Effective Photoimmunotherapy of Murine Colon Carcinoma Induced by the Combination of Photodynamic Therapy and Dendritic Cells

Photodynamic therapy (PDT) is a promising treatment of various malignant and nonmalignant disorders. It involves systemic administration of a photosensitizer that preferentially accumulates in transformed cells followed by an illumination of the tumor with a monochromatic and collimated beam of laser light (1). In the presence of oxygen, the laser light activates the photosensitizer and initiates a complex photochemical reaction that generates cytotoxic intermediates (2). The mechanism of higher photosensitizer retention in the tumor as compared with normal tissues has not been fully elucidated but probably results from the binding of Photofrin to low-density lipoproteins the receptors for which are expressed on tumor cells, better photosensitizer internalization at low pH, impaired lymphatic drainage, or a combination of these factors. Photofrin-based PDT has been approved by the FDA for treatment of early and late endobronchial non-small cell lung cancer in patients for whom surgery and radiotherapy are contraindicated, and for palliative treatment of advanced esophageal cancer (1, 3). Approval is pending for early-stage esophageal cancer in conjunction with Barrett’s esophagus. Another photosensitizer, temeporfin, has been approved in the European Union for the palliative treatment of patients with advanced head and neck cancer, and 5-aminolevulinic acid-based PDT is approved for the treatment of skin and head and neck cancers in many countries of the European Union. At least five other photosensitizers are in various stages of clinical trials.

Tumor destruction after PDT results from direct cytotoxic effects toward tumor cells and vascular damage, as well as from the induction of local inflammatory response (1, 4, 5). The relative contribution of all of these mechanisms is difficult to establish, but it seems that all of them are necessary for the successful outcome of the treatment. Extensive preclinical studies are being held to optimize each of these mechanisms for the most effective curative effects of PDT.

Accumulating evidence indicates that effective control over the PDT-treated tumors is exerted by the immune system. PDT is less effective in immunodeficient or in immune cells-depleted animals (6). Moreover, several immune-stimulating cytokines or immunomodulators and adoptive transfer of immune cells have been shown to potentiate the antitumor effectiveness of PDT (7, 8, 9, 10, 11). In some tumor models, PDT is even capable of controlling distant disease (12); but in others, this treatment was shown to increase the number of lung metastases (13). This latter observation should be kept in mind during translating the results of animal studies into a clinical setting. Despite the potential involvement of the immune system in the antitumor effects of PDT, this treatment modality is usually inefficient in the complete eradication and long-term control over disseminated tumors. Therefore, approaches that would exploit the effects of PDT for the induction of an effective systemic immunity are intensively being pursued.

Dendritic cells (DCs) are the professional antigen-presenting cells and the most effective inducers of adaptive immunity (14). These cells efficiently acquire antigens from apoptotic and necrotic tumor cells (15). This process is restricted to the immature stage of development, when DCs are effectively acquiring antigens but express low levels of MHC and costimulatory molecules. Immature DCs present antigens quite inefficiently. Additional signals, often referred to as danger signals, induce maturation, which transforms DCs into effective antigen-presenting cells that migrate to regional lymph nodes for the activation of T lymphocytes (16). Whereas a number of microbial products have been shown to provide maturation signals for immature DCs, it has not yet been unequivocally resolved what are the danger signals necessary for the effective stimulation of DCs interacting with dying tumor cells. Accumulating evidence indicates that some heat shock proteins (HSPs) might play this role (17, 18, 19).

The present study builds on the observations that PDT induces both necrotic and apoptotic death of tumor cells accompanied by oxidative stress and induction of HSPs. Therefore, PDT creates a unique environment that provides tumor antigens accompanied by “danger” signals that could trigger maturation signals for DCs. The aim of these studies was, therefore, to check whether the combination treatment with PDT followed by the administration of immature DCs could induce an effective antitumor immunity.

BALB/c mice, 8–12 weeks of age, were used in the experiments. Breeding pairs were obtained from the Institute of Oncology (Warsaw, Poland). All of the experiments with animals were performed in accordance with the guidelines approved by the Ethical Committee of the Medical University of Warsaw.

Photofrin was a generous gift of QLT PhotoTherapeutics, Inc. (Vancouver, BC, Canada). A vital nonphototoxic fluorescent dye, 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE), was purchased from Molecular Probes (Leiden, the Netherlands).

Murine Colon-26 (C-26, a poorly differentiated, immunogenic colon adenocarcinoma cell line) cells were obtained from Professor Czesław Radzikowski (Institute of Immunology and Experimental Medicine, Wrocław, Poland). Cells were cultured in RPMI 1640 (Invitrogen, Carlsbad, CA), supplemented with 10% heat-inactivated FCS, antibiotics, 50 μm 2-mercaptoethanol, and 2 mml-glutamine (all from Invitrogen), hereafter referred to as culture medium.

DCs were isolated and cultured according to the method of Inaba et al. (20) and as described previously (21). Briefly, DCs were obtained from bone marrow precursors by flushing femur, tibia, and humerus bones of 8–10-week-old BALB/c mice with cold PBS. RBCs were lysed using ammonium chloride. Cells (1 × 10⁷ cells/5 ml/well) were then cultured in 6-well plates (Nunc, Roskilde, Denmark) in culture medium, and 10 ng/ml granulocyte macrophage colony-stimulating factor (GM-CSF; PeproTech, EC, Ltd, London, United Kingdom). On the 2nd day, the culture medium was removed and fresh medium containing 10 ng/ml GM-CSF (PeproTech) and 5 ng/ml interleukin (IL)-4 (PeproTech) was added. The procedure was followed by replacing 75% of medium with a fresh one containing 10 ng/ml GM-CSF and 5 ng/ml IL-4 on day 4. Loosely adherent cells were replated on day 6 into new 6-well plates at a concentration of 5 × 10⁶ cells/5 ml/well in culture medium containing 10 ng/ml GM-CSF and 5 ng/ml IL-4. On day 8, the cells, collected by gently scraping (Cell Scraper, 23 cm; Nunc) the wells, were used for the experiments.

Individual C-26 tumors were excised and snap-frozen 24 h after PDT (10 mg/kg Photofrin; 90 J/cm² light dose). Several cryostat sections, 10 μm thick, were cut from each tumor. Some sections were stained with H&E routinely, the other sections underwent terminal deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) staining.

DNA fragmentation was detected by terminal deoxynucleotide transferase-based, in situ cell death detection kit (TUNEL; Boehringer Mannheim, Mannheim, Germany). The procedure was performed according to the manufacturer’s instructions.

For Western blotting and immunofluorescence studies, C-26 cells were cultured with 5 μg/ml Photofrin for 24 h before illumination. After a washing with PBS, the cells were illuminated with a 50-W sodium lamp (Philips) with a light filtered through a red filter to a final dose of 4.5 kJ/m², as described previously (22). After 4 h of culture in the fresh medium, the cells were washed with PBS and were lysed with radioimmunoprecipitation assay (RIPA) buffer (Tris base 50 mm, NaCl 150 mm, NP40 1%, sodium deoxycholate 0.25%, EDTA 1 mm) with protease inhibitors cocktail (Roche Diagnostics, Mannheim, Germany). Protein concentration was measured with the use of BCA protein assay (Pierce, Rockford, IL). Equal amounts of proteins were separated on 10% SDS-polyacrylamide gel, transferred onto polyvinylidene difluoride membranes, blocked with TBST [Tris-buffered saline (pH 7.4), 0.05% Tween 20] with 5% nonfat milk and 5% fetal bovine serum. The following antibodies were used for the 2-h incubation: mouse monoclonal antitubulin, goat polyclonal anti-HSP60, mouse monoclonal anti-HSP70, rabbit polyclonal anti-HSP90, goat polyclonal anti-HO-1, goat polyclonal anti-GRP78, goat polyclonal anti-GRP94, mouse monoclonal anti-HSP72/73, goat polyclonal anti-HSP27 at 1:1000 (all from Santa Cruz Biotechnology, Inc., Santa Cruz, CA). After extensive washing with TBST, the membranes were incubated for 45 min in corresponding alkaline phosphatase-coupled secondary antibodies (Jackson Immuno Research Inc. West Grove, PA). The color reaction was developed using NBT (p-nitro-blue tetrazolium chloride) and BCIP (5-bromo-4-chloro-3-indolyl-phosphate; Sigma). The immunofluorescent-microscopic studies were performed in cytospin preparations of the cells collected at 4 h after PDT. The specimens were air-dried, fixed in acetone (at −20°C), and then incubated in 5% normal donkey serum (Jackson Immunoresearch) in PBS for 1 h at room temperature to reduce nonspecific binding. The primary antibodies (the same as in Western blot studies) were applied overnight at 4°C, diluted 1:500 in 5% normal donkey serum in PBS, except for the anti-HSP72/73 antibody, which was used at the dilution of 1:100. Secondary detection reagents (all from Jackson Immunoresearch) were: donkey antirabbit or donkey antimouse antibodies conjugated to Cy3 fluorochrome, or biotinylated donkey antigoat antibody followed by streptavidin-Cy3. Cell nuclei were counterstained with Hoechst 33342 (Molecular Probes). After being mounted in Vectashield medium (Vector) and coverslipped, the specimens were examined under the Leica TCS SP2 laser confocal microscope (Leica, Mannheim, Germany), using the HeNe 543 nm, and MaiTai IR femto (double-photon operating at 780 nm) lasers, for Cy3 and Hoechst 33342 excitation respectively. The image-stacks of consecutive focal planes were collected with 0.5-μm z-interval, in 1024 × 1024 pixel format, using Plan Apo 63× oil, NA 1.32 objective, at no zoom. Subsequently, maximum intensity projections (extended focus images) were calculated from each fluorescence channel of the image stack and were stored as RGB (red-green-blue) images together with original image stacks. Special care was undertaken so as to keep the same instrument setting and operation-conditions while scanning control and PDT-treated samples.

C-26 cells were stained with 5 μm CFSE for 5 min., were washed three times with PBS containing 5% fetal bovine serum, and were incubated with Photofrin (5 μg/ml) for 24 h. After illumination with a laser light (10 kJ/m²), the cells were incubated for 4 h at 37°C and 5% CO₂. Then, immature bone marrow-derived DCs (1 × 10⁵) were added in a volume of 100 μl and cocultured with PDT-treated C-26 cells for another 4 h. The cells were then collected and washed three times in PBS. Surface expression of MHC class II molecules was determined by incubating the cells with R-phycoerythrin (R-PE)-conjugated rat antimouse I-A/I-E monoclonal antibody against MHC class II antigen (PharMingen, San Diego, CA). After a 30-min incubation at 4°C, the cells were washed three times in PBS containing 2% heat-inactivated fetal bovine serum and 0.1% sodium azide and were fixed with PBS containing 1% paraformaldehyde (Polysciences, Warrington, PA). A total of 10,000 cells were analyzed using a FACScan (Becton Dickinson Immunocytometry, San Jose, CA).

C-26 cells were stained with CFSE and were incubated with Photofrin (5 μg/ml) for 24 h. After illumination with a laser light (10 kJ/m²) the cells were incubated for 4 h at 37°C and 5% CO₂. Then, immature bone marrow-derived DCs (1 × 10⁵) were added in a volume of 100 μl and cocultured with PDT-treated C-26 cells for another 4 h. The coculture was subsequently incubated with R-PE-conjugated rat antimouse I-A/I-E monoclonal antibody against MHC class II antigen (PharMingen; 1 h on ice), plated into a 16-well chamber slides (Nunc), and kept at 37°C until the majority of the cells attached to the glass surface (about 90 min). Then, the cells were fixed in 2% paraformaldehyde in PBS, rinsed in PBS, and (after disassembling the chamber-slide), mounted in Vectashield medium (Vector) and coverslipped. The specimens were examined under the Leica TCS SP2 laser confocal microscope (Leica), using the 488-nm line of the Argon laser and the 543-nm line of the HeNe laser, for CFSE and phycoerythrin excitation, respectively. The detection settings were adjusted so as to prevent any appreciable cross-talk between the fluorochromes. Along the z-axis, usually 20–30 thin optical sections with a z-step of 0.5 μm were scanned in a 1024 × 1024 pixel format. Images were taken with a Plan Apo 63× oil, NA 1.4 objective lens at various zoom factors. Subsequently, maximum intensity projections (extended focus images) were calculated from each fluorescence channel of the image stack and stored as RGB (red-green-blue) images together with original image stacks. Further image processing and two-dimensional deconvolution was performed with the use of AutoDeblur/Autovisualize 9.1 software (AutoQuant, Inc., Watervliet, NY).

IL-12p40 and tumor necrosis factor (TNF) concentrations in culture supernatants were determined by commercially available ELISA kits (R&D Systems, Wiesbaden-Nordenstadt, Germany) according to the instructions of the manufacturer.

Cytotoxic activity of lymph node cells and splenocytes from mice treated with PDT and/or DCs (for details, see the following section) was tested in a standard (⁵¹Cr) release assay, as described previously (23). Lymph nodes and spleens were harvested and RBC-depleted single-cell suspensions were generated. In some cultures, recombinant IL-2 (Proleukin; specific activity of 18 × 10⁶ units/ml) at 20 units/ml was added for 72 h. To deplete specific lymphocyte subsets, we used a magnetic cell-separation system (MACS; Miltenyi-Biotec, Bergisch Gladbach, Germany). Removal of CD8⁺ and natural killer (NK) cells from splenocyte suspension was performed by passing magnetically labeled cells (with anti-CD8a and anti-NK MicroBeads) through a magnetic cell separator, according to the manufacturer’s protocol. Finally, 100-μl aliquots of cells were added to ⁵¹Cr-labeled C-26 cells. The cells were incubated either for 4 h or 18 h, the supernatants were collected, and the radioactivity of the ⁵¹Cr released from target cells was measured in a gamma counter (Wallac, Gaithersburg, MD). Maximum ⁵¹Cr release was determined in target cells treated with Triton X-100 at final concentration of 0.5%. The cytotoxicity was estimated as a cell lysis % according to the formula: cell lysis % = [(experimental release − spontaneous release)/(maximum release − spontaneous release)] × 100.

For in vivo experiments exponentially growing C-26 cells were harvested, resuspended in PBS at a concentration of 2 × 10⁵/20 μl of PBS and were injected into the footpad of the right hind limb of experimental mice. Tumor cell viability measured by trypan blue exclusion was always above 95%. Photofrin was administered i.p. at a dose of 10 mg/kg, 24 h before illumination with 630-nm light (day 6 after inoculation with tumor cells). Control mice received 5% dextrose. The light source was a He-Ne ion laser (LaserProject 2000, Warsaw, Poland). The light was delivered to tumors on day 7 after inoculation with tumor cells using a fiberoptic light delivery system. The power density at the illumination area, which encompassed the tumor and 1–1.5 mm of the surrounding skin, was ∼80 mW/cm². The total light dose delivered to the tumors was 90 J/cm². During the light treatment, mice were anesthetized with ketamine (87 mg/kg) and xylazine (13 mg/kg) and were restrained in a specially designed holder. On day 6 of the experiment, all of the mice were inoculated with 1 × 10⁵ C-26 cell in the left (contralateral) hind limb. DCs (1 × 10⁶) were injected into tumors growing in right hind limbs on days 7 (1 h after PDT) and 8. PBS was used as a control for DC injections. Local tumor growth was determined as described previously (24) by the formula: Tumor volume (mm³) = (longer diameter) × (shorter diameter)²/2.

Data were calculated using Microsoft Excel 98. Differences in in vitro cytotoxicity assays and in tumor volume were analyzed for significance by Student’s t test. Significance was defined as a two-sided P < 0.05.

Numerous previous studies have shown that PDT can induce both necrosis and apoptosis in tumor cells. The relative contribution of these processes to the overall kill of tumor cells depends mainly on the photosensitizer concentration and/or the light dose. To validate these observations in our tumor model, we inoculated mice with 1 × 10⁵ viable C-26 cells and then treated tumors with PDT (Photofrin was administered on day 6 after inoculation of tumor cells at a dose of 10 mg/kg, and the light was delivered on day 7 at a dose of 90 J/cm²). We observed that, at these defined conditions, Photofrin-based PDT could induce apoptosis in C-26 cells (Fig. 1, A and B). H&E staining of the tumor specimens revealed that the PDT-treated tumors contained regions of necrosis (Fig. 1, C and D). The tumors of mice treated with PDT were edematous and painful (mice tried not to use the treated limbs), indicating that the treatment induced local inflammation typical for the necrosis but not for apoptosis. In vitro treatment of C-26 with PDT also induced apoptosis in tumor cells (Fig. 1 E).

Because PDT causes an oxidative stress in treated tumor cells that leads to increased expression of several HSPs, we decided to investigate the influence of PDT on the expression of a panel of HSPs including those that were reported to influence the function of DCs. Confocal laser scanning microscopy studies performed with cells collected 4 h after PDT revealed that this treatment leads to a significant induction of HSP60, hem oxygenase 1 (HO-1, also referred to as HSP34), GRP78, and HSP90. In vitro treated C-26 cells also expressed slightly elevated levels of HSP27 and HSP72/73. No induction of HSP70 or GRP94 was observed (Fig. 2). Western blotting analysis confirmed these observations.

To induce an effective immune response, DCs need first to ingest tumor-derived material before processing it and presenting to T cells. Therefore, we decided to investigate whether DCs are able of endocytosing PDT-damaged tumor cells or tumor cell fragments. C-26 cells were stained with CFSE, a fluorescent dye that gives a strong and stable green fluorescence. CFSE-labeled C-26 cells were incubated with Photofrin for 24 h and were exposed to laser light. Coculture of CFSE-labeled and PDT-treated C-26 cells (but not untreated C-26 controls) with DCs resulted in an efficient uptake of tumor cells and/or tumor cell remnants, as demonstrated by confocal laser scanning microscopy (Fig. 3,A–C). Some DCs were observed to interact directly with lethally damaged tumor cells [there were numerous blebs and vacuoles at the surface of tumor cell (green) shown in Fig. 3,A1]. There were DCs that contained large (Fig. 3,A2) as well as small (Fig. 3A3 ) fragments of C-26 cells. The flow cytometry analysis confirmed observations with confocal laser scanning microscopy. Whereas only 3% of DCs cocultured with control C-26 cells contained green fluorescence (Fig. 3,B), 51% of such cells contained green fluorescence when cocultured with PDT-treated and CFSE-labeled C-26 cells (Fig. 3 D).

The next question was whether DCs become functionally activated after interaction with PDT-treated C-26 cells. One of the most important secreted mediators of activated DCs is IL-12, a cytokine with potent antitumor activity and the ability to polarize T helper 1 cell (Th1)/Th2 response. Unstimulated DCs produced three times more IL-12 than did DCs cocultured with control C-26 cells (107.93 ± 33.33 pg/ml and 37.38 ± 7.74 pg/ml, respectively). Importantly, DCs cocultured with PDT-treated C-26 cells produced 10 times more IL-12 than did unstimulated controls (1073.22 ± 39.34 pg/ml; Fig. 3 F). Altogether, the results of these studies unequivocally showed that DCs can interact with and efficiently endocytose PDT-treated tumor cells or tumor cell fragments and become functionally activated.

In the next step we decided to investigate whether intratumorally (i.t.) injected DCs are able to reach local lymph nodes, where in appropriate environment they could present antigens to naive T cells. CFSE-labeled DCs were injected into control or PDT-treated C-26 tumors growing in BALB/c mice. After 24 h, the local (popliteal) as well as distant (cervical) lymph nodes were isolated, and cell suspensions were analyzed with flow cytometry. Of 1 × 10⁶ DCs inoculated into untreated C-26 tumors, >0.2% managed to get into local and even less into distant lymph nodes (Table 1). Importantly, DCs injected into C-26 tumors at 1 h after PDT also migrated to the lymph nodes.

Although the number of DCs that must reach lymphatic tissue for an effective immune response is unknown, it was shown that i.v. administration of as few as 9 × 10³ DCs pulsed with tumor peptides had a measurable antitumor effect (25). The antigen-loading capacity of DCs after in vivo PDT was not determined in this study. However, the finding of effective endocytosis of PDT-treated C-26 cells and the capacity to migrate to the lymph nodes warranted additional in vivo experiments aimed at verifying the potential of DCs to induce adaptive immune response in the therapeutic schedule that might be easily adopted for the combination treatment.

Seven days after treatment (PDT followed by inoculation of 1 × 10⁶ DCs), regional lymph nodes and spleens were removed from mice, and the cytotoxic activity of the lymphocytes was measured in 4-h and 18-h cytotoxicity assays that measure, respectively, spontaneous cytotoxicity (attributed to NK cells) as well as specific cytotoxicity (attributed mainly to CD8⁺ T cells). Whereas there was no measurable stimulation of lymphocyte cytotoxicity in mice treated with PDT or DCs alone, inoculation of DCs into PDT-treated C-26 tumors growing in BALB/c mice resulted in a significant stimulation of lymph node cells’ cytotoxicity toward tumor cells (Fig. 4, A and B). A brief expansion of lymph node cells with IL-2 resulted in the stimulation of spontaneous cytotoxicity in all of the treated groups of mice and a 2-fold increase in the cytotoxicity measured after an 18-h incubation (Fig. 4,C and D). Interestingly, administration of DCs alone led to a 5-fold (as compared with controls) stimulation of cytotoxicity of spleen-derived lymphocytes (Fig. 4,E). A comparable cytotoxic activity was found in spleen cells obtained from mice treated with a combination of PDT and DCs. Depletion studies revealed that the cytotoxic activity (measured only in spleen cells because of the paucity of lymph node cells for the depletion studies) can be attributed to both NK and CD8⁺ T cells (Fig. 4,E). The culture supernatants of spleen lymphocytes from mice treated with DCs alone or in combination with PDT contained increased concentrations of TNF, a cytokine that is necessary for the induction of adaptive immunity and that is one of the mediators of NK and T-cell cytotoxicity (Fig. 4 F).

To test the concept that administration of DCs after PDT, a treatment that causes necrotic and apoptotic cell death accompanied by oxidative stress and HSP expression can induce an effective antitumor response, we used a transplantable and non-metastasizing C-26 colon adenocarcinoma growing in BALB/c mice. Tumor cells (2 × 10⁵) were inoculated into the right hind limb of experimental animals and were allowed to grow for 7 days before the PDT procedure. On day 6 (one day before laser illumination) mice were given injections of 1 × 10⁵ C-26 cells into the contralateral hind limb. DCs were administered at two doses – 1 h after PDT and 24 h later. This model enabled monitoring of the local growth of PDT-treated tumor (right hind limb) as well as the growth of an unmanipulated tumor (left hind limb) that grows at a distant site, mimicking metastasis. The combination treatment produced the strongest antitumor effects against tumors growing in the right hind limbs (Fig. 5,A). The statistical significance (P < 0.05; Student’s t test) was reached on days 16 and 18 after inoculation with C-26 cells. Remarkably strong antitumor effects were also observed in mice treated with the combination of PDT and DCs in the unmanipulated tumors; in six of seven mice, tumors completely disappeared (although the tumors were clearly visible on day 5 after the inoculation with C-26 cells; Fig. 5 B).

The main finding of this study is that a combination therapy approach using PDT and DCs was more effective than either procedure alone in producing antitumor effects. These observations may have significant translational importance for the development of improved clinical treatment regimens. PDT is a novel treatment modality used for the management of solid tumors and a variety of nonmalignant diseases (1). It is approved for use as a primary therapy for early-stage disease, as a palliation in advanced cancers, and as a surgical adjuvant in the treatment of lung, bladder, esophageal, head and neck, and gastric cancers in many countries. Moreover, PDT is extensively investigated in clinical trials in the treatment of other cancers including breast, colon, and bile duct cancers or brain tumors (1, 3). Likewise, DCs, despite their costly and time-consuming isolation and expansion procedures, are now being examined in several clinical trials. Whereas unmanipulated DCs have shown signs of activity in initial clinical trials, there is an intensive investigation aimed at optimizing their use for therapeutic purposes. Our results strongly support the consideration and development of protocols to evaluate the clinical efficacy of combining PDT with DCs.

One of the limitations of PDT is that this treatment modality is effective in the local control of the tumor with very rarely observed and unexplained (presumably immune-mediated) systemic antitumor effects. Therefore, combination strategies that could exploit unique properties of PDT for the induction of systemic antitumor effects are being intensively investigated. PDT induces both necrotic and apoptotic cell death that primarily follow oxidative stress generated by excited photosensitizers. PDT induces rapid and massive release of proinflammatory mediators liberated from cancer cell membranes, damaged endothelial cells, and tumor stroma (26, 27). Moreover, PDT-treated cells secrete a number of cytokines including TNF, IL-1β, and IL-6 (28, 29) that participate in the recruitment of neutrophils and other myeloid cells (30). This intense localized inflammation could be appropriately implemented as an initiating event for the induction of effective antitumor immunity. Indeed, a recent study demonstrated that PDT-generated tumor cell lysates are able to activate DCs and can induce antitumor immune response (31); and many studies show strengthened antitumor effectiveness of combinations of PDT with immunomodulators (9, 11).

Tumor cells dying via apoptotic or necrotic mechanisms are a rich source of antigens for processing and presentation by DCs. There are, however, conflicting reports on the immunomodulatory effects exerted by dying tumor cells on the antigen-presenting functions of DCs (32, 33, 34). The overwhelming data indicate that apoptotic tumor cells taken up by immature DCs provide antigens in a tolerogenic manner (35). In contrast, in situ killing of tumor cells by nonapoptotic mechanisms is associated with high immunogenicity (34). Exposure to heat stress before inducing apoptosis elevates expression of membrane HSPs (HSP72 and HSP60) in apoptotic tumor cells and converts them into more immunogenic cells that are effectively taken up by DCs (17). Inducible HSPs (HSP60, HSP70, and HSP90) activate monocytes and DCs and stimulate the expression of several cytokines such as IL-12 and IL-15 (18, 19), with potent immunoregulatory antitumor activities (36, 37). The immunoregulatory role of other HSPs is less well understood. HSPs chaperone antigenic peptides and channel them into the MHC class I presentation pathway of antigen-presenting cells (38, 39, 40). Additionally, some HSPs protect peptides processed by the proteasome from further degradation before being delivered by transporters associated with antigen processing (TAP) proteins to the endoplasmic reticulum for MHC class I loading (40, 41). Finally, down-regulation of HSP70 can lead to inefficient stimulation of the antitumor immune response, and transfection of tumor cell spheroids with HSP70 restores efficient antigen presentation (42). It is unclear how HSPs can stimulate maturation of DCs. Some observations indicate that HSP60 and HSP70 are endogenous stimuli for toll-like receptors that evolved to recognize pathogen-associated molecular patterns of infectious microorganisms (38). The induction of HSP expression may, therefore, provide danger signals required for efficient maturation of DCs.

In our studies, PDT induced significant amounts of several HSPs. Although correlative studies are never fully compelling, it is possible that some HSPs, induced by PDT, function as signature molecules for the activation of DCs maturation and induction of systemic antitumor response. Several previous studies revealed that stress proteins are expressed after PDT, including HSPs (HSP34, HSP60, HSP70, HSP90, and HSP110) and glucose-regulated proteins [GRPs (GRP74, GRP78, and GRP100); Refs. 43, 44, 45, 46, 47, 48]. Some of these proteins are presumed to be involved in rescue responses of cells after PDT (49). Our studies show that these rescue responses might be exploited for more effective tumor treatment. Indeed, coculture of DCs with PDT-treated tumor cells resulted not only in efficient endocytosis of tumor cells but also in a functional activation of antigen-presenting cells that produced a substantial amount of IL-12. Inoculation of immature DCs to the tumors treated with PDT resulted in efficient migration of these cells to both local and distal lymph nodes and stimulation of the cytotoxic activities of lymphocytes isolated from local lymph nodes. The migration of DCs injected into PDT-treated tumors was slightly, but insignificantly, decreased as compared with DCs inoculated into control tumors (Table 1). We can only speculate that this effect is a result of oxidative stress that is induced by the PDT procedure and that leads to decreased viability of injected cells. Moreover, both blood and lymphatic vessels are damaged by PDT, which could contribute to impaired DCs trafficking. Remarkably, the combination treatment leads to the potentiated antitumor response that is not limited to the treated tumor but is also effective in the control of the distant growth. Quite unexpectedly, we observed a stronger antitumor effect of the combination treatment against the tumors inoculated just before PDT into the contralateral footpads. This effect can be explained by the induction of concomitant immunity that is effective in the eradication of small tumor foci but is less efficient in the elimination of larger tumors.

Altogether, intratumorally injected DCs after PDT might prove advantageous in many aspects. Such treatment alleviates the need for the in vitro loading with tumor antigens, eliminates concerns regarding unpredictable trafficking of DCs injected via other routes, and allows DCs to acquire, process, and present tumor-derived material in the context of ongoing inflammation, which potentially renders the whole process more immunogenic. Additional studies are definitely necessary to optimize this combination treatment for more effective control of the primary tumor. Our studies are the first report of effective photoimmunotherapy that involves PDT and intratumoral administration of immature DCs. The feasibility of this treatment may warrant additional studies in the clinical setting.

We thank Adam Gołąb (Erco Leuchten. GmBH, Warsaw) for the construction of the sodium lamp for in vitro experiments, Anna Czerepińska and Elżbieta Gutowska for excellent technical assistance, the Foundation for Polish Science for financing the FACSCalibur, and Dr. Antoni Wrzosek for help with confocal microscopy.