an interesting chemical structure to develop PSs with improved features for clinical application, including PDT [9][10][11][12]. In addition, the use of suitably designed nanosystems for delivering PS molecules has been shown to be a promising approach in different preclinical anticancer PDT experiments [43][44][45]. In principle, loading the PS onto nanoparticles could favor its preferential accumulation at the target tissue, thanks to passive targeting, and protect the PS from degradation, thus, lowering the dosages and reducing the side effects.
In this work, we described, for the first time, the electrostatic loading of positively charged BODIPYs, namely 3 and 6, on negatively charged PMMA nanoparticles, affording reproducible nanoparticles, e.g., 3@nPMMA and 6@nPMMA, with diameters in the 140-200 nm range and with high colloidal stability. With respect to positively charged PMMA nanoparticles, previously described by our group [22,23], nPMMA displayed lower intrinsic toxicity, while maintaining suitable internalization capacity (Figures 4, 10 and 11) and ROS generation ability (Figure 3).
From the biological point of view, 3@nPMMA and 6@nPMMA exhibited intriguing photodynamic effects in both 2D-and 3D-cultured cancer cell lines. In monolayer cells, 3@nPMMA affected HCT116, SKOV3 and MDA-MB231 cellular viability at a similar extent to the free BODIPYs. On the other hand, a lower potency of 6@nPMMA compared to 6 was observed in all 2D-cultured cell lines. Internalization experiments, performed by taking advantage of the fluorescent BODIPY versions, 3f and 6f, partially supported the observed differences in phototoxic behaviors (Figure 4), confirming the close relationship between PSs accumulation and phototoxic effect [21,31,46].
The cellular response to PDT is closely associated with the type of photochemical reaction induced by the PS, e.g., type I or type II, and depends on the PS photophysical/photochemical nature, its subcellular localization, and the oxygen concentration. Type I sensitizers lead to the formation of superoxide and hydroxyl radicals. Type II sensitizers mainly generate singlet oxygen ( 1 O 2 ) [47]. BODIPYs are generally acknowledged as type II sensitizers. Indeed, compounds 3 and 6 confirmed their ability to produce 1 O 2 (Table 1) and ROS both when free and when loaded onto nPMMA (Figure 3 and Figure S7). Unfortunately, under our experimental conditions, both free and nPMMA-bound BODIPYs did not show a significant ROS production in the four tested cell lines ( Figure S10), probably due to the experimental settings.
It is generally recognized that, through the production of ROS and/or 1 O 2 , type I and type II photochemical reactions may trigger one or various cell death mechanisms, mainly apoptosis, necrosis and autophagy, that directly affect cancer cells and lead to tumor destruction [33,48,49]. Our results indicated that treatment with equitoxic concentrations of 3, 3@nPMMA, 6 and 6@nPMMA resulted in a different cell-type apoptotic response. In particular, the similar, or better, pro-apoptotic effects of the nPMMA-bound BODIPYs, compared to the free molecules, indicated a photodynamic advantage of the nPMMA formulations over free BODIPYs.
A key feature of an effective PDT treatment is the ability to accumulate and penetrate the inner and hypoxic core of the tumor mass. Indeed, the lower oxygen content and the non-homogeneous PS distribution would lead to reduced ROS generation and inefficient cell death. In this view, the use of 3D-cellular systems represents an increasingly considered model for mimicking the in vivo scenario, allowing gathering of important information on the potential pre-clinical transability of newly designed PDT systems [50,51]. Importantly, it was shown that within spheroids normoxic and hypoxic areas coexist, which make spheroids particularly useful for the study of new PSs [52].
Notably, in spheroids from HCT116 and MCF7 cells, 3@nPMMA and 6@nPMMA nanoparticles were significantly more potent than the free molecules ( Figure 9). Considering the comparable internalization degree of the formulations in the HCT116 and MCF7 3D models, the observed higher cytotoxic effect exerted by 3@nPMMA and 6@nPMMA suggested better in vivo performances of the nPMMA-bound BODIPYs.
Finally, the greater inhibition of cell migration observed, following the PDT treatment with 3@nPMMA and 6@nPMMA on MDA-MB231 and SKOV3 (Figure 6), whose high motility correlates with a greater degree of malignancy [53], could be ascribed to MMP2 and MMP9 inhibition (Figures 7 and 8). In fact, it is well known that the migratory behavior in tumor cells of epithelial origin is strongly related to the epithelial-to-mesenchymal transition (EMT) and is a harbinger of invasion and metastasis. A number of PSs have been reported to exhibit anti-migratory activity [53][54][55]. In addition, it has been reported that inhibition of EMT following PDT may occur through the inhibition of MMP2 and MMP9 [55,56] and account for the suppression of metastasis, thus ameliorating patient prognosis.

Conclusions
In conclusion, we reported the successful synthesis and characterization of negatively charged poly-methyl methacrylate nanoparticles as suitable and not intrinsically toxic carriers of positively charged BODIPYs. These novel nanosystems displayed effective PDT-mediated anticancer activity on a panel of solid tumor cell lines cultured both as 2Dand 3D-models. Most interestingly, compared to free molecules, BODIPY-bound PMMA nanoparticles were more effective in reducing cell viability in spheroids models, which better recapitulate the in vivo avascular tumor scenario. In addition, cell migration was significantly impaired upon BODIPY-bound nPMMA PDT treatment, and the inhibition of MMP2 and MMP9 was identified as a potential mechanism of the anti-migratory effect. These results, more evident on MDA-MB231 cells, which are widely recognized as less responsive to standard treatments and more prone to metastasize, support further investigations to validate this system into metastatic in vivo models of breast cancer. Finding alternative approaches to treat non-responsive primary tumors, while avoiding the spread of tumor cells to other sites, is crucial to increase the survival rate of patients and the quality of life experienced by them.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cancers15010092/s1, Figure S1. (A) Chemical synthesis of compounds 3f and 6f starting from derivatives 1a and 1b. (B) Absorbance and emission spectra of compound 3f and 6f. Figure S2. 1HNMR of compound 3f, (400 MHz, DMSO-d6). Figure S3. 13CNMR of compound 3f, (100 MHz, DMSO-d6). Figure S4. 1HNMR of compound 6f, (400 MHz, DMSO-d6). Figure S5. 13CNMR of compound 6f, (100 MHz, DMSO-d6). Figure S6. Graph showing the baseline profile of dichlorofluorescein (DCF) at different irradiation times. Figure S7. Graphs showing the ROS production of 3 and 6 displayed as absorption spectra of DCF, measured at different irradiation times (minutes). Figure S8. Effect of 24 h treatment with 3, 3@nPMMA, 6, and 6@nPMMA and nPMMA, 1 h irradiation and incubation, for 24 h in a drug free medium at 37 • C, on the migratory activity of SKOV3 cells. Scratch wounds were produced before the irradiation step and pictures were taken immediately after it (0) and after 24 h, through a camera connected to an Olympus IX81 microscope. The pictures shown are representative of three independent experiments. Figure S9. Effect of 24 h treatment with 3, 3@nPMMA, 6, and 6@nPMMA and nPMMA, 1h irradiation and incubation for 24 h in a drug free medium at 37 • C, on the migratory activity of MDA-MB231 cells. Scratch wounds were produced before the irradiation step and pictures were taken immediately after it (0) and after 24 h, through a camera connected to an Olympus IX81 microscope. The pictures shown are representative of three independent experiments. Figure S10. ROS levels in HCT116, SKOV3, MCF7 and MDA-MB231 cells following 24 h treatment with free-(green) or nPMMA-bound (red) 3 and 6 and 1 h irradiation. Fluorescence intensity was quantitated, based on the Median Fluorescence Intensity (MFI), and the results obtained in three independent experiments are reported in the graph (* p < 0.05, ** p < 0.01 vs. respective free-BODIPY).