h... a . . . “unfzkuwhhuwwfihg 7. a! ,3? u , 3:21) ., “9% $7 L fr...“ 13 I . .. no !. 1“, u . 1a.... 19:! 1.11. 55; um? v “.1; 4: .. _ 1?. a v ‘71.. .,.. .1. Ca 51’ . .x... ,. . I (_ ‘b‘ ‘ ‘6’.“ 90 V s: \3 “O 01 LIBRARIES ICHIGAN STATE UNIVERSITY EXIST LANSING, MICK 48824-1048 This is to certify that the thesis entitled XENOTRANSPLANTATION OF HUMAN PROSTATE CELL LINES: MODELS FOR STUDIES ON CANCER TREATMENT presented by AMANDA SUE RIVETTE has been accepted towards fulfillment of the requirements for the Master of degree in Zoology WW Major Professor’s Signature {fig/05' Date MSU is an Affinnative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE XENOTRANSPLANTATION OF HUMAN PROSTATE CELL LINES: MODELS FOR STUDIES ON CANCER TREATMENT By Amanda Sue Rivette A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Zoology 2005 ABSTRACT XENOTRANSPLANTATION OF HUMAN PROSTATE CANCER CELL LINES: MODELS FOR STUDIES ON CANCER TREATMENT By Amanda Sue Rivette Prostate cancer is the second leading cause of death from cancer in American men. Using a family of human prostate cancer cell lines developed in our laboratory, that mimic multiple steps in tumor progression, I developed xenograft models using the RWPEZ-W99, WPEl-NB26, and CTPE cell lines. When injected sub-cutaneously into immune-deficient mice, RWPEZ-W99 forms slow growing, WPEl-NBZ6 rapidly growing, and CTPE more aggressive, metastatic tumors. I derived two new cell lines, WPEl-NB26-64 and WPEl-NB26-65, from WPEl—NB26 tumors which show increased growth as compared to the parent WPEl-NB26 cells. Cells express cytokeratin 18, androgen receptor and prostate-specific antigen, establishing their prostatic epithelial origin. WPEl-NB26-65 is more invasive than WPEl-NB26 which, in part, may be associated with increased expression of matrix metalloproteinases, MMP-2 and MMP-9, which are barely detectable in WPEl-NB26 cells. Results also show that chemically modified tetracyclines, potential new drugs for cancer treatment, CMTs 2147 and 2215, inhibited cancer cell growth in vitro and reduced the size and number of RWPEZ-W99 tumors in vivo, which mimic the behavior of the majority of human prostate cancers. These results have important applications in studies on tumor progression and for evaluating the efficacy of new drugs for the treatment of prostate cancer. This thesis is dedicated to Patrick K. Grant, my loving and supportive husband iii ACKNOWLEDGEMENTS I owe a great deal of thanks to those that made possible or assisted me with the work in this thesis. I would like to first thank Dr. Mukta M. Webber, my graduate advisor, for her undying dedication, energetic effort, and constant words of encouragement while helping me complete the requirements for my degree. Dr. Webber, in collaboration with Dr. Richard J. Ablin (University of Arizona) and Dr. William E. Achanzar, (National Cancer Institute at National Institute of Environmental Health Sciences) gave me the opportunity to explore both in vitro and in vivo cancer research, gain experience working with mice, and publish some of my research results in reputable scientific journals. I thank Dr. Webber, Dr. Ablin, and Dr. Achanzar for these valuable and unique experiences which will undoubtedly have a major impact on my success in the future. I thank Dr. Hynda K. Kleinman, (National Institutes of Health) for her advice on performing xenografts. Committee members, Dr. Daniel E. Williams and Dr. Stephen C. Bromley, provided invaluable support and guidance. Their willingness to get involved with my thesis research is greatly appreciated. Dr. Jean A. Gaymer, another committee member, always offered suggestions when needed and did her best to train me toqperform different in vivo techniques and answer all of my questions. Animal care training was also provided by Dr. Sally Walshaw and Brenda Paschke, a veterinary technician at the University Laboratory Animal Resources. I would like to thank both Dr. Gaymer and Brenda for their time, patience, and support. The animal tissue samples were processed by the MSU. I—Iistopathology lab. I thank Dr. Charles Mackenzie along with members of the lab for their time and effort teaching me about tissue preparation, immunostaining, iv and pathology. I would also like to thank Erik J. Tokar and Brooke B. Ancrile for teaching me cell culture. A special thanks to Leanne Pasternak, Mary Tanski, and Elizabeth Grondin who assisted with daily animal care. I also thank Leslie S. Ovitt and Adam K. Keith for their time and assistance with formatting most of the figures and tables in this thesis. Finally and most importantly I would like to thank my husband and family. Thank you Patrick, Mom, Dad, Scott, Carrie, Matt, Melanie, Tony, Andrea, Grandma, Grandpa, and Nan for your unconditional love and support. TABLE OF CONTENTS LIST OF TABLES ............................................................................ xi LIST OF FIGURES ........................................................................... xiii ABBREVIATIONS ........................................................................... xxv OBJECTIVES .................................................................................. xxvii HYPOTHESES ................................................................................. xxviii PART 1 LITERATURE REVIEW .................................................................. 1 CHAPTER ONE THE PROSTATE AND NEOPLASTIC CONDITIONS OF THE PROSTATE. . ....2 Abstract ................................................................................. 3 Keywords .............................................................................. 3 Introduction ........................................................................... 3 Prostate structure and function ............................................. 3 Epithelial marker expression in prostate cell types ...................... 9 Response to androgen and expression of androgen receptor and PSA in prostate cell types ................................................ 9 Benign prostatic hyperplasia ................................................ 11 Prostate intraepithelial neoplasia (PIN) and prostate cancer ........... 13 Conclusions ........................................................................... 19 Literature cited ........................................................................ 21 CHAPTER TWO CHARACTERISTICS OF THREE HUMAN PROSTATE CANCER CELL LINES: PC-3, DU145, AND LNCaP ...................................................... 24 Abstract ................................................................................. 25 Keywords .............................................................................. 25 Introduction ........................................................................... 26 Source of cells ............................................................... 27 Source of PC—3 cells ................................................ 27 Source of DU145 cells ............................................. 27 Source of LNCaP cells .............................................. 28 Cell morphology ............................................................. 29 Epithelial origin .............................................................. 29 Response to androgen and the expression of androgen receptor and PSA .................................................................. 31 Production and response to growth factors .............................. 34 PC-3 cell line ........................................................ 35 vi DU145 cell line ...................................................... 36 LNCaP cell line ...................................................... 37 Adhesion properties ......................................................... 39 Proteases ...................................................................... 43 Invasion in vitro .................................................................................. 46 Conclusions ............................................................................ 48 Literature cited ........................................................................ 50 CHAPTER THREE RWPE-l CELL LINE AND ITS DERIVATIVES: RWPE—2, CTPE, AND THE MNU FAMILY OF CELL LINES .................................................... 56 Abstract ................................................................................. 57 Keywords .............................................................................. 57 Introduction ............................................................................ 58 Source of RWPE-l, RWPE2-W99, MNU, and the CT PE cell line. . .59 Cell morphology ............................................................. 62 Epithelial origin .............................................................. 62 Response to androgen and the expression of androgen receptor and PSA .................................................................. 63 Production and response to growth factors ............................... 66 Adhesion properties ......................................................... 67 Proteases ...................................................................... 71 Invasion in vitro .............................................................. 72 Conclusions ............................................................................ 76 Literature cited ........................................................................ 77 CHAPTER FOUR XENOTRANSPLANTATION OF HUMAN PROSTATE CANCER CELLS ....... 80 Abstract ................................................................................. 81 Keywords .............................................................................. 81 Introduction ............................................................................ 82 Intra-spleen injection ........................................................ 84 Inna-peritoneal injection ................................................... 86 Intravenous injection ........................................................ 89 Subcutaneous injection ...................................................... 91 Orthotopic injection .......................................................... 95 Surgical orthotopic implantation .......................................... 98 Models of bone metastasis ................................................. 100 Conclusions ........................................................................... 102 Literature cited ........................................................................ 104 CHAPTER FIVE TETRACYCLINES: APPLICATIONS IN INHIBITION OF TUMOR PROGRESSION AND METASTASIS ..................................................... 107 Abstract ................................................................................. 108 Keywords .............................................................................. 108 vii Introduction ............................................................................ 109 Chemical modifications of the tetracycline molecule ................... 112 CMT-3 inhibits cell proliferation .......................................... 114 Possible mechanisms of CMT-3 induced cytotoxicity .................. 117 CMT—3 decreases MMP production ....................................... 118 CMTs inhibit Matrigel invasion ............................................ 121 CMT-3 inhibits Dunning tumor growth and metastasis ................. 122 Phase I clinical trial of CMT-3 ............................................. 124 Conclusions ............................................................................. 126 Literature cited ......................................................................... 127 PART 2 ORIGINAL RESEARCH ................................................................... 130 CHAPTER SIX EVALUATION OF THE EFFICACY OF CHEMICALLY MODIFIED TETRACYCLINES (CMTs) AS AGENTS FOR THE TREATMENT OF PROSTATE CANCER: A PILOT STUDY USING CMT 2215 ........................ 131 Abstract ................................................................................. 132 Keywords .............................................................................. 133 Introduction ............................................................................ 133 Materials & Methods ........................................................ 135 In vitro studies ...................................................... 135 Cell culture general ........................................ 135 Dose response using a microplate assay ................ 135 In vivo studies ....................................................... 136 Mice ......................................................... 136 Animal maintenance ....................................... 136 Sucrose solution: vehicle for CMT 2215 ............... 139 Drug stock solutions ....................................... 139 Cells for injections ......................................... 140 Experimental groups ....................................... 140 Animal weights ............................................. 142 Tumor size and histology ................................. 142 Results ........................................................................ 142 In vitro studies ...................................................... 142 Effect of CMT 2215 on anchorage- dependent growth ...................................... 142 In vivo studies ....................................................... 144 Weight ...................................................... 144 Mice with R WPEZ-W99 cell xenografts ......... 144 Mice with CTPE cell xenografts ..................... 145 Tumor development ....................................... 146 R WPEZ—W99 cells ................................. 146 C TPE cells .......................................... 147 Histology .................................................... 147 Histology of R WPEZ-W99 tumors in viii control mice ................................................ 147 Histology of RWPEZ -W99 tumors in CMT-treated mice ...................................... 148 Histology of the C TPE tumors in control mice .................................... 149 Histology of the CTPE tumors in CMT-treated mice ............................. 151 Histology of C TPE tumor metastasis to the lung .................................................. 152 Discussion ..................................................................... 153 Acknowledgements .......................................................... 155 Literature cited ............................................................... 156 CHAPTER SEVEN THE EFFECTS OF CMT 2137 AND 2147 ON TUMOR GROWTH USING THE TUMORIGENIC RWPE2-W99 HUMAN PROSTATE CELL LINE ........... 157 Abstract ................................................................................. 158 Keywords .............................................................................. 158 Introduction ............................................................................ 159 Materials & Methods ........................................................ 160 In vitro studies ...................................................... 160 Cell culture general ........................................ 160 Dose response using a microplate assay ................ 160 In vivo studies ....................................................... 161 Mice ......................................................... 161 Animal maintenance ....................................... 162 Sucrose solution ............................................ 163 Drug stock solution ........................................ 163 Cells for injections ......................................... 164 Experimental groups ....................................... 164 Animal weights ............................................. 165 Tumor size and histology ................................. 166 Results ........................................................................ 166 In vitro studies ...................................................... 166 Effects of CMTs 2137 and 2147 on anchorage— dependent growth .......................................... 166 In vivo studies ....................................................... 167 Animal weight ............................................. 167 Tumor volume ............................................. 168 Histology ..................................... ' ............... 175 Discussion ..................................................................... 177 Acknowledgements .......................................................... 179 Literature cited ............................................................... 180 ix CHAPTER EIGHT SELECTION OF CELL LINES WITH ENHANCED IN VASIV E PHENOTYPE FROM XENOGRAFTS OF THE HUMAN PROSTATE CANCER CELL LINE WPEl-NB26 .................................................................................... 182 Abstract ................................................................................. 183 Keywords .............................................................................. 184 Introduction ............................................................................ 184 Materials & Methods ........................................................ 188 Materials ............................................................. 188 Methods .............................................................. 189 Cells and cell culture ...................................... 189 Growth in nude mice by subcutaneous injection. ......189 Growth in nude mice by intravenous injection... ......190 Selection of WPEl-NB26-64 and WPEl-NB26—65 cell lines ................................................. 190 Cell morphology in vitro ................................. 191 Immunostaining for cytokeratin expression ............ 191 Immunostaining for PSA and AR expression ......... 192 Anchorage-dependent growth in monolayer ........... 192 Invasion assay ............................................. 193 Collection of conditioned medium for MMP activity ................................................... 194 SDS-PAGE zymography ................................. 194 Statistical analysis .......................................... 195 Results ........................................................................ 196 Histology of xenografts in nude mice ............................ 196 Histology of metastases in nude mice ............................ 198 Cell morphology in vitro ........................................... 198 Immunostaining for cytokeratin expression ..................... 198 Expression of prostatic epithelial cell markers in cells... ......199 Anchorage-dependent growth in monolayer ..................... 200 Comparison of invasive ability ..................................... 203 Matrix metalloproteinase (MMP) expression .................... 203 Discussion ...................................................................... 204 Literature cited ................................................................. 210 Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 5.1 Table 7.1 LIST OF TABLES Expression of growth factors and their receptors in human prostatic carcinoma cell lines ............................................ 38 Response to exogenous growth factors ................................. 38 Relative levels of E-cadherin in prostate cells ....................... Expression of proteases and protease inhibitors in human ...42 prostate cancer cell lines .................................................. 45 Incidence of metastasis 6-8 weeks after intra-splenic injection of either PC-3, DU145, or LNCaP cells or their metastatic sublines in athymic mice .............................. 85 Incidence of metastasis after intra-peritoneal injection of PC-3 cells or metastatic sublines of PC-3 cells in athymic rmce ....................................................................... Metastatic potential of PC-3, DU145, or LNCaP cells in athymic mice following intravenous cell injection ................. Incidence of metastasis following subcutaneous injection of PC—3, DU145, or LNCaP cells in athymic mice ................. Incidence of metastasis following orthotopic injection of PC-3, DU145, or LNCaP cells in immune-suppressed mice. . . . . Incidence of metastasis to implanted human and host mouse tissue in SCID mice after tail vein injection of PC-3 or ...88 ...90 ...92 .97 LNCaP cells ............................................................... 101 vatotoxicity of DC and CMT-3 in prostate cells .................... This table shows the relationship between tumor volume, following injection of RWPE2-W99 cells, and percentage of tumors having the indicated size. Tumor volumes have been divided into four groups. The average tumor volume, range, and percent of tumors having the indicated size in each group are shown for control mice, and mice treated with CMT 2137 or CMT 2147 ............................................................ xi ..116 ...170 Table 7.2 This table is a summary of Table 7.1., comparing tumor volumes following injection of RWPE2-W99 cells in control mice and mice treated with CMT2137 or CMT2147. Data are shown as percent of tumors in each group having the indicated tumor volume .............................................................. 172 Table 7.3 In this table tumor volumes resulting from injection of RWPE2-W99 cells have been divided into nine groups which are arranged from the smallest to the largest. Data are shown as percent of tumors in each group having the indicated tumor volume .............................................................. 173 xii LIST OF FIGURES Images in this thesis are presented in color. Figure 1.1 Figure 1.2 Figure 1.3 Figure 1.4 Figure 1.5 Figure 1.6 Figure 1.7 The location of the prostate gland with respect to the remainder of the male reproductive and urinary anatomy (Modified from Starr and McMillian, 1997) ............ 4 Zonal anatomy of the prostate. There are three glandular zones and the anterior fibromuscular stroma (Ahmed et al., 1997) ..................................................... 5 Zonal anatomy of the prostate in anterior-posterior and sagittal planes showing central zone (CZ), peripheral zone (P2) and transition zone (T Z) (Kirby, 1996) .................. 6 The morphology and histology of the central and peripheral zones seen on coronal sections of normal human prostate (Modified from Aumuller,1983) ....................................... 7 Normal prostate gland. Simultaneous demonstration of cell specific markers, X 400. 1: PSA (secretory luminal cell type); 2: high molecular weight cytokeratins (basal cell type); 3: chromogranin A (neuro—endocrine cell type) (Bonkoff and Remberger, 1996) ....................................... 8 Benign prostatic hyperplasia. A. Low power view shows proliferation of glands. B. High-power view shows hyperplastic glands with two layers of cells: an inner columnar and an outer cuboidal or flattened (Cotran et al., 1994) ..................................................... 12 Model for PIN -carcinogenesis in the prostate (Kirby et al., 1996) ...................................................... 15 xiii Figure 1.8 Figure 1.9 Figure 2.1 Figure 2.2 Histology of human prostate tissue. Panels A-D depict hematoxylin-eosin stains, while panels E and F show immunohistochemical analyses. A: Low-power view showing the characteristic heterogeneity of prostate tissue, with this region containing a combination of BPH, PIN, and well-differentiated adenocarcinoma. B: High-power view of a region in panel A, showing details of BPH and PIN. The region of BPH has ducts surrounded by basal cells (arrows), which are not found in the region of PIN. The area of PIN shows a transition within the same duct between normal and atypical hyperchromatic cells that contain larger nuclei with prominent nucleoli. C: High-power view showing a nearby area of human prostate with well-differentiated adenocarcinoma that is invading the peri-neural space (N marks the position of the nerve fiber). Note that the carcinoma cells have large nuclei with very prominent nucleoli (arrows). D: View of a different prostate sample with high-grade PIN and a mixture of Gleason grade 4 and 5 carcinoma in the rest of the field. E: Immunohistochemical staining of PIN and carcinoma using anti-cytokeratin 8, which marks all of the epithelial cells. These PIN lesions have a cribiform pattern (arrows), but are still within the confines of a prostatic duct. F: Immunohistochemical staining of a tissue section containing both PIN and carcinoma using anti-cytokeratin 14, which marks the basal cells. Notably, the PIN displays inconsistent staining, whereas the carcinoma has no staining (Abate-Shen and Shen, 2000) ........................................... l6 Gleason grading system. The changes for each grade as assigned by Gleason are shown. A: Gleason grade 1; B: Gleason grade' 3; C: Gleason grade 4; D: Gleason grade 5 (Modified from Kirby, 1996) ............................................ 17 Expression of cytokeratin 8 in LNCaP cells is shown by the brown cytoplasmic stain, 400x ..................................... 30 Indirect avidin-biotin immunoperoxidase staining of LNCaP cells using mAb to PSA. a, cells stained with PSA antibody; b, control. Bar, 20pm. X 532 (W ebber et al., 1995) .................. 32 xiv Figure 2.3 Figure 2.4 Figure 2.5 Figure 3.1 Figure 3.2 Regulation of PSA expression in the LNCaP cell line. Various concentrations of dihydrotestosterone (DHT) were added to the LN CaP cell line. Twenty four h after treatment, total cellular RNA was prepared and 20 u g of total cellular RNA were subjected to Northern analysis. Relative PSA mRNA levels were determined by densitometrical quantification, and the control is defined as 1.0, DHT 0.1 (3.04), DHT 1.0 (3.96), and DHT 10.0 (4.54) (Modified from Hsieh et al., 1993) ..................................... 33 Western blot analysis of E-cadherin in prostate cells. LN=LNCaP cells; ml, normal prostate epithelial cells; PC3=PC-3 cells; DU=DU145 cells. For comparison purposes, LNCaP cells were analyzed at the same time as normal cells (left panel) and in a different analysis with the other two cell lines (right panel). Signals were quantitated by scanning densitometry of X-ray film. Exposure times were 2 min (left panel) and extended to 15 min (right panel) to increase sensitivity; 50 u g of total cellular protein were loaded in each lane, and probed with HECD-l monoclonal antibody. B-Galactosidase (116 kD) is the molecular weight marker for E-cadherin (124 kD) (Modified from Morton et al., 1993) .................................... 41 In vitro invasion of DU145 and LNCaP cells. Invasive ability of DU145 and LNCaP cell lines was examined by the Boyden chamber in vitro invasion assay. 400,000 cells were plated on each “Matrigel”-coated filter and allowed to invade for 24 h. The invasive ability of the highly invasive DU145 cell line was set at 100% invasion (Modified from Bello, 1996) .......... 48 Derivation of the MNU-transformed cell lines from RWPE-l, a HPV-18 immortalized human prostatic epithelial cell line. The 2A tumor was derived from treatment with MN U at 50 ug/ml and 3B tumor at 100 ug/ml (Webber et al., 2001) .......................................... 61 Characterization of RWPE-l cells. Proteins were detected by immunoperoxidase staining. (a) hematoxylin and eosin staining; (b) positive staining for PSA; (c) positive staining for nuclear androgen receptor. Cells for (b) and (c) were pretreated with 5nM mibolerone; (d) a control lacking primary antibody; (e) and (0 positive staining for cytokeratin 8 and 18, respectively. Scale bar is 20 M. X 625 (Modified from Bello et al., 1997) ................................ 64 XV Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Characterization of RWPE-2 cells. Proteins were detected by immunoperoxidase staining. (a) hematoxylin and eosin staining; (b) positive staining for PSA; (c) positive staining for nuclear androgen receptor. Cells for (b) and (c) were pretreated with 5nM mibolerone; (d) a control lacking primary antibody; (e) and (0 positive staining for cytokeratin 8 and 18, respectively. Scale bar is 20 pM. X 625 (Modified from Bello et al., 1997) ............................. 65 Characterization of MN U cell lines. Morphology of (hematoxylin and eosin stain): (a) RWPE-l, (b) WPEl-NA22, (c) WPEl-NB14, (d) WPEl-NBI 1, (e) WPEl-NB26 cells. F-h: PSA and androgen receptor expression in WPEl-NA22 cells treated with mibolerone, as detected by immunostaining; f, positive staining for PSA; g, positive staining for nuclear androgen receptor; and h, a control lacking primary antibody. Bar = 20 M (Webber et al., 2001) ..................................................... 66 Acinar morphogenesis by RWPE-l and WPEl-NB26 cells in 3-D Matrigel culture. (a) the non-tumorigenic RWPE-l cells form well organized acini of polarized cells around a central lumen, while WPEl-NB26 cells (b) form a disorganized cell mass, Bar = 25 um (Modified from Achanzar et al., in press) ............................. 69 The invasive ability of MN U cell lines compared with that of RWPE-l and DU-l45 cells by a modified Boyden chamber in vitro invasion assay. Cells were plated at 200,000 cells/chamber on a Matrigel-coated filter and allowed to invade for 24 h. +/- SEM. Two tailed t-test is shown as *P = 0.028, **P = 0.007, and ***P = 0.04 (Webber et al., 2001) ...................................................... 73 A comparison of the invasive ability of the three tumorigenic cell lines in vitro is shown where the invasive ability of WPEl-NB26 cells is taken as 100%. Cells were plated at 200,000 cells/Boyden chamber on Matrigel-coated filters and allowed to invade for 48 h. Results are plotted as i SD. *P = 0.1095, **P = 0.0008 (Achanzar et al., 2004) ........... 74 xvi Figure 3.8 Figure 5.1 Figure 5.2 Figure 5.3 A schematic diagram showing steps in the multistep process of carcinogenesis and tumor progression in the human prostate and the points possibly represented by RWPE-l, RWPE-2-W99, MN U, and CTPE cell lines in this progression. The sequence of progression from non- malignant RWPE-l cells to the highly malignant WPEl-NB26 cells: RWPE-1< WPEl-NA22< WPEl-NB14 < RWPE2-W99< WPEl—NB11< CTPE< WPEl-NB26 (Modified from Webber et al., 2001) .................................. 75 A schematic representation of tetracycline and the chemical modifications of tetracycline that generated the CMT-1, CMT-3, and CMT-8 compounds (Modified from Seftor et al., 1998) .................................... 113 Effect of doxycycline (DC) and CMT-3 on proliferation of prostate tumor cell lines. Tumor cells were incubated with various concentrations of DC or CMT-3 for 48 hours in complete culture medium. Cell proliferation activity, defined as synthesis of [3H]-thymidine-labeled DNA, was assayed by 2—hour pulse-labeling the cells with [3H]-thymidine as described in the text. Data presented are for three prostate cancer cell lines. Similar results were obtained for other cell lines. Vertical bars represent mean 1 SEM from four independent determinations (Lokeshwar, 1999) ..................................... 115 Zymographic detection of gelatinases secreted into the conditioned media from cultures treated with CMT-3 or doxycycline. Culture conditioned media (15 til/lane, equivalent to 5 x 103 cells) from TSU-PRl (a,b) and MAT LyLu (c,d) cells were separated by SDS-PAGE (8% polyacrylamide) on a gelatin-embedded (1 mg/rnl) gel and zymography. The positions of purified MMP-2 and MMP-9 are indicated. Note: the major fraction of MMP-2 from MAT LyLu (bottom) cell conditioned media was active (Mr ~64,000), whereas most TSU-PRl (top) MMP-2 was in the latent form (Mr 72,000) (Lokeshwar et al., 2002) .................................................. 119 xvii Figure 5.4 Inhibition of invasive potential of tumor cells by doxycycline (DC) and CMTs. Invasion of tumor cells through the Matrigel-coated filters was assayed following 48 hours of exposure to 5 ug/ml of each drug. Only the drug diluent (0.1% dimethyl sulfoxide) was added to control wells. Percentage of cells that invaded in the control (0.1% DMSO) wells varied from 12.5 i 6.4% for DU145 cells to 17 i 4.2 for MAT LyLu cells. 0.1% DMSO had negligible effect on invasion. Results presented are from three independent experiments (Lokeshwar, 1999) ......................................... 122 Figure 6.1 The facility, experimental design and equipment used for in vivo studies. 6.1a. Clinical Center Building; 6.1b. University Laboratory Animal Resources (ULAR) facility; 6.1e. room for housing immune-deficient mice (nude mice); 6.1d. laminar flow mouse cage rack. The cages were arranged in rows for the four groups of mice; row 1 = RWPE2-W99 controls; row 2 = RWPE2-W99 treated; row 3 = CT PE controls; row 4 = CTPE treated. Mice were housed, one mouse per cage, in autoclaved cages, and provided with autoclaved drinking water and irradiated food. 6.1e. laminar flow hood where gavage feeding was performed; 6.1f. Gavage feeding procedure. The control mice were fed 300 1.1.1 of a 5% sucrose solution in water by gavage. The treated mice were similarly fed with 0.675 mg or 2.25 mg of CMT 2215/mouse in 300 pl of a 5% sucrose solution starting 3 days prior to cell injection. Gavage feeding was performed daily for a total of 10 weeks ........................... 138 Figure 6.2 The effects of CMT 2215 on anchorage-dependent growth of RWPE2-W99 cells. Cells were plated in 96-well plates at a density of 10,000 cells per well and treated for 5 days. Results are plotted as percent of DMSO-treated control, :SEM. ...143 Figure 6.3 Average weight of control and treated mice injected with RWPE2-W99 cells. In the control group, four mice were given vehicle alone (5% sucrose solution in water) by gavage daily for 10 weeks. Four mice in the treated group were given 0.675 mg of CMT 2215/mouse daily by gavage for 10 weeks. The days on which gavage feeding was started, and cells injected, are shown .................................................. 144 xviii Figure 6.4 Average weight of control and treated mice injected with CTPE cells. In the control group, three mice were given vehicle alone (5% sucrose solution in water) by gavage daily for 10 weeks. Three mice in the treated group were given 2.25 mg of CMT 2215/mouse daily by gavage for 10 weeks. The days on which gavage feeding was started, and cells injected, are shown ........................................................ 145 Figure 6.5 a. and b. Nude mice (strain NCRNU-M male, homozygotes, ~8 weeks old from Taconic farms, Germantown, NY) were bilaterally injected subcutaneously with 250 pl of a cell suspension in Matrigel (cellszMatrigel volume 1:1) containing 1 million (a) RWPE2-W99 cells or (b) CTPE cells. Mice were sacrificed 10 weeks later. These mice were fed 300 pl of a 5% sucrose solution by gavage starting 3 days prior to cell injection. Gavage feeding was performed daily for 10 weeks. Arrows point to tumors ................................................... 146 Figure 6.6 Histology of the RWPE2-W99 tumors in control mice: Figure 6.6a shows a subcutaneous tumor (arrow). Under the skin, the tumor margin appears to be well defined and separate from the skin. Figure 6.6b shows tumorzadipose tissue interface with clear margins, and the tumor does not show invasion at this site. It is possible that if the animals are maintained for longer than 10 weeks, one may see invasion and metastasis. Figure 6.6c shows tumor histology at a higher magnification. Figure 6.6d shows (arrow) skeletal muscle cells amongst tumor cells. H & E stain ....................... 148 Figure 6.7 Histology of the RWPE2-W99 tumors in CMT-treated mice. Figure 6.7a shows an area of the tumor that does not appear to show any difference from the control tumor. Figure 6.7b shows a representative area with many vacuolated cells. Figure 6.7c shows (arrow) squamous metaplasia. There are also areas that show large lymphocytic infiltration (dark staining nuclei) (Figure 6.7d). Many areas showed what appear to be apoptotic cells (arrows) (Figure 6.7e). Such changes were not seen as frequently in the control tumors. H & E stain ....................................................... 149 xix Figure 6.8 Figure 6.9 Figure 6.10 Figure 7.1 Histology of CTPE tumors in control mice. The CTPE tumors are rapidly growing, invasive tumors and invasion was observed at the 10 week experimental period. Figures 6.8a shows a subcutaneous tumor with invasion into the sub-epidermal layer. The tumor has infiltrated into the dermis and does not have clear cut margins as can be seen in Figure 6.8b. Figures 6.8c and 6.8d show that the tumor cells are intermingled with skeletal muscle (M) (Figure 6.8c) and fat cells (FC) (Figure 6.8d). The tumor cell population is very heterogeneous with considerable variation in cell size (Figure 6.8e). H & E stain ................................................ 150 Histology of CTPE tumors in CMT-treated mice. This figure shows some features observed in CMT-treated CTPE tumors. A tumor with undifferentiated characteristics shown in Figure 6.9a suggests the aggressive nature of CTPE tumors. Invasion into skeletal muscle (M) is seen in Figure 6%. Cells which appear to be undergoing apoptosis are shown (arrows) in Figures 6.9c and 6.9d. Such cells were not seen as frequently in the control CT PE tumors. Areas with lymphocytic infiltration were observed in several tumors (Figure 6.9e). H & E stain ............................................... 152 Histology of the normal lung and of CTPE tumor metastasis to the lung. Figure 6.10a shows a normal area of the lung. One of the control mice (1/3) showed metastasis to the lung. Figures 6.10b and 6.10c are low magnification picture of the lung (L) showing metastatic tumors (T). Figure 6.10d is a higher magnification picture of the lung (L) showing lung tissueztumor (T) interface. The lung tissue has the alveoli represented by clear spaces against which the tumor tissue has a solid appearance. H & E stain ............................. 153 Laminar flow mouse cage rack. The cages were color-coded for the three groups of mice; yellow = controls; blue = 2137; red: 2147. Mice were housed, one mouse per cage, in autoclaved cages, and provided with autoclaved drinking water and irradiated food. The control mice were fed 300 pl of a 5% sucrose solution in water by gavage. The treated mice were similarly fed with 1.2 mg of CMT 2137 or 2147/mouse in 300 pl of a 5% sucrose solution starting 3 days prior to cell injection. Gavage feeding was performed daily for a total of 11 weeks .............................................. 162 Figure 7.2 Figure 7.3 Figure 7.4 Figure 7.5 The effects of CMT 2137 or CMT 2147 on anchorage dependent growth of RWPE2-W99 cells. Cells were plated in 96-well plates at a density of 10,000 cells per well and treated for 5 days. Results are plotted as percent of DMSO-treated control, :SEM ........................................... 167 Average weight of mice injected with RWPE2-W99 cells. This graph shows weight gain, over time, in the control mice given vehicle alone (5% sucrose solution in water) by gavage daily for 11 weeks, and in the mice gavage-fed with 1.2 mg/mouse of CMT 2137 or CMT 2147 in 5% sucrose solution. Mice were weighed at time zero and then weekly for 11 weeks. Beginning at time zero, mice were gavage-fed for three days prior to the injection of RWPE2-W99 human prostate tumorigenic cell line. The day on which gavage feeding was started (labeled as CMT), and the day when the cells were injected, are shown. RWPE2-W99-C = control; RWPE2-W99-37 = mice on CMT 2137; RWPE2-W99-47 = mice on CMT 2147 ................................ 168 Relationship between tumor volume, following injection of RWPE2-W99, and percentage of tumors having the indicated tumor size. This is a graphic representation of the data shown in Table 1. Tumor volumes have been divided into four groups. The effects of treatment with CMT 2137 or CMT 2147 are evident in this bar graph. Results indicate that overall, the treatment groups have a larger percentage of small tumors as compared to the control .............................. 171 Nude mice (N SWNU-M nu/nu strain) were bilaterally injected with 250 u] of a cell suspension in Matrigel (cell: Matrigel volume, 1:1) containing one million RWPE2-W99 cells. Mice were sacrificed 11 weeks later. This figures shows (from left to right) four representative mice with tumors from each of the three groups. A. Control mice (22-C, 25-C, 29-C, and 27-C); B. CMT 2137- treated mice (37-37, 34-37, 35-37, and 32-37); C. CMT 2147- treated mice (48—47, 50-47, 44-47, and 46-47); arrows point to small tumors ............................................................. 174 xxi Figure 7.6 Figure 8.1 Figure 8.2 H & E-stained sections showing tumor histology. Figures 7.63 and 7.6b: Histology of the RWE2-W99 tumors in control mice at low (Figure 7.6a) and high magnification (Figure 7.6b). The tumor appears to be an undifferentiated tumor with clear margins at the interface with connective and adipose tissue. In Figure 7.6b, many I mitotic figures can be seen (arrows). Figures 7.6c and 7.6d show a tumor from a mouse treated with 2137. Examination of the tumor at higher magnification (Figure 7.6d) shows little evidence of cells in mitosis, in contrast to the control tumors. Figures 7.6c and 7.6f show a tumor from a mouse treated with 2147. Examination of the tumor at higher magnification (Figure 7.61) again shows little evidence of cells in mitosis, in contrast to the control tumors. Bar = 10 pm ................................................................ 176 Derivation of MN U-transformed human prostate epithelial cell lines from RWPE-l, a non-tumorigenic human prostatic epithelial cell line, and the subsequent derivation of WPEl-NB26-64 and WPEl-NB26-65 cell lines. The 2A tumor was derived from treatment with MNU at 50 pg/ml and 3B at 100 pg/ml ...................................................... 186 8.2A. Nude mice (NSWNU-M) were bilaterally injected with 250 pl of a cell suspension in Matrigel (cell: Matrigel volume, 1:1) containing two million WPEl-NB26 cells. Mice were sacrificed 14 weeks later. This figure shows two mice with tumors from which the WPEl-NB26-64 and WPEl-NB26-65 cell lines were derived. Bar = 1 cm. 8.2B. Histological sections of the (a) WPEl-NB26-64 and (b) WPEl-NB26-65 tumors. H & E, Bar = 20 microns. 8.2C. Histological sections of mouse lung tissue: (a) Normal area of mouse lung tissue, (b) Necrotic WPEl-NB26 prostate tumor cells in a blood vessel (arrow) of mouse lung at 20 weeks after intravenous cell injection, (c) WPEl-NBZ6 cells (arrow) surrounded by hyperplastic, fibrous connective tissue in the lung of a mouse at 20 weeks after intravenous cell injection, (d) shows a higher magnification of the metastasis in (c). H & E, Bar = 20 microns. 8.2D. Morphology of (a) WPEl-NB26, (b) WPEl-NB26-64, (c) WPEl-NB26-65 cells. H & E, Bar = 20 microns ........................................................... 197 xxii Figure 8.3 Figure 8.4 Characterization of WPEl-NB26, WPEl-NB26-64, and WPEl-NB26-65 cells on the basis of cellular proteins. Proteins were detected by immunoperoxidase staining. (a-c) positive staining for cytokeratin 18, the inset in each is a control lacking primary antibody; (d-f) positive staining for cytokeratin 5/14, the inset in each is a control lacking primary antibody. Bar = 20 microns .................................... 199 Immunostaining for androgen receptor (Figure 8.4A) and PSA (Figure 8.4B) demonstrates androgen responsiveness and prostatic epithelial origin of WPEl-N B26, WPEl-NB26-64, and WPEl-NB26-65 cell lines. Cells were treated with 5 nM mibolerone for 6 days. Positive nuclear staining for AR is shown in 8.4A,a,b,c for all three cell lines. Panel a1 shows only weak nuclear staining in untreated control. Panel a2 and other insets are negative controls lacking primary antibody. Positive cytoplasmic staining for PSA is shown in 8.4B,a,b,c for all three cell lines. Panel a1 shows very weak staining in untreated control. Panel a2 and other insets are negative controls lacking primary antibody. Bar = 20 microns .................................... 200 xxiii Figure 8.5 Figure 8.6 8.5A. A comparison of the anchorage-dependent growth of WPEl-NB26, WPEl-NB26-64, and WPEl-NB26-65 cell lines. Cells were plated at densities of 625, 1250, 2500, 5000 and 10,000 cells per well in 96-well plates in complete KSFM. Plates were stained with MT'T five days after plating. Absorbance values were measured at 540 nm and plotted i: SEM. Results represent the average of 3 experiments. The growth of both WPEl-NB26-64 and WPEl-NB26-65 cell lines are significantly greater (p<0 .001) in comparison to the WPEl-NB26 cell line at each cell density using ANOVA. 8.5B. The invasive ability of WPEl-NB26-64 and WPEl -NB26-65 cell lines was compared with that of WPEl-NB26 cells by a modified Boyden chamber in vitro invasion assay. Cells were plated at 200,000 cells/ chamber on a Matrigel-coated filter and allowed to invade for 48 h. i SEM. The difference in the invasive ability of WPEl-NB26-65 as compared to WPEl-NB26 is very significant using ANOVA (p<0.001). 8.5C. Zymographic analysis of MMP expression in culture medium of WPEl-NB26, WPEl-NB26-64, and WPEl-NB26-65 cells. Samples of conditioned medium containing 8 pg protein each, were electrophoresed using 10% polyacrylamide gels. Lane 1, WPEl-NB26 cells; Lane 2, WPEl-NB26-64 cells; Lane 3, WPEl—NB26-65 cells. The gel is a representative of 3 independent experiments ................................................. 202 A schematic diagram showing steps in the multistep process of carcinogenesis and tumor progression in the prostate and the points represented by RWPE-l, the MNU cell lines, and the WPEl-NB26 tumor-derived cell lines, WPEl-NB26-64 and WPEl-NB26-65. The ability of WPEl-NB26 cells to form lung metastases after intravenous injection is also shown .................................................... 205 xxiv bFGF BPE BPH CCN U CMT DC DHT ECM EGF ELISA FACS FGF-R HGPIN HPV-18 IGF- 1 IGF— 1 -R Ki-MuSV KSFM ABBREVIATIONS Androgen receptor Basic fibroblast growth factor Bovine pituitary extract Benign prostatic hyperplasia methyl (2-chloroethyl)-3-cyclohexy-1-nitrosourea Chemically modified tetracycline Doxycycline Dihydrotestosterone Extracellular matrix Epidermal growth factor Enzyme-linked immunosorbent assay Fluorescence-activated cell sorter Fibroblast growth factor receptor Human adult bone Human adult lung High grade prostatic intraepithelial neoplasia Human papilloma virus-18 Insulin-like growth factor Insulin-like growth factor receptor Kirsten murine sarcoma virus Keratinocyte serum-free medium XXV MMPs MN U MTT PAL 1 PAl-2 PIN PSA SCID SOI TGF-a TGF-Bl TGF-B l -R TIMPs ULAR uPA Matrix metalloproteinases N—methyl-N—nitrosourea 3-[4,5-dimethylthiazol-2-yl]-2, 5-diphenyl tetrazolium bromide Plasminogen activator inhibitor type-l Plasminogen activator inhibitor type-2 Prostatic intraepithelial neoplasia Prostate specific antigen Severe combined immune deficiency Surgical orthotopic implantation Transforming growth factor-or Transforming growth factor-Bl Transforming growth factor-Bl receptor Tissue inhibitors of matrix metalloproteinases University laboratory animal resources Urokinase-type plasminogen activator xxvi OBJECTIVES . To develop xenograft models for testing drug efficacy . To evaluate the efficacy of chemically modified tetracyclines as agents for the treatment of prostate cancer in cell culture and on xenografts . To deve10p additional xenograft models using a cell line developed in our laboratory xxvii HYPOTHESES Chemically modified tetracyclines will inhibit: l. The growth of prostate cancer cells in culture 2. The growth of prostate cancer cells as xenografts in nude mice xxviii PART 1 LITERATURE REVIEW CHAPTER ONE THE PROSTATE AND NEOPLASTIC CONDITIONS OF THE PROSTATE Abstract The prostate gland is a part of the male reproductive system. Two major prostatic diseases include benign prostatic hyperplasia (BPH) and prostate carcinoma. BPH occurs so frequently in older men it is almost considered a normal aging process. Carcinoma of the prostate is the most frequently diagnosed cancer, excluding skin cancer, and the second leading cause of death in American men. Prostatic intraepithelial neoplasia (PIN) is a common precursor to prostate cancer and it is associated with progressive abnormalities, which are intermediate between normal prostate epithelium and cancer. The importance of recognizing PIN is based on its strong association with prostate carcinoma. Early detection and treatment may prevent progression to invasive and metastatic prostate cancer. Further studies utilizing in vitro human prostate cell lines and in vivo xenograft models that mimic multiple steps in prostate carcinogenesis, invasion and metastasis will assist in the understanding and treatment of the disease. Keywords Androgen receptor, basal cell, benign prostatic hyperplasia, Gleason score, prostate cancer, prostatic intraepithelial neoplasia, prostate specific antigen, secretory luminal cell Introduction Prostate Structure and Function: The prostate is an accessory sex gland in the male reproductive system which also includes the seminal vesicles and the bulbourethreal glands. A diagram of the male reproductive system is shown in Figure 1.1. urinary bladder vas deferens -. ‘ _ _ .. ' — seminal vesrcle ejaculatory ~- duct . - prostate gland prostatic —. urethra ° ,. l- penis I - —i- bulbourethral scrotum gland Figure 1.1 The location of the prostate gland with respect to the remainder of the male reproductive and urinary anatomy (Modified from Starr and McMillian, 1997) The accessory sex glands secrete fluids necessary for sperm movement. The human prostate gland lies posterior to the urinary bladder and surrounds the urethra. The prostatic urethra extends from the urinary bladder to the urethral sphincter and travels directly through the prostate. The prostate secretes 0.5-1m1 of fluid directly into the urethra through several small ducts during ejaculation (Dixon et al., 1999). The size of the prostate has been compared to that of a walnut. It is composed of nonglandular stroma and three glandular regions: the peripheral zone, transitional zone and central zone (Figure 1.2 and Figure 1.3). Transition zone Peripheral zone Figure 1.2 Zonal anatomy of the prostate. There are three glandular zones and the anterior fibromuscular stroma (Ahmed et al., 1997). T2 024 a T217 V “T If {In P2,” \ -" Figure 1.3 Zonal anatomy of the prostate in anterior-posterior and sagittal planes showing central zone (CZ), peripheral zone (P2) and transition zone (TZ) (Kirby, 1996). Prostatic stroma is a complex mixture of smooth muscle cells, fibroblasts, blood vessels, nerves and extracellular matrix and it is concentrated at the anterior surface of the prostate. In each zone of the prostate, prostatic acini are embedded in smooth muscle stroma (Kirby, 1996). The glands empty into the urethra independently and are classified as mucosal, submucosal, or main, depending on their location in the gland (Paulson, 2000). The mucosa] glands are located in the region immediately surrounding the urethra and are surrounded by the submucosal glands while the main prostatic glands are located in the outermost region of the prostate (Cormack, 2001). The peripheral zone in the normal prostate is the largest zone and takes up about 65% of the prostatic volume (Kirby, 1996). The peripheral zone extends around the posterolateral peripheral aspects of the gland. The histology of this zone is characterized by small, simple, acinar spaces lined by tall columnar secretory epithelial cells (Figure 1.4). centrol zone peripheral zone Figure 1.41 The morphology and histology of the central and peripheral zones seen on coronal sections of normal human prostate (Modified from Aumuller,1983). The second largest zone in the normal prostate is the central zone, which comprises about 25% of the prostatic volume. The central zone surrounds the ejaculatory ducts. Histologically, this zone can be identified by the presence of fairly large acini that are lined by low columnar cuboidal epithelium (Figure 1.4). The smallest zone in the normal prostate is the transition zone, it comprises only 5-10% of the prostatic volume. The transition zone is composed of two bilaterally symmetrical lobules found on the two sides of the prostatic urethra. In the three zones of normal prostate, prostatic epithelium is composed of two major cell populations; secretory luminal and basal cells. The basal cells and secretory luminal cells form a pseudo-stratified layer of cells that line the basement membrane in prostate acini (Figure 1.5). Figure 1.5 Normal prostate gland. Simultaneous demonstration of cell specific markers, X 400. 1: PSA (secretory luminal cell type); 2: high molecular weight cytokeratins (basal cell type); 3: chromogranin A (neuro-endocrine cell type) (Bonkoff and Remberger, 1996). Interdispersed with the secretory luminal and basal cells are also occasional neuroendocrine cells (Figure 1.5) (Lalani et al., 1997). Basal cells range from small, flattened cells to more cuboidal cells, whereas the morphology of secretory luminal cells is columnar. There are two morphological types of neuroendrocine cells: (1) open, flask- shaped cells with long slender extensions reaching the lumen, and (2) closed cells without luminal extensions (Abrahamsson et al., 1996). The three cell types found in the prostate differ in their marker expression and in their responses to hormonal regulation. Epithelial marker expression in prostate cell types: Basal cells exclusively express high molecular weight cytoskeletal proteins, cytokeratin 5, 14, and 15 (Nagle et al., 1987; De Marzo et al., 1998). While, the luminal cells express cytokeratins 8 and 18 (Brawer et al., 1985). Neuroendocrine cells have been shown to exhibit basal cell specific cytokeratin immunoreactivity and chromagranin A, a pan-endocrine marker (Bonkoff et al., 1994). Other evidence based on studies of androgen receptor and PSA expression in basal, secretory luminal, and neuroendocrine cells has indicated the presence of cell types with intermediate differentiation in normal and hyperplastic tissues (Bonkoff et al., 1994a). Response to androgen and the expression of androgen receptor and PSA in prostate cell types: Dihydrotestosterone (DHT) is ultimately responsible for the growth of prostate epithelial cells, it is produced from testosterone by an enzyme called S-a reductase. The effects of DHT on the development of the normal prostate gland and the growth of prostate tumors are mediated by the androgen receptor (AR). The three cell types differ in their response to hormones and combined with their epithelial marker expression, some intermediate differentiation has been observed between basal, secretory luminal and neuro-endocrine cells. Basal cells are generally considered to be androgen insensitive and at least a subpopulation may not express the androgen receptor and do not require androgen for survival. However, in some studies basal cells have been shown to express the androgen receptor focally (Bonkoff and Remberger, 1993). In both the developing and adult prostate coexpression of basal cell specific cytokeratins and PSA has also been detected in basal cells (De Marzo et al., 1998). The secretory luminal cells of the prostate express AR and utilize the hormone androgen as a growth, survival, and differentiation factor. Neuro-endocrine cells, similar to most basal cells, lack the AR and are not influenced by circulating androgens (Bonkoff et al., 1993). Yet, neuroendocrine cells have been shown to co-express PSA and chromagranin A (Bonkoff et al., 1994). So it appears that neuroendocrine cells share a common origin with both basal and secretory luminal cell types of the prostate. The expression of AR and the coexpression of basal and secretory luminal markers in some basal cells may also suggest that these two cell types share a common origin. Coupled with the observation that the proliferative activity in normal prostatic epithelium is localized in the basal cell layer, the presence of pluripotent stem cells in the basal cell layer provides a possible explanation for the development of each of these 3 cell types and intermediate cell types (Bonkoff et al., 1994a and b). Abnormal proliferation of cells in the basal cell compartment of the prostate is commonly associated with a condition called benign prostatic hyperplasia (BPH). 10 Benign prostatic hyperplasia: BPH is a common non-malignant condition of the prostate gland in older men. Approximately 50% of men over the age of 50 have symptoms of BPH (Wartinger et al., 1997). BPH results in continuous growth and increases the size of the prostate, but the rate at which the size increases declines over time (Ahmed et al., 1997). BPH has been found to originate from the formation of nodules in the transition zone of the prostate (Rous, 1988). Microscopically the appearance of nodules may be due mainly to glandular proliferation or muscular proliferation of the stroma (Cotran et al., 1994). During the early stages of BPH the nodules are small and, thus, do not disturb the architecture of the prostate, however, as the nodules grow the size of the prostate increases and the architecture is affected. An increase in cell proliferation that results in prostatic growth often causes difficulty and pain during urination due to obstruction of the urethra (Rous, 1988). In patients with BPH, surgery is commonly performed to enlarge the constricted channel or passage through the center of the prostate gland to improve urine flow (W artinger et al., 1997). The cause of BPH is unknown but it is likely related to the action of androgens. DihydrOtestosterone (DHT), which is derived from testosterone by the action of Sa-reductase, regulates prostatic growth. With aging in men, DHT accumulates in the prostate where it initiates cell proliferation (Cotran et al., 1994). The role of DHT in BPH is supported by observations in which an inhibitor of 5a -reductase was given to men with this condition. Treatment with 5d -reductase inhibitor reduced the DHT content of the prostate and resulted in a decrease in prostate volume as well as urinary obstruction (McConnell et al., 1992; Rittmaster, 1994). In aging men, estradiol levels 11 also increase (Cotran et al., 1994). There is evidence to suggest that DHT-mediated prostatic hyperplasia can be influenced by estrogen levels. In castrated animals, prostate hyperplasia can be induced by administration of androgens which can be enhanced by simultaneous administration of l7B-estradiol (Cotran et al., 1994). The hormonal imbalance of androgens and estrogens in older men may lead to a hyperplastic condition of the prostate. Microscopically, in BPH, the epithelium is characteristically arranged into numerous papillary buds and infoldings, which are more prominent than in the normal prostate (Cotran et al., 1994). Hyperplastic glands typically show two layers of cells, an inner columnar and an outer cuboidal or flattened layer (Figure 1.6). Foci of squamous metaplasia and small areas of necrosis are also changes frequently observed in BPH. 4; -, '. a, '— ‘npi'e . - , .. "r’fl’ ‘35. {Sr 4.1.!- s’ .r 7: ~..-s'-<;. 4 21 a . "i. 53*," .. H ‘ ‘ '1‘. :ii.’ ‘4 '. "LE5 as Fir‘s";>j' 5.: '. . fi' I. t. .. .\ 1;" 1.; \‘Vn Figure 1.6 Benign prostatic hyperplasia. A. Low power view shows proliferation of glands. B. High-power view shows hyperplastic glands with two layers of cells: an inner columnar and an outer cuboidal or flattened (Modified from Cotran et al., 1994). Although BPH and early prostate cancer share many of the same signs and symptoms, such as, an increase in the number of cells in the prostate and urinary obstruction, BPH is not considered to be a premalignant lesion, nor a precursor of prostate cancer. While BPH commonly occurs in the transition zone of the prostate, the neoplastic conditions of the prostate gland, prostatic intraepithelial neoplasia (PIN) and prostatic carcinoma, commonly occur in the peripheral zone of the prostate (McNeal et al., 1988; Bostwick, 1994). Prostate intraepithelial neoplasia (PIN) and prostate cancer: Prostate intraepithelial neoplasia (PIN) and prostate adenocarcinoma are malignant forms of prostate cancer. Of the 699,560 new cancer cases among American men in 2004, an estimated 230,110 of these will be prostate cancer (American Cancer Society, 2004). Therefore prostate cancers represent 33% of all diagnosed cancer cases, which is more than any other cancer in men except skin cancer. An estimated 29,500 deaths due to prostate cancer will occur in 2004 (American Cancer Society, 2004). Accounting for 10% of all cancer-related deaths, prostate cancer is the second leading cause of cancer death in men, after lung and bronchus cancers (American Cancer Society, 2004). The majority of prostate adenocarcinomas, more than 70%, arise in the peripheral zone (Kirby, 1996). Although less common, prostate carcinomas may also originate in the central and transitional zones. The central zone, which surrounds the ejaculatory ducts is the source of 10%, while the transition zone is the source of 20% of prostatic cancers (Ahmed et al., 1997). The majority of prostate cancers may arise in the 13 peripheral zone due to the presence of a high concentration of androgen receptors compared to the central and transition zones of the normal prostate (Kirby, 1996). It is not known whether there are distinct differences between carcinomas that arise in different zones of the prostate. Clinical studies suggest that PIN predates prostate adenocarcinoma by 10 or more years (Sakr et al., 1993). PIN has been observed in men in their 305 and is thought to be a precursor of prostate adenocarcinomas arising in the peripheral zone (Sharp et al., 2001; Webber et al., 1999). PIN is characterized by a morphology similar to that seen in prostate cancer in which the basal cells of the epithelium are absent (Kirby, 1996). PIN is also characterized by cytologic changes mimicking cancer, including nuclear and nucleolar enlargement (McNeal and Bostwick, 1986). A model of carcinogenesis in the prostate, depicting the development of pre-malignant PIN into prostate adenocarcinoma is shown in Figure 1.7. 14 A model for carcinogenesis in the prostate ._L.._ 7....._L, ._.____._.__.____t L..- __ ._c- _-. . _-_ .. . _- _.a..__ - -_._... _. - . ._ v_..h ..__.- _. _.-_. “a , -fi‘ .V— -___R C".\{.» ,1ng Carg'lrjorn‘a \C'" a. r l r T r i L g ‘ I l I I 7 ‘3"! ”l “J 'V‘ICI wil__(jff.'_3’f_jf SIT-'9’"? i lr‘ qt..l mlcrglpdasl|Ie / I . ' x ‘ t LUn‘Iflal \ ' . ‘ A ‘ (, _ “L ‘ secret-4.”, ' . , z: .2 . ‘ . l . . r. . _ cell 'ayer ' ~ ~' - - Basa- .‘Tei: !a.er ' Basemex‘l "‘ec‘oraW; . . I '\ .—-\ - \. a r- a, 7 - \ -. _ \ - I ’ I GRADE 1 Lari-4L): 2 i Lag-30E 3 } .- P'ostdtu‘ W" .ie-U't'w =51 "also 1< .-, Figure 1.7 Model for PIN-carcinogenesis in the prostate (Kirby et al., 1996). The diagnosis of PIN as low, moderate or high grade is established by increasing proliferation and cytological changes (Kirby, 1996). Low grade PIN retains an intact basal cell layer, whereas high grade prostate intraepithelial neoplasia (HGPIN) and early invasive cancer are characterized by progressive basal cell layer disruption (Bostwick, 1994). In an autopsy study of 249 cases, HGPIN was observed in 77% of the prostates with cancer, but in only 24% of prostates without cancer (Sakr et al., 1994). This demonstrates the strong association between PIN and prostatic carcinoma. Most prostate cancer lesions are heterogeneous and multifocal (Abate-Shen and Shen, 2000). For example, benign glands, preneoplastic (PIN) foci, and neoplastic foci of varying severity can all be observed in one region of prostate tissue (Figure 1.8). 15 Figure l.8 Histology of human prostate tissue. Panels A-D depict hematoxylin-eosin stains, while panels E and F show immunohistochemical analyses. A: Low- power view showing the characteristic heterogeneity of prostate tissue, with this region containing a combination of BPH, PIN, and well-differentiated adenocarcinoma. B: High-power view of a region in panel A, showing details of BPH and PIN. The region of BPH has ducts surrounded by basal cells (arrows), which are not found in the region of PIN. The area of PIN shows a transition within the same duct between normal and atypical hyperchromatic cells that contain larger nuclei with prominent nucleoli. C: High-power View showing a nearby area of human prostate with well-differentiated adenocarcinoma that is invading the peri-neural space (N marks the position of the nerve fiber). Note that the carcinoma cells have large nuclei with very prominent nucleoli (arrows). D: View of a different prostate sample with high- grade PIN and a mixture of Gleason grade 4 and 5 carcinoma in the rest of the field. E: Immunohistochemical staining of PIN and carcinoma using anti- cytokeratin 8, which marks all of the epithelial cells. These PIN lesions have a cribiform pattern (arrows), but are still within the confines of a prostatic duct. F: Immunohistochemical staining of a tissue section containing both PIN and carcinoma using anti-cytokeratin 14, which marks the basal cells. Notably, the PIN displays inconsistent staining, whereas the carcinoma has no staining (Abate-Shen and Short, 2000). Although cytologically prostate cancer is variable, generally nuclei are large and vacuolated and contain one or more large nucleoli. Microscopically, besides the absence of the basal cell layer, another characteristic of prostate cancer is the observation of glands growing ‘back to back’ with no intervening stroma (Kirby, 1996). In well- differentiated tumors, the glandular pattern is apparent, however in some poorly differentiated tumors the glandular pattern is only visible upon careful examination (Cotran et al., 1994). In such cases the tumor cells tend to grow in cords, sheets, or nests with varying amounts of stromal production (Cotran et al., 1994). To assist with the prognosis of prostate cancer and as a way to effectively communicate with other pathologists, the Gleason grading system was developed (Kirby, 1996). Using the Gleason grading system the pattern of the tumor is assigned two grades (ranges from l-5) which are added together to obtain the Gleason score (ranges from 2-10). The changes for each grade as assigned by Gleason are shown in Figure 1.9. c . D Figure 1.9 Gleason grading system. The changes for each grade as assigned by Gleason are shown. A: Gleason grade 1; B: Gleason grade 3; C: Gleason grade 4; D: Gleason grade 5 (Modified from Kirby, 1996). Gleason grade 1 tumors are comprised of small, uniform glands exhibiting minimal nuclear changes; the nodules typically possess well-defined borders. Gleason grade 3 is the most common grade of prostate cancer (Kirby, 1996). These tumors show a high degree of variation in architecture, glandular size, shape and regularity; infiltrative borders are generally found. Gleason grade 4 and 5 represent more aggressive tumors with cytologic atypia, extensive infiltrative borders, seminal vesicle extension, and/ or metastatic spread. It is generally accepted that human prostatic carcinomas are slow growing. Prostate capsular invasion occurs early in invasive prostatic cancer, local and direct spread of cancer is at first mostly posterior. Direct local spread of cancer cells often leads to ureteric obstruction with invasion frequently shown in the fatty tissue between the prostate and rectum, the seminal vesicles, and the bladder (Rose, 1985). Metastatic spread of prostate cancer to the bone is the main cause of morbidity among cancer patients, other common sites of metastases are the lymph nodes and liver (Cotran et al., 1994). The bones commonly involved, in descending order of frequency, are lumbar spine, proximal femur, pelvis, thoracic spine and ribs (Cotran et al., 1994). In contrast to prostate cancer, benign tumors and low grade PIN do not metastasize to other organs. There are several methods for prostate cancer detection which include: palpation, transurethral resection, transrectal ultra sound, and serum PSA levels. Blood tests to detect PSA and/ or the digital rectal examination are most commonly performed to screen for prostate cancer. Most PSA produced by prostate cells remains within the prostate, while only a small amount gets absorbed into the bloodstream. When a carcinoma develops in the prostate the patient’s PSA levels tend to rise. If elevated PSA levels or a 18 prostatic nodule are detected upon screening then a biopsy is taken to make the diagnosis of cancer. Early prostate cancer has few or no symptoms, but with more advanced disease individuals may experience difficulty or pain during urination which are also symptoms caused by benign conditions (American Cancer Society, 2004). Surgery and radiation may be used to treat early-stage prostate cancer. Supplemental therapies for early-stage disease include: hormonal therapy, chemotherapy, and radiation (or combinations of these treatments) (American Cancer Society, 2004). These therapies may also be used to treat metastatic disease. The extent of prostate cancer can be determined by CT scans, magnetic resonance imaging, or pelvic lymphadenectomy. Instead of immediately treating prostate cancer, an individual that may have a limited life span or a less aggressive stage of the disease, may also elect to just ‘watch and wait’. Many factors seem to play a role in the development of prostate cancer, including environmental influences, race, age, nutrition, cigarette smoking and family history (American Cancer Society, 2004; Plaskon et al., 2003). Although some studies have suggested a genetic linkage in the familial aggregation of prostate cancer, researchers have had difficulty identifying a single susceptibility gene (Simard et al., 2003). Efforts to assist with the treatment of prostate cancer are currently directed towards the - development of potential markers of malignant potential for prostate cancer. Conclusions The phenotypic and genetic abnormalities which result in the development of BPH and prostatic adenocarcinoma are being intensely investigated. The goal of many investigators is to develop a diagnostic method which can be readily employed to 19 accurately predict the behavior of an individual cancer. Due to the latent onset of HGPIN and its strong association with prostate cancer, early detection and treatment may prevent progression to invasive and metastatic prostate cancer. Further studies utilizing in vitro human prostate cell lines and in vivo xenograft models that mimic multiple steps in prostate carcinogenesis, invasion and metastasis will assist in the understanding and treatment of the disease. 20 Literature cited Abate-Shem, C. and Shen, M.M.: Molecular genetics of prostate cancer. Genes & Development 14:2410-2434, 2000. Abrahamsson, P.: Neuroendocrine differentiation and hormone-refractory prostate cancer. Prostate Supplement 623-8, 1996. Ahmed, M.M., Lee, CT. and Oesterling, J.E.: Current trends in the epidemiology of prostatic diseases: benign hyperplasia and adenocarcinoma. In: Prostate: Basic and Clinical Aspects, R.K. Naz (ed.), Boca Raton, FL, CRC Press, pp.3-25, 1997. American Cancer Society: Cancer Facts and Figures 2004, Atlanta, GA, pp.16-l7, 2004. Aumuller, ,G.: Morphologic and endocrine aspects of prostatic function. Prostate 4:195 214, 1983. Brawer, M.K., Peehl, D.M., Stamey, T.A., and Bostwick, D.G.: Keratin immunoreactivity in the benign and neoplastic human prostate. Cancer Research 45:3663-3667, 1985. Bonkoff, H. and Remberger, K.: Differentiation pathways and histogenetic aspects of normal and abnormal prostatic growth: A stem cell model. Prostate 28298-106, 1996. Bonkoff, H. and Remberger, K.: Widespread distribution of nuclear androgen receptors in the basal cell layer of the normal and hyperplastic human prostate. Virchows Arch [A] 422:35-38, 1993. Bonkoff, H., Stein, U., and Remberger, K.: Multidirectional differentiation in the normal, hyperplastic, and neoplastic human prostate. Simultaneous demonstration of cell specific epithelial markers. Human Pathology 25:42-46, 1994a. Bonkoff, H., Stein, U. and Remberger, K.: The proliferative function of basal cells in the normal and hyperplastic human prostate. Prostate 24:114-118, 1994b. Bonkoff, H., Stein, U., and Remberger, K.: Androgen receptor status in endocrine paracrine cell types of the normal, hyperplastic, and neoplastic human prostate. Virchows Arch [A] 423:291-294, 1993. Bostwick, D.G.: High grade prostatic intraepithelial neoplasia. Cancer Supplement 75:1823-1836, 1994. 21 Campbell, N.A., Reece, IE. and Mitchell, L.G.: Animal reproduction. In: Biology, Fifth Edition, Benjamin Cummings (ed.), Menlo Park, CA, pp.9l3-935, 1999. Cormack, D.H.: Essential Histology, Baltimore, Lippincott, Williams and Wilkins, pp. 463, 2001. Cotran, R.S., Kumar, V. and Robbins, S.L.: Male genital system. In: Robbins Pathologic Basis of Disease, Fifth Edition, F.J. Schoen (ed), Philidelphia, W.B. Saunders Company, pp. 1023-1031, 1994. De Marzo, A.M., Nelson, W.G., Meeker, AK, and Coffey, D.S.: Stem cell features of benign and malignant prostate epithelial cells. Journal of Urology 160:2381 2392, 1998. Dixon, J .S., Chow, RH. and Gosling, J .A.: Anatomy and function of the prostate gland. In: Textbook of Prostatitis, J .C. Nickel (ed.), Oxford, UK, Isis Medical Media Ltd., pp.33-46, 1999. Kirby, R.S., Christmas, T.J., and Brawer, M.: Anatomical and pathological considerations. In: Prostate Cancer, New York, Mosby, pp. 3-21, 1996. Lalani, E., Laniado, M.E., and Abel, P.D.: Molecular and cellular biology of prostate cancer. Cancer and Metastasis Reviews 16:29-66, 1997. McConnell, J .D. et al.: Finasteride, an inhibitor of Su-reductase, suppresses prostatic dihydrotestosterone in men with benign prostatic hyperplasia. Journal of Endocrinological Metabolism 74:505-508, 1992. McNeal, 1E. and Bostwick, D.G.: Intraductal dysplasia: a premalignant lesion of the prostate. Human Pathology 17:64-71, 1986. McNeal, J .E., Redwine, E.A., Freiha, RS. and Stamey, T.A.: Zonal distribution of prostatic adenocarcinoma: Correlation with histologic pattern and direction of spread. American Journal of Surgical Pathology 12:897-906, 1988. Nagle, R.B., Ahmann, F.R., McDaniel, K.M., Paquin, M.L., Clark, V.A., and Celniker, A.: Cytokeratin characterization of human prostatic carcinoma and its derived cell lines. Cancer Research 47:281-286, 1987. Paulson, D.F.: Male reproductive system. In: Histology and Cell Biology, Fourth Edition, New York, McGraw-Hill, pp.279—289, 2000. Plaskon, L.A., Penson, D.F., Vaughan, TL. and Stanford, J .L.: Cigarette smoking and risk of prostate cancer in middle-aged men. Cancer Epidemiology Biomarkers and Prevention 12:604-609, 2003. 22 Rittmaster, R.S.: Finasteride. New England Journal of Medicine 330: 120-125, 1994. Rose, E.F.: Neoplasms of the genital conduit system. In: Genitourinary Oncology, D.A. Culp and SA. Loening (eds), Philadelphia, Lea & Febiger, pp. 382-464, 1985. Rous, S.N.: Normal anatomy and normal function. In: The Prostate Book: Sound Advice on Symptoms and Treatment, New York, Norton and Company, pp.19-26, 1988. Sakr, W.A., Haas, G.P., Cassin, J .J ., Pontes, J .E., Crissman, J .D.: The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. Journal of Urology 150:379-385, 1993. Sharp, R.M., Bello-DeOcampo, D., Quader, S. and Webber, M.M.: N-(4-hydroxyphenyl) retinamide (4—HPR) decreases neoplastic properties of human prostate cells: an agent for prevention. Mutation Research 496: 163-170, 2001. Simard, J., Dumont, M., Labuda, D., Sinnett, D., Meloche, C., El-Alfy, M., Berger, L., Lees, E., Labrie, F. and Tavtigian, S.V.: Prostate susceptibility genes: lesions learned and challenges posed. Endocrine-Related Cancer 10:225-259, 2003. Starr, C. and McMillian, B., (eds): Human Biology, Belmont, CA, Wadsworth, pp. 1-531, 1997. Wartinger, D.D., Webber, M.M., Chu, W.W., and Bello-Deocampo, D.: Benign tumors of the prostate. In: Prostate Health Watch. Webber, M.M., Wartinger, D.D., Williams, D.E. (eds), pp. 2-6, 1997. Webber, M.M., Bello-DeOcampo, D., Quader, S., DeOcampo, N ., Metcalfe, W.S. and Sharp, R.M.: Modulation of the malignant phenotype of human prostate cancer cells by N-(4-hydroxyphenyl)retinamide (4-HPR). Clinical & Experimental Metastasis 17:255-263, 1999. 23 CHAPTER TWO CHARACTERISTICS OF THREE HUMAN PROSTATE CANCER CELL LINES: PC-3, DU145, AND LN CaP 24 Abstract The processes of invasion and metastasis in prostate cancer are complex. To increase our understanding of these complex processes, several in vi vo models, developed from well characterized cell lines, would be useful. Invasive tumor cells show an increase in the expression of growth factors, alter their cell:cell and cellzmatrix interactions, and express proteases that degrade the extracellular matrix. Each of these invasive characteristics have been studied in the PC-3, DU145, and LNCaP human prostate cancer cell lines. Based on data collected from in vitro studies, all three cell lines are invasive and may serve as suitable models for advanced and metastatic prostate cancer. The LNCaP cell line, however, also carries an androgen receptor mutation. LNCaP cells are therefore useful for studying: i) prostate cancers carrying androgen receptor (AR) mutations and ii) response of prostate cancer cells to androgens, anti- androgens, other hormones, and drugs and agents, whose effects may be mediated by AR in prostate cancers carrying AR mutations. While data collected from a xenograft model, using PC-3 or DU145 cell lines, would be useful and applicable to prostate cancer patients with advanced prostate cancer that does not carry AR mutations. Keywords Cell line, DU145, invasion, LNCaP, metastatic, PC-3, prostate cancer 25 Introduction Metastatic spread is the main cause of death in prostate cancer patients. It is, therefore, important to develop in vi vo models for studying prostate cancer metastasis using human cell lines. A few prostate cancer cell lines of human origin are currently available. Some of these cell lines may be suitable for in vivo studies, because their invasive behavior has been studied in vitro. Cell morphology, differentiation, regulation of growth, and expression of adhesion molecules and proteases all contribute to the invasive behavior of cancer cells. Knowledge of these characteristics of cells and their invasive behavior in vitro would enable us to select a suitable model for studying advanced and metastatic prostate cancer. The use of human prostate cancer cell lines in immune-suppressed mice to generate xenografts, permits a direct comparison between the histopathology and molecular biology of the patient-derived specimen and resulting tumors in mice. Since studies using xenografts are an important step before clinical trials of new drugs can be conducted for cancer treatment, the xenograft model should mimic the human disease as closely as possible in order to collect relevant and applicable data. Xenograft models of human prostate cancer can be used to correlate the invasive behavior of a cell line in vitro with its invasive and metastatic behavior in vivo. Xenograft models are particularly useful for testing new drugs for chemotherapy. To date most of the studies in the field of human prostate cancer have focused only on three human prostate cancer cell lines, PC-3, DU-145, and LNCaP, because they were developed in the period from 1977 to 1980 and became readily available. Other human prostate cancer cell lines developed more recently are not readily available at the present time. All cell lines should be considered when planning research using xenograft models 26 to select the most appropriate cells for collecting data prior to human trials. The objective of my own research is to develop xenograft models using newly developed human prostate cancer cell lines. The first step in selecting a cell line for in vivo studies is to become familiar with the invasive characteristics of the cell line. In this review the characteristics of the PC-3, DU145, and LNCaP human prostate cancer cell lines and their potential applications, advantages, as well as limitations as xenograft models, are discussed. Source of cells: Source of PC-3 cells PC-3 cells were derived from a 62 year old Caucasian male (Kaighn et al., 1979). His symptoms included urinary retention, weight loss and anemia. Biopsy of the prostate revealed poorly differentiated prostatic adenocarcinoma. Next the patient underwent bilaterial orchiectomy and was treated with diethylstilbestrol, which only transiently improved his condition. Four months later the patient’s condition worsened despite cryotherapy and he died. Tumor tissue was taken from a rib of the patient shortly after his death for in vitro culture to initiate the PC-3 cell line. Autopsy also revealed abundant tumor in the bone marrow and metastasis in the adrenal. The first report of the PC-3 cell line was published in 1978 (Kaighn et al., 1978). Source of DU145 cells In 1975, a 69 year old white male was admitted to Durham Veterans Administration Hospital with widespread metastatic carcinoma of the prostate (Mickey et 27 ti al., 1980). He also had a three year history of lymphocytic leukemia. He was treated with diethylstilbestrol, but subsequent evaluation showed further metastatic spread to the central nervous system. He then underwent bilateral orchiectomy, transurethral resection of the prostate and parieto-occipital craniotomy to excise a tumor mass. Tissue removed from the metastatic central nervous system lesion was taken for in vitro culture to initiate the DU 145 cell line. The brain metastasis was identified as a moderately differentiated adenocarcinoma with foci of poorly differentiated cells. There was brief improvement of central nervous system function but the symptoms recurred and the patient died in 1976 with pneumonia and septicemia. Autopsy results showed tumor metastasis to the right femoral neck, vertebral column, periaortic and celiac nodes, liver, lungs, and brain. The first report of the DU145 cell line was published in 1977 (Mickey et al., 1977). Source of LNCaP cells LNCaP cells were derived in 1977 from a patient with metastatic carcinoma in a lymph node. The patient was diagnosed one year earlier with moderately differentiated adenocarcinoma of the prostate and showed minimal response to hormone therapy and no response to chemotherapy (Horoszewicz et al., 1980). Initially the patient was treated with oral estrogen, but six months later disseminated bone metastases were found. Next the patient underwent orchiectomy which only resulted in temporary response. After experiencing pain in his right leg he was admitted to Roswell Park Memorial Institute and was treated with methyl (2-chloroethyl)-3-cyclohexy-1-nitrosourea (CCNU), which is an alkalating agent and Estracyt, a combination of estrogen and mechlorethamine. One month later he was diagnosed with metastatic carcinoma in a lymph node, the material 28 aspirated for biopsy was used to initiate LNCaP cell culture in vitro. Hydronephrosis was also observed at this time due to increased pressure on the ureter by the metastatic lymph node. As a result the patient was treated with cis-platinum at which time his disease progressed rapidly. A high dosage of the anti-inflammatory drug, Decadron, temporarily improved his condition but the patient died six months after admission and one and a half years after diagnosis. The first report of the LNCaP cell line was published in 1980 (Horoszewicz et al., 1980). Prostate cancer arises from epithelial cells lining the glands of the prostate. Therefore, the first step in the process of selecting a cell line to study prostate cancer is to establish that the cells are of epithelial origin. To confirm epithelial origin, one would observe cell morphology and expression of epithelial cell markers. Cell morphology: Both PC-3 and DU145 cell lines show typical, polygonal, epithelial morphology. The LNCaP cell line does not have distinct epithelial morphology. The LNCaP cells have a more spindle-shaped and elongated morphology, which is sometimes observed in invasive prostate cancer cells (W ebber et al., 2001). Epithelial origin: The cytoskeletal proteins, cytokeratins, serve as an important marker for establishing epithelial origin of cells. To insure that the cells have an epithelial origin, immunocytochemistry is commonly performed. Secretory luminal cells of the prostate express cytokeratins 8 and 18 and the basal epithelial cells express cytokeratins 5 and 14. 29 PC-3, DU145, and LNCaP cells are all positive for cytokeratins 8 and 18, but negative for cytokeratins 5 and l4 by immunohistochemistry (Figure 2.1) (Mitchell et al., 2000). In another study the basal cell marker, cytokeratin 5, was detected in PC-3 and DU145 cell lines (van Bokhoven et al., 2003). The expression of cytokeratin 5 in PC-3 and DU145 cells is consistent with data collected from human prostate samples. In regressed and therapy-resistant prostate cancers, an increase in cytokeratin 5-positive tumor cells was noted when compared with untreated carcinomas (Gil Diez de Medina et al., 1998). PC-3 and DU145 cell lines are both apparently derived from secretory luminal epithelial cells and may also contain cells that represent either an intermediate phenotype or dedifferentiated luminal cells, while the LNCaP cell line is apparently derived from secretory luminal epithelial cells. Figure 2.1 Expression of cytokeratin 8 in LNCaP cells is shown by the brown cytoplasmic stain, 400X (Modified from Mitchell et al., 2000). 30 The next step in characterizing a cell line is to establish that it is indeed of prostatic origin, by determining that cells express some prostate specific marker proteins. One such protein is prostate specific antigen (PSA). Response to androgen and the expression of androgen receptor and PSA: PSA is a component of the seminal fluid and serves as a marker for prostatic origin of cells. Normal prostatic epithelial cells are stimulated to grow by androgen and they secrete PSA in response to androgen exposure. Most studies show that PC-3 cells do not express PSA or AR and are androgen insensitive, however, weak staining for PSA has been reported (Kaighn et al., 1979; Garde et al., 1993; Mitchell et al., 2000). Some studies show that the DU145 cell line does not express androgen receptor and is therefore, hormone-insensitive and does not express PSA (Paulson et al., 1977; Garde et al., 1993; Mitchell et al., 2000). However other studies have shown the presence of androgen binding sites in the DU145 cell line using a radioligand binding assay, as well as, immunohistochemistry (Carruba et al., 1994; Brolin et al., 1992). Regardless of whether or not these cell lines express androgen receptor, PC-3 and DU145 are androgen-insensitive. The ability of PC-3 and DU145 cells to grow in the presence or absence of androgen is consistent with the growth of the tumor cells in both patients. Neither orchiectomy, nor diethylstilbestrol treatment slowed the growth of the prostate cancer in either patient. Since both the DU145 and PC-3 cell lines are hormone insensitive, they show increased expression of several growth factors to assist in their growth and survival. 31 LNCaP cells respond to androgen and it stimulates their growth, as well as, cytoplasmic PSA expression (Figures 2.2 and 2.3) (Bems et al., 1986; Hsieh et al., 1993; Webber et al., 1995; Mitchell et al., 2000). PSA is also constitutively secreted by LNCaP cells even in the absence of an androgen stimulus. This may be due to a mutation in the androgen receptor. Figure 2.2 Indirect avidin-biotin immunoperoxidase staining of LNCaP cells using mAb to PSA. a, cells stained with PSA antibody; b, control. Bar, 20pm. (Webber etal., 1995). 32 , DHT 1.0 nM g _'j DHT10.0 nM if Control 1, '_ DHT 0.1nM . a PSA ’ Figure 2.3 Regulation of PSA expression in the LNCaP cell line. Various concentrations of dihydrotestosterone (DHT) were added to the LNCaP cell line. Twenty four h after treatment, total cellular RNA was prepared and 20 pg of total cellular RNA were subjected to Northern analysis. Relative PSA mRNA levels were determined by densitometrical quantification, and the control is defined as 1.0, DHT 0.1 (3.04), DHT 1.0 (3.96), and DHT 10.0 (4.54) (Modified from Hsieh et al., 1993). The androgen receptor mutation in the LNCaP cells makes the cells responsive not only to androgens but also to anti-androgens, estrogen, and progesterone (Schuurmans et al., 1991; Jiang et al., 2004). The patient that was the source of the LNCaP cell line was treated with estrogen to counter the effects of androgen on prostate cell growth, but six months later bone metastases were found. The altered ligand responsiveness observed may have assisted in the ligand-independent activation of the androgen receptor after estrogen treatment, thus, permitting tumor progression in the patient. In addition to hormones, growth factors can also assist in androgen-independent cell growth and survival. 33 Production and response to growth factors: Growth factors enable cells to maintain local homeostasis and adapt to their biological microenvironment. The secretion of growth factors allows cells to control promotion or inhibition of cellular proliferation and many other functions, such as, cell differentiation and increased or decreased expression of PSA and androgen receptor (Henttu et al., 1993). A number of growth factors produced by epithelial and stromal cells stimulate prostate cell growth and proliferation. Epidermal growth factor (EGF) and transforming growth factor-a (TGF-a) are stimulatory factors secreted by epithelial cells that share a receptor on epithelial cells. However, TGF-u is only expressed during embryonic development in normal cells. A variety of tumor types have been found to secrete TGF—a, and not the normal EGF (Connolly and Rose, 1990). Stromal derived growth factors that have a stimulatory effect on prostate cells include basic fibroblast growth factor (bFGF), and insulin-like growth factor (IGF). The receptors for these growth factors reside at the cell membrane of both epithelial and stromal cells. Transforming growth factor-[31 (T GF-Bl) inhibits epithelial cell growth, however, some prostate cancer cells may lose the ability to be inhibited by TGF-Bl or show a decrease in growth inhibitory response. Growth factors may be secreted by cells which may affect their own behavior in an autocrine manner, or that of a neighboring target cell in a paracrine manner. One characteristic of progression from a normal to a malignant phenotype is reflected in the increased rate of cell proliferation caused by the autocrine secretion of growth factors (Culig et al., 1994). Increased expression of growth factors may also enhance invasion in 34 prostate cancer cells (Jarrard et al., 1994). PC-3, DU145, and LNCaP cells secrete and respond to some of their own growth factors, therefore, unlike normal prostate epithelial cells, these cell lines are not as dependent on the local microenvironment for cell growth and survival. The increased expression of these growth factors permits these cells to survive in various microenvironments. A comparison of growth factor secretion and receptor expressiOn in PC-3, DU145, and LNCaP cell lines is shown in Table 2.1. A comparison of growth factor response in PC-3, DU145, and LNCaP cell lines is shown in Table 2.2. PC-3 cell line PC-3 cells exhibit low or undetectable levels of both the receptors and their ligands, for example, epidermal growth factor receptors (EGF-R) and its ligands, EGF and TGF—a (Carruba et al., 1994). Addition of exogenous EGF or exogenous TGF-o. to PC-3 cells does not affect cell growth in monolayer cultures (J arrard et al., 1994; Wilding et al., 1988). PC-3 cells show expression of TGF—Bl and TGF-Bl receptors and addition of TGF-B to PC-3 cells inhibits colony formation in soft agarose (Carruba et al., 1994; Jakowlew et al., 1997). PC-3 cells also express growth factors that are normally expressed by stromal cells. PC-3 cells produce measurable amounts of bFGF and large amounts of FGF-R mRNA (Nakamoto et al., 1992). Addition of recombinant bFGF to the culture media, does not enhance DNA synthesis in the PC-3 cell line (N akamoto et al., 1992). Low levels of another stromal-derived growth factor, IGF-l, was observed in conditioned media from PC-3 cells (Kimura et al., 1996). These cells express functional, high affinity 35 IGF-l receptors, but do not show a growth stimulatory response to exogenous IGF-l (Kimura et al., 1996). PC-3 cells may not show a response to exogenous IGF-I because they express almost twice the amount of IGF-II compared to DU145 and LNCaP cells, which may contribute to the autocrine action of IGF-I via the IGF-R (Kimura et al., 1996). DU145 cell line Irnmunofluorescent staining of DU145 cells shows intense staining for TGF-a, but to a lesser extent for EGF (Carruba et al., 1994). Intense staining for EGF—R was also shown in DU145 cells. DU145 cells show very little response to exogenous TGF-a or EGF (Carruba et al., 1994, MacDonald and Habib, 1992). DU145 cells do not show TGF-Bl expression, but they do express TGF-Bl receptors and their growth is inhibited in response to exogenous TGF-B (Carruba et al., 1994; Jakowlew et al., 1997). DU145 cells, similar to PC-3 cells, express stromal—derived growth factors. DU145 cells produce measurable amounts of bFGF and large amounts of FGF-R (Nakamoto et al., 1992). DU145 cells also show a growth stimulatory response upon addition of recombinant bFGF to the culture medium. These cells express functional, high affinity IGF-l receptors and produce low levels of IGF-I (Kimura et al., 1996). Addition of exogenous IGF-l at low concentrations stimulates the growth of DU145 cells (Lee et al., 2003). In this cell line the overproduced bFGF and IGF-1 activate their respective receptors in an autocrine loop. 36 LN CaP cell line LNCaP cells express EGF and TGF-a, as well as their receptors (Carruba et al., 1994). Both exogenous TGF-a and EGF stimulate the growth of LNCaP cells (MacDonald and Habib, 1992; Wilding et al., 1989). LNCaP cells express TGF-Bl receptor and show weak expression of TGF-B (Carruba et al., 1994; Jakowlew et al., 1997). The growth of LNCaP cells is inhibited upon addition of exogenous TGF-B (Jakowlew et al., 1997). LNCaP cells do not express bFGF, but they do show low levels of FGF mRNA and a growth stimulatory response to recombinant bFGF (N akarnoto et al., 1992). These cells express functional, high affinity IGF-l receptors and low levels of IGF-1 (Kimura et al., 1996). Addition of exogenous IGF-l stimulated the growth of LN CaP cells in a dose dependent manner (Lee et al., 2003). In this cell line the overproduced IGF-l activates the receptor in an autocrine loop. 37 Table 2.1 Expression of growth factors and their receptors in human prostatic carcinoma cell lines LN CaP DU 145 PC -3 Growth factor expression/secretion EGF ++ ++ + TGF-a ++ +++ + bFGF ND ++ + IGF-l + + + TGF-B (+) ND ++ Receptor expression EGF/TGF-u-R ++ +++ + FGF-R + +++ +++ IGF-l-R ++ ++ ++ TGF-B-R ++ ++ ++ (+), weakly positive; +, low measurable levels; +++, high levels; ND, not detectable (Modified from Webber et al., 1997). Table 2.2 Response to exogenous growth factors LN CaP DU145 PC-3 EGF + O O TGF-a + O O bFGF + + 0 IGF-l + + O TGF-B - - - +, stimulatory effect; -, inhibitory effect; 0, no response (Modified from Webber et al., 1997). 38 DU145 and PC-3 cells show autocrine regulation of several growth factors, including bFGF and IGF-1, which contributes to the malignant phenotype of these two cell lines. LNCaP cells only show autocrine regulation of IGF-1. This deregulation of growth in favor of proliferation not only aids in the survival of cells in various environments, but it has also been shown to decrease acinar forming ability in 3-D culture (Bello-DeOcampo et al., 2001b). Adhesion proteins, such as, integrins and cadherins are important for acinar morphogenesis and therefore cell organization and cell polarity. Adhesion properties: Polarized epithelial cells form the glandular compartment of the prostate and apically secrete their products into a lumen. The maintenance of cell polarity depends on cellzcell and cellzextracellular matrix (ECM) interactions. These interactions are also an important component of cell motility. Inte grins provide the link between cells and the ECM whereas cadherins are responsible for cellzcell adhesion. Invasive prostate cancer cells have decreased or abnormal expression of both integrins and cadherins, which assist in their dissemination from the prostate (Bello-DeOcampo et al., 2001b ; Achanazar et al., 2004). Invasive behavior of prostate cancer cells has been shown to be inversely correlated with the ability to undergo acinar morphogenesis (Bello-DeOcampo et al., 2001b). The non-tumorigenic, RWPE-l cell line forms acini at a high frequency (100%) in 3-D Matrigel cultures and the cells are not invasive using a Boyden chamber in vitro invasion assay. On the other hand, DU145 cells failed to form acini in (three- 39 dimensional) 3-D Matrigel cultures and were highly invasive in the Boyden chamber in vitro invasion assay (Bello—DeOcampo et al., 2001b). Normal acinar morphogenesis in RWPE-l cells was found to be associated with the expression of both (16 and [31 integrins at the basal and baso-lateral surfaces of the cells (Bello-DeOcampo et al., 2001b). The DU145, P03, and LNCaP cell lines were all found to express a relatively similar pattern of a6 and 131 integrins, as compared to each other. In prostate carcinoma the pattern of the (16 and Bl-integrin subunits, which in normal cells, are restricted to the basal and baso-lateral surfaces, are distributed diffusely throughout the cytoplasmic membrane (Knox et al., 1994). Indirect immunofluroescence microscopy shows PC-3, DU145, and LNCaP cells contain (16 in focal regions on the cell surface (W itkowski et al., 1993). DU145 and LNCaP cells also contain [31 in focal regions on the cell surface, however PC-3 cells show a diffuse cytoplasmic membrane pattern of B 1. The abnormal expression of a 6 and [31 integrins is likely to be responsible for the loss of ability to undergo acinar morphogenesis (Bello-DeOcampo et al., 2001b). In addition, this may assist in the invasion of these cells through the basement membrane. Other adhesion proteins, such as, cadherins also play an important role in the progression from a non- invasive to an invasive phenotype. Cadherins, in addition to integrins, play an important role in maintaining cellzcell adhesion, cell shape, cell polarization and acinar morphogenesis (Bussemakers et al., 1996; Webber et al., 1997, Bello-DeOcampo et al., 2001a, Bello-DeOcampo et al., 2001b). Altered expression of the protein, E-cadherin, has been observed in about 50% of prostate cancer cases (Achanazar et al., 2004; Hayward et al., 1998). In a study of E-cadherin expression in human tumors, expression of E-cadherin in metastatic deposits 40 was generally reduced or absent, but some metastatic deposits were found to have strong E-cadherin staining (Umbas et al., 1992). The LNCaP cell line was derived from a metastatic deposit but, based on results of Western blot analysis, these cells show strong expression of E-cadherin (Morton et al., 1993). The PC-3 and DU145 cell lines were also derived from a metastatic deposit, but they have less E-cadherin expression than the LNCaP cell line and cultured normal prostate epithelial cells (Figure 2.4 and Table 2.3). These data are consistent with results of immunocytochemistry using these three cell lines. The LNCaP cell line shows strong, positive, homogenous E-cadherin expression in 100% of cells, while E—cadherin staining of PC-3 and DU145 cells show heterogenous expression (Mitchell et al., 2000). DU LN m LN PC3 .- fifivst Figure 2.4 Western blot analysis of E-cadherin in prostate cells. LN =LNCaP cells; ml, normal prostate epithelial cells; PC3=PC-3 cells; DU=DU145 cells. For comparison purposes, LNCaP cells were analyzed at the same time as normal cells (left panel) and in a different analysis with the other two cell lines (right panel). Signals were quantitated by scanning densitometry of X-ray film. Exposure times were 2 min (left panel) and extended to 15 min (right panel) to increase sensitivity; 50 pg of total cellular protein were loaded in each lane, and probed with HECD-l monoclonal antibody. B-Galactosidase (116 kD) is the molecular weight marker for E-cadherin (124 kD) (Modified from Morton et al., 1993). 41 Table 2.3 Relative levels of E-cadherin in prostate cells* Cells E-cadherin Normal** 1.0 LNCaP 1.1 DU 145 0.1 PC-3 0.6 *Levels determined by scanning densitometry of autoradiographs of Western blots. **All values are relative to levels found in cultured normal prostate epithelial cells and represent the average values from 3 separate experiments (Morton et al., 1993). Besides E-cadherin, other classical cadherins include, P- and N -cadherin. PC-3 cells, in addition to E-cadherin, express N- and P-cadherin while DU145 cells express increased levels of P-cadherin (Bussemakers et al., 2000; Tran et al., 1999). The increased, abnormal expression of P- or N-cadherin, also referred to as cadherin switching, resulting in the loss of cadherin homeostasis, has been associated with the acquisition of an invasive phenotype (Achanazar et al., 2004). For example, in comparison to the non-tumorigenic human prostate epithelial cell line, RWPE-l, which mimics cadherin expression in normal cells, the related, yet invasive human prostate cancer cell lines, RWPE2-W99, WPEl-NB26 and CTPE, all express varying levels of P- and N-cadherins (Achanazar et al., 2004). RWPE2-W99 and WPEl—NB26 cells express lower than normal levels of P-cadherin, but increased levels of N-cadherin as compared to RWPE-l cells. While two of the three invasive cell lines show low P-cadherin and increased levels of N-cadherin, CTPE cells show increased levels of 42 P-cadherin and undetectable levels of N -cadherin. Cadherin switching and heterogeneity of cadherin expression observed in these cell lines mimics cadherin switching and heterogeneous cadherin expression observed in human prostate cancers (Achanazar et al., 2004). A loss or increased expression of one or more cadherins can disrupt homeostasis resulting in a change in cell adhesion, shape, polarization, and motility, and thus, may contribute to an invasive phenotype. Adhesive functions of cadherins also involve the normal expression of catenins. These proteins form a complex with the cytoplasmic portion of cadherin molecules and couple them to actin cytoskeletons of epithelial cells. Further studies are necessary to confirm the adhesive properties of cadherin expression and its involvement in the invasive behavior in these cell lines. The invasive potential of these cell lines can also be attributed to the expression of proteases such as matrix metalloproteinases (MMPs). Proteases: To metastasize prostate cancer cells must first degrade the ECM. While integrins are responsible for adhering cells to the ECM, proteases are responsible for degradation of the ECM. Type IV gelatinases, MMP-2 and MMP-9, are secreted as zymogens and upon activation they degrade type IV collagen in the basement membrane. The activity of MMPs is modulated by proenzyme activation and expression of their inhibitors, the tissue inhibitors of matrix metalloproteinases (TIMPs). Although both normal and neoplastic cells produce MMPs and other proteases, only malignant cells are invasive. In normal cells homeostasis is maintained between MMPs and their inhibitors. Analysis of 43 conditioned medium by gelatin zymography and enzyme assays show that both benign and neoplastic prostate tissues secrete latent MMP-9 and latent and active forms of MMP-2 (Lokeshwar et al., 1993). However conditioned medium from malignant prostate explants contained a higher proportion of the active form of MMP-2. Significant amounts of TlMP-l and TIMP-2 were secreted by adult prostate and benign prostate tissues, but TIMP-l levels were markedly reduced in conditioned medium from neoplastic tissues and TIMP-2 was absent (Lokeshwar et al., 1993). These data show that increasing levels of active MMP activity and decreasing levels of TIMPs are associated with prostate carcinoma. Another protease produced by cancer cells is urokinase-type plasminogen activator (uPA), it can activate plasminogen to plasmin which can degrade many ECM proteins and activate collagenases. uPA can also activate collagenases or directly degrade basement membrane components (Reith and Rucklidge, 1992; Webber and Waghray, 1995). Serum levels of u-PA in prostate cancer patients with metastasis are higher than those in healthy controls and in prostate cancer patients without metastasis (Miyake et al., 1999). Thus, the serine protease, u-PA, is considered to play a crucial role in the degradation of the extracellular matrix leading to tumor cell invasion and metastasis (W aghray and Webber, 1995; Webber and Waghray, 1995). Two proteins responsible for the negative regulation of uPA include, plasminogen activator inhibitor type-1 (PAH) and plasminogen activator inhibitor type-2 (PAI-2). While little is known about the expression of PAI-2, PAI-l is undetectable in cells of some aggressive malignancies (Lyon et al., 1995; Soff et al., 1995). A comparison of protease and 44 protease inhibitor expression in PC-3, DU 145, and LNCaP cell lines is shown in Table 2.4. Table 2.4 Protease and protease inhibitor expression in human prostate cancer cell lines LNCaP DU145 PC-3 Reference(s) MMP-2 0/+ 0 0 Webber, et al., 1996; MMP-9 0/+ 0 0 Dong et al., 2001 TIMP-l + ++ ++ Zhang etal., 2002 TIMP-2 + ++ ++- u-PA 0/+ +++ +++ Waghray and Webber, 1995; Hoosein et al., 1991 PAI-l 0 0 + Lyon et al., 1995 PAT-2 0 0 + 0, undetectable; +, trace; -H-, low; +++, high expression Trace levels of MMP-2 and MMP-9 were secreted by LNCaP cells, however gelatinase activity was undetectable in conditioned medium from the DU145 cell line (W ebber et al., 1996). Gelatinase activity was also undetectable in the PC-3 cell line (Dong et al., 2001). Although low or no expression of MMPs was observed in LNCaP, DU145 and PC-3 cells, mRNA expression of TIMP-1 is lower in all three cell lines compared to benign prostatic tissue (Zhang et al., 2002). LN CaP cells express lower levels of TIMP-2 mRNA compared to benign prostatic tissue, whereas PC-3 and DU145 cells express higher levels of TIMP-2. LNCaP cells expressed the least amount of 45 TIMP-1 and TIMP-2 mRNA compared to DU145 and PC-3 cells. Additional testing for active TIMP-1 and TIMP-2 expression should be performed on these three cell lines. DU145 and PC-3 cells also secrete u-PA, which could contribute to their invasive potential. DU145 cells express five times more extracellular, secreted u-PA activity than the tested normal prostatic epithelial cells (W aghray and Webber, 1995). Three different assays performed on the LNCaP, PC-3, and DU145 cell lines, show very low levels of urokinase secretion by LNCaP cells and higher urokinase levels in PC-3 compared to DU145 cells (Hoosein et al., 1991). The production of plasminogen activator inhibitors was very low or undetectable in these three cell lines. Only a small amount of PAH and PAI-2 protein was detectable in the PC-3 cell line (Lyon et al., 1995). To compare the expression of proteases with the invasive potential of a cell line in vitro, a reconstituted basement membrane is used to examine invasive behavior. Invasion in vitro: The invasive potential of DU145 and PC-3 cells was determined by their ability to invade in vitro in a Matrigel invasion assay. Matrigel serves as a reconstituted basement membrane. Invasion by cancer cells through the basement membrane in vivo is one of the first steps in the progression to metastasis (W ebber et al., 1995). Normal prostate cells are attached to the basement membrane and are not invasive. The invasive ability of DU145 and LNCaP cells was determined using the in vitro Boyden chamber assay (Bello, 1996). The invasive ability of PC-3 cells was also determined using the in vitro Boyden chamber assay, although the technique varied from the assay with DU145 and LNCaP cells (Fong et al., 1992). The invasion assay performed by (Bello, 1996) used 46 filters with a smaller pore size, a different chemoattractant and cells were allowed to invade for twenty four hours instead of six hours. (Bello, 1996) also used a different technique to quantify invasive cells, which involved staining the nuclei of cells on the bottom of the filter and extracting dye in addition to counting viable cells in the chamber well. In comparison to DU145 cells, LNCaP cells showed 19% invasion (Figure 2.5) (Bello, 1996). In comparison to DU145 cells, PC-3 cells were about twice as invasive (Fong et al., 1992). So the PC-3 cell line is the most invasive cell line in vitro followed by the DU145 cell line and then the LN CaP cell line. The invasive ability observed in vitro in these cell lines is also observed in vi vo. Nude mice injected subcutaneously with PC-3 cells often develop metastatic tumors or show invasion into surrounding tissues, but metastatic tumors are rare in mice injected subcutaneously with LNCaP cells (Kozlowski etal., 1984; Sato et al., 1997; Rembrink et al., 1997). In a subcutaneous xenograft model of DU145, no macroscopic metastases were visible, however intra-splenic injection of DU145 cells, but not LNCaP cells, did result in the formation of metastasic tumors (Devi et al., 2002; Pettaway et al., 1996; Kozlowski et al., 1984). 47 LNCaP ‘ DU145 0% 20% 40% 60% 80% 100% Invasion expressed as percent of DU145 Figure 2.5 In vitro invasion of DU] 45 and LNCaP cells. Invasive ability of DU145 and LNCaP cell lines was examined by the Boyden chamber in vitro invasion assay. 400,000 cells were plated on each “Matrigel”-coated filter and allowed to invade for 24 h. The invasive ability of the highly invasive DU145 cell line was set at 100% invasion (Modified from Bello, 1996). Conclusions Derivation of the PC-3, DU145, and LNCaP cell lines from metastatic deposits demonstrates their metastatic potential. Both PC-3 and DU 145 cell lines are androgen- insensitive and show autocrine regulation of growth. PC-3 and DU145 cells show varying levels of cadherin expression and PC-3 cells show abnormal (1601 integrin expression. The PC-3 cell line expresses higher urokinase levels and is about twice as invasive as DU145 cells in the in vitro Boyden chamber assay. These cell lines would be useful for studies on androgen insensitive cell growth for developing treatment strategies for prostate cancer patients in which the cancer cells do not respond to hormone therapy. 48 The LNCaP cell line is androgen-sensitive and also shows autocrine regulation of growth. In comparison to PC-3 and DU145 cells, LNCaP cells show strong expression of E-cadherin and low uPA expression. The LNCaP cell line, in vitro, is the least invasive of the three cell lines. Although LNCaP cells are androgen-sensitive, the cells also carry a mutated androgen receptor, which is only observed in a minority of prostate cancer patients. Therefore, the results of some studies with this cell line may not be applicable to all of prostate cancers as many prostate cancer patients do not possess a mutated androgen receptor. The same AR gene mutation described in the LN CaP cell line has been observed in 6 of 24 (25%) prostate tissue specimens derived from patients with advanced prostate cancer (Gaddipati et al., 1994). The LNCaP cell line would be particularly useful for developing treatment strategies for prostate cancer patients in which the cancer cells have a mutated androgen receptor. To better understand prostate cancer and to develop treatments for the majority of prostate cancer patients, a prostate cancer cell line that is androgen-sensitive and has normal androgen receptor expression would be useful. The WPEl-NB26 cell line, deve10ped in our laboratory (W ebber et al., 2001) is such a cell line. It is for this reason that I used WPEl-NB26 cells for developing a xenograft model of prostate cancer. 49 Literature cited Achanzar, W.E., Lamar, P.C., Tokar, E.J., Rivette, A.S., Bello-DeOcampo, D., Prozialeck, W.C., Webber, M.M., and Walkes, M.P.: Human prostate cell lines mimic heterogeneity of cadherin expression in human prostate cancer. UroOncology 4: 15-25, 2004. Bello, D.: In vitro invasion. In: Characterization of immortalized and malignant adult human prostate epithelial cell lines: PWR-lE, RWPE-l, and RWPE-2. Master of Science Thesis, Michigan State University, East Lansing, MI, USA, pp.69-78, 1996. . Bello-DeOcampo, D., Kleinman, H.K., DeOcampo, N .D., and Webber, M.M.: Laminin 1 and (16131 integrin regulate acinar morphogenesis of normal and malignant human ‘ prostate epithelial cells. Prostate 46: 142-153, 2001a. Bello-DeOcampo, D., Kleinman, H.K., and Webber, M.M.: The role of 0.6131 integrin and EGF in normal and malignant acinar morphogenesis of human prostatic epithelial cells. Mutation Research 480-481:209-2l7, 2001b. Bems, E.M.J.J., de Boer, W., and Mulder, E.: Androgen-dependent growth regulation of and release of specific protein(s) by the androgen receptor containing human prostate tumor cell line LN CaP. Prostate 9:247-259, 1986. Brolin, J ., Skoog, L., and Ekrnan, P.: Immunohistochemistry and biochemistry in detection of androgen, progesterone, and estrogen receptors in benign and malignant human prostatic tissue. Prostate 20:281-295, 1992. Bussemakers, M.J.G., and Schalken, J.A.: The role of cell adhesion molecules and proteases in tumor invasion and metastasis. World Journal of Urology 14: 151 -156, 1996. Bussemakers, M.J.G., Van Bokhoven, A., Tomita, K., Jansen, C.F.J., and Schalken, J .A.: Complex cadherin expression in human prostate cancer cells. International Journal of Cancer 85:446-450, 2000. Carruba, G., Leake, R.E., Rinaldi, F., Chalmers, D., Comito, L., Sorci, C., Pavone Macaluso, M., and Castagnetta, L.A.M.: Steroid-growth factor interaction in human prostate cancer. 1. Short-term effects of transforming growth factors on growth of human prostate cancer cell lines. Steroids 59:412-420, 1994. Connolly, J.M., and Rose, D.P.: Production of epidermal growth factor and transforming growth factor-0. by the androgen-responsive LNCaP human prostate cancer cell line. Prostate 16:209-218, 1990. 50 Culig, Z., Hobisch, A., Cronauer, M.V., Radmayr, C., Trapman, J., Hittmair, A., Bartsch, B., and Klocker, H.: Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. CancerResearch 54:5474-5478, 1994. Devi, G.R., Oldenkamp, J.R., London, CA, and Iverson, P.L.: Inhibition of human chorionic gonadotropin B-subunit modulates the mitogenic effect of c-myc in human prostate cancer cells. Prostate 52:200-210, 2002. Dong, Z., Nemeth, J., Cher, M.L., Palmer, K.C., Bright, RC, and Fridman, R.: Differential regulation of matrix metalloproteinase-9, tissue inhibitor of metalloproteinase-1 (TIMPl) and TIMP-2 expression in co-cultures of prostate cancer and stromal cells. International Journal of Cancer 93:507-515, 2001. Pong, C., Sutkowski, D.M., Kozlowski, J .M., and Lee, C.: Utilization 0f the boyden chamber to further characterize in vitro migration and invasion of benign and malignant human prostatic epithelial cells. Invasion Metastasis 12:264-274, 1992. Gaddipati, J .P., McLeod, D.G., Heidenberg, H.B., Sesterhenn, I.A., Finger, M.J., Moul, J .W., and Srivastava, S.: Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancers. Cancer Research 54:2861 2864, 1994. Garde, S.V., Sheth, A.P., Porter, AT, and Pienta, K.J.: A comparative study on expression of prostatic inhibin peptide, prostate acid phosphatase and prostate specific antigen in androgen independent human and rat prostate carcinoma cell lines. Cancer Letters 70:159-166, 1993. Gil Diez de Medina, S., Salomon, L., Colombel, M., Abbou, C.C., Bellot, J., Thiery, J .P., Radvanyi, F., Van der Kwast, TH, and Chopin, D.K.: Modulation of cytokeratin subtype, EGF receptor, and androgen receptor expression during progression of prostate cancer. Human Pathology 29: 1005-1012, 1998. Hayward, S.W., Grossfeld, G.D., Tlsty, TD, and Cunha, 6.: Genetic and epigenetic influences in prostate carcinogenesis (Review). International Journal of Oncology 13:35-47, 1998. Henttu, P., and Vihko, R.: Growth factor regulation of gene expression in the human prostatic carcinoma cell line LNCaP. Cancer Research 53:1051-1058, 1993. Hoosein, N.M., Boyd, D.D., Hollas, W.J., Mazar, A., Henkin, J., and Chung. L.W.K.: Involvement of urokinase and its receptor in the invasiveness of human prostatic carcinoma cell lines. Cancer Communications 3:255-264, 1991. 51 Horoszewicz, J .S., Leong, S.S., Chu, T.M., Wajsman, Z.L., Friedman, M., Papsidero, L., Kim, U., Chai, L.S., Kakati, S., Arya, SK, and Sandberg, A.A.: The LNCaP cell line-A new model for studies on human prostatic carcinoma. In: Progress in Clinical and Biological Research, Models for Prostate Cancer, Murphy, G.P. (ed.), Buffalo, NY, Alan R. Liss Inc., 37:67-84, 1980. Hsieh, J ., Wu, H., Gleave, M.E., von Eschenbach, AC, and Chung, L.W.K.: Autocrine regulation of prostate-specific antigen gene expression in a human prostatic cancer (LNCaP) subline. Cancer Research 53:2852-2857, 1993. J arrard, K.F., Blitz, B.F., Smith, R.C., Patai, BL, and Rukstalis, D.B.: Effect of epidermal growth factor on prostate cancer cell line PC3 growth and invasion. Prostate 24:46-53, 1994. J akowlew, S.B., Moody, T.W., and Mariano, J .M.: Transforming growth factor-beta receptors in human cancer cell lines: analysis of transcript, protein, and proliferation. Anticancer Research 17:1849-1860, 1997. J iang, J., Wang, L.F., Fang, Y.H., Jin, ES, and J in, W.S.: Proliferative response of human prostate cancer cell to hormone inhibited by androgen receptor antisense RNA. Chinese Medical Journal 117:684-688, 2004. Kaighn, M.E., Lechner, J.F., Narayan, K.S., and Jones, L.W.: Prostate carcinoma: Tissue culture cell lines. National Cancer Institute Monograph 49:17-21, 1978. Kaighn, M.E., Narayan, S., Ohnuki, Y., Lechner, J .F., and Jones, L.W.: Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Investigative Urology 17:16-23, 1979. Kimura, G., Kasuya, J ., Giannini, S., Honda, Y., Mohan, S., Kawachi, M.H., Akimoto, M., and Fujita-Yamaguchi, Y.: Insulin-like growth factor (IGF) system components in human prostatic cancer cell-lines: LNCaP, DU145, and PC-3 cells. International Journal of Urology 3:39-46, 1996. Knox, J .D., Cress, A.E., Clark, V., Manriquez, L., Affinito, K.S., Dalkin, BL, and Nagle, R.B.: Differential expression of extracellular matrix molecules and the alpha 6-integrins in the normal and neoplastic prostate. American Journal of Pathology 145:167-174, 1994. Kozlowski, J .M., Fidler, I.J., Campbell, D., Xu, 2., Kaighn, M.E., and Hart, I.R.: Metastatic behavior of human tumor cell lines grown in the nude mouse. Cancer Research 44:3522-3529, 1984. 52 Lee, H., Pienta, K.J., Kim, W., and Cooper, C.R.: The effect of bone-associated growth factors and cytokines on the growth of prostate cancer cells derived from soft tissue versus bone metastases in vitro. International Journal of Oncology 222921 -926, 2003. Lokeshwar, B.L., Selzer, M.G., Block, N.L., and Gunja-Smith, Z.: Secretion of matrix metalloproteinases and their inhibitors (tissue inhibitor of metalloproteinases) by human prostate in explant cultures: reduced tissue inhibitor of metalloproteinase secretion by malignant tissues. Cancer Research 53:4493-4498, 1993. Lyon, P.B., See, W.A., Xu, Y., and Cohen, M.B.: Diversity and modulation of plasminogen activator activity in human prostate carcinoma cell lines. Prostate 27:179-186, 1995. MacDonald, A., and Habib, F .K.: Divergent responses to epidermal growth factor in hormone sensitive and insensitive human prostate cancer cell lines. British Journal of Cancer 65: 177-182, 1992. Mickey, D.D., Stone, K.R., Wunderli, H., Mickey, G.H., and Paulson, D.F.: Characterization of a human prostate adenocarcinoma cell line (DU145) as a monolayer culture and as a solid tumor in athymic mice. In: Progress in Clinical and Biological Research, Models for Prostate Cancer, Murphy, G.P. (ed.), Buffalo, NY, Alan R. Liss Inc., 37:67-84, 1980. Mickey, D.D., Stone, K.R., Wunderli, H, Mickey, G.H., Vollmer, RT, and Paulson, D.F.: Heterotransplantation of a human prostatic adenocarcinoma cell line in nude mice. Cancer Research 37:4049-4058, 1977. Mitchell, S., Abel, P., Ware, M., Stamp, G., and Lalani, E.N.: Phenotypic and genotypic characterization of commonly used human prostatic lines. British Journal of Urology International 85:932-944, 2000. Miyake, H., Hara, 1., Yamanaka, K., Arakawa, S., and Karnidono, S.: Elevation of urokinase type plasminogen activator and its receptor densities as new predictors of disease progression and prognosis in men with prostate cancer. International Journal of Oncology 14:535-541, 1999. Morton, R.A., Ewing, C.M., Nagafuchi, A., Tsukita, Shoichiro, and Isaacs, W.B.: Reduction of e-cadherin levels and deletion of the a-catenin gene in human prostate cancer cells. Cancer Research 53:3585-3590, 1993. Nakamoto, T., Chang, C., Li, A., and Chodak, G.W.: Basic fibroblast growth factor in human prostate cancer cells. Cancer Research 52:571-577, 1992. 53 Paulson, D.F., Stone, K.R., Mickey, D.D., Bonar, R.A., and Wunderli, H.: Development and application of basic research techniques in bladder cancer research. Cancer Research 37:2969-2973, 1977. Pettaway, C.A., Pathak, S., Greene, G., Ramirez, E., Wilson, M.R., Killion, J.J., and Fidler, I.J.: Selection of highly metastatic variants of different human prostatic carcinomas using orthotopic implantation in nude mice. Clinical Cancer Research 221627-1636, 1996. Reith, A., and Rucklidge, G.J.: Invasion of brain tissue by primary glioma: evidence for the involvement of urokinase-type plasminogen activator as an activator of type IV collagenase. Biochemistry and Biophysical Research Communications 186:348-354, 1992. Rembrink, K., Romijn, J .C., van der Kwast, T.H., Rubben, H., and Schroder, F.H.: Orthotopic implantation of human prostate cancer cell lines: a clinically relevant animal model for metastatic prostate cancer. Prostate 31:168-174, 1997. Sato, N., Gleave, M.E., Bruchovsky, N ., Rennie, P.S., Beraldi, E., and Sullivan, L.D.: A metastatic and androgen-sensitive human prostate cancer model using intraprostatic inoculation of LNCaP cells in SCID mice. Cancer Research 57:1584-1589, 1997. Schuurmans, A.L.G., Bolt, J., Veldscholte, J ., and Mulder, E.: Regulation of growth of LNCaP human prostate tumor cells by growth factors and steroid hormones. Journal of Steroid Biochemical and Molecular Biology 40: 193-197, 1991. Soff, G.A., Sanderowitz, J ., Gately, S., Verrusio, E., Weiss, 1., Brem, S., and Kwaan, H.C.: Expression of plasminogen activator inhibitor type 1 by human prostate carcinoma cells inhibits primary tumor growth, tumor-associated angiogenesis, and metastasis to lung and liver in an athymic mouse model. Journal of Clinical Investigations 96:2593-2600, 1995. Tran, N.L., Nagle, R.B., Cress, A.E., and Heimark, R.L.: N-cadherin expression in human prostate carcinoma cell lines. American Journal of Pathology 155:787 -798, 1999. Umbas, R., Schalken, I.A., Aalders, T.W., Carter, B.S., Karthaus, H.F.M., Schaafsma, H.B., Debruyne, F.M.J., and Isaacs, W.B.: Expression of the cellular adhesion molecule, E-cadherin, is reduced or absent in high grade prostate cancer. Cancer Research 52:5104-5109, 1992. Waghray, A., and Webber, M.M.: Retinoic acid modulates extracellular urokinase-type plasminogen activator activity in DU-145 human prostatic carcinoma cells. Clinical Cancer Research 1:747-753, 1995. 54 Webber, M.M., Bello, D., and Quader, S.: Irnmortalized and tumorigenic adult human prostatic epithelial cell lines: Characteristics and applications part 2. Tumorigenic cell lines. Prostate 30:58-64, 1997. Webber, M.M., Quader, S.T.A., Kleinman, H.K., Bello-DeOcampo, D., Storto, P.D., Bice, G., DeMednonca-Calaca, W., and Williams, D.E.: Human cell lines as and in vitro/in vivo model for prostate carcinogenesis and progression. Prostate 47:1 -l3, 2001. Webber, M.M., and Waghray, A.: Urokinase-mediated extracellular matrix degradation by human prostatic carcinoma cells and its inhibition by retinoic acid. Clinical Cancer Research 1:755-761, 1995. Webber, M.M., Waghray, A., and Bello, D.: Prostate specific antigen (PSA), a serine protease, facilitates human prostate cancer invasion. Clinical Cancer Research 1:1089-1094, 1995. Webber, M.M., Waghray, A., Bello, D., and Rhim, J .S.: Mini review: proteases and invasion in human prostate epithelial cell lines: implications in prostate cancer prevention and intervention. Radiation Oncology Investigations 3:358-362, 1996. Wilding, G., Valverius, E., Knabbe, C., and Gelmann, E.P.: Role of transforming growth factor-a in human prostate cancer cell growth. Prostate 15:1-12, 1989. Wilding, G., Zugmeier, G., Knabbe, C., Valverius, E., Flanders, K., and Gelmann, E.P.: The role of transforming growth factors a and B in human prostate cancer cell growth. Endocrinology 29:241, 1988. Witkowski, C.M., Rabinovitz, I., N agle, R.B., Affinito, K.D., and Cress, A.E.: Characterization of integrin subunits, cellular adhesion and tumorigenicity of four human prostate cell lines. Journal of Cancer Research and Clinical Oncology 119:637-644, 1993. van Bokhoven, A., Varella-Garcia, M., Korch, C., Johannes, W.U., Smith, E.E., Miller, H.L., Nordeen, S.K., Miller, G.J., and Lucia, M.S.: Molecular characterization of human prostate carcinoma cell lines. Prostate 57:205-225, 2003. Zhang, J., Jung, K., Lein, M., Kristiansen, G., Rudolph, B., Hauptmann, S., Schnorr, D., Loening, SA, and Lichtinghagen, R.: Differential expression of matrix metalloproteinases and their tissue inhibitors in human primary cultured prostatic cells and malignant prostate cell lines. Prostate 50:38-45, 2002. 55 CHAPTER THREE RWPE-l CELL LINE AND ITS DERIVATIVES: RWPE2-W99, CTPE, AND THE MNU FAMILY OF CELL LINES 56 Abstract Most studies in prostate cancer research have been conducted using cell lines derived from metastatic deposits. To assist in the treatment and prevention of prostate cancer, cell lines that represent earlier stages of prostate cancer are needed. A cell line that mimics normal prostate cell behavior is also necessary for comparison in studies using malignant prostate cell lines. Therefore, several cell lines, all derived from the same parental, RWPE-l cell line, that represent multiple steps in prostate carcinogenesis and progression, have been generated and extensively characterized. These cell lines all sliow unique characteristics to study prostate cancer. The parent, RWPE-l cells, behave much like normal prostate cells while the derived cell lines all show varying degrees of malignant cell behavior. Several characteristics of abnormal cell behavior have been observed during prostate carcinogenesis and tumor progression. Such characteristics include changes in: cell morphology, response to growth factors, as well as, expression of cytoskeletal proteins, adhesion molecules, and proteases. The objective of this chapter is to describe the development and derivation of a family of cell lines, and their characteristics. Further, their applications in studies of prostate cancer prevention and treatment and the development of xenograft models are explored. Keywords Cell line, CTPE, prostate cancer, RWPE-l, RWPE2-W99, WPEl-NA22, WPEl-NBl 1, WPEl 1-NB14, and WPE 1 -NB26 57 Introduction To assist in the prevention of tumor progression and metastatic spread, which is the main cause of death in prostate cancer patients, human cell lines that represent early events in carcinogenesis and tumor progression would be useful. Cell lines that mimic behavior of normal human prostate cells are also necessary to serve as controls in vitro or as standardized models in vivo when studying prostate cancer progression. However, normal human epithelial cells require immortalization to provide a practical system for in vitro studies (Bello et al., 1997; Webber et al., 1996a). Irnmortalization is an important step in the process of carcinogenesis. Irnmortalized cells can be used to study carcinogenesis as well as normal prostatic epithelial cell functions, differentiation, and modulation by growth factors, hormones, and other agents (W ebber et al., 1996b). The use of normal, human, immortalized prostate cells along with other human prostate cell lines in vitro or in vivo allows investigators to obtain comparable data for solving problems in prostate carcinogenesis and metastasis. The three most commonly studied cell lines; PC-3, DU145, and LNCaP, were all derived from metastatic deposits in prostate cancer patients (Kaighn et al., 1979; Mickey et al., 1980; Horoszewicz et al., 1980). They are, therefore, more appropriate for in vitro and in vivo studies on advanced prostate cancer. In this review characteristics of a normal prostatic epithelial cell line, RWPE-l, and two RWPE-l-derived transformants, RWPE2—W99 and CT PE, as well as, a family of cell lines transformed by N-methyl-N-nitrosourea (MNU), will be described. Potential applications and advantages of these cells lines as appropriate models of prostate cancer will also be discussed. 58 Source of RWPE-l, RWPE2-W99, MNU, and the CTPE cell line: RWPE-l cells were isolated from the prostate of a 54 year-old Caucasian man undergoing radical prostatectomy (Bello et al., 1997). These cells were immortalized with the human papilloma virus-18 (HPV-18) genome using a plasmid vector and Polybrene-induced DNA transfection followed by shock with dimethyl sulfoxide (Rhim et al., 1994). HPV-18 was used for immortalization because cells are more likely to retain growth and differentiation characteristics of their normal cells of origin and they are non-tumorigenic ,(Bello et al., 1997; Woodworth et al., 1990; Yankaskas et al., 1993). HPVs are also the most common sexually transmitted disease, and are implicated in the etiology of several cancers (Chen et al., 1993; Webber et al., 1997). The RWPE-l cell line was further transformed by the Ki-ras oncogene to obtain the RWPE-2 cell line (Bello et al., 1997; Webber et al., 1997). To transform RWPE—l cells, the Kirsten murine sarcoma virus (Ki-MuSV) containing an activated Ki-ras oncogene and a helper virus, baboon endogenous virus, were used (Rhim et al., 1994). RWPE2-W99 cells were derived from a colony of RWPE-2 cells growing in agar that was selected for high Ki-ras expression. Both, the presence of ras mutations and high-risk HPV DNA sequences, have been linked at a relatively high frequency in Japanese men with prostate cancer (Anwar et al., 1992). A family of cell lines, the MNU cell lines, was also generated from the RWPE-l cells (Figure 3.1) (W ebber et al., 2001). RWPE-l cells were treated with MNU, a chemical carcinogen, at 50 or 100 pg/ml. Carcinogen-exposed cells were injected subcutaneously in nude mice and tumors were removed 10 weeks after cell injection. Cells from these tumors were grown in culture to give rise to 2A (50 pg/ml MNU) and 59 3B (100 pg/ml MNU) cells. These cells were again injected subcutaneously into nude mice and tumors were collected and plated in culture to expand the cell population. Cells were then plated in soft agar. Individual colonies were isolated and expanded and gave rise to the MNU cell lines which all share a common lineage; These cell lines include: WPEl-NA22, WPEl-NB14, WPEl-NBl l, and WPEl-NB26 (Figure 3.1). 60 Human prostatic epithelial cells Immortalized with HPV-18 RWPE-l cell line Treated with MNU Transformed cells Injected into nude mice First generation tumors 2 3B Cells / Grown in culture and injected into nude mice 2A2 3B1 3B2 Second generation tumors 1 l \ \Plated in agar. colonies isolated WPEl-NA22 WPEl-NB14 WPEl-NBll WPEl-NB26 \ J V -The following MNU cell lines were established: WPEl-NA22, WPEl-NB14, WPEl-NBll, WPEl-NB26 -Cells were injected into nude mice and third generation tumors were obtained Figure 3.1 Derivation of the MN U-transforrned cell lines from RWPE-l, a HPV-18 immortalized human prostatic epithelial cell line. The 2A tumor was derived from treatment with MNU at 50 pg/ml and 3B tumor at 100 pg/ml (Webber etal., 2001). RWPE-l cells were also transformed by in vitro cadmium exposure for 8 weeks to obtain the CTPE cell line (Achanazar et al., 2001). Cadmium, a known human carcinogen, has been implicated in prostate cancer etiology and cancer progression (W aalkes et al., 1997). Cadmium is also an effective carcinogen in rats, and the rat prostate was found to be a target for cadmium carcinogenesis. This suggests that 61 occupational or environmental cadmium exposure is a risk factor for the development of prostate malignancies. So, CTPE, RWPE2-W99, and the MNU family of cell lines, each have many applications in the study of prostate cancer, such as, identifying molecular changes associated with cadmium, Ki-ras or MN U. To confirm epithelial origin of these cell lines, cell morphology and the expression of cytokeratins were examined. Cell morphology: RWPE-l and RWPE2-W99 cells have a typical, polygonal, epithelial morphology (Figures 3.2a and 3.3a) (Bello et al., 1997). The morphology of the MNU cell line, WPEl-NA22, closely resembles that of RWPE-l cells (Figure 3.4a and b) (W ebber etal., 2001). However, the cell morphology of another MNU cell line, WPEl-NB26, was more elongated compared to RWPE-l cells (Figure 3.4e). The morphology of the other two MNU cell lines, WPEl-NB14 and WPEl-NBI 1, was in between polygonal and elongated (Figure 3.4c and d). Epithelial origin: Cytokeratins are cytoskeletal proteins that are important markers expressed by epithelial cells. Secretory luminal epithelial cells of the prostate express cytokeratins 8 and 18,while basal epithelial cells of the prostate express cytokeratins 5 and 14. Both RWPE-l and RWPE-2 cells express cytokeratins 8 and 18 (Figures 3.2e and f and 3.3e and f), which establishes their epithelial origin (Bello et al., 1997). All of the MNU cell lines show expression of cytokeratins 8 and 18 confirming their epithelial origin (W ebber 62 et al., 2001). CTPE cells also express cytokeratin 18 (W ebber, M.M., personal communication). Several additional markers, such as, the expression of androgen receptor (AR) and prostate specific antigen (PSA) are also necessary to confirm prostatic epithelial origin of the cell lines. Response to androgen and the expression of androgen receptor and PSA: Prostatic epithelial cells express androgen receptor and in response to androgen exposure, they express PSA. Both RWPE-l and RWPE-2 cell lines show a growth response to a synthetic androgen, mibolerone, and express PSA and AR, as assessed by immunocytochernistry (Figures 3.2 and 3.3, b-d) (Bello etal., 1997). CTPE cells also express PSA in response to androgen exposure (W ebber, M.M., personal communication). The cell lines of the MNU family show a growth stimulatory response to mibolerone and express PSA and AR, however, the expression pattern of AR varies amongst the cell lines (Figure 3.4f-h) (Webber et al., 2001). WPEl-NBll and WPEl-NB26 cells were found to show homogeneous, strong nuclear staining while the WPEl-NA22 and WPEl-NB 14 cells were found to show heterogeneous nuclear staining. The ability of RWPE-l, RWPE2-W99, CTPE and the MNU farme of cell lines to respond to androgens, confirms their prostatic origin. Cell response to growth factors is an important aspect of cell behavior. 63 .0 a, fi- . I. it... i 0‘ ’1 . ' 031‘“ ~ 1 . . f ‘1 r a —.‘ -..aa- b - e- C o I C d D‘&\‘g I.“ '. r. Figure 3.2 Characterization of RWPE-l cells. Proteins were detected by immunoperoxidase staining. (a) hematoxylin and eosin staining; (b) positive staining for PSA; (0) positive staining for nuclear androgen receptor. Cells for (b) and (c) were pretreated with 5nM mibolerone; (d) a control lacking primary antibody; (e) and (0 positive staining for cytokeratin 8 and 18, respectively. Scale bar is 20 pM. X 625 (Modified from Bello et al., 1997). 64 Figure 3.3 Characterization of RWPE-2 cells. Proteins were detected by immunoperoxidase staining. (3) hematoxylin and eosin staining; (b) positive staining for PSA; (c) positive staining for nuclear androgen receptor. Cells for (b) and (c) were pretreated with 5nM mibolerone; (d) a control lacking primary antibody; (e) and (0 positive staining for cytokeratin 8 and 18, respectively. Scale bar is 20 pM. X 625 (Modified from Bello et al., 1997). 65 Figure 3.4 Characterization of MN U cell lines. Morphology of (hematoxylin and eosin stain): (a) RWPE-l, (b) WPEl-NA22, (c) WPEl-NB14, (d) WPEl-NBl 1, (e) WPEl-N B26 cells. F-h: PSA and androgen receptor expression in WPEl-NA22 cells treated with mibolerone, as detected by irnmunostaining; f, positive staining for PSA; g, positive staining for nuclear androgen receptor; and h, a control lacking primary antibody. Bar = 20 pM (Webber et al., 2001). Production and response to growth factors: Malignant cells tend to secrete and respond to their own growth factors in an autocrine manner, to assist in their growth and survival in various microenvironments. For this reason, when malignant cells are treated with exogenous growth factors in vitro, they may not show a marked response. Epidermal growth factor (EGF) stimulates epithelial cell growth while transforming growth factor—B (TGF—B) inhibits epithelial cell growth, however, some prostate cancer cells may lose the ability to be inhibited by TGF—B or show a decrease in growth inhibitory response. 66 RWPE-l cells show a response, similar to normal cells, to EGF and TGF-B (Bello et al., 1997). The MNU family of cell lines show a growth stimulatory response to exogenous EGF and an inhibitory response to TGF-B, however, the response varied amongst the four cell lines (W ebber et al., 2001). The WPEl-NA22 and WPEl-NB14 cells were more responsive to the growth stimulatory effects of EGF than the WPEl-NBll and WPEl-NB26 cells. WPEl-NB26 cells were also the least responsive to the inhibitory effects of TGF—B, with WPEl-NBll being the most responsive to TGF-B. Cell response to growth factors has not yet been published for the CTPE cell line. These data suggest that the RWPE-l cell line responds to growth factors similar to normal cells, with the other cell lines being less responsive, and showing varying degrees of responsiveness. The WPEl-NB26 cell line shows the least amount of responsiveness to exogenous growth factors, and thus, is considered to represent the most malignant cell line of the MNU family in terms of growth factor response. Deregulation of growth in favor of proliferation has been correlated with loss of cell organization (Bello-DeOcampo et al., 2001b). Loss of cell organization has also been correlated with abnormal integrin and cadherin expression. Adhesion properties: In glandular tissues, such as the prostate, cellzmatrix and cellzcell adhesions mediated by integrins and cadherins respectively, result in cell polarization and organization into acini. Invasive behavior of prostate cancer cells has been shown to be inversely correlated with the ability to undergo acinar morphogenesis (Bello-DeOcampo et al., 2001b). RWPE-l cells, like normal cells, organize into acini with distinct lumens 67 lined by a single layer of cells in 3-D Matri gel culture (Figure 3.5a) (Bello-DeOcampo et al., 2001b). However, the tumorigenic, RWPE-2 cells remain as single cells or form small aggregates (W ebber et al., 1997). The ability of the RWPE-l cell line to polarize and form acini is consistent with their normal growth regulation and strong, basal expression of the (1681 integrin (Bello-DeOcampo et al., 2001b). In 3-D Matrigel culture, the malignant, WPEl-NB26 cells form solid masses of disorganized cells (Figure 3.5b) (Bello-DeOcampo et al., 2001b). In the WPEl-NB26 cell line, [3; integrin expression is strong and positive, yet, diffuse. Another explanation for the loss of cell organization is the lack of or, integrin staining observed in cell masses of WPEl-NB26 (Bello-DeOcampo et al., 2001b). These results demonstrate that while ()6 integrin expression decreases or is lost in WPEl-NB26 cells, [31 expression is altered so that its expression is no longer confined basally, but is diffusely expressed throughout the cell surface. The abnormal expression of (1681, correlates with the loss of ability of the RWPE-2 and WPEl-NB26 cell lines to undergo cell organization and acinar formation. CTPE cells in 3-D Matrigel culture show some polarization compared to RWPE2-W99 and WPEl-NB26 cells, which do not show any signs of cell organization in 3-D culture, but the expression of (16131 is not known in the CTPE cell line (Achanazar et al., 2004). 68 Figure 3.5 Acinar morphogenesis by RWPE—l and WPEl-NB26 cells in 3-D Matrigel culture. (a) the non-tumorigenic RWPE—l cells form well organized acini of polarized cells around a central lumen, while WPEl-NB26 cells; (b) form a disorganized cell mass, Bar = 25 pm (Modified from Achanzar et al., 2004). Cadherins, in addition to integrins, play an important role in maintaining cellzcell adhesion, cell shape, cell polarization and acinar morphogenesis (Bussemakers et al., 1996; Webber et al., 1997, Bello-DeOcampo eta1., 2001a, Bello-DeOcampo et al., 2001b). Altered expression of the protein, E-cadherin, has been observed in about 50% of prostate cancer cases (Achanazar et al., 2004; Hayward et al., 1998). Non- tumorigenic, RWPE-l cells, and the more malignant, WPEl-NB26 cells, exhibit strong and uniform plasma membrane localization of E-cadherin using immuno-fluorescence, but the staining was weak in RWPE2-W99 cells and heterogeneous in CTPE cells (Achanzar et al., 2004). These results show that only two of the three malignant human prostate cell lines tested, RWPE2—W99 and CTPE, express decreased, variable levels E-cadherin and its expression is abnormally localized. 69 Besides E-cadherin, other classical cadherins include, P- and N-cadherin. The increased, abnormal expression of P- or N-cadherin, also referred to as cadherin switching, resulting in the loss of cadherin homeostasis, has been associated with the acquisition of an invasive phenotype (Achanazar et al., 2004). In comparison to RWPE-l cells, which mimic cadherin expression observed in normal human prostate epithelial cells, the malignant RWPE2-W99, WPEl-NB26 and CTPE cells, all express varying levels of P- and N-cadherins (Achanazar et al., 2004). RWPE2-W99 and WPEl-NB26 cells express lower than normal levels of P-cadherin, but increased levels of N-cadherin as compared to RWPE-l cells. The levels of N -cadherin in the RWPE2-W99 and WPEl-NB26 cell lines are 2.8 and 8—fold higher than the level observed in RWPE-l cells respectively. While two of the three malignant cell lines show low P-cadherin and increased levels of N-cadherin, CT PE cells show increased levels of P-cadherin and undetectable levels of N -cadherin. The level of P-cadherin in the CTPE cell line is 2-fold higher that the level observed in the RWPE-l cell line. Cadherin switching and heterogeneity of cadherin expression observed in these cell lines rnirrrics cadherin switching and heterogeneous cadherin expression observed in human prostate cancers (Achanazar et al., 2004). These results are also consistent with cadherin expression observed in other human prostate cancer cell lines such as PC-3 and DU145 (Bussemakers et al., 2000; Tran et al., 1999). A loss or increased expression of one or more cadherins can disrupt homeostasis resulting in a change in cell adhesion, shape, polarization, and motility, and thus, may contribute to an invasive phenotype. 70 Invasive prostate cancer cells have decreased or abnormal expression of both integrins and cadherins, which assist in their dissemination from the prostate (Bello-DeOcampo et al., 2001b; Achanazar et al., 2004). Abnormal expression of adhesion molecules such as, integrins and cadherins, facilitate cell invasion by allowing cells to detach and move into the surrounding tissues. To degrade the extracellular matrix and invade surrounding tissues, cancer cells secrete proteolytic enzymes. Proteases: Although both normal and neoplastic cells produce matrix metalloproteinases (MMPs) and other proteolytic enzymes, only malignant cells are invasive. Normal cells may express proteases during such processes as tissue remodeling, but this expression is transient. Many human prostate cancers show elevated secretion of proteolytic enzymes, MMP-2 and MMP-9, which can degrade the extracellular matrix and thus assist in the spread of tumor cells (Lokeshwar et al., 1993; Festuccia et al., 1996). Another protease, urokinase-type plasminogen activator (u-PA), is also secreted by normal cells but higher levels of expression are associated with metastatic prostate cancer (W aghray and Webber, 1995). u-PA can degrade the extracellular matrix directly, but it can also activate MMP-2 and MMP-9, that are involved in the degradation of type IV collegen in the basement membrane (W ebber and Waghray, 1995). RWPE-l and RWPE2-W99 both secrete detectable levels of MMP-2 and MMP-9 (W ebber et al., 1996b). However, RWPE-2 cells secrete higher levels of both MMP-2 and MMP-9 than RWPE-l cells. RWPE-2 cells also produce considerably higher levels 71 of u-PA compared to RWPE-l cells and normal prostatic epithelium (W ebber et al., 1996b). CTPE cells secrete higher levels of active MMP-2 and MMP-9 compared to the RWPE-l cell line (Achanzar et al., 2001). The levels of protease expression are not known for the family of MNU cell lines, however their invasive behavior in vitro has been characterized. The invasive potential of a cell line can also be examined in vitro using the Boyden chamber invasion assay. Invasion in vitro: The invasive ability of a cell line can be examined in vitro using a reconstituted basement membrane and a Boyden chamber. In this assay, Matrigel serves as a reconstituted basement membrane and is composed of several matrix proteins. The Boyden chamber assay is a useful in vitro test, since invasion by cancer cells through the basement membrane in vivo is one of the first steps in the progression to metastasis. The invasive ability of the MNU cell lines was determined using the Boyden chamber invasion assay and was compared with invasion by DU145 cells. The invasive ability of a cell line derived from a metastatic deposit, DU-145, was set at 100%. In comparison to this cell line, the RWPE-l and MNU cell lines showed different invasive abilities (Figure 3.6) (Webber et al., 2001). 72 DU-145 WPEl-NB26 WPEl-NBll WPEl-NB14 WPEl-NA22 RWPE-l I 100 A I'—"" 95* d j—t 73** d:::i—- 30... .3 9 ll 1 0 20 40 60 80 100 120 Percent Invasion Figure 3.6 The invasive ability of MN U cell lines compared with that of RWPE-l and DU-145 cells by a modified Boyden chamber in vitro invasion assay. Cells were plated at 200,000 cells/chamber on a Matrigel-coated filter and allowed to invade for 24 h. +/- SEM. Two tailed t-test is shown as *P = 0.028, **P = 0.007, and ***P = 0.04 (Webber et al., 2001). The WPEl-NA22 cells showed low invasion (9%), which was not significantly higher than that of the non-tumorigenic, non-invasive, RWPE-l cells ( 1%), in comparison to the highly metastatic DU-145 cell line. WPEl-NB 14 and WPEl-NBll cells showed 30% and 73% invasion respectively, and WPEl-NB26 showed the highest invasion (95%). In another in vitro Boyden chamber assay, the invasive potential of WPEl-NB26 was set at 100% (Achanzar et al., 2004). In comparison to the WPEl-NB26 cells, CTPE cells show 78% invasion and RWPE2-W99 cells show 55% invasion (Figure 3.7). 73 WPE1-N826 [ _ V : 100% q 0) .s . ‘ ~ : CTPE, -‘ ' . ——i78°/.* U 1 RWPez-wss . a ‘ , . j—issvo“ 0% 20% 40% 60% 80% 100% % Invasion Figure 3.7 A comparison of the invasive ability of the three tumorigenic cell lines in vitro is shown where the invasive ability of WPEl-NB26 cells is taken as 100%. Cells were plated at 200,000 cells/Boyden chamber on Matrigel-coated filters and allowed to invade for 48 h. Results are plotted as i SD. *p = 0.1095, **p = 0.0008 (Achanzar et al., 2004). ' RWPE-l cells are non-tumorigenic when injected subcutaneously in nude mice while RWPE2-W99 cells form slow growing tumors (Achanzar et al., 2004). After sub- cutaneous injection into nude mice, the WPEl-NA22 cells form the slowest growing tumors amongst the MN U family of cell lines (W ebber et al., 2001). Both WPEl-NB14 and WPEl-NBll cells form tumors of intermediate size, and WPEl-NB26 cells form the largest tumors that are invasive. CTPE cells form large rapidly growing tumors that are highly invasive when injected subcutaneously in nude mice (Achanzar et al., 2004). 74 On the basis of their characteristics, the cell lines in this paper have been ranked in the following order of increasing malignancy: RWPE-l cell line at the non-malignant end of the process of carcinogenesis, WPEl-NA22 showing the lowest invasive ability, followed by WPEl-NB14, RWPE2-W99, and WPEl-NBll showing intermediate, but progressively increasing invasive abilities, and CTPE and WPEl-NB26 cell lines showing the greatest invasive ability (Figure 3.8). Carcinogenesis and Progression Luminal Cells my 3“ F 1...... rate at «is ill 0.13.6 lg‘h heat with Basement . H"; \—y—l H—i " K-‘A‘ Membrane RWPE-l WPE]- WPEl I ( r KW Normal Immortalized NA22 'NBM , ‘ ,E RWPE2-W99 £15:- CTPE WPE]- N826 Figure 3.8 A schematic diagram showing steps in the multistep process of carcinogenesis and tumor progression in the human prostate and the points possibly represented by RWPE-l, RWPE-2-W99, MNU, and CT PE cell lines in this progression. The sequence of progression from non-malignant RWPE-l cells to the highly malignant WPEl-NB26 cells: RWPE-1< WPEl-NA22< WPEl-NB14< RWPE2-W99< WPEl-NB11< CTPE< WPEl-NB26 (Modified from Webber et al., 2001). 75 Conclusions Each of the human prostate cell lines discussed have some unique characteristics for studying prostate carcinogenesis and cancer progression. The RWPE-l cell line has characteristics of normal prostate tissue even though the cells have been immortalized. RWPE-l cells show epithelial morphology, respond to growth factors, polarize and form acini in vitro and are non-tumorigenic in vivo. RWPE—l cells may, therefore, be useful as a comparison in studies with other prostate cell lines, especially its related cell lines. The behavior of RWPE2-W99 and MNU cell lines, as far as their tumorigenicity and invasive ability in vivo is concerned, is consistent with their malignant characteristics in vitro. The RWPE-2 and MNU cell lines show varying degrees of malignant characteristics which permits studies on prostate carcinogenesis, prostate cancer progression, and testing agents for chemoprevention and treatment of early stage prostate cancer. The CTPE cell line, another derivative of the parent RWPE-l cell line, serves as a useful model of cadmium- induced prostate cancer in men. The fact that all of these cell lines are derived from the same parental, RWPE-l cells, allows one to examine molecular events associated with prostate carcinogenesis and tumor progression. The prostate cell lines used in my study include: RWPE2-W99, WPEl-NB26, and CT PE. 76 Literature cited Achanzar, W.E., Lamar, P., Tokar, E.J., Rivette, A.S., Bello-DeOcampo, D., Prozialeck, W.C., Webber, M.M., and Waalkes, M. P.: Human prostate cell lines mimic heterogeneity of cadherin expression in human prostate cancer. UroOncology 4:15-25, 2004. Achanzar, W.E., Kiwan, B.A., Liu, J ., Wuader, S.T., Webber, M.M., and Waalkes, M.P.: Cadmium-induced malignant transformation of human prostate epithelial cells. Cancer Research 61:455-458, 2001. Anwar, K., Nakakuki, K., Shiraishi, T., Naiki, H., Yatani, R., and Inuzuka, M.: Presence of ms oncogene mutations and human papillomavirus DNA in human prostate carcinomas. Cancer Research 52:5991-5996, 1992. Bello, D., Webber, M.M., Kleinman, H., Wartinger, DD, and Rhim, J .S.: Androgen responsive adult human prostatic epithelial cell lines immortalized by human papillomavirus 18. Carcinogenesis 18:1215-1223, 1997. Bello-DeOcampo, D., Kleinman, H.K., DeOcampo, ND, and Webber, M.M.: Laminin 1 and (1431 integrin regulate acinar morphogenesis of normal and malignant human prostate epithelial cells. Prostate 462142-153, 2001a. Bello-DeOcampo, D., Kleinman, H.K., and Webber, M.M.: The role of 06131 integrin and EGF in normal and malignant acinar morphogenesis of human prostatic, epithelial cells. Mutation Research 480-4811209-217, 2001b. Bussemakers, M.J.G., Van Bokhoven, A., Tomita, K., Jansen, C.F.J., and Schalken, J .A.: Complex cadherin expression in human prostate cancer cells. International Journal of Cancer 85:446-450, 2000. Chen, T., Pecoraro, G., and Defendi, V.: Genetic analysis of in vitro progression of human papillomavirus-transfected human cervical cells. Cancer Research 53: 1 167-1 171, 1993. DeMarzo, A.M., Nelson, W.G., Meeker, A.K., and Coffey, D.S.: Stem cell features of benign and malignant prostate epithelial cells. Journal of Urology 160:2381 2392, 1998. Festuccia, C., Bologna, M., Vicentini, C., Tacconelli, A., Miano, R., Violini, S., and Mackay, A.R.: Increased matrix metalloproteinase-9 secretion in short-term tissue cultures of prostatic tumor cells. International Journal of Cancer 69:386 393, 1996. 77 Horoszewicz, J .S., Leong, S.S., Chu, T.M., Wajsman, Z.L., Friedman, M., Papsidero, L., Kim, U., Chai, L.S., Kakati, S., Arya, SK, and Sandberg, A.A.: The LNCaP cell line-A new model for studies on human prostatic carcinoma. In: Progress in Clinical and Biological Research, Models for Prostate Cancer, Murphy, G.P. (ed.), Buffalo, NY, Alan R. Liss Inc., 37:67-84, 1980. Kaighn, M.E., Narayan, S., Ohnuki, Y., Lechner, J .F., and Jones, L.W.: Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Investigative Urology 17:16-23, 1979. Lokeshwar, B.L., Selzer, M.G., Block, N .L., and Gunja—Smith, Z.: Secretion of matrix metalloproteinases and their inhibitors (tissue inhibitor of metalloproteinases) by human prostate in explant cultures: reduced tissue inhibitor of metalloproteinase secretion by malignant tissues. Cancer Research 53:4493-4498, 1993. Mickey, D.D., Stone, K.R., Wunderli, H., Mickey, G.H., and Paulson, D.F.: Characterization of a human prostate adenocarcinoma cell line (DU145) as a monolayer culture and as a solid tumor in athymic mice. In: Progress in Clinical and Biological Research, Models for Prostate Cancer, Murphy, G.P. (ed.), Buffalo, NY, Alan R. Liss Inc., 37:67-84, 1980. Rhim, J .S., Webber, M.M., Bello, D., Lee, M.S., Amstein, P., Chen, L., and Jay, G.: Stepwise immortalization and transformation of adult human prostate epithelial cells by a combination of HPV-18 and v-Ki-ras. Proceedings of the National Academy of Sciences U.S.A. 91:11874-11878, 1994. Tran, N.L., N agle, R.B., Cress, A.E., and Heimark, R.L.: N-cadherin expression in human prostate carcinoma cell lines. American Journal of Pathology 155:787 -79& 1999. Waalkes, M.P., Rehm, S., Coogan, T.P., and Ward, J .M.: Role of cadmium in the etiology of cancer of the prostate. In: J .A. Thomas and H.D. Colby (eds.), Endocrine Toxicology, Ed. 2, pp. 227-243. Washington, DC: Taylor & Francis, 1997. Waghray, A., and Webber, M.M.: Retinoic acid modulates extracellular urokinase-type plasminogen activator activity in DU-145 human prostatic carcinoma cells. Clinical Cancer Research 1:747-753, 1995. Webber, M.M., Quader, S.T.A., Kleinman, H.K., Bello-DeOcampo, D., Storto, P.S., Bice, G., DeMendonca-Calaca, W., and Williams, D.: Human cell lines as an in vitro/in vivo model for prostate carcinogenesis and progression. Prostate 47: 1-13, 2001. 78 Webber, M.M., Bello, D., Kleinman, H.K., and Hoffman, M.P.: Acinar differentiation by non-malignant immortalized human prostatic epithelial cells and its loss by malignant cells. Carcinogenesis 18:1225-1231, 1997. Webber, M.M., Bello, D., Kleinman, H..,K Wartinger, D.D., Williams, DE, and Rhim, J .S.: Prostate specific antigen and androgen receptor induction and characterization of an immortalized adult human prostatic epithelial cell line. Carcinogenesis 17: 1641- 1646, 1996a. Webber, M.M., Bello, D., and Quader, S.: Immortalized and tumorigenic adult human prostatic epithelial cell lines: characteristics and applications part 1. Cell markers and immortalized nontumorigenic cell lines. Prostate 29:386-394, 1996b. Webber, M.M., and Waghray, A.: Urokinase-mediated extracellular matrix degradation by human prostatic carcinoma cells and its inhibition by retinoic acid. Clinical Cancer Research 1:755—761, 1995. Woodworth, C.D., Waggoner, S., Barnes, W., Stoler, M.H., and DiPaolo, J .A.: Human cervical and foreskin epithelial cells immortalized by human papillomavirus DNAs exhibit dysplastic differentiation in vivo. Cancer Research 50:3709-3715, 1990. Yankaskas, J .R., Haizlip, J .E., Conrad, M., Koval, D., Lazarowski, E., Paradiso, A.M., Rinehart, C.A. Jr, Sarkadi, B., Schlegel, R., and Boucher, R.C.: Papilloma virus immortalized tracheal epithelial cells retain a well-differentiated phenotype. American Journal of Physiology 264:C1219-C1230, 1993. 79 CHAPTER FOUR XENOTRANSPLANTATION OF HUMAN PROSTATE CANCER CELLS 8O Abstract The metastatic process is not only determined by the characteristics of the tumor, but also by its surrounding microenvironment. The implantation site of tumors cells has been shown to influence both the rate of tumorigenesis, as well as the rate and pattern of metastasis in immune—suppressed mice. This illustrates the importance of tumor-stromal cell interaction and emphasizes the critical role of epigenetic factors in influencing prostate cancer growth. Although intrinsic tumor cell factors are important in determining the incidence of metastatic spread, it is also evident that simple manipulation of the route of tumor cell injection can allow maximum expression of the metastatic phenotype. The orthotopic site, the prostate, appears to result in metastasis to several organs following cell injection or tissue implantation while other sites may lack or only show metastasis to few organs. The metastatic phenotype of the cell lines discussed in this chapter are different and they are not affected in the same way by alteration of injection site in immune-suppressed mice. Therefore one should evaluate each xenograft before designing a study, to select an appropriate model for prostate cancer. Keywords Cell lines, DU145, injection, prostate cancer, LN CaP, metastasis, PC-3, subline, technique, xenograft 81 Introduction In vivo models are important for studying the behavior, growth and gene expression of tumor tissue in a more natural environment that cannot be easily mimicked in vitro. Metastatic spread to other tissues, angiogenesis, and interaction with stromal cells are lacking in vitro. In addition, most animals rarely develop spontaneous prostate cancer metastasis. For example, in non-human primates prostatic carcinoma is almost non-existent (Waters et al., 1998; Hughes and Lang, 1978). Although canine models show a rate and pattern of prostate cancer progression and metastasis similar to humans, the lack of control over the population of dogs that will develop prostate cancer and subsequent bone metastasis makes the use of this model difficult (N avone et al., 1999; Waters et al., 1998). Rodent models have been developed for a variety of cancers and tumors may be induced or implanted in them that grow rapidly and often metastasize, but the cells are nevertheless of rodent origin. With the use of immune-suppressed mice as a model of prostate cancer, it is possible to study human prostate cells, such as, patient derived specimens or established prostate cell lines, in an in vivo environment. When cells or tissues are transplanted from one species to another it is called a xenograft. Human tumor xenotransplantation began after the discovery of a mutant mouse with low immunity. The nude mouse, a hairless mutant, has a normal complement of bone marrow-dependent B-cells, but lacks a thymus, which is essential for the production of T—cells and, therefore, has a deficient cell-mediated immune response (Rygaard and Povlsen, 1982). Another immune-suppressed mouse model for xenotransplantation, consists of mice with severe combined immune deficiency (SCID) (Bosma et al., 1983). The SCID mutation selectively impairs the differentiation of 82 lymphopoietic stem cells and as a result they are deficient in immunologic functions mediated by B— and T-cells. Therefore, SCID mice are even more immune-deficient than nude mice allowing a higher percentage of tumor engraftment, enhanced tumor growth rate, and less tumor regression of human tumors (Kim, 1996; Sato et al., 1997). However their enhanced immune-deficiency, as compared to nude mice, also makes SCID mice more susceptible to infection. In 1977 the first transplantable human prostate xenograft, PC-82, was established (van Weerden and Romijn, 2000). Small fragments of human prostate cancer tissue were transplanted subcutaneously in athymic, nude mice. Many prostate cancer xenografts have been established since then. Using xenograft models of prostate cancer it is possible to determine the influence of the microenvironment on gene expression, growth, and behavior of tumor cells within the prostate gland and other organ sites. Hormonal status may also be altered in prostate cancer xenografts to study the importance of androgen- independence in tumor progression and metastasis. Xenograft models are particularly useful for testing new drugs for chemotherapy. Since studies using xenografts are an important step before clinical trials of new drugs can be conducted for cancer treatment, the xenograft model should mimic the human disease as closely as possible in order to collect relevant and applicable data. Xenograft models of prostate cancer that represent various aspects of human prostate cancer exist, however most lack metastasis to bone (van Weerden and Romijn, 2000). A xenograft model of prostate cancer that metastasizes to bone would be useful because metastatic spread to the bone is the main cause of morbidity among prostate cancer patients. Another obstacle with xenograft models is the sustained growth of 83 human prostate cells in immune-suppressed mice. Therefore, several approaches have been attempted to overcome such obstacles. These approaches include different implantation sites, implantation of a high number of cells, implantation of cultured pieces of tissue rather than a cell suspension, and the use of in viva-selected cell lines or sublines. These approaches will further be discussed with the aim of identifying appropriate xenograft models of human prostate cancer and metastasis. This information will be useful when selecting a xenograft model for prostate cancer studies, as key features and their utility in understanding the mechanisms of prostate carcinogenesis and metastasis will be provided for several xenograft models. Intra-spleen injection: Most investigators use the technique of intrasplenic cell injection to examine the metastatic potential of a cell line. The spleen filters foreign particles from the bloodstream and also eliminates damaged blood cells. Blood leaving the spleen empties into the splenic vein which then empties into the hepatic portal vein before it enters the liver (Paulsen, 2000). Transplantation of human prostate cancer cells to the intrasplenic site in a nude mouse requires anesthesia and surgery. After anesthetizing the mouse an incision is made in the left flank through the skin and peritoneum to expose the spleen for cell injection. The incidence of metastasis after the intra-splenic injection of either PC-3, DU145, LNCaP cells or their metastatic sublines is shown in Table 4.1. 84 Table 4.1 Incidence of metastasis 6-8 weeks after intra-splenic injection of either PC-3, DU145, or LNCaP cells or their metastatic sublines in athymic mice. Lung Liver Abdomen* Reference DU145 30% 0 0 Kozlowski et al., 1984 PC-3 80% 100% 85% Kozlowski et al., 1984 PC-3 14% 14% 57% Sherwood et al., 1990 431-P 20% 60% 50% Shevrin et al., 1989 LNCaP NR 0 NR Pettaway et al., 1996 LN CaP-LN3 NR 40% NR Pettaway et al., 1996 *Includes tumor ascites NR, Not reported Intra-splenic injection of PC-3 cells (5X105) resulted in 16/20 mice developing lung metastases, 20/20 mice developing liver metastases and 13/20 mice were found to have tumor ascites (Kozlowski et al., 1984). After intra—splenic injection of DU145 cells (5X105) only 3/10 mice developed pulmonary metastases while metastases to the liver were not observed (Kozlowski et al., 1984). In another study using the same technique, injection of PC-3 cells (1X106) in athymic mice was found to show metastasis in the liver 1/7, lung 1/7, and diaphragm 4/7 (Sherwood et al., 1990). Although a higher cell number was used compared to Kozlowski et a1. 1984, the incidence of metastases observed did 85 not increase. These results demonstrate the inconsistency of this injection technique. Instead of increasing the cell number some investigators develop sublines to obtain a higher incidence of metastasis after cell injection. For example, intra-splenic injection of a PC-3 subline, 431—P (1X106), which was obtained from the 16‘h passage of PC—3 cells injected subcutaneously in athymic mice, resulted in liver metastases in 6/ 10 mice, intra- ‘abdominal tumor growth in 5/ 10 mice, and lung lesions in 2/ 10 mice (Shevrin et al., 1989). A subline of LNCaP cells also shows greater metastatic potential than the parental LNCaP cell line. Intra-splenic injection of LNCaP (2X106) did not result in any visible tumors in the spleen, pancreas, or liver (Pettaway et al., 1996). However, liver metastases were observed in 4/10 mice after intrasplenic injection of the LNCaP metastatic subline, LNCaP-LN3. LNCaP-LN3 was derived from a tumor from a lymph node after cell injection of LNCaP cells in the prostate. The intra-peritoneal cavity is another site used by investigators to examine the metastasic behavior of a cell line. Intra-peritoneal injection: Metastases resulting from the intra-peritoneal route of tumor cell injection have been attributed to improved tumor vascularity due to the absence of the restrictive fibrous sheath around the primary tumor (Morrissey et al., 1980; Takahashi et al., 1978). Another advantage of this site is that anesthesia is not necessary for cell injection in athymic mice. 86 Athymic mice given an intraperitoneal injection of PC-3 cells (1X106) showed metastases to the lung (1/7) and liver (4/7) three weeks after cell injection (Ware et al., 1985). In comparison to the parental PC-3 cell line, it’s two sublines, 1-LN and clone 4, were both found to show a higher incidence of metastases in athymic mice given an intraperitoneal cell injection (Table 4.2). The l-LN cell line was recovered from a lymph node metastasis in a PC-3 tumor bearing mouse and clone 4 is a clonal derivative of l-LN. In addition, both l-LN and clone 4 cells formed solid tumor masses that adhered to the lining of the peritoneal cavity in all mice, ascitic fluid was found in 3/7 l-LN injected mice. Another subline of PC-3 cells, 431-P, also resulted in a high incidence of abdominal tumor growth after intra-peritoneal cell injection (Shevrin et al., 1989). Intra- abdominal tumor growth and malignant ascites developed in 17/22 mice and lung metastases were observed in 6/22 mice (Table 4.2). 87 Table 4.2 Incidence of metastasis after intra-peritoneal injection of PC-3 cells or metastatic sublines of PC-3 cells in athymic mice. Lung Liver Abdomen Reference PCB 14% 57% 0 Ware et al., 1985 l-LN 71% 71% 100% Ware et al., 1985 clone 4 80% 80% 100% Ware et al., 1985 431-P 29% 0% 77%* Shevrin et al., 1989 *Includes tumor grth and ascitic fluid The lack of tumor cell spread after intra—peritoneal cell injection may be due to the mice developing a high number of abdominal metastases. This is supported by the results of implantation experiments with sublines with preferential metastatic abilities. Organ- targeted sublines showed an increase in non-specific metastasis after intra-peritoneal injection in SCID mice (Wang and Steams, 1991). The intra-venous cell injection technique, in contrast to the intrasplenic and intra- peritoneal injection techniques, places cells directly in circulation via the tail vein, so the growth of metastatic tumors may be more dependent on cell behavior rather than the selected microenvironment. 88 Intra-venous injection: Before reaching the general circulation, cells injected via the tail vein must first pass through the lung, therefore, the lungs are the most common site of metastatic tumor growth following intravenous cell injection. Human prostate cancer cells transfected with a marker gene and injected intravenously in athymic, nude mice, show a high number of micrometastases in the lungs compared to liver, bone, kidney, and brain tissues one hour after cell injection (Holleran et al., 2002). Cells injected via the tail vein behave differently than when they are placed at another site such as in the spleen or the intraperitoneal cavity. For example, the incidence of lung metastases decreases and the incidence of metastatic tumor growth in other organs are more common following intrasplenic or intraperitoneal cell injection in comparison to intravenous cell injection (Ware et al., 1985; Shevrin et al., 1989). Following intravenous cell injection in athymic mice, both PC-3 and DU145 cells colonize the lungs, however, LNCaP cells do not metastasize (Kozlowski et al., 1984; Ware et al., 1985; Pettaway et al., 1996) (Table 4.3). 89 Table 4.3 Metastatic potential of PC-3, DU 145, or LN CaP cells in athymic mice following intravenous cell injection. Reference(s) PC-3 ++ Kozlowski et al., 1984; Ware et al., 1985 DU145 + Kozlowski et al., 1984 LNCaP O Pettaway et al., 1996 O, non-metastatic; +, 1-20% metastatic; ++, 20-50% metastatic Sublines of LNCaP cells, generated from repeated in viva selection of LNCaP cells, are also non—metastatic following intravenous cell injection in athymic mice, while sublines of PC-3 cells show a higher incidence of lung metastasis in comparison to the parental PC-3 cells (Pettaway et al., 1996). Sublines of PC-3 cells have also been generated that preferentially metastasize at ~80% efficiency to the lumbar vertebrae, the mandibular region of the right cheek, the rib cartilage, and the right front knee bone in SCID mice (Wang and Stearns, 1991). These sublines of PC-3 cells, which preferentially metastasize, were continuously grown and selected both in vitro and following intravenous cell injection in SCID mice. Metastases to bone have been observed following intravenous injection of a PC-3 subline, 43l-P, but this required occlusion of the inferior vena cava during cell injection (Shevrin et al., 1988). Although sublines may show a high incidence of metastasis, experiments with P03 and DU145 cells, which have not undergone any selection processes, have resulted in a relatively low incidence of lung metastasis. 90 In separate groups of mice given injections of either DU145 of PC-3 (1X106) cells via the lateral tail vein, only 1 out of fifteen (6.6%) mice showed microscopic metastases in the lungs (Kozlowski et al., 1984). In another study, tail vein injection of PC—3 cells (1X106) resulted in a higher incidence of lung metastases, 47% (Ware et al., 1985). The difference in the incidence of lung metastases observed in the two studies using PC-3 cells could be due to a technical error such as misinjection. In a recent study application of colloidal gold labeled with an isotope followed by quantification in tissue samples by neutron activation was shown to be a valid method for quantifying the tail vein injection technique (Groman and Reinhardt, 2004). To avoid or reduce technical errors during intravenous cell injection one could use the colloidal gold method or consider other techniques for cell injection. The least difficult technique for cell injection, the subcutaneous method, results in very little experimental error since the site is readily accessible for cell injection as well as observation and measurement of the subsequent tumor. Subcutaneous injection: The subcutaneous cell injection technique does not require surgery or anesthesia although some investigators use anesthesia to ensure the accuracy of cell injection. The subcutaneous cell injection technique also allows one, not only to observe tumor growth, but also to measure the resulting tumor. Therefore, the subcutaneous xenograft model may be useful for studies which involve hormonal manipulation or screening of potential treatments that inhibit or slow tumor growth for additional testing in clinical trials. The 91 presence or lack of metastasis following subcutaneous cell injection of PC-3, DU145, or LNCaP cells is shown in Table 4.4. Table 4.4 Incidence of metastasis following subcutaneous injection of PC-3, DU145, or LN CaP cells in athymic mice. Lung Lymph node Reference PC-3 20% 60% Rembrink et al., 1997 DU145 0% 0% Devi et al., 2002; Mickey et al., 1977 LNCaP 0% 0% Rembrink et al., 1997; Stephenson et al., 1992 After subcutaneous cell injection of PC-3 or DU145 (~2X105) cells, 8/9 (89%) and 4/6 (67%) of the mice respectively developed tumors (T rikha et al., 1998). Although PC-3 cells had a higher take rate compared to DU145 cells, no metastases were observed in any of the mice after 8-14 weeks. In addition, subcutaneous injection of a higher number of DU145 cells does not result in metastasis (Devi et al., 2002; Mickey et al., 1977). DU145 cells placed at the subcutaneous site also do not show accelerated tumor growth compared to animals without hormone treatment, which corresponds to their hormone- insensitive behavior in vitro (Mickey et al., 1977). In contrast to DU145 cells, subcutaneous injection of a higher number of PC-3 cells (1X106), results in lung and lymph node metastases in 1/5 and 3/5 mice respectively, six weeks after cell injection (Rembrink et al., 1997). 92 LNCaP cells do not show signs of metastasis, regardless of the cell number, after subcutaneous cell injection (Rembrink et al., 1997; Stephenson et al., 1992). In these two studies, testing the in viva growth of LNCaP cells following subcutaneous cell injection, the tumor take rate was less than 10%. However, the tumor take rate was 68% when LNCaP cells were mixed with Matrigel for subcutaneous cell injection (Lim et al., 1993). Furthermore the growth of LNCaP cells can be manipulated by castration of LNCaP- bearing athymic nude mice (Lim et al., 1993). Castration leads to involution of the tumor and stabilization of serum PSA level. This xenograft model using LNCaP cells may be useful as a model of hormone-sensitive human prostate cancer. But the results collected from studies using the hormone-sensitive model may only be applicable to human prostate cancers in which the cells express an androgen receptor mutation, as observed in LNCaP cells. Sublines of human prostate cancer cells such as PC-3 and LNCaP, are eight times more metastatic and twice as tumorigenic, respectively, in comparison to the parental cell lines following subcutaneous cell injection (Kozlowski et al., 1984; Pettaway et al., 1996). Among the three most commonly tested cell lines, PC-3, DU145, and LNCaP, only the PC-3 cell line metastasizes after subcutaneous cell injection. Therefore, as a subcutaneous xenograft model, the PC-3 cell line may be useful for evaluating treatments that may prevent or slow metastasis from the primary tumor, whereas LNCaP and DU145 cells may be more useful for evaluating treatments that may slow or inhibit primary tumor growth. For ideal studies on primary tumor growth, a family of cell lines exists that represents a progression of tumor growth when injected subcutaneously in nude mice from small, slow-growing tumors to large, fast-growing tumors that are invasive. 93 Included in this family of cell lines is the parental, non-tumorigenic, RWPE-l cell line which serves as a control in vitro or as a standardized model in viva when studying prostate cancer progression. RWPE-l cells were isolated from the normal prostate of a 54 year-old Caucasian man undergoing radical prostatectomy because of cystectomy for bladder cancer and immortalized with the human papilloma virus-18 (HPV-18) (Bello et al., 1997; Rhim et al., 1994). Several tumorigenic cell lines, the MNU cell lines, were derived from RWPE-l by transformation with N—methyl—N—nitrosourea (MNU) (W ebber et al., 2001). These cell lines include: WPEl-NA22, WPEl-NB14, WPEl-NBl l, and WPEl-NB26. RWPE-l cells were also transformed by Ki-ras or cadmium exposure to obtain the RWPE-2 and CTPE cell lines, respectively (Bello et al., 1997; Achanazar et al., 2001). Cells were tested for tumorigenicity by subcutaneously injecting 5X105 cells with Matrigel in athymic, nude mice. As indicated in previous experiments, the RWPE-l cell line does not form tumors in nude mice, and when injected with Matrigel, the cells organized similarly to normal glands in viva. All of the MNU cell lines were capable of forming tumors when injected into nude mice. The WPEl-NA22 cells were found to form the slowest growing tumors, followed by WPEl-NB14 and WPEl-N B1 1 cells, which formed tumors of intermediate size, while WPEl-NB26 cells formed the largest tumors (W ebber et a1, 1997). The MNU cell lines are unique because they show progression of characteristics from non-tumorigenic, to low, and then to a high level of malignancy and mimic different stages of carcinogenesis and progression as they occur in the human prostate. 94 The related, RWPE-2 cells form small, slow-growing tumors in mice following subcutaneous cell injection and provide a model for prostate cancers in which the cells show increased expression of Ki-ras (Bello et al., 1997). CTPE cells not only produce tumors, but these cells are invasive when inoculated (1X106) subcutaneously into nude mice (Achanazar et al., 2001). Tumors arose in 18/20 mice within six weeks after inoculation with CTPE cells and 80% of these tumors invaded into the subdermal muscle, fat, or connective tissue. These results indicate the highly aggressive nature of the CT PE cells and should lead to a better understanding of the mechanisms involved metastasis, as well as, in cadmium-induced prostatic malignancies. The six cell lines; WPEl-NA22, WPEl—NBI4, WPEl-NBI 1, WPEl-NB26, RWPE—Z, and CTPE, all share a common lineage and represent a unique and relevant model which mimics stages in prostatic intra- epithelial neoplasia and progression to invasive cancer and can be used to study carcinogenesis, progression, intervention, and chemoprevention (W ebber etal., 2001). The subcutaneous site is easily accessible for cell injection and most human prostate cancer cell lines form tumors, but for prostate cancer cells it does not correspond with their anatomic origin or orthotopic site, the prostate. Orthotopic Injection: In 1992, the technique of orthotopic transplantation was reintroduced as a means of inducing spontaneous metastasis originating from the prostate (Fu et al., 1992; Stephenson et al., 1992; van Weerden et,al., 2000). Several investigators have observed a higher incidence of metastasis following orthotopic cell injection compared to subcutaneous cell injection of PC-3, DU145, and LNCaP cells in immune-suppressed 95 mice Waters et al., 1995; Pettaway et al., 1996; Sato et al., 1997; Rembrink et al., 1997; Trikha et al., 1998). Cell injection at the orthotopic site in immune-suppressed mice requires anesthesia, surgery, and technical experience. Some drugs used by investigators for anesthetizing mice prior to orthotopic cell injection include; methoxyflurane, nembutal, and tribromoethanol. After anesthetizing the mouse, an incision is made in the abdominal wall to expose the prostate for cell injection. Cells suspended in Hank’s balanced salt solution, serum free medium (SFM), or Ham’s medium have been administered with a 28 or 30 gauge needle in 2040 v.1 into the prostate in immune- suppressed mice. Successful cell injection is usually confirmed by the absence of leakage from the prostate or visualizing the expansion of the prostate or both. After cell injection the prostate gland is placed back inside the abdominal wall and the incision is either closed with nylon or silk sutures or wound clips. The incidence of metastasis following orthotopic cell injection of PC-3, DU145, or LNCaP cells is shown in Table 4.5. 96 Table 4.5 Incidence of metastasis following orthotopic injection of PC-3, DU145, or LN CaP cells in immune-suppressed mice. Time+ Lung Lymph node Other Reference(s) PC-3 ~200,000 8-14 NR 0% 0% Trikha et al.,1998 5 x 105 9 10% 100% 18%a Waters et al.,1995 1 x 106 7 100% 100% NR Rembrink et al.,1997 DU145 ~200,000 7 NR 50% 100%b Trikha et al., 1998 LNCaP 1 x 106 13 0% 57% NR Rembrink et al.,l997 2 x 106 14 NR 28% NR Pettaway et al.,l996 +, weeks a, kidney metastasis b, peritoneal metastasis NR, not reported In SCID mice given orthotopic injections of either PC-3 or DU145 (~200,000) cells, metastasis was only observed in mice given DU145 cells (Trikha et al., 1998). Seven weeks following orthotopic cell injection of DU145 cells, all four mice inoculated were dead, whereas some mice given orthotopic injections of PC-3 cells survived twice as long. Upon examination, all mice inoculated with DU145 cells were positive for peritoneal invasion and two of the mice also showed lymph node metastasis (T rikha et al., 1998). In another study using the orthotopic cell injection technique, but a higher number of PC-3 cells (5 x 105), 10/ 10 mice and 1/10 mice showed lymph node and lung metastasis, respectively, and in 2 athymic mice, metastatic tumors were also observed in the kidney after 8-9 weeks (Waters et al., 1995). The difference in the metastatic potential of PC-3 cells compared to DU145 cells following orthotopic cell injection may 97 be due to experimental error or DU145 cells may be more metastatic because orthotopic injection of less cells resulted in metastasis. Regardless, both PC-3 and DU145 cells metastasize following orthotopic cell injection in immune-suppressed mice. LNCaP cells are also metastatic following orthotopic cell injection, however, the mice are maintained for more than 90 days, which is approxiamately twice as long as mice given PC-3 or DU145 cells (Rembrink et al., 1997; Pettaway et al., 1996). Not only are the mice less likely to survive longer than about 9 weeks, but a lower cell number results in a higher incidence of metastasis in mice given orthotopic cell injections of PC-3 or DU145 cells compared to LNCaP cells (Table 4.5). Therefore, LNCaP cells are considered the least invasive compared to PC-3 and DU145 cells following orthotopic cell injection. While orthotopic cell injection results in a higher incidence of metastasis compared to subcutaneous cell injection, another technique, surgical orthotOpic implantation (80]), results in a higher incidence of metastasis compared to orthotopic cell injection (Fu et al., 1992; Wang et al., 1999). Surgical Orthotopic Implantation (801): Surgical orthotopic implantation is a technique which involves the implantation of tissue pieces in the prostate. This technique does require anesthesia and surgery, similar to the orthotopic cell injection technique; however, the use of tissue pieces instead of a cell suspension allows one to avoid spillage outside of the prostate during cell injection and, therefore, limit artificial metastasis (An et al., 1998). One of the most common anesthetics for S01 is isoflurane inhalation. Following anesthesia, an incision is made in the abdominal wall to expose the prostate. An incision is then made in the prostate 98 capsule for implantation of tumor tissue. Tissue pieces are collected from tumors that grow following subcutaneous cell injection in immune-suppressed mice. After the tumor tissue has been implanted in the prostate, the prostate capsule is closed with sutures and subsequently the abdominal wall. One side effect associated with 801 is urinary obstruction due to large local tumor growth (Fu et al., 1992; An et al., 1998; Wang et al., 1999). Hydronephrosis or swelling of the kidneys as a result of urinary obstruction, has also been observed (Fu et al., 1992; An et al., 1998; Wang et al., 1999). After orthotopic implantation of tissue pieces of PC-3, each about 1 mm3 in size, local growth and metastasis to the bladder and kidney, as well as, distant metastases to the lymph nodes were observed. In mice implanted with DU145, tumor tissue was only found to invade the. lamina propria of the urinary bladder. Although distant metastases were not observed in mice implanted with DU145, hydronephrosis due to urinary blockage was observed by both locally growing tumors of PC-3 and DU145 (Fu et al., 1992). ' As observed following 801 of PC-3, LNCaP tumors were often found to show invasion to the seminal vesicles, the bladder, and the lower abdominal wall (An et al., 1998; Wang et al., 1999). Distended urinary bladder and hydronephrosis were also frequently seen (An et al., 1998; Wang et al., 1999). Microscopic examination of tissue sections from mice implanted with PC-3 or LNCaP, demonstrated that both groups of mice were found to have similar incidences of lung and lymph node metastases (An et al., 1998; Wang et al., 1999). The results of these studies show that using the SOI technique one can obtain a high incidence of local tumor growth in the prostate, as well as, distant metastases. However, even though PC-3, DU145, and LNCaP all appear to be highly 99 tumorigenic and invasive following SOI, metastatic spread to bone was not observed. Models of prostate cancer with metastasis to bone would be useful for prostate cancer studies since metastatic deposits develop in bone before metastases to soft viscera become apparent (Bubendorf et al., 2000). Models of Bone Metastasis: More than 80% of prostate cancer patients develop bone metastases, and are generally associated with poor prognosis (Linehan et al., 1992). Since xenograft models of prostate cancer rarely show metastasis to bone, PC-3 or LNCaP cells were injected directly into the femur medullas of nude mice to compare their intraosseal growth (8005 et al., 1997). PC-3 and LNCaP tumors both colonized the bone marrow within a week. PC-3 tumors eventually broke through the bone cortex, invaded the surrounding tissues, and metastasized to the regional lymph nodes, however, LNCaP remained localized within the bone and appeared to regress and die after displacing the normal bone marrow cells. The different growth requirements of these two cell lines may explain the regression of LNCaP cells and the survival of PC-3 cells in mouse bone. As a more useful and ideal model of human prostate cancer metastasis to bone, investigators transplant human tissue, bone or lung, into immune-suppressed mice prior to human prostate cancer cell injection. This permits one to study the interaction between tumor cells and a human organ environment. Human adult bone (HAB), human adult lung (HAL), or mouse bone was transplanted subcutaneously in nude mice prior to intravenous cell injection of PC-3 or LNCaP (Yonou et al., 2001). Eight weeks after intravenous cell injection several organs 100 were evaluated for metastases. The incidence of metastasis to implanted human and host mouse tissue is shown in Table 4.6. Table 4.6 Incidence of metastasis to implanted human and host mouse tissue in SCID mice after tail vein injection of PC-3 or LN CaP cells. Human Bone“ Human Lunga Mouse Bone" Mouse Bone Mouse Lung PC-3 1 3/20 0/20 1/ 20 3/20 5/20 LNCaP 7/20 0/ 20 0/ 15 O/ 20 2/20 (Yonou et al. Cancer Research 61:2177-2182, 2001.) a, transplanted human, adult tissue or host mouse tissue In this model of bone metastasis, very few tumors developed when murine fetal bone was used suggesting that homing of human prostate cancer cells to bone is human-specific. PC-3 and LNCaP cells preferentially metastasized to HAB over HAL, which reflects the clinical features of prostate cancer (Yonou et al., 2001). PC-3 and LNCaP cells were derived from metastatic deposits, so to assist in understanding the mechanism of prostate cancer metastasis under conditions similar to those in the human body, additional models using clinical specimens or cell lines isolated from primary tumors may be more useful. 101 Conclusions Each xenograft method has unique properties which provide opportunities to identify the multiple molecular pathways in prostate cancer and metastasis. Intrasplenic cell injection of prostate cancer cells usually results in metastasis to the liver and abdominal cavity. Following intraperitoneal cell injection, mice develop a high incidence of abdominal metastases. Intravenous cell injection, without manipulation of the vena cava, is capable of resulting in a high incidence of lung metastasis and with manipulation of the vena cava, it is also possible to achieve metastasis to bone using this technique. The subcutaneous cell injection technique seems most promising for studies testing drugs that may slow or inhibit primary tumor growth. Distant metastasis can be observed using orthotopic cell injection or tissue implantation, however metastasis to bone has not been observed. For such studies, the femur of immune-suppressed mice may be used as the site of cell injection or the xenograft model with human bone may be useful. The panel of xenografts available make excellent models for molecular and genetic analysis, gene discovery, and for testing new therapies. Hypotheses generated by experimentation with xenografts can be correlated with clinical data and also tested in transgenic and knockout models to increase our ability to prevent and control prostate cancer. Similar to the relevance of a broad panel of xenograft models, it is essential to establish various metastatic sublines, which follow the preferential spread to bone as observed in the patient. Metastatic model systems will enable us to study the requirements for tumor cells to metastasize and grow in several organs including bone and hopefully lead to therapies targeting this process. The combination of multiple 102 resources and models should lead to advances in our ability to prevent and treat prostate cancer. 103 Literature cited Achanzar, W.E., Diwan, B.A., Liu, J ., Quader, S.T., Webber, M.M., and Waalkes, M.P.: Cadmium-induced malignant transformation of human prostate epithelial cells. Cancer Research 61:455-458, 2001. An, Z., Wang, X., Geller, J ., Moossa, A.R., and Hoffman, R.M.: Surgical orthotopic implantation allows high lung and lymph node metastatic expression of human prostate carcinoma cell line P03 in nude mice. Prostate 34: 169-174, 1998. Bello, D., Webber, M.M., Kleinman, H., Wartinger, DD, and Rhim, J .S.: Androgen responsive adult human prostatic epithelial cell lines immortalized by human papillomavirus 18. Carcinogenesis 18:1215-1223, 1997. Bosma, G.C., Custer, RF. and Bosma, M.J.: A severe combined immunodeficiency mutation in the mouse. Nature 301:527-530, 1983. Bubendorf, L., Schopfer, A., Wagner, U., Sauter, G., Moch, H., Willi, N, Gasser, T.C., and Mihatsch, M.J.: Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Human Pathology 31:578-583, 2000. Devi, G.R., Oldenkamp, J .R., London, C.A., and Iversen, P.L.: Inhibition of human chorionic gonadotropin B-subunit modulates the mitogenic effect of c-myc in human prostate cancer cells. Prostate 53:200-210, 2002. Fu, X., Herrera, H., and Hoffman, R.M.: Orthotopic growth and metastasis of human prostate carcinoma in nude mice after transplantation of histologically intact tissue. (Letter to the editor) International Journal of Cancer 52:987-990, 1992. Groman, E.V. and Reinhardt, C.P.: Method to quantify tail vein injection technique in small animals. Contemporary Topics 43:35-38, 2004. Holleran, J.L., Miller, C.J., Edgehouse, N.L., Pretlow, T.P., and Culp, L.A.: Differential experimental rnicrometastasis to lung, liver, and bone with lacZ-tagged CWR22R prostate carcinoma cells. Clinical & Experimental Metastasis 19:17-24, 2002. Hughes, HO and Lang, C.M.: Basic principles in selecting animal species for research projects. Clinical Toxicology 13:611-621, 1978. Kozlowski, J .M., Fidler, I.J., Campbell, D., Xu, 2., Kaighn, E.M., and Hart, I.R.: Metastatic behavior of human tumor cell lines grown in the nude mouse. Cancer Research 44:3522-3529, 1984. 104 Lim, D.J., Liu, X., Sutkowski, D.M., Braun, E.J., Lee, C., and Kozlowski, J .M.: Growth of an androgen-sensitive prostate cancer cell line, LNCaP, in nude mice. Prostate 22:109-118, 1993. Linehan, W.M., Long, J .P., Steeg, PS, and Gnarra, J .R.: Metastatic models and molecular genetics of prostate cancer. Journal of the National Cancer Institute 84:914-915, 1992. Navone, N .M., Logothetis, C.J., von Eschenback, A.C., and Troncoso, P.: Model systems of prostate cancer: uses and limitations. Cancer and Metastasis Reviews 17:361-371, 1999. Paulsen, D.F.2 Lymphoid system. In: Histology & cell biology: examination & board review, fourth edition. New York, McGraw-Hill, pp.163-180, 2000. Pettaway, C.A., Pathak, S., Greene, G., Ramirez, E., Wilson, M.R., Killion, J .J ., and Fidler, 1.1.: Selection of highly metastatic variants of different human prostatic carcinomas using orthotopic implantation in nude mice. Clinical Cancer Research 221627-1636, 1996. Rembrink, K., Romijn, J .C., van der Kwast, T.H., Rubben, H., and Schroder, F.H.: Orthotopic implantation of human prostate cancer cell lines: a clinically relevant animal model for metastatic prostate cancer. Prostate 31:168-174, 1997. Rhim, J .S., Webber, M.M., Bello, D., Lee, M.S., Amstein, P., Chen, L., and Jay, G.2 Stepwise immortalization and transformation of adult human prostate epithelial cells by a combination of HPV-18 and v-Ki-ras. Proceedings of the National Academy of Sciences USA. 91: 1 1874-1 1878, 1994. Rygarrd, J. and Povlsen, C.O.: Athymic (nude) mice. In: The mouse in biomedical research, volume IV. Foster, H.L., Small, DJ. and Fox, J .G. (eds), New York, Academic Press, pp. 51-67, 1982. Sato, N., Gleave, M.E., Bruchovsky, N ., Rennie, P.S., Beraldi, E., and Sullivan, L.D.: A metastatic and androgen-sensitive human prostate cancer model using intraprostatic inoculation of LNCaP cells in SCID mice. Cancer Research 57:1584-1589, 1997. Sherwood, E.R., Ford, T.R., Lee, C., and Kozlowski, J .M.: Therapeutic efficacy of recombinant tumor necrosis factor 0. in an experimental model of human prostatic carcinoma. Journal of Biological Response Modifiers 9:44-52, 1990. Shevrin, D.H., Gomy, KL, and Kukreja, S.C.: Patterns of metastasis by the human prostate cancer cell line PC-3 in athymic nude mice. Prostate 15: 187-194, 1989. 105 Shevrin, D.H., Kukreja, S.C., Ghosh, L., and Lad, T.E.: Development of skeletal metastasis by human prostate cancer in athymic nude mice. Clinical & Experimental Metastasis 6:401-409, 1988. Stephenson, R.A., Dinney, C.P.N., Gohji, K., Ordonez, N .G., Killion, J .J ., and Fidler, I.J.: Metastatic model for human prostate cancer using orthotopic implantation in nude mice. Journal of the National Cancer Institute 84:951-957, 1992. 8005, G., Jones, R.F., Haas, GP, and Wang, C.Y.: Comparative intraosseal growth of human prostate cancer cell lines LNCaP and PC-3 in the nude mouse. Anticancer Research 17:4253-4258, 1997. Trikha, M., Raso, E., Cai, Y., Fazakas, Z., Paku, 8., Porter, A.T., Timar, J ., and Honn, K.V.: Role of aHbB3 integrin in prostate cancer metastasis. Prostate 35: 185-192, 1998. van Weerden, W.M. and Romijn, J .C.: Use of nude mouse xenograft models in prostate cancer research. Prostate 43:263-271, 2000. Wang, X., An, Z., Geller, J ., and Hoffman, R.M.: High-malignancy orthotopic nude mouse model of human prostate cancer LNCaP. Prostate 392182-186, 1999. Wang, M. and Stearns, M.B.: Isolation and characterization of PC-3 human prostatic tumor sublines which preferentially metastasize to select organs in SCID. mice. Differentiation 482115-125, 1991. Ware, J .L., Lieberman, A.P., Webb, K.S., and Vollmer, R.T.: Factors influencing phenotypic diversity of human prostate carcinoma cells metastasizing in athymic nude mice. Exploratory Cell Biology 532163-169, 1985. Waters, D.J., Sakr, W.A., Hayden, D.W., Lang, C.M., McKinney, L., Murphy, G.P., Radinsky, R., Ramoner, R., Richardson, RC, and Tindall, D.J.: Workgroup 4: spontaneous prostate carcinoma in dogs and nonhuman primates. Prostate 36:64 67, 1998. Webber, M.M., Quader, S.T.A., Kleinman, H.K., Bello-DeOcampo, D., Storto, P.D., Bice, G., DeMednonca-Calaca, W., and Williams, D.E.2 Human cell lines as and in vitro/in viva model for prostate carcinogenesis and progression. Prostate 47:1 13, 2001. Yonou, H., Yokose, T., Kamijo, T., Kanomata, N., Hasebe, T., Nagai, K., Hatano, T., Ogawa, Y., and Ochiai, A.: Establishment of a novel species- and tissue-specific metastasis model of human prostate cancer in humanized non-obese diabetic/severe combined immunodeficient mice engrafted with human adult lung and bone. Cancer Research 61:2177-2182, 2001. 106 CHAPTER 5 TETRACYCLINES: APPLICATIONS IN INHIBITION OF TUMOR PROGRESSION AND METASTASIS 107 Abstract Tetracycline analogs or chemically modified tetracyclines (CMTs) are potential compounds for preventing prostate cancer progression and metastasis. They have been shown to inhibit cell proliferation and invasion through Matrigel, in the in vitro invasion assay, of several prostate cancer cell lines, as well as cause a decrease in matrix metalloproteinase (MMP) production and activity in vitro. MMPs are important enzymes involved in prostate cancer progression and metastasis. A possible cytotoxic mechanism of CMTs may include the induction of programmed cell death, but the most important feature of CMTs for cancer treatment is their ability to inhibit MMPs. In viva, CMTs inhibit tumor incidence, tumor growth, and metastasis to the lungs in Copenhagen rats given subcutaneous injection of MAT LyLu, Dunning rat prostate cancer cells. In rats given an intravenous cell injection of MAT LyLu tumor cells and treated daily by gavage with CMT-3, both an increase in survival and a decrease in metastasis were observed. In a phase I clinical trial of CMT-3, patients with advanced refractory metastatic cancers were given a daily dose of CMT-3. Disease stabilization continued for more than 61 days in patients with certain malignancies, including sarcomas and a metastatic Sertoli-Leydig cell tumor of the ovary. These results suggests that additional screening of CMT compounds could lead to the identification of compounds which show greater efficacy in the treatment of prostate cancer. Keywords Chemically modified tetracyclines, invasion, matrix metalloproteinases, metastasis, tetracylines 108 Introduction Metastatic spread of cancer, a major obstacle in curing cancer, remains to be overcome. Therefore, an increased understanding of the steps that take place during metastasis as well as the design and use of therapeutic strategies to inhibit these steps are needed. Once class of molecules that may play a role in the process of metastasis are, matrix metalloproteinases (MMPs). Therefore, control of MMP activity has generated considerable interest as a possible target to inhibit tumor progression, invasion, and metastasis (Chambers and Matrisian, 1997; Wojtowicz-Praga et al., 1997). One group of compounds, tetracycline antibiotics, which have long been recognized as useful adjuncts in the treatment of periodontal diseases, have been shown to inhibit MMP activity, and therefore, may be useful for inhibiting tumor progression (Golub et al., 1991). Steps in the process of metastasis, where MMPs are thought to be involved, include the following: escape of cells from the primary tumor, intravasation (entry of cells into the blood or lymphatic circulation), survival and transport in the circulation, arrest in distant organs, extravasation (escape of cells from the circulation), and growth of cells to form secondary tumors in the new organ environment (Chambers and Matrisian, 1997; Fidler, 1991; Liotta et al., 1993). Tumor cells then interact with the stromal components of the new organ, which results in either the elimination of tumor cells or their colonization due to stimulation of cell proliferation and angiogenesis (Lokeshwar et al., 1999). MMPs can be broadly subdivided into three classes based on their substrate specificity: collagenases, stromelysins, and gelatinases (W ojtowicz-Praga et al., 1997). MMPs are secreted as zymogens and upon activation degrade basement membrane 109 components and facilitate tissue destruction. Their activity is modulated by proenzyme activation and expression of their inhibitors, tissue inhibitors of matrix metalloproteinases (TIMPs). Activation of MMPs requires the proteolytic removal of the pro-domain located at its amino terminus end. This cleavage breaks the cysteine/zinc ion interaction and instead allows water molecules to interact with zinc, which is necessary for activation (Chu, 1998). MMPs may be activated by other proteolytic enzymes, such as uPA and other MMPs (Fridman et al., 1995; Webber and Waghray, 1995). Although both normal and neoplastic cells produce MMPs and other proteinases, only malignant cells are invasive (Lokeshwar et al., 1993). Localized degradation of ECM occurs where the expression of active proteolytic enzymes is higher than their natural inhibitors, TlMPs. An imbalance of secretion between MMPs and TIMPs in prostatic carcinoma has been observed (Lokeshwar et al., 1993). TlMPs were not detected in conditioned medium from primary prostate carcinoma and activated gelatinases were not detected in the conditioned medium from normal adult prostate explants. An increased expression of the gelatinases, MMP-2 and MMP-9, whose substrate is type IV collagen, have been shown to be associated with malignant progression of prostate cancer (Liotta et al., 1977; Liotta et al., 1980; Webber et al., 1996). Studies of human prostate tumor tissue have shown that levels of both MMP-2 and MMP—9 are low in normal prostate and organ-confined tumors with Gleason sum of 5 or lower, whereas they were highly expressed in tumors with Gleason sum of 8-10 (Wood et al., 1997). Of significant interest is the increased expression of activated MMP-2 in prostate cancer progression (Lokeshwar et al., 1993; Stearns and Stearns, 1996). 110 Anti-MMP or anti-collagenase activity of tetracyclines makes these molecules useful for the treatment of cancer but also other diseases where tissue destruction takes place, such as, rheumatoid arthritis, skin lesions, corneal ulcers, and periodontitis (Golub et al., 1991). Most prostate cancer patients initially undergo some type of androgen ablation treatment to slow tumor growth. However, deprivation of androgens to prostate almost always leads to the onset of a more aggressive, metastatic, hormone-refractory incurable phase of the disease (Newling, 1996). To treat the metastatic phase of the disease, drugs that inhibit the metastatic process and do not discriminate between androgen-sensitive and androgen-insensitive prostate tumor cells are needed (Lokeshwar et al., 1999). Inhibitors of MMP activity, such as tetracycline analogs, are drugs with such a potential. Following administration, tetracyclines distribute widely throughout the body and into tissues and secretions, including the prostate and urine, however their half life is in the range of 6-12 hours (Kapusnik-Uner et al., 1995). Besides the need for continuous administration, other limitations of tetracyclines include, antibiotic resistance and toxicity as a result of long term exposure. Therefore, the tetracycline molecule has been chemically modified in multiple ways, generating a new family of compounds called CMTs (chemically modified tetracyclines). The derivation of some of the CMT drugs, as well as, results of some studies testing the effects of CMTs on prostate cancer will be further discussed. 111 Chemical modifications of the tetracycline molecule: One of the first modifications to the tetracycline molecule involved removal of the dimethylamino group from the carbon-4 position of the “A” ring, resulting in the CMT called 4-de-dimethylaminotetracycline (CMT-l) (Figure 5.1). This modification eliminated the drug’s antimicrobial activity but did not reduce the ability of the drug to block the activity of collagenases (Golub et a1. 1991). 112 N(CH3)2 Tetracycline Molecule Group(s) Removed Position(s) from Tetracyline CMT-1 -N(CH3)2 (4) CMT-3 -N(CH3)2; CH3(OH) (4; 6) CMT-8 -N(CH3)2; OH (4; 6) Figure 5.1 A schematic representation of tetracycline and the chemical modifications of tetracycline that generated the CMT-l, CMT-3, and CMT-8 compounds (Modified from Seftor et al., 1998). Also shown in Figure 5.1, are the chemical structures of subsequent CMTs along with tetracycline. Each of the CMTs were shown to inhibit collagenase activity in vitro, however, when the carb0nyl oxygen at carbon 11 and the hydroxyl group at carbon 12 were removed from tetracycline to produce CMT-5 (pyrazole derivative), the collagenase-inhibitory activity of the molecule was lost (Golub et al., 1991). Thus, the two side chains at carbon 11 and carbon 12, are considered to be involved with anticollagenase activity. 113 CMT-3 inhibits cell proliferation: The effect of CMTs on in vitro cell proliferation of prostate cancer cells varied greatly (Lokeshwar et al., 2002). In the three cell lines tested, LNCaP, TSU-PRl, and MAT LyLu, all CMTs, except CMT—7, were significantly cytotoxic. CMT-3 was the most cytotoxic tetracycline analogue tested. Therefore, most prostate cancer studies only test the effects of CMT-3. In another study, following a 48 hour incubation period with a I range of concentrations of doxycycline or CMT-3, inhibition of cell proliferation of PC-3, DU145, and MAT LyLu cells was dose dependent (Figure 5.2) (Lokeshwar, 1999). CMT-3 was significantly more potent than doxycycline; a 5-fold lower concentration of CMT-3 compared to doxycycline was needed to decrease cell proliferation by 50%. 114 120 DU 145 +CMT-3 -r-DC ‘1-1.‘ .L A V Y' 1 Cell proliferation % of control 60 ‘ 0.1 - 10 . 10.0 100.0 Cell proliferation I % of control 60" 0 L , , if 2 0.1 1.0 10.0 100.0 Drugs, rig/ml l MAT LyLu --CMT-3 ‘ ' cur-DC 120 Cell proliferation % of control 6" ‘. V v v V' V vv 0.1 ' V ' 1:0 10.0 100.0 Drugs, lug/ml Figure 5.2. Effect of doxycycline (DC) and CMT-3 on proliferation of prostate tumor cell lines. Tumor cells were incubated with various concentrations of DC or CMT-3 for 48 hours in complete culture medium. Cell proliferation activity, defined as synthesis of [3H]-thymidine-labeled DNA, was assayed by 2-hour pulse-labeling the cells with [3H]-thymidine as described in the text. Data presented are for three prostate cancer cell lines. Similar results were obtained for other cell lines. Vertical bars represent mean i SEM from four independent determinations (Lokeshwar, 1999). 115 In several prostate cancer cell lines, CMT-3 was found to be 8-fold more effective than doxycycline at inhibiting cell survival in vitro (Table 5.1). CMT-3 was also 10-fold more effective at inhibiting clonogenic survival of two human prostate cancer cell lines (Lokeshwar, 1999). The possible mechanisms of CMT-3 induced toxicity will be further discussed. Table 5.1 Cytotoxicity of DC and CMT-3 in Prostate Cells 61502 Cell line' DC CMT-3 ALVA 101 (4) 16.67 : 1.3b 3.1 x 0.34 BPH-l (3) 9.68 z 2.45 4.78 i 1.68 CaP 139(1) 18.7137 9.3:2.11 DU 145 (8) 19.8 i 4.25 2.3 i 0.53 LNCaP (5) 6.3 i 1.35 2.29 :1: 0.96 MAT Lylu (7) 9.09 i 2.95 2.36 1 0.86 PC-3 (5) 16.55 i 1.06 4.8 t 0.96 TSU PR-l (5) 18.64 25.1 6.7 $1.2 1 Numbers of replicate experiments are given in parentheses. 2 Growth inhibition was calculated from linear regression of the dose-response curves generated for each experiment using log (dose) vs. cell proliferation (% of control). Correlation coefficient (r) was always 20.95 (negative). Results are presented as mean i SEM ug/ml (1 rig/ml = 2.2 pm) of at least 3 G150 values calculated from each experiment. (Lokeshwar et al., 2002). 116 Possible mechanisms of CMT-3 induced cytotoxicity: Many anti-tumor drugs which inhibit cell proliferation also induce apoptosis or programmed cell death (Lokeshwar, 1999). This has been shown with CMT-3. Culture media from cells incubated with various concentrations of CMT-3 were assayed for . soluble nucleosomes resulting from apoptosis (Lokeshwar et al., 1998). CMT-3 was found to induce apoptosis in >80% of the cells in all seven prostate cancer cell lines tested and induction of apoptosis was both dose and time dependent (Lokeshwar et al., 1998; Lokeshwar, 1999). Only at 5-10 fold higher concentration was doxycycline able to induce similar levels of apoptosis as CMT-3 treated cells. In another study the minimum incubation time required for CMT-3 to induce apoptosis was similar to the time range of maximum expression in cellular [OH'] and detectable depolarization of the mitochondria (Lokeshwar et al., 2002). These results suggest that CMT-3-induced apoptosis is associated with production of free radicals and depolarization of the mitochondria. In fact mitochondrial depolarization is frequently observed in cells undergoing apoptosis (Lemasters et al., 1998). In addition to the induction of apoptosis, cell cycle progression was blocked in prostate cancer cell cultures treated with CMT-3 (Lokeshwar et al., 2002). A significant increase in the accumulation of cells at the Go/Gl phase was observed; from ~50% in the control cell population to up to 85% in cells treated with CMT-3. Similarly, a decrease in the S-phase fractions was observed in PC-3, DU145 and LNCaP, which is indicative of the inhibition of cell cycle regulators (Lokeshwar et al., 2002). The molecular mechanisms of the cytotoxic effects of CMT-3 are still under study. Another proposed 117 mechanism to explain the ability of CMT-3 to inhibit cell growth, invasion and metastasis may be due to inhibition of MMPs. CMT-3 decreases MMP production: MMPs are important enzymes involved in prostate cancer progression and metastasis. It has been previously suggested that tetracycline’s inhibitory effect on MMPs may involve the drug’s well-known ability to bind metal ions like zinc, which are required by the MMPs to maintain their prOper conformation and hydrolytic activity (Golub et al., 1983 and 1991). In support of this, addition of excess zinc has been shown to overcome the inhibition of gelatinases and collagenase by doxycycline (Lee et al., 1992; Yu et al., 1992). The amount of MMP synthesized and secreted by prostate cancer cells is lower in monolayer cultures treated with CMT-3 as compared to untreated cells (Lokeshwar et al., 2002). MP activity in serum-free conditioned medium from drug-treated rodent and human prostate cancer cells, MAT LyLu and TSU-PRl respectively, were analyzed by zymography (Lokeshwar et al., 2002). TSU-PRl cultures predominantly secreted latent forms of MMP-2 and MMP-9 while MAT LyLu cells secreted activated MMP-2 (62kDa) but little MMP-9 (Figure 5.3). 118 MMP-9 MMP-2 015102050 015102050 CMT3 (pg/ml) Doxycycline (pg/ml) MMP-9 MMP-2 0 1 5 10 20 50 0 1 5 10 20 50 CMT3 (pg/ml) Doxycycline (pg/ml) Figure 5.3 Zymographic detection of gelatinases secreted into the conditioned media from cultures treated with CMT-3 or doxycycline. Culture conditioned media (15 til/lane, equivalent to 5 x 103 cells) from TSU-PRl (a,b) and MAT LyLu (c,d) cells were separated by SDS-PAGE (8% polyacrylamide) on a gelatin- embedded (1 mg/ml) gel and zymography. The positions of purified MMP-2 and MMP-9 are indicated. Note: the major fraction of MMP-2 from MAT LyLu (bottom) cell conditioned media was active (Mr ~64,000), whereas most TSU-PRl (top) MMP-2 was in the latent form (Mr 72,000) (Lokeshwar et al., 2002). Incubation of TSU-PRl and MAT LyLu cell lines with CMT—3 decreased the secretion of MMPs in both cell lines in a dose-dependent manner (Lokeshwar et al., 2002). Similar to other results, CMT-3 was more effective than doxycycline; cells treated with CMT-3 secreted significantly less MMP than the cells treated with doxycycline. In this experiment treatment of prostate cancer cells with CMT-3 resulted in a decrease in the amount of both MMP-2 and MMP-9. 119 To establish further that the observed decreases in MMP levels in conditioned medium were indeed due to the drug-induced inhibition of MMP production/secretion, protein levels of MMPs in CaP 139 cells were measured by an enzyme-linked irnmunosorbent assay (ELISA) that uses a monoclonal anti-MMP-2 antibody (Lokeshwar et al., 2002). CaP 139 cells were derived from a primary human prostate tumor. A dose- dependent decrease in the levels of secreted MMP-2 was observed by the CaP 139 cells treated with CMT-3 or doxycycline. In CaP 139 cells MMP-2 levels decreased by 51% and 74% at 20 rig/ml doxycycline and 10 rig/ml CMT-3, respectively. ELISA kits were also used to measure the protein levels of the natural inhibitors of MMPs, TIMPs, in Ca 139 cells treated with CMT-3 (Lokeshwar et al., 2002). The decreases in TIMP-1 and TIMP-2 levels were 33% and 10.27%, respectively at 10 ug/ml CMT-3. These data show that both TIMP-1 and TEMP-2 levels were much less inhibited by CMT-3 than MMP-2. This suggests that CMT-3 reduces invasive activity of tumor cells, not only by inhibiting MMP synthesis and secretion, but also by not affecting TIMP levels significantly. The lack of total inhibition of MMP activity in tumor cells by CMT-3 could be due to the production of other proteinases, which are capable of activating MMPs, but are not inactivated by CMTs (Lokeshwar et al., 2002). One such proteinase is urokinase-type plasminogen activator (uPA) (W ebber and Waghray, 1995). It has been reported that CMTs do not inhibit uPA secretion or activity (Chang et al., 1996). 120 CMTs inhibit Matrigel invasion: The invasive activity of two human prostate cancer cell lines, PC-3 and DU145, and one rodent prostate cancer cell line, MAT LyLu, were assayed following 48 hours of exposure to 5 ug/ml of each CMT (CMT-l, 2, 3, 4, 6, 7 and 8) or doxycycline. The percentage of cells that invaded through Matrigel-coated filters in the control wells were used to determine percent inhibition of Matrigel invasion in treated wells for each cell line tested. As shown in Figure 5.4, CMT-3 was the most potent inhibitor of invasive activity of P03, DU145 and MAT LyLu cells while CMT-6 was the least potent inhibitor (Lokeshwar, 1999). 121 ..L O O C O .- o- g 2:. E 00145 > 80' 5:5; , I: 5:? '5; 2 @PCJ .- ' v i '5 3:33 :2 = D :-: :55: :52 E m MAT LYLU It ‘ 1'. 504 1:3- I; .. a: : '6' 35' 53- E 2 E -2 .- I :v 2 is 2:: E- : § 9 no 0 .21 - it? - c 40* 2;. '=' = o 5;; 333. E i: . if: 2 :32; a ,2 :5; .9 2:51 5 ii : 3:3 .‘. 'm- .‘.‘ I: . - J. 'c 20.. -. f2 :121 E T-‘5 - = ;. :3: c :5. - a: 0:0: - 3 = - ‘4; ,:' "" = 513' = “‘ 321 E g E E 3:43 333 ‘Q = :52 - _ 3:1: 5 :2 :3:3 :1 E5 .. , = E: = 2 2;: O .. - - .. .. :- .., .. .. I- . .. - p 0- :- .'.' - 4.4 ~v ,. . — c - ". - u 0 - H - .- .. - '3. - .. - L“ " ‘.' - W ~ '.' - "' 4': - ‘l '-'- - ’t‘ '.'. - 4.. - '7: 3:1 ’.' - TI; - " ‘3 - 4‘ I.“ — 1:. -.~. = jg u‘u. - :’ - 1.. 0‘. .... .n - a; — ‘1‘. a. 0‘ 3 f 32?: E .11 .712: g I: :21: E if; .:. 2:". T 1:3: __ iii 123 = E: 2-: E ii. 3:3 DC CMT-1 CMT-2 CMT-3 CMT-4 CMT-8 CMT-7 CMT-8 Drugs, 5p glml Figure 5.4 Inhibition of invasive potential of tumor cells by doxycycline (DC) and CMTs. Invasion of tumor cells through the Matrigel-coated filters was assayed following 48 hours of exposure to 5 rig/ml of each drug. Only the drug diluent (0.1% dimethyl sulfoxide) was added to control wells. Percentage of cells that invaded in the control (0.1% DMSO) wells varied from 12.5 :t 6.4% for DU145 cells to 17 :1: 4.2 for MAT LyLu cells. 0.1% DMSO had negligible effect on invasion. Results presented are from three independent experiments. (Lokeshwar, 1999). Doxycycline and CMT-2 were minimally effective as inhibitors of invasion of the Dunning rat MAT LyLu prostate cancer cells. These results show that CMT-3 is not only the most cytotoxic CMT but it is also the most effective inhibitor of Matrigel invasion by prostate cancer cells. CMT-3 inhibits Dunning tumor growth and metastasis: The androgen-insensitive Dunning rat MAT LyLu cell line was chosen for in viva studies in Copenhagen rats. These cells are highly tumorigenic upon subcutaneous 122 injection of as low as 5X104 cells and metastasize to the lymph nodes and lungs only 12 days post cell injection (Isaacs et al., 1986). In rats bearing subcutaneous tumors and given treatment (40 mg/kg) for 21 days with a daily oral gavage of CMT-3 or doxycycline, tumor incidence and tumor growth rate were only significantly reduced in the CMT-3 gavage-fed group (Lokeshwar, 1999). A regression or disappearance of palpable tumor was observed in CMT-3 treated groups, but not in the control or doxycycline treated groups. Spontaneous metastasis to the lungs was reduced significantly in groups treated with doxycycline or CMT-3, the number of metastatic foci were reduced to 49.7% and 41.2% of control, respectively. These results could only be obtained if the tumor cell inoculum was lowered from 1X106 cells/site to 1X105 cells/site. When rats were given subcutaneous injections of 1X106 cells/site, tumor incidence and tumor growth were not affected by oral administration of CMT-3 or doxycycline. In this experiment, rats began treatment with the test agent (CMT-3 or doxycycline) on the same day as cell injection. In another study, pre-dosing tumor-bearing rats with CMT-3 at 40 mg/kg with daily gavage for 7 days and using the tumor cell inoculum of 2X105 cells/site resulted in a remarkable reduction in tumor incidence and significant tumor remission (Lokeshwar et al., 2002). More than 90% of the rats in control and doxycycline-treated groups developed tumors in 3 independent experiments. In contrast, the incidence of rats developing tumors in CMT-3-treated groups was significantly lower (55 i 9%) than that for control or doxycycline-treated groups. In two separate experiments tumor regression was observed in 20% and 30% of CMT-3-treated groups. This enhanced efficacy of 123 CMT-3 upon pre-dosing suggests that CMT-3 treatment will be effective if administered prior to the appearance of clinical signs. In rats given an intravenous cell injection of MAT LyLu tumor cells (5X104), daily treatment by gavage with CMT-3 resulted in both an increase in survival and a decrease in skeletal and soft tissue metastasis (Lokeshwar, 1999; Selzer et al., 1999). Treatment with CMT-3 also resulted in a delayed onset or a total lack of paraplegia following intravenous cell injection compared to control animals. As observed in the subcutaneous rat model of prostate cancer, CMT-3 was most effective when fed to rats by gavage beginning several days prior to cell injection. Several in viva models of aggressive prostate cancer described above have demonstrated the efficacy of CMT-3 against tumor incidence, tumor growth, and tumor metastasis to soft or skeletal tissue. In both subcutaneous and intravenous rat models of prostate cancer, treatment with CMT-3 by gavage did not have any adverse effects on the animals indicating its safe nature. The enhanced efficacy of CMT-3 upon pre-dosing and oral bioavailability, with minimal adverse reactions within a tolerable dose, suggests that CMT-3 could be used as an adjuvant to hormone ablation or radiation therapy in prostate cancer. Phase I clinical trial of CMT-3: Besides prostate cancer, CMTs have also been shown to exhibit in vitro and in viva anti-tumor invasion and metastasis potential in many aggressive types of tumor, including breast cancer and melanoma (Meng et al., 2000; Seftor et al., 1998). Because of its interesting mechanism of action and potent preclinical activity, COL-3 or CMT-3 124 was entered into phase 1 testing (Rudek et al., 2001; Lokeshwar et al., 2002). A study conducted by the Investigational Drug Branch, Cancer Therapy Evaluation Program at the National Cancer Institute evaluated maximum tolerated dose and dose-limiting toxicities of CMT-3 in patients with refractory solid tumors (Rudek et al., 2001). In this phase I clinical trial of CMT-3, patients with advanced refractory metastatic cancer were given a daily dose of CMT-3. Eight patients had stable disease at the first 2-month follow-up and continued on-study for more than 61 days. One patient with hemangioendothelioma experienced disease stabilization for more than 26 months. This patient had not received any prior cytotoxic chemotherapy, suggesting that MMP inhibitors are more effective if given early in the course of treatment. Of seven patients with tumors of nonepithelial origin, three (43%) showed some degree of clinical benefit from COL-3. These patients had disease stabilization for more than 6 months and included three women with hemangioendothelioma, metastatic Sertoli-Leydig cell tumor of the ovary and fibrosarcoma metastatic to the lung. Those patients that demonstrated disease stabilization, were also shown to have a significant reduction in plasma MMP-2 levels. MMP-2 levels decreased with increasing cumulative dose of COL-3 in many of the patients with drug-induced toxicity to a greater degree than the reduction seen in patients with stable disease. The reason for this association and the mechanism by which COL-3 inhibits the production of MMP-2 is not clear. It is also not clear whether MMP-2 production was from vascular endothelial cells, tumor cells, or both. The major dose-limiting toxicity of CMT-3 was photosensitivity. Drug-induced phototoxicity was observed in 40-70% of the patients receiving CMT-3 at a dose greater than or equal to 70 mg/mZ/day. Based on the results of this study a daily dose of 36 mg/m2 was 125 recommended for a phase II clinical trial. However, a dose of 70 mg/mz/day may be considered if diligent sun precautions are used. Conclusions CMTs could be an effective therapy for prostate cancer. They have been shown to inhibit cell proliferation and Matrigel invasion of several prostate cancer cell lines, as well as, cause a decrease in matrix metalloproteinase (MMP) production and activity in vitro. A decrease in the incidence of metastasis was observed in several different rat tumor models of prostate cancer where animals received a daily dose of CMT-3. Results obtained from many in vitro and in viva cancer studies using CMT-3 support the concept that the ability of CMTs to inhibit MMPs is an effective approach to reducing tumor growth and metastasis. Most important is the ability of CMTs to inhibit the production and activity of MMP-2 because increased expression of MMP-2 is of significant interest in prostate cancer progression. Combined with their cytotoxic property and little systemic toxicity, CMTs may have great potential as anticancer drugs. Therefore, the screening of additional CMTs and future clinical trials may be necessary to identify compounds which show greater efficacy in the treatment of prostate cancer. 126 Literature cited Chambers, A.F., and Matrisian, L.M.z Changing views of the role of matix metalloproteinases in metastasis. Journal of the National Cancer Institute 89:1260-1270, 1997. Chang, K., Rani, A.S., Chang, K.: Plasminogen activator activity is decreased in rat gingival during diabetes. Journal of Periodontal Research 672743-747, 1996. Chu, W.: Matrix metalloproteinases in tissue remodeling, regeneration and prostate cancer. Masters of Science Thesis, Michigan State University, East Lansing, MI, USA. pp.55-70, 1998. Fidler, 1.] .2 The biology of human cancer metastasis. Acta Oncologica 302668-675, 1991. Fridman, R., Toth, M., Pena, D., and Mobashery, 5.: Activation of progelatinase B (MMP-9) by gelatinase A (MMP-2). Cancer Research 55:2548-2555, 1995. Golub, L.M., Ramamurthy, N .S., and McNamara, T.F.: Tetracyclines inhibit connective tissue breakdown: new therapeutic implications for an old family of drugs. Critical Reviews in Oral Biology and Medicine 22297-322, 1991. Golub, L.M., Lee, H.M., Lehrer, G., Nemiroff, A., McNamara, T.P., Kaplan, R. and Ramamurthy, N .S.: Minocycline reduces gingival collagenolytic activity during diabetes: preliminary observations and a proposed new mechanism of action. Journal of Periodontal Research 182516-526, 1987. Isaacs, J .T., Isaacs, W.B., Feitz, W.F. and Scheres, J .: Establishment and characterization of seven Dunning rat prostatic cancer cell lines and their use in developing methods for predicting metastatic abilities of prostatic cancers. Prostate 9:261 -281, 1986. Kapusnik-Uner, J .E., Sande, M.A., and Chambers, H.F.2 Tetracyclines, chloramphenicol, erythromycin, and miscellaneous antibacterial agents. In: Goodman & Gilman’s The Pharmacological Basis of Therapeutics, Wonsiewicz, M.J. and McCurdy, P. (eds.), New York, The McGraw-Hill Companies, Inc., 9th Editionzl 123-1130, 1995. Lee, H.M., Hall, J .E., Sorsa, T., Simon, S. and Golub.L.M.: Doxycycline inhibits PMN -mediated tissue breakdown in culture. Journal of Dental Research (special issue) 71:245, 1992. 127 Lemasters, J.J., Nieminen, A.L., Quian, T., Trost, L.C., Elmore, S.P., Nishimura, ., Crowe, R.A., Cascio, W.E., Bradham, C.A., Brenner, D.A., and Herman, B.: The mitochondrial permeability transition in cell death: a common mechanism in necrosis, apoptosis and autophagy. Biochimica et Biophysica Acta 13662177 -196, 1998. Liotta, L.A., Kleinerman, J ., Catanzaro, P., and Rynbrandt, D.: Degradation of basement membrane by murine tumor cells. J ounal of the National Cancer Institute 58:1427-1431, 1977. Liotta, LA. and Stetler-Stevenson, W.G.: Priniciples of molecular cell biology of cancer: cancer metastasis. In: Cancer: principles and practive of oncology. DeVita, V.T., Hellman, S., and Rosenberg, S.A. (eds.), Philidelphia, Lippincott, 134-149, 1993. Liotta, L.A., Tryggvason, K., Garbisa, 5., Hart, 1., Foltz, C.M., Shafie, S.: Metastatic potential correlates with enzymatic degradation of basement membrane collagen. Nature 284267-68, 1980. Lokeshwar, B.L., Selzer, M.G., Zhu, B., Block, N .L. and Golub, L.M.2 Inhibition of cell proliferation, invasion, tumor growth and metastasis by an oral non-antimicrobial tetracycline analog (COL-3) in a metastatic prostate cancer model. International Journal of Cancer 982297-309, 2002. Lokeshwar, B.L.: MMP inhibition in prostate cancer. Annals New York Academy of Sciences 878:271-289, 1999. Lokeshwar, B.L., Houston-Clark, H.L., Selzer, M.G., Block, N.L., and Golub, L.M.2 Potential application of a chemically modified non-antimicrobial tetracycline (CMT-3) against metastatic prostate cancer. Advances in Dental Research 12:97 -102, 1998. Lokeshwar, B.L., Selzer, M.G., Block, N .L., and Gunja-Smith, Z.2 Secretion of matrix metalloproteinases and their inhibitors (tissue inhibitor of metalloproteinases) by human prostate in explant cultures: reduced tissue inhibitor of metalloproteinase secretion by malignant tissues. Cancer Research 53:4493-4498, 1993. Meng, Q., Xu, J ., Goldberg, I.D., Rosen, E., Greenwald, R., and Fan, 8.: Influence of chemically modified tetracyclines on proliferation, invasion, and migration properties of MDA-MB-468 human breast cancer cells. Clinical & Experimental Metastasis 182139-146, 2000. Newling, D.W.2 The management of hormaone refractory prostate cancer. European Journal of Urology 29:69-74, 1996. 128 Rudek, M.A., Figg, W.B., Dyer, V., Dahut, W., Turner, M.L., Steinberg, S.M., Liewehr, D.J., Kohler, D.R., Pluda, J .M., and Reed, E.2 Phase I clinical trial of oral COL-3, a matrix metalloproteinase inhibitor, in patients with refractory metastatic cancer. Journal of Clinical Oncology 192584-592, 2001. Seftor, R.B.B., Seftor, B.A., De Larco, J E Kleiner, D.E., Leferson, J ., Stetler-Stevenson, W.G., McNamara, T.F., Golub, L.M., and Hendrix, M.J.C.2 Chemically modified tetracyclines inhibit human melanoma cell invasion and metastasis. Clinical & Experimental Metastasis 162217-225, 1998. Selzer, M.G., Zhu, B., Block, N.L., and Lokeshwar, B.L.: CMT-3, a chemically modified tetracycline, inhibits bony metastases and delays the development of paraplegia in a rat model of prostate cancer. Annals New York Academy of Sciences 8782678- 682, 1999. Stearns, M. and Stearns, M.E.2 Evidence for increased activated metalloproteinase 2 (MMP-2a) expression associated with human prostate cancer progression. Oncology Research 8269-75, 1996. Webber, M.M., and Waghray, A.: Urokinase-mediated extracellular matrix degradation by human prostatic carcinoma cells and its inhibition by retinoic acid. Clinical Cancer Research 1:755-761, 1995. Webber, M.M., Waghray, A., Bello, D., and Rhim, J .S., Mini review: proteases and invasion in human prostate epithelial cell lines: implications in prostate cancer prevention and intervention. Radiation Oncology Investigations 32358-362, 1996. Wojtowicz-Praga, S.M., Dickson, R.B., and Hawkins, M.J .2 Matrix metalloproteinase inhibitors. Investigational New Drugs 15:61-75, 1997. Yu, L.P., Smith, G.N., Hasty, K.A. and Brandt, K.D.2 Doxycycline inhibits type XI collagenolytic activity of extracts from human osteoarthritic cartilage and of gelatinase. Journal of Rheumatology 18:1450-1452, 1992. 129 PART 2 ORIGINAL RESEARCH 130 CHAPTER SIX EVALUATION OF THE EFFICACY OF . CHEMICALLY MODIFIED TETRACYCLINES (CMTs) AS AGENTS FOR THE TREATMENT OF PROSTATE CANCER: A PILOT STUDY USING CMT 2215 131 Abstract Chemically modified tetracyclines (CMTs) may be effective chemotherapeutic agents for prostate cancer. Results obtained from in vitro and in viva studies, in addition to a phase I clinical trial, suggest a potential use for CMTs as an oral, nontoxic drug to treat metastatic prostate cancer and other cancers. Additional screening of CMTs may lead to the identification of compounds which show greater efficacy in the treatment of prostate cancer. Thus, in this chapter, I examine the ability of CMT 2215, to inhibit the growth of two human prostate epithelial cell lines, RWPE2-W99 and CTPE, in viva. RWPE2-W99 forms slow growing tumors when injected into nude mice and mimics the behavior of the majority of primary human prostate cancers, while CTPE forms rapidly growing, aggressive tumors, and represents the more aggressive, late stage of tumor progression. Treatment of RWPE2-W99 cells grown in monolayer cultures with 50 jig/ml 2215 caused ~65% growth inhibition. In viva, I assessed the ability of CMT 2215 to inhibit growth and reduce the size and number of tumors produced after subcutaneous injection of RWPE2-W99 or CTPE cells in immune-suppressed mice. Using the slow growing RWPE2-W99 cells, treatment with CMT 2215 (0.675 mg/ml) caused a decrease in tumor volume, but CMT 2215 (2.25 mg/ml) appeared to have no effect on tumor growth in the aggressive CTPE cells. Based on the results of these in vitro and in viva studies, I selected the RWPE2-W99 model for additional studies using other CMTs. 132 Keywords Chemically modified tetracycline, CTPE, gavage, prostate cancer, RWPE2-W99, subcutaneous, tumor Introduction Chemically modified tetracyclines (CMTs) have been shown to inhibit cell proliferation and Matrigel invasion of several prostate cancer cell lines, as well as, cause a decrease in matrix meta110proteinase production and activity in vitro (Lokeshwar, 1999; Lokeshwar et al., 2002). In male Copenhagen rats, given subcutaneous injection of MAT LyLu cells, treatment with CMT-3 by gavage inhibited tumor incidence and reduced the tumor growth rate (Lokeshwar et al., 1999). In Copenhagen rats given an intravenous injection of MAT LyLu cells, treatment with CMT-3 decreased the frequency of tumor metastasis to soft or skeletal tissue and also resulted in an increase in survival (Lokeshwar, 1999; Selzer et al., 1999). Other CMTs are now being extensively investigated because of their increased efficacy as compared to their natural derivatives. In this study CMT 2215 was tested for its effects on the growth of the tumorigenic RWPE2-W99 and highly aggressive CT PE human prostate cancer cell lines (Achanzar et al., 2004; Achanzar et al., 2001; Hello et al., 1997; Webber et al., 1997a). These two cell lines are related because they share a common origin. The following describes the process by which the RWPE2-W99 cell line was developed. Human prostate epithelial cells were derived from the peripheral zone of a normal human prostate and immortalized with a single copy of human papilloma virus-18 (HPV-18) DNA to give rise to the RWPE-l cell line (Bello et al., 1997; Webber et al., 1997a). RWPE-l cells were then 133 transformed by v-Ki-ras, giving rise to the RWPE-2 cell line (Bello et al., 1997; Webber et al., 1997a). The transformed RWPE-2 cells are tumorigenic and can grow in soft agar in an anchorage-independent manner. In order to select cells that showed high Ki-ras expression, RWPE-2 cells were grown in agar and colonies were screened for Ki-ras expression. One of these colonies was expanded and it gave rise to the RWPE2-W99 cell line. I used this cell line for in vitro studies and for in viva studies to assess the efficacy of CMT 2215 to inhibit tumor growth when RWPE2-W99 cells were grown as a xenograft in nude mice. RWPE2-W99 cells represent an early stage of prostate cancer progression. In addition, I used the CTPE cell line, which represents a more aggressive, rapidly growing tumor. The CT PE cell line was derived from RWPE-l cells by chronic exposure to cadmium, a carcinogen (Achanzar et al., 2001). Both RWPE2-W99 and CTPE cell lines are tumorigenic and CTPE cells have also been shown to be invasive and metastatic following subcutaneous injection in nude mice (Achanzar et al., 2004; Achanzar et al., 2001; Bello et al., 1997; Webber et al., 1997a and b). Since both cell lines are related and mimic different stages of human prostate cancer progression, they are useful for testing agents for the prevention and treatment of prostate cancer. Such agents include chemically modified tetracyclines (CMTs). The objectives of this study were to determine the ability of CMT 2215 to inhibit cell growth in vitro and tumor growth in viva of these two human prostate cancer cell lines. This study was conducted to develop a xenograft model for assessing the efficacy of CMTs for the treatment of prostate cancer. In order to determine dose levels that might be used for in viva studies, the effects of the test agent on the growth of RWPE2-W99 cells were first tested in cell culture. 134 Materials & Methods In vitro studies Cell culture: Both RWPE2-W99 and CTPE cells were grown in complete keratinocyte serum- free medium (KSFM) containing 50 rig/ml bovine pituitary extract (BPE), 5 ng/ml epidermal growth factor (EGF) and 1X antibiotic/antimycotic solution. Cultures were maintained at 370C in a humidified atmosphere containing 5% C02 and subcultured weekly. Dose response using a microplate assay: RWPE2-W99 cells were plated, six replicate wells per treatment, in complete keratinocyte serum-free medium (K-SFM) containing 50 ug/ml bovine pituitary extract (BPE) and 5 ng/ml epidermal growth factor (EGF), 10,000 cells/well in 96-well plates and allowed to attach for 48 h at which time medium was changed to medium containing varying concentrations of the test agent. The test agent, CMT 2215, was dissolved in DMSO. The final concentration of the DMSO vehicle in the culture medium was 0.1%. Treatment groups consisted of untreated control, vehicle-treated control, and CMT 2215 at doubling dilutions from 0.39 jig/ml to 50 ug/ml. Cells received fresh CMT treatment every 48 h for 5 days, receiving a total of three treatments. At the end of the 5-day treatment, plates were processed using the MTT [3-(4,5-dimethyl thiazol-2-yl)-2,5- diphenyl tetrazolium bromide) assay described previously (W ebber et al., 2001). Results represent the average of two experiments. 135 In viva studies Mice: Eight week old, albino male mice, nu/nu strain (NSWNU-M, homozygotes) (Taconic farms, Inc., Germantown, NY), were used in this study. This strain of mice is the standard athymic model for the National Cancer Institute (NCI) studies as well as many pharmaceutical and institutional oncology screening programs. Mice were socialized for four days after arrival from Taconic. The mice were provided with autoclaved tap water to drink and fed a complete, irradiated diet (Teklad 7904, manufactured by Harlan, Madison, WI). To ensure the health of the mice, their physical condition and food and water intake were examined daily. For enrichment purposes the mice were given nestlettes once a week. Animal maintenance: The mice were maintained at the University Laboratory Animal Resources (ULAR) facility in a clean room dedicated for this experiment (Figure 6.1). The room was maintained at 720 -740 F and on a twelve-hour light-twelve hour dark schedule. The mice were housed individually in autoclaved cages on paper chip bedding in a laminar flow rack. The Clinical Center is one of several buildings on campus that houses animals (Figure 6.1a). The entrance to the ULAR facility of the Clinical Center Building (Figure 6.1b), animal room (Figure 6.1c) and laminar flow cage rack (Figure 6.1d) in which the mouse cages are separated by rows for the four groups of mice, are shown. Investigators and caretakers were required to wear the following upon entrance to the room: Bonnet, 136 mask, booties, sterile gloves, and sterile gown. All procedures involving mice were performed in a laminar flow hood. The drug (CMT 2215) was administered to mice by gavage feeding. Gavage feeding was done in the laminar flow hood under yellow light to protect the CMT. A laminar flow hood (Figure 6.1e), where cell injections, daily feeding by gavage (Figure 6.11) using a sterile feeding tube, weighing, and cage changes take place is also shown. This study was conducted with the approval of the All University Committee on Animal Use and Care (AUCAUC) and all guidelines were followed. 137 Figure 6.1 The facility, experimental design and equipment used for in viva studies. 6.1a. Clinical Center Building; 6.1b. University Laboratory Animal Resources (ULAR) facility; 6.1c. room for housing immune-deficient mice (nude mice); 6.1d. laminar flow mouse cage rack. The cages were arranged in rows for the four groups of mice; row 1 = RWPE2-W99 controls; row 2 = RWPE2-W99 treated; row 3 = CTPE controls; row 4 = CTPE treated. Mice were housed, one mouse per cage, in autoclaved cages, and provided with autoclaved drinking water and irradiated food. 6. 1 e. laminar flow hood where gavage feeding was performed; 6. l f. Gavage feeding procedure. The control mice were fed 300 [.11 of a 5% sucrose solution in water by gavage. The treated mice were similarly fed with 0.675 mg or 2.25 mg of 2215/mouse in 300 u] ofa 5% sucrose solution starting 3 days prior to cell injection. Gavage feeding was performed daily for a total of 10 weeks. 138 Sucrose solution: vehicle for CMT 2215: The control mice were fed 300 111 of a 5% sucrose solution in water by gavage. A 5% sucrose (Sigma, Cat. No. S-5016) solution was prepared in de-ionized water and autoclaved. Aliquots of 1.5 ml/tube were prepared for gavage feeding of control mice and stored in a -200 C freezer. Gavage feeding was performed daily for a total of 10 weeks. This solution was used as the vehicle for CMT 2215. Drug stock solutions: CMT 2215 was provided by Innapharrna lnc./Tetragenex Pharmaceuticals, Inc., Park Ridge, NJ (from ACROS Organics, NJ), stored at 40to 60 C, and protected from the light. A low-dose and a high-dose stock solution of CMT 2215 was prepared. The low- dose stock solution (6.75 mg/ml) was prepared in sterile 5% sucrose and further diluted to obtain a 2.25 mg/ml (0.675 mg/ 300 (11) solution and filter-sterilized using a 0.22 um pore size filter. The high-dose stock solution of CMT 2215 (7.5 mg/ml or 2.25 mg/ 300 111) was also prepared in sterile 5% sucrose solution and filter-sterilized using a 0.22 um pore size filter. Aliquots of both low-dose and high-dose CMT 2215 of 1.5 ml/ brown Eppendorf tube were prepared and stored in a -200 C freezer, in boxes to protect from light, until needed. The treated mice were fed 300 111 of a 5% sucrose solution, containing 0.675 mg or 2.25 mg of 2215/mouse by gavage, starting 3 days prior to cell injection. 139 Cells for injections: A sterile cell suspension of RWPE2-W99 (Bello et al., 1997; Webber eta1., 1997a) or CTPE cells was prepared in basal keratinocyte serum-free growth medium and mixed with an equal volume of Matrigel (W ebber et al., 2001) to obtain 4 million cells/ml. All steps with Matrigel were performed on ice. The cell suspension was kept on ice and taken to the animal facility. A 1.0 cc syringe with a 23 gauge needle (or 25 gauge for CTPE) was used to inject 250 pl of the cell suspension containing one million cells, per inoculation site. Before injection the mice were swabbed with alcohol at the injection site. Cells were injected subcutaneously and bilaterally on the dorso-lateral side. The mice were kept on a heating pad during the procedure to maintain body temperature. Experimental groups: The study involved four groups of mice. Group 1: RWPEZ- W99 controls: Four mice were gavage fed (Figure 11) with 300 ul of 5% sucrose solution in water daily for three days prior to being injected with RWPE2-W99 cells. After cell injection, daily gavage feeding with sucrose solution continued for 10 weeks. A fresh vial of sucrose was used each day. Group 2: RWPEZ- W99 CM T #2215 low dose treated mice: Four mice were gavage fed, each receiving 300 111 of 5% sucrose solution in water containing CMT 2215 daily for three days prior to being injected with RWPE2-W99 cells. Group 2 mice, with average body weight of ~27 g, received 0.675 mg of 140 CMT 2215/mouse. After cell injection, daily gavage feeding with CMT 2215 in sucrose solution continued for 10 weeks. A fresh vial of CMT solution was used each day. Each day’s supply was thawed just before use and the vials were kept in cardboard boxes to prevent light exposure. Group 3: CT PE controls: Three mice were gavage fed with 300 11.1 of 5% sucrose solution in water daily for three days prior to being injected with CT PE cells. After cell injection, daily gavage feeding with sucrose solution continued for 10 weeks. A fresh vial of sucrose was used each day. (Note: Groups 3 and 4 only had 3 mice each because two mice became dehydrated and expired prior to start of this experiment.) Group 4: C TPE CMT 2215 high dose treated mice: Three mice were gavage fed with 300 111 of 5% sucrose solution in water containing CMT 2215 daily for three days prior to being injected with CTPE cells. Group 4 mice, with average body weight of ~30 g, received 2.25 mg of CMT 2215/mouse. After cell injection, daily gavage feeding with CMT 2215 in sucrose solution continued for 10 weeks. A fresh vial of CMT solution was used each day. Each day’s supply was thawed just before use and the vials were kept in cardboard boxes to prevent light exposure. 141 Animal weights: Animals were weighed at the start of the experiment and then weekly until the end of the experiment, and the weights were recorded. Mice were given a physical examination daily, and their food and water intake were also monitored daily. The average weight of mice on day zero was ~27 g for treatment groups 1 and 2 and ~30 g for treatment groups 3 and 4. Tumor size and histology: When tumors became palpable, their size was measured periodically in two dimensions (length and width), using digital calipers, and all measurements were recorded. Tumor volume (TV) was calculated using the formula: TV = a X bZ/Z where a is the longest dimension and b is the width (Nemeth et al., 1999). Upon termination of the experiment at 10 weeks, mice were sacrificed using C02, and photographed for tumor size. Tumors were dissected, fixed in 10% buffered formalin and processed for histology. Tumor sections were cut at Sum thickness. In addition, abdominal wall, liver and lungs were examined for visible tumor metastasis. Lungs, liver and other areas showing invasion were also fixed for histology. Results: In vitro studies Effect of CMT 2215 on anchorage-dependent growth In order to examine the effects of CMT 2215, concentrations from 0.39 to 50 ug/ml were tested. Two independent experiments, using six replicate wells/treatment, were conducted. Results shown in Figure 6.2 represent the average ofthese two experiments. From these data. it is evident that RWPE2-W99 cells show a dose- dependent inhibition of growth at concentrations higher than 1.56 rig/ml for CMT 2215. Approximately 65% growth inhibition was observed at 50 rig/ml. The leo for 2215 was ~35.74 rig/ml when cells were plated at 10.000 cells/well. 125%l 1000/04 ‘Hx _ ! Cell “‘7 growth 75%, \ (percent \ 0f \\ 0 control) 50 A) \ 250/04 ID 50 = ~35.74 pg/ml 00/0 . . . . . . . . . . - . . I . . . 0 0.39 0.78 1.56 3.13 6.25 12.5 25 50 CMT #2215 concentration (pg/ml) Figure 6.2 The effects of CMT 2215 on anchorage-dependent growth of RWPE2-W99 cells. Cells were plated in 96-well plates at a density of 10,000 cells per well and treated for 5 days. Results are plotted as percent of DMSO-treated control, iSEM. 143 In vivo studies Weight: .Mice with R WPEZ- W99 cell xenografts Animals were weighed weekly. Figure 6.3 shows the average weight of control and CMT 2215-treated mice, injected with RWPE2-W99 cells. The treated mice received CMT 2215 at 0.675 mgz”mouse by gavage, daily. Results show that there was no difference in the average weight of mice between the control and treated groups over a 10 week period. 401 Weight 35 (grams) Cells injected __ r “ . .7- 304 5 _ " 3' . RWPE2-W99 Control '- f5 RWPE2-W99 Treated CMT #2215 252.2... 2W- l 2 3 4 5 6 7 8 9 10 11 Time(weeks) Figure 6.3 Average weight of control and treated mice injected with RWPE2-W99 cells. In the control group, four mice were given vehicle alone (5% sucrose solution in water) by gavage daily for 10 weeks. Four mice in the treated group were given 0.675 mg of CMT 2215/mouse daily by gavage for 10 weeks. The days on which gavage feeding was started, and cells injected, are shown. 144 Mice with C TPE cell xenografts Figure 6.4 shows the average weight of control and CMT 2215-treated mice injected with CTPE cells. The treated mice received CMT 2215 at 2.25 mg/mouse by gavage, daily. Results show that the treated mice have a slightly lower average weight as compared to control mice and the shape ofthe curves at 10 weeks suggests a divergence in weight. 40- 35‘ .4.— Cells injected //-*\,//' (grams) . . 302 5 —o— CTPE Control CTPE Treated CMT#2215 25 -.T.V.-4a.v.r.er., l 2 3 4 5 6 7 8 9 10 Time (weeks) Figure 6.4 Average weight of control and treated mice injected with CTPE cells. In the control group, three mice were given vehicle alone (5% sucrose solution in water) by gavage daily for 10 weeks. Three mice in the treated group were given 2.25 mg of CMT 2215/mouse daily by gavage for 10 weeks. The days on which gavage feeding was started, and cells injected, are shown. 145 Tumor development: RWPE2- W99 cells Figure 6.5a shows tumors in mice injected with RWPE2-W99 cells. Mice in Figure 6.5a are controls that were given a 5% sucrose solution. Mice treated with CMT 2215 were given a daily dose of 0.675 mg/mouse by gavage. Tumor measurements show an average tumor volume of 3.2 cc for the RWPE2-W99 controls and 1.8 cc for the CMT 2215- treated mice that received a daily dose of 0.675 mg/mouse. These preliminary results suggest that using the slow growing RWPE2-W99 model, treatment with CMT 2215 caused a decrease in tumor volume. The average tumor size from treated animals was approximately half the average size of tumors from the controls. RWPE2-W99 Cells CTPE Cells Figure 6.5 Nude mice (strain NCRNU-M male, homozygotes, ~8 weeks old from Taconic farms, Germantown, NY) were bilaterally injected subcutaneously with 250 pl of a cell suspension in Matrigel (cellszMatrigel volume 121) containing 1 million cells. Figure 6.5a shows mice given injections of RWPE2-W99 cells. Figure 6.5b shows mice given injections of CTPE cells. Mice were sacrificed 10 weeks later. These mice were fed 300 pl of a 5% sucrose solution by gavage starting 3 days prior to cell injection. Gavage feeding was performed daily for 10 weeks. Arrows point to tumors. 146 C T PE cells Figure 6.5b shows tumors in mice injected with CTPE cells. Mice in Figure 6.5b are controls that were given a 5% sucrose solution. Mice treated with CMT 2215 were given a daily dose of 2.25 mg/mouse by gavage. Tumor measurements show an average tumor volume of 6.6 cc for the CTPE controls and 8 cc for the CMT 2215-treated mice. These preliminary results suggest that using the fast growing and invasive CTPE model, treatment with CMT 2215 did not show an effect on average tumor volume. It should be noted that in the CTPE experiment, only three mice per group were used. Histology: Histology of the RWPEZ- W99 tumors in control mice The RWPE2-W99 tumors are slow growing. It is possible that for this reason, the test agent may have shown a greater effect on tumor volume than on the fast growing CTPE tumors. Figure 6.6 shows RWPE2-W99 tumors in control mice. Figure 6.6a shows a subcutaneous tumor. Under the skin, the tumor margin appears to be well defined. Figure 6.6b shows the interface between the adipose tissue and the tumor with clear margins, and does not show invasion at this site. It is possible that if the animals are maintained for longer than 10 weeks, one may see invasion and metastasis. Figure 6.6c shows tumor histology at a higher magnification. Figure 6.6d shows skeletal muscle cells amongst tumor cells. 147 Figure 6.6 Histology of the RWPE2-W99 tumors in control mice: Figure 6.6a shows a subcutaneous tumor (arrow). Under the skin, the tumor margin appears to be well defined and separate from the skin. Figure 6.6b shows tumor2adipose tissue interface with clear margins, and the tumor does not show invasion at this site. It is possible that if the animals are maintained for longer than 10 weeks, one may see invasion and metastasis. Figure 6.6c shows tumor histology at a higher magnification. Figure 6.6d shows skeletal muscle cells amongst tumor cells. H & E stain. Histology of the RWPE2- W99 tumors in CM T -treated mice Figure 6.7 shows RWPE2-W99 tumors in CMT 2215-treated mice. Figure 6.7a shows an area of the tumor that appears similar to the control tumor. Figure 6.7b shows a representative area with many vacuolated cells. Figure 6.7c shows squamous metaplasia. There are also areas that show large lymphocytic infiltration (Figure 6.7d). Many areas showed what appeared to be apoptotic cells (Figure 6.7e). Such changes were not seen as frequently in the control tumors. 148 Figure 6.7 Histology of the RWPE2-W99 tumors in CMT-treated mice. Figure 6.73 shows an area of the tumor that does not appear to show any difference from the control tumor. Figure 6.7b shows a representative area with many vacuolated cells. Figure 6.7c shows (arrow) squamous metaplasia. There are also areas that show large lymphocytic infiltration (dark staining nuclei) (Figure 6.7d). Many areas showed what appear to be apoptotic cells (arrows) (Figure 6.7e). Such changes were not seen as frequently in the control tumors. H & E stain. Histology of the CT PE tumors in control mice The CTPE tumors are rapidly growing, invasive tumors and invasion can be seen at the 10 week experimental period (Figure 6.8). Figures 6.83 shows a subcutaneous tumor showing invasion into the sub-epidermal layer. The tumor does not have clear cut 149 margins as can be seen in Figure 6.8b. Figures 6.8c and 6.8d show that the tumor cells are intermingled with skeletal muscle (Figure 6.8c) and fat cells (Figure 6.8d). The tumor cell population is very heterogeneous with considerable variation in cell size (Figure 6.86). Figure 6.8 Histology of CTPE tumors in control mice. The CTPE tumors are rapidly growing, invasive tumors and invasion was observed at the 10 week experimental period. Figures 6.8a shows a subcutaneous tumor with invasion into the sub-epidermal layer. The tumor has infiltrated into the dermis and does not have clear cut margins as can be seen in Figure 6.8b. Figures 6.8c and 6.8d show that the tumor cells are intermingled with skeletal muscle (M) (Figure 6.8c) and fat cells (FC) (Figure 6.8d). The tumor cell population is very heterogeneous with considerable variation in cell size (Figure 6.8c). H & E stain. 150 Histology of the C T PE tumors in CM T -treated mice Figure 6.9 shows some features seen in CMT-treated CTPE tumors. A tumor with undifferentiated characteristics, shown in Figure 6.9a, suggests the aggressive nature of CTPE tumors. Invasion into skeletal muscle is seen in Figure 6%. Cells, which appear to be undergoing apoptosis, are seen in Figures 6.9c and 6.9d. Such cells were not seen as frequently in the control CTPE tumors. Areas showing lymphocytic infiltration were seen in several tumors (Figure 6.9e). The appearance of apoptotic cells in tumors from treated mice is an observation that needs to be further examined and may have some significance when assessing the effects of the CMT. 151 4.“ _ 23?. .122” fig": . '7"‘:.’" : l‘.k‘ “a:‘§§ M "£8: 5“. .r "(I \a. Figure 6.9 Histology of CTPE tumors in CMT-treated mice. This figure shows some features observed in CMT-treated CTPE tumors. A tumor with undifferentiated characteristics shown in Figure 6.9a suggests the aggressive nature of CTPE tumors. Invasion into skeletal muscle (M) is seen in Figure 6%. Cells which appear to be undergoing apoptosis are shown (arrows) in Figures 6.9c and 6.9d. Such cells were not seen as frequently in the control CTPE tumors. Areas with lymphocytic infiltration were observed in several tumors (Figure 6.9e). H & E stain. Histology of C T PE tumor metastasis to the lung One control mouse (1/3) showed metastasis to the lung (Figure 6.10). Figure 6.103 shows a normal area of the lung. Figure 6.10b and 6.10c are low magnification pictures of the lung showing metastatic tumors. Figure 6.10d is a higher magnification 152 picture of the lung showing lung tissueztumor interface. The lung tissue has the alveoli represented by clear spaces against which the tumor tissue has a solid appearance. One treated mouse (1/3) also showed tumor adhering to the lung. Figure 6.10 Histology of the normal lung and of CTPE tumor metastasis to the lung. Figure 6.10a shows a normal area of the lung. One of the control mice (1/3) showed metastasis to the lung. Figures 6.1% and 6.10c are low magnification picture of the lung (L) showing metastatic tumors (T). Figure 6.10d is a higher magnification picture of the lung (L) showing lung tissueztumor (T) interface. The lung tissue has the alveoli represented by clear spaces against which the tumor tissue has a solid appearance. H & E stain. Discussion This study was conducted primarily to develop methods for conducting in vivo experiments using CMTs. The treatment groups are very small, hence, the results are suggestive, and should be considered as such. Eight mice were given subcutaneous 153 injections of RWPE2-W99 cells and six were given subcutaneous injections of CTPE cells. Half of the mice given subcutaneous injections for each cell line were treated with CMT 2215 and the other half were treated with 5% sucrose daily by gavage for 10 weeks. The preliminary results suggest that in the slow growing RWPE2-W99 model, treatment with CMT 2215 (2.25 mg/ml) caused a decrease in tumor volume. The average tumor size from treated animals was approximately half the average size of tumors from the controls. Metastasis was not observed in any of the mice given subcutaneous injections of RWPE2-W99 cells. In contrast to RWPE2-W99, tumors produced in control mice by CTPE cells following subcutaneous injection were ~2-fold larger. Despite the fact that mice with CTPE tumors were given a high-dose (7.5 mg/ml) of CMT 2215, tumor measurements show an average tumor volume of 6.6 cc for the CT PE controls and 8.0 cc for the CMT 2215-treated mice. Thus, using the fast growing and invasive CTPE model, treatment with CMT 2215 did not show a growth inhibitory effect on average tumor volume. Although CMT 2215 appeared to have no effect on the tumor growth of CTPE cells, the tumors collected from both groups of CMT-treated mice had many sections showing what appeared to be apOptotic cells. Such changes were not seen as frequently in the control tumors. Induction of apoptosis could be one of the mechanisms by which CMTs inhibit tumor growth. The RWPE2-W99 tumors are slow growing. It is possible that for this reason, the test agent may have shown a greater effect on tumor growth than on the fast growing CTPE tumors. On the other hand, studies have confirmed the highly invasive and metastatic behavior of CTPE cells in viva (Achanzar et al., 2001). As a more appropriate model of human prostate cancer, the RWPE2-W99 cell line, which is slow growing, was 154 selected to conduct a second study in nude mice to test the effects of two other chemically modified tetracyclines, CMT 2137 and CMT 2147 (see Chapter 7). Acknowledgements Chemically modified tetracycline 2215 was provided by Tetragenex Pharmaceuticals, Inc., Park Ridge, NJ. Suppport for this project by Tetragenex is acknowledged. 155 Literature cited Achanzar, W.E., Lamar, P.C., Tokar, E.J., Rivette, A.S., Bello-DeOcampo, D., Prozialeck, W.C., Webber, M.M. and Waalkes, M.P.: Human prostate cell lines mimic heterogeneity of cadherin expression in human prostate cancer. UroOncology 4:15-25, 2004. Achanzar, W.E., Diwan, B.A., Liu, J ., Quader, S.T., Webber, M.M. and Waalkes, M.P.: Cadmium-induced malignant transformation of human prostate epithelial cells. Cancer Research 612455-458, 2001. Bello, D., Webber, M.M., Kleinman, H.K., Wartinger, DD, and Rhim, J .S.: Androgen responsive adult human prostatic epithelial cell lines immortalized by human papillomavirus 18. Carcinogenesis 18:1215-1223, 1997. Nemeth, J .A., Harb, J .F., Barroso, Jr., U., He, 2., Grignon, DJ. and Cher, M.L.2 Severe combined irnrnunodeficient-hu model of human prostate cancer metastasis. Cancer Research 59:1987-1993, 1999. Webber, M.M., Bello, D., Kleinman, H..,K and Hoffman, M.P.: Acinar differentiation in non-malignant immortalized human prostatic cells and its loss by malignant cells. Carcinogenesis 1821225-1231, 1997a. Webber, M.M., Bello, D., and Quader, S.: Immortalized and tumorigenic adult human prostatic epithelial cell lines. Characteristics and applications. Part 2. Tumorigenic cell lines. The Prostate 30258-64, 1997b. Webber, M.M., Quader, S., Kleinman, H..,K Bello-DeOcampo, D., Storto, P.D., Bice, G., de Mendonca-Calaca, W., and Williams, D.E.2 An in vitro/in viva model of human cell lines for prostate carcinogenesis and tumor progression. The Prostate 4721-13, 2001. 156 CHAPTER SEVEN THE EFFECTS OF CMT 2137 AND 2147 ON TUMOR GROWTH USING THE TUMORIGENIC RWPE2-W99 HUMAN PROSTATE CELL LINE 157 Abstract Chemically modified tetracyclines (CMTs) may be effective chemotherapeutic agents for prostate cancer. Results obtained from in vitro and in viva studies, in addition to a phase I clinical trial, suggest a potential use of CMTs as an oral, nontoxic drug to treat metastatic prostate cancer and other cancers. Additional screening of CMTs may lead to the identification of compounds which show greater efficacy in the treatment of prostate cancer. Thus, in this chapter, I examine the ability of two new CMTs, CMT 2137 and 2147, to inhibit the growth of a human prostate epithelial cell line, RWPE2-W99 both in vitro, and as a xenograft in viva. Treatment of RWPE2-W99 cells grown in monolayer cultures with 2137 (50 ug/ml) caused ~25% growth inhibition, while treatment with 2147 (50 ug/ml) caused ~60% growth inhibition. In viva, the ability of CMT 2137 and 2147, to reduce the size and number of tumors produced after subcutaneous injection of RWPE2-W99 cells in immune-suppressed mice, was examined. The mice treated with the test agents had a higher percentage of tumors in the small category (11-110 mm3) as compared to the control mice. Based on the results of these in vitro and in viva studies, both CMT 2137 and 2147 demonstrate potential as anticancer drugs and warrant further study. Keywords Chemically modified tetracycline, gavage, prostate cancer, subcutaneous, tumor 158 Introduction Chemically modified tetracyclines (CMTs) have been shown to inhibit cell proliferation and Matrigel invasion of several prostate cancer cell lines, as well as, cause a decrease in matrix metalloproteinase production and activity in vitro (Lokeshwar, 1999; Lokeshwar et al., 2002). In male Copenhagen rats, given subcutaneous injection of MAT LyLu cells, treatment with CMT-3 by gavage inhibited tumor incidence and reduced the tumor growth rate (Lokeshwar et al., 1999). In Copenhagen rats given an intravenous injection of MAT LyLu cells, treatment with CMT-3 decreased the frequency of tumor metastasis to soft or skeletal tissue and also resulted in an increase in survival (Lokeshwar, 1999; Selzer et al., 1999). Other CMTs are now being extensively investigated because of their increased efficacy as compared to their natural derivatives. Based on the pilot study with CMT 2215 (see chapter 6), a protocol was developed to conduct a study in nude mice to test the effects of two other chemically modified tetracyclines, CMT 2137 and CMT 2147. These two CMTs were first tested for their effects on the growth of RWPE2-W99 cells in vitro and were shown to inhibit cell proliferation. The tumorigenic human prostate epithelial cell line RWPE2-W99 (Bello et al., 1997; Webber et al., 1997a) was used to examine the potential of CMTs in vitro and in viva. Using RWPE2-W99 cells to produce tumors in immune-deficient mice is a relevant model because it mimics the majority of human prostate cancers which are slow growing (Webber et al., 1997b; Webber et al., 2001). The following describes the process by which the RWPE2-W99 cell line was developed. Human prostate epithelial cells were derived from the peripheral zone of a normal human prostate and immortalized with a single copy of human papilloma virus-l8 159 (HPV-l8) DNA to give rise to the RWPE-l cell line (Bello et al., 1997; Webber et al., 1997a). RWPE-l cells were then transformed by v-Ki-ras, giving rise to the RWPE-2 cell line (Bello et al., 1997; Webber et al., 1997a). The transformed RWPE-2 cells are tumorigenic and can grow in soft agar in an anchorage-independent manner. In order to select cells that showed high Ki-ras expression, RWPE-2 cells were grown in agar and colonies were screened for Ki-ras expression. One of these colonies was expanded and gave rise to the RWPE2-W99 cell line. Using the RWPE2-W99 cell line, the objectives of this study were to determine the ability of CMT 2137 and 2147 to inhibit cell growth in vitro and tumor growth in viva using RWPE2-W99 cells which form slow growing tumors when injected into immune-suppressed mice. Materials & Methods: In vitro studies Cell culture general: RWPE2-W99 cells were grown in complete keratinocyte serum-free medium (KSFM) containing 50 ug/ml bovine pituitary extract (BPE), 5 ng/ml epidermal grth factor (EGF) and 1X antibiotic/antimycotic solution. Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO; and subcultured weekly. Dose response using a microplate assay: RWPE2-W99 cells were plated, six wells per treatment, in complete keratinocyte serum-free medium (K-SFM) containing 50 ug/ml bovine pituitary extract (BPE) and 5 ng/ml epidermal growth factor (EGF), at 10,000 cells/well in 96-well plates and 160 allowed to attach for 48 h at which time medium was changed to medium containing varying concentrations of the test agent. The test agents, CMT 2137 and 2147 were dissolved in DMSO. The final concentration of the DMSO vehicle in the culture medium was 0.1%. Treatment groups consisted of untreated control, vehicle-treated control, CMT 2137 or CMT 2147 at doubling dilutions from 0.39 jig/ml to 50 ug/ml. Cells received fresh CMT treatment every 48 h for 5 days, receiving a total of three treatments. At the end of the 5-day treatment, plates were processed using the MTT [3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay as described previously (W ebber et al., 2001). Two or three replicate experiments were conducted for each test agent. In vivo studies Mice: Eight week old, albino male mice, nu/nu strain (N SWNU-M, homozygotes) (Taconic farms, Inc., Germantown, NY), were used in this study. This strain of mice is the standard athymic model for the National Cancer Institute (NCI) studies as well as many pharmaceutical and institutional oncology screening programs. Mice were socialized for five days after arrival from Taconic. The mice were provided with autoclaved tap water to drink and fed a complete, irradiated diet (T eklad 7904, manufactured by Harlan, Madison, WI). For enrichment purposes the mice were given nestlettes once a week. To ensure the health of the mice, their physical condition and food and water intake were examined daily. 161 Animal maintenance: The mice were maintained at the University Laboratory Animal Resources (ULAR) facility in a clean room dedicated for this experiment. The room was maintained at 720 -740 F and on a twelve-hour light-twelve hour dark schedule. The mice were housed individually in autoclaved cages on paper chip bedding in a laminar flow rack. The animal room and laminar flow cage rack, where the mouse cages are color- coded for the three groups of mice, are shown in Figure 7.1. Figure 7.1 Laminar flow mouse cage rack. The cages were color-coded for the three groups of mice; yellow = controls; blue 2 2137; red: 2147. Mice were housed, one mouse per cage, in autoclaved cages, and provided with autoclaved drinking water and irradiated food. The control mice were fed 300 111 of a 5% sucrose solution in water by gavage. The treated mice were similarly fed with 1.2 mg of 2137 or 2147/mouse in 300 1.11 of a 5% sucrose solution starting 3 days prior to cell injection. Gavage feeding was performed daily for a total of 11 weeks. 162 Investigators and caretakers were required to wear the following upon entrance to the room: bonnet, mask, booties, sterile gloves, and sterile gown. All procedures involving mice were performed in a laminar flow hood. The drug was administered to mice by gavage feeding. Gavage feeding was done in the laminar flow hood under yellow light to protect the CMTs. Cell injections, daily feeding by gavage using a sterile feeding tube weighing, and cage changes took place in a laminar flow hood. This study was conducted with the approval of the All University Committee on Animal Use and Care (AUCAUC) and all guidelines were followed. Sucrose solution: The control mice were fed 300 pl of a 5% sucrose solution in water by gavage. A 5% sucrose (Sigma, Cat. No. S-5016) solution was prepared in de-ionized water and autoclaved. Aliquots of 2.5 ml/tube were prepared for gavage feeding of mice and stored in a -200 C freezer. Gavage feeding was performed daily for a total of 11 weeks. This 5% sucrose solution was also used as the vehicle for CMTs 2137 and 2147. Drug stock solutions: Chemically modified tetracycline: CMTs 2137 (Lot No. 14-77-1) and 2147 (Lot No. 14-95-2) were provided by Innapharma Inc/Tetragenex Pharmaceuticals, Inc., Park Ridge, NJ (from RSA Corporation). Test reagents were stored at 40 to 60 C, and protected from light. The treated mice were fed 300 p1 of a 5% sucrose solution by gavage starting 3 days prior to cell injection with 1.2 mg of 2137 or 2147/mouse. Stock solutions containing 4 mg/ml of 2137 or 2147 were prepared in sterile 5% sucrose and 163 filter-sterilized using a 0.22 pm pore size filter. Aliquots of 2.5 ml in sterile brown glass vials were prepared and stored in a -200 C freezer, in boxes to protect from light, until needed. Gavage feeding was performed daily for a total of 11 weeks. Cells for injections: A sterile cell suspension of RWPE2-W99 cells was prepared in keratinocyte serum-free basal medium (Bello et al., 1997; Webber et al., 1997a) and mixed with an equal volume of Matrigel (W ebber et al., 2001) to obtain four million cells/ml. All steps with Matrigel were performed on ice. The cell suspension was kept on ice and taken to the animal facility. A 1.0 cc syringe with a 25 gauge needle was used to inject 250 pl of the cell suspension containing one million cells per inoculation site. Before the cells were injected, the mice were placed under methoxyflurane anesthesia and swabbed with alcohol at the injection site. The cells were injected subcutaneously and bilaterally on the dorso-lateral side. The mice were kept on a heating pad during the procedure to maintain body temperature. Experimental groups: The study involved three groups of mice. Mice were randomly selected and placed in three groups. Group I: RWPEZ- W99 controls: Starting on day zero, ten mice were gavage fed with 300 pl of 5% sucrose solution in water, daily for three days, prior to being injected with RWPE2-W99 cells. After cell 164 injection, daily gavage feeding with sucrose solution continued for a total of 11 weeks. Fresh vials of sucrose were used each day. Group 2: RWPEZ- W99 CMT 2137: Twelve mice were gavage fed, each mouse receiving 300 pl of 5% sucrose solution in water containing 1.2 mg of CMT 2137, daily for three days prior to being injected with RWPE2-W99 cells. Daily gavage feeding, at the same dose, continued after cell injection, for a total of 11 weeks from day zero. Fresh vials of the test agent solution were used each day. Each day’s supply was thawed just before use and the vials were kept in cardboard boxes to prevent light exposure. Group 3: RWPE2- W99 CMT 2147: Twelve mice were gavage fed, each mouse receiving 300 pl of 5% sucrose solution in water containing 1.2 mg of CMT 2147 daily for three days prior to being injected with RWPE2-W99 cells. Daily gavage feeding, at the same dose, continued after cell injection, for a total of 11 weeks from day zero. Fresh vials of the test agent solution were used each day. Each day’s supply was thawed just before use and the vials were kept in cardboard boxes to prevent light exposure. Animal weights: Animals were weighed at the start of the experiment and then weekly until the end of the experiment, and the weights were recorded. Mice were given a physical 165 examination daily, and their food and water intake were also checked daily. The average weight of mice on day zero was 28.2 g. Tumor size and histology: When tumors became palpable, their size was measured periodically in two dimensions (length and width), using digital calipers, and all measurements were recorded. Tumors were measured and photographed at the termination of the experiment at 11 weeks. Tumor volume (TV) was calculated using the formula: TV = a X bZ/Z where a is the longest dimension and b is the width (Nemeth et al., 1999). Tumors were dissected, fixed in 10% buffered formalin and processed for histology. Tumor sections were cut at 5pm thickness. In addition, abdominal wall, liver and lungs were examined for visible tumor metastasis. Results In vitro studies Anchorage—dependent growth In order to examine the effects of CMT 2137 and 2147, concentrations from 0.39 to 50 pg/ml were tested using the RWPE2-W99 cell line. Two or three independent experiments, using six replicate wells/treatment, were conducted. Results shown in Figure 7.2 represent the average of three experiments. From these data, it is evident that RWPE2-W99 cells show a dose-dependent inhibition of growth at concentrations higher than 25 and 3.13 pg/ml for CMT 2137 and 2147, respectively. Approximately 25% inhibition was observed at 50 pg/ml for CMT 2137 and 60% growth inhibition at 166 50 pg/ml for CMT 2147. The len for 2147 was ~31.26 pgx’ml when cells were plated at 10,000 cells/well. 125% ‘ ..,-f" 1000/0 "““"---..__ __ ,___ ---"’Pfi.— \ ._. “.2 ‘_._- 2-2:»- . Cell \\ growth 75% . (percent of control) 50% ID 50 = 31.26 pg/ml 25 /° -- CMT 2137 .1. . CMT 2147 00/0 . . u w r T F 0 0.39 0.78 1.56 3.13 6.25 12.5 25 50 CMT concentration (pg/ml) Figure 7.2 The effects of CMT 2137 or CMT 2147 on anchorage dependent grth of RWPE2-W99 cells. Cells were plated in 96-well plates at a density of 10,000 cells per well and treated for 5 days. Results are plotted as percent of DMSO- treated control, iSEM. In vivo studies Animal weight: Animals were weighed weekly. Figure 7.3 shows the average weight of control mice and those treated with CMT 2137 or 2147. The treated mice received 1.2 mg ofthe test agent daily for l 1 weeks. Results show that there was no apparent difference in the average weight of mice between the control and treated groups up to about 9 weeks. At 1 l 167 weeks, the control mice weighed 32.3 g, the 2137-treated mice weighed 32.7g, while the 2147-treated mice weighed 31.6g. 33 . "“-e-____} j 9» I ,o.-- -' .e- _ .‘.."'H N: .- t 1; 1 Cells 3‘ 4": Weight injected ring-file i ( rams) . ' _. ’4 I g 29 j: . + RWPE2-W99-C 5!: £11" —- RWPE2-W99-37 I ‘ -** RWPE2-W99-47 27 CMT 25 . . . r . . . . 0 l 2 3 4 5 6 7 8 9 10 11 Time (weeks) Figure 7.3 Average weight of mice injected with RWPE2-W99 cells. This graph shows weight gain, over time, in the control mice given vehicle alone (5% sucrose solution in water) by gavage daily for l 1 weeks, and in the mice gavage-fed with 1.2 mg/mouse of 2 l 37 or 2147 in 5% sucrose solution. Mice were weighed at time zero and then weekly for l 1 weeks. Beginning at time zero, mice were gavage-fed for three days prior to the injection of RWPE2-W99 human prostate tumorigenic cell line. The day on which gavage feeding was started (labeled as CMT), and the day when the cells were injected, are shown. RWPE2-W99-C = control; RWPE2-W99-37 = mice on 2137; RWPE2-W99-47 = mice on 2147. Tumor volume: Tumors were measured at the time of sacrifice. Tumor measurement results are shown in Tables 7.1, 7.2 and 7.3 and Figure 7.4. Table 7.1 shows the relationship between tumor volume and percent of tumors having the indicated size. Tumor volumes 168 were divided into four groups. There is considerable variation in tumor size within each group. The size interval in the first group (ll-110 m3) is smaller than the other groups. This was done to enable recording of the smallest tumors. While only 8% of the tumors in control mice were in the smallest tumor volume category (ll-110 mm3), 34% and 21% of the tumors in mice treated with 2137 or 2147, respectively, were in the smallest tumor volume category (Figure 7.4). Furthermore, while 25% of the tumors in the control mice were in the largest tumor volume category, none or only 5% (one tumor) of the tumors were in the largest category in 2137 or 2147 treated mice, respectively. These data are summarized in Table 7.2. If the tumors of the two small categories are combined, 66% of the control tumors fall within this group, while 77% and 84% of the tumors fall in this group for 2137 or 2147 treated mice, respectively. Similarly, when the two large categories are combined, 34% of control, 23% of 2137 and 16% of 2147-treated mice, fall into the large tumor volume category. In Table 7.3, tumors have been grouped, according to volume into nine groups which again show that the mice treated with the test agent have no or very few tumors in the largest categories while 25% of the tumors in the control mice are in the largest categories. Representative tumors of varying sizes from control and treated mice are shown in Figure 7.5. These results suggest that the two test agents caused a decrease in tumor size. 169 Table 7.1 This table shows the relationship between tumor volume, following injection of RWPE2-W99 cells, and percentage of tumors having the indicated size. Tumor volumes have been divided into four groups. The average tumor volume, range, and percent of tumors having the indicated size in each group are shown for control mice, and mice treated with 2137 or 2147. Tumor Control 2137 2147 Volume (M’) Average Range % of Average Range % of Average Range % of tumors tumors tumors 11-1 10 24 24 8% 41 12-73 34% 82 72-92 21% 11 1.510 196 1 16-298 58% 262 125-489 43% 242 127-451 63% 511-910 519 519 9% 728 663-825 23% 691 628-754 1 1% >911 1214 1118-1272 25% 0 0 0% 1462 1462 5% 170 7 D % 8 0% 513% Percent 4 0% of tumors 3 0 % 2 [1% 10% [1% o o o N o o o N o o o N :5 5x 9's #9 INN SON 9" 4‘5" INN [bk 9'\ P’s KN '\ KN '4‘ K" KN KN KN KN '\ v; N <3 \ 9; Control 2137 2147 Tumor volume (mm3) Figure 7.4 Relationship between tumor volume, following injection of RWPE2-W99 cells, and percentage of tumors having the indicated tumor size. This is a graphic representation of the data shown in Table 7.1. Tumor volumes have been divided into four groups. The effects of treatment with 2137 or 2147 are evident in this bar graph. Results indicate that overall, the treatment groups have a larger percentage of small tumors as compared to the control. 171 Table 7.2 This table is a summary of Table 7.1., comparing tumor volumes of RWPE2-W99 cells in control mice and mice treated with 2137 or 2147. Data are shown as percent of tumors in each group having the indicated tumor volume. Tumor volume % of tumors having the indicated size (films) Control 2137 2147 11-110 8 34 21 111-510 58 43 63 511-910 9 23 11 >911 25 0 5 172 Table 7.3 In this table tumor volumes resulting from injection of RWPE2-W99 cells have been divided into nine groups which are arranged from the smallest to the largest. Data are shown as percent of tumors in each group having the indicated tumor volume. Tumor Volume % of mice with tumors with the indicated size (”‘3’ Control 2137 2147 11-60 8 24 0 61-110 0 10 21 111-310 58 33 58 311-510 0 10 5 511-710 9 9 6 711-910 0 14 5 911-1110 0 0 0 1111-1310 25 0 0 1311-1510 0 0 5 173 Control INN01137 O INN01147 46—47 48—47 50-47 Figure 7.5 Nude mice (NSWNU-M nu/nu strain) were bilaterally injected with 250 pl of a cell suspension in Matrigel (cell: Matrigel volume, 121) containing one million RWPE2-W99 cells. Mice were sacrificed 1 1 weeks later. This figures shows (from left to right) four representative mice with tumors from each of the three groups. A. Control mice (22-C, 25-C, 29-C. and 27-C); B. 2137- treated mice (37-37, 34-3 7, 35-37, and 32-37); C. 2147-treated mice (48-47, 50-47, 44-47, and 46-47); arrows point to small tumors. 174 Histology: A histological examination of the tumors shows that RWPE2-W99 tumors are of an undifferentiated type (Figure 7.6). RWPE2-W99 produces slow growing tumors which, after 11 weeks, did not show marked signs of invasion into the surrounding tissue at this time. The tumors show clear boundaries between the tumor and the adjacent connective tissue, fat or muscle. There were no obvious signs of visible metastatic tumors. However, it is possible that if the mice were maintained for a longer period, signs of invasion and metastasis may become apparent. 175 Figure 7.6 H & E-stained sections showing tumor histology. Figures 7.6a and 7.6b: Histology of the RWPE2—W99 tumors in control mice at low (Figure 7.63) and high magnification (Figure 7.6b). The tumor appears to be an undifferentiated tumor with clear margins at the interface with connective and adipose tissue. In Figure 7.6b, many mitotic figures can be seen (arrows). Figures 7.6c and 7.6d show a tumor from a mouse treated with 2137. Examination of the tumor at higher magnification (Figure 7.6d) shows little evidence of cells in mitosis, in contrast to the control tumors. Figures 7.6c and 7.6f show a tumor from a mouse treated with 2147. Examination of the tumor at higher magnification (Figure 7.61) again shows little evidence of cells in mitosis, in contrast to the control tumors. Bar = 10 pm. 176 A preliminary examination of the histological slides suggests that while the control tumors have an abundance of mitotic figures (Figure 7.6b), mitotic figures were seen less frequently in tumors from treated mice. Further evaluation would be required to substantiate this observation. However, this observation appears to be consistent with the smaller tumor size in treated mice. Discussion: Because mortality from some common cancers, such as, lung, breast and prostate, using chemotherapy, has not significantly decreased, it is necessary and sensible to explore other strategies to treat cancer (Spom and Sub, 2000; Quader et al., 2001). The main cause of death among prostate cancer patients is metastasis and bone is one of the most frequent sites to support metastatic tumors. Members of the tetracycline family of antibiotics have potential treatment value for bone metastasis; they inhibit cancer cell proliferation and they are also potent MMP inhibitors (Duivenvoorden et al., 2002). In rats given intravenous cell injection of the Dunning MAT LyLu tumor cells, daily treatment by gavage with CMT-3 resulted in both an increase in survival and a decrease in skeletal and soft tissue metastasis (Selzer et al., 1999). Members of the tetracycline family not only offer potential for the treatment of metastatic cancer, but also primary tumor growth. In a MAT LyLu model of prostate cancer, treatment of rats with CMT-3 by gavage decreased tumor incidence and growth (Lokeshwar, 1999). The growth inhibitory nature of CMTs, has been demonstrated against BPH-1, LNCaP, DU145, PC-3, TSU-Prl and MAT LyLu cell lines (Lokeshwar et al., 1998; Lokeshwar, 1999). In 177 vitro, CMTs have also been shown to induce dose- and time-dependent apoptosis in several tumor cell lines (Lokeshwar et al., 1998; Lokeshwar, 1999). The mode of action of CMTs is often related to their capacity to bind the critical Zn2+ ions in the active centers of the MMP molecules, thus inhibiting MMP activity (Golub et al., 1992; Sorsa et al., 1998). CMT-3 has been shown to reduce invasive activity of tumor cells by specifically inhibiting MMPs without significantly inhibiting their natural inhibitors (Lokeshwar et al., 2002). Matrix metalloproteinases have been repeatedly implicated in tumor cell invasion and metastasis. However, recent evidence suggests that MMPs also play an important role in the establishment and growth of the primary tumor (Chambers and Matrisian, 1997). This is supported by experiments using other MMP inhibitors to block MMP activity. These matrix metalloproteinase inhibitors caused significant inhibition of primary tumor growth both at the subcutaneous and orthotopic site in animal models (Conway et al., 1996; Lokeshwar, 1999; Wang et al., 1994). Similarly in this study, treatment of mice bearing subcutaneous tumors with a CMT, an MMP inhibitor, caused a decrease in tumor size. Consistent with the ability of CMT 2137 and 2147 to inhibit anchorage-dependent growth of RWPE2-W99 in vitro, they were also shown to cause a decrease in tumor size in viva. The mice treated with the test agents had a higher percentage of tumors in the small category (1 1-510 m3) and a lower percentage of tumors in the large category (>500 mm3) as compared to the control mice. It is interesting to note that, in addition to the smaller tumor size in treated mice, tumor sections from the control mice showed a large number of mitotic figures. Mitoses appeared to be less frequent in tumor sections 178 from mice treated with the test agents. The observed cellular effects and the mechanism by which the test agents exert their effects on tumor growth need to be further explored. MMPs have been shown to play an important role in supporting the growth of primary tumors, thus, if CMTs have a negative effect on MMP production, as do other members of the tetracycline family, then this could explain the small tumor size and decreased mitoses observed in mice treated with CMT 2137 and 2147. Although, CMT 2147 appears to be more effective at inhibiting the growth of RWPE2-W99 in cell cultures than CMT 2137, both CMT 2137 and 2147 were shown to inhibit tumor growth in viva. Therefore, CMT 2137 and 2147 demonstrate potential as anticancer drugs and warrant further screening. Acknowledgements Chemically modified tetracyclines, CMT 2137 and 2147, were provided by Tetragenex Pharmaceuticals, Inc., Park Ridge, NJ. Suppport for this project by Tetragenex is acknowledged. 179 Literature cited Bello, D., Webber, M.M., Kleinman, H.K., Wartinger, DD, and Rhim, J .S.: Androgen responsive adult human prostatic epithelial cell lines immortalized by human papillomavirus 18. Carcinogenesis 18:1215-1223, 1997. Chambers, AF. and Matrisian, L.M.2 Changing views of the role of matrix metalloproteinases in metastasis (review). Journal of the National Cancer Institute 89:1260-1270, 1997. Conway, J .G., Trexler, S.J., Wakefield, J.A., Marron, B.E., Emerson, D.L., Bickett, D.M., Deaton, D.N., Garrison, D., Elder, M., McElroy, A., Willmott, N., Dockerty, A.J.P. and McGeehan, G.M.: Effect of matrix metalloproteinase inhibitors on tumor growth and spontaneous metastasis. Clinical & Experimental Metastasis 142115-124, 1996. Duivenvoorden, W.C.M., Popovic, S.V., Lhotak, S., Seidlitz, Hirte, H.W., Tozer, R.G. and Singh, G: Doxycycline decreases tumor burden in a bone metastasis model of human breast cancer. Cancer Research 62:1588-1591, 2002. Golub, L.M., Suomalainen, K. and Sorsa, T.: Host modulation by tetracyclines and their chemically-modified non-antimicrobial derivatives. Current Opinions in Dentistry 2280-90, 1992. Lokeshwar, B.L., Houston-Clark, H.L., Selzer, M.G., Block, N.L., and Golub, L.M.2 Potential application of a chemically modified non-antimicrobial tetracycline (CMT-3) against metastatic prostate cancer. Advances in Dental Research 12:97 -102, 1998. Lokeshwar, B.L.2 MMP inhibition in prostate cancer. Annals New York Academy of Sciences 8782271-289, 1999. Lokeshwar, B.L., Selzer, M.G., Zhu, B., Block, N .L. and Golub, L.M.2 Inhibition of cell proliferation, invasion, tumor growth and metastasis by an oral non-antimicrobial tetracycline analog (COL-3) in a metastatic prostate cancer model. International J oumalof Cancer 98:297-309, 2002. Nemeth, J .A., Harb, J .F., Barroso, Jr., U., He, 2., Grignon, DJ. and Cher, M.L.: Severe combined immunodeficient-bu model of human prostate cancer metastasis. Cancer Research 59:1987-1993, 1999. Quader, S.T.A., Bello-DeOcampo, D., Williams, D.E., Kleinman, H..,K Webber, M.M.: Evaluation of the chemopreventative potential of retinoids using a novel in vitro human prostate carcinogenesis model. Mutation Research 496:153-161, 2001. 180 Selzer, M.G., Zhu, 8., Block, N .L., and Lokeshwar, B.L.2 CMT-3, a chemically modified tetracycline, inhibits bony metastases and delays the development of paraplegia in a rat model of prostate cancer. Annals New York Academy of Sciences 878:678-682, 1999. Sorsa, T., Ding, Y., Salo, T., Lauhio, A., Teronen, O., Ingman, T., Ohtani, H., Andoh, N., Takeha, S. and Konttinen, Y.T.2 Effects of tetracyclines on neutrophil, gingival and salivary collagenases. Annals New York Academy of Science 7322112-131, 1994. Sorsa, T., Ramamurthy, N.S., Vemillo, A.T., Zhang, X., Konttinen, Y.T., Rifldn, BR, and Golub, L.M.2 Functional sites of chemically-modified tetracyclines: inhibition of the oxidative activation of human neutrophil and chicken osteoclast promatrix metalloproteinase. Journal of Rheumatology 252975-982, 1998. Sporn, MB. and Suh, N .2 Chemoprevention of cancer. Carcinogenesis 212525-530, 2000. Wang, X., Fu, X., Brown, P.D., Crimmin, M.J. and Hoffman, R.M.: Matrix metalloproteinase inhibitor BB-94 (Batimastat) inhibits human colon tumor growth and spread in a patient like orthotopic model in nude mice. Cancer Research 54:4726-4728, 1994. Webber, M.M., Bello, D., Kleinman, H.K., and Hoffman, M.P.: Acinar differentiation in non-malignant immortalized human prostatic cells and its loss by malignant cells. Carcinogenesis 18:1225-1231, 1997a. Webber, M.M., Bello, D., and Quader, S.: Immortalized and tumorigenic adult human prostatic epithelial cell lines. Characteristics and applications. Part 2. Tumorigenic cell lines. The Prostate 30258-64, 1997b. Webber, M.M., Quader, S., Kleinman, H.K., Bello-DeOcampo, D., Storto, P.D., Bice, G., de Mendonca-Calaca, W., and Williams, D.E.: An in vitro/in viva model of human cell lines for prostate carcinogenesis and tumor progression. The Prostate 4721-13, 2001. 181 CHAPTER EIGHT SELECTION OF CELL LINES WITH ENHANCED INVASIVE PHENOTYPE FROM XENOGRAFTS OF THE HUMAN PROSTATE CANCER CELL LINE WPEl-NB26 Abstract Prostate cancer is the second leading cause of death from cancer in American men and metastasis is the main cause of death in prostate cancer patients. In order to better understand the disease and to accelerate development of new therapies, in viva models that reflect different disease stages are needed. A family of cell lines, that mimic multiple steps in cancer development and progression, have been developed in our laboratory. The parent, non-tumorigenic, RWPE-l cell line was derived by immortalization with human papillomavirus-18 (HPV-18) (Bello et al., 1997; Webber et al., 1997a). Several tumorigenic cell lines, the MNU cell lines, were derived from RWPE-l by transformation with N-methyl-N—nitrosourea (MNU) (W ebber et al., 1997c). In a tumor progression model, WPEl-NB26 is the most malignant MN U-transformed cell line and it forms metastases in the lungs of nude mice after intravenous injection. Two new cell lines, WPEl-NB26-64 and WPEl-NB26-65, showing more malignant characteristics than the parent WPEl-NB26 cell line, were derived from tumors that resulted from subcutaneous injection of WPEl-NB26 cells into nude mice. The WPEl-NB26-64 and WPEl-NB26-65 cell lines show an increase in anchorage- dependent growth and invasive ability as compared to the parent WPEl-NB26 cells. While the parent WPEl-NB26 cells express barely detectable levels, the new cell lines produce increased levels of matrix metalloproteinase (MMP) MMP-2 and detectable levels of MMP-9. By immunostaining, all three cell lines were positive for cytokeratins Ck5/ 14 and Ckl8. These cell lines, having the same lineage, represent another step in the multi-step process of tumor progression and provide novel and useful cell models for 183 studies on tumor progression and for drug development for the treatment of prostate cancer. Keywords Cell line, metastasis, matrix metalloproteinase, N-methyl-N-nitrosourea, prostate cancer Introduction Prostate cancer is the most common cancer (excluding skin cancer) in men in the United States of America (American Cancer Society, 2004). Prostate cancers show heterogeneity in their composition and diversity in behavior, thus, xenograft models of prostate cancer that reflect different disease stages of prostate cancer are necessary. Xenograft models pemiit one to directly compare the histopathology and molecular biology of the patient-derived specimen and the resulting xenograft tumor. Xenografts have been widely used in cancer studies and some of their applications include: i) validation and testing the utility of therapeutic agents for clinical trials ii) determination of the influence of the microenvironment on gene expression, growth, and behavior of tumor cells within the prostate and other organ sites and iii) studies on the importance of hormonal status and of androgen-independence in tumor progression and metastasis. The majority of commonly used human prostate cancer cell lines have been derived from bi0psies of metastatic prostate cancer and thus, are more appropriate for studies on advanced disease and progression to metastasis (W ebber et al., 1996a; Webber et al., 1997b). Therefore, human prostate cancer cell lines that represent early events in carcinogenesis, tumor progression and metastasis, have been developed in our laboratory. 184 Included in this family of cell lines is the parental, non-tumorigenic, RWPE-l cell line which serves as a control in vitro or as a standardized model in viva when studying prostate carcinogenesis and cancer progression (Bello et al., 1997; Webber et al., 1997a; VVebberetaL,1997c) RWPE-l cells were isolated from the peripheral zone of the normal prostate of a 54 year-old Caucasian man undergoing radical cystoprostatectomy for bladder cancer and immortalized with a single copy of the human papilloma virus-l8 (HPV-18) DNA (Bello et al., 1997; Webber et al., 1997a). The RWPE-l cells, although immortalized, have retained properties exhibited by normal prostate epithelial cells in viva, such as, the ability to undergo acinar morphogenesis in 3-dimensiona1 (3D) Matrigel cultures and in viva upon subcutaneous cell injection in athymic mice. Furthermore, they have the ability to produce PSA upon exposure to androgen (Bello et al., 1997; Webber eta1., 1997a; Bello-DeOcampo et al., 2001a; Bello-DeOcampo et al., 2001b). A family of cell lines, the MNU cell lines, was generated from RWPE-l cells (Figure 8.1) (W ebber et al., 1997c) 185 Human prostatic epithelial cells Immortalized with HPV-18 RWPE-1 cell line Treated with MNU Transformed Cells Injected into nude mice First generation tumors / \ 2A 313 I /\ Cells grown in culture, injected into mice Second generation tumors 1 l 2A2 381 382 / / \ \ Cells grown in culture, plated in agar, colonies isolated, injected into mice / l \ \ Third generation tumors / / \ \ WPEl-NAZZ WPEl-NB‘M WPE1-N811 WPE1-N826 L l The following MNU cell lines were established: . WPEl-NA22, WPEl-NB14, WPEl-NBll, WPEl-NB26 Cells injected into mice Fourth generation tumors WPE1-N826-64 VW’E1-NBZ6—65 Figure 8.1. Derivation of MNU-transformed human prostate epithelial cell lines from RWPE-l, a non-tumorigenic human prostatic epithelial cell line, and the subsequent derivation of WPEl-NB26-64 and WPEl-NB26-65 cell lines. The 2A tumor was derived from treatment with MN U at 50 pg/ml and 3B at 100 pg/ml (Modified from Webber et al., 1997c). 186 RWPE-l cells were treated with MN U, a chemical carcinogen, at 50 or 100 pg/ml. Carcinogen-exposed cells were injected subcutaneously in nude mice and tumors were removed 10 weeks after cell injection. Cells from these tumors were grown in culture to give rise to 2A (50 pg/ml MNU) and 3B (100 pg/ml MNU) cells. These cells were again injected subcutaneously into nude mice and tumors were collected and plated in culture to expand the cell population. Cells were then plated in soft agar. Individual colonies were isolated and expanded and gave rise to the MNU cell lines, which all share a common lineage. These cell lines include: WPEl-NA22, WPEl-NB14, WPEl-NBI 1, and WPEl-NB26 (Figure 8.1). The RWPE-l and the MN U cell lines are unique because they show progression of characteristics from non-tumorigenic, to low, and then to a high level of malignancy which mimic different stages of carcinogenesis and progression as they occur in the human prostate. On the basis of their in vitro and in viva characteristics the MNU cell lines rank in the following order of increasing malignancy: WPEl-NA22 showing the lowest invasion and tumorigenicity, the WPEl-NB14 and WPEl—NBll showing intermediate, and WPEl-NB26 the greatest invasion and tumorigenicity (Webber et al., 1997c). In the present study we show that intravenous injection of WPEl-NB26 cells into nude mice results in the production of distant metastases. We were also successful in isolating two new cell lines from tumors resulting from subcutaneous injection of WPEl-NB26 cells. These two new cell lines designated, WPEl-NB26-64 and WPEl-NB26-65, may be more tumorigenic and metastatic than the parental WPEl-NB26 cell line based on the results presented here. The most significant differences between the parental and tumor-derived cell lines are the increased growth and expression of 187 MMPs in tumor-derived cell lines. The three cell lines, WPEl-NB26, WPEl-NB26-64, and WPEl-NB26-65, described here, provide a model system for studying prostate carcinogenesis, matrix metalloproteinase expression and tumor progression, including metastasis. These cell lines may also be used for testing agents for the treatment of localized and metastatic prostate cancer. Materials & Methods Materials: Keratinocyte serum-free medium (KSFM) with supplements of bovine pituitary extract (BPE) and epidermal growth factor (EGF) (Gibco/In Vitrogen, Grand Island, NY); antibiotic/antimycotic mixture (Gibco/In Vitrogen, Grand Island, NY); Dulbecco’s phosphate buffered formalin (DPBS) (Mallinckrodt Baker, Inc., Phillipsburg, NJ), Matrigel (Collaborative Biomedical Products, Bedford, MA); donor calf serum (DCS) (Intergen, Purchase, NY); trypsin/EDTA (Gibco/In Vitrogen, Grand Island, NY); bovine serum albumin (BSA) (Pierce, Rockford, IL); Falcon T-75 flasks (Becton Dickinson Labware, Franklin Lakes, NJ); Falcon T-25 flasks (Becton Dickinson Labware, Franklin Lakes, NJ); 96-well flat bottom plates (Becton Dickinson Labware, Franklin Lakes, NJ), 4-chamber slides (Becton-Dickinson Labware, Franklin Lakes, NJ); MoAb to cytokeratin 5/14 (Cat. #M0630, Dako, Carpinteria, CA); MoAb to cytokeratin 18 (Cat. #C8541, Sigma, St. Louis, MO); MoAb to AR (Cat. #sc-7305, Santa Cruz Biotechnology, Santa Cruz, CA), MoAb to PSA (Cat. #M0750, Dako, Carpinteria, CA); 3-[4,5-dimethyl thaizol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) (Sigma, St. Louis, MO); Vectastain Elite ABC peroxidase kit (Vector Labs, Burlingame, CA); 3,3’- 188 diaminobenzidine (DAB) substrate kit (Vector Labs, Burlingame, CA); Glacial acetic acid (EM SCIENCE, Gibbstown, NJ); Nucleopore filters (8.0 pm pore size, 13 mm diameter, Fisher Scientific, Chicago, IL); Boyden chambers (Neuro Probe, Inc., Cabin John, MD); HEMA-3 kit (Fisher Scientific); Coomassie brilliant blue R-250 (Gibco/In Vitrogen, Grand Island, NY); Laernrnli sample buffer (BIO-RAD, Hercules, CA); Mini- protean II apparatus (BIO-RAD, Hercules, CA); triton-X (BIO-RAD, Hercules, CA); Bis- Acrylarnide (BIO-RAD, Hercules, CA) Methods Cells and cell culture: The WPEl-NB26 cell line was derived from RWPE-l cells by exposure to N-methyl-N-nitrosourea (MN U). The WPEl-NB26-64 and WPEl-NBZ6-65 cell lines were selected from WPEl-NB26 cells that formed subcutaneous tumors following cell injection in athymic mice. The cells were grown in complete keratinocyte serum-free medium (KSFM) containing 50 pg/ml bovine pituitary extract (BPE), 5 ng/ml epidermal growth factor (EGF) and 1X antibiotic/antimycotic solution. Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO; and subcultured weekly. Growth in nude mice by subcutaneous injection: Cells were tested for tumorigenicity by subcutaneously injecting a cell suspension of two million WPEl-N B26 cells in 250 pl of basal KSFM (without BPE or EGF) containing an equal volume of Matrigel. Cells were injected bilaterally on the dorso- lateral sides of three male, athymic (NSWNU-M) (Taconic Inc., Germantown, NY) nu/nu 189 mice. When tumors became palpable, their size was measured periodically in two dimensions (length and width), using digital calipers, and all measurements were recorded. Tumors were measured and photographed at the termination of the experiment at 14 weeks. Tumor volume (T V) was calculated using the formula: TV = a X 122/2 where a is the longest dimension and b is the width as described by Nemeth et al., (1999). Tumor tissue was removed aseptically from the site of injection and divided into two; one part was processed for cell culture and the other fixed in 10% phosphate-buffered formalin and processed for histology. Growth in nude mice by intravenous injection: Cells were tested for their metastatic ability by intravenously injecting a cell suspension of one million WPEl-NB26 cells in 100 pl of phosphate-buffered saline. Cells were injected intravenously using the tail vein of six male, athymic (NSWNU-M) (Taconic Inc., Germantown, NY) nu/nu mice under anesthesia. Lung and liver tissue was removed from the animals at the time of sacrifice, fixed in 10% phosphate-buffered formalin, and processed for histology. This study was conducted with the approval of the All University Committee on Animal Use and Care (AUCAUC) and all guidelines were followed. Selection of WPEl-NB26-64 and WPEl-NB26-65 cell lines: WPEl-NB26 cells were injected subcutaneously on the left and right dorso-lateral sides of nude mice, and the resultant tumors were harvested by aseptic technique. Tumors were removed 14 weeks after cell injection from two nude mice, #64 and #65, 190 one tumor each and divided into two pieces. One half was fixed in 10% phosphate- buffered formalin for histology and the other half was rinsed with distilled H20 containing 2X antibiotic/antimycotic solution. Tumor tissue for in vitro culture was minced into 1 mm size pieces and digested in full strength trypsin/EDTA to facilitate release of cells. The tumor tissue was placed in the incubator for 2 hours and the tissue suspension was triturated every 30 min. Trypsin/EDTA was neutralized with 2% donor calf serum (DCS) and removed after centrifugation. The pellet was resuspended in complete KSFM and plated in 60 mm plates. Epithelial cell cultures were expanded and designated as WPEl-NB-26-64 and WPEl-NB-26-65 cell lines. Cell morphology in vitro: Cells plated and grown in 4-well chamber slides were rinsed twice with D-PBS and then fixed with 10% phosphate-buffered formalin for five min. Chamber slides were then rinsed in distilled water, and stained with hematoxylin for 1.5 min, and rinsed in 1% Glacial Acetic water to remove excess stain. Following a rinse in tap water, the slides were rinsed with 70% and 95% ethanol prior to staining with Eosin (2 min). Slides were dehydrated in two changes of 100% ethanol, xylol (121, 100% ethanol: xylene), and xylene and mounted in Permount. Immunostaining for cytokeratin expression: Protein expression was detected by a modified avidin-biotin immunoperoxidase Vector protocol, using monoclonal antibodies. The antibody dilutions were made in normal horse serum. The primary antibody dilution used for detection of Ckl8 191 expression was 1:500 and for Ck5/14, 12100. The following sequential steps were conducted at room temperature and cells were rinsed twice with PBS between steps after application of the primary antibody: cells were blocked with normal horse serum for l h; incubated with the appropriate specific antibody for 1 h, followed by biotinylated secondary antibody (12200) for 30 min; treated with 3% H202 for 3 min to quench endogenous peroxidase activity; incubated with the avidin-biotin complex for 30 min; developed with DAB and dehydrated and mounted on alcohol washed slides. Negative controls included incubation with PBS in place of the primary antibody. Immunostaining for PSA and AR expression: Cells grown in chamber slides for immunostaining were pretreated in KSFM medium containing 5 nM mibolerone, a non-metabolizable androgen, for 4 days, beginning 48 h after plating. Absolute ethanol was used as the vehicle for mibolerone. Controls consisted of ethanol-treated cultures. Cells on 4-well chamber slides were processed as described above. The primary antibody dilutions, incubation temperature and times were: PSA, 1250, 4°C, 24 h; AR, 12100, room temperature, 2 h. Negative controls included incubation with PBS in place of the primary antibody. Anchorage-dependent growth in monolayer: Cells were plated in 96-well plates at densities of 625, 1250, 2500, 5000, and 10000 cells/well in 200 pl of KSFM, using 12 wells per cell density, and allowed to attach for 48 h at which time medium was changed. Cells were allowed to grow for five days with medium change every 48 h. Plates were stained with MTT as previously 192 described (W ebber et al., 1997c). Absorbance was measured at 540 nm with a Titertek micr0plate reader. Results represent the average of three experiments. Invasion assay: The invasive ability of WPEl-NB26, WPEl-NB26-64 and WPEl-NB26-65 cell lines was examined using Matrigel coated filters in the Boyden chamber assay (W ebber et al., 1997c). The day prior to running the assay, cells were plated at a density of four million cells in two T-75 flasks. NucleOpore filters were coated with 25 pg Matrigel in 50 pl of distilled water and left to dry overnight at room temperature under sterile conditions. Cells were lifted 24 h after plating, by incubation with 1 pM EDTA (~9 min). Cells were dislodged by tapping, suspended in KSFM medium containing 0.1% BSA, and recovered by centrifugation. NIH/3T3 cell conditioned medium (212 pl) served as the chemo-attractant in the bottom of the Boyden chamber. A cell suspension, containing 200,000 cells in 200 pl medium, was added to the top chamber of five replicate chambers, allowed to remain for 15 min, then overlayed with 650 pl of the medium containing 0.1% BSA. The invasion assay was carried out for 48 hrs at 37°C. The filters were then fixed and stained with HEMA-3 kit. After the non-migrated cells on the Matrigel-coated side of the filter were wiped off, the stain was extracted with 0.1 N HCl and absorbance was measured at 620 nm using a Titertek microplate reader. Percent absorbance for the tumor-derived cell lines were calculated and compared with the parental WPEl-NBZ6 cell line. The results shown represent the average of two experiments. 193 Collection of conditioned medium for MMP activity: WPEl-NB26, WPEl-NBZ6-64, and WPEl-NB26-65 cells were plated in complete KSFM at one million cells/60 mm culture plate and placed in the incubator. After 48 h, the medium was removed and cells were washed once with D-PBS. The medium was replaced with basal KSFM (without BPE or EGF). The conditioned medium was collected 48 h later and centrifuged to remove cell debris. The supernatant was mixed with 2.0% Triton X-100 to obtain a final concentration of 0.01% and analyzed by SDS-PAGE zymography to detect MMP-2 and MMP-9 activity. SDS-PAGE zymography: Analysis of the levels of MMPs released into serum-free medium by WPEl-NBZ6, WPEl-NBZ6-64, and WPEl-NB26-65 cells was performed using SDS- PAGE zymography. Conditioned media were collected by plating cells in complete KSFM at a density of one million cells/60 mm plate. After 48 h, the medium was slowly removed and each plate was washed once with 3 ml of D-PBS. Next, 2.2 ml basal KSFM (no (BPE or EGF) was added to each dish. After 48 h, the conditioned media were removed from each dish, added individually to sterile 15 ml centrifuge tubes and centrifuged at 2500 rpm for 5 minutes to remove cell debris. The supernatant was transferred to new sterile 15 ml centrifuge tubes. To each tube 2.0% Triton-X 100 was added at a final concentration of 0.01%, the media were gently vortexed to mix, and aliquoted into sterile 200 pl Eppendorf tubes. The protein content of each sample was determined using the Lowry High protein assay. Samples of conditioned media were stored at -80°C until needed. 194 To determine the MMP-2 and MMP-9 activity of these cell lines, 0.75 mm thick 10% acrylamide gels, containing 0.1% gelatin were prepared for SDS—PAGE zymography. The gels were allowed to polymerize overnight at 40C. Conditioned media containing 8 pg of protein in 4 p1 of Laemmli Sample Buffer were loaded into each well. Using a Bio-Rad Mini Protean II apparatus, the gels were electrophoresed (4°C) at 150 volts in approximately 800 ml of IX running buffer (24 g Tris base, 115.2 g Glycine Bio- Rad), pH 8.3. After electrophoresis for 1.5 h, the gels were soaked in 2.5% Triton X-100 on a shaker for one hour, changing the solution after 30 minutes, to eliminate 'SDS. The gels were then rinsed with distilled water and placed in Tris-HCl soaking buffer (50 mM Tris-HCl, 200 mM NaCl, 5 mM CaClz, 0.02% Brij-35, Bio-RAD), pH 7.5, to renature protease activity. After 24 h incubation in soaking buffer at 37°C, the gels were rinsed in distilled water in preparation for staining. The gels were stained for 20 minutes with 0.25% Coomassie Brilliant Blue R-250 stain and then de-stained after which clear bands of digested gelatin were visible against a blue background. The gels were briefly rinsed in distilled water and scanned. The area of each band was measured densitometrically to determine MMP-2 and MMP-9 activity. Statistical analysis: All of the results in this study were obtained from at least two independent experiments. Results are expressed as averages with standard error of the mean (SEM). Differences between the means were considered significant if P<0.05. The results obtained from the in vitro anchorage-dependent growth and Boyden chamber invasion assays of each cell line were compared using one-way analysis of variance (AN OVA) 195 and Tukey-Kramer multiple comparison tests. GraphPad InStat 3 was used for these analyses. Results Histology of xenografts in nude mice: Tissue samples were collected 14 weeks after subcutaneous injection of cells into nude mice and prepared for histology. WPEl-NB26 cells formed large tumors in all mice injected subcutaneously (Figure 8.2A). The average volume of the tumors from which WPEl-NBZ6-64 and WPEl-NB26-65 cells were derived, were approximately 280.0 and 460.0 mm3, respectively. Figure 8.2B shows the histology of the two tumors. 196 Figure 8.2. 8.2A. Nude mice (NSWNU-M) were bilaterally injected with 250 pl of a cell suspension in Matrigel (cell: Matrigel volume, 121) containing two million WPEl-NBZé cells. Mice were sacrificed 14 weeks later. This figure shows two mice with tumors from which the WPEl-NB26-64 and WPEl-NB26-65 cell lines were derived. Bar = 1 cm. 8.23. Histological sections of the (a) WPEl-NB26-64 and (b) WPEl-NB26-65 tumors. H & E, Bar = 20 microns. 8.2C. Histological sections of mouse lung tissue: (a) Normal area of mouse lung tissue, (b) Necrotic WPEl-NB26 prostate tumor cells in a blood vessel (arrow) of mouse lung at 20 weeks after intravenous cell injection, (c) WPEl-NBZ6 cells (arrow) surrounded by hyperplastic, fibrous connective tissue in the lung of a mouse at 20 weeks after intravenous cell injection, (d) shows a higher magnification of the metastasis in (c). H & E, Bar = 20 microns. 8.2D. Morphology of (a) WPEl-NBZ6, (b) WPEl-NBZ6-64, (c) WPEl-NB26-65 cells. H & E, Bar = 20 microns. 197 Histology of metastases in nude mice: Tissue samples were collected 140 days after intravenous injection of WPEl-NB26 cells in nude mice. Histopathology of the lungs of nude mice showed 2/5 (40%) mice injected intravenously with one million WPEl-NB26 cells, established metastatic tumors. Figure 8.2C shows histology of a normal area of the lung (Figure 8.2C,a), necrotic tumor cells in a blood vessel in the lung (Figure 8.2C,b), and of metastatic tumors in lung tissue of mice (Figure 8.2C,c-d) given intravenous cell injection of WPEl-NB26 cells. The metastatic tumor shown in 8.2c and 8.2d is surrounded by fibrous tissue formed in reaction to the presence of tumor cells. Cell morphology in vitro: Stepwise changes in cell morphology have been observed in the MNU family of cell lines, from typical epithelial cells of the WPEl-NA22 cell line showing low tumorigenicity, to a more elongated morphology of the WPEl-NB26 cell line showing high tumorigenicity, with the other cell lines having an intermediate morphology (W ebber et al., 1997c). Similarly, both tumor-derived cell lines, WPEl-NB26-64 and WPEl-NB26-65, show a more elongated and spindle-shaped morphology compared to WPEl-NB26 cells (Figure 8.2D). Immunostaining for cytokeratin expression: The expression of Ck18 and Ck5/14, two epithelial cell markers, was examined by immunocytochemistry. Results show that WPEl-NB26, WPEl-NB26-64, and 198 WPE1-NBZ6-65 cells all express Ck18 and Ck5/l4 (Figure 8.3), which confirms their common epithelial origin. WPE1-N826 ,1 i“ WPE1-N826-64 a] WPE1-flames CK5I14 ' ‘11 e M ~_______} 1 r l 4 ‘2 2 . _ ,I‘rbr’r‘i ‘ NT???» Q i U"? _; Figure 8.3. Characterization of WPE1-NB26, WPE1-NB26-64, and WPE1-NBZ6-65 cells on the basis of cellular proteins. Proteins were detected by immunoperoxidase staining. (a-c) positive staining for cytokeratin 18, the inset in each is a control lacking primary antibody; (d-f) positive staining for cytokeratin 5/14, the inset in each is a control lacking primary antibody. Bar = 20 microns. Expression of prostatic epithelial cell markers in cells: The expression of AR and PSA, two prostatic epithelial cell markers, was 199 examined by immunocytochemistry. Results show nuclear staining for AR (Figure 8.4A), as well as, cytoplasmic expression of PSA (Figure 8.43) in WPE1-N326, WPE1-NBZ6-64, and WPE1-NBZ6-65 cells. An increase in AR and PSA expression was induced in all three cell lines after treatment with 5nM mibolerone for six days. WPE1 -N 826 WPE1-N826-64 WPE1-N826-65 A. , v. ‘o ,5 0 .. I a: < B.' <. a) a. Figure 8.4. Immunostaining for androgen receptor (Figure 8.4A) and PSA (Figure 8.4B) demonstrates androgen responsiveness and prostatic epithelial origin of WPE1-N826, WPE1-NBZ6-64, and WPE1-NBZ6-65 cell lines. Cells were treated with 5 nM mibolerone for 6 days. Positive nuclear staining for AR is shown in 8.4A,a,b,c for all three cell lines. Panel al shows only weak nuclear staining in untreated control. Panel a2 and other insets are negative controls lacking primary antibody. Positive cytoplasmic staining for PSA is shown in 8.4B,a,b,c for all three cell lines. Panel a1 shows very weak staining in untreated control. Panel a2 and other insets are negative controls lacking primary antibody. Bar = 20 microns. Anchorage-dependent growth in monolayer: Anchorage-dependent growth of the tumor-derived cell lines, WPE1-NBZ6-64 and WPE1-NBZ6-65 in monolayer culture is shown in Figure 8.5A and is compared with that of the parental WPE1-NB26 cell line. Both tumor derived cell lines, 200 WPE1-NB26-64 and WPE1-NBZ6-65, grow more rapidly than the parental WPE1-NB26 cell line. The difference in the growth of WPE1-NB26-64 and WPE1-NB26-65 cells, in comparison to WPE1-N826 cells, is very significant at each cell density (p<0.001). 201 3’ 1.2 - 75‘ 1'0 ‘ --<>---WPEl-NBZ6 =1 —O—WPEl-NB26-64 ¢ 0.8 '- 30: ---A-- WPE1-NBZG-65 8 0.6 - 9’0 5 -" / "g 0.4 - .z' U) .D < 0.2 - 0,0 . . . T . . r . . 0.625 1.25 2.5 5 10 Number of Cells per Well (X 1000) B. WPE1-NB26 WPE1-NB26-64 it. WPE1-NBZ6-65 133% I I T I 0 50 100 150 Percent Invasion ’ Figure 8.5. 8.5A. A comparison of the anchorage-dependent growth of WPE1-NB26, WPE1-NB26-64, and WPE1-NB26-65 cell lines. Cells were plated at densities of 625, 1250, 2500, 5000 and 10,000 cells per well in 96-well plates in complete KSFM. Plates were stained with MTT five days after plating. Absorbance values were measured at 540 nm and plotted i SEM. Results represent the average of 3 experiments. The growth of both WPE1-NB26-64 and WPE1-NB26-65 cell lines are significantly greater (p