UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of immune cells in photodynamic therapy Krosl, Gorazd 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-19602X.pdf [ 8.04MB ]
Metadata
JSON: 831-1.0087999.json
JSON-LD: 831-1.0087999-ld.json
RDF/XML (Pretty): 831-1.0087999-rdf.xml
RDF/JSON: 831-1.0087999-rdf.json
Turtle: 831-1.0087999-turtle.txt
N-Triples: 831-1.0087999-rdf-ntriples.txt
Original Record: 831-1.0087999-source.json
Full Text
831-1.0087999-fulltext.txt
Citation
831-1.0087999.ris

Full Text

THE ROLE OF IMMUNE CELLS IN PHOTODYNAMIC THERAPY by Gorazd Krosl B.Sc. University of Ljublana, Slovenia, 1981 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Pathology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1996 ©Gorazd Krosl, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It js .understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of H Q b Q c ^ y The University of British Columbia Vancouver, Canada Date Oove - o q W i i e , ) t c t ct6 DE-6 (2788) ABSTRACT The mechanisms of tumor destruction by photodynamic therapy (PDT) have proven to be very complex. The evidence from published work suggests that host factors may have an important role in both the uptake of photosensitizers and the effect of PDT. Focusing on clinically established photosensitizer Photofrin®, the role of tumor associated macrophages (TAMs) in accumulation of the drug using an experimental murine tumor mode! was examined in this thesis. To provide additional information about the inflammation processes that develop in the tumor after PDT treatment, the host cell infiltration into the treated tumor during and after the photodynamic light delivery was also investigated. The hypothesis that enhancing the host immune response can increase the curative potential of PDT was tested with two PDT-immunotherapy protocols. Photosensitizer accumulation in tumors is an important factor in the destruction of malignant lesions by PDT. The role of tumor associated macrophages in the accumulation of Photofrin in mouse SCCVII squamous cell carcinoma was studied by two different approaches. First, in vitro uptake and retention of Photofrin in mouse peritoneal macrophages were compared to that in SCCVII cells. Second, the accumulation of Photofrin in SCCVII tumor cells and tumor associated macrophages in vivo was studied using flow cytometry analysis. In all experiments, it was found that macrophages have a superior capacity for Photofrin uptake in comparison to that observed in SCCVII tumor cells. The in vivo accumulation of Photofrin showed consistently higher levels of the drug in TAMs compared to SCCVII malignant cells throughout the observation period (up to 96 hours past Photofrin administration). At 24 hours after Photofrin administration, the drug content was 74±15 (SD) ng/ixg cell protein for TAMs compared to 15+2 ng/|ig cell protein for SCCVII tumor cells. In Ill addition, factors that modify macrophage phagocytic activity also influenced the ability of these cells to accumulate Photofrin. The infiltration of inflammatory cells is a central event in an acute inflammatory reaction. The major cellular populations contained in SGCVII tumors were analyzed at various time points after PDT. The cell populations were identified by staining with monoclonal antibodies against leukocyte membrane markers using indirect immunoperoxidase technique or by flow cytometry. Photofrin-based PDT induced profound changes in the levels of different cell populations comprising the treated tumor. The most immediate change was a dramatic increase in the content of neutrophils, which occurred within 5 min. after the onset of light treatment. This was followed by an increase in mast cell.numbers, while monocyte/macrophage infiltration followed between 0 and 2 h after PDT. The examination of in vitro cytotoxicity of macrophages against SCCVII tumor cells revealed a pronounced increase in cytolysis mediated by TAMs harvested from the tumor at 2 h after PDT compared to that seen with the TAMs recovered from an untreated tumor. Such increased cytotoxicity of TAMs suggests that PDT induces an acute inflammatory infiltration of myeloid cells into the treated tumor, that is associated with functional activation of immune cells. This inflammatory reaction may provide an appropriate environment for activation of immune cells, and therapy with immunostimulators may effectively enhance its contribution to tumor destruction by, PDT. Two different PDT-immunotherapy protocols were tested for their ability to improve the curative potential of PDT. In the first case, treatment of SCCVII tumor bearing mice with glucan SPG which raised the relative content of Mac-1 positive cells infiltrating the tumor was examined. The tumor cure rate following Photofrin-based PDT (25 mg/kg Photofrin; 60J/cm2 light) increased approximately 3 times if PDT was preceded by the SPG therapy. In vitro colony iv formation assay revealed a greater killing of tumor cells in groups pre-treated with SPG, and this enhancement was particularly pronounced when the tumor excision was delayed to 8 h after PDT. In another part of the study, the SCCVII tumor cells were genetically engineered to produce the cytokine granulocyte-macrophage colony-stimulating factor. (GM-CSF). 1x107 lethally irradiated GM-CSF producing cells were injected under the parental SCCVII tumor, followed by PDT (25 mg/kg Photofrin; 60 J/cm2 light) 48 h later. This therapy resulted in a dramatic increase in tumor cures (62% cure compared to 0% for PDT alone). Similarly, the POT with 10 mg/kg Photofrin and 150 J/cm2 light (that resulted in 6% tumor cure) was enhanced to 75% cure by the GM-CSF immunotherapy administered 3 times in 2 day intervals starting at 2 days before the light treatment. A comparable effect of GM-CSF immunotherapy was observed in the combination with benzoporphyrin derivative mediated PDT. The immunotherapy alone did not have a visible effect on tumor growth, and did not significantly affect peripheral white blood cell counts. However, a marked increase in TAM cytotoxicity against syngeneic tumor cells was detected. V TABLE OF CONTENTS ABSTRACT... . . . . ............... ..II TABLE OF CONTENTS V LIST OF TABLES .....VIII LIST OF FIGURES —IX ABBREVIATIONS............... ......X LIST OF PUBLICATIONS........... ...XIII ACKNOWLEDGMENTS .....XIV CHAPTER 1. INTRODUCTION............. 1 1.1 Overview and thesis outline .". ...1 1.2 Historical perspective..... 3 1.3 PDT in clinical trials 4 1.4 Tumor localization of photosensitizers ...8 1.5 Mechanism of action 10 1.5.1 Photosensitization 10 1.5.2 Cellular effects 12 1.5.3 The effect of PDT on tumor ........14 1.5.4 PDT triggered inflammatory response 17 1.5.5. PDT and immunotherapy 20 1.6 Specific aims of the thesis ; ; 21 CHAPTER 2. THE ROLE OF TUMOR INFILTRATING MACROPHAGES IN PHOTOFRIN ACCUMULATION IN THE TUMORS... 23 2.1 Summary .23 2.2 Introduction ,-. 24 2.3 Materials and methods -.25 2.3.1 Murine tumor model ....25 2.3.2 Measurement of Photofrin uptake 26 2.3.2.1 Harvesting macrophages and Photofrin exposure in vitro 26 2.3.2.2 Effect of plasma proteins on Photofrin . accumulation and retention 27 2.3.2.3 Fluorescence analysis of tumor cells... 27 2.3.2.4 Photofrin measurement from cell extracts 28 2.3.2.5 Determination of Photofrin content by radiolabeled drug...... : 29 2.3.2.6 Separation of macrophages by differential cell attachment 29 2.3.3 Statistical analysis...... 30 2.4 Results 30 2.4.1 The uptake of Photofrin by murine peritoneal macrophages and SCCVII tumor cells in vitro ...30 2.4.2 The effect of serum on Photofrin uptake by SCCVII cells and macrophages... 34 2.4.2.1 The effect of plasma proteins on uptake and retention of Photofrin, by SCCVII cells and macrophages ..36 2.4.3 Photofrin uptake by peritoneal macrophages is inhibited by cytochalasin-B ......37 2.4.4 The uptake of Photofrin by SCCVII tumor cells and tumor associated macrophages in vivo ..........41 2.4.4.1 The effect of separation method on Photofrin content in SCCVIItumorcells and TAMs 46 2.4.4.2 The effect of enzymatic digestion on Photofrin content in tumor cells and macrophages 46 2.5 Discussion 47 CHAPTER 3. INDUCTION OF IMMUNE CELL INFILTRATION INTO MURINE SCCVII TUMOR BY PHOTOFRIN BASED PHOTODYNAMIC THERAPY..... .......5 6 3.1 Summary. ..56 3.2 Introduction. ". 57 3.3 Materials and methods 59 3.3.1 Tumor model.... 59 3.3.2 Photodynamic therapy 59 3.3.3 Indirect immunoperoxidase and Wright staining 60 3.3.4 Identification of blood derived tumor infiltrating cells .61 3.3.5 Macrophage cytotoxicity against tumor cells .....63 3.4 Results.... 65 3.4.1 Changes in tumor cellular content 65 3.4.2 Hoechst 33342 staining of circulating leukocytes. 70 3.4.3 Cytotoxicity of TAMs after PDT .75 3.5 Discussion 77 CHAPTER 4 . POTENTIATION OF PHOTODYNAMIC THERAPY BY IMMUNOTHERAPY THE EFFECT OF SCHYZOPHILLAN (SPG) .80 4.1 Summary 80 4.2 Introduction: ...r. .....80 4.3 Materials and methods .82 4.3.1 Tumor model, drugs......... -82 4.3.2 Analysis of immune cell infiltration ......83 4.3.3 Determination of Photofrin levels in tumor tissue/cells 84 4.3.5 Clonogenic assay ........84 4.4 Results and discussion .......85 4.4.1 Photofrin tumour levels 85 4.4.2 Leukocyte infiltration of SCCVII tumor ......86 4.4.3 The effect of SPG on tumor cell clonogenicity after PDT ......89 4.4.4 The effect on tumor recurrence ...91 CHAPTER 5. POTENTIATION OF PHOTODYNAMIC THERAPY ELICITED ANTITUMOR RESPONSE BY LOCALIZED TREATMENT WITH GRANULOCYTE-MACROPHAGE COLONY STIMULATING FACTOR ......9 4 5.1 Summary •• •••94 5.2 Introduction •.; 95 5.3 Materials and Methods.. 97 5.3.1 Tumor model, cell lines 97 5.3.2 Generation of GM-CSF retroviral vector...:..... 98 5.3.3 Virus production and infection of SCCVII tumor cells ........99 5.3.4 Northern blot analysis 100 5.3.5 GM-CSF assay 100 5.3.6 Photodynamic treatment 101 5.3.7 GM-CSF immunotherapy ......102 5.3.8 Macrophage cytotoxicity 103 5.4 Results 104 5.4.1 GM-CSF production by SCCVII/JzGM-CSF cells. ............104 5.4.2 Blood leukocyte content in GM-CSF treated mice ..........107 5.4.3 The effect of GM-CSF treatment on growth of SCCVII tumor..;....108 5.4.4 The effect of GM-CSF treatment on cytotoxic activity of TAM 111 5.4.5 The effect of GM-CSF treatment on the response of tumors to PDT ........112 5.5 Discussion.... .'. .116 CHAPTER 6. CONCLUSIONS AND FUTURE DIRECTIONS...........120 6.1 Conclusions..... ..........120 6.2 Future directions ..........123 CHAPTER 7. REFERENCES.. .........125 LIST OF TABLES Table 1.1 Photophysical and photochemical reactions of photosensitization.. 11 Table 2.1 Effect of presence of non-adherent cells from peritoneal exudate on the uptake of [14C]-polyhematoporphyrin by murine peritoneal macrophages 34 Table 2.2 Effects of cytochalasin B on in vivo uptake of Photofrin by peritoneal macrophagesa 38 Table 2.3 Photofrin content in tumor cells fraction and TAM fraction, derived from SCCVII tumor and separated by differential substrate attachment. „ 46 Table 2.4 The effect of exposure to a mixture of trypsin, collagenase, and DNAase on Photofrin levels in tumor cells and macrophages 47 Table 4.1 The effect of SPG on Photofrin content in SCCVII tumor.... 86 Table 4.2 The effect of SPG on Photofrin content in FcPr and FcR+ cell populations of SCCVII tumor 89 Table 4.3 The effect of SPG pretreatment on clonogenicity of SCCVII tumor cells isolated from PDT-treated subcutaneously growing tumors 90 Table 5.1 Blood leukocyte content in GM-CSF treated mice.. .107 ix LIST OF FIGURES Figure 2.1 Photofrin uptake by peritoneal macrophages and S C C V I I tumor cells in vitro .32 Figure 2.2 Uptake of [14C]-polyhematoporphyrin by peritoneal macrophages and S C C V U tumor cells in vitro... .....33 Figure 2.3 Effect of serum on Photofrin uptake by peritoneal . macrophages in vitro... 35 Figure 2.4 Effects of plasma proteins on uptake and retention of Photofrin by peritoneal macrophages in vitro..... 37 Figure 2.5 Retention of Photofrin in macrophages and tumor cells under in vitro conditions. 40 Figure 2.7 Photofrin fluorescence intensity in tumor cells (FcR") and host cells (FcR+) 41 Figure 2.8 Photofrin content in two populations of S C C V I I tumor cells 43 Figure 2.9 The ratios of Photofrin content determined in FcR+ : FcR' cell populations derived from SCCVII tumors. 45 Figure 3.1 Levels of major cellular populations in S C C V I I tumor before and after PDT ; 66 Figure 3.2 S C C V I I tumor contents of neutrophils and mast cells 69 Figure 3.3 Distribution of Hoechst 33342 fluorescence in representative S C C V I I tumors determined by flow cytometry. 72 Figure 3.4 . Infiltration induced by PDT of Hoechst 33342 labeled cells from the blood into the SCCVII tumor....;. ....73 Figure 3.5 The content of various populations among tumor resident cells (Hoechst 33342 negative) and newly infiltrated cells (Hoechst 33342 positive) in SCCVII tumors at 2 hours post PDT .........74 Figure 3.6 Cytotoxicity of tumor associated macrophages (TAMs), selected from PDT treated and non-treated SCCVII tumors, against in vitro cultivated SCCVII cells 76 Figure 4.1 The effect of SPG therapy on immune cell content of PDT non-treated (a) and PDT-treated (b) SCCVII tumor 88 Figure 4.2 The effect of SPG treatment on tumor eradication by PDT 92 Figure 5.1 Northern analysis of the total RNA isolated from SCCVII/JzGM-CSF and SCCVII/Jzneo cells.... ....,„.....105 Figure 5.2 GM-CSF bioassay.... .106 Figure 5.3 The effect of GM-CSF treatment on growth of established SCCVII tumor 109 Figure 5.4 Effect of GM-CSF treatment on inoculated SCCVIItumor ..110 Figure 5.5 The effect of tumor localized GM-CSF treatment on cytotoxic activity of TAM 112 Figure 5.6 The effect of localized GM-CSF treatment on the control of SCCVII tumor by Photofrin mediated PDT. ...114 Figure 5.7 The effect of localized GM-CSF immunotherapy on the control of SCCVII tumor by BPD mediated PDT... ...115 ABBREVIATIONS ALA 5-aminolevulinic acid AlCISPc Chloraluminum sulfonated phthalocyanine AlPcS Aluminum phthalocyanine BGG Bacillus of Calmette and Guerin BPD Benzporphyrin derivative BSA Bovine serum albumin cDNA Complementary deoxyribonucleic acid CFU Colony forming unit CP Corynebacterium parvum DNA Deoxyribonucleic acid E. coli Escherichia coli EDTA Ethylenediaminetetraacetic acid FACS Fluorescence activated cell sorter FBS Fetal bovine serum FcR Fc receptor FITC Fluorescin isothyocyanate GM-CSF Granulocyte-macrophage colony stimulating factor HBSS Hank's balanced salt solution HDL High density proteins IgG imunoglobulin gamma IL Interleukin LDL Low density proteins LTR Long terminal repeat m-THPC m-tetrahydroxyphenylchlorin MEM Minimum.essential medium mRNA Messenger ribonucleic acid NaCI Sodium chloride NaH2PC"4 Sodium phosphate, monobasic NCS Newborn calf serum NPe6 Chlorin e6 PBS Phosphate buffered saline PDT Photodynamic therapy PE Phycoerythrin PMN Polymorphonuclear leukocytes RNA Ribonucleic acid ' SD Standard deviation SDS Sodium dodecyl sulphate SF Surviving fraction xii Schizophillan Tumor associated macrophage T cell receptor Tumor necrosis factor-a Tetraphenylporphine XIII Publications containing material covered in this thesis 1. Korbelik, M., Krosl, G., Olive, P.L, Chaplin, D.J, (1991) Distribution of Photofrin between tumour cells and tumour associated macrophages. Br. J. Cancer 64, 508-512. 2. Korbelik, M., Krosl, G., Chaplin, D.J. (1991) Photofrin uptake by murine macrophages. Cancer Res. 51, 2251-2255. 3. Krosl, G., Korbelik, M. (1994) Potentiation of photodynamic therapy by immunotherapy: the effect of schizophyllan (SPG). Cancer Lett. 84, 43-49. 4. Krosl, G., Korbelik, M., Dougherty, G.J. (1995) Induction of immune cell infiltration into murine S C C V I I tumour by photofrin-based photodynamic therapy. Br. J. CancerT\, 549-555. 5. Krosl, G., Korbelik, M., Krosl, J., Dougherty, G.J. (1996) Potentiation of photodynamic therapy elicited antitumor response by localized treatment with granulocyte-macrophage colony stimulating factor. Cancer Res. 56: 3281-3286. xiv ACKNOWLEDGMENTS I am very grateful to my supervisor Dr. Mladen Korbelik for his unlimited support, constant availability, and excellent guidance throughout this project. I would also like to thank Dr. Graeme Dougherty for his advice and constructive criticism that helped me at the critical points during my work. I owe special thanks to Dr. David Chaplin for encouragement and helpful advice. I am grateful to Dr. Branko Palcic for his support and the interest he always showed for my work. I would like to thank the members of my supervisory committee for advice and helpful discussions during our meetings that helped to formulate sharp and focused questions to be answered by this project. In addition I would like to thank the staff of Cancer Imaging, Terry Fox Laboratory and Medical Biophysics for their companionship and assistance throughout the years I spent at the Cancer Agency. I would like to acknowledge expert advice I received from Dr. Ralph Durand. My thanks also to Dr. Andrew Minchinton for unlimited use of the laboratory and equipment. I would also like to thank Sandra Lynde for the technical and friendly support that she had given me in the laboratory. Nancy LePard and Denise McDougal were very helpful in the experiments involving flow cytometry analysis. Last, but not least it was Jana's love, patience and understanding that supported and helped me to carry this work through. The work on this thesis was supported by the Research Golincil of Slovenia, British Columbia Health Research Foundation, and Medical Research Council of Canada. Introduction 1 CHAPTER 1 Introduction 1.1 Overview and thesis outline Photodynamic therapy (PDT) requires three simultaneously present components for its cytotoxicity: 1)- a light-sensitive drug (photosensitizer); 2) light of an appropriate wavelength, consistent with absorption characteristics of photosensitizer; and 3) molecular oxygen. The therapy involves a sequential administration of photosensitizer and light for the treatment of diseased tissue: The mechanism of cytotoxicity involves the conversion of molecular oxygen to an excited form called singlet oxygen by energy transfer from the excited photosensitizer to the oxygen. This form of extremely reactive oxygen is believed to be responsible for most of the PDT-mediated cytotoxicity. The ideal drug for PDT of solid cancers would have to satisfy five conditions : 1) no dark toxicity; 2) selective uptake and/or retention by tumor; 3) high production of cytotoxic oxygen species; 4) excitation by light of sufficient tissue penetration; and 5) photostability (Pass 1993). However, such a photosensitizer does not exist at this time and intense research is underway for a drug that would satisfy most of the listed conditions. Accumulation of photosensitizers in tumor tissue is an important determinant for selective destruction of malignant lesions by PDT. Photofrin®, an improved and purified version of hematoporphyrin derivative (HPD) is the most widely used and studied photosensitizer. Much research has been done on its pharmacokinetics and factors that influence its uptake and retention in malignant tumors, but a number of these factors still remain to be elucidated. As reviewed, Photofrin accumulates at high levels in tumors, at the sites that have Introduction 2 high content of elements of the reticuloendothelial system, at sites of inflammation, and in traumatized tissue. The destruction of tumor cells by PDT. is believed to originate from singlet oxygen mediated oxidative damage to vital cellular components. However, as will be reviewed in this Chapter, only a fraction of tumor cells are killed by such "direct" phototoxic damage. It is known that PDT induces a cascade of events responsible for so called "secondary damage" characterized by vascular stasis, blood vessel leakage and hemorrhagic necrosis and can ultimately lead to tumor cure. In this process a variety of biological mediators are released from malignant and host cells in treated tumor which appear to have an important role in the development of tumor necrosis. Reports in the literature indicate that it is possible, by modulating the immune status of a tumor bearing animal, to influence the response of tumor to PDT. Based on these reports, we have hypothesized that the immune cells of a tumor bearing host play a significant role in the accumulation of Photofrin in tumor tissue and in the outcome of PDT. To test this hypothesis we first compared the in vitro uptake of Photofrin by the cells of an experimental murine tumor to that of murine macrophages. We also tested the influence of different serum proteins on Photofrin uptake and retention in malignant cells and macrophages cultured in vitro. We then proceeded to analyze the in vivo levels of the photosensitizer in malignant tumor cells and tumor associated macrophages (TAMs) at different time points after Photofrin administration. Next we tested the hypothesis that host immune cells are involved in tumor response to PDT. Since the reaction to PDT closely resembles the acute inflammatory reaction, we examined the possibility that PDT induces infiltration of circulating immune celis at the treated site. To study this, we employed flow Introduction 3 cytometry analysis and immunoperoxidase staining of cells that were recovered from the tumor site at different time points after PDT. To differentiate between the newly arrived and resident immune cells we selectively stained the cells in circulation using a supravital nuclear stain Hoechst 33342. The results have shown, that PDT indeed induces a massive and regulated infiltration of myeloid cells into the treated site. In the last part of the study we attempted to enhance the efficacy of PDT by modulating the immune system of the host. The results indicate that immunotherapy given in a proper sequence with PDT can significantly improve the rate of tumor cure. Taken together these results support our initial hypothesis that host cells have a significant role in the accumulation of Photofrin in tumors and that they also participate in the eradication of tumors treated by PDT. 1.2 Historical perspective PDT has developed into one of the clinically accepted therapies of malignant disease in the last 20 years. The idea of treating tumors with photosensitizers and light, however, is much older. It was already known at the beginning of this century, that a combination of light and a topically applied aqueous solution of eosin results in regression of some cutaneous tumors (Tappenier and Jesionek 1903). The first observation of porphyrin fluorescence in human and animal tumors was reported by Policard (Policard 1924) who attributed the reddish fluorescence to endogenous porphyrins that accumulated after infection of neoplastic tissue by hemolytic bacteria. Selective retention of hematoporphyrin (HP) in primary and metastatic tumors was first observed in 1942 by Auler and Banzer (Auler and Banzer 1942), who observed HP fluorescence in the lymph nodes. Later it was noted that hematoporphyrins and Introduction 4 metallohemafoporhyrins are selectively retained in neoplastic, embryonic and regenerating tissues in rodents (Figge et al. 1948). In 1960, Lipson and coworkers (Lipson and Baldes 1960) improved upon the tumor-localizing properties of HP by treating it with acetic and sulfuric acid. They named the improved drug hematoporphyrin derivative (HPD) and demonstrated that it is selectively retained by malignant and proliferating tissues. A year later they performed the first endoscopic determination of malignant lesions by detecting HPD fluorescence in the respiratory and upper digestive tract (Lipson et al. 1961). Lipson was also the first to actually treat a patient for metastatic chest wall breast cancer with PDT (Lipson et al. 1966). Although the tumor was not cured, there was objective evidence of a cytotoxic effect of PDT (Lipson et al. 1966). Kelly et al. reported the destruction of recurrent bladder cancer by HPD and light delivered through fiber optics (Kelly et al. 1975). The first phase l/ll clinical trials were reported in 1977 and 1979 by Dougherty et al. (Dougherty et al. 1977; Dougherty et al. 1979), who also identified the tumor localizing fraction of HPD (Dougherty et al. 1984) which was characterized as dihematoporphyrin ether/ester (DHE) (Kessel 1986). Dougherty et al. also explored the mechanism and applications of HPD-mediated PDT (Dougherty 1984; Dougherty 1987) for treatment of diverse animal and human malignancies. 1.3 PDT in clinical trials. Since these first trials, PDT has developed into a promising modality forthe treatment of malignant lesions. It has the potential to be used both as a primary and palliative treatment for a variety of malignant and non-malignant conditions. PDT has now received regulatory approval for use in cancer treatment in Canada, Japan, and some European countries. Clinically, the most utilized drug is Photofrin® (porfimer sodium), an improved and purified preparation of. Introduction 5 the original HPD, however, a new generation of photosensitizers is being intensely tested. In addition to Photofrin, several other photosensitizers have entered clinical studies. Protoporphyrin IX, derived from its precursor 8-aminolaevulonic acid (ALA), is being studied for the treatment of superficial basal cell carcinomas (Kennedy et al. 1990; Cairnduff et al. 1994). Benzoporphyrin derivative mono acid (BPD) (Richter et al. 1991) is being studied for the treatment of cutaneous tumors and several non-malignant conditions. (Lui et al. 1993; Lui 1994) . The aspartic acid derivative of chlorin e6 (NPe6) (Volz and Allen 1992) and m-tetrahydroxyphenylchlorin (m-THPC), which has been noted for the low light and drug doses needed to achieve the tumor necrosis, is also being investigated in the treatment of cutaneous malignancies (Ris et al. 1992). Controlled phase III clinical trials using Photofrin mediated PDT for bladder, lung, and esophageal carcinomas were initiated in late 1980s. Photofrin based PDT for bladder carcinoma is now approved for use in Canada, while Photofrin PDT for esophageal, endobronchial, stomach, and cervical carcinoma is approved for use in Japan. In addition to the recent FDA approval for PDT treatment of esophageal cancer in USA, this modality has been regulatory approved in a number of European countries. Evidence from clinical studies indicate that PDT can have a beneficial role in the management of many other malignant conditions, including tumors of the bladder, bronchus, esophagus, skin, head and neck, cervix and colorectal cancers, metastatic breast carcinoma and ocular tumors (Dougherty 1989; Pass 1993). This review will outline only a few of these studies. A large number of lung cancer patients have received PDT in many different centres. The most promising use of PDT appears to be for early stage lung cancer (reviewed by Lam (Lam 1994)). In a Phase ll-lil clinical trial in Japan, Introduction 6 89% of patients with early stage lung cancer showed complete response (Furuse et al. 1993). Studies at Mayo Clinic (Cortese et al. 1992; Eddel and Cprtese 1992) showed that PDT may be an alternative to surgery in patients with early stage lung cancer. In a study at the British Columbia Cancer Agency, a group of patients with mostly larger tumors was treated with PDT (Lam 1994). The results suggested that larger tumors, especially bronchial stump tumors are difficult to treat with PDT because of difficulties in light delivery. PDT in combination with external radiotherapy may also produce very good results in palliative therapy in patients with obstructive bronchial cancers (Lam 1994). PDT for cancer of the esophagus was shown to be potentially curative in patients with early tumors of either squamous or glandular histology (Tian et al. 1985; Marcon 1994). In a study by Hier et al. (Hier et al. 1993), it was shown that palliative PDT for obstructive esophageal cancers compares favorably with traditional Nd-YAG thermal ablation. PDT appears to have an added advantage of less frequent treatment and higher efficacy (Marcon 1994). PDT utilizing protoporphyrin IX which is synthesized in situ from either topically or systemically administered ALA, has been developed by Kennedy et al. (Kennedy et al. 1990). They obtained encouraging results with ALA cream applied to superficial basal and squamous cell carcinomas subsequently illuminated with 600-800 nm light (Kennedy and Pottier 1992). Similarly good results were obtained by Cairnduff et al. (Cairnduff et al. 1994) for Bowen's disease. BPD, a second generation photosensitizer has an absorption peak at 690 nm, which results in better light penetration. It has faster tissue clearance compared to Photofrin (Richter et al. 1991). In the studies of Lui et al. (Lui et al. 1993; Lui 1994) better results were obtained in treatment of cutaneous lesions with BPD-mediated PDT compared to Photofrin-mediated PDT. In addition to Introduction 7 malignant lesions, BPD-mediated PDT has shown to induce a positive response in the treatment of psoriasis (Hruza et al. 1993; Lui 1994) human papilloma virus infections (Abramson et al. 1992; Abramson et al. 1994), and vascular malformations (Orenstein et al. 1990). The aspartic acid derivative of chlorin e6 (NPe6) (Nelson et al. 1987) is also being studied in clinical trials with cutaneous lesions. Preliminary results indicate very good responses with minimal adverse reactions (Volz and Allen 1992). Bonnett and Berenboum first described m-tetrahydroxyphenylehlorin (m-THPC) (Bonnett and Berenbaum 1989). In a study by Ris et al. (Ris et al. 1992), the authors noted a good sparing of normal tissues and favorable tumor responses. Of special interest is the application of PDT in combination with surgery for peritoneal seeding of ovarian tumors (DeLaney et al. 1993) and diffuse intra-thoracic cancers including malignant mesothelioma, or AIDS-related Kaposi's sarcoma (Dougherty et al. 1992). High dose chemotherapy with autologous bone marrow transplantation can be an effective strategy for the treatment of some malignant disorders. Unfortunately, autologous grafts may be contaminated with malignant cells. While there are several techniques available for purging of bone marrow, PDT is an interesting possibility. Both, Photofrin and BPD have been extensively studied and show promise in this regard (Mulroney et al. 1994). The great advantage of this new modality is that it can be repeated many times since there was no dose-tolerance build-up noted so far. Patients with recurrence after chemotherapy, radiation, or immunotherapy can still be effectively treated by PDT. The only side.effect for Photofrin-mediated PDT is cutaneous photosensitivity, which may last for 4 weeks. New generation Introduction 8 photosensitizers, like BPD, which have a faster skin clearance compared to Photofrin, may not cause prolonged skin photosensitivity. 1.4 Tumor localization of photosensitizers Hematoporphyrin derivative, the first established photosensitizer, was initially known as a tumor; localizing agent. Its in situ fluorescence is currently being employed for detection and delineation of various malignancies with newly developed fluorescence imaging and non-imaging devices (Lam et al. 1990; Andersson-Engels et al. 1992; Lam et al. 1993). This tumor localizing property of photosensitizers is also important for selective destruction of tumors by PDT. However, even in early studies, it was noted that porphyrins accumulate at high levels in the lymph nodes (Auler and Banzer 1942) and healing wounds (Figge et al. 1948). Several studies with Photofrin and aluminum phthalocyanine sulfonate (AlPcS) have shown that considerable levels of both types of photosensitizers are retained in the organs that are rich in reticuloendothelial cell components (Gomer and Dougherty 1979; Kessel 1986; Chan et al. 1988; Henderson and Bellnier 1989). It is now known that many photosensitizers are retained at high levels in the liver, kidney, lymphatic organs, inflammatory sites, and atherosclerotic plaques (Spears et al. 1983; Nseyo et al. 1985; Bellnier and Henderson 1992). In addition, the tumor vs normal tissue (muscle, skin) ratio of most photosensitizers, including Photofrin, BPD or phthalocyanines is rarely above 2:1 (Bellnier and Dougherty 1989; Tarlau et al. 1990; Bellnier and Henderson 1992; Ma et al. 1992). Several hypothesis have been put forward to explain the retention of photosensitizers within tumors. Changes in cellular and tissue pH have been proposed to have an effect on Photofrin biodistribution (Pottier and Kennedy 1990). Repeated glucose administration, which induced a transient decrease of Introduction 9 tumor pH in experimental animals, resulted in increased retention of Photofrin and enhanced effect of PDT (Thomas et al. 1987; Peng et al. 1991) and seem to support this hypothesis. The role of plasma lipoproteins and albumin in the delivery of photosensitizers to the malignant lesion and their uptake by the cells comprising the tumor has also been proposed (Kessel 1986). Cancer cells may have augmented expression of the low-density lipoprotein (LDL) receptor expression, and since lipoproteins are major carriers of lipophilic porphyrins in the bloodstream (Jori et al. 1984), photosensitizer/LDL complex uptake by LDL receptors was proposed for porphyrin localization in tumors. However, not all photosensitizers which bind to serum LDL are good tumor localizers; an example is protoporphyrin (Kongshaug et al. 1989). The increased permeability of tumor vasculature has also been proposed to play a role in sensitizer accumulation (Bugelski et al. 1981). However, the increased interstitial pressure may also affect the extravasation of photosensitizers (Jain 1989). In this respect, work in our laboratory (Korbelik and Krosl 1994) has shown how proximity to the blood supply affects the levels of different photosensitizers in the cells of a murine fibrosarcoma. It was demonstrated that there are marked differences among photosensitizers in their capacities to reach tumor cells distant from the vasculature. The lack of difference with regard to diffusion potential between di- and tetrasulphonated forms of tetraphenylporphine ( T P P S 2 and T P P S 4 , respectively), and disulphonated and tetrasulphonated derivatives of aluminum phthalocyanine ( A I P C S 2 and A I P C S 4 , respectively) implied that the lipophilicity of the sensitizers was not a decisive factor, nor were their affinities for binding to plasma proteins. It is possible that the factor that determines the distribution of photosensitizers within the tumor is their affinity for cellular or other structures within the tumor. Peng et al. (Peng et al. 1991) have studied the distribution of several photosensitizers with different Introduction 10 hydrophobicity within a human tumor growing in nude mice. They found that. A I P C S 4 and T P P S 4 accumulate mainly in acellular tumor stroma. Mono- and disulfonated forms of these photosensitizers were found retained more selectively in cells than in acellular tissue components. Since the study of Bugelski et al. (Bugelski et al. 1981), it has been known that elevated Photofrin levels accumulate in tumor associated, macrophages (TAMs). Jori (Jori 1989) suggested that porphyrins, phthalocyanines, and other photosensitizers that can remain in highly aggregated form in vivo are phagocytized to a large degree by TAMs. Following the initial studies described in Chapter 2, the distribution of various photosensitizers between malignant and host cell populations of experimental tumors has been extensively studied by our group (Korbelik and Krosl 1995; Korbelik and Krosl 1995; Korbelik and Krosl 1995; Korbelik and Krosl 1996). It was shown that there is a marked heterogeneity in photosensitizer levels between the different cell populations that comprise a tumor. The consistent observation in all tumor models examined was that by far the highest cellular levels of photosensitizers are accumulated in a subpopulation of TAMs which is characterized by a high granularity and highly elevated expression of interleukin (IL)-2 receptor, indicating that activated TAMs may be responsible for accumulation of significant portion of retained photosensitizer (Korbelik and Krosl 1995). 1.5 Mechanism of action 1.5.1 Photosensitization Photosensitization refers to a light-activated process in which the presence of a chromophore leads to changes in a substrate that otherwise would not take place (Grossweiner 1989). Porphyrins and photosensitizers in general have a common mechanism of action. The major pathways by which an excited Introduction 11 photosensitizer can transfer energy are depicted in Table 1.1. After absorbing light of appropriate wavelength the photosensitizer is converted from stable ground state (P) to a short lived excited state (1P*) also known as the singlet state. The excited molecule may then undergo conversion to a longer-lived excited state known as the triplet state (3P*). This triplet state may become involved in forming cytotoxic species or may release energy by other processes including fluorescence decay (Grossweiner 1994). Table 1.1 Photophysical and photochemical reactions of photosensitization P + hv -> 1P* light absorption 1P* -> 3P* intersystem crossing 1P* -> hv' + P fluorescence 3P* + 0 2 -> w + P energy transfer 3P* '.-> P non-radioactive decay 10 2* + S -> S(O) reaction mediated by singlet oxygen 3P* + S -> p- + S(O) Type I reaction P- + o2 -> p + O 2 -Where: P = Photosensitizer in ground state hv = . Light quantum 1P* = Excited singlet state of photosensitizer 3P* = Excited triplet state of photosensitizer 0 2 = Oxygen in the ground (triplet) state 1 o 2* = Excited singlet state of oxygen S = Tissue target (substrate) S(O) = Oxidized tissue target . Modified from: Dougherty, T.J. (1989) Photodynamic therapy: Status and potential. Oncology 3, 67-78. The pathways of photosensitization in biological systems are divided into two broad categories (Foote 1984). Type I process refers to a mechanism where excited photosensitizer reacts directly with a substrate involving hydrogen or electron transfer to yield transient radicals. This process can take place under anoxic conditions. In the alternative Type II mechanism, the Introduction . ' 12 energy transfer from the excited photosensitizer in triplet state to molecular oxygen takes place, leading to singlet oxygen (1C*2*) which further reacts with substrates susceptible to oxidation (S(O)). Photodynamic processes are; believed to take place via a Type II pathway, since when oxygen is not present, or is present at levels less than 2%, cells are resistant to PDT (Lee See et al. 1984; Mitchell et al. 1985). This is supported by the fact that singlet oxygeawas observed in vivo during the PDT of an experimental mouse tumor (Parker 1986) although these results seem to be in dispute. Further evidence for oxygen requirements for PDT was provided by Tromberg etal. (Tromberg et al. 1990) who observed that (presumably due to its consumption) oxygen tension in a tumor becomes irreversibly low with high. PDT fluences. Foster and coworkers (Foster et al. 1991) suggested that fractionating light delivery may partially overcome the oxygen depletion at high fluences. It should be noted that the possibility of Type I reaction occuring in PDT under certain conditions cannot be ruled out, For example, this type of reaction may be favoured in systems where the sensitizer is bound to an easily oxidized biomolecule, (Foote 1984). It was also suggested that a competition between photo-oxidation and photo-reduction of porphyrin photosensitizers can take place, as evidenced by formation of a chlorin-type molecule following photoirradiation (Rotomskis et al. 1996). 1.5.2 Cellular effects Singlet oxygen lifetime and diffusion distance in a cellular environment are extremely limited by its avid reactivity with cell constituents. It was estimated that the diffusion distance of 1C>2* in cells is about 0.1 u.m (Moan and Berg 1991). The cell damage will therefore occur very close to the site of 10*2* generation; hence the intracellular localization of photosensitizer will dictate the Introduction 13 type of cellular damage that will occur during PDT. For anionic and lipophilic photosensitizers (like Photofrin), major target are cellular membranous structures. PDT induced damage to plasma membrane is indicated by changes in cellular shape (swelling, formation of blebs) (Moan et al. 1982), changes to microvilli as seen by electron microscopy (Moan et al. 1982; Yanhua et al. 1990), and depolarization of the plasma membrane (Specht and Rodgers 1990). Candide et al. observed PDT induced alteration of endoplasmic reticulum associated enzymes (Candide et al. 1989). Hilf and coworkers reported that cytosolic enzymes were much less affected by PDT compared to enzymes associated with the mitochondrial membrane (Hilf et al. 1984). Mitochondrial enzymes that are impaired the most are those involved in electron transport, oxidative phosphorylation, and ATP/ADP and phosphate/H+ translocation (Hilf et al. 1984; Hilf et al. 1984; Atlante et.al. 1989; Atlante et al. 1990; Yamamoto and Kawanishi 1991). These latter changes result in a decreased rate of oxidative phosphorylation and decreased cellular ATP levels (Hilf et al. 1986). In addition to the damage to cellular membranous structures described above, Berg and coworkers observed that cells exposed to PDT using porphyrin and phthalocyanine photosensitizers accumulate in mitosis (Berg et al. 1990; Berg and Moan 1992). In the case of T P P S 4 the non-polymerized form of microtubules was identified as a. target, which prevented the organization of spindle apparatus. Depolymerization of microtubules induced by PDT in vitro in human endothelial cells was also observed (Sporn and Foster 1992). This damage results in disruption of plasma membrane and cytoskeleton which causes the endothelial cells to contract, exposing the underlying basement membrane (Fingar et al. 1993). This effect on endothelial cells is probably one of the important factors in the development of the vascular response to PDT discussed later in this chapter. Introduction 14 Hydrophilic photosensitizers seem to accumulate preferentially in lysosomes (Moan et al. 1989; Specht and Rodgers 1990). Damage to these . structures with ensuing release of proteolytic enzymes is probably a major cause of PDT induced cell death with NPes and T P P S 4 (Moan et al. 1989; Roberts and Berns 1989; Roberts et al. 1989). Mitochondria are most likely the major target for cationic photosensitizers and protoporphyrin IX, which seem to preferentially localize in these organelles (Salet and Moreno 1990). Preferential intracellular localization of lipophylic photosensitizers leads to photoperoxidation of membrane cholesterol and other unsaturated phospholipids (Henderson and Dougherty 1992). Cell leakage, which ultimately leads to loss of cell integrity, was observed to start almost immediately after the onset of light treatment (Henderson and Donovan 1989). The damage to plasma membrane is accompanied by the release of a variety of substances including metabolites of arachidonic acid, and lipid fragments. Metabolites of arachidonic acid, produced via cyclooxygenase or lipoxygenase pathway (Samuelsson 1982), were detected after PDT in vitro (Henderson and Donovan 1989; Henderson et al. 1992) or in vivo (Fingar et al. 1990; Fingar et al. 1991). 1.5.3 The effect of PDT on tumor The development of macroscopic and microscopic tissue damage following PDT has been studied with several photosensitizers, including Photofrin (Reed et al. 1989; Reed et al. 1989; Reed et al. 1989), phthalocyanines and NPe6 (Nelson eta l . 1988), and purpurins (Morgan et al. 1990). The general phenomenon accompanying PDT seems to be vessel occlusion. The time interval between the treatment and this vascular response varies with different photosensitizers and from tumor to tumor (Henderson and Dougherty 1992). Introduction 15 Clumps of red blood cells can be seen in the tumor vasculature almost immediately after the onset of light treatment, which is followed by transient vascular dilatation, constriction and eventual blood stasis leading into hemorrhage (Fingar et al. 1993). The gross edema that accompanies these vascular effects is usually seen immediately after the completion of light treatment and increases for some time post PDT (Lim et al. 1986; Kerdel et al. 1987). Once the vessel occlusion is complete it is irreversible and is followed by ischemic tissue necrosis (Henderson and Fingar 1987). Tumor disappearance usually occurs within 1-2 days after PDT (Pass 1993). The importance of vascular occlusion events has been demonstrated in studies by Henderson et al. (Henderson et al. 1985) and Star and coworkers (Star et al. 1986). Two major determinants contribute to PDT induced destruction of tumors. The first is the degree of tumor cell destruction by the cytotoxic action of excited photosensitizer, and the second is vascular photosensitization that leads to vascular collapse and general nutrient deprivation and tumor hypoxia (Henderson and Dougherty 1992), The relative contribution of each these two factors depends on the type of photosensitizer used. For Photofrin, where vascular shutdown commences immediately after the onset of light treatment, the direct cell kill is typically responsible for approximately 20% of tumor cell death (Henderson et al. 1985; Krosl and Korbelik 1994). However, further cell death proceeds immediately, and its kinetics is similar to that caused by anoxia (Henderson et al. 1985). Monosulfonated aluminum phthalocyanine (AlPcSi) was reported to kill 10 times more tumor cells than Photofrin by direct phototoxic action while secondary damage proceeds immediately due to the shutdown of the tumor vasculature. With bacteriochlorophyll-a, the direct cell kill is limited by its rapid photobleaching (Henderson et al. 1991), and reduces the clonogenic Introduction 16 fraction of tumor cells for approximately 50%. However, additional cell death does not commence until 2-3 h later, when the shutdown of tumor vasculature occurs. The vascular effects described above are probably triggered by several mechanisms. One is the direct photodynamic damage to endothelial cells. This results in damage to their cytoskeleton (Sporn and Foster 1992) and contraction and rounding which exposes the underlying basement membrane. The contact between the exposed basement membrane and platelets triggers the clotting cascade and results in the formation of thrombus and is accompanied by the action of vasoactive molecules that were identified to be released during and after PDT (Fingar et al. 1993). Another mechanism most likely involves PDT triggered release of vasoactive substances from the cells comprising the tumor. Administration of indomethacin, which is an inhibitor of cyclooxygenase, was shown to block many of the observed vascular effects (Fingar et al. 1990; Fingar et al, 1992; Hampton and Selman 1992). Fingar et al. also tested several specific thromboxane inhibitors and observed that the tumor response inversely correlated with the inhibition of vessel stasis (Fingar et al. 1993). The interesting observation made by Fingar and coworkers in this study was that some initial tumor response was seen even when the synthesis of thromboxane was completely blocked. This is not the case when indomethacin is used, which blocks the production of both prostaglandins and thromboxanes. These results have suggested that thromboxane plays an important role in the vascular effects of PDT, but is not the only mediator responsible for overall tumor response to PDT (Fingar et al. 1992). Another factor that leads to vascular damage is the inflammatory reaction induced by PDT in treated tumors. Introduction 17 1.5.4 PDT triggered inflammatory response The relationship between PDT and the host response was first proposed by Elmets at al. (Elmets and Bowen 1986) who described immunological suppression observed in mice that underwent PDT. Lynch and coworkers then established that PDT.induced suppression of contact hypersensitivity is induced by macrophages, and that it can be adoptively transferred to naive mice (Lynch et al. 1989; Lynch et al. 1989). The ability of spleen cells from PDT treated mice to act as stimulators in mixed lymphocyte response was significantly impaired (Lynch et al. 1989). It was suggested that this immunosuppression may be mediated by cyclooxygenase pathway metabolites of arachidonic acid, since it was ameliorated by a continuous low dose indomethacin treatment. Ferrario and Gomer (Ferrario and Gomer 1990) reported that PDT induces acute lethality in some mouse strains. The histological profiles of different organs from dead mice indicated that this effect was consistent with a traumatic shock syndrome. The protective action of anti-inflammatory drugs such as indomethacin, aspirin, and antihistamine suggested,that prostaglandins, thromboxanes and histamine are associated with the lethality observed, in PDT treated mice. As mentioned above, these inflammatory mediators were all observed to be released after PDT in vitro and in vivo. In addition to the pro-inflammatory metabolites of arachidonic acid, PDT induces the release of other biologically active molecules. PDT treatment causes degranulation of mast cells (Kerdel et al. 1987), and induces macrophages to release tumor necrosis factor (TNF-a) in a dose dependent fashion (Evans et al. 1990). TNF-a, IL-1p\ and IL-2 have been detected in the urine of patients with urinary bladder carcinoma treated with PDT (Nseyo et al. 1990). Kick and coworkers recently described findings that in vitro PDT of HeLa cells induces increased messenger RNA expression and protein synthesis of Introduction 18 pro-inflammatory cytokine IL-6 (Kick et al. 1995). The accumulation of T lymphocytes and plasma cells that has been observed subsequent to PDT of bladder carcinoma (Shumaker and Hetzel 1987) and increased adherence of polymorphonuclear leukocytes (PMN) to venules during PDT (Fingar et al. 1992) suggest that the host immune system responds immediately and strongly to the PDT. -With the observed edema and erythema that develop during and after the light irradiation, the host response to PDT resembles an acute inflammatory reaction. - , The importance of the host response to PDT in the control of the tumor can be recognized in the fact that the treatment with anti-inflammatory drugs reduces the effect of PDT (Fingar et al. 1993). While treatment with Flunarizine (inhibitor of thromboxane release from platelets) did inhibit the release of thromboxane from platelets, it did not prevent PDT induced vessel leakage. Fingar et al. (Fingar et al. 1993) hypothesized that accumulated PMN could release enough eicosanoids to induce such, vessel leakage. If the entire pathway for production of prostaglandins and thromboxanes is blocked by inhibitors then the tumor response to PDT is diminished more strongly than with thromboxane inhibitors alone. This suggests not only that the vascular response to the release of eicosanoids is important for the effect of PDT, but also that the role of the members of prostaglandin series in modulation of inflammatory response and immune modulation must not be ignored. The inflammatory reaction that develops subsequent to PDT may be stronger in tumor tissue than.in surrounding normal tissue. Star et al. (Star et al. 1986) observed that no hemorrhage developed in the blood vessels adjacent.to PDT treated tumors. In addition, the PDT treatment of normal tissue (rat ears) resulted in only slightly larger damage than that induced in tumors in observation chambers, despite the fact that Photofrin content in the normal Introduction 19 tissue was six times higher than in the observation chambers. No vascular occlusion or hemorrhage was observed after A I P C S 2 - and ALA-mediated PDT of.normal rat femoral arteries (Grant et al. 1994). The blood vessels in PDT field did loose the endothelial cell layer but remained patent. Even more striking was the complete absence of an inflammatory reaction (Grant etal. 1994). These observations imply that the PDT induced inflammatory response is much stronger in the tumor compared to normal tissue. One of the reasons for this effect could be the composition of malignant cell membranes, which contain alkylether derivatives of phospholipids, that are present in very low levels in the membranes of normal cells (Snyder and Wood 1969; Howard et al. 1972). The PDT induced activation of membrane-bound phospholipases (Agarwal et al. 1993) triggers the release of alkyl-lipid derivatives that were shown to be more potent macrophage stimulating agents than inflammation derived lipid metabolites from non-malignant cells (Yamamoto and Ngwenya 1987). The initial PDT damage to cellular membranes, which is not necessarily lethal, may trigger a cascade of events that lead.to tumor destruction (Yamamoto et al. 1992; Korbelik and Krosl 1994). We have shown that in vitro, PDT treatment of the human tumor cells A549 markedly enhanced their killing by macrophages. X-ray treatment of A549 cells did not similarly affect macrophage mediated killing of these cells. On the other hand, very limited macrophage toxicity was noted against PDT treated normal kidney cells. This suggests, that macrophages recognize PDT induced damage on PDT treated malignant cells (Korbelik and Krosl 1994). The activation of macrophages induced by PDT may be an additional factor in the "secondary effects." In addition to their enhanced cytotoxicity to PDT damaged tumor cells mentioned above, macrophages can act as antigen presenting cells. They can present putative tumor antigens to T lymphocytes Introduction 20 and induce the development of specific antitumor immunity. This possibility is supported by the observation by Canti et al. (Canti et al. 1994), that aluminum phthalocyanine mediated PDT induced in cured mice a specific immunity against subsequent challenge by the parental tumor. 1.5.5. PDT and immunotherapy The inflammatory reaction triggered by PDT offers perspectives for adjuvant treatment with immunomodulating agents. Myers and coworkers (Myers et al. 1989) have successfully combined PDT and immunotherapy with Corynebacterium parvum (CP) vaccine of the murine carcinoma MRT-2. They observed that the timing and the dose of immunotherapy administration is very important for the efficacy of combined therapy.. Their results suggest that the vaccination with high dose CP which resulted in acute inflammatory infiltration should be given after PDT. In contrast, low dose CP vaccination that elicits a slower inflammatory response with predominant T cell infiltration, enhanced the effect of PDT if it was given before the light treatment (Myers et al. 1989). These results suggest that PDT can produce an environment where tumor specific immunity may develop. Similarly positive results were obtained in PDT of experimental murine bladder carcinoma combined with Bacillus Calmette and Guerin (BCG) vaccination (Cho et al. 1992). Dima et al. successfully combined PDT with endotoxin (Dima et al. 1994), and Bellnier (Bellnier 1991) showed that combination of PDT with TNF-a administration significantly enhances the tumor response to PDT. Positive initial results have also been obtained with PDT in combined treatment with IL-7 secreting tumor cells (Dougherty et al. 1992) and serum vitamin D 3 binding protein (Korbelik et al. 1995), It seems that due to its inflammatory character, PDT is a particularly good candidate for optimized adjuvant immunotherapy. Introduction 21 1.6 Specific aims of the thesis In summary, evidence available at the time when work on this thesis was initiated, implicated cells of the macrophage lineage in the localization of photosensitizers (including Photofrin) in solid tumors. In addition, the nature and development of the tumor response to PDT indicated the existence of an acute inflammatory response with involvement of polymorphonuclear and monocyte/macrophage cellular components. Several questions emerged at the beginning of my project: 1. What is the role of tumor associated macrophages in the localization of Photofrin in tumors? 2. Is the reaction that develops in tumor after PDT characterized by an inflammatory cell infiltration? If so, is this associated with functional ., •• activation (tumoricidal activity) of inflammatory cells? 3. If host cells are involved in tumor reaction to PDT, will immunotherapy designed to activate inflammatory cells enhance the PDT effect? The following specific aims were formulated to answer these questions: 1. To examine the uptake of Photofrin in malignant and macrophage populations contained in an experimental murine tumor, in order to determine if tumor-associated macrophages have increased Photofrin accumulation compared to malignant cell population. 2. To determine if Photofrin-mediated PDT of an experimental murine tumor is associated with inflammatory cell infiltration and to investigate the antitumor cytotoxic activity of macrophages recovered from the tumor site following PDT. Introduction 22 3. To determine if immunotherapy, designed to activate the tumor infiltrating and/or res ident immune c e l l s , pa r t i cu l a r l y t hose of monocyte/macrophage lineage, improves the curative potential of PDT. The work carried out to answer these questions is presented in the following chapters. Photofrin in murine macrophages CHAPTER 2 23 The role of tumor infiltrating macrophages in Photofrin accumulation in tumors1 2.1 Summary The uptake of Photofrin by mouse peritoneal macrophages was studied in vitro and compared to the levels of the drug in SCCVII tumor cells accumulated under the same conditions. We also studied the Photofrin levels in the cells isolated from murine SCCVII tumor from mice that had previously received the drug. The cellular content of Photofrin in peritoneal macrophages was measured either by a fluorometric assay or by using 14C-labeled drug. The level of Photofrin in the cells comprising the SCCVII tumor growing in C3H/HeN mice was determined using fluorescence activated cell sorter (FACS). In addition, the cells obtained from excised tumors were stained for the presence of Fc receptors using fluorescein isothiocyanate (FITC) labeled anti-mouse antibody which enabled us to discriminate between Fc receptor negative (i.e. malignant SCCVII cells) and Fc receptor positive (predominantly tumor associated macrophages [TAM]). The Photofrin content in malignant SCCVII cells and TAMs sorted by FACS was also determined by chemical extraction. The data from all experiments demonstrate that macrophages have a much greater capacity for Photofrin uptake than SCCVII cells. The Photofrin content at 24 hours after drug administration (25 mg/kg) measured 420 ± 90 ng/u.g (SD), 74 + 15 ng/u,g , and 15 + 2 ng/u,g of cell protein for peritoneal macrophages, 'The data presented in this chapter have been published in following papers: 1. Korbelik, M., Krosl, G., Olive, P.L., Chaplin, D.J. (1991) Distribution of Photofrin between tumour cells and tumour associated macrophages. Br. J. Cancer64, 508-512. 2. Korbelik, M., Krosl, G., Chaplin, D.J. (1991) Photofrin uptake by murine macrophages. Cancer Res. 51, 2251-2255. Photofrin in murine macrophages 24 TAMs, and SCCVII cells, respectively. Factors that influence macrophage activity also influenced the accumulation of the drug in these cells. Analysis of in vivo accumulation of Photofrin in SCCVII tumor cells and TAMs comprising the growing tumor showed consistently higher Photofrin levels in TAMs throughout the 96 hour observation period. In vivo accumulated material in TAMs, SCCVII cells and peritoneal macrophages showed very similar in vitro clearance rate. 2.2 Introduction One of the most important properties of Photofrin that makes it a clinically successful photosensitizer is its localization in tumors (Henderson 1989; Lin 1989). The effectiveness of tumor destruction by PDT is related not only to the tissue distribution of the administered photosensitizer, but also to its distribution within the tumor (Henderson and Bellnier 1989). In recent years there has been increasing evidence that Photofrin and other photosensitizers specifically accumulate in the macrophages which infiltrate tumors; One of the earliest observations was published by Bugelski and coworkers (Bugelski et al. 1981). They found high levels of HPD located in macrophages scattered throughout a mouse tumor. In addition, they observed that the normal tissues showing the most prominent accumulation of photosensitizers (liver, kidney, spleen, lymph nodes, and skin) are all characterized by the presence of cells of mononuclear phagocyte system, also called the reticuloendothelial system (Bugelski et al. 1981). Chan et al. (Chan et al. 1988) have used flow cytometric analysis and cell sorting to study the levels of chloraluminum sulfonated phthalocyanine (AlCISPc) in cellular fractions of a mouse colorectal carcinoma. Based on photosensitizer fluorescence, they separated populations of high and low photosensitizer Photofrin in murine macrophages 25 content from suspensions of tumor cells after in vivo drug administration. They identified macrophages among the cells with high drug content. Henderson and Bellnier (Henderson and Bellnier 1989) suggested that macrophages exhibit very high affinity for Photofrin and may contribute significantly to the overall effect of PDT. Consequently, Henderson and Donovan (Henderson and Donovan 1989) demonstrated PDT induced release of prostaglandin E 2 from mouse peritoneal macrophages treated with Photofrin based PDT in vitro. Evans et al. (Evans et al. 1990) showed increased TNF-a secretion by macrophages treated with PDT in vitro. In this work we have examined the uptake of Photofrin by murine peritoneal macrophages in vitro and in vivo. We have also compared the uptake and retention of Photofrin in macrophages and tumor cells. By using fluorescence activated cell sorting we studied photosensitizer levels in cells from excised, disaggregated tumors. Using a dual staining technique we identified a subset of cells that showed an increased uptake and retention of the photosensitizer. . 2.3 Materials and methods 2.3.1 Murine tumor model Female C3H/HeN mice, 9-11 weeks of age, were used in all experiments. The mice were kept in Joint Animal Facility and fed standard rodent chow. The maintenance and tumor implantation of SCCVII murine squamous cell carcinoma was performed as described in detail earlier (Chaplin et al. 1987). Intramuscularly growing SCCVII tumor was excised and thoroughly minced with two scalpels. For maintenance, the resulting tissue was repeatedly passed through 18 and 20 gauge needles, respectively, and resuspended in phosphate buffered saline (PBS) (Sigma Chemical Co., St. Louis, MO, USA) at a 1:4 (tissue : PBS, v/v) ratio. 100 uL of tissue suspension was then injected into the Rhotofrin in murine macrophages 26 thigh muscles of an anaesthetized mouse. For tumor implantation, minced tumor was enzymatically dissociated into single cell suspension using the following enzyme cocktail: 0.02 % Collagenase I, 0.015% DNAse.l, (Sigma) and 0.125 % Trypsin (Difco Laboratories, Detroit, Ml, USA) in PBS, and incubated for 30 minutes at 37°C with gentle rotation. Approximately 200 mg of tumor tissue was digested in 5 mL of this cocktail. The resulting cell suspension was then filtered through a nylon mesh filter with pores of 80 |im and washed 2x with Eagle's minimum essential medium (MEM) (Gibco, Grand Island, NY, USA), supplemented with 10% fetal bovine serum (FBS) (Gibco). Cells were then counted and cell concentration adjusted to 1x107 cells/mL. Thirty \iL of this cell suspension was injected subcutaneously over the shaved sacral region of anaesthetized mouse. Tumors were used for experiments when they attained a size of 500-800 mg (wet weight). 2.3.2 Measurement of Photofrin uptake 2.3.2.1 Harvesting macrophages and Photofrin exposure in vitro Peritoneal macrophages were collected from 9-11 week old C3H/HeN mice. The mice received one intraperitoneal injection 0.5 mL of thyoglycollate broth (Difco Laboratories, Detroit, Ml, USA) at 3 days before macrophage collection. After the sacrifice, the peritoneum was washed with 5 mL of ice-cold (0°C) PBS. The lavages from three to five mice were pooled, cells were pelleted by centrifugation and resuspended in Eagle's MEM supplemented with 10% FBS (all by GIBCO) at a concentration of 1-2x106 cells/mL. The cells were transferred into 60 mL tissue culture dishes (Falcon, Becton Dickinson and Company, Lincoln Park; NJ, USA) and incubated for 2h in a complete growth medium at 37°C. Nonadherent cells were then washed away and Photofrin was added to the samples at given concentrations. The growth medium with Photofrin in murine macrophages 27 1% FBS was used during the exposure to Photofrin, unless stated otherwise. The exact number of cells per sample was determined by substracting the number of nonattached cells from the total number of nucleated cells originally plated. The macrophage preparations used in this study consisted of 90-95% macrophages as judged by the expression of nonspecific esterase activity. 2.3.2.2 Effect of plasma proteins on Photofrin accumulation and retention In the experiments testing the effects of individual plasma proteins, macrophages were exposed to Photofrin (40 u.g/ml_) in the presence of low density lipoproteins (LDL), high density lipoproteins (HDL) (both at 0.2 mg/mL), albumin (5 mg/mL), or in protein free medium. Before cell exposure, the Photofrin-protein mixture was first incubated for 30 min. at 37°C to allow for the optimal interaction between Photofrin and the proteins. For Photofrin clearance incubations, the macrophages were washed free of the drug and plasma proteins and further incubated in a complete medium for 24 hours. In all cell culture experiments, the pH was maintained at 7.2-7.4. To prevent photodestruction of Photofrin, all procedures were performed in subdued light and culture dishes were wrapped in aluminum foil. 2.3.2.3 Fluorescence analysis of tumor cells In all experiments Photofrin® (kindly provided by Quadralogic Phototherapeutics, Inc., Vancouver BC, Canada) was used as a photosensitizer. Mice received 25 mg/kg of Photofrin by injecting the appropriate volume of the original Photofrin solution (2.5 mg/mL) via the tail vein. At the indicated times after the Photofrin administration,, mice were sacrificed and tumors excised, weighed, and single cell suspensions prepared as described in section 2.3.1. The cell concentration was then adjusted to 1-2x10 7 cells/mL in MEM supplemented with 10% FBS. Fluorescein Photofrin in murine macrophages 28 isothiocyanate (FITC) goat anti-mouse IgG (whole molecule) (Sigma) was then added to the cell suspension at 1:200 (v/v) final dilution, as described previously (Olive 1989). The cells were then incubated at 37°C for 3-5 minutes and then washed 2x with ice cold MEM. This procedure was shown to effectively stain Fc receptor bearing (FcR+) cells in tumor cell suspensions. Anti-mouse IgG binds to either Fc receptor or to mouse antibody which is occupying the Fc receptor on the FcR+cell (Lindsay et al. 1982; Olive 1989). The cells were analyzed by a dual laser fluorescence activated cell sorter (FACS 440, Becton-Dickinson, Mountainview, CA, USA). Fc receptor positive (FcR+) and Fc receptor negative (FcR-) cells were simultaneously analyzed for the intensity of the Photofrin fluorescence. FITC was excited by a 488 nm laser and the emission was measured through a 530 ± 15 nm bandpass filter. Photofrin® was excited by UV lines (350-360 nm) and emission recorded through a 635 nm longpass filter. 1-2x104 cells were analyzed per sample in these measurements. In addition, 0.5-1.0x106 FcR+ and FcR- cells per sample were sorted for independent determination of Photofrin content using the chemical extraction of porphyrin material from the cells. 2.3.2.4 Photofrin measurement from cell extracts. Following centrifugation of the cells which were collected from the FACS as described in the previous section, 3 mL of ScintiGest tissue solubiliser (Fisher Scientific, Fair Lawn, NJ, USA) at 1:10 (v/v) aqueous dilution was added to the cell pellet. The hydrolysis of porphyrin material was facilitated by incubating the samples overnight at 60°C. As a result, all Photofrin components are converted into a more uniform material with optimal fluorescence yield (Brown and Vernon 1990; Korbelik and Hung 1991). Fluorescence of the cell extracts was measured by a System 3 Scanning Spectrofluorometer (Farrand Optical Co:, Photofrin in murine macrophages 29 Valhalla, NY, USA). The fluorescence intensity recorded was at the maxima for excitation and emission, at 405 and 625 nm, respectively. The actual Photofrin concentration was calculated using standard calibration curves obtained with known Photofrin concentrations. 2.3.2.5 Determination of Photofrin content by radiolabeled drug [14C]-polyhematoporphyrin (which is identical to Photofrin), 4 (iCi/mL (1 Ci/mol), was purchased from Leeds Radioporphyrins (University of Leeds, Industrial services Ltd., Leeds, England); the labeled atoms are located within the porphyrin ring. Cells were exposed to [14C]-polyhematoporphyrin under the same in vitro conditions as for the other Photofrin uptake experiments. The cell extracts in ScintiGest were also prepared the same way as for the fluorimetric assay. Before radioactivity measurement, the cell extracts were neutralized with glacial acetic acid. H 2 O 2 (30%) was the added (0.5 mL/1.5 mL cell extract), followed by a scintilation fluid (ScintiVerse Bio-HP; Fisher Scientific Co.). Radioactivity was measured in an .LKB 1214 liquid scintilation spectrometer. The concentration of photosensitizer in the samples was determined using a calibration curve prepared from radioactivity counts obtained with the ScintiGest-treated solutions of known concentrations of radiolabelled drug. 2.3.2.6 . Separation of macrophages by differential cell attachment Macrophages have the capacity of fast and firm attachment to a plastic substrate, while SCCVII tumor cells need a much longer time for firm substrate attachment. This difference can be exploited as a means of obtaining separate populations of tumor cells and TAM (Russell et al. 1980). Single cell suspensions in MEM + 10% FBS were prepared from SCCVII tumor as described above. The cells were plated in 10 cm tissue culture dishes (Falcon, Becton Dickinson and Comp., Lincoln Park, NJ, USA) at a concentration of 5-Photofrin in murine macrophages 30 6 x 1 0 6 cells per dish and left 5 min. in the CO2 incubator at 37°C. The supernatant was then removed and non-attached cells rinsed with two washes using 5 mL of the medium. The attached cells consisted almost exclusively of macrophages. Over 8 5 % of these cells were positively stained for non-specific esterase, and the morphology, of some non-positively stained cells suggested that they were macrophages. The non-attached cells in the original supernatant and the two washings were pooled together and transferred into second 10 cm dish. The macrophages remaining in the cell suspension were allowed to attach to the bottom surface of this second dish by 10 min. incubation at 37°C. The non-attached cells remaining suspended in the medium were finally plated in the third 10 cm tissue culture dish, or they were taken immediately for determination of Photofrin content. The tumor cell population obtained by this procedure contained no positively stained cells for non-specific esterase. 2.3.3 Statistical analysis The significance of the difference between two mean Values was in most cases carried out using an unpaired t-test assuming a normal sample distribution. A pooled variance was used with assumption that samples were drawn from populations with equal Variance. P-values smaller that 0.05 were considered statistically significant. Standard deviation was employed as a mesaure of dispersion. 2.4 Results 2.4.1 The uptake of Photofrin by murine peritoneal macrophages and SCCVII tumor cells in vitro. The results of Photofrin uptake by peritoneal macrophages and S C C V I I tumor cells are shown in Figure 2.1. Measurement of drug fluorescence in Photofrin in murine macrophages 31 live cells (Figure 2.1 A) showed much better accumulation of Photofrin in tumor cells than in macrophages. The uptake curves of both cell types show that saturation is reached after the initial 2-3 hours of exposure to the drug. However, quite different results were obtained when Photofrin fluorescence in the same cells was measured in the cell extracts (Figure 2.1 B). The uptake capacity of macrophages was in this case clearly much superior to that seen in tumor cells. The shape of the curves remained basically unchanged. Photofrin in murine macrophages 32 3 "ai o CO o CD Q l <D O C CD O CO CD o 3 - • — Macrophages - • — SCCVII n 1 — T 1 1 T 1 1 r - —r— 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 Uptake time [h] 3 JO 75 o CD o CD o_ CD o c CD O , CO CD O D 1 5 Uptake time [h] r 2 0 1 2 5 Figure 2.1 Photofrin uptake by peritoneal macrophages and SCCVII tumor cells in vitro. Cells were exposed to Photofrin (40 u-g/mL) in Eagle's MEM with 1% FBS for the time intervals indicated. A, Photofrin fluorescence measured in suspension of live cells; B, Photofrin' fluorescence in cell extracts. Bars indicate SD. Photofrin in murine macrophages 33 Photofrin uptake by macrophages and S C C V I I tumor cells following exposure to different photosensitizer concentrations was also measured, t h e macrophages and tumor cells were exposed to [ 1 4C]-polyhematoporphyrin for 24 hours and the results are expressed per cell protein to compensate for differences in cell size. Determination of cellular protein content based on the Lowry method showed that SCCV I I tumor cells have on average 3 times more cellular protein than macrophages (278 versus 92 pg/cell, respectively). The data presented in Figure 2.2 clearly demonstrate that macrophages, compared to tumor cells have significantly greater capacity for photosensitizer uptake. 5 10 Photofrin exposure Tjig/ml] 15 Figure 2.2 , Uptake of [14C]-polyhematoporphyrin by peritoneal macrophages and SCCVII tumor cells in vitro. Cells were exposed for 24 hours to different concentrations of the radiolabeled drug in Eagle's MEM with 1% FBS. Radioactivity in the samples was measured using a scintillation counter. Bars indicate SD. Photofrin in murine macrophages 34 The effect of the presence of nonadherent cells from the peritoneal exudate on photosensitizer uptake by macrophages is shown in Table 2.1. Two hours after the initiation of in vitro culture, the non-adherent cells were washed away in one group of samples, while in the other samples these cells were not removed. All the samples were then incubated for 24 hours in seruni-free growth medium and exposed to [1 4C]-labeled photosensitizer. The presence of non-adherent cells, was obviously stimulating the photosensitizer uptake by macrophages, which was 40-50% greater than in the samples where non-adherent cells were washed away shortly after the initiation of in vitro culture. Table 2.1 Effect of presence of non-adherent cells from peritoneal exudate on the uptake of [14C]-polyhematoporphyrin by murine peritoneal macrophages. . . Photosensitizer content 3 (ng/u.g cell protein) Non-adherent cells not present 42 + 3 Non-adherent cells present 60 ±3 a Determined after 4 hours of exposure to 10 u.g/mL (0.016 |xGi/mL) of [ 1 4 C ]-polyhematoporphyrin in serum-free medium, using a scintillation counter. The values for photosensitizer content for the two groups are statistically different (P<0.01), as determined by Student's t test. Values are mean + SD. 2.4.2 The effect of serum on Photofrin uptake by SCCVII cells and macrophages. In experiments testing the effect of serum on photosensitizer uptake, macrophages were always cultured in the presence of. non-adhering cells. As shown in Figure 2.3 the presence of 1% FBS strongly inhibited sensitizer uptake by both S C C V I I cells and macrophages. Little additional effect on uptake was seen at higher serum concentrations. The inhibitory effect of serum was more pronounced with macrophages than with tumor cells (a reduction of Photofrin in murine macrophages 35 approximately V-io, compared to a reduction of V4 - V5 of the sensitizer level seen in serum-free medium with tumor cells). -^ 5 CD O O 4 CD _N V-< '(Ii c CD CO 5 1 • Macrophages • SCCVII cells 6 r T H i l l ! Br T w 0 "53 CO °1 2 1— •*—> c |1 0 0 c ~ 8 0 0 % 6 CD co 4 N jSU §"2 CD O CO co 0 E, 0% 1 % 5% 10% Percent serum 0 % 1% . 5% 1 0 % Percent of FBS in culture media Figure 2.3 Effect of serum on Photofrin uptake by peritoneal macrophages in vitro. Cells were exposed to [14C]-polyhematoporphyrin (10 ug/mL; 0.016 uCi/mL) for 4 h in Eagle's MEM with 0, 1, 5 or 10% FBS in the presence of non-adhering cells from peritoneal exudate. SCCV I I tumor cells exposed to the drug under the same conditions were examined for comparison. The cellular sensitizer content was determined as for Figure 2.2. . Inset: Ratios (macrophages : tumor cells) of the sensitizer content per cellular protein calculated from the drug uptake levels obtained in the presence of different serum concentrations. Bars indicate SD. This is best illustrated by following the ratio macrophages/tumor cells of the sensitizer levels on per cellular protein basis (Figure 2.3, inset). The ratio value, Photofrin in murine macrophages 36 which in the absence of serum was 11.0 +1.4 (SD), was reduced by almost 5 0 % with 1% serum present. With higher serum concentrations, the ratio declined even further, although much less prominently. . 2.4.2.1 The effect of plasma proteins on uptake and retention of Photofrin by SCCV I I cells and macrophages. The effects of different plasma proteins on the uptake and subsequent retention of Photofrin are shown in Figure 2.4. The inhibition of Photofrin uptake (compared to protein free samples) was obtained with all the proteins, with the greatest effect shown by albumin. The ensuing retention of the drug in macrophages was not affected to any greater degree by the interaction with plasma proteins during the Photofrin uptake period, although the samples with LDL and albumin showed somewhat inferior Photofrin retention. As the net result of these differences in Photofrin uptake and retention in peritoneal macrophages, the levels of the drug that remained in the cells were the lowest in albumin and LDL samples. Photofrin in murine macrophages 37 ^ 100 -. c. CD c o o CD > JO 2 0 _ C 0 CD DC 8 0 £ 60 -O •—» o 40 o CO d CM CM • Uptake H Uptake and clearance 0 0 d I CD CO 5^ in CM MEM. LDL HDL ALBUMIN Figure 2.4 Effects of plasma proteins on uptake and retention of Photofrin by peritoneal macrophages in vitro. Cells were exposed to Photofrin (40 ug/mL) for 1 hour in Eagle's MEM with LDL (0.2 mg/mL), HDL (0.2 mg/mL), or albumin (5 mg/mL). or in the absence of protein. For the ensuing retention of the drug, the cells were incubated for 24 hours in drug-free medium with 10% FBS. Photofrin content was determined from the cell extracts using the fiuordmetric assay. Photofrin uptake values relative to the uptake values obtained with protein-free samples and percentage of the retained drug following the clearance incubation are indicated in the uptake and clearance columns, respectively. Bars indicate SD 2.4.3 Photofrin uptake by peritoneal macrophages is inhibited by cytochalasin-B. In vivo uptake of Photofrin by peritoneal macrophages was also examined. In this case, Photofrin was administered i.v. and the mice were sacrificed at different times after the drug injection. Peritoneal macrophages were harvested and the photosensitizer content was determined in the cell extracts by the fluorometric assay. The data (Table 2.2) show that Photofrin accumulates faster during the initial hours following drug administration. Subsequently, it took 20 hours to double the cellular level reached at 4 hours after administration Photofrin in murine macrophages 38 of the drug. The results of Photofrin measurement in macrophages obtained from mice which received cytochalasin B immediately before Photofrin are also shown in the same table. It is evident that the uptake of the photosensitizer by, peritoneal macrophages in vivo was strongly inhibited by cytochalasin B (p < 0.01). Table 2.2 Effects of cytochalasin B on in vivo uptake of Photofrin by peritoneal macrophages8. • Photofrin content (ng/pig cell protein) Time after drug No With . administration (h) cytochalasin B cytochalasin B 4 210 + 40° 24 420 ± 90 a Photofrin, 25 mg/kg, i.v.; cytochalasin B, 15 mg/kg, i.p. Photofrin content was determined from cell extracts by the fluorometric assay. b Mean ±SD. c p < 0.01 Retention of Photofrin (accumulated in vivo) in peritoneal macrophages, tumor associated macrophages, and S C C V I I tumor cells under in vitro conditions is depicted in Figure 2.5. The clearance curves for all three cell types are remarkably similar. They show some loss of Photofrin material from cells between the 1st and 4th hour of clearance incubation, and a very limited further clearance between 4 and 24 hours of incubation. All three cell types retained 70-80% of Photofrin after 24 hours of clearance period. The initial values of Photofrin per |ig of cel l protein in peritoneal macrophages and TAM were 420 ± 90 and 74 + 13 ng, respectively, and 15 + 2 ng in tumor cells. Calculated per cell protein, the levels of Photofrin in TAM were approximately 60 ± 30° 70 ± 40° Photofrin in murine macrophages 39 15 times higher than in tumor cells. Retention of Photofrin accumulated in vitro in SCCVII tumor cells is also shown (Figure 2.5.b). In contrast to the other three clearance curves shown in Figure 2.5, the rate of Photofrin loss from the cells was in this case evidently greater during the initial hours of incubation. Photofrin in murine macrophages 40 1 Peritoneal macrophages Tumor assoc. macrophages 0 L_i_l i • i • i • i • i • i • ' • i .1 , I 0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 Clearance time [h] • Uptake in vitro • Uptake in vivo • • • • i • ' • ' • ' 0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 22 2 4 2 6 Clearance time [h] Figure 2.5 Retention of Photofrin in macrophages and tumor cells under in vitro conditions. A, retention of the drug accumulated in vivo in peritoneal and tumor associated macrophages; B; retention of the drug accumulated either in vivo or in vitro in SCCVII tumor cells.. The clearance incubation was in drug free medium with 10% FBS. Photofrin retention was calculated from the drug contents determined from the cell extracts using the fluorometric assay. Bars, SD. Photofrin in murine macrophages 4.1 2.4.4 The uptake of Photofrin by SCCVII tumor cells and tumor associated macrophages in vivo. In next series of experiments Photofrin fluorescence in FcR+ and FcPr cell populations from S C C V I I tumors excised at different times after the administration of photosensitizer to mice was determined by flow cytometry analysis (Figure 2.7). 2 r CO 1.5 -o 1 -o 3 O JO 0.5 -1 2 2 4 3 6 4 8 6 0 7 2 8 4 ,9 6 Time after Photofrin administration (hours) Figure 2.7 Photofrin fluorescence intensity in tumor cells (FcR") and host cells (FcR+). Cells were sorted from SCCVII tumor cell suspension and fluorescence intensity measured by FACS 440. The measurement was performed for a series of tumors (three or more per point) excised at different times post administration of Photofrin (25 mg/kg). " = Photofrin level significantly lower compared to that at12.hours. (p<0.01). Bars represent standard deviations. Photofrin in murine macrophages 42 The Photofrin levels in both cell populations reached a maximum 12 h after the drug administration. This was followed by a steady decrease in Photofrin levels in both cell populations at 2, 3, and 4 days after the administration of the photosensitizer. The decrease was statistically significant for F c R + cell population (p < O.Oi). The Photofrin level in F cR + populations was constantly higher than in FcR- populations at all time intervals examined. The results of an independent method of Photofrin content measurement employing chemical extraction of porphyrin material from F cR + and FcR' cells sorted by F A C S are presented in Figure 2.8. Similar to the results obtained by flow cytometry (Figure 2.7), constantly higher levels of the photosensitizer were detected in the FcR+ tumor cell population. The Photofrin content reached its highest level at 24 h post administration of the drug. During the next 2 days the Photofrin levels decreased in both cell populations. Compared to the value at 24 h, the decrease was statistically significant at 96 h with FcR+ cells. Photofrin in murine macrophages 43 5 r F c R + FcR " 0 1 2 24 3 6 48 6 0 7 2 8 4 9 6 Time after Photofrin administrat ion (hours) Figure 2.8 Photofrin content in two populations of S C C V I I tumor cells; FcR+ and FcR". The cells were separated and sorted by FACS and the Photofrin content was determined by fiuorometric assay from the cell extracts. The FACS measurement of Photofrin fluorescence in the same samples is shown in Figure 2.7. **= Photofrin level significantly lower than at 24 h (p<0.05). Bars are SD. The results for Photofrin measurements shown in Figures 2.7 and 2.8 are expressed on per cell basis and thus do not take into account a marked difference in size and protein content that exists between cells found in F c R + and FcR- populations. For FcR+ cells, the measurements in Figure 2.8 with protein content calculated per u,g cell protein gives 0.38 and 0.18 ng of Photofrin per jig of cell protein at 24 and 96 h, respectively. The values at 24 and 96 h for FcR- cells are 0.05 and 0.04 ng of Photofrin per |ig of cell protein, respectively. The analysis of Photofrin content per weight of cell protein clearly emphasizes the higher levels of the photosensitizer accumulation in F c R + populations. Photofrin in murine macrophages 44 The ratios of the Photofrin levels determined in FcR+ and FcR' populations expressed on per cell basis, are shown in Figure 2,9. The ratio values derived from flow cytometry analysis are always greater than 1.7, and reach 2.7 at 12 h post Photofrin administration. Compared to the value at 12 h, the ratios at 1 and 24 h are significantly lower, while at the other time points the difference is not statistically significant. The comparative ratio values derived from the porphyrin extraction method shown on the same graph are very similar, except that the maximal value (3.0) is already reached at the 4 h time point. Compared to this maximal value, the ratios at 48, 72, and 96 h are significantly lower (p<0.005), while at the other time points the difference is not statistically significant. Photofrin in murine macrophages 45 rr o o o + rr o 4.0 r 3.5 -3.0 2.5 2.0 1.5 •£ 1 -0 0.5 0.0 • Flow cytometry • .Extraction method i I I 1 4 12 2 4 4 8 7 2 Time after Photofrin administration (h) 96 Figure 2.9 The ratios of Photofrin content determined in FcR + : FcR" cell populations derived from SCCVII tumors The ratio values were derived from the data shown in Figure 2.7 (flow cytometry analysis) and Figure 2.8 (chemical extraction of porphyrin from the cells). Values significantly lower than the values at 4 h (cell extracts), or 12 h (flow cytometry analysis): **p<0.005; *.= p<0.01. The calculation, based on weight of cell protein of the FcR+ / FcR- ratios of Photofrin content obtained with the extraction method, give essentially the values shown in Figure 2.9 multiplied by a factor 3.02, average ratio of protein content in FcR- cells compared to FcR+ cells. Such analysis shows that FcR+ cells can accumulate 9.2 times more Photofrin on a per cell protein basis at the peak time, i.e. at 4 h post Photofrin administration. Photofrin in murine macrophages 46 2.4.4.1 The effect of separation method on Photofrin content in SCCVII tumor cells and TAMs. The results of Photofrin content measurement in populations of S C C V I I tumor derived cells separated by the differential attachment separation technique are shown in Table 2.3. The TAM population obtained with this separation technique again showed much higher levels of Photofrin than the tumor cell fraction. The measurement, with cells harvested at 24 h post drug administration, showed a Photofrin level in the tumor cell fraction very similar to that obtained with F A C S sorted FcR" cells (Figure 2.8, 24 h). The photosensitizer level in the TAM fraction, however, was over two times higher than in FACS sorted FcR + cells, i.e. 7.37 u,g compared to 3.6 fig of Photofrin per 108 cells. Table 2.3 Photofrin content in tumor cells fraction and TAM fraction, derived from SCCV I I tumor and separated by differential substrate attachment. Photofrin content Ratio (mg/10 8 cells) (TAM : tumor cells) Tumor cell fraction i.50± 0.17a 4.9 ±0.9 TAM fraction 7.37 ±1.30 Tumors excised 24 h after Photofrin administration (25 mg/kg)". Photofrin content determined by porphyrin extraction method followed by fluorometric measurement (see Materials and Methods). a Standard deviations given. 2.4.4.2 The effect of enzymatic digestion on Photofrin content in tumor cells and macrophages. The issue of a possible disproportionate photosensitizer loss from different cells during the exposure to the enzymatic digestion procedure used for dissociation of cells from tumor tissue was analyzed in a separate in vitro experiment. Three different types of cells, S C C V I I tumor cells, S C C V I I tumor Photofrin in murine macrophages 47 TAM, and peritoneal macrophages from .C3H/HeN mice, were exposed to Photofrin in vitro, and then harvested either by using a rubber policeman, or by employing the enzymatic procedure identical to that used for dissociating cells from tumor tissue. The macrophages (both TAM and peritoneal) accumulated more Photofrin than the tumor cells (Table 2.4) under these in vitro conditions. The loss of cellular Photofrin following enzymatic digestion was approximately 6 0 % with tumor cells, and around-75% with both types of macrophages. Table 2.4 The effect of exposure to a mixture of trypsin, collagenase, and DNAase on Photofrin levels in tumor cells and macrophages. Photofrin content (mg / 1 0 8 cells) Mechanical Enzymatic treatment detachment SCCVI I tumor ce l ls 3 10+1.8° 3.8 ±0.3 TAM 62 ± 18 - 15±3 Peritoneal macrophages 8 0 + 1 5 21 ±1 Cells were exposed to 10 ug Photofrin/mL in growth medium (MEM with 1% FBS) for 24 h at 37°C. Subsequently, they were rinsed with PBS, and then harvested using either a rubber policeman, or by the enzymatic digestion procedure identical to that used in dissociation of cells from tumor tissue. Photofrin content was determined by Photofrin extraction from washed cell pellets, followed by fluorometric measurement. aTumor cells and TAM were selected from SCCVII tumors by the differential attachment procedure (see Materials and Methods); the tumor cells were cultivated in vitro for 2 weeks before the experiment. Cultures with peritoneal macrophages were isolated from C3H mice as described in Materials and Methods. bStandard deviations given. 2.5 Discussion It was suggested by Jori (Jori 1989) that porphyrins, phthalocyanines, and other photosensitizers that can remain in a highly aggregated form in vivo, are phagocytized to a large degree by TAMs. The evidence presented in this study demonstrates that macrophages have a capacity for Photofrin accumulation vastly superior to that of SCCVI I tumor cells. In addition, it is shown that factors Photofrin in murine macrophages 48 that modify macrophage activity also influence Photofrin uptake by these cells. Cytochalasin B, a known inhibitor of phagocytosis (Axline and Reaven 1974), reduced in vivo uptake of Photofrin by peritoneal macrophages (Table 2.2). In contrast, the presence of non-adherent cells from peritoneal lavage (mainly lymphocytes) induces macrophage activation (Yamamoto et al. 1988) and stimulates Photofrin uptake by peritoneal macrophages in vitro (Table 2.1). All of these results support the assumption that Photofrin uptake by macrophages is dominated by phagocytic internalization of a, highly aggregated drug component. The interaction of Photofrin with human plasma proteins and complete serum inhibits the uptake of this drug by macrophages, as it does with tumor and other cells (Korbelik et al. 1990; Korbelik and Hung 1991). However, the presence of serum inhibits the photosensitizer uptake to a greater extent in macrophages than it does in tumor cells (Figure 2.3). The above can be interpreted as a consequence of disaggregation of Photofrin material induced by serum and its components (Korbelik and Hung 1991).. This disaggregation of the drug affects its uptake by macrophages more (fewer aggregates to be phagocytized) than its uptake by tumor cells. Photofrin-plasma protein interaction affects the ensuing retention of the Photofrin material in macrophages less than in tumor cells (Korbelik and Hung 1991). In the experiments with individual plasma proteins (Figure 2.4), the macrophages have presumably still phagocytized the same type of drug material, i.e., the remaining Photofrin aggregates. Tumor cells in such circumstances, however, take up altered drug material with more poorly retaining Photofrin species, released upon interaction with plasma proteins (Korbelik and Hung 1991). Both peritoneal macrophages and TAMs have shown a very good retention of Photofrin accumulated in vivo, and similar retention qualities were exhibited Photofrin in murine macrophages 49 by tumor cells (Figure 2.5). In these experiments it was not possible to determine cellular Photofrin contents immediately after collecting cells from the mice, because of the time needed for separation (15-20 min. for peritoneal macrophages, and more than 1 h for TAMs and tumor cells). However since the clearance of Photofrin from these cells is very limited, there is no indication that there could be a significant loss of the drug from the cells during the isolation procedure. The clearance of Photofrin from peritoneal macrophages (Figure 2.4), tumor cells (Figure 2.5 B), and other cells (Korbelik and Hung 1991) following the uptake of the drug in vitro is much greater than for the Photofrin accumulated in vivo. In the former case, the clearance curves are characterized by an early phase of rapid Photofrin loss during the initial 1-3 hours of the clearance incubation, followed by a phase of further limited loss of the drug from the cells (Henderson et al. 1983; Korbelik and Hung 1991). It can be assumed, therefore, that in the cells collected after 24 hour accumulation of Photofrin in vivo the initial phase of fast Photofrin clearance has mostly been completed. One of reliable methods for discrimination between tumor cells and host cells in solid tumors is based on the presence of surface receptors for the Fc portion of immunoglobulin on the membranes of the host cells and the absence of these receptors on the membranes of tumor cells (Wood and Gollahon 1977; Lindsay et al. 1982). This principle was successfully employed by Olive (Olive 1989) to separate tumor and host cells derived from murine S C C V I I tumor by flow cytometry using the FITC-conjugated anti-mouse IgG. Over 9 5 % of FcR+ (i.e. IgG positive) cells in S C C V I I tumor were identified as TAM (Olive 1989). The FcR- fraction contains tumor cells contaminated with 5-10% diploid cells, presumably host cells devoid of Fc receptor. In this study we have combined the above method for separation of host and tumor cells with simultaneous Photofrin in murine macrophages 50 excitation and fluorescence measurement of photosensitizer Photofrin using a dual laser F A C S apparatus. This enabled the examination of Photofrin distribution between malignant tumor cells (FcR negative) and host cells (FcR positive) contained in the murine SCCV I I tumor following in vivo administration of the photosensitizer. The flow cytometry analysis of cellular Photofrin content in FcR+ and FcR' populations Comprising the S C C V I I tumor demonstrated that, at least up to 4 days after Photofrin administration, there is more drug accumulated in F c R + than in FcR' cells (Figure 2.7). Taking into account the content of host cells in S C C V I I tumor (-40%), the data indicate that most of the photosensitizer material is. in fact contained in TAMs, not in the tumor cells. This is particularly pronounced between 4 and 24 h after Photofrin administration when there is up to 3 times more photosensitizer in FcR+ than in FcR - cells on per cell basis. Based on a per cell protein calculation the ratio is greater than 9 times for F c R + over FcR ' cells. The measurement of Cellular Photofrin content by flow cytometry was verified by an independent measurement of the drug content in sorted cells using chemical extraction of porphyrin followed by the fluorometric assay. This method includes hydrolysis and disaggregation of porphyrin material, which results in conversion of all photosensitizer components into a more uniform material with optimal fluorescence yield (Brown and Vernon 1990; Korbelik and Hung 1991). This is not the case with Photofrin fluorescence measurements in live cells during flow cytometry analysis, which registers the fluorescence intensity strongly dominated by highly fluorescing monomeric species, and reflects to a much lower degree the presence of highly aggregated and less-fluorescing species of this drug (Moan and Sommer 1983; Brown and Vernon Photofrin in murine macrophages 51 1990). The total Photofrin content in the cells not determined by the flow cytometry analysis was thus assessed by the extraction method. The comparison of the results obtained by these two methods of Photofrin measurement (Figure 2.9) does not reveal any major discrepancies; both methods show higher levels of the sensitizer in F cR + then in FcR - cells. The only possible difference may be observed in the time when the maximal value for FcR +/FcR- ratio in Photofrin levels is reached. This peak time is observed earlier for the extraction method (4 h post administration) than with flow cytometry analysis (24 h post administration); however, this difference cannot be fully supported by statistical calculation (see Figure 2.9). This possible difference is not unexpected, since at its highest levels (attained 4 h post administration) much of the Photofrin material in the cells could be in the highly aggregated form and thus underestimated by the flow cytometry. A few hours later, following intracellular dissociation of its aggregates, Photofrin can reach its strongest fluorescence. Nevertheless, Figure 2.9 suggests that in general there is no substantial difference in Photofrin content as revealed by the two methods (fluorescence in the live cells and fluorescence following the extraction procedure). This implies that the form in which the photosensitizer is presented to cells under in vivo conditions is.not the same as in vitro (Figure 2.1). In fact, an alternate possibility to the existence of aggregated species in vivo would be complexes of non-aggregated photosensitizer associated with various macromolecules. There seem to exist two potential impediments to the FACS method used in this study. The procedure for dispersion of the tumor into single cell suspension takes time (more than 1 hour) and exposes cells to digestion enzymes; all of which could result in some loss of photosensitizer from the cells. The other potential problem is photodestruction (photobleaching) of Photofrin by exposure Photofrin in murine macrophages 52 to the strong excitation light (FACS laser). In a control experiment, in which TAMs were separated,from tumor cells by taking advantage of their much more rapid attachment to the plastic substrate; a 4.9 times higher level of Photofrin was detected in TAMs enriched population compared to the tumor cell population (Table 2.3). In this case FACS was not used, and thus there could be no Photofrin.photobleaching. Compared to the TAM population sorted by FACS , the macrophages selected by differential attachment procedure may be enriched in activated TAMs. The most likely reason for this difference is that not all macrophages contained in FcR+ fraction would attach equally well to the substrate. The non-attaching TAMs, which are probably less active and accumulate less Photofrin, are lost in the differential attachment procedure. This factor seems most likely to be responsible for the difference iii the Photofrin levels found in TAM populations selected by these two procedures. The enzymatic procedure used for obtaining single cell suspension, from tumors could induce a differential loss of Photofrin from TAMs and. tumor cells. The Photofrin loss seen in the experiment designed to address this issue (Table 2.4) was greater in macrophages than in tumor cells. This suggests that higher -Photofrin levels in F cR + cells compared to FcR - cells (Figures 2.7-2.9) could not be an artifact induced by the enzymatic digestion procedure. It should also be noted that the enzymatic treatment is much harsher to isolated cells attached to the tissue culture dish than to the cells contained in pieces of tumor tissue. It can be assumed that the photosensitizer loss from the cells induced by.the enzymatic procedure before the flow cytometry analysis was substantially lower than that seen in Table 2.4. This is also inferred by the retention measurements, which suggest that Photofrin material accumulated in wVo.in cellular populations of the SCCV I I tumor (at 24 h after the drug administration) is bound more firmly to the cells than following brief in vitro exposure to the Photofrin in murine macrophages 53 drug. In the latter case, the drug may be bound loosely to the outer cellular membrane, where it is more readily removed by the enzymatic treatment. The data presented in this study clearly show that in SCCV I I murine tumor much higher levels of Photofrin are accumulated in TAMs than in tumor cells. This result strongly supports previously published but less direct evidence and suggestions by others (Bugelski et al. 1981; Chan et al. 1988; Henderson and Donovan 1989; Jori 1989)! In tumors with a high macrophage content the preferential accumulation of Photofrin in TAMs could significantly contribute to the localization of Photofrin in tumor tissue. The fact that most of the photosensitizer material could be accumulated in the TAM population has obviously important implications for PDT. Light energy dependent production of tumor necrosis factor alpha (TNF-a) by murine macrophages has been demonstrated after in vitro PDT using Photofrin (Evans et al. 1990). Release of prostaglandin E from peritoneal macrophages after Photofrin-based PDT has also been reported (Henderson and Donovan 1989). There are strong indications that PDT-induced immunosuppression (Lynch et al. 1989) and lethality due to TNF-a induced cahexia following PDT are mediated by macrophages (Darling et al. 1990; Ferrario and Gomer 1990). PDT-induced release of a variety of bioactive substances from TAMs , either by their destruction or activation can have profound effects on tumor cells, tumor vasculature, and other host cells. Recently, an alternative mechanism for selective photosensit izer accumulation in TAMs in vivo was proposed by Hamblin and Newman (Hamblin and Newman 1994). The authors hypothesize that TAMs accumulate photosensitizers which are bound to low density lipoproteins (LDL) and other plasma proteins via the LDL receptors and scavenger receptors. Photofrin in murine macrophages 54. Photofrin accumulation in malignant and host cell populations of various tumors was more recently studied in depth in our laboratory. In an extension to the initial work described above, we have performed more detailed analysis of different host cell populations infiltrating murine tumors. The study of Photofrin distribution in the murine fibrosarcoma FsaR growing in syngeneic C3H/HeN mouse or severe combined immunodeficient (SCID) mouse (Korbelik and Krosl 1995) has shown that there is marked heterogeneity in the cellular content of Photofrin within the tumor. Photofrin content in some TAMs was not higher.than in malignant cells. However, a subpopulation of macrophages characterized by high expression of CD25 (IL-2 receptor) and high granularity accumulated on average almost 13 times more Photofrin than malignant cells. A study with various tumor models of different histological types and degrees of differentiation, including chemically induced squamous cell carcinoma in Syrian hamster and a spontaneous adenocarcinoma that arose in C3H mouse, has shown very similar results (Korbelik and Krosl 1996). While the overall levels of Photofrin varied in different tumors, a subpopulation of TAMs that accumulated by far the highest cellular levels of photosensitizer was found in all of the examined tumors. Based on the results presented in this Chapter and the findings described above, it can be concluded that Photofrin accumulation in solid tumors does not depend on host species, the histological type or degree of differentiation of the tumor, or immunogenicity. Moreover, no difference was noted in Photofrin distribution between hamster and murine tumors despite the difference in the blood LDL content between hamster (similar to human) and mouse (low LDL content) (Chapman 1986). In bur view, it is the activation state of TAMs and their relative content in the tumor that is the determining factor for the degree of accumulation of Photofrin in solid tumors. This is also consistent with the Photofrin in murine macrophages 55 observation of photosensitizer accumulation in areas of inflammation (Nseyo et al. 1985). The evidence that in human malignancies TAMs mostly localize in a lesion's periphery (Svennevig and Svaar 1979; Bucana et al. 1992) gives support to our hypothesis that the TAM population that accumulates the highest cellular levels of Photofrin is responsible for the tumor fluorescence of this photosensitizer which delineates the tumor from normal tissues. PDT induced host cell infiltration 56 CHAPTER 3 Induction of immune cell infiltration into murine SCCVII tumor by Photofrin based photodynamic therapy1 3.1 Summary Cellular populations in the squamous cell carcinoma SCCV I I , growing in C3H/HeN mice given Photofrin, were examined at various time intervals during the photodynamic light treatment and up to 8 hours later. Cell populations present within excised tumors were identified by monoclonal antibodies directed against cell type specific membrane markers using a combination of indirect immunoperoxidase and Wright staining or by flow cytometry. Photofrin based photodynamic therapy (PDT) induced dramatic changes in the level of different cellular populations contained in the treated tumor. The most pronounced was a rapid increase in the content of neutrophils, which increased 200 fold within 5 min. after the initiation of light treatment. This was followed immediately by an increase in the level of mast cells, while another myeloid cell type, most likely monocytes, invaded the tumor between 0 and 2 hours after PDT. Macrophages harvested from S C C V I I tumor 2 h post PDT exhibited pronounced increase in tumoricidal activity against autochtonous SCCV I I tumor cells. It seems, therefore, that PDT induced both the recruitment and functional activation of host-derived myeloid cells. 'The data presented in this chapter have been published in the following paper: -.1. Krosl, G., korbelik, M., Dougherty, G.J. (1995) Induction of immune cell infiltration into murine SCCVII tumour by photofrin-based photodynamic therapy. Br. J. CancerTK, 549-555. PDT induced host cell infiltration 57 3.2 Introduction A combined effect of several cancer tissue destroying mechanisms is responsible for tumor regression following photodynamic therapy (PDT). (n addition to direct killing of tumor.cells by phototoxic action and ischemic necrosis secondary to the collapse of the vascular system, there are indications of the involvement of PDT induced immune reaction (Bellnier and Henderson 1992; Pass 1993). An understanding of this rather complex interaction is made more difficult by the fact that the events leading to the vascular damage are different with different photosensitizers (Henderson and Fingar 1994). The most abundant lesion induced in PDT treated tumor is the phototoxic damage to the surface membranes of tumor cells. We have hypothesized that this subtle initial and not necessarily lethal damage may trigger a chain of events leading to tumor eradication (Korbelik and Krosl 1994). The invoked damage is probably the peroxidation of membranous lipids (Thomas et al. 1987), which prompts a rapid (< 1 min.) activation of membranous phosphol ipases (Agarwal et al. 1993) for accelerated degradation of phospholipids. It should be noted that very similar events occur in cell membranes with the initiation of inflammation by microbial infection or by some other types of tissue injury (Chien et al. 1978; Yamamoto and Ngwenya 1987). The membrane damage discussed above can be thus described as PDT induced inflammatory cellular damage. Inflammatory immune responses induced in cancerous tissues may be of different nature and more intense than those induced in normal t issues (Yamamoto and Ngwenya 1987); this may very well be the case with the reaction triggered by PDT in solid tumors. The lipid composition of tumor cell membranes is different than that of normal cell membranes (Snyder and Wood 1969; Howard et al. 1972). Fragments released from tumor cell membranes PDT induced host cell infiltration 58 damaged by PDT include a variety of lysophospholipids and alkylglycerols (Yamamoto et al. 1988), as well as arachidonic acid and its metabolites (Henderson and Donovan 1989;. Bellnier and Henderson 1992; Fingar e t a l . 1992), all of which can serve as highly potent, stimulatory signals for the amplification of the inflammatory reaction. They are powerful chemotactic and stimulatory agents for immune cells, or highly active vasomodulatory mediators (Zurier 1982; Yamamoto and Ngwenya i987; Yamamoto et al , 1988). A strong inflammatory reaction may be the most important contributor to the destruction of vasculature in PDT treated tumor. Very high doses of inflammatory products of cancerous tissues can cause immunosuppression, which was observed following PDT (Lynch et al. 1989). On the other hand, at different dose levels these same agents cause immune stimulation (Ngwenya and Yamamoto 1986; Yamamoto and Ngwenya 1987). It is therefore of a critical importance to better understand the PDT induced inflammatory/immune reaction in order to develop improved strategies for therapeutic benefit. The accumulation of inflammatory cells is a central event in the inflammatory process. Yet, there, is very little information available on the infiltration of these cells into the treated tumor during and after photodynamic light delivery. Obtaining this information was the main objective of this study. Focusing on the clinically established photosensitizer Photofrin, we, have chosen as tumor model the squamous cell Carcinoma S C C V I I growing subcutaneously in C3H/HeN mouse. The photodynamic light treatment was performed 24 hours after the administration of Photofrin. The results demonstrate a rapid, massive and regulated infiltration of various immune cells into the tumor site during and immediately following PDT. PDT induced host cell infiltration 59 3.3 Materials and methods 3.3.1 Tumor model Female C3H/HeN mice, age 10-12 weeks, were used for experiments. The S C C V I I squamous cell carcinoma was maintained by intramuscular passage as described previously (Korbelik 1993). This is a weakly immunogenic tumor which originated spontaneously in the abdominal wall of a C3H mouse (Suit et al. 1985) For experiments, tumors were induced by injecting 3 x 1 0 5 cells, obtained by enzymatic digestion of intramuscularly growing tumor, subcutaneously over the sacral region of the back of anesthetized mice. Prior.to tumor inoculation all the hair was removed from the injection site by shaving. Tumors were used for experiments two weeks after the inoculation at which point their largest diameter was 7-8 mm and thickness 3-4 mm. 3.3.2 Photodynamic therapy Photofrin® (Quadralogic Technologies Phototherapeutics Inc., Vancouver, B.C., Canada) was administered at 25 mg/kg intravenously 24 hours before light treatment. A tunable light source (Photon Technology International Inc., Model A5000) equipped with a 1 kW Xenon arc bulb and infrared filter was used to deliver 630+ 10 nm light to the tumor by a liquid light guide (Oriel Corp., Stratford, CT, USA) with a 5 mm core diameter. The output light power was 35 mW. Mice were restrained unanesthetized in specially designed holders. A dose of 60 J c n r 2 (average fluence rate 45 mW/cm) was used in all experiments and the average treatment time was 24 min; the exception was the analysis of tumors for the effects occurring before the full light dose was delivered. Tumor temperature measured using a hypodermic thermocouple (YSI, Yellow Springs, OH, USA) increased to 38-39°C at the end of the light treatment. PDT induced host cell infiltration 60 3.3.3 Indirect immunoperoxidase and Wright staining At various times after light treatment, mice were sacrificed by cervical dislocation, tumors were then excised and minced using two scalpels. Tumor tissue was then enzymatically dissociated by a 30 min. treatment in a mixture of dispase, collagenase and DNAse as described in detail previously (McBride et al. 1992; Korbelik 1993). Briefly, minced tumor tissue was resuspended in 5 mL of HBSS and 300 |iL of each 4 mg/mL Collagenase I, 3 mg/mL DNAse (both from Sigma), and 25 mg/mL Dispase (Boehringer, Meincheim, Germany) was added. After 30 min. of gentle mixing the resulting cell suspension was filtered through a 100p:m nylon mesh to remove any remaining tissue clumps and washed with Eagle's Minimum Essential Medium (EMEM) containing 1 0 % fetal bovine serum (FBS, HyClone Laboratories Inc., Logan, UT, USA). CD45 (panleukocyte marker), and ap-TCR (T cell receptor) were detected using supernatants from hybridomas YE1.21 and HB218 that were generously provided by Dr. Fumio Takei (Terry Fox Laboratory). F4/80 antigen (marker for macrophages) was stained with hybridoma supernatant HB198 from A T C C . Cells were stained with an indirect immunoperoxidase technique as previously described (Dougherty et al. 1986). Briefly, acetone fixed cytospin preparations of cell suspensions were overlaid with 100[iL of hybridoma supernatant and incubated for 30 min. at room temperature. Cytospins were then washed 3x in HBSS, and left in last wash for 5 minutes. Rabbit anti-rat IgG F(ab')2 fragment conjugated with horseradish peroxidase (Sigma Chemical Co. , St. Louis, MO, USA) was used as a secondary antibody. Staining dilutions as to manufacturer recommendations were made in HBSS + 2 % FBS immediately prior to use. Cells were then overlaid with 100 |iL of secondary antibody and incubated for an additional 30 min. at room temperature. After 3 washes in HBSS the cytospins were immersed in peroxidase substrate dilution (Sigma) prepared PDT induced host cell infiltration 61 according to manufacturers recommendation and incubated until sufficient staining had developed. Slides were then counterstained with Harris hematoxylin, air dried and covered. At least 200 positive cells were scored on each slide. The levels of neutrophils and mast cells were determined using Wright stain (Accustain, Sigma) to manufacturers recommendations. For this purpose, 10 [it of a suspension which contained 5 x 10 4 cells was layered onto glass slides (3 slides per sample), which were left to dry in air before Wright staining. 3.3.4 Identification of blood derived tumor infiltrating cells The bisbenzimide dye Hoechst 33342 purchased from Sigma was used to identify newly infiltrated leukocytes in tumors exposed to PDT. Since this fluorescent nuclear stain is cleared rapidly from the plasma of mice following intravenous administration (Olive et al. 1985), it is possible to prevent the blood flow through the tumor for the time needed for the dye to disappear from the circulation. Tumor bearing mice injected with Hoechst 33342 (16 mg/kg, i.v.) were divided into two groups. In the first group, the tumors were clamped before the injection in order to prevent Hoechst 33342 from reaching the tumors. In the second group, the tumors were not clamped. The clamp was removed after 10 min. to allow re-perfusion of the tumor. Half of the mice from each group were exposed to photodynamic light treatment at 30 min after the Hoechst 33342 injection. The mice were sacrificed at 2 hours after the termination of light treatment, the tumors were excised and disaggregated into single cell suspensions as described above. The cells were then stained with monoclonal antibody to mouse CD45, or a combination of anti-mouse Gr1 (myeloid marker) and F4/80. PDT induced host cell infiltration 62 The antibodies to CD45 and Gr1, labeled with phycoerythrin (PE) were purchased from PharMingen (San Diego, CA), while fluorescein isothiocyanate (FITC) conjugated anti F4/80 was obtained from Serotec Canada Ltd. (Toronto, Ontario, Canada). The staining was performed using a modification of the procedure described earlier (Dougherty et al. 1989). Each sample (total volume 100 u.L) conta ined 0.5-1 x 1 0 6 cel ls suspended in H B S S (Sigma) supplemented with 2 % FBS, to.which the antibodies were added at a dilution recommended by the supplier. The samples, kept in subdued light, were incubated on ice (0°C) for 30 min. The cells were then washed twice in HBSS + 2 % FBS using centrifugation. For each sample, 1 0 4 cells were analyzed by flow cytometry on a Coulter Epics Elite ESP apparatus (Coulter Electronics Ltd.). The 488 nm laser was used to excite FITC and PE. The emission of FITC and PE was recorded through 530 ± 15 and 580+ 10 nm bandpass filters, respectively. Hoechst 33342 was excited by the UV laser and its fluorescence measured through a 449 + 5 nm bandpass filter. Light scatter signals, (forward and side scatter) were also recorded and used for gating out dead cells, erythrocytes and debris. The PDT treated cells of S C C V I I tumor have shown decreased autofluorescence in the 530-580 nm region compared to non-treated cells. As the cells die, their forward light scatter signal will drop substantially. We have not seen any significant difference in the staining signal (Hoechst 33342, FITC or PE fluorescence intensity) between the cells from a non-treated tumor and those cells from PDT treated tumors that still show the forward scatter signal within the values for live cells (although some of them will eventually die). The objective of the above described flow cytometry analysis, indirect immunoperoxidase and Wright staining, was to identify and determine the levels of major cellular populations contained in SCCV I I tumor. The total yield PD J induced host cell infiltration 63 of viable cells per gram of tumor tissue was determined immediately after the single cell suspension was obtained from previously weighed tumor tissue. The cells were counted using a hemacytometer, with Trypan blue staining used to eliminate dead cells. Yields of individual cell populations were calculated from their relative shares in the tumor cell suspension. The determination of the levels of mast cells (and in some cases neutrophils), whose incidence was very low, was based on the fact that a known number of cells were deposited on the glass slide. It was therefore not necessary to count the cells that were not mast cells (or neutrophils), which enabled scoring in average 1.5 x 1 0 5 cells per sample. 3.3.5 Macrophage cytotoxicity against tumor cells The target cells were obtained from an enzymatically digested S C C V I I tumor and cultivated in vitro for 2-3 weeks, which resulted in the elimination of non-malignant cells from the culture. The growth medium was RPMI-1640 (HyClone) supplemented with 1 0 % FBS. The cells were labeled by exposure to [3H-methyl]thymidine (2.0 Ci/mmol, NEN, Du Pont Canada Inc.) at 2 uCi/mL for 24 hours in cell growth medium. Next, they were washed twice with cell growth medium and left to incubate further at 37°C to facilitate elimination of the radioactive label from the cytoplasm. Actinomycin D (Sigma) was added 2 hours later in a final concentration of 1.5 mg/mL. Four hours later, the cells were washed, trypsinized and transferred into a 24-well plate (Falcon 3047, Becton Dickinson and Comp., Lincoln Park, NJ, USA) where they were admixed with the effector cells. Macrophages from PDT treated and non-treated S C C V I I tumors were harvested using a modification of the differential attachment procedure described earlier (Korbelik et al. 1991). A known number of cells obtained by PDT induced host cell infiltration 64 the above described enzymatic dissociation of tumor tissue (suspended in RPMI-1640 + 1 0 % FBS) was transferred into the wells of a 24-well plate and : incubated for 30 min. at 37°C. The medium was then collected, the attached cells overlaid with 0.5 mL of Trypsin-EDTA solution (Sigma) and incubated for 30 seconds at room temperature. After that, the Trypsin-EDTA solution was collected and the wells were washed vigorously three times with 1 mL of HBSS. The washout containing removed cells was collected each time. The number of tumor associated macrophages (TAMs) isolated by this procedure (ceils remaining in the wells) was determined by subtracting the number of cells removed by washings from the number of plated cells. Before admixing the target cells, the effector cells were incubated with lipopolysaccharide (LPS) from E. colt 011 :B4 (Sigma) at 0.1 ng/mL in RPMI-1640 + 1 0 % FBS for 24 hours. The FBS used in these experiments was inactivated by heating at 56°C for 30 min. The effector and target cells were admixed at a 20:1 ratio, each well containing 2 x 1 0 5 TAMs and 1 x 1 0 4 tumor cells. All testing was done in quadruplicates (4 wells for each TAM population). The plates with effector and target cells were incubated 72 hours at 37°G, before the supernatants were collected from the wells and radioactivity counted using an LKB 1214 liquid scintillation spectrometer. The total radioactivity incorporated into the target cells was determined from the cells transferred directly into the scintillation vials. Spontaneous release of [3H]-thymidine from the target cells (in the absence of effector cells) was determined for each group of samples; its levels ranged between 10 and .15%. This value was subtracted from the experimental release in the calculation of the percent of [3H]rthymidine release from the target cells caused by the effector cells. This assay was described in more detail elsewhere (Korbelik and Krosl 1994). POT induced host cell infiltration 65 The data presented are based on the analysis of at least 6 identically treated tumors. 3.4 Results 3.4.1 Changes in tumor cellular content Mice bearing SCCV I I tumors were given Photofrin (25 mg/kg, i.v.) and the tumors were irradiated with red light (60 J/cm 2) 24 h later. This PDT treatment results in the average tumor cure rate of 1 8 % (Krosl and Korbelik 1994). The effect of such treatment on the cell content of the S C C V I I tumor is shown in Figure 3.1. Indirect immunoperoxidase staining with monoclonal antibodies was used to identify the malignant cell population (CD45"), TAMs (F4/80+) and T cells (TCR+). The remaining immune cells infiltrating the SCCV I I tumor were described as "other myeloid cells"; most, but not all of these, stained positively for the M a d myeloid marker, and appeared by morphological criteria to be members of the monocytic lineage. It should be noted that the neutrophils occasionally present in these slides were not included in the analysis. We.have observed that nearly all of the neutrophils present in single cell suspensions obtained from SCCV I I tumors are destroyed during the preparation of cytospins for the indirect immunoperoxidase staining. These cells, which are known, to be especially fragile, are too sensitive for the centrifugal forces used in the cytospin preparations. PDT induced host cell infiltration 66 Control 0 hours post PDT 4 hours post PDT . sVv .% .^ . i . . \ .s .h .s .-. • h > h * H > > > > > > > > > " L . \ . % . - . . \ . \ . \ . -CO o a o t-l l x i o 9 r l x lO 8 lx lO 7 « l x lO 6 i—i U l x lO 5 i - \* • CD45" F4/80+ a(3TCR+ • Neutrophils e Other myeloid cells Control 0 hours 2 hours 4 hours Time after light treatment Is 8 hours Figure 3.1 Levels of major cellular populations in SCCVII tumor before and after PDT. The mice were sacrificed at various time intervals relative to the PDT treatment (Photofrin 25 mg/kg, 60 J/cm 2 light), and the cells dissociated from the tumors analyzed by the indirect immunoperoxidase staining, except for neutrophils which were determined, in Wright, stained preparations/The inserted pie graphs depict relative contributions of the examined cell populations in control tumors and tumors excised either immediately or 4 hours after the termination of photodynamic light treatment. The data are average values from a group of 6 or more identically treated mice. The bars show SD. PDT induced host cell infiltration 67 Because of this problem, separate aliquots of single cell suspensions were always taken for Wright staining. Instead of cytospin centrifugation, 10u,L of concentrated cell suspensions were gently layered onto glass slides and left to dry before Wright staining was performed. In this way, neutrophils were preserved, and they were easily identified by the characteristic shape of their nuclei. The number of neutrophils in relation to the other immune and non-immune cells was also determined in these preparations. In addition, mast cells were also identified by the Wright staining. The above described combination of the indirect immunoperoxidase staining (with neutrophils selectively eliminated) and Wright staining of slides specially prepared to preserve neutrophils enabled reliable determination of major cellular populations found in the SCCVI I tumor. The columns in Figure 3.1 represent cell yields for the main populations recovered from tumors at indicated times after the completion of light treatment. Their proportions in the total cell mass are given in the pie graphs 2. The data in the histogram section demonstrates that the yield of individual cell populations decreased markedly at 4 hours post PDT, and even more so at 8 hours post PDT. However, important changes were also seen at 0 hours (immediately after the termination of light treatment) and at 2 hours post PDT. The most dramatic occurrence noted at 0 hours is a 100-fold increase in the neutrophil content compared to the control tumor. In addition, the level of "other myeloid cells" significantly decreased relative to the level in the controls at this time point. In contrast, the neutrophil content dropped markedly at 2 hours post PDT, while the yield of "other myeloid cells" increased. All these changes are statistically significant (p < 0.01). They also affected the proportions among the major cell 2The values for total number of cells per gram of tumor tissue for specific cell populations were calculated using tumor cell yields and determined proportions of these cell populations. PDT induced host cell infiltration 68 populations. As shown in the pie graphs, the percentage of malignant cells decreased at 0,hours compared to the controls (due to markedly increased percentage of neutrophils), but it decreased even further at 4 hours post PDT. More than half of the cells at this latter time point were F4/80+, and there was also a high proportion (-25%) of "other myeloid cells". No substantial changes in the percentage of T lymphocytes were seen during the observation period. The striking changes in neutrophil numbers detected at 0 hours post PDT, prompted us to examine the neutrophil content in tumors during the photodynamic light treatment, t h i s analysis (based on Wright stained preparations), which included also mast cells, is shown in Figure 3.2. As early as 2 min. after the onset of light delivery, the number of neutrophils increased from 1.6 x 1 0 6 cells/g tumor tissue (control tumors) to 2.4 x 1 0 8 cells/g tumor tissue. Three min. later the neutrophil levels were even higher, but at 12 min. into the light delivery of the total light dose) they already showed a marked decline. The neutrophil content then continued to decrease more slowly. It should be emphasized that the weight of tumors used in this study was 7-10 times lower than 1 g, the weight used conventionally (and also in the presentation of this data) in the calculations of cell yields. This is important to mention, because the total'number of neutrophils in a mouse is lower than 3.2 x 10 8 , the peak level of these cells that would be contained in a 1 g tumor. An increase in the number of mast cells was observed at 5 min. into the light treatment and continued during the light delivery and beyond, reaching the highest value at 30 min. post PDT (Figure 3.2). A similarly high level of mast cells was detected at 2 and 4 hours after PDT (data not shown). Even at these levels the mast cell content in the tumors was much lower than the content of the other cell populations shown in Figure 4.1, and thus they could not affect the percent distributions given in the pie graphs. It should be noted that the scale PDT induced host cell infiltration 69 for mast cells (right ordinate in Figure 3.2) has much lower values than the scale for neutrophils (left ordinate), as the levels of these cells were several logs lower than the neutrophil levels. Duration of light treatment Neu t roph i l s Mast cel ls Time after onset of light treatment [minutes] Figure 3.2 SCCVII tumor contents of neutrophils and mast cells. The neutrophil, and mast cell numbers were determined in Wright stained preparations, at various times during and after photodynamic light delivery . the PDT treatment was as in Fig. .3.1. The bars show SD. PDT induced host cell infiltration 70 Photofrin administered to SCCV I I tumor bearing mice (24 hours earlier) not combined with photodynamic light treatment, or light treatment of tumors in the absence of Photofrin administration, produced no effect on the content of tumor cell populations (data not shown). 3.4.2 Hoechst 33342 staining of circulating leukocytes The changes in the cellular populations present within S C C V I I tumors presented in Figures 4.1 and 4.2 cannot be explained without a contribution, at least in part, from newly arrived cells from the circulation. With neutrophils and mast cells, whose incidence in non-treated tumors is below 1% (neutrophils) or even below 0 .01% (mast cells), their dramatic increase in the tumor can be explained only by the infiltration of cells from the circulation. However, the situation is less clear with cells whose levels in non-treated tumors are much higher, and a possible PDT induced infiltration would not result in a multi-fold increase in their total tumor content. In such case, the relative increase in one type of cells may result also from selective killing of the other cell types. In the next series of experiments we used the fluorescent dye Hoechst 33342 to selectively label immune cells in the blood stream and to determine their presence in the tumor after PDT, as described in Materials and Methods. The limitation of this type of experiment is that the time interval between the Hoechst 33342 injection and tumor excision cannot be extended longer than 2-3 hours, because at later times the dye levels in the labeled cells markedly decreased. Based on this consideration, and on the result indicating a significant increase in "other myeloid cells" between 0 and 2 hours post PDT (Figure 3.1), we have chosen 2 hours post PDT as the time point in these experiments. The combination of GR1 and F4/80 antibodies in the two color flow cytometry seemed to us the best solution for the identification of myeloid PDT induced host cell infiltration 71 cells that are not mature macrophages. The majority of cells stained as Gr1 + F4/80" can be assumed to be monocytes, since we know that tumor levels of other G r 1 + cells (neutrophils and other granulocytes) at that time interval are low (1-2%). In our experience, the Gr1 antibody served better for the identification of myeloid cells than the M a d ; not all G r 1 + cells stained positively for M a d . The representative examples of Hoechst 33342 fluorescence in cells obtained from differently treated tumors are shown in Figure 3.3, while the average values from groups of identically treated tumors are depicted in Figure 3.4. With control samples, it can be seen that around 3 5 % of cells from the undamped tumors were Hoechst 33342 positive, while only a small cell fraction from the clamped tumors were within the gate for Hoechst 33342 positive staining. It seems that the clamping itself induced a minor influx of Hoechst 33342 positive cells, due probably to a reaction stemming from temporary vessel occlusion. The results with the tumors growing in mice not injected with Photofrin exposed to the photodynamic light treatment were very similar to those with the control tumors. In the undamped PDT treated tumors, the percentage of Hoechst 33342 positive cells was similar to that seen with the undamped control tumors. In contrast, the level of Hoechst 33342 positive cells in the clamped PDT treated tumor (more than 2 0 % of the cells were newly infiltrated) was much higher than that observed with the non-treated clamped tumors. This finding demonstrates that PDT treatment induced infiltration of Hoechst 33342 positive cells from the circulation. Shuch was not the case with the light treatment in the absence of the photosensitizer. P D T induced host cell infiltration 72 log Hoechst fluorescence Figure 3.3 Distribution of Hoechst 33342 fluorescence in representative SCCVII tumors determined by flow cytometry. The tumors treated with PDT (as in Fig. 1) were excised at 2 hours after the termination of light treatment. The graphs show Hoechst 33342 fluorescence per cell in arbitrary units on the logarithmic scale (abscissa) and cell number on a linear scale (ordinate), a) Cells from an undamped tumor excised from a control mouse not injected with Hoechst 33342 (gate B defining Hoechst 33342 positive cells and gate E delimiting Hoechst 33342 negative cells were set in this graph for all the other cell samples); b) cells from an undamped control tumor excised from a mouse given Hoechst 33342; c) cells from a control tumor that was clamped before the mouse was given Hoechst 33342; d) cells from a photodynamic light treated tumor that was clamped before the mouse was administered Hoechst 33342, but which had not received Photofrin; e) cells from an undamped tumor treated with Photofrin based PDT that was growing in a mouse given Hoechst 33342; f) cells from a tumor treated with Photofrin based PDT that was clamped before the mouse was injected with Hoechst 33342. PDT induced host cell infiltration 73 4 0 r _co "c5 o a> 3 0 > 'co o a. 2 0 C O o CD O X g 1 0 o CD Q_ 0 I 1 I • Clamped • • Undamped Con t ro l Light only PDT treated Figure 3.4 Infiltration induced by PDT of Hoechst 33342 labeled cells from the blood into the SCCVII tumor. Average values are shown for the clamped and undamped tumors (shaded and unshaded columns, respectively) exemplified in Fig. 3.3. The bars are SD. . . The reconstruction of cellular composition of PDT treated tumor at 2 hours post treatment, showing the contribution of resident and newly infiltrating cells, is shown in the histogram section of Figure 3.5. From the data it is evident that Gr1 + F4/80" cel|s are the major component of the PDT induced infiltrate at this time interval. Approximately % of the Gr1+ F4/80" cells present in the tumor are those that invaded after PDT. PDT induced host cell infiltration 74 Hoechst positive cells Hoechst negative cells (2 hours post PDT) (2 hours post PDT) CD45" F4/80+ GR1+F4/80" Other CD45+ Figure 3.5 The content of various populations among tumor resident cells (Hoechst 33342 negative) and newly infiltrated cells (Hoechst 33342 positive) in SCGVII tumors at 2 hours post PDT. Relative contributions of these cell populations are shown in the pie graphs. The experimental details were as in Figs. 3.3 and 3.4. The bars show SD. PDT induced host cell infiltration 75 The pie graphs in Fig. 3.5 show the percent distribution of major cellular populations separately for newly infiltrated and resident cells. Over 8 0 % of newly infiltrated cells were Gr1+ F4/80', while - 1 0 % were F4/80+. The values for the percentage of the remaining Hoechst 33342 positive cells (other CD45+, CD45" ) were at the levels which fall within the experimental error, and the presence of these cells cannot be supported by statistical evaluation. The percent distribution of tumor resident cells (Hoechst 33342 negative) shows that the share of malignant cells (CD45") is very similar to that in the non-treated tumors (pie graph in Figure 4.1). In spite of different methods (indirect immunoperoxidase staining vs flow cytometry) and somewhat different combinations of antibodies used, the results suggest that the percent composition of resident immune cells has also not drastically changed at 2 hours post PDT compared to the non-treated tumor. Given the fact that the total cell yield was reduced by approximately half at 2 hours after PDT, these results indicate that the level of killing of different types of resident leukocytes and malignant cells was similar. 3.4.3 Cytotoxicity of TAMs after PDT The occurrence of a pronounced invasion of leukocytes into PDT treated tumor may, together with the other inflammatory changes, affect the activity of immune cells directed against the tumor. To test this, we harvested TAMs contained in SCCV I I tumor using a differential attachment technique employed in our earlier studies (Korbelik et al., 1991). Staining with monoclonal antibodies showed that over 9 0 % of the cells selected in this way were Mac1 + and - 7 0 % were F4/80+. The cytotoxicity of these cells against in vitro cultured S C C V I I malignant cells was examined. The results (Figure 4.6) show that TAMs harvested from a PDT treated SCCVI I tumor at 2 hours after the treatment PDT induced host cell infiltration 76 were almost five times more effective in the cytolysis of target cells than the TAMs from non-treated tumor. 3 0 r 2 5 S 2 0 CO <D CD C 15 -Iv 1 0 X CO 5 -0 T A M s from control tumour T A M s from PDT treated tumour Figure 3.6 Cytotoxicity of tumor associated macrophages (TAMs), selected from PDT treated and non-treated S C C V I I tumors, against in vitro cultivated SCCV I I cells. The cultured (malignant) SCCVI I cells, were labeled with [3H]thymidine and admixed with macrophages harvested from SCCVII tumors as described in Materials and Methods section. The treatment by Photofrin based PDT of SCCVII tumors used for TAMs isolation was as in Figure 3.1, with the tumor excision performed at 2 hours post PDT. The bars show SD. PDT induced host cell infiltration 77 3.5 Discussion Dramatic changes occur in the cellular composition of S C C V I I tumors treated with Photofrin based PDT. The most striking change is the massive invasion of neutrophils into the tumor that starts rapidly (within 2 min.) after the onset of photodynamic light treatment. Within 5 min., the neutrophil content in the treated tumor increased 200 fold, before decreasing once again to the levels seen in non-treated tumors between 2 and 4 hours after PDT. Mast cells were the other type of immune cells that started to invade the tumor very early after the onset of photodynamic light treatment (within 5 min.). The number of these cells in non-treated SCCV I I tumor is very low (about 1,000 cells in a 100 mg tumor) but their levels were more than 5 times higher between 1/2 and 4 hours after PDT. Mast cells are powerful mediators of inflammatory response, and their recruitment, in spite of the low tumor content, could have a pronounced effect in the PDT treated tumor. Selective labeling, of circulating leukocytes enabled the detection of the PDT induced infiltration by another myeloid cell type into treated SCCV I I tumors between 0 and 2 hours post PDT. These cells expressed the Gr1 antigen (a common myeloid cell marker) and most of them stained negatively io the F4/80 antigen (a marker for mature macrophages). Based on this fact, and oh the morphological examination, we concluded that the majority of these cells were most likely monocytes. The percentage of malignant cells in S C C V I I tumor decreased markedly during the photodynamic light treatment due to the massive neutrophil invasion. The proportion of these cells in the tumor decreased even further during the next several hours following light delivery because of newly infiltrating monocytic cells. However, the total number of malignant cells in the tumor PDT induced host cell infiltration 78 started to decrease substantially only after 2 hours post PDT, presumably due to rapidly developing tumor tissue ischemia (Henderson et al. 1985). The results based on the identification of resident malignant and immune cells in S C C V I I tumor in the experiments with Hoechst 33342 suggest that within the first two hours after PDT all these cells exhibit a similar sensitivity to the lethal effects of PDT. The only exception may be the inactivation of monocytes, suggested by the reduction in tumor content of "other myeloid cells" observed at 0 hours post PDT. A fast disappearance of newly infiltrated neutrophils from the treated tumor indicates that these cells were killed relatively quickly. These cells, as well as "other myeloid cells" might have been localized in the area of tumor vasculature already extensively damaged during the light treatment. In contrast, the percent distribution of cellular populations at 4 hours post PDT (pie graph in Figure 4.1) infers that newly infiltrated monocytes may outlive most of the tumor resident cells. Taken together, the above results portray a typical scenario of acute inflammatory infiltration of myeloid cells at the affected site. The rapid and massive neutrophil invasion is promptly accompanied by the arrival of mast cells. Massive release of chemotactic substances from degranulating mast cells (Kerdel et al. 1987), dying neutrophils, as well as from damaged membranes of tumor cells (alluded to in Introduction), is the probable impelling force behind another wave of infiltration, this time involving monocytes, that follows 1-2 hours later. There was no evidence for the participation of lymphocytes in these PDT induced tumor infiltration events during the observation period of this study (up to 8 hours post PDT). The inflammatory reaction in PDT treated tumor appears to have similarities with the cutaneous inflammation, identified as the main factor in PDT induced PDT induced host cell infiltration 79 skin phototoxicity, where neutrophils and mast cells also have a critical role (Lim 1989). The data demonstrating a pronounced increase in the tumoricidal activity of TAMs (Fig. 3.6) offers evidence that this inflammatory response is actually associated with the functional activation of immune cells. In a related study with cultured macrophages and tumor cells, we have demonstrated that in vitro PDT treatment of tumor cells (but not normal cells) potentiates their killing by macrophages (Korbelik and Krosl 1994). The implication is that potentially repairable damage induced by PDT in tUmor cells triggers macrophage mediated tumoricidal activity. Such activity of non-specific immune cells may lead in a later phase to the development of a T cell specific immune activity. The ingestion of PDT damaged or killed tumor cells by macrophages, which are antigen presenting cells, can result in tumor antigen presentation with the consequent induction of tumor specific immunity (Yamamoto et al. 1992). Such a development of PDT induced immunopotentiation is suggested by the work of Ganti and co-workers (Canti et al. 1994). A number of reports documented enhancement of tumor control by combining PDT with a variety of immunotherapy regimens (Myers et al. 1989; Bellnier 1991; Dougherty et al. 1992; Dima et al. 1994; Krosl and Korbelik 1994). This supports the idea that PDT induced immune reaction may be amplified and directed towards more effective tumor destruction. Much work remains to be done in exploring the effects of PDT on cellular composition of tumors treated with different Photofrin / light dose combinations, examining the effects in different tumor models and with other photosensitizers. Each of these aspects requires a substantial experimental effort. SPG potentiation of PDT 80 CHAPTER 4 Potentiation of photodynamic therapy by immunotherapy: the effect of schizophillan (SPG)1 4.1 Summary Treatment of squamous cell carcinoma (SCCVI I ) bearing mice with the immunostimulant Schizophillan (SPG) raised the relative content of M a d positive host cells infiltrating the tumor and increased Photofrin retention in these tumors. In vitro colony formation assay following Photofrin based photodynamic therapy (PDT) in vivo revealed a greater killing of tumor cells in the SPG pre-treated group, particularly pronounced when tumor excision was delayed until 8 hours after PDT. The tumor cure rate increased approximately three fold when PDT was preceded by the S P G therapy. In contrast, the administration of SPG after PDT had no benificial effect on tumor control. 4.2 Introduction Tumor destruction and cure after Photofrin based photodynamic therapy (PDT) is a result of several contributing factors, including the direct cytotoxicity to tumor cells, vascular stasis with the ensuing ischemia, and immune reaction (Bellnier and Henderson 1992). Primary events, which initiate changes in tumor microvasculature leading to irreversible vascular occlusion and eventually to tissue necrosis in PDT treated tumor (Fingar et al. 1990) have not been clearly identified. Early, not necessary lethal effects of PDT on multiple targets including endothelial cells (Ben-Hur et al. 1988; Sporn and Foster 1 The data presented in this chapter have been published in following paper: 1. Krosl, G. & Korbelik, M. (1994). Potentiation of photodynamic therapy by immunotherapy: the effect of schizophyllan (SPG). Cancer Letters, 84, 43-49. SPG potentiation of PDT 81 1992), mast cells and macrophages (Henderson and Bellnier 1989), and platelets (Fingar et al. 1990) may be involved. They trigger the release of powerful physiological mediators, including thromboxane and other eicosanoids, histamine, and various cytokines. In addition to their procoagulant and vasoconstricting activity, these mediators are the initiators of a strong local inflammatory response which is itself capable of destroying the affected tissue. The inflammatory reaction induced in a PDT treated tumor is characterized by a rapid and massive invasion of neutrophils occurring within minutes after the initiation of the light, treatment, and followed later by mast cel ls and monocytes/macrophages arriving from, the circulation (Krosl et al. 1995). The development of a strong inflammatory reaction in response to PDT suggests that the tumor infiltrating leukocytes may play a role in tumor eradication by PDT. The observation that tumoricidal activity of macrophages is markedly elevated when the target tumor cells are pretreated with PDT (Korbelik et al. 1993) suggests that the damaged membrane structures on PDT treated tumor cells can be recognized by the macrophages. The inflammatory reaction induced by PDT can also be a source of immune cell activation. In this respect, we have reported that tumor associated macrophages (TAM) isolated from the PDT treated tumors show increased tumoricidal activity compared to the TAM isolated from non-treated tumors (Korbelik et al. 1993). We have suggested that the effect of PDT can be potentiated by adjuvant immunotherapy (Korbelik et al. 1991). This is supported by the studies with several PDT-immunotherapy combinations reported by different authors (Myers et al. 1989; Bellnier 1991; Dougherty et al. 1992). In this work, we have examined the effect of a non-specific immunostimulant Schizophyllan (SPG) administered either before or after PDT. This glucan isolated from Schizophyllum comunae (Norisuye et al. 1980) is used in clinical cancer SPG potentiation of PDT 82 therapy (Okamura et al. 1986; Nakano et al. 1993). Leukocytes isolated from mice treated with S P G showed increased response to cytokines and to the mitogen concavalin A (Tsuchiya et al. 1989). It was found that S P G directly stimulates macrophage-mediated antitumor activity (Sugawara et al. 1984). The improved tumor control seen with mice treated with radiotherapy in combination with SPG correlated with the increased infiltration into the tumor of immune cells, especially macrophages (Inomata et al. 1987). 4.3 Materials and methods 4.3.1 Tumor model, drugs C3H/HeN female mice were used for experiments when they were 10-12 weeks old. Murine squamous cell carcinoma (SCCVII) was maintained by bi-weekly intramuscular passages as described, elsewhere (Korbelik 1993). For experiments the tumors were implanted subcutaneously by injecting 3 x 10^ cells in a 0.03 mL volume over the sacral region of the back of the mouse. PDT was performed when the tumors reached 6-8 mm in size, measuring the largest diameter. Schizophillan (SPG), produced by Kaken Pharmaceutical Ltd. (Tokyo, Japan) was kindly provided by Dr. T. Inomata. Mice treated with S P G were given 9 consecutive daily doses of 15mg/kg of the drug intramuscularly injected into the hind leg. For each type of experiment, a control group with mice receiving the treatment with SPG-free saline was also examined. In all the cases, treatment with the control saline showed no effect on measured parameters. Photofrin was obtained from QuadraLogic Technologies Phototherapeutics Inc. (Vancouver, BC., Canada). Mice were administered a dose of 25 mg/kg of Photofrin intravenously 24 hours before the dose of 60 J / c m 2 (630 ± 10 nm) SPG potentiation of POT : 83 light was delivered from a tunable light source equipped with the 1kW Xenon bulb (model A 5000, Photon Technology International Inc.). The power output was 30 mW. The light treatment was delivered either 48 hours after the last S P G injection or 24 hours before the first daily dose of SPG. In the former case, the SPG therapy was initiated 4 days after the tumor inoculation. Following PDT, the tumors were observed every second day and the 3 orthogonal diameters were measured using a caliper. Tumor volume was 4 X 7t calculated by a formula V = x(axbxc) where a, b, and c represent the 3 6 orthogonal diameters. The mice were sacrificed if one of the diameters reached 12 mm. Each treatment group consisted of at least 6 mice. The experiment was done three times. 4.3.2 Analysis of immune cell infiltration Tumors from mice treated with SPG and from the non-treated animals were excised, minced, and then enzymatically digested for 30 min. at 37°C with 0 .02% Collagenase (type IV, Sigma Chemical Co. , St. Louis, Mo,, USA), 6 . 0 1 5 % DNAse (type I, Sigma), and 0 . 1 % dispase (Boerhinger, Mannheim, Germany) (McBride et al. 1992). The cell suspensions obtained in this way were washed twice in Hank's balanced salt solution (Sigma) supplemented with 2 % Fetal Bovine Serum (FBS, HyGlone Laboratories Inc., Logan, UT, USA). The cytospin preparations prepared from the cell suspensions were fixed with acetone and stained for M a d or CD45 (T200) antigen using, an indirect immunoperoxidase technique (Korbelik et al. 1992).. Hybridoma TIB 128 producing anti-Mad antibody was purchased from A T C C (Maryland, USA). Monoclonal antibody YE1.21 directed against the T-200 antigen (CD45) was generously provided by Dr. Fumio Takei (Terry Fox Laboratory, B.C. Cancer Research Centre). The secondary antibody used was a rabbit anti-rat SPG potentiation of PDT 84 IgG conjugated with horseradish peroxidase (Sigma), diluted as per manufacturer specifications. At least 200 positive cells were scored on each Slide. 4.3.3 Determination of Photofrin levels in tumor tissue/cells The excised tumors were weighed; minced and left at 60°C overnight in ScintiGest (Fisher Scientific Co., Fair View, NJ, USA). Two mL of ScintiGest plus 0.2 mL of double distilled water was used per 0.T mg of tumor tissue. Alternatively, the tumors were first dissociated into single cell suspensions by the enzymatic digestion procedure as for the analysis of immune cell infiltration. The cells were counted, pelleted by centrifugation and left at 60°C overnight in 1:10 (vol/vol) ScintiGest dilution in distilled water. Photofrin concentration in lysates of tumor tissue or cells lysates was determined by a flubrometric assay as described.earlier (Korbelik et al. 1991). Lysates obtained from the tumors of mice that had not received Photofrin served as the background contrbls. 4.3.5 Clonogenic assay Clonogenicity of SCCV I I cells isolated from the tumors was assessed using a soft agar clonogenic assay (Chaplin et al. 1986; Chaplin et al. 1987; Korbelik et al. 1992). The tumors were excised either immediately following PDT or at 8 hours post PDT and cell suspensions were prepared by the enzymatic digestion, procedure. Cel l viability and cell yield per gram of tumor tissue were determined. A known number of cells resuspended in Eagle's Minimal Essential Medium (GIBCO Laboratories, Grand Island, NY, USA) supplemented with 1 6 % of FBS was plated on the soft agar layered into the petri dish. SPG potentiation of PDT 85 Tumor cell colonies were counted two weeks later and the clonogenicify of the plated cells was. determined based on the plating efficiency and cell yield (Henderson et al. 1985). The results are presented as colony forming units per gram of tumor (CFU), which is derived as follows: CN Cell yield (CY) for the tumour is calculated by the formula: CY = where CN is total number of cells, and TW is tumour weight in grams. Plating efficiency (PE) is derived: PE = N c a l Ncell . where NCoi is number of colonies, and N c e i i represents number of cells plated. PEt Surviving fraction SF is then calculated: SF = —-PEc where PEt represents plating efficiency of treated sample, and PE C is plating efficiency of a control. Colony forming units per gram of tumor tissue are then calculated: CFU = SFx^^L CYC where CYt is cell yield of treated sample and CY C is cell yield of a control. 4.4 Results and discussion 4.4.1 Photofrin tumour levels The effect of S PG therapy on Photofrin level in S C C V I I tumor determined 24 hours after the administration of the photosensitizer (25 mg/kg, i.v.) is shown in Table 4.1. It can be seen that the S P G treatment results in a small, but statistically significant increase in Photofrin content per gram of tumor tissue. An even greater effect of S PG was seen with another mouse model, Lewis lung carcinoma growing subcutaneously in C57/BI mice (Korbelik et al. 1991). This study has concentrated on S C C V I I tumor, since Lewis lung carcinoma is not SPG potentiation of PDT 86 { • • • suitable for the experiments with tumor cure as an endpoint due to the lung metastases that develop in the early stages of the growth of this tumor. Table 4.1 The effect of SPG on Photofrin content in SCCVII tumor |ig Photofrin per |ig Photofrin per 10^ gram of tumor cells. Control tumors 7.58 + 0.28 0.251 ±0.031 S P G treated tumors 9.71 ±0.13* 0.288 ±0.036 Tumor bearing mice received 9 daily injections of SPG (15 mg/kg i.m.) or the control solvent. Photofrin (25 mg/kg) was administered one day after the last SPG treatment. The mice were sacrificed 24 hours after receiving Photofrin. The tumors were excised, minced and dissolved either immediately in ScintiGest, or they, were first enzymatically dissociated into single cell suspensions. Photofrin levels in the tumor tissue or in cell lysates were .determined using the fluorometric assay. The results of a representative experiment (mean ± SD, six mice per group) are shown. * = significantly different (p < 0.001) then the control group. The results of measurements of Photofrin content per cell are also shown in Table 4.1. In this case, S C C V I I tumors were minced and cell suspensions obtained by the enzymatic digestion procedure. In contrast to the Photofrin tumor tissue levels, the average cellular level of the photosensitizer was not significantly increased in the S P G treatment group. The cell yield per gram of tumor tissue does not seem to be affected by the S P G therapy, as shown in Table 4.2. However, the composition of cell populations in the S C C V I I tumor may have changed as a consequence of the immunomodulating effect of SPG . This could affect the average Photofrin tumor level, because the level of this photosensitizer is several times higher in tumor associated macrophages (TAM) than in parenchymal malignant cells of SCCVI I tumor (Korbelik et al. 1991). 4.4.2 Leukocyte infiltration of SCCVII tumor Cellular composition of S C C V I I tumors was analyzed using monoclonal antibodies that can identify major populations of host immune cells infiltrating the tumor. The results (Figure 4.1a) show that the tumors treated with S P G SPG potentiation of PDT 87 have an increased content of cells that are positively stained with the antibody to M a d , the antigen that is specific for macrophages and granulocytes. This may reflect either increased expression of the integrin identified by M a d by cellular populations already present in the tumor,"or S P G induced infiltration of M a d positive cells into the tumor. The latter was reported to be the case following the S P G treatment of the MM46 tumor in C3H mouse (Inomata et al. 1987). The analysis using a pan-leukocyte marker CD45, also shown in Figure 4.1a, suggests that the total immune cell content in the SCCV I I tumor was not significantly affected by the SPG therapy. In our related studies (Korbelik et al. 1992), we have observed that PDT treatment of SCCV I I tumor has a marked effect on tumor infiltrating leukocyte populations. An example of this can also be seen in Figure 4.1b. In this case, the tumors were excised at 4 hours after the photodynamic light treatment; at later time intervals rapidly developing necrosis may affect the quality of staining for membrane antigens. It is shown that the content of both CD45 positive and M a d positive cell populations at 4 hours post PDT is increased when compared to the levels in the PDT non-treated tumor depicted in Figure 4.1a (p<0.01). With the S P G treatment before Photofrin based PDT, the M a d positive cell population reached 6 0 % of the total tumor cell content. Thus, the incipient M a d + level in un-treated tumors was more than doubled by the apparently additive effect of S PG and PDT. The levels of CD45 positive cells were again, on average, slightly higher but not statistically different, than in the corresponding S P G non-treated group (Figure 4.1b). The administration of Photofrin without the light treatment or light treatment without Photofrin showed no effect on the tumor leukocyte content (not shown). SPG potentiation of PDT 88 8 0 r 6 0 -_C0 "53 o © 4 0 CO o Q _ o 2 0 0 CD45+ Mac-1 + 7 5 -o "E _co o > CO o Q . 5 0 -2 5 -0 B CD45+ Mac-1 + Figure 4.1 The effect of SPG therapy on immune cell content of PDT non-treated (a) and PDT-treated (b) SCCVII tumor. Tumor bearing mice received 9 daily injections of SPG (15 mg/kg, i.m). Mice treated with PDT were given Photofrin (25 mg/kg, i.v.) 24 hours after the last SPG injection which was followed by 60 J/cm 2 light treatment 24 hours later. The tumors were excised either 24 hours after the last SPG injection (A), or 4 hours after the.photodynamic light treatment (B). The percentage of cells positive to CD45 and M a d antigens was determined using the indirect immunoperoxidase staining. * = significantly different than the No SPG group (p < 0.05). SPG potentiation of PDT 89 Using a flow cytometry technique described earlier (Korbelik et al. 1991) we have also analyzed Photofrin levels in Fc receptor (FcR) positive cell population (predominantly TAM) and in FcR negative cell population (predominantly malignant cells) of SCCVI I tumor. The results revealed no significant difference (p > 0.05) for both of these populations between the S P G treated and S P G non-treated groups (Table 4.2). Table 4.2 The effect of SPG on Photofrin content in FcR" and FcR+ cell populations of SCCVII tumor Photofrin content [u.g/106 cells] FcR" FcR+ Untreated 0.251 ± 0.031 0.727 ± 0.051 S P G 0.288 ±0.036 0,815 ±0.093 Tumor bearing mice received 9 daily injections of SPG (15 mg/kg i.m.) or the control solvent. Photofrin (25 mg/kg) was administered one day after the last SPG treatment. The mice were sacrificed 24 hours after receiving Photofrin. The tumors were excised, minced and enzymaticaliy dissociated into single cell suspensions. Photofrin levels in the FcR + and FcR" cell populations were determined as explained in Chapter 2. The results of a representative experiment (mean + SD, six mice per group) are shown. 4.4.3 The effect of SPG on tumor cell clonogenlcity after PDT Survival of tumor cells determined by colony formation assay in vitro after Photofrin based PDT in vivo is shown in Table 4.3. The tumors were excised either immediately after the termination of the light treatment, or at 8 hours post PDT. In the former case data reflect direct killing of tumor cells by PDT, while the excision of tumors with 8 hour delay gives information on indirect lethal effects of PDT (Henderson et al. 1985). Direct killing: The cell yield with S C C V I I tumors excised immediately after the PDT was not different than with PDT non-treated tumors. However, the plating efficiency of these cells decreased from 0.26 to 0.13 due to the direct killing of tumor cells by PDT. The plating efficiency of tumor cells in SPG treated SPG potentiation of PDT 90 group was not different than with SPG non-treated tumors; however, the yield of these cells was lower, which resulted in a somewhat lower value for the relative CFU/g of tumor. Indirect killing: Due to substantial decrease in cell yield, only approximately 2 % of tumor cells from the SPG non-treated group survived at 8 h post PDT. The cell yield was even more drastically reduced in the SPG treated group, which resulted in the additional decrease in survival of tumor cells (p < 0.01, compared to the S P G non-treated group). This data suggests that the effect of S P G on PDT inflicted indirect killing of tumor cells is much more pronounced than on the direct killing. Table 4.3 The effect of SPG pretreatment on clonogenicity of S C C V I I tumor cells isolated from PDT-treated subcutaneously growing tumors. Treatment Plating efficiency cell yield /g of tumor relative CFU/g of tumor Non-treated controls 0.26 ±0.04 1.3 ± 0.2x10 8 1.00 S P G control 0.22 ±0.08 1.2 ± 0.3x10 8 0.78 ±0.1 PDT only, Oh 0.13 ±0.07 1.4 ± 0.2x10 8 0.54 ± 0.07 SPG+PDT, Oh 0.13 ±0.07 8.0 ± 2.0x10 7 0.39 ±0.12 PDT only, 8h 0.15 ±0.03 4.6 ± 2.5x10 8 0.02 ± 0.011. SPG+PDT, 8h 0.10 ±0.02 1.2 ± 0.7x10 6 * 0.0036 ± 0.0022* Tumor bearing mice received 9 daily injections of SPG (15 mg/kg i.m.). Photofrin (25 mg/kg) was administered one day after the last SPG treatment. Light treatment (60 J/cm2) was performed 24 hours after the Photofrin injection. Tumors were excised at 0 or 8 hours after PDT and enzymatically disaggregated as described in Materials and Methods. Known number of viable cells were plated for the soft agar clonogenic assay and the colonies were scored 2 weeks later. Twelve mice were included in each group. The results are mean ± SD. * = significantly different than the PDT only group (p < 0.01). SPG potentiation of PDT 91 4.4.4 The effect on tumor recurrence The effect of S P G therapy administered either before or after Photofrin based PDT with S C C V I I tumor regrowth or cure as endpoint is shown in Figure 4.2. The SPG therapy alone has no significant effect on tumor growth (Figure 4.2 insert). The PDT treatment alone, under the conditions of this experiment, gave a 1 7 % cure rate. The cure in this assay is defined as no sign of tumor regrowth at 90 days after the treatment. The experience with SCCV I I tumor model is that no tumor recurrence after this time period is equivalent to tumor cure. The SPG therapy administered before PDT markedly enhanced the effect of PDT, increasing the cure rate to around 5 0 % (p < 0.05, using Wilcoxon Rank Sum test). In contrast, the SPG treatment administered after PDT had no significant effect on the outcome of PDT. SPG potentiation of PDT 92 1 0 0 r 8 0 -CD O CD CD 1 000 C O E E " 1 0 0 CD E o > 10 jmor 1 I- 1 i . i . * . i . i . i . i . i . i 1 3 5 7 9 1 1 1 3 1 5 1 7 Days after implant 6 0 -~ 4 0 -2 0 -0 SPG + PDT PDT only PDT + SPG 0 2 4 6 8 10 12 14 16 1 8 2 0 2 2 9 0 9 2 Days after PDT Figure 4.2 The effect of SPG treatment on tumor eradication by PDT. Mice bearing SCCVII tumor received Photofrin (25 mg/kg, i.v.) followed 24 hours later by. 60 J/cm 2 light treatment. The SPG therapy (9 daily injections of 15 mg/kg, i.m.) was administered either before PDT or it was initiated 24 hours after PDT. The PDT only group is also shown. The insert shows growth of non-treated tumors (*) and tumors treated with the SPG protocol only (+); the bars are + SD. Eighteen mice (pooled from the three experiments) were included in each treatment group. SPG potentiation of PDT 93 The experimental evidence presented in this Chapter demonstrates that immunomodulating action of SPG potentiates the effect of PDT. This agent increased primarily the extent of indirect killing of tumor cells by PDT, probably by elevating the level of tumor infiltrating macrophages before and after PDT, which may have enhanced the tumoricidal activity of these cells and augmentated the non-specific immune reaction. This enhancement, however, appears to require prolonged treatment with SPG before PDT, as no effet could be detected when SPG was administered after the photodynamic treatment. The results of this work support the suggestions on the importance of immune reaction in tumor control by PDT. Immunotherapy seems to be a promising approach for improving the clinical effectiveness of PDT in cancer treatment. The effect of GM-CSF on PDT 94 CHAPTER 5 Potentiation of photodynamic therapy elicited antitumor response by localized treatment with granulocyte-macrophage colony stimulating factor3 5.1 Summary Murine squamous cell carcinoma ( S C C V I I ) cells were genetically engineered to produce murine granulocyte-macrophage colony stimulating factor (GM-CSF). GM-CSF immunotherapy, based on the peritumoral injection of lethally irradiated GM-CSF producing cells, was examined as adjuvant to photodynamic therapy (PDT) treatment of this tumor. The GM-CSF immunotherapy administered 3 times in 48 h intervals starting 2 days before the light treatment substantially improved the curative effect of Photofrin mediated PDT. A comparable effect of GM-CSF immunotherapy was observed in the combination with benzporphyrin derivative mediated PDT. The tumor localized GM-CSF immunotherapy alone had no obvious effect on the growth of parental S C C V I I tumor. This treatment did not significantly alter the differential peripheral white blood cell count, and appeared not to affect tumor leukocyte infiltration. However, GM-CSF treatment did increase the cytotoxic activity of tumor associated macrophages against S C C V I I tumor cells. It appears, therefore, that tumor localized immune stimulation by GM-CSF amplifies PDT induced immune reaction, which has a potentiating effect on tumor control. 3The results presented in this chapter have been published in following paper: 1. Krosl, G., Korbelik, M., Krosl, J., and Dougherty G.J. (1996) Potentiation of photodynamic therapy-elicited antitumor response by localized treatment with granulocyte-macrophage colony stimulating factor. Cancer Res 56: 3281-3286. The effect of GM-CSF on PDT 95 5.2 Introduction Eradication of tumors treated by PDT is a result of combined effects on both tumor cells and various host derived cell types (Henderson and Dougherty 1992). In addition to direct killing induced by photooxidative damage to vital cellular structures (Penning et al. 1994), tumor cells appear to be inactivated not only by ischemia secondary to the damage to tumor vasculature, but also by the integrated tumoricidal activity of non-specific and specific immune effector cells (Korbelik and Krosl 1996). The host response triggered by the PDT treatment is dominated by a strong tumor localized acute inflammatory reaction associated with the functional activation of tumor resident and newly arrived leukocytes (Chapter 3 and (Krosl et al. 1995)). Neutrophils, mast cells, monocytes and macrophages have been suggested to participate in the antitumor activity in this early, phase post PDT treatment (Krosl e t a l . 1995; Korbelik and Krosl 1996). The inflammation of tumor tissue induced by PDT is accompanied by intense tumor destruction followed by the.release and phagocytosis of tumor cell debris. This creates conditions for the processing and presentation of tumor antigens by activated professional antigen presenting cells (macrophages) resulting in development of tumor specific immunity (Yamamoto et al. 1992; Korbelik and Krosl 1996). Indeed, .experimental evidence obtained in our laboratory (Korbelik et al. 1996) and elsewhere (Canti et al. 1994) supports the hypothesis that specific T cell mediated immune responses facilitate the eradication of tumors following PDT. Tumor regression induced by radiotherapy, chemotherapy or surgical excision is generally not associated with induction of systemic immunity. In contrast, PDT appears to have the capacity to reset the tumor-host relationship from tumor dominated to antitumor oriented. This justifies the development of specific strategies for the potentiation of the antitumor effect of PDT by adjuvant The effect of GM-CSF on PDT 9.6 immunotherapy. One obvious type of immunofherapeutic intervention that could be considered is to augment the activity of myeloid effector cells at the treated site. Neutrophils which massively and rapidly accumulate in PDT treated tumors (Krosl et al. 1995), have the capacity to inflict substantial damage on tumor tissue by. releasing reactive oxygen metabolites and a number of other tissue destructive mediators, or by attracting other immune cells to the tumor site. Macrophages isolated from PDT treated tumors were shown to exhibit elevated tumoricidal activity (Krosl et al. 1995). Increased macrophage activity after PDT in vitro has also been documented (Yamamoto et al. 1991; Yamamoto et al. 1992). The activation of these cells by systemic treatment by a specific macrophage activating factor generated from vitamin D 3 binding protein was shown to potentiate the curative effect of PDT (Korbelik et al. 1995). Macrophages were shown to release tumor necrosis factor-alpha (TNF-a) following PDT treatment (Evans et al. 1990). In addition, macrophages appear to preferentially recognize and destroy PDT treated tumor cell targets (Korbelik and Krosl 1994). Cross talk among granulocytes, macrophages and lymphocytes, in which the participants receive and/or deliver regulatory messages through the/ secretion of cytokines and other regulatory factors builds up and amplifies immune reaction (Colombo et al. 1992). The localized production of some cytokines within the tumor microenvironment can prevent or inhibit tumor growth by a T-cell independent mechanism, while other cytokines can stimulate a potent systemically-acting T cell mediated antitumor immune response (Tepper and Mule 1994). One of the latter cytokines is granulocyte-macrophage colony stimulating factor (GM-CSF), which is a key regulator controlling the maturation and function of granulocytes and monocytes/macrophages (Metcalf 1991), This glycoprotein stimulates the proliferation and differentiation of dendritic and other The effect of GM-CSF on PDT 97 antigen presenting cells, and is a potent enhancer of antigen presentation by these cells (Grabbe et al. 1992; Grabbe et al. 1994; Sallusto and Lanzavecchia 1994). The protection against challenge with parental tumor induced by immunization with GM-CSF producing tumor cell vaccine was clearly shown to be mediated by CD4+ and CD8+ T cells (Dranoff et al. 1993; Levitsky. et al. 1994). This cytokine is one of the immunotherapeutic agents whose effectiveness is currently under intense investigation both clinically and with in vivo tumor models (Dranoff et al. 1993; Tepper and Mule 1994). Since systemic treatment with high doses of GM-CSF may lead to profound perturbations in hemopoiesis and induce serious side effects (Metcalf 1991), and intravenously injected GM-CSF has relatively short half-life (~2 hours), it would be preferable if a means could be developed to achieve a sustained local release of this cytokine at the tumor site. This can be accomplished by the constitutive release of GM-CSF by peritumorally injected lethally irradiated tumor cells transfected with the gene encoding this cytokine (Dranoff et al. 1993). Using such an approach, we"introduced the murine GM-CSF cDNA into SCCV I I tumor cells (mouse squamous cell carcinoma) and, in combination with PDT, these genetically engineered cells were employed for the GM-CSF immunotherapy of mice bearing the parental non-infected SCCV I I tumor. -Data observed indicate that S C C V I I tumor localized GM-CSF treatment markedly elevates the effectiveness of PDT in this tumor model. 5.3 Materials and Methods 5.3.1 Tumor model, cell lines The poorly immunogenic murine squamous cell carcinoma SCCV I I (Suit et al. 1985) growing in C3H/HeN mice (10-12 weeks old females) was used in this The effect of GM-CSF on PDT 98 study as a tumor model. Maintenance and implantation of subcutaneously growing SCCV I I tumors has been described in detail in Chapter 3. S C C V I I tumor cells were cultured in Dulbeco's modified Eagle's medium (DMEM) containing a high glucose concentration (4.5 g/L) with 10 % fetal bovine serum (FBS; HyClone Laboratories Inc., Logan, UT, USA). The ecotropic GP+E-86 retrovirus packaging cell line (Markowitz et al. 1988) kindly provided by Dr. A. Banks (Columbia University, New York, NY, USA) was maintained in DMEM with high glucose content and 1 0 % newborn calf serum (HyClone), supplemented with 15 |j,g/mL hypoxanthine, 250 (ig/mL xanthine, and 25 |ig/mL mycophenolic acid (HXM selective medium). B6SutA, an IL-3 and/or GM-CSF dependent murine hematopoietic cell line was grown in RPMI-1640 media containing 1 0 % FBS (both HyClone) and 10 ng/mL of C O S cell derived IL-3 (produced in house at the terry Fox Laboratory). The FBS used for growing B6SutA cells was pretested for its ability to support growth and differentiation of murine bone marrow cells. 5.3.2 Generation of GM-CSF retroviral vector The murine retroviral vector used was based on upon the JzenTkneo vector described elsewhere (Damen et al. 1992). Briefly, a 495 base pair (bp) Stul-Hinfl fragment containing the entire coding region of murine GM-CSF cDNA was isolated from the plasmid pGM19 GM-CSF (kindly provided by Dr. M Gorth; Ludwig Institute for Cancer Research, Melbourne, Australia), and blunted and subcloned into a Smal site of pAX114 (Kay and Humpries 1991). The GM-CSF cDNA was then reisolated as 507 bp Sai l fragment and subcloned into the Xhol site of JzenTkneo upstream of the neo r gene, such that GM-CSF cDNA expression is controlled by the myeloproliferative sarcoma long terminal repeat (LTR), with an expected chimaeric GM-GSF/neo r mRNA of 3.5 kb. All restriction The effect of GM-CSF on PDT 99 enzymes used were from Canadian Life Technologies (Burlington, ONT, Canada). 5.3.3 Virus production and infection of SCCVII tumor cells The general strategy used to generate retrovirus producing cell lines was as outlined by Miller et al. (Miller et al. 1985). JzenGM-CSF Tkneo plasmid DNA was transfected by calcium phosphate precipitation method (Ausubel et al. 1994) into GP+E-86 cells. Transfected cells were selected based on their. acquired resistance to neomycin by incubation in DMEM containing 0.8 mg/mL (active weight) of neomycin analog G-418 (Canadian Life Technologies) and were subsequently used as producers of recombinant retroviral particles encoding GM-CSF. Subconfluent monolayers of GP+E-86/JzGM-CSF and control GP+E-86/Jzneo cells were washed twice with DMEM, overlaid with DMEM + 1 0 % FBS and irradiated to 12 Gy, using an x-ray source (Philips RT250, 250 kV, 0.5 mm Cu). Virus containing cell culture supernatant was collected 24 hours later, filtered through low protein binding filters (0.2 \im, Mil lex-G; Mill ipore, Mississauga, ONT, Canada) and immediately used for infection of S C C V I I tumor cells. Low density cultures of SCCV I I tumor cells were infected by three consecutive 8 h incubations with virus containing supernatant in the presence of 6 u.g/mL polybrene. After the third incubation, the cells were washed twice with DMEM and cultured in DMEM + 1 0 % FBS for additional 24 h. The infected cells were then selected in medium containing 0.8 mg/mL of G-418. Individual antibiotic resistant colonies were scrapped off the dish, expanded in G-418 containing medium and tested for production of GM-CSF. The effect of GM-CSF on PDT 100 5.3.4 Northern blot analysis SCCVII/JzGM-CSF cells (infected with JzenGM-CSFTkneo retrovirus) and SCCVII/Jzneo cells (infected with JzenTkneo retrovirus) were grown in 10 cm tissue culture dishes in the absence of selection until they reached 80-90% confluence. The cells (approximately 5x10 6 ) were lyzed in TRIzol reagent (Canadian Life Technologies) and total cellular RNA was isolated following the manufacturers recommendations. For Northern blot analysis, 10 jig of total RNA were separated on 1% agarose, 5% formaldehyde gel and transferred by blotting to a Zetaprobe nylon membrane (BioRad, Mississauga, ONT, Canada). The membrane was then prehybridized for 1 h in 5 0 % formaldehyde, 0.5 mmol/L N a H 2 P 0 4 , 2.5 mmol/L EDTA, 5% sodium dodecyl sulfate (SDS), and 1 mg/mL bovine serum albumin (BSA) at 42°C and subsequently hybridized overnight under the same conditions with cDNA probes that were 3 2 P labeled by the random primer method (Feinberg and Vogelstein 1984). Membranes were washed twice at 55°C, first with 2xSSPE, 0 .3% SDS (2xSSPE: 0.3 mol/L NaCI, 20 mmol/L N a H 2 P 0 4 ) 2 mmol/L EDTA; pH 7), then with 1x SSPE , 0 .5% SDS , and finally with 0.3x S SPE , 1%SDS. The membranes were then exposed to Kodak X-Omat AR film (Eastman-Kodak, Rochester, NY, USA). 5.3.5 GM-CSF assay 1x10 6 SCCVII/JzGM-CSF and SCCVII/Jzneo cells were incubated in 6 c m , tissue culture dishes in 5 mL of DMEM + 1 0 % FBS in the absence of selection for 24 h. The media were then harvested, and filtered through 0.22 urn Millex-G filters. Serial 5 0 % dilutions of this conditioned medium were prepared by diluting the culture supernatant with RPMI medium containing 0 . 1 % BSA (Stem Cell Technologies, Vancouver, BC, Canada). The effect of GM-CSF on PDT 101 Exponentially growing B6SutA cells were washed twice with RPMI medium containing 0 . 1 % BSA, and 2x10 4 cells in 80 uL of RPMI with 0 . 1 % BSA were dispensed per U shaped well of 96 multiwell plate (Linbro: (CN, Mississauga, ONT, Canada). Serial dilutions of conditioned media (20 u.L) were added to duplicate wells to give a final volume of 0.1 mL/well. After 18 h incubation at 37°C, io |iL of a 50 uCi/mL solution of [3H]-thymidine (2.0 Ci/mmol; NEN , DuPont Canada Inc.) diluted in RPMI medium was added to each well to give the final activity of 1 u,Ci/well. After 2 h incubation at 37°C, the contents of the wells were harvested onto filtermats and counted using LKB Betaplate Harvester and Liquid Scintillation Counter (LKB Wallac, Turku, Finland). 5.3.6 Photodynamic treatment Tumors were used for experiments one week after inoculation, at which point their largest diameter was 5-6 mm. During the light treatment the mice were restrained unanasthetized in specially designed holders. Two different photosensitizers were used for photodynamic therapy of SCCV I I tumors. Photofrin based PDT consisted of intravenous administration of 10 mg/kg of Photofrin® (porfimer sodium) (QuadraLogic Technologies Phototherapeutics Inc., Vancouver, BC, Canada) followed by 150 J/cm 2 of 630 nm light delivered 24 h later. Benzporphyrin derivative (BPD) based PDT was performed by intravenous administration of 2.5 mg/kg of BPD monoacid ring A (veterporfin, a liposomal preparation also from QuadraLogic) followed by 690 nm light (120 J/cm 2) 4 hours later. The light treatment for Photofrin-based PDT (630+10 nm) was delivered from a tunable light source equipped with 1 kW xenon bulb (model A 5000; Photon Technology International Inc.) through a 5 mm core diameter light guide (2000A, Luminex, Germany). The. power density at the treatment area was 75 The effect of GM-CSF on PDT 102 mW/cm2. The BPD-based PDT was performed using a 250 mW diode laser (SDL7422-H1, Spectra Diode Labs, San Jose, CA, USA), with 690 nm light delivered through a 2000A light guide. The power at the end of the light guide was 110 mW, and the power density at the irradiated area was 140 mW/cm2. The thickness of treated tumors never exceeded 3.5 mm, which ensured their transillumination by the monodirecfional light beams of 630 or 690 nm. In conjunction with the tissue light scattering, this achieved a high degree of light dose homogeneity within the treated lesions. The PDT light doses used with both photosensitizers were to assert, for the PDT alone groups, the level of tumor response that is close to the curative treshold (0-10% cures). Tumor temperature never reached more than 39°C during the light treatment as measured by a hypodermic thermocouple (YSI, Yellow Springs, OH , USA). Mice were observed for tumor regrowth every second day. Three orthogonal tumor diameters were measured by a vernier caliper. The mice were considered cured if they were tumor free 90 days after the PDT. 5.3.7 GM-CSF immunotherapy The cultures of cytokine producing SCCV I I tumor cells were expanded and harvested when they reached approximately 80-90% confluence. Cells were then washed three times with 50 mL of cold (0°C) HBSS, to remove serum proteins. The cell concentration was adjusted to 1x10 8 cells/mL in HBSS. The cell suspensions in 5 mL plastic tubes held.on ice were.then irradiated with x-rays to 50 Gy. Aliquots (0.1 mL) of this suspension, containing 1x10 7 irradiated cells were injected into each mouse. In all protocols, 1x10 7 irradiated SCCVII/JzGM-CSF or SCCVII/Jzneo cells were injected per application. Mice were briefly anesthetized by inhalation of Metofane (Janssen Pharmaceutica, Mississauga, ONT, Canada), and the The effect of GM-CSF on PDT 103 subcutaneously growing tumor was lifted and the cell suspension slowly injected under the tumor using 26 gauge needle. The tumor was then gently released such that no cells were lost through the needle track. The first treatment was given 48 hours before light treatment (i.e. 24 h before Photofrin administration), second treatment was immediately following the light delivery, and the last treatment was at 48 h after PDT. When the tumor became unpalpable at 48 h post PDT, the cells were injected subcutaneously as close as possible to the tumor site. 5.3.8 Macrophage cytotoxicity Mice bearing SCCV I I tumors were injected peritumorally with IxTO 7 lethally irradiated SCCVII/JzGM-CSF or SCCVII/Jzneo cells. The mice were sacrificed 48 hours later, tumors excised and dissociated into single cell suspension by an enzymatic procedure used routinely in our laboratory (Korbelik 1993). Tumor associated macrophages (TAMs) were isolated from this suspensions by a modification of the differential attachment procedure described previously (Korbelik et al. 1991; Krosl et al: 1995). Briefly, the ability of macrophages to attach firmly to a plastic substrate much more rapidly than the other cells present in S C C V I I tumor was exploited for obtaining highly purified TAM populations. Tumor cell suspension was resuspended in RPMI medium supplemented with 1 0 % FBS, and aliquots containing 6 .9x10 5 (of which approximately 2x10 5 were expected to be TAMs) were transferred into wells of a 24-well plate. After 30 min. incubation at 37°C, the medium was replaced with 0.5 mL Trypsin-EDTA solution (Sigma Chemical Co., St. Louis, MO, USA) and incubated for 30 seconds at room temperature. This was followed by washing the attached cells vigorously three times with 1 mL of HBSS . The washout containing removed cells was collected and pooled for each sample. The effect of GM-CSF on PDT 104 The number of TAMs remaining in wells was determined by subtracting the number of cells removed by washing from the number of plated cells. The TAMs were then incubated with lipoplysaccharide (LPS) from E. coli 011:b4 (Sigma) at 10 u.g/mL in RPMI + 1 0 % FBS for 24 h. The target cells (SCCVII tumor cells maintained in vitro for 6-8 weeks) were added to the wells containing TAMs at a ration 20:1. Before that, the target cells were pre-labeled by exposure to [3H-methyl]thymidine (2.0 Ci/mmol, NEN) at 2 |iCi/mL for 24 h in cell growth medium and then pre-incubated for 4 h with Actinomycin-D (1.5 ixg/mL, Sigma). The samples with effector and target cells were incubated 72 h at 37°C and the supernatants were then collected for the measurement of radioactivity released from killed SCCV I I cells as described in detail earlier (Korbelik and Krosl 1994). All experiments were carried out in quadruplicate (four wells for each TAM population). 5.4 Results 5.4.1 GM-CSF production by SCCVII/JzGM-CSF cells Northern blot analysis of total RNA isolated from SCCVI I/ JzGM-CSF and SCCVI I/ Jzneo cells is shown in Figure 5.1. The upper panel confirms the presence of GM-CSF mRNA in S C C V I I cells transduced with the JzGM-CSF retrovirus, and shows that this message is not present in control cells infected with JzenTkneo virus. The lower panel shows that in S C C V I I / J z G M - C S F derived RNA the neo r probe recognized a transcript of approximately 3500 bp initiated within the retroviral LTR and encoding both GM-CSF and neo r , whereas the SCCVII/Jzneo cells expressed shorter neo r message initiated by the internal Tk promoter! The effect of GM-CSF on PDT 105 A A GM-CSF n e o Figure 5.1 Northern analysis of the total RNA isolated from SCCVII/JzGM-CSF and SCCVII/Jzneo cells. The production of GM-CSF protein by SCCVII/JzGM-CSF cells was tested by a bioassay using cytokine dependent B6SutA cells. The conditioned media from SCCVII/JzGM-CSF cells induced proliferative response of B6SutA cells, while the conditioned media from both, parental S C C V I I cells and SCCVII/Jzneo cells showed no support of the growth of B6SutA cells (Figure 5.2). A comparison to the response achieved with known concentrations of recombinant GM-CSF (Fig. 5.2 inset), indicates that SCCVII/JzGM-CSF cell conditioned medium contained approximately 400 ng/mL of GM-CSF released per 106 cells over a 24 h period. The effect of GM-CSF on PDT 106 3 0 0 0 0 CD § 2 5 0 0 0 E Q_ O 2 0 0 0 0 c o 2 1 5 0 0 0 o o o c I CO 1 0 0 0 0 5 0 0 0 0 -o c 2 0 0 0 CO 1 2 3 ng GM-CSF SCCVII/JzGM-CSF SCCVII parental SCCVII/Jz neo J 0.05 0.1 0.15 0.2 0.25 Condit ioned medium (% of total volume) Figure 5.2 GM-CSF bioassay B6SutA cells were incubated overnight in serial dilutions of conditioned medium from 24 h culture of 1x10 6 SCCVII/JzGM-CSF, SCCVII/Jzneo, and parental SCCVII cells. The responder cells were then pulsed with [3H]-thymidine for 4 hours and the amount of incorporated nucleotide was registered on a scintillation counter. Inset: Proliferative response of B6SutA cells to recombinant GM-CSF. The effect of GM-CSF on PDT 107 5.4.2 Blood leukocyte content in GM-CSF treated mice Blood smears were prepared at 2, 4, and 6 days following single or multiple treatment with 1x0 7 lethally irradiated GM-CSF producing S C C V I I cells administered subcutaneously in the dorsal area of C3H/HeN mice (otherwise the site of SCCV I I tumor implantation). Following Wright staining, a differential white blood cell count was performed based on 200 nucleated cells per smear (five mice per group). The only difference noted (Table 5.1) was an increase in the average monocyte content from 7±3% in control mice to 13±6% in mice examined 6 days after the last of three injections of SCCVI I/ JzGM-CSF cells that were given in 48 h intervals; this change however, was not statistically significant. The blood content of neutrophils, other granulocytes and lymphocytes were unchanged. Injection of SCCV I I / J zneo cells had no detectable effect on total and differential blood cell counts. Table 5.1 Blood leukocyte content in GM-CSF treated mice Treatment Days Lym a % . Neutro % . Mo % Baso & Eo % -Control 70 + 6° 22 ±7 7 + 3 1 + 2 2 c 69 ±9 2 1 + 5 8 + 3 1.5+1 SCCVII/Jzneo 4 '. 71 ±7 2 1 + 4 7 + 4 1±1 6 72 ±6 20 ±5 6 ± 3 1.5 ± 1 2 67 ± 7 22 + 3 10±5 1+1 SCCVII/JzGM-CSF 4 67 + 9 23 ± 4 9±5 1 ± 1.5. 6 65 ±8 22 ±5 13 ±6 1 ± 1 2 66 + 6 22 ± 6 12 ±5 0.8 ± 1 SCCVII/JzGM-CSF 3x 4 60 + 4. 24 + 7 15 + 6 1 ± 1 6 61 ±8 24 ±9 13 ± 6 2 + 2 Mice were injected either with GM-CSF producing SCCVII cells or control SCCVII cells infected with JzenTkneo virus. Blood samples for Wright staining were taken at 2 days after the last injection. a All differential counts were done manually on 200 cells. b Results are the mean + SD for 5 mice per group. c Time after single GM-CSF treatment or after the last of three GM-CSF treatments. The effect of GM-CSF on PDT 108 5.4.3 The effect of GM-CSF treatment on growth of SCCVII tumor Mice bearing a parental S C C V I I tumor were injected peritumorally with 1x10 7 lethally irradiated SCCVII/JzGM-CSF cells at the time when the largest tumor diameter reached 5-6 mm. The tumor volumes were then determined every second day. All treated mice developed edema around the tumor, which persisted for 2-4 days after the injection of GM-CSF releasing cells. This treatment, however, had no significant effect on the rate of growth of the SCCVI I tumor (Figure 5.3). The injection of lethally irradiated SCCV I I / J zneo or parental S C C V I I cells also had no effect in the growth of S C C V I I tumor (data -not shown), In other experiments, 1x10 7 lethally irradiated SCCVI I/ JzGM-CSF cells were injected subcutaneously into the area of tumor implantation two days after the inoculation of parental SCCV I I cells. Even in this situation, when the tumor mass is very small (similarly as in the case with the reduced tumor burden after PDT) the GM-CSF treatment had no visible effect in tumor growth (Figure 5.4). The effect of GM-CSF on PDT 109 0.1 I—i • 1 • • • 1 i l—i i i J i_J . I 4 6 8 10 12 14 16 18 2 0 Days after tumor implant . Figure 5.3 The effect of GM-CSF treatment on growth of established S C C V I I tumor. Mice bearing SCCVII tumors 5-6 mm in diameter were injected peritumorally with 1x10 7 lethally irradiated SCCVII/JzGM-CSF cells. Tumor volumes were determined every second day. Each group consisted of six mice. Bars are SD. The effect of GM-CSF on PDT 110 0.1 I 1 I- ' • • ' i_J • I i I i I i I i I 4 6 8 10 12 14 16 18 2 0 Days after tumor implant Figure 5.4 Effect of GM-CSF treatment on inoculated S C C V I I tumor. Mice were injected subcutaneously with 1x10 7 lethally irradiated SCCVII/JzGM-CSF cells 2 days after the inoculation of parental SCCVII tumor. Tumor volume was determined every 2 days. Bars represent SD. The effect of GM-CSF treatment on the levels of tumor infiltrating leukocyte populations in the S C C V I I tumor was also examined. Mice bearing S C C V I I tumors (of a similar size as used for PDT) were injected peritumorally with 1x10 7 lethally irradiated SCCVII/JzGM-CSF cells, the tumors were excised 48 h later and analyzed for the content of malignant and other cell populations. This was determined by flow cytometry analysis of cells dissociated from the tumor tissue and stained with monoclonal antibodies against specific membrane markers which serve for the identification of various types of tumor infiltrating leukocytes. The experimental protocols for such analysis of various tumor models (including S C C V I I tumor) were developed in our earlier studies (Korbelik and Krosl 1995; Korbelik and Krosl 1996). As previously reported (Korbelik and Krosl 1996), the majority of cells in SCCV I I tumor are cancerous The effect of GM-CSF on PDT 111 (61%) i.e. CD45", the second largest population are TAMs (29%) i.e. F4/80+ while there are 5% of other myeloid cells (GR1+F4/80-) and 5% T lymphocytes (CD3+). These average values for the incidence of major cell populations in S C C V I I tumor remained unchanged following the above described treatment with GM-CSF releasing cells. 5.4.4 The effect of GM-CSF treatment on cytotoxic activity of TAM Mice bearing S C C V I I tumor were injected peritumorally with 1 x 1 0 7 of lethally irradiated SCCVII/JzGM-CSF of SCCVII/Jzneo cells. The tumors were excised 48 h later and TAMs were harvested for the analysis of their cytotoxicity against syngeneic in vitro cultured malignant SCCV I I cells. The measurement of [3H]thymidine released from killed target cells (Figure 5.5) revealed that, in comparison to TAMs isolated from non-treated or SCCV I I / J zneo treated S C C V I I tumors, the TAMs obtained following tumor localized GM-CSF treatment exhibit 3-4 fold increase in tumoricidal activity. The effect of GM-CSF on PDT 112 3 0 r 2 5 -'S 2 0 CO CD CD C TJ 15 ->, 1 0 -I CO 0 Con t ro l Jz neo Jz GM-CSF Figure 5.5 The effect of tumor localized GM-CSF treatment on cytotoxic activity of TAM. S C C V I I tumor bearing mice were injected with 1 x 1 0 7 of lethally x-ray irradiated S C C V I I / J z G M - C S F of SCCVII / Jzneo cells. The tumors were excised and T A M s isolated 48 h later. Cultured malignant S C C V I I cells were labeled with [ 3H]thymidine and admixed to T A M s at 1:20 ratio (1 x 1 0 4 S C C V I I cells vs. 2 x 1 0 5 T A M s per sample). Following 72 h incubation, the supernatants were collected for the measurement of radioactivity released from killed target cells. * = statistically significant difference compared to controls (p < 0.001). Bars indicate S D . 5.4.5 The effect of GM-CSF treatment on the response of tumors to PDT. Three groups of 16 mice were included in the experiment designed to investigate the effect of GM-CSF immunotherapy on Photofrin mediated PDT. In two groups, 1 x 1 0 7 lethally irradiated S C C V I I / J z G M - C S F cel ls or SCCVII/Jzneo cells were injected peritumorally 48 h before the photodynamic The effect of GM-CSF on PDT 113 light delivery, immediately after the light treatment and 48 h later. The third group received PDT only. Complete tumor ablation was observed within 2 days after light delivery in all three groups (Figure 5.6). However, the tumors started to regrow in the PDT only group and the PDT+Jzneo group at 8-10 days post PDT. By day 20, only one mouse (out of 16) showed no signs of tumor recurrence in both these groups. In PDT+JzGM-CSF group, however, the tumor free period lasted at least two weeks. Although tumor regrowth occurred afterwards in some mice, the overall cure rate for this group was 7 5 % . Prolonged tumor free period and significantly increased survival of tumor bearing mice in Photofrin mediated PDT combined with G M - C S F immunotherapy indicates that such a treatment modality offers increased effectiveness of tumor control. The effect of GM-CSF on PDT 114 0 4 8 12 16 2 0 2 4 9 2 9 6 Days relative to PDT Fjgure 5.6 The effect of localized GM-CSF treatment on the control of SCCV I I tumor by Photofrin mediated PDT. Mice bearing SCCVI I tumor received Photofrin (10 mg/kg, i.v..) 24 hours before the light treatment with 150 J/cm2. Some mice were also injected peritumorally with lethally x-ray irradiated SCCVII/JzGM-CSF or SCCVII/Jzneo cells in three 2 day intervals, i.e. at 48 h before PDT, immediately following light treatment, and 48 h after PDT. The mice were observed afterwards for signs of tumor recurrence. Each treatment group contained 16 mice. The difference of cure rate between GM-CSF+PDT and PDT alone groups is statistically significant (p < 0.0005).; / The same GM-CSF immunotherapy protocol was also examined in combination with BPD mediated PDT of SCCVI I tumor. The result are depicted in Figure 5.7. In all three groups, the tumors regressed and become unpalpable within several days after PDT. However, visible signs of recurrence appeared by day 8 and 10 in PDT only and PDT+Jzneo groups, respectively. All mice in these two groups showed tumor regrowth by day 18 after PDT. In The effect of GM-CSF on PDT 115 PDT+JzGM-CSF group, there was no tumor recurrence until day 12, with 5 0 % of animals remaining tumor free at the end of the observation period (96 days). PDT PDT + Jz neo PDT + JzGM-CSF I I I I I L L I L I I I l _ i • i 0 2 4 6 8 101 2 1 4 1 6 1 8 2 0 2 2 24 92 94 Days relative to PDT Figure 5.7 The effect of localized GM-CSF immunotherapy on the control of SCCVII tumor by BPD mediated PDT SCCVII tumor bearing mice received BPD (2.5 mg/kg, i.v.) at 4 h before the light treatment with 110 J/cm 2. Two groups of mice were treated with SCCVII/JzGM-CSF or SCCVII/Jzneo cells as described in Fig. 5.6. Other details were also as in Fig 5.6. The difference in the cure rate between GM-CSF and PDT alone groups is statistically significant (p < 0.0005). The effect of GM-CSF on PDT 116 5.5 Discussion In order to investigate the effect of localized GM-CSF treatment on the control of tumors treated with PDT, we introduced the gene for murine GM-CSF into SCCV I I tumor cells. Northern blot analysis and the bioassay with GM-CSF dependent B6SutA cells confirmed that GM-CSF transduced cells (denoted SCCVI I/ JzGM-CSF) express GM-CSF mRNA and secrete biologically active form of this cytokine. Stable secretion of relevant cytokines from infected tumor cells following antiproliferative doses of x-ray irradiation (as used in the present study) was reported to continue under in vitro conditions for at least 7-10 days (Tepper and Mule 1994). Single or . multiple subcutaneous injections of lethally irradiated SCCVII/JzGM-CSF cells exerted no significant effect on the leukocyte numbers in peripheral blood of treated mice. The peritumoral injections of GM-CSF releasing cells markedly increased cytotoxicity of TAMs against autochtonous SCCVI I cells (Figure 5.5), but this treatment had no obvious effect on the rate of growth of SCCV I I tumor (Figures 5.3 and 5.4). As demonstrated in the number of studies, a large tumor burden normally overcomes the immunostimulation induced by vaccination with cytokine gene-modified tumor cells (Tepper and Mule 1994). On the other hand, Dranoff era/. (Dranoff et al. 1993) showed that the vaccination of mice with GM-CSF producing B16 melanoma cells induces the rejection of B16 melanoma inoculum implanted 3 days earlier, while we were unable to demonstrate a similar effect with SCCVII/JzGM-CSF cells given two days after the S C C V I I tumor implant. This finding may reflect the lower antigenicity of SCCVI I tumor compared to the B16 melanoma. It should also be noted that the tumor inoculum used by Dranoff et al. (Dranoff et al. 1993) was lower than the inoculum used in the present study (5 x 10 4 vs. 5 x 10 5 cells). The effect of GM-CSF on PDT 117 In preliminary studies (Krosl et a l , 1995), we examined a variety of treatment schedules with GM-CSF releasing cells in search for the most effective protocol for combined application with PDT. Based on this evaluation, we opted for three peritumoral injections of lethally irradiated SCCV I I / J zGM-CSF cells spaced in 48 h intervals and begun two days before PDT. It should be noted, however, that even one such injection (given 48 h before PDT) proved effective in the enhancement of PDT related tumor control (Krosl et al. 1995). Much less effective was GM-CSF therapy initiated after PDT. , Since the cytokine production by SCCVI I / JzGM-CSF cells is mediated through their regular metabolic activity, there is no basis to expect an increased release of GM-CSF from these cells by a light-activated mechanism. On the contrary, photosensitizer accumulation in the cells subtumorally injected before its administration and consequent induction of PDT-mediated photodamage can negatively affect the metabolism of these cells, resulting in decreased GM-CSF release. The results presented in this Chapter demonstrate that tumor localized GM-CSF treatment substantially improves the curative effect of PDT in the S C C V I I tumor model. This potentiating effect of GM-CSF was highly pronounced irrespective of whether Photofrin of BPD were employed as photosensitizers for PDT. The enhancement level with both photosensitizers was equivalent to a -2.5 fold increase in the light dose for PDT used alone. The SCCV I I cells transfected with the control retroviral vector encoding for neo r only were used to demonstrate that: i) peritumoral injections perse do not affect the PDT response, and ii) localized treatment with lethally irradiated tumor cells transfected with the retroviral vector are not inducing an artifact in the evaluation of tumor response to PDT. The effect of GM-CSF on PDT 118 Based on available knowledge of the immunomodulating properties of GM-CSF, several putative mechanisms can be proposed to account for the observed effects of treatment with this cytokine on the antitumor activity of PDT. The increased local concentration of GM-CSF attained by the treatment protocol used in the present study may: i) upregulate the expression of leukocyte adhesion molecules and consequently the extravasation of inflammatory cells into tumor tissue; ii) potentiate specifically the tumoricidal activity of neutrophils and/or monocytes-macrophages; iii) induce release of a number of secondary cytokines and other molecules (including e.g. TNF-a and interleukin-2) involved in mediating the antitumor effect; iv) facilitate the presentation of tumor antigens released following PDT treatment and thereby activate specific T cell mediated immune responses; v) promote the inactivation of remaining islets of viable tumor cells, unhindered by the tumor burden, as PDT effectively de-bulked the tumor mass; vi) diminish the immunosuppressive effect (Bilyk and Holt 1993) induced by PDT (Lynch et al. 1989; Lynch et al. 1989). The ongoing research in our laboratory is designed to address specifically the contribution of the above listed mechanisms. The fact that the potentiation of the antitumor effect of PDT was observed with the S C C V I I tumor which is poorly immunogenic is particularly encouraging with respect to potential clinical ramifications. It remains to be determined whether the potentiating effect of GM-CSF treatment is equally pronounced or different with PDT treated immunogenic tumors. In Chapter 4 it was shown that the response of the S C C V I I tumor to PDT can be augmented by multiple intramuscular injections of glucan S P G administered before the light treatment. The response of tumors characterized by different degrees of immunogenicity was shown to be enhanced by a variety of other nonspecific immune stimulants, including B C G (Cho et al. 1992), Corynebacterium parvum vaccine (Myers et al. 1989), mycobacterial cell wall The effect of GM-CSF on PDT 119 extract (Korbelik and Krosl 1996) and endotoxin (Dima et al. 1994), as well as with specific immune agents such as the macrophage activating factor GcMAF (Korbelik et al. 1995) and cytokines TNF-a (Bellnier 1991) and interleukin-7 (Dougherty et al. 1992). It appears, therefore, that due to its unique inflammatory/immune character, PDT is highly responsive to adjuvant immunotherapy. Immunotherapy regimens tailored and optimized for particular malignant lesions may produce outstanding gains in PDT mediated tumor control. The treatment with PDT, serving as a strong acute inflammatory insult associated with de-bulking of tumor mass,, may emerge as a model system of value for characterization of different aspects of the immune reaction. Conclusions 120 CHAPTER 6 Conclusions and future directions 6.1 Conclusions The goal pursued in this research project was to provide a better insight into the role of the host in the tumor response to PDT. The reports available at that time (Bugelski et al. 1981; Elmets and Bowen 1986; Lynch et al. 1989; Lynch et al. 1989; Myers et al. 1989; Evans et al. 1990) and the evidence the PDT induces release of vasoactive and proinflammatory mediators (Ben-Hur et al. 1988; Henderson and Donovan 1988; Henderson and Donovan 1989; Nseyo et al. 1990; Fingar et al. 1991; Fingar et al. 1992) indicated that cells of host immune system may be actively involved in the processes that lead to tumor destruction by PDT. This provided the rationale for initiating our studies on the role of host immune cells in the response of tumors to PDT. Our initial work concentrated on the distribution of Photofrin between the malignant and host cells of the murine S C C V I I tumor. The in vitro studies showed that not only the uptake of Photofrin by macrophages is markedly higher compared to S C C V I I cells, but also that influencing the phagocytic activity of macrophages affects their ability to accumulate the photosensitizer (Tables 2.1 and 2.2). An interesting result from these studies was the demonstration of the inhibitory effect of serum and different plasma proteins on Photofrin accumulation in S C C V I I cells as well as in macrophages (Figure 2.3). We interpreted this to be a consequence of disaggregation of Photofrin material by the serum components (Korbelik and Hung 1991). An alternate explanation would invoke internalization of photosensitizer-serum protein complexes through specific endocytic receptors which are known to be abundant on macrophage surface membrane. However, these in vitro results Conclusions 121 do not support the hypothesis that macrophages accumulate mainly LDL bound Photofrin via the LDL and scavenger receptor pathways as proposed by Hamblin et al. (Hamblin and Newman 1994). The studies on the distribution of different photosensitizers in various tumor models that were done in our laboratory after the initial experiments described in Chapter 2 (Korbelik and Krosl 1995; Korbelik and Krosl 1995; Korbelik and Krosl 1996) support the hypothesis that the state of macrophage activation determines the accumulation of Photofrin in these cells. This is also in concordance with the observations of porphyrin accumulation in areas of inflammation, atherosclerosis and healing wounds (Figge et al. 1948; Spears et al. 1983; Nseyo et al. 1985). It is now established that an intense inflammatory reaction follows PDT treatment of solid tumors. The release of pro-inflammatory mediators such as metabolites of arachidonic acid, cytokines and degradation products of membranous lipids have been implicated in this process (Pass 1993). We have shown that Photofrin-based PDT initiated a massive and regulated infiltration of inflammatory cells into the treated tumor. Such an infiltration could facilitate the effect of PDT by several interconnected mechanisms. The swift and massive infiltration of neutrophils is probably the initiating event in the host response to PDT. The ability of activated neutrophils to kill tumor cells is well known (Fady et al. 1990; Midorikama et al. 1990). The release of superoxide by neutrophil membrane-associated NAPDH oxidase (Rossi 1986) and liberation of proteases and non-oxidative toxins by these cells is likely to play an important role in the, observed blood vessel damage and the resulting hemorrhagic necrosis (Gallin 1989). Although the numbers of infiltrating mast cells observed in PDT treated S C C V I I tumors are much lower than those determined for neutrophils, mast cells are known to produce a variety of cytokines such as IL-4, IL-5, IFN-y, and TNF-a (Harvima et al. 1994). The macrophages that were Conclusions 122 recovered from the tumor site after PDT showed markedly increased cytotoxicity against tumor cells (see Figure 3.6). The intense tumor cell destruction and release of dead cell fragments, that are likely phagocytized by resident and infiltrating macrophages, build up the conditions for the efficient presentation of tumor antigens to T lymphocytes. The pro-inflammatory cytokines mentioned above, documented to be released after PDT (Nseyo et al. 1990), and mediators released by infiltrating neutrophils and macrophages may provide an appropriate milieu for T cell activation (Colombo et al. 1992; LLoyd and Oppenheim 1992). This proinflammatory and immunostimulatory nature of PDT was exploited in the experiments with combined PDT-immunotherapy treatment that are described in Chapter 4 and Chapter 5. Significantly enhanced tumor control obtained with PDT in combination with GM-CSF and S P G immunotherapies indicates that the immune system can be induced to eradicate tumor cells that would otherwise escape PDT treatment. The constituitive release of cytokines by genetically engineered tumor cells, as was used in the immunotherapy with GM-CSF (Chapter 5), is an intensely researched area in tumor biology (Colombo et al. 1992; Tepper and Mule 1994). Localized treatment of tumors with exogenous cytokines was shown to induce antitumor reaction which can lead to tumor regression (Pericle et al. 1992). In our experimental system, the treatment with either of the tested immunotherapies alone did not affect tumor progression (Figure 4.1 and Figure 5.3). The enhanced tumor control that was achieved with the PDT-immunotherapy combinations compared to PDT alone is probably the result of a stronger PDT-immunotherapy induced host response. The results presented in Table 4.2 clearly show that the SPG immunotherapy significantly enhanced the secondary effects of PDT. In the experiments with both immunotherapies it was Conclusions 123 shown that the timing of immunotherapy relative to PDT is of a critical importance. S P G requires repeated administration for priming of macrophages (Sugawara et al. 1984; Tsuchiya et al. 1989), and is therefore ineffective if given after PDT. GM-CSF enhances proliferation and antigen presentation of A P C s (Heufler et al. 1987; Witmer-Pack et al. 1987; Grabbe et al. 1992; Grabbe et al. 1994) and together with TNF-a induces the increased expression of MHC class II and costimulatory molecules on A P C s (Grabbe et al. 1992; Sallusto and Lanzavecchia 1994). Continuous presence of GM-CSF during and after PDT (Figure 4.6) seemed to have a stronger potentiating effect on PDT compared to the single GM-CSF treatment (Figure 4.5) It has been reported that curative PDT can induce a protective immunity against the secondary challenge with the same tumor (Canti et a|. 1994), while Myers et al. (Myers et al. 1989) observed that the treatment with low dose Corynebacterium parvum vaccine which induces primarily lymphocyte infiltration in tumor site amplifies the effect of PDT. Therefore, it can be hypothesized that GM-CSF immunotherapy enhances the specific immune response against the tumor cells which was initiated by PDT. 6.2 Future directions The role of the inflammatory reaction in PDT mediated tumor control is well established but it needs to be better characterized. It is known that treatment with inhibitors of inflammation, such as drugs that block the metabolism of arachidonic acid, reduce the tumor cure (Fingar et al. 1993). Additional studies in our laboratory will focus on the importance of leukocyte-endothelial cell interactions in tumor control by PDT. Antibodies that can block leukocyte-endothelial cell adhesion (Albelda et al. 1,994) will be used in order to manipulate inflammatory cell extravasation pathways. These results may Conclusions 124 provide important information about the development of PDT induced inflammatory reaction and its importance for PDT-mediated tumor cure. Results of enhanced tumor control with PDT-immunotherapy combination compared to PDT alone, presented in this thesis warrant additional studies on the role of immune system in the response to PDT. The possible involvement of specific immunity in the curative potential of GM-CSF immunotherapy adjuvant to PDT can be inferred from the fact that GM-CSF induces T-cell mediated immunity against the tumors (Dranoff e t a l . 1993; Levitsky et al. 1994). Our observation that the anitumor effect of PDT is greatly diminished in the severe combined immunodeficient (SCID) mice, which lack the specific immune response mediated by T and B lymphocytes, in comparison to syngeneic Balb/c mice provide an additional support for this hypothesis. Adoptive transplantation of T lymphocytes or their subsets from normal mice into SCID mice at different times relative to PDT will give us more specific answers in regard to the involvement of specific effector cells in PDT tumor control. The depletion of T lymphocyte subsets in normal mice before or after PDT will provide additional data with respect to the importance of T cells during and after PDT. We expect that at least for some tumors the presence of specific immune cells will be shown to be of a crucial importance for tumor control. A better knowledge about the contribution of these cells in the processes that lead to PDT-mediated tumor destruction will help in designing more effective immunotherapy protocols for combination with PDT. References 125 CHAPTER 7 References Abramson, A. L, M. J . Shikowitz, V. M. Mullooly, B. M. Steinberg, C. A. Amelia and H. R. Rothstein (1992). "Clinical effects of photodynamic therapy on recurrent laryngeal papillomas." Arch. Otolar. Head Neck Surg 118: 25-9. Abramson, A. L, M. J . Shikowitz, V. M. Mullooly, B. M. Steinberg and R. B. Hyman (1994). "Variable light-dose effect on photodynamic therapy for laryngeal papillomas." Arch. Otolar. Head Neck Surg. 120: 852-5. Agarwal, M. L, H. E. Larkin, S. I. A. Zaidi, H. Mukhtar and N. Oleinick (1993). "Phospholipase activation triggers apoptosis in photosensitized mouse lymphoma cells." Cancer Res. 53: 5897-5902. Albelda, S. M., C. W. Smith and P. A. Ward (1994). "Adhesion molecules and inflammation injury." FASEB J. 8: 504-512. Andersson-Engels, S., R. Berg, J . Johansson, S. Svanberg, U. Stenram and K. Svanberg (1992). Laser spectroscopy in medical diagnostics. Photodynamic Therapy, Basic Principles and Clinical1 Applications. (B. W. Henderson and T. J . Dougherty, ed.) New York, Marcel Dekker Inc. pp. 387-424. Atlante, A., S. Passarella, E. Quagliariello, G. Moreno and C. Salet (1989). "Haematoporphyrin derivative (Photofrin II) photosensitization of isolated mitochondria: inhibition of ADP/ATP translocator." J. Photochem. Photobiol. B: Biol. 4: 35-46. Atlante, A., S. Passarella, E. Quagliariello, G. Moreno and C. Salet (1990). "Carrier thiols are targets of Photofrin II photosensitization of isolated rat liver mitochondria." J. Photochem. Photobiol. B: Biol. 7: 21-32. References 126 Auler, H. and G. Banzer (1942). "Untersuchungen uber die rolle der porphyrine bei geschwulstkranken menshen und tieren." Z. Krebsforsch. 53: 65-68. Ausubel, F. M., R. Brent, R.'E. Kingston, D. D. Moore, J . G. Seidman, J . A. Smith and K. Struhl, Ed. (1994). Transfection of DNA into eukaryotic cells. Current protocols in molecular biology. New York, Greene Publishing Associates and Wiley-lnterscience. Axline, S. G. and E. P. Reaven (1974). "Inhibition of phagocytosis and plasma membrane mobility of the cultivated macrophage by cytohalasin B." B. J. Cell Biol. 62: 647-659. Bellnier, D. A. (1991). "Potentiation of photodynamic therapy in mice with recombinant human tumor necrosis factor-oc." J. Photochem. Photobiol. B: Biol. 8: 203-210. Bellnier, D. A. and T. J . Dougherty (1989). "The time course of cutaneous porphyrin photosensitization in the murine ear." Photochem. Photobiol. 49: 369-372. Bellnier, D. A. and B. W. Henderson (1992). Determinants of photodynamic tissue destruction: Photodynamic Therapy, Basic Principles and Clinical Applications. (B. W. Henderson and T. J . Dougherty, ed.) New York, Marcel Dekker Inc. pp. 117-127. Ben-Hur, E., E. Heldman, S. E. Crane and I. Rosenthal (1988). "Release of clotting factors from photosensitized endothelial cells: a possible trigger for blood vessel occlusion by photodynamic therapy." FEBS Lett 236: 105-108. Berg, K. and J . Moan (1992). "Mitotic inhibition by phenylporphines and tetrasulfonated aluminium phthalocyanine in combination with light." Photochem. Photobiol. 56: 333-339. References 127 Berg, K., J . Moan, JVC. Bommer and J . W. Winkelman (1990). "Cellular inhibition of microtubule assembly by photoactivated sulphonated meso-tetraphenylporphines." Int. Jr Radial Biol. 58: 475-487. Bilyk, N. and P. G. Holt (1993). "Inhibition of the immunosuppressive activity of resident pulmonary macrophages by granulocyte/macrophage colony-stimulating factor." J. Exp. Med. 177: 1773-1777. Bonnett, R. and M. Berenbaum (1989). "Porphyrins as photosensitizers." CIBA Found. Symp. 146: 40-53. Brown, S. B. and D. I. Vernon (1990). The quantitative determination of porphyrins in tissues and body fluids: applications in studies of photodynamic therapy. Photodynamic therapy of neoplastic disease. (D. Kessel, ed.) Boca Raton, C R C Press, Inc. pp. 109-128. Bucana, C D , A. Fabra, R. Sanchez and I. J . Fidler (1992), "Different patterns of macrophage infiltration into allogenic-murine and xenogenic-human neoplasms growing in nude mice." Am. J. Pathol. 141: 1225-1236. Bugelski, P. J . , C. W. Porter and T. J . Dougherty (1981). "Autoradiographic distribution of hematoporphyrin derivative in normal and tumor tissue of the mouse." Cancer Res. 41: 4606-4612. Cairnduff, F., M. R. Stringer, E. J . Hudson, D. V. Ash and S. B, Brown (1994). "Superficial photodynamic therapy with topical 5-aminolaevulonic acid for superficial primary and secondary skin cancer." Br. J. Cancer 69: 605-608. Candide, C , J . C. Maziere, R. Santus, C. Maziere, P. Morliere, J . P. Reyftmann, S. Goldstein and L Dubertret (1989). "Photosensitization of Wi26-VA4 transformed human fibroblasts by low density lipoprotein loaded with the anticancer porphyrin mixture photofrin II: evidence for endoplasmic reticulum alteration." Cancer Lett. 44: 157-161. References 128 Canti, G., D. Lattuada, A. Nicolin, P. Taroni, G. Valentini and R. Cubeddu (1994). "Antitumor immunity induced by photodynamic therapy with aluminum disulfonated phthalocyanines and laser light." Anti-Cancer Drugs 5: 443-447. Chan, W.-S., J . F. Marshall, G. J . F. Lam and I. R. Hart (1988). "Tissue uptake, distribution, and potency of the photoactive dye chloraluminum , sulfonated phthalocyanine in mice bearing transplantable tumors." Cancer Res. 48: 3040-3044. Chaplin, D. J . , R. E. Durand, I. J . Stratford and T. C. Jenkins (1986). "The radiosensitizing and toxic effects of RSU-1069 on hypoxic cells in a murine tumor." Int. J. Radiation Oncology Biol. Phys. 12: 1091-1095. Chaplin, D. J . , P. L. Olive and R. E. Durand (1987). "Intermittent blood flow in a murine tumour: radiobiological effects." Cancer Res. 47: 597-601. Chapman, M. J . (1986). "Comparative analysis of mammalian plasma lipoproteins." Methods Enzymol. 128: 569-573. Chien, K. R., J . Abrams, A. Serioni, J . T. Martin and J . K. Farber (1978). "Accelerated phospholipid degradation and association membrane dysfunction in irreversible ischemic liver cell injury." J. Biol. Chem. 256: 4809-4817. Cho, Y., R. C. Straight and J . H. Smith (1992). "Effects of photodynamic therapy in combination with intravesical drugs in a murine bladder tumor model." Journal of Urology 147: 743-746. Colombo, M. P., A. Modesti, G. Parmiani and G. Forni (1992). "Perspectives in Cancer research: local cytokine availability elicits tumor rejection and systemic immunity through granulocyte-T-lymphocyte cross-talk." Cancer Res. 52: 1-5. References 129 Cortese, D. A., E. S. Edell, M. D. Silvastrein, K. Offord, V. F. Trastek, P. C. Pairolero, M. S. Allen and C. Dechamps (1992). An evaluation of the effectiveness of photodynamic therapy (PDT) compared to surgical resection in early stage roentgenographically occult lung cancer. Photodynamic Therapy and Biomedical Lasers. (P. Spinelli, M. Dal Fante and R. Marchesini, ed.) Amsterdam, Elsevier Science Publishers B. V. pp. 15-22. Damen, J . , A. L. F. Mui, P. Hughes, K. Humpries and G. Krystal (1992). "Erythropoeitin-induced tyrosine phosphorylation in a high erythropoeitin receptor-expressing lymphoid cell line." Blood 80: 1923-1932. Darling, G., D. L. Fraker, J . C. Jensen, C. M. Gorschboth and J . A. Norton (1990). "Cachectic effects of recombinant human tumor necrosis factor in rats." Cancer Res. 50: 4008-4013. DeLaney, T. F., W. G. Sindelar, Z. Tochner, P. D. Smith, W. S. Friauf, G. Thomas, L. Dachowski, J . W. Cole, S. M. Steinberg and E. Glatstein (1993). "Phase I study of debulking surgery and photodynamic therapy for disseminated intraperitoneal tumors." Int. J. Radiation Oncology Biol. Phys. 25: 445-457. Dima, V. F., V. Vasiliu, D. Laky, M. D. lonescu and S. V. Dima (1994). "Treatment of rat Walker-256 carcinosarcoma with photodynamic therapy and endotoxin irradiated with high energy electrons." SPIE 2078: 547-557. Dougherty, G. J . , C. A. Allen and N. M. Hogg (1986). Applications of immunological techniques to the study of the tumour-host relationship. Handbook of Experimental Immunology. (D. M. Weir, L. A. Herzenberg, C. Blakwell and L. A. Herzenberg, ed.) Oxford, Blackwell Scientific Publications Ltd. pp. 125.1-125.12. Dougherty, G. J . , S. T. Dougherty, R. J . Kay, P. Lansdorp and R. K. Humphries (1989). "Identification and characterization^ 114/A10, an antigen highly expressed on the surface of murine myeloid and erythroid progenitor cells and IL-3 dependent cell lines." Exp. Hematol. 17: 877-882. References 130 Dougherty, G. J . , J . D. Thecker, W. H. McBride, G. Krosl and M. Korbelik (1992). "Effect of immunization with genetically modified tumor cells on tumor recurrence following photodynamic therapy." Lasers Med. Sci. 7: 226. Dougherty, T. J . (1984). "Photoradiation therapy." Urol. Supl. 23: 61. Dougherty, T. J . (1987). "Photosensitizers: Therapy and detection of malignant tumors." Photochem. Photobiol. 45: 874-889. Dougherty, T. J . (1989). "Photodynamic therapy: Status and potential." Oncology 3: 67-78. Dougherty, T. J . , D. Boyle, K. R. Weishaupt, C. Gomer, D. Borcicky, J . H. Kaufman and G. Grindey (1977). Phototherapy of human tumors. \ Research in Photobiology. (A. Castellani, ed.) New York, Plenum Publishing Corp. pp. 435-466. Dougherty, T. J . , G. Lawrence, J . H. Kaufman, D. Boyle, K. R. Weishaupt and A. Gbldfarb (1979). "Photoradiation in the treatment of recurrent breast carcinoma." JNCI62: 231-237. Dougherty, T. J . , T. S. Mang, A. M. Regal, H. Takita, Z. Bernstein, B. D. Wilson and M. McAuley (1992). "New potential clinical applications of photodynamic therapy: Intrathoracic PDT and treatment of AIDS-related Kaposi's sarcoma." Lasers Med. Sci. 7: 215. Dougherty, T. J . , W. R. Potter and K. R. Weishaupt (1984). The structure of the active component of hematoporphyrin derivative. Porphyrin localization and treatment of tumors. (D. R. Dorion and C. J . Gomer, ed.) New York, Alan R Liss. pp. 301-314. References 131 Dranoff, G., E. Jaffee, A. Lazehby, P. Golumbek, H. Leyitsky, K. Brose, V. Jackson, H. Hamada, D. Pardoll and R. C. Mulligan (1993). "Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colonyTStimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity." Proc. Natl. Acad. Sci. USA 90: 3539-3543. Eddel, E. S. and D, A. Cortese (1992). "Photodynamic therapy in the management of early superficial squamous cell carcinoma as an alternative to surgical resection." Chest 102: 1319-1322. Elmets, C. A. and K. D. Bowen (1986). "Immunological supression in mice treated with hematoporphyrin derivative photoradiation." Cancer Res. 46: 1608-1611. Evans, S., W. Matthews, R. Perry, D. Fraker, J . Norton and H. I. Pass (1990). "Effect of photodynamic therapy oh tumor necrosis factor production by murine macrophages." JNCI82: 34-39. Fady, C , D. Reisser and F. Martin (1990). "Non-activated rat neutrophils kill syngeneic colon tumor cells by the release of low molecular wight factor." lmmunobiology\8~\: 1-12. Feinberg, A. P. and B. Vogelstein (1984). "A technique for radiolabelling DNA restriction endonuclease fragments to high specific, activity." Anal. Biochem. 132: 6-13. Ferrario, A. and C. T. Gomer (1990). "Systemic toxicity in mice induced by localized porphyrin photodynamic therapy." Cancer Res. 50: 539-543. Figge, F. H. J . , G. S. Weiland and M. L. O. J . (1948). "Cancer detection and therapy. Affinity of neoplastic, embryonic and traumatized tissue for porphyrins and metalloporphyrins." Proc. Soc. Exp. Biol. Med. 68: 640-641. References 132 Fingar, V. H., K. A. Siegel, J . T. Wieman and K. W. Doak (1993). "The effects of thromboxane inhibitors on the microvasculature and tumor response to photodynamic therapy." Photochem. Photobiol. 58: 393-399. Fingar, V. H., T. J . Weiman, S. A. Wiehle and P.'.B. Cerrito (1992). "The role of microvascular damage in photodynamic therapy: The effect of treatment on vessel constriction, permeability and leukocyte adhesion." Cancer Res. 52; 4914-4921. Fingar, V. H , T. J . Wieman and k. W. Doak (1990), "Role of thromboxane and prostacyclin release on photodynamic therapy-induced tumor destruction." Cancer Res. 50: 2599-2603. Fingar, V. H., T. J . Wieman and K. W. Doak (1991). "Machanistic studies of PDT induced vascular damage: evidence that eicosanoids mediate the . process." Int.J. Radial Biol. 60: 303-309. Foote, C. S. (1984). Mechanisms of photooxygenation. Porphyrin localization and treatment of tumors. (D. R. Dorion and C. J . Gomer, ed.) New York, Alan R. Liss Inc. pp. 3-18. Foster, T. H., R. S. Murant, R. G. Bryant, R. S. Knox, S. L. Gibbson and R. Hilf (1991). "Oxygen consumption and diffusion effects in photodynamic therapy." Radial Res. 126: 296-303. Furuse, K., M. Fukuoka, H. Kato, T. Horai, K. Kubota, N. Kodama, Y. Kusunoki, N. Takifuji, T. Okunaka and C. Konaka (1993). "A prospective phase II study on photodynamic therapy with photofrin II for centrally located early-stage lung cancer. The Japan Lung Cancer Photodynamic Therapy Study Group [see comments]." J. Clin. Oncol. 11: 1852-1857. Gallin, J . I. (1989). Inflammation. Fundamental Immunology. (W. E. Paul, ed.) New York, Raven Press, pp. 721-733. References 133 Gomer, C. J . and T. J . Dougherty (1979). "Determination of [3H]- and [ 1 4C]Hemathoporphyrin derivative distribution in malignant and normal tissue." Cancer Res. 39: 146-151. Grabbe, S.,S. Bruvers, S. Beissert and R. D. Granstein (1994). "Interferon-y inhibits tumor antigen presentation by epidermal antigen-presenting cells." J. Leukoc. Biol. 55: 695-701. Grabbe, S., S. Bruvers, A. M. Lindgren, J . Hosoi, K. C. Tan and R. D. Granstein (1992). "Tumor antigen presentation by epidermal antigen-presenting . cells in the mouse: modulation by granulocyte-macrophage colony-stimulating factor, tumor necrosis factor a, and ultraviolet irradiation." J . Leukoc. Biol. 52: 209-217. Grant, W. E., P. M. Speight, A . J . MacRobert, C. Hopper and S. G. Bown (1994). "Photodynamic therapy of normal rat arteries after photosensitization , using disulphonated aluminum phthalocyanine and 5-aminblaevulonic acid." Br. J. Cancer 70: 72-78. Grossweiner, L. I. (1989). The Science of Photobiology. New.York, Plenum Press. •' ' : Grossweiner, L. I. (1994). The Science of Phototherapy. Boca Raton, C R C Press, Inc. Hamblin, M. R. and L. E. Newman (1994). "On the mechanism of the tumour-localising effect in photodynamic therapy." J. Photochem: Photobiol. B: Biol. 23: 3-8" Hampton, J . A. and S. H. Selman (1992). "Mechanisms of cell killing in photodynamic therapy using a novel in vivo drug/in vitro light culture system." Photochem. Photobiol. 56: 235-243. Harvima, I. T., L. Horsmanheimo, A. Naukkarinen and M. Horsmanheimo rf (1994). "Mast cell proteinases and cytokines in skin inflammation. [Review]."/Arc/7. Dermato/. Res. 287: 61-67, ; References 134 Henderson, B. W. (1989). "Photodynamic therapy: coming of age." Photodermathology 6: 200-211. Henderson, B. W. and D. A. Bellnier (1989). "Tissue localization of photosensitizers and the mechanism of photodynamic tissue destruction. [Review]." CIBA Found. Symp. 146: 112-25. Henderson, B. W., D. A. Bellnier, B. Ziring and T. J . Dougherty (1983). Aspects of the cellular uptake and retention of hematoporphyrin derivative and their correlation with the biological response to PRT in vitro. Porphyrin Photosensitization. (D. Kessel and T. J . Dougherty, ed.) New York, Plenum Publishing Corporation, pp. 129-137. Henderson, B. W. and J . M. Donovan (1988). "Prostaglandin release from peritoneal macrophages and tumor cells after photodynamic therapy." Photochem. Photobiol. 47: 76S. Henderson, B. W. and J . M. Donovan (1989). "Release of prostaglandin E2 from cells by photodynamic treatment in vitro." Cancer Res. 49: 6896-6900. Henderson, B. W. and T. J . Dougherty (1992). "How does photodynamic therapy work?. [Review]." Photochem. Photobiol. 55: 145-157. Henderson, B. W. and V. H. Fingar (1987). "Relationship of tumor hypoxia and response to photodynamic treatment in an experimental mouse tumor." Cancer Res. 47: 3110-3114. Henderson, B. W. and V. H. Fingar (1994). "The role of vascular photodamage in photodynamic therapy." Photochem. Photobiol. 59: 1S-2S. Henderson, B. W., B. Owczarczak, J . Sweeney and T. Gessner (1992). "Effects of photodynamic treatment of platelets or endothelial cells in vitro on platelet aggregation." Photochem. Photobiol. 56: 513-521. References 135 Henderson, B. W., A. B. Sumlin, B. L. Owczarczak and T. J . Dougherty (1991). "Bacteriochlorophyll-a as photosensitizer for photodynamic treatment of transplantable murine tumors." J. Photochem. Photobiol. B: Biol. 10: SOS-SIS. Henderson, B. W., S. M. Waldow, T. S. Mang, W. R. Potter, P. B. Malone and T. J . Dougherty (1985). "Tumor destruction and kinetics of tumor cell death in two experimental mouse tumors following photodynamic therapy." Cancer Res. 45: 572-576. Heufler, C , F. Koch and G. Schooler (1987). "Granulocyte/macrophage colony stimulating factor and interleukins mediate the maturation of murine epidermal Langerhans cells into potent immunostimUlatory dendritic cells." J . Exp. Med. 167: 700-705. Hier, S. K., K. Rotham, L. M. Hier and W. S. Rosenthal (1993). "Phototherapy of esophageal malignancies." Lasers Surg. Med. Suppl. 5: 51 (Abstract). Hilf, R., R. S. Murant, U. Narayanan and S. L. Gibson (1986). "Relationship of mitochondrial function and cellular adenosine triphosphate levels to hematoporphyrin derivative-induced photosensitization in R3230AC mammary tumors." Cancer Res. 46: 211-217. Hilf, R., D.-B. Smail, R. S. Murant, P. B. Leakey and S. L. Gibson (1984). "Hematoporphyrin derivative-induced photosensitivity of mitochondrial succinate dehydrogenase and selected cytosolic enzymes of R3230AC mammary adenocarcinomas of rats." Cancer Res. 44: 1483-1488. Hilf, R., N. W. Warne, D. B. Smail and S. U Gibson (1984). "Photodynamic inactivation of selected intracellular enzymes by hematoporphyrin derivative and their relationship to tumor cell viability in vitro." Cancer Lett. 24: 165-172. . Howard, B. V., H. P. Morris and J . M. Bailey (1972). "Ether-lipids, glycerol phosphate dehydrogenase and growth rate in tumors and cultured cells." Cancer Res. 32: 1533-1538. References 136 Hruza, L, H. Lui, N. Kollias, J . Winberly, P. Dwyer, M. Gagel, V. Salvatori and R. Anderson (1993). "Photodynamic therapy of psoriasis using benzoporphyrin derivative monoacid ring A (BPD-MA): Preliminary observations." Photochem, Photobiol. BIS: A 09S. Inomata, T., Y. Ogawa, F. Hamada, S. Yoshida, T. Maeda and H. Seguchi (1987). "Effects of radiotherapy and Sizophyllan (SPG) on MM46 tumor " inoculated into C3H/He mice." Radiation System Research (Tokyo) 2: 216-219. Jain, R. K. (1989). "Delivery of novel therapeutic agents in tumors: physiological barters and strategies." JNCI81: 570-576. Jori, G. (1989). "In vivo transport and pharmacokinetic behaviour of tumor photosensitizers." CIBA Found. Symp. 146: 78-86. Jori, G., M. Beltramini, E. Reddi, B. Salvato, A. Pagnan* L. Ziron, L. Tomio and T: Tsanov (1984). "Evidence for a major role of plasma lipoproteins as hematoporphyrin carriers in vivo." Cancer Lett. 24: 291-297. Kay, R. and R. K. HUmpries (1991). "New.vectors and procedures for isolating cDNAs encoding cell surface proteins by epression in C O S cells." Meth. Molec. Cell. Biol. 2: 254-265. Kelly, J . F., M. E. Snell and M. C. Berenbaum (1975). "Photodynamic destruction of human bladder carcinoma." Br. J. Cancer 31: 237-244. Kennedy, J . C. and R. H. Pottier (1992). "Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. [Review]." J. Photochem. Photobiol. B: Biol. 14: 275-92. Kennedy, J . C , R. H. Pottier and D. C. Pross (1990). "Photodynamic therapy with endogenous protoporphyrin IX: basic principles and present clinical experience." J. Photochem. Photobiol. B: Biol. 6: 143-148. References 137 Kerdel, F . A., N. A. Soter and H. W. Lim (1987). "In vivo mediator release and degranulation of mast cells in hematoporphyrin derivative-induced phototoxicity in mice." J. Invest. Dermatol. 88: 277-280. Kessel, D. (1986). "Porphyrin-lipoprotein association as a factor of porphyrin localization." Cancer Lett. 33: 183-188. Kessel, D. (1986). "Proposed structure of the tumor localizing fraction of hematoporphyrin derivative." Photochem. Photobiol. 44: 193-196. Kick, G., G. Messer, A. Goetz, G. Plewig and P. Kind (1995). "Photodynamic therapy induces expression of interleukin 6 by activation of AP-1 but not N F - K B DNA binding." Cancer Res. 55: 2373-2379. Kongshaug, M., J . Moan and S. B. Brown (1989). "The distribution of porphyrins with different tumour localising ability among human plasma proteins." Br. J. Cancer59: 184-188. Korbelik, M. (1993). "Distribution of disulfonated and tetrasulfonated aluminum phthalocyanine between malignant and host cell populations of a murine fibrosarcoma." J . Photochem. Photobiol. B: Biol. 20: 173-181. Korbelik, M. and J . Hung (1991). "Cellular delivery and retention of Photofrin II: the effects of interaction with human plasma proteins." Photochem. Photobiol. 53: 501-510. Korbelik, M., J . Hung, S. Lam and B. Palcic (1990). "The effects of low density lipoproteins on uptake of Photofrin II into cultured cells." Photochem. Photobiol. 51: 191-196. Korbelik, M. and G. Krosl (1994). "Cellular levels of photosensitisers in tumours: the role of proximity to the blood supply." Br. J. Cancer 70: 604-10. Korbelik, M. and G. Krosl (1994). "Enhanced macrophage cytotoxicity against tumor cells treated with photodynamic therapy." Photochem. Photobiol. 60: 497-502. References 138 Korbelik, M. and G. Krosl (1995). "Accumulation of benzoporphyrin derivative in malignant and host cell populations of the murine RIF tumor." Cancer Lett. 97: 249-254. Korbelik, M. and G. Krosl (1995). "Photofrin accumulation in malignant and host cell populations of murine fibrosarcoma." Photochem. Photobiol. 62: 162-168. Korbelik, M. and G. Krosl (1995). "Tumor associated macrophages (not tumor cells) are the determinants of photosensitizer tumor localization." SPIE 2371: 344-346. Korbelik, M. and G. Krosl (1996). "Photofrin accumulation in malignant and host cell populations of various tumours." Br. J. CancerlZ: 506-513. Korbelik, M. and G. Krosl (1996). Photosensitizer distribution and photosensitized damage of tumor tissues. Basic principles of Phototherapy. (G. Jori, H. Young and H. Honigsman, ed.) Berlin, Springer Verlag. pp. in press. Korbelik, M., G. Krosl and D. J . Chaplin (1991). "Can PDT be potentiated by immunotherapy?" SPIE 1616: 192-198. Korbelik, M., G. Krosl and D. J . Chaplin (1991). "Photofrin uptake by murine macrophages." Cancer Res. 51: 2251-2255. Korbelik, M., G. Krosl and D. J . Dougherty (1993). "PDT elicited tumoricidal activity of tumor associated macrophages." Photochem. Photobiol. 57: 91S. . Korbelik, M., G. Krosl, G. J . Dougherty and J . D. Chaplin (1992). "Host cell infiltration into PDT treated tumor." SPIE 1645: 124-130. References 139 Korbelik, M., G. Krosl, J . Krosl and G. J . Dougherty (1996). "Modulation of tumor response to photodynamic therapy in severe combined immunodeficient (SCID) mice by adoptively transferred lymphoid cells." SPIE 2675: 156-162. Korbelik, M., G. Krosl, V. R. Naraparayu and N. Yamamoto (1995). "The effect of enzymatically generated macrophage activating factor (GcMAF) on the tumor response to photodynamic therapy." Photochem. Photobiol. 61: 97S. Korbelik, M., G. Krosl, P. L. Olive and D. J . Chaplin (1991). "Distribution of. Photofrin between tumour cells and tumour associated macrophages." Br. J. Cancer64: 508-512. Krosl, G. and M. Korbelik (1994). "Potentiation of photodynamic therapy by immunotherapy: the effect of schizophyllan (SPG)." Cancer Lett. 84: 43-49. Krosl, G., M. Korbelik and G. J . Dougherty (1995). "Induction of immune cell infiltration into murine SCCVII tumour by photofrin-based photodynamic therapy," Br. J. Cancer 71: 549-555. . Krosl, G., M. Korbelik, J . Krosl and G. J . Dougherty (1995). "Potentiation of photodynamic therapy by granulocyte-macrophage colony stimulating factor immunotherapy." SP/E 2371: 347-350. . Lam, S. (1994). "Photodynamic therapy of lung cancer. [Review]." Seminars in Oncology 21 (6 Suppl 15): 15-19. Lam, S., C. MacAulay, J . C. LeRiche, J . Qu, G. Krosl, H. Zeng, M. Korbelik and B. Palcic (1993). "Fluorescence imaging of pre-malignant and malignant tissue." SPIE 1881: 160-167. Lam, S., B. Palcic, D. McLean, J . Hung, M. Korbelik and A. E. Profio (1990). "Detection of lung cancer using low dose Photofrin II." Chest 97: 333-337.; • References 140 Lee See, K., I. J . Forbes and W. H. Betts (1984). "Oxygen dependency of photocytotoxicity with haematoporphyrin derivative." Photochem. Photobiol. 39: 631-634. Levitsky, H. I., A. Lazenby, R. J . Hayashi and D. M. Pardoll (1994). "In vivo priming of two distinct antitumor effector populations: the role of MHC Class I expression." J. Exp. Med. 179: 1215-1224. Lim, H. W. (1989). "Role of mediators of inflammation and cells in porphyrin-induced cytotoxicity." SPIE1065: 28-33. Lim, H. W., M. Hagan and I. Gigli (1986). "Phototoxicity induced by hematoporphyrin derivative in C5- defficient, mast cell-defficient, and leukopenic mice." Photochem. Photobiol. 44: 175-180. Lin, C.-W. (1989). Selective localization of photosensitizers in tumors: a review of the phenomenon and possible mechanisms. Photodynamic therapy of neoplastic disease. (D. Kessel, ed.) Boca Raton, FL, C R C Press, Inc. pp. 79-101. Lindsay, J . M., L. Manning and G. W. Wood (1982). "Exclusive binding of immunoglobulin to Fc gamma receptors on macrophages in 3-methylcholanthrene-induced murine tumors." UNCI 69: 1163-1174. Lipson, R. L. and E; J . Baldes (1960). "The photodynamic properties of a particular hematoporphyrin derivative." Arch. Dermatol. 82: 508-516. Lipson, R. L., M. J . Gray and E. J . Baldes (1966). Hematoporphyrin derivative for detection and management of cancer. Ninth Internat. Cancer Congr. Tokyo, Japan, 393. Lipson, R. L., B. E. J . and A. M. Olsen (1961). "Hematoporphyrin derivative: A new aid for endoscopic detection of malignant disease." J. Thorac. Cardiovasc. Surg. 42:623-629. References 141 LLoyd, A. R. and J . J . Oppenheim (1992). "Poly's lament: the neglected role of the polymorphonuclear neutrophil in the efferent limb of the immune system." Immunol. Today 13: 169-171. Lui, H. (1994). "Photodynamic therapy in dermatology with porfimer sodium and benzporphyrin derivative: an update." Seminars in Oncology 21 (6): 11-14. Lui, H., L. Hruza, N. Kollias, J . Wimberly, M. Gagel, D. McLean, V. Salvatori and R. Anderson (1993). "Photodynamic therapy of malignant skin tumors with benzporphyrin derivative monoacid ring a (BPD-MA)." Photochem. Photobiol. 57: 109S. Lynch, D. H., S. Haddad, C. J . Jolles, V. J . King, M. J . Ott, B. Robertson and R. C. Straight (1989). "Immunologic effects of peritoneal photodynamic treatment." SPIE 1065: 55-65. Lynch, D. H., S. Haddad, K. J . Vernon, M. J . Ott, R . C Straight and C. J . Jolles (1989). "Systemic immunosupression induced by photodynamic therapy (PDT) is adoptively transferred by macrophages." Photochem. Photobiol. 49: 453-458. Ma, L. W., J . Moan and Q. Peng (1992). "Effects of light exposure on the uptake of photofrin II in tumors and normal tissues." Int. J. Cancer52: 120-123. Marcon, N. R. (1994). "Photodynamic therapy and cancer of the esophagus." Seminars in Oncology 21 (6 Suppl 15): 20-23. Markowitz, D., S. Goff and A. Bank (1988). "A safe packaging line for gene transfer: separating viral genes on two different plasmids." J. Virol. 62: 1120-1124. McBride, W. H., J . D. Thacker, S. Comora, J . Economou, D. Kelley, S. M. Dubinett, D. Hogge and G. J . Dougherty (1992). "Genetic modification of a murine fibrosarcoma to produce IL7 stimulates host immune infiltration and tumor immunity." Cancer Res. 52: 3931-3937. References 142 Metcalf, D. (1991). "Control of granulocytes and macrophages: molecular, cellular and clinical aspects." Science 254: 529-533 Midorikama, Y., T. Yamashita and F. Sendo (1990). "Modulation of the immune response to transplanted tumors in rats by selective depletion of neutrophils in vivo using a monoclonal antibody: aborgation of specific transplantation resistance to chemical neutrophils in vivo." Cancer Res. 50: 6243-6247. Miller, A. D., M. F. Law and I. M. Verma (1985). "Generation of helper-free amphotropic retroviruses that transduce a dominant-acting, methotrexate-resistant dihydrofolate reductase gene." Mol. Cell. Biol. 5: " 431-437.; Mitchell, J . B., S. McPherson, W. DeGraff, J . Gamson, A. Zabell and A. Russo (1985). "Oxygen dependence of hematoporphyrin derivative-induced photoinactivation of Chinese-hamster cells." Cancer Res. 45: 2008-2011. Moan, J . and K. Berg (1991). "The photodegradation of porphyrins in cells can be used to estimate the lifetime of singlet oxygen." Photochem. Photobiol. 53: 549-553. . Moan, J . , K. Berg, E. Kvam, A. Western, Z. Malik, A. Ruck and H. Schneckenburger (1989). "Intracellular localization of photosensitizers. [Review]." CIBA Found. Symp. 146: 95-107. Moan, J . , K. Berg, E. Kvam, A. Western, Z. Malik, A. Ruck and H. Schnekenburger (1989). Intracellular localization of photosensitizers. • Photosensitizing compounds: their chemistry, biology and clinical use. (G. Bock and S. Harnett, ed.) Chicester, UK, Wiley Interscience. pp. 95-. 107. Moan, J . , J . V. Johannessen, T. Christehsen, T. Espevik and J . B. McGhie (1982). "Porphyrin-sensitized photo-inactivation of human cells in vitro." Am. J. Pathol. 109: 184-192. References 143 Moan, J . and S. Sommer (1983). "Uptake of the components of hematoporphyrin derivative by cells and tumours." Cancer Lett. 21: 167-174. Morgan, A. R., G. M. Garbo, R. W. Keck, L. D. Eriksen and S. H. Selman (1990). "Metallopurpurins and light: Effect on transplantable rat tumors and murine skin." Photochem. Photobiol. 51: 589-592. Mulroney, C. M., S. Gluck and A. D. Ho (1994). "The use of photodynamic therapy in bone marrow purging. [Review]." Seminars in Oncology 21(Suppl 15): 24-27. Myers, R. C., B. H. S. Lau, D. Y. Kunihira, R. R. Torrey, J . L. Woolley and J . Tosk (1989). "Modulation of hematoporphyrin derivative-sensitized phototherapy with corynebacterium parvum in murine transitional cell carcinoma." Urology 3: 230-235. Nakano, T., K. Oka, T. Sugita and H. Tsunemoto (1993). "Antitumor activity of Langerhans cells in radiation therapy for cervical capcer and its modulation with SPG administration." In Vivo 7: 257-264. Nelson, J . S., L H. Liaw, A. Orenstein, W. G. Roberts and M. W. Berns (1988). "Mechanism of tumor destruction following photodynamic therapy with, hematoporphyrin derivative, chlorin, and phthalocyanine." JNCI80: 1599-605. ; ' Nelson, J . S., W. G. Roberts and M. W. Berns (1987). "In vivo studies on the utilization of mono-L-aspartyl chlorin (NPeg) for photodynamic therapy." Cancer Res. 47: 4681-4685. Ngwenya, B. Z. and N. Yamamoto (1986). "Effects of inflammation products on immune systems: lysophosphatidylcholine stimulates macrophages." Cancer Immunol. lmmunother.2"\: 174-182. References 144 Norisuye, K., T. Yanaki and H. Fujita (1980). "Triple helix of Schizophyllum communae polysaccharide in aqueous solution." J. Polym. Sci. 18: 547-558. Nseyo, U. O., T. J . Dougherty, D. Boyle, W. Potter, L. S. Englander, R. P. Huben and J . E. Pontes (1985). "Experimantal photodynamic therapy of canine bladder." J. Urol A 33: 311-315. Nseyo, U. O., R. K. Whalen, M. R. Duncan, B. Berman and S. L. Lundahl (1990). "Urinary cytokines following photodynamic therapy for bladder cancer." Urology 36: 167-171. Okamura, K., M. Suzuki, T. Chihara and 1. others (1986). "Clinical evaluation of Schizophyllan combined with irradiation in patients with cervical cancer. A randomized control study." Cancer 58: 865-872. Olive, P. L. (1989). "Distribution, oxygenation, and clonogenicity of macrophages in a murine tumor." Cancer Commun. 1: 93-97. Olive, P. L, D. J . Chaplin and R. E. Durand (1985). "Pharmacokinetics, binding and distribution of Hoechst 33342 in spheroids and murine tumors." Br. J. Cancer 52: 739-746. Orenstein, A., J . S. Nelson, L. H. Liaw, R. Kaplan, S. Kimel and M. W. Berns (1990). "Photochemotherapy of hypervascular dermal lesions: a possible alternative to photothermal therapy?" Lasers in Surgery & Medicine 10: 334-343. Parker, J . G. (1986). "Optical detection of the photodynamic production of singlet oxygen in vivo." Lasers. Surg. Med. 6: 258. Pass, H. I. (1993). "Photodynamic therapy in oncology: Mechanisms and clinical use." JNCI85: 443-456. References 145 Peng, Q., J . Moan and L. S, Cheng (1991). "The effect of glucose administration on the uptake of photofrin II in a human tumor xenograft." Cancer Lett: 58: 29-35. Peng, Q., J . Moan, G. Farrants, H. E. Danielsen and C. Rimington (1991). "Localization of potent photosensitizers in human tumor LOX by means of laser scanning microscopy." Cancer Lett. 58(1-2): 17-27. Penning, L. C , K. Tijssen, P. J . Boegheim, J . van Steveninck and T. M. A. R. Dubbleman (1994). "Relationship between photodynamically induced, damage to various cellular parameters and loss of echogenicity in different cell types with hematoporphyrin derivative as sensitizer." Biochem. Biophys. Acta 1221: 250-258. Pericle, F., F. Di Pierro and G. Forni (1992). Clinical trials with local administration of lymphohaematopoietic growth factors. Lymphohaematopoietic Growth Factors II. (R. Martelmasnn, ed.) Berlin, Springer-Verlag. pp. Policard, A. (1924). "Etudes sur le aspects offerts par des tumeurs experimentales examinees a la lumiere de Woods." CR Soc. Biol. 91: 1423-1425. Pottier, R. and J . C. Kennedy (1990). "The possible role of ionic species in selective biodistribution of photochemotherapeutic agents toward neoplastic tissue." J. Photochem. Photobiol 8:1-1.6. Reed, M. W., D. A. Schuschke, D. M. Ackermann, J . I. Harty, T. J . Wieman and F. N. Miller (1989). "The response of the rat urinary bladder microcirculation to photodynamic therapy." J . Urol. 142(3): 865-868. Reed, M. W., T. J . Wieman, K. W. Doak, C. G. Pietsch and D. A. Schuschke (1989). "The microvascular effects of photodynamic therapy: evidence for a possible role of cyclooxygenase products." Photochem. Photobiol. 50: 419-423. References 146 Reed, M. W., T. J . Wieman, D. A. Schuschke, M. T. Tseng and F. N. Miller (1989); "A comparison of the effects of photodynamic therapy on normal and tumor blood vessels in the rat microcirculation." Radial Res. 119: 542-552. Richter, A. M., S. Yip, E. Waterfield, P. M. Logan, C. E. Slonecker and J . G. Levy (1991). "Mouse skin photosensitization with benzoporphyrin derivatives and Photofrin: macroscopic and microscopic evaluation." Photochem. Photobiol. SZ: 281-6. Ris, H.-B., H. J . Altermatt, B. Nachbur, J . C. M. Stewart, Q. Wang, C. K. Lim, R. Bonnett and U. Althus (1992). Clinical evaluation of photodynamic therapy with mTHPC for chest malignancies. Photodynamic therapy and biomedial lasers. (P. Spinelli, M. Dal Fante and R. Marchesini, ed.) Amsterdam, Elsevier Science Publishers B. V. pp. 421-425. Roberts, W. G. and M. W. Berns (1989). "In vitro photosensitization I. Cellular uptake and subcellular localization of mono-L-aspartyl chlorin e6, chloro-aluminum sulfonated phthalocyanine, and photofrin II." Lasers. Surg. Med. 9: 90-101. Roberts, W. G., L. H. Liaw and M. W. Berns (1989). "In vitro photosensitization II. An electron microscopy study of cellular destruction with mono-L-aspartyl chlorin e6 and photofrin II." Lasers. Surg. Med. 9: 102-108. Rossi, F. (1986). "The superoxide forming NAPDH oxidase of phagocytes: nature, machanisms of activation and function." Biochim. Biophys. Acta 853: 65-69. Rotomskis, R., S. Bagdonas and G. Streckyte (1996); "Spectroscopic studies of photobleaching and photoproduct formation of porphyrins used in tumour therapy." J. Photochem. Photobiol. B: Biol. 33: 61-67. References 147 Russell, S. W., G. Y. Gillespie and J . L. Pace (1980). "Evidence for mononuclear phagocytes in solid neoplasms and appraisal of their nonspecific cytotoxic capabilities." Contemporary Topics in ImmunobiologylO: 143-"166. Salet, C, and G. Moreno (1990). "New trends in photobiology. . Photosensitization of mitochondria. Molecular and cellular aspects." J . Photochem. Photobiol. B: Biol. 5: 133-150. Sallusto, F. and A. Lanzavecchia (1994). "Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colbny-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor a." J. Exp. Med. 179: 1109-1118. Samuelsson, B. (1982). Prostaglandins, thromboxanes and leukotrienes: biochemical pathways. Prostaglandins and related lipids. (T..J. Powles, R. S. Bockman, K. V. Honn and P. Ramwell, ed.) New York, Alan R. Liss. pp. 1-19. Shumaker, B. P. and F. W. Hetzel (1987). "Clinical laser photodynamic therapy in the treatment of bladder carcinoma." Photochem, Photobiol. 4: 899-901. Snyder, F. and R. Wood (1969). "Alkylalk-1-enyl-ethers of glycerol in lipids from normal and neoplastic tissues." Cancer Res. 29: 251-257. Spears, J . R., J . Serur, D. Shropshire and S. Paulin (1983). "Fluorescence of experimental atheromatous plaques with hematoporphyrin derivative." J. Clin. Invest. 71: 395-399: Specht, K. G. and M. A. J . Rodgers (1990). "Depolarization of mouse myeloma cell membranes during photodynamic action." Photochem. Photobiol. 51: 319-324. References 148 Sporn, L. A. and T. H. Foster (1992). "Photofrin and light induce microtubule depolymerization in cultured human endothelial cells." Cancer Res. 52: 3443-3448. Star, W. M.j P. A. Marjnissen, S. van den Berg-Blok, J . A. C. Versteeg, K. A. P. Franken and H. S. Reinshold (1986). "Destruction of rat mammary tumor and normal tissue microcirculation by hematoporphyrin derivative photoradiation observed in vivo in sandwich observation chambers." Cancer Res. 46: 2532-2540. Sugawara, I., K. Lee and M. Wong (1984). "Schyzophyllan (SPG)-treated macrophages and anti-tumor activities against syngeneic and allogeneic tumor cells. I. Characteristics of S PG treated macrophages." Cancer Immunol. Immunother. 16: 137-144. Suit, H. D., R. S. Sedlacek, G. Silver and D. Dosoretz (1985). "Pentobarbital anesthezia and the response of tumor and normal tissue in the C3Hf/Sed mouse to radiation." Radial Res. 104: 47-65. Svennevig, J . and H. Svaar (1979). "Content and distribution of macrophages and lymphocytes in solid malignant tumors." Int. J. Cancer 24: 754-758. Tappenier, H. and A. Jesionek (1903). "Therapeutische versuche mit fluoreszirenden stoffen." Muench Med. Wochenschr. 1: 2042-2044. Tarlau, C. J . , H. Barr, A. J . MacRobert and S. G. Brown (1990). Relative merits of porphyrin and phthalocyanine sensitization for photodynamic therapy. Photodynamic Therapy of Neoplastic Disease. (D. Kessel, ed.) Boca Raton, FL, C R C Press, pp. 263-277. Tepper, R. I. and J . J . Mule (1994). "Experimental and clinical studies of cytokine gene-modified tumor cells. [Review]." HGT5: 153-64. Thomas, J . P., R. D. Hall and A. W. Girotti (1987). "Singlet oxygen intermediacy in the photodynamic action of membrane bound hematoporphyrin derivative." Cancer Lett. 36: 295-302. References 149 Tian, M. E., S. L. Qui and Q. Ji (1985). "Preliminary results of hematoporphyrin derivative-laser treatment for 13 cases of early esophageal carcinoma." Adv. Exp. Med. Biol. 193: 21-25. Tromberg, B. J . , S. Kimel, A. Orenstein, S. J . Barker, J . Hyatt, J . S. Nelson, W. G. Roberts and M. W. Berns (1990). "Tumor oxygen tension during photodynamic therapy." J. Photochem. Photobiol. B: Biol. 5(1): 121-126. Tsuchiya, Y., M. Igarashi, M. Inoue and K. Kumagai (1989). "Cytokine ralated immunomodulationg activities of an anti-tumor glucan, Sizofiran (SPG)." J. Pharamcobio-Dyn. 12: 616-625. Volz, W. and R. Allen (1992). Cutaneous phototoxicity of NPeg in man. Photodynamic therapy and biomedical lasers. (P. Spinelli, M. Dal Fante and R. Marchesini, ed.) Amsterdam, Elsevier Science Publishers B. V. pp. 446-448. Witmer-Pack, M. D.s W. Oliver, J . Valinsky, G. Schulerand R. M. Steinman (1987). "Granulocyte/macrophage colony stimulating factor is essential for the viability and function of cultured mouse Langerhans cells." J. Exp. Med. 166: 1484-1489. Wood, G. W. and K. A. Gollahon (1977). "Detection and quantitation of macrophage infiltration into primary human tumors with the use of cell-, surface markers." JNCI59: 1081-1087. Yamamoto, K. and S. Kawanishi (1991). "Oxidation of specific SH protein of mitochondria by photodynamic action of hematoporphyrin. Relevance to uncoupling of oxidative phosphorylation." Biochem. Pharmacol. 42(5): 1087-1092. Yamamoto, N., S. Homma, T. W. Sery, L. A. Donoso and K. J . Hoober (1991). "Photodynamic immunopotentiation: in vitro activation of macrophages by treatment of peritoneal cells with haematoporphyrin derivative and light." Eur. J. Cancer27(4): 467-471. References 150 Yamamoto, N., K. Hoober, N..Yamamoto and S. Yamamoto (1992). "Tumoricidal capacities of macrophages photodynamically activated with hematoporphyrin derivative." Photochem. Photobiol. 56(2): 245-250, Yamamoto, N. and B. Z. Ngwenya (1987). "Activation of mouse peritoneal macrophages by lysophospholipids and ether derivatives of neutral lipids . and phospholipids." Cancer Res. 47: 2008-2013. Yamamoto, N., D. A. ST Claire, S. Homma and B. Z. Ngwenya (1988). "Activation of mouse macrophages by alkylglycerols, inflammation products of cancerous tissues." Cancer Res. 48:6044-6049. Yanhua, D., O. Xiaolin, L. Zongkang and L. Hong (1990). "Scanning electron microscopic observations on photodynamic effects in vitro." Laser Med. Sci. 5: 375-379. Zurier, R. B. (1982). Prostaglandins and inflammation. Prostaglandins. (J. B. Lee, ed.) New York, Elsevier, pp. 91-112. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0087999/manifest

Comment

Related Items