Berberine Photo-Activation Potentiates Cytotoxicity in Human Astrocytoma Cells through Apoptosis Induction

Photodynamic therapy (PDT) has recently attracted interest as an innovative and adjuvant treatment for different cancers including malignant gliomas. Among these, Glioblastoma (GBM) is the most prevalent neoplasm in the central nervous system. Despite conventional therapeutic approaches that include surgical removal, radiation, and chemotherapy, GBM is characterized by an extremely poor prognosis and a high rate of recurrence. PDT is a physical process that induces tumor cell death through the genesis and accumulation of reactive oxygen species (ROS) produced by light energy interaction with a photosensitizing agent. In this contribution, we explored the potentiality of the plant alkaloid berberine (BBR) as a photosensitizing and cytotoxic agent coupled with a PDT scheme using a blue light source in human established astrocytoma cell lines. Our data mainly indicated for the combined BBR-PDT scheme a potent activation of the apoptosis pathway, through a massive ROS production, a great extent of mitochondria depolarization, and the sub-sequent activation of caspases. Altogether, these results demonstrated that BBR is an efficient photosensitizer agent and that its association with PDT may be a potential anticancer strategy for high malignant gliomas.


Introduction
Glioblastoma (GBM, WHO grade IV) is the most common and aggressive type of malignant primary brain tumor in humans [1]. When adopting the current treatment options, i.e., surgery tumor ablation, chemo-and/or radio-therapy protocols, relatively unchanged in the last decade [2], the disease inevitably progresses, however, and relapses within few months post diagnosis [3]. Despite a multitude of research efforts, ongoing therapeutic strategies, including for example first-line chemotherapeutic compounds as temozolomide or carmustine, have not determined a significantly higher efficacy in clinics mostly due to limited pharmacokinetics and to GBM intrinsic and acquired resistance to the treatments [4]. In this regard, the large extent of genetic and phenotypic heterogeneity among GBMs [5] has constituted a biological hurdle for the efficacy of the therapeutic agents toward tumor cells [6]. Therefore, exploiting new strategies that might overcome the cell resistance and heterogeneity features will be crucial to improve the quality of life and survival time of GBM patients [7]. Recently, clinical applications of photodynamic therapy (PDT) in oncologic contexts including GBM have gained increased attention [8][9][10].

Cell Culture and Chemicals
High grade astrocytoma established cell lines (i.e., T98G, U87-MG, U373-MG, and U138-MG) were obtained from the American Type Culture Collection (Manassas, VA, USA); Res186 low grade pediatric astrocytoma and rat normal astrocytes were respectively provided by Dr. M. Bobola (University of Washington, Seattle, WA, USA) and Prof. S. Schinelli (University of Pavia, Italy) and respectively described in [22,23]. Cells were routinely grown as monolayers at 37 • C in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/mL of penicillin and 100 µg/mL of penicillin-streptomycin (all reagents from Euroclone, Milan, Italy), under atmosphere controlled at 5% CO 2 .
Berberine (BBR, 99% purity, Sigma-Aldrich, St. Louis, MO, USA) was resuspended in dimethyl sulfoxide (DMSO) and administered at different concentrations (i.e., 20, 50, 100, 200 µg/mL) for 4 h. Next, cells were washed with PBS and incubated for 24 h post treatment (p.t.) with complete media without BBR before trypsinization. Control-mock administration experiments (using the same DMSO concentration adopted for BBR solutions) were performed to exclude DMSO-induced cytotoxicity (data not shown). To stain mitochondria in living cells, MitoTracker Deep Red FM (Thermofisher, Waltham, MA, USA), a far red-fluorescent dye (abs/em 644/665 nm) was employed as follows: cells were incubated with 10 nM dye at 37 • C for 45 min and then visualized by inverted fluorescent and confocal microscopes or trypsinized for cytofluorimetric analysis. The pan-caspase inhibitor Z-VAD(OMe)-FMK, purchased from Cell Signaling (Danvers, MA, USA), was used at 100 µM for 4 h. After this time interval, media were removed following PBS washing and fresh media without BBR/Z-VAD(OMe)-FMK was provided. Finally, cells were evaluated by cytofluorimetric analysis after 24 h p.t. To evaluate differences in the autophagy flux, the expression of LC3B-II isoform was assayed following bafilomycin A1 (Sigma-Aldrich) administration (10 nM at 4 h p.t.) and finally protein expression visualized by immunoblotting.

Cell Viability Assays
Cells were seeded at a density of 8 × 10 3 cells/well in 96-well plates in a volume of 200 µL for 24 h and then treated with several concentrations of BBR (20,50, 100, 200 µg/mL for 4 h). After 24 h p.t., 20 µL of Cell Titer One Aqueous Solution (Promega, Madison, WI, USA) was added in each well and incubated for 2 h at 37 • C. Then, absorbance was measured using a microplate reader (Sunrise, Tecan, Männedorf, Switzerland) at a wavelength of 492 nm. All experiments were performed in triplicate with independent assays.

BBR Photo-Stimulation
Cells were seeded at a density of 10 5 cells in 30 mm plates for 24 h. Cells (with an approximate 80% confluence) were then incubated for 4 h with BBR (200 µg/mL). Four treatment groups were considered: Group 1 (control, untreated), Group 2 (LED-irradiated control), Group 3 (BBR-only incubation), Group 4 (BBR + LED, incubated for 4 h with BBR and irradiated). After BBR incubation, cells were washed with PBS and 1 mL of fresh culture medium was then added to each plate. A blue LED source (Safe Imager 2.0 Blue-Light Transilluminator, Invitrogen, Carlsbad, CA, USA), operating at 447 nm and 1.2 mW/cm 2 of intensity, set for 4 min of application, was used for irradiation. All experiments were performed in triplicate independent assays.

Muse Cytofluorimetric Assays
A Muse cytofluorimeter and the corresponding assays (Luminex, Austin, TX, USA) were adopted for analysis of Annexin V (Annexin V and Cell Dead kit), caspases activation (Caspase 3/7, and MultiCaspase kits), evaluation of Reactive Oxygen Species (ROS) (Oxydative Stress kit), mitochondria depolarization (MitoPotential kit), autophagy activation (Autophagy LC3-Antibody Based kit) in T98G cells as untreated (NT), exposed to led-only stimulation (LED), BBR (200 µg/mL for 4 h) administration and with the combined LED+BBR scheme, all evaluated after 24 h p.t. For cytofluorimetric evaluations, trypsinized cells, derived from 10 5 cells in 30 mm plates grown for 24 h with an approximate 80% confluence, were analyzed according to the manufacturer's specifications of each assay. Experiments were performed in duplicates or triplicates, evaluating 2000 cells for each cytofluorimetric assay.

Microscopy Analysis
T98G cells were seeded into 35 mm glass-bottom petri dishes at a density of 10 4 /mL, and allowed to recover for 24 h prior to analysis.
Phase contrast and fluorescence images were obtained using an inverted microscope (Nikon Eclipse TS100) using a 40× and a 100× plan fluor oil immersion objective, respectively as described [24]. For BBR fluorescence detection, the UV2A (Nikon, Tokyo, Japan) block filter (355/50 nm; dichroic mirror = 400, barrier filter = 410) was employed. BBR and mitochondria were also visualized using a Nanolive 3D CELL EXPLORER-fluo microscope (MEDIA System Lab, Macherio, Italy) using fluorescence channels FITC (excitation peak at 491 nm and an emission peak at 516 nm) and Cy5 (excitation peak at 646 nm and an emission peak at 664 nm), respectively for BBR and MitoTracker.
For confocal microscopy, BBR and mitochondria fluorescence were imaged using a TCS SP8 X system (Leica, Wetzlar, Germany) with excitation peak at 488 nm and emission at 500-600 nm for BBR and excitation peak at 641 nm and emission at 650-750 nm for MitoTracker, with a 63X oil objective (NA = 1,4, zoom 2).
For transmission electron microscopy (TEM) analysis, T98G cells (10 6 ) were grown in DMEM medium in 90 mm plates. Following LED and/or BBR (200 µg/mL for 4 h) stimulation for 4 min, after additional 24 h p.t., cells were harvested by centrifugation at 800 rpm for 5 min and fixed with 2.5% glutaraldehyde in PBS, for 2 h at room temperature as described [25]. Cells were then rinsed in PBS (pH 7.2) overnight and post-fixed in 1% aqueous OsO 4 for 1 h at room temperature. Cells were pre-embedded in 2% agarose in water, dehydrated in acetone, and finally embedded in epoxy resin (Electron Microscopy Sciences, EM-bed812). Ultrathin sections (60-80 nm) were collected on nickel grids and stained with uranyl acetate and lead citrate. The specimens were observed with a JEM 1200 EX II (JEOL, Peabody, MA, USA) electron microscope, equipped with a MegaView G2 CCD camera (Olympus OSIS, Tokyo, Japan) and operating at 120 kV. The morphology of mitochondria (at least 20 for each sample) was then analyzed by two independent evaluators.

DNA Extraction and Electrophoretic Analysis
Genomic DNA was isolated from 10 6 T98G cells (i.e., untreated; LED-only; BBR 200 µg/mL for 4 h; BBR combined with led stimulation) as follows: cell pellets were gently resuspended into 50 µL of TES lysis buffer (Tris-OH pH 7.5 10 mM; EDTA pH 8.0, 10 mM; SDS 0.5% (v/v), all from Sigma-Aldrich), supplemented with 10 µL of RNAse Cocktail (Thermofisher), mixed well by flipping the tip of the tubes without vortexing and incubating for 90 min at 37 • C. Then, 10 µL of proteinase K (Sigma-Aldrich) was added following an incubation of 50 • C for 90 min. DNA was then quantified by a Qubit fluorimeter using DNA HS Assay kit (Thermofisher) and finally loaded into 2% (w/v) agarose gel. Electrophoresis was performed at 15 V for 5 h and visualized using blue LED transilluminator (Safe Imager 2.0 Blue-Light Transilluminator, Invitrogen).

Clonogenic Assay
To evaluate plating efficiency, a clonogenic survival assay was performed as described [27]. Briefly, 24 h after BBR and/or PDT stimulation, cells were trypsinized and seeded into 6-well plates (10 4 cells per well), incubated for two weeks at 37 • C and then fixed with ethanol. Cells were stained with 0.5% (v/v) Crystal Violet (Sigma-Aldrich), and colonies that contained more than 50 cells were automatically counted using Im-ageJ colony-counter (https://imagej.nih.gov/ij/plugins/colony-counter.html, accessed on 19 April 2020). The number of clones/well was calculated and normalized to the corresponding control samples. Each experiment was performed in triplicate.

Statistical Analysis
The data were analyzed using the statistics functions of the MedCalc statistical software version 18.11.6. (http://www.medcalc.org, accessed on 15 February 2021). The ANOVA test differences were considered statistically significant when p < 0.05.

BBR Effect and Cellular Internalization in Cancer and Normal Astrocyte Cells
According to published results, reduction in astrocytoma cell line viability was observed after 24 h administration of BBR in a range of 20-200 µg/mL [28][29][30]. Moreover, the compound was reported to be efficiently internalized into the cells after 3 h incubation [19]. Here, different BBR concentrations were administered for 4 h and subsequently cells were analyzed at 24 h post treatment (p.t.).
Specifically, by means of MTS assays aimed at assessing cell viability, the cytotoxicity of BBR in different low-(i.e., Res186) and high-grade (T98G, U87-MG, U373-MG, and U138-MG) astrocytoma cells lines was investigated and compared to the effects of this compound in rat primary astrocytes ( Figure 1). U138-MG) astrocytoma cells lines was investigated and compared to the effects of this compound in rat primary astrocytes ( Figure 1). As reported, BBR affected astrocytoma cells viability in a dose-dependent manner, while normal astrocyte cells showed only a slight reduction of viability after the alkaloid administration doses. In particular, when treated with 200 μg/mL BBR for 4 h, T98G cells exhibited the lowest average viability (i.e., 19.50%) compared to Res186 (39.44%), U87-MG (44.37%), U373-MG (72.36%), and U138-MG (72.37%); on the contrary, normal rat astrocytes showed 85.13% viability under these conditions.
To evaluate if the cytotoxic effect of BBR in cancer cells was dependent on uptake efficiency, BBR internalization was analyzed by flow cytometry (Amnis ImageStream MkII, Luminex). For this purpose, T98G cells and normal rat astrocytes were treated with 200 μg/mL BBR for 4 h and visualized after 24 h p.t. As reported in Figure 2, normal rat astrocytes displayed a lower percentage of BBR-positive fluorescent cells compared to T98G cells (i.e., 58.32 and 97.03%, respectively). Moreover, the intensity of fluorescence was higher in the astrocytoma cell line. As reported, BBR affected astrocytoma cells viability in a dose-dependent manner, while normal astrocyte cells showed only a slight reduction of viability after the alkaloid administration doses. In particular, when treated with 200 µg/mL BBR for 4 h, T98G cells exhibited the lowest average viability (i.e., 19.50%) compared to Res186 (39.44%), U87-MG (44.37%), U373-MG (72.36%), and U138-MG (72.37%); on the contrary, normal rat astrocytes showed 85.13% viability under these conditions.
To evaluate if the cytotoxic effect of BBR in cancer cells was dependent on uptake efficiency, BBR internalization was analyzed by flow cytometry (Amnis ImageStream MkII, Luminex). For this purpose, T98G cells and normal rat astrocytes were treated with 200 µg/mL BBR for 4 h and visualized after 24 h p.t. As reported in Figure 2, normal rat astrocytes displayed a lower percentage of BBR-positive fluorescent cells compared to T98G cells (i.e., 58.32 and 97.03%, respectively). Moreover, the intensity of fluorescence was higher in the astrocytoma cell line.
Flow cytometry analysis (Amnis ImageStream MkII) was then employed to evaluate differences of BBR accumulation in nuclei of T98G cells ( Figure S1). Following BBR administration (200 µg/mL for 4 h) and DRAQ5 nuclear staining, the percentage of cells displaying nuclear co-localization signals was 14.78%. Furthermore, in agreement with that previously described by Agnarelli et al. [28], a higher BBR concentration (i.e., 500 µg/mL) significantly increased BBR nuclear internalization, resulting in 29.04% of BBR-positive cells. To shed light on the cytoplasmic localization of BBR in T98G cells, fluorescent and confocal microscopy were used to assess BBR and mitochondria co-localization signals following administration of 200 µg/mL BBR for 4 h. At 24 h p.t, cells showed extensive co-localized signals ( Figure 3).

Percentages of focused cells with BBR internalization spots and Internalization
Erode median parameters are reported. BBR internalization was analyzed by "Internalization" wizard with the following gating strategy: single cells were gated (using Area/Aspect Ratio Intensity, R1); R1 was then gated for focused cells using the "Gradient-RMS" feature and finally Ch2 intensity was measured with "Internalization" feature and an "Erode" mask applied. R2 gate (not shown) identified focused cells. Examples of cell images of R3 (negative control, i.e., cells without BBR) and R4 gates are shown.
Flow cytometry analysis (Amnis ImageStream MkII) was then employed to evaluate differences of BBR accumulation in nuclei of T98G cells ( Figure S1). Following BBR administration (200 μg/mL for 4 h) and DRAQ5 nuclear staining, the percentage of cells displaying nuclear co-localization signals was 14.78%. Furthermore, in agreement with that previously described by Agnarelli et al. [28], a higher BBR concentration (i.e., 500 μg/mL) significantly increased BBR nuclear internalization, resulting in 29.04% of BBR-positive cells. To shed light on the cytoplasmic localization of BBR in T98G cells, fluorescent and

Percentages of focused cells with BBR internalization spots and Internalization
Erode median parameters are reported. BBR internalization was analyzed by "Internalization" wizard with the following gating strategy: single cells were gated (using Area/Aspect Ratio Intensity, R1); R1 was then gated for focused cells using the "Gradient-RMS" feature and finally Ch2 intensity was measured with "Internalization" feature and an "Erode" mask applied. R2 gate (not shown) identified focused cells. Examples of cell images of R3 (negative control, i.e., cells without BBR) and R4 gates are shown.
confocal microscopy were used to assess BBR and mitochondria co-localization signals following administration of 200 μg/mL BBR for 4 h. At 24 h p.t, cells showed extensive colocalized signals (Figure 3).

BBR Associated with PDT Induces Apoptosis
BBR cytotoxicity and the combined effect of PDT stimulation was investigated in normal rat astrocytes, low-(Res186) and high-grade astrocytoma cells (i.e., U87-MG, U138-MG, T98G, and U373-MG). Among the investigated cells, the T98G cells displayed at 200 μg/mL BBR the highest cytotoxic PDT induced effect ( Figure 4). Based on these results, deeper cellular and molecular investigations were therefore focused on T98G cells.

BBR Associated with PDT Induces Apoptosis
BBR cytotoxicity and the combined effect of PDT stimulation was investigated in normal rat astrocytes, low-(Res186) and high-grade astrocytoma cells (i.e., U87-MG, U138-MG, T98G, and U373-MG). Among the investigated cells, the T98G cells displayed at 200 µg/mL BBR the highest cytotoxic PDT induced effect ( Figure 4). Based on these results, deeper cellular and molecular investigations were therefore focused on T98G cells. . Cell Viability Assays after BBR and/or PDT (LED) administration. Normal rat astrocytes, low-(i.e., Res186) and high-grade (U87-MG, T98G, U373-MG. and U138-MG) astrocytoma cell lines were treated with different concentrations of BBR (0, 20, 50, 100, 200 μg/mL for 4 h, white bars) and exposed to LED stimulation for 4 min (grey bars). Colorimetric MTS analysis was carried out after 24 h p.t. All experiments were performed in triplicates with independent assays. Asterisks indicated statistical significance compared to untreated cells (p < 0.05, Anova One-way).
Since BBR was previously reported to induce apoptotic and autophagy programmed cell death pathways in different cancer cells [28][29][30], a preliminary evaluation of Annexin V and LC3 expression in T98G cells was performed. Apoptosis induction was primarily assayed through cytofluorimetric Annexin V experiments in T98G cells. Cells treated with increasing BBR concentrations (i.e., 20, 50, 100, 200 μg/mL for 4 h) or exposed to a combined BBR and led stimulation (4 min) treatment (BBR + LED) were assayed at 24 h p.t. Untreated (NT) and LED-only treated cells were used as controls ( Figure S2). Remarkably, the combined scheme (BBR + LED) resulted in a dose-dependent reduction in the fraction of live cells associated with a significant increase in early and late apoptotic events (in particular, using 200 μg/mL BBR: BBR coupled with LED stimulation = 46.68% total apoptosis; BBR = 29.82%; LED = 9.18%; untreated cells = 7.52%).
The above mentioned BBR and PDT treatment resulted also in affecting T98G longterm viability: at two weeks after a 4-h exposure to 100 or 200 μg/mL BBR following PDT scheme, clonogenic assays showed a dose-dependent decrease in the number of T98G colonies ( Figure S3).
In relation to the autophagic pathway, no significant variations in LC3B expression by Muse cytofluorimetric and immunoblotting analysis in the presence of bafilomycin A1 (10 nM) were scored in BBR-led stimulated T98G cells compared to untreated ones ( Figure  S4). . Cell Viability Assays after BBR and/or PDT (LED) administration. Normal rat astrocytes, low-(i.e., Res186) and high-grade (U87-MG, T98G, U373-MG. and U138-MG) astrocytoma cell lines were treated with different concentrations of BBR (0, 20, 50, 100, 200 µg/mL for 4 h, white bars) and exposed to LED stimulation for 4 min (grey bars). Colorimetric MTS analysis was carried out after 24 h p.t. All experiments were performed in triplicates with independent assays. Asterisks indicated statistical significance compared to untreated cells (p < 0.05, Anova One-way).
Since BBR was previously reported to induce apoptotic and autophagy programmed cell death pathways in different cancer cells [28][29][30], a preliminary evaluation of Annexin V and LC3 expression in T98G cells was performed. Apoptosis induction was primarily assayed through cytofluorimetric Annexin V experiments in T98G cells. Cells treated with increasing BBR concentrations (i.e., 20, 50, 100, 200 µg/mL for 4 h) or exposed to a combined BBR and led stimulation (4 min) treatment (BBR + LED) were assayed at 24 h p.t. Untreated (NT) and LED-only treated cells were used as controls ( Figure S2). Remarkably, the combined scheme (BBR + LED) resulted in a dose-dependent reduction in the fraction of live cells associated with a significant increase in early and late apoptotic events (in particular, using 200 µg/mL BBR: BBR coupled with LED stimulation = 46.68% total apoptosis; BBR = 29.82%; LED = 9.18%; untreated cells = 7.52%).
The above mentioned BBR and PDT treatment resulted also in affecting T98G longterm viability: at two weeks after a 4-h exposure to 100 or 200 µg/mL BBR following PDT scheme, clonogenic assays showed a dose-dependent decrease in the number of T98G colonies ( Figure S3).
In relation to the autophagic pathway, no significant variations in LC3B expression by Muse cytofluorimetric and immunoblotting analysis in the presence of bafilomycin A1 (10 nM) were scored in BBR-led stimulated T98G cells compared to untreated ones ( Figure S4).  According to the indication of an apoptotic involvement, flow cytometric analysis of BBR and BBR-PDT treated cells evidenced an increase in nuclear fragmentation processes compared to untreated and LED-only treated controls ( Figure 6). According to the indication of an apoptotic involvement, flow cytometric analysis of BBR and BBR-PDT treated cells evidenced an increase in nuclear fragmentation processes compared to untreated and LED-only treated controls ( Figure 6). Trypsinized cells were analyzed for bright field (Ch04) and for nuclear fluorescence (Ch05) using DRAQ5 dye (Invitrogen) at 60× magnification. R3 and R4 gates indicated integrity and nuclear fragmentation signals, respectively. Mean Spot Count Peak values (i.e., 0.323 and 1.367) and cell images representative of R3 and R4 gates are reported. R1 and R2 gates (not shown) identified focused cells.
To further confirm the nuclear fragmentation patterns induced by BBR and PDT stimulation, genomic DNA was extracted from untreated, LED-only, BBR-only and BBR coupled with PDT stimulated T98G cells. Gel electrophoresis analysis showed marked DNA fragmentation patterns in BBR and BBR-LED treated cells, compared to untreated and LED stimulated ones (Figure 7). To further confirm the nuclear fragmentation patterns induced by BBR and PDT stimulation, genomic DNA was extracted from untreated, LED-only, BBR-only and BBR coupled with PDT stimulated T98G cells. Gel electrophoresis analysis showed marked DNA fragmentation patterns in BBR and BBR-LED treated cells, compared to untreated and LED stimulated ones (Figure 7). To assess the activation of apoptosis following BBR administration and PDT stimulation, T98G cells were examined for the expression of the main apoptotic markers. As reported in Figure 8, immunoblotting analysis highlighted a clear decrease in the amount of the examined pro-caspases (i.e., 3, 7, and 9) accompanied by the appearance of the respective cleaved forms, mainly evident in the combined BBR-LED treatment along with the cleavage of PARP-1 full length protein. In addition, a marked reduction in the expression of the anti-apoptotic proteins Bcl-2 and Bcl-xL was reported following BBR administration and with the combined BBR-LED treatment. To assess the activation of apoptosis following BBR administration and PDT stimulation, T98G cells were examined for the expression of the main apoptotic markers. As reported in Figure 8, immunoblotting analysis highlighted a clear decrease in the amount of the examined pro-caspases (i.e., 3, 7, and 9) accompanied by the appearance of the respective cleaved forms, mainly evident in the combined BBR-LED treatment along with the cleavage of PARP-1 full length protein. In addition, a marked reduction in the expression of the anti-apoptotic proteins Bcl-2 and Bcl-xL was reported following BBR administration and with the combined BBR-LED treatment.
Furthermore, two caspase-specific cytofluorimetric assays (i.e., Muse Caspase 3/7 and Muse MultiCaspase, Luminex), reported respectively in Figures S5 and S6, clearly documented the marked and significant increase in caspase activation following the blue-LED photo-activation of BBR in T98G cells, accompanied by a drastic reduction in viable cells. Of note, according to the reported scores, LED stimulation alone did not affect cell viability. ) treated, and BBR + LED exposed T98G cells were evaluated after 24 h p.t. by immunoblotting. For each sample, 40 μg of total protein extracts was loaded into 10% (for PARP-1) and 12% polyacrylamide gels. All the experiments were performed in triplicate with independent assays. A representative experiment is shown. Molecular weights in KDa are reported. Furthermore, two caspase-specific cytofluorimetric assays (i.e., Muse Caspase 3/7 and Muse MultiCaspase, Luminex), reported respectively in Figures S5 and S6, clearly documented the marked and significant increase in caspase activation following the blue-LED photo-activation of BBR in T98G cells, accompanied by a drastic reduction in viable cells. Of note, according to the reported scores, LED stimulation alone did not affect cell viability.
To further verify the involvement of the apoptotic process following PDT stimulation, the experimental scheme of BBR administration in presence/absence of photo-activation was combined with the provision of the specific synthetic peptide Z-VAD(OMe)-FMK, a cell permeable, irreversible pan-caspase inhibitor that displays non-cytotoxic effects [31]. Specifically, the peptide was administered at 100 μM, 4 h after BBR administration and/or PDT stimulation. T98G cells were then evaluated after 24 h p.t., in viability and in caspase expression using the cytofluorimetric Muse MultiCaspase assay (Luminex). As illustrated in Figure 9, the pan-caspase inhibitor significantly rescued cell viability to 63.20% compared the corresponding treatment (BBR and PDT, without Z-VAD(OMe)-FMK) whose overall viability declined to 5.15%. An opposite and significant trend was scored considering caspase+/dead cell events (58.40 vs 18.75%, respectively). ) treated, and BBR + LED exposed T98G cells were evaluated after 24 h p.t. by immunoblotting. For each sample, 40 µg of total protein extracts was loaded into 10% (for PARP-1) and 12% polyacrylamide gels. All the experiments were performed in triplicate with independent assays. A representative experiment is shown. Molecular weights in KDa are reported.
To further verify the involvement of the apoptotic process following PDT stimulation, the experimental scheme of BBR administration in presence/absence of photo-activation was combined with the provision of the specific synthetic peptide Z-VAD(OMe)-FMK, a cell permeable, irreversible pan-caspase inhibitor that displays non-cytotoxic effects [31]. Specifically, the peptide was administered at 100 µM, 4 h after BBR administration and/or PDT stimulation. T98G cells were then evaluated after 24 h p.t., in viability and in caspase expression using the cytofluorimetric Muse MultiCaspase assay (Luminex). As illustrated in Figure 9, the pan-caspase inhibitor significantly rescued cell viability to 63.20% compared the corresponding treatment (BBR and PDT, without Z-VAD(OMe)-FMK) whose overall viability declined to 5.15%. An opposite and significant trend was scored considering caspase+/dead cell events (58.40 vs 18.75%, respectively). To better discern the effect of the photo-activation scheme within T98G cells, the cytometric Muse Oxidative Stress assay (Luminex) was employed to simultaneously determine the count and percentage of cells undergoing oxidative stress based on the intracellular detection of superoxide radicals (ROS) using dihydroethidium, a well-characterized reagent extensively used to detect ROS [31]. As schematized in Figure 10, ROS positive cells in the combined scheme evaluated at 24 h p.t. reached 85.13%, compared to 37.90% To better discern the effect of the photo-activation scheme within T98G cells, the cytometric Muse Oxidative Stress assay (Luminex) was employed to simultaneously determine the count and percentage of cells undergoing oxidative stress based on the intracellular detection of superoxide radicals (ROS) using dihydroethidium, a well-characterized reagent extensively used to detect ROS [31]. As schematized in Figure 10  Next, to evaluate differences in the sub-cellular localization of BBR, particularly within mitochondria, following PDT stimulation, fluorescent microscopy and flow cytometry (Amnis ImageStream MkII, Luminex) examinations were used to visualize and quantify BBR and mitochondria co-localization signals following BBR administration (200 μg/mL for 4 h) with/without LED stimulation and finally analyzed at 24 h p.t. The results, reported in Figure S7, indicated roughly similar co-localization ratios between the investigated samples (i.e., 2.424 and 2.396, respectively for BBR and BBR and LED combined scheme), suggesting that rather than differences induced by LED stimulation in the subcellular localization of BBR, photochemical processes might be responsible for the more pronounced cytotoxic effect produced by the combination of BBR and light stimulation.
Then, since ROS over-production leads to mitochondria dysfunction as widely documented [32,33], the functionality of mitochondria following BBR and PDT stimulation Next, to evaluate differences in the sub-cellular localization of BBR, particularly within mitochondria, following PDT stimulation, fluorescent microscopy and flow cytometry (Amnis ImageStream MkII, Luminex) examinations were used to visualize and quantify BBR and mitochondria co-localization signals following BBR administration (200 µg/mL for 4 h) with/without LED stimulation and finally analyzed at 24 h p.t. The results, reported in Figure S7, indicated roughly similar co-localization ratios between the investigated samples (i.e., 2.424 and 2.396, respectively for BBR and BBR and LED combined scheme), suggesting that rather than differences induced by LED stimulation in the subcellular localization of BBR, photochemical processes might be responsible for the more pronounced cytotoxic effect produced by the combination of BBR and light stimulation.
Then, since ROS over-production leads to mitochondria dysfunction as widely documented [32,33], the functionality of mitochondria following BBR and PDT stimulation was evaluated in T98G cells, using the specific Muse MitoPotential cytofluorimetric assay (Luminex). As a result, a prominent mitochondrial depolarization was scored in the combined treatment scheme, with in particular a highly significant increase in both mean death/depolarized and in total depolarized cells (52.80 and 70.32%, respectively), compared to BBR (10.98 and 25.62%), LED treated (18.00 and 29.34%) and untreated cells (16.61 and 22.24%) ( Figure 11). Again, LED-only stimulation exhibited similar mitochondria depolarization percentages of untreated cells. Furthermore, BBR and the combined treatment with PDT caused significant mitochondrial damage compared to untreated and LED-only treated cells. was evaluated in T98G cells, using the specific Muse MitoPotential cytofluorimetric assay (Luminex). As a result, a prominent mitochondrial depolarization was scored in the combined treatment scheme, with in particular a highly significant increase in both mean death/depolarized and in total depolarized cells (52.80 and 70.32%, respectively), compared to BBR (10.98 and 25.62%), LED treated (18.00 and 29.34%) and untreated cells (16.61 and 22.24%) ( Figure 11). Again, LED-only stimulation exhibited similar mitochondria depolarization percentages of untreated cells. Furthermore, BBR and the combined treatment with PDT caused significant mitochondrial damage compared to untreated and LED-only treated cells. To confirm mitochondrial damage and the subsequent activation of caspase pathways, a flow cytometry quantification of the release of Cytochrome c from the mitochondria of apoptotic cells was performed using Amnis ImageStream MkII. As reported in Figure 12, untreated (NT) or LED-only stimulated cells (LED) demonstrated higher levels of Cytochrome c staining while BBR-treated (BBR) and BBR-LED stimulated (BBR + LED) cells which have released their Cytochrome c from the mitochondria to the cytoplasm showed significant reduced staining intensity when probed with an anti-Cytochrome c DyLight 650 conjugated antibody.
is reported (B). Asterisks indicated statistical significance compared to untreated cells (p < 0.05, Anova One-way).
To confirm mitochondrial damage and the subsequent activation of caspase pathways, a flow cytometry quantification of the release of Cytochrome c from the mitochondria of apoptotic cells was performed using Amnis ImageStream MkII. As reported in Figure 12, untreated (NT) or LED-only stimulated cells (LED) demonstrated higher levels of Cytochrome c staining while BBR-treated (BBR) and BBR-LED stimulated (BBR + LED) cells which have released their Cytochrome c from the mitochondria to the cytoplasm showed significant reduced staining intensity when probed with an anti-Cytochrome c DyLight 650 conjugated antibody.  Lastly, ultrastructural analysis by TEM, reported in Figure 13, revealed swollen mitochondria with clear damage to their cristae and a reduction in the mitochondrial matrix density in correspondence with BBR and BBR LED-combined treatments.
Lastly, ultrastructural analysis by TEM, reported in Figure 13, revealed swollen mitochondria with clear damage to their cristae and a reduction in the mitochondrial matrix density in correspondence with BBR and BBR LED-combined treatments.

Discussion
Among glial-derived cancers, GBM is the most aggressive and common adult primary intracranial neoplasm with extremely poor prognosis. Conventional treatments for GBM, as surgery, radiotherapy, and chemotherapy, produced relatively modest effects towards the patients' prognosis. Consequently, new therapeutic strategies are nowadays urgently required [4]. In this regard, increasing evidence has demonstrated that PDT, a two-stage treatment that combines light energy with a drug (photosensitizer), is significantly effective against different types of cancers [34], including malignant brain tumors [8,9,35,36]. Importantly, in the search for novel photosensitizing agents, the natural plant-

Discussion
Among glial-derived cancers, GBM is the most aggressive and common adult primary intracranial neoplasm with extremely poor prognosis. Conventional treatments for GBM, as surgery, radiotherapy, and chemotherapy, produced relatively modest effects towards the patients' prognosis. Consequently, new therapeutic strategies are nowadays urgently required [4]. In this regard, increasing evidence has demonstrated that PDT, a two-stage treatment that combines light energy with a drug (photosensitizer), is significantly effective against different types of cancers [34], including malignant brain tumors [8,9,35,36]. Importantly, in the search for novel photosensitizing agents, the natural plant-derived alkaloid BBR has demonstrated a broad spectrum of efficacy [37][38][39] as well as in conjunction with PDT anticancer schemes [19].
To test the potentiality of this compound in PDT applications in GBM cells, internalization studies were first performed to confirm that BBR tend to accumulate at higher concentrations in the tumor compared to normal cells, possibly due to the BBR affinity to low density lipoproteins (LDL) that act as carriers of BBR moieties [19]. In tumor cells, indeed, the increase of cholesterol catabolism results in an over-expression of LDL receptors (specifically, B/E receptors). It was previously reported that LRP are abundant in different GBM cell lines [40][41][42]. For example, in U87-MG glioma cells, Andreazza and collaborators [19] estimated the binding of 400 BBR molecules per LDL moiety. It has been documented that photosensitizers, with cationic charges as BBR, can localize in mitochondria [43] due to the influence of the mitochondrial membrane potential as well as the lipid bilayer composition of the membrane [44]. It has also been assayed that carcinoma cells preferentially accumulate and retain certain cationic dyes to a much greater extent than most normal cells [45]. This evidence might therefore explain the significant difference we scored in BBR cellular uptake, comparing normal rat astrocytes with T98G GBM cells. Furthermore, in agreement with Lin et al. [46], a nearly 100% uptake was observed when tumor cells were incubated with BBR for a time interval between 2 and 4 h.
Different studies reported that BBR mostly affected the mitochondrial function of cancer cells, due to its selective accumulation in these organelles, through the interaction with the adenine nucleotide translocator [47]. On the other hand, we confirmed in T98G cells that a relatively high BBR concentration (i.e., 500 µg/mL) induced a significant accumulation of the alkaloid even in the nuclei compartments, as also documented in the U343 GBM cell line [29]. Importantly, the intracellular distribution of PS, including therefore BBR, is one of the key determinants of PDT efficacy, and its final subcellular localization is well correlated with the specific induced mechanism of cell death [48,49]. We therefore employed a specific BBR concentration evaluating through cytofluorimetric analysis its prevalent accumulation within the cytoplasm of T98G cells. Next, considering that a light device with a blue LED source produced an excited state of BBR [50], we employed a similar energy stimulation to evaluate the PDT effect in T98G cells, one of the most employed established GBM cell lines. Furthermore, as documented by Caverzan and collaborators [51] among several tested GBM cell lines, T98G displayed the higher gene-transcription levels of antioxidant enzymes accompanied by the smallest endogenous oxidative-stress levels; these cells can be therefore reasonably considered as paradigmatic substrates characterized by a relatively high level of intrinsic resistance to the photo-activation process.
It was previously reported that BBR treatment activated apoptosis enhancing procaspase-3 cleavage and therefore caspase-3 activity in several tumor cells [28,[52][53][54], including U251 and U87-MG GBM cell lines [28,30]. Conversely, Floriano and collaborators [55] did not report significant differences in the activation of the key executioner of apoptosis caspase 3 in cervical carcinoma cells, leading to the hypothesis that the effect of BBR on key caspase activity may be cell specific. Regarding these considerations, we primarily confirmed in vitro evidence reporting that BBR induced apoptosis of T98G cells by means of the activation of mitochondrial/caspase-dependent pathways in agreement with previous results [28,56]. As initial hallmarks of the apoptosis involvement, we highlighted genomic DNA and nuclei fragmentation patterns, subsequently accompanied by caspase activation through immunoblotting and cytofluorimetric independent assays. Importantly, we also reported a significant reduction in the expression of two main anti-apoptotic proteins, i.e., Bcl-2 and Bcl-xL. In this regard, it is important to underline that the reduction of Bcl-2 through the use of specific antagonists, represents a potential strategy for the treatment of tumors [57]. In addition, we reported a massive increase in intracellular production of ROS and of mitochondrial membrane depolarization events accompanied by a significant release of mitochondrial Cytochrome c following BBR and light stimulation. Electron microscopy analyses documented that BBR and the combined treatment with photo-stimulation produced a marked damaged of mitochondria, with disrupted and discontinuous outer membranes and cristae. Again, the identification of an evident mitochondria vulnerability might be a promising direction for the generation of antitumor drug development [58].
Of importance, according to our documented results, these apoptotic features were further amplified in T98G cells following the photosensitizing scheme, while the control treatment (i.e., with LED-only stimulation) was not effective and nearly identical to the untreated samples. Particularly, through the use of a specific pan-caspase inhibitor, i.e., Z-VAD(OMe)-FMK already assayed in T98G cells [59], we further confirmed the involvement of the apoptotic process within the induced cytotoxicity in T98G cells. This overall evidence strongly supported the concept that BBR can constitute a valid substrate in PDT stimulation schemes. Additional results highlighted similar cytotoxic effect of the developed BBR-PDT protocol in low-(Res186 and Res 259) and high-grade (U87-MG, U373-MG, and U138-MG) glioma cells.
On the other hand, our investigation did not highlight a marked involvement of the autophagy process following BBR and irradiation. First, direct optical microscope examinations did not reveal the presence of cytoplasmic vacuoles resembling autophagosomes and specific cytofluorimetric and immunoblotting analysis of LC3 expression, one of the most validated autophagy markers according to current guidelines [60], indicated slight but not significant variations following the combined BBR-LED treatment. Differently, other contributions reported autophagy activation in response to photodynamic stimulation at 24 h p.t., albeit in cervical carcinoma cancer cells [56]. On the other hand, the autophagy activation was highlighted in U343 GBM cells but in the absence of caspase-3-dependent apoptosis [16]. However, we cannot exclude that the activation of the autophagy process might be strictly related to the BBR concentration adopted and to the physical parameters of the irradiation scheme. It is also important to consider that the crosstalk between autophagic and apoptotic signaling pathways is complex and not always discernible in their specific contribution. A detailed study reported that an autophagic survival response was triggered in the presence of a relatively low level of photo-damage and that the programmed cell death fate was rescued when apoptosis was blocked [61]. This experimental scenario might be similarly consistent to that revealed in our study. On another level of complexity, some chemotherapeutics but not photosensitizers per se, induced both pathways, namely temozolomide in established glioma cells [62] and givinostat in glioma stem cells [63]. Consequently, in our specific experimental scheme, based on the synergic effect of BBR and PDT, additional studies might be necessary to better decipher at the molecular levels the players involved in the apoptotic/autophagy interplay.
In relation to GBM, it was recently reported that IDH1 mutation and the related production of 2-HG can induce oxidative stress, autophagy, and apoptosis [64]. Based on these premises, it might be interesting to evaluate if IDH1 mutated cells, for instance R132H polymorphism, induces additional and cumulative PDT effects (i.e., production of ROS and programmed cell death induction pathways) compared to cells exhibiting the wildtype allele. As an additional perspective, in vivo studies are scheduled to determine the possibility that BBR combined with PDT and combined with optical nanofiber devices could be an effective therapeutic agent for the management of malignant gliomas.