The purpose of the present study was to investigate the mechanism of photodynamic therapy (PDT) supplemented with exogenously added 5-aminolevulinic acid (ALA) on human urothelial cancer (UC). Moreover, we aimed to determine whether the therapeutic effects of ALA-based PDT (ALA-PDT) for UC could be enhanced by deferoxamine (DFX), an inhibitor of ferrochelatase. The efficiency of ALA-PDT on these cells was analyzed using flow cytometry and the type of cell death was also assessed. The ALA-PDT promoting effect of DFX was examined on both UC cells and human umbilical vein endothelial cells (HUVEC). The ALA-PDT decreased levels of mitochondrial membrane potential and induced cell death mainly via apoptosis in these cells. Moreover, inhibition of ferrochelatase by DFX led to an increase of protoporphyrin IX (PpIX) accumulation and enhanced the effect of ALA-PDT on UC cells. We further investigated the effect of DFX on in vivo PDT with a tumor-bearing animal model and found that DFX efficiently enhanced tumor cell apoptosis. ALA-PDT induced death of neovascular endothelial cells in tumors but did not affect small vessel endothelial cells in normal tissues surrounding the tumor. Furthermore, DFX enhanced inhibition of neovascularization. These results demonstrated ALA-PDT dominantly induced apoptosis over necrosis by direct action on UC as well as via antiangiogenic action on neovacular endothelial cells, suggesting that the therapeutic damage by ALA-PDT could be kept to a minimum in the surrounding normal tissues. In addition, increased accumulation of PpIX by DFX could enhance this effectiveness of ALA-PDT.
Urothelial carcinoma (UC) of the urinary bladder is the fourth most common malignancy in the United States. In Japan, approximately 16 500 new cases are diagnosed and 50 000 endoscopic surgeries are performed each year. The standard treatment for non-muscle invasive cancer is transurethral resection; in contrast, radical cystectomy is indicated for operable invasive bladder cancer, whereas chemotherapy offers the only viable therapeutic option for distant metastasis and locally advanced bladder cancer. In fact, the major therapeutic strategy for bladder cancer is the conventional surgical procedure, leading to a lower quality of life and a less favorable prognosis. Therefore, the development of novel therapeutic strategies are necessary if we are to improve the clinical outcome for patients with bladder cancer.
Recently, photodynamic diagnosis (PDD) and therapy (PDT) mediated by 5-aminolevulinic acid (ALA) as a new-generation photosensitive substance has received much attention. ALA-mediated PDD (ALA-PDD) is currently widespread in clinical use, especially for the intraoperative diagnosis of brain tumor, prostate cancer and bladder cancer, to confirm the precise surgical margin and prevent overlooking endoscopic invisible lesions, leading to a better prognosis. In contrast, ALA-mediated PDT (ALA-PDT) is not yet widely applied in the clinical setting except for the standard treatment for skin pre-malignancies. The excess accumulation of a photoactive substance, protoporphyrin IX (PpIX), biosynthesized from ALA in mitochondria, is the linchpin to the mechanism of ALA-PDD and ALA-PDT in various cancer cells. We previously reported ALA-induced accumulation of PpIX was increased by deferoxamine (DFX), an iron chelator, in a time- and concentration-dependent manner in histiocytic lymphoma cells[9-11] and UC cells.[12, 13] When a photosensitive substance, PpIX, excessively accumulates in the mitochondria of cancer cells and is irradiated with visible light of the specific wavelength at a low output, reactive oxygen species (ROS), predominantly singlet oxygen 1O2, are produced and injure the cells.[14, 15] In addition to this direct cytotoxic action on cancer cells, the antitumor effects of ALA-PDT is caused partly by intratumoral antiangiogenic effects. In various types of cancer, excessive PpIX accumulation is observed. In particular, in the urothelium, tumor selectivity of PpIX is increased nine- to 16-fold. Moreover, bladder cancer is relevant to angiogenesis and usually shows a hypervascular appearance. These biological features demonstrate bladder UC is a good indication for ALA-PDD and ALA-PDT with antiangiogenic effects. Therefore, we investigated the antitumor and antiangiogenic effect of ALA-PDT in human UC. Moreover, the improvement and enhancement of photodynamic therapeutic effects by the addition of DFX on cell death were examined as a preliminary experimental study toward clinical implementation in an attempt to optimize PDT with a lower concentration of ALA, so as to avoid the cytotoxic side-effects on normal cells.
Materials and Methods
The ALA was provided by Cosmo Bio Co. Ltd. (Tokyo, Japan). An iron chelator, DFX, was provided by Novartis Pharma Co. Ltd. (Tokyo, Japan) and suspended in saline as a stock solution. Tetramethylrhodamine-ethyl-este (TMRE), which is a potentiometric fluorescent dye for the detection of mitochondrial membrane potential, and PpIX were obtained from Sigma Chemical Co. (St Louis, MO, USA). N-methyl protoporphyrin IX (Frontier Scientific Inc., Logan, UT, USA) was obtained from Funakoshi (Tokyo, Japan). 10-Nonyl acridine orange (NAO), which is a fluorescent dye compound used for the detection of mitochondria and liposomes, was obtained from Molecular Probes (Eugene, OR, USA). All other chemicals were of analytical grade and obtained from Nacalai Tesque (Kyoto, Japan). The NAO and TMRE were dissolved in DMSO and stored in aliquots (4°C) until use.[9-13]
Cell lines and culture conditions
The highly metastatic human bladder carcinoma cell line 253J B-V is a subline derived through orthotopic cycling of the non-tumorigenic human UC cell line 253J-P. 253J B-V was used as subcutaneous tumors for the therapy experiments. The cell line was grown as a monolayer culture in modified Eagle's Minimal Essential Medium supplemented with 10% fetal bovine serum (FBS), vitamins, sodium pyruvate, l-glutamine, non-essential amino acids and penicillin streptomycin. Human umbilical vein endothelial cells (HUVEC) were obtained from Clonetics and grown in endothelial growth medium-2 BulletKit (Clonetics, San Diego, CA, USA) (37°C, 5% CO2). Cell cultures were established in 75 cm2 flasks and kept in a humidified atmosphere (37°C, 5% CO2).
Experimental schedule of in vitro PDT
The light source was a Na-Li lamp, TheraBeam VR-630 (Ushio Lighting Inc., Tokyo, Japan) and the wavelength of exposing light was 550–740 nm, preferentially 630 and 670 nm. The power density was 20–65 mW/cm2 at a distance from the light source of 20–5 cm. 253J B-V cells (4 × 104 cells/mL) and HUVEC (5 × 105 cells/mL) were cultured with FBS for 24 h and incubated with ALA without FBS for 3 h. The cells were treated with light irradiation for 10 min, cultured with FBS for 12 h, and then harvested for cell analysis. During this process, the temperature was constantly maintained at 37°C.
Detection of PpIX in cells cultured in the presence of ALA
Cells were seeded in six-well plates and cultured with serum-free cultured medium containing 0.5 mM of ALA for 1.5 h. The cells were washed with culture medium and stained with 10 nM NAO for 15 min (37°C). These cells were washed with PBS and fluorescence of NAO and PpIX was observed using fluorescence microscopy (Axiovert 200; Carl Zeiss AG, Oberkochen, Germany) with a 100 W halogen lamp. Fluorescence images were taken using a highly light-sensitive thermo-electrically cooled charge-coupled device camera (ORCAII-ER, Hamamatsu, Japan). The filter combinations were 450 nm excitation filter, 510 nm beam splitter and a 515–565 nm emission filter for NAO, a fluorescent dye compound for the detection of mitochondria; G: 365 nm excitation filter, a FT: 580 nm beam splitter and an up LP: 590 nm emission filter for PpIX.[9-13]
Flow cytometry of cellular PpIX
Of 253J B-V cells (4 × 104 cells/mL) and HUVEC (5 × 105 cells/mL) were grown on tissue culture plates and incubated for 24 h. The ALA was diluted in RPMI-1640 medium to make a stock solution of 1 M and added to a final concentration of 0.1–0.5 mM, and then incubated with cells for 3 h. After incubation with ALA, the cells were washed with PBS (without Ca2+ and Mg2+) and scraped off with a rubber policeman. After 10 min of centrifugation at 73g, the medium was decanted and 0.5 mL of PBS (without Ca2+ and Mg2+) was added. The cell suspension was measured using a Fluorescence-Activated Cell Sorter (Becton Dickinson FACS Calibur, Mountain View, CA, USA). In all, 20 000 cells were measured in each sample (excitation 488 nm, emission 650 nm).
Animals and ectopic implantation
Male athymic BALB/cA Jc1-nu nude mice were obtained from Clea Japan Inc. (Osaka, Japan). The mice were maintained in a laminar-airflow cabinet under pathogen-free conditions and used at 6–8 weeks of age.
Cultured 253J B-V cells (60–70% confluent) were prepared for injection. Mice were anesthetized with methoxyflurane. For ectopic implantation, viable tumor cells (1 × 106/0.2 mL) in Hank's balanced salt solution were implanted into the back subcutaneous tissue. The tumors were monitored until they reached a diameter of 5–7 mm. The animals tolerated the surgical procedure well and no anesthesia-related deaths occurred.
In vivo PDT with or without DFX in a tumor-bearing animal model
Thirteen days after exotopic implantation, the in vivo PDT was performed. In each ALA-based group, ALA (50 mg/kg) and DFX (100 mg/kg) were administrated intraperitoneally for 90 min prior to light irradiation. In all PDT groups, a total laser dose of 100 J/cm2 at 635 nm, the absorption band of PpIX, was irradiated at a dose rate of 100 mW/cm2 without the absorption band of photoprotoporphryn (670 nm) using a dual-color laser diode system (m&m Co. Ltd., Tokyo, Japan). Twenty-four hours after irradiation, histological examination of the removed tumors was performed.[11, 12]
Mice were killed by cervical dislocation. The necropsied tumors were fixed in 20% formalin for 24 h. The specimens were dehydrated with graded ethanols, embedded in paraffin and 4-μm sections were obtained and adhered to ProveOn Plus Microscope Slides (Fisher Scientific, Pittsburgh, PA, USA) for immunohistochemistry (IHC) and TUNEL assay. Sections (4-μm thick) were also stained with hematoxylin and eosin (HE) and the estimated volume of tumors was microscopically calculated using three axes (X, Y, Z) using the formula of π/6XYZ to evaluate the antitumoral effects on the HE slide.
For the TUNEL assay, tissue sections (5-μm thick) of formalin-fixed, paraffin-embedded specimens were deparaffinized in xylene, rehydrated in graded alcohol and transferred to PBS. The slides were rinsed twice with distilled water with BRIJ detergent (Fisher Scientific) and treated with a 1:500 proteinase K solution (20 μg/mL) for 15 min; endogenous peroxidase was blocked with 3% hydrogen peroxide in PBS for 12 min. The samples were washed three times with DW/BRIJ and incubated for 10 min with terminal-deoxynucleotidal-transferase (TDT) buffer. Excess TDT buffer was drained and the samples were incubated for 18 h (4°C) with terminal transferase and biotin-16-dUTP. The samples were rinsed four times with TB buffer and incubated for 30 min (37°C) with peroxidase-conjugated streptavidin (1:400). The slides were rinsed with PBS and incubated for 5 min with diaminobenzidine (Research Genetics Inc, AL, USA). The sections were washed three times with PBS, counterstained with Gill's hematoxylin and washed three times with PBS. The slides were mounted with a Universal Mount (Research Genetics Inc, AL, USA).
For the immunohistochemistry, tissue sections (5-μm thick) of formalin-fixed, paraffin-embedded specimens were deparaffinized in xylene, rehydrated in graded alcohol and transferred to PBS. Endogenous peroxidases were blocked by incubation in 3% hydrogen peroxide in PBS for 12 min. The samples were washed three times with PBS and incubated for 20 min at room temperature with a protein-blocking solution containing 5% normal horse serum and 1% normal goat serum in PBS (pH 7.5). Excess blocking solution was drained and the samples were incubated for 18 h (4°C) with the appropriate dilution (1:100) of monoclonal mouse anti-single stranded DNA (ssDNA) (Enzo Life Science, Inc, Tokyo, Japan) or rat monoclonal anti-CD31 antibody (Pharmingen, San Diego, CA, USA). The samples were rinsed four times with PBS and incubated for 60 min with the appropriate dilution of the secondary antibody, peroxidase-conjugated anti-mouse Immunoglobulin G (H+L) (DakoCytomation Co. Ltd., Kyoto, Japan). The slides were rinsed with PBS and incubated for 5 min with diaminobenzidine. The sections were washed three times with PBS, counterstained with Gill's hematoxylin and washed three times with PBS. The slides were mounted with Universal Mount (Research Genetics Inc).
Quantification necrosis and apoptosis in cancer cells
Proliferation and apoptosis of cancer cells were determined using immunohistochemistry with anti-ssDNA and the TUNEL assay. The tissue was recorded using a cooled CCD Optotronics Tec 470 camera (Optotronics Engineering, Goletha, CA, USA) linked to a computer and digital printer (Sony Co., Tokyo, Japan). The density of proliferative cells and apoptotic cells was expressed as an average number of the five highest areas identified within a single ×200/field.
Quantification of microvessel density
Microvessel density was determined using immunohistochemistry with anti-CD31 antibodies. Clusters of stained endothelial cells distinct from adjacent microvessels, tumor cells or other stromal cells were counted as one microvessel. The tissue was recorded using a cooled CCD Optotronics Tec 470 camera linked to a computer and digital printer. The density of microvessels was expressed as an average number of five highest areas identified within a single ×200/field.
253J-BV cells (4 × 104/mL) were incubated with 0, 0.5 and 1.0 mM ALA for 3 h in FBS-free culture medium and exposed to light (29 mW/cm2) for 10 min. The cells were incubated in culture medium containing FBS for 6 h. The cells were stained with 1 μM Hoechst 33342 for 10 min and FITC-Annexine-V for 15 min. Hoechst 33342 and FITC-Annexine-V-stained 253J-BV cells were observed under fluorescence microscopy.[11, 12]
The statistical differences in the amount of cell apoptosis and necrosis within the tumors were analyzed with the Mann–Whitney test. The incidence of tumors and estimated tumor volume were statistically analyzed with the Chi-squared test. A value of P < 0.05 was considered significant.
Therapeutic effects by in vitro PDT on UC cells and HUVEC
We assessed the effects using in vitro PDT on 253J-BV cells and HUVEC measured using scattergram, and FACScan with Annexin-V, TMRE and propidium iodide (PI) stain.
To examine the morphological transformation by PDT, Hoechst 33342 and FITC-Annexine-V-stained 253J-BV cells were observed under fluorescence microscopy. ALA concentration-dependent nuclear aggregation and phosphatidylserine externalization were observed (data not shown).
The scattergram showed that ALA concentration-dependent cell damages were increased by photodynamic action in ALA-treated 253J-BV cells (Fig. 1). FACScan with annexin-V and TMRE-stained cells revealed ALA concentration-dependent cell damages were caused by depolarization in the mitochondrial membrane in ALA-treated 253J-BV cells. FACScan with annexin-V and PI-stained cells revealed apoptosis-preferred cell death was increased in an ALA concentration-dependent manner by photodynamic action in ALA-treated 253J-BV cells.
Next, we sought optimal conditions for the therapeutic procedure such as irradiation time, distance and culture time-dependent cell damage by PDT in ALA-treated 253J-BV cells measured using a scattergram and FACScan with Annexin-V and PI stain. Apoptosis-preferred cell death was increased in relation to irradiation time, distance and culture time by photodynamic action in ALA-treated 253J-BV cells (data not shown).
The scattergram showed ALA concentration-dependent cell damages were increased by photodynamic action in ALA-treated HUVEC (Fig. 2). FACScan with annexin-V and TMRE-stained cells revealed ALA concentration-dependent cell damages were caused by depolarization in the mitochondrial membrane in ALA-treated HUVEC. FACScan with annexin-V and PI-stained cells revealed apoptosis-preferred cell death was increased in an ALA concentration-dependent manner by photodynamic action in ALA-treated HUVEC.
Additional effects of DFX to in vitro PDT of UC cells and HUVEC
Next we assessed the additional effects of DFX on PDT in ALA-mediated PpIX-accumulated 253J-BV cells (Fig. 3) and HUVEC (Fig. 4) measured using a scattergram and FACScan with Annexin-V, TMRE and PI stain.
The scattergram showed that cell damage was enhanced with increased concentration of DFX by photodynamic action in ALA-treated 253J-BV cells and HUVEC. FACScan with annexin-V and TMRE-stained cells revealed cell damage caused by depolarization in the mitochondrial membrane was enhanced by DFX in a concentration-dependent manner in ALA-treated 253J-BV cells and HUVEC. FACScan with annexin-V and PI-stained cells revealed apoptosis-preferred cell death was enhanced DFX concentration dependently by the photodynamic action in ALA-treated 253J-BV cells and HUVEC.
Therapeutic effects of in vivo PDT with/without DFX in a tumor-bearing animal model
Next we assessed the range of degenerated lesions treated with in vivo PDT in an established tumor of 253J-BV cells growing ectopically in the subcutis of athymic nude mice and measured using a Nikon Digital Sight DS-L1 (Nikon Co. Ltd., Tokyo, Japan). To detect apoptosis and necrosis, immunohistochemistry with anti-ssDNA antibody and TUNEL assay were used (Fig. 5, Table 1).
There were no therapeutic effects in any of the control groups with or without DFX, including the no-ALA-treated and no-light-irradiated group (n = 5), the no-ALA-treated and light-irradiated group (n = 5) and the ALA-treated and no-light-irradiated group (n = 5). The ratio (mean ± SD [range]) of the range of whole degenerated lesion to that of whole tumor in a maximal 2-D area of the tumor was 61.5 ± 23.0 (29.7–85.8)% in the ALA-treated and light-irradiated group (n = 5). The ratio (mean ± SD) of the range of apoptotic lesion to that of whole tumor was 44.9 ± 16.3 (21.7–62.6)% (calculated value) and was significantly higher than that of necrotic lesion, which was 16.6 ± 6.0 (8.0–23.2)%, P < 0.05. For in vivo PDT with DFX, the ratio (mean ± SD [range]) of the range of whole degenerated lesion was 72.4 ± 10.6 (60.7–81.3) and apoptotic lesion was 52.9 ± 7.7 (44.3–59.4), and was significantly increased compared with those of in vivo PDT without DFX (P < 0.05). Drug-induced bodyweight loss was not significantly different among any groups (Fig. 5, Table 1).
Antiangiogenic effects of in vivo PDT in a tumor-bearing animal model
To evaluate the therapeutic effects on angiogenesis, we next assessed the microvessel density subsequent to immunohistochemistry with anti-CD31 antibodies (Fig. 6).
Figure 6a shows the effect of PDT on intratumoral neovascularization. The number of intratumoral neovascularization counted per ×200/field was 16.9 ± 6.2 in the no-ALA-treated and no-light-irradiated group (n = 7), 24.8 ± 8.0 in the no-ALA-treated and light-irradiated group (n = 12) and 21.5 ± 5.1 in the ALA-treated and no-light-irradiated group (n = 12). Those values were significantly reduced to 6.9 ± 3.7 in the ALA-treated and light-irradiated group (n = 10, P < 0.01) and 4.4 ± 1.1 in the group of ALA-treated and light-irradiated with DFX (n = 6, P < 0.01). Figure 6b shows the effect of PDT on microvessels in normal skin surrounding the tumor. The number of microvessels in normal skin counted per ×200/field was not significantly changed from 14.0 ± 2.7 and 10.3 ± 4.5 in the no-ALA-treated and no-light-irradiated group (n = 3) and ALA-treated and light-irradiated group (n = 3), respectively, to 10.8 ± 4.4 in the ALA-treated and light-irradiated group (n = 10) and 10.1 ± 3.0 in the group of ALA-treated and light-irradiated with DFX (n = 6). These results revealed the antiangiogenic effects of in vivo PDT is limited to the cancerous lesion of human UC in a xenograft animal model.
5-Aminolevulinic acid is a natural amino acid that has existed in animals and plants for 3.6 billion years and is a common precursor of hemoglobin and chlorophyll. The photodynamic technology mediated by ALA is based on the tumor-selective accumulation of a photoactive substance, PpIX, biosynthesized from ALA in mitochondria. The phenomena involved in PpIX biosynthesis and metabolism are considered to be due to the common biological characteristic of cancers: preference for anaerobic metabolism of various abnormal enzyme activities and cancers, which is called the Warburg effect. Therefore, in various types of cancer, PpIX accumulation is observed excessively. When excessive PpIX accumulated in the mitochondria of cancer cells is irradiated with visible light of the specific wavelength at a low output, ROS, predominantly singlet oxygen 1O2, are produced and injure the cells.[14, 15] This is the mechanism of ALA-PDT. Therefore, PDT is a novel and promising therapeutic strategy based on the common biological characteristic of cancers and is expected as a less invasive and effective therapy clinically applicable for various types of cancer.
The UC cells have high tumor selectivity of PpIX and are relevant to angiogenesis. These biological features support that UC of the urinary bladder is a good indication for ALA-PDD and ALA-PDT. In fact, several preliminary clinical studies of ALA-PDT have previously been demonstrated for bladder cancer.[21, 22] Despite clinical studies of ALA-PDT for persistent cases, a favorable outcome on safety and efficacy was obtained. However, trial methods such as the wavelength of light excitation for cancer and the time of end-points were different between these clinical studies, which is controversial. Therefore, it is necessary to elucidate a further mechanism of action and to optimize the therapeutic modality in accordance with the mechanism in ALA-PDT.
Previously, we demonstrated ALA-mediated PpIX synthesis in human UC cells was regulated by the process of ALA uptake, ALA conversion to PpIX and metabolism of accumulated PpIX to heme. In particular, PpIX accumulation was increased by DFX through the inhibition of heme synthesis in a time- and concentration-dependent manner in human histiocytic lymphoma cells[9-11] and UC cells.[12, 13] Thus, we anticipated the cellular PpIX accumulation upregulated by the addition of DFX led to improvement of clinical practicability of PDD and PDT. In the present study, we take advantage of the effects of DFX to ALA-PDT. As expected, the results of the present study revealed increased photodynamic efficacy by DFX enhanced cell death dominantly via apoptosis over necrosis in UC cells and vascular endothelial cells.
These antitumor effects of ALA-PDT and the enhancement of photodynamic therapeutic effects by the addition of DFX were also demonstrated in human UC using a tumor-bearing animal model. Moreover, it was revealed that the antiangiogenic effects of in vivo PDT was limited to the cancerous lesion of human UC, but not to the normal skin surrounding the tumor. This means our results can contribute to increasing the therapeutic accuracy of ALA-PDT by avoiding a few cytotoxic side-effects on normal cells. This is because the vascular endothelial cells of neovascular vessels in tumor tissue are different in biological character from that of mature capillary vessels in normal tissue without inflammation. In particular, the increase in cellular proliferative potential in vascular endothelial cells of intratumorneovascularization seems to cause excess accumulation of PpIX, resulting in selective photodynamic action onto angiogenic endothelial cells in tumor tissues.
In conclusion, we demonstrated that ALA-PDT induced direct antitumor actions and antiangiogenic actions dominantly via apoptosis over necrosis. In particular, prevention of heme production and PpIX accumulation by DFX enhances the effectiveness of ALA-PDT by its action. Therefore, the present study suggests that ALA-PDT is a novel therapeutic modality for various types of cancer and less markedly influences the normal tissue surrounding the tumor, and is a low-invasive, less harmful treatment for cancer.
The authors thank Eri Sakurai and Chiaki Kawada for their valuable technical assistance.