Lasers in Surgery and Medicine 46:319–334 (2014)
Photodynamic Therapy With the Novel Photosensitizer Chlorophyllin f Induces Apoptosis and Autophagy in Human Bladder Cancer Cells Du Lihuan, MD,1 Zheng Jingcun, PhD, MD,2 ** Jiang Ning, MD,2 Wang Guozeng, MD,2 Chu Yiwei, PhD,3 Lin Wei, MD,3 Qian Jing, MD,3 Zhang Yuanfang, MD,4 and Chen Gang, PhD, MD1* 1 Department of Urology, Jin Shan Hospital, Fudan University, Shanghai 201508, China 2 Department of Urology, Gongli Hospital, Shanghai 200135, China 3 Department of Immunology, Fudan University, Shanghai 2000032, China 4 Department of Urology, Hua Shan Hospital, Fudan University, Shanghai 200040, China
Background and Objectives: Our group recently synthesized a new, low-cost photosensitizer, chlorophyllin f. In this study, the effects of chlorophyllin f-mediated photodynamic therapy (PDT) and its potential mechanisms were examined in human bladder cancer cells. Materials and Methods: MitoTracker1 Green probe and LysoTracker1 Green probe were used to label mitochondria and lysosomes, revealing the intracellular localization of chlorophyllin f in 5637 and T24 cells by confocal laser scanning microscopy (CLSM). The cells were treated with chlorophyllin f-mediated PDT; the photo-cytotoxicity of chlorophyllin f was monitored using the Cell Counting Kit-8 assay, and apoptosis was measured by Annexin V-FITC/PI dual staining. Western blotting, transmission electron microscopy (TEM), and staining with Cyto-ID1 Autophagy Detection dye, monodansylcadaverine (MDC) and acridine orange were performed to assess autophagy. The role of autophagy was examined by measuring cell viability and apoptosis in both cell lines pretreated with the autophagy inhibitor 3-methyladenine (3-MA). Results: Chlorophyllin f showed affinity for mitochondria and lysosomes. It exhibited significant photocytotoxicity, resulting in a maximum of 86.51% and 84.88% cell death in 5637 and T24 cells, respectively. Additionally, chlorophyllin f-mediated PDT (f-PDT) also induced a significantly higher percentage of apoptosis in treated cells compared to the control groups (P < 0.05). Moreover, the expression of Beclin1 protein and the proportion of LC3-II:LC3-I in both cell lines significantly increased after f-PDT. Autophagy, characterized by an increase in the formation of Cyto-ID1 Autophagy Detection dye-labeled autophagosomes, MDC fluorescent dye-labeled autophagic vacuoles and acridine orange-labeled acidic vesicular organelles (AVOs), was observed in f-PDT-treated cells. TEM also revealed doublemembrane autophagosome structures 1 hour after f-PDT. Most importantly, when pretreated with 3-MA, the two cell lines showed more significant photo-cytotoxicity and apoptotic cell death compared to those exposed to f-PDT alone (P < 0.05). Conclusion: Chlorophyllin f-mediated PDT exerts antitumor activity by inducing apoptosis and autophagy, and ß 2014 Wiley Periodicals, Inc.
most importantly, autophagy inhibition enhances f-PDTmediated apoptotic cell death. These results suggest that chlorophyllin f is a new, effective photosensitizer and that the combination of f-PDT with autophagy inhibitors may be an attractive therapeutic strategy against human nonmuscle invasive bladder cancer. Lasers Surg. Med. 46:319– 334, 2014. ß 2014 Wiley Periodicals, Inc. Key words: autophagy; apoptosis; bladder cancer; chlorophyllin f; lysosome; mitochondria; photodynamic therapy INTRODUCTION Bladder cancer remains a commonly diagnosed malignancy of the urinary tract. In 2012, there were 73,510 newly diagnosed cases and 14,880 deaths [1]. Non-muscle invasive bladder cancer (NMIBC) accounts for 75–80% of newly diagnosed bladder cancer [2], and its natural history is marked by a high probability of recurrence and disease progression. Clinically, NMIBC recurs in 50–80% of patients, and up to 30% of patients initially diagnosed with NMIBC will progress to a muscle invasive disease. Currently, the standard treatment for NMIBC is the intravesical instillation of Bacillus Calmette-Guerin (BCG) after the transurethral resection of the bladder tumor (TURBt) [3]. However, BCG therapy is often associated with considerable side effects, such as decreased vesical
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. Contract grant sponsor: National Nature Science Foundation of China; Contract grant number: 81272845. *Correspondence to: Chen Gang, PhD, MD, Department of Urology, Jinshan Hospital, Fudan University, 1508 Longhang Road, Shanghai 201508, China. E-mail: [email protected] **Correspondence to: Zheng Jingcun, PhD, MD, Department of Urology, Gongli Hospital, 219 Miaopu Road, Shanghai 200135, China. E-mail: [email protected] Accepted 2 January 2014 Published online 24 January 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/lsm.22225
320
LIHUAN ET AL.
compliance and even bladder contracture. Moreover, it is estimated that there are approximately 30% of NMIBC patients who either fail to respond to this therapy or suffer recurrent disease within 5 years [4]. Recently, there has been increased interest in alternative or new adjuvant treatments such as PDT, which can be effective even in BCG-refractory bladder cancer and may provide a potential alternative for patients who are indicated for cystectomy [5]. PDT is being investigated as a minimally invasive and innovative treatment modality for NMIBC. It involves the administration of a photosensitizer (PS), which when activated with light of a specific wavelength in the presence of molecular oxygen, will generate reactive oxygen species in photosensitized target tissues or cells [6]. The PS is considered to be the most important parameter for PDT. The second-generation PS such as 5-Aminolevulinicacid (5-ALA) has been central to the clinical application of PDT thus far. Recently, our research group developed a novel PS belonging to the family of chlorophyll derivatives, chlorophyllin f, which is also a secondgeneration PS. It has the following benefits: (1) it is a pure compound (purity > 97%);(2) it has a stable chemical structure (Fig. 1) similar to meso-tetra (hydroxyphenyl) porphyrin(m-THPC), which guarantees that all our experiments can be performed under visible light; (3) it is rapidly cleared from tissues or cells; (4) it is easily dissolved in aqueous solutions; (5) it has good optical properties, with an absorption peak of 653.5 nm, making it well suited for PDT; (6) it can be obtained from the traditional Chinese herb, Excrementum bombycis, and has a much lower production cost than 5-ALA, which is the semi-synthetic compound derived from natural products, with the trade name Levulan1 [7]. Taken together, the distinct advantages of chlorophyllin f implored us to assess its phototoxicity against bladder cancer cells. Autophagy is a catabolic process that involves the engulfment of proteins and entire organelles within double-membrane vesicles, the autophagosomes, followed by fusion with lysosomes, where the cargoes are delivered for degradation and recycling [8]. Autophagy is now widely implicated in pathophysiological processes, for example, cancer, metabolic and neuro-degenerative disorders, cardiovascular and pulmonary diseases, and in physiological
Fig. 1. Chemical structure of chlorophyllin f sodium.
responses to exercise and aging [9]. Interestingly, autophagy is also involved in cell death following photodynamic therapy. David Kessel’s group was the first to demonstrate that PDT induces only autophagy to kill DU145 cells, whereas both apoptosis and autophagy occurred in L1210 cells after 9-capronyloxytetrakis porphycene (CPO)-mediated photodamage [10]. However, the exact role of autophagy in PDT is still disputed. On one hand, autophagy might play a prosurvival role in PDT, as it was reported that the shRNA knockdown of autophagy-related protein 7 (ATG7) enhanced the cytotoxicity of photosensitizers in murine hepatoma 1c1c7 cells [11]. On the other hand, autophagy may serve as a cell death mechanism after PDT because blocking autophagy through the knockdown of ATG7 increased the survival of MCF-7 cells exposed to PDT [12]. Collectively, these studies suggest that PDT very likely induces autophagy and apoptosis and that autophagy might play different roles in different cell types and with different photosensitizers. In this study, we first aimed to present the effects of chlorophyllin f-mediated PDT on human bladder 5637 and T24 cells. We further investigated the mechanisms of f-PDT for the two cell lines, asking the following questions: can f-PDT modulate apoptosis and autophagy in 5637 and T24 cells? What is the relationship between autophagy and apoptotic cell death by f-PDT for the two cell lines? Does the induction of autophagy have a pro-survival or a pro-death role in this process? MATERIALS AND METHODS Preparation of Chlorophyllin f Chlorophyllin f was prepared according to our patent specifications (No. CN 200510024984.8). We first extracted 200 g crude chlorophyll from E. bombycis, and dissolved the crude product with 800 ml 80% acetone, and then added 30 ml of 20% NaOH into the solution to hydrolyze the samples for 40 minutes. Next, we distilled the liquid to remove acetone, cooled the mother liquor to room temperature and adjusted the pH to 10 with 10% HCl. Then, 300 ml gasoline was added and shaken for 5 minutes to remove impurities. This wash was repeated three times. The aqueous layer was transferred to a glass beaker, and 10% HCl was added to adjust the pH to 3.0. Then, the amorphous e6 crystals were precipitated. We obtained the crystals by filtration and dried them at 808C. The solid was dissolved in 80% acetone and then subjected to chromatography in a silica gel column. The pure e6 was eluted with a 50:25:1 mixture of acetone: methanol: formic acid (v/v/v). Next, we dissolved 10 g e6 in 250 ml pyridine, and 30 ml 25% KOH solution was added. After 1 hour, the solution was refluxed under nitrogen gas for 1 hour. After removing the pyridine by distillation, we added 750 ml distilled water to the mother liquor, adjusted the pH to 3.0 using 10% HCl, and the f crystal was precipitated. The crystal was dried, dissolved and purified by column chromatography with the same method used for e6. Chlorophyllin f was
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
dissolved with RPMI-1640 medium (without 10% FBS) and stored at 48C. Cancer Cells and Cell Culture Human bladder cancer cell lines 5637 and T24 were purchased from the Shanghai Institutes of Biological Science and grown in RPMI-1640 medium supplemented with 10% FBS plus 1% streptomycin and gentamycin (Thermo Scientific, Peking, China) in 5% CO2 at 378C in plastic culture dishes. 1
O2 Assays
We used 1,3-diphenylisobenzofuran (DPBF), a sensitive probe of reactive oxygen species (ROS), to detect the 1O2 produced in the system. The decrement of its fluorescence, which reflects the relative yield of 1O2, was measured using a spectrometer (Hitachi, Chiyoda, Tokyo, Japan, F-2500). In the photochemical experiment, 5 mM DPBF was first mixed with 2 mM chlorophyllin f and then irradiated by a 40 mW filtered red laser from a projector for different times of 15, 30, 45, 60, and 75 seconds. The DPBF photodegradation of chlorophyllin f in aqueous solutions was evaluated as the relative capacity for 1O2 production. Subcellular Location of Chlorophyllin f Cells were cultured on coverslips placed in 35 mm dishes at a density of 1 105 cells per dish. When they reached 80% confluence, 5637 and T24 cells were co-cultured with chlorophyllin f, which was dissolved with RPMI-1640 medium (without 10% FBS), for 4 hours and incubated for 30 minutes with 100 or 150 nM MitoTracker1 Green probe (Invitrogen, Eugene, Oregon), respectively. After incubation, the staining solution was replaced by fresh culture medium (without 10% FBS), and cells were observed by CLSM (Olympus FV300). The excitation wavelength for chlorophyllin f was 400 nm. The fluorescence excitation and emission wavelengths of the lysosome probe were 488 and 516 nm, respectively. For the LysoTracker1 Green probe (Invitrogen), the procedure was similar to that used for the MitoTracker1 Green probe. The two cell lines were incubated with chlorophyllin f for 2 hours and then for 30 minutes with 50 nM (T24) or 75 nM (5637) LysoTracker1 Green probe. The fluorescence excitation and emission wavelengths of the lysosome probe were 504 and 511 nm, respectively. Photosensitization When 5637 and T24 cells reached approximately 80% confluence, they were incubated with different concentrations of chlorophyllin f for 2 hours at 378C, and the PS solution was then replaced by fresh complete culture medium (with 10% FBS). The cells were then irradiated with 650 nm laser light at an irradiation power of 40 mW/ cm2. The time of irradiation was adjusted to yield the desired loss of cell viability as determined by Cell Counting Kit-8 assays. After irradiation, cells were incubated at 378C as specified.
321
Cell Counting Kit-8 Assay The assay was performed according to the manufacturer’s instructions (Dojindo Laboratories, Kumamoto, Japan, EQ829). Briefly, 5637 and T24 cells were seeded in 96-well plates at a density of 1.0 104 cells in 100 ml medium per well and cultured overnight. After removing the culture medium, cells were incubated with different concentrations of chlorophyllin f for 2 hours. 5637 cells were treated with 2.5, 5, and 10 mg/ml chlorophyllin f, and T24 cells were treated at concentrations of 1, 2, and 4 mg/ml. After incubation, the PS solution was replaced by complete culture media, and the cells were irradiated with a 650-nm 40 mW/cm2 laser light for different times (50 and 100 seconds). Three control groups, including blank control, cells treated with PS without laser light, and cells exposed to laser light without PS, were established in each cell line. The photo-damaged cells were then incubated at 378C in 5% CO2. After 12 hours, 10 ml staining solution was added into each well and incubated with the cells for an additional 2 hours. Finally, the absorbance (optical density, OD) was measured at 450 nm with a micro-plate reader (Infinite M200 pro, TECAN). Each group included three replicates. Measurement of Cell Apoptosis The detection of apoptotic cells after f-PDT was performed by flow cytometry using the AnnexinV-FITC Apoptosis Detection kit (Invitrogen) in accordance with the manufacturer’s instructions. Both 5637 and T24 cells were cultured overnight in six-well plates at a density of 2 105 cells per well. At 12 hours after PDT treatment, cells were collected by 0.025% trypsin (Thermo Scientific) and washed twice with ice-cold PBS. After re-suspension with 500 ml binding buffer, the treated cells were stained with 5 ml AnnexinV-FITC and 5 ml PI at 378C for 15 minutes in the dark. Finally, the fluorescence of each group was immediately determined with a CyAn ADP analyzer (Beckman Coulter, Beckman). Western Blot Analysis The PDT-treated cells seeded in 60-mm dishes at a density of 3 105 cells per dish were first washed with PBS and then harvested in lysis buffer containing protease and phosphatase inhibitors at the indicated times (0, 0.5, 1, 2, and 4 hours after PDT). Samples containing equal amounts of protein (40 mg) were resolved on 12% SDS-polyacrylamide gels, transferred onto PVDF membranes (Millipore, Billerica, MA), and probed sequentially with primary antibodies against Beclin1 (Cell Signaling, 3738), LC3 (Sigma, L7543), and GAPDH (Kangchen, KC-5G5) overnight at 48C. The membranes were then incubated with secondary antibody at room temperature for 1 hour. Finally, immunoreactive bands were visualized using a chemiluminescence (ECL) kit (CW Bio). Cyto-IDW Autophagy Detection Dye Staining This staining was performed in accordance with the manufacturer’s instructions. Briefly, both 5637 and T24 cells were seeded into 24-well plates at a density of
322
LIHUAN ET AL.
5 10 cells per well and cultured overnight. At 1 hour after chlorophyllin f-mediated photosensitization, the treated cells were stained with Cyto-ID1 Autophagy Detection dye (ENZO) plus Hoechst 33342 nuclear dye (ENZO) at 378C for 30 minutes in the dark. Stained cells were then photographed with a CLSM (Leica, TCS-SP8). The excitation and emission wavelengths for Cyto-ID1 Autophagy Detection dye were 463 and 534 nm, respectively. The wavelengths for Hoechst 33342 nuclear dye were 350 and 461 nm, respectively. The bladder cancer cells with no exposure to f-PDT were used as negative controls. 4
MDC Staining Autophagic vacuoles were stained with the autofluorescent agent MDC (Sigma). Both 5637 and T24 cells were cultured overnight in 24-well plates at a density of 5 104 cells per well. At 2 hours post-irradiation, both cell lines were incubated with 50 mM MDC in RPMI-1640 medium supplemented with 10% FBS at 378C for 15 min. After washing, the intracellular MDC fluorescence was immediately observed (excitation/emission wavelengths: 335/498 nm) with a CLSM (Leica, TCS-SP8). The negative controls were as described above in Cyto-ID1 Autophagy Detection dye staining. Acridine Orange Staining Briefly, at 4 hours after irradiation, chlorophyllin f-PDTtreated cells, which were seeded in 24-well plates at a density of 5 104 cells per well and cultured overnight before PDT, were stained with 1 mM acridine orange (Sigma) in RPMI-1640 medium containing 10% FBS at 378C for 15 min. Green (510–536 nm) and red (650 nm) fluorescence emission from stained cells illuminated with blue (488 nm) excitation light was measured by a CLSM (Leica, TCS-SP8). The negative controls were also set as described above. Transmission Electron Microscopy
IL). Values are expressed as the means standard error of the mean (SEM). The unpaired two-tailed t test was used for the comparison of differences between different groups, which were considered to be significant if P < 0.05. All experiments were repeated at least three times. RESULTS Spectral Analysis of Chlorophyllin f We performed a spectral analysis of chlorophyllin f. The absorbance spectrum of photosensitizer chlorophyllin f showed that it has a strong absorption peak at wavelengths 653.5, 599, 501.5, and 398.5 nm (Fig. 2). Because the 398.5 nm absorption peak of chlorophyllin f was the strongest, we chose 400 nm as the excitation laser channel for CLSM. Because the penetration depth of laser light is proportional to its wavelength, we chose 650 nm for the irradiation wavelength in the chlorophyllin f-mediated photosensitization. 1
O2 Yield of Chlorophyllin f
When DPBF and chlorophyllin f are present in the aqueous solution, the PS molecules could produce singlet oxygen (1O2) upon the absorption of photons, and 1O2 then oxidizes DPBF into o-dibenzoylbenzene, which does not fluoresce [13,14]. Therefore, the decrement in the DPBF fluorescence in the system reflects the 1O2 yield of chlorophyllin f after light irradiation. 1O2 is the excited state of O2 and plays an important role in PDT. As shown in Figure 3, the fluorescence intensity of DPBF in chlorophyllin f solutions decreased with the light irradiation from 15 to 75 seconds. The degradation rate is proportional to the 1O2 yield. This result indicates that longer irradiation of chlorophyllin f solutions can produce more incident photons to increase 1O2 production. More importantly, the result indicates that chlorophyllin f is an effective photosensitizer and that it can generate 1O2 and ROS after light excitation.
Briefly, 5637 and T24 bladder cancer cells seeded in six-well plates at a density of 2 105 cells per well were exposed to chlorophyllin f-mediated PDT. Then, 1 hour after PDT, the treated cells were harvested and fixed in 2.5% phosphatebuffered glutaraldehyde, pH 7.4, for 2 hours. Thereafter, cells were washed three times in 0.1 M phosphate buffer and fixed with 1% OsO4 buffer for 20 min. Subsequently, cells were washed with 0.1 M phosphate buffer, dehydrated through an ascending ethanol gradient series, and finally embedded in pure acetone with embedding liquid (2:1) overnight and pure embedding liquid for 3 hours. Next, samples were solidified at 378C overnight, 458C for 12 hours and 608C for 24 hours. Ultrathin (70–90 nm) sections were cut on an ultra-microtome (KB-I), stained with 3% lead citrate plus uranyl acetate, and photographed using a transmission electron microscope (Philips CM20). Statistical Analysis The statistical significance of differences between groups was assessed using SPSS 16.0 (SPSS, Inc., Chicago,
Fig. 2. Absorption spectrum of chlorophyllin f.
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
323
revealing a granular intracellular distribution. In addition, the MitoTracker1 Green probe, which specifically stains mitochondria and emits green fluorescence light when excited by the 488 nm wavelength channel, was found to be restricted to the cytoplasm in 5637 and T24 cells (Fig. 4b and f, respectively). An overlay of the chlorophyllin f and MitoTracker1 Green probe images showed that the signal distributions for the two agents were almost identical in both 5637 and T24 cells (Fig. 4d and h, respectively). We next analyzed the two cell lines co-incubated with chlorophyllin f and the LysoTracker1 Green Probe by CLSM at 400 and 488 nm, respectively. As can be clearly observed in Figure 5, after a 2-hour incubation in 5637 and T24 cells, the red signal distribution of chlorophyllin f was almost identical to the green fluorescence of the LysoTracker1 Green Probe (Fig. 5a–d and e–h, respectively). Fig. 3. Fluorescence intensity decreasing of diphenylisobenzofuran (DPBF) in chlorophyllin f (2 mM) aqueous solutions with the irradiation of red light (irradiation power: 40 mW/cm2).
Chlorophyllin f Is Located in Both Mitochondria and Lysosomes As shown in Figure 4c and g, after 4 hours incubation with 5637 and T24 cells, chlorophyllin f excited by the 400nm wavelength channel emitted red fluorescent light
Cell Viability of 5637 and T24 Cells After f-PDT The photodynamic effects of chlorophyllin f in 5637 and T24 cells were determined using the Cell Counting Kit-8 assay. The growth inhibition rate was calculated as follows: 100% (OD value of the control group OD value of treatment group)/OD value of the control group. Figure 6a and b show the growth inhibition rates in 5637 and T24 cells after chlorophyllin f-mediated PDT.
Fig. 4. Intracellular fluorescence with confocal laser scanning microscopy (CLSM) in 5637(a–d) and T24 cells (e–h). The bright field micrograph shows the distribution of 5637 and T24 cells in the dishes (a, e), while the green fluorescence micrograph illustrates the localizations of MitoTracker1 Green probes, which are identical to those of the mitochondria. Arrowheads indicate punctate fluorescence in the mitochondria of the two cell lines (b, f), the auto-fluorescence of chlorophyllin f (c, g) and the overlapped fluorescence of the mitochondria tracker and chlorophyllin f (d, h) (magnification, 1,200)
324
LIHUAN ET AL.
Fig. 5. Intracellular distribution of chlorophyllin f with CLSM in 5637(a–d) and T24 cells (e–h). The bright field micrograph shows the distribution of 5637 and T24 cells in the dishes (a, e), while the arrowheads indicate the green punctate fluorescence representing the locations of LysoTracker1 Green probes which are identical to those of the lysosomes in 5637 and T24 cells (b, f), the auto-fluorescence of chlorophyllin f (c, g) and the overlapped fluorescence of the lysosome tracker and chlorophyllin f (d, h) (magnification, 1,200.
In 5637 cells, a statistically significant difference of OD values detected at a 450-nm wavelength was observed in all treatment groups compared to the blank control group (P < 0.05). The 4 mg/ml chlorophyllin f treatment group with a light dose of 4 J/cm2 experienced much lower OD values than either the 2 or 1 mg/ml treatment groups (P < 0.05); it was also significantly different from
the 4 mg/ml chlorophyllin f treatment group with a light dose of 2 J/cm2. However, both the laser alone and the chlorophyllin f alone groups did not show any difference from the blank control group (P > 0.05). Similar results were observed in the different groups with T24 cells. In 5637 cells, treatment with 1 mg/ml chlorophyllin f-mediated PDT resulted in only 25.4 and 33.92% cell death
Fig. 6. Phototoxic effects of different concentrations of chlorophyllin f, which were exposed to different doses of laser light, on 5637 and T24 cells. Cell growth inhibition of different treatments in 5637 and T24 cells (a, b).
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
325
Fig. 7. Effects of 3-MA, f-PDT on apoptotic cell death in 5637(a–c) and T24 (d–f) cells. R2: necrosis rate; R3: late apoptosis rate; R4: normal cell rate; and R5: early apoptosis rate. (a) 5637 cells without f-PDT and 3-MA; (b) 5637 cells with f-PDT; (c) 5637 cells with f-PDT and 3-MA; (d) T24 cells without f-PDT and 3-MA; (e) T24 cells with f-PDT; (f) T24 cells with f-PDT and 3-MA. P < 0.01, P < 0.05.
at the light doses of 2 and 4 J/cm2, respectively (Fig. 6a). However, a concentration of 4 mg/ml of chlorophyllin f appeared to be an effective dose, which induced a remarkable 70.09 and 84.88% cell death with light doses of 2 and 4 J/cm2, respectively. A 2 mg/ml chlorophyllin f-mediated PDT resulted in 51.21 and 57.38% cell death with light doses of 2 and 4 J/cm2, respectively. The situation was similar in T24 cells (Fig. 6b). Treatment with 2.5 mg/ml of chlorophyllin f and PDT induced only 31.79% and 34.03% cell death at light doses of 2 and 4 J/cm2, respectively, whereas a concentration of 10 mg/ml chlorophyllin f-mediated PDT induced 72.89 and 86.51% cell death, respectively. In the two cell lines, both the laser alone groups and chlorophyllin f without light groups did not show any cell toxicity. Chlorophyllin f-Mediated PDT Induces Apoptotic Cell Death Chlorophyllin f-mediated PDT remarkably induced 50.61 1.66% and 54.3 1.32% cell apoptosis in 5637 and T24 cell lines, respectively. The apoptosis rates in the PDT groups of the two cell types were significantly higher than those in the corresponding control groups (P < 0.01),
in which the rates were only 4.05 0.12% and 11.31 0.71% for 5637 and T24 cells, respectively (Fig. 7). Chlorophyllin f-Mediated PDT Induces Autophagy We first examined the expression levels of two autophagy-related proteins, Beclin1 and LC3, in f-PDT-treated cells to determine whether f-PDT induced autophagy and when autophagy occurred after f-PDT. Beclin1, a Bcl-2 homology3 (BH3) domain-only protein, is a key regulator of autophagosome formation and autolysosome fusion [15]. Figure 8 clearly shows that both f-PDT-treated cell types have a notable increase in Beclin1 expression levels compared with blank control cells, beginning at 1 hour after PDT and lasting for 3 hours. We next tracked the conversion of LC3-I to LC3-II, which indicates autophagic activity [16], by Western blotting. When autophagy is induced, LC3-I is conjugated to phosphatidylethanolamine (PE) to form LC3-II, the 16kDa form of the LC3 protein that is specific for the membranes of autophagosomes. Therefore, the conversion of LC3-I to LC3-II is correlated with autophagic reflux. In Figure 8, the conversion of LC3-I to LC3-II was remarkably increased in both 5637 and T24 cells at 0.5 hour after PDT.
326
LIHUAN ET AL.
Fig. 8. Autophagy responses were measured by Western-blotting to determine the Beclin1 protein levels and the LC3 I to LC3 II conversion in 5637 and T24 cells after f-PDT.
Similar to Beclin1, a significantly higher ratio of LC3-II: LC3-I in chlorophyllin f-PDT-treated cells also lasted for 3.5 hours. We then assessed autophagy induction by the fluorescence microscopy of Cyto-ID1 Green autophagy dye, a novel fluorescent probe that is reported to co-localize with RFP-LC3 in HeLa cells [17] and has been developed to facilitate the investigation of autophagy [18–20]. It only weakly stains lysosomes but can specifically stain autophagosomes in live cells, emitting green fluorescence and forming punctate structures. Cells with such positive structures are counted as autophagy-positive cells. We also used Hoechst 33342 nuclear dye to stain the nucleus. Figure 9 clearly shows that 1 hour after f-PDT, the treated cells had obvious increases in green punctate structures distributed in perinuclear regions and focally throughout the cytoplasm in 5637 and T24 cells (Fig. 9a and c, respectively), whereas these structures were not observed in negative control cells (Fig. 9b and d). The nucleus in all the cells emitted blue fluorescence. We further determined the formation of autophagosomes by monitoring the distribution of the fluorescent dye MDC. This dye has been used to detect the presence of latestage autophagosomes and other acidic vesicles [21]. From Figure 10, we could see clearly that 2 hours after PDT there was a significant increase in the number of MDC-labeled green autophagic vacuoles, which appeared as distinct dotlike structures distributed in the cytoplasm or localized in the perinuclear region in both f-PDT-treated 5637 and T24
cells compared with the corresponding untreated cells. The control cells showed a primarily uniform diffuse distribution of MDC. Autophagy can also be detected by the intracellular detection of acidic cellular organelles. Therefore, we used the lysosomotropic agent acridine orange, which emits bright red fluoresces in acidic vesicles, indicating the formation of autolysosomes (the late stage of autophagy), and emits green fluorescence in the cytoplasm and nucleus. From Figure 11, we could see clearly that both 5637 and T24 cells showed a significant increase of red fluorescence from AVOs, which appeared as red dots in the cytoplasm at 4 hours after f-PDT treatment (Fig. 11a and c, respectively), while there were no such structures in the control cells but only weak red fluorescence that originated from late endosomes with interior acidic pH (Fig. 11b and d). All cells exhibited the green fluorescence of the acridine orange dye in the cytoplasm and nucleus. Furthermore, the f-PDT-treated cells were morphologically examined by TEM, which is considered the “gold standard” for autophagy [22,23]. Figures 12b and d show that the control cells had normal morphologies of their cytoplasm and organelles. However, a high level of double membrane-bound vacuoles marked by the arrowheads was observed inside 5637 and T24 cells 1 hour after f-PDT (Fig. 12a and c, respectively). The vacuoles were found with engulfed bulk cytoplasm and cytoplasmic organelles and served as a typical determination of autophagosomes. These findings cannot be observed in control cells. Inhibition of Autophagy Enhances f-PDT-Induced Cell Death To determine whether f-PDT-induced autophagy contributes to cell survival or death, a commonly used autophagy inhibitor, 3-MA (5 mM), which can specifically inhibit the two important factors in autophagy-related pathways-class I and class III PtdIns 3Ks, was preadministered for 1 hour in 5637 and T24 cells prior to f-PDT. The Cell Counting Kit-8 assay was performed to detect the effect of 3-MA on cell survival in both 5637 and T24 cells after f-PDT. In 5637 cells, a statistically significant difference was observed in the 3-MA þ f-PDT group compared to the f-PDT group (P < 0.05); both treatment groups experienced much lower OD values than 3-MA groups (P < 0.05). Similar results were observed in the different groups of T24 cells. Figure 13a and b show the cell inhibition for each group of 5637 and T24 cells. In 5637 cells, the cell death rates were 93.52%, 83.15%, and 6.73% for the 3-MA þ f-PDT group, f-PDT group and 3-MA group, respectively (Fig. 13a). In T24 cells, these rates were 95.51%, 86.19%, and 5.78%, respectively (Fig. 13b). No cell death was observed in either control group. To validate the effect of autophagy inhibition on cell apoptosis after f-PDT, AnnexinV-FITC/PI dual staining was performed. As shown in Figure 7, flow cytometric analysis clearly demonstrated that the proportions of
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
327
Fig. 9. Autophagy was assessed by the appearance of punctate vesicle structures in both 5637 and T24 cells via Cyto-ID fluorescence staining. Cells with punctate structures, which represent autophagosomes, were counted as autophagy-positive cells. The nucleus in all the cells emitted blue fluorescence. The arrowheads denote green punctate structures that were distributed in the peri-nuclear region and throughout the cytoplasm in 5637 (a) and T24 cells (c). a: 5637 cells with f-PDT; (b) 5637 cells without f-PDT; (c) T24 cells after f-PDT; (d) T24 cells without f-PDT. Scale bar, 10 mm. P < 0.05.
AV-positive cells and AV/PI-positive cells, which are considered apoptotic cells, significantly increased in the f-PDT plus 3-MA group compared with the f-PDT only group (P < 0.05). Specifically, f-PDT plus 3-MA remarkably induced 69.45 2.14% apoptosis, whereas the f-PDT group showed only 50.61 1.66% apoptosis in 5637 cells; the apoptotic rates in T24 cells were 77.69 1.75% and 54.3 1.32%, respectively (Fig. 7). DISCUSSION Bladder cancer is the most expensive cancer to treat, mainly due to the frequency of follow-up that is necessary for patients with NMIBC because of the high probability of recurrence and progression [24]. PDT is a minimally invasive and promising therapy for NMIBC because the bladder is accessible by endoscopy and the tumors are often limited to the mucosa or sub-mucosa [25]. In a phase I study, PDT using hexaminolevulinate (HAL) was shown to
be technically feasible and safe and may offer an alternative in the therapy of NMIBC [26]. As the most important parameter in PDT, an ideal photosensitizer should be biologically stable and minimally toxic and should selectively accumulate in target tissue, all while, most importantly, maintaining photochemical efficiency [27]. Thus, significant effort is now being directed toward the synthesis of novel photosensitizers that fulfill the necessary chemical, physical and biological requirements of an ideal photosensitizer. We recently developed a new photosensitizer, chlorophyllin f, which was extracted from the traditional Chinese herb E. bombycis and would meet all the requirements of an ideal photosensitizer for PDT as its patent specifications have stated. Therefore, it may be a novel photosensitizer, and more suitable than 5-ALA for developing countries such as China, as it can be obtained from abundant crude materials and costs less to prepare than 5-ALA [7].
328
LIHUAN ET AL.
Fig. 10. Autophagy responses were measured by the appearance of autophagic vacuoles, using monodansylcadaverine (MDC) staining. The thick arrowheads indicate the punctate fluorescence in the autophagic vacuoles, distributing within the cytoplasm or in the perinuclear region in 5637 (a) and T24 cells (c). And all the control cells show mainly a uniformly diffuse distribution of MDC. a: 5637 cells after f-PDT; (b) 5637 cells without f-PDT; (c) T24 cells with f-PDT; (d) T24 cells without f-PDT. Scale bar, 10 mm.
In this study, we performed a spectral analysis of this new photosensitizer and found it had an absorption peak of 653.5 nm, which guarantees good penetration depth for excitation laser light. Therefore, chlorophyllin f-mediated PDT could penetrate the mucosa or sub-mucosa of the bladder tissue, making it suitable for the therapy of NMIBC. Furthermore, we also monitored the 1O2 yield of our novel photosensitizer after light irradiation because, in most cases, the key ROS of PDT is 1O2, which is the lowest excited electronic state of oxygen. PDT exerts its cytotoxicity mainly through the generation of ROS. Interestingly, we found that chlorophyllin f had a high potential to produce 1O2 upon light excitation, which further showed that it would be an effective photosensitizer. Next, we performed a cell viability assay to evaluate the cytotoxic effect of chlorophyllin f-mediated PDT in 5637 and T24 cells. We found that the effective dose of chlorophyllin f for 5637 cells is a concentration of 4 mg/ml and that of the irradiation laser is 4 J/cm2 (40 mW/cm2 for
100 seconds). Under these circumstances, f-PDT resulted in a remarkable 84.88% cell death in 5637 cells. The most effective doses of PS and light energy for T24 cells are 4 mg/ml and 4 J/cm2, respectively, inducing 86.51% cell death. Together, these data reveal that chlorophyllin f has high photochemical efficiency in 5637 and T24 cells, and its photocytotoxicity in different cell lines is dependent on both the PS and the light energy doses during PDT. Again, these results confirm that chlorophyllin f would be an effective photosensitizer for the therapy of NMIBC. As cancer cells have a much higher accumulation of photosensitizers than normal cells, PDT is to some degree a tumor-selective treatment modality. Additionally, ROS, the cytotoxic agents generated by specific light interaction with photosensitized target tissues or cells, exist within a very limited distance (
Photodynamic Therapy With the Novel Photosensitizer Chlorophyllin f Induces Apoptosis and Autophagy in Human Bladder Cancer Cells Du Lihuan, MD,1 Zheng Jingcun, PhD, MD,2 ** Jiang Ning, MD,2 Wang Guozeng, MD,2 Chu Yiwei, PhD,3 Lin Wei, MD,3 Qian Jing, MD,3 Zhang Yuanfang, MD,4 and Chen Gang, PhD, MD1* 1 Department of Urology, Jin Shan Hospital, Fudan University, Shanghai 201508, China 2 Department of Urology, Gongli Hospital, Shanghai 200135, China 3 Department of Immunology, Fudan University, Shanghai 2000032, China 4 Department of Urology, Hua Shan Hospital, Fudan University, Shanghai 200040, China
Background and Objectives: Our group recently synthesized a new, low-cost photosensitizer, chlorophyllin f. In this study, the effects of chlorophyllin f-mediated photodynamic therapy (PDT) and its potential mechanisms were examined in human bladder cancer cells. Materials and Methods: MitoTracker1 Green probe and LysoTracker1 Green probe were used to label mitochondria and lysosomes, revealing the intracellular localization of chlorophyllin f in 5637 and T24 cells by confocal laser scanning microscopy (CLSM). The cells were treated with chlorophyllin f-mediated PDT; the photo-cytotoxicity of chlorophyllin f was monitored using the Cell Counting Kit-8 assay, and apoptosis was measured by Annexin V-FITC/PI dual staining. Western blotting, transmission electron microscopy (TEM), and staining with Cyto-ID1 Autophagy Detection dye, monodansylcadaverine (MDC) and acridine orange were performed to assess autophagy. The role of autophagy was examined by measuring cell viability and apoptosis in both cell lines pretreated with the autophagy inhibitor 3-methyladenine (3-MA). Results: Chlorophyllin f showed affinity for mitochondria and lysosomes. It exhibited significant photocytotoxicity, resulting in a maximum of 86.51% and 84.88% cell death in 5637 and T24 cells, respectively. Additionally, chlorophyllin f-mediated PDT (f-PDT) also induced a significantly higher percentage of apoptosis in treated cells compared to the control groups (P < 0.05). Moreover, the expression of Beclin1 protein and the proportion of LC3-II:LC3-I in both cell lines significantly increased after f-PDT. Autophagy, characterized by an increase in the formation of Cyto-ID1 Autophagy Detection dye-labeled autophagosomes, MDC fluorescent dye-labeled autophagic vacuoles and acridine orange-labeled acidic vesicular organelles (AVOs), was observed in f-PDT-treated cells. TEM also revealed doublemembrane autophagosome structures 1 hour after f-PDT. Most importantly, when pretreated with 3-MA, the two cell lines showed more significant photo-cytotoxicity and apoptotic cell death compared to those exposed to f-PDT alone (P < 0.05). Conclusion: Chlorophyllin f-mediated PDT exerts antitumor activity by inducing apoptosis and autophagy, and ß 2014 Wiley Periodicals, Inc.
most importantly, autophagy inhibition enhances f-PDTmediated apoptotic cell death. These results suggest that chlorophyllin f is a new, effective photosensitizer and that the combination of f-PDT with autophagy inhibitors may be an attractive therapeutic strategy against human nonmuscle invasive bladder cancer. Lasers Surg. Med. 46:319– 334, 2014. ß 2014 Wiley Periodicals, Inc. Key words: autophagy; apoptosis; bladder cancer; chlorophyllin f; lysosome; mitochondria; photodynamic therapy INTRODUCTION Bladder cancer remains a commonly diagnosed malignancy of the urinary tract. In 2012, there were 73,510 newly diagnosed cases and 14,880 deaths [1]. Non-muscle invasive bladder cancer (NMIBC) accounts for 75–80% of newly diagnosed bladder cancer [2], and its natural history is marked by a high probability of recurrence and disease progression. Clinically, NMIBC recurs in 50–80% of patients, and up to 30% of patients initially diagnosed with NMIBC will progress to a muscle invasive disease. Currently, the standard treatment for NMIBC is the intravesical instillation of Bacillus Calmette-Guerin (BCG) after the transurethral resection of the bladder tumor (TURBt) [3]. However, BCG therapy is often associated with considerable side effects, such as decreased vesical
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. Contract grant sponsor: National Nature Science Foundation of China; Contract grant number: 81272845. *Correspondence to: Chen Gang, PhD, MD, Department of Urology, Jinshan Hospital, Fudan University, 1508 Longhang Road, Shanghai 201508, China. E-mail: [email protected] **Correspondence to: Zheng Jingcun, PhD, MD, Department of Urology, Gongli Hospital, 219 Miaopu Road, Shanghai 200135, China. E-mail: [email protected] Accepted 2 January 2014 Published online 24 January 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/lsm.22225
320
LIHUAN ET AL.
compliance and even bladder contracture. Moreover, it is estimated that there are approximately 30% of NMIBC patients who either fail to respond to this therapy or suffer recurrent disease within 5 years [4]. Recently, there has been increased interest in alternative or new adjuvant treatments such as PDT, which can be effective even in BCG-refractory bladder cancer and may provide a potential alternative for patients who are indicated for cystectomy [5]. PDT is being investigated as a minimally invasive and innovative treatment modality for NMIBC. It involves the administration of a photosensitizer (PS), which when activated with light of a specific wavelength in the presence of molecular oxygen, will generate reactive oxygen species in photosensitized target tissues or cells [6]. The PS is considered to be the most important parameter for PDT. The second-generation PS such as 5-Aminolevulinicacid (5-ALA) has been central to the clinical application of PDT thus far. Recently, our research group developed a novel PS belonging to the family of chlorophyll derivatives, chlorophyllin f, which is also a secondgeneration PS. It has the following benefits: (1) it is a pure compound (purity > 97%);(2) it has a stable chemical structure (Fig. 1) similar to meso-tetra (hydroxyphenyl) porphyrin(m-THPC), which guarantees that all our experiments can be performed under visible light; (3) it is rapidly cleared from tissues or cells; (4) it is easily dissolved in aqueous solutions; (5) it has good optical properties, with an absorption peak of 653.5 nm, making it well suited for PDT; (6) it can be obtained from the traditional Chinese herb, Excrementum bombycis, and has a much lower production cost than 5-ALA, which is the semi-synthetic compound derived from natural products, with the trade name Levulan1 [7]. Taken together, the distinct advantages of chlorophyllin f implored us to assess its phototoxicity against bladder cancer cells. Autophagy is a catabolic process that involves the engulfment of proteins and entire organelles within double-membrane vesicles, the autophagosomes, followed by fusion with lysosomes, where the cargoes are delivered for degradation and recycling [8]. Autophagy is now widely implicated in pathophysiological processes, for example, cancer, metabolic and neuro-degenerative disorders, cardiovascular and pulmonary diseases, and in physiological
Fig. 1. Chemical structure of chlorophyllin f sodium.
responses to exercise and aging [9]. Interestingly, autophagy is also involved in cell death following photodynamic therapy. David Kessel’s group was the first to demonstrate that PDT induces only autophagy to kill DU145 cells, whereas both apoptosis and autophagy occurred in L1210 cells after 9-capronyloxytetrakis porphycene (CPO)-mediated photodamage [10]. However, the exact role of autophagy in PDT is still disputed. On one hand, autophagy might play a prosurvival role in PDT, as it was reported that the shRNA knockdown of autophagy-related protein 7 (ATG7) enhanced the cytotoxicity of photosensitizers in murine hepatoma 1c1c7 cells [11]. On the other hand, autophagy may serve as a cell death mechanism after PDT because blocking autophagy through the knockdown of ATG7 increased the survival of MCF-7 cells exposed to PDT [12]. Collectively, these studies suggest that PDT very likely induces autophagy and apoptosis and that autophagy might play different roles in different cell types and with different photosensitizers. In this study, we first aimed to present the effects of chlorophyllin f-mediated PDT on human bladder 5637 and T24 cells. We further investigated the mechanisms of f-PDT for the two cell lines, asking the following questions: can f-PDT modulate apoptosis and autophagy in 5637 and T24 cells? What is the relationship between autophagy and apoptotic cell death by f-PDT for the two cell lines? Does the induction of autophagy have a pro-survival or a pro-death role in this process? MATERIALS AND METHODS Preparation of Chlorophyllin f Chlorophyllin f was prepared according to our patent specifications (No. CN 200510024984.8). We first extracted 200 g crude chlorophyll from E. bombycis, and dissolved the crude product with 800 ml 80% acetone, and then added 30 ml of 20% NaOH into the solution to hydrolyze the samples for 40 minutes. Next, we distilled the liquid to remove acetone, cooled the mother liquor to room temperature and adjusted the pH to 10 with 10% HCl. Then, 300 ml gasoline was added and shaken for 5 minutes to remove impurities. This wash was repeated three times. The aqueous layer was transferred to a glass beaker, and 10% HCl was added to adjust the pH to 3.0. Then, the amorphous e6 crystals were precipitated. We obtained the crystals by filtration and dried them at 808C. The solid was dissolved in 80% acetone and then subjected to chromatography in a silica gel column. The pure e6 was eluted with a 50:25:1 mixture of acetone: methanol: formic acid (v/v/v). Next, we dissolved 10 g e6 in 250 ml pyridine, and 30 ml 25% KOH solution was added. After 1 hour, the solution was refluxed under nitrogen gas for 1 hour. After removing the pyridine by distillation, we added 750 ml distilled water to the mother liquor, adjusted the pH to 3.0 using 10% HCl, and the f crystal was precipitated. The crystal was dried, dissolved and purified by column chromatography with the same method used for e6. Chlorophyllin f was
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
dissolved with RPMI-1640 medium (without 10% FBS) and stored at 48C. Cancer Cells and Cell Culture Human bladder cancer cell lines 5637 and T24 were purchased from the Shanghai Institutes of Biological Science and grown in RPMI-1640 medium supplemented with 10% FBS plus 1% streptomycin and gentamycin (Thermo Scientific, Peking, China) in 5% CO2 at 378C in plastic culture dishes. 1
O2 Assays
We used 1,3-diphenylisobenzofuran (DPBF), a sensitive probe of reactive oxygen species (ROS), to detect the 1O2 produced in the system. The decrement of its fluorescence, which reflects the relative yield of 1O2, was measured using a spectrometer (Hitachi, Chiyoda, Tokyo, Japan, F-2500). In the photochemical experiment, 5 mM DPBF was first mixed with 2 mM chlorophyllin f and then irradiated by a 40 mW filtered red laser from a projector for different times of 15, 30, 45, 60, and 75 seconds. The DPBF photodegradation of chlorophyllin f in aqueous solutions was evaluated as the relative capacity for 1O2 production. Subcellular Location of Chlorophyllin f Cells were cultured on coverslips placed in 35 mm dishes at a density of 1 105 cells per dish. When they reached 80% confluence, 5637 and T24 cells were co-cultured with chlorophyllin f, which was dissolved with RPMI-1640 medium (without 10% FBS), for 4 hours and incubated for 30 minutes with 100 or 150 nM MitoTracker1 Green probe (Invitrogen, Eugene, Oregon), respectively. After incubation, the staining solution was replaced by fresh culture medium (without 10% FBS), and cells were observed by CLSM (Olympus FV300). The excitation wavelength for chlorophyllin f was 400 nm. The fluorescence excitation and emission wavelengths of the lysosome probe were 488 and 516 nm, respectively. For the LysoTracker1 Green probe (Invitrogen), the procedure was similar to that used for the MitoTracker1 Green probe. The two cell lines were incubated with chlorophyllin f for 2 hours and then for 30 minutes with 50 nM (T24) or 75 nM (5637) LysoTracker1 Green probe. The fluorescence excitation and emission wavelengths of the lysosome probe were 504 and 511 nm, respectively. Photosensitization When 5637 and T24 cells reached approximately 80% confluence, they were incubated with different concentrations of chlorophyllin f for 2 hours at 378C, and the PS solution was then replaced by fresh complete culture medium (with 10% FBS). The cells were then irradiated with 650 nm laser light at an irradiation power of 40 mW/ cm2. The time of irradiation was adjusted to yield the desired loss of cell viability as determined by Cell Counting Kit-8 assays. After irradiation, cells were incubated at 378C as specified.
321
Cell Counting Kit-8 Assay The assay was performed according to the manufacturer’s instructions (Dojindo Laboratories, Kumamoto, Japan, EQ829). Briefly, 5637 and T24 cells were seeded in 96-well plates at a density of 1.0 104 cells in 100 ml medium per well and cultured overnight. After removing the culture medium, cells were incubated with different concentrations of chlorophyllin f for 2 hours. 5637 cells were treated with 2.5, 5, and 10 mg/ml chlorophyllin f, and T24 cells were treated at concentrations of 1, 2, and 4 mg/ml. After incubation, the PS solution was replaced by complete culture media, and the cells were irradiated with a 650-nm 40 mW/cm2 laser light for different times (50 and 100 seconds). Three control groups, including blank control, cells treated with PS without laser light, and cells exposed to laser light without PS, were established in each cell line. The photo-damaged cells were then incubated at 378C in 5% CO2. After 12 hours, 10 ml staining solution was added into each well and incubated with the cells for an additional 2 hours. Finally, the absorbance (optical density, OD) was measured at 450 nm with a micro-plate reader (Infinite M200 pro, TECAN). Each group included three replicates. Measurement of Cell Apoptosis The detection of apoptotic cells after f-PDT was performed by flow cytometry using the AnnexinV-FITC Apoptosis Detection kit (Invitrogen) in accordance with the manufacturer’s instructions. Both 5637 and T24 cells were cultured overnight in six-well plates at a density of 2 105 cells per well. At 12 hours after PDT treatment, cells were collected by 0.025% trypsin (Thermo Scientific) and washed twice with ice-cold PBS. After re-suspension with 500 ml binding buffer, the treated cells were stained with 5 ml AnnexinV-FITC and 5 ml PI at 378C for 15 minutes in the dark. Finally, the fluorescence of each group was immediately determined with a CyAn ADP analyzer (Beckman Coulter, Beckman). Western Blot Analysis The PDT-treated cells seeded in 60-mm dishes at a density of 3 105 cells per dish were first washed with PBS and then harvested in lysis buffer containing protease and phosphatase inhibitors at the indicated times (0, 0.5, 1, 2, and 4 hours after PDT). Samples containing equal amounts of protein (40 mg) were resolved on 12% SDS-polyacrylamide gels, transferred onto PVDF membranes (Millipore, Billerica, MA), and probed sequentially with primary antibodies against Beclin1 (Cell Signaling, 3738), LC3 (Sigma, L7543), and GAPDH (Kangchen, KC-5G5) overnight at 48C. The membranes were then incubated with secondary antibody at room temperature for 1 hour. Finally, immunoreactive bands were visualized using a chemiluminescence (ECL) kit (CW Bio). Cyto-IDW Autophagy Detection Dye Staining This staining was performed in accordance with the manufacturer’s instructions. Briefly, both 5637 and T24 cells were seeded into 24-well plates at a density of
322
LIHUAN ET AL.
5 10 cells per well and cultured overnight. At 1 hour after chlorophyllin f-mediated photosensitization, the treated cells were stained with Cyto-ID1 Autophagy Detection dye (ENZO) plus Hoechst 33342 nuclear dye (ENZO) at 378C for 30 minutes in the dark. Stained cells were then photographed with a CLSM (Leica, TCS-SP8). The excitation and emission wavelengths for Cyto-ID1 Autophagy Detection dye were 463 and 534 nm, respectively. The wavelengths for Hoechst 33342 nuclear dye were 350 and 461 nm, respectively. The bladder cancer cells with no exposure to f-PDT were used as negative controls. 4
MDC Staining Autophagic vacuoles were stained with the autofluorescent agent MDC (Sigma). Both 5637 and T24 cells were cultured overnight in 24-well plates at a density of 5 104 cells per well. At 2 hours post-irradiation, both cell lines were incubated with 50 mM MDC in RPMI-1640 medium supplemented with 10% FBS at 378C for 15 min. After washing, the intracellular MDC fluorescence was immediately observed (excitation/emission wavelengths: 335/498 nm) with a CLSM (Leica, TCS-SP8). The negative controls were as described above in Cyto-ID1 Autophagy Detection dye staining. Acridine Orange Staining Briefly, at 4 hours after irradiation, chlorophyllin f-PDTtreated cells, which were seeded in 24-well plates at a density of 5 104 cells per well and cultured overnight before PDT, were stained with 1 mM acridine orange (Sigma) in RPMI-1640 medium containing 10% FBS at 378C for 15 min. Green (510–536 nm) and red (650 nm) fluorescence emission from stained cells illuminated with blue (488 nm) excitation light was measured by a CLSM (Leica, TCS-SP8). The negative controls were also set as described above. Transmission Electron Microscopy
IL). Values are expressed as the means standard error of the mean (SEM). The unpaired two-tailed t test was used for the comparison of differences between different groups, which were considered to be significant if P < 0.05. All experiments were repeated at least three times. RESULTS Spectral Analysis of Chlorophyllin f We performed a spectral analysis of chlorophyllin f. The absorbance spectrum of photosensitizer chlorophyllin f showed that it has a strong absorption peak at wavelengths 653.5, 599, 501.5, and 398.5 nm (Fig. 2). Because the 398.5 nm absorption peak of chlorophyllin f was the strongest, we chose 400 nm as the excitation laser channel for CLSM. Because the penetration depth of laser light is proportional to its wavelength, we chose 650 nm for the irradiation wavelength in the chlorophyllin f-mediated photosensitization. 1
O2 Yield of Chlorophyllin f
When DPBF and chlorophyllin f are present in the aqueous solution, the PS molecules could produce singlet oxygen (1O2) upon the absorption of photons, and 1O2 then oxidizes DPBF into o-dibenzoylbenzene, which does not fluoresce [13,14]. Therefore, the decrement in the DPBF fluorescence in the system reflects the 1O2 yield of chlorophyllin f after light irradiation. 1O2 is the excited state of O2 and plays an important role in PDT. As shown in Figure 3, the fluorescence intensity of DPBF in chlorophyllin f solutions decreased with the light irradiation from 15 to 75 seconds. The degradation rate is proportional to the 1O2 yield. This result indicates that longer irradiation of chlorophyllin f solutions can produce more incident photons to increase 1O2 production. More importantly, the result indicates that chlorophyllin f is an effective photosensitizer and that it can generate 1O2 and ROS after light excitation.
Briefly, 5637 and T24 bladder cancer cells seeded in six-well plates at a density of 2 105 cells per well were exposed to chlorophyllin f-mediated PDT. Then, 1 hour after PDT, the treated cells were harvested and fixed in 2.5% phosphatebuffered glutaraldehyde, pH 7.4, for 2 hours. Thereafter, cells were washed three times in 0.1 M phosphate buffer and fixed with 1% OsO4 buffer for 20 min. Subsequently, cells were washed with 0.1 M phosphate buffer, dehydrated through an ascending ethanol gradient series, and finally embedded in pure acetone with embedding liquid (2:1) overnight and pure embedding liquid for 3 hours. Next, samples were solidified at 378C overnight, 458C for 12 hours and 608C for 24 hours. Ultrathin (70–90 nm) sections were cut on an ultra-microtome (KB-I), stained with 3% lead citrate plus uranyl acetate, and photographed using a transmission electron microscope (Philips CM20). Statistical Analysis The statistical significance of differences between groups was assessed using SPSS 16.0 (SPSS, Inc., Chicago,
Fig. 2. Absorption spectrum of chlorophyllin f.
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
323
revealing a granular intracellular distribution. In addition, the MitoTracker1 Green probe, which specifically stains mitochondria and emits green fluorescence light when excited by the 488 nm wavelength channel, was found to be restricted to the cytoplasm in 5637 and T24 cells (Fig. 4b and f, respectively). An overlay of the chlorophyllin f and MitoTracker1 Green probe images showed that the signal distributions for the two agents were almost identical in both 5637 and T24 cells (Fig. 4d and h, respectively). We next analyzed the two cell lines co-incubated with chlorophyllin f and the LysoTracker1 Green Probe by CLSM at 400 and 488 nm, respectively. As can be clearly observed in Figure 5, after a 2-hour incubation in 5637 and T24 cells, the red signal distribution of chlorophyllin f was almost identical to the green fluorescence of the LysoTracker1 Green Probe (Fig. 5a–d and e–h, respectively). Fig. 3. Fluorescence intensity decreasing of diphenylisobenzofuran (DPBF) in chlorophyllin f (2 mM) aqueous solutions with the irradiation of red light (irradiation power: 40 mW/cm2).
Chlorophyllin f Is Located in Both Mitochondria and Lysosomes As shown in Figure 4c and g, after 4 hours incubation with 5637 and T24 cells, chlorophyllin f excited by the 400nm wavelength channel emitted red fluorescent light
Cell Viability of 5637 and T24 Cells After f-PDT The photodynamic effects of chlorophyllin f in 5637 and T24 cells were determined using the Cell Counting Kit-8 assay. The growth inhibition rate was calculated as follows: 100% (OD value of the control group OD value of treatment group)/OD value of the control group. Figure 6a and b show the growth inhibition rates in 5637 and T24 cells after chlorophyllin f-mediated PDT.
Fig. 4. Intracellular fluorescence with confocal laser scanning microscopy (CLSM) in 5637(a–d) and T24 cells (e–h). The bright field micrograph shows the distribution of 5637 and T24 cells in the dishes (a, e), while the green fluorescence micrograph illustrates the localizations of MitoTracker1 Green probes, which are identical to those of the mitochondria. Arrowheads indicate punctate fluorescence in the mitochondria of the two cell lines (b, f), the auto-fluorescence of chlorophyllin f (c, g) and the overlapped fluorescence of the mitochondria tracker and chlorophyllin f (d, h) (magnification, 1,200)
324
LIHUAN ET AL.
Fig. 5. Intracellular distribution of chlorophyllin f with CLSM in 5637(a–d) and T24 cells (e–h). The bright field micrograph shows the distribution of 5637 and T24 cells in the dishes (a, e), while the arrowheads indicate the green punctate fluorescence representing the locations of LysoTracker1 Green probes which are identical to those of the lysosomes in 5637 and T24 cells (b, f), the auto-fluorescence of chlorophyllin f (c, g) and the overlapped fluorescence of the lysosome tracker and chlorophyllin f (d, h) (magnification, 1,200.
In 5637 cells, a statistically significant difference of OD values detected at a 450-nm wavelength was observed in all treatment groups compared to the blank control group (P < 0.05). The 4 mg/ml chlorophyllin f treatment group with a light dose of 4 J/cm2 experienced much lower OD values than either the 2 or 1 mg/ml treatment groups (P < 0.05); it was also significantly different from
the 4 mg/ml chlorophyllin f treatment group with a light dose of 2 J/cm2. However, both the laser alone and the chlorophyllin f alone groups did not show any difference from the blank control group (P > 0.05). Similar results were observed in the different groups with T24 cells. In 5637 cells, treatment with 1 mg/ml chlorophyllin f-mediated PDT resulted in only 25.4 and 33.92% cell death
Fig. 6. Phototoxic effects of different concentrations of chlorophyllin f, which were exposed to different doses of laser light, on 5637 and T24 cells. Cell growth inhibition of different treatments in 5637 and T24 cells (a, b).
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
325
Fig. 7. Effects of 3-MA, f-PDT on apoptotic cell death in 5637(a–c) and T24 (d–f) cells. R2: necrosis rate; R3: late apoptosis rate; R4: normal cell rate; and R5: early apoptosis rate. (a) 5637 cells without f-PDT and 3-MA; (b) 5637 cells with f-PDT; (c) 5637 cells with f-PDT and 3-MA; (d) T24 cells without f-PDT and 3-MA; (e) T24 cells with f-PDT; (f) T24 cells with f-PDT and 3-MA. P < 0.01, P < 0.05.
at the light doses of 2 and 4 J/cm2, respectively (Fig. 6a). However, a concentration of 4 mg/ml of chlorophyllin f appeared to be an effective dose, which induced a remarkable 70.09 and 84.88% cell death with light doses of 2 and 4 J/cm2, respectively. A 2 mg/ml chlorophyllin f-mediated PDT resulted in 51.21 and 57.38% cell death with light doses of 2 and 4 J/cm2, respectively. The situation was similar in T24 cells (Fig. 6b). Treatment with 2.5 mg/ml of chlorophyllin f and PDT induced only 31.79% and 34.03% cell death at light doses of 2 and 4 J/cm2, respectively, whereas a concentration of 10 mg/ml chlorophyllin f-mediated PDT induced 72.89 and 86.51% cell death, respectively. In the two cell lines, both the laser alone groups and chlorophyllin f without light groups did not show any cell toxicity. Chlorophyllin f-Mediated PDT Induces Apoptotic Cell Death Chlorophyllin f-mediated PDT remarkably induced 50.61 1.66% and 54.3 1.32% cell apoptosis in 5637 and T24 cell lines, respectively. The apoptosis rates in the PDT groups of the two cell types were significantly higher than those in the corresponding control groups (P < 0.01),
in which the rates were only 4.05 0.12% and 11.31 0.71% for 5637 and T24 cells, respectively (Fig. 7). Chlorophyllin f-Mediated PDT Induces Autophagy We first examined the expression levels of two autophagy-related proteins, Beclin1 and LC3, in f-PDT-treated cells to determine whether f-PDT induced autophagy and when autophagy occurred after f-PDT. Beclin1, a Bcl-2 homology3 (BH3) domain-only protein, is a key regulator of autophagosome formation and autolysosome fusion [15]. Figure 8 clearly shows that both f-PDT-treated cell types have a notable increase in Beclin1 expression levels compared with blank control cells, beginning at 1 hour after PDT and lasting for 3 hours. We next tracked the conversion of LC3-I to LC3-II, which indicates autophagic activity [16], by Western blotting. When autophagy is induced, LC3-I is conjugated to phosphatidylethanolamine (PE) to form LC3-II, the 16kDa form of the LC3 protein that is specific for the membranes of autophagosomes. Therefore, the conversion of LC3-I to LC3-II is correlated with autophagic reflux. In Figure 8, the conversion of LC3-I to LC3-II was remarkably increased in both 5637 and T24 cells at 0.5 hour after PDT.
326
LIHUAN ET AL.
Fig. 8. Autophagy responses were measured by Western-blotting to determine the Beclin1 protein levels and the LC3 I to LC3 II conversion in 5637 and T24 cells after f-PDT.
Similar to Beclin1, a significantly higher ratio of LC3-II: LC3-I in chlorophyllin f-PDT-treated cells also lasted for 3.5 hours. We then assessed autophagy induction by the fluorescence microscopy of Cyto-ID1 Green autophagy dye, a novel fluorescent probe that is reported to co-localize with RFP-LC3 in HeLa cells [17] and has been developed to facilitate the investigation of autophagy [18–20]. It only weakly stains lysosomes but can specifically stain autophagosomes in live cells, emitting green fluorescence and forming punctate structures. Cells with such positive structures are counted as autophagy-positive cells. We also used Hoechst 33342 nuclear dye to stain the nucleus. Figure 9 clearly shows that 1 hour after f-PDT, the treated cells had obvious increases in green punctate structures distributed in perinuclear regions and focally throughout the cytoplasm in 5637 and T24 cells (Fig. 9a and c, respectively), whereas these structures were not observed in negative control cells (Fig. 9b and d). The nucleus in all the cells emitted blue fluorescence. We further determined the formation of autophagosomes by monitoring the distribution of the fluorescent dye MDC. This dye has been used to detect the presence of latestage autophagosomes and other acidic vesicles [21]. From Figure 10, we could see clearly that 2 hours after PDT there was a significant increase in the number of MDC-labeled green autophagic vacuoles, which appeared as distinct dotlike structures distributed in the cytoplasm or localized in the perinuclear region in both f-PDT-treated 5637 and T24
cells compared with the corresponding untreated cells. The control cells showed a primarily uniform diffuse distribution of MDC. Autophagy can also be detected by the intracellular detection of acidic cellular organelles. Therefore, we used the lysosomotropic agent acridine orange, which emits bright red fluoresces in acidic vesicles, indicating the formation of autolysosomes (the late stage of autophagy), and emits green fluorescence in the cytoplasm and nucleus. From Figure 11, we could see clearly that both 5637 and T24 cells showed a significant increase of red fluorescence from AVOs, which appeared as red dots in the cytoplasm at 4 hours after f-PDT treatment (Fig. 11a and c, respectively), while there were no such structures in the control cells but only weak red fluorescence that originated from late endosomes with interior acidic pH (Fig. 11b and d). All cells exhibited the green fluorescence of the acridine orange dye in the cytoplasm and nucleus. Furthermore, the f-PDT-treated cells were morphologically examined by TEM, which is considered the “gold standard” for autophagy [22,23]. Figures 12b and d show that the control cells had normal morphologies of their cytoplasm and organelles. However, a high level of double membrane-bound vacuoles marked by the arrowheads was observed inside 5637 and T24 cells 1 hour after f-PDT (Fig. 12a and c, respectively). The vacuoles were found with engulfed bulk cytoplasm and cytoplasmic organelles and served as a typical determination of autophagosomes. These findings cannot be observed in control cells. Inhibition of Autophagy Enhances f-PDT-Induced Cell Death To determine whether f-PDT-induced autophagy contributes to cell survival or death, a commonly used autophagy inhibitor, 3-MA (5 mM), which can specifically inhibit the two important factors in autophagy-related pathways-class I and class III PtdIns 3Ks, was preadministered for 1 hour in 5637 and T24 cells prior to f-PDT. The Cell Counting Kit-8 assay was performed to detect the effect of 3-MA on cell survival in both 5637 and T24 cells after f-PDT. In 5637 cells, a statistically significant difference was observed in the 3-MA þ f-PDT group compared to the f-PDT group (P < 0.05); both treatment groups experienced much lower OD values than 3-MA groups (P < 0.05). Similar results were observed in the different groups of T24 cells. Figure 13a and b show the cell inhibition for each group of 5637 and T24 cells. In 5637 cells, the cell death rates were 93.52%, 83.15%, and 6.73% for the 3-MA þ f-PDT group, f-PDT group and 3-MA group, respectively (Fig. 13a). In T24 cells, these rates were 95.51%, 86.19%, and 5.78%, respectively (Fig. 13b). No cell death was observed in either control group. To validate the effect of autophagy inhibition on cell apoptosis after f-PDT, AnnexinV-FITC/PI dual staining was performed. As shown in Figure 7, flow cytometric analysis clearly demonstrated that the proportions of
PHOTODYNAMIC THERAPY WITH THE CHLOROPHYLLIN F
327
Fig. 9. Autophagy was assessed by the appearance of punctate vesicle structures in both 5637 and T24 cells via Cyto-ID fluorescence staining. Cells with punctate structures, which represent autophagosomes, were counted as autophagy-positive cells. The nucleus in all the cells emitted blue fluorescence. The arrowheads denote green punctate structures that were distributed in the peri-nuclear region and throughout the cytoplasm in 5637 (a) and T24 cells (c). a: 5637 cells with f-PDT; (b) 5637 cells without f-PDT; (c) T24 cells after f-PDT; (d) T24 cells without f-PDT. Scale bar, 10 mm. P < 0.05.
AV-positive cells and AV/PI-positive cells, which are considered apoptotic cells, significantly increased in the f-PDT plus 3-MA group compared with the f-PDT only group (P < 0.05). Specifically, f-PDT plus 3-MA remarkably induced 69.45 2.14% apoptosis, whereas the f-PDT group showed only 50.61 1.66% apoptosis in 5637 cells; the apoptotic rates in T24 cells were 77.69 1.75% and 54.3 1.32%, respectively (Fig. 7). DISCUSSION Bladder cancer is the most expensive cancer to treat, mainly due to the frequency of follow-up that is necessary for patients with NMIBC because of the high probability of recurrence and progression [24]. PDT is a minimally invasive and promising therapy for NMIBC because the bladder is accessible by endoscopy and the tumors are often limited to the mucosa or sub-mucosa [25]. In a phase I study, PDT using hexaminolevulinate (HAL) was shown to
be technically feasible and safe and may offer an alternative in the therapy of NMIBC [26]. As the most important parameter in PDT, an ideal photosensitizer should be biologically stable and minimally toxic and should selectively accumulate in target tissue, all while, most importantly, maintaining photochemical efficiency [27]. Thus, significant effort is now being directed toward the synthesis of novel photosensitizers that fulfill the necessary chemical, physical and biological requirements of an ideal photosensitizer. We recently developed a new photosensitizer, chlorophyllin f, which was extracted from the traditional Chinese herb E. bombycis and would meet all the requirements of an ideal photosensitizer for PDT as its patent specifications have stated. Therefore, it may be a novel photosensitizer, and more suitable than 5-ALA for developing countries such as China, as it can be obtained from abundant crude materials and costs less to prepare than 5-ALA [7].
328
LIHUAN ET AL.
Fig. 10. Autophagy responses were measured by the appearance of autophagic vacuoles, using monodansylcadaverine (MDC) staining. The thick arrowheads indicate the punctate fluorescence in the autophagic vacuoles, distributing within the cytoplasm or in the perinuclear region in 5637 (a) and T24 cells (c). And all the control cells show mainly a uniformly diffuse distribution of MDC. a: 5637 cells after f-PDT; (b) 5637 cells without f-PDT; (c) T24 cells with f-PDT; (d) T24 cells without f-PDT. Scale bar, 10 mm.
In this study, we performed a spectral analysis of this new photosensitizer and found it had an absorption peak of 653.5 nm, which guarantees good penetration depth for excitation laser light. Therefore, chlorophyllin f-mediated PDT could penetrate the mucosa or sub-mucosa of the bladder tissue, making it suitable for the therapy of NMIBC. Furthermore, we also monitored the 1O2 yield of our novel photosensitizer after light irradiation because, in most cases, the key ROS of PDT is 1O2, which is the lowest excited electronic state of oxygen. PDT exerts its cytotoxicity mainly through the generation of ROS. Interestingly, we found that chlorophyllin f had a high potential to produce 1O2 upon light excitation, which further showed that it would be an effective photosensitizer. Next, we performed a cell viability assay to evaluate the cytotoxic effect of chlorophyllin f-mediated PDT in 5637 and T24 cells. We found that the effective dose of chlorophyllin f for 5637 cells is a concentration of 4 mg/ml and that of the irradiation laser is 4 J/cm2 (40 mW/cm2 for
100 seconds). Under these circumstances, f-PDT resulted in a remarkable 84.88% cell death in 5637 cells. The most effective doses of PS and light energy for T24 cells are 4 mg/ml and 4 J/cm2, respectively, inducing 86.51% cell death. Together, these data reveal that chlorophyllin f has high photochemical efficiency in 5637 and T24 cells, and its photocytotoxicity in different cell lines is dependent on both the PS and the light energy doses during PDT. Again, these results confirm that chlorophyllin f would be an effective photosensitizer for the therapy of NMIBC. As cancer cells have a much higher accumulation of photosensitizers than normal cells, PDT is to some degree a tumor-selective treatment modality. Additionally, ROS, the cytotoxic agents generated by specific light interaction with photosensitized target tissues or cells, exist within a very limited distance (
