Oncostatin M: origins of a pleiotropic cytokine.

Cytokines comprise the molecular communication networks that allow cells to respond to their environments. Oncostatin M belongs to the interleukin-6 family of cytokines and has diverse effects on many different cell types. In various model systems, it has been shown to influence cell cycle progression, developmental regulation, and the inflammatory process. Therefore, it is difficult to classify oncostatin M as one type of effector molecule. The origins and regulation of oncostatin M expression have not been extensively examined. The goal of this dissertation was to identify and characterize oncostatin M expression in normal human tissues. Primary human monocytes secrete oncostatin M following activation with cytokines or bacterial components. Agonist treatment of monocytes results in activation of the Janus-kinase signal transducer and activator of transcription (JAK-STAT) pathway. Specifically, STAT5b was found to bind and activate the oncostatin M promoter at a STAT binding site located -180 base pairs before the transcriptional start site. In addition, oncostatin M was produced by epithelial cells as indicated by immunohistochemistry and in situ hybridization. In contrast to monocytes, primary epithelial cells appeared to synthesize oncostatin M constitutively. Oncostatin M expression was detected only in primary cells, not in transformed cell lines. This would suggest, with previous reports, that a lack of oncostatin M expression is associated with the transformed phenotype. In summary, the origins of oncostatin M in the normal human system were found to be activated monocytes, which synthesize oncostatin M following activation, and specific types of epithelial cells, which synthesize oncostatin M constitutively.

ONCOSTATIN M: ORIGINS OF A PLEIOTROPIC CYTOKINE by Kimberly Mae Dahms A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Experimental Pathology Department of Pathology The University of Utah August 2000 Copyright © Kimberly Mae Dahms 2000 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Kimberly Mae Dahms This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. //.67l'c , i 4(z( lOO THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: I have read the dissertation of Kimbery Mae Dahms in its final form and have found that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory committee and is ready for submission to The Graduate School. Date ~Id~ Thomas M. Mel re Chair: Supervisory Committee Approved for the Major Department ~~~.~ Chair Approved for the Graduate Council -- TABLE OF CONTENTS ABSTRACT ...................................................................................................................... .iv ACKNOWLEDGMENTS ................................................................................................... v Chapters 1. INTRODUCTION ......................................................................................................... 1 2. REGULATION OF ONCOSTATIN M PRODUCTION BY PRIMARY HUMAN MONOCYTES ............................................................................................ 12 Abstract .................................................................................................................. 12 Introduction ............................................................................................................ 13 Materials and Methods ........................................................................................... 14 Results .................................................................................................................... 18 Discussion .............................................................................................................. 26 References .............................................................................................................. 32 3. EPITHELIAL CELL PRODUCTION OF ONCOSTATIN M .................................... 35 Abstract .................................................................................................................. 35 Introduction ............................................................................................................ 35 Materials and Methods ........................................................................................... 38 Results .................................................................................................................... 42 Discussion .............................................................................................................. 5 8 References .............................................................................................................. 62 4. SUMMARY AND PERSPECTIVES ........................................................................... 66 ABSTRACT Cytokines comprise the molecular communication networks that allow cells to respond to their environments. Oncostatin M belongs to the interleukin-6 family of cytokines and has diverse effects on many different cell types. In various model systems, it has been shown to influence cell cycle progression, developmental regulation, and the inflammatory process. Therefore, it is difficult to classify oncostatin M as one type of effector molecule. The origins and regulation of oncostatin M expression have not been extensively examined. The goal of this dissertation was to identify and characterize oncostatin M expression in normal human tissues. Primary human monocytes secrete oncostatin M following activation with cytokines or bacterial components. Agonist treatment of monocytes results in activation of the Janus-kinase signal transducer and activator of transcription (JAK-STA T) pathway. Specifically, ST AT5b was found to bind and activate the oncostatin M promoter at a STAT binding site located -180 base pairs before the transcriptional start site. In addition, oncostatin M was produced by epithelial cells as indicated by immunohistochemistry and in situ hybridization. In contrast to monocytes, primary epithelial cells appeared to synthesize oncostatin M constitutively. Oncostatin M expression was detected only in primary cells, not in transformed cell lines. This would suggest, with previous reports, that a lack of oncostatin M expression is associated with the transformed phenotype. In summary, the origins of oncostatin M in the normal human system were found to be activated monocytes, which synthesize oncostatin M following activation, and specific types of epithelial cells, which synthesize oncostatin M constitutively. ACKNOWLEDGMENTS My graduate school experience has been a most enjoyable one. I wish to thank my mentor, Tom McIntyre, to whom I am deeply indebted for his benevolence in sharing a place in his lab with me, and for giving me the opportunity to start traveling down the road of discovery in science. I also wish to thank my advisory committee members, Raymond Daynes, Lorise Gahring, Janis Weis, and Guy Zimmennan, for their time, guidance, and teachings. Steve Wiley and John Weis, though not on my committee, were extremely generous with both time and infonnation. Indeed, the members of Experimental Pathology have all helped me to begin my scientific education. Special thanks are due to Sean Davies and Paul Neilsen, my fellow graduate students, who gave extraordinary support, and who are two of the kindest people I know. I would also like to thank Aaron Pontsler, who patiently taught me the basics of molecular biology. ZhengMing Wang and Kurt Albertine deserve appreciation for their generosity in facilitating the immunohistochemical and in situ hybridization studies. Progression through the bureaucracy of graduate school was expertly handled by Kim Cash, and I greatly appreciate her assistance. I would like to thank Rachel Tennyson, who has been a truly wonderful friend throughout my time in graduate school. Furthennore, I wish to acknowledge and thank George Klannann, an excellent scientist, who has been both an auxiliary advisor and my best friend. Most importantly, I wish to express my greatest thanks and appreciation to my parents, who have supported me in every way possible throughout this experience, and who always be the most superb mentors. CHAPTER 1 INTRODUCTION Cytokines Cytokines compose the information networks that allow cells to sample and respond to their environments. They are important for both development and regulation of many human systems. Disregulation of the cytokine network causes or exacerbates a variety of pathologies, including asthma, septicemia, acute/adult respiratory distress syndrome, and atherosclerosis. To halt this disregulation therapeutically, the cytokine interactions must be identified. Therefore, it is important to understand the cytokine production and response systems because of their critical role in the regulation of inflammatory and immune systems. Discovery of Oncostatin M Oncostatin M was discovered in 1986 in the search for anti-tumor therapies (1). It is secreted by a phorbol ester differentiated promyelocytic cell line and by phytohemagglutinin activated T cells (2). The purification of oncostatin M, as well as its name, is based on its ability to inhibit the growth of A375 melanoma cells, and it was originally suggested that oncostatin M might be used to inhibit the growth of transformed cells. However, subsequent research has shown that its effects are far more widespread than originally expected. Structure Following the discovery of oncostatin M, research focused on identification of oncostatin M complementary DNA (eDNA), genomic coding sequence, and the basic 2 structure of the protein. The cDNA for oncostatin M contains 1839 base pairs. Oncostatin M is encoded by three exons and the gene is located on human chromosome 22q12 (3). The precursor protein form of oncostatin M is 252 amino acids. The 25 N­terminal amino acids comprise a signal peptide that is cleaved during the secretory process (4, 5) (Figure 1.1). Thirty-one amino acids are cleaved from the C-terminus to yield the mature oncostatin M protein, which is approximately 28 kilo Daltons (kDa). The mature secreted form is active as a monomer and has one N-linked glycosylation site (1, 4, 6, 7). In addition, the protein has two disulfide bonds, only one of which is required for activity (8). The structure of oncostatin M is predicted to be a four a.-helix bundle, growth-hormone-like structure, which it shares with the interleukin-6 (lL-6) family members (9). Oncostatin M belongs to the IL-6 family of cytokines which includes, with IL-6 and oncostatin M, cardiotrophin-I (CT-I), interleukin-II (lL-II), leukemia inhibitory factor (LIF), and ciliary neurotrophic factor (CNTF) (9, 10). These cytokines all share a common receptor subunit, gpl30 (glycoprotein 130 kDa) (11, 12). gp130 in combination with an a. or ~ receptor subunit comprise more specific receptor complexes. These subunits preferentially bind certain cytokines, but multiple cytokines have been shown to bind to each unit. For example, oncostatin M can bind to a complex of oncostatin M receptor ~ subunit (OSMR~) and gp 130, or the LIF receptor J3subunit (LIFRJ3) and gp130, or gp130 alone (Figure 1.2) (11, 13, 14). Signal Transduction The JAK-STA T Pathway The oncostatin M receptor complexes are associated with a number of different signal transduction pathways. The Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathways are the predominant signal transduction systems activated 3 Exon 1 Exon 2 Exon3 Genomic: Protein: signal peptide mature peptide C-terminal peptide Figure 1.1: Schematic representation of oncostatin M genomic region and protein. Oncostatin M is comprised of three exons, which encode a 252 amino acid precursor. A 25 amino acid precursor peptide is cleaved from the N-terminus, and a 31 amino acid peptide is cleaved from the C-terminus, resulting in the mature 1 96 amino acid protein. .., low affinity complex LlFR[3 OSMR[3 OSMRII .... _----...... OSMRI", high affinity complexes Figure 1.2: Oncostatin M receptor complexes. Oncostatin M can bind to gp 130 alone or in complex with LIFRJ3 (low affinity complex) or OSMRJj (high affinity complex). The individual subunits have been found to be specifically expressed on different cell types, conferring oncostatin M responsiveness in a tissue specific fashion. 4 following IL-6 cytokine family receptor activation (Figure 1.3). There are currently three known JAK family members: JAK1, JAK2, and TYK2. These proteins are activated by tyrosine phosphorylation following interaction with an active receptor subunit. The JAKs in turn activate STAT family members, of which there are currently seven members: STAT1, STAT2, STAT3, STAT4, STAT5a, STAT5b, and STAT6. Following phosphorylation, ST A Ts dimerize and translocate to the nucleus where they bind to specific sites in the promoter regions of responsive genes (10, 15). STAT bindings sites are also known as gamma interferon activated sites, or GAS, elements. The consensus sequence of these elements is TTNNNNNAA. The middle and flanking nuc1eotides are hypothesized to provide specificity for individual STAT family members (16). Active STATs can induce synthesis of JAK-STAT pathway inhibitors that allow the pathways to be deactivated (17-19). Oncostatin M has been shown to activate STATs 1,3, and 5 (14, 20). Activated Receptor Pathways In addition to activation of the JAK-STAT pathways, oncostatin M activates other signal transducers. General levels of tyrosine phosphorylation increase in cells following oncostatin M treatment (21), and JAK-ST AT activation causes stimulation of multiple mitogen activated protein kinase (MAPK) pathway members (22, 23). The connection between MAPK family members and active JAKs may be through JAKlgp130 association with the adapter molecules, Grb2 and She. These adapters connect to the MAPK pathway proteins (24, 25). In addition, it was recently reported that oncostatin M alters peroxisome proliferator-activated receptor (PPAR) activity (26, 27). These signal transduction cascades regulate transcription of multiple gene families. Therefore, the effects of oncostatin M may be more diverse than those currently identified. ) ligand binds to receptor complex STATs are phosphorylated and activated STATs dimerize and translocate STA Ts tind to responsive gene promoters and activate transcription ~~:A:~1--+-F--- (nucleus) 5 Figure 1.3: Schematic representation of the JAK-ST AT pathway. After ligand binding, receptor activation results in JAK phosphorylation and activation. Active JAKs phosphorylate ST A Ts, which dimerize and translocate to the nucleus. Within the nucleus, ST A Ts bind to specitic sequences in the promoters of responsive genes resulting in the activation of transcription. Cytostatic Effects Known Effects of Oncostatin M Cell Cycle Regulation 6 Oncostatin M was identified and named based on its ability to inhibit the growth of the A375 melanoma cell line (l, 2). Oncostatin M also induces differentiation or cytostasis of myeloid cell lineages, breast cancer cell lines, ovarian cancer, hepatocytes, and glioma cell lines (28-32). These reports show the desirable effects of oncostatin M as a potential anti-tumor agent. Mitogenic Effects. In 1986 Nair et aL discovered that oncostatin M s effects were not limited to cell cycle inhibition. They found oncostatin M acts as an autocrine growth factor for Kaposi s sarcoma derived cells (33, 34). In addition, oncostatin M induces mitogenesis of smooth muscle cells, and inhibits the differentiation of embryonic stem cells and hematopoietic cells in vitro (31, 35, 36). Recently, oncostatin M was identified as an angiogenic factor. It induces human microvascular endothelial cells to grow and form vessel-like structures in culture (37, 38). Therefore, based on these reports, oncostatin M does not ubiquitously function as a cytostatic agent. Regulation of Inflammation Inflammatory Effects Subcutaneous injection of recombinant oncostatin M in mice produces localized inflammation, characterized at the microscopic level by leukocyte intiltrates and tissue destruction. In vitro, oncostatin M induces endothelial cells to express markers of the inflammatory response. Endothelial cells have high levels of oncostatin M binding activity and express both gp 130 and OSMR~ receptor subunits (unpublished observations). Primary human umbilical vein endothelial cells express the adhesion molecule E-selectin and secrete interleukin-6 within four hours of exposure to oncostatin 7 M. Both of these responses are associated with leukocyte recruitment and adhesion to endothelial cells (39). Anti-inflammatory Effects Oncostatin M upregulates matrix protease inhibitor expression and stimulates glycosaminoglycan synthesis in epithelial cells and fibroblasts. Inhibition of matrix proteases and glycosaminoglycan production may interfere with leukocyte extravasation in the early stages of inflammation, or it may begin the resolution of inflammation. In addition, oncostatin M treatment of primary mononuclear cells reduces the amount of tumor necrosis factor (TNF) they release following lipopolysaccharide (LPS) exposure (40). These reports present oncostatin M as an anti-inflammatory cytokine. In summary, oncostatin M's effects depend upon the cell type in contact with the cytokine. Origins of Oncostatin M Most of the research on oncostatin M is devoted to identification of its effects on various cell types. Its origins have not been as intensely studied. The goal of this thesis project was to determine where oncostatin M is produced in normal human systems. It was logical to examine primary leukocytes, as oncostatin M was purified from a phorbol ester differentiated promyeloid cell line, and also from T cells. Of the primary human leukocytes examined in this project, oncostatin M was expressed at the highest levels by activated monocytes. Oncostatin M mRNA was produced by monocytes only after activation with bacterial components or cytokines. The expression of oncostatin M was controlled by ST A TSb binding to a STAT binding site in the oncostatin M promoter region. In addition, fully differentiated columnar epithelium fron1 colon, cornea, and bronchioles also expressed oncostatin M. The presence of oncostatin M mRNA in these cells was confirmed by in situ hybridiztion. None of the transformed epithelial cells examined made detectable levels of oncostatin M. However, primary human bronchial 8 epithelial cells did synthesize oncostatin M mRNA. It is possible that oncostatin M expression has been lost in transformed epithelial cells, but that in normal epithelial cells oncostatin M is associated with or responsible for cytostasis. Summary The data in this thesis, in combination with the reports of others, present a complex picture of the regulation of oncostatin M. Oncostatin M is synthesized following myeloid cell interaction with cytokines or bacterial products. In addition, oncostatin M is present in epithelial cells, as indicated by immunohistochemistry and in situ hybridization, where it may function in cytostasis or in recruitment of leukocytes to a site of damaged epithelium. References 1. Zarling, 1. M., M. Shoyab, H. Marquardt, M. B. Hanson, M. N. Lioubin, and G. J. Todaro. 1986. Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells. Proc Natl Acad Sci USA 83:9739. 2. Brown, T. 1., M. N. Lioubin, and H. Marquardt. 1987. Purification and characterization of cytostatic lymphokines produced by activated human T lymphocytes. Synergistic antiproliferative activity of transforming growth factor beta 1, interferon­gamma, and oncostatin M for human melanoma cells. J Immunol 139:2977. 3. Rose, T. M., M. 1. Lagrou, I. Fransson, B. Werelius, O. Delattre, G. Thomas, P. 1. De long, G. 1. Todaro, and 1. P. Dumanski. 1993. The Genes for Onocstatin M (OSM) and Leukemia Inhibitory Facotr (LIF) Are Tightly Linked on Human Chromosome 22. Genomicf) 17: 136. 4. Malik, N., 1. C. Kallestad, N. L. Gunderson, S. D. Austin, M. G. Neubauer, V. Ochs, H. Marquardt, 1. M. Zarling, M. Shoyab, C. M. Wei, and et al. 1989. Molecular cloning, sequence analysis, and functional expression of a novel growth regulator, oncostatin M. Mol Cell BioI 9:2847. 5. Linsley, P. S., 1. Kallestad, V. Ochs, and M. Neubauer. 1990. Cleavage of a hydrophilic C-terminal domain increases growth-inhibitory activity of oncostatin M. !vIol Cell BioI 10: 1882. 6. Linlsey, P. S., 1. Kallestad, V. Ochs, and M. Neubauer. 1990. Cleavage of a hydrophilic C-terminal domain increases growth-inhibitory activity of oncostatin Nt. /\tfol Cell BioI 10:1882. 7. Malik, N., C. Clegg, and M. Shoyab. 1992. Cloning and expression of the bovine homologue of the cytokine oncostatin M. FASEB J 6:A 1671. 9 8. Kallestad, J. C., M. Shoyab, and P. S. Linsley. 1991. Disulfide bond assignment and identification of regions required for functional activity of oncostatin M. J Bioi Chern 266:8940. 9. Rose, T. M., and A. G. Bruce. 1991. Oncostatin M is a member of a cytokine family that includes leukemia- inhibitory factor, granulocyte colony-stimulating factor, and interleukin 6. Proc Natl A cad Sci USA 88:8641. 10. Heinrich, P. C., I. Behrmann, G. Muller-Newen, F. Schaper, and L. Graeve. 1998. Interleukin-6-type cytokine signalling through the gp 130/JaklST AT pathway. Biochern J 334:297. 11. Gearing, D. P., M. R. Comeau, D. J. Friend, S. D. Gimpel, C. J. Thut, J. McGourty, K. K. Brasher, 1. A. King, S. Gillis, B. Mosley, and et al. 1992. The IL-6 signal transducer, gp 130: an oncostatin M receptor and affinity converter for the LIF receptor. Science 255: 1434. 12. Linsley, P. S., M. Bolton-Hanson, D. Hom, N. Malik, 1. C. Kallestad, V. Ochs, J. M. Zarling, and M. Shoyab. 1989. Identification and characterization of cellular receptors for the growth regulator, oncostatin M. J BioI Chem 264:4282. 13. Mosley, B., C. De Imus, D. Friend, N. Boiani, B. Thoma, L. S. Park, and D. Cosman. 1996. Dual Oncostatin M (OSM) Receptors. Cloning and characterization of an alternative signaling subunit conferring osm-specific receptor activation. J BioI Chem 271:32635. 14. Auguste, P., C. Guillet, M. Fourcin, C. Olivier, J. Veziers, A. Pouplard­Barthelaix, and H. Gascan. 1997. Signaling of type II oncostatin M receptor. J Bioi Chern 272: 15760. 15. Heim, M. H. 1999. The Jak-ST A T pathway: cytokine signalling from the receptor to the nucleus. J Recept Signal Transduct Res 19:75. 16. O'Shea, J. J. 1997. Jaks, STATs, Cytokine Signal Transduction, and Immunoregulation: Are We There Yet? Immunity 7: 1. 17. Starr, R., T. A. Willson, E. M. Viney, L. J. Murray, J. R. Rayner, B. J. Jenkins, T. 1. Gonda, W. S. Alexander, D. Metcalf, N. A. Nicola, and D. 1. Hilton. 1997. A family of cytokine-inducible inhibitors of signalling. Nature 387:917. 18. Naka, T., M. Narazaki, M. Hirata, T. Matsumoto, S. Nlinamoto, A. Aono, N. Nishilnoto, T. Kajita, T. Taga, K. Yoshizaki, S. Akira, and T. Kishimoto. 1997. Structure and function of a new STAT -induced STAT inhibitor. Nature 387:924. 19. Starr, R., and D. J. Hilton. 1999. Negative regulation of the JAKJSTA T pathway. Bioess(J}'s 21:47. 20. Faris, M., B. Ensoli, N. Stahl, G. Yancopoulos, A. Nguyen, S. Wang, and A. Nel. 1996. Differential activation of the extracellular signal-regulated kinase, Jun kinase and Janus kinase-Stat pathways by oncostatin M and basic fibroblast growth factor in AIDS-derived Kaposi's sarcoma cells. Aids 10:369. 10 21. Schieven, G. L., J. C. Kallestad, T. J. Brown, J. A. Ledbetter, and P. S. Linsley. 1992. Oncostatin M induces tyrosine phosphorylation in endothelial cells and activation ofp62yes tyrosine kinase. J ImmunoI149:1676. 22. Amaral, M. C., S. Miles, G. Kumar, and A. E. Nel. 1993. Oncostatin-M stimulates tyrosine protein phosphorylation in parallel with the activation of p42MAPKIERK-2 in Kaposi's cells. Evidence that this pathway is important in Kaposi cell growth. J Clin Invest 92:848. 23. Yin, T., and Y. C. Yang. 1994. Mitogen-activated protein kinases and ribosomal S6 protein kinases are involved in signaling pathways shared by interleukin-ll, interleukin-6, leukemia inhibitory factor, and oncostatin M in mouse 3T3-L 1 cells. J Bioi Chem 269:3731. 24. Chauhan, D., S. M. Kharbanda, A. Ogata, M. Urashima, D. Frank, N. Malik, D. W. Kufe, and K. C. Anderson. 1995. Oncostatin M induces association of Grb2 with Janus kinase JAK2 in multiple myeloma cells. J Exp Med 182:1801. 25. Giordano, V., G. De Falco, R. Chiari, I. Quinto, P. G. Pelicci, L. Bartholomew, P. Delmastro, M. Gadina, and G. Scala. 1997. Shc mediates IL-6 signaling by interacting with gp 130 and Jak2 kinase. J Immunol 158:4097. 26. Zhou, Y. C., and D. J. Waxman. 1999. STAT5b down-regulates peroxisome proliferator-activated receptor alpha transcription by inhibition of ligand-independent activation function region-l trans-activation domain [In Process Citation]. J Bioi Chem 274:29874. 27. Zhou, Y. C., and D. J. Waxman. 1999. Cross-talk between janus kinase-signal transducer and activator of transcription (JAK-ST A T) and peroxisome proliferator­activated receptor- alpha (PPARalpha) signaling pathways. Growth honnone inhibition of pparalpha transcriptional activity mediated by stat5b. J Bioi Chem 274:2672. 28. Bruce, A. G., I. H. Hoggatt, and T. M. Rose. 1992. Oncostatin M is a differentiation factor for myeloid leukemia cells. J Immunol 149: 12 71. 29. Halfter, H., R. Lotfi, R. Westermann, P. Young, E. B. Ringelstein, and F. T. Stogbauer. 1998. Inhibition of growth and induction of differentiation of glioma cell lines by oncostatin M (OSM). Growth Factors 15:135. 30. Grant, S. L., and C. G. Begley. 1999. The oncostatin M signalling pathway: reversing the neoplastic phenotype? Mol Med Today 5:406. 31. Kinoshita, T., T. Sekiguchi, M. J. Xu, Y. Ito, A. Kamiya, K. Tsuji, T. Nakahata, and A. Ivtiyajima. 1999. Hepatic differentiation induced by oncostatin M attenuates fetal liver hematopoiesis. Proc Natl A cad Sci USA 96: 7265. 32. Douglas, A. I'v1., S. L. Grant, G. A. Goss, D. R. Clouston, R. L. Sutherland, and C. G. Begley. 1998. Oncostatin M induces the differentiation of breast cancer cells. lnt J Cancer 75:64. 33. Nair, B. C., A. L. DeVico, S. Nakamura, T. D. Copeland, Y. Chen, A. Patel, T. O'NeiL S. Oroszlan, R. C. Gallo, and M. G. Samgadharan. 1992. Identification of a major growth factor for AIDS-Kaposi's sarcoma cells as oncostatin M.Science 255:1430. 11 34. Cai, J., P. S. Gill, R. Masood, P. Chandrasoma, B. Jung, R. E. Law, and S. F. Radka. 1994. Oncostatin-M is an autocrine growth factor in Kaposi's sarcoma. Am J Pathol 145: 74. 35. Grove, R. 1., C. Eberhardt, S. Abid, C. Mazzucco, J. Liu, P. Kiener, G. Todaro, and M. Shoyab. 1993. Oncostatin M is a mitogen for rabbit vascular smooth muscle cells. Proe lVatl Aead Sei USA 90:823. 36. Rose, T. M., D. M. Weiford, N. L. Gunderson, and A. G. Bruce. 1994. Oncostatin M (OSM) inhibits the differentiation of pluripotent embryonic stem cells in vitro. Cytokine 6:48. 37. Vasse, M., J. Pourtau, V. Trochon, M. Muraine, J. P. Vannier, H. Lu, J. Soria, and C. Soria. 1999. Oncostatin M induces angiogenesis in vitro and in vivo. Arterioseler Thromb Vase BioI 19:1835. 38. Wijelath, S., B. Carlsen, T. Cole, J. Chen, S. Kothari, and W. P. Hammond. 1997. Oncostatin M induces basic fibroblast growth factor expression in endothelial cells and promotes endothelial cell proliferation, migration and spindle morphology. J Cell Sci 110:871. 39. Modur, V., M. J. Feldhaus, A. S. Weyrich, D. L. Jicha, S. M. Prescott, G. A. Zimmerman, and T. M. McIntyre. 1997. Oncostatin M is a proinflammatory mediator. In vivo effects correlate with endothelial cell expression of inflammatory cytokines and adhesion molecules. J Clin Invest 100: 158. 40. Wallace, P. M., J. F. MacMaster, K. A. Rouleau, T. J. Brown, J. K. Loy, K. L. Donaldson, and A. F. Wahl. 1999. Regulation of inflammatory responses by oncostatin M. J ImmunoI162:5547. CHAPTER 2 REGULATION OF ONCOST A TIN M EXPRESSION BY PRIMARY HUMAN MONOCYTES VIA STAT5B Abstract The basic inflammatory response contributes to a wide variety of pathologies including atherogenesis, adult/acute respiratory distress syndrome, and septicemia. One of the many effects of the cytokine oncostatin M is induction of the endothelial pro inflammatory response, and immunohistochemistry shows it to be present in leukocytes infiltrating chronically inflamed arterial walls. This study demonstrates that oncostatin M is predominantly produced by primary human monocytes stimulated with the cytokines interleukin-3, granulocyte macrophage-colony stimulating factor, and the endotoxin lipopolysaccharide. In contrast, secretion of oncostatin M by monocytes was not stimulated in response to interleukin-5, leukemia inhibitory factor, ciliary neurotrophic factor, or interleukin-6, which are recognized by cellular receptors that share a common subunit with that of oncostatin M. Analysis of the human oncostatin M promoter region in primary human monocytes showed the importance of a STAT-like binding site -180 base pairs from the putative transcription start site. Gel shift analysis confirmed transcription factor binding to this sequence in cytokine-activated monocytes. and supershift analysis identified the transcription factor as ST AT5b. The STAT -like site at -100 bp, important for regulating oncostatin M expression in n1urine cells, was not responsible for the induction of oncostatin M expression in human leukocytes. The finding that monocytes secrete oncostatin M following stimulation with int1ammatory 13 agonists, together with its known effects on endothelial cells, indicates the involvelllent of this cytokine in the inflammatory response. Introduction Cytokine expression in the inflammatory response is temporally and spatially regulated by the presence of pathogens and by other cytokines. The interleukin-6 (lL-6) family of cytokines is involved in both inflammatory and immune responses. This family of cytokines includes oncostatin M (OSM), leukemia inhibitory factor (LIF), ciliary neurotrophic factor (CNTF), interleukin-11 (IL-11), and cardiotrophin-1 (CT -1) (1). Collectively and individually these cytokines have truly pleiotropic effects, including activation of expression of genes leading to both pro- and anti-inflammatory effects, cellular differentiation, and mitogenesis (2-4). Oncostatin M was originally purified from a promyelocytic cell line that was forced to differentiate into a macrophage-like cell type by phorbol ester treatment (5), and also from T cells activated with phytohemagglutanin for an extended period (6). It was found to act as an anti-tumor agent, inhibiting growth of the A375 melanoma cell line. Subsequently, it was found to activate proinflammatory properties of endothelial cells, inducing them to express adhesion molecules and inflammatory cytokines (7). Immunohistochemistry for oncostatin M showed that it is coexpressed with TNFa by cells infiltrating a chronically inflamed human aorta (7). This observation raised the question of the origin of oncostatin M expression in the inflammatory environment, and the agonists responsible for induction of its expression. Recently, it was reported to be produced by human peripheral blood monocytes following lipopolysaccharide treatment (8). However~ it is unlikely that lipopolysaccharide was the inducing agent in the chronically intlamed aorta, because the tissue was not from a septic patient. Therefore~ agonists other than endotoxin are likely to be responsible for induction of oncostatin N1 expression at sites of chronic innammation. 14 This study shows the ability of a variety of cytokines, as well as endotoxin, to induce oncostatin M expression by primary human monocytes and macrophages. The oncostatin M promoter region was examined by deletion analysis and site directed mutagenesis. A STAT binding site 180 base pairs before the transcriptional start site was responsible for activity of the human oncostatin M promoter. This site specifically bound ST A T5b as shown an electrophoretic mobility supershift assay. These findings demonstrate that oncostatin M is synthesized by monocytes in response to paracrine signaling factors, in addition to endotoxin, and characterize an important transcriptional control mechanism. Materials and Methods Reagents Ml99 was fronl Bio\\lhittaker (Walkersville, MD), human serum albumin was a product of the American Red Cross (Washington, DC). Cytokines, monoclonal anti­oncostatin M (MAB295) and affinity purified biotinylated polyclonal anti-oncostatin M (AF-295-NA) antibodies were obtained from R&D Systems (Minneapolis, MN). Lipopolysaccharide (LPS) from Escherichia coli strain 0111 :B4 and protease inhibitors for nuclear extractions, were obtained from Sigma (St. Louis, MO), except for Pefabloc SC, which was purchased from Boehringer Mannheim (Indianapolis, IN). Oligonucleotides used for electrophoretic mobility shift assays were synthesized by Operon (Alameda, CA), and all supershifting antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Leukocyte Isolation and Culture Human peripheral blood monocytes were isolated fronl healthy donors as previously described (9). rv1ononuclear cells were separated from whole blood by density gradient centrifugation, and monocytes were purified by countercurrent elutriation. Platelets were isolated using the method of Hamburger and McEver (10). In briet~ human 15 blood was drawn into acid-citrate-dextrose (ACD) (7 ml ACD/42 ml blood) and was centrifuged (200xg for 20 minutes) to obtain platelet-rich plasma. Platelet-rich plasma was recentrifuged (500xg for 20 minutes), the supernatant discarded, and the platelet pellet was resuspended in Ca2+ and Mg2+ free Hank s balanced salt solution (HBSS) (Bio Whittaker). Neutrophils were isolated from venous blood using dextran sedimentation and Ficoll-Hypaque gradients as described (11). PMN were suspended in HBSS with 5 mg/ml human serum albumin (HSA). Oncostatin M Protein Secretion Assays Purified leukocytes were incubated at a density of 5x 106 cells/ml in M 199 supplemented with 2% human serum albumin with or without the stated agonists for 20 hours, at which time supernatants were collected. Cells were maintained in suspension by rocking at 37°C unless otherwise stated. Secreted oncostatin M was detected by sandwich enzyme linked immunosorbant assay (ELISA) using a monoclonal antibody for capture and an affinity purified-biotinylated polyclonal antibody for detection. Specificity of these antibodies for oncostatin M was confirmed by western blotting (data not shown). Detection of Oncost at in M Messenger RNA Total RNA was isolated from cells using TRIzol (Gibco-BRL, Grand Island, NY) as per the manufacturer's directions. Complementary DNA (cDNA) was generated from PMA treated U937 cell RNA using an oligo dT primer and Moloney murine leukemia virus (MML V) reverse transcriptase (Gibco BRL). The cDNA was then subjected to peR using the following primers for oncostatin M: "forward" primer 5'­CGGAATTCATATGGGGGTACTGCTCACAC- 3', "reverse" pnmer 5'­CGGAATTCTGCTCTCGAGGCT ACCG-3'. PCR products were resolved by electrophoresis on a 1% agarose gel and visualized by ethidiun1 bromide staining. Human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers used as a positive 16 control for RNA/cDNA loading were: 5 -TTTCT AGACGGCAGGTCAGG-3 , and 5 - CCCATGGCAAATTCCATGGC-3. Primers were designed to amplify sections of different exons to prevent detection of genomic contaminants. Primers were synthesized by the University of Utah DNA/peptide resource facility. The PCR was 35 cycles of the following progran1: 53°C anneal, 72°C extend, and 95°C denature. Oncostatin M Promoter Reporter Constructs The genomic 5' flanking region was obtained from cosmid N5E 11 (a kind gift of Dr. Jan Dumanski, Karolinska Hospital, S-1 040 1 Stockholm, Sweden). The region was sequenced by the University of Utah Sequencing Core Facility. Oncostatin M promoter­luciferase reporter plasmids were created by the insertion of 1754 base pairs of the human oncostatin M promoter upstream of the luciferase gene in the Promega's pGL3 reporter vector (Madison, WI). Restriction sites were engineered into the -180 and -1 00 STAT binding sites, to create the oncostatin M promoter site directed mutants. The mutagenized promoters were cloned into the pGL3 reporter vector. The -180 site was changed from TTCGAAGAA to AGATCTGAA. The -100 site was changed from TTCCCAGAA to GCGGCCGCA. Reporter plasmids were transfected into primary human monocytes with the Biolistic PDS-I000/He Particle Delivery System (Bio-Rad Laboratories, Hercules, CA). Transfections were performed as per the manufacturer's preliminary instruction manual with vacuum at 25 mmHg, a rupture disk of 1550 PSI on level 1.6 J.lM gold particles, 1 J.lg of DNA per 5x 106 monocytes. Monocytes were plated on 35 mm culture dishes, allowed to settle from the media for 30 minutes. A majority of the cell-free media was aspirated, the culture dish positioned at level two, and the cells were transfected by "tiring" the PDS-I000/He. The cells were resuspended in fresh media, with or without 500 ng/mL of IL-3 or LPS and incubated for 18 hours. Luciferase production was measured with Promega's Dual-Luciferase Assay System 17 (E 191 0, Madison, WI) and the ML3000 Microtiter Plate Luminometer (Dynatech Laboratories, Chantilly, VA). ST ATSb Western Blot Monocyte nuclear extracts were made as previously described (12). Briefly, Sx 106 suspended primary human monocytes were treated for 30 minutes with the stated agonists. The cells were pelleted by centrifugation at 3S0xg, resuspended in 1 ml cold PBS and centrifuged again. The cell pellet was resuspended in 1 ml buffer "A" (10 mM HEPES pH 7.9, 1.S mM MgCh, 10 mM KCl, 1 mM dithiothreotal (DTT), 1 mg/ml Pefabloc, 10 Jig/ml aprotinin, 10 Jig/mlleupeptin, 0.1 % NP40), and rocked at 4°C for 10 nlinutes. Nuclei were pelleted by centrifugation at 1900xg at 4°C for S minutes. The nuclear pellet was resuspended in buffer "B" (20 mM HEPES pH 7.9, 2S% glycerol, 420 mM NaCI, 1.S mM MgCh, 0.2 mM EDTA, 1 mM DTT, 1 mg/ml Pefabloc, 10 Jig/ml aprotinin, 10 Jig/ml leupeptin) and incubated on ice for 30 minutes, vortexing every 10 minutes. Cellular membranes were pelleted from the nuclear extracts by centrifugation at 20,000xg at 4°C for 10 minutes. The resulting supernatants (nuclear extracts) were stored at -70°C until use. Nuclear extract proteins (10 Jig) were electrophoresed on a 12% SDS-polyacrylamide gel, that was transferred to Immobilon ™ membrane. The membrane was probed with anti-ST ATSb (as per Santa Cruz western blot recommendations ). Electrophoretic Mobility Shift Assay and Supershift Nuclear extracts were made as described for the western blot procedure, above. For electrophoretic mobility shift assays, 3 Jig of total nuclear extract protein were incubated with SO,OOO cpm 32p labeled oligonucleotide with or without stated antibodies in binding buffer (2 Jig poly dr-dC, 10 mM Tris HCl pH 7.S, SO mM KCl, 1 mM DTT, 200 Jig/ml BSA, and lS% glycerol). Protein-DNA complexes were resolved by electrophoresis on a S% nondenaturing Tris-Borate-EDTA (TBE) polyacrylamide gel, 18 and visualized by phosphorimagery (Molecular Dynamics, Sunnyvale, CA). For supershift assays, 2 to 8 Jlg of anti-ST A T5b (N-20, Santa Cruz) antibody were added to the binding reactions. For multiple antibody supershift assays, 6 Jlg of each antibody were added to the binding reactions. Results Human Monocytes and Monocyte Derived Macrophages Secrete Oncostatin M Oncostatin M was originally purified from phorbol ester differentiated, n1acrophage-like U937 cells, and previous observations from this laboratory demonstrate that oncostatin M is present in infiltrating leukocytes that have properties consistent with a macrophage phenotype (7). Experiments in this study examined the ability of isolated human monocytes and macrophages to synthesize and secrete oncostatin M as a function of time. Monocytes were purified by elutriation to avoid an adhesion dependent purification step that can induce numeroils changes in these cells (13). The purified monocytes were treated in fresh serum-free medium supplemented with human albumin (2%) and either 500 ng/ml lipopolysaccharide, interleukin-3 (IL-3), or granulocyte colony-macrophage stimulating factor (GM-CSF). After 20 hours in culture in the presence or absence of these agonists, supernatants were assayed for oncostatin M by sandwich ELISA. Incubation of the monocytes in serum free media avoided antibody cross reaction with serum components. Monocytes secreted a small amount of oncostatin M during a 20-hour period when maintained in suspension or when allowed to adhere to tissue culture wells (Figure 2.1). Freshly-isolated suspended monocytes treated with lipopolysaccharide secreted five times the amount of oncostatin M produced by untreated cells, while adherent monocytes at the initial stage of differentiation into macrophages increased their oncostatin M secretion by less than 1 fold. The maximal oncostatin M secretion induced by IL-3 and GM-CSF was only half the maximal lipopolysaccharide induced oncostatin M secretion. 2 1.75 -....J -fn .s 1.2 ~ c: '';::; C'!l 1i) 0 0 c: 0 Days in Culture: 1 '-v--" suspension 19 o media -LPS IIIL-3 SGM-CSF 1 4 8 12 16 - adherent Figure 2.1: Cytokines and endotoxin induce oncostatin M production by cultured human monocytes, Monocytes, isolated by elutration, were cultured in suspension (day 1 only) or on tissue culture plates for 16 days. Each day, 20 hours prior to sample collection, culture media containing serum was removed and fresh, serum free, M199/2% HSA was added to the cells (such that the cell density was approximately 5xl06 cells/ml). Graphed values indicate mean oncostatin M production values ± standard deviation derived from triplicate samples, following 20 hours of monocyte incubation with: media alone (white bars), lipopolysaccharide (LPS, black bars), interleukin-3 (IL-3, cross-hatched bars), or granulocyte macrophage-colony stimulating factor (GM-CSF, striped bars). All agonists were at 200 ng/ml. 10 J.lg/ml polymyxin B sulfate was included in the media of all samples except those with LPS. Supernatants were collected and oncostatin M concentration was determined by ELISA. 20 Culturing monocytes on tissue culture plastic allowed them to differentiate into macrophages. This differentiation resulted in a profound change in the pattern of oncostatin M secretion. By 8 days in culture, oncostatin M secretion by untreated monocytes decreased to undetectable levels. Monocyte stimulation with cytokines (IL-3 and GM-CSF), which initially resulted in a doubling of oncostatin M secretion compared to non-activated monocytes, also declined to insignificant levels after the cells were in culture for 8 days. Nevertheless, lipopolysaccharide remained a powerful agonist for oncostatin M secretion throughout the 16 days of culture. These data show that cytokine­stimulated oncostatin M secretion is primarily a function of monocytes and nascent monocyte-derived macrophages. In contrast, lipopolysaccharide is a strong agonist for oncostatin M secretion by both monocytes and monocyte-derived macrophages. Other leukocytes were examined for their ability to synthesize oncostatin M (7). Neither ELISA, nor RT -PCR detected oncostatin M production by activated primary human lymphocytes, platelets, neutrophils as great as that of monocytes. Either these leukocytes were unable to synthesize oncostatin M in these experimental conditions or an appropriate agonist was not found (data not shown). It was necessary to establish that cytokine induction of oncostatin M synthesis was not due to endotoxin contamination, given the extreme sensitivity of monocytes to endotoxin. This was accomplished in two ways. The cytokines IL-3 and GM-CSF induced maximal synthesis of oncostatin M by monocytes at approximately 100 ng/ml (Figure 2.2A). Neither interleukin-6, an oncostatin M family member that shares the gp 130 component of the oncostatin M receptor, nor another family member, ciliary neurotrophic factor, induced oncostatin M secretion. Additionally, interleukin-5 was incapable of inducing oncostatin M synthesis and secretion, in spite of the fact it shares a common receptor subunit with IL-3 and GM-CSF. The maximum oncostatin M synthesis did not exceed 1.2 ng/ml, regardless of additional cytokine. 1 ng/ml to 100 ng/ml of lipopolysaccharide were able to induce a n1aximal oncostatin M secretion of 3 ng/ml, 21 double that achieved with cytokine activation (Figure 2.2B). Since the maximum amount of oncostatin M synthesis induced by cytokines was less than half that affected by lipopolysaccharide activation, it is unlikely endotoxin contamination of the cytokines is responsible for the induction of oncostatin M secretion in cytokine treated cells. This observation was supported by adding 10 Jlg/ml of polymyxin B, which inactivates lipopolysaccharide, to the monocyte cultures. There was no alteration in oncostatin M production upon addition of polymyxin B to the cytokine-monocyte cultures (data not shown). Inducible oncostatin M synthesis by monocytes was controlled at the transcriptionalleve1. Oncostatin M messager RNA was synthesized by monocytes only after 60 minutes of cytokine or lipopolysaccharide stimulation, as detected by RT -PCR. No oncostatin M message was reliably detected in unactivated cells (Figure 2.2C). Others have shown that y-interferon has a priming effect on monocytes, causing them to increase their activation responses to stimuli such as endotoxin or cytokines (14). However, y-interferon coincubation or preincubation did not reproducibly alter oncostatin M production by cytokine or endotoxin stimulated monocytes (data not shown). The Oncostatin M Promoter Requires a 242 Base Pair Region for Activity The genomic region 1754 base pairs (bp) 5' to the transcription start site was obtained and sequenced as described in ItMethods. U Generation of promoter deletion reporter constructs allowed definition of essential promoter regions that regulate oncostatin M production. A 242 base pair region 5 of the transcription start site conferred inducibility in response to both IL-3 and to lipopolysaccharide. Regions 5' to this area were completely dispensable (Figure 2.3). Two potential STAT binding sites, likely regulatory targets as both IL-3 and GM-CSF activate multiple members of the JAK-STAT pathways, were located -100 and -180 base pairs 5' to the transcriptional start sequence (15). Each site was mutagenized separately (Figure 2.4A). Disruption of the 22 A 1.4 -e-IL-3 1.2 -ll-GM-CSF ~ 1 0') c: -0.8 :2 c: ~0.6 8 g 0.4 0.2 IL-5,IL-6, 0 LlF, CNTFO 10.4 10.3 10-2 10.1 10° 101 102 103 B 3 cytokine (ng/ml) 2.5 i0 ') 2 S ~ c: 1.5 +-c=o' en 00 1 c: 0 0.5 10-1 10° 101 102 103 LPS (nglml) Figure 2.2: Primary human monocytes make oncostatin M in response to both bacterial components and cytokines. Monocytes were treated for 18 hours in suspension with various cytokines or lipopolysaccharide (LPS). Interleukin-3 (IL-3, circles), granulocyte macrophage-colony stimulating factor (GM-CSF, squares) (shown in graph A), or lipopolysaccharide (LPS, squares) (shown in graph B) were incubated in monocyte cultures at the indicated concentrations. Supernatants were analyzed for oncostatin M content by ELISA, as described in ItMaterials and Methods." The error bars indicate mean values ± standard devation of the triplicate sample values. These data are representative of three experiments. (C) Monocyte oncostatin M mRNA production was determined by RT-PCR. Monocytes were incubated with media, LPS, IL-3, GM-CSF, or oncostatin M (OSM) for the times indicated. Each agonist was at a concentration of 500 ng/ml. RNA was purified, transcribed to cDNA and subjected to PCR for oncostatin M or GAPDH (indicated under "Primers:") as described in "Materials and Methods." GAPDH amplification was used as a positive control to show equal RNA loading. PCR reactions that did not contain cDNA did not amplify products (data not shown). These results are representative of three experiments. 23 c Agonist: o· 20' 60· 120' Primers: OSM media _ ......... GAPDH . .. - OSM LPS • ---- GAPDH OSM IL-3 ---- GAPDH -. OSM GM-CSF , . --_.- GAPDH ... OSM OSM ... ~ GAPDH -1754 STAT binding sites -1BO -100 "'/ a media -LPS lSI1L-3 600 BOO Relative Light Units 24 Figure 2.3: The oncostatin M promoter region 262 base pairs 5' to the transcriptional start site is essential for pronl0ter activity. 1754 base pairs of the genomic region 5' to the oncostatin M gene were cloned into the pGL3 firefly luciferase reporter vector. Deletions of the promoter region were made by restriction digestion, followed by ligation. Primary human monocytes, were transfrected by particle bombardment as described in "Materials and Methods." Firefly luciferase reporter plasmids were cotransfected with cytomegalovirus promoter driven Renilla luciferase reporter plasmids (1 Jlg each) into 5x 1 06 monocytes/35mm dishltransfection. Following transfection, monocytes were cultured in suspension for 18 hours with or without agonists: media alone (media, white bars), lipopolysaccharide at 500 ng/ml (LPS, black bars), or interleukin-3 at 500 ng/ml (IL-3, striped bars). Following incubation, cell lysates were assayed for firefly luciferase production and normalized to Renilla luciferase production. Normalized values are reported as mean ± standard error of the mean. These data are representative of two experiments. 25 A STAT binding site mutations -1754 /~ -180 altered: .. 100 altered: From: TTCGAAGAA TTCCCAGAA To: AGATCTGAA GCGGCCGCA B 1 -180 -100 sites mutated Figure 2.4: The -180 oncostatin M promoter STAT binding site is essential for inducible promoter activity. STA T binding sites were altered by PCR based mutagenesis as described in "Materials and Methods" (A). Monocytes were transfected with the luciferase reporter vector, pGL3 containing no promoter insert (no promoter), the -1754 5' genomic region (full promoter), the full promoter with the -180 STAT binding site mutagenized (-180), or the full promoter with the -1 00 STAT binding site mutagenized. The reporter plasmids were transfected into monocytes as previously described. Transfected monocytes were treated with interleukin-3 (IL-3) at 500 ng/ml for 18 hours in suspension cultures. Cells were lysed and luciferase production was determined. Firefly luciferase luminesence values were normalized to cotransfected Renilla luciferase luminesence values and reported as Relative Light Units (B). Bars indicate mean values ± standard error of the mean. Data are representative of three experiments. 26 -180 site abolished IL-3 dependent transcription, but disruption of the -100 site resulted in only a modest decrease in stimulation (Figure 2.4B). Electrophoretic mobility shift assay (EMSA) showed an oligonucleotide corresponding to the -180 site interacted with factors present in the nuclei of activated monocytes. These factors were absent in unactivated monocytes. Corresponding to its lack of effect in the reporter assay, the oligonucleotide based on the -100 site formed fewer complexes with nuclear material from activated cells as indicated by decreased band intensity in an EMSA (Figure 2.5). Formation of shifted complexes with each oligonucleotide was not abolished by an excess of irrelevant oligonucleotide, but was disrupted by the presence of a specific oligonucleotide competitor. Complex formation did not occur when oligonucleotides based on the mutagenized STAT sites were used as targets in the EMSA (data not shown). ST A T5B Binds the -180 STAT-like Binding Site Because both IL-3 and GM-CSF activate multiple members of the STAT family (15), it was necessary to identify the transcription factor in activated monocytes that bound to the -180 STAT -like binding site. Western blotting showed ST A T5b was inducibly translocated to the nucleus following IL-3 treatment of monocytes for 30 minutes (Figure 2.6). In addition to IL-3, GM-CSF and oncostatin M itself induce STAT5b binding to the -180 oligonucleotide (Figure 2.7). STATl, STAT3, STAT5a, and ST A T5b were examined for binding to the -180 STAT binding site oligonucleotide by electrophoretic mobility supershift assay. Only the anti-ST A T5b antibody produced a supershifted complex, showing that ST A T5b binds to the essential -180 site in the oncostatin M promoter region (Figure 2.8). Discussion The purification and nomenclature of oncostatin M was based on its ability to inhibit the growth of the A375 melanoma cell line. Subsequently, oncostatin M has been STAT Binding Site Oligonucleotide: agonist: shifted oligo ~ unbound oligo ,..,. unlabeled competitor: 27 -180 -100 ( __ ----A------~ (_ _ ----A-----~\ media IL-3 media IL-3 ~~~~ .,. + + Figure 2.5: The -180 ST AT binding site binds more activated monocyte nuclear extract protein than the -100 STAT binding site. Nuclear extracts from monocytes activated with interleukin-3 (IL-3) at 500 nglml for 30 minutes, were incubated with 32p labeled 23mer oligonucleotides corresponding to the -180 or the -100 STAT binding sites. Protein­DNA complexes were resolved on a 4% TBE native acrylamide gel. A 50-fold molar excess of the oligonucleotide used in the binding reaction was added to show specificity of the shifted complex (ItUnlabeled competitorlt). western blot: anti .. STAT5b monocyte vehicle IL-3 treatment: _ ~ STAT5b ~ non-specific 28 Figure 2.6. Interleukin-3 induced ST A T5b nuclear translocation in primary human monocytes. Monocytes were activated with interleukin-3 (IL-3) at 200 ng/ml for 30 minutes. Nuclear extracts were made as described in Materials and- Methods. Ten Jl g of nuclear extract protein was electrophoresed on a 12% SDS polyacrylamide gel and transferred to Immobilon membrane. The membrane was probed with polyclonal anti­ST A T5b, as per the manufacturers recommendations (Santa Cruz). Anti-ST A T5b binding was detected with a horse radish peroxidase labeled secondary antibody visualized by enhanced chemiluminesence and autoradiography. 29 agonists: buffer IL-3 GM-CSF OSM bsuapsiecr sshhiifftt .... •• ~• ~·."'. ..·ti••l .... ' tIIIfi· , ,:;,:. ,.~ - <, - , anti-STAT5b (lJg): 2 8 2 8 2 8 2 8 cuonmlapbeetilteodr : + + + + Figure 2.7. Interleukin-3 (IL-3), granulocyte macrophage-colony stimulating factor (GM-CSF), and oncostatin M (OSM) induce STAT5b binding to the -180 oncostatin M promoter STAT binding site. Monocytes were treated for 30 minutes with 200 ng/ml IL- 3, GM-CSF, or OSM before nuclear extracts were made as described in the "Materials and Methods. II Binding reactions were prepared as in Figure 2.6, with the addition of 2 Jlg or 8 Jlg of anti-ST AT5b antibody. Reactions were resolved on a 4% polyacrylamide, non-denaturing, TBE gel. Unlabeled -180 oligonucleotide competitor was added to show the specificity of the shifted com.pl', _e x (u_nlabe led competitor) . .i~ supershift ~ .. basic shift ~ 1 3 5a 5b \.. J ---------------~-------- anti-STAT antibodies no antibody unlabeled competitor oligonucleotide Figure 2.8. STAT5b, and not STATl, 3, or 5a, binds to the -180 STAT binding site. Monocytes were activated with interleukin-3 (IL-3) at 200 ng/ml for 30 minutes. Nuclear extracts and electrophoretic mobility shift assay were performed as described in "Materials and Methods." Six Jlg of STAT 1, STAT3, STAT5a, or STAT5b antibodies were added to the binding reactions prior to resolution on a 4% polyacrylamide non­denaturing, TBE gel. Fifty-fold molar excess of the -180 oligonucleotide was added to the binding reaction to show specificity of the shifted complex. 30 shown to be involved in many physiological processes, including differentiation of myeloid cells (16, 17) and T cell development in an athymic environment (18). Transgenic overexpression of bovine oncostatin M in mice, under the control of a thymus specific promoter, leads to a lethal autoimmune phenotype and dramatically altered lymphoid tissue development (19). In addition, oncostatin M induces cytokine, protease inhibitor, and proteoglycan expression by a variety of different cell types (7, 19-29). Oncostatin M is, therefore, a pluripotent immune cell modulator. However, despite progress in defining its effects, knowledge about its production in humans is limited. Given the complexity and diversity of effects that oncostatin M has on different types of cells and in different systems, characterization of leukocyte production of oncostatin M would clarity which human tissues are actually exposed to oncostatin M. The cytokines IL-3 and GM-CSF were able to induce equivalent levels of oncostatin M secretion by monocytes, and oncostatin M itself was able to induce production of its own mRNA. Attempts to measure oncostatin M induced protein production by immunoprecipitation from metabolically labeled cells were not successfuL IL-3 and GM­CSF share a common receptor subunit with interleukin-5, however, interleukin-5 was unable to induce oncostatin M secretion. Similarly, interleukin-6 and ciliary neurotrophic factor share the common gp130 subunit with oncostatin M, yet oncostatin M production following treatment with either of these cytokines was not detected. These data indicate the importance of the cytokines multiple receptor subunits in specific signal transduction and gene expression. Monocyte adhesion to tissue culture plastic decreased the oncostatin M secretion induced by lipopolysaccharide, but does not affect the cytokine induced secretion in the first 24 hours of culture (Day 1, Figure 2.1). Adhesion to tissue culture plastic may acti vate antagonistic signal transduction pathways in monocytes that diminish lipopolysaccharide induced signaling, but do not interfere with cytokine induced signal 31 transduction. For this reason, the experiments in this study use monocytes isolated by elutriation to avoid the effects that adhesion to plastic has on the cells. There are precedents for adhesion-based modulation of monocyte activation. Monocyte integrin interaction following adhesion to plastic leads to calcium fluctuations and cytoskeletal rearrangement (13, 30) and multiple cytokine mRNA levels are altered (31). Therefore, some change in the expression of oncostatin M following monocyte adhesion to plastic was not unexpected. In addition, there was a major decrease, over time, in the ability of cytokines to induce oncostatin M secretion. The phenotypic changes monocytes undergo in the process of becoming macrophages in culture must account for this observation, but this was not due to the loss of the capacity to synthesize oncostatin M as lipopolysaccharide induced oncostatin M expression did not exhibit the same decrease as the cytokine treated cells. Differences in oncostatin M expression by cytokine or LPS treated adherent and non adherent cells may be due to the modulation of other pathways in addition to the JAK-STATSb pathway, or differential regulation of the JAK-STATSb system. Both IL-3 and GM-CSF activate multiple members of the JAK-STAT pathway, including: STAT1, STAT3, ST ATSa, STATSb, or heterodimers of these (15). Therefore, identification of ST A TSb as the sole transcription factor that binds to the -180 site indicates the specificity in activation and subsequent binding of STAT family members. The observation that STATSb binds to the -180 binding site as opposed to the -100 site in the human oncostatin M promoter is different than that found in the murine model of oncostatin M expression, which showed the -100 site was the promoter target site for STAT binding (32). One explanation for this could be that differences exist between the DNA binding sites of the murine STA TSb and human ST A TSb. According to Lin et aI., however, there is only one amino acid difference in the DNA binding domain of human and mouse ST A TSb (33), It is unknown whether this difference in ST A TSb between the two species could lead to a significant difference in STA TSb 32 binding affinity. Any differences in binding must be in the transcription factor itself, or the signaling pathway leading to its activation, as the sequences of the murine and the human -180 and -100 STAT binding sites are identical. Based on the complex nature of oncostatin M's effects, identification of the signal transduction pathways activated in monocytes in response to cytokine treatment, and understanding the oncostatin M promoter regulation, will help determine where and when oncostatin M could be expressed in humans in vivo. Oncostatin M is truly a pleiotropic cytokine whose physiological roles must be elucidated in order to comprehend its widespread effects and its potential as a therapeutic agent. References 1. Rose, T. M., and A. G. Bruce. 1991. Oncostatin M is a member of a cytokine family that includes leukemia- inhibitory factor, granulocyte colony-stimulating factor, and interleukin 6. Proc Natl A cad Sci USA 88:8641. 2. Gadient, R. A., and P. H. Patterson. 1999. Leukemia inhibitory factor, Interleukin 6, and other cytokines using the G P 130 transducing receptor: roles in inflammation and injury. Stem Cells 17: 127. 3. Wolvekamp, M. C., and R. L. Marquet. 1990. Interleukin-6: historical background, genetics and biological significance. Immunol Lett 24: 1. 4. Hirano, T. 1992. Interleukin-6 and its relation to inflammation and disease. Clin Immunol Immunopathol 62:S60. 5. Zarling, J. M., M. Shoyab, H. Marquardt, M. B. Hanson, M. N. Lioubin, and G. J. Todaro. 1986. Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells. Proc Nat! A cad Sci USA 83:9739. 6. Brown, T. J., M. N. Lioubin, and H. Marquardt. 1987. Purification and characterization of cytostatic lymphokines produced by activated human T lymphocytes. Synergistic antiproliferative activity of transforming growth factor beta 1, interferon­gamma, and oncostatin M for human melanoma cells. J Immunol 139:2977. 7. Modur, V., M. J. Feldhaus, A. S. Weyrich, D. L. Jicha, S. M. Prescott, G. A. Zimn1erman, and T. M. McIntyre. 1997. Oncostatin M is a proinflammatory mediator. In vivo effects correlate with endothelial cell expression of inflammatory cytokines and adhesion molecules. J Clin Invest 100: 158. 8. Wallace, P. M., J. F. Macrv1aster, K. A. Rouleau, T. J. Brown, J. K. Loy, K. L. Donaldson, and A. F. Wahl. 1999. Regulation of inflammatory responses by oncostatin M. J Immunol 162:5547. 33 9. Elstad, M. R., S. M. Prescott, T. M. McIntyre, and G. A. Zimmerman. 1988. Synthesis and release of platelet-activating factor by stimulated human mononuclear phagocytes [published erratum appears in J Immunol 1988 Jul 1;141(1):352]. J Immunol 140:1618. 10. Hamburger, S. A., and R. P. McEver. 1990. GMP-140 mediates adhesion of stimulated platelets to neutrophils. Blood 75:550. 11. Zimmerman, G. A., A. D. Renzetti, and H. R. Hill. 1984. Granulocyte adherence in pulmonary and systemic arterial blood samples from patients with adult respiratory distress syndrome. Am Rev Respir Dis 129: 798. 12. Weyrich, A. S., T. M. McIntyre, R. P. McEver, S. M. Prescott, and G. A. Zimmerman. 1995. Monocyte tethering by P-selectin regulates monocyte chemotactic protein- 1 and tumor necrosis factor-alpha secretion. Signal integration and NF- kappa B translocation [see comments]. J Clin Invest 95:2297. 13. Rosales, C., and R. L. Juliano. 1995. Signal transduction by cell adhesion receptors in leukocytes. J Leukoc BioI 57: 189. 14. Speert, D. P. 1992. Macrophages in bacterial infection. In The Macrophage. C. E. Lewis, and J. O. D. McGee, eds. IRL Press, New York, p. 222. 15. Kirito, K., M. Uchida, M. Yamada, Y. Miura, and N. Komatsu. 1997. A distinct function of STAT proteins in erythropoietin signal transduction. J BioI Chem 272: 16507. 16. Bruce, A. G., 1. H. Hoggatt, and T. M. Rose. 1992. Oncostatin M is a differentiation factor for myeloid leukemia cells. J Immunol 149: 12 71. 17. Tanigawa, T., N. Nicola, G. A. McArthur, A. Strasser, and C. G. Begley. 1995. Differential regulation of macrophage differentiation in response to leukemia inhibitory factor/oncostatin-M/interleukin-6: the effect of enforced expression of the SCL transcription factor. Blood 85:379. 18. Clegg, C. H., J. T. Rulffes, P. M. Wallace, and H. S. Haugen. 1996. Regulation of an extrathymic T-cell development pathway by oncostatin M. Nature 384:261. 19. Malik, N., B. Evans, B. W. Greenfield, R. A. Shapiro, M. Hanson, and M. Shoyab. 1995. Autocrine/paracrine induction of tissue inhibitor of metalloproteinase- 1 in Chinese hamster ovary cells by oncostatin M. Matrix Bioi 14:677. 20. Boutten, A., P. Venembre, N. Seta, J. Hamelin, M. Aubier, G. Durand, and M. S. Dehoux. 1998. Oncostatin M is a potent stimulator of alpha1-antitrypsin secretion in lung epithelial cells: modulation by transforming growth factor-beta and interferon-gamma. Am J Respir Cell Mol BioI 18:511. 21. Cichy, J., J. Potempa, R. K. Chawla, andJ. Travis. 1995. Regulation of alpha 1- antichymotrypsin synthesis in cells of epithelial origin. FEBS Lett 359:262. 22. Cichy, J., J. Potempa, R. K. Chawla, and J. Travis. 1995. Stimulatory effect of inflammatory cytokines on alpha 1- antichymotrypsin expression in human lung-derived epithelial cells. J Clin Invest 95:2729. 34 23. Cichy, 1., S. Rose-John, and J. Travis. 1998. Oncostatin M, leukaemia-inhibitory factor and interleukin 6 trigger different effects on alpha1-proteinase inhibitor synthesis in human lung-derived epithelial cells. Biochem J 329:335. 24. Duncan, M. R., A. Hasan, and B. Berman. 1995. Oncostatin M stimulates collagen and glycosaminoglycan production by cultured normal dermal fibroblasts: insensitivity of sclerodermal and keloidal fibroblasts. J Invest Dermatol 104: 128. 25. Nemoto, 0., H. Yamada, M. Mukaida, and M. Shimmei. 1996. Stimulation of TIMP-1 production by oncostatin M in human articular cartilage [see comments]. Arthritis Rheum 39:560. 26. Richards, C. D., T. 1. Brown, M. Shoyab, H. Baumann, and J. Gauldie. 1992. Recombinant oncostatin M stimulates the production of acute phase proteins in HepG2 cells and rat primary hepatocytes in vitro. J ImmunoI148:1731. 27. Richards, C. D., M. Shoyab, T. J. Brown, and J. Gauldie. 1993. Selective regulation of metalloproteinase inhibitor (TIMP-1) by oncostatin M in fibroblasts in culture. J Immunol 150:5596. 28. Richards, C. D., C. Kerr, M. Tanaka, T. Hara, A. Miyajima, D. Pennica, F. Botelho, and C. M. Langdon. 1997. Regulation of tissue inhibitor of metalloproteinase-1 in fibroblasts and acute phase proteins in hepatocytes in vitro by mouse oncostatin M, cardiotrophin-1, and IL-6. J Immunol 159:2431. 29. Sallenave, J. M., G. M. Tremblay, J. Gauldie, and C. D. Richards. 1997. Oncostatin M, but not interleukin-6 or leukemia inhibitory factor, stimulates expression of alpha1-proteinase inhibitor in A549 human alveolar epithelial cells. J Interferon Cytokine Res 17:337. 30. Lin, T. H., C. Rosales, K. Mondal, J. B. Bolen, S. Haskill, and R. L. Juliano. 1995. Integrin-mediated tyrosine phosphorylation and cytokine message induction in monocytic cells. A possible signaling role for the Syk tyrosine kinase. J BioI Chem 270:16189. 31. Sporn, S. A., D. F. Eierman, C. E. Johnson, 1. Morris, G. Martin, M. Ladner, and S. Haskill. 1990. Monocyte adherence results in selective induction of novel genes sharing homology with mediators of inflammation and tissue repair. J Immunol 144:4434. 32. Yoshimura, A., M. Ichihara, I. Kinjyo, M. Moriyama, N. G. Copeland, D. J. Gilbert, N. A. Jenkins, T. Hara, and A. Miyajima. 1996. Mouse oncostatin M: an immediate early gene induced by multiple cytokines through the JAK-STAT5 pathway. EmboJ 15:1055. 33. Lin, J. X., J. Mietz, W. S. Modi, S. John, and W. J. Leonard. 1996. Cloning of human Stat5B. Reconstitution of interleukin-2-induced Stat5A and Stat5B DNA binding activity in COS-7 cells. J BioI Chem 271:10738. CHAPTER 3 ONCOST A TIN M EXPRESSION BY DIFFERENTIATED EPITHELIAL CELLS Abstract Epithelial cells produce a variety of cytokines. Some of these cytokines facilitate leukocyte recruitment following epithelial damage, while others are thought to prevent excessive damage to the epithelial layer during pathogenesis. Oncostatin M is a cytokine that has the capacity to do both. It has previously been characterized as a cytokine produced by activated leukocytes. Data in this study demonstrate constitutive oncostatin M production by several types of epithelium, including corneal, bronchial, and intestinal epithelial cells by immunohistochemistry and in situ hybridization. This expression is specific to differentiated epithelial cells and primary cells in culture, as transformed cell lines do not express detectable oncostatin M message. Oncostatin M expression varies with maturation, position, and type of epithelial cell. Columnar bronchial epithelial cells express oncostatin M while proximal alveolar epithelial cells appear to have lower levels of the protein. These data, in conjunction with other reports, suggest that oncostatin M is associated with differentiated cell cycle arrest in certain epithelial cell types. In addition, in the event of epithelial damage, oncostatin M may be released leading to the recruitment of leukocytes and the induction of repair mechanisms. Introduction Oncostatin M (OSM) is a 28 kDa protein that belongs to the pleiotropic interleukin-6 (IL-6) family of cytokines (1, 2). Family members include, in addition to oncostatin M and IL-6, leukemia inhibitory factor (LIF), ciliary neurotrophic factor 36 (CNTF), interleukin-ll (IL-ll), and cardiotrophin-l (CT-l) (3). The members share a four a-helix bundle predicted structural motif, and modular receptor components. The receptors for these cytokines contain one constant receptor subunit, gp 130, usually in combination with a second cytokine-specific receptor subunit (4-9). There is some promiscuity between receptor subunits and cytokine binding. Oncostatin M can bind to gp130 alone, with low affinity, or complexes of gpI30-0SMR~ or gpI30-LIFR~, with high affinity (2, 3). The oncostatin M receptor components are expressed in various degrees and combinations on a variety of cell types including lung epithelial cells, hepatocytes, fibroblasts, and leukocytes (4, 8, 10, 11). Leukocyte Recruitment Oncostatin M can induce endothelial cells to produce hallmarks of the inflammatory response, specifically to express adhesion molecules and cytokine expression (12), each of which lead to leukocyte recruitment from the blood stream and subsequent activation (13). Recently, it has been reported that oncostatin M induces angiogenesis in human microvascular cells, an important step toward recovery following pathogenic damage to a vascularized region (14, 15). Therefore, oncostatin M can modulate the inflammatory response, and promote vascular repair. Epithelium The oncostatin M receptor components are expressed at comparatively high levels on various types of epithelial cells (4, 8). In response to oncostatin M alveolar epithelium and fibroblasts produce protease inhibitors (16, 17). Oncostatin M has similar effects on synovial fibroblasts and can induce dermal fibroblasts to make collagen and glycosaminoglycans (18-21), Cumulatively, these data indicate epithelial cells bind and respond to oncostatin M in a manner that activates systems designed to reduce and recover from extracellular matrix damage. 37 Growth Arrest Oncostatin M is also involved in differentiation and cell cycle arrest. Its original purification, both from phorbol ester (PMA) differentiated U937s and from phytohemagglutinin (PHA) treated T cells, was based on its ability to inhibit the growth of A375 melanoma cells (22, 23). Recently, part of the molecular mechanism for this inhibition has been identified. Oncostatin M activates OIG37, p21, and p27/Kipl in certain cell types, leading to growth arrest. These factors are thought to block cell cycle progression by inhibiting the cyclins or the cyclin-dependent kinases (24-26). Thus, both epithelial and endothelial cells have the capacity to respond to oncostatin M. In addition, oncostatin M has the ability to activate endothelium to recruit leukocytes, as well as induce angiogenesis following damage. It induces epithelial cells and fibroblasts to generate and protect extracellular matrix components, and modulates progression through mitosis in several different cell types. This information suggests oncostatin M may act in a constitutive manner on differentiated cells to promote cell cycle arrest, while an acute release of the protein leads to activation of the inflammatory response and the initiation of its resolution. Sources of Oncostatin M To date, the known cellular sources of oncostatin M are activated leukocytes primary cells and cell lines, and Leydig cells (3, 27). In leukocytes, previous studies have shown its mRNA is synthesized only after stimulation with cytokines, bacterial products, or chemical agonists (22, 23)(Chapter 2). Data in this study demonstrate that oncostatin M is constitutively synthesized by several types of columnar epithelial cells. Oncostatin M production was detected in situ and in primary bronchial epithelial cells in culture, but not in transformed epithelial cell lines. Thus, differentiated epithelial cells may express oncostatin M in conjunction with cell cycle arrest. In addition, to having a cytostatic function, oncostatin M may induce 38 leukocyte recruitment to a region of epithelial damage and initiate damage protection and repair mechanisms. Materials and Methods Immunohistochemistry for Oncostatin M Human tissue specimens were collected at the University of Utah, Salt Lake City, Utah, and at LDS Hospital, Salt Lake City, Utah. Rabbit corneas were a kind gift of Dr. Heidi Bazan, Louisiana State University Eye Center, New Orleans, USA. Five J.1m sections were deparaffinized with Americlear and rehydrated with serial ethanol/H20, rinsed in distilled water and then treated with Citra (HK068-9K, Biogenex, San Ramon, CA) as per the manufacturers instructions for antigen retrieval. Sections were permeabilized and endogenous peroxidases were quenched with methanol (20% solution, M1775, Sigma, St. Louis, MO), H20 2 (3%, H-I009, Sigma), and Triton X-IOO (1%, X- 100, Sigma). After blocking with 5% rabbit serum for 30 minutes at room temperature (S-5000, Vector, Burlingame, CA), the tissue was incubated with 333 ng/ml polyclonal, affinity purified goat anti-oncostatin M (AF-495-NA, R&D Systems, Minneapolis, MN) or isotype matched nonspecific goat antibody (HK406-5G, Biogenex), was incubated on the tissue overnight at 4°C. For anti-STAT5b immunohistochemistry, 5% goat serum was used to block nonspecific antibody binding. One J.1g1ml biotin labeled rabbit anti­goat (BA-500, Vector) or goat anti-rabbit (BA-IOOO, Vector) antibodies were incubated on the sections at room temperature for 50 minutes. Vector's ABC kit (PK-6100) and diaminobenzidine tetrahydrochloride (DAB, D-4293, Sigma), VIP (SK-4600, Vector), or NovaRed (SK-4800, Vector) were used for chromogenic detection. Sections were counterstained with hematoxylin (Gill s No.3, Sigma). Detection of Oncost at in M mRNA in situ The oncostatin M coding sequence to be used as a template for the RNA probe was obtained by RT-PCR from a U937 cDNA library, and was cloned into pCR 2.1 39 (K2000-0I, Invitrogen, Carlsbad, CA) under control of the T7 promoter. Sense and anti­sense mRNA oncostatin M probes were DIG labeled with the Genius 4 kit (Boehringer Mannheim, Indianapolis, IN). Five ~m thick sections were deparaffined in Americ1ear (C4200-l, Scientific Products), and rehydrated with serial ethanol dilutions (100%, 95%, 70%) and DEPC treated water. The sections were treated with proteinase K (Boehringer Mannheim) at 6 ).lg/ml for 90 minutes at room temperature then rinsed with DEPC treated water. Subsequently, the sections were rinsed in 0.2x SSC buffer with gentle stirring for 10 minutes at room temperature and dehydrated in 70%/95%/100% ethanol baths (1 minute incubation in each dilution). The sections were air dried for 30 minutes and then exposed to the hybridization mixture (buffer: 40% deionized formamide, 40/0 dextran sulfate, 0.8x Denhardt's solution, 80mM DTT, 16mM Tris HCl, pH 7.5, 4mM EDTA, 240 mM NaCl; 500 ).lglml tRNA, 1 ng/~ll oncostatin M probe). Sections were incubated at 85°C for 10 minutes to denature RNA then hybridized at 50°C for 16 hours. The sections were rinsed with 4x SSCIl OmM DTT for 1 hour at room temperature, incubated in 5% ultrapure formamide/2x SSC/20mM DTT for 30 minutes at 50°C, and then rinsed with Ix NTE buffer (lOOmM Tris-HCl, pH 7.4, 5M NaCl, 10mM EDTA) for 15 minutes at 37°C. The sections were then washed in 2x SSC for 5 minutes at room temperature, 0.1 x SSC for 15 minutes at room temperature, buffer 1 (1 OOmM maleic acid, 150mM NaCl) for 5 minutes at room temperature, and blocked in buffer 2 (50/0 sheep serum, 0.3% Triton-X 100, diluted into buffer 1) for 30 minutes at room temperature. aDIG-alkaline phosphatase conjugate or aDIG-peroxidase conjugates (Boehringer Mannheim) at 1 :500 dilution in buffer 2, was incubated on the sections for 2 hours at room temperature. The slides were washed with buffer 1, and subsequently washed with buffer 3 (100 mM Tris­HCl pH 9.5, 100 mM NaCl, 50 mM MgCb). Antibody was detected with either 3,3'­DAB substrate solution or Vector VIP ™ substrate solution. Enzymatic color development was arrested with 10 mM Tris-HCI pH 8.0, 1 mM EDTA. Sections were counterstained with hematoxylin or methyl green (H-3402, Sigma). 40 The protocols for collecting human lung tissue were approved by the Institutional Review Board Committees at the University of Utah Hospital and LDS Hospital, Salt Lake City, Utah. The protocol for collecting human colonic tissue was approved by the Institutional Review Board Committee at the University of Utah Hospital, Salt Lake City, Utah. Cell Culture Cell Lines Bovine corneal cells (2048-CRL) and colon epithelial cell lines from ATCC (Manassas, VA), were cultured in recommended media. Basic media were from BioWhittaker (Walkersville, MD), supplemented with serum (Hyclone, Logan, UT). Additional chemical supplements were from Sigma. The HCA-7 cell line was a kind gift of Susan E. Kirkland (ICRF, England). Caco-2 cells, obtained from ATCC (37-HTB) were cultured in uncoated 6 well plates or plates that were coated with fibronectin (Sigma) at 1.5 J.lg/cm2 in water or gelatin (Sigma) at 0.1 % in water for 1 hour at 37°C, or on collagen I coated transwells (3418, Costar, Cambridge, MA) until they reached 80% confluence. Primary Human Bronchial Epithelial Cells Normal human bronchial epithelial cells were purchased from Clonetics (CC- 0225 San Diego, CA), and cultured in the recommended BEGM (bronchial epithelial growth medium, CC-3170, Clonetics). Six well tissue culture plates (Falcon) were coated with collagen I (Sigma) at 50 J.lg/ml in water for 20 hours at 4°C, prior to plating cells. Cells were treated with interleukin-l ~ or interleukin-3 (both from R&D Systems) at 250 nglml for 20 hours prior to harvest. For experiments, cells were harvested at 80% confluence or 7 days postconfluence. 41 Detection of Oncostatin M Messenger RNA Total RNA was isolated from cells using TRIzol (Gibco-BRL, Grand Island, NY) as per the manufacturer's instructions. Complementary DNA (cDNA) was generated from RNA using an oligo dT primer and MMLV reverse transcriptase (Gibco BRL). To prevent detection of genomic contaminants primers were designed to amplify regions from two different exons. cDNA was examined by PCR using human oncostatin M primers: "forwardtl 5' -CGGAA TTCA T A TGGGGGT ACTGCTCACAC-3', and "reverse" 5' -CGGAA TTCTGCTCTCGAGGCT ACCG-3 '. Human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers, used as a positive control for RNA/cDNA loading, were: 5'-TTTCTAGACGGCAGGTCAGG-3', and 5'-CCCATGGCAAATTCCATGGC- 3'. To detect oncostatin M in bovine epithelial cells, the following primers were used: "forward" primer 5'-CAGCGTATGCAGAGAACACTG-3', and "reverse" primer 5'­CAAAGGCT ACCGGGCGGCTGGC-3'. Primers used for bovine GAPDH controls were: "forward" primer 5' -GTTCCAGTA TGA TTCCACCCAC-3', and "reverse" primer 5' -GCAGTTGGTGGTGCAGGAGG-3: PCR products were resolved by electrophoresis on a 1 % agarose gel and visualized by ethidium bromide staining. al-anti-trypsin primers, used as a positive control for Caco-2 differentiation status, were: 5'- GACAGA T ACA TCCCACCA TGA-3' and 5' -AGGCTTCTGAGTGGT ACAACTT -3'. Primers were synthesized by the University of Utah DNA/peptide resource facility. PCR was 35 cycles of the following program: 94°C denaturation, 55°C anneal, and 72°C extension. Detection of Oncostatin M Protein Cells were lysed at Ix I 06 cells/ml in lysis buffer (PBS, with 1 mg/ml Pefabloc, 10 ~glml aprotinin, and 1 0 ~g/rrl1 leupeptin). Leupeptin and aprotinin were from Sigma, Pefabloc was from Boehringer Mannheim. Oncostatin M was detected by sandwich enzyme linked immunosorbant assay (ELISA) using a monoclonal antibody for antigen capture (MAB295, R&D Systems). An affinity purified-biotinylated polyclonal ""~e""~ ~~" !~'~'1L~ il ~"& 42 antibody was used for detection (AF-295-NA, R&D Systems). Streptavidin-biotin was from BioSource, and all other chemicals were from Sigma. Results Oncostatin M Protein is Present in Epithelium An affinity purified, biotinylated polyclonal antibody was used to examine several types of tissue for oncostatin M expression. Immunohistochemistry for oncostatin M on normal, noninflammed, human lung sections showed differentiated columnar epithelial cells constitutively expressed oncostatin M. The oncostatin M protein localized specifically to columnar bronchiolar epithelium (Figure 3.1 A and 3.1 B), and oncostatin M staining was less intense in the alveolar epithelium (Figure 3.1 E and 3.1 G). Lack of staining with a nonspecific isotype matched antibody showed this reaction was antigen­dependent (Figures 3.1B, 3.1D, 3.1F, and 3.lH). Therefore some, but not all, epithelial cells express oncostatin M at high levels in the absence of a discemable stimulus. These data revealed that epithelial cells in specific microenvironments or sites of maturation express high levels of oncostatin M, while alveolar epithelial cells that have a different function and morphology, have comparatively lower levels of oncostatin M. Colonic epithelium also expressed oncostatin M constitutively. Oncostatin M is continuously present in the epithelial cells along the length of the villus of mouse colon (brown chromogen, Figure 3.2A). None of the surrounding cell types stained for oncostatin M, and no staining was observed when an isotype matched control primary antibody was used in the same conditions (Figure 3.2B). Oncostatin M was also present human colon epithelial cells (brown chromogen, Figure 3.2C, 3.2E, and 3.2F). These sections clearly showed that oncostatin M staining was localized to the epithelium and is not present in stromal cells. Again, the antibody binding was specific, as no staining was 43 Figure 3.1: Oncostatin M is present in human lung epithelium. 5 Jlm human lung sections were prepared and stained as described in "Materials and Methods:' The presence of oncostatin M is indicated by the brown chromogen, DAB. Bronchiolar epithelium is shown in panels A and C. Alveolar epithelium is shown in panels E and G. An isotype matched, nonspecitic goat antibody, used at the same concentration as the oncostatin M antibody, did not react with the tissue (B, 0, F, and H). All sections were cowlterstained with hematoxylin (blue). 44 Figure 3.2: Oncostatin M is present in colon epitheliun1. Staining for oncostatin M was performed as described in "Materials and Methods." The presence of oncostatin M in human colon sections is indicated by the brown chromogen, DAB (A, C, D). Mouse colon sections also stained for oncostatin M, panel E. Specificity of oncostatin M immunohistochemistry is shown by a lack of staining with a nonspecific, isotype matched antibody (B, F). Sections were counterstained with methyl green (C, D), or hematoxylin (A, B, E, F). 45 present when a nonspecific primary antibody was used in the same conditions (Figure 3.2D). Oncostatin M was also present in rabbit eye epithelium, specifically in the corneal epithelium (Figure 3.3B and 3.3D), the bulbar conjunctiva (Figure 3.3F), and Descemet's endothelium (Figure 3.3H). Anti-oncostatin M antibody binding was specific as a nonspecific isotype matched control antibody did not result in staining (Figures 3.3C, 3.3E, 3.3G, and 3.31). Oncostatin M mRNA is Synthesized by Epithelium As previously described, leukocyte and Leydig cell expression of oncostatin M has been characterized; but oncostatin M expression by other cell types has not yet been shown. It was necessary to confinn that the oncostatin M protein detected in the various epithelial cells was, in fact, synthesized by these epithelial cells. The possibility existed that the oncostatin M produced by activated tissue macrophages, dendritic cells, or infiltrating mononuclear cells had simply been endocytosed by the epithelial cells. This is especially problematic because epithelial cells express oncostatin M receptors and these receptor complexes can be internalized (2, 4, 8). To demonstrate the capacity of epithelial cells to synthesize oncostatin M mRNA, in situ hybridization using the complete oncostatin M coding sequence to make mRNA probes showed that the epithelial cells that contained oncostatin M protein also expressed oncostatin M mRNA (Figure 3.4). Thus, lung bronchiolar epithelium stained intensely for oncostatin M message when probed with the strand complementary to oncostatin M mRNA (the anti-sense strand) (Figure 3.4A) and not when the nonspecific (sense strand) probe was used (Figure 3.4B). The epithelium of alveoli did not stain as intensely as the bronchiolar epithelium in the same section, corresponding to the immunohistochemistry data (Figures 3.4C and 3.4D). A similar pattern was seen when colon (Figures 3.4E and 3.4F) was stained with oncostatin M probes capable of hybridizing to oncostatin M mRNA and not with the nonspecific probe. In addition, oncostatin M mRNA was detected in corneal epithelium and bulbar A cornea ,Descemet's / endothelium anti-OSM Aq nonspecific Ab B c D H 46 Figure 3.3: Corneal epithelium contains oncostatin M. The cornea and its epithelial layers are represented schematically in panel A. Oncostatin M immunohistochemistry was performed as described in "Materials and Methods. II Oncostatin M staining is indicated by the brown chromogen, DAB (B, D, F, H). A nonspecific, isotype matched antibody did not result in staining (C, E, G, I). Corneal epithelium is shown in panels B, C, D, and E. Bulbar conjunctiva is shown in panels B, C, F, and G. Descemet's endothelium is shown in panels B, C, H, and L 47 Figure 3.4: Oncostatin M nlRNA is present in human columnar epithelial cells. Anti­sense oncostatin M mRNA ("OSM mRNA") hybridization is indicated by the presence of the brown chromogen, DAB. Panels A and B: bronchiolar epithelium. Panels C and D: bronchiolar and alveolar epithelium. Panels E and F: colonic epithelium. Sense oncostatin M mRNA ("nonspecific"), used as a negative control to show the specificity of the anti-sense in situ probe, did not react with the epithelium (B, D, F). All sections were counterstained with hematoxylin (blue). 48 conjunctiva (purple chromogen Figure 3.5A and 3.5C). Again, this hybridization was specific because the negative control mRNA probe did not bind to sequential, matched sections (Figure 3.58 and 3.5D). The proximity of epithelium and leukocytes, particularly in the colon, made it important to demonstrate that oncostatin M mRNA was produced by epithelial cells and not other stromal cells. The dendricyte, of myeloid lineage, could be responsible for oncostatin M production. However, double antibody immunohistochemistry for oncostatin M and a dendritic cell marker (B7-2) (28) showed cells expressing B7 -2 (purple chromogen), do not express oncostatin M (brown chromogen) (Figure 3.6A and 3.6B). Also, dendritic cells generated in several different cultures from human peripheral blood monocytes (by the methods of (29) or (28» did not produce oncostatin M mRNA as determined by RT -PCR (Figure 3.6C). Therefore, even though dendricytes are of the myeloid lineage, they are not a source of oncostatin M in colonic tissue. Epithelial Cell Lines Do Not Make Oncostatin M An epithelial cell line would permit in vitro studies of the pathways and promoter elements involved in oncostatin M expression by epithelial cells, and allow comparison with the regulation of oncostatin M expression in monocytes. Immunohistochemistry and in situ hybridization showed corneal epithelium and Descemet's endothelium expressed oncostatin M (Figure 3.3); however, a bovine corneal epithelial cell line did not express oncostatin M mRNA as detected by RT-PCR (Figure 3.7). Furthermore, 18 hours of LPS or IL-3 treatment (both at 500 ng/ml), agents that induce oncostatin M expression by monocytes, did not induce expression of detectable levels of oncostatin M mRNA. RNA isolation, cDNA production, and PCR were successful as indicated by the amplification of a constitutively expressed gene, GAPDH. Since it was possible that the bovine corneal cell line lacked the ability to synthesize oncostatin M, other epithelial cell lines were examined. Eleven different colon epithelial-like cell lines were assayed for oncostatin Nt 49 A B c " Figure 3.5: Oncostatin M mRNA is present in corneal epithelial cells as indicated by in situ hybridization. Hybridization conditions are described in Materials and Methods. Anti-sense oncostatin M mRNA ("OSM mRNAIt) hybridization is indicated by the presence of the purple chromogen, VIP. Corneal epithelium is shown in panels A and B. Bulbar conjunctiva is shown in panels C and D. The hybridization reaction was specific as the sense mRNA probe ("nonspecific") did not produce staining (B, D). All sections were counterstained with methyl green. 50 ... ..~ ..8L !'Ja .. :' ~iIf' "',* OSM: brown (DAB) B7 -2: purple (VI P) Counterstain: green (Methyl Green) Figure 3.6: Unactivated dendritic cells do not make oncostatin M. Panels A and B show double antibody immunohistochemistry on human colon sections. Dendritic cells, which express B7-2 (indicated by the purple chromogen, VIP), do not colocalize with cells that express oncostatin M (indicated by the brown chromogen, DAB). Sections were counterstained with methyl green. Panel C: RT-PCR was performed on unactivated dendritic cell RNA from three separate cultures, as described in "Materials and Methods." RNA from monocytes, activated with 200 ng/ml interleukin-3 for 18 hours prior to RNA isolation, was used as a positive control to show the PCR for oncostatin M was successful. GAPDH amplification showed equal loading of RNA/eDNA. Negative control (-) indicates RT-PCR done without eDNA. DNA bands were resolved by electrophoresis on a 1 % agarose gel, stained with ethidium bromide. 51 OSM GAPDH Figure 3.7: Transformed bovine corneal epithelial cells do not synthesize oncostatin M mRNA, with or without stimuli. Cells were treated for 18 hours with 500 ng/ml LPS or IL-3 prior to RNA isolation. RT -PCR for oncostatin M was done as described in "Materials and Methods." GAPDH amplification was used as a positive control for RT­PCR to show equal loading of RNA/cDNA. The RT-PCR reactions were free on contaminants as indicated by the absence of bands in the no cDNA (-) lane. DNA bands were resolved by electrophoresis on a 1 % agarose gel, stained with ethidium bromide. 52 production. However, none of these lines expressed detectable levels of oncostatin M mRNA, as determined RT -PCR (Figure 3.8). Monocytes activated with IL-3 or LPS at 500 ng/ml for 18 hours expressed oncostatin M mRNA, and served as a positive control for the oncostatin M RT -PCR reaction. GAPDH amplification showed the presence of the colon cell line cDNA in each sample. The lack of oncostatin M expression by multiple cell lines suggests that oncostatin M is produced only by fully differentiated, nontransformed epithelial cells. Loss of oncostatin M expression may be associated with or contribute to transformation. The Caco-2 human colon carcinoma cell line can be forced to express hallmarks of differentiation by various methods, the simplest of which allows cells to grow past confluence until they form microdomes (30). Other methods involve growing cells on extracellular matrix components or on transwells to more closely mimic the in vivo environment (31). Caco-2 cells grown in each of these culture conditions did not express oncostatin M as determined by RT -PCR (Figure 3.9), or by ELISA (data not shown). Several different primer sets were used to amplify oncostatin M, in an effort to detect potentially alternatively spliced oncostatin M transcripts. These primers were designed to bind in the middle of individual exons and amplify across introns to avoid amplification of sequences from any genomic contamination in the cDNA preparation. Functionality of these primer sets is shown by the amplification of oncostatin M bands from activated monocyte RNA (monos+IL-3 (+». None of the culture conditions used showed amplification of oncostatin M by any of the primer sets (Figures 3.9A, 3.9B, and 3.9C). Success of the RT -peR reaction and equal loading of Caco-2 cDNA was shown by GAPDH amplification (Figure 3.9E). al-anti-trypsin, an indicator for Caco-2 differentiation (31, 32), was present in the Caco-2 cells in all culture conditions, showing the cells were differentiated as evidenced by an independent marker (aI-AT, Figure 3.90). ~ (j) o J: a a.. « C!J 53 Figure 3.8: Oncostatin M mRNA is not synthesized by transfonned colon cell lines. Colon cell RNA was harvested when the cells were 80% confluent. For positive controls, primary human monocytes were treated for 18 hours with 500 ng/ml LPS or IL-3 prior to RNA isolation. R T -PCR for oncostatin M was perfonned as described in "Materials and Methods." GAPDH amplification was used as a positive control for RT-PCR to show equial loading of RNA/eDNA. The RT -PCR reactions were not contaminated as indicated by the absence of bands in reactions without target eDNA (-). DNA bands were resolved by electrophoresis on a 1 % agarose gel, stained with ethidium bromide. 54 A .-.- ' - - - ~ B OSM c D E Figure 3.9: Differentiated, transformed colon carcinoma cells do not express oncostatin M mRNA. Caco-2 cells were cultured on extracellular matrix proteins: fibronectin, gelatin, or collagen I coated transwells; to confluence or 3 days postconfluence. Following RNA isolation and cDNA production, three different primer sets were used to detect oncostatin M expression by RT-PCR. For positive control RNA, primary human monocytes were treated for 18 hours with 500 ng/ml IL-3 prior to RNA isolation. Primer sets were designed to amplify the entire known oncostatin M coding sequence (A), a section of oncostatin M from the middle of exon 1 to the middle of exon 2 (B), or a section from the middle of exon 2 to the middle of exon 3 (C). aI-antitrypsin amplification was used as a marker of Caco-2 differentiation status (D). GAPDH amplification showed success of the RT -PCR reactions and equal RNA/cDNA loading (E). The RT-PCRs were not contaminated as shown in the (-) lane, without target cDNA. DNA bands were resolved by electrophoresis on a 1 % agarose gel, stained with ethidium bromide. Primary Bronchial Epithelial Cells Make Oncostatin M mRNA 55 The search for cultured epithelial cells capable of making oncostatin M was expanded to include primary human cells. Normal human bronchial epithelial cells from Clonetics, cultured on collagen I, were examined by RT -PCR for oncostatin M expression. Low levels of oncostatin M message were detected by RT-PCR (Figure 3.1 0). Stimulation with interleukin-l ~ or interleukin-3 did not augment oncostatin M expression by these cells. Bands corresponding to oncostatin M did not appear in the interleukin-3 stimulated cells grown to 80% confluence, nor in the buffer treated cells grown 7 days beyond confluence. The faintness of the bands indicates that the RT-PCR is at the limit of detection for the conditions used. It is likely that the two samples that failed to produce oncostatin M bands in the RT -PCR did so because their original oncostatin M mRNA content fell below the detectable limit. As in previous experiments, amplification of an oncostatin M band from activated monocyte RNA and amplification of a GAPDH band from each sample showed the RT-PCR reaction worked. Therefore, oncostatin M expression by epithelial cells is restricted to primary cells grown on appropriate extracellular matrices. Moreover, in these cells, oncostatin M mRNA appears to be in low abundance, but constitutively present. ST A T5b is Not Responsible for Expression of Oncostatin M in Epithelium The apparent constitutive production of oncostatin M by several epithelial cell types suggests the signal transduction pathways leading to induction of oncostatin M expression in epithelial cells are different from those of leukocytes, as leukocytes express oncostatin M only after a cytokine or bacterial stimulus. The transcription factor, STAT5b, activates monocyte expression of oncostatin M (Chapter 2) (33). Immunohistochemistry for the transcription factor STAT5b (red chromogen) showed it was not present in all cells in the corneal epithelium (Figure 3.11 A) or bronchiolar epithelium (Figure 3.11 C and 3.11 E). As a positive control, sections of inflamed lung 80% 7 days confluent post confluent GAPDH """"'-S99bp 56 Figure 3.10: RT-PCR for oncostatin M mRNA in normal human bronchial epithelial cells. RNA was harvested from epithelial cells at 80% confluence, or 7 days postconfluence as described in "Materials and Methods." Prior to harvest, cells were treated with interleukin 1~ (IL-1~) or interleukin-3 (IL-3) at 250 ng/ml for 20 hours. RNA was subjected to RT -PCR performed with oncostatin M primers designed to amplify the oncostatin M coding sequence, spanning exon 1 to exon 3 (754 base pairs), or GAPDH (599 base pairs). Amplified DNA bands were resolved on a 1 % agarose gel stained with ethidium bromide. IL-3 activated monocyte RNA was used as a positive control for the oncostatin M PCR ('tmono+IL-3 (+)"). The RT-PCRs were free of contaminating eDNA as shown by the absence of a band in reactions without target eDNA (-). 57 A B Figure 3.11: ST A T5b is not constitutively present in unactivated corneal or bronchiolar epithelial cells, but is present in bronchiolar epithelium and mononuclear cells in inflammed lung. Immunohistochemistry for ST A T5b was done as described in "Materials and Methods." Presence of STAT5b is indicated by the red chromogen (Vector Red), as shown in panels A, C, E, and G. Panels A and B: rabbit corneal epithelium. Panels C, D, E, and F: human bronchiole from normal, non-inflamed lung tissue. Panels G and H: inflammed human lung (ARDS). An isotype matched irrelevant antibody (anti-insulin) did not bind to the sections, showing the specificity of the oncostatin M immunohistochemistry reaction (B, D, F, and H). All sections were counterstained with hematoxylin (blue). 58 were examined with same procedure to show that infiltrating mononuclear cells stained intensely for ST A T5b protein as anticipated. Direct comparison of panels showing oncostatin M and ST A T5b staining shows diffuse ST A T5b in the cytoplasm of several of the surface corneal epithelial cells (red chromogen, Figure 3.12C) but did not correlate completely with cells expressing oncostatin M protein (brown chromogen, Figure 3.12A), or mRNA (purple chromogen, Figure 3.12B). Bronchiolar epithelial cells showed a similar result. Cells shown to express oncostatin M protein and message did not express ST A T5b, and no ST A T5b appeared in the nuclei of these cells (Figure 3.1 3.12F, and 3.12G). Thus epithelial cells can express ST A T5b, but this expression does not correlate with oncostatin M protein or message expression. Inflamed lung tissue has active mononuclear cells, that should have both ST A T5b and oncostatin M. Therefore, these mononuclear cells may act as a postive control for both ST A T5b and oncostatin M staining in tissue sections from inflammed lung. Bronchiolar epithelium in inflammed lung sections expressed oncostatin M protein (Figure 3.13). In addition, they also contained STAT5b, which was absent or present only at low levels in normal epithelium. This observation suggests that ST A T5b induces inflammatory signals in epithelial cells in mononuclear cells, but the presence of nuclear ST A T5b and oncostatin M production is correlative in mononuclear cells it is not in epithelial cells. Discussion The diverse effects of oncostatin M both in developing and mature organisms, leads to the question of how oncostatin M expression is temporally and spatially controlled in vivo. It has previously been shown that oncostatin M is secreted by activated primary monocytes, T cells, and neutrophils (Chapter 2)(23, 34), and now this study shows that oncostatin M is constitutively expressed by differentiated epithelial cells A OSM protein: brown, counterstain: blue E OSM protein: brown counterstain: blue --8 OSM mANA: purple counterstain: green F OSM mANA: brown counterstain: blue ~ C STAT5b: red counterstain: blue G STAT5b: red counterstain: blue 59 Figure 3.12: Oncostatin M expression does not correlate exclusively with the presence of ST A T5b in corneal epithelium or norn1al human lung epithelium. Imn1unohistochemistry and in situ hybridization for oncostatin M, and immunohistochemistry for ST A T5b were performed as described in "Materials and Methods." Presence of oncostatin M protein is shown by the brown chromogen, DAB (A, E). Oncostatin M mRNA ( OSM mRNA ) is epithelium is indicated by the purple chromogen, VIP (B), or the brown chromogen, DAB (F). Presence of ST A T5b is shown by the red chromogen, Vector Red (C, G). Rabbit corneal epithelium is shown in panels A, B, and C. Human bronchiolar epithelium, from noninflammed lung, is presented in panels D, E, and F. All sections were counterstained with hematoxylin (blue), except for the section in panel B, which is stained with methyl green. 60 bronchiole alveoli Figure 3.13: Oncostatin M and STAT5b are expressed at different intensities by various cells in inflammed human lung tissue. Sections from inflammed human lung (ARDS) were examined by immunohistochemistry. Staining for oncostatin M and ST A T5b was performed as described in "Materials and Methods." Oncostatin M staining is indicated by the brown chromogen, DAB (A, B, and C). ST A T5b expression is shown by the red chromogen, Vector Red (D, E, and F). All sections were counterstained with hematoxylin (blue). 61 in situ, without activation as is required by leukocytes. The immunohistochemical analyses show columnar epithelial cells of the cornea, bronchioles, and colon produce oncostatin M. Epithelial cells have previously been shown to produce a variety of cytokines, both constitutively and following activation (33, 35-38). The possibility existed that the oncostatin M detected in the epithelial cells was secreted by neighboring tissue macrophages or dendritic cells and was merely endocytosed by the epithelial cells. Cells in the colon of myeloid origin, a potential source of oncostatin M, are the dendricytes. However, in situ hybridization showed that oncostatin M mRNA was present only in the epithelial cells. In addition, unstimulated dendritic cells generated in culture did not produce oncostatin M (28, 29). These data show that oncostatin M is produced by the epithelial cells and not unactivated tissue myeloid cells. For many systems, cell lines provide an excellent model to study the regulation of cytokine expression. The major concern with transformed cell lines is that they have lost some of the characteristics of primary cells. Nonetheless, multiple cell lines were examined for oncostatin M expression. A bovine corneal epithelial cell line and eleven different transformed colon cell lines did not express oncostatin M as determined by ELISA (not shown) and RT-PCR. To minimize the possibility the RT-PCR was not an1plifying an alternatively spliced transcript, several different primers sets were used in the reactions. Addition of agonists known to induce oncostatin M expression by monocytes did not induce oncostatin M expression by epithelial cell lines. For some cells, oncostatin M is a cytostatic agent (22, 23) that activates cell cycle inhibitors (24-26), thus the expression of oncostatin M could be associated with terminal differentiation of epithelium. The colon carcinoma epithelioid Caco-2 cell line was differentiated in culture, but never expressed detectable levels of oncostatin M. In summary, no cell line tested produced oncostatin M mRNA regardless of culture conditions. Whether the lack of oncostatin M production is in part responsible for, or 62 simply associated with transformation in these cells remains unknown. Normal human bronchial epithelial cells did, however, produce oncostatin M mRNA at low levels, showing a difference between cells in situ, primary cells, and transformed cell lines. Oncostatin M expression by differentiated epithelial cells of the colon, cornea, and bronchioles, but not in transformed cell lines suggest oncostatin M is associated with cytostasis. A second or additional role for oncostatin M may be to modulate epithelial repsonses to pathogenic damage. As reported by others, epithelial cells respond to oncostatin M in a variety of ways, including secretion of the protease inhibitors al-anti-trypsin, aI-protease inhibitor, and tissue inhibitor of metalloproteinase-1 (TIMP-1) (17, 19-21,39-41). The expression of oncostatin M by some, but not all, epithelial cells suggest local, autocrine control of the functions. For example, should lung epithelium become damaged, the columnar epithelial cells have the potential to release oncostatin M to the surrounding areas and induce the release of protease inhibitors. This may function in a protective fashion to minimize lung epithelial damage and help maintain airway integrity. Identification of epithelial expression of oncostatin M contributes to the knowledge of where this pleuripotent cytokine may be present in the body. References 1. Rose, T. M., and A. G. Bruce. 1991. Oncostatin M is a member of a cytokine family that includes leukemia- inhibitory factor, granulocyte colony-stimulating factor, and interleukin 6. Proc Natl Acad Sci USA 88:8641. 2. Heinrich, P. C., 1. Behrmann, G. Muller-Newen, F. Schaper, and L. Graeve. 1998. Interleukin-6-type cytokine signalling through the gp 130/Jak/ST AT pathway. Biochem J 334:297. 3. Bruce, A. G., P. S. Linsley, and T. M. Rose. 1992. Oncostatin M. Prog Growth Factor Res 4: 157. 4. Auguste, P., C. Guillet, M. Fourcin, C. Olivier, J. Veziers, A. Pouplard­Barthelaix, and H. Gascan. 1997. Signaling of type II oncostatin M receptor. J Bioi Chem 272: 15760. 5. Cichy, J., S. Rose-John, and E. Pure. 1998. Regulation of the type II oncostatin M receptor expression in lung- derived epithelial cells. FEBS Lett 429:412. 63 6. Gadient, R. A., and P. H. Patterson. 1999. Leukemia inhibitory factor, Interleukin 6, and other cytokines using the GP130 transducing receptor: roles in inflammation and injury. Stem Cells 17: 12 7. 7. Gearing, D. P., M. R. Comeau, D. J. Friend, S. D. Gimpel, C. J. Thut, J. McGourty, K. K. Brasher, J. A. King, S. Gillis, B. Mosley, and et al. 1992. The IL-6 signal transducer, gp130: an oncostatin M receptor and affinity converter for the LIF receptor. Science 255:1434. 8. Mosley, B., C. De Imus, D. Friend, N. Boiani, B. Thoma, L. S. Park, and D. Cosman. 1996. Dual Oncostatin M (OSM) Receptors. Cloning and characterization of an alternative signaling subunit conferring osm-specific receptor activation. J Bioi Chem 271:32635. 9. Watanabe, D., R. Yoshimura, M. Khalil, K. Yoshida, T. Kishimoto, T. Taga, and H. Kiyama. 1996. Characteristic localization of gp130 (the signal-transducing receptor component used in common for IL-6/IL-II/CNTF/LIFIOSM) in the rat brain. Eur J Neurosci 8: 1630. 10. Linsley, P. S., M. Bolton-Hanson, D. Horn, N. Malik, 1. C. Kallestad, V. Ochs, J. M. Zarling, and M. Shoyab. 1989. Identification and characterization of cellular receptors for the growth regulator, oncostatin M. J Bioi Chem 264:4282. 11. Thoma, B., T. A. Bird, D. J. Friend, D. P. Gearing, and S. K. Dower. 1994. Oncostatin M and leukemia inhibitory factor trigger overlapping and different signals through partially shared receptor complexes. J Bioi Chem 269:6215. 12. Modur, V., M. J. Feldhaus, A. S. Weyrich, D. L. Jicha, S. M. Prescott, G. A. Zimmeffi1an, and T. M. McIntyre. 1997. Oncostatin M is a proinflammatory mediator. In vivo effects correlate with endothelial cell expression of inflamnlatory cytokines and adhesion molecules. J Clin Invest 100: 158. 13. Prescott, S. M., T. M. McIntyre, and G. A. Zimmerman. 1999. Platelet-Activating Factor: A Phospholipid Mediator of Inflammation. In Inflammation: Basic Principles and Clinical Correlates. J. I. Gallin, and R. Snyderman, eds. Lippincott Williams and Wilkins, Philadelphia, p. 387. 14. Vasse, M., J. Pourtau, V. Trochon, M. Muraine, J. P. Vannier, H. Lu, J. Soria, and C. Soria. 1999. Oncostatin M induces angiogenesis in vitro and in vivo. Arterioscler Thromb Vasc Bioi 19:1835. 15. Wijelath, E. S., B. Carlsen, T. Cole, 1. Chen, S. Kothari, and W. P. Hammond. 1997. Oncostatin M induces basic fibroblast growth factor expression in endothelial cells and promotes endothelial cell proliferation, migration and spindle morphology. J Cell Sci 110:871. 16. Cichy, J., S. Rose-John, and J. Travis. 1998. Oncostatin M, leukaemia-inhibitory factor and interleukin 6 trigger different effects on alpha I-proteinase inhibitor synthesis in human lung-derived epithelial cells. Biochem J 329:335. 17. Cichy, J., J. Potempa, R. K. Chawla, and J. Travis. 1995. Stimulatory effect of inflammatory cytokines on alpha 1- antichymotrypsin expression in human lung-derived epithelial cells. J Clin Invest 95:2729. 64 18. Duncan, M. R., A. Hasan, and B. Berman. 1995. Oncostatin M stimulates collagen and glycosaminoglycan production by cultured normal dermal fibroblasts: insensitivity of sclerodermal and keloidal fibroblasts. J Invest Dermatol 104: 128. 19. Richards, C. D., C. Kerr, M. Tanaka, T. Hara, A. Miyajima, D. Pennica, F. Botelho, and C. M. Langdon. 1997. Regulation of tissue inhibitor of metalloproteinase-1 in fibroblasts and acute phase proteins in hepatocytes in vitro by mouse oncostatin M, cardiotrophin-l, and IL-6. J ImmunoI159:2431. 20. Richards, C. D., and A. Agro. 1994. Interaction between oncostatin M, interleukin 1 and prostaglandin E2 in induction of IL-6 expression in human fibroblasts. Cytokine 6:40. 21. Richards, C. D., M. Shoyab, T. 1. Brown, and 1. Gauldie. 1993. Selective regulation of metalloproteinase inhibitor (TIMP-1) by oncostatin M in fibroblasts in culture. J ImmunoI150:5596. 22. Zarling, 1. M., M. Shoyab, H. Marquardt, M. B. Hanson, M. N. Lioubin, and G. 1. Todaro. 1986. Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells. Proc Natl Acad Sci USA 83:9739. 23. Brown, T. 1., M. N. Lioubin, and H. Marquardt. 1987. Purification and characterization of cytostatic lymphokines produced by activated human T lymphocytes. Synergistic antiproliferative activity of transforming growth factor beta 1, interferon­gamma, and oncostatin M for human melanoma cells. J Immunol 139:2977. 24. Nakayama, K., T. Hara, M. Hibi, T. Hirano, and A. Miyajima. 1999. A novel oncostatin M-inducible gene OIG37 forms a gene family with MyDl18 and GADD45 and negatively regulates cell growth. J Bioi Chem 274:24766. 25. Bellido, T., C. A. O'Brien, P. K. Roberson, and S. C. Manolagas. 1998. Transcriptional activation of the p21(WAF1,CIP1,SDIl) gene by interleukin-6 type cytokines. A prerequisite for their pro- differentiating and anti-apoptotic effects on human osteoblastic cells. J Bioi Chem 273:21137. 26. Kortylewski, M., P. C. Heinrich, A. Mackiewicz, U. Schniertshauer, U. Klingmuller, K. Nakajima, T. Hirano, F. Horn, and 1. Behrmann. 1999. Interleukin-6 and oncostatin M-induced growth inhibition of human A375 melanoma cells is STAT­dependent and involves upregulation of the cyclin-dependent kinase inhibitor p27/Kip1. Oncogene 18:3742. 27. De Miguel, M. P., 1. Regadera, F. Martinez-Garcia, M. Nistal, and R. Paniagua. 1999. Oncostatin M in the normal human testis and several testicular disorders. J Clin Endocrinol Metab 84:768. 28. Romani, N., N. Bhardwaj, M. Pope, F. Koch, W. 1. Swiggard, U. O'Doherty, M. D. Witmer-Pack, L. Hoffman, G. Schuler, K. Inaba, and R. M. Steinman. 1996. Chapter 156: Dendritic Cells. In Weir's Handbook of Experimental Immunology, Vol. IV: The Intergrated Immune System. D. M. Wier, L. A. Herzenberg, L. A. Herzenberg, and C. D. Blackwell, eds. Blackwell Science, Can1bridge, p. 156.1. 29. Randolph, G. 1., S. Beaulieu, S. Lebecque, R. M. Steinman, and W. A. Muller. 1998. Differentiation of Monocytes into Dendritic Cells in a Model of Transendothelial Trafficking. Science 282:480. 65 30. Field, F. J., E. Albright, and S. N. Mathur. 1987. Regulation of cholesterol esterification by micellar cholesterol in CaCo-2 cells. J Lipid Res 28: 1057. 31. Basson, M. D., G. Turowski, and N. J. Emenaker. 1996. Regulation of human (Caco-2) intestinal epithelial cell differentiation by extracellular matrix proteins. Exp Cell Res 225:301. 32. Perlmutter, D. H., J. D. Daniels, H. S. Auerbach, K. De Schryver-Kecskemeti, H. S. Winter, and D. H. Alpers. 1989. The alpha I-antitrypsin gene is expressed in a human intestinal epithelial cell line. J Bioi Chem 264:9485. 33. Yoshimura, A., M. Ichihara, I. Kinjyo, M. Moriyama, N. G. Copeland, D. J. Gilbert, N. A. Jenkins, T. Hara, and A. Miyajima. 1996. Mouse oncostatin M: an immediate early gene induced by multiple cytokines through the JAK-ST AT5 pathway. EmboJ 15:1055. 34. Grenier, A., M. Dehoux, A. Boutten, M. Arce-Vicioso, G. Durand, M. A. Gougerot-Pocidalo, and S. Chollet-Martin. 1999. Oncostatin M production and regulation by human polymorphonuclear neutrophils. Blood 93: 1413. 35. Wirtschafter, J. D., L. K. McLoon, J. M. Ketcham, R. J. Weinstock, and J. C. Cheung. 1997. Palpebral conjunctival transient amplifying cells originate at the mucocutaneous junction and their progeny migrate toward the fornix. Trans Am Ophthalmol Soc 95:429. 36. Shirota, K., L. LeDuy, S. Y. Yuan, and S. Jothy. 1990. Interleukin-6 and its receptor are expressed in human intestinal epithelial cells. Virchows Arch B Cell Pathol Incl Mol PathoI558:303. 37. Larsson, B. M. 1999. Gram positive bacteria induce IL-6 and IL-8 production in human alveolar macrophages and epithelial cells. Inflammation 23:217. 38. Yoshida, Y. 1999. Reactive onxygen intermediates stimulate interleukin-6 production in human bronchial epithelial cells. Am J Physiol 276:L900. 39. Cichy, J., J. Potempa, R. K. Chawla, and J. Travis. 1995. Regulation of alpha 1- anti chymotrypsin synthesis in cells of epithelial origin. FEBS Lett 359:262. 40. Richards, C. D., C. Langdon, F. Botelho, T. J. Brown, and A. Agro. 1996. Oncostatin M inhibits IL-l-induced expression of IL-8 and granulocyte- macrophage colony-stimulating factor by synovial and lung fibroblasts. J ImmunoI156:343. 41. Sallenave, J. M., G. M. Tremblay, J. Gauldie, and C. D. Richards. 1997. Oncostatin M, but not interleukin-6 or leukemia inhibitory factor, stimulates expression of alphal-proteinase inhibitor in A549 human alveolar epithelial cells. J Interferon Cytokine Res 17:337. CHAPTER 4 SUMMARY AND PERSPECTIVES Models of cytokine interaction networks become more complex each year. As the scientific community works to identify the complex cytokine communication pathways between cells, integration of resulting data becomes a greater challenge. The effects of each cytokine depend upon the microenvironment examined. To organize the expanding field, cytokines have been grouped into families based on their effects, structures, and receptors. Members of the interleukin-6 family of cytokines have extremely diverse effects. In some cases, as with interleukin-6 family member oncostatin M, some of the reported effects appear to be dichotomous. Different research groups have reported developmental, proinflammatory, anti-inflammatory, mitogenic, and non­mitogenic effects for oncostatin M (1, 2). A possible explanation for these observations is that oncostatin M production is temporally and spatially restricted, and the environments examined are not constitutively in contact with the cytokine. This thesis project has focused on the origins of oncostatin M expression. Identifying the environments in which oncostatin M is produced provides a better understanding of the variety of effects attributed to the molecule in vivo. Regulation of Oncostatin M Production by Monocytes Oncostatin M was originally purified from a promyeloid cell line that secretes oncostatin M following extended phorbol ester treatment (3). Data in this thesis show cytokines and bacterial products induce oncostatin M secretion by primary human monocytes. Oncostatin M mRNA was not constitutively present in the cells, it was synthesized only after monocytes encountered agonist. Specifically, interleukin-3 (IL-3) 67 and granulocyte macrophage-colony stimulating factor (GM-CSF) activate monocytes to produce oncostatin M. However, interleukin-5 (IL-5), an IL-3/GM-CSF family menlber that shares a common receptor subunit with these cytokines, did not induce production of oncostatin M by monocytes. In addition, oncostatin M itself was able to initiate production of its own mRNA. As with IL-5, other oncostatin M family members did not have the capacity to activate oncostatin M secretion. These results indicate specific receptor complexes are required for activation of oncostatin M synthesis by monocytes. As a group the cytokines found to induce oncostatin M synthesis are those that may be present at sites of inflammatory and immune responses. Bacteria frequently enter at a site of tissue damage. The bacterial component, lipopolysaccharide (LPS), initiated significant oncostatin M production, which reproducibly exceeded the amount produced following cytokine stimulation. One possible explanation for this is that once monocytes encounter bacterial components, they release factors, which along with LPS, feedback to the monocyte to modulate the intensity of the oncostatin M production response. Promoter Regulation/Signal Transduction IL-3, GM-CSF, and oncostatin M have all been shown to activate members of the Janus kinase (JAK) - signal transducer and activator of transcription (STAT) pathways. In addition, some groups have shown that receptors that interact with JAK, or JAKs themselves, can also bind and activate adapter proteins like Grb, which can lead to the activation of the MAP kinase pathway (4, 5). Therefore, a number of different transcription factors may potentially bind the oncostatin M promoter region and induce transcription. In this study, the genomic region 5' of the oncostatin M gene was sequenced and examined for potential transcription factor binding sites. Based on the knowledge that the cytokines shown to activate oncostatin M activate JAK-ST A T pathways, promoter examination began with two potential ST AT binding sites -180 base 68 pairs, and -100 base pairs to the transcriptional start site. Site directed mutagenesis of the sites showed the - 180 STAT binding site was essential for cytokine induced oncostatin M promoter activity. These results indicated STAT involvement. However, reports indicate a four different family members can be activated by IL-3 and OM-CSF (6). A more specific analysis, using electrophoretic mobility supershift assay, showed that ST A T5b bound to the promoter at the -180 site, but the other probably candidate ST A Ts did not. Therefore, oncostatin M expression by monocytes is tightly controlled. Cytokines binding to their specific receptor complexes result in ST A T5b binding to the oncostatin M promoter, and subsequent transcription and secretion. Oncostatin M Production by Epithelial Cells Immunohistochemistry showed oncostatin M was present in certain types of epithelial cells, in addition to activated mononuclear cells. However, unlike monocytes, epithelial cells synthesized oncostatin M constitutively. in situ hybridization showed epithelial cells made oncostatin M mRNA. Double antibody immunohistochemistry showed the epithelial cells, and not tissue macrophages or dendritic cells made oncostatin M. Based on the apparent constitutive production of oncostatin M by epithelial cells, regulation of oncostatin M expression is different from that of monocytes. Multiple epithelial cell lines were examined for oncostatin M expression, but none of the cell lines tested made detectable oncostatin mRNA or protein. RT -PCR with mid-exon primers showed no indication of full length or truncated oncostatin M mRNA expression. These data suggest cellular transformation is associated with an absence of oncostatin M production. Normal human bronchial epithelial cells in culture did produce low levels of onocstatin M mRNA. The summary of these observations suggest that oncostatin M is only produced by differentiated epithelial cells in vivo, or by primary cells in culture. Based on others reports of oncostatin M's ability to induce cell cycle arrest, the epithelial 69 cell lines examined for oncostatin M expression may have lost the ability to produce oncostatin M. The oncostatin M detected in human tissue sections may be confirmed by future work done by PixCell™, a system that allows individual cells to be ren10ved from tissue sections and subjected to RT -PCR. In addition, in situ PCR experiments may also be done to confirm the presence of oncostatin M mRNA in epithelial cells in normal human tissues. In addition, the normal human bronchial epithelial cells provide a model system in which the signal transduction pathway leading to oncostatin M expression by epithelial cells may be studied. The individual effects of oncostatin M indicate it has therapeutic potential. However, a comprehensive examination of the available information on oncostatin M suggests, because its effects are diverse, further investigation of the contradictory functions reported. The origins of oncostatin M production, as shown in this thesis project, suggest its expression is under strict control, localized to nontransformed cells with a specific morphology and location. References 1. Bruce, A. G., P. S. Linsley, and T. M. Rose. 1992. Oncostatin M. Prog Growth Factor Res 4: 157. 2. Heinrich, P. C., 1. Behrmarm, G. Muller-Newen, F. Schaper, and L. Graeve. 1998. Interleukin-6-type cytokine signalling through the gp130/JakiSTAT pathway. Biochem J 334:297. 3. Zarling, J. M., M. Shoyab, H. Marquardt, M. B. Hanson, M. N. Lioubin, and G. 1. Todaro. 1986. Oncostatin M: a growth regulator produced by differentiated histiocytic lymphoma cells. Proc Natl Acad Sci USA 83:9739. 4. Chauhan, D., S. M. Kharbanda, A. Ogata, M. Urashima, D. Frank, N. Malik, D. W. Kufe, and K. C. Anderson. 1995. Oncostatin M induces association of Grb2 with Janus kinase JAK2 in multiple myeloma cells. J Exp Med 182:1801. 5. Giordano, V., G. De Falco, R. Chiari, 1. Quinto, P. G. Pelicci, L. Bartholomew, P. Delmastro, M. Gadina, and G. Scala. 1997. Shc mediates IL-6 signaling by interacting with gpl30 and Jak2 kinase. J ImmunolI58:4097. 70 6. Kirito, K., M. Uchida, M. Yamada, Y. Miura, and N. Komatsu. 1997. A distinct function of STAT proteins in erythropoietin signal transduction. J BioI Chern 272:16507.