THE ROLE OF PKC EPSILON IN BREAST CANCER CELL SURFACE POLYFIBRONECTIN ASSEMBLY AND PULMONARY METASTASIS A Dissertation Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy by Lynn Yu Ling Huang May 2009 © 2009 Lynn Yu Ling Huang ROLE OF PKC EPSILON IN BREAST CANCER CELL SURFACE POLYFIBRONECTIN ASSEMBLY AND PULMONARY METASTASIS Lynn Yu Ling Huang, Ph. D. Cornell University 2009 Fibronectin polymerized and assembled into globules on the surface of breast cancer cells has been shown to mediate metastasis to the lungs. Disruption of the binding between these tumor-cell surface-associated fibronectin, or polymeric fibronectin (polyFn), and dipeptidyl peptidase (DPPIV) negatively impacts breast cancer lung colonization. This dissertation was designed to characterize the molecular mechanisms underlying assembly of surface coated polyFn to understand the regulation of breast cancer cell surface polyFn. In order to investigate the signaling pathways involved in polyFn assembly, we utilized pharmacological inhibitors for various signaling pathways that have been proposed to regulate cellular function. Mutant plasmid expression, siRNA knockdown of proteins and biochemical analyses were carried out to identify the particular kinase involved in regulating polyFn. The ability of rat breast cancer cells to assemble polyFn was impaired by disruption of PKCε function. Knockdown of fibronectin also abolished polyFn formation. Cells with knockdown of either PKCε or fibronectin both had diminished ability to colonize the lungs in rat tail vein injection metastasis model. To further substantiate the role of PKCε in lung metastasis-mediating polyFn, mass spectrometry analysis was performed on the PKCε complex. Two proteins were identified and tested with corresponding inhibitors to relate cellular function to polyFn assembly. Detailed biochemical analyses, immunofluorescent microscopy, mutant protein expression and pharmacological inhibition demonstrated that the subcellular localization of PKCε complex with actin cytoskeleton has a great impact on polyFn assembly. A focal adhesion kinase was also identified to be involved. Therefore, PKCε exhibits its effects on polyFn by controlling the activation of downstream signaling proteins. In summary, the presence of endogenous cellular fibronectin and catalytically competent PKCε is a prerequisite of successful polyFn assembly. Proper subcellular localization of PKCε-actin complex facilitates PKCε downstream signaling, most likely to Pyk2, promoting the formation of polyFn surface formation. These results present a novel potential signaling pathway for therapeutical intervention in breast cancer metastasis. BIOGRAPHICAL SKETCH Lynn Yu Ling Huang was born on April 11, 1980, in South Bend, Indiana, USA. She attended a year of elementary school in California and received the remainder of elementary education in the bilingual department of National Experimental High School (NEHS) in Hsinchu, Taiwan, where she was instructed mostly in English. She transferred to the regular high school department of NEHS for preparation of the national college entrance exam and was accepted to the Department of Veterinary Medicine of National Taiwan University, Taipei, Taiwan. She completed veterinary education with one year of internship while working as a part time intern veterinarian at the animal hospital ward and graduated with a Bachelor of Veterinary Medicine (B.V.M.) in 2003. After graduation, she worked as a lab assistant researcher in Academia Sinica, Taipei, Taiwan and National Tsing-Hua University and passed the national examination for licensed veterinarians. In May 2004, she began her doctoral studies in the lab of Dr. Bendicht U. Pauli in the Cancer Biology Laboratories of the Department of Molecular Medicine, College of Veterinary Medicine, Cornell University, Ithaca, NY. iii For my lovely family members Who have stood behind me through the ups and downs of my life Who Always had faith in me iv ACKNOWLEDGMENTS First and foremost I would like to thank the chairman of my graduate committee, Dr. Bendicht U. Pauli, whose expertise and support has made an outstanding research environment for my dissertation studies. His scientific counseling has been invaluable during the years of study. Without Dr. Pauli’s utmost support, understanding, and encouragement, I would not have been able to complete the work presented here. I am also grateful for the guidance and support from the remaining members of my committee, Drs. James Casey, Roy Levine, and Robert Weiss. I would like to extend my thanks to previous and current lab members of the “Pauli lab”, who has helped created a wonderful environment to learn research techniques and exchange ideas. I thank Dr. Hung-Chi Cheng for assisting me during my earliest years to gain a solid ground in biochemistry and cellular assays, providing the basis for my research in this lab. I also thank Dr. Randolph C. Elble for guiding me through the various approaches in plasmid construction and offering precious advice on graduate life. I thank Dr. Mossaad Abdel-Ghany for his active help and input in various biochemical techniques. I thank Dr. Keyi Liu, who has offered not only scientific interactions but also valuable insights in life. I thank Dr. Roy Levine for the wonderful support as a lab neighbor and for providing advice on experimental design. I thank Dr. James Casey for offering rich encouragements and advice. I also want to thank the lab members of Dr. Jun-lin Guan, Dr. Richard Cerione and Dr. Yung-fu Chang, who have offered accommodating assistance to the lab and at various points of my project. Finally, I would like to thank Dr. Rodman Getchell for his support and guidance on operating the FACS machine, an integral part of my research. v TABLE OF CONTENTS BIOGRAPHICAL SKETCH DEDICATION ACKNOWLEDGMENTS TABLE OF CONTENTS LIST OF FIGURES LIST OF TABLES LIST OF ABBREVIATIONS LIST OF SYMBOLS CHAPTER I BACKGROUND AND SIGNIFICANCE A. STRUCTURE OF FIBRONECTIN B. FIBRONECTIN SPLICE VARIANTS AND FUNCTIONS (1) EDA DOMAIN (2) EDB DOMAIN (3) V REGION C. PHENOTYPES OF FIBRONECTIN KNOCKOUT MICE D. CELL SPECIFIC DELETION OF FIBRONECTIN E. FIBRONECTIN IN MAMMARY DEVELOPMENT F. ROLE OF FIBRONECTIN IN BREAST CANCER G. FIBRONECTIN MATRIX ASSEMBLY iii iv v vi x xii xiii xv 1 2 4 7 7 8 8 10 11 12 13 vi (1) INTERACTIONS OF FN AND INTEGRINS (2) INTERACTION OF FN AND OTHER MATRIX COMPONENTS (3) FN SELF ASSOCIATION (4) INTERACTION OF FN WITH DPPIV 15 16 16 17 H. FN IN EPITHELIAL TO MESENCHYMAL TRANSITION AND 19 INVASION I. FN IN METASTASIS 21 J. SITE-SPECIFIC METASTASIS 23 K. POLYMERIC FN ON SUSPENDED BREAST CANCER CELLS 27 MEDIATES LUNG METASTASIS L. SIGNALING EVENTS IN FN ASSEMBLY 28 M. SUMMARY AND GOALS OF THIS STUDY 34 N. REFERENCES 35 CHAPTER II PKCε MEDIATES FIBRONECTIN ASSEMBLY IN MTF7 TO PROMOTE PULMONARY METASTASIS 48 A. ABSTRACT 49 B. INTRODUCTION 49 C. MATERIAL AND METHODS 52 vii D. RESULTS 59 E. DISCUSSIONS 85 F. REFERENCES 90 CHAPTER III POLYMERIC FIBRONECTIN ASSEMBLY ON THE SURFACE OF BREAST CANCER CELLS IS DEPENDENT ON SUBCELLULAR LOCALIZATION OF PKC ΕPSILONACTIN COMPLEX AND PYK2 ACTIVITY 95 A. ABSTRACT 96 B. INTRODUCTION 97 C. MATERIAL AND METHODS 99 D. RESULTS 104 E. DISCUSSIONS 122 F. REFERENCES 131 CHAPTER IV CONCLUDING DISCUSSIONS 134 A. SUMMARY OF RESULTS 135 B. FUTURE EXPERIMENTS 137 C. REFERENCES 142 viii APPENDIX ROLE OF FIBRONECTIN IN MAMMARY GLAND MORPHOGENESIS A. ABSTRACT B. INTRODUCTION C. MATERIAL AND METHODS D. RESULTS E. DISCUSSIONS F. REFERENCES 144 145 146 148 152 171 179 ix Figure 1.1 Figure 1.2 Figure 1.3 Figure 1.4 Figure 1.5 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 2.7 Figure 2.8 Figure 2.9 Figure 2.10 Figure 3.1 Figure 3.2 Figure 3.3 LIST OF FIGURES Schematic diagram of modular structure of FN dimer. 3 Fibronectin alternative splice sites and sites of integrin 5 receptor and cell binding. Major steps of Fn fibril assembly. 14 The main steps in the formation of a metastasis. 25 Model of Fn matrix assembly. 29 PolyFn globules expressed on the surface of suspended 60 (blood-borne) MTF7L breast cancer cells. Inhibitors of protein synthesis and secretion prevent polyFn 63 assembly. Effect of siRNA knockdown of cFn on polyFn assembly in 65 the presence of 20% FBS. Fn and PKCε siRNA-knockdown inhibit pulmonary 67 metastasis. Inhibitors of PLC and PKC diminish polyFn assembly. 70 Effect of various PKC inhibitors on polyFn assembly. 73 Effect of RD-PKCδ, RD-PCη, and RD-PKCε expression and 76 PKCδ and PKCε siRNA-knockdown on polyFn assembly. Analysis of stable clones selected from siRNA-PKCε 78 transfected MTF7L cells. Membrane translocation and phosphorylation of PKCε. 81 Effect of PMA on PKCε and polyFn assembly. 83 PKCε associates withβ-actin. 105 PKCε colocalization with F-actin in activated MTF7 cells is 108 time and cytoskeletal integrity-dependent. PKCε and actin association. 111 x Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 4.1 Figure A.1 Figure A.2 Figure A.3 Figure A.4 Figure A.5 Figure A.6 Figure A.7 Figure A.8 Figure A.9 PolyFn assembly is dependent on cytoskeletal integrity. 113 PKCε actin binding motif is not essential for polyFn 116 assembly. PYK2 kinase activity is required for polyFn assembly. 120 Inhibitors of myosin ATPase activity had no effect on polyFn 124 assembly. Diagram of cellular processes in polyFn assembly. 128 Cancer cell secreted factors in polyFn assembly. 139 Conditional Knockout of Fn in Mammary Epithelia. 154 Mammary Gland Whole Mounts from Fnfl+/+ and FnMEp–/– 157 mice. Quantitative Aspects of the Effects of Fn-knockout on MG 159 postnatal development. Horizontal Section of Mammary Glands from Fnfl+/+ and 162 FnMEp–/– Mice Stained with H.&E. Mammary Gland Volume Density of the Mammary Fat Pad. 164 Cell Proliferation Assay. 166 MG Immunohistochemistry. 169 Neovascularization of the Lobulo-Alveolar Compartment in 172 Fnfl+/+ Mice. Fn-Signaling via Itgb1 and FAK during MG Development in 174 15-day Pregnant Fnfl+/+, but not FnMEp–/– Mice. xi Table 1 LIST OF TABLES Stimulators tested for polyFn assembly. 138 xii LIST OF ABBREVIATIONS ABD Actin binding domain DPPIV Dipeptidylpeptidase IV ECM Extracellular matrix EDA Extradomain A EDB Extradomain B EGF Epidermal growth factor EMT Epithelial to mesenchymal transition ERK Extracellular signal-regulated protein kinase FACS Fluorescence Activated Cell Sorting F-actin Filamentous actin FN, Fn Fibronectin GFP Green fluorescent protein Igtb1 Beta1 integrin MAP Mitogen-activated protein MDA-MB-231 Human breast adenocarcinoma cell line MEp Mammary epithelial cell MFP Mammary fat pad MG Mammary gland MMTV Mouse mammary tumor virus MTF7 Rat breast carcinoma cell line PKC Protein Kinase C PolyFn Polymeric fibronectin PYK2 Proline rich kinase 2 RD-PKC Regulatory domain of PKC xiii RGD RT-PCR SDS-PAGE TEB TGFβ1 VEGF V region WT Arginine-glycine-aspartic acid Reverse trascriptase polymerase chain reaction sodium dodecyl sulfate polyacrylamide gel electrophoresis Terminal end bud Transforming growth factor β1 Vascular endothelial growth factor Variable region Wild type xiv LIST OF SYMBOLS α …………… Alpha β …………… Beta δ …………… Delta ε …………… Epsilon η …………… Eta θ …………… Theta ζ …………… Zeta xv CHAPTER I FIBRONECTIN IN CANCER PROGRESSION AND METASTASIS 1 INTRODUCTION The extracellular matrix (ECM) supports specific interactions with cell surface receptors and is essential for tissue function. Fibronectin (Fn), an abundant ECM glycoprotein, is involved in a variety of cellular functions and disease processes. Fn regulates cellular functions such as cell adhesion, cell migration and cell proliferation. Fn is necessary for proper embryonic development and tissue repair and disregulation of Fn leads to diseases of the skin, liver, lung, cartilage and bones as well as cancer progression. The integral role of Fn in maintenance of normal cell function cannot be undermined. In this review, we will take a look at the structure and domains of Fn, consequences of Fn deletion, Fn matrix assembly and provide a brief overview of Fn function in cancer progression and metastasis. STRUCTURE OF FIBRONECTIN Fibronectin (Fn) is a glycoprotein that exists as a dimer comprised of two 220250 kDa subunits connected at the C-termini by disulfide bonds (1,2) (Fig 1.1). Each subunit consists of repeating modules (Fn repeats) that includes type I, type II and type III repeats. Type I repeats contain two disulfide bonds and contain approximately 40 amino acids; type II repeats are about 60 amino-acid residues in length and contains two intrachain disulfide bonds; and type II repeats are 90 residues in length and have no disulfide bonds. Fn is comprised of 12 type I repeats, 2 type II repeats, 15-17 type III repeats and a variable(V) region not homologous to other areas on Fn (1,2). These domains contain multiple binding sites for heparin and for various cell surface receptors that modulate cellular function. 2 Magnusson, M. K. et al. (1998) Arterioscler Thromb Vasc Biol 18(9), 1363-1370 Figure 1.1: Schematic diagram of modular structure of Fn dimer (2). 3 The 2 type II repeats that undergo alternative splicing are termed EDA (ED, extradomain) and EDB (Fig 1.2). Fn can be subdivided into 2 forms based on solubility: the more soluble plasma Fn and the less soluble cellular FN. Plasma FN (pFn), which is synthesized in hepatocytes, has a simple splicing pattern and contains neither EDA nor EDB while cellular Fn (cFn), which is synthesized in tissues, contains either or both of EDA and EDB and is a more heterogenous group of Fn isoforms as a consequence of cell-type and species-specific alternative splicing (3). In pFn, only one of the subunits contains Variable (V) region whereas most cFn contains V region (1,2). Individual type III repeats have only 20-40% identity in amino acid sequence but a high degree of structural homology. Type III repeats are organized into 7 antiparallel β strands with exposed loops within these strands. Fn domains may be structurally modified, which can expose cryptic sites induced by mechanical forces stretching the Fn molecule, proteolysis, or incorporation of one or both of EDA and EDB (4). Exposure of cryptic site allows Fn molecule self association (5) and has a potential role in serving as nucleation sites for Fn polymerization and fibril assembly (6). FIBRONECTIN SPLICE VARIANTS AND FUNCTIONS Fibronectin was one of the first genes to be reported to be alternatively spliced and is one of the best studied models of alternative splicing (4). During transcription, the EDA and EDB exons can be included or excluded from Fn mRNA while the V region (also called IIICS, type III connecting segment) can be completely included(V120), partially included, or excluded(V0), depending on species (4) (Fig 1.2). On top of alternatively spliced forms, in zebrafish, 2 Fn genes are present (FN1a 4 White, E. S. et al. (2008) J Pathol 216(1), 1-14 Figure 1.2: Fibronectin alternative splice sites and sites of integrin receptor and cell binding: Splicing of EDA and EDB is similar in all species (total inclusion or exclusion) while splicing of the variable region is species specific (4). 5 and FN1b) (7). However, the total number of Fn isoforms still correlates with evolutionary scale. The 2 independent genes in zebrafish produce only 1 V-region variant in FN1a and 2 V variants in FN1b and a total of 5 Fn isoforms (8). In other species, frogs and chickens have 2 V variants (generates up to 6 Fn isoforms), rats and mice have 3 (produces 12 Fn isoforms) and humans have 5 V variants that generates up to 20 Fn isoforms (4). Incorporation of the alternatively spliced region in Fn is increased during embryonic development and decreases dramatically after birth and with aging (4,8,9). A comparison of fetal and nonfetal fibroblasts shows a loss of EDB inclusion in nonfetal cells (10). Analysis of rat embryo mRNA reveals that the entire V region (V120) is included in most FN mRNA while rats in later developmental stages contain V0, V95 and V120, indicating developmentally regulated splicing (11). RNA from mice of different ages after birth up to 24 months revealed the increased inclusion of V0 and decreased inclusion of V120 with aging, while V95 showed no change (12). Under certain situations, the “embryonic” splicing pattern is re-established in adults, such as during tissue repair, tissue fibrosis and angiogenesis. In skin wound healing, EDA and EDB are present in the Fn mRNA in the cells at the base of the wound (13). During pheripheral nerve regeneration, V25 inclusion was increased while V0 levels remain low (14). Cell type specificity of alternative splicing patterns can be seen more clearly in adults. In large vessel wall endothelial cells, inclusion of EDA but not EDB is seen in Fn. The same EDA+EDB- splice form exists within the uterine endometrium. On the other hand, chondrocytes contain EDB+EDA- Fn mRNA (9). The differential distribution of different Fn isoforms certainly suggests different roles for each of the isoforms in cellular function. 6 (1) EDA DOMAIN EDA domain has numerous functions which include cell adhesion, wound healing, matrix assembly, dimer formation, protein secretion, cell differentiation, tissue injury and inflammation, cell cycle progression and various disease processes (4). EDA+ Fn is implicated in disease processes such as rheumatoid arthritis (15). A low concentration of pFn and EDA-only polypeptide were synergistic for cell attachment and spreading and lead to cytoskeletal reorganization and focal contact formation compared to either substrate used alone(16). EDA+ Fn was found abundantly expressed only in the ECM of scarred human eyes and not normal conjunctiva (17), indicating a role in wound healing. In airway epithelial cells that secrete Fn from both apical and basal surfaces, EDA+ Fn were only secreted apically, demonstrating a role in directing secretion (18). On the other hand, levels of EDA+ Fn are elevated in plasma and tissues of patients with certain disorders such as pulmonary fibrosis (19), psoriasis, rheumatoid arthritis, diabetes and cancer (4), but the functional role of EDA+ Fn has not be clarified. (2) EDB DOMAIN EDB+ Fn has a positive role in matrix assembly, as it is more efficiently incorporated into Fn matrix, along with EDA+ Fn, than EDA-EDB-V-Fn and V+Fn (20). Studies on EDB-/- mice revealed no significant phenotype (21-23), but a mild in vitro effect on matrix assembly and proliferation (23). EDB-/- embryonic fibroblasts grew slower and produced shorter and thinner FN fibrils compared to normal cells (23). On the other hand, EDB+ Fn is lost from many cell types upon completion of development and its expression correlates closely to cell proliferation (9). Recently, EDB+ Fn was found to reside in neoplastic tissue blood vessels but absent from mature/adult tissues, highlighting its role in angiogenesis (24). EDB+ Fn may 7 potentiate and assist in vascular development, suggested by the findings that EDB+ Fn can increase VEGF expression and endothelial proliferation (24), even though in EDB-/- mice a compensatory pathway might exist for angiogenesis. EDB+ Fn is upregulated in the serum of diabetic retinopathy patients (25) and is expressed in hematologic tumors such as Hodgkin and non-Hodgkin lymphoma (26). Future research on these disease models of EDB+ Fn will be able to elucidate the other potential cellular functions of this Fn splice variant. (3) V REGION The V region of Fn contains integrin recognition sites that promote cell spreading and migration (9,20). Human Fn contains two cell binding sites within the V region: LDV in V25 (CS1) and REDV in V31 (CS5), both sequence are recognized by α4β1 integrin. LDV is also found in Xenopus, rat and chicken Fn. REDV is not conserved and is RGDV in rat and bovine. The V region mediates tumor invasion, motility and spreading of squamous cell carcinoma (27) and hepatocellular carcinoma (28). Up regulation of V25 has been shown to enhance nerve repair (29). Increased exclusion of the V region has been observed in early disease processes in disease models of lupus nephritis and immune complex nephritis (30). In human primary fibroblasts, the V region induces apoptosis and concomitant transcriptional decreases of c-Myc, p53, p21 and bcl-2 (31). Other roles in Fn secretion and clot formation have also been suggested (32). PHENOTYPES OF FIBRONECTIN KNOCKOUT MICE Targeted mutations in the Fn gene in different mouse models have been used to complement experimental data by Fn expression. The first Fn-null mice were 8 generated by removing the translation initiation codon and part of the 5’exon to disrupt the Fn gene (33). Mice with heterozygous mutation (+/-) are fertile and normal, even with only half the normal level of FN (33). However, homozygous mutations were embryonic lethal. Fn is not required for blastocyst formation and abnormalities in the Fn-null embryos first appear after the onset of gastrulation, which includes shortened anterior-posterior axis, mesoderm deficiencies and distorted ectoderm. These embryos have an abnormal or no heart, no notochord and no somites (33). Heart defects develop depending on genetic background of Fn-null mice: the C57/BL6 background developed a less severe heart and blood vessel phenotype in which the heart tube forms normally while the 129S4 background is affected earlier in development and causes cardia bifida (34). More recently, 5 novel modifier genes were identified on chromosome four that were linked to the difference in embryonic heart development by comparing the above two genetic backgrounds in mice with high throughput SNP genotyping (35). The specific functions of these genes are yet to be determined. Fn-null EDB mutations (EDB-cDNA fusion replacement, no introns) produced similar embryonic lethal and mesoderm defects as the Fn-null mice (36) while EDBnull mutation (deletion of EDB exon) created no obvious phenotype in various healing, organogenesis, tumorigenesis and thrombosis models (21-23). Conditional pFn knockout using ubiquitous Mx-cre recombinase revealed increased neuronal apoptosis and brain infarction following cerebral ischemia treatment but no effect on wound healing or hemostasis (37). Roles for pFn in thrombosis, angiogenesis and bacterial virulence were also suggested in similar pFn knockout models (38-40). Deletion of EDA exon produced mice with reduced atherosclerosis but no differences in tumor growth or thrombus formation was detected (22,23,41). Mice without the EDA exon in the FN protein (EDA-null) and mice having constitutive inclusion of the 9 EDA exon had reduced life spans, while EDA-null mice had wound healing defects (42). Mice engineered with constitutive inclusion of EDA into the FN molecule had increased thrombosis and thromboembolism, suggesting a pro-thrombotic role for EDA+Fn (43). Roles of EDA+Fn in motor coordination and lung fibrosis were also suggested (19,44). On the other hand, double knockout of EDA and EDB was embryonic lethal with cardiovascular defects similar but less severe than that of FN-null mice. Deletion of both EDA and EDB exons had no effect on protein synthesis, secretion or cell surface deposition of Fn, indicating embryonic lethality was due specifically to the lack of EIIIA and EIIIB exons from Fn (35). Disruption of the cell binding site RGD by mutation to RGE generated was embryonic lethal at E10 (45), suggesting that the majority of Fn cellular function is mediated through RGD interaction with integrins. Assembly of Fn was normal but embryos had shortened posterior trunk, absent tail bud-derived somites and severe vascular defects similar to the phenotype of α5integrin deficient mice (45). Although these total knockouts have provide valuable clues for Fn function, with newly available technology and well characterized promoters, tissue specific Fn deletion models will be able to further dissect the specific roles of Fn in various disease models. CELL-SPECIFIC DELETION OF FIBRONECTIN Cell-specific deletion of Fn is used to determine roles of Fn in different organs or cell types without the detrimental effects of whole mice Fn knockout. Mice engineered to constitutively express EDA+Fn had a significant decrease of Fn levels in plasma and most tissues while hepatocytes modified to produce EDA+ Fn have normal extracellular matrix Fn levels but secrete less soluble Fn (42). A liver-specific EDA- 10 exon deletion in these mice restored Fn levels in both plasma and tissues, indicating that a fraction of tissue Fn is plasma derived (46). Future work will be able to reveal the other functional significance of Fn in mouse knockout models. FIBRONECTIN IN MAMMARY GLAND DEVELOPMENT The histologic structure of mammary glands varies depending on sex, age and physiologic status. The basic mammary gland structure consists of lobules that consist of several branching intralobular ducts that end in alveoli (acini). During pregnancy and lactation, the mammary gland undergoes striking structural and physiologic change, such as acinar growth, as a result of the synergistic action of various hormones, which includes estrogen, progesterone, prolactin and human placental lactogen (47). Fn levels are regulated by estrogen and progesterone and change considerably during mammary gland development, increasing to 3-fold at puberty and maintains at high level during pregnancy and lactation (48). Fn is expressed around the basal cell surface of acini but downregulated in the late stages of acinar morphogenesis (49). A recent paper of mammary cell 3D culture shows that fibronectin expression regulates mammary epithelial cell growth. Fn in matrigel increases acinar size and Fn supplementation in the medium alters growth arrest state of mammary epithelial cells and increases proliferation and luminal filling. Dysregulated Fn expression affects proper acinar morphogenesis by altering mammary cell proliferative state and prevents lumen formation in mammary gland (49). However, proteolytic fragments of Fn suppress growth and promote apoptosis of tumorigenic mammary epithelial cells through matrix protease dependent pathways (50). ECM proteins like Fn can also influence cell proliferation through regulation of integrin interaction with hormones and growth factors. 11 ROLE OF FIBRONECTIN IN BREAST CANCER The interstitial ECM in normal adult mammary tissue mostly lacks fibronectin while increased fibronectin is noted in the stroma of benign mammary hyperplasia and various types of mammary tumors and high levels of fibronectin in tissue sections correlate with decreased patient survival (49). Subtracted cDNA libraries comparing non-breast tissue, normal breast tissue and breast tumors show a threefold expression of Fn in breast tumors (51). Gene expression profile of primary invasive ductal carcinoma of the breast and axillary node with metastasis demonstrated that fibronectin was upregulated in both varieties of isolated tumor compared to disease free mammary epithelium (52). Fibronectin is not only differentially expressed between normal and tumor breast tissue but also differentially upregulated when compared to other cancers such as lung squamous cell cancer, lung adenocarcinoma, and renal cell cancer (53). In fact, fibronectin is most prominently expressed in cell lines selected for enhanced lung colonization and is known as a “pro-metastatic” gene (51). FN1 is also upregulated in a highly lung metastatic breast cancer cell variant isolated from early lung metastasis (54). Bioactive Fn fragments released during remodeling of mammary tissue in post-lactational involution have also been associated with increased breast cancer invasiveness and metastasis to the lung, liver and kidney (55). This result correlates with the high occurrence of metastases in pregnancyassociated breast cancer. These studies strongly indicate a role for fibronectin in breast cancer progression and metastasis. 12 FIBRONECTIN MATRIX ASSEMBLY Fibronectin mediates a wide range of cellular functions in normal cells which includes cellular interactions with the extracellular matrix (ECM), cell adhesion, migration, growth and differentiation (1). The single question of utmost importance regarding the function of fibronectin is the mechanism by which cells convert it from a relatively compact soluble protomer to an insoluble form of multimers visible by fluorescence microscopy. This is a highly regulated process that involves interaction between fibronectin and cell surface receptors, matrix components, as well as itself. The ability to interact with the variety of molecules to accomplish the formation of a complex meshwork is conferred in the unique structure and domains within fibronectin (1). Fibronectin is secreted as a dimer composed of two subunits of 250-kDa fibronectin molecules linked covalently near the C-termini and after secretion, fibronectin is assembled into a fibrillar matrix through interactions with cell surface integrin receptors, other matrix components or via self association (Fig 1.3). The incorporation into the matrix occurs in a two step process. The first step is binding of fibronectin in a deoxycholate (DOC) soluble pool that is reversible. The second step forms a DOC insoluble pool and results in a multimeric insoluble matrix. (1,56). The assembly process is integrin-dependent and fibronectin-integrin interactions trigger sequential events involving extracellular comformational activation of fibronectin and intracellular organization of the actin cytoskeleton. During assembly, the conformational changes of fibronectin expose fibronectin-binding sites and promote intermolecular interactions that result in fibril formation (57). 13 Mao, Y. et al. (2005) Matrix Biol 24(6), 389-399 Figure 1.3: Major steps of Fn fibril assembly: (A) Soluble Fn binds to α5β1 integrin via its cell binding domain. (B) Fn binding to integrins and other receptors (pink bars) induces reorganization of the actin cytoskeleton (green lines) and activates intracellular signaling complexes (silver circles). (C) Fn–Fn interactions results in fibrils formation. (57) 14 (1) INTERACTIONS OF FN AND INTEGRINS Fibronectin is recognized by at least 10 different integrins. The first of such isolated is the α5β1 integrin, often referred to as the classical fibronectin receptor. It appears to be the main integrin attaching cells to fibronectin and has a partial role in controlling the assembly of fibronectin matrix around cells. (1,58,59). The region spanning the center of the Fn polypeptide chain in the ninth and tenth type III repeats binds to the α5β1 integrin (60). Extensive research had narrowed down the essential sequence for cell adhesion to several minimal integrin-recognition sequences. Two critical amino acid sequences in Fn are required for α5β1 integrin binding: an ArgGly-Asp (RGD) sequence and a Pro-His-Ser-Arg-Asn (PHSRN) sequence (Fig 1.2). The well known RGD sequence is located in the tenth type III repeat. The binding to this simple tripeptide sequence depends on many factors including the flanking residues, the three dimensional presentation of the sequence and individual characteristics of the integrin-binding pockets (1). The synergy site PHSRN in the Fn repeat III9 mediates binding of Fn to α5β1 integrin via interactions with α5. Binding of Fn to α5β1 integrin can also be achieved through interactions with N-terminal fragment that contains I1-9 and II1-2 and trigger distinct intracellular signals from those elicited through ligation with RGD (1). This PHSRN site has been implicated in regulating matrix metalloproteinase (MMP)-1-dependent invasion of human breast cancer and mammary epithelial cells (61). Other integrins that have been shown to bind fibronectin are α3β1, α4β1, αvβ1, αIIbβ3, αvβ3 αvβ6, α4β7 and α?β8 (62). Many of the other fibronectin binding integrins bind to the RGD sequence, but the α4β1 integrin binds to fibronectin at a different site. The α4β1 binding site in fibronectin resides in the CS1 or V25 site that is located in the alternatively spliced V region of Fn (63-66). Studies have implicated functional differences between α4β1 and α5β1 in fibronectin matrix assembly (67, 68). 15 (2) INTERACTIONS OF FN AND OTHER MATRIX COMPONENTS Fibronectin mediates a wide variety of cellular functions through the binding of molecules other than integrins. These include heparin, collagen/gelatin, and fibrin. The interacting domains of Fn to these molecules have been identified by proteolytic fragmentation or recombinant DNA analyses (8,69,70). Fn contains two major heparin-binding domains with the strongest binding site at the C-terminus (Heparin II) and a weaker binding site at the N-terminus (Heparin I). The Heparin II site also binds to glycosaminoglycan, chondroitin sulfate while Heparin I mediates Fn interactions with bacteria. The heparin binding domains of Fn is involved in cell adhesion in some cell types (1, 71). The collagen binding domain includes I6-9 and II1-2 and binds denatured collagen (gelatin) more effectively than native collagen. This domain has been implicated more in the clearance of collagenous materials than as a mediator of cell adhesion (1). The fibrin binding sites of Fn are Fibrin I and Fibrin II. Interactions at these sites have a role in cell adhesion and cell migration (1). (3) FN SELF ASSOCIATION Matrix assembly requires self association of fibronectin and involves domains within fibronectin that are highly specialized for Fn-Fn interactions (3,8,69,72). Two regions of fibronectin has been implicated in Fn binding (self association) and matrix assembly (3). The 70-kDa amino-terminal region was shown to be essential for matrix assembly while the type I repeats I1-5 within this region confer fibronectin binding activity. The 70-kDa amino-terminal fragment was shown to have saturable binding to fibroblast monolayers and to be required for de novo assembly of newly synthesized recombinant fibronectins into the extracellular matrix. A fragment containing type I repeats I1-5 can be covalently cross linked to fibronectin at the cell surface while deletion of the five repeats drastically reduced the binding ability of the fragment to 16 fibroblasts. An excess of the 70-kDa fragment and antibodies against repeats I1-5 had the ability to inhibit matrix assembly. A second site involved in fibronectin binding exists in a fragment spanning the first two type II repeats (II1-2) adjacent to the gelatin binding domain. The 56-kDa fragment spanning this region and antibodies raised against repeats within it also had the ability to inhibit fibronectin binding and matrix formation (3). Other regions involved in matrix assembly include the cell binding domain (3,67,73) and the carboxyl-terminal cysteines that form the interchain disulfide bonds in the Fn dimer (74). (4) INTERACTION OF FIBRONECTIN WITH DIPEPTIDYLPEPTIDASE IV (DPPIV) DPPIV (CD26) is a membrane bound ectopeptidase that modulates the activities of several cytokines, chemokines and hormones via its unique proteolytic activity preferentially cleaving N-terminal X-Pro or X-Ala dipeptides (75-77). CD26/DPPIV also exerts biological functions unrelated to its dipeptidase activity. Not only can DPPIV form heterodimers with FAP-alpha, associate with plasminogen 2, ADA, CD45, and CXCR4, it can also bind to the ECM proteins fibronectin and collagen (77). Binding of fibronectin to hepatocytes in suspension was first observed by Johansson utilizing an 85kDa cell binding peptide that lacks the gelatin- and heparinbinding domains. The 85kDa peptide inhibited initial attachment of hepatocytes to immobilized fibronectin and bound to the cells in suspension at 4 degrees C in a timedependent, saturable, and partially reversible reaction (62). Piazza et al. later reported that initial binding of fibronectin to the cell surface where it was subsequently crosslinked is accelerated by inhibitors of DPPIV (78). Kinetic studies revealed that increasing amounts of iodinated Fn were irreversibly incorporated into large covalent, 17 nonreducible complexes that fail to enter the stacking gel during SDS-PAGE. Nitrocellulose binding assays using 125I-labelled DPP IV purified from rat hepatocytes demonstrated a direct interaction of DPP IV with fibronectin. Fibronectin was found to bind DPP IV at a site distinct from its exopeptidase substrate recognition site (78). Studies were performed in this lab to identify the endothelial cell adhesion molecule responsible for binding multimeric fibronectin presented on the surface of metastatic breast cancer cells (polyFn), which mediates lung colonization, showed the responsible molecule to be the DPPIV expressed on rat lung capillary endothelia (79). Such adhesion could be inhibited in vivo and in vitro by soluble DPPIV and by antibody to Fn but not by soluble Fn (79). Immobilized DPPIV pulls down Fn multimers but not protomers from 35[S]-Met-labeled MTF7 cell extracts. MTF7 cells in suspension accumulate 125[I]-labeled Fn on their surface in a manner similar to that described above and in adherent cells, resulting in deoxycholate-insoluble and nonreducible multimers, seemingly forming a transglutaminase cross-linked matrix. These multimers assembled on tumor cell surface can be visualized as distinct globules under a fluorescent microscope (79). Soluble DPPIV binds to Fn-coated plastic (immobilized Fn) but such binding is not inhibited by soluble Fn, justifying the physiological relevant condition where the cancer cells are able to bind to endothelial DPPIV through its surface associated polyFn in a fibronectin rich environment in the plasma (79). A detailed study in our lab using proteolytic fragments and maltose-binding protein fusion proteins found that the DPPIV-Fn interaction to be mediated by the consensus motif T (I/L)TGLX (P/R)G (T/V)X on fibronectin. These DPPIV-binding sites are located in type III repeats 13, 14, and 15 (Fn III13, -14, and -15, respectively) 18 of fibronectin. Peptides spanning the DPPIV-binding domain of FNIII14 blocked DPPIV/polyFn adhesion and suppressed pulmonary metastasis (80). FN IN EPITHELIAL TO MESENCHYMAL TRANSITION AND INVASION Tumor cells that migrate must first undergo a process known as epithelial to mesenchymal transition (EMT) to alter their phenotype into a less adhesive, more motile and spindle shaped cell that resembles cells from mesenchymal origin (81,82). EMT is a normal physiologic process that occurs during embryonic development, organogenesis, tissue growth and wound-healing and repair (83) but is also involved in the pathogenesis of many diseases and tumorigenesis. In breast cancer, EMT has been found to occur in 18% of tumors in vivo (84). EMT is defined as a variable proportion of tumor cells that express mesenchymal markers like vimentin and tenascin (84). EMT is involved in the appearance of different breast carcinoma types (85). Partial or complete loss of E-cadherin during carcinoma progression, which leads to poor prognosis, has been reported (85). Snail, a gene regulating EMT in Drosophila gastrulation, is expressed in areas of heterogenous tumors that are devoid of Ecadherin and is also found in all ductal invasive carcinomas with lymph node involvement (86). On the other hand, twist, another embryonic regulator, is highly expressed in invasive lobular carcinoma (87). Suppression of twist also inhibits the lung metastatic ability of highly metastatic mammary carcinoma (87). Areas of tumor lesions harboring poorly differentiated tumor cells that exhibited mesenchymal features in mammary sections of Cripto-1 transgenic mice were stained positive for Fn as well as integrins αv,α3, β1 and β3, known Fn receptors (88), strongly correlating EMT to Fn. A blocking antibody of α5β1 integrin signaling resulted in a more epithelial morphology in EGF-treated cells while other integrin antibodies tested have 19 no effects (89). Inhibition of α5β1 integrin signaling also blocked EGF-mediated increase of EMT markers vimentin and snail and decrease of E-cadherin while application of Fn increased EGF-induced effects on EMT markers. These results indicate that α5β1 integrin signaling and Fn can modulate EGF-induced EMT in cervical cancer cells (89). α5β1 integrin has also been implicated in TGF 1-induced EMT in mammary epithelial cells (90). Direct evidence of EMT involvement in breast cancer has also recently been shown in three different oncogene-driven mouse mammary tumor models (91). These models found stromal fibroblasts associated with myc-induced tumors were of epithelial origin and these cells lacked cytokeratin and Ecadherin expression while expressing mesenchymal specific markers vimentin and FN (91). However, EMT is not required for breast cancer invasiveness and metastasis because mice and human tumors lacking EMT also had metastatic potential (91). EMT presents one mechanism by which cancer progress to a more aggressive phenotype. Invasion of cells into a tissue can be a part of tumor progression or a part of normal cell processes. In cancer, invasion of cancerous cells into normal cells leads to metastasis. In inflammatory response or wound healing, migration of cells to the site of infection or wound site is a normal response (92). Fn in the matrix can promote or prevent invasion. In the case where invasive tissue is migrating into an acellular extracellular matrix, such as during embryogenesis or wound repair, Fn promotes invasion. Migration fails when fibronectin is ablated, when cell-Fn interaction is blocked by antifunctional antibodies to Fn or integrin, and when RGD-containing peptide is used, and this typically leads to reduced invasion (92). In developing fetus, when Fn accumulation at the intimal cushion (smooth muscle in the center of arteries that keep arteries open) extracellular matrix is reduced, vascular smooth muscle invasion is blocked and the ductus fails to close properly after birth (93). If invasion occurs towards densely cellular tissue, Fn prevents invasion. Fn contributes to the 20 stabilization of the organization of tissues, such as in a developing amniote heart, where Fn stabilizes the contact boundary between the myocardium and the epicardial mesenchyme. Loss of crucial mediators of adhesion may result in pathological disorganization of tissues that leads to aggressive malignant invasion (92). An anti-β1 blocking antibody inhibited invasion of both HT-1080 fibrosarcoma cells and MDAMB-231 breast carcinoma cells through Matrigel basement membrane (94,95). Antiα5 blocking antibody also inhibited MDA-MB-231 invasion in the presence of chemoattractant but not in the absence of it (95). In this case, the α5 integrin may be functioning as a chemoattractant and not adhesion receptor to promote migration of cells towards Fn (95). Similarly, in cervical cancer, EGF stimulated cancer cell invasiveness was abolished by α5β1 integrin blocking antibody (89). Comparison of normal and fibrotic lung fibroblasts revealed that Fn signaling through α5β1 integrin mediates migration/invasion while Fn signaling through α4β1 integrin suppresses migration/invasion (96). The α4 integrin has also been found to be inversely correlated to invasive potential in B16 melanoma cells (97). The functional roles of Fn in EMT, cell migration and invasion tightly correlate with its role in tumor progression. FIBRONECTIN IN METASTASIS The role of Fn in metastasis cannot be simply stated. In some cases, Fn contributes to metastasis and is upregulated in metastatic variants of cancer cells (98105). In other cases, Fn is found to be negatively correlated to metastasis (51,95,106110). Various studies on the same type of cancer have pointed towards different functional consequences of Fn expression. There are generally two broad classes of in vivo metastasis models. The “experimental” metastasis models involve injection of tumor cell suspensions into 21 mice and quantitation of the number or sizes of the metastatic colonies at the metastatic organ. This model studies the later steps of metastasis, skipping the early steps of detachment from the primary tumor mass and invasion into the vessels. The “spontaneous” metastasis models are more complicated and involve the implantation of tumors or cells into mice. This model measures the earlier steps including detachment, intravasation leading to the final steps of adhesion, tumor arrest and migration into the secondary metastatic organ (95). Fn can have a negative role in metastasis. Fn has been shown to suppress transformed phenotypes in some studies. Overexpression of pFn in human fibrosarcoma cells leads to suppression of tumor features such as a flattened morphology, reduced cell migration and reduced tumor growth (98). Similarly, inverse correlation has also been shown between Fn expression and metastatic potential. Comparison between non-metastatic prostate cancer cell subtype and metastatic subtype (99) and between moderate-metastatic and high-metastatic murine mammary adenocarcinoma subtypes (100) showed decreased Fn expression in non-metastatic or lower metastatic subtypes. Independent experiments that perturb Fn-integrin interactions also showed negatively correlation of Fn and metastasis. Injection of mice with RGD-containing peptides reduced experimental metastasis of murine melanoma cells (101). Intraperitoneal administration of super-Fn (induced multimeric form of fibronectin with enhanced adhesive properties) into mice reduced lung colonization in both experimental and spontaneous metastasis models of KRIB osteosarcoma, C8161 melanoma or HT-29 carcinoma cells (102). Studies using the MM3 murine mammary adenocarcinoma cell line showed inverse correlation between Fn expression and metastatic potential. The highly metastatic MM3 does not express Fn protein or mRNA while the parental M3 moderately expresses Fn and forms Fn ECM (103). Global gene expression comparison between the two cell lines revealed only two 22 genes significantly down regulated in MM3 , one of which is Fn, reduced by 19 fold (104). LMM3 tumor cells, derived from MM3 spontaneous lung metastasis, like its parent cell line also does not express Fn. Expression of wild type Fn and Fn lacking RGD, which does not form ECM fibrils, in LMM3 cells is shown to be sufficient to reduce the metastatic potential of cancer cells (105). These studies provide a model in which Fn has a negative impact on cancer metastasis. However, many studies also strongly suggest a role of Fn in metastasis. Global gene profile revealed 3-fold expression of Fn in breast tumors (51), up to 28 fold in colorectal tumors (106), 2.83 fold in head and neck squamous cell carcinoma (107). Fn expression was positively correlated with lymph node involvement and mortality risk of breast cancer (108). In an experimental metastasis model studying B16-F10 murine melanoma cells, the synthetic peptide GRGDS, derived from Fn IIICS cellbinding domain, specifically inhibited the number of lung colonies and results show that the peptide may be interfering with multiple fibronectin-mediated events at different steps of the metastatic process (109). A peptide consisting of Fn central cellbinding domain and heparin-binding domain (CBD-HepII) was able to suppress lung metastasis of B16 melanoma (110) and hepatic metastasis of hepatocarcinoma (111). The importance of Fn-Tumor cell interactions in metastasis can be further supported by the studies that showed the abilities of anti-α5 and anti-β1 monoclonal antibodies to inhibit experimental metastasis in MDA-MB-231 cells. In spontaneous metastasis models, treatment of mice bearing mammary carcinoma transplants with anti-β1 polyclonal antibodies reduced the size of the metastatic colonies but not the number, possibly indicating inhibition of tumor cell interations with the surrounding extracellular matrix (95). 23 SITE SPECIFIC METASTASIS In the process of hematogenous metastasis, cancer cells break away from the primary tumor, enter the blood stream, arrest at the vasculature of a secondary organ and either form colonies at the endothelium adhesion site (112) or extravasate to form tumor colonies completing the metastatic cascade (113) (Fig 1.4). Studies have followed labeled tumor cells through the course of hematogenous spread and found that cells initially arrest in the microvasculature of the first organ they encounter. At this site, most tumor cells die (114) and a few survive either to form colonies at that site or colonize another tumor specific metastatic organ (115). Metastasis was once believed to be a mechanistic phenomenon but evidence supports involvement of organ specific molecules in the specific recognition of tumor cells to secondary organs of colonization. This is known as the “seed-and-soil hypothesis (190).The fact that tumors have select specificity for certain organs during metastasis is supported by clinical assessment (116). For instance, prostatic carcinomas and small cell carcinomas of the lungs preferentially colonize bones and the brain, respectively, on the other hand, breast carcinomas most frequently metastasize to the lungs, but also to liver, bones, brain, and adrenals (79). Tracking of GFP tagged tumor cells found that the cells were mainly in pre-capillary arterioles, which had diameters larger than tumor cells, indicating that cells were not merely arrested by capillary trapping. The same study also found extravasation to be a rare event, with no intact cells surviving extravascularly for more than 48 hours, although occasional presence of invadopodia were noted intravascularly (112). Although the extravascular metastasis model is based on the fact that loss of cell contact leads to accelerated cell death in blood circulation, anchorage independent growth in suspension in tumor cells and facilitation of cell proliferation by endothelial attachment have been shown, which 24 Fidler, I. J. (2003) Nat Rev Cancer 3(6), 453-458 Figure 1.4: The main steps in the formation of a metastasis. (190) 25 further supports the intravascular metastasis model (112). Opposing to the extravascular metastasis model, which tumor cells extravasate at the endothelial attachment site to form metastatic foci extravascularly, the intravascular metastasis model supports that metastatic colonies originate from intravascular cells, form micrometastases within the vasculature, and finally the colonies will outgrow the vessels and disrupt the vascular walls (112). Vascular arrest of cancer cells mediated by “organotypic” molecules is responsible for the initial selection of an organ for metastasis (117-119). For example, the lung endothelial cell adhesion molecule-1 (Lu-ECAM-1; bCLCA2, bovine calcium activated chloride channel 2) characterized in this laboratory selectively binds lungmetastatic melanoma cells and its expression on the endothelia of pulmonary venules correlates with the formation of melanoma metastases at these sites (120). Monoclonal antibodies against Lu-ECAM-1 inhibit colonization of the lungs by lung-metastatic murine B16 melanoma cells but have no effect on lung colonization by other types of lung-metastatic cancer cells indicating a tumor cell specific adhesion molecule that mediates lung colonization (79,120). Colonization of the lungs by human breast cancer cells has been shown to be mediated by breast cancer β4 integrin and endothelial hCLCA2, the human counterpart of bCLCA2 (121). The CLCA family members comprise of one group of endothelial tumor homing receptors that facilitate vascular arrest of tumor cells. Utilizing outside out luminal membrane vesicles, we also show that more vesicles isolated from rat lung microvascular endothelia binds to lung metastatic rat breast carcinoma cells compared to nonmetastatic counterparts (79,122,123). Vesicles isolated from nonmetastatized organ (e.g., hind leg muscle) showed no binding preference for lung-metastatic or nonmetastatic mammary carcinoma cells. Monoclonal antibodies against the lung-derived endothelial vesicles inhibit specific 26 adhesion of the vesicles to lung-metastatic breast cancer cells. The antibody recognizes a 110-kDa membrane glycoprotein which was found to be dipeptidyl peptidase IV (DPP IV; also known as CD26 or gp110) (123). Studies have previously identified association between cell surface expression of Fn and lung metastasis of rhabdomyosarcoma cells and the DPP IV ligand was identified as tumor cell surfaceassociated Fn (124). Fibronectin assembles on breast cancer cell surfaces into multiple, randomly dispersed globules from cellular and plasma Fn. The level of Fn surface expression correlates to the tumor cells' ability to bind to DPP IV and metastasize to the lungs. Soluble DPPIV blocks DPPIV/Fn mediated adhesion and metastasis while anti-DPPIV and anti-Fn mAb blocks adhesion. Adhesion is unaffected by soluble plasma FN (pFn) and indicates the ability of cancer cells to arrest through this mechanism under high plasma Fn conditions in the blood stream. The ability of many cancer cells to capture Fn molecules on their surface and to augment such deposits by Fn self-association during passage in the blood suggests that DPP IV/Fn binding may be a relatively common mechanism for lung metastasis (79). The CLCA/β4 interaction and the DPPIV/Fn interaction demonstrate the importance of vascular addresses in the adhesion step in metastasis. POLYMERIC FIBRONECTIN ON SUSPENDED BREAST CANCER CELLS MEDIATES LUNG METASTASIS The molecular foundation of the original immobilization of Fn on the surface of breast cancer cells is poorly understood. It seems that rat breast cancer cells growing as a solid tumor mass in syngeneic animals support the synthesis of a Fncontaining matrix (116). After tumor cells escape the primary tumor, they become blood-borne and travel passively with the blood stream to other organ sites (113,119). 27 In the pFn-rich environment of the plasma, tumor cells decorated with a light coat of cellular Fn use this Fn scaffold to attain pFn molecules for the buildup of a prominent Fn coat that is visualized as multiple, densely distributed Fn globules by immunocytochemistry. This suggests that the initial Fn binding to the cell surface occurs around focal adhesion points from which polymerization is then initiated by Fn self-association (125-128). Such adhesion points are anticipated to be sites of integrin clusterings, most likely the classic Fn receptor α5β1 (67,129,130). An alternative mechanism may be that a cellular Fn-α5β1 complex forms intracytoplasmically (e.g. in Golgi vesicles) and serves as a scaffold for build up upon transport to and incorporation into the plasma membrane (79). The Fn buildup on cancer cell surfaces is comparable with that reported first for normal confluent, adherent cell layers of fibroblasts (56). A 4 hour incubation of lung-metastatic breast cancer cells with 125I-pFn results in cell surface-associated Fn that can be harvested as DOC-soluble and DOC-insoluble fractions. The multimeric Fn on breast cancer cells and normal hepatocytes, both grown in suspension, progressively converted to nonreducible, seemingly covalently bonded Fn complexes (131) not observed on anchorage-dependent normal fibroblasts (56). This conversion is possibly mediated by a cancer cell-associated transglutaminase activity, similar to hepatocytes (131). The similarities between the Fn assembly in suspended, bloodborne breast cancer cells and adherent fibroblasts suggest that commons receptors and signaling molecules might be shared between the two pathways. SIGNALING EVENTS IN FIBRONECTIN ASSEMBLY Fibronectin matrix assembly involves regulated intracellular signaling events that include binding to integrins, organization of the actin cytoskeleton and stimulation 28 of cell contractility and activation of kinase cascades (132) (Fig 1.5). A number of studies have established that binding of Fn by specific integrins is critical in initiating Fn matrix assembly. Fn fragments that contain the RGD integrin binding site or antibodies recognizing the integrin binding site inhibited Fn matrix assembly in cultured cells and developing amphibian embryos (133-138). Also, antibodies to α5β1 integrin decreased the deposition of Fn into extracellular matrix by fibroblasts (72,133,139,140). The role of Fn-binding integrins in Fn matrix assembly has been extensively studied in Chinese hamster ovary (CHO) cells. Overexpression of α5β1 in CHO cells increased Fn deposition in extracellular matrix (141) while CHO B2 cells deficient in α5 did not assemble plasma Fn into the extracellular matrix (67). Transfecting the CHO B2 cells with a full-length α5 chain completely restored fibrillar Fn matrix assembly (67). These studies determined an important role of α5β1 integrin in supporting Fn matrix assembly by CHO cells. On the other hand, cells are also able to support Fn matrix assembly through other integrins. One example of such is fibroblastic cells derived from α5 integrin null mutant embryos being able to assemble a Fn matrix in the absence of α5β1 (142). The primary role of α5β1 integrin in Fn matrix assembly seems to involve the initiation step of Fn matrix assembly, since Fn mutants lacking the α5β1 integrin binding site could not be assembled into Fn matrix if not in the presence of native Fn (74,143). Integrin links extracellular matrix (ECM) to cytoplasmic molecules mediating initiation of intracellular pathways that control cellular function. One cytoplasmic signaling molecule that is phosphorylated in response to integrin binding to ECM is FAK. FAK controls cell adhesion and spreading, migration, cell survival and proliferation. FAK directly interacts with downstream molecules such as Src, phosphatidylinositol 3-kinase (PI3-kinase) and Grb7 and these molecules have been implicated in regulation of early Fn fibrillogenesis (5,144-146). Cells lacking Src 29 Mao, Y. et al. (2005) Matrix Biol 24(6), 389-399 Figure 1.5: Model of Fn matrix assembly. (57) 30 family kinases, cells treated with Src and PI3-kinase inhibitors, and FAK-null cells and cells with increased Src expression and activity are all associated with a decrease in fibronectin assembly (8,147). These proteins are regulated by phosphorylation on specific tyrosine residues and phosphorylation events may vary depending on the cell type (5). Integrin ligation can also trigger activation of certain MAP kinases, including extracellular signal-regulated kinase (ERK) 1 members and JNK (JUN kinase) (148151). Different types of MAP kinases have distinct pathways and functions. ERK activation have been linked to EGF growth factor response pathways, while JNK is not activated by EGF in most cells but is activated by inflammatory cytokines or stress response such as UV irradiation (151,152-154). One study using inhibitors found that herbimycin A and genistein blocked the activation of both types of MAP kinases induced by cell binding to FN coated beads, while Cytochalasin D inhibited the ERK pathway and have no effect on SAPK/JNK1 pathway. The study showed that integrins can mediate activation of two distinct serine/threonine kinase pathways with different kinetics and different responses to disruption of the actin cytoskeleton (154). A subset of cytoskeletal proteins including tensin, α-actinin, talin and vinculin does not seem to require tyrosine phosphorylation to accumulate locally after integrin ligation (154,155). In contrast, another set of cytoskeletal proteins including paxillin and F-actin, and a group of signaling proteins such as Ras, phospholipase Cγ and ERK require tyrosine phosphorylation for accumulation (155). Cytoskeletal responses to integrin ligation can be studied using beads coated with molecules that interact with integrins. Fibronectin-coated beads induced the local accumulation of a variety of cytoskeletal molecules, including vinculin, talin, α-actinin, and F-actin (154,156-159). This further underlies the importance of cytoskeletal participation in the process of fibronectin matrix assembly. 31 Oncogenes can also affect fibronectin assembly through regulation of integrin function and localization. Ras proteins, which are small GTPases, link cell surface receptors to intracellular effector pathways regulating cell proliferation, differentiation, and survival (155). Ras proteins exert cell functions through the activation of several downstream effectors, including Raf, Rac, phosphatidylinositol 3-kinase (PI3K), and Ral (160). Through Raf, Ras activates MAPK/ERK kinase (MEK), which then phosphorylates ERKs (161-163). Activation of Raf-1 impedes the ability of α5ß1 integrin to mediate Fn matrix assembly (165). The suppression is similar to what occurs in v-src-transformed cells and correlates with activation of ERK. HT1080 human fibrosarcoma cells possess one activated N-ras allele and can be stimulated to assemble Fn matrix by activation of integrins using Mn2+ or ß1 integrin-activating antibody or by inhibition of Ras signaling through ERK (165). In summary, mutation of at least two oncogenes, ras and src, have negative effects on Fn matrix assembly and, in some cells, exert their effects through a common downstream effector, ERK. Integrin ligand binding affinity can also be regulated via inside-out signaling, in which intracellular components are able to control extracellular matrix assembly by modulation of ligand preference through interaction with the cytoplasmic tail of integrins. (133,166-170). Constitutively active R-Ras expressed in CHO cells with αIIb3 integrin activated the integrin and caused an increase in Fn matrix assembly (171), whereas activation of Raf-1, probably via the MAPK pathway, suppresses integrin activation and Fn matrix assembly (164). One report shows that overexpression of ILK in epithelial cells dramatically stimulated integrin-mediated Fn matrix assembly (133). ILK, a serine/threonine kinase, binds to the cytoplasmic domains of both β1 and β3 integrins and phosphorylates the β1 cytoplasmic domain in vitro (172). These studies show that cells control integrin activity and Fn matrix 32 assembly via interactions of cytoplasmic regulatory proteins with integrin cytoplasmic domains (133). However, integrins do not act alone in transmembrane signaling in Fn matrix assembly (132). Cells with impaired proteoglycan synthesis showed defective Fn matrix assembly, and this has been attributed to reduced activity of syndecans, transmembrane proteoglycans that can bind Fn (173, 174). The major syndecan in fibroblasts, syndecan-2, has been shown to have an indirect regulatory effect on matrix assembly through truncation studies. A truncated version of syndecan-2 lacking a part of its cytoplasmic tail blocks Fn assembly in a dominant negative way. On the other hand, syndecan-2 without its entire cytoplasmic tail had no effect on matrix formation (175). Another syndecan, syndecan-4, has a more direct effect on Fn assembly. Syndecan-4 cooperates with integrins in regulation of Rho-dependent cell adhesion, spreading and actin organization (176). Concomitant ligation of α5β1 integrin and syndecan-4 increases active Rho(177) and phosphorylated FAK and Rho and FAK, when both stimulated, have been shown to favor Fn matrix assembly (178). Another important signaling molecule that has been implicated in fibronectin assembly is PKC. (179-182). PKC activation has been shown to increase Fn synthesis and fibrillogenesis in a number of cell types, cinluding pulmonary fibroblasts, retinal pigment epithelial cells, vascular smooth muscle cells, osteoblasts, hyperglycemic mesangial cells and Xenopus cells (132, 182, 183-187). Activation of PKC increases I125-labeled Fn binding to the cell surface while inhibition of PKC resulted in a rapid decrease in Fn binding (179). The PKC activator TPA increases extracellular assembly of Fn fibril while the PKC inhibitor forskolin decreased assembly. Moreover, TPA increased α5 integrin clustering and surface expression while PKC inhibitors inhibited clustering; implicating that PKC regulates the dynamic assembly of Fn through modulation of integrin distribution (180,181). PKC also binds to the cytoplasmic tail 33 of clustered syndecan-4 (188-189) and may act on Fn assembly through syndecan-4. The strong connection of PKC with integrins and syndecan could potentially be the pathway regulating cell surface polyFn assembly in blood-borne cancer cells. SUMMARY AND GOALS OF THIS STUDY In summary, gene expression profiling of breast cancer of different stages and metastatic preference shows that fibronectin is upregulated in breast cancer. Fibronectin interacts with integrins, matrix components and itself during matrix assembly through its integrin interaction sites, collagen, heparin or fibrin binding sites, and self association sites, respectively. Cellular fibronectin has distinct domains from plasma fibronectin and may play a different role in fibronectin assembly. Cells are also able to interact with fibronectin and transduce inside-out or outside-in signals through different integrin receptors. Fibronectin also interacts with DPPIV. In breast cancer cells, multimeric fibronectin (polyFn) assembled on the surface of the cells interact with DPPIV on lung endothelial cells facilitating lung vascular arrest and promotes lung metastasis. The level of polyFn surface expression correlates with cancer cells’ ability to metastasize to lungs. The intracellular mediator of the process of polyFn assembly has not been identified, and a further investigation into the regulatory mechanism of fibronectin assembly on the surface of these breast cancer cells might bring forward new insights and novel strategies into the potential targets available for anti-metastatic therapy in breast cancer patients. The goal of this study is to examine the related pathways previously known to be associated in signaling pathways of adherent cell fibronectin matrix assembly in order to discover the molecule(s) regulating polyFn assembly and breast cancer cell lung metastasis. 34 REFERENCES 1. Pankov, R., and Yamada, K. M. (2002) J Cell Sci 115(Pt 20), 3861-3863 2. Magnusson, M. K., and Mosher, D. F. (1998) Arterioscler Thromb Vasc Biol 18(9), 1363-1370 3. Aguirre, K. M., McCormick, R. J., and Schwarzbauer, J. E. (1994) J Biol Chem 269(45), 27863-27868 4. White, E. S., Baralle, F. E., and Muro, A. F. (2008) J Pathol 216(1), 1-14 5. Geiger, B., Bershadsky, A., Pankov, R., and Yamada, K. M. (2001) Nat Rev Mol Cell Biol 2(11), 793-805 6. Vakonakis, I., Staunton, D., Rooney, L. M., and Campbell, I. D. (2007) Embo J 26(10), 2575-2583 7. Sun, L., Zou, Z., Collodi, P., Xu, F., Xu, X., and Zhao, Q. (2005) Matrix Biol 24(1), 69-77 8. Hynes RO. Fibronectins (1st edn). Springer-Verlag: New York, 1990. 9. Ffrench-Constant, C. (1995) Exp Cell Res 221(2), 261-271 10. Borsi, L., Balza, E., Allemanni, G., and Zardi, L. (1992) Exp Cell Res 199(1), 98-105 11. Pagani, F., Zagato, L., Vergani, C., Casari, G., Sidoli, A., and Baralle, F. E. (1991) J Cell Biol 113(5), 1223-1229 12. Chauhan, A. K., Iaconcig, A., Baralle, F. E., and Muro, A. F. (2004) Gene 324, 55-63 13. Ffrench-Constant, C., Van de Water, L., Dvorak, H. F., and Hynes, R. O. (1989) J Cell Biol 109(2), 903-914 14. Mathews, G. A., and Ffrench-Constant, C. (1995) J Neurobiol 26(2), 171-188 15. Przybysz, M., Borysewicz, K., and Katnik-Prastowska, I. (2009) Rheumatol Int 16. Xia, P., and Culp, L. A. (1995) Exp Cell Res 217(2), 517-527 17. Meyer-Ter-Vehn, T., Grehn, F., and Schlunck, G. (2008). Mol Vis 14, 136-41. 35 18. Wang, A., Cohen, D. S., Palmer, E., and Sheppard, D. (1991) J Biol Chem 266(24), 15598-15601 19. Muro, A. F., Moretti, F. A., Moore, B. B., Yan, M., Atrasz, R. G., Wilke, C. A., Flaherty, K. R., Martinez, F. J., Tsui, J. L., Sheppard, D., Baralle, F. E., Toews, G. B., and White, E. S. (2008) Am J Respir Crit Care Med 177(6), 638-645 20. Guan, J. L., Trevithick, J. E., and Hynes, R. O. (1990) J Cell Biol 110(3), 833847 21. Fukuda, T., Yoshida, N., Kataoka, Y., Manabe, R., Mizuno-Horikawa, Y., Sato, M., Kuriyama, K., Yasui, N., and Sekiguchi, K. (2002) Cancer Res 62(19), 5603-5610 22. Astrof, S., Crowley, D., George, E. L., Fukuda, T., Sekiguchi, K., Hanahan, D., and Hynes, R. O. (2004) Mol Cell Biol 24(19), 8662-8670 23. Matuskova, J., Chauhan, A. K., Cambien, B., Astrof, S., Dole, V. S., Piffath, C. L., Hynes, R. O., and Wagner, D. D. (2006) Arterioscler Thromb Vasc Biol 26(6), 1391-1396 24. Khan, Z. A., Chan, B. M., Uniyal, S., Barbin, Y. P., Farhangkhoee, H., Chen, S., and Chakrabarti, S. (2005) Angiogenesis 8, 183-96. 25. George, B., Chen, S., Chaudhary, V., Gonder, J., and Chakrabarti, S. (2009) Curr Eye Res 34(2), 134-144 26. Sauer, S., Erba, P. A., Petrini, M., Menrad, A., Giovannoni, L., Grana, C., Hirsch, B., Zardi, L., Paganelli, G., Mariani, G., Neri, D., Durkop, H., and Menssen, H. D. (2009) Blood 113(10), 2265-2274 27. Kapila, Y. L., Niu, J., and Johnson, P. W. (1997) J Biol Chem 272(30), 1893218938 28. Oyama, F., Hirohashi, S., Sakamoto, M., Titani, K., and Sekiguchi, K. (1993) Cancer Res 53(9), 2005-2011 29. Dufour, S., Duband, J. L., Humphries, M. J., Obara, M., Yamada, K. M., and Thiery, J. P. (1988) Embo J 7(9), 2661-2671 30. Bergijk, E. C., Baelde, H. J., Kootstra, C. J., De Heer, E., Killen, P. D., and Bruijn, J. A. (1996) J Pathol 178(4), 462-468 36 31. Kapila, Y. L., Wang, S., Dazin, P., Tafolla, E., and Mass, M. J. (2002) J Biol Chem 277(10), 8482-8491 32. Wilson, C. L., and Schwarzbauer, J. E. (1992) J Cell Biol 119(4), 923-933 33. George, E. L., Georges-Labouesse, E. N., Patel-King, R. S., Rayburn, H., and Hynes, R. O. (1993) Development 119(4), 1079-1091 34. George, E. L., Baldwin, H. S., and Hynes, R. O. (1997) Blood 90(8), 30733081 35. Astrof, S., Kirby, A., Lindblad-Toh, K., Daly, M., and Hynes, R. O. (2007) Mech Dev 124(7-8), 551-558 36. Georges-Labouesse, E. N., George, E. L., Rayburn, H., and Hynes, R. O. (1996) Dev Dyn 207(2), 145-156 37. Sakai, T., Johnson, K. J., Murozono, M., Sakai, K., Magnuson, M. A., Wieloch, T., Cronberg, T., Isshiki, A., Erickson, H. P., and Fassler, R. (2001) Nat Med 7(3), 324-330 38. Ni, H., Yuen, P. S., Papalia, J. M., Trevithick, J. E., Sakai, T., Fassler, R., Hynes, R. O., and Wagner, D. D. (2003) Proc Natl Acad Sci U S A 100(5), 2415-2419 39. Yi, M., Sakai, T., Fassler, R., and Ruoslahti, E. (2003) Proc Natl Acad Sci U S A 100(20), 11435-11438 40. Nyberg, P., Sakai, T., Cho, K. H., Caparon, M. G., Fassler, R., and Bjorck, L. (2004) Embo J 23(10), 2166-2174 41. Tan, M. H., Sun, Z., Opitz, S. L., Schmidt, T. E., Peters, J. H., and George, E. L. (2004) Blood 104(1), 11-18 42. Muro, A. F., Chauhan, A. K., Gajovic, S., Iaconcig, A., Porro, F., Stanta, G., and Baralle, F. E. (2003) J Cell Biol 162(1), 149-160 43. Chauhan, A. K., Kisucka, J., Cozzi, M. R., Walsh, M. T., Moretti, F. A., Battiston, M., Mazzucato, M., De Marco, L., Baralle, F. E., Wagner, D. D., and Muro, A. F. (2008) Arterioscler Thromb Vasc Biol 28(2), 296-301 44. Chauhan, A. K., Moretti, F. A., Iaconcig, A., Baralle, F. E., and Muro, A. F. (2005) Behav Brain Res 161(1), 31-38 37 45. Takahashi, S., Leiss, M., Moser, M., Ohashi, T., Kitao, T., Heckmann, D., Pfeifer, A., Kessler, H., Takagi, J., Erickson, H. P., and Fassler, R. (2007) J Cell Biol 178(1), 167-178 46. Moretti, F. A., Chauhan, A. K., Iaconcig, A., Porro, F., Baralle, F. E., and Muro, A. F. (2007) J Biol Chem 282(38), 28057-28062 47. Luiz Carlos Uchoa Junqueira, Jose Carneiro, Robert O. Kelley. Basic Histology. Appleton & Lange; 8th edition (July 1, 1996) 48. Haslam, S. Z., and Woodward, T. L. (2003) Breast Cancer Res 5(4), 208-215 49. Williams, C. M., Engler, A. J., Slone, R. D., Galante, L. L., and Schwarzbauer, J. E. (2008) Cancer Res 68(9), 3185-3192 50. Schedin, P., Strange, R., Mitrenga, T., Wolfe, P., and Kaeck, M. (2000) J Cell Sci 113 ( Pt 5), 795-806 51. Jiang, Y., Harlocker, S. L., Molesh, D. A., Dillon, D. C., Stolk, J. A., Houghton, R. L., Repasky, E. A., Badaro, R., Reed, S. G., and Xu, J. (2002) Oncogene 21(14), 2270-2282 52. Zucchi, I., Mento, E., Kuznetsov, V. A., Scotti, M., Valsecchi, V., Simionati, B., Vicinanza, E., Valle, G., Pilotti, S., Reinbold, R., Vezzoni, P., Albertini, A., and Dulbecco, R. (2004) Proc Natl Acad Sci U S A 101(52), 18147-18152 53. Amatschek, S., Koenig, U., Auer, H., Steinlein, P., Pacher, M., Gruenfelder, A., Dekan, G., Vogl, S., Kubista, E., Heider, K. H., Stratowa, C., Schreiber, M., and Sommergruber, W. (2004) Cancer Res 64(3), 844-856 54. Bandyopadhyay, A., Elkahloun, A., Baysa, S. J., Wang, L., and Sun, L. Z. (2005) Cancer Biol Ther 4(2), 168-174 55. McDaniel, S. M., Rumer, K. K., Biroc, S. L., Metz, R. P., Singh, M., Porter, W., and Schedin, P. (2006) Am J Pathol 168(2), 608-620 56. McKeown-Longo, P. J., and Mosher, D. F. (1983) J Cell Biol 97(2), 466-472 57. Mao, Y., and Schwarzbauer, J. E. (2005) Matrix Biol 24(6), 389-399 58. Pytela, R., Pierschbacher, M. D., and Ruoslahti, E. (1985) Proc Natl Acad Sci U S A 82(17), 5766-5770 59. Hay, E.D. (1991) Cell Biology of the Extracellular Matrix. 2nd Ed, Springer, 340-351. 38 60. Akiyama, S. K. (1996) Hum Cell 9(3), 181-186 61. Jia, Y., Zeng, Z. Z., Markwart, S. M., Rockwood, K. F., Ignatoski, K. M., Ethier, S. P., and Livant, D. L. (2004) Cancer Res 64(23), 8674-8681 62. Johansson, S. (1985) J Biol Chem 260(3), 1557-1561 63. Wayner, E. A., Garcia-Pardo, A., Humphries, M. J., McDonald, J. A., and Carter, W. G. (1989) J Cell Biol 109(3), 1321-1330 64. Guan, J. L., and Hynes, R. O. (1990) Cell 60(1), 53-61 65. Mould, A. P., Wheldon, L. A., Komoriya, A., Wayner, E. A., Yamada, K. M., and Humphries, M. J. (1990) J Biol Chem 265(7), 4020-4024 66. Komoriya, A., Green, L. J., Mervic, M., Yamada, S. S., Yamada, K. M., and Humphries, M. J. (1991) J Biol Chem 266(23), 15075-15079 67. Wu, C., Bauer, J. S., Juliano, R. L., and McDonald, J. A. (1993) J Biol Chem 268(29), 21883-21888 68. Wu, C., Fields, A. J., Kapteijn, B. A., and McDonald, J. A. (1995) J Cell Sci 108 ( Pt 2), 821-829 69. Mosher, D.F. (1989) Fibronectin. San Diego: Academic Press, Inc. 70. Yamada, K. M. and Clark, R. A. F. (1996). Provisional matrix. In The Molecular and Cellular Biology of Wound Repair (ed. R. A. F. Clark), pp. 5193. New York: Plenum Press. 71. Mostafavi-Pour, Z., Askari, J. A., Whittard, J. D., and Humphries, M. J. (2001) Matrix Biol 20(1), 63-73 72. Mosher, D. F. (1993) Curr. Opin. Struct. Biol. 3, 214-222 73. Akiyama, S. K., Yamada, S. S., Chen, W. T., and Yamada, K. M. (1989) J Cell Biol 109(2), 863-875 74. Schwarzbauer, J. E. (1991) J Cell Biol 113(6), 1463-1473 75. Fleischer, B. (1994) Immunol Today 15(4), 180-184 76. Gorrell, M. D., Gysbers, V., and McCaughan, G. W. (2001) Scand J Immunol 54(3), 249-264 39 77. Havre, P. A., Abe, M., Urasaki, Y., Ohnuma, K., Morimoto, C., and Dang, N. H. (2008) Front Biosci 13, 1634-1645 78. Piazza, G. A., Callanan, H. M., Mowery, J., and Hixson, D. C. (1989) Biochem J 262(1), 327-334 79. Cheng, H. C., Abdel-Ghany, M., Elble, R. C., and Pauli, B. U. (1998) J Biol Chem 273(37), 24207-24215 80. Cheng, H. C., Abdel-Ghany, M., and Pauli, B. U. (2003) J Biol Chem 278(27), 24600-24607 81. Boyer, B., Valles, A. M., and Edme, N. (2000) Biochem Pharmacol 60(8), 1091-1099 82. Savagner, P. (2001) Bioessays 23(10), 912-923 83. Perez-Pomares, J. M., and Munoz-Chapuli, R. (2002) Anat Rec 268(3), 343351 84. Petersen, O. W., Nielsen, H. L., Gudjonsson, T., Villadsen, R., Rank, F., Niebuhr, E., Bissell, M. J., and Ronnov-Jessen, L. (2003) Am J Pathol 162(2), 391-402 85. Vincent-Salomon, A., and Thiery, J. P. (2003) Breast Cancer Res 5(2), 101106 86. Blanco, M. J., Moreno-Bueno, G., Sarrio, D., Locascio, A., Cano, A., Palacios, J., and Nieto, M. A. (2002) Oncogene 21(20), 3241-3246 87. Yang, J., Mani, S. A., Donaher, J. L., Ramaswamy, S., Itzykson, R. A., Come, C., Savagner, P., Gitelman, I., Richardson, A., and Weinberg, R. A. (2004) Cell 117(7), 927-939 88. Strizzi, L., Bianco, C., Normanno, N., Seno, M., Wechselberger, C., WallaceJones, B., Khan, N. I., Hirota, M., Sun, Y., Sanicola, M., and Salomon, D. S. (2004) J Cell Physiol 201(2), 266-276 89. Lee, M. Y., Chou, C. Y., Tang, M. J., and Shen, M. R. (2008) Clin Cancer Res 14(15), 4743-4750 90. Maschler, S., Wirl, G., Spring, H., Bredow, D. V., Sordat, I., Beug, H., and Reichmann, E. (2005) Oncogene 24(12), 2032-2041 40 91. Trimboli, A. J., Fukino, K., de Bruin, A., Wei, G., Shen, L., Tanner, S. M., Creasap, N., Rosol, T. J., Robinson, M. L., Eng, C., Ostrowski, M. C., and Leone, G. (2008) Cancer Res 68(3), 937-945 92. Armstrong, P. B., and Armstrong, M. T. (2000) Biochim Biophys Acta 1470(2), O9-20 93. Mason, C. A., Bigras, J. L., O'Blenes, S. B., Zhou, B., McIntyre, B., Nakamura, N., Kaneda, Y., and Rabinovitch, M. (1999) Nat Med 5(2), 176182 94. Yamada, K. M., Kennedy, D. W., Yamada, S. S., Gralnick, H., Chen, W. T., and Akiyama, S. K. (1990) Cancer Res 50(15), 4485-4496 95. Akiyama, S. K., Olden, K., and Yamada, K. M. (1995) Cancer Metastasis Rev 14(3), 173-189 96. White, E. S., Thannickal, V. J., Carskadon, S. L., Dickie, E. G., Livant, D. L., Markwart, S., Toews, G. B., and Arenberg, D. A. (2003) Am J Respir Crit Care Med 168(4), 436-442 97. Qian, F., Vaux, D. L., and Weissman, I. L. (1994) Cell 77(3), 335-347 98. Akamatsu, H., Ichihara-Tanaka, K., Ozono, K., Kamiike, W., Matsuda, H., and Sekiguchi, K. (1996) Cancer Res 56(19), 4541-4546 99. Schalken, J. A., Ebeling, S. B., Isaacs, J. T., Treiger, B., Bussemakers, M. J., de Jong, M. E., and Van de Ven, W. J. (1988) Cancer Res 48(8), 2042-2046 100. Muro, A., Puricelli, L., Kornblihtt, A. R., and Bal de Kier Joffe, E. (1991) Invasion Metastasis 11(5), 281-287 101. Humphries, M. J., Olden, K., and Yamada, K. M. (1986) Science 233(4762), 467-470 102. Pasqualini, R., Bourdoulous, S., Koivunen, E., Woods, V. L., Jr., and Ruoslahti, E. (1996) Nat Med 2(11), 1197-1203 103. Werbajh, S. E., Urtreger, A. J., Puricelli, L. I., de Lustig, E. S., Bal de Kier Joffe, E., and Kornblihtt, A. R. (1998) FEBS Lett 440(3), 277-281 104. Urtreger, A. J., Werbajh, S. E., Verrecchia, F., Mauviel, A., Puricelli, L. I., Kornblihtt, A. R., and Bal de Kier Joffe, E. D. (2006) Oncol Rep 16(6), 14031410 41 105. Urtreger, A., Porro, F., Puricelli, L., Werbajh, S., Baralle, F. E., Bal de Kier Joffe, E., Kornblihtt, A. R., and Muro, A. F. (1998) Int J Cancer 78(2), 233241 106. Zhang, L., Zhou, W., Velculescu, V. E., Kern, S. E., Hruban, R. H., Hamilton, S. R., Vogelstein, B., and Kinzler, K. W. (1997) Science 276(5316), 12681272 107. Al Moustafa, A. E., Alaoui-Jamali, M. A., Batist, G., Hernandez-Perez, M., Serruya, C., Alpert, L., Black, M. J., Sladek, R., and Foulkes, W. D. (2002) Oncogene 21(17), 2634-2640 108. Ioachim, E., Charchanti, A., Briasoulis, E., Karavasilis, V., Tsanou, H., Arvanitis, D. L., Agnantis, N. J., and Pavlidis, N. (2002) Eur J Cancer 38(18), 2362-2370 109. Humphries, M. J., Yasuda, Y., Olden, K., and Yamada, K. M. (1988) Ciba Found Symp 141, 75-93 110. Gong, W., Liu, Y., Huang, B., Lei, Z., Wu, F. H., Li, D., Feng, Z. H., and Zhang, G. M. (2008) Biochem Biophys Res Commun 367(1), 144-149 111. Liu, Y., Huang, B., Yuan, Y., Gong, W., Xiao, H., Li, D., Yu, Z. R., Wu, F. H., Zhang, G. M., and Feng, Z. H. (2007) Int J Cancer 121(1), 184-192 112. Al-Mehdi, A. B., Tozawa, K., Fisher, A. B., Shientag, L., Lee, A., and Muschel, R. J. (2000) Nat Med 6(1), 100-102 113. Fidler, I. J. (1997) in Cancer: Principles and Practice of Oncology (DeVita, V. T., Hellman, S., and Rosenbvery, S. A., eds), 5th Ed., pp. 135-152, LipincottRaven Publishers, Philadelphia 114. Weiss, L., and Dimitrov, D. S. (1986) J Theor Biol 121(3), 307-321 115. Yeatman, T. J., and Nicolson, G. L. (1993) Semin Surg Oncol 9(3), 256-263 116. Sugarbaker, E. V. (1981) Cancer Biol. Rev. 2, 235-278 117. Albelda, S. M. (1993) Lab Invest 68(1), 4-17 118. Lafrenie, R. M., Buchanan, M. R., and Orr, F. W. (1993) Cell Biophys 23(1-3), 3-89 119. Pauli, B. U., and Lin, H. (1997) in Encyclopedia of Cancer (Bertino, J. R., ed), Vol. 1, pp. 464-476, Academic Press, New York 42 120. Zhu, D., Cheng, C. F., and Pauli, B. U. (1992) J Clin Invest 89(6), 1718-1724 121. Abdel-Ghany, M., Cheng, H. C., Elble, R. C., and Pauli, B. U. (2001) J Biol Chem 276(27), 25438-25446 122. Johnson, R. C., Augustin-Voss, H. G., Zhu, D. Z., and Pauli, B. U. (1991) Cancer Res 51(1), 394-399 123. Johnson, R. C., Zhu, D., Augustin-Voss, H. G., and Pauli, B. U. (1993) J Cell Biol 121(6), 1423-1432 124. Korach, S., Poupon, M. F., Du Villard, J. A., and Becker, M. (1986) Cancer Res 46(7), 3624-3629 125. Hocking, D. C., Smith, R. K., and McKeown-Longo, P. J. (1996) J Cell Biol 133(2), 431-444 126. Chernousov, M. A., Fogerty, F. J., Koteliansky, V. E., and Mosher, D. F. (1991) J Biol Chem 266(17), 10851-10858 127. Morla, A., and Ruoslahti, E. (1992) J Cell Biol 118(2), 421-429 128. Sakai, K., Fujii, T., and Hayashi, T. (1996) J Biochem 119(1), 58-62 129. Hynes, R. O. (1987) Cell 48(4), 549-554 130. Pierschbacher, M. D., and Ruoslahti, E. (1984) Nature 309(5963), 30-33 131. Fellin, F. M., Barsigian, C., Rich, E., and Martinez, J. (1988) J Biol Chem 263(4), 1791-1797 132. Wierzbicka-Patynowski, I., and Schwarzbauer, J. E. (2003) J Cell Sci 116(Pt 16), 3269-3276 133. Wu, C., Keightley, S. Y., Leung-Hagesteijn, C., Radeva, G., Coppolino, M., Goicoechea, S., McDonald, J. A., and Dedhar, S. (1998) J Biol Chem 273(1), 528-536 134. Wu, C., Keivens, V. M., O'Toole, T. E., McDonald, J. A., and Ginsberg, M. H. (1995) Cell 83(5), 715-724 135. Darribere, T., Koteliansky, V. E., Chernousov, M. A., Akiyama, S. K., Yamada, K. M., Thiery, J. P., and Boucaut, J. C. (1992) Dev Dyn 194(1), 6370 43 136. Darribere, T., Guida, K., Larjava, H., Johnson, K. E., Yamada, K. M., Thiery, J. P., and Boucaut, J. C. (1990) J Cell Biol 110(5), 1813-1823 137. McDonald, J. A., Quade, B. J., Broekelmann, T. J., LaChance, R., Forsman, K., Hasegawa, E., and Akiyama, S. (1987) J Biol Chem 262(7), 2957-2967 138. Wu, C., Hughes, P. E., Ginsberg, M. H., and McDonald, J. A. (1996) Cell Adhes Commun 4(3), 149-158 139. Fogerty, F. J., Akiyama, S. K., Yamada, K. M., and Mosher, D. F. (1990) J Cell Biol 111(2), 699-708 140. Roman, J., LaChance, R. M., Broekelmann, T. J., Kennedy, C. J., Wayner, E. A., Carter, W. G., and McDonald, J. A. (1989) J Cell Biol 108(6), 2529-2543 141. Giancotti, F. G., and Ruoslahti, E. (1990) Cell 60(5), 849-859 142. Yang, J. T., Rayburn, H., and Hynes, R. O. (1993) Development 119(4), 10931105 143. Sechler, J. L., Takada, Y., and Schwarzbauer, J. E. (1996) J Cell Biol 134(2), 573-583 144. Schlaepfer, D. D., Hauck, C. R., and Sieg, D. J. (1999) Prog Biophys Mol Biol 71(3-4), 435-478 145. Zhao, J. H., and Guan, J. L. (2000) Prog Mol Subcell Biol 25, 37-55 146. Sechler, J. L., and Schwarzbauer, J. E. (1998) J Biol Chem 273(40), 2553325536 147. Olden, K., and Yamada, K. M. (1977) Cell 11(4), 957-969 148. Chen, Q., Kinch, M. S., Lin, T. H., Burridge, K., and Juliano, R. L. (1994) J Biol Chem 269(43), 26602-26605 149. Morino, N., Mimura, T., Hamasaki, K., Tobe, K., Ueki, K., Kikuchi, K., Takehara, K., Kadowaki, T., Yazaki, Y., and Nojima, Y. (1995) J Biol Chem 270(1), 269-273 150. Schlaepfer, D. D., Hanks, S. K., Hunter, T., and van der Geer, P. (1994) Nature 372(6508), 786-791 151. Derijard, B., Hibi, M., Wu, I. H., Barrett, T., Su, B., Deng, T., Karin, M., and Davis, R. J. (1994) Cell 76(6), 1025-1037 44 152. Davis, R. J. (1994) Trends Biochem Sci 19(11), 470-473 153. Oehlen, B., and Cross, F. R. (1994) Curr Opin Cell Biol 6(6), 836-841 154. Miyamoto, S., Teramoto, H., Coso, O. A., Gutkind, J. S., Burbelo, P. D., Akiyama, S. K., and Yamada, K. M. (1995) J Cell Biol 131(3), 791-805 155. Yamada, K. M., and Miyamoto, S. (1995) Curr Opin Cell Biol 7(5), 681-689 156. Grinnell, F., and Geiger, B. (1986) Exp Cell Res 162(2), 449-461 157. Mueller, S. C., Kelly, T., Dai, M. Z., Dai, H. N., and Chen, W. T. (1989) J Cell Biol 109(6 Pt 2), 3455-3464 158. Plopper, G., and Ingber, D. E. (1993) Biochem Biophys Res Commun 193(2), 571-578 159. Lewis, J. M., and Schwartz, M. A. (1995) Mol Biol Cell 6(2), 151-160 160. Marshall, C. J. (1996) Curr Opin Cell Biol 8(2), 197-204 161. Alessi, D. R., Saito, Y., Campbell, D. G., Cohen, P., Sithanandam, G., Rapp, U., Ashworth, A., Marshall, C. J., and Cowley, S. (1994) Embo J 13(7), 16101619 162. Kyriakis, J. M., App, H., Zhang, X. F., Banerjee, P., Brautigan, D. L., Rapp, U. R., and Avruch, J. (1992) Nature 358(6385), 417-421 163. Shin, I., Kim, S., Song, H., Kim, H. R., and Moon, A. (2005) J Biol Chem 280(15), 14675-14683 164. Hughes, P. E., Renshaw, M. W., Pfaff, M., Forsyth, J., Keivens, V. M., Schwartz, M. A., and Ginsberg, M. H. (1997) Cell 88(4), 521-530 165. Brenner, K. A., Corbett, S. A., and Schwarzbauer, J. E. (2000) Oncogene 19(28), 3156-3163 166. Hynes, R. O. (1992) Cell 69(1), 11-25 167. Schwartz, M. A., Schaller, M. D., and Ginsberg, M. H. (1995) Annu Rev Cell Dev Biol 11, 549-599 168. Ginsberg, M. H., Du, X., and Plow, E. F. (1992) Curr Opin Cell Biol 4(5), 766771 45 169. O'Toole, T. E., Katagiri, Y., Faull, R. J., Peter, K., Tamura, R., Quaranta, V., Loftus, J. C., Shattil, S. J., and Ginsberg, M. H. (1994) J Cell Biol 124(6), 1047-1059 170. Dedhar, S., and Hannigan, G. E. (1996) Curr Opin Cell Biol 8(5), 657-669 171. Zhang, Z., Vuori, K., Wang, H., Reed, J. C., and Ruoslahti, E. (1996) Cell 85(1), 61-69 172. Hannigan, G. E., Leung-Hagesteijn, C., Fitz-Gibbon, L., Coppolino, M. G., Radeva, G., Filmus, J., Bell, J. C., and Dedhar, S. (1996) Nature 379(6560), 91-96 173. Chung, C. Y., and Erickson, H. P. (1997) J Cell Sci 110 ( Pt 12), 1413-1419 174. Mercurius, K. O., and Morla, A. O. (2001) BMC Cell Biol 2, 18 175. Klass, C. M., Couchman, J. R., and Woods, A. (2000) J Cell Sci 113 ( Pt 3), 493-506 176. Saoncella, S., Echtermeyer, F., Denhez, F., Nowlen, J. K., Mosher, D. F., Robinson, S. D., Hynes, R. O., and Goetinck, P. F. (1999) Proc Natl Acad Sci U S A 96(6), 2805-2810 177. Wilcox-Adelman, S. A., Denhez, F., and Goetinck, P. F. (2002) J Biol Chem 277(36), 32970-32977 178. Midwood, K. S., and Schwarzbauer, J. E. (2002) Mol Biol Cell 13(10), 36013613 179. Somers, C. E., and Mosher, D. F. (1993) J Biol Chem 268(30), 22277-22280 180. Yang, R. S., Tang, C. H., Ling, Q. D., Liu, S. H., and Fu, W. M. (2002) Mol Pharmacol 61(5), 1163-1173 181. Lin, W., Wang, S. M., Huang, T. F., and Fu, W. M. (2002) Connect Tissue Res 43(1), 22-31 182. Tang, C. H., Yang, R. S., Huang, T. H., Liu, S. H., and Fu, W. M. (2004) Mol Pharmacol 66(3), 440-449 183. Kaiura, T. L., Itoh, H., and Kent, K. C. (1999) J Surg Res 84(2), 212-217 184. Lee, B. H., Park, R. W., Choi, J. Y., Ryoo, H. M., Sohn, K. Y., and Kim, I. S. (1996) Biochem Mol Biol Int 39(5), 895-904 46 185. Lin, S., Sahai, A., Chugh, S. S., Pan, X., Wallner, E. I., Danesh, F. R., Lomasney, J. W., and Kanwar, Y. S. (2002) J Biol Chem 277(44), 4172541735 186. Osusky, R., Soriano, D., Ye, J., and Ryan, S. J. (1994) Curr Eye Res 13(8), 569-574 187. Singh, L. P., Andy, J., Anyamale, V., Greene, K., Alexander, M., and Crook, E. D. (2001) Diabetes 50(10), 2355-2362 188. Horowitz, A., and Simons, M. (1998) J Biol Chem 273(40), 25548-25551 189. Oh, E. S., Woods, A., and Couchman, J. R. (1997) J Biol Chem 272(13), 8133- 8136 190. Fidler, I. J. (2003) Nat Rev Cancer 3(6), 453-458 47 CHAPTER II PKC ΕPSILON MEDIATES POLYMERIC FIBRONECTIN ASSEMBLY ON THE SURFACE OF BLOOD-BORNE RAT BREAST CANCER CELLS TO PROMOTE PULMONARY METASTASIS* * Huang L, Cheng HC, Isom R, Chen CS, Levine RA, Pauli BU. Protein kinase C epsilon mediates polymeric fibronectin assembly on the surface of blood-borne rat breast cancer cells to promote pulmonary metastasis. J Biol Chem. 2008 Mar 21;283(12):7616-27. (Huang L: immunofluorescence, FACS, biochemistry; Cheng HC: initial drug screening; Isom R, Chen CS: constructs; Levine R: siRNA constructs advice) 48 ABSTRACT Malignant breast cancer cells that have entered the blood circulation from primary mammary fat pad tumors or are grown in end-over-end suspension culture assemble a characteristic, multi-globular polymeric fibronectin (polyFn) coat on their surfaces. Surface polyFn is critical for pulmonary metastasis, presumably by facilitating lung vascular arrest via endothelial dipepidylpeptidase IV (DPP IV; CD26). Here, we show that cell-surface polyFn assembly is initiated by the state of suspension, is dependent upon the synthesis and secretion of cellular Fn, and is augmented in a dose- and time-dependent manner by plasma Fn (pFn). PolyFn assembly is regulated by PKCε, which translocates rapidly and in increasing amounts from the cytosol to the plasma membrane and is phosphorylated. PolyFn assembly is impeded by select inhibitors of this kinase, i.e., bisindolylmaleimide I, Ro-32-0432, Gö6983, and Rottlerin, by the PMA-mediated and time-dependent loss of PKCε protein and decreased plasma membrane-translocation and, more specifically, by stable transfection of lung-metastatic MTF7L breast cancer cells with siRNA-PKCε and dominant-negative PKCε constructs (e.g., RD-PKCε). Inability to assemble a cell surface-associated polyFn coat by knockdown of endogenous Fn or PKCε impedes cancer cells from metastasis to the lungs. The present studies identify a novel regulatory mechanism for polyFn assembly on blood-borne breast cancer cells and depict its effect on pulmonary metastasis. INTRODUCTION Fibronectin (Fn) is a “pro-metastatic” gene that is overexpressed in several malignancies (1-7) and, most prominently, in cancer cell lines selected for enhanced 49 lung colonization (8-11). This pro-metastatic role of Fn is multifaceted, affecting several steps of the metastatic cascade by modulating cell adhesion, motility/invasion, cell cycle progression, and cell survival (reviewed in refs. 12-16). By analyzing cancer cells that had entered the blood circulation from malignant breast cancers implanted into the mammary fat pad of rats or mice, we discovered that blood-borne tumor cells were decorated with a unique, multi-globular coat of polymeric Fn (polyFn) (9,17). PolyFn aggregates appeared to arise from focal accumulations of endogenous, cell surface-immobilized (‘linearized’) Fn, which served as scaffolds for further Fn-selfassembly from Fn recruited from blood plasma (pFn) (9). Aggregates typically became increasingly deoxycholate-insoluble as they increased in size with time of incubation of suspended cancer cells in serum-containing medium in vitro. Biochemically, aggregates impressed as prominent, insoluble (covalently bonded) Fn polymers sitting on top of the stacks of SDS-polyacrylamide gels and, immunocytochemically, as large globules randomly dispersed over the entire cancer cell surface (9). This “cluster arrangement” of polyFn was shown to have the following functional implications: First, the conversion of Fn from the globular state of soluble Fn to the “linearized” state of insoluble, surface-associated Fn aggregates is associated with exposure of a novel, cryptic binding domain for the lung endothelial cell addressin dipeptidylpeptidase IV (DPP IV) (9,17-21). This DPP IV-binding domain is present as consensus motif in each of the 13th, 14th, and 15th type III repeats of Fn (17). Second, the DPP IV binding specificity for linearized (polymeric), but not for globular (soluble) pFn allows tumor cell adhesion to endothelial DPP IV in the presence of high pFn concentrations (9). Third, the large Fn aggregates on cancer cell surfaces allow multiple binding interactions with endothelial DPP IV molecules, thereby generating adhesion strengths between cancer cells and endothelial cells that are able to withstand the rigors of hemodynamic shear stresses (9). The importance of 50 the Fn/DPP IV-docking mechanism is substantiated by our discovery that synthetic peptides directed against the DPP IV-binding domain in the 13th, 14th, or 15th type III repeats of Fn as well as a polypeptide encompassing the bulk of the extracellular domain of DPP IV dramatically impeded pulmonary metastasis in the MT rat breast cancer model (19,20). These data are consistent with our finding that colonization of the lungs was greatly diminished in Fischer 344/CRJ rats, in which DPP IV is mutated causing a significantly decreased DPP IV protein expression in pulmonary endothelia (21), as well as in DPP IV(-/-) mice (19). DPP IV(-/-) mice injected with lungmetastatic cancer cells lived significantly longer than their wild-type counterparts, an outcome granted by the formation of significantly fewer and smaller lung colonies (Cheng, unpublished data). Although we have firmly established a critical dependence between the cancer cells’ ability to assemble an insoluble, globular polyFn-surface coat and lung colonization in an experimental metastasis model (9,17), we still do not know whether blood-borne cancer cells use their own cellular Fn (cFn) or rely on ubiquitous pFn to assemble their polyFn surface coat, how cancer cells regulate the polyFn build-up, and how Fn cell surface deposits are transformed into covalently bonded, insoluble aggregates (9). To answer some of these questions we examined polyFn genesis in lung-metastatic MTF7L rat breast cancer cells subjected to end-over-end (EoE) suspension culture in serum- or pFn-containing medium, which together induce and augment the build-up of a polyFn surface coat similar to that observed on tumor cells that have entered the blood circulation (9). The data presented here show that assembly of the “pro-metastatic”, multi-globular polyFn surface coat on MTF7L breast cancer cells depends upon the synthesis and secretion of endogenous, cellular Fn (Fn1: EDA+EDB+IIICS120-Fn) and is regulated by membrane-translocation and Ser/Thr-phosphorylation of protein kinase C epsilon (PKCε). Inhibitors of protein 51 synthesis, protein secretion, and novel PKC isoforms (nPKCs) as well as transfection of MTF7L cells with the PKCε regulatory domain (RD) or PKCε siRNA species all substantially decrease the cancer cells’ ability to assemble a polyFn surface coat. Functionally, inability to assemble polyFn is associated with failure to colonize the lungs. Together, our studies provide novel insights of the regulation of Fn in suspended (blood-borne) breast cancer cells and provide a renewed appreciation for the previously recognized role of Fn in metastasis (1-7). MATERIAL AND METHODS Antibodies and Reagents Rabbit anti-PKCδ, -PKCε, PKCη, PKCθ, PKCζ and anti-hemagglutinin tag (HA-tag) polyclonal antibodies, and mouse anti-Fn (raised against a region in the human Fn-EDA domain; human- mouse- and rat-specific) monoclonal antibody (antiFn[EDA]) were from Santa Cruz Biotechnology (Santa Cruz, CA), rabbit anti-Fn polyclonal antibodies that recognized pFn and cFn from both bovine and rat (antiFn[pan]: does not crossreact with fibrinogen, vitronectin, laminin, collagen type IV) from Sigma (St. Louis, MO), rabbit anti-PKCε polyclonal antibodies from Upstate (Lake Placid, NY), mouse anti-PKCε used for immunoprecipitation from BD Biosciences (San Jose, CA), phycoerythrin (PE)-conjugated donkey anti-rabbit, PEconjugated goat anti-mouse, horseradish-peroxidase (HRP)-conjugated donkey antirabbit, and HRP-conjugated goat anti-mouse antibodies from Jackson ImmunoResearch (West Grove, PA), and rabbit anti-PKCα, -vinculin, and -actin from Dr. Guan (Cornell University). Pertussis toxin (Gαi inhibitor), PD98059 (MEK1/2 inhibitor), SU6656 and PP2 (Src family kinase inhibitors), Wortmannin and LY294002 (PI3K inhibitors), JAKI (Janus-family kinase 1/2 inhibitor), and Y27632 52 (ROCK1/2 inhibitor), U73122 (inhibitor of phosphatidyl-inositol-specific phospholipase C [PI-PLC]), and the PKC inhibitors calphostin C, Gö6976, HBDDE, bisindolylmaleimide I (BIM I), Gö6983, BIM XI (Ro-32-0432), Rottlerin, Brefeldin A, monensin, and cycloheximide were from EMD Chemicals (San Diego, CA). Fetal bovine serum (FBS) was purchased from Gemini Bio-Products (Woodland, CA). Fnfree FBS (FFS) was generated by successive gelatin- and anti-Fn antibody affinity chromatography (22). All other chemicals and reagents were from Sigma. Cell Cultures MTF7L cells were derived from a lung-metastasis generated by tail-vein injection of MTF7 breast cancer cells (obtained from Dr. D.R. Welch, University of Alabama at Birmingham, Birmingham, AL) into Fischer 344 rats. At an i.v. inoculation dose of 2x105 cell per rat, MTF7L cells consistently produce in excess of 400 lung colonies. Cells were grown in culture in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% heat-inactivated fetal bovine serum (FBS). For EoE suspension culture, MTF7L cells were grown to 80 to 90% confluence, then removed from the growth surface by trypsinization [0.25% trysin, 0.02% EDTA in phosphatebuffered saline (PBS); 10 min, 37°C], washed twice in DMEM containing 10% FBS, and subjected to EoE suspension culture for 1 hour (or as indicated) in 2ml-centrifuge tubes in DMEM plus 20% FBS at a concentration of 5x106 cells/ml (9,17,20). Tumor cells were used for all experiments within 10 passages from frozen stocks that were tested for metastatic performance immediately prior to freezing. For metabolic labeling, MTF7L cells in logarithmic growth phase were labeled with [35S]-methionine (0.33 mCi/3-ml) in methionine-free DMEM (both from MP Biomedicals, Solon, OH) containing 20 µM methionine and 10% dialyzed, Fn-free FBS as previously described (9). For [32P]-labeling, cells were serum-starved 53 overnight, then incubated for 4h in phosphate-free DMEM containing 100 µCi/ml of [32P]orthophosphate (ICN Biochemicals, Irvine, CA), washed in three changes of PBS. Labeled cells were subjected to EoE suspension culture as describe above, and immediately processed for biochemical analyses. Plasmids Constructs, Transfection, and Selection The constructs wtPKCε, RD-PKCε, and RD-PKCη cloned into pEGFP-N1 were obtained from Dr. C. Larsson (Lund University, Malmö, Sweden) (23) and wtPKCδ and RD-PKDδ cloned into pcDNA3.1 from Dr. D. Mayer (Deutsches Krebsforschungszentrum, Heidelberg, Germany) (24). For siRNA knockdown of protein expression the following nucleotide (nt.) sequences were cloned into pRNAU6-hygro vector (GenScript, Piscataway, NJ): nt. sequence #1 [5’acatgagactggtggctat-3’ (NM_019143.1, nt. 681-699)] and nt. sequence #2 [5’aacaaatctcctgcctgggac-3’ (NM_019143.1, nt. 4452-4472)] (25) for rat Fn1; nt. sequence #3 [5’-atggtagtgttcaatggc-3’ (NM_017171.1, nt. 194-211)] (26); nt. sequence #4 [5’-ccaactctattgctgcttc-3’ (NM_017171.1, nt. 1603-1621)] for rat PKCε; and nt. sequence #5 [5’-aactcgcacatagcgactctg-3’] for the non-specific control sequence. The siRNA plasmid pKD™-PKCδ-v6 was purchased from Upstate (Lake Placid, NY). All plasmid constructs were verified by double-stranded sequencing. MTF7L cells grown to 70% confluence were transiently transfected with above vector constructs or vector alone using LipofectAMINE™ Plus as described by the manufacturer (Invitrogen, Carlsbad, CA). Transfection rates assessed by expression of GFP that is either tagged to the cDNA of interest or co-transfected at a ratio of 1:50 with the cDNA of interest were 20-30%. Cells were used in the various assays 48h after transfection unless otherwise stated. Stable clones were obtained by hygromycin 54 selection (750µg/ml). In some cases, hygromycin-selected clones were further selected by fluorescence-activated cell sorting (FACS) for optimal expression. Flow Cytometry FACS was used to quantify Fn expression on MTF7L breast cancer cell surfaces (9,17). Tumor cells that had been subjected to EoE suspension culture were washed twice in DMEM containing 1% bovine serum albumin (BSA), then incubated with rabbit anti-Fn[pan] antibody diluted 1:100 in PBS containing 1%BSA (PBSBSA) for 1h at 4°C . After washing in PBS-BSA, tumor cells were stained with phycoerythrin (PE)-conjugated donkey anti-rabbit antiserum in PBS-BSA for 1h at 4°C and fixed in 2% paraformaldehyde in PBS. In select experiments, cells were stained with mouse anti-Fn[EDA] (diluted 1:50) and PE-conjugated goat anti-mouse antiserum. FACS analysis was performed on a Coulter Epics Profile (Coulter Electronics, Hialeah, FL). Nonspecific fluorescence was accounted for by incubating tumor cells with non-immune rabbit serum instead of primary antibody. To quantify the effect of overexpressed or knocked-down proteins on polyFn assembly, we generated bivariate distributions of red fluorescence (y-axis: cells stained with anti-Fn antibodies and PE-conjugated secondary antibodies) and green fluorescence (x-axis: same cells expressing GFP-tagged protein or co-transfected with GFP and cDNA of interest). The levels of polyFn expression in the cell population that emitted high GFP fluorescence were taken as a reflection of the effect of the transfected cDNA on polyFn assembly. To assess the effect of inhibitors of cell signaling, tumor cells were incubated with inhibitor 30-min prior to (adherent) and throughout EoE suspension culture, then subjected to polyFn quantification as described above. Controls were tumor cells incubated in equimolar inhibitor solvent concentration. 55 Semiquantitative RT-PCR Analyses Total RNA was prepared from MTF7L grown as adherent monolayers or in EoE suspension cultures by extraction with Trizol as described by the manufacturer (Invitrogen). For every experimental sample, total RNA was quantified both spectrophotometrically and electrophoretically, and amounts were adjusted so that 1µg was reverse-transcribed (SuperScript reverse transcriptase, Invitrogen). cDNA was subjected to PCR (93oC, 30"; 55oC, 30"; 72oC, 30"; 35 cycles) using Taq DNA polymerase (Invitrogen) and primer sets derived from rat Fn1 (NM_019143), PKCδ (NM_133307), PKCε (NM_017171), PKCη (NM_031085), PKCθ (XM_341553), and PKCζ (ΝΜ_022507). Controls were run in the absence of reverse transcriptase. GAPDH served as reference standard. Cell Fractionation MTF7L cells and transfectants thereof were incubated in EoE suspension culture in DMEM containing 20% FBS for the indicated periods of time. Cells were washed in PBS, collected by centrifugation and resuspended in 0.5 ml ice-cold buffer A [50mM Tris (pH 8.0), 150mM NaCl, 1mM EDTA, 1mM EGTA, 1mM NaF, 0.1mM NaVO4, 20µM leupeptin, 0.1% aprotinin, and 1mM PMSF]. Cells were disrupted by two 15-sec cycles of sonication at 4°C using a microprobe sonicator at maximum power. After removal of unbroken cells and nuclei by centrifugation (500xg for 5 min at 4°C), supernatants were centrifuged at 16,300xg for 15 min at 4°C. The resulting supernatant was designated the cytosolic fraction. The pellet was solubilized in 0.5ml buffer A containing 1% Triton X-100 for 1 hour at 4°C EoE, then centrifuged at 16,300xg for 15 min at 4°C. The detergent-solute was designated the membrane fraction. Twenty micrograms of protein from both the cytosolic and membrane 56 fractions were separated by SDS–PAGE (10%), transferred to nitrocellulose membrane at 4°C, and probed by Western blotting as described (9). Cell Lysis, Immunoprecipitation, Western Blotting, and Autoradiography Cells were extracted with lysis buffer (50mM Tris-HCl, pH 7.4, 150mM NaCl, 1mM EDTA, 1mM EGTA, 1mM benzamidine chloride, 1mM PMSF, 2 µg/ml leupeptin, 0.27 TIU/ml aprotinin, 0.1mM sodium vanadate, and 1% Triton X-100) for 1h at 4°C (9). Total cell lysates or cytosolic and membrane fractions were subjected to: (i) SDS-PAGE (~20-50µg protein) and Western blotting, using anti-Fn[pan], antiFn[EDA], or various PKC isoform specific antibodies, HRP-conjugated donkey antirabbit or goat anti-mouse secondary antibodies and ECL for detection of bound antibody as described (9); (ii) immunoprecipitation with anti-Fn[pan], anti-PKCε or anti-PKCδ antibodies (27). Immunoprecipitates obtained from lysates of unlabeled, [32P]-orthophosphate-, or [35S]-methionine labeled cells were separated by SDS-PAGE (6-12% polyacrylamide) and analyzed by autoradiography (radio-labeled samples) or blotted to nitrocellulose membranes and probed with either anti-Fn[pan], antiFn[EDA], anti-PKC isoform-specific antibodies, or anti-pSer and anti-pThr antibodies (9). Tumor Cell Proliferation Assay MTF7L cells and clones thereof were seeded into 96-well microtitration plates (500 cells/well) and incubated in DMEM containing 10% FBS. Cell growth was monitored in daily intervals for up to 4 days. At the end of each incubation period, tumor cells were fixed with 2% paraformaldehyde in PBS and then stained with 0.5% crystal violet in 20% methanol as described (21). Absorbance was read on a 57 microplate reader (Bio-Teck Instruments) at 562nm and graphed as a function of time of incubation. Isolation of Blood-Borne Cancer Cells from Tumor-Bearing Rats MTF7L cancer cells (1x106 cells/50µl DMEM) were injected into the 4th (left+right) mammary fat pads of six 6-week-old, female Fischer 344 rats. At a tumor diameter of ~2cm, rats were anesthetized by the intraperitoneal injection of sodium pentobarbital (65mg/kg body weight) and blood collected by cardiac puncture. Pooled, EDTA-treated blood was transferred to precooled 50-ml centrifuge trubes containing 15ml of OncoQuick tumor enrichment medium below a porous barrier (Greiner BioOne, Longwood, FL) and centrifuged at 1600g for 20min at 4°C in a swing-out rotor as described by Rosenberg et al. (28). After a second round of centrifugation of the fluid in the upper compartment, cells were washed with PBS-BSA and stained with anti-Fn antibodies as described above. Tumor cells were readily differentiated from contaminant blood mononuclear cells by size and intensity of anti-Fn staining in comparison with preparations from MTF7L cell-spiked blood. Lung Colony Assays To determine the effects of Fn- or PKCε-knockdown on the lung colony efficiency, selected MTF7L clones as well as wt or vector-transfected MTF7 cells (2x105 cells/0.3ml DMEM/rat) were injected into the lateral tail vein of 5-week-old, female Fischer344 rats (5-9 rats/experiment) as described (9,21). Rats were sacrificed 30 days after tumor cell injection. Means and standard deviations of the lung weights, the number of lung colonies, and the colony diameters were determined for each cell variant (21). 58 RESULTS Endogenous Fn Is Required for PolyFn Assembly on Suspended Breast Cancer Cells Breast cancer cells entering the blood circulation assemble a characteristic surface coat of globular polyFn as evidenced by anti-Fn[pan] staining of cancer cells isolated from the blood of mammary tumor-bearing rats (Fig 2.1A). This phenomenon is mimicked when cancer cells are grown in EoE suspension culture in the presence of serum (9). To examine whether polyFn assembly is mediated by exogenous pFn alone or whether it critically depends on the synthesis of endogenous cFn, we incubated MTF7L breast cancer cells, which express high levels of Fn1 (EDA+EDB+CSIII120), in EoE suspension culture in DMEM containing either 20% complete FBS or Fn-free fetal bovine serum (FFS). FFS-treated tumor cells stained with either anti-Fn[pan] (Fig 2.1B) or anti-Fn[EDA] (Fig 2.1B, inset) exhibited similar numbers of polyFn globules on their surfaces as FBS-treated cells (Fig 2.1C). However, the addition of complete FBS to the suspension medium significantly augmented the size of individual polyFnglobules relative to those generated by Fn-free FBS (FFS) (Fig 2.1B&C). The serumeffect on polyFn assembly appeared to be mediated only in part by pFn as shown by FACS analysis of MTF7 cells incubated in EoE suspension culture in the presence of (i) 20% FBS (Fig 1D&E), (ii) 6µg/ml pFn (amount present in 20% FBS [22]), or (iii) 20% FFS (Fig 1E). Both 20% FFS and pFn (6µg/ml) were significantly less effective in promoting polyFn assembly than FBS (Fig 2.1E). The potency of FFS to promote polyFn assembly could be restored to that of FBS by the addition of 6µg/ml pFn (Fig 2.1E, inset). Thus, FFS appeared to promote surface deposition of cellular Fn (Fn1) on suspended MTF7L cells, while pFn in a time- and dose-dependent manner contributed 59 Figure 2.1: PolyFn globules expressed on the surface of suspended (blood-borne) MTF7L breast cancer cells: (A) MTF7L breast cancer cells were isolated from blood of tumor-bearing rats, using the OncoQuick tumor enrichment medium. The mononuclear cell fraction separated from red blood cells, platelets, and polymorphonuclear leukocyte fractions was washed several times in PBS, then stained with rabbit anti-Fn[pan] antibodies followed by PE-conjugated donkey anti-rabbit IgG antibodies and observed under a fluorescent microscope. There are multiple Fn globules randomly dispersed over the cancer cell surface. (B & C) MTF7L cells subjected to EoE suspension culture in DMEM+20% Fn-free FBS (FFS) (B) or DMEM+20% complete FBS (C). Both are incubated for 1h at 37°C, then stained as described under (A). Notice, the same numerical density of Fn stipples are observed in (B) and (C), but a larger stipple size is present in (C) mimicking those observed in (A). C-Inset: Same treatment, but stained with anti-Fn[EDA]. (D) FACS analysis of MTF7L cells incubated EoE for 1h in DMEM+20% FBS and stained with antiFn[pan] antibodies as described under (A). (E) FACS analysis of MTF7L cells incubated EoE for 1h in DMEM+6µg/ml pFn (green), DMEM+20% FFS (red), or DMEM+20% FBS (black). Notice that tumor cells treated with DMEM+20% FBS (containing ~6µg/ml pFn) exhibit a higher mean fluorescence than tumor cells treated with DMEM+20% FFS or DMEM+6µg/ml pFn. E-Inset: Tumor cells incubated with 20% FFS+6µg/ml (red) and 20% FBS (green) show identical polyFn assembly. (F) FACS analysis of MTF7L cells treated for the indicated periods of time with DMEM+20%FBS. There is a time-dependent increase in polyFn assembly. Bars in AC: 50µm 60 61 to augmentation of the Fn polymers by Fn-Fn self assembly (Fig 2.1F). To further test the hypothesis that surface deposition of a scaffold of endogenous cFn was important for the initiation of polyFn assembly, we examined whether inhibitors of protein synthesis (e.g., cycloheximide) (29) or secretion (e.g., monensin, Brefeldin A) (30,31) would affect polyFn assembly on suspended MTF7L breast cancer cells. Cycloheximide, monensin, and Brefeldin A all dramatically impeded polyFn assembly on MTF7L breast cancer cells in the presence of 20% FBS (Fig 2.2). Since these compounds might have also impaired surface expression of Fn receptors, we used RNA interference to knockdown endogenous cFn in MTF7L (Fn1) cells and to examine whether the specific inhibition of MTF7L-cFn would affect polyFn assembly. Biochemically, stable clones derived from siRNA-Fn transfected cells exhibited a significant decrease in the expression cFn protein as depicted by Western blotting (Fig 2.3A; clones cl1 & cl2) and by a dramatic decrease in insoluble, metabolically labeled polyFn residing on top of the stack of the polyacrylamide gel (Fig 2.3B, clone cl1). Functionally, knockdown of endogenous, cellular Fn was associated with impaired surface polyFn assembly on MTF7L cells as shown by FACS (Fig 2.3C) and immunocytochemistry (Fig 2.3D&E), albeit all tumor cells were incubated in the presence of 20%FBS containing ~6µg/ml pFn. An unspecific nucleotide sequences had no affect (Fig 2.3A-E: mclus). Stable siRNA-Fn clones were also used to test their lung metastatic potential in lung colony assays. In accordance with the level of knockdown of endogenous Fn, lung colonization, assessed by the averages of lung weights, colony numbers, and colony diameters, was significantly decreased relative to wtMTF7L cells or tumor cells that were stably transfected with an unspecific nucleotide sequence (Fig 2.4; siFn-cl1). 62 Figure 2.2: Inhibitors of protein synthesis and secretion prevent polyFn assembly: Adherent MTF7L rat breast cancer cells were exposed for 30 min to 10µg/ml of cycloheximide (A), 1µM of monensin (B), or 5µM of Brefeldin A (C), then subjected to EoE suspension cultured in DMEM+20%FBS in the presence of the respective drug or drug solvent (S). After a 1-hr incubation period, cells were stained with anti-Fn[pan] antibodies as described in figure 2.1A (control: MTF7L cell stained with non-immune rabbit IgG [rIgG]). Tumor cells treated with cycloheximide, monensin, and Brefeldin A exhibit a dramatic reduction in the cell surface-associated polyFn, relative to those treated with drug solvent. 63 64 Figure 2.3: Effect of siRNA knockdown of cFn on polyFn assembly in the presence of 20% FBS: Hygromycin-resistant clones were selected from MTF7L breast cancer cells transfected with the rat Fn1 nt. sequences #1 (clones cl1 and cl2), rat Fn1 nt. sequence #2 (clone cl3), or unspecific nt. sequence #5 (multi-clone mclus) all cloned into pRNAU6-Hygro for siRNA targeting as described in Material and Methods. (A) Anti-Fn[pan] Western blot from lysates of siFn clones cl1 and cl3, and the contol clone mclus. (B) MTF7L cells metabolically labeled with [35S]-methionine as described in Material and Methods and ref. 9 were subjected to EoE culture in DMEM+20% FBS for 4h, then extracted in lysis buffer containing 1% NP-40 (9). Lysates were immunoprecipitated with anti-Fn[pan] and immunoprecipitated proteins separated by SDS-PAGE (6%) under non-reducing conditions. Notice the significant reduction in the amount of anti-Fn[pan] immunoprecitable material (polyFn) residing on top of the stacking gel. (C) FACS analysis of MTF7L siFn clones cl1, cl2, and cl3 (green) and the control clone mclus (black: control), cultured and stained with antiFn[pan] as described in figure 1A. Mc=mean fluorescence of multi-clone mclus (back); MsiFn= mean fluorescence of siFn clones cl1, cl2, and cl3 (green). (D &E) Micrographs from MTF7L siRNA-Fn clone cl1 (D) and control-clone mclus (E), both processed and stained with anti-Fn[pan] as described in figure 2.1A. Notice a significantly decreased polyFn assembly in MTF7L siFn clone cl1 (C), relative to MTF7L control clone mclus (D), even though both clones (D&E) were incubated with DMEM+20% FBS for the same period of time (1h at 37°C) in EoE suspension culture. Bars in D&E: 50µm. 65 66 Figure 2.4: Fn and PKCε siRNA-knockdown inhibit pulmonary metastasis: Cells (2x105 cells/0.3ml DMEM/rat) derived from MTF7L clone mclus (control clone), clone siFn-cl1 (see Fig 2.3) and MTF7L clone siPKε-cl1 (see Fig 2.8) were injected into the lateral tail vein of female Fischer344 rats as described in Material and Methods. Numbers of lung colonies were significantly decreased in rats injected with MTF7L-siFn-cl1 (*) and, most prominently, with MTF7L-siPKCε-cl1 (*) relative to rats injected MTF7L-mclus cells. Colony diameters were reduced by approximately 5fold in MTF7L-siFn-cl1 (*) and 2-fold in MTF7L-siPKCε-cl1 relative to MTF7-mclus, in which the coalescence of individual colonies to large aggregates might have led to an inflated estimate of the colony diameter. [Results presented in this figure may reflect to some extent a reduced in vitro growth rate of 29% for MTF7L-siFn-cl1 and 12% for MTF7L-siPKCε-cl1 cells relative to MTF7-mclus observed after a 4-day culture period in vitro]. *, Student’s t-test: p<0.01. 67 68 PolyFn Assembly Occurs in a PKC-Dependent Manner To identify the signaling cascade that regulates polyFn assembly, we screened a diverse group of inhibitors of cell signaling for their ability to impede polyFn assembly as determined by routine FACS analyses of anti-Fn[pan] stained MTF7L breast cancer cells subjected for 1 hour to EoE suspension culture in FBS (20%)containing medium in the presence or absence of inhibitor. These inhibitors included pertussis toxin, PD98059, SU6656, PP2, Wortmannin, LY294002, JAKI, and Y27632. None of these inhibitors had any effect on the assembly of the polyFn surface coat of MTF7L breast cancer cells (data not shown). Consistent with the negative result obtained with the ROCK1/2 inhibitor Y27632 is the finding that the Rho activity, associated with Fn-matrix assembly and stress fiber formation in adherent cells (32,33), steadily decreased during the polyFn assembly phase on suspended breast cancer cells. Moreover, transfection with constitutively active Rho (RhoAQ63L) suppressed polyFn assembly, while transfection with dominant negative Rho (RhoAT19N) appeared to promote polyFn assembly (Cheng and Pauli, manuscript in preparation). In contrast, polyFn assembly was dramatically reduced by U73122, an inhibitor of phospholipase C (PLC) (34) and the PKC inhibitor calphostin C, which competes with the diacylglyceral and phorbol ester binding site of conventional PKCs (cPKCs) and nPKCs (35) (Fig 2.5A). Together, these findings suggest involvement of a cPKC or nPKC isoform, but not an atypical PKC (aPKC) isoform, presumably acting downstream of PLC in polyFn assembly on MTF7L cancer cell surfaces (36). PKCε Is the Mediator of polyFn Assembly and Metastasis Our attempts to identify the PKC isoform responsible for polyFn assembly were preceded by determining the c- and nPKC expression levels in MTF7L breast cancer cells. As reported for adherent, metastatic MT tumor variants by Kiley et al. 69 Figure 2.5: Inhibitors of PLC and PKC diminish polyFn assembly: (A) MTF7L breast cancer cells were incubated in the presence of U73122 (PLC inhibitor) and calphostin C (inhibitor of cPKCs and nPKCs) at the indicated concentrations or inhibitor solute (DMSO, control) for 30min, then subjected to EoE suspension culture for 1h in DMEM+20%FBS+inhibitor (or inhibitor solute [S]). Cells were then stained for Fn surface expression as described in figure 2.1A and analyzed by FACS (negative control: cells stained with non-immune rabbit IgG). Strong, dose-dependant inhibition of polyFn assembly was observed with both U73122 and calphostin C. (B) Expression of c- and nPKC isoforms in MTF7L cells: MTF7L breast cancer cells were subjected to EoE suspension culture for 1h in DMEM+20%FBS. One half of the cells were extracted with Trizol and RNA purified as described by the manufacturer (Invitrogen), the other half was extracted with lysis buffer. One microgram of purified RNA was subjected to RT-PCR (35 cycles) using PKC isoform-specific primers (upper panel). GAPDH served as reference message. Controls were samples run in the absence of reverse transcriptase (RTase). Cell lysates were subjected to SDS-PAGE and Western blotting with isoform-specific anti-PKC antibodies (lower panel). Strong message levels were recorded for PKCα, δ, and ε, moderate for PKCη, and weak for PKCθ. Protein expression was strong for PKCα, δ, and ε, weak for PKCη, and non-detectable for PKCθ. 70 71 (37), suspended MTF7L cells strongly express PKCα, PKCδ, and PKCε, both at the mRNA and protein levels (Fig 2.5B). PKCη mRNA is expressed moderately and protein weakly, and PKCθ mRNA weakly and protein non-detectably (Fig 2.5B). Next, we used select PKC inhibitors to narrow the spectrum of PKC isoforms involved in polyFn assembly. Failure of high doses of the cPKC inhibitors Gö6976 and HBDDE (38) to prevent polyFn assembly ruled out participation of PKCα in polyFn assembly (Fig 2.6A). In contrast, inhibitors that in addition to cPKCs also inhibited nPKCs, including BIM I, Gö6983, BIM XI (Ro-32-0432), and Rottlerin (39-43), decreased polyFn assembly in dose-dependant manners (Fig 2.6B). Albeit these inhibitor data do not provide conclusive evidence of the nPKC isoform involved in polyFn assembly, PKCδ (preferentially, but not specifically, inhibited by Rottlerin, ref. 43) and PKCε (preferentially inhibited by BIM XI) are our most likely candidates for involvement in polyFn assembly based on high protein expression levels and inhibitor activity. To examine the roles of PKCδ, PKCε and PKCη (nPKC isoforms previously associated with breast cancer metastasis, see refs. 44-46) in polyFn assembly, we transfected MTF7L cells with the regulatory domains (RD) of these isoforms or knocked down their expression by PKC isoform-specific siRNA oligonucleotides. The RD domains of PKCδ (co-transfected with GFP at a ratio of 1:50), PKCη-GFP and PKCε-GFP were transiently expressed in MTF7L breast cancer cells. Forty-eight hours after transfection, cancer cells were subjected to EoE suspension culture in serum-containing medium for 1 hour, then stained with anti-Fn antibodies, and Fn visualized by PE-conjugated secondary antibodies. Cells were analyzed by bivariate, dual-color (red-green) FACS. Among the three nPKC isoforms, only RD-PKCε inhibited polyFn assembly as depicted by a significant down-shift in the number of 72 Figure 2.6: Effect of various PKC inhibitors on polyFn assembly: (A) Inhibitors of cPKCs such as Gö6976 and HBDDE have no effect on polyFn assembly. (B) However, other PKC inhibitors such as BIM I, BIM XI, Gö6983, and Rottlerin inhibit polyFn assembly in a dose-dependant manner. Experiments were conducted as described in figure 2.5A. Significant differences in polyFn assembly by BIM XI at 100nM and 500nM and Rottlerin at 3µM and, more notably, at 100µM provide strong clues of a possible involvement of PKCε or PKCδ, given the fact that these isoforms are also strongly expressed in suspended MTF7L cells. S, drug solvent. 73 74 GFP-positive cells that strongly stained with anti-Fn antibody, i.e., only 21% of tumor cells that expressed high levels of GFP also expressed high levels of polyFn, relative to 56% in vector transfected cells (Fig 2.7A). The RD sequences of PKCη had no effect on polyFn assembly, while the RD sequence of PKCδ caused an increase in the number of strongly GFP-positive cells that expressed high levels of polyFn (Fig 2.7A), suggesting that PKCδ might exert a negative regulatory role in polyFn assembly. A similar effect on polyFn assembly to that of RD-PKCδ and RD-PKCε was observed when MTF7L cells were co-transfected transiently with GFP and siRNAPKCδ or siRNA-PKCε and subjected to bivariate FACS analyses 48h or 72h after transfection. Again, knockdown of PKCε suppressed polyFn assembly, while knockdown of PKCδ resulted in increased polyFn assembly on MTF7L cell surfaces (Fig 2.7B). These results were confirmed by the analysis of stable MTF7L clones with “siRNA-knockdown” of PKCε. PKCε-knockdown resulted in a significant decrease in polyFn assembly relative to a control clone transfected with an unspecific siRNA nucleotide sequence (Fig 2.8A). As expected, the FACS-measured decrease in polyFn paralleled decreased PKCε protein expression (Fig 2.8B) as well as decreased amounts of polyFn residing on top of the stack of polyacrylamide gels (Fig 2.8C, top), generated from anti-Fn[pan]-immunoprecipitates of metabolically labeled siRNAPKCε and control clones. The biochemically identified decrease in polyFn was reflected in decreased surface-association of polyFn on MTF7L cells stained with antiFn[pan] (Fig 2.8C, bottom). Functionally, impaired polyFn assembly in siRNA-PKCε clones is associated with a dramatic decrease in lung colonization (Fig 2.4). 75 Figure 2.7: (A) Effect of RD-PKCδ, RD-PCη, and RD-PKCε on polyFn assembly: MTF7L breast cancer cells grown to a density of 70-80% were transiently transfected with the regulatory domains (RD) of PKCδ, PKCη, and PKCε as described in Material and Methods. Forty-eight hours after transfection, cells were subjected to EoE suspension culture in DMEM+20%FBS for 1h at 37°C, then stained with anti-Fn[pan] antibodies as described in Figure 2.1A and immediately analyzed by bivariate, dualcolor [red fluorescence (FnF) vs. green fluorescence (GF)] FACS. Gates were selected based on the staining of wtMTF7L cells with non-immune rabbit IgG and Fn staining of vector-transfected MTF7L cells. Cells in the upper right quadrant of the scatter gram (=strong expressers of RD-PKC and Fn) were expressed as percent of the sum of cells in the right upper and lower quadrants (=strong RD-PKC expressers: gray area), thereby providing a quantitative measure of the inhibitory activity of each RD-PKC isoform on polyFn assembly. Data show inhibitory activity of RD-PKCε, stimulatory activity of RD-PKCδ, and no effect of RD-PKCη. (B) Effect of PKCδ and PKCε siRNA-knockdown on polyFn assembly: MTF7L breast cancer cells grown to a density of 70-80% were transiently co-transfected with either pRNAU6-HygrosiRNA-PKCε (sequence #3), pKD™-PKCδ-v6, or pRNAU6-Hygro-siRNA-us (unspecific nt. sequence #5) and GFP at a cDNA ratio of 50:1 as described in Material and Methods. Forty-eight hours after transfection, tumor cells were analyzed by FACS. Using a similar gate setting as in (A) [MTF7L cells stained with non-immune rabbit IgG and MTF7L cells transfected with siRNA-us and stained with anti-Fn[pan] antibodies], only 33% of the siPKCε-transfectants exceeded the ~50% threshold in polyFn assembly of siRNA-US transfectants, while in siPKCδ-transfectants this value was 73%. Fn-F, Fn fluorescence; GFP-F, GFP fluorescence. 76 77 Figure 2.8: Analysis of stable clones selected from siRNA-PKCε transfected MTF7L cells: siRNA-PKCε clones (cl1 and -cl2) were hygomycin-selected from MTF7L cells transfected with the nt. sequence #3, siRNA-PKCε-cl3 from MTF7L cells transfected with the nt. sequence #4, andcontrol clone mclus from MTF7L cells transfected with the unspecific nt. sequence #5 [see Material and Methods]: (A) Clones PKCε-cl1 and -cl2 and control clone mclus subjected to EoE suspension culture in DMEM+20% FBS for 1h were stained with anti-Fn[pan] and analyzed by FACS as described in Material and Methods, then analyzed by FACS. PolyFn assembly by MTF7L clones PKCε-cl1 and PKCε-cl2 was significantly decreased relative to the mclus control clone. (B) Western blot analysis for PKCε− and vinculin-expression in lysates from siPKCε-cl1, -cl2 and -cl3 and the mclus clone: PKCε protein is decreased in siPKCε clones relative to the unspecific siRNA-mclus control clone. (C) Lysates from MTF7L clones mclus and siPKCε-cl1 labeled with [35S]-methionine and processed as described in Fig 2.3B were subjected to anti-Fn[pan]immunoprecipitation followed by SDS-PAGE and autoradiography. There is a significant decreased in polyFn residing on top of the stacking gel in siPKCε-cl1 relative to the mclus control, mimicking the degree of tumor cell surface-associated polyFn of these clones stained with anti-Fn[pan] (C, bottom). 78 79 Plasma Membrane Translocation and Activation of PKCε Promote and PMA Suppresses polyFn Assembly The process of polyFn assembly was associated with a time-dependent translocation of PKCε to the plasma membrane. In adherent MTF7L cells, PKCε was present predominantly in the cytosolic fraction (Fig 2.9A: Time 0), but translocated in increasing amounts to the plasma membrane when MTF7L cells were subjected to EoE suspension culture in the presence of pFn-containing serum (Fig 2.9A: Time 1560 min). Peak values for membrane-associated PKCε were observed at 30 min of incubation in EoE suspension culture and remained high for the remainder of the 1h incubation period. Translocation of PKCε to the plasma membrane was associated with a noticeable gain of its molecular weight, presumably caused by the observed serine- and threonine phosphorylation of PKCε. PKCε phosphorylation, again, was most prominent at 30 min of EoE suspension culture, then rapidly declined to barely detectable levels at 60 min (Fig 2.9B). This decline was accompanied with a backward shift to the lower molecular weight form of PKCε (Fig 2.9A). The PKC inhibitor BIM XI totally blocked PKCε phosphorylation (Fig 2.9B). This result was confirmed when 32P-labeled MTF7L were examined for PKCε-phosphorylation (Fig 2.9C). Control experiments conducted with PKCδ yielded no phoshorylation of this PKC isoform (Fig 2.9D). Interestingly, PMA (1µM) caused a time-dependent loss of PKCε protein from both the cytosolic and membrane fractions of MTF7L cells (Fig 2.10A). The PMA-mediated loss of PKCε protein, and to a slightly lesser extent of PKCδ protein, was most striking after an 18h incubation period of tumor cells with PMA. No effect was observed for the PMA-insensitive aPKC isoform PKCζ (Fig 2.10B). Concomitantly, loss of PKCε protein and decreased plasma membrane translocation in PMA-treated cells resulted in a dose-dependent, diminished polyFn assembly (Fig 2.10C). 80 Figure 2.9: Membrane translocation and phosphorylation of PKCε: (A) PKCε translocates in increasing amounts from the cytosol to the plasma membrane during 0, 15, 30, and 60 min of EoE suspension culture of MTF7L in DMEM+20% FBS. Notice a slightly higher molecular weight of PKCε harvested from the membrane fraction (M) relative to that from the cytosolic fractions (C). (B) Time course of PKCε phosphorylation: MTF7L were incubated in EoE suspension culture for the indicated periods of time in DMEM+20% FBS in the presence or absence of BIM XI (500nM). At the end of the indicated time periods, cells were lysed, and lysates subjected to antiPKCε immunoprecipitation. Precipitates were probed by Western blotting with antiPKCε, anti-pSer, and anti-pThr antibodies. Maximal phosphorylation of PKCε was observed at 30 min of EoE suspension culture. Thereafter, the PKCε phosphorylation decreased to barely detectable levels at 60 min. PKCε of MTF7 cells treated with BIM XI stained weakly or non-detectably with anti-pSer and anti-pThr. (C & D) [32P]incorporation into PKCε (C) and PKCδ (D): MTF7 cells labeled for 4h in phosphatefree DMEM containing 100µCi/ml of [32P]orthophosphate as described in Material and Methods were subjected for 30min to EoE suspension cultures in DMEM+20µg/ml pFn (lane 2), DMEM+20%FBS (lane 3), and DMEM+20%FBS+500nM BIM XI (lane 4) [control: [32P]orthophosphate-labeled, adherent cells incubated for 30min in DMEM+20%FBS (lane 1)]. Anti-PKCε (C) and anti-PKCδ (D) immunoprecipitates were analyzed by autoradiography. RG, radiograph; WB, Western blot; IP, immunoprecipitation. 81 82 Figure 2.10: Effect of PMA on PKCε and polyFn assembly: (A) PMA treatment (1µM) of MTF7L cells caused a time-dependent loss of PKCε from both cytosolic and membrane fractions relative to untreated cells, in which PKCε decreased in the cytosolic fraction, but increased in the membrane fraction with time of incubation. (B) MTF7L cells pretreated with 1µM PMA for 18h, then subjected to EoE suspension culture in DMEM+20%FBS+1µM PMA exhibited decreased amounts of PKCδ protein and, particularly, PKCε protein relative to cells treated without PMA, while protein expression of the atypical PKC isoform PKCζ was unaffected by PMA (control). (C) PMA caused a dose-dependent decrease in polyFn assembly on MTF7L cells subjected to EoE suspension culture in DMEM+20%FBS for 1h. (S=PMA solvent). 83 84 DISCUSSION The role of Fn in cancer progression is controversial. Early studies suggest that Fn may act as a tumor suppressor gene, promoting differentiation and suppressing proliferation, migration, invasion, and metastasis (reviewed in refs. 13-16). In accordance, expression of pFn in Fn-negative LMM3 murine mammary cancer cells reduced the rate of migration as well as spontaneous and experimental metastasis (47), and forced expression of the α5β1 integrin, which was thought to capture secreted cFn at the cell surface, convert it into fibrils and then deposit it in the extracellular matrix, reduced motility and tumorigenicity of transfected tumor cells (48). These findings are contrasted by cDNA or oligonucleotide microarray analyses that attempted to identify genes whose expression was associated with malignant behavior of breast cancers. In several of these studies, Fn (Fn1: EDA+EDB+IIICS120) was found to be consistently overexpressed in invasive/metastatic breast tumors in toto as well as in tumor cells isolated by laser capture microdissection from a primary invasive ductal carcinoma and an axillary node harboring metastatic breast cancer (5-7). Interestingly, ErbB2 overexpression, observed in 25-30% of invasive ductal breast carcinomas and correlated with a poor clinical prognosis, was associated with Fn1 overexpression in 11 out of 36 cases (49). Overexpression of Fn was also noticed in cancer cell lines that exhibited an invasive/metastatic phenotype, including MDA-MB-231, MeWo-70W, PC3-ML, 4T1, LLC1, B16-F10, K7M2, MTF7, and RPC-2 (17), and was most dramatic in cancer cell lines that were subjected to a selection process for enhanced lung metastatic performance (9,11,17) or in highly metastatic clones derived from a rabdomyosarcoma (8) or B16 melanoma (10), but not in non-metastatic clones (8). Moreover, expression of the metastasis suppressor gene nm23 in MDA-MB-435 breast cancer cells resulted in a down regulation of genes associated with adhesion and 85 motility, including Fn1 (50). On the other hand, neither the poor prognosis signature obtained by large-scale transcript profiling of human breast cancers (51,52) nor the gene expression profile that was associated with lung-metastasis of selected MDAMB-231 clones (i.e., lung-metastatic signature) contained Fn (53). A possible explanation for this result may be related to the multi-functional role of Fn in tumor progression, requiring high levels of Fn expression throughout tumor progression. For example, in early stages of progression Fn promotes survival, cell cycle progression, and angiogenesis and in later stages motility/invasiveness and dissemination and implantation in secondary organs (reviewed in refs. 12-16). Support for this notion can be found in the observation that tumor cell lines that do not express Fn such as MCF-7 cells exhibit poor tumorigenicity, growth, and angiogenesis (54), while overexpresssion of c-Jun in MCF7 cells stimulated Fn synthesis and, concomitantly, increased tumorigenicity and invasiveness (55,56). In recent years our laboratory has been interested in the role of Fn in breast cancer lung metastasis. This interest was triggered by the fact that a survey of all our highly lung-metastatic breast cancer cell lines, including MTF7, MTF7L, MDA-MD231 and 4T1 cells, revealed high levels of Fn expression (17) and by the observation that tumor cells isolated from the blood of tumor-bearing animals exhibited a characteristic cell surface coat of multi-globular polyFn. This phenotype could be reproduced in vitro, when cancer cells were subjected to EoE suspension culture in the presence of serum (9,17). As indicated in the introduction, this unique surface expression of Fn was responsible for the docking of MTF7 to the lung endothelial addressin DPP IV (9,17-21) and for facilitating lung colonization (9,17,20,21). Interference with the polyFn/DPP IV adhesion by monoclonal antibodies (9), peptides directed against the DPP IV-binding domains in the 13th, 14th and 15th type III repeats of Fn (17), and a polypeptide representing the extracellular domain of DPP IV (20) 86 abrogated metastatic colonization of the lungs. Here, we have presented some of the molecular underpinnings that regulate polyFn assembly in a rat model of hematogenous dissemination of breast cancer cells. We show that polyFn assembly critically depends upon the expression of endogenous cFn (Fn1) and is regulated by PKCε. This is evidenced by a significantly decreased polyFn assembly in cells transfected with either siRNA-PKCε or dominant-negative RD-PKCε, and in cells treated with select PKC inhibitors. During a 1-hour polyFn assembly phase, PKCε translocates in increasing amounts from the cytosol to the plasma membrane. This translocation is accompanied by PKCε Ser/Thr-phosphorylation and is crucial for successful polyFn assembly, since overexpression of dominant-negative RD-PKCε, which blocks DAG-mediated plasma membrane-docking and activation of endogenous PKCε (57,58), prevents polyFn assembly. In contrast, both RD-PKCδ and siRNA knockdown of PKCδ promoted polyFn assembly, mimicking the previously recognized opposing effects of PKCε and PKCδ in ischemic injury of the heart (59) and suggesting a dominant-negative regulatory effect of PKCδ over PKCε in MTF7L cells. A PKCε regulatory effect was also observed for PMA, which in a dose- and time-dependent manner inhibited PKCε protein synthesis and, in accordance, impaired polyFn assembly on MTF7L cells in a manner similar to that reported for other celltypes (60,61). Together, PMA/DAG competition for plasma membrane-docking, overall loss of PKCε protein, and competition between related PKC isoforms appear to be primarily responsible for the observed decrease in polyFn assembly on cancer cell surfaces. The role of PKCε in polyFn assembly appears to be dualistic in that it may facilitate surface expression of endogenous, cellular Fn by promoting Fn exocytosis (62-66) and protein synthesis (67). In accordance, inhibitors of protein synthesis (e.g., cycloheximide) and exocytosis (e.g., monensin and Brefeldin B) strongly impeded 87 polyFn assembly (Fig 2), albeit cells were kept in suspension culture in the presence of high concentrations of pFn. However, at this writing we have not identified any of the PKCε partner proteins that are required in regulating these processes. Nonetheless, inhibition of PKCε has been shown to trap integrin β1, a possible receptor for cellular Fn, in a CD81-positive intracellular compartment and electron microscopy demonstrated the co-localization of PKCε and integrin β1 on vesicular membranes (68). Translocation of the integrin β1 from cytosolic vesicles to the plasma membrane appears to involve the PKCε-mediated phosphorylation of the cytoplasmic tail of β1integrin at threonine 788/789 (69) or cytoskeletal intermediaries such as vimentin (70) and actin (62). Accordingly, interaction of PKCε with actin has been linked to the formation of invadopodia-like structures, increased pericellular metalloproteinase activity and, ultimately, invasiveness and metastasis in PKCε-transformed NIH3T3 fibroblasts (71). Whether the concomitant downregulation of RhoC in MDA-MB-231 cells, in which PKCε expression was knocked down by RNA interference resulting in a cell phenotype that was significantly less proliferative, invasive, and metastatic (45), is intertwined in PKCε-actin remodeling is unclear and needs further investigation. Supporting evidence for such role may be deduced from the thrombin-induced PKCεRho-actin complex formation in actin reorganization in myofibroblasts (72). However, in our system of suspended (blood-borne) breast cancer cells, the Rho activity is significantly suppressed and expression of constitutively active Rho suppresses polyFn assembly (Cheng and Pauli, manuscript in preparation). In conclusion, our studies suggest that surface expression of polyFn occurs in PKCε-dependant manner presumably involving the transport of Fn-receptor complexes from cytosolic vesicles to the plasma membrane and that such complexes serve as scaffolds for the subsequent augmentation of polyFn aggregates on tumor cells during their journey in Fn-rich blood plasma. This notion is supported by a 88 dramatic inhibition of polyFn assembly by inhibitors of exocytosis and by siRNAknockdown of PKCε, that occurs even when cancer cells are exposed to high exogenous Fn concentrations and express several integrin and syndecan Fn-receptors on their surfaces (Cheng, unpublished data). Thus, we have disclosed here a novel regulatory principle for the recognized “pro-metastatic” role of Fn that may spur the design of new anti-metastatic therapies. 89 REFERENCES 1. Zhang L, Zhou W, Velculescu VE, Kern SE, Hruban RH, Hamilton SR, Vogelstein B, and Kinzler KW. (1997) Science 276, 1268-1272 2. Bittner M, Meltzer P, Chen Y, Jiang Y, Seftor E, Hendrix M, Radmacher M, Simon R, Yakhini Z, Ben-Dor A, Sampas N, Dougherty E, Wang E, Marincola F, Gooden C, Lueders J, Glatfelter A, Pollock P, Carpten J, Gillanders E, Leja D, Dietrich K, Beaudry C, Berens M, Alberts D, and Sondak V. (2000) Nature 406, 536-540 3. Al Moustafa AE, Alaoui-Jamali MA, Batist G, Hernandez-Perez M, Serruya C, Alpert L, Black MJ, Sladek R, and Foulkes WD. (2002) Oncogene 21, 26342640 4. Jiang Y, Harlocker SL, Molesh DA, Dillon DC, Stolk JA, Houghton RL, Repasky EA, Badaro R, and Xu J. (2002) Ongoene 21, 2270-2282 5. Amatschek S, Koenig U, Auer H, Steinlein P, Pacher M, Gruenfelder A, Dekan G, Vogl S, Kubista E, Heider KH, Stratowa C, Schreiber M, and Sommergruber W. (2004) Cancer Res. 64, 844-856 6. Hao X, Sun B, Hu L, Lähdesmäki H, Dumire V, Feng Y, Zhang SW, Wang H, Wu C, Wang H, Fuller GN, Symmans WF, Shumulevich I, and Zhang W. (2004) Cancer 100, 1110-1122 7. Zucchi I, Mento E, Kuznetsov VA, Scotti M, Valsecchi V, Simionati B, Vicinanza E, Valle G, Pilotti S, Reinbold R, Vezzoni P, Albertini A, and Dulbecco R. (2004) Proc. Natl. Acad. Sci. U S A 101, 18147-18152 8. Korach S, Poupon MF, Du Villard JA, and Becker M. (1986) Cancer Res. 46, 3624-3629 9. Cheng HC, Abdel-Ghany M, Elble RC, and Pauli BU. (1998) J. Biol. Chem. 273, 24207-24215 10. Clark EA, Golub TR, Lander ES, and Hynes RO. (2000) Nature 406, 532-535 11. Bandyopadhyay A, Elkahloun A, Baysa SJ, Wang L, and Sun LZ. (2005) Cancer Biol. Ther. 4, 168-172, 2005 12. Hynes RO, and Yamada KM. (1982) J. Cell Biol. 95, 369-377 90 13. Humphries MJ, Obara M, Olden K, and Yamada KM. (1989) Cancer Invest. 7, 373-393 14. Rouslahti E. (1999) Adv. Cancer Res. 67, 1-20 15. Danen EHJ, and Yamada KM. (2001) J. Cell Physiol. 189, 1-13 16. Labat-Robert J. (2002) Cancer Biol. 12, 187-195 17. Cheng HC, Abdel-Ghany M, and Pauli BU. (2003) J. Biol. Chem. 278, 2460024607 18. Johnson RC, Augustin-Voss HG, Zhu D, and Pauli BU. (1991) Cancer Res. 51, 394-399 19. Johnson RC, Zhu D, Augustin-Voss HG, and Pauli BU. (1993) J. Cell Biol. 121, 1423-1432 20. Abdel-Ghany M, Cheng HC, Levine RA, and Pauli BU. (1998) Invasion & Metastasis 18: 35-43 21. Cheng HC, Abdel-Ghany M, Zhang S, and Pauli BU. (1999) Clin. Exp. Metastasis 17, 609-615 22. Hayman EG and Ruoslahti E. (1979) J. Cell Biol. 83, 255-259 23. Zeidman R, Löfgren B, Pahlman S, and Larsson C. (1999) J. Cell Biol. 145, 713726 24. De Servi B, Hermani A, Medunjanin S, and Mayer D. (2005) Oncogene 24, 4946-4955 25. Sakai T, Larsen M, and Yamada KM. (2003) Nature 423, 876-881 26. Cheng JJ, Wung BS,Chao YJ, and Wang DL. (2001) J. Biol. Chem. 276, 3136831375 27. Abdel-Ghany M, Cheng HC, Elble RC, and Pauli BU. (2001) J. Biol. Chem. 276, 25438-25446 28. Rosenberg R, Gertler R, Friedrichs J, Fuehrer K, Dahm M, Phelps R, Thorban S, Nekarda H, and Siewert JR. (2002) Cytometry 49, 150-158 29. Barr RK, and Bogoyevitch MA. (2001) Int. J. Biochem. Cell Biol. 33, 10471063 91 30. Dinter A, and Berger EG. (1998) Histochem. Cell Biol. 109, 571-590 31. Nebenfuhr A, Ritzenthaler C, and Robinson DG. (2002) Plant Physiol. 130, 1102-1108 32. Narumiya S, Ishizaki T, and Watanabe N. (1997) FEBS Lett. 410, 68-72 33. Hall A. (2005) Biochem. Soc. Trans. 33(Pt 5), 891-895 34. Gloyna W, Schmitz F, and Seebeck J. (2005) Regul. Pept. 125, 179-184 35. Valledor AF, Xaus J, Comalada M, Soler C, and Celada A. (2000) J. Immunol. 164, 29-37 36. Munakata M, Stamm C, Friehs I, Zurakowski D, Cowan DB, Cao-Danh H, McGowan FX Jr, and del Nido PJ. (2000) Ann. Thorac. Surg. 73, 1236-1245 37. Kiley SC, Clark KJ, Duddy SK, Welch DR, and Jaken S. (1999) Oncogene 18, 6748-6757 38. Schreckenberg R, Taimor G, Piper HM, and Schluter KD. (2004) Cardiovasc. Res. 63, 553-560 39. Wilkinson SE, Parker PJ, and Nixon JS. (1993) Biochem. J. 294, 335-337 40. Gschwendt M, Müller HJ, Kielbassa K, Zang R, Kittstein W, Rincke G, and Marks F. (1994) BBRC 199, 93-98. 41. Di-Capua N, Sperling O, and Zoref-Shani E. (2003) J. Neurochem. 84, 409-12 42. Imamdi R, de Graauw M, and van de Water B. (2004) J. Pharmacol. Exp. Ther. 311, 892-903 43. Tapia JA, Jensen RT, and Garcia-Marin LJ. (2006) Biochim. Biophys. Acta 1763, 25-38 44. Kiley SC, Clark KJ, Goodenough M. Welch DR, and Jaken S. (1999) Cancer Res. 59, 3230-3228 45. Pan Q, Bao LW, Kleer CG, Sabel MS, Griffith KA, Teknos TN, and Merajver SD. (2005) Cancer Res. 65, 8366-8371 46. Masso-Welch PA, Winston JS, Edge S, Darcy KM, Ascxh H, Vaughan MM, and Ip MM. (2001) Breast Cancer Res. Treat. 68, 211-223 92 47. Urtreger A, Porro F, Puricelli L, Werbajh S, Baralle FE, De Joffe EB, Kornblihtt AR, and Muro AF. (1998) Int. J. Cancer 78, 233-241 48. Giancotti FG, and Ruoslahti E. (1999) Science 285, 1028-1032 49. Mackay A, Jones C, Dexter T, La Silva R, Bulmer K, Jones A, Simpson P, Harris RA, Jat PS, Neville AM, Reis LFL, Lakhani S, and O’Hare MJ. (2003) Oncogene 22, 2680-2688 50. Zhao H, Jhanwar-Uniyalk M, Datta PK, Yemul S, Ho L, Khitrov G, Kupershmidt I, Pasinetti GM, Ray T, Athwal RS, and Achary MP. (2004) Int. J. Cancer 109, 65-70 51. van 't Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, Peterse HL, van der Kooy K, Marton MJ, Witteveen AT, Schreiber GJ, Kerkhoven RM, Roberts C, Linsley PS, Bernards R, and Friend SH. (2002) Nature 425, 530-536 52. Ramaswamy S, Ross KN, Lander ES, and Golub TR. (2003) Nat. Genet. 33, 4954 53. Minn AJ, Gupta GP, Siegel PM, Bos PD, Shu W, Giri DD, Viale A, Olshen AB, Gerald WL, and Massague J. (2005) Nature 436, 518-524 54. Levenson AS, Thurn KE, Simons LA, Valiceasa D, Jarrett J, Ospio C, Jordan VC, Volpert OV, Satcher, Jr. RL and Gartenhaus RB. (2005) Cancer Res. 65, 10651-10656 55. Smith LM, Wise SC, Hendricks DT, Sabichi AL, Bos T, Reddy P, Brown PH, and Birrer MJ. (1999) Oncogene 18, 6063-7070 56. Reinehart-Kim J, Johnston M, Birrer M, and Bos T. (2000) Int. J. Cancer 88, 180-190 57. Lopez-Andreo MJ, Gomez-Fernandez JC, and Coirbalan-Garcia S. (2003) Mol. Biol. Cell 14, 4885-4895 58. Stahelin RV, Digman MA, Medkova M, Ananthanarayanan B, Melowic HR, Rafter JD, and Cho W. (2005) J. Biol. Chem. 280, 19784-19793 59. Chen L, Hahn H, Wu G, Chen CH, Liron T, Schechtman D, Cavallero G, Bancil L, Olli R, Dorn II GW, and Mochly-Rosen D. (2001) Proc. Natl. Acad. Sci. USA 98, 111114-111119 60. Ways DK, Messler BR, Garris TO, Qin W, Cook PP, and Parker PJ. (1992). Cancer Res. 52, 5604-5609 93 61. Fujihara M, Connolly N, Ito N and Suzuki T. (1994) J. Immunol. 15, 1898-906 62. Chun J, Auer KA, Jacobson BS. (1997) J. Cell Physiol. 173, 361-370 63. Hoy M, Berggren PO, and Gromada J. (2003) J. Biol. Chem. 278, 35168-35171 64. Mendez CF, Leibiger IB, Leibiger B, Hoy M, Gromada J, Berggren PO, and Bertorello AM. (2003) J. Biol. Chem. 278, 44753-4477 65. Jerdeva GV, Yarber FA, Trousdale MD, Rhodes CJ, Okamoto CT, Dartt DA, and Hamm-Alvarez SF. (2005) Am. J. Physiol. Cell Physiol. 289, C1052-C1068 66. Park YS, Hur EM, Choi BH, Kwak E, Jun DJ, Park SJ, and Kim KT. (2006) J. Neurosci. 26, 8999-9005 67. Lee HB, Yu MR, Song JS, and Ha H. (2004) Kidney Int. 65, 1170-1179 68. Ivaska J, Whelan RD, Watson R, and Parker PJ. (2002) EMBO J. 21, 3608-3619 69. Stawowy P, Margeta C, Blaschke F, Lindschau C, Spencer-Hansch C, Leitges M, Biagini G, Fleck E, and Graf K. (2005) Cardiovasc. Res. 67, 50-59 70. Ivaska J, Vuoriluoto K, Huovinen T, Izawa I, Inagaki M, and Parker PJ. (2005) EMBO J. 24, 3834-3845 71. Tachado SD, Mayhew MW, Wescott GG, Foreman TL, Goodwin CD, McJilton MA, and Terrian DM. (2002) J. Cell Biochem. 85, 785-7697 72. Bogatkevich GS, Tourkina E, Abrams CS, Harley RA, Silver RM, and Ludwicka-Bradley A. (2003) Am. J. Physiol. Lung Cell Mol. Physiol. 285, L334-L343 94 CHAPTER III POLYMERIC FIBRONECTIN ASSEMBLY ON THE SURFACE OF BREAST CANCER CELLS IS DEPENDENT ON SUBCELLULAR LOCALIZATION OF PKC ΕPSILON-ACTIN COMPLEX AND PYK2 ACTIVITY* * Huang L, Pauli BU. Polymeric fibronectin assembly on the surface of breast cancer cells is dependent on subcellular localization of PKCε-actin complex and PYK2 activity. To be submitted to Experimental Cell Research. 95 ABSTRACT Lung colonization by breast cancer cells is mediated, at least in part, by interaction of lung endothelial dipeptidyl-peptidase IV (DPPIV; CD26) with breast cancer cell surface polymeric fibronectin (polyFn). PolyFn on the surface of MTF7L rat breast cancer cells has been shown previously to be regulated by protein kinase C epsilon (PKCε). Here, we introduce β-actin and proline-rich tyrosine kinase-2 (Pyk-2) as partners of PKCε in orchestrating the successful polyFn assembly on MTF7L cells. Mimicking conditions of cancer cells disseminating in the blood, MTF7L cells subjected to end-over-end suspension culture exhibit an increasing accumulation of PKCε and F-actin in the subplasmalemmal space beneath emerging polyFn surface aggregates. Translocation of these proteins and polyFn assembly is abrogated by Cytochalasin D, an actin barbed-end capping drug, and to a lesser extent latrunculin B, an actin monomer sequestering drug. PKCε-actin interaction is dependent on PKCε catalytic activity, as shown by treatment of suspended cancer cells with the PKCε inhibitor bisindolylmaleimide XI (BIMXI; Ro-32-0432), but does not involve the unique PKCε actin binding domain. Catalytic activity of Pyk2 is required for successful assembly of polyFn, as shown by the absence of polyFn assembly upon transfection of cancer cells with kinase dead Pyk2. The present study provides new insights into the regulation of the PKCε-mediated polyFn assembly and pulmonary metastasis of breast cancer cells that may harbor new therapeutic potentials. 96 INTRODUCTION Cancer cells isolated from blood of rats or mice from malignant breast tumors implanted into the mammary fat pad are adorned with a multi-globular coat of polymeric fibronectin (polyFn) (1). These polyFn aggregates increase in size with time of incubation in suspension and assemble into increasingly deoxycholate-insoluble polymers that can be seen in the top of the stacks of SDS-polyacrylamide gels (1) and on the surface of cancer cells with immunofluorescent staining. Conversion of Fn from the globular (soluble) to the linearized state (insoluble, surface-associated) exposes a cryptic domain in Fn responsible for binding to DPPIV on lung endothelial cells. The DPPIV domain of Fn has been identified as a consensus motif in each of the 13th, 14th and 15th type III repeats of Fn. DPPIV binding is specific for the linearized form of Fn and not for the globular form, facilitating cancer cell adhesion in the presence of high Fn concentrations in the plasma under physiological conditions (1). The cluster arrangement of polyFn aggregates presents multiple binding sites for interaction with DPPIV that strengthens cancer cell and endothelial cell adhesion and enables cancer cell vascular arrest under hemodynamic shear stress (1). These findings are consistent with our studies with rodent models showing reduced lung colonization in Fischer 344/CRJ rats with mutated DPPIV and diminished pulmonary endothelia expression (2) and in DPPIV-/- mice (3). We have recently elucidated one of the PKCs responsible in mediating polyFn assembly on breast cancer cells. In the rat model of hematogenous dissemination of breast cancer cells, polyFn assembly and lung metastasis critically depends on expression of endogenous cellular Fn (cFn) and is regulated by PKCε (4). Pharmacological inhibition, expression of dominant negative forms of PKCε and knockdown of PKCε lead to decreases in the levels of polyFn on cancer cell surface 97 and lung colonization (4). Several possible pathways may account for the ability of PKCε to mediate polyFn formation. PKCε may regulate polyFn assembly on the cell surface by phosphorylation of cytoskeletal regulators, through signaling to other signaling pathways and/or control extracellular matrix receptors (5-11). To promote substrate recognition, PKC may directly interact with its substrates (substrates that interact with C-kinase, STICK) or bind to proteins that facilitate substrate phosphorylation (receptors for activated C-kinase, RACK). STICKs localize to a range of specific cellular compartments which includes plasma membrane, cortical cytoskeleton, focal contacts, nucleus and vesicles (5) and allow PKC to perform specialized function. Several PKC substrates, such as myristoylated alanine-rich protein kinase C substrate (MARCKS), have a role of anchoring the actin cytoskeleton to the plasma membrane (6). On the other hand, PKCε may also directly phosphorylate kinases that are able to activate polyFn receptors. PKCε has been shown to play a role in the activation of Pyk2 (7), which is able to bind the phosphorylated integrin beta3 cytoplasmic tail (8) and promote integrin ligand binding ability. Lastly, PKCε may control integrin function. PKCε is involved in inside-out signaling for beta integrin mediated cell adhesion (9-11) and RACK1 has been shown to be one of the mediators of binding between integrin beta chains and PKCε. Whether direct or indirect, integrin activation may indeed be an important pathway in PKCεmediate polyFn assembly remains to be seen. In order to further elucidate the role of PKCε in polyFn assembly, we have utilized matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI TOF/ m/s) fingerprinting to identify proteins that reside in the PKCε complex under cancer cell-activating conditions. Two cytoskeletal proteins, β-actin, cytoplasmic-2 and myosin IIa (Myh9) have been identified in PKCε immunoprecipitates. Focusing on the contribution of β-actin-2, we report that polyFn 98 assembly is dependent on PKCε/actin interaction and cytoskeletal integrity. With time of incubation of cancer cell in suspension, PKCε and F-actin translocated in increasing amounts in the subplasmalemmal space at the base of globular polyFn aggregates. This translocation is associated with activation and phosphorylation of Pyk2. Disruption of the PKCε/actin association by actin polymerization blocking agents such as Cytochalasin D and kinase-dead Pyk2 prevent polyFn assembly. Together, our study presents a role of PKCε in the regulation of polyFn assembly through interaction with the actin cytoskeleton and activation of Pyk2. MATERIAL AND METHODS Antibodies and Reagents Rabbit and mouse anti-PKCε polyclonal antibody and mouse anti-βactin antibody for western blotting were from Santa Cruz Biotechnology (Santa Cruz, CA), rabbit anti-Fn polyclonal antibodies that recognized pFn and cFn from both bovine and rat (anti-Fn[pan]: does not crossreact with fibrinogen, vitronectin, laminin, collagen type IV) were from Sigma (St. Louis, MO), mouse anti-PKCε used for immunoprecipitation was from BD Biosciences (San Jose, CA), phycoerythrin (PE)conjugated donkey anti-rabbit, horseradish-peroxidase (HRP)-conjugated donkey anti-rabbit, and HRP-conjugated goat anti-mouse antibodies were from Jackson ImmunoResearch (West Grove, PA), rabbit anti-Pyk2 antibody werefrom Upstate (Lake Placid, NY), and phospho-specific anti-Pyk2(Tyr402) were from Cell Signaling (Danvers, MA). PKC inhibitor BIM XI (Ro-32-0432) was from EMD Chemicals (San Diego, CA). Actin inhibitor Cytochalasin D and Latrunculin B was from Sigma (St. Louis, MO). Fetal bovine serum (FBS) was purchased from Gemini Bio-Products (Woodland, CA). All other chemicals and reagents were from Sigma. 99 Cell Cultures MTF7L cells were derived from a lung-metastasis generated by tail-vein injection of MTF7 breast cancer cells (obtained from Dr. D.R. Welch, University of Alabama at Birmingham, Birmingham, AL) into Fischer 344 rats. At an i.v. inoculation dose of 2x105cells/0.3ml PBS per rat, MTF7L cells consistently produce in excess of 400 lung colonies. Cells were grown in culture in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% heat-inactivated fetal bovine serum (FBS). For end-over-end (EoE) suspension culture, MTF7L cells were grown to 80 to 90% confluence, then removed from the growth surface by trypsinization [0.25% trypsin, 0.02% EDTA in phosphate-buffered saline (PBS); 10 min, 37°C], washed twice in DMEM containing 10% FBS, and subjected to EoE suspension culture for 1 hour (or as indicated) in 2ml-centrifuge tubes in DMEM plus 20% FBS at a concentration of 5x106 cells/ml (1,13). Tumor cells were used for all experiments within 10 passages from frozen stocks that were tested for metastatic performance immediately prior to freezing. Plasmids Constructs, Transfection, and Selection The construct Wild type PKCε (PKCε−WT) cloned into pEGFP-N1 was obtained from Dr. C. Larsson (Lund University, Malmö, Sweden) (12). PKCε with actin binding domain 223-228 deleted (PKCε−ΔABDε) was constructed from PKCε−WT using Quickchange XL Site-directed mutagenesis kit from Stratagene (La Jolla, CA). Wild type FAK, kinase dead FAK (K454R), wild type Pyk2 and kinase dead Pyk2 (K457R) cloned into pKH3 vector were obtained from Dr. Jun-Lin Guan (Cornell University, Ithaca, NY). All plasmid constructs were verified by doublestranded sequencing. 100 MTF7L cells grown to 70% confluence were transiently transfected with above vector constructs or vector alone using ExGen 500 (Fermentas, Glen Burnie, MD) as described by the manufacturer. Transfection rates were assessed by expression of GFP that is either tagged to the cDNA of interest or co-transfected at a ratio of 1:50 with the cDNA of interest, which were 30-40 %. Cells were used in the various assays 24h after transfection unless otherwise stated. Flow Cytometry FACS was used to quantify Fn expression on MTF7L breast cancer cell surfaces (1, 13). Tumor cells that had been subjected to EoE suspension culture were washed twice in DMEM containing 1% bovine serum albumin (BSA), then incubated with rabbit anti-Fn[pan] antibody diluted 1:100 in PBS containing 1%BSA (PBSBSA) for 1h at 4°C. After washing in PBS-BSA, tumor cells were stained with phycoerythrin (PE)-conjugated donkey anti-rabbit antiserum in PBS-BSA for 1h at 4°C and fixed in 2% paraformaldehyde in PBS. FACS analysis was performed on a Coulter Epics Profile (Coulter Electronics, Hialeah, FL). Nonspecific fluorescence was accounted for by incubating tumor cells with non-immune rabbit serum instead of primary antibody. To quantify the effect of overexpressed proteins on polyFn assembly, we generate bivariate distributions of red fluorescence (y-axis: cells stained with anti-Fn antibodies and PE-conjugated secondary antibodies) and green fluorescence (x-axis: same cells expressing GFP-tagged protein or co-transfected with GFP and cDNA of interest). The levels of polyFn expression in the cell population that emitted high GFP fluorescence were taken as a reflection of the effect of the transfected cDNA on polyFn assembly. X-axis gates were selected based on the staining of MTF7L cells with non-immune rabbit IgG. Y-axis gates were selected based on 50% Fn staining intensity of the vector transfected GFP-positive cells. Cells 101 in the upper right quadrant of the scatter gram (strong expressers of PKCε and Fn) were expressed as percent of the sum of cells in the right upper and lower quadrants (strong PKCε expressers), providing a quantitative measure of the inhibitory activity of transfected plasmids or pharmacological inhibitors. To assess the effect of inhibitors of cell signaling, tumor cells were incubated with inhibitor 30-min prior to and throughout EoE suspension culture in DME/FBS, then subjected to polyFn quantification as described above. Controls were tumor cells incubated in equimolar inhibitor solvent concentration. Cell Lysis, Immunoprecipitation, and Western Blotting Cells were extracted with lysis buffer (50mM Tris-HCl, pH 7.4, 150mM NaCl, 1mM EDTA, 1mM EGTA, 1mM benzamidine chloride, 1mM PMSF, 2 µg/ml leupeptin, 0.27 TIU/ml aprotinin, 0.1mM sodium vanadate, and 1% Triton X-100) for 1h at 4°C (9). Total cell lysates were subjected to: (i) SDS-PAGE (~20-50µg protein) and Western blotting, using anti-Fn[pan], anti-PKCε, anti-PYK2, phospho-specific anti-PYK2 antibodies, anti-β-actin, HRP-conjugated donkey anti-rabbit or goat antimouse secondary antibodies and ECL for detection of bound antibody as described (1); (ii) immunoprecipitation with anti-PKCε antibodies (14). Immunoprecipitates obtained from cell lysates were separated by SDS-PAGE (8-10% polyacrylamide) and blotted to nitrocellulose membranes and probed with either anti-PKCε, anti-βactin or anti-pSer and anti-pThr antibodies (1). Mass Spectrometry PKCε monoclonal antibody was conjugated to protein-G (PG) agarose beads by incubating antibody with beads for 2 hours at room temperature. PG beads were washed with PBS, washed with borate buffer (0.2M Sodium Borate pH9.0), then 102 incubated with borate/DMP buffer (0.2M Sodium Borate pH9.0, 20mM dimethyl pimelimidate) for 1 hour at room temperature. After incubation, the beads were incubated with 0.2M ethanolamine twice then incubated in 0.2M ethanolamine for 2 hours room temperature. Beads were transferred to a column and washed with 1% OG lysis buffer (1% Octylglucoside, 50mM Tris-HCl pH 7.4, 150mM NaCl, 1mM CaCl2, 2mM MgCl2, 1mM PMSF, 1mM benzamidine, 0.01 % aprotinin, 0.1% BSA). PKCε was immunoprecipitated on the column from lysates of MTF7L cells activated for 30 minutes in 20%FBS/DME in suspension and immunoprecipitates resolved by SDS– PAGE on 8% or 10% gels. Proteins were visualized with Colloidal Coomassie blue for MALDI TOF m/s, and protein bands were excised and sent for mass spectrometry analysis on MALDI-TOF/TOF 4700 Analyzer (Applied Biosystems) by The Proteomics and Mass Spectrometry Core Facility, Cornell University. Filamentous actin assay An actin binding spin-down assay kit, for non-muscle actin (Cytoskeleton Inc, Denver, CO) was used to examine the interaction between PKCε and actin. Briefly, filamentous actin was allowed to form in compatible buffer from globular actin and purified PKCε was added to the solution to react for 30 min. The reaction mixture was ultracentrifuged and separated into supernatant (S) and pellet (P) fractions. S and P fractions were resolved by SDS-PAGE, stained with Coomassie blue and Western probed with anti- PKCε. Confocal microscopy Confocal microscopy was used to monitor cytoskeletal actin structure in MTF7L breast cancer cells. Tumor cells that had been subjected to EoE suspension culture were washed twice in DMEM containing 1% bovine serum albumin (BSA), 103 permeabilized with permeabilization buffer (0.5% TX-100, 20mM HEPES, 3mM MgCl2, 50mM NaCl, 300mM sucrose, 0.02% NaN3) at room temperature for 10 min, then incubated with phalloidin—fluorescein (to stain for filamentous actin) or mouse anti-PKCε diluted 1:50 or 1:300, respectively, in PBS containing 1%BSA (PBS-BSA) for 1h at 4°C . After washing in PBS-BSA, tumor cells stained with phalloidin were fixed in 2% paraformaldehyde in PBS and those stained with PKCε were washed twice with 0.1% Tween20 in PBS, stained with PE conjugated anti-mouse antiserum for 1h at 4°C, washed twice, then fixed. In select experiments, cells were stained with rabbit anti-Fn as described. Coverslips were mounted onto glass slides using Prolong Antifade (Molecular Probes, Eugene, OR) and viewed by laser scanning confocal microscopy (LSCM). An Olympus Fluoview 500 confocal laser scanning imaging system equipped with argon, krypton, and He-Ne lasers on an Olympus IX70 inverted microscope with a PLAPO 60× objective was used (Olympus America, Inc., Melville, NY). Confocal images were processed using Adobe Photoshop 6.0 (Adobe Systems, Inc., Mountain View, CA). RESULTS Mass spectrometry of PKCε complex in MTF7L reveals two cytoskeletal interacting proteins To identify potential interacting proteins within the PKCε signaling complex, we used a PKCε immunoaffinity column. Proteins that co-immunoprecipitated with PKCε were visualized with Coomassie staining (Fig 3.1A). The bands were excised from the SDS-PAGE gels and subjected to MALDI TOF m/s analysis. Analysis revealed two cytoskeletal proteins that were associated with PKCε, myosin IIa and βactin-2. In this paper, we shall examine the contribution of β-actin-2 to PKCε- 104 Figure 3.1: PKCε associates with β-actin: (A) Coomassie blue staining and western blot of PKCε complex: MTF7L cells were suspended in 20% FBS/DMEM for 30 minutes then lysed and immunoaffinity purified on an anti-PKCε antibody conjugated protein G column. Bound proteins were resolved by SDS-PAGE, the gel was stained with Colloidal Coomassie blue, and protein bands were excised for MALDI TOF m/s analysis. Corresponding fractions were ran in duplicate for Western blotting. (B) β-actin co-immunoprecipitates with PKCε. Immunoaffinity-purified PKCε fractions #E3 and #E4 from lysed MTF7L cells treated as described above were Western probed with anti-β-actin antibodies. PKCε and β-actin co-immunoprecipitate, while corresponding control mIgG fraction is negative for both PKCε and β-actin. (C,D) PKCε binds F-actin. Immunoaffinity purified PKCε was subjected to an Factin binding assay using Biochem Actin binding spin-down assay kit as described in Material and Methods. Coomassie blue staining (CB) reveals F-actin co-sedimentation of positive control α-actinin, no co-sedimentation for negative control bovine serum albumin (BSA) (C) and PKCε co-sedimentation with F-actin in both eluted fractions E3 and E4 of purified PKCε (D), as confirmed by Western blotting (WB). S, supernatant. P, pellet. 105 106 mediated polyFn assembly. To confirm the results of MALDI TOF m/s analysis, we used anti-βactin antibody to specifically detect βactin in immunoprecipitates of PKCε. βactin co-immunoprecipitates (co-IP) with PKCε, but not with control mouse(m)IgG nonspecific immunoprecipitates (Fig 3.1B). To determine whether PKCε binds to functional actin filaments, an in vitro filamentous actin (F-actin) binding assay was performed. Results reveal that PKCε does indeed bind to F-actin, suggesting a cytoskeletal involvement in PKCε-mediated polyFn assembly (Fig 3.1C). PKCε and F-actin co-immunoprecipitate and colocalize at the subplasmalemmal space beneath polyFn aggregates PKC subcellular localization determines the specificity of kinase action on cellular function. Interactions with actin-binding proteins, RACKs or direct interaction with actin can position activated PKC at specific membrane domains in proximity of suitable substrate proteins, and further promote substrate phosphorylation. To determine whether PKCε and actin colocalizes at a certain cellular compartment, we utilized fluorescent confocal microscopy to examine the redistribution of activated PKCε in MTF7L cells following suspension culture in FBS. MTF7L cells were treated in EoE suspension in 20% FBS and endogenous PKCε was stained with red fluorescence while filamentous actin (F-actin) was stained with green fluorescence with phalloidin-fluorescein. Staining of PKCε and F-actin and their co-localization is monitored as a function of time in EoE suspension culture (Fig 3.2). At 5 min of EoE suspension culture, much of the PKCε and F-actin are dispersed in a punctate manner throughout the cytoplasm with some F-actin lining the subplasmalemmal space. There is no co-localization of the two molecules. At 10 min of incubation, PKCε and F-actin 107 Figure 3.2: PKCε colocalization with F-actin in activated MTF7L cells is time and cytoskeletal integrity-dependent: MTF7L cells subjected to EoE suspension for 5, 10 or 30 minutes were stained with anti-PKCε or phalloidin-FITC (for F-actin) and analyzed by confocal microscopy. Confocal images of cells were merged and colocalized images isolated. MTF7L cells show increasing, time-dependent colocalization of PKCε-F-actin in the subplasmalemmal region. 108 109 have both co-segregated to the subplasmalemmal space, where they co-localize in a punctate manner reminiscent of the distribution of polyFn globules on the cancer cell surface. At 30 min of incubation, co-localization of PKCε and F-actin in the subplasmalemmal space is further pronounced. Confocal microscopy data were complemented by biochemical analyses. MTF7L cells subjected to EoE suspension culture for various periods of time were lysed, and lysates subjected to immunoprecipitation with anti-PKCε antibody. Western blotting of anti-PKCε immunoprecipitates shows that the association of PKCε and actin peaks at 15 minutes, then decreases drastically afterwards (Fig 3.3A). As we have shown previously (4), PKCε phosphorylation peaks at 15 minutes in suspension, correlating to the time point at which PKCε and actin binding occurs at maximum as detected by Western blotting. To determine whether PKCε activity is required for this PKCε-actin association, the inhibitor BIMXI was used at PKCε-inhibitory concentration. BIMXI caused a significant decrease in the association of PKCε and actin after drug inhibition (Fig 3.3B), indicating that PKCε activity is critical in mediating binding of PKCε to the actin cytoskeleton and activation of PKCε most possibly occurs before PKCε-actin association. F-Actin cytoskeletal integrity is necessary for polyFn assembly and proper PKCε localization The presence of an intact cytoskeleton has been shown to be critical for Fn matrix assembly in adherent cells (15). The ability of cytoskeleton to control fibronectin assembly may be mediated through regulation of the integrin functional state by interaction with cytoplasmic tails of β integrin (15) or to support proper PKC signaling complex localization (16). To examine whether cytoskeletal integrity is 110 Figure 3.3: PKCε and actin association: (A) PKCε-actin association is timedependent. MTF7L cells are treated with FBS for the time points 0, 15, and 30 minutes and PKCε was immunoprecipitated and detected by Western blotting. Blots were subjected to staining with anti-β-actin antibody or anti-PKCε antibody. (B) PKCε -actin association is inhibited by BIMXI. MTF7L cells were trypsinized and treated BIMXI (500nM, PKCε inhibitor) for 15 minutes and suspended in 20% FBS/DMEM for 0, 15 and 30 minutes, and processed as above. 111 112 Figure 3.4: PolyFn assembly is dependent on cytoskeletal integrity: Actin depolymerizing drugs, Cytochalasin D (A) and Latrunculin B (B), block polyFn assembly. MTF7L rat breast cancer cells were exposed for 30min to 1uM, 10uM and 20uM of Cytochalasin D or 200nM and 1uM of Lantrunculin B, and then subjected to EoE suspension culture for 1 hour in the presence of the drug. Cells were stained for poly-Fn and analyzed by FACS, showing a dose-dependent inhibition of polyFn assembly that is confirmed densitometrically and by immunocytochemistry. (C) Cells treated with solvent (DMSO) or Cytochalasin D (CytD), were EoE cultured for 30 minutes in 20% FBS, and then immunostained and processed as described in Figure 3.2. Data show co-localization of PKCε and F-actin in the subplasmalemmal space of DMSO-treated cancer cells and cytoplasmic clusters of PKCε/F-actin complexes in Cytochalasin D-treated cells. 113 114 critical for polyFn assembly of breast cancer cells during suspension, we used two inhibitors known to disrupt actin polymerization, Cytochalasin D and Latrunculin B. Both inhibitors were able to block polyFn assembly in MTF7L cells in dose-dependent manners (Fig 3.4A&B), with Cytochalasin D having a more prominent effect. Cytochalasin D impedes actin polymerization by binding to barb ends of actin filaments, blocking addition of monomeric actin to actin filaments, while Latrunculin B blocks actin polymerization by binding to monomeric actin (17,18). The distinct mechanisms of the two inhibitors may explain the differences in the inhibitory activity on polyFn assembly. To determine whether disturbance of actin cytoskeletal organization would perturb the subcellular localization of the PKCε-actin complex, we used the stronger inhibitor of polyFn assembly, Cytochalasin D. Cytochalasin D treatment resulted in increased cytoplasmic localization of the PKCε-actin binding complex (Fig 3.4C), which is paralleled by diminished polyFn. By altering the subcellular localization of PKCε-actin, Cytochalasin may render PKCε unable to activate downstream signaling components leading to inhibited polyFn assembly. These results suggest that an intact cytoskeletal structure is necessary for and supports the proper localization of the PKCε complex to transduce signals leading to polyFn assembly. Deletion of proposed Protein kinase Cε actin-binding site is insufficient to block polyFn assembly A unique actin binding domain of PKCε that is homologous to other actin binding proteins has been identified and localized to amino acids 223-228, located between the C1A and C1B domains (19). A mutant protein lacking this actin binding site is able to perturb proper neurite formation but does not alter its subcellular 115 Figure 3.5: PKCε actin binding motif is not essential for polyFn assembly: MTF7L breast cancer cells were transfected with wild type PKCε or PKCε deleted of its actin binding domain (cloned into pEGFR-N1), immunostained and analyzed for poly-Fn on FACS. X-axis gates were selected based on the staining of MTF7L cells with non-immune rabbit IgG. Y-axis gates were selected based on 50% Fn staining intensity of the vector transfected GFP-positive cells. Cells in the upper right quadrant of the scatter gram (strong expressers of PKCε and Fn) were expressed as percent of the sum of cells in the right upper and lower quadrants (strong PKCε expressers), providing a quantitative measure of the inhibitory activity of PKCε−△ABD (A). Cell lysates from transfected cells were blotted with anti-GFP antibody to show equal expression levels (B) and also immunoprecipitated with anti-GFP and blotted with actin to show the actin binding activity of each construct (C). (D) Adenovirusmediated expression of PKCε−△ABD shows identical results. AdWTε, adenoviral PKCε−WT. AdABDε, adenoviral PKCε−△ABD. 116 117 localization in cells compared to the wild type protein (20). We generated a PKCε construct lacking amino acids 223-228 (PKCε− △ ABD) in order to investigate whether deletion of this domain will negatively affect PKCε mediated polyFn assembly. Surprisingly, transfection of PKCε−△ABD into MTF7L cells failed to block surface expression of polyFn (Fig 3.5A). This was not due to differential expression of the transfected proteins, as shown by equal GFP-tagged protein expression (Fig 3.5B). Overexpression of PKCε−△ABD from adenoviral vectors with over 90% transduction rate in MTF7Ls also failed to have any effect on polyFn (Fig 3.5D). These data suggest that the proposed actin binding domain does not have a direct role for PKCε regulated polyFn assembly in MTF7L cells. We then examined whether the mutant PKCε, PKCε− △ ABD, is still able to bind β-actin. Immunoprecipitation of the transfected protein PKCε− △ ABD revealed that this mutant PKCε still binds β-actin, but seems to have weaker affinity compared to wild type PKCε (Fig 3.5C). It is possible that this residual actin binding still plays a role in supporting polyFn assembly, and that under these circumstances, an alternative pathway or accessory proteins exists in MTF7L cells that is able to compensate for the loss of the actin binding domain in PKCε. The fact that PKCε−△ABD was previously found to localize to the same compartments as wild type PKCε (20) also further highlights the importance of proper PKCε localization to the cortical cytoskeleton. Pyk2 kinase activity is required for polyFn assembly Pyk2 has been known to have a role in actin cytoskeleton organization in processes such as cell adhesion and cell migration of several cell types (21). The autophosphorylation and activation of Pyk2 is stress fiber-dependent (21) and its activation is PKCε-dependent in myocytes (7). These facts prompted us to investigate whether or not Pyk2 kinase activity plays a role in polyFn assembly. A Pyk2 kinase 118 dead (K475A) mutant (Pyk2-KD) was used to transfect MTF7L cells in comparison to Pyk2 wild type (Pyk2-WT). Wild type and kinase dead mutant of the other focal adhesion kinase member, FAK, were also transfected in MTF7 cells for comparison. 24 hours after transfection, the cells were subjected to EoE suspension culture for 1h, then stained with anti-Fn antibody to detect surface assembled polyFn. Results show that the introduction of kinase dead Pyk2 partially impaired the ability of the cells to form surface Fn aggregates (Fig 3.6A&B), demonstrating the active role of Pyk2 during the course of polyFn formation on the cancer cell surface. In contrast, kinasedead FAK had no effect on polyFn assembly (Fig 3.6A). To further examine whether or not Pyk2 autophosphorylation correlates with the course of polyFn assembly in our assay, MTF7L cells were serum-starved overnight and incubated in an EoE suspension culture in FBS for the indicated time points. Cell lysates were probed for phosphorylated Pyk2 content (pPyk2). The time course experiment indicated that pPyk2 peaked at 10 minutes of incubation (Fig 3.6C), coinciding with the peak of PKCε phosphorylation as previously shown (4). The close occurrence of phosphorylation of PYK2 and PKCε suggests that one protein could be regulating the phosphorylation of the other. Since a previous study in rat ventricular myocytes has shown that PKCε regulates Pyk2, we were eager to know whether or not diminished PKCε will have an effect on Pyk2 activation. Indeed, the PKCε inhibitor BIMXI concentration markedly decreased phosphorylation of Pyk2 (Fig 3.6D). Similar results were obtained when cells were treated with Cytochalasin D (Fig 3.6D). This suggests that Pyk2 activation is downstream of PKCε and that abnormal subcellular localization of PKCε under disrupted cytoskeletal organization will abrogate PKCε downstream signaling to Pyk2. Salicylate, the known Pyk2 inhibitor, is able to abrogate the phosphorylation of the PKCε mediated Pyk2 phosphorylation (22). We tested this inhibitor for its effects 119 Figure 3.6: Pyk2 kinase activity is required for polyFn assembly: (A&B) MTF7L breast cancer cells were transiently transfected with wild type or kinase-dead FAK and Pyk2 and 24h later subjected to EoE suspension culture in DMEM+20%FBS for 1h at 37℃. Cells expressing equal amounts of wild type and kinase-dead FAK and Pyk2 were stained with anti-Fn antibodies and analyzed by dual-color FACS as described. (A) and Fn-immunocytochemistry (B). Kinase-dead Pyk2 but not FAK impairs polyFn assembly. (C) Time-dependent Pyk2 phosphorylation: Pyk2-transfected MTF7L cells were suspended for the indicated time periods in DMEM+20%FBS, then lysed. Lysates were separated by SDS-PAGE, blotted onto nitrocellulose and probed with phosphor-specific Pyk2(Y402) antibody or anti-Pyk2 antibody. (D) BIMXI and Cytochalasin D (CytD) inhibit Pyk2 phosphorylation: MTF7 cells were trypsinized and treated with BIMXI (500nM) or CytD(1μM) for 15 min, then subjected to EoE suspension in 20% FBS/DMEM for 0,15, and 30 min, lysed and lysates resolved by SDS-PAGE (10% gel) and Western probed with anti-Pyk2 and phosphor-specific antiPyk2 antibodies. (E) MTF7L cells pretreated with sodium salicylate (10 or 20mM) for 30 min, then subjected to EoE suspension (1h; DMEM+20%FBS) were stained with anti-Fn and analyzed by FACS. The Pyk2 inhibitor salicylate inhibits polyFn assembly in a dose-dependent manner. 120 121 on polyFn and found that it abolished polyFn formation at the proposed working concentrations (Fig 3.6E). DISCUSSION Early work on PKCε has linked it to oncogenic transformation of cells. The oncogenic activity has been reported in fibroblasts, colonic cell lines, prostate epithelial cell lines and breast cancer cell lines (23-26). In mice, epidermis-specific transgenic overexpression of wild type PKCε resulted in development of metastatic squamous cell carcinomas (27). High-density tissue microarray analysis revealed significant positive correlation between PKCε and tumor histologic grade, while knockdown of PKCε reduced breast cancer cell invasiveness and lung metastasis (26). In our previous study, impairment of PKCε in rat breast cancer cells by a dominant negative construct and specific drug inhibitors leads to decreased polyFn assembly and reduced lung colonization (4). The converging signaling events of PKCε and the actin cytoskeleton is a potential mechanism underlying these oncogenic processes. Many approaches have been used to identify PKC binding partners in order to understand the role of PKC in different cellular functions (28). Mass spectrometry results have showed that PKCε associates with proteins including β’Cop, cytokeratins, cardiac myofibrils, matrin3, transferin, Rac GAP1 and vimentin. (16,29,30). Many of the binding proteins identified within the PKCε complex are putative PKC substrates such as MARCKS and STICKS, while others are not phosphorylated by PKC and may function as anchoring proteins or regulators of substrate phosphorylation, facilitating integration of PKC signaling into other pathways (5,16,19,29). Our current study presents two cytoskeletal interacting partners of PKCε potentially involved in mediating polyFn assembly. This finding is consistent with a previous study 122 performed in fibroblasts (28), which found that both actin and myosin IIa coimmunoprecipitated with PKCε but are not substrates of PKCε. We have tested the myosin ATPase inhibitors 2,3-Butadione monoxime (31) and blebbistatin (32) and it seems that myosin ATPase activity is not involved in our cellular function in polyFn assembly (Fig 3.7 A&B). Although PKCε does not directly regulate myosin phosphorylation (28), it is still possible that PKCε is able to modulate myosin activity non-catalytically or that myosin can regulate PKCε activity. The role of myosin IIa in the PKCε complex may also be to bridge PKCε to actin stress fibers or to strengthen the association of PKCε and F-actin. PKC activity is regulated through differential recruitment by PKC isotypespecific binding proteins localized in different subcellular domains, which facilitates PKC localization to activate proteins in a particular cellular compartment (19,30,33). Subcellular localization is a key event in PKC functional and regulatory specificity. We show that both PKCε activity and cytoskeletal integrity is required for PKCε-actin association in MTF7L cells and that PKCε increasingly associates with actin in a time dependent manner. Inhibition of PKCε activity diminishes this response, suggesting that the catalytic component of PKCε could play a role in either its initial binding to actin or strengthen the binding interaction and that PKCε must be activated before binding to the actin cytoskeleton. Perturbing the cytoskeleton by Cytochalasin D prevents PKCε translocation from the cytoplasm to the plasma membrane, resulting in abrogation of polyFn assembly on tumor cell surfaces. Therefore, we propose that upon PKCε activation, PKCε binds to intact F-actin cytoskeleton, which sustains PKCε activity and PKCε translocation to the plasma membrane and facilitates downstream activation of Pyk2 to promote successful polyFn assembly. Interaction between PKCε and F-actin has been shown to activate PKCε (19,34) and the PKCε actin binding domain has been mapped to amino acids 223-228. 123 Figure 3.7: Inhibitors of myosin ATPase activity had no effect on polyFn assembly: MTF7L cells pretreated with blebbistatin (100μM or 200μM) (A) and 2,3Butadione monoxime (BDM) (20 μM) (B) for 30 min, then subjected to EoE suspension (1h; DMEM+20%FBS) were stained with anti-Fn and analyzed by FACS. 124 125 In our results, deletion of the proposed actin binding site of PKCε was insufficient to block the formation of polyFn on breast cancer cells and PKCε activation seems to be a prerequisite for actin binding. Immunoprecipitation of wtPKCε-GFP and PKCε−△ ABD -GFP revealed residual levels of actin binding. The residual actin binding may be due to the existence of an alternate actin binding site of PKCε or through alternative pathways via supporting proteins such as other actin-binding proteins within the cells that can bind PKCε to the cytoskeleton. On the other hand, the fact that PKCε−△ABD overexpression failed to abrogate polyFn assembly could be due to the fact that PKCε−△ABD still retains the wild type protein subcellular localization (20) and that subcellular localization is the crucial factor in the relationship of actin cytoskeleton and PKCε in this study. The proximity of both wild type PKCε and PKCε−△ABD, which both possess the catalytic domain, to the cytoskeleton and/or plasma membrane compartment may be the main feature responsible for downstream signaling leading up to polyFn assembly. Hence, the mere absence of the actin-binding site may not be truly sufficient to induce a significant effect on polyFn formation when the mutant protein maintains proper localization that supports PKCε to activate the necessary proteins at its cytoskeletal signaling site. Pyk2 and FAK are highly homologous in sequence and share a similar structural organization (35). Despite of the high similarity, the two proteins seems to be differentially regulated. Although phosphorylation of FAK and Pyk2 can both be controlled by integrin engagement, Pyk2 response to soluble factors such as angiotensin II is dramatically stronger in certain cells, while FAK is mainly regulated by integrin-mediated cell adhesion (35). Pyk2 and FAK also associate with the cytoskeleton and with integrin signaling complexes by complexing to proteins such as Src kinases, p130 Cas, paxillin and Graf1 (36). FAK C-terminal domain binds the integrin-associated protein talin while Pyk2 binds the actin-associated protein gelsolin 126 (37), which can further account for their differential functions. In this study, we show that Pyk2 is the main adhesion kinase responsible for downstream signaling from PKCε to polyFn assembly of MTF7L cells, indicated by the ability of kinase dead Pyk2 and the Pyk2 inhibitor salicylate to impair polyFn formation. Inhibition of PKCε by BIMXI, which also blocks polyFn assembly (4), inhibits time-dependent activation of Pyk2 in MTF7L cells. The relationship of PKCε and Pyk2 is consistent with a previous study which shows that constitutively active PKCε increases endothelin-1 induced Pyk2 phosphorylation while kinase inactive PKCε inhibited Pyk2 phosphorylation (7). Pyk2 has been implicated in integrin inside-out signaling by associating with beta3 integrins (38) and this could be a possible pathway by which Pyk2 regulates polyFn assembly. The phosphorylated beta3 integrin cytoplasmic tail provides a binding site for Pyk2 (8). Pyk2 has the FERM domain that contains a phosphotyrosine binding domain (PTB) domain similar to talin, which is also able to bind beta3 tail and trigger activation for ligand binding (39). Cellular signal induces the conformational change of talin allowing it to bind to the beta integrin tail and unclasps the complex between the alpha and beta integrin cytoplasmic tails, which causes a conformational shift in the extracellular head domain and renders it more extended for high affinity ligand binding (39). Pyk2 binding to the integrin cytoplasmic tails could have a similar impact on the polyFn receptor and activate the integrins for fibronectin binding and subsequent polyFn assembly. Results obtained in this study are summarized in Fig 3.8A. In our working model (Fig 3.8B), we propose that in MTF7L cells subjected to EoE suspension culture PKCε is activated and phosphorylated and associates with F-actin in the cytoplasm, presumably at secretory vesicle located in the vicinity of the Golgi complex (Fig 3.2, arrows). The PKCε-actin complex then translocates in increasing 127 Figure 3.8: Diagram of cellular processes in polyFn assembly: (A) Summary of effects of various treatments performed in this study, providing the relationship of subcellular localization of PKCε, actin and polyFn. (B) Signaling pathway in polyFn assembly. Soluble ligands in the plasma activates PKCε through receptor binding, PKCε translocates to the membrane and cortical cytoskeleton, activates PYK2 in proximity. Activated PYK2 activates the beta integrin cytoplasmic tail domain and extracellular domain of integrin is rendered extended for ligand binding. Phosphorylated protein marked with “p”. 128 129 amounts (10-30min) to the subplasmalemmal space, which is prevented by Cytochalasin D as shown by the intracellular accumulation of PKCε/actin complexes (Fig 3.4). Whether the PKCε-mediated activation of Pyk2 happens immediately following complexing of PKCε/F-actin in the cytosolic compartment or at the plasma membrane in unclear. However, strong phosphorylation of Pyk2 as early as 1-5 min after subjecting MTF7L cells to EoE suspension culture (Fig 3.6C), i.e., prior to the appearance of PKCε/F-actin complexes in the subplasmalemmal space (Fig 3.2), suggests that Pyk2 is activated in the cytosol. At Golgi-derived secretory vesicles (Fig 3.2, arrows), PKCε and/or Pyk2 could well serve as activator(s) of β integrins, collaborating with talin to promote integrin activation and clustering (39) and to initiate the assembly of Fn/integrin complexes in secretory vesicles. As secretory vesicles fuse with plasma membrane, Fn is exposed at the cell surface, where it serves as a scaffold for Fn-Fn self association with plasma Fn into insoluble polymeric complexes (1). Such mechanism is supported by siRNA Fn-knockdown experiments in which polyFn is dramatically reduced, albeit cancer cells are exposed to high plasma Fn concentration in EoE suspension culture (4). Similarly, human MCF7 breast cancer cells that do not express Fn yet have a similar β1 integrin expression profile as MTF7L (40) are unable to assemble polyFn on their surfaces in medium containing 20% FBS or plasma Fn (Pauli et al., personal communication). Together, our data provided novel information of the polyFn assembly process, that in previous studies from our laboratory has been shown to be an integral part of cancer cells’ ability to colonize the lungs (1,2,4). 130 REFERENCES 1. Cheng, H. C., Abdel-Ghany, M., Elble, R. C., and Pauli, B. U. (1998) J Biol Chem 273(37), 24207-24215 2. Cheng, H. C., Abdel-Ghany, M., Zhang, S., and Pauli, B. U. (1999) Clin Exp Metastasis 17(7), 609-615 3. Johnson, R. C., Zhu, D., Augustin-Voss, H. G., and Pauli, B. U. (1993) J Cell Biol 121(6), 1423-1432 4. Huang, L., Cheng, H. C., Isom, R., Chen, C. S., Levine, R. A., and Pauli, B. U. (2008) J Biol Chem 283(12), 7616-7627 5. Jaken, S., and Parker, P. J. (2000) Bioessays 22(3), 245-254 6. Disatnik, M. H., Boutet, S. C., Pacio, W., Chan, A. Y., Ross, L. B., Lee, C. H., and Rando, T. A. (2004) J Cell Sci 117(Pt 19), 4469-4479 7. Bayer, A. L., Heidkamp, M. C., Howes, A. L., Heller Brown, J., Byron, K. L., and Samarel, A. M. (2003) J Mol Cell Cardiol 35(9), 1121-1133 8. Butler, B., and Blystone, S. D. (2005) J Biol Chem 280(15), 14556-14562 9. Stawowy, P., Margeta, C., Blaschke, F., Lindschau, C., Spencer-Hansch, C., Leitges, M., Biagini, G., Fleck, E., and Graf, K. (2005) Cardiovasc Res 67(1), 50-59 10. Besson, A., Wilson, T. L., and Yong, V. W. (2002) J Biol Chem 277(24), 22073-22084 11. Palmantier, R., George, M. D., Akiyama, S. K., Wolber, F. M., Olden, K., and Roberts, J. D. (2001) Cancer Res 61(6), 2445-2452 12. Zeidman, R., Lofgren, B., Pahlman, S., and Larsson, C. (1999) J Cell Biol 145(4), 713-726 13. Cheng, H. C., Abdel-Ghany, M., and Pauli, B. U. (2003) J Biol Chem 278(27), 24600-24607 14. Abdel-Ghany, M., Cheng, H. C., Elble, R. C., and Pauli, B. U. (2001) J Biol Chem 276(27), 25438-25446 131 15. Wu, C., Keivens, V. M., O'Toole, T. E., McDonald, J. A., and Ginsberg, M. H. (1995) Cell 83(5), 715-724 16. Keenan, C., and Kelleher, D. (1998) Cell Signal 10(4), 225-232 17. McAvey, B. A., Wortzman, G. B., Williams, C. J., and Evans, J. P. (2002) Biol Reprod 67(4), 1342-1352 18. Wakatsuki, T., Schwab, B., Thompson, N. C., and Elson, E. L. (2001) J Cell Sci 114(Pt 5), 1025-1036 19. Prekeris, R., Mayhew, M. W., Cooper, J. B., and Terrian, D. M. (1996) J Cell Biol 132(1-2), 77-90 20. Zeidman, R., Troller, U., Raghunath, A., Pahlman, S., and Larsson, C. (2002) Mol Biol Cell 13(1), 12-24 21. Wu, S. S., Jacamo, R. O., Vong, S. K., and Rozengurt, E. (2006) Cell Signal 18(11), 1932-1940 22. Wang, Z., and Brecher, P. (2001) Hypertension 37(1), 148-153 23. Tachado, S. D., Mayhew, M. W., Wescott, G. G., Foreman, T. L., Goodwin, C. D., McJilton, M. A., and Terrian, D. M. (2002) J Cell Biochem 85(4), 785-797 24. Perletti, G. P., Concari, P., Brusaferri, S., Marras, E., Piccinini, F., and Tashjian, A. H., Jr. (1998) Oncogene 16(25), 3345-3348 25. Wu, D., Foreman, T. L., Gregory, C. W., McJilton, M. A., Wescott, G. G., Ford, O. H., Alvey, R. F., Mohler, J. L., and Terrian, D. M. (2002) Cancer Res 62(8), 2423-2429 26. Pan, Q., Bao, L. W., Kleer, C. G., Sabel, M. S., Griffith, K. A., Teknos, T. N., and Merajver, S. D. (2005) Cancer Res 65(18), 8366-8371 27. Jansen, A. P., Verwiebe, E. G., Dreckschmidt, N. E., Wheeler, D. L., Oberley, T. D., and Verma, A. K. (2001) Cancer Res 61(3), 808-812 28. England, K., Ashford, D., Kidd, D., and Rumsby, M. (2002) Cell Signal 14(6), 529-536 29. Csukai, M., Chen, C. H., De Matteis, M. A., and Mochly-Rosen, D. (1997) J Biol Chem 272(46), 29200-29206 132 30. Xu, T. R., and Rumsby, M. G. (2004) FEBS Lett 570(1-3), 20-24 31. Soeno, Y., Shimada, Y., and Obinata, T. (1999) Cell Tissue Res 295(2), 307-316 32. Shu, S., Liu, X., and Korn, E. D. (2005) Proc Natl Acad Sci U S A 102(5), 1472-1477 33. Csukai, M., and Mochly-Rosen, D. (1999) Pharmacol Res 39(4), 253-259 34. Prekeris, R., Hernandez, R. M., Mayhew, M. W., White, M. K., and Terrian, D. M. (1998) J Biol Chem 273(41), 26790-26798 35. Zheng, C., Xing, Z., Bian, Z. C., Guo, C., Akbay, A., Warner, L., and Guan, J. L. (1998) J Biol Chem 273(4), 2384-2389 36. Ivankovic-Dikic, I., Gronroos, E., Blaukat, A., Barth, B. U., and Dikic, I. (2000) Nat Cell Biol 2(9), 574-581 37. Mitra, S. K., Hanson, D. A., and Schlaepfer, D. D. (2005) Nat Rev Mol Cell Biol 6(1), 56-68 38. Pfaff, M., and Jurdic, P. (2001) J Cell Sci 114(Pt 15), 2775-2786 39. Qin, J., Vinogradova, O., and Plow, E. F. (2004) PLoS Biol 2(6), e169 40. Morini, M., Mottolese, M., Ferrari N., Ghiorzo, F., Buglioni, S., Mortarini, R., Noonan, D.M., Natali, P.G., and Albini, A. (2000) Int. J. Cancer 87(3), 336-342 133 CHAPTER 4 CONCLUDING DISCUSSIONS 134 PKC EPSILON-MEDIATED REGULATION OF POLYMERIC FIBRONECTIN ASSEMBLY Overexpression of fibronectin (Fn) has been demonstrated in invasive/metastatic breast tumors and tumor cells from primary invasive ductal breast carcinomas (1-3) as well as cell lines such as MDA-MB-231 and MTF7, and is especially prominent in cell lines selected for enhanced lung metastatic ability (4,5). Survey of all our highly lung-metastatic breast cancer cell lines revealed high levels of Fn expression (4). Tumor cells that were isolated from the blood of tumor-bearing animals are characterized by a cell surface coat of multi-globular polyFn, which could be reproduced on the cell surface of cancer cells subjected to suspension culture in the presence of serum. This surface coat of polyFn mediates the docking of breast cancer cells MTF7 to the lung endothelial addressin DPP IV and facilitates lung colonization (4,6-8). In this study, the mechanism of polymeric fibronectin formation on the surface of breast cancer cells was investigated. We found that protein synthesis and secretion is crucial for the polyFn assembly process and knockdown of endogenous cellular fibronectin impedes both the formation of surface polyFn and pulmonary metastasis of the cancer cells. PolyFn assembly is regulated by PKCε. Pharmocological inhibition of PKC, transfection of dominant negative regulatory domain mutant of PKCε and knockdown of PKCε protein all decreased surface polyFn in MTF7 cells. PKCε translocates to the membrane in increasing amounts during the course of polyFn assembly, which is accompanied by phosphorylation of PKCε on serine and threonine residues. In contrast, transfection of dominant negative PKCδ and knockdown of PKCδ promoted polyFn assembly, in accordance with the proposed opposing effects of PKCε and PKCδ (9). 135 In order to further understand how PKCε reglates polyFn assembly, we utilized mass spectrometry to identify possible regulating factors residing within the PKCε complex formed during suspension incubation. Two cytoskeletal proteins were found within the PKCε complex, β-actin and myosin IIa. Myosin ATPase activity is not required for polyFn assembly, as evidenced by lack of effect of pharmacological inhibitors. On the other hand, β-actin associated with PKCε in a time-dependent and PKC activity-dependent manner, and filamentous actin associated with PKCε increasingly at the subplasmalemmal space. The actin depolymerizing drug, Cytochalasin D, disrupted formation of polyFn and perturbed the subplasmalemmal localization of the PKCε-actin complex. The images of PKCε-actin complexes within the cells showed a few cytoplasmic granule-like existence while images from Cytochalasin D treated cells demonstrated a tremendous amount of cytoplasmic complexes. Since PKCε has been implicated in various exocytic processes (10-12), these observations raise the question of whether the PKCε-actin complexing could be involved in exocytosis of fibronectin or polyFn receptors. One of the focal adhesion kinase family members, Pyk2, was also shown to be involved in polyFn assembly. Transfection of a kinase dead mutant of Pyk2 and pharmacological inhibition with the Pyk2 inhibitor sodium salicylate both decreased surface polyFn assembly, and time-dependent phosphorylation of Pyk2 was blocked by the PKCε inhibitor BIMXI and Cytochalasin D, suggesting that Pyk2 phosphorylation was downstream of cytoskeletal integrity and PKCε activity. Pyk2 possess the same FERM domain by which talin, a cytoplasmic cytoskeletal protein that regulates integrin ligand binding affinity via inside-out signaling (13), binds integrin cytoplasmic tails. Pyk2 has also been shown to bind to β3 integrin cytoplasmic tails (14). Therefore, we propose in our working model that PKCε regulates polyFn assembly by signaling downstream to Pyk2 and allowing Pyk2 to 136 relay the signal to β3 integrin cytoplasmic tails, in turn altering integrin ligand binding affinity to bind fibronectin, allowing cells to form a fibronectin scaffold that facilitates fibronectin self-association and build up of polymeric fibronectin globules on the surface of cancer cells. The soluble factor responsible for triggering the polyFn assembly cascade has not been identified. This factor could be a cytokine secreted by the cancer cells and could act in an autocrine manner. The factors tested in this study are listed in Table 1. Negative results may indicate either: the factor was not involved, the factor was insufficient to trigger an active response, the concentration used was not optimal, or the activating conditions were not optimal. The conditioned medium isolated from MTF7 culture was able to induce an increase in polyFn assembly (Fig 4.1A). Two soluble factors were identified in the conditioned medium of MTF7 cells, monocyte chemotactic protein-1 (MCP-1) and tissue inhibitor of metalloproteinase (TIMP-1) (Fig 4.1B), and MCP-1 was tested and shown to be unable to increase polyFn compared to control (Table 1). Future analysis of the conditioned medium of MTF7 cells with cytokine arrays that test for a greater number of cytokines than the 18cytokine array used may be helpful in identifying the “trigger” factor. Future investigations on the receptor(s) of polyFn may present evidence to support the detailed mechanism in PKCε regulation of poly Fn assembly. The first candidate to test would be β3 integrin, binding target of Pyk2. To test the hypothesis that β3 integrin is the polyFn receptor subunit, the first step is to confirm the expression of β3 integrin in MTF7 cells. Subsequently, imaging studies can be done to visualize the colocalization of β3 integrin and polyFn globules. A possible interaction may be further characterized by attempting to co-immunoprecipitate β3 integrin and polyFn in biochemical analysis. To further investigate the requirement for inside-out signaling in β3 integrin activation, mutants of β3 integrin cytoplamic tail, which has 137 Table 1: Stimulators tested for polyFn assembly. Reagents tested negative as stimulator of polyFn assembly* Drug/reagent Lysophosphatidic acid (LPA) Phosphatidylinositol(3,4) bisphosphate (diC8) Inositol hexakisphosphate (InsP6) m-3M3FBS, phospholipase C activator Epidermal growth factor (EGF) Insulin-like growth factor (IGF-1) Stromal cell-derived factor alpha (SDF-1) Concentration used 5μM, 10 μM, 20 μM, 10μg/ml 200μM 25 μM, 50 μM 100ng/ml 100ng/ml, 200ng/ml, 500ng/ml 300ng/ml, 600ng/ml Platelet derived growth factor (PDGF) Transforming growth factor beta (TGFβ) Interleukin-8 (IL-8) 20gn/ml, 100ng/ml 10pg/ml, 20pg/ml, 50pg/ml, 500pg/ml 25ng/ml, 50ng/ml Thrombin Sphingosine-1-phosphate (S1P) Platelet-derived growth factor (PDGF) Monocyte chemoattractant protein-1 (MCP-1) 5units, 10units 500nM, 2.6μM 50ng/ml, 100ng/ml, 500ng/ml, 1μg/ml 100 ng/ml *comparison between MTF7 cells from 1 hour suspension in the presence of no drug, drug + no Fn, drug + Fn (20 μg/ml) in DMEM 138 Figure 4.1: Cancer cell secreted factors in polyFn assembly. (A) MTF7 cell overnight conditioned medium (CDME) was used to treat cells in EoE suspension culture, in comparison to medium (DME). Cells were stained with anti-Fn and analyzed with FACS. Results show cells treated with conditioned medium had an increased ability to assemble surface polyFn. (B) Rat cytokine array (Raybiotech, Norcross GA) of MTF7 CDME shows secretion of monocyte chemotactic protein1 (MCP-1) and tissue inhibitor of metalloproteinase (TIMP-1). 139 140 been shown to be crucial in integrin inside-out signaling (15), can be used in transfection experiments. Lastly, effects of knockdown of β3 integrin protein can also be determined. Clarification of the α integrin subunit partner of the possible polyFn β3 integrin receptor could further be investigated by subcellular imaging, biochemistry and mutants. Fibronectin binding sites of β3 integrin could also be blocked using peptides or blocking antibodies. The involvement of the small GTPases such as Rho, Rac and Cdc42 would also be of interest due to their roles in cytoskeletal organization (16). To test the involvement of the small GTPases Rac and Cdc42 in polyFn assembly, dominant negative or constitutive mutants of Rac and Cdc42 can be transfected into MTF7 cells. Rac or Cdc42 activity can further be characterized by activity assays testing for GTPform binding to glutathione S transferase–p21-activated kinase (GST-PAK). Functional correlating effects of mutants on polyFn assembly in conjunction with activated Rac or Cdc42 in polyFn assembly phase would indicate a role for a Rho GTPase in surface polyFn formation. Moreover, the possible involvement of a cytokine such as CXCL1(Groα;Cinc1), which is downregulated 8-fold in PKCεdepleted MTF7 cells relative to control cells, must also be considered. In support of this notion, CXCL1 binding to its receptor CXCR2 could have been responsible for the binding of RACK1 with Gβγ subunit of heterotrimeric G proteins and the observed translocation of activated PKCε to the plasma membrane (17). 141 REFERENCES 1. Amatschek, S., Koenig, U., Auer, H., Steinlein, P., Pacher, M., Gruenfelder, A., Dekan, G., Vogl, S., Kubista, E., Heider, K. H., Stratowa, C., Schreiber, M., and Sommergruber, W. (2004) Cancer Res 64(3), 844-856 2. Hao, X., Sun, B., Hu, L., Lahdesmaki, H., Dunmire, V., Feng, Y., Zhang, S. W., Wang, H., Wu, C., Wang, H., Fuller, G. N., Symmans, W. F., Shmulevich, I., and Zhang, W. (2004) Cancer 100(6), 1110-1122 3. Zucchi, I., Mento, E., Kuznetsov, V. A., Scotti, M., Valsecchi, V., Simionati, B., Vicinanza, E., Valle, G., Pilotti, S., Reinbold, R., Vezzoni, P., Albertini, A., and Dulbecco, R. (2004) Proc Natl Acad Sci U S A 101(52), 18147-18152 4. Cheng, H. C., Abdel-Ghany, M., and Pauli, B. U. (2003) J Biol Chem 278(27), 24600-24607 5. Bandyopadhyay, A., Elkahloun, A., Baysa, S. J., Wang, L., and Sun, L. Z. (2005) Cancer Biol Ther 4(2), 168-174 6. Cheng, H. C., Abdel-Ghany, M., Elble, R. C., and Pauli, B. U. (1998) J Biol Chem 273(37), 24207-24215 7. Abdel-Ghany, M., Cheng, H., Levine, R. A., and Pauli, B. U. (1998) Invasion Metastasis 18(1), 35-43 8. Cheng, H. C., Abdel-Ghany, M., Zhang, S., and Pauli, B. U. (1999) Clin Exp Metastasis 17(7), 609-615 9. Chen, L., Hahn, H., Wu, G., Chen, C. H., Liron, T., Schechtman, D., Cavallaro, G., Banci, L., Guo, Y., Bolli, R., Dorn, G. W., 2nd, and Mochly-Rosen, D. (2001) Proc Natl Acad Sci U S A 98(20), 11114-11119 10. Ivaska, J., Whelan, R. D., Watson, R., and Parker, P. J. (2002) Embo J 21(14), 3608-3619 11. Mendez, C. F., Leibiger, I. B., Leibiger, B., Hoy, M., Gromada, J., Berggren, P. O., and Bertorello, A. M. (2003) J Biol Chem 278(45), 44753-44757 12. Park, Y. S., Hur, E. M., Choi, B. H., Kwak, E., Jun, D. J., Park, S. J., and Kim, K. T. (2006) J Neurosci 26(35), 8999-9005 13. Qin, J., Vinogradova, O., and Plow, E. F. (2004) PLoS Biol 2(6), e169 14. Butler, B., and Blystone, S. D. (2005) J Biol Chem 280(15), 14556-14562 142 15. Hynes, R. O. (2002) Cell 110(6), 673-687 16. Heasman, S. J., and Ridley, A. J. (2008) Nat Rev Mol Cell Biol 9(9), 690-701 17. Chen, S., Dell, E. J., Lin, F., Sai, J., and Hamm, H. E. (2004) J Biol Chem 279(17), 17861-17868 143 APPENDIX CONDITIONAL KNOCKOUT OF FIBRONECTIN IN MAMMARY EPITHELIUM LEADS TO SEVERE LOBULO-ALVEOLAR HYPOPLASIA IN THE MOUSE MAMMARY GLAND* * Liu K, Cheng L, Flesken-Nikitin A, Huang L, Nikitin AY and Pauli BU. Conditional Knockout of Fibronectin in Mammary Epithelium Leads to Severe Lobulo-Alveolar Hypoplasia in the Mouse Mammary Gland. To be submitted. (Huang L: Biochemistry, mice mating, genotyping, mammary gland extraction) 144 ABSTRACT Fibronectin (Fn) plays an important part in the branching morphogenesis of salivary gland, lung, and kidney. Here, we examine the effect of the conditional knockout of Fn in the mammary epithelium [FnMEp–/–] on postnatal mammary gland development, using Cre-loxP mediated gene knockout technology. Our data show that FnMEp–/– mice exhibit (i) a moderate retardation in outgrowth and branching of the ductal tree as well as a reduced number of terminal end buds (TEBs) in adolescent mice (5 weeks old) that became statistically indifferent to control mice [FnMEp+/–; Fnfl+/+] in virgin adult mice (16 weeks old), and (ii) severe lobulo-alveolar hypoplasia / aplasia in 15-day pregnant FnMEp–/– mice, resulting in inability to lactate and nurse. Both heterozygous FnMEp+/– and Fnfl+/+ control mice exhibit normal MG development and Fn expression. Lack of lobulo-alveolar differentiation during pregnancy of FnMEp– /– mice is associated with significantly decreased rates of MEp proliferation, decreased numbers of MEps that had undergone ‘epithelial mesenchymal transition’ and invaded the surrounding fat pad, and decreased neovascularization, which was paralleled by a dramatic decrease in Vegf expression. Concomitant downregulation of β1 integrin (Itgb1) and decrease autophosphorylation of FAK suggest that this pathology is mediated, at least in part, by disruption of Fn/Intgb1/FAK signaling pathway. This notion is supported by the fact that conditional knockouts of Itgb1 and FAK generate the same pathology as Fn-deletion. 145 INTRODUCTION Fibronectin (Fn) is an adhesive glycoprotein that is present in extracellular matrices (insoluble form) and in body fluids (soluble form; e.g., blood plasma). It exists as a dimer composed of nearly identical ~250 kDa subunits linked covalently near their C-termini by a pair of disulfide bonds. Each monomer consists of 3 types of repeats: 12 type I, 2 types II, and 15 type III (1-3). Although Fn molecules are the product of a single gene, the resulting protein can exist in multiple forms arising from alternative splicing of a single pre-mRNA (~20 variants in humans). Splice sites are EDA (Extra Domain A), EDB (Extra Domain B), and IIICS (type III connecting segment; also called variable [V] region) (4). This multi-domain structure harboring binding sites for cell adhesion receptors (e.g., integrins, syndecans, CD44, CD26), matrix macromolecules (e.g., collagens, hyaluronic acid, proteoglycans, tenascin, thrombospondin, fibrinogen), secreted, biologically important molecules (e.g., factor XIIIa, tissue transglutaminase, fibrinogen, heparin), as well as for Fn-Fn selfassociation (5-8) makes it a perfect candidate for a ‘multi-task’ role. As such, it anchors the extracellular matrix to the cell cytoskeleton via adhesion receptors and, in doing so, relays signals from the milieu outside the cell to its inside. This so-called outside-in signaling accounts for the many functions with which Fn has been associated including cell adhesion, differentiation, growth, and migration (6,9,10). Thus, Fn is critical in vertebrate development, tissue morphogenesis and homeostasis, wound healing, angiogenesis, and cancer (1-4,9-13). Fn knockout is embryologically lethal, with abnormalities becoming manifest first at the onset of gastrulation (for details see 14,15). Many organs, including glands, lung and kidney are formed during embryonic development by epithelial branching through a process of repetitive epithelial cleft and 146 bud formations. In an elegant organ culture study, Yamada and associates (16) showed that Fn is essential for cleft formation during the initiation of epithelial branching in the submaxillary gland. Fn messenger RNA and fibrils were expressed transiently and focally in forming epithelial cleft regions. Knockdown of Fn by small interfering RNA and by inhibition with anti-Fn or anti-integrin antibodies (anti-α5, -α6, and -β1) blocked cleft formation and branching, while exogenous Fn accelerated cleft formation and branching. Mechanistically, the authors implied that fibrillar Fn mediates cleft formation in branching morphogenesis by downregulating cadherins at cell-cell junctions, thereby converting cell-cell to cell-matrix adhesions. Although a similar scenario has not been established for the mammary gland (MG), the functional role of ECM proteins, including Fn, collagen type I, and laminin, in MG postnatal development has been studied more recently (17). Among the three ECM proteins only the Fn levels changed appreciably, increasing threefold between puberty and sexual maturity and remaining high during pregnancy and lactation. This temporal expression of Fn coincides with the expressions of hepatocyte growth factor (HGF; produced by mammary fibroblasts) (17-19), estrogen and progesterone (both shown to regulate Fn expression) (20). HGF levels rise at midpuberty and are maximal at sexual maturity and remain high during pregnancy, while estrogen is driving ductal development during puberty, and estrogen and progesterone mediate proliferation, ductal branching and alveologenesis at sexual maturity and during pregnancy (21,22). Moreover, blockade of AP-1 transcription factors, which inhibits expression of AP-1dependent genes (e.g., Fn, cyclin D, c-myc, TIMP-1, vimentin) suppresses branching and budding and reduces gland tree size and fat pad occupancy in developing MGs (23). More details of the genes involved in postnatal MG development have been provided by several, excellent review articles by Hennighausen and associates (24-26). Together these studies suggest that the temporal expression patterns of growth 147 regulatory molecules (e.g., cyclin D, c-myc, HGF, EGF), matrix macromolecules (e.g., Fn) and steroid hormones modulate mammary epithelial proliferation and morphogenesis by interacting Fn/integrin-, growth factor-, and steroid hormonesignaling pathways, which ultimately may also harbor new clues of the role of these factors, particularly Fn, in mammary cancer progression (27). In this report, we present data of the role of Fn in the postnatal development of the mammary gland, using mice with MMTV-Cre-loxP conditional knockout of Fn in the mammary epithelium (FnMEp–/–). Our whole-mount, histologic, immunohistochemical, morphometric, biochemical, and functional data show that these mice exhibit (i) moderate retardation in outgrowth and branching of the ductal tree in virgin mice, and (ii) lack of alveolar budding and lobule formation during pregnancy and lactation, resulting in inability to lactate and nurse. Both heterozygous FnMEp+/– and Fn-floxed (Fnfl+/+) control mice exhibited near normal and normal MG development and Fn expression, respectively. Decreased mammary epithelial cell (MEp) proliferative rates in FnMEp–/– mice and altered signaling via the β1 integrin (Itgb1) Fn receptor and focal adhesion kinase may have accounted, at least in part, for the observed result. MATERIAL AND METHODS Conditional Knockout of Fn in Mouse Mammary Epithelium Heterozygous, male Fnfl+/– carrying loxP sites within the 5’ untranslated region at the Msc1 site and within the first intron at the Nhe1 site of the Fn gene were generated in Dr. Fässler laboratory (Max-Planck-Institute, Martinsried, Germany) (28). These mice were backcrossed with wild type FVB mice to produce heterozygous 148 Fnfl+/– mice. The heterozygous Fnfl+/– mice were crossed with siblings to generated homozygous Fnfl+/+ mice. To introduce Cre-recombinase [an enzyme that cleaves loxP sites, thereby removing the start codon, signal sequence and the exon/intron border of exon 1 of Fn to generate the Fn-null allele 28)] that is under control of the mammary epithelial cell-specific MMTV LTR promoter into Fnfl+/+ mice, Fnfl+/+ females were paired with male MMTV-Cre mice with an FVB genetic background (supplied by Dr. Guan, University of Michigan) to ultimately obtain female mice with the genotype MMTV-Cre+ Fnfl+/+ and the phenotype FnMEp–/–. Offspring with the desired genotypes were identified by PCR analysis of tail DNA, using the same primer sets as described by Fässler and associates (28). Control mice were heterozygous FnMEp+/– and Fnfl+/+ mice. Mice were analyzed at 5- and 16-weeks of age and at day 15 of pregnancy (13 to 14 weeks of age). To assure selectivity of Cre-recombinase activity in MEp, we stained frozen sections (5 µm thick) of the MGs from offspring of homozygous MMTV-Cre and Rosa26STOPloxPLacZ mice (expression of β-galactosidase is possible only after Cremediated deletion of a stop codon flanked by loxP sites; Jackson Laboratory) with Xgal solution as described by Bonnerot et al. (29). Sections were counterstained with Nuclear Fast Red. Controls were MGs from Rosa26STOPloxPLacZ mice. To further assess organ-specificity of the MMTV-Cre-mediated Fn knockout in MGs, DNA was extracted from lung (Lu), heart (He), kidney (Ki), mammary gland (MG), salivary gland (SG), uterus (Ut) and liver (Li) of FnMEp-/- mice. PCR was performed with the primers designed from regions flanking the Fn1 lox-P sites. Fn deletion results in a 320 bp band, while floxed Fn1 revealed a 1,200 bp band. 149 Mammary Gland Whole Mount Preparation, Staining, and Analyses The left 4th inguinal mammary fat pad (MFP) was excised essentially as described by Tiran and Elson (30). In brief, excised MFPs were spread onto glass slides and fixed with Tellyesniczky’s fixative overnight. MFPs were then soaked in three changes of 100% acetone each for 6h, followed by 2h-treatments with each 100%, 95% and 70% ethanol, and stained with 0.2% carmine red overnight. After rinsing in water, MFPs were dehydrated sequentially in 50%, 70%, 95%, and 100% ethanol each for 2h, cleared with methyl salicylate overnight and examined under a dissecting microscope. MG whole mounts were evaluated as follows: (i) Extent of ductal outgrowth (maximal outgrowth beyond MFP lymph node measured in millimeters; (ii) number of ductal branch points per unit area of MFP; and (iii) number of terminal end buds (TEBs) in MGs from virgin mice (virgin mice 5- and 16-weeks old) as described by Shen et al. (23); and (iv) size and number of alveolar lobules from MGs of pregnant and lactating mice. A minimum of 7 mice per each age-category and genotype were analyzed. Histologic and Morphometric Techniques The right 4th inguinal MFP was excised as described above, spread onto glass slides and fixed with phosphate-buffered 4% paraformaldehyde overnight at 4°C. MFPs were then subjected to routine paraffin-embedding, and 4-µm thick, planar sections stained with hematoxylin & eosin (H.&E.). Sections were evaluated as follows: (i) assessment of the tissue architecture and branching pattern; and (ii) estimation of the volume density (VMG) of the glandular structures occupying the MFP as described by Weibel et al. (31) and Pauli et al. (32), using computer-enlarged photographs from entire MFP sections and superimposed test lattices of 1-cm2 150 subunits: VMG=Pi (number of line intersection points falling on MG components) / PT (total number of intersection points falling on MG components + MFP stroma) (32). A minimum of 7 mice per each age-category and genotype were analyzed. Immunohistochemistry Paraformaldehyde-fixed sections from the right 4th inguinal MFP mounted on poly-L-lysine coated glass slides were used for immunohistochemical analyzes. Prior to staining with primary antibody (mouse anti-Fn monoclonal antibody directed against the EDA domain of human Fn1: 1:50; overnight; 4°C) (Santa Cruz Biotechnology), 5-µm thick, deparaffinized sections were treated sequentially with 1% pepsin (Sigma) in 10 mM HCl (pH 2.0; 30 min; 37°C) for antigen unmasking, M.O.M. Ig blocking reagent (M.O.M. Immunodetection kit; Vector Laboratories, Burlingame, CA) for blocking endogenous immunoglobulin, and normal horse serum (R.T.U. Vectastain® Universal Elite ABC kit; Vector Laboratories) for blocking unspecific binding of the secondary antibody. Detection of bound primary antibody was by biotinylated horse anti-mouse secondary antibody, ready-to-use, stabilized ABC reagent (R.T.U. Vectastain® Universal Elite ABC kit) and the peroxidase substrate 3amino-9-ethyl carbozole (AEC; Invitrogen, Carlsbad, CA). Sections were washed vigorously with PBS between each step of this procedure and counterstained with hematoxylin (5 seconds; room temperature). Control section were stained normal mouse IgG instead of primary antibody. Assessment of MEp Proliferative and Apoptotic Rates The MEp proliferative rate was estimated by BrdU assay. In brief, mice were injected intraperitoneally with 0.1 ml/10 gm body weight of BrdU solution (Invitrogen) 2 hours prior to sacrifice. BrdU-positive cells were detected by immunohistochemistry using anti-BrdU antibody (1:150 dilution; 1.5 h; 37°C) 151 (Invitrogen) essentially as described above. The proliferative rate was expressed as the percent BrdU-positive cells of the total number of MEps counted (~1,000 cells/MG counted). The MEp apoptotic rate was determined by the cleaved caspase 3 immunohistoichemical assay (23). The anti-cleaved caspase 3 antibody was from BD Biosciences (San Jose, CA) and was used at a dilution of 1:150 as described above. The percent cleaved caspase 3-positive cells of the total number of MEps (~1,000 cell counted) was used as an expression of the MEp-apoptotic rate. Biochemical Analyses To assess the role of the downstream Fn-signaling partners in the lobulo- alveolar differentiation of the MG, MG were homogenized in lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM benzamidine chloride, 1 mM PMSF, 2 µg/ml leupeptin, 0.27 TIU/ml aprotinin, 0.1 mM sodium vanadate, and 1% Triton X-100) and incubated for 1h at 4°C (33). Total cell lysates were subjected to SDS-PAGE (~20-50µg protein) and Western blotting (33), using anti-Fn (Sigma), anti-ItgB1 (Sana Cruz Biotechnology), anti-focal adhesion kinase (FAK) (Biomol), anti-FAK(pY397) (Biomol), HRP-conjugated donkey anti-rabbit or goat anti-mouse secondary antibodies and ECL for detection of bound antibody as described (33). RESULTS MMTV-Cre Mediated Recombination Mice used in this study had the following genotypes: (i) Fnfl+/+, (ii) MMTV- Cre+Fnfl+/– (phenotype: FnMEp+/–) and (iii) MMTV-Cre+Fnfl+/+ (phenotype: FnMEp–/–). 152 These genotypes were identified by PCR using tail-derived genomic DNA and primer sets that flanked the floxed sites of the Fn gene as described in Material and Methods. The same set of primers was also used to PCR-probe DNA isolated from various organs including lung, heart, kidney, liver, uterus, mammary gland, and salivary gland in order to examine which organs exhibited Cre recombinase activity. In MMTVCre+Fnfl+/+ mice only mammary gland and salivary gland exhibited strong Cre activity as visualized by deletion of the floxed Fn-segment (320bp band) in these organs (Fig A.1A). In contrast, heterozygous MMTV-Cre+Fnfl+/– mice exhibited both Fn-species (i.e., deleted [320bp] and undeleted [1.2kb] floxed-Fn) and Fnfl+/+ mice revealed only undeleted, floxed Fn as expected (data not shown). To further assess the selectivity of Cre recombinase activity and expression in MEp, we stained frozen sections from the mammary gland of MMTV-Cre+- Rosa26STOPloxPLacZ and Rosa26STOPloxPLacZ (control) mice with X-gal. X-gal staining was only observed in the mammary epithelium from MMTV-Cre+- Rosa26STOPloxPLacZ mice, where Cre recombinase activity removed a floxed stop colon to allow expression of LacZ (Fig A.1B). No Xgal staining was observed in non-epithelial tissues. Age-matched, Cre-negative control mice revealed no X-gal staining of the mammary epithelium (Fig A.1C). To examine Fn protein expression we selected mammary gland from 15-day pregnant Fnfl+/+ and FnMEp–/– mice since at stage mammary glands of normal mice express relatively high levels of Fn (17) and since most of the MFP is occupied by glandular tissue thereby keeping the amount contributed by the stromal compartment at a minimum. As expected, Fnfl+/+ mice exhibit strong Fn expression, while Fn expression in FnMEp–/– mice varied from approximately 0 to approximately 40% (Fig A.1D). 153 Figure A.1: Conditional Knockout of Fn in Mammary Epithelia: (A) DNA was extracted from lung (Lu), heart (He), kidney (Ki), mammary gland (MG), salivary gland (SG), uterus (Ut) and liver (Li) of MMTV-Cre+Fnfl+/+ mice. PCR was performed with the primers designed from the regions flanking the Fn loxP sites. Fn deletion results in a 320bp band, while floxed Fn in a 1200bp band. (B) Frozen section from the MG of a 16-week old MMTV-Cre/Rosa26STOPloxPLacZ mouse stained with XGal solution (Fermentas) and counterstained with Fast Red. Notice the select expression of Cre-recombinase activity in MEp, but not surrounding stromal cells (C) X-Gal staining of a frozen section from the mammary gland of a 16-week old Rosa26STOPloxPLacZ control mouse. X-Gal staining is negative. (D) Fn protein expression in Fnfl+/+ and FnMEp–/–mice. 154 155 Analyses of MFP Whole Mounts Carmine red-stained MFPs from each age-group and genotypes of mice were analyzed qualitatively and quantitatively. At five weeks of age, mice with MMTVCre-mediated deletion of floxed Fn [FnMEp–/–] exhibited a significant retardation in ductal outgrowth of the MG relative to heterozygous FnMEp+/– and Fnfl+/+ mice (Figs A.2 and A.3A&B). Also decreased were the number of ductal branch points and the number of terminal end buds (TEBs) relative to the two control groups [FnMEp+/–; Fnfl+/+] (Fig A.3C&D). At 16 weeks of age, ductal outgrowth and number of branch points in FnMEp–/– mice were still decreased, but statistically indifferent from those of control FnMEp+/– and Fnfl+/+ mice (Fig A.3B&C). The most dramatic changes in MG postnatal development were observed at day 15 of pregnancy. The MG of FnMEp–/– mice exhibited normal ductal development and branching extending from the nipple to the terminal ductules (Fig A.2), but lobulo-alveolar differentiation, typically observed at this stage of pregnancy in normal mice, was totally missing (Fig A.2). Together, these data show that Fn deletion in MEp caused a minor retardation in ductal outgrowth and number of branch points in adolescent FnMEp–/– mice relative to the two age-matched control groups [FnMEp+/–; Fnfl+/+]. However, at 16 weeks of age MG ductal outgrowth and branching in FnMEp–/– mice had caught-up to those in agematched control mice (Fig A.3 B&C), suggesting that Fn-deletion might have slowed the rate of MEp proliferation. 156 Figure A.2: Mammary Gland Whole Mounts from Fnfl+/+ and FnMEp–/– mice: At 5 weeks, there is a noticeable retardation in the ductal outgrowth, decreased branching, and decreased occupancy of the MFP with glandular elements in FnMEp–/– relative to Fnfl+/+ mice. At 16 weeks (virgin mice; v16w), ductal outgrowth and branching are almost identical and there is only a minor decrease in mammary fat pad (MFP) occupancy in FnMEp–/– relative to Fnfl+/+ mice. In pregnant mice at day 15, there is a total lack in lobulo-alveolar differentiation in FnMEp–/– mice (see higher magnification) and a normal lobulo-alveolar differentiation in Fnfl+/+ mice (see higher magnification). There is a more or less identical density and branching pattern of MG ducts that extend to terminal ductules in Fnfl+/+ and FnMEp–/– mice. Bar: 1mm. No noticeable differences were observed between heterozygous FnMEp+/– and Fnfl+/+ and wild-type FVB mice (data not shown). 157 158 Figure A.3: Quantitative Aspects of the Effects of Fn-knockout on MG postnatal development: MG whole mounts were used to measure outgrowth of ductal branches beyond the fat pad lymph node (A & B) and to count the number of branch points of outgrowing ducts (C) in 5- and 16-week-old, Fnfl+/+, heterozygous FnMEp+/-, and homozygous FnMep-/- virgin mice. Also counted was the number of terminal end buds (TEBs) (D) in 5-week old mice from the three groups of mice. Notice retarded ductal outgrowth and decreased numbers of branch points and TEBs in FnMep-/- mice, relative to FnMEp+/- and Fnfl+/+ mice at 5 weeks, but no significant differences at 16 weeks. 159 160 Histological Analyses Histological examination of planar sections prepared from mammary glands of mice of the same groups complemented our findings from MG whole mounts. At 5 weeks of age, there was a slightly decreased number of ductal cross-sections in FnMEp– /– mice relative to Fnfl+/+ mice (Fig A.4 & Fig A.5A). However, this did not result in a statistically significant difference in the MG/MFP volume percent. Similar observations were made in 16-week old virgin mice. Glandular cross-sections were more frequent and there was a more pronounced branching in Fnfl+/+ relative to FnMEp– /– mice (Fig A.4 & Fig A.5B). Again, dramatic differences were only observed in 15day pregnant females. There was normal lobulo-alveolar differentiation in Fnfl+/+ mice versus a rudimentary to a total lack of lobulo-alveolar differentiation in FnMEp–/– mice (Fig A.4 & Fig A.5D&E). This observation was reflective of a significant decrease in MG/MFP volume percent in the latter group of mice (Fig A.5 C-E). No statistical differences were observed between Fnfl+/+ and FnMEp+/– mice. To examine whether the difference in the MG/MFP volume percentages between Fnfl+/+ and FnMEp–/– mice was due to changes in the rates of proliferation and apoptosis of mammary epithelial cells in FnMEp–/– mice, we performed a BrdU incorporation and cleaved caspase-3 assays. Results show a significant decrease in the rate of proliferation of MEps in FnMEp–/– mice relative to Fnfl+/+ mice (Fig A.6), but no significant differences in the rate of apoptosis (data not shown). Immunohistochemical Analyses Anti-Fn staining of mammary gland planar sections from 5 and 16-week old virgin mice of the three animal groups yielded no reportable differences. However, in 15-day pregnant mice, there were intriguing differences between Fnfl+/+ and FnMEp–/– mice. At low magnification, there was little or no Fn-staining in MG sections from 161 Figure A.4: Horizontal Section of Mammary Glands from Fnfl+/+ and FnMEp–/– Mice Stained with H.&E.: At 5 weeks, there is a decreased number of ductal crosssections in FnMEp–/– mice relative to their heterozygous counterparts. At 16 weeks (virgin 16w), MGs from FnMEp–/– mice show ductal cross-sections (D) and branching ductules (d) and no alveolar development, while MGs from Fnfl+/+ mice exhibit a higher density of ductal cross-sections and indications of alveolar differentiation (a). At 15 days of pregnancy, MGs of FnMEp–/– exhibit good ductal branching, ducts and ductules are collapsed and contain no milk, lobulo-alveolar differentiation is lacking. Newborn pups die within 2 days following birth. In contrast, Fnfl+/+ mice show extensive lobulo-alveolar differentiation and ducts (D) and ductules (d) are filled with milk. Pups develop normally. Images are depicted at identical magnifications. 162 163 Figure A.5: Mammary Gland Volume Density of the Mammary Fat Pad: Five micron thick paraffin sections were prepared from the mammary gland of 5 adolescent 5-week old mice (A), 5 adult virgin 16-week old mice (B), and 5 15-day pregnant mice (C) with the phenotypes Fnfl+/+, FnMEp+/- , and FnMep-/-. The MG volume density of the mammary fat pad was determined by standard morphometric techniques, using computer-enlarged photographs from entire MFP sections and superimposed test lattices of 1-cm2 subunits: VMG=Pi (number of line intersection points falling on MG components) divided by PT (total number of intersection points falling on MG components + MFP stroma) (150). Statistically significant differences in the volume density of MGs were only observed in 15-day pregnant mice, where the MG volume density in FnMep-/- was dramatically decreased relative to the MG volume densities in Fnfl+/+ and FnMEp+/- mice (C). This difference in MG volume densities is readily seen in H.&E.-stained sections from Fnfl+/+(D) and FnMEp+/-(E) mice. D & E Insets: Fn protein expression; actin loading control. 164 165 Figure A.6: Cell Proliferation Assay: BrdU-staining of mammary gland section from a 10-day pregnant Fnfl+/+ control mouse (A) and a 10-day pregnant FnMep-/- mouse (B), both injected 2h prior to euthanasia with 100µg BrdU/gm of BW. There is a significant decrease in the number of BrdU-positive MEps in FnMep-/- mice relative to control Fnfl+/+ mice (counts are from 200 MEps in each of 3 mice per category) (C). 166 167 FnMEp–/– mice, showing only rudimentary lobulo-alveolar differentiation (Fig A.7A), while MG sections from Fnfl+/+ mice exhibited a haze of Fn staining around small ductules and newly formed alveoli (Fig A.7B). Staining was particularly strong around small blood vessels. At high magnification at the tip of sprouting ductules, epithelial cells lost their cuboidal shape and assumed a spindle- mesenchymal-like shape as they penetrated surrounding connective tissue shape in a manner reminiscent of epithelialmesenchymal transition of invading cancer cells (Fig A.7C, arrows). At the same time, the cell body of spindle-shaped epithelial cells stained positive for Fn. Such staining was especially prominent at the tip of the invading cell. In other areas of lobuloalveolar differentiation epithelial cells separating from the tubular confines of alveoli and sprouting into neighboring tissues stain positively for Fn (Fig A.7D&E). These processes are reminiscent of budding alveoli and are accompanied by strong Fn expression. Also noticeable is a strong Fn staining reaction around newly formed blood vessels of the lobulo-alveolar compartment. Such staining was absent around mature blood vessels outside the lobule. Neovascularization of Mammary Gland Lobules Lobulo-alveolar differentiation in 15-day pregnant Fnfl+/+ mice was associated with a flurry of newly formed blood vessels that was associated with Fn staining at their periphery (Fig A.8; arrows). In contrast, age-matched FnMEp–/– mice of the same stage of pregnancy exhibited rudimentary alveolar differentiation with few blood vessels and little or no Fn-staining (Fig A.7A). To examine whether the increased neovascularization was associated with increased expression of vascular endothelial cell growth factor (Vegf) expression, we extracted total RNA from Fnfl+/+ and FnMEp–/– mice and determined Vegf expression levels relative to the housekeeping gene Gapdh, using RT-PCR. Data show a more than 8-fold decrease in Vegf expression in 168 Figure A.7: MG Immunohistochemistry: Five µm-thick sections from MGs of 10day pregnant FnMep-/- (A) and Fnfl+/+(B-E) mice were stained with anti-Fn polyclonal Ab. At low magnification (A & B), there is little or no Fn-staining in MG sections from FnMep-/- mice showing only rudimentary lobulo-aveolar differentiation (A), while MG sections from Fnfl+/+mice exhibit a haze of Fn-positive staining around small ducts and particularly within areas of early lobulo-alveolar differentiation that is especially prominent around small blood vessels (B). At higher magnification of MG sections from Fnfl+/+ mice (C-E), there are Fn-positive cells that have undergone epithelialmesenchymal transition at the tip of a sprouting terminal ductule (arrows, C), Fnpositive epithelial cells “dropping out” from the alveolar lining (arrows D) and forming new alveolar structures (arrows E). 169 170 mammary glands from 15-day pregnant FnMEp–/– mice over Fnfl+/+ mice Fnfl+/+ mice (Fig A.8). Fn, Integrin β1, and Focal Adhesion Kinase - Mediators of Lobulo-Alveolar Morphogenesis? Conditional knockout of either Fn, integrinβ1 (Itgb1) (35,36), or focal adhesion kinase (FAK) (37) in mammary epithelium appeared to have a similar effect on mammary gland postnatal morphogenesis. They all caused failure of lobuloalveolar differentiation during pregnancy. To test whether Itgb1 and FAK might act downstream of Fn, we examined expression of these proteins and in the case of FAK, the activation status, in the mammary glands of Fnfl+/+ and FnMEp–/– mice. Indeed, knockout of Fn triggered a decrease in Itgb1 expression and autophosphorylation of FAK, but no decrease in FAK protein expression (Fig A.9). DISCUSSION Fibronectin has been introduced as a multi-domain glycoprotein that regulates a variety of cellular functions including differentiation, proliferation, apoptosis, migration (1-3). As such it is involved in embryologic and postnatal development (1416), tissue homeostasis, wound healing, angiogenesis, and cancer (1-10). Here, we have analyzed the effects of conditional knockout of Fn in mammary epithelial cells on postnatal development of the mammary gland. Studies of adolescent (5 weeks) and adult virgin (16 weeks) mice of the three phenotypes Fnfl+/+, FnMEp+/–, and FnMEp–/– yielded quantitative differences in the expansion and differentiation of the ductal tree within the mammary fat pad between groups, but no significant differences in the branching pattern. Noteworthy are statistically decreased distances of branch 171 Figure A.8: Neovascularization of the Lobulo-Alveolar Compartment in Fnfl+/+ Mice: Histologic section of the lobulo-alveolar compartment of a 15-day pregnant Fnfl+/+ mouse exhibits a high vascular density (arrows), that is absent in 15-day pregnant FnMep-/- mice (see Fig. A.7A). This result is mimicked by a strong message for Vegf in MGs from Fnfl+/+ mice and a very weak Vegf message in MGs from FnMep/- mice. 172 173 Figure A.9: Fn-Signaling via Itgb1 and FAK during MG Development in 15-day Pregnant Fnfl+/+, but not FnMEp–/– Mice: MGs (4th; inguinal) were extracted in lysis buffer and subjected to Western blotting with anti-Fn, anti-Itgb1, anti-FAK, and antiFAK(pY397). Fnfl+/+ MGs exhibit strong Fn, Itgb1, and FAK(pY397) expression, while FnMep-/- MGs fail to activate FAK and have somewhat reduced Itgb1 expression. 174 175 outgrowth, numbers of branch points, and number of TEBs in 5- week old FnMEp–/– mice. These differences were still present in 16-week old FnMEp–/– mice, but were less pronounced and statistically not significant due to variability between samples asvisualized by large standard deviations in our statistical analyses, likely due to the mosaic Cre expression (34). Dramatic, statistically sound differences were observed only in pregnant mice, where the mammary glands of FnMEp–/– mice exhibited severe deficiencies in the typical lobulo-alveolar differentiation observed in the two control groups, i.e., Fnfl+/+ and FnMEp+/– mice (see Fig A.5). The degree of lobulo-alveolar impairment in FnMEp–/– mice ranged from severe hypoplasia to aplasia and was closely associated with the amount of Fn protein recovered from these glands. Failure of lobulo-alveolar differentiation resulted in lack of milk production and early death of newborn pups (<2days). Our data mimic to a large extent those reported for mice with the conditional knockout of the β1 integrin (Itgb1) (35,36) and focal adhesion kinase (FAK) (37) in the mammary epithelium. Fn-knockout in mammary epithelium, like Itgb1- (35,36) and FAK-knockout (37), did not affect the pattern of ductal outgrowth and branching, albeit both were slowed down somewhat in FnMEp–/– and FAKMEp–/– mice. In pregnant mice, deletion of Fn, Itgb1, and FAK left only a skeleton of ducts and ductules (Fig A.2) with lack of lobulo-alveolar differentiation and ensuing inability of females to produce milk and nurse their pups (35-37). These data suggest that Fn may operate proximal of Itgb1 and FAK (38). In support of this notion Fn deletion was followed by significant decreases in Itgb1 expression and autophosphorylation of FAK, while FAK protein expression remained constant. Disruption of Fn-Itgb1-FAK interactions and associated malformation of focal adhesions (35) may have accounted for the observed decreased rate of proliferation that was most prominent in 16-week old virgin and 15day pregnant FnMEp–/– mice. Such impaired proliferation has also been noticed in 176 Itgb1- and FAK-knockouts in the mammary epithelium as well as in mammary glands of mice that expressed a dominant-negative chimeric form of Itgb1 (39). Moreover, consistent with our in vivo observation, mammary epithelial cells isolated from Itgb1MEp–/– mice expressed decreased levels of tyrosine phosphorylation at residue 397 of FAK relative to mammary epithelial cells from Itgb1-floxed mice (35). In the mammary gland of Fn MEp–/– mice branching initiation, branch outgrowth, elongation and ramification seem to occur independent of Fn, albeit epithelial-mesenchymal transition (EMT), which involves expression of Fn, has been implicated to be responsible for the expansion (invasion) of the ductal tree within the mammary fat pad (40). However, our morphologic studies are more indicative of “cohort-type” of invasion of ducts into the mammary fat pad, rather than a “mesenchymal-type” of invasion (41) that involves individual spindle-shape cells of files of cells to penetrate MFP tissue. In contrast, at sites of alveologenesis epithelial cells arising from terminal ductules and primary alveolar buds seem to separate themselves from the epithelial lining and enter the mammary fat pad. These cells are spindly (mesenchymal) in shape and are strongly Fn positive and appear highly motile (42). The driving forces behind these processes are not well established, but could well be mediated by Fn itself, which has been shown to exert an EMT promoting effect (43). Differentiation of the newly created lobuloalveolar compartment could then be achieved by Jak2/Stat5a-signaling, reversing EMT and promoting differentiation of epithelial cells as shown by accompanying increased expression of E-cadherin, zonula occludens-1 and cytokeratins 8 and 18 (44). Lobulo-alveolar differentiation in the mammary gland of Fnfl+/+ mice is associated with an elaborate neovascularization process, as shown by the formation of vascular honeycombs around the alveolar structures (45) and high levels of Vegf expression. Such neovascularization is missing in FnMEp–/– mice and, correspondingly, 177 Vegf expression is low. This result may be in response to extensive, new, lobuloalveolar growth in Fnfl+/+ mice and lack thereof in and FnMEp–/– mice. However, several studies suggest a more direct involvement of Fn in Neovascularization. For example, (i) rat coronary neovascularization during prenatal and early postnatal development was preceded by the deposition of Fn suggesting that Fn may provide a primary scaffolding for the migration of primordial endothelial cells/angioblasts (46); (ii) Retinal vasculogenesis follows the spread of astrocytes into the nerve fiber layer of the retina and the deposition of Fn in the zone of vasculogenesis (47); (iii) Proliferative diabetic retinopathy, a condition which entails proliferation of microvascular endothelial cells and retinal neovascularization is associated with upregulation of Fn extra domain B (EDB) (48); (iv) Exposure of endothelial cells to EDB+-Fn or EDBpolypeptide significantly increased Vegf expression and promotes endothelial proliferation and tube formation in vivo (49). In summary, we have present novel evidence of the role of Fn in postnatal mammary gland morphogenesis. Conditional knockout of Fn in mammary epithelium leads to severe lobulo-alveolar hypoplasia/aplasia. Concomitant downregulation of Itgb1 and decrease autophosphorylation of FAK suggest that this pathology is mediated, at least in part, by disruption of Fn/Itgb1/FAK signaling pathway. This notion is supported by the fact that conditional knockouts of Itgb1 and FAK generate the same pathology as Fn-deletion (35-37). Moreover, a more that 6-fold downregulation of Vegf and significantly impaired neovascularization in FnMEp–/– mice further compromised lobulo-alveolar differentiation. 178 REFERENCES 1. Mosher DF (1989) Fibronectin. San Diego, CA: Academic Press, Inc. 2. Hynes RO (1990) Fibronectin. New York, NY: Springer-Verlag 3. Pankov P and Yamada KM (2002) J. Cell Sci., 115, 3861-3963 4. ffrench-Constant C (1995) Exp. Cell Res., 221, 261-271 5. Johansson S, Svineng G, Wennerberg K, Armulik A, and Lohikangas L (1997) Front. Biosci., 2, 126-146 6. Danen EHJ and Yamada KM (2001) J. Cell. Physiol., 189, 1-13 7. Couchman JR, Austria MR and Woods A (1990) J. Invest. Dermatol., 94, 7S14S 8. Wierzbicka-Patynowski I and Schwarzbauer JE (2003) J. Cell Sci., 116, 32693276 9. Kumar CC (1998) Signaling by integrin receptors. Oncogene, 17, 1365-1373. 10. Geiger B, Bershadsky A, Pankov R and Yamada KM (2001) Nat. Rev. Mol. Cell. Biol., 2, 793-805 11. Yamada KM and Clark RAF (1996) Provisional matrix. In: The Molecular and Cellular Biology of Wound Repair (ed. Clark RAF), pp. 51-93. New York: Plenum Press. 12. Cheng HC, Abdel-Ghany M and Pauli BU (2003) J. Biol. Chem., 278, 2460024607 13. Miyamoto S, Katz BZ, Lafrenie RM and Yamada KM (1998) Ann. NY Acad. Sci., 857, 119-129 14. George EL, George-Labouesse EN, Patel-King RS, Rayburn H and Hynes RO (1993) Development, 119, 1079-1091 15. Watt FM and Hodivala KJ (1994) Curr. Biol., 4, 270-272 16. Sakai, T., Larsen, M., and Yamada, K. M. (2003). Nature 423, 876-81 17. Haslam, S. Z., and Woodward, T. L. (2003). Breast Cancer Res 5, 208-15 179 18. Wang Y, Selden AC, Morgan N, Stamp GW and Hodgson HJ (1994) Am. J. Pathol., 144, 675-682 19. Yang Y, Spitzer E, Meyer D, Sachs M, Niemann C, Hartmann G, Weidner KM, Bichmeier C and Birchmeier W (1995). J. Cell Biol., 131, 215-226 20. Woodward TL, Mienaltowski A, Modi R, Bennet JM and Haslam SZ (2000) Endocrinol., 142, 3214-3222 21. Woodward TL, Xie JW and Haslam SZ (1998) J. Mammary Gland Biol. Neopl., 3, 117-131 22. Fendrick JL, Raafat AM and Haslam SZ (1998) J. Mammary Gland Biol. Neopl., 3, 7-22 23. Shen, Q., Zhang, Y., Uray, I. P., Hill, J. L., Kim, H. T., Lu, C., Young, M. R., Gunther, E. J., Hilsenbeck, S. G., Chodosh, L. A., Colburn, N. H., and Brown, P. H. (2006). Dev Biol 295, 589-603 24. Hennighausen L and Robinson GW (2001) Dev. Cell, 1, 467-475 25. Shillingford JM and Hennighausen L (2001) Trends Endocrinol. Metab., 12, 402-408 26. Hennighausen L and Robinson GW (2005) Nat. Rev. Mol. Cell Biol., 6, 715725 27. Labat-Robert J (2002) Cancer Biol., 12, 187-195 28. Sakai T, Johnsoin KJ, Murozono M, Sakaio K, Magnuson MA, Wieloch T, Cronberg T, Isshiki A, Erickson HP and Fässler R (2001) Nat. Med., 7, 324330 29. Bonnerot C, Nicolas JF (1993) Methods Enzymol. 225:451-469 30. Tiran Z and Elson A (2003) Meth. Enzymol., 366, 124-132 31. Weibel ER and Bolender RP (1973) Stereological Techniques for electron microscopic morphometry. In: Principles and Techniques of Electron Microscopy, (ed. Hayat MA), Vol. 3, 237-296. 32. Pauli BU, Cohen SM and Weinstein RS (1983) J. Urol., 129, 646-652 33. Abdel-Ghany M, Cheng HC, Elble RC, Pauli BU. (2002) J Biol Chem. 277, 34391-3400 180 34. White DE, Kurpios NA, Zuo D, Hassell JA, Blaess S, Mueller U, Muller WJ. (2004) Cancer Cell, 6, 159-170 35. Naylor, M. J., Li, N., Cheung, J., Lowe, E. T., Lambert, E., Marlow, R., Wang, P., Schatzmann, F., Wintermantel, T., Schuetz, G., Clarke, A. R., Mueller, U., Hynes, N. E., and Streuli, C. H. (2005) J Cell Biol 171, 717-728 36. Li, N., Zhang, Y., Naylor, M. J., Schatzmann, F., Maurer, F., Wintermantel, T., Schuetz, G., Mueller, U., Streuli, C. H., and Hynes, N. E. (2005) Embo J 24, 1942-53 37. Nagy, T., Wei, H., Shen, T. L., Peng, X., Liang, C. C., Gan, B., and Guan, J. L. (2007) J Biol Chem 282, 31766-76 38. Mitra SK, Schlaepfer DD. (2006) Curr Opin Cell Biol., 18, 516-523 39. Faraldo MM, Deugnier MA, Lukashev M, Thiery JP, Glukhova MA (1998) EMBO J. 17, 2139-2147 40. Fata JE, Werb Z, Bissell MJ. (2004) Breast Cancer Res. 6, 1-11 41. Friedl P and Wolf K. (2003) Nat. Rev. Cancer 3:362-374, 2003 42. Larsen M, Wei C, Yamada KM. (2006) J Cell Sci. 119, 3376-3384 43. Taliana L, Evans MD, Ang S, McAvoy JW. (2006) Mol Vis. 12, 1233-1242 44. Sultan AS, Brim H, Sherif ZA. (2008) Cancer Sci. 99, 272-279 45. Djonov V, Baum O, Burri PH. (2003) Cell Tissue Res. 314, 107-117 46. Rongish, B. J., Hinchman, G., Doty, M. K., Baldwin, H. S., and Tomanek, R. J. (1996) J Mol Cell Cardiol 28, 2203-15 47. Jiang, B., Liou, G. I., Behzadian, M. A., and Caldwell, R. B. (1994) J Cell Sci 107 ( Pt 9), 2499-508 48. Khan ZA, Cukiernik M, Gonder JR, Chakrabarti S. (2004) Invest Ophthalmol Vis Sci. 45, 287-295 49. Khan, Z. A., Chan, B. M., Uniyal, S., Barbin, Y. P., Farhangkhoee, H., Chen, S., and Chakrabarti, S. (2005)Angiogenesis 8, 183-96 181