@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Microbiology and Immunology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Norman, Nadine Elizabeth"@en ; dcterms:issued "2010-11-01T17:39:56Z"@en, "1990"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """A large amount of effort has gone into the elucidation of the mechanism of suppression of the immune system. This level of immunoregulation has been demonstrated to be mediated by both antigen-nonspecific and antigen-specific protein factors elicited by leukocytes. In this work, two different modes of immunosuppression were investigated. First, an attempt was made to purifiy an antigen-nonspecific protein factor, SIF (Suppressor Inducer Factor) secreted by the Natural Suppressor cell line M1-A5. M1-A5 culture supernatants were subjected to ion exchange chromatogrphy (IEC) and fast protein liquid chromatography (FPLC). Bioactivity of eluted fractions was determined by the plaque forming cell assay and followed through the purification. Reducing SDS-PAGE of selected fractions suggested that bands with Mr's of >110 KD and/or 55 KD were mediating the suppressive activity. In addition, an assay was developed to further investigate the mode of action of SIF. Second, the binding of two antisera raised to components associated with murine antigen-specific suppression was studied using human peripheral leucocytes and several human tumour cell lines. Anti-p80 and anti-p30 binding was found to be variable (within a range) and to involve two populations of human mononuclear cells. Subsequently, it was found that all CD3⁺ (T cells), CD19⁺ (B cells) and neutophils expressed both the p80 and p30 determinants. Four human leukemic cell lines were found to express varying levels of the p80 and p30 determinants. Cell lysates from each of the cell lines were subjected to Western blot analyses using anti-p30. The results showed that anti-p30 binds to a major band of 42 KD and minor bands of 60 and 80 KD in all lysates. In addition, a 25 KD band was observed in RAJI and CEM-CM3 lysates only. Thus, it appears that HuT 78 cells sythesize but are unable to express the p30-containing, 42 KD molecule on the cell surface. No firm conclusions can be made with respect to the biochemical nature or mechanisism of action of either of the two suppressor factors studied in this work. Research into the mechanisms of suppression of the immune system is complex and multi-faceted, and it seems that for now, there will remain a gap in our overall understanding of immune regulation."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/29691?expand=metadata"@en ; skos:note "PARTIAL CHARACTERIZATION AND PURIFICATION OF A MURINE SUPPRESSOR-INDUCER FACTOR SECRETED BY A NATURAL SUPPRESSOR CELL LINE and CHARACTERIZATION OF THE BINDING OF ANTISERA RAISED TO MURINE ANTIGEN-SPECIFIC SUPPRESSIVE MATERIAL IN HUMAN PERIPHERAL LEUKOCYTES BY NADINE ELIZABETH NORMAN B.Sc, University of Toronto, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Microbiology We accept this thesis as conforming to the required standard: THE UNIVERSITY OF BRITISH COLUMBIA OCTOBER 1990 © Nadine Elizabeth Norman, 1990. In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT A large amount of effort has gone into the elucidation of the mechanism of suppression of the immune system. This level of immunoregulation has been demonstrated to be mediated by both antigen-nonspecific and antigen-specific protein factors elicited by leukocytes. In this work, two different modes of immunosuppression were investigated. First, an attempt was made to purifiy an antigen-nonspecific protein factor, SIF (Suppressor Inducer Factor) secreted by the Natural Suppressor cell line M1-A5. M1-A5 culture supernatants were subjected to ion exchange chromatogrphy (IEC) and fast protein liquid chromatography (FPLC). Bioactivity of eluted fractions was determined by the plaque forming cell assay and followed through the purification. Reducing SDS-PAGE of selected fractions suggested that bands with Mr's of >110 KD and/or 55 KD were mediating the suppressive activity. In addition, an assay was developed to further investigate the mode of action of SIF. Second, the binding of two antisera raised to components associated with murine antigen-specific suppression was studied using human peripheral leucocytes and several human tumour cell lines. Anti-p80 and anti-p30 binding was found to be variable (within a range) and to involve two populations of human mononuclear cells. Subsequently, it was found that all CD3+ (T cells), CD19+ (B cells) and neutophils expressed both the p80 and p30 determinants. Four human leukemic cell lines were found to express varying levels of the p80 and p30 determinants. Cell lysates from each of the cell lines were subjected to Western blot analyses using anti-p30. The results showed that anti-p30 binds to a major band of 42 KD and minor bands of 60 and 80 KD in all lysates. In addition, a 25 KD band was observed in RAJI and CEM-CM3 lysates only. Thus, it appears that HuT 78 cells sythesize but are unable to express the p30-containing, 42 KD molecule on the cell surface. No firm conclusions can be made with respect to the biochemical nature or mechanisism of action of either of the two suppressor factors studied in this work. Research into the mechanisms of suppression of the immune system is complex and multi-faceted, and it seems that for now, there will remain a gap in our overall understanding of immune regulation. iii TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEGEMENTS ix MAIN INTRODUCTION: An Overview of Immune Function 1 Regulation of Immune Responses by Lymphocytes 1 Immunosuppression 2 An Overview of Research to be Presented 3 SECTION 1: ANTIGEN NON-SPECIFIC SUPPRESSION SECTION 1 - INTRODUCTION: An Overview of Antigen-Nonspecific, or Natural Suppression 4 Sites of NS Cell Activity 4 Possible Mode of Action of NS Cells 5 Antigen Non-Specific Suppressor Factors 6 NS Cell Activity in Response to Tumours 7 Suppressor Inducer Cells 7 Isolation of the NS Cell Line M1-A5 8 In Vitro Bioactivity of M1-A5 Cells 8 Biochemical Purification of M1-A5-Derived Suppressor-lnducer Factor (SIF) 9 Summary of Work to be Presented 11 SECTION 1 - MATERIALS AND METHODS: Cell Lines 13 Harvesting Supernatants 13 Biochemical Purification of SIFa and SIFp* from M1-A5 Supernatant 13 Sodium Dodecyl Sulfate - Polyacrylamide Gel Electrophoresis (SDS-PAGE) 14 Screening for Bioactivity - The Suppressor Cell Assay 14 Plaque Forming Cell (PFC) Assay 15 Antibody Culture (Ab C) Assay 15 Mycoplasma Testing and Cell Line Curing 16 iv SECTION 1 - RESULTS: Culturing WEHI-3B and M1-A5 Cell Lines 1 7 SIF Activity in the Antibody Culture Assay 1 7 Preincubation of KLH-lmmune Splenocytes with M1-A5 Supernatant 1 9 Biochemical Purification of Suppressor-lnducer Factor (SIF) From M1-A5 Serum-free Supernatant 2 2 Reducing SDS-PAGE of Pooled FPLC Fractions 3 1 Mycoplasma Infection of the M1-A5 Cell Line 31 Mycoplasma Testing and Antibiotic Curing of M1-A5 Cells 3 3 SECTION 1 - DISCUSSION: 3 4 SECTION 1 - CONCLUSIONS: 4 1 SECTION 2: ANTIGEN-SPECIFIC IMMUNOSUPPRESSION SECTION 2 - INTRODUCTION: An Overview 4 2 Attempts to Correlate Ts Cell Surface Phenotype with Function 4 2 Purification of T Cell-Derived Suppressor Factors (TsF's) 4 4 Molecular Interactions in Ts-Mediated Suppression 4 5 The P815 Tumour, Ts Cells and the MAb B16G 4 6 Isolation of A10, A Ts Cell Hybridoma 4 7 Fd11: A Ts Hybridoma Specific for the Antigen Ferredoxin (Fd) 4 7 Anti-p80 and Anti-p30: Antisera Made to Components of Affinity-Enriched Fd11F 4 8 Summary of Work to be Presented 4 9 SECTION 2 - MATERIALS AND METHODS: Antisera 5 0 Antibodies 5 0 Harvesting Human Peripheral Blood Mononuclear Cells From Normal Whole Blood 5 0 Staining of Peripheral Blood Mononuclear Cells With Antisera and MAbs 5 0 Isolation of CD3, CD4 and CD8 Positive PBMN Cells For Staining With Antisera 5 2 Screening CD19 + Cells for Anti-p80/Anti-p30 Reactivity 5 2 Isolation of Granulocytes and Screening for Reactivity with p80/p30 Antisera 5 2 Cell Suface Studies With Human Lymphocytic Cell Lines 5 3 Western Blot Analysis of Leukemic Cell Lysates 5 3 V SECTION 2 - RESULTS: Anti-p80 and Anti-p30 Reactivity With Murine Cells 5 4 Screening Human PBMN Cells For Anti-p80 and Anti-p30 Binding 5 4 Expression of the P80 and P30 Determinants by Human T Lymphocytes 5 8 The Binding of Anti-p80 and Anti-p30 to the Surface of Human B Lymphocytes 6 1 Expression of the P80 and P30 Determinants by Human Granulocytes 6 1 Screening of Human Lymphocytic Cell Lines With Anti-p80 and Anti-p30 6 6 Western Blot Analysis of Human Cell Lines 6 6 SECTION 2 - DISCUSSION: 7 8 SECTION 2 - CONCLUSIONS: 8 5 CLOSING DISCUSSION 8 7 LITERATURE CITED 8 8 LIST OF TABLES SECTION 1: I: SUPPRESSION OF THE PFC RESPONSE BY FPLC FRACTIONS SECTION 2: 11: SCREENING OF HUMAN PERIPHERAL LYMPHOCYTES FOR ANTI-P80/ANTI-P30 BINDING III: BINDING OF ANTI-P30/ANTI-P80 AND CELL SURFACE CHARACTERIZATION OF HUMAN LYMPHOCYTIC CELL LINES vii LIST OF FIGURES SECTION 1: 1a. INDUCTION OF SUPPRESSION IN NORMAL SPLEEN CELL PREPARATIONS BY M1-A5 SUPERNATANT 10 1 b. SUPPRESSION OF THE PLAQUE FORMING CELL ASSAY BY M1-A5 SUPERNATANT 1 2 2. GROWTH CURVES FOR WEHI-3B CELLS SEEDED AT DIFFERENT DENSITIES 1 8 3. SUPPRESSION OF KLH ANTIBODY PRODUCTION BY M1-A5 SUPERNATANT 2 0 4. THE EFFECT OF PREINCUBATION OF M1-A5 SUPERNATANT WITH KLH-IMMUNE SPLENOCYTES ON THE KLH- ANTIBODY CULTURE ASSAY 2 1 5. ELUTION AND BIOACTIVITY PROFILES FROM ION EXCHANGE CHROMATOGRAPHY OF M1-A5 SUPERNATANT: 2 4 5a. Profile from Round 1 of DEAE-Sephacel Chromatography 2 5 5b. Profile from Round 2 of DEAE-Sephacel Chromatography 2 6 5c. Profile from Round 1 of S-Sepharose Chromatography 2 7 6. SAMPLE FPLC ELUTION PROFILES OF POOLED, BIOACTIVE ION EXCHANGE FRACTIONS 2 8 6a. D1 Profile (Fraction #4-7, DEAE-Round 1) 2 9 6b. D2a Profile (Fraction #3-7, DEAE-Round 2) 2 9 6c. D2b Profile (Fraction #9-11, DEAE-Round 2) 2 9 7. SDS-PAGE OF SELECTED FRACTIONS OBTAINED FROM FPLC 3 2 SECTION 2: 8. SAMPLE PROFILE (a) OF NORMAL HUMAN MONONUCLEAR CELLS STAINED WITH NRS (b), ANTI-P80 (c) AND ANTI-P30 (d). 55 & 56 9. ISOLATED HUMAN CD3 + CELLS BIND ANTI-P80 AND ANTI-P30 5 9 PROFILES OF GaRIg (a), NRS (b), anti-p80 (c) and anti-p30 (d) 6 0 viii 1 0 . DOUBLE LABELLING OF HUMAN MONONUCLEAR CELLS WITH ANTI-CD19-PE AND ANTI-P80/ANTI-P30 6 2 PROFILES OF GaRIg (a), NRS (b), Anti-p80 (c) and Anti-p30 (d) 6 3 1 1 . CELLSUFACE ANALYSIS OF HUMAN GRANULOCYTES USING ANTI-P80 AND ANTI -P30 6 4 PROFILES OF GaRIg (a), NRS (b), Anti-p80 (c) and Anti-p30 (d) 6 5 12 . ANTI-P80 AND ANTI-P30 REACTIVITY WITH SEVERAL HUMAN LEUKEMIC CELL LINES 6 7 GATING (i) AND PROFILES OF NRS (ii), ANTI-P80 (iii) and ANTI-30 (iv) FOR: MOLT-4 (12a) 6 8 C E M - C M 3 (12b) 6 9 HuT 78 (12c) 7 0 RAJI (12d) 71 1 3 . TITRATION OF ANTI-P80 AND ANTI-P30 ON THE SURFACE OF LEUKEMIC HUMAN CELL LINES 7 2 TITRATION OF ANTI-P80 (13a) 7 2 TITRATION OF ANTI-P30 (13b) 7 3 14. SDS-PAGE OF HUMAN CELL LINE LYSATES 7 5 1 5 . WESTERN BLOTS OF HUMAN CELL LINE LYSATES WITH ANTI-P30 7 7 ix ACKNOWLEDGEMENTS First I would like to thank Dr. Julia Levy and all of the people in her laboroatory for their continual support and useful discussions regarding these experiments. Special thanks to Janice North and Doug Carlow for their help and advice in the cell surface studies, and to Dr. Barbara Pope and her laboratory for screening fractions by PFC Assay in Halifax. On a more personal note, I would like to pay tribute to my friends and family in Ontario who always manage to offer support and encouragement over the miles in trying times. And finally, thanks to D. Mitchell, for his friendship and constant faith in my abilities. 1 MAIN INTRODUCTION AN OVERVIEW OF IMMUNE FUNCTION The immune system has been functionally divided into the innate immune system and the adaptive immune system. In most cases, innate immunity comprises the first line of defense against invading microorganisms, however, if these defenses are overwhelmed or unable to function, the adaptive immune system is activated to fight the infection. The two distinguishing features of the adaptive immune system are specificity and memory, both of which are mediated by T and B lymphocytes. Activated B cells secrete bifunctional molecules called immunoglobulins (Ig) or antibodies (Ab) that bind to specific antigen (via the Fab domain), to Fc receptors on phagocytes and fix complement (via the Fc domain). T cells can be further subdivided into cytotoxic/ suppressor T lymphocytes (CTL/Ts) and helper/inducer T lymphocytes (Th). Upon activation, T cells secrete protein factors (lymphokines) some of which act nonspecifically and others that bind specific antigen. In concert, the products released by antigen activated lymphoid and accessory cells (such as antigen-presenting cells which present antigen to T and B lymphocytes) regulate both the level and type of response generated to a specific antigen. REGULATION OF IMMUNE RESPONSES BY T LYMPHOCYTES Although a great deal of effort has resulted in a widely accepted mechanisim of induction of immune responses, relatively little has been learned about the manner in which one particular response is chosen over another, eg. cell mediated (CTL, or NK cells) versus humoral (Abs), or the way in which reponses are suppressed in normal and pathological states. Most immune responses are the result of interactions between subsets of lymphocytes with regulatory and/or effector functions. T lymphocytes, in addition to their cytolytic functions (CTL), participate in immune response regulation by either inducing (Th), amplifiying (Th and Ts) or diminishing (Ts) specific immune cell responses. Analysis of human and murine T lymphocyte subpopulations has been facilitated by the availability of monoclonal antibodies that define subset-specific cell surface determinants. Use of these reagents coupled with flow cytometric techniques has allowed the detection, 2 quantitation and isolation of subsets of cells expressing one, two and sometimes three specific cell surface markers. The most important division of the human T cell population has been into CD4 + and CD8 + subsets (rev. in Reinherz et al., 1980), the murine homologs of which were originally designated L3T4 and Ly2 respectively (Dialynas et al., 1 983). By in vitro analyses the CD8+ subset was found to mediate most cytotoxic (CTL) and suppressor (Ts) functions and the CD4+ subset was shown to provide helper/ inducer effects required for the differentiation of B cells into Ig-secreting cells and also for the maturation of CD8+ CTL and CD8+ Ts (Reinherz et al., 1980). Although the correlation of CD4 /CD8 expression with biologic function has been useful, there exists a more consistent association with the class of major histocompatibility complex, or MHC (HLA in humans and H-2 in mouse) alloantigens used for antigen recognition (Englemen et al., 1981). Thus CD8+ T cells respond to antigen in association with Class I MHC proteins (A, B and C in humans and K, D and Qa in mouse) and CD4+ cells respond to antigen in association with Class II MHC proteins (DP, DQ and DR in humans and l-A and l-E in mouse). Characterization of the polypeptide hormones or lymphokines (LK) secreted by activated helper T cells (Th) and antigen-presenting cells (APC) has done much to advance our understanding of the inductive mechanisms for both T and B cell mediated immunity. Upon activation by antigen expressed in the context of class II MHC proteins, CD4+ T cells synthesize and secrete a variety of growth-promoting and differentiation-inducing lymphokines such as IL-2, IL-4, IFN-y and BCGF, whereas activated CD8 + CTL cells secrete products which are ultimately cytotoxic to their targets and CD8 + Ts secrete product(s) that have yet to be clearly characterized. IMMUNOSUPPRESSION The topic of immunosuppression is currently controversial (Moller, 1988 for eg.). Although it has been repeatedly demonstrated in vivo and in vitro, and in many different systems, that suppression of the immune system exists, it is curious that the mechanism(s) remain elusive. If taken together, all of the published work on immunosuppression suggest that it is a very heterogeneous phenomenon. Reports have shown that it involves dispartate molecules and cells ranging from antigen-specific suppressor factors and antigen-specific T cells to common cytokines and nonspecific 3 protein factors elicited from non-lymphoid hemopoietic cells (rev. in Teillaud and Fridman, 1989). Briefly, the major problems encountered in studies on suppression have been in cloning suppressor cell lines, isolating and characterizing suppressor factors and identifying distinct cell surface determinants to further define their population. AN O V E R V I E W O F R E S E A R C H T O B E P R E S E N T E D In this work, different aspects of two distinct yet related types of suppression were studied. First, an attempt was made to purify and partly characterize an antigen-nonspecific suppressor inducer factor (SIF) secreted by a third population LGL cell line (M1-A5) derived from the spleen of a tumor-bearing mouse. Second, two antisera (anti-p80 and anti-p30) raised against reduced components of a partially characterized murine antigen-specific suppressor factor (Fd11F) were used to determine their reactivity with normal human peripheral lymphocytes and several lymphoid leukemic cell lines and cell lysates. As the work to be presented involves research within two different areas of immuno-suppression this thesis has been divided into two main sections entitled \"Antigen Nonspecific Suppression\" (Section 1) and \"Antigen-Specific Suppression\" (Section 2). 4 S E C T I O N 1: A N T I G E N - N O N S P E C I F I C S U P P R E S S I O N INTRODUCTION AN OVERVIEW OF ANTIGEN-NONSPECIFIC. OR NATURAL SUPPRESSION. Natural suppression (NS) is defined as the ability of an LGL-like population of null cells to suppress immune functions in an antigen nonspecific, MHC nonrestricted manner (Saffran and Singhal, 1990 and Maier era/, 1986). The cell surface phenotype of NS cells derived from various locations was found to be similar to that reported for cloned natural killer (NK) cells (Thy-1 + , Ly-5 +, asialo-GM1 + , slg-, Lyt-1\", Lyt-2\", L3T4\", la\" , MAC-1') (Brooks et al, 1982 and Schwadron et al., 1989), yet cloned NS cells are unable to kill NK cell targets such as YAC-1 (Hertel-Wulff er al, 1984 ). A recent report attempting to sort out the relationship between NK and NS cells was only able confirm that these two cell types are related yet distinct (Saffran and Singhal, 1990). To further confuse the origin of NS cells, several cloned NS cell lines have been shown to harbor rearranged and expressed T cell receptor (TCR) genes. Thus even though the cell surface phenotype of NS cells is distinct from that of any mature T cell characterized to date, it is now assumed by some groups that NS cells are of the T cell lineage (Hertel-Wulff et al., 1987). SITES OF NS CELL ACTIVITY NS activity was observed first in murine (Singhal et al, 1972 and Duwe and Singhal, 1979) and later, human (Bains et al, 1982 and Mortari et al, 1986) bone marrow (BM). BM-derived natural NS cells were reported to inhibit the in vitro generation of IgM Ab responses (Mortari et al., 1986 for eg.), mitogen activation (Maier et al, 1985) and MLR's (Sugiura et al, 1988 for eg.). In vivo, BM-derived NS cells were able to prevent graft versus host disease (GVHD) during the generation of murine-mixed allogeneic chimeras (Sykes et al, 1988). Cells with the functional and phenotypic characteristics of NS cells have been found in a number of locations and resulting from a variety treatments (rev. in. Maier er al., 1989) Definite sites in addition to bone marrow (BM), include fetal and newborn 5 lymphoid tissues, and spleens after total lymphoid irradiation (TLI), cyclophosphamide (CY) treatment and during chronic graft versus host disease (GVHD). NS cell activity has also been reported to be associated with processes such as pregnancy, tumor progression and treatment with radioactive Strontium ( 8 9 Sr) . All of the locations in which NS cell activity has been observed have two things in common. The first is that they are all sites of hematopoiesis. For example, fetal and newborn liver and spleen, as well as BM throughout life, are hematopoietic organs, and after TLI, spared progenitors from BM auto-transfuse the spleen which subsequently becomes a site of rapid cell turnover. Also, during GVHD in irradiated recipients, there is considerable proliferation of donor cells in the host spleen and after 8 9 S r treatment which detroys BM, the spleen takes over as the primary hematopoietic organ. Finally, some tumours cause increased proliferation of certain stem cell populations which may correlate with the presence of NS cells. Thus, one function of NS cells may be to regulate cell growth in hematopoietic organs - possibly by suppression of growth factor production or target cell activation (Soderberg, 1985). The second characteristic which most of the sites of NS cell activity have in common is that they are locations in which the induction of tolerance can occur. It is known that the fetal/newborn period is the point in development where self-tolerance is established and aquired tolerance can be induced. (Billingham etal., 1953 and Streilein, 1979). Also, it has been suggested that BM is a site throughout life where tolerance can be induced (Hurme and Sihvola, 1985 for eg.). During pregnancy, there exits a transient state of tolerance between the mother (host) and fetus (graft) and immunosuppressive and/or cytoreductive treatments (such as TLI or CY) produce environments conducive to tolerance induction. In contrast, reasons for the state of tolerance that has been observed in some tumor-bearing animals have not been offered. POSSIBLE MODE OF ACTION OF NS CELLS NS cells have been reported to be stimulated by T cell lymphokines such as IFNy (Maier et al., 1985b and 1986 and Holda et al., 1986), thus it can be theorized that T cells activated (in the appropriate environment) by self or foreign antigens release lymphokines which stimulate NS cells. NS cells in turn, may mediate the suppression of the development of mature T cell clones directed against that antigen. Some experiments have shown that Ts (T suppressor) cells are refractory to NS cell-mediated down 6 regulation, whereas Th (T helper) cells are responsive (Okada et al., 1982). This would prevent sensitization to the antigen (Th) and ensure nonresponsiveness during future antigenic challenges by allowing the expansion of antigen-specific Ts clones. ANTIGEN-NONSPECIFIC SUPPRESSOR FACTORS A few groups have described the existence and purification of antigen-nonspecific immunosuppressive factors released from NS or other cell types. Two examples of purified nonspecific suppressor factors are SIRS (Soluble Immune Response Suppressor) and IBF (Immunoglobulin Binding Factor). Both are released by murine activated T cells (SIRS specifically by Ly 1\"2+ T cells) and nonspecifically inhibit in vitro antibody production to both T-dependent and T-independent antigens. SIRS also inhibits in vitro B cell proliferation but does not inhibit CTL induction or MLR responses (Aune and Pierce, 1984). In addition, SIRS has been shown to have a molecular weight between 48 and 67 KD and require macrophage oxidation for activation. IBF has been demonstrated to consist of FcR's shed from the surface of alloantigen-activated, FcR + T cells (Lowy, er al., 1983). This factor (chains of 38 and 18 KD) mediates suppression by binding to the Fc portion of IgG and blocking complement activation. There have also been examples of NS cell activity shown to be mediated by a secreted product. One example is the NS factor described by Hertel-Wulff and Strober (1988). A cloned NS cell line was isolated from the spleens of mice given TLI. Upon activation, these cells released a protein factor between 50 and 100 KD which was able to suppress the murine MLR. This NS factor was shown to have a mode of action which lacked antigen specificity and MHC restriction. In addition, it did not appear to affect B cell prolifer-ation (the factor was unable to suppress mitogenic B cell stimulation). Although there have been reports on a diverse array of antigen-nonspecific suppressor factors, there has yet to be agreement on the molecular nature and function of these cytokines in the suppression of immune responses. Few believe that natural suppression is a phenomenon acting in isolation. Rather, it is thought that this level of immune regulation acts in concert with that of antigen-specific suppression mediated by Ts cell circuits and their soluble products. 7 NS CELL ACTIVITY IN RESPONSE TO TUMORS NS cell activity has been shown on several occasions to be associated with a decreased immune response to tumours. TLI and GVHD-induced NS cells have been demonstrated to inhibit the in vitro generation of tumour-specific CTL (cytotoxic T lymphocytes) (Okada and Strober, 1982 for eg.). In addition, natural killer (NK) cells, which are thought to be important in limiting cancer metastases in vivo, seem to exist in a reciprocal relationship with NS cells (Maier et al., 1990). Thus, NS cells may suppress NK cell activity, or in light of their cytological and phenotypic similarities, may represent a different stage of differentiation of the NK cell lineage. Although antigen-specific suppressor circuits have been described in tumor-bearing animals (Schatten et al., 1984 and Green era/., 1983), less is known about antigen-nonspecific or natural suppression (NS). Several cell types such as B cell blasts, tumor-activated T cells and non-T cells from tumor-bearing mice have the ability to induce NS cells and it is postulated that this results in an additional level of control superimposed on antigen-specific circuits (rev. in Pope, 1985). SUPPRESSOR INDUCER CELLS As part of an ongoing study of suppressor T cell circuits associated with the growth of M-1 (methylcholanthrene-induced) fibrosarcomas (Pope et al., 1976 and 1978), Pope described the existence of suppressor inducer cells that secrete soluble factor(s) capable of recruiting and activating suppressor effector T cells from unprimed splenocytes. The activated effector cells were found to be Thy-1+, Lyt 1 +2 + and had the ability to suppress the anti-IgM response to a T cell-dependent antigen (SRBC). As the antibody response to a T cell independent antigen (DNP-LPS), and that mediated by Nu/Nu spleen cells (activated by exogenously added growth factors), were not suppressed, Pope suggested that the mode of action of activated Ts might be inhibition of the production of lymphokines by Th cells. This theory has since been supported by others working in different suppressor systems and with C a + + flux studies (Malkovsky et al., 1982 for eg. & Utsunomiya et al., 1990). 8 I S O L A T I O N O F T H E N S C E L L L INE M1-A5 Almawi and Pope in 1985, reported the isolation of an antigen-nonspecific suppressor inducer cell line (M1-A5). M1-A5 was generated from the spleen of a mouse bearing an advanced M-1 fibrosarcoma. When 2 x 104 M1-A5 cells were cocultured with 106 normal spleen cells (NSC), and 5 x 105 sheep red blood cells (SRBC) for 5 days, subsequent challenge with SRBC resulted in a 99% suppression of antibody synthesis (as assessed by the plaque forming cell (PFC) assay of Cunningham and Szenberg, 1968). This effect was shown to titrate with decreasing numbers of M1-A5 cells cocultured. M1-A5 cells displayed a similar morphology (large granular lymphocyte or LGL) and cell surface phenotype (Thy-1-, Lyt-1 \"2\", slg*. MAC-1\", Ly-5+ and asialo-GM1+) to natural suppressor (and NK) cells. In addidtion, M1-A5 was able to lyse the natural cytotoxic (NC) cell target WEHI-164, and to a small extent cultured M-1 fibrosarcoma cells, but not the NK cell target YAC-1. Tritiated thymidine uptake experiments showed that M1-A5 cells required IL-3, but not IL-1, IL-2 or GM-CSF to proliferate. Cell-free supernatant (shown to be nonsuppressive) from the murine, IL-3-producing, promyelocytic cell line WEHI-3B (Lee et al., 1982), was used to support the growth of M1-A5 cells cultured in vitro. IN VITRO B I O A C T I V I T Y O F M1-A5 C E L L S In order to investigate whether the mechanism of suppression by M1-A5 cells involved direct cellular contact or the secretion of soluble suppressor factor(s), Marbrook chambers were used. M1-A5 cells (in the outer chamber) were separated from NSC (in the inner chamber) by dialysis tubing (mol. cutoff of 12 KD). After 20 hours, potentially activated NSC were then cocultured with SRBC and assayed for antibody producing cells as above. It was found that 2.5 x 106 M1-A5 cells incubated in the outer chamber were able, via the secretion of a factor of less than or equal to 12 KD, to suppress the PFC response of 106 NSC by approximately 50% (Almawi and Pope, 1985). Further, increasing numbers of M1-A5 cells activating the same number of NSC resulted in a greater decrease in PFC response. Additional experiments showed that a minimum preincubation of 8 hours was needed for M1-A5 cells to activate NSC in Marbrook vessels. Longer incubation periods resulted in only a slight enhancement of suppression. Also, incubating NSC for a minimum of 3 9 hours in 50% dialyzed M1-A5 supernatant (SN) resulted in 50% suppresssion of the PFC response, an effect which titrated through to 90% SN (Fig. 1a and Discussion). Finally, the suppressor inducer factor (named SIF) was shown to be protein in nature as a loss of bioactivity was observed upon heat, acid, base and trypsin treatments (Almawi and Pope, 1 985). In 1987, Pope further characterized the range, potency and kinetics of SIF activity. She found that M1-A5 cells and S/N were able to 1)suppress the generation of CTL's, 2)inhibit T cell proliferation in reponse to alloantigens (MLR) and 3)suppress the generation of antibody producing cells (PFC). However, neither M1-A5 nor SIF were able to suppress the proliferative T cell response to Concanavilin A (Con A). Suppression of the MLR, the PFC response and the generation of CTL was only observed when M1-A5 cells or SN was added early (in the first day of culture for MLR and PFC and at the start of culure for CTL). In the same report, the mechanisim of M1-A5-mediated suppression was shown to differ from another IL-3 dependent cell line FDC-P1. FDC-P1 was directly suppressive in all of the assays but the spent SN from these cells was completely inactive. Almawi and Pope (1987) found that addition of acetylsalicylic acid (ASA) or ibuprofen (which blocks prostaglandin synthesis) to M1-A5 cells in culture effectively abolished the ability of M1-A5 cells and SN to induce suppression. Exogenously added prostaglandins PGE-|, E 2 and l 2 (but not F 2 a or D2), were found to restore the ability of M1-A5 cells to mediate suppression but could not replace the suppressor inducer activity of SN from ASA or ibuprofen treated cultures. Further, Almawi etal. (1987a) found that cyclic AMP was the second messenger in the PGE-mediated modulation of SIF secretion. BIOCHEMICAL PURIFICATION OF MI-A5 DERIVED SUPPRESSOR INDUCER FACTOR (SIR Almawi et al., (1987b) initially purified 100 u.g of SIF from 6 liters of M1-A5 cell free SN by Sephadex G-100 size fractionation and ion exchange chromatography. Significant levels of bioactivity were found in peaks eluting around M r 70 KD, termed SI Fa , and M r 5.5 KD, termed SIFfi. When SIF a and p* were subjected to ion exchange chromatography, SIFfJ was shown to bind to the cationic exhanger QAE-Sephadex and elute in a single peak, whereas SIFoc activity was found in the void volume in addition to the eluate of a DEAE-Sephacel column. Reducing SDS-PAGE of the bioactive QAE fraction (SIFfJ) showed one protein band with a M r of approximately 6KD. Overall, the purifi-10 F I G U R E 1a : I N D U C T I O N O F S U P P R E S S I O N I N N O R M A L S P L E E N C E L L P R E P A R A T I O N S B Y M1-A5 S U P E R N A T A N T . Taken from Pope et al., 1987. 5 X 10 5 NSC were incubated with increasing amounts of dialyzed (Mol. wt. cutoff-12 KD) M1-A5 supernatant for at least 4 hours, cultured in vitro with 1 X 10 6 untreated NSC and S R B C for 5 days and assayed for the number of plaque forming cells per culture by the method of Cunningham and Szenberg (1968). Percent suppression was calculated from cultures in which NSC had been incubated with RPMI medium instead of supernatant. Each point represents the mean of three independent tests. 11 cation resulted in a 104 fold increase in activity for SIF(3 and only a 4 fold increase in activity for SIFct. In the same report, SIFoc and p\" were demonstrated to have a genetically nonrestricted mode of action and the equal ability to suppress the antibody response of unprimed splenocytes to SRBC in vivo (Almawi et al., 1988a). Also in 1988 (b), Almawi et al. reported on the large scale purification of SIFct and (3 from 15 litres of M1-A5 SN using subsequent steps of Sephadex G-100 gel filtration, cationic (a) and anionic (|3) exchange chromatography and two rounds of reverse phase high performance liquid chromatography (RP-HPLC). Fractions were screened by a modified PFC assay which resulted in an immense increase in sensitivity (Fig. 1b and Discussion). Overall, 40 u,g of SIFp* with a 3 x 104 fold increase in activity and 5 u.g of SI Fa with a 1.6 x 103 fold increase in activity were purified. Final purity was assesed by the migration of the factors as a single band on SDS-polyacrylamide gels and elution from the HPLC column as sharp, single peaks. SECTION 1: SUMMARY OF WORK TO BE PRESENTED The above purifcation process was modified (see below) and set up in our laboratory with the hope of obtaining enough purified SIFP to get the protein sequence from an internal peptide (in the previous purification, the N-terminus was shown to be blocked) and to begin the production of a monoclonal antibody in rat. Simultaneously, an in vitro assay already in use in the laboratory to study suppression of the antibody response to the antigen ferredoxin was adapted for studying SIF action. 12 LOG DILUTION OF M1-A5 S/N FIGURE 1b: SUPPRESSION OF THE PLAQUE-FORMING CELL ASSAY BY M1 -A5 SUPERNATANT. Taken from Almawi ef al., 1988b. Naive spleen cells were cultured in vitro with SRBC and log dilutions of M1-A5 supernatant for five days and then assayed for the number of plaque forming cells per culture as above Percent suppression was calculated from cultures in which RPMI media was added instead of suppressor sample. Each point represents the average of three independent experiments. 13 SECTION 1: MATERIALS AND METHODS CELL LINES; M1-A5 and WEHI-3B cells were obtained from the laboratory of Dr. B. Pope (Dalhousie Univeristy, Halifax, Nova Scotia) and were grown in Flow RPMI 1640 supplemented with 1% and 5% heat inactivated fetal bovine serum respectively in a humidified, 37°C, 5% C 0 2 incubator. M1-A5 growth media was supplemented with 20% WEHI-3B cell-free supernatant as a source of IL-3. Both cell lines were grown in large (175cm2) Nunc tissue culture flasks; M1-A5 upright and WEHI-3B flat. HARVESTING SUPERNATANTS: When cultured WEHI-3B cells reached a density of 5 x 10 5 cells per ml, the cell suspension was centrifuged and the supernatant filtered (0.2 UMTI filter) for addition to the M1-A5 growth media. Excess supernatant was occasionally frozen at -70°C for future use. When the M1-A5 cells reached a density of 5 x 105 - 1 x 10 6 cells per ml, the supernatant was aseptically removed from the cells resting on the bottom with a 25 ml pipette. Cells were harvested, washed three times with sterile PBS, resupended at 5 x 105 cells per ml in serum-free RPMI and incubated at 37°C, 5% C 0 2 overnight. The next day cell free supernatant was harvested and frozen at -70°C until approximately 3.0 litres was collected. BIOCHEMICAL PURIFICATION OF SIFa AND SIFp FROM M1-A5 SN: M1-A5 serum free SN was thawed at room temperature and applied via gravity feed to DEAE-Sephacel and S-Sepharose ion exchange columns (Pharmacia) set up in tandem and equilibrated with PBS at 4°C. Columns were washed separately with an equal volume of cold PBS and eluted with 150 ml continuous salt gradients (0.15M to 0.95M NaCl) at 0.5-0.7 ml/min. 10 ml fractions were collected with a Pharmacia Frac 100 fraction collector and the amount of protein in each determined by Micro-Lowry Assay. The fractions were dialyzed at 4°C against PBS and tested for bioactivity by the Plaque Forming Cell (PFC) Assay, (B. Pope, Halfax). Active fractions were pooled, concentrated on a bed of sucrose (Fisher Spectra Por dialysis tubing, MWt. Cutoff of 3500 Dal) and back dialyzed in 50mM N H 4 H C O 3 (2L x 12 hours twice). Pooled fractions were lyophilized, and resuspended in PBS. The pooled, concentrated, bioactive fractions were next subjected to FPLC at room temperature. Approximately 4.5 mg of protein was applied to an FPLC Superose 12 HR column (Pharmacia) per run. The column was eluted with 30.0 ml of PBS and 0.5 ml fractions were collected and pooled from three to four 14 successive runs. Fractions were frozen at -70°C and assayed for bioactivity by PFC (Dr. Pope). SODIUM DODECYL SULFATE - POLYACRYLAMIDE GEL ELECTROPHORESIS (SDS-PAGE): The content and purity of selected fractions obtained from FPLC was assessed using 12% (w/v) acrylamide (Bio-Rad) gels in a mini-protean II slab gel apparatus (Bio-Rad). Samples were dialyzed against 50 mM ammonium bicarbonate, lyophilized and approximately 2.0 u.g was resuspended in 10-15 U.I of SDS sample buffer (10% glycerol, 2% SDS, 6.25 mM Tris-HCl buffer and bromphenol blue). Material was reduced by the addition of dithiothreitol (DTT) to 10 mM, and heating to 100°C for 4-5 minutes, and then cooled to room temperature for loading. Separated proteins in the gel were fixed in 50% methanol/10% glacial acetic acid/40% dH 20 for 30 min. and rehydrated by incubation in 10% methanol/10% glacial acetic acid/80% dH 20 (in a microwave for 1 min.) followed by 100% dH20 for 5 minutes at room temperature. The gel was next treated with 33 u,l of 100 mM DTT in 100 ml dH20 for 1 min. in a microwave and then 1 ml of 10% silver nitrate in 99 ml dH20 for 15-20 min. at room temperature. Silver-binding material was visualized by the addition of 40-50 ml of developing reagent (40 ml of 15% Na 2C0 3/60 ml of H20 and 200 u,l of 37% formaldehyde) for a variable period of time. Developing was stopped by the addition of 3-4 ml of glacial acetic acid. Gels were stored in dH20 and subsequently dried on a Bio-Rad slab gel dryer. SCREENING FOR SIF BIOACTIVITY -THE MODIFIED SUPPRESSOR CELL ASSAY (Dr. B. Pope. Halifax, N.S.): Fresh sheep red blood cells (SRBC) were collected aseptically in Alsever's solution and stored in aliquots at 4°C for up to one week. Before use SRBC's were washed 2-3 times (or until supernatant was clear) with PBS. Fresh guinea pig serum was used as a source of complement after being adsorbed with 1 ml of packed SRBC per 10 ml serum for 30 minutes on ice. The SRBC were removed by centrifugation and the serum was stored in aliquots and frozen at -70°C. Spleens from 6-8 week old BALB/c or DBA/2J mice were removed aseptically, pressed through a metal screen into 10 ml of serum-free RPMI, washed once (10 min. x 400 g), and resuspended in 5 ml of RPMI per spleen. After a viability count, cells were diluted to 1.2 x 107 per ml in RPMI. Suppressor samples were added to the wells of flat-bottom, 96-well microtiter trays (Dynatech-lmmulon 2) in duplicate or triplicate and brought to 100 ul with 15 medium. 100 jil of medium alone was added to positive control wells. 100 ul of the normal spleen cell (NSC) suspension (1.2 x 106 cells) and 50 (il of a mixture of 50% FBS, 25% RPMI 1640, 12.5% 2 mM 2-mercaptoethanol and 12.5% SRBC (at 4 x 107 cells per ml) were added to bring the final volume per well to 250 ul. The cell mixture was then incubated at 37°C, 5% C 0 2 for 5 days and assessed by PFC assay (see below). PLAQUE FORMING CELL (PFC) ASSAY: (Cunningham and Szenberg, 1968) SRBC-primed SC (from above) were harvested from wells with a pasteur pipette and duplicate or triplicate tests pooled. After centrifugation (10 min x 400g) the pellet was resuspended in 300 or 450 ul (150 ul per well) of PBS. 25 ul of this mixture was diluted with medium to give a volume of 150 ul. 50 ul of a 25% suspension of washed, packed SRBC in RPMI + 5% FBS and 50 ul of guinea pig complement diluted 1:1 in RPMI were added to give a final volume of 250 ul. 65 ul of this final solution was added to previously prepared Cunningham slides which were then sealed with melted vaseline to prevent evaporation. Slides were incubated at 37°C for 1 hour and plaques counted against the light of an X-ray box. The number of PFC per culture was determined by multiplying the number of PFC per chamber (mean of 2-3 chambers) by the appropriate dilution factors. ANTIBODY CULTURE (Ab C) ASSAY: 6-8 week old, female DBA/2J mice were primed subcutaneously (s.e.) with 50 ug of kehole limpet hemocyanin (KLH, Calbiochem) emulsified in an equal volume of complete Fruend's Adjuvant (CFA). After 21 days, animals were boosted s.e. with 50 ug KLH emulsified in incomplete Freund's adjuvant. Serum anti-KLH levels were evaluated by ELISA as follows; Dynatech-lmmulon 2, 96-well microtiter plates were coated with 100 ng KLH per well in sodium bicarbonate buffer (left at 4°C, overnight). Serum obtained from tail vein blood of primed animals was added at various dilitions in duplicate to the KLH-coated wells and incubated for 1-1.5 hours at 37°C. Plates were developed using goat anti-mouse antibody conjugated to alkaline phosphatase (37°C x 1 hour, Jackson Labs) and paranitrophenyl-phosphate (Sigma) substrate (37°C x 15-30 min), and read on a Titertek Microplate Reader at 405 nm. The spleens of animals with the highest serum levels of anti-KLH antibodies were aseptically harvested and pressed through a sterile metal screen with the aid of the plunger from a 3 ml syringe into 10 ml of DME with 10% FBS. Splenocytes were pelleted, resupended in 10 ml medium and divided into two 5 ml portions. One portion recieved a 250 ug per spleen per ml in vitro pulse of KLH in 16 medium and the other an equal volume of media alone. Suspensions were incubated for 2 hours at 37°C with 5% C 0 2 and periodic tapping to resuspend settled out cells. After two washes with medium and a viability count, cells were resuspended in RPMI 1640 with 10% FBS and 5 x 10\"5 M 2-mercaptoethanol at a density of 2.5 x 106 cells per ml. 2 ml (5 x 106 cells) of each suspension (KLH-pulsed and media-pulsed) were added to sterile Nunc 24-well culture plates. An aliquot of M1-A5 serum-free supernatant was added to both the KLH-pulsed and medium-pulsed wells. An equal volume of medium alone was added to cells in duplicate and constituted positive controls. Plates were incubated for 7 days in a humidified, 37°C, 5% C 0 2 incubator. On day 7, 1.5 ml of culture supernatant from each well was removed and stored in 0.02% sodium azide at 4°C, or frozen at -20°C. The level of anti-KLH antibody produced was evaluated by testing 100 u.l of each supernatant in a KLH ELISA (as described above). To test the relevance of time of factor addition to the assay, serum-free supernatant was incubated with KLH-primed splenocytes for 6 hours at 37°C before the KLH or medium pulse. MYCOPLASMA TESTING AND CELL LINE CURING: Cells in the laboratory were periodically tested for Mycoplasma infection by cytochemical analysis using the flourescent dye 4-6-diamine-2-phenyl indole dihydrochloride or DAPI (Boeringher Mannheim). Dr. Pope's laboratory, in addition to the DAPI method, occasionally sent cell lines to the Victoria General Hospital Diagnostic Laboratory (Halifax, Nova Scotia) for their more sensitive Mycoplasma detection system. The cell pellet from a 150 ml culture of cells was washed, resuspended in Mycoplasma growth media and after an appropriate incubation period, assayed for the presence of the organism by plating out samples of supernatant. Three M1-A5 subcultures were cured of Mycoplasma using the BM-Cycline (Boeringher Mannheim) antibiotic kit. Samples of treated cultures were sent to the General Hospital to determine the effecincy of the procedure by the method described above. 17 SECTION 1: RESULTS CULTURING WEHI-3B AND M1-A5 CELL LINES: M1-A5 cells have been shown to require the hemopoietic growth factor lnterleukin-3 (IL-3)(Pope et al., 1986). The mouse promyelocytic cell line WEHI-3B (Lee et al., 1982) constitutively secretes IL-3; thus, filtered WEHI supernatant was used to support the growth of M1-A5 cells in vitro. During the large scale culturing of M1-A5, it was observed that if the WEHI cells were allowed to reach a density exceeding 1 x 106 per ml they began to secrete a substance which initially inhibited cell growth and finally caused autolysis and lysis of the M1-A5 cells. This was not due to exhaustion of the nutrient supply in the growth medium, as seeding WEHI cells into fresh medium at too high a density (5 x 105 cells per ml) did not produce a plateau in the exponential decline of the growth curve. Fig. 2 shows several growth curves for the WEHI-3B cell line. Each curve represents a different seeding density at day 0. The culture which was started with 5 x 105 cells per ml immediately began an exponential decrease in cell number with virtually no viable cells remaining after 3 days. The culture seeded at 1 x 105 cells per ml gave a growth curve which increased to 7 x 105 cells per ml (day 2), plateaued (day 2-4) then exponentially decreased. Those cultures seeded at 1 x 104 and 5 x 104 cells per ml both displayed normal exponential increases in cell number to 7 x 105 per ml then declined as that observed with the culture seeded at 5 x 105 cells per ml. From these curves, and assessing the response of the M1-A5 cells to various samples of WEHI-3B supernatant (data not shown), it was determined that the optimal conditions for harvesting WEHI supernatant was from cultures seeded at 1 or 5 x 104 cells per ml and allowed to reach a density of 5 x 105 cells per ml. SIF ACTIVITY IN THE ANTIBODY CULTURE ASSAY: The Antibody Culture (Ab C) Assay has been in use in this laboratory for several years to screen for suppressive ability of the antigen-specific, murine suppressor factor Fd11F. Fd11F is specific for the nominal antigen ferredoxin, thus the mouse immunization schedule and Ab C Assay have been developed to optimize production of anti-Fd antibodies by Fd-immune splenocytes (R. Chu, 1989 for eg.). Since SIF has previously 18 DAYS AFTER SEEDING FIGURE 2: GROWTH CURVES FOR WEHI-3B CELLS SEEDED AT DIFFERENT DENSITIES. WEHI-3B cells were cultured in RPMI supplemented with 10% FCS in a humidified, 37°C, 5% C 0 2 incubator. Cell viability was assessed and quantitated by light microscopy using a hemocytometer and Eosin Y stain every 24 +/- 2 hours. The number of cells per ml plotted at each 24 hour time interval represents the average of four values. 19 been shown to have a mode of action that is antigen-dependent yet nonspecific, the molecule kehole limpet hemocyanin (KLH) was chosen (for it's availability and immunogenicity) as the antigen to adapt the Ab C Assay to study SIF activity. KLH-primed mouse splenocytes were pulsed in vitro with KLH or medium alone and subsequently cultured for 7 days with serum-free M1-A5 or control medium. Anti-KLH levels were evalulated by ELISA using KLH-coated plates. As can been seen in Fig. 3, M1-A5 serum-free supernatant suppresses 11% of the anti-KLH response at a dilution of 10\"8 and 93% at a dilution of one in 4. From this suppression curve, an E D 5 0 of 10\"2 was obtained. Thus, the sensitivity for detecting SIF activity in the antibody culture assay is 100 times more sensitive than that obtained by Dr. Pope in the original suppressor assay (Fig. 1a) and 10,000 fold less sensitive than the straight PFC assay used to screen for SIF activity in the biochemical purification (Fig. 1b). Although the suppressive effect does titrate in all three of the assays, the shape of the curves differ significantly (FIG.'s 1a, 1b & 3). PREINCUBATION OF KLH-IMMUNE SPENOCYTES WITH M1-A5 SN To investigate whether the SIF-induced suppression of the KLH Ab response was dependent on the incubation time of factor with splenocytes prior to antigen pulse, KLH-immune splenocytes were divided into two equal portions. One portion was preincubated with suppressor sample for 6 hours prior to the in vitro KLH pulse and the other was given the antigen-pulse directly and then cultured with SIF sample. The 6 hour preincubation time was chosen because Pope et al. (1987) have shown that a minimum of 8 hours is required for washed M1-A5 cells to synthesize and secrete enough SIF to activate the suppressor population of NSC in Marbrook vessels and that SIF already in SN only needs 4 hours. Since in this experiment the SIF sample is already in the supernatant, 6 hours was chosen to ensure more than enough time for induction of suppression in the assay. After 7 days, anti-KLH levels were compared for both groups of cultured splenocytes. From the results presented in Fig. 4, the E D 5 0 of the supernatant not preincubated was 5 x 10 - 3 whereas the maximum level of suppression achieved with the preincubated sample was 22% at the lowest dilution (0% at all other dilutions). Since the positive control anti-KLH levels from cells preincubated and not preincubated with M1-A5 supernatant ( O D 4 0 5 of 0.82 vs. 0.89 in the KLH ELISA respectively) differed by only 8%, the 6 hour preincubation step is not significantly affecting the functional capability of KLH-immune splenocytes. 20 100 n L O G DILUTION OF M1-A5 SUPERNATANT FIGURE 3: SUPPRESSION OF KLH ANTIBODY PRODUCTION BY M1-A5 SUPERNATANT. The spleens of KLH-immune mice were harvested, mashed through a metal screen into DME/10% FCS media, pulsed with KLH for 2 hours and washed. Activated splenocytes were incubated in duplicate with various dilutions of M1-A5 supernatant for 7 days at which point culture supernatants were collected and the amount of anti-KLH antibodies in each quantitated by KLH-ELISA (each dilution tested in triplicate). Percent suppression was calculated from cultures pulsed with antigen but incubated with DME media instead of supernatant. Each point represents the average of the values obtained in two separate experiments. 21 100 -i 80 -(f> (/> ui CC Q. a. (/> z 111 o CC LU CL 60 -40 -20 -+ 8HR PREINC. NO PREINC. LOG DILUTION OF M1-A5 SUPERNATANT FIGURE 4: THE EFFECT OF PREINCUBATION OF M1-A5 SUPERNATANT WITH KLH-IMMUNE SPLENOCYTES ON THE KLH ANTIBODY CULTURE ASSAY. KLH-immune splenocytes were harvested and split into two portions. One portion was preincubated with M1-A5 supernatant for 8 hours prior to the KLH pulse and the other was pulsed, washed and put into culture with M1-A5 supernatant directly. All cultures were set up in duplicate. After 7 days, culture supernatants were analzyed as in FIG. 3. Percent suppression of preincubated SC was calculated from cultures in which SC were preincubated with DME media instead of supernatant and that of untreated SC, as in FIG. 3. Each point represents the results of one experiment. 22 BIOCHEMICAL PURIFICATION OF SUPPRESSOR INDUCER FACTOR (SIR FROM M1-A5 SERUM-FREE SUPERNATANT: The previously described SIF purification process (Almawi etal., 1988b) was set up in our lab in an attempt to obtain enough purified SIFp from M1-A5 cell culture SN's to synthesize a monoclonal antibody in rat. Several modifications were made to this procedure. First, the initial Sephadex G-100 size separation (gel exclusion chromatography) was omitted and serum-free M1-A5 SN was applied directly to the ion exchange columns. The rationale for this change was the high binding capacity and resilience of modern ion exchange column matrices and charged groups. The second alteration replaced three rounds of reverse-phase high pressure liquid chromatography (RP-HPLC) with one round of FPLC. This change was made because purified SIFJ3 was found to be blocked at the N terminus when sent for protein sequencing -presumably due to the harsh conditions imposed by the acetonitrile elutions. FPLC fractions were eluted with PBS. M1-A5 cells grown in 75% RPMI 1640, 5% FBS, 20% WEHI-3B cell-free SN (as a source of IL-3) were harvested, washed with PBS and incubated at a density of 5 x 106 cells/ml for 16-20 hours in serum free RPMI 1640. Once 2.5-3.0 litres of cell- and serum free M1-A5 supernatant had been collected it was passed over anionic and cationic ion exchange columns set up in tandem. The anionic (upper) exchange column consisted of DEAE-Sephacel and had been previously shown to bind SIFfJ but not SIFct (W. Almawi, unpublished). The flowthrough from this column was directed over the cationic (lower) exchange column containing S-Sepharose which was known to bind SIFcc. Columns were washed separately with PBS and eluted with 150 ml 0.15M to 0.95M NaCl continuous salt gradients. Ten ml fractions were collected and tested for protein content by Micro-Lowry Assay. One large protein peak eluting betweeen 0.31 M and 0.57M NaCl (Fractions #4-7) was obtained in the first round with a shoulder at 0.63M (Fraction #8) and two smaller peaks eluting between 0.31-0.42M (Fraction #4) and 0.42-0.57M NaCl (Fractions #6,7) were obtained in the second round of DEAE-Sephacel chromatography. One very large protein peak eluted from the S-Sepharose column between 0.42 and 0.84M NaCl (Fractions #6-12) with a shoulder at 0.36 (Fraction #5). The second set of S-Sepharose fractions unfortunately were lost. Technical difficulties were experienced in attempts to set up the PFC Assay practised in the laboratory of Dr. B. Pope at Dalhousie University. In order to continue with the 23 purification process as speedily as possible, aliquots of the 10 ml fractions eluted from ion exhange columns were dialyzed against PBS and sent to Dr. Pope's laboratory where they were screened for bioactivity by PFC assay. The elution profiles, and bioassy results for both the DEAE-Sephacel (SIFP) and S-Sepharose (SIFa) runs are shown in Fig. 5a, b, and c. In Fig. 5a, it can be seen that the highest levels of bioactivity are found within the main protein peak (Fractions #4-7) but not in the shoulder (Fraction #8). In Fig.5b, most of the bioactivity is found in Fractions #3-7, with lower amounts of protein in Fraction #5 and especially #3. Also in the second round, significant levels of suppression (greater than 40%) were observed in Fractions #9 and 11 (56% and 54% respectively). These fractions however, contained much lower amounts of protein (0.55 and 0.10 mg). In Fig. 5c it can be seen that there are no clear peaks of suppressive activity in the fractions eluted from the cationic exchange column. The protein fraction eluting at 0.73M N a + (Fraction #11) suppressed the PFC response of NSC by 44% and contained a significant amount of protein (3.9 mg). Fraction #15 gave the same level of bioactivity (44%)yet contained trace amounts of protein (0.004 mg). The previously documented SIF purification protocol (Almawi et al., 1987) has shown SIFP to be 2500 times more active as SIFa. Given this, plus the virtual lack of significant levels of bioactivity in the S-Sepharose fractions obtained in this work, only the bioactive fractions obtained from the DEAE-Sephacel (anionic exhanger) column were taken to the next step of the purification procedure. Anionic exchange fractions comprising the peak suppressive activity (Round 1, Fraction #4-7, and Round 2, Fraction #3-7 and 9-11) were pooled (to become D1, D2a and D2b respectively), concentrated on a bed of sucrose, back-dialyzed against 50 mM NH 4 HC0 3 and lyophilized. Pellets were resuspended in PBS and subjected to FPLC using a Superose 12 HR column (approximately 4.5 mg/run). The column was eluted with PBS at room temperature (30 ml in 60 minutes, collecting 0.5 ml fractions). Given that each FPLC run produced identical profiles (data not shown), fractions from successive runs were pooled. As with the dialyzed ion exchange fractions, an aliquot of each FPLC fraction was sent to Dr. Pope in Halifax to be screened for bioactivity by PFC assay. Representative FPLC elution profiles are shown in Fig. 6 and the corresponding bioassay results are given in Table I. In all three sets of FPLC, the highest levels of SIF activity were found in the earliest eluting fractions, with lower levels scattered throughout the profile. 24 FIGURE 5: ELUTION AND BIOACTIVITY PROFILES FROM ION EXCHANGE CHROMATOGRAPHY OF M1-A5 SUPERNATANT. Results from two separate rounds of DEAE-Sephacel chromatography (a - Round 1 and b - Round 2) and one round of S-Sepharose chromatography (c) are presented. Three litres of serum-free M1-A5 SN was applied via gravity feed to DEAE-Sephacel (anionic exchanger) and S-Sepharose (cationic exchanger) columns set up in tandem and equilibriated with PBS at 4°C. Columns were washed and eluted separately with a 150 ml continuous salt gradient from 0.15 M (PBS) to 0.95 M NaCl (a 1M solution in PBS). 10 ml fractions were collected, dialyzed against PBS and screened for bioactivity by PFC assay (Dr. Pope - various dilutions were cultured with naive spenocytes and SRBC for 5 days). Percent suppression was calculated from cultures incubated with PBS instead of eluted fractions, and each value given represents the average from three wells and two slide chambers. The amount of protein in each well was determined by Micro Lowry assay. 25 MOLARITY OF ELUTANT FIGURE 5a: ELUTION AND BIOACTIVITY PROFILE FOR ROUND 1 OF DEAE-SEPHACEL CHROMATOGRAPHY (DP FIGURE 5b: ELUTION AND BIOACTIVITY PROFILE FOR ROUND 2 OF DEAE-SEPHACEL CHROMATOGRAPHY (D2). 27 FIGURE 5c: ELUTION AND BIOACTIVITY PROFILE FROM S-SEPHAROSE CHROMATOGRAPHY. 28 FIGURE 6: S A M P L E F P L C ELUTION PROFILES OF P O O L E D , BIOACTIVE ION E X C H A N G E F R A C T I O N S . DEAE-eluted fractions #4-7, Round 1 (D1) , #3-7, Round 2 (D2a) and #9-11, Round 2 (D2b) were each pooled and subjected to FPLC using a Superose 12 HR column. D1: (a), D2a: (b) and D2b: (c). Approx-imately 4.0 mg of protein was applied to the column per run. Protein was eluted with 30 mis of PBS over a 60 minute time interval and 0.5 ml fractions were collected and pooled from subsequent runs (in a and b). 30 TABLE I: SUPPRESSION OF THE PFC RESPONSE BY FPLC FRACTIONS: D1 represents pooled DEAE fractions #4-7 from round 1, D2a represents pooled DEAE fractons #3-7 from round 2 and D2b represents #9-11 from round 2. Eluted FPLC fractions comprising distinct peaks were pooled and an aliquot of each frozen at -70°C and sent to Halifax to be screened by PFC assay. Percent suppression is given for samples diluted 1 in 20 in the assay and was ca lcu la ted from the number of PFC 's obtained in P B S versus suppressor sample-treated cultures. Each value represents the average of three values (#PFC's) from duplicate assays. D1 D2a FRACTION # % SUPPRESSION FRACTION # % SUPPRESSION 14,15 79 18-20 28 22-24 0 25,26 0 28,29 5 33-35 55 36,37 15 38,39 0 40-43 0 44-46 6 57,58 21 14-16 18-20 21-23 25, 26 27-29 32-35 36, 37 38-41 42, 43 45-46 55-57 97 55 0 0 45 29 0 0 0 24 44 D2b FRACTION# % SUPPRESSION 14-16 17-19 20-23 24-27 28-31 32-35 36, 37 38 92 79 25 49 36 44 16 78 62 55 3 39-41 42-44 45-48 31 REDUCING SDS-PAGE OF POOLED FPLC FRACTIONS: Eluted FPLC fractions comprising distinct protein peaks were pooled, dialyzed against 50 mM ammonium bicarbonate, lyophilized and subjected to reducing SDS-PAGE. A representative gel is presented in Fig. 7. There are several bands in common between lanes containing bioactive material. As was mentioned, the earliest eluting peak in all three FPLC runs (Fractions 14 and 15 from D1 and 14-16 from D2a and D2b) contained the highest levels of bioactivity. The proteins contained within D2b (14-16) appear to be mostly degraded and thus cannot be analyzed. The first peaks from D1 and D2a however, both can be seen to contain bands with Mr's of >110 KD, 70 KD, 55 KD and 40 KD. Of these four bands, the 55 KD and possibly the 110 KD bands are unique to lanes containing bioactive material. FPLC peak D1(25, 26), was shown to have zero bioactivity, yet contain larger amounts of many different proteins. For this reason, it is difficult to confirm the presence or absence of the 70 KD and 55 KD bands. However, it is evident that this peak does contain the 40 KD, as well as a small amount of > 110 KD protein (see discussion). Peak D2a(18-20), was able to suppress the PFC response by 55% (Table I). This peak is seen to also contain the >110 KD, the 70 KD and the 40 KD protein bands, but the 55 KD band is not evident. Peak D2a(27-29) is shown to be comprised of large amounts of different proteins, and had some bioactivity (45%). This peak does not show bands with Mr's of >110 KD, or 70 KD but does contain the the 50 KD protein. The remaining lanes contain material with zero bioactivity and none of the above discussed four bands are present. In summary, although the suppressor activity in FPLC-eluted material cannot be definitely assigned to any one protein, bands with Mr's of >110 KD and 55 KD were only observed in lanes containing bioactive material (except the small amount of >110 KD that can be seen in D1(25, 26), see discussion). MYCOPLASMA INFECTION OF THE M1-A5 CELL LINE As the above described purification process was being carried out, M1-A5 cells started to exhibit a different growth pattern, dysplasia and increased (over 50%) cell death. The first observed change was a greatly decreased doubling time coincident with an increase in cytoplasmic granularity. Eventually cells in culture began to crenate slightly and occassionally formed large multi-nucleated syncitia. Simultaneously, the characteristic 32 FIGURE 7: SDS-PAGE OF SELECTED FRACTIONS OBTAINED FROM FPLC. Eluted fractions comprising protein peaks were pooled, dialyzed against 50 mM ammonium bicarbonate, lyophilized and subjected to reducing (10 mM DTT for 5 minutes at 100°C), SDS-PAGE using 10% acrylamide. Protein bands were visu-alized by staining with 0.1% AgN03. D1, D2a and D2b represent pooled, bio-active, DEAE-Sephacel fractions #4-7 from round 1, #3-7 from round 2 and #9-11 from round 2 respectively that were subjected to FPLC. 33 suppression of the serum-free culture supernatant became irregular in both the KLH Antibody Culture and the PFC Assays. Shortly thereafter, Dr. Pope's laboratory communicated to us that the M1-A5 cell line had come back positive for an animal strain of Mycoplasma. Several attempts were made in Halifax to cure the cell stocks (which also proved to test positive) of the Mycoplasma infection. MYCOPLASMA TESTING AND ANTIBIOTIC CURING OF M1-A5 CELLS: When the M1-A5 cell line was originally isolated in the lab. of Dr. B. Pope, cells were tested for Mycoplasma both in the lab. (by DAPI immunofluorescent staining ) and at a nearby hospital (Mycoplasma culturing) on a regular basis. All results had been negative. Here at UBC, our lab. also does routine testing of all cell lines using the DAPI method. Our M1-A5 stock tested negative three times by these methods. Once Dr. Pope received word from the hospital labs that her current and frozen stocks were contaminated, an aliquot of our stock was sent and also found to be positive. The circumstances of this seemingly sudden pan-contamination are curious, as the hospital diagnostic technologists assured Dr. Pope that there had been no change in the sensitivity of the detection method in the time that she had been sending cells there for testing. Dr. Pope's lab. undertook attempts to cure M1-A5 of Mycoplasma with the commercially available antibiotic kit from Boeringher Mannheim. The treatment was successful, however, coincident with losing the Mycoplasma, cell supernatants lost the ability to suppress the PFC response. Three independently cured M1-A5 cultures gave the same result. Given our inability to prove that the previously observed suppression was caused by products other than those which were /Wycop/asma-derived, Dr. Pope decided to discard the M1-A5 system and begin the process of isolating a new natural suppressor cell line. In light of her decision, and the Mycoplasma problem, our lab decided to follow her lead and leave this project. 34 S E C T I O N 1: D I S C U S S I O N . Experiments investigating the mode of action of SIF have been interpreted to suggest that it acts by inducing suppressor cells present in normal murine splenocyte preparations in an antigen nonspecific fashion (hence the name Suppressor Inducer Factor - SIF). In the original suppressor cell assay of Almawi and Pope (1985) and Pope (1987), normal splenocytes exposed to M1-A5 supernatant in Marbrook vessels were co-cultured with twice the number of untreated normal spleen cells (NSC) and SRBC. After 5 days, the cell mixture was tested for its' ability to launch a B cell response against SRBC (by PFC Assay). Pope explains the observed suppression by presuming that SIF (in the M1-A5 SN) induces a suppressor cell population(s) or network which in turn inhibits the secondary immune response of the spleen cells to SRBC. The level of suppression obtained in these experiments was from 55-67%, depending on the number of M1-A5 cells or concentration of the M1-A5 supernatant (see Fig. 1a) in the outer chamber of the Marbrook vessel in the activation step. Considering that one third of the cells in the suppressor assay culture are SIF-treated and two thirds are untreated, the fact that the level of suppression obtained by PFC Assay is greater than 33% (ie. one third) suggests that SIF does not act directly on effector cells (B or T cells). Partly in support of this is Pope's assertion that blastogenesis is not affected by SIF treatment because (a) normal cell numbers were obtained in control and suppressor tests and (b) SIF treatment had no effect on splenocytes ability to respond to Con A (Pope, 1987). However, Con A has no effect on B cell blastogenesis, thus this observation does not rule out the possibility of abnormal B cell proliferation, IgH gene class-switiching and/or inhibition of antibody production. It would have been useful to have tested the ability of SIF to suppress the splenocyte response to LPS, which selectively activates B cells or PWM which activates both T and B cells. It is possible that SIF somehow alters the functional ability of B cells directly. Some mechanisms might be inhibition of the B cell response to T cell-derived B cell growth factors such as BCGF and IL-3. Experiments by Almawi and Pope (1985) and Pope (1987) have suggested that SIF-induced suppresssor cells act via the release of suppressor factor(s). It was observed that SIF-activated NSC (incubated in the inner chamber of a Marbrook vessel) secreted a substance(s) (less than 12 KD) which was able to suppress the PFC response of untreated NSC incubated in the outer chamber. Two attempts have been made to purify enough SIF material to obtain a protein sequence, begin production of a monoclonal antibody and study 35 its mode of action. In this work, a third attempt was made to purify SIFf5, and a new assay was adapted to study suppression of antibody responses by SIF. For the results presented in Fig. 1b and this purification work, the suppressor assay discussed above was simplified in order to screen suppressor samples faster and with more ease. Instead of preactivation of NSC in Marbrook vessels, just a straight PFC assay was done. SIF sample was added directly to a mixture of NSC and SRBC and incubated for 5 days before assaying for the presence of antibody producing cells. This change in assay procedure should not alter the normal action of SIF, but as expected, did increase the sensitivity of the assay. However, this increase was unexpectedly large in magnitude (see dilution of SN that gives 50% suppression in Fig. 1a versus 1b). Some of this difference in sensitivity can be explained by examining each protocol. In the basic PFC assay used by Almawi et al (1988b) to screen for SIF activity, SIF is added to all cells in culture for the duration of the five day incubation with antigen. Thus, the entire culture is 'activated' and SIF has the potential to continue 'inducing' for as long as its activity is stable in serum-containing media. In Pope's suppressor assay (Pope et al., 1987), one third of the NSC preparation is treated, washed and then co-cultured with the remaining two thirds of the NSC's (untreated). Thus any observed suppression of the 2° Ab response to SRBC 5 days later must be mediated by the suppressor cell population within the initially activated NCS preparation. It would have been interesting to observe the effect of activating increasing numbers of NSC (and co-culturing these with decreasing numbers of untreated cells) on the magnitude of suppression. In an attempt to study the nature of SIF action further, KLH-immune splenocytes were activated by antigen and subsequently cultured for 7 days with various dilutions of M1-A5 serum-free SN in the Ab C assay. Anti-KLH titers were determined by KLH-ELISA. From Fig. 4, the E D 5 0 of M1-A5 SN in this assay was found to be 10\"2. Thus this assay is more than 100 times more sensitive than the original suppressor assay of Pope er al. and 10,000 times less sensitive than the PFC assay used to screen for SIF activity. Both the Antibody Culture (Ab C) and the Plaque Forming Cell (PFC) Assay measure the ability of primed murine splenocytes to respond to furthur antigenic challenge by quantitating the B cell response. However, these two tests also have significant differences. In the PFC assay, suppression of the secondary response to antigen is measured after SIF (or activated SC) is introduced in the priming reaction and in the Ab C Assay, suppression of the tertiary response to antigen is measured with no interference in the primary or secondary response (occurs in vivo). It would appear from this and the results presented in Fig.'s 1 & 3, that SIF, whatever it's exact mode of action, is much 36 more efficient at effecting suppression of the B cell primary responses (ie. priming) than secondary or tertiary responses. If the priming reaction is inhibited in vitro in the PFC assay, suppression of the response to subsequent challenge with antigen would be observed. Alternately, it was possible that a longer incubation period was required for SIF and unactivated (but immune) splenocytes in order to see an optimal effect. SIF is added to culture directly after the in vitro KLH boost in the Ab C Assay. Given that a tertiary response is being elicited, this protocol may not have allowed sufficient time for SIF to induce the suppressor cell population and/or for this population to act on target cells before too much stimulation or activation of immune splenocytes had occurred. In order to determine whether the time of additon of SIF or M1-A5 supernatant to splenocytes in the Ab C assay was a factor in the magnitude of the resultant inhibition, an experiment was done in which one half a preparation of KLH-immune splenocytes was preincubated for six hours with SN and then pulsed with KLH and cultured, and the other half pulsed and cultured with the SN immediately. The results (Fig. 4) clearly indicate that the preincubation step abolished most of the inhibition of the anti-KLH response. This loss of suppresssion with preincubation would suggest that some antigen must be present in order for SIF to act. It would have been interesting to know whether the injection of purified, bioactive M1-A5 material with KLH into syngeneic mice in the priming stage of the immunization schedule would have any effect in the resultant anti-KLH levels in this assay. It is also reasonable to question the stability of SIF in serum-containing media. In the PFC assay, with antigen present, SIF may act early to induce the suppressor population and thereafter be inactive in the culture. If this were the case, in the preincubated Ab C assay, SIF may have been unable to act on the immune SC preparation without antigen present. The factor may then have been degraded or rendered inactive by serum components before the in vitro antigen pulse. There are also differences between the Ab C and PFC Assays that are inherent in their modes of quantitation that would most likely give different results regardless of the mode of action of the factor being tested. In the PFC Assay, one activated antibody-producing cell forms 1 plaque, thus an accurate measure of the magnitude of the B cell response is obtained by quantifying the number of plaques produced in each test. The directness of this assay results in increased sensitivity. In the Ab C Assay, the relative amount of specific antibody produced (not antibody-producing cells) in each test culture is determined by sandwich ELISA. Thus, this assay measures the magnitude of the B cell response in a more indirect manner, resulting in decreased sensitivity. In addition, there are many more steps in which variability can be introduced, from the immunizations through to the 37 antigen-coating and developing of ELISA test plates. It would be very difficult to quantitate what proportion of the large difference in sensitivity observed in this work (Fig. 1b versus Fig. 3) is inherent in the protocols of these assays versus a reflection of the mode of action of SIF. M1-A5 cell cultures were expanded in order to begin the purification of SIF from spent cell SN. During this time, it was observed that if the WEHI-3B cells were allowed to overgrow slightly before harvesting the SN for addition to M1-A5 growth media, they released a product which was autolytic and potent. This factor, when added (at 1 in 5 dilution) to M1-A5 media subsequently caused complete death of the culture within 2 days. When WEHI cells were seeded into fresh media, the remaining live cells did not recover. From several WEHI-3B growth curves representing cultures seeded at increasing cell densities (Fig. 2), it was found that the optimal cell concentration to harvest SN to support the growth of M1-A5 cells was 5 X 105 cells per ml from cultures seeded at 1-5 X 104 cells per ml. The previously described process for purifying SIF a and (3 from serum-free culture supernatants involved size fractionation on a Sephadex G-100 column followed by ion exchange chromatography and two rounds of RP-HPLC of bioactive fractions (Almawi et al., 1988b). The size and composition of the bioactive HPLC fraction(s) was determined by SDS-PAGE using a 5-15% acrylamide gradient. When the SIFfi preparation was sent out of the lab. for sequencing, it was reported that the amino terminus of the peptide was blocked. This was thought to be caused by the harsh conditions of acetonitrile elution during the two rounds of RP-HPLC and unfortunately there was not enough purified material to effect the sequencing of separated tryptic peptides. Due to the high binding capacity of the more contemporary ion exchange column matricies, it was decided that the size fractionation step of the purification process could be omitted and serum-free culture supernatants were added directly to DEAE-Sephacel and S-Sepharose columns set up in tandem. Eluted, bioactive fractions were pooled in peaks and subjected to FPLC, and eluted bioactive FPLC fractions comprising distinct peaks were pooled and a sample from each subjected to SDS-PAGE to determine the composition and level of purity. The lack of distinct peaks and significant levels of bioactivity obtained in the S-Sepharose eluate and the fact that purified SI Fa has been shown to be much less active than SIFp prompted the decision to continue with the purification of SIFp* only. The bioactivity eluted with the main protein peak in both rounds of DEAE-Sephacel chromatography (D1 and D2a). Suppression was also seen in two fractions eluting at higher molarities in the second round (D2b) but these contained lower amounts of protein. 38 Peaks of bioactivity were pooled and subjected to FPLC using a Superose 12 HR column. The highest levels of bioactivity were found to reside in the earliest eluting peaks in all three rounds (D1, D2a and D2b, see Fig. 6 and Table I) of FPLC. This was suprising given that the molecular weight of purified SIFJ3 has been determined to be 6 KD (Almawi et al., 1987 and 1988b). The bioactive component of this protein peak was possibly in the form of large multimers, or alternately, associated with another protein. Material within bioactive and inactive FPLC-eluted peaks were subjected to reducing SDS-PAGE (Fig. 7). The first peak from D1 and D2a had several protein bands in common (Mr's of >110 KD, 70 KD, 55 KD and 40 KD). Of these, there is only the suggestion that the 55 KD, and possibly the >110 KD bands are unique to lanes containing suppressive activity. Peak D1(25, 26), although inactive in the PFC assay at a 1 in 20 dilution, was found to contain many different size proteins. Thus, eventhough there is a small amount of the >110KD band in this lane, if this is the protein which mediated the observed suppression in other peaks, it's activity may have been diluted out or altered in this peak by another component. Although the protein concentration and bioactivity of the serum-free M1-A5 supernatant was not determined at the start of this purification, 17.60 mg (Fractions #4-7) of pooled bioactive material was obtained in the first round of anion exchange (DEAE) chromatography and 18.35 mg (Fractions #3-7) plus 1.05 mg (Fractions #9-11) were obtained in the second round. The average E D 5 0 of D2a was 4.74 ug, that is, approximately 4.74 ug of the pooled sample gives 50% inhibition of the antibody response of murine NSC to SRBC (as determined by PFC Assay). Although much less protein was obtained in D2b (1.05 mg), this pool gave an average E D 5 0 of 0.340 ug, indicating a 14-fold higher level of activity in this sample. A similar titration of bioactivity was not done in the screening of the first set of DEAE fractions, thus the E D 5 0 of D1 cannot be determined for comparison. However, if one compares the amount of suppression obtained at a similar dilution (1 in 20) in D1 and D2a, and knowing the protein concentration of each, the activity (expressed as percent suppression per ug protein) can be determined. The recovery of protein from the FPLC Superose 12 HR column (Pharmacia) has been documented as being from 80-85% (Pharmacia Superose 12 HR product file). For these purposes, the lowest value was chosen (80%). As the FPLC monitor integrates the area beneath each eluted protein peak (elution time or volume versus absorbance at 280 nm) and presents this value as percent of protein loaded onto the column, the protein concentration of each peak can be determined. Unfortunately, limiting dilutions of each 39 FPLC fraction were not tested by the PFC Assay. Thus, the increase in activity from the bioactive ion exchange fractions to the FPLC fractions cannot be determined by standard methods (ie. comparing ED 5 0 ' s or units of activity per amount protein). Instead, as described above, the percent suppression per (ig protein (determined from a sample diluted 1 in 20 in the assay) was used to determine which protein peaks contained the highest levels of activity.and compare activity between steps of the purification. To further complicate the analysis, in 2 of the 3 most active FPLC peaks (Peak #1 in all three runs), the level of activity had not even begun to titrate at the 1 in 20 dilution (the only common dilution between all samples in the purification) tested, thus no accurate quantitation could be made. In FPLC profile D1, 112 ug of protein eluted in the first peak. Of this amount, 11.20 ug (a 1 in 5 dilution in the PFC assay) and 5.60 ug (a 1 in 10 dilution) suppressed the PFC response by 95 and 90% respectively. When 2.80 ug (a 1 in 20 dilution) was tested, 79% suppression was observed. As the bioactivity had begun to titrate, Peak 1 of D1 can be said to have an activity of 28.2% per ug protein. Peak #1 from D2a contained much higher amount of protein (700 ug). 70 ug (1 in 5 dilution in the PFC assay), 35 ug (1 in 10) and 17.5 ug (1 in 20) all gave from 97 to 100% suppression. The bioactivity contained in this sample had not begun to titrate at the dilutions tested and thus a calculation of percent suppression per ug protein would be meaningless. If the relative amounts of protein tested from Peak #1 in D1 and D2a at the 1 in 20 dilution are compared (2.8 versus 17.5 ug), it is not suprising that a decrease in the amount of suppression was not observed. In the third FPLC run (D2b), 1.05 mg of protein was separated. Coincidently, approximately the same amount of protein eluted in the first peak of this run (115 ug) as in the first run (112 ug) eventhough greater than 4 times the amount of protein was loaded in the latter. 1 in 5 (11.5 ug) and 1 in 10 (5.75 ug) dilutions of the material in this peak both gave 98% suppression of the PFC response, whereas the 1 in 20 dilution (2.87 ug) gave 92%. If this 6% decrease in activity with one doubling dilution was the start of a titration, the activity of this first Peak would be 32.1 percent per ug protein, which compares well with the value of 28.2 obtained from 2.8 ug of Peak #1 in D1. In addition, peak D2b(17-19) and D2b(38) were found to have activities of 20.4 and 53.7 percent per ug protein respectively. Lower levels of suppression were also observed scatterred throughout all three elution profiles (Table I) During the purification process presented in this work, M1-A5 stocks and cultures were found to be infected with an animal strain of Mycoplasma. Since curing cultures of the infection resulted in a complete loss of suppressive activity in three separate 40 attempts, the origin of the moiety mediating the above discussed bioactivity was brought into question. Dr. Pope maintains that the characteristics of suppression by M1-A5 SN became irregular a few months before the above determination was made. It is quite possible that the problems encountered in the culturing of M1-A5 cells in this lab (slow doubling time, large amount of cell death etc.) are the direct result of the Mycoplasma infection. In addition, it is also possible, though less likely, that all of the suppression observed in the M1-A5 system has been mediated by products of Mycoplasma origin. Thus the purification presented in this work may have been guided by following the bioactivity levels of a /Wycop/asma-derived product. SECTION 1: CONCLUSIONS 41 Given the antigen-nonspecific, and MHC nonrestricted mode of action of the partially characterized Suppressor Inducer Factor (SIF) secreted by the isolated NS cell line M1-A5, an attempt was made to purify the p* form to homogeneity for biological study, amino acid sequencing and the synthesis of a monoclonal antibody. Following a modified version of the previously described SIF purification protocol, an attempt was made to purify SIFJ3 using ion exchange chromatography, and FPLC. Although, the level of bioactivity was followed through these two steps, in the final analysis of FPLC eluted proteins by SDS-PAGE and PFC assay, suppressive ability could not be definitively assigned to any one protein band (although there are possibilities which could have been further investigated) and high levels of activity (percent suppression per u.g protein) were found to be scattered throughout the profiles. Whether this reflects a mixture of M1-A5 and Mycoplasma derived products was not determined. In addition, an in vitro assay was adapted for studying suppression by M1-A5 SN. M1-A5 SN in the Ab C assay gave a suppression curve from which an E D 5 0 of 10\"2 was obtained. In addition, it was found that preincubating SN with immune SC prior to antigen pulse and culture virtually abolished all suppressive ability. These results were discussed and compared with data obtained in earlier work using different assays. It can be concluded that suppression by M1-A5 SN requires the presence of antigen, yet is most pronounced when present during the generation of specific T cell clones and memory cells (ie priming or generation of the primary response). The origin and nature of the soluble product(s) studied in this work remain unknown, as this line of research has been discontinued. 42 SECTION 2: ANTIGEN-SPECIFIC IMMUNOSUPPRESSION INTRODUCTION Active investigation into the phenomenon of antigen-specific suppression began in the mid-seventies with several groups reporting the existence of antigen-specific Ts cells and soluble \"suppressive\" factors. For example, spleen and thymus extracts from mice primed with hapten-carrier conjugates were shown to contain a suppressive factor which bound specific antigen, reacted with anti-MHC serum and had a MHC-restricted mode of action (Kapp etal., 1974 and 1976 and Takemori and Tada, 1975). This restriction, which mapped to the I region of the mouse MHC, and was shown to be distinct from that mediated by l-A and l-E products, was named l-J (Murphy et al., and Tada et al., 1976). Shortly thereafter, Cantor and Gershon (1979) postulated the existence of distinct subsets of Ts cells which could be distinguished according to their expression of various cell surface markers and individual roles in effecting suppression. Yamauchi et al. (1981a & b) confirmed that suppression involved several distinct Ts cell types. Further, they showed that this suppression was mediated by the release of soluble factors (TsF's) which functioned to activate the specific Ts subsets in a sequential manner. This, and other studies has led to the general acceptance of the existence of antigen-specific Ts cell circuits. In the most widely accepted pathway (rev. in Germain and Benacerraf, 1981) of antigen-specific suppression, a Ts1 (suppressor inducer) cell stimulated by antigen releases TsF-1, an antigen- or idiotype-specific suppressor factor that activates a Ts2 (transducer or regulatory) cell. It in turn releases TsF-2, which is thought to be anti-idiotypic, and cause activation of the Ts3 cell (effector). The Ts3 cell is then thought to act (directly or via antigen-/idiotype-specific TsF-3 secretion) on Th cells to ultimately cause down-regulation of B cell and CTL function. ATTEMPTS TO CORRELATE Ts CELL SURFACE PHENOTYPE WITH FUNCTION Initial cell surface characterization of the components of murine suppressor T cell circuits described Ly1 +2 _, I-J+ cells that functioned as inducers and activated Ly1 +2 + regulatory cells. These cells in turn stimulated Ly1\"2+, l-J + effector cells which were 43 distinct from the Ly1\"2+, l-J- CTL population (Cantor and Gershon, 1979). In support of this work Yamauchi et al. (1981a and b) reported that Ly1+2_ (Ts1) cells released a factor (TsF-1) which bound antigen and could stimulate Ly2+ cells. They showed that the final step in the path was mediated by Ly1\"2+ (Ts3) cells via the release of effector TsF (or TsF-3). Other groups have reported similar T cell circuits bridged by the release of antigen-binding TsF's (Taniguchi and Tokuhisa, 1980, Hausman et al., 1986 and rev. in Tada, 1984). Flow cytometric studies of human Ts1 suppressor inducer cells using various monoclonal antibodies combined with biological assays demonstrated that CD4+ T cells could be separated into three distinct subpopulations: (1)CD4+ Leu8+ CD29\"(4B4) CD45R +(2H4), (2)CD4+ Leu8+ CD29+ CD45R\" and (3)CD4+ Leu8\" CD29+ CD45R\" (Mohagheghpour et al., 1983, and Damle et al., 1987). All three subpopulations, upon stimulation with alloantigen proliferated and produced IL-2. However, only the latter two subsets were able to respond to soluble antigenic stimuli and only the second group was capable of inducing CD8+ Ts (Damle et al., 1985, 1986 and 1987). The first subset was subsequently found to represent a circulating pool of naive (have not yet interacted with specific antigen) T lymphocytes and not a distinct CD4 + sublineage as originally assumed. Expression of the CD29 and CD45R markers was shown to reflect different stages of immunological differentiation from naive (CD29-CD45R+) to memory (CD29+CD45R-) CD4+ cells (Sauders et al., 1988). Although the expression of the Leu8 determinant did not seem to be affected by the state of activation or differentiation of the cell, this marker was later further characterized and found to be expressed on 75% of peripheral T cells as well as some B cells, monocytes and neutrophils (Rich et al., 1986). The authors however still maintain that the Leu8+ and Leu8\" subsets of CD4+ CD29+CD45R\" T cells mediate suppressor-inducer and helper functions respectively (Kansas et al., 1985 for eg.). Despite the fact that both CTL and Ts3 (suppressor effector) cells are CD8 + , the antigen-mediated activation of each is known to occur by a different mechanism. CD8+ CD28+ CTL are stimulated by direct contact with activated APC or altered self (ie viral or allogenic determinants expressed on autologous cells) and ultimately are cytotoxic to their target, whereas CD8+CD28' Ts are activated by indirect contact (via soluble TsF) with antigen-stimulated CD4 + suppressor-inducer cells (Damle et al., 1984 and 1985). 44 PURIFICATION OF T CELL-DERIVED SUPPRESSOR FACTORS (TsF'sl Only very small quantities of TsF were detected in the extracts from murine lymphoid organs in early suppression studies, and it eventually became critical to purify TsF's to homogeneity to further characterize their nature. Since biochemical purifications require large amounts of starting material, attempts were made to isolate Ts cell clones and make Ts cell hybridomas, that would constitutively synthesize and secrete TsF into their growth media. Some groups attempted to obtain Ts cell clones by adding exogenous IL-2 to in vitro cultures enriched for Ts cells (Nakauchi et al., 1984 for eg.) and others fused enriched Ts cell preparations with the thymoma BW5147 to obtain Ts cell hybridomas (Kapp et al., 1980 for eg.). Using these approaches several groups (Taniguchi et al., 1979 and 1980, Kitamara et al. and Nakauchi et al., 1984, Kapp et al., 1980 , Krupen et al., 1982, Turck et al., 1986, Saito and Taniguchi, 1984, Webb er al., 1989, Fresno et al., 1981 and Hausman er al., 1986 for eg.) reported on the purification and characterization of antigen-specific TsF's, yet to date none of the genes encoding such factors have been cloned. There is no firm consensus with respect to size (from 19 to 90 KD Mr), structure (single versus two-chain and disulfide bonded versus noncovalent association) serology (reactivities with l-J antisera and anti-TCR reagents) or mode of action (MHC restricted or nonrestricted) of TsF's (Steele era/., 1987a). There is however, tentative agreement as to the nature of TsF-1 (suppressor-inducer factor). TsF-1 is thought to be composed of at least two molecules, a T cell derived antigen binding molecule (T-ABM) and an antigen-nonspecific accessory molecule which bears I-J determinants. TsF-1's are: (1)antigen specific (2)display Igh but not MHC restriction (3) function during the inductive but not effector phase of the suppression response (4) suppress both delayed type hypersensitivity (DTH) and plaque forming cell (PFC) responses and (5)bind to antigen and isotype-specific anti-TsF antibodies (from Kuchroo et al, 1990). One group has isolated (in the form of Ts cell hybridomas) all three orders of Ts cells and TsF's that participate in suppressing the immune response in the NP (dinitrophenyl) hapten system (Hausman era/., 1986 and Kuchroo era/., 1990). The NP-TsF-1 was found to be idiotypic (binds to NP), express l-J determinants and have a nonrestricted mode of action. The NP-TsF-2 also expressed l-J determinants, but had a restricted mode of action and was anti-idiotypic (binds to anti-NP Abs). Finally, NP-TsF-3, like TsF-1, was found to be idiotypic (binds to NP), and l-J +. With respect to biological function, 45 TsF-1 could suppress the contact sensitivity (CS) to NP if it was administered to the animal at the time of antigen-priming, TsF-2 could suppress the same response at any time in previously primed animals, and TsF-3, which was postulated as being primed simultaneouly with Th cells, had the ability to not only suppress the CS response to NP but also to other antigens (bystander suppresssion) (Hausman etal., 1986). As mentioned above, other Ts/TsF pathways are reviewed by Tada (1984). Very recently, Kuchroo et al. (1990) working in the NP system has made Ts hybridomas using a new fusion partner (BW100) that is functionally TCRa(3 _. These should help to characterize and clarify the nature of Ts antigen-receptor interactions. MOLECULAR INTERACTIONS IN Ts-MEDIATED SUPPRESSION Although fairly well characterized in B cells, CTL and Th, the nature of the receptor-ligand interaction that mediates Ts activity is poorly understood. Work from several laboratories (Zheng et al, 1988 and 1989 and Fairchild et al, 1988 for eg.) have finally confirmed that Ts cells interact with antigen via the standard TCRafi complex. Early studies on Ts hybridomas indicated that there was no CD3/TCR cell surface expression nor functional TCRct or p RNA transcripts in Ts - suggesting that these cells utilize some other form of antigen receptor. A possible reason for the conflicting reports and discussions in regard to this matter was offered initially by Weiner er al, who in 1988 found that 1-3% of Ts hybridoma cells in culture expressed the CD3 marker. Upon cloning these cells, as well as several CD3\" cells from the same culture, it was found that only the CD3 + cells were capable of sythesizing detectable amounts of factor. The CD3 molecules were subsequently shown to be associated with a functional TCRap\\ The above has been repeated with the same results on another apparently CD3\" hybridoma (Kutchroo et al, 1988). In this latter study, it was also shown that modulation of the CD3/TCR complex from the cell surface effectively removed antigen binding by the cell suggesting that no other antigen-binding moiety resided on the Ts cell surface. Thus it is possible that as Ts hybridomas are maintained for extended periods in culture, fewer cells express TCR (possibly due to chromosome loss). This may in turn correlate with a decrease in TsF production and be a contributing factor in the difficulties that have been experienced in their biochemical purification. The effector leg of antigen-specific suppression circuits are thought to act by blocking the activation of B cells and CTL by inhibiting the ability of CD4+ helper/ inducer cells to 46 promote their growth and differentiation (Rich et al, 1986). Recent C a + + flux studies have suggested that the critical event in this final step of the suppression pathway is the inhibition of early Th membrane signal transduction (Utsunomiya etal, 1990). As was mentioned previously, the genetically restricted mode of action of some of the described TsF's was found to map to the I region of the mouse MHC (Murphy ef al., 1976), yet the expressed determinant appeared to be distinct from l-A and l-E. Subsequently, the entire I region was analyzed at the molecular level and the gene(s) encoding l-J was not found (Kronenberg, 1983). Thus the question of: \"What is l-J?\" still remains. One group has suggested that l-J is a receptor for l-A products and another that l-J is actually a TCRa(3 determinant (Green and Zheng, 1989). Whatever l-J is, it has been demonstrated to play an important role in Ts systems in some strains of mice, and thus may be involved in the control of suppression of immune responses. The finding that Ts cells do express conventional TCRap* has raised the question of whether there is any relationship between Ts TCR and antigen-specific TsF's. Work by several groups (rev. in Green and Zheng, 1989) has suggested that TCR genes, especially a , play an imporant role in mediating the fine antigen-specificity of TsF's. Since there are important differences between TCR and TsF's, if this theory turns out to be true, Green concedes that the mechanism of TsF generation will likely involve alternate RNA splicing or post-translational modification of TCR-gene products. It is interesting that alternate splicing of TCR RNA has already been observed by different groups (Behlke and Loh, 1986 for eg.). THE P815 TUMOR. Ts CELLS AND THE MAb B16G Several groups have studied the nature of Ts cells in tumor-bearing animals (Fisher and Kripke, 1982, Granstein et al., 1984 and Mills and North, 1985 for eg.). Working with the murine mastocytoma P815, Takei et al. (1976) observed that the presence of Ts cells in DBA/2 tumor-bearing animals correlated with an increase in tumor growth. In 1978, Takei et al. demonstrated the presence of a soluble factor (TsF) in the thymocytes of tumor-bearing DBA/2 mice that was able to suppress the in vitro generation of CTL to P815 tumor cells. The above TsF was affinity-purified over a P815 membrane extract column and used to synthesize the TsF-specific monoclonal antibody B16G (Maier etal., 1983). Injection of B16G into mice slowed the growth of not only the P815 tumor, but also the chemically 47 induced syngeneic fibrosarcoma M-1. B16G was also found to be nonspeci-fically immunoenhancing and to selectively remove a subset of regulatory T cells from a thymocyte preparation. In light of the above observations, it was postulated that B16G recogizes an invariant epitope on TsF's. ISOLATION OF A10. A Ts CELL HYBRIDOMA In order to further study and characterize the nature of TsF's, P815-stimulated thymocytes were fused with the thymoma BW5147 and the antigen-specific hybridoma A10 was isolated (Steele era/., 1985). A soluble factor (A10F) released by A10 cells was bound by P815 and B16G affinity columns and was able to accelerate the growth of the P815 tumor, but not other tumors (M-1 or L1210). Affinity enriched A10F appeared as 3 distinct protein bands of 140, 80 and 45 KD when subjected to reducing SDS-PAGE (Chan et al., 1988 for eg.). Of these, only the 140 and 80 KD proteins had suppressive ability. The 80KD band was thought to dimerize (to the 140KD band) and a minor protein, appearing occasionally at 30KD was assumed to be a 80KD breakdown product. Protein sequence data from the N-terminus of the 140 and 80 KD bands showed identity whereas that from the 30KD band was distinct. As the antigen to which A10F was specific was uncharacterized, it was difficult to further characterize the biochemical and biological nature of this TsF in vitro. Fd11: A Ts HYBRIDOMA SPECIFIC FOR THE ANTIGEN FERREDOXIN fFd) The immune respone to the antigen ferredoxin (Fd) has been an important line of investigation in this lab (rev. in Chu, 1989). Fd is a prokaryotic protein of an electron-carrier chain and has been well characterized biochemically and structurally (Jensen, 1974 and Graves et al., 1985) For these reasons and previous work demonstrating the existence of Fd-specific Ts cell circuits in mice classified as nonresponders to Fd (Weaver et al., 1983 and Singhai et al., 1984), this antigen was chosen for use in the synthesis of a new Ts hybridoma (Fd11). Fd11, like A10, was found to secrete a factor (Fd11F) which bound to Fd and B16G affinity columns, and was able to suppress the in vitro and in vivo antibody response of Fd immune splenocytes (Steele et al., 1987 and Chu, 1989). Using a Ca++-sensitive reagent (Quin-2), Steele showed that the addition of Fd activates Fd11 cells in culture. 48 To purify Fd11F to homogeneity, Fd11 cell-free SN was subjected to steps of B16G-affinity chromatography, followed by fast protein liquid chromatography (FPLC) using a Superose 12 HR column (Chu, 1989). Material eluting in the void volume (Mol. wt. >150 KD) of the Superose column was found by ELISA to bind Fd and by in vitro Ab Culture Assay to have suppressive ability. Reducing SDS-PAGE of the bioactive FPLC eluate gave protein bands of approximately 80 and 30 KD, as well as some minor bands. However, some 80 KD material was also seen on SDS gels of FPLC eluates from a control cell line (BW5147, the fusion partner for Fd11) SN. In an attempt to clarify which component(s) of the reduced eluate were responsible for the antigen-binding and suppressive activity, the same FPLC eluates were run out on preparative SDS gels. Protein eluted from gel slices was tested for binding to Fd (by ELISA) and bioactivity (by in vitro Ab Culture). Antigen binding was observed in 80, 30 and 16 KD bands from the Fd11F preparation whereas none was observed from the BW5147 preparation. None of the gel slice eluates showed consistent immunosup-pressive ability. The 80, 30 and 16 KD components of the above Fd11F preparation were shown to react with an antisera raised to the 30 KD moiety eluted from a gel slice from an B16G-affinity enriched Fd11F preparation (see below). From these results, the author asserts that the both the 80 and 30 KD components of the Fd11 preparation are components of Fd11F. P80 AND P30: ANTISERA MADE TO COMPONENTS OF AFFINITY-EN RICH ED Fd11 F Preparative gels were run on B16G affinity-enriched Fd11F and the 30 KD band excised. Eluted 30 KD protein was used to prime (0.1-2.0 u.g in 50% CFA) and repeatedly boost NZW rabbits (North et al., 1988). The anti-p30 titer of the immunized rabbit serum was evaluated by Fd11F ELISA. Normal rabbit serum for use as a control for nonspecific binding was obtained from the blood of rabbits prior to immunization. Anti-p80 was made by the same procedure starting with the 80 KD band from the same Fd11F preparation. This antiserum was shown to have a high background (nonspecific binding) and thus results obtained with this reagent were not reported. Using flow cytometry, anti-p30 was shown to bind to the surface of Fd11and A10 Ts hybridomas (although with less intensity to the latter), but not to several other Th and Tc hybridomas nor the surface of the thymoma BW5147 (North era/., 1988). The p30 49 marker was found to be down regulated on the surface of Fd11 cells in response to treatment with phorbol esters. In five different strains of mice, anti-p30 was shown to react with approximately 4% of thymic T cells and 10% of splenic T cells (purified over nylon wool). To further characterize the population of murine p30+ cells, nylon wool-purified T cells were double labelled with either anti-L3T4 or Ly 2 monoclonals and anti-p30 serum and analyzed on a Coulter EPICS machine. North showed that twice as many L3T4+ cells were p30+ as Ly 2+ cells. Also, after complement mediated lysis of L3T4 and Ly 2 positive cells, the remaining double negative population was enriched for p30 binding (30% positive). North et al. also tested the ability of anti-p30 to block the suppressive effect of Fd11F in the in vitro Ab Culture Assay. It was demonstrated that anti-p30 alone had the ability to enhance the anti-Fd Ab response, and Fd11F alone had the ability to suppress the response (as had been shown previously). Further, the enhancing effect of anti-p30 was shown to be dominant over the suppressive effect of Fd11F when both were added to the same culture. Although anti-p30 was tested and found to not react with cells from some non-lymphoid tissues (brain, liver, kidney) and bone marrow, it was not tested on B cells or immune cells of the myeloid lineage. SUMMARY OF WORK TO BE PRESENTED Using anti-p30 and anti-p80, the population of human periheral blood mononuclear cells staining positive was characterized with respect to other cell surface markers. In addition, cell surface studies and Western blot analyses were done on several p80+ and /or p30+ lymphoid leukemic cell lines. 50 SECTION 2: MATERIALS & METHODS ANTISERA: P30 and P80 antisera were prepared as described in North et al., 1987. Briefly, Fd11F was affinity-purified using the monoclonal antibody B16G from spent Fd11 (a Ts cell hybridoma-R. Chu, 1989 for eg.) culture supernatants. Eluted protein, when subjected to reducing SDS-PAGE gives two distinct bands at 80 and 30 KD. NZW rabbits were primed and boosted subcutanesously with 0.1-2.0 u,g of protein eluted from 80 and 30 KD preparative gel slices. Serum antibody titers were evaluated by ELISA using Fd11 F-coated plates. Sera was stored frozen at -70°C or in 0.02% Sodium Azide at 4°C. Normal rabbit serum samples were harvested preimmunization and used as controls for studies using the matched immune sera. ANTIBODIES: Phycoerythrin (PE) conjugated anti-Leu 2a/2b (CD 8), anti-Leu 3a (CD 4), anti-Leu 4 (CD 3) and anti-Leu 12 (CD 19) and fluoroscein isothiocyanin (FITC)-conjugated anti-TCRoc(3-1 were purchased from Becton Dickinson (Mississauga, Ont.) and used at 1 in 30 final dilution. Goat anti-rabbit Ig-FITC (Fab2) was obtained from Cappel through Cooper Biomedical (Malvern, PA) and used at 1 in 50 dilution. All antibodies were stored at 4°C in the dark. HARVESTING HUMAN PERIPHERAL MONONUCLEAR CELLS FROM NORMAL WHOLE BLOOD: Blood was collected into 10.0 ml sterile tubes containing Lithium heparin (Becton Dickinson), pooled and diluted to twice the initial volume with PBS at 37°C. 10.0 ml aliquots of the mixture was carefully layered onto 3.0 ml of room temperature Ficoll Hypaque (Pharmacia, density = 1.077 grams per ml), in 15.0 ml polyethylene tubes. Tubes were spun at 3.2 g for 10 minutes or until good separation of serum, mononuclear cells, and granulocytes/erythrocytes was effected. The mononuclear layer (between serum and erythrocytes) was harvested with a pasteur pipette and washed twice with 12.0 ml of warm PBS (3.2 g for 7 minutes). Pellets were pooled and washed a final time and the cells resuspended in 1.0-2.5 ml of PBS. Cell number and viability was assessed with a hemocytometer. STAINING OF PERIPHERAL BLOOD MONONUCLEAR (PBMN^ CELLS WITH ANTISERA & MAbs: An appropriate dilution of antibody/antisera was added to1.0-2.0 x 106 MN cells in 5.0 ml polyethylene tubes and brought to 200 uJ with cold PBS. All 51 incubations were on ice for 30-40 minutes in the dark and all washes were with 4.5 ml of cold PBS and 3.2 g x 7 minutes. Serum binding was visualized by a 1 in 50 dilution of Goat anti-rabbit Ig-FITC. The washed pellet was resuspended in 300 ul cold PBS. If samples were to be analyzed (FACScan) or sorted (FACStar)the same day, stained mixtures were stored on ice and in the dark. If samples were to be analyzed the following day by EPICS (Coulter EPICS C-Flow Cytometer, Coulter Electronics, Hialeah, FL), they were fixed in 1-1.5 ml of 1% paraformaldehyde and stored in the dark at 4°C. ISOLATION OF CD3. CD4 and CDB POSITIVE PBMN CELLS FOR STAINING WITH ANTISERA: MN cells from 40 ml of fresh blood were harvested and washed as described above. All incubations were for 20-30 minutes on ice and in the dark and between steps cells were washed twice with 4.5 mis cold PBS. 15 x 107 cells were incubated at 1.0 x 10 7 cells per ml with 1)1 in 20 dilution of NRS, 2)1 in 50 dilution of Goat anti-rabbit Ig-FITC 3)1 in 20 NRS again. Stained cells were then split into three equal portions and treated with 1 in 30 dilutions of anti-CD3-PE, anti-CD4-PE and anti-CD8-PE respectively. Cells were sorted by FACStar and live CD3+ (1st), CD4+ (2nd) and CD8+ (3rd) cells were collected on ice in FCS. CD3, CD4 and CD8 positive cells were pelleted, washed and split into five (CD3+), four (CD4+) or three (CD8+) equal portions in round-bottom Microtiter plates. Each population was screened for binding with anti-p80, anti-p30, and NRS (at 1 in 20 dilutons) The goat anti-rabbit Ig-FITC secondary Ab was preadsorbed with normal mouse serum (1:1, 15 min. on ice) to remove any cross-reactivity with the murine, anti-human PE conjugated reagents already on the cell surface. Samples were analyzed by FACScan. SCREENING CD19+ CELLS FOR P80/P30 REACTIVITY: MN cells were isolated and pretreated with a 1 in 15 dilution of NHS and then incubated with the following reagents: 1)NRS, anti-p80 or anti-p30 at 1 in 20, 2)Goat anti-rabbit-FITC at 1 in 50 (preadsorbed with NHS), 3)1:1 NMS/NRS at 1 in 20 and 4)anti-CD19-PE at 1 in 10. All incubations were for 20-30 minutes on ice and in the dark. Between steps cells were washed with 2 x 4.5 mis of cold PBS and samples were analyzed by FACScan. ISOLATION OF GRANULOCYTES AND SCREENING FOR P80/P30 REACTIVITY: Red blood cells (RBC's) in pellets from two tubes containing ficoll-hypaque-separated whole blood were lysed by repeated incubations in 3 volumes of Tris ammonium cholride 52 for 10 minutes at 37°C. Between treatments cells were washed with warm PBS and the amount of lysis evaluated by visual inspection of the WBC pellet. Once a pellet was obtained that was relatively free of RBC, cells were pretreated with 1 in 15 normal human serum (at 107 cells/ml to block FcR's) followed by NRS, anti-p80 or anti-p30 at 1 in 20 and then goat anti-rabbit Ig-FITC at 1 in 50. Cells were washed between steps and analyzed as above. CELL SURFACE STUDIES WITH HUMAN LYMPHOCYTIC CELL LINES: CELLS: The human cell lines MOLT-4, HuT 78, CEM-CM3 and RAJI were all obtained from the American Type Culture Collection (Rockville, MD). Cells were grown in RPMI (Flow) supplemented with 10% FCS (Gibco) in a humidified, 37°C, 5% C 0 2 incubator. STAINING: Cultured cells in the logarithmic phase were harvested and washed three times with cold PBS. 1.0-1.5 x 106 cells were stained per test and analyzed by EPICS-C flow cytometry as described above. In the initial double label studies, PE-conjugated reagents were added first and the antisera/NRS second. WESTERN BLOT ANALYSIS OF LEUKEMIC CELL LYSATES: LYSATES: Cultured cells in the logarithmic phase were harvested, washed three times with cold (0°C) PBS, resuspended in 1.0 ml of lysis buffer (20 mM Tris HCI-pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100 and 0.5% sodium deoxycholate in ddH20) and vortexed. The resultant aggregate of cells and cell debris was partially broken up by repeated passage through an 18, followed by a 22 gauge syringe. The mixture was incubated at 4°C for 30 minutes, centrifuged for 20 minutes (15 K) and the protein content of the supernatant determined by Micro Lowry Assay. 10 and 20 u.g of each lysate were subjected to reducing SDS-PAGE as is described in Section 1 Materials and Methods. WESTERN BLOTS: 35, 50 and 80 u,g of lysate from each of MOLT-4, HuT 78, CEM-CM3 and RAJI were subjected to reducing (10 mM DTT/100°C x 5 minutes) SDS-PAGE (12% acrylamide). Separated proteins were electro-blotted onto pre-wetted nitrocellulose filters with Towbin Transfer Buffer (25 mM Tris, 192 mM glycine, 0.1% SDS and 20% v/v methanol, pH 8.3). Gels were stained with silver nitrate to determine the relative efficiency of protein transfer. Nitrocellulose blots were washed in NET buffer (15 mM NaCl, 0.5 mM EDTA, 5 mM Tris HCI-pH 7.4, 0.025% gelatin, 0.005% Triton X-100) and incubated overnight with gentle shaking in a 1 in 500 dilution of anti-p30 or normal rabbit serum in NET buffer (50 ml final volume). The next morning, blots were washed with 4 x 50 mis of NET buffer and incubated with a 1 in 500 dilution of sheep anti-rabbit 53 Ig with shaking for 1.5-2.0 hours. After 4 x 50 ml washes with NET buffer, 10 ml per blot of l125-protein A was added to 30-40 ml of NET buffer and incubated for 2 hours with shaking. After 5 x 50 mis washes with NET buffer, filters were air-dried and put with X-ray film (X-omat, Kodak) at -70°C for 3-5 days. 54 SECTION 2: RESULTS ANTI-P80 AND ANTI-P30 REACTIVITY WITH MURINE CELLS Fd11F, a T cell-derived suppressor factor purified from Fd11 (a Ts cell hybridoma) cell supernatants over a B16G-affinity column gives distinct and unique protein bands at 80, 30, and occasionally 45-50 KD with reducing SDS polyacrylamide gel electrophoresis (PAGE). Chu (1989) has shown antigen-binding and suppressive activity associated with the 80 and 30 KD bands. Subsequently, North efa/(1988) eluted the 80 and 30 KD components from preparative gel slices and raised antisera in rabbits. Anti-p30 was shown to bind to the surface of 4% of murine thymocytes and 10% of splenic T cells. In addition, both CD4+ and CD8+ T subsets contained p30+ populations and CD4-CD8 - cells showed enhanced p30 reactivity (30%+). Finally, anti-p30 was demonstrated to abrogate the suppressive effect of affinity-purified Fd11F in vitro. SCREENING HUMAN PBMN CELLS FOR ANTI-P80 AND ANTI-P30 BINDING. An initial investigation was undertaken to determine the reactivity of anti-p80 and anti-p30 sera with normal human peripheral blood mononuclear (PBMN) cells. This involved collecting blood from four individuals and isolating and staining the mononuclear (MN) cells with normal rabbit serum (a control for serum components nonspecifically binding to cell surface proteins), anti-p80 and anti-p30. Binding to the cell surface was detected by the addition of goat anti-rabbit Ig-FITC and flow cytometry. Initial studies of this type invariably demonstrated the existence of a small population of anti-p80/30 intensely staining cells and more variably, a larger population (usually the remainder of the cell population, ie. 65-70%) of weakly staining cells (Fig. 8 for eg.). Given the values obtained for the percent p30 positive T cells in the mouse system, this smaller, intensely-staining population may have represented a specific suppressor-related subset of human MN cells. The range of mononuclear cells intensely staining with anti-p80 serum was found to be 5.6% to 29.5%, whereas the range with anti-p30 serum was lower, 3.1% to 9.8% with both sets of data corrected for nonspecific binding of normal rabbit serum at the same dilution (Table II). It is interesting that both of the individuals who showed significantly higher percentages of high and low affinity anti-p80 55 FIGURE 8: SAMPLE PROFILE OF NORMAL HUMAN MONONUCLEAR CELLS (a) STAINED WITH NORMAL RABBIT SERUM (b), ANTI-p80 (c) AND ANTI-p30 (d). Mononuclear cells were separated from fresh whole blood over ficoll-hypaque and incubated with 1 in 20 dilutions in PBS of NRS, anti-p80 or anti-p30. Cell surface binding was detected by goat anti-rabbit Ig-FITC (1 in 50 in PBS) and samples were analyzed in an EPICS-C flow cytometer gated for lymphocytes. :4 Fig, 8a; Flow Cy-tometer F| Gating. r~' -.'•>>.v.v.. '1 • • : ; ; : ^ ^ : : : : : ; : S ; : : v : Fig, 9p: ap80 + GaRlg-FITC. l±J ti i ! 1 jit, 3 / ' ' 1 MIN MAX COUNT L F L i PERCENT HEAN SD. X HPCV s 1.824 1.377 96 2.4 i .073 1.061 76.7 2 13.43 1023. 12S5 38.3 19.78 2.01 27.6 3 i .454 12.14 2628 65.8 5.848 i.469 18.5 4 i .024 1023. 3981 1BB.B 8.181 2.270 18.5 1 Fia. 8b: NRS + GaRlg-FITC. FigT W; (i u ap30 + GaRIg- T FITC. T„FT 1 MIN MAIS COUNT PERCENT MEAN SO 1 HPCV 1 1.024 1.377 4727 89.7 1,867 1.051 70,6 2 10.43 1023. 93 1.8 17.11 2.01 0.67 3 1.454 10.14 426 8 .1 3.B76 1.836 \" 4,18 4 1.024 1B23. 5270 100 .B 1.225 1.686 70.6 MIN MAX COUNT LFLI PERCENT MEAN SD I HPCV 1 1 .024 1.377 2387 2 2 . 2 1,183 1,093 7 9 , 2 L 12.43 1023. 1116 10,4 2 3 . 2 8 2 , 3 7 4b, i 1.454 10.14 7B9S 6 5 , 9 3 , 6 9 5 1.700 6 2 , 3 4 i . 0 2 4 1023. 1S770- 100.8 3 . 3 9 6 2 . 6 8 0 7 9 , 2 N 57 TABLE II: SCREENING OF HUMAN PERIPHERAL LYMPHOCYTES FOR anti-p80/anti-p30 BINDING: Mononuclear cells were harvested from the fresh whole blood of several healthy donors over ficoll- hypaque. Cells were incubated with 1 in 20 dilutions of NRS, anti-p80 or anti-p30 followed by GaRlg-FITC at 1 in 50. See Fig. 8 for details. Statistics were obtained from the amount of binding above that of the negative control (NRS). DONOR* ANTIBODY % POSITIVE3 1 2 3 4 5 ocp80 II II II II 5.6 6.6 27.4 10.5 29.5 1 2 3 4 5 ctp30 II II II II 3.5 3.6 8.6 3.1 9.8 aAII fluorescence values corrected for binding of normal rabbit serum. 58 and anti-p30 binding are atopic. Upon repeated testing of the same individual (data not shown), the percentage of p80 and p30 intensely staining cells was shown to vary, sometimes by as much as 10%. Thus the results indicate that two different populations of anti-p80 and anti-p30 positive cells exist in the mononuclear fraction of human blood. Each of these populations express the p80 and p30 markers at different densities or, express a different form of these proteins for which the specific Abs in the sera have more or less affinity. In addition, there is variability in anti-p80 and anti-p30 binding between individuals and at different times within the same individual. EXPRESSION OF THE P80 AND P30 DETERMINANTS BY HUMAN T LYMPHOCYTES, In order to begin characterizing these populations of human anti-p80/anti-p30 positive mononuclear cells, CD 3, CD 4 and CD 8 positive cells were isolated from PBMN cells pretreated with NRS and goat anti-rabbit Ig-FITC by cell sorting and stained with each of anti-p80, anti-p30, NRS and goat anti-rabbit Ig-FITC alone (preadsorbed with normal mouse serum). The resulting FACScan profiles are presented in Fig. 9 where it can be seen that CD3 + cells display a 1 log increase in anti-p80 binding and a slightly smaller increase in anti-p30 binding over NRS. Thus T cells express lower levels of anti-p30 than anti-p80, or the anti-p30 sera has a lower affinity for T cells than the anti-p80 sera. A similar result was obtained for the CD 4 + and CD 8 + subsets (data not shown). In addition, there is a small population of CD3 + cells (2.1% with anti-p80 and 1.5% with anti-p30) which stain at a much higher intensity (1.5-2 logs increased binding over background) than the rest of the preparation. A slightly larger population showed this same level of binding in the human screen discussed above. Whether these two intensely staining populations represent the same cell subset (potentially of T cell origin) with a high affinity for the antisera, or artefactual binding is not known. The antisera profiles of CD4 + cells gave 4.5 and 3.5% intense staining with anti-p80 and anti-p30 respectively whereas less than 1% of CD8 + positive cells stained at this level with both antisera (data not shown). 59 FIGURE 9: ISOLATED HUMAN CD3 + CELLS BIND ANTI-P80 AND ANTI-P30. Human mononuclear cells were isolated as in Fig. 8, and treated with 1)NRS (1 in 20), 2)GaRlg-FITC (1 in 50), 3)NRS (1 in 20) and 4)ccCD3-PE (1 in 30). All incubations were for 20-30 minutes on ice and in the dark. Cells were washed with 4.5 mis of ice cold PBS between each step. Stained cells were sorted by FACStar and CD3+ cells were collected in FCS. CD3+ cells were then washed and screened for reactivity with NRS (b), anti-p80 (c) and anti-p30 (d) as has been described. GaRlg-FITC (preadsorbed with NMS) was used at 1 in 50 dilution and samples were anlysed by FACScan. GaRlg-FITC alone (a). Fia. 9a: Flow Cy-tometer Gating. Fia. 9b: NRS + GaRlg-FITC. ' ' • i i i i i i i i i i i i i i i i i i i i e S Q Q 4 0 0 S Q 0 'eae 1 iaac F o r u i r t l S o t t k r Fig. 9c: ap80 + GaRlg-FITC. ae a a (A 3 L. a <* c I -I »• ' •• ' PP? t t pP? • ( PP? r • PP? t t \\ i a a F j q 9 d , < ' i ' i ' i i i i T [ ' I i n u n ) — r i a B I B 1 I B 8 I B 3 I B * ap30 + GaRlg-FITC. 61 THE BINDING OF ANTI-P80 AND ANTI-P30 TO THE SURFACE OF HUMAN B LYMPHOCYTES. Since the data indicated that human CD3+ MN cells bound anti-p80 and anti-p30 (to a lesser extent), it was of interest to investigate whether peripheral B cells showed any reactivity with these reagents. PBMN cells were double labelled with NRS, anti-p80 or anti-p30 (followed by goat anti-rabbit Ig-FITC) and anti-CD19-PE. Fig. 10 shows two different populations of MN cells which are binding the CD19-PE antibody. From a subsequent experiment, it was determined that the more intensely staining cells were specifically binding anti-CD19 and thus it was concluded that the weaker staining population was nonspecific or artefactual. The CD19+ cells show approximately the same level of antisera binding (1 log increase over NRS with anti-p80) as the CD3 + cells with lower levels of anti-p30 binding (log intensity was taken from the centre of each treated CD19 + population as with NRS and anti-p30 samples staining was more variable). This experiment suggests that peripheral B lymphocytes also express the p80 and p30 determinants. A B cell subpopulation staining with an intensity greater than that observed for the main population was not observed in this experiment. EXPRESSION OF THE P80 AND P30 DETERMINANTS BY HUMAN GRANULOCYTES. Since the p80 and p30 antisera were shown to bind to peripheral T and B lymphocytes, an experiment was done to investigate whether cells of the myeloid lineage also expressed this marker. Granulocytes were obtained by repeated Tris ammonium chloride treatment of RBC pellets from a sample of whole blood separated over Ficoll-hypaque. The resultant WBC pellet was washed and pretreated first with normal human serum (NHS) to block FcR's and second with normal rabbit serum (NRS) to adsorb out any nonspecific binding to cells by serum components. A portion of the cells stained with anti-CD19-PE and anti-CD3-PE showed that the preparation contained very few contaminating lymphocytes. The remaining cells were screened for NRS, anti-p80, and anti-p30 binding. As can be seen in Fig. 11, even though the Goat anti-rabbit Ig-FITC secondary Ab (preadsorbed with NHS) is binding to the human Abs on the surface of the granulocytes, addition of NRS (negative control) does not shift the peak. However, addition of anti-p80 or anti-p30 (followed by GaRlg-FITC) does shift the entire peak to the right by one log. In this experiment, equal levels of binding were obtained with both antisera. Thus it appears that 62 •FIGURE 10: DOUBLE LABELLING OF HUMAN MONONUCLEAR CELLS WITH ANTI-CD19 AND ANTI-P80/ANTI-P30. Human mononuclear cells were isolated as in Fig. 8, and pretreated with a 1 in 15 dilution of NHS. Cells were washed and incubated with reagents in the following order. 1)NRS (b), anti-p80 (c) or anti-p30 (d) at 1 in 20 2)goat anti-rabbit Ig-FITC at 1 in 50 3)NMS and NRS each at 1 in 20 and 4)anti-CD19-PE. All incubations were from 20-30 minutes on ice and in the dark. Cells were washed twice with 4.5 mis of ice cold PBS between steps and stained populations were analyzed by FACScan with the gating set for lymphocytes (a). Fluorescence 2 (y-axis) represents binding of anti-CD19-PE (red) and Fluorescence 1 (x-axis) repre-sents binding of goat anti-rabbit Ig-FITC (green). Fia.10a: GaRlg-FITC only. CO a a c a t a 3 I B 1 —i i 111 m i r 1 B B ' l B * fi'i iirh' — i — i 111 mi I B * ' i B 3 i I I I I inig IBTBB E B B [2T3B I B P l u o r a i e t t t D Q 1 Fio.10c: ap80 + GaRIg- j j 1 0*^ FITC. H I B 1 H B B K B B (TOB ISfBB — I I 111 II II • V r f m W , w , r r i i i i i i | 1 m i I B ' i B 1 ' l B a ' i B 3 r i u o r i t D t n e a 1 I B I B a-J. Fia.10b: a a NRS + GaRlg-FITC. c a \" I B 8 ' I B 1 ^ 1 B B \" 1 B ' i I I I I m i ' i i 111 m i I B I l l l l M i l 1 I I I I III 8 ' 1 B 3 I B B B E B B FSTBB B B B BT I B ' r i u o r a s B B n o s 1 Fig.lOd: „, 1 BJ ap30 + S GaRlg-FITC. o S*BS-a 3 I B I B 1 — i i i l i i i i f '^PT'l 'f i i i i i ' i i i n n i i — i i i i i m p A I B B ' I B 1 ' I B 3 ' I B 3 ' I B 4 lETBB E B B pFBB E B B r i u o r t t e i r i D i 1 64 FIGURE 11: CELL SURFACE ANALYSIS OF HUMAN GRANULOCYTES USING ANTI-P80 AND ANTI-P30. Human granulocytes were isolated from red blood cells in the pellets of a ficoll-hypaque separated whole blood by repeated 10 minute incubations (at 37°C) with Tris ammonium chloride. The degree of remaining RBC's was assessed by pelleting cells between incubations. Isolated granulocytes were washed and screened for contaminating lymphocytes with ccCD3-PE and aCD19-PE. Cells were pretreated with a 1 in 15 dilution of NHS and then incubated with NRS (b), anti-p80 (c) or anti-p30 (d) at 1 in 20 dilutions, followed by GaRlg-FITC (preadsorbed with NHS) at 1 in 50. GaRlg-FITC alone (a). Samples were analyzed by a FACScan gated for granulocytes.. 12Q l a 0 B . • l*BB • . F E B , , l B B B . , ItBBB F i g . 1 1 a : • J G a R I g -F I T C o n l y . a 3 L. a t> c 3 a U I B * 12B F i o . 1 1 d : ap30 + G a R l g -F I T C . D U (fl 3 L a « J c 3 a U PP? • , KtBB , , |6BB , |BBB , , j lBBB i B B IB 1 IB' r i u D r a s e g n c B 1 66 neutrophils (the main cell type in the granulocyte population), in addition to lymphocytes express the p80 and the p30 cell surface markers. SCREENING OF HUMAN LYMPHOCYTIC CELL LINES WITH ANTI-P80 AND ANTI-P30. To furthur study the binding characteristics of anti-p80 and anti-p30, several human lymphocytic cell lines (obtained from the ATCC) were tested with limiting dilutions of both antisera. A sample anti-p80/anti-p30/NRS EPICS-C profile, showing the 1 in 20 dilution from the titration series is shown for each of the cell lines in Fig. 12a-d and the titration curves for MOLT-4, CEM-CM3 (both isolated from human T cell leukemias), HuT 78 (isolated from a patient with Sezary Syndrome) and RAJI (isolated from a Burkitt's Lymphoma) with anti-p80 and anti-p30 are shown in Fig. 13a and 13b respectively. Since 1 in 20 is the dilution of sera used throughout the cell surface studies, the percent of cells staining at this dilution was used to evaluate which lines were anti-p80+ and anti-p30+. HuT 78 is the only cell line which shows strong binding (between 90 and 100% of cells were positive-even at 1 in 80 dilution) of the anti-p80 antiserum. MOLT-4, CEM-CM3 and RAJI lines all gave between 20 and 50% binding of anti-p80 at 1 in 20. Conversely, with anti-p30, 90-100% of both MOLT-4 and CEM-CM3 cells, 70-80% of RAJI cells, and less than 10% of HuT 78 cells stained positve. These cell lines were also tested for other T and B cell markers, the results of which are presented in Table III. WESTERN BLOT ANALYSIS OF HUMAN CELL LINES. In an attempt to characterize the marker to which the p30 antiserum was binding, Western Blot analyses of leukemic cell lysates were undertaken. Since HuT 78 did not bind the anti-p30 serum at any dilution and anti-p80 bound to all of the cell lines to varying extents, to avoid ambiguity, anti-p30 was chosen over anti-p80 in the analysis, and HuT 78 cell lysates were used as a negative control. Cell lysates were subjected to reducing SDS-PAGE (Fig. 14) to show that the protein concentration was approximately equal in all lysates and that good separation of bands was being achieved. 35, 50 and 80 u\\g of each cell lysate were separated and blotted onto 67 FIGURE 12: ANTI-P80 AND ANTI-P30 REACTIVITY WITH SEVERAL HUMAN LEUKEMIC CELL LINES. MOLT-4 (a), CEM-CM3(b), HuT 78(c) and RAJI (d) cells in the logarithmic phase were harvested, washed with cold PBS and incubated with NRS (ii), anti-p80 (iii) or anti-p30 (iv) at 1 in 20 dilutions followed by GaRlg-FITC at 1 in 50. Incubations were for 30-40 minutes on ice and in the dark. Cells were washed twice with 4.5 mis of ice cold PBS between steps and analyzed by EPICS-C. The gated cell populaton is shown in (i). Fia.12a lUi Flow Cy-tometer Gating. Fl S i :4 c 0 Fia.12a U (iii): N T ap80 + GaRlg-FITC. jferftf-Jn.rfiltl-ft.:!. L F L 1 4j MIN MAX COUNT PERCENT MEAN SD I HPCV 1 1.024 10.42 2618 65.4 6.983 1.392 16.4 n i. 90,28 1023. 19 0.5 453.2 2.2 •j 10.43 87.82 1368 34.1 14.34 1.43 16.7 4 1.024 1023. 4006 100.0 9.112 1.759 16.4 c 0 Fig. 12a y (ii): N TP NRS + GaRlg-FITC. MIN MAX COUNT L F L 1 PERCENT MEAN 1.024 10.42 90.28 1023. 10.43 87.82 1.B24 J023. 2V39 2 17 2959 0.1 0.6 106.2 28.12 1.440 1.1 1.99 1.521 HPCV 8.25 0.40 oo Fia.12a U (iv): N ap30 + GaRlg-FITC. T 1, 1 i .SBfl l i Jk_ 3'r PERCENT MIN MAX COUNT MEAN 3D I HPCV 1 1.024 10.42 237 8.2 7.885 1.365 30.2 2 90.28 1023. 33 1.1 208.6 2.6 0.53 3 10.43 87.82 2623 90.6 21 . 0 0 1.56 13.0 4 1.024 1023. 2896 100.0 20.61 1.80 13.0 Fia.12b Oil Flow Cy-tometer Gating. 1 / :llt*t \") y ss Fig.12b ULU-I ap80 + GaRlg-FITC. MIN MAX L F L l COUNT PERCENT MEAN 1 1.024 1.377 2 1.454 7.141 3 7.341 1023. 4 1.024 1023. 7 1404 546 2009 0.3 69.9 27.2 1.119 4.834 10.43 5.988 1.106 1.328 1.78 1.700 4 SD X HPCV 53.1 15.3 17.6 15.3 NRS + GaRlg-FITC. MIN MAX COUNT L F L l PERCENT MEAN SD 7. HPCV 1 1.824 1.377 558 26.2 1.126 1.182 88.9 2 1.454 7.141 1494 71.2 2.288 1.383 26.3 3 7.341 1023. 10 0.5 20.86 1.87 0.57 4 1.024 1823. 2099 188.0 1.980 1.547 88.9 Fig.12b Q OsOi N ap30 + GaRlg-FITC. L F L l MIN MAX COUNT PERCENT MEAN SD X HPCV 1 1.024 1.377 3 0.1 1.130 1.113 84.9 2 1.454 7.141 122 5.4 5.183 1.352 1.33 3 7.341 1023. 2142 94.1 17.26 1.62 11.3 4 1.024 1023. 2276 100.0 16.07 1.75 11.3 Fia.12c Oil Flow Cy-tometer Gating. o o Fig.12c UiiU ap80 + GaRlg-FITC. C 0 u MIN MAX L F L l B..Li.lrWtt.t.uad.jL.„faMJ.t_tJt!a 1 i.024 16.93 2 97.90 1023. 3 16.95 97.83 4 i.024 1023. COUNT PERCENT MEAN SD 8 0.4 4.703 2.957 211 10.8 248.1 2.5 1.40 1728 88.8 46.25 1947 100,0 54.97 2.84 •/; HPCV 84.9 6.47 6.47 4 Fig.12c UiU NRS + GaRlg-FITC. C 0 u N T MIN MAX L F L l COUNT PERCENT MEAN 1.824 16.93 97.90 1023. 16.95 97.83 1.824 1023. 1061 8 21 1090 97.3 8.141 0.7 2i2.7 1.9 31.93 SD 1. HPCV 1 - 427 2.5 1.82 8.77 8.23 2.84 8.77 >i o c 0 Fig.12p u (iv): N T ap30 + f GaRlg-FITC. L F L l MIN MAX COUNT PERCENT MEAN SD I HPCV i i.024 i6.93 i 917 93.7 18.58 1.39 5.92 2 97.98 1823. 12 8.6 315.2 2.7 16.95 97.83 184 8.7 23.96 1.56 17.4 4 1.824 1823. 2113 108.0 11.59 1.64 5.92 Fia.12d ( U i Flow Cy-tometer Gating. Fia.12d y 1111X1 N ap80 + GaRlg-FITC. HIN HAX COUNT L F L l PERCENT HEAN SD X HPCV 1 1.824 1.377 6 8.4 1.185 1.897 53.1 2 1.454 7.141 787 42.0 5.080 1.319 15.9 3 7.341 1023. 935 55.5 16.66 2.86 14.9 4 1.824 1023. 1684 100.0 9.833 2.744 14.9 Fig. i2d f> NRS + GaRlg-FITC. HIN MAX COUNT L F L l PERCENT MEAN SD I HPCV 1 1.824 1.377 13 8.8 1.150 1.117 63.7 2 1.454 7.141 1491 86.5 3.854 1.389 9.48 3 7.341 1023. 288 12.1 21.46 2.94 1.44 4 1.824 1823. 1723 100.0 4.713 2.140 9.40 c 0 Fia.12d u Uvil N T ap30 + GaRlg-FITC. MIN HAX COUNT L F L l PERCENT HEAN SD I HPCV 1 1.024 i.377 2 0.1 1.866 1.838 84.9 2 1.454 7.141 145 8.7 5.307 1.322 29.7 3 7.341 1823. 1518 90.2 19.83 2.67 14.6 4 1.824 1023. 1674 188.8 17.45 2.77 14.6 72 o H 1 1 1 1 1 1 — • r 0 20 40 60 80 RECIPROCAL DILUTION OF ANTISERA FIGURE 13a: TITRATION OF ANTI-P80 ON HUMAN LEUKEMIC CELL LINES. FIGURE 13: TITRATION OF ANTI-P80 (a) AND ANTI-P30 (b) ON THE SURFACE OF MOLT-4, CEM-CM3, HuT 78 AND RAJI HUMAN CELL LINES. Cells were harvested and treated as in FIG. 12. Cells were incubated with doubling dilutions (from 1 in 10 through to 1 in 80) of NRS, anti-p80 or anti-p30. GaRlg-FITC was used at 1 in 50. and incubations were for 30-40 minutes in the dark and on ice. Cells were washed between steps with 2 x 4.5 mis cold PBS and analyzed by EPICS-C. Plotted points have been corrected for NRS binding at the same dilution. 73 FIGURE 13b: TITRATION OF ANTI-P30 ON HUMAN LEUKEMIC CELL LINES. 74a TABLE III: BINDING OF ap30/ap80. AND CELL SURFACE CHARACTERIZATION OF HUMAN LYMPHOCYTIC CELL LINES. Cells in the logarithmic phase were harvested, washed and incubated with rabbit antiserum followed by goat anti-rabbit Ig-FITC or directly conjugated monoclonal antibodies specific for various cell surface markers. See Fig. 13 for details. Values given as percent positive (%+) FITC is the amount of binding above control serum in green fluorescence and %+ P E , the percent positive with red fluorescence (phycoerythrin). CELL LINE ANTIBODY DILUTION %+(FITC/PE) MOLT-4 ap80 1/10 1/20 1/40 1/80 1/10 1/20 1/40 1/80 1/10 1/20 1/30 1/30 1/30 94.3 91.3 29.7 3.7 71.8 34.2 3.0 1.1 0.0 0.0 2.1 85.1 59.0 ap30 B16G WT-31 aCD3 aCD8 aCD4 HuT 78 ap80 1/10 1/20 1/40 1/80 1/10 1/20 1/40 1/80 1/16 1/30 1/30 1/30 97.0 97.0 89.4 68.0 12.4 6.7 5.0 3.8 91.5 96.2 33.4 98.2 ap30 It WT-31 aCD3 aCD8 aCD4 CONT.. 74b TABLE III CONT, CELL LINE ANTIBODY DILUTION %+(FITC/PE) CEM-CM3 ap80 1/10 66.7 1/20 26.7 1/40 26.0 1/80 18.5 ap30 1/10 97.3 1/20 93.6 1/40 93.5 1/80 64.3 WT-31 1/20 4.2 aCD3 1/30 18.8 aCD8 1/30 2.0 aCD4 1/30 90.9 RAJI ap80 1/10 73.7 1/20 42.2 1/40 37.1 1/80 36.6 ap30 1/10 87.3 1/20 78.1 1/40 52.4 1/80 26.4 aCD19 1/30 89.5 WT-31 1/20 2.8 aCD8 1/30 3.2 aCD4 1/30 3.5 39-FIGURE 14: SDS-PAGE OF HUMAN CELL LINE LYSATES. The protein content in each of the cell lysates was determined by Micro Lowry Assay. 10 and 20 p.g of each lysate was reduced (10 mM DTT/100°C x 5 min.) and loaded onto a 12% acrylamide mini-gel with prestained protein standards (Bio Rad). Proteins were separated and stained with 0.1% silver nitrate. 76 nitrocellulose. Blots were incubated with a 1 in 500 dilution of anti-p30 or NRS (followed by sheep anti-rabbit Ig) or sheep anti-rabbit Ig alone. Reactive protein bands were visualized by incubating blots with Protein A coupled I125 and autoradiography at -70°C. In Fig. 15, photos of an autorad obtained from l125-labelled blots containing 80 |ig of each cell lysate are shown. It can be seen that anti-p30, but not NRS (or SceRIg, data not shown) binds to one main protein band with an M r of approximately 42KD in all cell lysates and to a more minor band with an M r of 25 KD in RAJI and CEM-CM3 lysates only. Two distinct but weaker bands were observed at 60 and 80 KD in all cell lines (the 80 KD band is not visible in the enlarged photo in MOLT-4 or RAJI lanes). Also, in each lysate, high molecular weight material which did not enter the gel to a significant extent was shown to bind anti-p30. The 42 KD band was not visible on blots containing 35 u.g of HuT 78 lysate whereas it was clearly visible with the same amount of lysate from other cell lines (data not shown). Although anti-p30 did react with a protein of 25 KD in two of the cell lines, it did not bind with any material around 30 KD. 77 NRS ANTI-P30 FIGURE 15: WESTERN BLOTS OF HUMAN CELL LINE LYSATES WITH ANTl-P30. Reducing SDS gels containing 80 | ig of each cell line lysate were run, blotted onto nitrocellulose and incubated with a 1 in 500 dilution in NET buffer of NRS or anti-p30. Binding was assessed by the addition of sheep anti-rabbit Ig (1 in 500) followed by protein A-conjugated 1 2 5 l and autoradiography for 5 days at -70°C. 78 SECTION 2: DISCUSSION Given the observation that antisera raised to reduced components of the affinity-enriched T suppressor factor Fd11F binds to distinct populations of murine T lymphocytes, these reagents were tested for reactivity with human lymphoid and myeloid peripheral blood cells. Initial studies showed two positive populations of PBMN cells, one strongly positive and one weakly positive (Fig. 8). Further, the percentage of intensely staining cells (from 5.6% to 29.5% with anti-p80 and from 3.1% to 9.8% with anti-p30) was found to be variable both between individuals and at different time points within the same individual (Table II). The observed variability may be a result of harsh in vitro conditions imposed on the cells, or reflect a characteristic of the p80/p30 positive population(s). For example, lymphocytes from different individuals may vary in their sensitivity to ficoll-hypaque, being centrifuged and pipetted repeatedly or being incubated on ice. Potentially, any of these treatments could cause cell surface changes that might result in increased (most likely) or decreased binding of antisera. Alternately, it has been observed that the number of peripheral B lymphocytes varies both between individuals and on a daily basis within the same individual (Clark, P. et al., 1986 for eg.), thus it seemed conceivable that the observed variability was a naturally occurring phemonenon. After several attempts to double label MN cell preparations with anti-T cell marker Abs and anti-p80/p30 failed due to high amounts of nonspecific binding, a different method was used which involved collecting cells that bound anti-CD3-PE but not NRS (visualized by GaRlg-FITC) by cell sorting, thus eliminating from the preparation those cells which bound serum components, and staining with NRS, anti-p80 and anti-p30. The results (Fig. 9) indicated that all of the CD3 + (as well as CD4 + and CD8+) cells expressed the p80 marker(s) and most (at least 75%) of the CD3 + cells expressed p30 determinant(s). Thus peripheral T cells express lower levels of p30 than p80, or alternately the p30 antisera has a lower affinity than the p80 antisera for T cells. Since the range of peripheral T cells (65-70%) present in MN cell preparations corresponded well with the number of cells found in the weakly staining peak in the initial screening experiments, and given the difficulties experienced with background in trying to double label this population, it seemed probable that this peak, which was originally assumed to reflect nonspecific binding of the immune serum, actually represented T cells. In this experiment it was also found that a subpopulation of T cells (2.1% with anti-p80 and 79 1.5% with anti-p30) bound the antisera to a much greater extent-possibly correlating with the intensely staining population observed in the human screening study. It was of interest to investigate whether the p80 and p30 determinants were expressed on all MN cells of the lymphoid lineage or only on the T cell subset. Since there are much lower numbers of peripheral B cells than peripheral T cells and the cell surface characteristics of B cells predispose them to binding antibody (FcR+), an attempt was not made to isolate CD19+ cells from MN cells by flow cytometry. Instead, this population was double labelled directly with antisera (and GocRlg-FITC) and anti-CD19-PE. Although there was some artefactual or nonspecific red fluorescence evident in these profiles (Fig. 10), once the more intensely staining population was shown to be the one binding anti-CD19-PE by a subsequent experiment, it was concluded that B lymphocytes bound anti-p80 and anti-p30 at levels similar to that observed with T lymphocytes (ie. 1 log increase in binding over NRS with anti-p80 and slightly less of an increase with anti-p30). Given that cells of the human lymphoid lineage had been shown to express both the p80 and p30 markers, cells of the myeloid lineage were next investigated. Granulocytes were harvested, pretreated with NHS to block free FcR's and screened for anti-p80 and anti-p30 binding. The gated population displayed binding to the goat anti-rabbit Ig-FITC secondary, indicating that this anti-rodent Ig reagent also recognized human Ig bound to the granulocyte cell surface. Aside from this, it was still clearly evident that this cell population was p80+ and p30+ as a full log higher binding over NRS was observed with both immune sera (Fig. 11). Treatment of granulocytes with anti-CD3 and anti-CD19 showed that very few lymphocytes contaminated the cell preparation. The finding that neutrophils, the major cell type of the myeloid lineage bind anti-p80, anti-p30 and goat anti-rabbit Ig-FITC, raised the possibility that the variability in intensely staining cells in initial experiments was due to varying numbers of contaminating granulocytes. What appeared to be intensly staining cells may have been a reflection of the binding of antisera on top of the already high amount of binding with goat anti-rabbit Ig. Eventhough MN cell preparations were not pretreated with NHS, neutrophil FcR's are known to exist in constant equilibrium with serum Ig, and thus it is conceivable that these cells were effectively slg + as in the granulocyte experiment. In retrospect, it would have been useful to have consistently tested MN cell preparations with a granulocyte- or neutrophil-specific MAb to determine the degree of contamination. Also, if this were the case, the same intensely staining population should have been observed with the NRS treated MN cells although at a relatively lower level (since GaRIg-80 FITC is used as the secondary Ab). The origin of the cell population which stains intensely with anti-p80 and anti-p30 sera remains uncertain. From the data presented, it appears that the larger, more weakly staining population of MN cells observed in the initial screenings is comprised of T cells and most likely B cells (both bind approximately the same levels of antisera). Since the T cell population also contained small subsets of intensely p80+(2.1%) and p30+(1.5%) cells, these may represent the true p80 and/or p30 positive population, thus suggesting that lower levels of binding represent cross-reactivity with another related or unrelated human cell surface protein or alternately, that the immune sera contain antibodies to a ubiquitous leucocyte marker in addition to the Fd11 F-associated products. This latter explaination is plausible, given the finding that the bands excised from preparative gels have been shown to contain more than one protein species by isoelectric focussing (IEF, R. Chu, 1989). The observation that all of the profiles obtained with anti-p80 and anti-p30 were very similar in levels of binding and range of activity, suggest that p80 and p30 are of a common origin, ie. are epitopes on the same protein or molecular complex. Prior to the discovery that cells other than T cells expressed p80 and p30, a study was undertaken in an attempt to characterize the molecular size of the cell surface marker(s) being recognized. Four leukemic cell lines were ordered from the ATCC; MOLT-4, isolated from a mature T lymphocytic leukemia, CEM-CM3 isolated from an acute T lymphoblastic leukemia, HuT 78, from a patient with Sezary Syndrome (a cutaneous T cell lymphoma) and RAJI, from a Burkitt's Lymphoma (B cell). 90% of HuT 78 cells were shown to bind anti-p80 even at low dilution (1 in 80). In contrast, the other three cell lines all dis-played much lower levels of reactivity. Conversely, MOLT-4, CEM-CM3 and RAJI all bound anti-p30, whereas HuT 78 cells were effectively p30\" (as assessed at a 1 in 20 dilution). Given the clarity of this latter result with anti-p30, this reagent was chosen to screen Western blots of cell lysates obtained from each of these cell lines. The same amount of each cell lysate was analyzed by Western blot analysis using anti-p30/NRS, sheep anti-rabbit Ig and Protein A-conjugated 1 2 5 l . Anti-p30 was shown to bind to one major band with an approximate M r of 42 KD in all cell lysates and to a minor band of 25 KD in RAJI and CEM-CM3 lysates only. In addition, weaker binding to bands of 60 KD, 80 KD and high molecular weight material was observed in all lysates. The blot containing 35 u,g of each lysate suggested that the 42 KD band was missing in HuT 78 cells, however it is present in equal intensity to that observed in the lysates from the other cell lines on 50 and 80 u,g blots (Fig. 15). Only the 25 KD band is absent in HuT 78 lysates. However, this band is also missing 81 in MOLT-4 lysates. The cell surface data obtained with anti-p30 (Fig. 13b) show that CEM-CM3 express the highest levels of the p30 determinant(s) whereas HuT 78 cells express virtually none. At the lowest dilution tested (1 in 80), it can be seen that very few MOLT-4 cells (6%) bind anti-p30 whereas close to 30% of RAJI cells do (approx. 70% of CEM cells are positive at this dilution). Given the fact that a dilution of 1 in 500 of anti-p30 was used in the Western blot analysis, the cell surface data do correlate with the Western blot data. It would be more convincing if there were a visible increase in intensity of this band in the CEM-CM3 lane (expresses higher levels of p30) versus the RAJI lane. Also, as was mentioned, the 42 KD, not the 25 KD band is the major band visible on anti-p30 blots. If the 42 KD band contains the protein of interest, in light of the anti-p30 cell surface data, it can be concluded that HuT 78 cells synthesize, but do not express the p30 determinant(s). If the molecule bearing p30 is normally expressed on the cell surface in a complex with one or more other molecules, deletion or mutation of one molecule may completely abolish cell surface expression of the other. A well-documented example of this is the CD3/TCR complex. In fact, the MOLT-4 cell line utilized in this study has a TCRocC gene deletion and is cell surface CD3\" as well as TCRapV. Leukemic cell lines are known to harbour substantial mutations and chromosomal aberrations. In fact, HuT 78 has been shown to have a translocation of 2q34 to 3' of c-myc on chromosome 8. An additional point of interest regarding HuT 78 cells is that they grow very slowly in culture. The ATCC recomends the addition of rlL-2 to cultures to increase the doubling time. It is possible that the molecule bearing the p30 determinant is synthesized by HuT 78 cells but not expressed on the cell surface unless the cell is activated. It would be interesting to determine the reactivity of HuT 78 cells grown in cultures with exogenously added IL-2 with anti-p30. The lack of HuT 78 p30 surface expression may also be due to a defect in intracellular processing or transport. The observation that anti-p30 did not bind to any 30 KD material in lysates is not that surprising given cross-species variation and the above mentioned uncertainty of the content of the priming antigen preparation in the production of the antisera. In humans, the 25 KD band may contain the the murine equivalent of the p30 determinant. Alternately, the 25 KD band may represent nonspecific binding of the immune serum and the 42 KD band may contain the protein of interest. It is possible that the material to which anti-p30 was raised against was an antigen-binding breakdown product of Fd11F. Since Fd11F is secreted and/or cell surface, it is likely that very little would be found in lysates obtained from healthy p30+ cells. If the 42 KD band represents specific anti-p30 82 binding, evidence for this would have been the presence of a distinct, or much more intense band in HuT 78 lysates (given the conclusion that these cells are unable to express or secrete the p30+ product, it would have to be broken down intracellularly). On the other hand, if the 25 KD band is of significance, then HuT 78 cells synthesize low levels of p30+ protein (similar to MOLT-4 cells), or none at all. Other Western blot analyses done in this laboratory on Fd11 and A10 (Ts hybridomas) lysates have also shown a lack of reactivity of anti-p30 with a 30 KD protein band (M. MacDonald, unpublished). The binding of anti-p30 to 80 and 42 KD proteins does correspond somewhat with data obtained in the murine system, in that reduced Fd11F affinity-enriched preparations have been shown to be comprised of bands with Mr's of 80 KD, 45 KD (more variable) and 30 KD (R. Chu, 1989 for eg.). The binding of anti-p30 to an 80 KD, 60 KD and possibly 25 KD bands in cell lysates may represent cross-reactivity with cellular proteins. Given that the 80 KD band present in affinity-enriched murine Fd11F preparations has been shown to harbour antigen-binding and suppressive capabilities, it may be a coincidence that a protein of approximately the same size is cross reacting with the anti-p30 serum. It is however possible, that the human equivalent of Fd11F is expressed on the cell surface as a multimer of 42, 80 and/or 60 KD components. Only limited comparisons can be drawn between the work presented here and the anti-p30 murine cell surface labelling studies of North er al. (1988). The staining protocol used in the one- and two-colour analysis of murine thymocytes was very similar to that used in the initial screening of human MN cells with anti-p30/anti-p80 and profiles with similar characteristics were obtained. However, the protocol used in this work to further characterize the p80/p30+ human population which involved removing NRS-reactive cells with goat-anti-rabbit Ig-FITC and collecting CD3 + cells in a cell sorter before screening for reactivity with the antisera has shown that all human peripheral T cells bind anti-p80 and anti-p30 (to a lesser extent). Also, although North et al. screened several other murine tissues with anti-p30, B lymphocytes and cells of the myeloid lineage were not tested. An important difference between these two sets of experiments is that murine T cells were obtained from the spleen and thymus, whereas human leucocytes were obtained from peripheral blood. This may be significant given the fact that cells of the immune system express specific receptors or adhesion molecules when in ciruclation and in response to injury or infection (Springer, T., 1990). This is addressed further (in light of recent findings) below. The lower levels of antisera binding to whole MN cell preparations versus that 83 obtained with isolated T cells is a paradox. In retrospect, it would seem that a better control rabbit serum than that obtained from pre-immunization bleeds would have been immune serum (raised against a molecule known not to be expressed by the test popu-lation). It appears that one of the components in the immune sera sticks and masks the p80 and p30 determinants in whole MN cell preparations, giving lower levels of binding and making statistical and two colour analysis very difficult. However, binding of anti-p80 and anti-p30 to cell lines was reproducible and always at significant levels over that of NRS. This may be due to higher levels of p80 and p30 cell surface expression on postive tumour cell lines. An alternate solution to this paradox may relate to the state of cell activation as was briefly discussed for HuT 78 lysates. Since anti-CD3 antibodies are known to stimulate T cell blastogenesis (rev. in Altman et al., 1990), the pan-T reactivity of isolated T cells may be due to activation-induced up-regulation of p80/p30 expression. Eventhough CD4 and CD8 molecules are known to be associated with T cell signal transduction and activation, the fact that alt sorted CD4 + and CD8 + cells pretreated as above also expressed p80/p30 argues against this theory somewhat. In addition, the p30/p80 determinants were also found to be expressed by resting cells of the myeloid lineage (unless neutrophils are activated by in vitro manipulation). Anti-p80 and anti-p30 were raised against antigen-binding moieties present in Fd11F preparations affinity-enriched from a Ts1 hybridoma (Fd11) culture supernatant using the TsF framework antibody B16G. This information, considering the fact that only a proportion of peripheral blood CD4 + T lymphocytes function as suppressor-inducer cells, raises the question of the nature and function of the p80/p30 determinant in the human (and possibly the murine) immune system. Any model which postulates the function of the p80/p30 protein must take into account its possibly ubiquitous expression by peripheral white blood cells (monocytes, basophils, eosinophils and mast cells were not specifically tested). A recent review by T. Springer (1990) contains a discussion of a family of immune system adhesion molecules named 'selectins'. Selectins have varied functions in adhesion yet all appear to be involved with the regulation of leucocyte binding to endothelium at the sites of inflammation. One of the human molecules within this family, LAM1 (for leucocyte adhesion molecule 1), has been shown to bear the epitopes recognized by antibodies raised to MEL-14 (a murine lymphocyte homing receptor) and Leu 8 (a fairly ubiquitous cell surface marker of circulating white blood cells which has also been shown to define suppressor-inducer subpopulations in the human CD4+ T cell subset - see 84 Introduction). It is interesting that upon immunoprecipitation, the Leu 8 marker was originally described as having a molecular weight of 80 KD. The LAM1 molecule has recently been immuno-precipitated from peripheral T cells and found to be 74 KD under reducing conditions, whereas LAM1 precipitated from neutrophils was 95 KD (Tedder et al., 1990). Selectins have an N-terminal, Ca + +-dependent, lectin-like domain, an epidermal growth factor-like domain and short consensus repeats similar to that observed for C3/C4 complement binding proteins (Springer, 1990). There are several similarities between LAM1 and Fd11F preparations in addition to molecular size. First, PMA stimulation of LAM1-expressing peripheral thymocytes results in rapid modulation of this marker from the cell surface as was observed for Fd11 cells using anti-p30. Further, lymphocytes and neutrophils have been demonstrated to actively shed LAM1 into the culture media upon activation. In light of these similarities, the percent p30+ T cells obtained by North et al., using murine, spleen- and thumus-derived T cells versus the virtual pan-reactivity of anti-p30 with human, peripheral T lymphocytes could be significant and relate to the presence of Leu 8 on the surface of cells in the periphery and it's absence on the surface of cells out of ciruclation. It would be interesting to test cells from human lymphoid organs and from murine peripheral blood with anti-p80 and anti-p30 as well as anti-Leu 8 or LAM1. Although specific antigen-binding and suppressive ability have obviously not been assayed for selectins, the possibility exists that Fd11F, and even suppression of the immune system in general somehow involves solublized adhesion molecules which bind to target receptors and block cell to cell interactions. Antigen-specificity potentially could be obtained as Green believes, by utilizing alternately spliced TCR transcripts. Idiotype-bearing molecules may even associate with adhesion molecules intracellular^ and be secreted upon activation by antigen. Whatever the function of the molecule(s) bearing the p80 and p30 determinants is in humans, it must relate to it's near ubiquitous expression on peripheral white blood cells and it's origin from a preparation enriched by binding to a TsF-framework-specific monoclonal antibody. Finally, it should be noted that the possibility exists that the human cell surface and cell lysate binding of these antisera represent interspecies cross-reactivity with related or even unreleated (human tissues other than those from the hematopoietic system were not tested in this work) cell surface protein(s). 85 SECTION 2: CONCLUSIONS Antisera raised in rabbits against two distinct components of an affinity enriched antigen-specific, murine TsF were used to screen subsets of human peripheral white blood cells. Anti-p80 and anti-p30 were both shown to bind to the surface of two distinct populations of mononuclear cells on the basis of different levels of binding. With anti-p80, on average 65% of the preparation showed lower levels of binding and 15% gave higher levels. With anti-p30, approximately 50% of the MN population showed weak binding and 6% gave strong binding. An experiment which isolated the T lymphocyte population from nonspecific binding to NRS and non T cells was able to clearly show that this subset of PBMN cells expresses both the p80 and p30 cell surface determinants (although the level of binding was slightly lower with the latter). These antisera were also shown to bind in a similar manner (same increase in binding over background) to peripheral B lymphocytes as to peripheral T lymphocytes. Finally, it was shown that human neutrophils express the p80 and p30 determinants at equal levels, even though the goat anti-rabbit Ig-FITC displayed significant binding to human Abs (bound to FcR's) on the granulocyte cell surface. In an attempt to characterize the determinant(s) to which these antisera were binding in humans, several human, leukemic cell lines were obtained and screened for reactivity. MOLT-4, CEM-CM3, HuT 78 and RAJI all bound anti-p80 to varying extents. All of these lines expressed high levels of p30 at a dilution of 1 in 20 except HuT 78, which was effectively p30_. With Hut 78 as a negative control, a Western blot analysis of cell lysates was undertaken with anti-p30 and these four cell lines. Anti-p30 was found to bind to a major band of 42 KD, to minor bands at 60 and 80 KD and to high molecular weight material in all cell lysates. In addition a band of 25 KD was observed in CEM-CM3 and RAJI lysates only. If the minor bands are ignored, the cell suface and Western blot data suggest that HuT 78 cells synthesize but do not express the p30-expressing 42 KD protein. This could be due to an intracellular processing or transport defect or a mutation resulting in the lack of association with a component of a multimeric complex. However, since HuT 78 cells are p30', and the 25 KD band is close in size to 30 KD and absent in HuT 78 lysates, the 42 KD band may represent strong interspecies cross-reactivity with Abs contained within the anti-p30 serum, and the 25 KD band may represent specific binding to a suppression-related protein. 86 Finally, given what seems to be ubiquitous expression of the p80 and p30 markers on peripheral leucocytes, it is interesting that very recently, a molecule (Leu 8) of 80 KD, that is expressed on most peripheral white blood cells, and known to be associated with the suppressor-inducer subset of CD4 + T cells, has recently been identified as LAM1 a new member of the 'selectin' family of adhesion molecules of the immune system. 87 CLOSING DISCUSSION The work presented in this thesis has involved different aspects of two different modes of immunosuppression. Section 1 dealt with the further characterization and partial purification of an antigen-nonspecific suppressor inducer factor (SIF) secreted by an LGL, third population cell line. In addition, the mode of action of SIF was investigated and discussed for different assay systems. This work was discontinued shortly after the discovery was made that the M1-A5 cell line was infected with M y c o p l a s m a and that the suppressive phenomena could not be separated from the presence of the organism. Section 2 utilized antisera raised to components of a murine, antigen-specific suppressor factor secreted by a Ts cell hybridoma to investigate their reactivity with normal, human, peripheral leucocytes and several human, lymphoid leukemic cell lines. In addition, an attempt was made to characterize the protein expressing the p80 and p30 epitopes using cell line lysates and Western blot analysis. Although many difficulties were experienced in obtaining the results presented in this work, similar problems have been encountered by many other workers in the field of immunosuppression. Most of the assays required to characterize bioactivity of suppressor factors are variable and dependent on many different elements acting in concert and often in vitro. In addition, suppressor cells are extremely difficult to clone and purified suppressor factors seem to be an unbelievably diverse group of molecules to mediate the fine level of control that has been shown to exist in the immune system. Putting the technical difficulties of this field of research aside, it is still curious that with the current advanced state of technology, so little about the basic mechanism of suppression of the immune system is actually understood. There seems to be no other way to increase our knowledge in this area but to keep 'following leads'. I believe that once the general mechanism of immune down-regulation has been clearly laid out, important and exciting advances in this field will not be far behind. LITERATURE CITED Almawi, W. Y. and Pope, B. L , 1985. Induction of suppression by a murine nonspecific suppressor-inducer cell line (M1-A5). Cellular Immunology 96:199. Almawi, W. Y. and Pope, B. L , 1986. Induction of suppression by a murine nonspecific suppressor-inducer cell line (M1-A5). II. The role of prostaglandins. J. Immunology 136:1982. Almawi, W. Y., Murphy, M. G. , Ogbaghhebriel, A. and Pope, B. L , 1987a. Cyclic AMP as the second messenger for prostaglandin E in modulating suppressor cell-activation by natural suppressor/cytotoxic cells. Int. J. Immunopharmacology 9:697. Almawi, W. Y., Dolphin, P. J . and Pope, B. L , 1987b. Induction of suppression by a murine nonspecif ic suppressor-inducer cell line (M1-A5). III. Partial purification of the suppressor cell-inducing factors. J. Mol. Cell. Immunology 3 : 1 5 7 . Almawi, W., Y . and Pope, B. L , 1988: Enhancement of tumor growth by suppressor factors derived from a natural suppressor/cytotoxic cell line. Transplantation 4 5 : 9 9 8 . Almawi, W. Y., Dolphin, P. J . , Thies, R, L , McMaster, R. W., Levy, J . G. and Pope, B. L., 1988b. Suppressor cell-inducing factor: A new lymphokine secreted by a natural suppressor cell line with natural cytotoxic activity. I. Purification to apparent homogeneity and initial characterization of suppressor-cell inducing factor. J. Immunology 141:2529. Altman, A., Coggeshall, K. M. and Mustelin, T., 1990. Molecular events mediating T cell activation. Advances in Immunology 48:227. Aune, T. M. and Pierce, C. W., 1984. Mechanism of SIRS action at the cellular and biochemical level. Lymphokines 9:257. Bains, M. A., McGarry, R. C. and Singhal, S . K., 1982. Regulatory cells in human bone marrow: Suppression of an in vitro primary antibody response. Cellular Immunology 74:150. 89 Behlke, M. A. and Loh, D. Y., 1986. Alternative splicing of murine T cell receptor J3-chain transcripts. Nature 322:379. Brooks, C. G., Kuribayashi, K., Sale, G. E. and Henney, C. S., 1982. Characterization of five cloned murine cell lines showing high cytolytic activity against YAC-1 cells. J. Immunology 128:2326. Cantor, H. and Gershon, R. K., 1979. Immunological Circuits: Cellular composition. Fed. Proc. 38:2058. Chan, A., Tench Stammers, A., North, J . , Steele, J . K., Chu, N. R. and Levy, J . G., 1988. Characterization of antigen-binding molecules from T suppressor hybridomas. Intern. Rev. Immunology 3:175. Clark, P., Normansell, D. E., Innes, D. J . and Hess, C. E., 1986. Lymphocyte subsets in normal bone marrow. Blood, 67:1600. Chu, N. R. Characterization of a T lymphocyte-derived, antigen-binding molecule with s u p p r e s s i v e activity. PhD. Diss., University of British Columbia, 1989. Cunningham, A. and Szenberg, A., 1968. Further improvements in the plaque technique for detecting single antibody forming cells. Immunology 14:599. Damle, N. K., Mohagheghpour, N. and Engleman, E. G., 1984. Soluble antigen primed inducer T cells activate antigen-specific suppressor T cells in the absence of antigen pulsed accessory cells: Phenotypic definition of suppressor inducer and suppressor effector cells. J. Immunology 132:644. Damle N. K., Mohagheghpour, N., Kansas, G. S., Fishwild, D. M. and Engleman, E. G., 1985. Immunoregulatory T cell circuits in man: Identification of a distinct subpopulation of the helper/inducer lineage that amplifies the development of alloantigen-specific suppressor T cells. J. Immunology 134:125. Damle, N. K., 1986. Suppressor T lumphocytes in man: Facts and views. Year in Immunology 2:60. 90 Damle, N. K., Childs, A. L. and Doyle, L. V., 1987. Immunoregulatory T lymphocytes in man: Soluble antigen-specific suppressor-inducer T lymphocytes are derived from the CD4+CD45Prp80+ subpopulation. J. Immunology 139:1501. Dialynas, D. P., Quan Z. S., Wall K. A., Pierres A., Quintans, J. , Loken M. R., Pierres, M.and Fitch, M. W., 1983. Characterization of the murine T cell surface molecule, designatedl_3T4, identified by monoclonal antibody GK1.5: Similarity of L3T4 to the human Leu-3/T4 molecule. J. Immunology 131:2445. Duwe, A. K. and Singhal, S. K., 1979. The immunoregulatory role of bone marrow. I. Suppression of the induction of the antibody response to T-dependent and T-indepen-dent antigens by cells in the bone marrow. Cellular Immunology 43:362. Engleman, E. G., Benike C. J., Grumet, F. C. and Evans R. L, 1981. Activation of human T lymphocyte subsets: Helper and suppressor/cytotoxic T cells recognize and respond to distinct histocompatability antigens. J. Immunology 127:2124. Fairchild, R. L, Kubo, R. T. and Moorhead, J.W., 1988. Soluble factors in tolerance and contact sensitivity to 2.4-dinitrofluorobenzene in mice. IX. A monoclonal T cell suppressor molecule is structurally and serologically related to the a/(3 T cell receptor. J. Immunolgy 141:3342. Fisher, M. S. and Kripke, M. L , 1982. Suppressor T lymphocytes control the development of primary skin cancers in ultraviolet light irradiated mice. Science 216:1133. Fresno, M., McVay-Boudreau, L. and Cantor, H., 1981. Antigen-specific T lymphocyte clones II. Purification and biological characterization of an antigen-specific suppressive protein synthesized by cloned T cells. J. Exp. Medicine 153:1260. Germain, R. N. and Benacerraff, B., 1981. A single major pathway of T lymphocyte interactions in antigen-specific immune suppression. Scand. J. Immunology 13:1. Granstein, R. D., Parrish, J. A., McAuliffe, C , Waltenbaugh, C. and Greene, M. I., 1984. Immunologic inhibition of ultraviolet radiation-induced tumor suppressor cell activity. Science 224:615. Graves, M. C , Mullenbach, G. T. and Rabinowitz, J. C , 1985. Cloning and nucleotide sequence determination of the Clostridium pasteurianum ferredoxin gene. Proc. Natl.Acad. Sci., USA 82:1653. 91 Green, D. R., Flood, P. M. and Gershon, R. K., 1983. Immunoregulatory T cell pathways. Ann. Rev. Immunology 1:439. Green, D. R. and Zheng, H., 1989. Antigen-specific regulatory T-cell factors and the T cell receptor, in the 26th Forum in Immunology. Res. in Immunolgy 140:294. Hausman, P. B., Sherr, K. H. and Dorf, M. D., 1986. Regulation of hapten-specific T and B cell respnses by suppressor cells. Concepts in Immunopathol. 3:38. Hertel-Wulff, B., Okada, S., Oseroff, A. and Strober, S., 1984. In vitro propagation and cloning of murine natrual suppressor (NS) cells. J. Immunology 133:2791. Hertel-Wulff, B., Lindsten, T., Davis, M. M. and Strober, S., 1987. Rearrangement and expression of T cell receptor genes in cloned natural suppressor cell lines. J. Exp. Medicine 166:1168. Holda, J. H., Maier, T. and Claman, H. N., 1986. Natural suppressor activity in graft-vs.-host spleen and normal bone marrow is augmented by IL-2 and interferon-y. J. Immunology 137:3538. Hurme, M. and Sihvola, M., 1985. In vivo activation of autoreactive cytotoxic T lymphocytes in bone marrow after transplantation. Evidence for prethymic specificity of T cells? J. Immunology 135 :1108. Janeway, C. A., Tada, T., Pereira, P., Larsson-Sciard, E. L, Coutinho, A. and Bandeira, A., 1988. Editorial Discussions (3) in reply to G. Moller, 1988, Scand. J. Immunology 27:621. Jensen, L. H., 1974. X-ray structural studies of ferredoxin and related electron carriers. Ann. Rev. Biochem. 43:461. Kansas, G. S., Wood, G. S., Fishwild, D. M. and Engleman, E. G., 1985. Functional characterization of human T lymphocyte subsets distinguished by monoclonal anti-Leu8. J. Immunology 134:2995. Kapp, J. A., Pierce, C.W., De La Croix, F. and Benacerraf, B., 1976. Immunosupressive factor(s) extracted from lymphoid cells of nonresponder mice primed with L-glutamic acid 6 0 -L-alanine 3 0 -L-tyrosine 1 0 (GAT). I. Activity and antigenic specificity. J. Immunology 116:305. Kapp, J . A., Pierce, C. W., Schlossman, S. and Benacerraf, B., 1978. Genetic control of immune responses in vitro. V. Stimulation of suppressor T cells in nonresponder mice by the terpolymer L-glutamic a c i d 6 0 - L - a l a n i n e 3 0 - L - t y r o s i n e 1 0 (GAT). J. Exp. Medicine 140:649. Kapp, J . A., Araneo, B. A. and Clevinger, B. L , 1980. Suppression of antibody and T cell proliferative response to L-glutamic a c i d 6 0 - L - a l a n i n e 3 0 - L - t y r o s i n e 1 0 by a specific monoclonal T cell factor. J. Exp. Medicine 160:1012. Kitamara, K., Nakauchi, H., Koyasu, S. , Yahara, I., Olumura, K. and Tada.T., 1984. Characterization of an antigen-specific suppressive factor derived from a cloned suppressor effector T cell line. J. Immunology 133:1371. Kronenberg, M., Steinmetz, M., Kobori, J . , Kraig, E., Kapp, J . A., Pierce, C. W., Sorenson, C. M., Suzuki, G. , Tada, T. and Hood, L., 1983. RNA transcripts for l-J polypeptides are apparently not encoded between the l-A and l-E subregions of the murine major histocompatibility complex. Proc. Natl. Acad. Sci., USA 80:5704. Krupen, K., Araneo, B. A., Brink, L , Kapp, J . A., Stein, S., Weider, K. J . and Webb, D. R., 1982. Purification and characterization of a monoconal T cell suppressor factor specific for p o l y ( L - G l u 6 0 - L - A l a 3 0 - L - T y r 1 °). Proc. Natl. Acad. Sci., USA 79:1 2 5 4 . Kuchroo, V. K., Steele, J . K., Billings, P. R., Selvaraj, P. and Dorf, M. E., 1988. Expression of CD3-associated antigen-binding receptors on suppressor T cells. Proc. Natl. Acad. Sci., USA 85:9209. Kuchroo, V. J . , Steele, J . K., O'Hara, R. M., Jayaraman, S., Selvaraj, P., Greenfield, E., Kubo, R. T. and Dorf, M. E., 1990. Relationships between antigen-specific helper and inducer suppressor T cell hybridomas. J. Immunology 145:438. Kupfer, A. and Singer, S. J . , 1989. Cell biology of cytotoxic and helper T cell functions Microscopic studies of single cells and cell couples. Annual Rev. Immunology 7 :309. Lee, J . C , Hapel, A. J . and Ihle, J . N., 1982. Constitutive productin of a unique lymphokine (IL-3) by the WEHI-3 cell line. J. Immunology 128:2393. Lowy, I., Brezin, C , Neauport-Sautes, J . T. and Fridman, W. H., 1983 Proc. Natl. Acad. Sci., USA 80:2323. Maier, T., Tench Stammers, A. and Levy, J . G., 1983. Characterization of a monclonal antibody directed to a T cell suppressor factor. J. lmmunolgy\\2'\\:'\\ 843. Maier, T., Holda, J . H. and Claman, H. N., 1985. Graft-vs.-host reactions (GVHR) across minor murine histocompatiability barriers. II. Development of natural suppressor cell activity. J. Immunology 135:1644. Maier, T., Holda, J. H. and Claman, H. N.: Natural suppressor (NS) cells: Members of the LGL regulatory family. Immunology Today 7:312, 1986. Maier, T., Holda, J. H. and Claman, H. N.: Natural Suppressor Cells. Immunity to Cancer II. Alan R. Liss Inc., New York:1989. Maier, T., Holda, J. H., Lee Choi, K. and Claman, H. N.: The characterization and functions of the natural suppressor cell systems. In Reynolds, C. W., Wiltrout, R. H (eds): Functions of the Natural Immune System. Plenum Publishing Corp., New York:1990. Malkovsky, M., Asherson, G. L., Stockinger, B. and Watkins, M. C , 1982. Nonspecific inhibitor released by T acceptor cells reduces the production of interleukin-2. Nature 300:652. Mills, C. D. and North, R. J. , 1985. Ly1+ suppressor cells inhibit the expression of passively transferred anti-tumor immunity by suppressing the generation of cytolytic T cells. Transplantation 39:202 Mohagheghpour, N., Benike, C. J., Kansas, G. S., Bieber, C. and Engleman, E. G., 1983. Activation of antigen-specific suppressor T cells in the presence of cyclosporin requires interactions between T cells of inducer and suppressor lineage. J. Clin. Invest. 72:2092. Moller, G., 1988. Do suppressor T cells exist? Scand. J. Immunology 27:247. Mortari, F., Bains, M. A. and Singhal, S. K., 1986. Immunoregulatory action of human bone marrow. Identification of suppressor cells possessing OKM1, SSEA-1 and HNK-1 antigens. J. Immunology 137:1133. 94 Murphy, D. B., Herzenberg, L A., Okumura, K., Herzenberg, L. A. and McDevitt, H. O., 1976. A new I subregion (l-J) marked by a locus (la-4) controlling surface determinants on suppressor T lymphocytes. J. Exp. Medicine 144:699. Nakauchi, H., Ohno, I., Kim, M., Okumura, K. and Tada, T., 1984. Establishment and functional analysis of a cloned, antigen-specific suppressor effector T cell line. J. Immunology 132:88. North, J., Weaver, M. S., Stammers, A. T., Chu, N. R., Chan, A. and Levy, J. G., 1988. Identification of a T cell subset using a rabbit antibody raised against a T suppressor molecule. J. Immunology 141:2268. Okada, S. and Strober, S., 1982. Spleen cells from adult mice given total lymphoid irradiation (TLI) or from newborn mice have similar regulatory effects in mixed lymphocytereaction (MLR). II. Generation of antigen-specific suppressor cells in the MLR after the addition of speen cells from newborn mice. J. Immunology 129:1 892. Pope, B. L., Whitney, R. B., Levy, J. G. and Kilbum, D. G., 1976. Suppressor cells in the spleens of tumor-bearing mice: Enrichment by centrifugation on hypaque-ficoll and characterization of the suppressor population. J. Immunology 116:1342. Pope, B. L., Whitney, R. B. and Levy, J . G., 1978. Two distinct populations of suppressor cells in the spleens of mice bearing methylcholanthrene-induced tumors. J. Immunology 120:2033. Pope, B. L., 1985. Activation of suppresor T cells by low-molecular weight factors secreted by spleen cells from tumor-bearing mice. Cellular Immunology 93:364. Pope, B. L., Rowden G. and Djeu, J. Y., 1986. Suppressor of a suprssor cell inducing factor by an interleukin-3 dependent cell line with natural cytotoxic activity. Cellular Immunology 100:97. Pope, B. L., 1987. Secretion of a suppressor cell-inducing factor by an interleukin-3 dependent cell line with natural cytotoxic activity. II. Range, potency and kinetics of suppressive activity. Immunobiology 174:107. Reinherz, E. L. and Schlossman, S. F., 1980. The differentiation and function of human T lymphocytes. Cell 19:821. 95 Rich, R. R.p ElMasry, M. N. and Fox, E. J., 1986. Human suppressor T cells: Induction, differentiation and regulatory functions. Human Immunology 17:369. Saffran, D. C. and Singhal, S., 1990. Further characterization of murine bone marrow-derived natural suppressor cells. Potential relationships between NS and NK/LAK activities. Cellular Immunology 128:301. Saito, T. and Taniguchi, M., 1984. Chemical features of an antigen-specific suppressor T cell factor composed of two polypeptide chains. J. Mol. Cell. Immunology 1:137. Salter, R. D., Benjamin, R. J. , Wesley, P. K., Buxton, S. E., Garrett, T. P., Clayberger, C , Krensky, A. M., Norment, A. M., Liftman, D. R. and Parham, P., 1990. A binding site for the T cell coreceptor CD8 in the a 3 domain of HLA-A2. Nature 345:41. Sanders, M. E., Makgoba, M. W. and Shaw, S. W., 1988. Human naive and memory T cells: Reinterpretation of helper-inducer and suppressor-inducer subsets. Immunology Today 9:195. Schatten, S., Drebin, J. A., Perry, L. L., Chung, W. and Green, M. I., 1984. Regulation of the immune response to tumor antigens. X. Activation of third order suppressor T cells that abrogate anti-tumor immune responses. J. Immunology 133:1064. Schwadron, R. B., Palathumpat, V. and Strober, S., 1989. Natural suppressor cells derived from adult spleen and thymus. Transplantation 48:107. Singhai, R., Weaver, M., Sikora, L. and Levy, J . G., 1984. Evidence for the presence of idiotype-bearing regulatory T cells in which idiotype expression does not show linkage to either Igh alleles or the MHC. Immunology 51:743. Singhal, S. K., King, S. and Drury, T., 1972. Anitbody-inhibiting acitivty of bone marrow cells in vitro. Int. Arch. Allergy 43:934. Springer, T. A., 1990. Adhesion receptors of the immune system. Nature 346:425. Steele, J. K., Tench Stammers, A. and Levy, J. G., 1985. Preliminary characterization of human T cell suppressor factor (HTsF) isolated from tonsil cells by monoclonal antibody immunoabsorbance. J. Immunology 135:1201. 96 Steele, J . K., Kawasaki, H., Kuchroo, V. K., Minami, M., Levy, J . and Dorf, M. E., 1987a. A monoclonal antibody raised to tumor-specific T cell derived suppressor factors also recognizes T suppressor inducer factors of the 4-hydroxy-3-nitrophenyl acetyl hapten suppressor network. J. Immunology 139:2629. Steele, J . K., Chu, N. R., Chan, A., North, J . and Levy, J . G., 1987b. Isolation of an antigen-specific T suppressor factor that suppresses the in vivo immune response of DBA/2 mice to ferredoxin. J. Immunology 139:469. Sugiura, K., Inaba, M., Ogata, H., Yasumizy, R., Inaba, K., Good, R. A. and Ikehara, S., 1988. Wheat germ agglutinin-positive cells in a stem cell-enriched fraction of mouse bone marrow have potent natural suppressor activity. Proc. Nat. Ac. Sci. 85 :4282. Sykes, M., Eisenthal, A. and Sachs, D. H., 1988. Mechanism of protection from graft versus host disease in murine mixed allogeneic chimeras. J. Immunology 1 4 0 : 2 9 0 3 . Tada, T., Taniguchi, M. and David, C. S., 1976. Properties of the antigen-specific T cell factor in the regulation of the antibody response of the mouse. IV. Special subregion assignment of the gene(s) that codes for the suppressive T-cell factor in the H-2 histocompatability complex. J. Exp. Medicine 144:713. Tada, T.: Help, suppression and specific factors. In Fundamental Immunology, ed. W. E. Paul, Raven Press, New York:1984. Takei, F., Levy, J . G. and Kilburn, D. G., 1976. In vitro induction of cytotoxicity against syngeneic mastocytoma and its suppression by spleen and thymus cells of tumor-bearing mice. J. Immunology 116:288. Takei, F., Levy, J . G . and Kilburn, D. G. , 1978. Characterization of a soluble factor that specifically suppresssed the in vitro generation of cells cytotoxic for syngeneic tumor cells in mice. J. Immunolgy 120:1218. Takemori, T. and Tada, T., 1975. Properties of antigen-specific suppressive T-cell factor inthe regulation of antibody response of the mouse. I. In vivo activity and immunochemical characterizations. J. Exp. Medicine 142:1241. 97 Taniguchi, M., Saito, T. and Tada, T., 1979. Antigen-specific suppressive factor produced by a transplantable l-J bearing T-cell hybridoma. Nature 278:227. Taniguchi, M. and Tokuhisa, T., 1980. Cellular consequenes in the suppression of antibody response by the antigen-specific T cell factor. J. Exp. Medicine 124:1241. Tedder, T. F., Penta, A. C , Levine, H. B. and Freedman, A. S., 1990. Expression of the human leucocyte adhesion molecule, LAM1: Identity with the TQ1 and Leu-8 differentiation antigens. J. Immunology 144:532. Teillaud, J. L. and Fridman, W. H., 1989. Discussion of 26th Forum in Immunology. Research in Immunology 140:344. Turck, C. W., Kapp, J. A. and Webb, D. R., 1986. Structural analyses of a monoclonal heterodimeric suppressor factor specific for L-glutamic acid-L-alanine-L-tyrosine. J. Immunology 137:1904. Turner, J . M., Brodsky, M. H., Irving, B. A., Levin, S. D., Perlmutter, R. M. and Liftman, D. R., 1990. Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs. Cell 60:755. Weaver, M., Singhai, R., Sikora, L. and Levy, J. G., 1983. Identification of an idiotypic marker of a major regulatory T cell of the immune response in B10.BR mice to ferredoxin. J. Exp. Medicine 157:285. Webb, D. R., Devens, B. H., Kraig, E., Kapp, J. A., Pierce, C. W. and Sorensen, C. M., 1989. More speculations on specific suppression. Immunology Today 10:326. Weiner, D. B., Liu, J. , Hanna, B., Bluestone, J. A., Coligan, J. E., Williams, W. W. and Greene, M. I., 1988. CD3-associated heterodimeric polypeptides on suppressor hybridomas define biologically active inhibitory cells. Proc. Natl. Acad. Sci., USA. 85:6077. Yamauchi, K., Murphy, D., Cantor, H. and Gershon, R. K., 1981a. Analysis of antigen-specific, Ig-restricted cell-free material made by I-J+ Ly-1 cells (Ly-1 TsiF) that induces ly-2+ cells to express suppressive activity. Eur. J. Immunology 11:905. Yamauchi, K., Murphy, D., Cantor, H. and Gershon, R. K., 1981b. Analysis of an antigen-specific, H-2 restricted cell-free product(s) made by \"I-J-\" Ly 2 cells (Ly-2 TsF) that suppresses Ly-2 cell-depleted spleen cell activity. Eur. J. ImmunologyM-.WS. Zheng, H., Boyer, M., Fotedar, A., Singh, B. and Green, D. R., 1988. An antigen-specific helper T cell hybridoma produces an antigen-specific suppressor inducer molecule with identical antigenic fine specificity. J . Immunology 140:1351. "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0302160"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Microbiology and Immunology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Partial characterization and purification of a murine suppressor-inducer factor secreted by a natural supressor cell line and characterizaton of the binding of antisera raised to murine antigen-specfic suppressive material in human peripheral leukocytes"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/29691"@en .