Acceleration of Murine Systemic Lupus Erythematosus (SLE) by Exposure to the Organochlorine Pesticide Chlordecone

Material Information

Acceleration of Murine Systemic Lupus Erythematosus (SLE) by Exposure to the Organochlorine Pesticide Chlordecone
Copyright Date:


Subjects / Keywords:
Apoptosis ( jstor )
Autoimmunity ( jstor )
B lymphocytes ( jstor )
Cells ( jstor )
Cytokines ( jstor )
Estrogens ( jstor )
Lymphocytes ( jstor )
Lymphoid tissue ( jstor )
Macrophages ( jstor )
Secretion ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright the author. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:


This item is only available as the following downloads:

Full Text




Copyright 2005 by Fei Wang


To my mother Xiaoxian Wa ng, and my father Rongde Wang.


iv ACKNOWLEDGMENTS I would like to give my deepest gratitude to my mentors, Dr. Stephen M. Roberts and Dr. Eric S. Sobel, for all the time and effort they have invested on me. I am very grateful to them for having faith in me and accepting me to be their graduate student, and I also cherish the years under their guidance. Their keen insights in the fields of toxicology and rheumatology have made my research much easier, and their consideration and trust have made me, a foreign graduate student, feel at home. Dr. Roberts and Dr. Sobel have been both mentor s and friends for me. I also want to acknowledge my committee members Dr. Kath leen T. Shiverick, Dr. Dietmar W. Siemann, and Dr. Nancy D. Denslow. They ha ve assisted me to gr ow gradually by their great suggestions and comments over the last several years. I also want to thank Dr. Laurence Morel for many enthusiastic sugge stions. Dr. Christopher M. West, my previous menter at biochemistry, provided me wonderful trainings in biochemistry and molecular biology. My special thanks also go to my colleagues in the laboratory, both past and present. In particular, I want to thank Edward J. Butfiloski, the lab manager and Research Associate in Dr. Sobel’s lab, w ho taught me most of the f undamentals when I started, and someone I can always count on when I face an unexpected technical problem. John W. Munson, the research manager in Dr. Roberts’ lab, has always been ready to help me whenever I needed.


v I want to thank my parents from the botto m of my heart. Without their love and support, I would never have reached this far.


vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT.....................................................................................................................xi ii CHAPTER 1 INTRODUCTION........................................................................................................1 Pathogenesis of Lupus..................................................................................................1 Breaking Tolerance...............................................................................................2 Apoptosis and Its Role in Autoimmunity..............................................................3 Cytokines in the Induction of Autoimmunity........................................................4 The Role of Estrogen in Lupus.....................................................................................5 Organochlorine Pesticide Chlordecone........................................................................9 Health Hazard Information of Chlordecone...............................................................10 Neurotoxicity.......................................................................................................10 Carcinogenesis of Chlordecone...........................................................................11 Reproductive Toxicity.........................................................................................12 Immunotoxicity...................................................................................................12 Estrogenicity of Chlordecone.....................................................................................14 2 MATERIALS and METHODS..................................................................................25 Experimental Animals................................................................................................25 Test Materials and Treatments....................................................................................25 Flow Cytometry..........................................................................................................26 Apoptosis Assay.........................................................................................................26 Proliferation Assay.....................................................................................................27 CD4 T Cell Cytokine Determination..........................................................................28 cDNA Preparaion........................................................................................................28 Real-time PCR............................................................................................................29 ELISA for Autoantibody Titers..................................................................................29 Determination of Serum Prolactin..............................................................................30 Producing and Labeling Apoptotic Cells....................................................................31


vii Clearance of Apoptotic Cells......................................................................................31 Statistical Analysis . .....................................................................................................32 3 COMPARISON OF SPLENIC LYMP HOCYTE PHENOTYPIC CHANGES WITH TREATMENT OF CHLO RDECONE AND ESTROGEN............................34 Introduction.................................................................................................................34 Germinal Centers.................................................................................................35 Marginal Zone B cells.........................................................................................37 Results........................................................................................................................ .39 Chlordecone Exposure Caused Only Sli ght Changes in Spleen, Uterus and Body Weight Compared to Estradiol...............................................................39 Estradiol, but Not Chlordecone, Ch anged Lymphocyte Percentages..................40 Both Chlordecone and Estradiol Trea tments Activated Splenic B Cells............40 Chlordecone Exposure Enhanced Germinal Center Reactions...........................41 Examination of Expression Profile of Genes Involved in Germinal Center Reactions..........................................................................................................42 Exposure to Chlordecone and Estradio l Decreased the Rates of Germinal Center Apoptosis Without Affecting the Proliferation of Lymphocytes.........43 In Contrast to Estradiol, Chlordec one Treatment Did Not Alter Splenic B Cell Subsets......................................................................................................43 Estradiol, But not Chlordecone, Increased the Mature Plasma Cell Population.44 Estradiol, but Not Chlordecone, Enhanced Anti-dsDNA and Anti-ssDNA Autoantibody Titers in Serum..........................................................................45 Estradiol, but Not Chlordecone , Activated Splenic T Cells................................45 Estradiol, but Not Chlordecone, Increa sed the Regulatory T Cell Population....46 Estradiol Treatment Enhanced the Expression of T Cell Receptor V , but Not V Chains........................................................................................................46 Chlordecone and Estradiol Treatments Enhanced Bcl-2 Expression on CD4 T Cells.................................................................................................................47 Chlordecone and Estradiol Treatmen ts Reduced CD4 T Cell Apoptosis Without Affecting Proliferation.......................................................................47 Chlordecone Increased the Percentage of Macrophage Population, but Not Dendritic Cell Population in Spleen................................................................47 Discussion...................................................................................................................48 4 COMPARISION OF SPLENIC B C ELL GENE EXPRESSION AND CD4 T CELL CYTOKINE SECRETION BY CHLORDECONE AND ESTROGEN TREATMENT............................................................................................................82 Introduction.................................................................................................................82 Selected Genes.....................................................................................................83 Cytokines.............................................................................................................86 Results........................................................................................................................ .91 Negative Selection by Magnetic Bead s Resulted in Highly Purified Populations of B and CD4 T Cells...................................................................91 Chlordecone Enhanced Bcl-2, Shp-1, FAS Expression in B Cells.....................92


viii Estradiol Decreases IFNExpression in B Cells...............................................93 Neither Chlordecone Nor Estradio l Changed the Expression of Fc RIIb, TNF, TGF, Ly5, IL-6 and IL-2.................................................................93 CD3 and/or CD28 Stimulation Significan tly Increased Cytokine Secretion......93 Normalizing the Cytokine Levels........................................................................95 Both Chlordecone and Estradiol Signifi cantly Increased Secretion of the Proinflammatory Cytokines TNFand IL-2........................................................95 Chlordecone, but Not Estradiol, Increase d the Secretion of Pro-inflammatory Cytokines IFNand GM-CSF........................................................................96 Estradiol, but Not Chlordecone , Increased IL-10 Secretion................................96 Estradiol, but Not Chlordec one, Caused a Significant Decrease in IL-4 Level..96 Discussion...................................................................................................................97 5 COMPARISON OF CHLORDECON E AND ESTROGEN EFFECTS ON MACROPHAGE FUNCTIONS AND PEPTIDE HORMONE PROLACTIN SECRETION............................................................................................................117 Introduction...............................................................................................................117 Macrophage and Autoimmunity........................................................................117 Toxic Effect on Macrophage.............................................................................118 Prolactin and Autoimmunity.............................................................................118 Results.......................................................................................................................1 20 Both Chlordecone and Estradiol Reduced Peritoneal Macrophage Clearance of Apoptotic Cells..........................................................................................120 Chlordecone Treatment did Not A ffect the Proliferation of RAW 267.4 Cells...........................................................................................121 Both Clordecone and Estradiol E nhanced IL-10 Secretion in RAW 267.4 Cells...............................................................................................................122 Chlordecone Exposure Decreased Prolactin Secretion in Marked Contrast to the Effects of Estradiol...................................................................................122 Estradiol, but Not Chlordecone, Significantly Enhanced Prolactin Receptor Gene Expression in B Cells and CD4 T Cells...............................................123 Discussion.................................................................................................................124 6 GENERAL DISCUSSION AND PERSPECTIVE..................................................135 LIST OF REFERENCES.................................................................................................148 BIOGRAPHICAL SKETCH...........................................................................................174


ix LIST OF TABLES Table page 2-1 List of primers used in real-time PCR......................................................................33 3-1 Introduction of phenotypic markers.........................................................................60 6-1 Summary of the results in the studies.....................................................................145


x LIST OF FIGURES Figure page 1-1 Structures of 17 -estradiol and chlordecone............................................................19 1-2 Time to development of elevated au toantibodies in female NZB/NZW F1 mice treated with chlordecone..........................................................................................20 1-3 Enhanced renal disease and immune complex deposition in ovariectomized, chlordecone-treated mice afte r 8 weeks of treatment...............................................21 1-4 Anti-chromatin autoanti body titers in female BA LB/c mice treated with chlordecone..............................................................................................................22 1-5 Uterine hypertrophy in ovariec tomized NZB/NZW F1 mice..................................23 1-6 Potential pathogenesis of lupus development..........................................................24 3-1 Transverse section of spleen white pulp..................................................................61 3-2 Chlordecone and estradiol effects on mice body, spleen and uterus weight.............62 3-3 Chlordecone and estradiol effect s on splenic lymphocyte populations...................63 3-4 Analysis of chlordecone and estrad iol effects on B cell activation markers CD44, CD69, co-stimulation marker B7.2, and class II molecule I-A(d) expression in splenic B cells....................................................................................64 3-5 Enlarged germinal center in both ch lordecone and estradiol-treated mice..............65 3-6 Analysis of chlordecone and estradio l effects on the expression of chemokine receptors CXCR4 and CXCR5 in total B ce lls and germinal center B cells from the spleen..................................................................................................................66 3-7 Analysis of chlordecone and estradio l effects on the expression of chemokine receptors CXCR5 in CD4 T cells from the spleen...................................................67 3-8 Analysis of chlordecone and estradiol effects on the expression of Bcl-2 in B cells.......................................................................................................................... .68


xi 3-9 Analysis of chlordecone and estradio l effects on the expression of the cell adhesion molecules ICAM-1 and VCAM-1 in the total B cells and the germinal center B cells............................................................................................................69 3-10 Analysis of chlordecone and estradiol e ffects on the expression of the inhibitory receptor Fc RIIb in the germinal center B cells.......................................................70 3-11 Analysis of B cell apoptosis and prol iferation from chlordecone and estradioltreated mice..............................................................................................................71 3-12 Analysis of chlordecone and es tradiol effects on B cell subsets..............................72 3-13 Analysis of chlordecone and estradio l effects on the percentage of mature plasma cells in spleen...............................................................................................73 3-14 Analysis of chlordecone and estradio l effects on the serum titers of anti-dsDNA and anti-ssDNA antibodies.......................................................................................74 3-15 Analysis of chlordecone and estradiol effects on the expression of the activation marker CD69 on CD4 and CD8 T cells...................................................................75 3-16 Analysis of chlordecone and estrad iol effects on the CD4 T cell subsets................76 3-17 Analysis of chlordecone and estrad iol effects in the regulatory CD4 T cell population.................................................................................................................77 3-18 Analysis of chlordecone and estradiol effects on the T cell receptor revision.........78 3-19 Analysis of chlordecone and estradiol effects on the expression of Bcl-2 in CD4 T cells.......................................................................................................................7 9 3-20 Analysis of chlordecone and estrad iol effects on CD4 T cell apoptosis and proliferation..............................................................................................................80 3-21 Analysis of chlordecone and estradio l effects on macrophage and dendritic cell populations...............................................................................................................81 4-1 Increased gene expression of Shp1 and Bcl-2 by chlordecone treatment in purified splenic B cells...........................................................................................108 4-2 Analysis of gene expression of FAS and FAS ligand by chlordecone treatment in purified splenic B cells...........................................................................................109 4-3 Comparison of chlordecone and estr adiol effects on the expression of IFNon splenic B cells.........................................................................................................110 4-4 Comparison of chlordecone and estr adiol effects on the expression of Fc RIIb, TNF, TGFand Ly5 in splenic B cells.............................................................111


xii 4-5 Comparison of chlordecone and estradio l effects on the expression of IL-2 and IL-6 in splenic B cells............................................................................................112 4-6 Analysis of the secretion of the pro-inflammatory cytokine TNFand IL-2 by CD4 T cells.............................................................................................................113 4-7 Comparison of chlordecone and estradio l effects on the secretion of cytokines IFNand GM-CSF by the CD4 T cells................................................................114 4-8 Comparison of chlordecone and estradiol effects on the secretion of cytokine IL10 by the CD4 T cells.............................................................................................115 4-9 Comparison of chlordecone and estradiol effects on the secretion of cytokine IL4 by cultured CD4 T cells......................................................................................116 5-1 Chlordecone and estradiol treatments im paired the clearance of apoptotic cells by peritoneal macrophages but not B1 cells...........................................................130 5-2 Chlordecone or estradio l treatment did not affect the proliferation of RAW 267.4 cell line...................................................................................................................131 5-3 Comparison of chlordecone and estradiol effects on the secretion of cytokine IL10 by the RAW 267.4 cell line...............................................................................132 5-4 Comparison of chlordecone and estrad iol effects on serum prolactin level in ovariectomized NZB/NZW F1 mice a nd unovariectomized BALB/c mice..........133 5-5 Comparison of chlordecone and estradio l effects on the gene expression level of prolactin receptor on sp lenic B and CD4 T cells....................................................134


xiii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ACCELERATION OF MURINE SYSTEMIC LUPUS ERYTHEMATOSUS (SLE) BY EXPOSURE TO THE ORGANOCHLO RINE PESTICIDE CHLORDECONE By Fei Wang December 2005 Chair: Stephen M. Roberts Major Department: Pharmacology and Therapeutics The weakly xenoestrogenic pesticid e chlordecone can accelerate lupus development in ovariectomized NZB/NZW F1 mice, an effect similar to that produced by estrogen. A series of experiments were conducted to determine whether chlordecone produces effects on autoimmunity by functioning as an estrogen mimic. Two-month-old, ovariectomized NZB/NZW F1 mice were implanted subcutaneously with 60-day sustained-release pellets cont aining 17-beta estradiol (0.05 mg/pellet) or chlordecone (1.0 or 5 mg/pellet) for six weeks. Mice implant ed with pellets containing matrix only served as controls. Lymphocyte phe notypic changes were studied by flow cytometry, gene expression changes on purified splenic B cel ls were measured by real-time PCR, and CD4 T cell cytokine secretion was tested by a multi-cytokine de tection system. Both chlordecone and estrogen activated B cells and enhanced germinal center (GC) reactions. Both chemicals increased Bcl-2 expression on splenic B and CD4 T cells, and reduced apoptosis w ithout affecting pro liferation. On the other hand, estradiol


xiv but not chlordecone, significantly increased mature plasma cells, changed the B cell subsets, activated T cells and changed CD4 T cell subsets. At the gene level, both chlordecone and estrogen increased B cell Shp-1, Bcl-2, and Fas expression, and IFNexpression was decreased. Neither chlordecone nor estradiol changed TNF, Fc RIIb, TGF, Ly5, Fas ligand, IL-2 and IL-6 expression in B cells. On the other hand, although both chlord econe and estradio l increased TNFand IL-2 secretion by CD4 T cells, only chlordecone tr eatment increased the secretion of IFNand GM-CSF. Estradiol treatment increased IL-10, while inhibiting IL-4 secretion. Both chlordecone and estradiol treatm ents reduced peritoneal macrophage clearance of apoptotic cells ex vivo , and increased IL-10 secretion in a concentrationdependent manner by RAW 264.7 cel ls. However, estrogen dramatically increased serum prolactin levels 10to 20-fold, as m easured by RIA, while chlordecone showed a marked dose-dependent decrease. Furt hermore, estrogen, but not chlordecone, significantly increased B and CD4 T cell prol actin receptor gene expression level. These findings suggest that while chlo rdecone and estradiol may share some pathways leading to enhanced survival of autoreactive lymphocytes, there are important differences, and chlordecone is not mediating it s effects indirectly th rough stimulation of prolactin-mediated signaling.


1 INTRODUCTION Systemic lupus erythematosus (SLE) is a chronic, systemic autoimmune disorder characterized by exacerbations and remissions of varying intensity and duration. The immune system that normally protects the body from bacteria, viruse s, and other foreign substances mistakenly directs an attack agai nst the body's own healthy tissues in SLE. Its clinical manifestations are almost inva riably accompanied by the presence of autoantibodies directed at a wide arra y of self-components including cell surface structures (surface protei ns and phospholipids on lymphocytes) and intracellular molecules (DNA, histone, and RNA) (von M uhlen et al., 1995; Mohan et al., 1993). Although the cause of lupus is still uncertain, current evidence suggests that a combination of factors plays a role in th e development of autoimmunity in SLE, including genetic, environmen tal, hormonal, and viral influences (D'Cruz, 2000; Roubinian et al., 1978). Pathogenesis of Lupus Genetic studies in mice indicate that pred isposition to lupus-like diseases can be divided broadly into three categories: (1 ) defective peripheral tolerance caused by alterations in the expression of genes regula ting cell survival and d eath, such as reduced expression of Fas/Fas ligand, Bim, PTEN, IL -2/IL-2R, and overexpression of Bcl-2; (2) defective clearance of dead and dying cel ls, and immune complexes; and (3) reduced threshold of lymphocyte activation, such as deficiencies of lyn, Shp-1, CD22, and FCR 2b (Kim et al,. 2003).


2 Breaking Tolerance In the autoimmune diseases characte rized by autoantibody-mediated pathology, breaking the tolerance of B cells is an im portant issue for unders tanding the underlying mechanisms of the disease (Ohashi et al., 2002). B cell toleran ce is established via several mechanisms. First, during lymphocyte developmen t, B cells reactive with selfantigens undergo clonal deletion that effectively removes autoreactive cells by apoptosis. This deletion occurs in the bone marrow at a pre-B to immature B-cell transitional stage (Nemazee et al., 1989; Hartley et al., 1991). Secondly, self-tolerance can also be achieved by functionally altering self-reactive cells instead of physically eliminating them. In B cells, a period of exposure to a relatively weak, co stimulator-deficient antigenic stimulus renders the B cell much more difficult to activate into proliferation and antibody secretion by a subseque nt immunogenic antigen challe nge, a state called anergy (Goodnow et al., 1988). Third, B-cell tolera nce can also occur in newly formed bone marrow cells through receptor editing, a form of receptor processing that markedly alters the Ig-variable (V)-region ge nes expressed by B cells and, consequently, changes the specificity of the surface Ig (Tiegs et al., 1993; Radic et al., 1993). Failures in B cell tolerance play an important ro le in the generation of high le vels of autoantibodies, as is seen in human SLE and repr esentative mouse models. Two important events are cri tical in breaking of lymphocyt e tolerance. One is the maturation state of the antigen presenting ce lls (APCs), and the other is the amount of self antigen that is detected by the immune system (Ohashi et al., 2002). If increased levels of self antigen expre ssion occur in the absence of signals that promote full APC maturation, then tolerance will occu r. If, however, self antigen is detected in the presence of pro-inflammatory signals or other events that promote APC maturation, then tolerance


3 will be broken and autoimmunity will arise. The maturation of APCs , such as dentritic cells and macrophages, has been reported to relate to signals thr ough Toll-like receptors and the Tyro 3 family of tyrosine kinases (Tyro 3, Axl and Mer) (Ohashi et al., 2002). Changes in APC signaling and maturation break tolerance and lead to autoimmunity. Moreover, genetic studies indicate a st rong and multigenic predisposition to the development of SLE. Several genetic loci from NZB/NZW F1 mice, such as sle1, sle2, and sle3, have been identified to confer susceptibility to break tolerance when transferred to the C57BL/6 background (Wakeland et al., 1999; More l et al., 1998). Apoptosis and Its Role in Autoimmunity Apoptosis, or programmed cell death, is a genetically controlled process with the characteristic features of cell crimple, chromatin condensation, DNA fragmentation, and apoptotic body formation (Alberts et al., 2002). It is initiated by tw o principal pathways the extrinsic pathway is activated by the li gation of death recepto rs, and the intrinsic pathway emerges from mitochondria. Apopt otic bodies are composed of nucleolus bodies and organelles (Alberts et al., 2002). Several studi es in SLE have revealed increased lymphocyte apoptosis as well as pe ripheral blood mononuclear cells (Emlen et al., 1994; Grondal et al., 2002; Lorenz et al., 1997). Apoptotic cells are usually removed by macr ophages in the early phase of apoptotic cell death (Savill et al., 1993). This process induces neithe r inflammation nor an immune response. In the immune system, especia lly at the thymus, the bone marrow, and the germinal centers of lymph nodes and splee n, specialized phagoc ytes can engulf the apoptotic cells so efficiently that few can be detected (Savill et al ., 2002). This instant removal of apoptotic cells offers the n ecessary space for the heavily proliferating thymocytes or centroblasts, and is a prereq uisite for a proper function of thymus, bone


4 marrow, and lymph nodes. It has been known for a long time that macrophages from patients with SLE have an impaired phagocytic activity for yeast and bacteria (Svensson, 1980; Hurst et al., 1984; Salmon et al., 1984). In addition, in vitro differentiated macrophages from a subgroup of SLE patients show a significantly reduced phagocytosis of apoptotic cells (Beyer et al., 2002). With reduced clearance, the dead cells accumulate, lose their membrane integrity, danger signals are released, and nuclear antigens become accessible in an inflammatory context. In times of increased apoptosis, tolerance can be broken, and a chronic inflam mation results, which then can lead to an autoimmune reaction against nuclear cons tituents. Many adaptor molecules and receptors are involved in the clearance of dying cells. Complement components, serum DNase I, phosphatidylserine, and modified gl ycoproteins participate crucially in the clearance of apoptotic and necrotic cells (Mevorach et al., 1998). Cytokines in the Induction of Autoimmunity Cytokines are essential mo lecules involved in the di fferentiation, maturation and activation of cells and thus, by nature , have a significant influence on the immunoinflammatory response. In autoimm une diseases, cytokines may not only be involved in the generation of the aberration of immune regul ation, but also in the local inflammatory processes that ultimately lead to tissue destruction. Cytokines can be produced by many cell types including lym phocytes, monocytes, macrophages, dendritic cells etc., but the predominant producers are helper T cells (Th) and macrophages (Janeway et al., 2005). CD4 T cells secrete much more cyto kines than CD8 T cells, and they can be further subdivided into Th1 and Th2 cells based on the patterns of cytokine release. Th1 cytokines, such as IL2 and IFN, activate T cells and macrophages and are critical for cell-mediated immunity and in flammation, whereas Th2 cytokines, such as


5 IL-4, IL-5, IL-6, and IL-10, promote an tibody production by B cells and humoral immunity (Mok et al., 2003). Autoimmune cond itions can be classifi ed according to their dependence on Th1 or Th2 responses; however , SLE shares both Th1 and Th2 response characteristics. From the previous study, many cytokines, such as IL-6, IL-10, IFN, and TNF, showed elevated serum concentration le vels (Kim et al., 1987; Gabay et al., 1997; Grondal et al., 2000). Rep eated administration of IFNto an experimental lupus model can intensify the diseas e (Engleman et al., 1981). Some mechanisms for IFNÂ’s effects in autoimmunity include activating macrophages and leading to enhanced antigen presentation, which includes autoantigens, in creasing expression of class I and class II MHC antigens and elaboration of additional cytokines and inflammatory enzymes including iNOS. Engelman et al. (1979) and Jacob et al. (1987) re ported that treatment with antibodies against IFN leads to an amelioration of the disease. The Role of Estrogen in Lupus While SLE can be seen in either sex at a ny age, females are at a much greater risk than males, with a female to male ratio of 10:1 (Lahita, 1999). The highest incidence of SLE occurs in women at their childbearing years (Lahita, 1996). Several lines of evidence suggest that sex horm ones, especially estrogen, ma y be a predisposing factor contributing to the female predilection or to exacerbations of SLE (Cutolo et al., 1995; Sanchez-Guerrero et al., 1997; Rood et al., 1998). Estrogen has been implicated as an enhancer of humoral immunity (Wilder, 1998). Estrogens are aromatized steroid horm ones synthesized from a cholesterol backbone and produced predominantly in th e ovary, although some aromatization may occur in adipose tissue and brain. Its struct ure is shown in Figure 1-1, Panel A. The potential consequence of exposure to estr ogens on the initiation or exacerbation of


6 immunological and autoimmune diseases, such as SLE, has been studied primarily from clinical observations of cyclic SLE fl ares, exacerbations by oral contraceptive administration, and abnormal steroid metabolis m. The most direct evidence to date linking increased estrogen levels to lupus development comes from the recently completed Safety of Estrogen in Lupus Erythematosus National Assessment (SELENA) clinical trial. Buyon JP et al. (2005) found that a shortcourse of hormone replacement therapy (HRT), during which the levels of circulating 17ß-estradiol reached about one fifth of peak menstrual cycle leve ls, significantly increased the risk for mild to moderate flares in menopausal women with SLE. In the mouse, several lines of evidence have shown that estrogen is an important factor: (1) female lupus-prone mice develop more severe disease and succumb at an earlier ag e than do males (Roubini an et al. 1978); (2) ovariectomy (removal of ovary so that no endogenous estrogen will be secreted) attenuates the disease; (3) injection of exogenous estroge n in females or castration in males accelerates lupus development (Sobel et al., 2005); and (4) A ndrogen treatment of females decreases autoantibody levels, diminish es renal disease, and improves survival (Roubinian et al. 1979). Elevated estrogen during pregnancy can affect lymphopoiesis. Smithson et al. (1994) confirmed that an increased level of estrogen can reduce the number of B cell precursors in the bone marrow, and Medina et al. (1994) revealed that pregnancy doses of estrogen impair the ability of pr o-B cells to progress to pre-B cell stage. At a later stage of B cell development, Grimaldi et al . (2001) found that estrogen can alter the distribution of B cell subsets on a BALB/c transgenic R4A2b mouse model, with a marked reduction in the percentage of transi tional type 1 and significant increase in the


7 mature marginal zone B cells. In this mous e model, elevation of serum estrogen level to 70-100 pg/mL, which is similar to that observe d during the follicular and luteal phases of the estrus cycle, can break the tolerance of high-affinity, naïve autoreactive B cells arising in the bone marrow. Characterization at the molecu lar level of the anti-DNA B cells revealed that these au torreactive B cells are rescued from negative selection with either rapid transit from the T1 to the T2 stag e and/or a lack of deletion at the T2 stage, and are activated in estradiol-treated mice. Most of these rescued, naïve, high-affinity, anti-DNA B cells differentiated to marginal z one B cells, and some to follicular B cells (Bynoe et al., 2000). Most of the estrogen responses are medi ated by intracellular estrogen receptors, although signal pathways that do not require estr ogen receptors have also been identified. Some of the signal pathways involved in regulating B cells by estrogen have been studied. It has been well know n that estrogen can enhance th e anti-apoptotic gene Bcl-2 expression in B cells, which is speculated to rescue the autoreactive B cells that would normally be deleted at an immature stage of B cell development (Bynoe et al., 2000). Grimaldi et al. 2002 also found estrogen can increase the level of CD22 and Shp-1 on B cells. The molecule CD22 is an inhibitory regulator of BCR signa ling (Carter et al., 1991; O’Rourke et al., 1 998), while Shp-1 inhibits BCR-mediated sigals by dephosphorylating molecules in the BCR signa ling pathway (Doody et al., 1995; Pani et al., 1995). The elevated level of CD22 a nd Shp-1 may raise the threshold for BCR crosslinking that is required for the dele tion of autoreactive B cells (Grimaldi et al., 2005).


8 The effects of estrogen on lymphocyte a nd macrophage cytokine secretion have been studied. Several previous studies have demonstrated the presence of estrogen receptors on lymphocytes and macrophages (Cohen et al., 1983; Daniel et al., 1983, Novotny et al., 1983; Paavonen et al., 1981; Ansar et al., 1985; Hu et al., 1988; Flynn, 1986; Schreiber et al., 1988). Hu et al. (1988) and Cutolo et al. (1993) reported estradiol treatment on rat peritoneal cells can increase IL-1 levels. Ot her studies showed decreased expression of IL-6 by estradiol treatment (Ga lien et al., 1997; Ray et al. 1987; Manolagas et al., 1995). This inhibition has been re vealed to involve estrogen receptor and a transcription factor NF B (Galien et al., 1997). Estrogen binds to estrogen receptor to form a complex, and this complex then binds to the NF B p65 to form a suppressor. This suppressor can bind to the IL-6 promotor a nd inhibit IL-6 production. Sarvetnick and Fox (1990) also showed that estrogens enhanced IFNproduction by lymphocytes. Estrogen can increase the synthesis and s ecretion of prolactin, a peptide hormone mostly produced by the anterior pituitary (Bole-Feys ot et al., 1998). This stimulation by estrogen is through its inhi bition of hypothalamic dopaminergic suppression (McMurray, 2001). Dopaminergic suppression of adenohypophys eal prolactin synthesis and secretion is crucial to its pharmacologi cal manipulation. Estrogen and prolactin share a reciprocal endocrinologic relationship, a nd estrogenic stimulation of prolactin secretion is, presumably, the primary cause for the highe r mean serum prolactin concentration in women compared to men (McMurray, 2001). In the NZB/NZW F1 mouse model of lupus, prolactin accelerates disease by stim ulating both cell and humoral-based immunity, whereas high physiologic levels of estrogen, when its prolactin stimulating properties were abrogated by dopaminergic agonism w ith bromocriptine, actually suppressed


9 autoimmune disease development (Elbourne et al. 1998). In the non-autoimmune BALB/c transgenic R4A2b mouse model, Peeva et al. (2000) also found bromocriptine can restore tolerance that was broken by estr ogen treatment only. A lthough estradiol plus bromocriptine-treated mice show an expansion of transgene-expressing 2b-producing B cells, which is similar to the mice treated onl y with estradiol, these rescued autoreactive B cells are present in a state that is nonr esponsive to physiological activation signals. Organochlorine Pesticide Chlordecone Chlordecone (C10Cl10O) is the common name for the chlorinated insecticide, decachloroocatahydro-1,3,4-methene-2H-cyclobuta(cd)pentalen-2-one, which is commercially available under the trade name of Kepone (Figure 1-1, Panel B). It is a tanto-white, sand-like, odorle ss chemical first introduced into agriculture as an insecticide in 1958 and was prohibited by EPA in 1978. It was widely used for twenty years for leafeating insects, ants and cockroaches, and as a larvicide for flies. Chlordecone has a very stable structure with a molecular wei ght of 490.68. The primary process for the degradation of chlordecone in soil or sediments is anaer obic biodegradation, and the process goes on very slowly. The stability of chlordecone contributes to the long persistence and presence in environmental sites where it was sprayed. In humans, the fate of chlordecone involves uptake by the liver, enzymatic reduction to chlordecone alcohol by a hepatic cytosolic aldo-keto reductase named chlordecone reductase, conjugation with glucuronic acid, partial c onversion to unidentified polar forms, and excretion of these metabolites mainly as glucuronide conjugates into bile (DHHS/ATSDR, Toxicological Profile for Mirex and Chlordecone, 1995).


10 Health Hazard Information of Chlordecone In 1975, industrial carelessness during the manufacture of chlordecone brought this agent to the attention of toxi cologists, when 76 of 148 worker s in a factory in Hopewell, Virginia, developed a severe neurologic syndrome known as Kepone shakes (WHO, Environmental Health Criteria Document No. 43: Chlordecone (143-50-0)). Acute exposure to chlordecone can irritate the ey es, nose, and throat. Chronic exposure can cause tremors, altered gait, poor coordina tion, slurred speech, muscle twitch, poor memory, behavioral changes, oc ular flutter (opsoclonus), visu al disturbances, arthralgia, headache, chest pains, weight loss, hepato megaly, and splenomegaly (Larson et al., 1979). Chlordecone may be a carcinogen to hum ans, as chronic exposure to chlordecone has been reported to cause liver cancer in animals and long-term e xposure to chlordecone can also cause adverse effect s on the reproductive system and kidneys (Sirica et al., 1989; Linder et al., 1983; Larson et al., 1979). Neurotoxicity Exposure to high-dose of chlordecone can cau se serious tremor in humans (Taylor, 1985). Electron microscopic examination f ound damage to Schwann cells, including membranous inclusion and cytoplasmic folds, prominent endoneural collagen pockets, vacuolization of unmyelinated fibers, focal degeneration of axons with condensations of neurofilaments and neurotubules, focal inter-la mellar splitting of my elin sheaths, the formation of myelin bodies, and a complex in folding of inner mesaxonal membranes into axoplasm (Phillips et al., 1985). The mechan ism of chlordecone effects on the nervous system has been studied in animal models. Wang et al. (1981) repor ted that chlordeconeinduced neurotoxicity correlate d closely with both brain a nd plasma concentrations of chlordecone. Fujimori et al. (1982) found significant decreases in whole brain and


11 striated dopamine levels in chlordecone-treat ed mice exhibiting tremors. Hoskins et al. 1982 suggested that tremors induced by chlo rdecone might be due to chlordeconeinduced calcium deficiency in brain synaptos omes, as significant di fferences in calcium content and subcellular distribution were f ound between chlordecone-treated and control mice. Jordan et al. (1981) reported that the brain synaptosomal sodium-potassium ion and oligomycin sensitive magnesium ion ATPases were significantly decreased in chlordecone-treated rats, and a linear relationship was observed between the decreased in ATPase activities and tremor activity, sugges ting the inhibition of the ATPase system by chlordecone in the brain might be related to the production of the neurotoxic symptoms. Hong et al. (1982) suggested that the hypothalamo -pituitary axis may be the primary neural target to the chlordecone-elicited decr ease in pituitary (Met5)-enkephalin level. Carcinogenesis of Chlordecone Chlordecone is carcinogenic in rats and mice and induces malignant tumors in the liver and other organs (Reuber, 1979). The mo st direct correlation between chlordecone and tumor production was revealed by a carcinogenesis bioassay conducted by the National Toxicology Program. In this study, Os borne-Mendel rats and B6C3 F1 mice of both sexes were used, and two doses of ch lordecone that were well tolerated by the animals were administered. A significant in crease in the inciden ce of hepatocellular carcinomas was found in both rats and mice. The time to detection of the first hepatocellular carcinoma observed at death was shorter for chlordecone-treated mice than control mice, and in both sexes and both specie s, the time appeared inversely related to the dose. In chlordecone-treated mice and ra ts, extensive hyperplasia of the liver was also found. However, chlordecone treatment did not increase the incidence of tumors other than in the liver compared to controls.


12 Reproductive Toxicity ChlordeconeÂ’s adverse effects on the re productive system include damaging the developing fetus, decreasing fe rtility in males and females, damaging the testes and decreasing sperm production and viability (Sir ica et al., 1989). Animal study revealed that chlordecone treatment can induce increas ed atresia among large follicles in mouse uterus (Swartz et al. 198 9). Gellert et al. (1979) also re ported that female rat offspring exposed prenatally to chlordecone exhibite d persistent vaginal estrus, anovulation, and tonic levels of serum estradiol. The repr oductive failure caused by chlordecone exposure was largely due to an effect in females characterized by prolonged follicle-stimulating hormone (FSH) and estrogen stimulation, induc ing constant estrus, large follicles and absence of corpora lutea, but with levels of luteinizing hormone (LH) subminimal for ovulation. Reduction of preovul atory LH offered an explanation for chlordeconeÂ’s reproductive deficits, but can not acount for the decreased fertility (U phouse et al., 1986). Immunotoxicity Despite the known adverse effects of ch lordecone described above, very few studies have been reported on chlordecone eff ects on human or animal immune functions. Chetty et al. (1993) reported an immunomodulatory effect of chlordecone by increasing spleen weight and plaque forming cells in ra ts. The original attention to the connection between pesticides and lupus came directly from the farmworkers who once toiled in the pesticide-laced muck farms off Lake Apopka region of Florida. They complained for years of common symptoms of unusal rashes, swelling a nd arthritic conditions, and others who grew up near the farms noticed common ailments too. At least 50 cases of lupus have been documented in the area ar ound the Lake Apopka which is much higher than the incidence of this disease nation wide.


13 Representative pesticides, including DDT , methoxychlor and chlordecone, were studied in our labs for their potentials to accelerate autoimmunity (Sobel et al., 2005). The most extreme effects were seen in chlordecone-treated mice. A lupus prone NZB/NZW F1 mouse strain was used in the tests. This mouse model of autoimmunity manifests disease similar to human SLE in its development of autoantibodies, immune complex glomerulonephritis, and earlier disease development in the female mouse. To reduce the influence by endogenous estradiol, mice were ovariectomized in the beginning of study. Exposure to chlordecone at 30 µg/mg/day significantly decreased the time to onset of renal impairment compared with the control group (Sobel et al., 2005). In an expanded study in which lower doses were us ed (0.167 to 16.7 µg/m g/day), chlordecone treatment caused a clear dose -related early appearance of elevated anti-double-strand DNA autoantibody titers (Figure 1-2) that corresponded with increased BUN and proteinuria (Figure 1-3, Pa nel C) as well as subsequent development of glomerulonephritis (Figure 1-3, Panel A and B) , which is the most serious manifestation of lupus in this mouse model. Immunohist ofluorescence confirme d early deposition of immune complexes in kidneys of mice treated with chlordecone (Figure 1-3, Panel D). In a follow-up study (Sobel et al., in press), the effect of chronic ch lordecone treatment on SLE was evaluated in ovary-inta ct NZB/NZW F1 female mice, as well as in female mice from a non-autoimmune BALB/c mouse strai n. In this study, chlordecone shortened significantly the time to onset of elevated autoantibody titers and renal disease in a dosedependent manner. The doses required to produce this e ffect were similar to those observed to accelerate SLE development in ovariectomized females. These observations confirmed the ability of chroni c chlordecone to influence the process of SLE. However,


14 treatment of female BALB/c mice with chlordecone for up to one year did not produce elevated autoantibody titers or any renal diseas e (Figure 1-4), suggesting an inability of chlordecone to cause a break in tolerance in this strain. This demonstrates the importance of genetic background for this effect. There was also no evidence for toxic effects, as these mice appeared healthy at the end of the experiment. Estrogenicity of Chlordecone Recently there has been consid erable interest in the abil ity of environmental agents to produce adverse health effects secondary to endocrine disrupti on, particularly through estrogenic or anti-estrogenic effects. Several studies found insecticides, such as methoxychlor, chlordecone, endosulphan, toxaph ene and dieldrin, exhibit estrogen-like activities (Soto et al., 1994). Estrogenicit y is defined as the property of producing biologic responses qualitatively similar to those produced by the endogenous hormone, 17 -estradiol. Hallmark estrogenic reponses in mammals include increased uterine weight and vaginal epithelia cornifica tion, but there are many others, including development of secondary sexua l characteristics, and regulation of the estrous/menstrual cycle. Estrogenic chemicals do not always sh are any chemical structural resemblance to the prototypical estrogen17 -estradiol, but evoke agonist or antagonist responses possibly through a comparable mechanism of action (Lee et al ., 2004). Estrogenic responses to toxic chemicals may be elicite d by direct binding to and activation of intracellular estrogen receptors, leading to ch anges in gene expression characteristic of endogenous estrogen exposure (Daston et al ., 1997). The estrogen signaling pathway may also be independent of estrogen recepto rs (Das et al., 1997) via other response proteins, such as a novel orphan receptor me mber of the nuclear receptor superfamily (Enmark et al., 1996). Toxicants may also i ndirectly produce an es trogenic response by a


15 number of different mechanisms, such as in creasing estrogen synthe sis (e.g., peroxisome proliferators inducing aromatase activity, thereby increasing ci rculating estradiol levels), or changing the rate of estrogen degradation (Lee et al., 2004). Chlordecone has reportedly a weak capabi lity to bind to the estrogen receptor in the uterus. Our study has also confirmed a weak binding of chlordecone to estrogen receptors in the spleen (unpublished observa tions). Okubo et al. (2004) reported an estrogenic effect of chlordecone on MCF-7 ce ll proliferation that was suppressed by the antiestrogen ICI 182,780. Hodges et al. (2000) also reported chlordecone estrogenic effects on Eker rat uterine leiomyoma-driv ed cells by stimulating proliferation and transcription of vitellogenin estrogen-response element via the estrogen receptor, and inducing the expression of an endogenous estrogen-responsive gene the progesterone receptor. Classic uterus hypotrophy assay also shows a chlordecone estrogenic effect by significantly increasing uterus weight. Our st udy verified that a high dose of chlordecone (30 mg per implantable, sustained-release pe llet) significantly incr eased uterus weight, (Sobel et al., 2005); however at low doses that clearly accelerate autoimmunity in NZB/NZW F1 mouse model, chlo rdecone showed little or no uterotrophic effects (Figure 1-5; Figure 3-2, Panel D). These data s uggest that exposure to chlordecone can accelerate autoimmunity in the context of a genetic predisposition, but may not be directly linked to estrogenic pot ency as measured in reproductive tissue. There are at least two possible inte rpretations for the different uterine hypertrophy findings: 1) Chlordecone accelerates autoimmunity through its estrogenic effects. The chlordecone doses required to produce estr ogenic effects on the immune system are lower than the doses required to produce estr ogenic effects in uterotropi c assays; and 2) Chlordecone


16 accelerates autoimmunity, wholly or in part, through an estrogen-independent mechanism. We hypothesized that chlordecone influen ces autoimmunity by functioning as an estrogen mimic. Pilot studies showed bindi ng of chlordecone to estrogen receptors on immunocytes. Although the binding affinity of chlordecone was low, receptor binding might nevertheless result in accelerated autommunity through the same mechanism [albeit with higher doses] as estradiol. If th is is the case, chlordecone and estradiol would be expected to produce similar effects on es trogen-responsive genes in immunocytes at doses relevant to effects on autoimmunity, as well as similar phenotypic changes in immunocytes germane to lupus. Changes in cytokine expression t hought to influence the onset and progression of lupus would also be ex pected to be the same. There is evidence that the effects of estrogen on lupus are in fact mediated through increased levels of prolactin (Elbourne et al. 1998) . A similar effect by chlord econe to increase prolactin would also be consistent with the hypothesis. The hypothesis was tested by comparing chlo rdecone and estradiol effects in the NZB/NZW F1 mouse model in the following respects: 1) Expression of selected estrogen-res ponsive genes in B cells. B cells were selected as logical targets for estrogenic effects to modulate autoimmunity, and the comparison focused on genes: a) known to be either upor down-regulated by estradiol and b) plausibly related to immune dysfunction leading to autoimmunity. 2) Phenotypic changes in T and B cells associated with autoimmunity. As indicated in Figure 1-6, proce sses leading to a break in im mune tolerance, production of autoantibodies, and development of clinical disease are complex. While several potential


17 contributing factors to lupus development have been identified, none has yet been demonstrated to be a single critical effect . Consequently, comparisons between estradiol and chlordecone were made with respect to several phenotypic markers. The markers examined included some shown previously to be changed by estradiol, as well as others that are related to lupus for which the e ffects of estradiol [and chlordecone] were unknown. 3) Changes in cyotkine expression. As w ith changes in phenotypic expression in lymphocytes, estradiol and chlordecone were compared with respec t to their effects on CD4 T cell cytokine secretion, which is t hought to play an impor tant role in the development and progression of lupus. 4) Effects on macrophage clearance of apopt otic bodies. Diminished clearance of apoptotic bodies has been suggested as incr easing the risk of autoimmunity by increasing the autoantigen burden (see Figure 1-6). There is evidence that organochlorine pesticides suppress macrophage function, and this might be a means by which chlordecone and estradiol accelerate autoimmunity. The e ffects of estradiol and chlordecone on macrophage clearance of apoptotic b odies were examined and compared. 5) Effects on prolactin. As noted above, estradiol effects on autoimmunity have been postulated to be mediated through increa sed prolactin. The eff ects of estradiol and chlordecone on circulating prolactin levels, as well as expression of prolactin receptor on lymphocytes, were compared. All comparisons were made with doses of chlordecone and estradiol shown in previous studies in our laboratories to accelerate the development of lupus. The hypothesis would be considered supported if estradiol and chlordecone produced the


18 same, or essentially similar effects. Litt le or no similarities would suggest that chlordecone influences autoimmunity by a diffe rent mechanism than estradiol. If some of the same effects were observed, this woul d indicate that chlordecone is not functioning simply as an estrogen mimic, but would leav e open the possibility that chlordecone and estradiol both affect autoimmunity by the sa me mechanism. Effects shared by estradiol and chlordecone could provide clues to th at mechanism to guide future research.


19 A B Figure 1-1. Structures of (A) 17 -estradiol and (B) chlordecone.


20 A. IgG anti-dsDNA autoantibody titers B. IgG anti-chromatin autoantibody titers Figure 1-2. Time to development of elevat ed autoantibodies in female NZB/NZW F1 mice treated with chlordecone. Anti -dsDNA (A) and anti-chromatin (B) autoantibody titers were measured over time in mice implanted with 60-day sustained release pellets containing chlordecone at the specified amount. Controls received pellets without chlo rdecone. Pellets were replaced every 60 days over the course of the experiment. The time to development of elevated autoantibody titers in mice treated wi th pellets containing 1 or 5 mg chlordecone was significantly less than controls for both specificities (p = 0.005 for anti-dsDNA and p = 0.001 for anti-chromatin autoantibody titers (N=15-20 per group) (Sobe l et al., in press)


21 A. Proliferative glomerulonephritis B. Proteinuria level C. Immunofluorescence staining fo r IgG immune complexes on kidney Figure 1-3. Enhanced renal disease and imm une complex deposition in ovariectomized, chlordecone-treated mice after 8 weeks of treatment with control pellets or pellets containing chlordecone (1 mg/pel let) or 17ß-estradiol (0.05 mg/pellet; n = 6/group). Frequency of appearance of proliferative glomerulonephritis (A) and proteinuria (B) was significa ntly increased in both chlordeconetreated and 17ß-estradiol-treated mi ce. Immunofluorescence staining (magnification, 200 ) for IgG in kidney was shown in (D). The staining was absent in control mice, but present in mice treated with either chlordecone or 17ß-estradiol after 8 weeks of treatment; later, when renal disease developed in controls, a similar extent of i mmunofluorescence stai ning was observed. (Sobel et al., 2005)


22 Figure 1-4. Anti-chromatin auto antibody titers in female BALB/c mice treated with chlordecone. Anti-chromatin autoantibody titers were measured in terminal blood samples taken after one year of chlordecone treatment. Chlordecone doses reflect the amount of chlordecone delivered via sustained-release pellet every 60 days. There was no significan t difference in autoantibody titers between any of the chlordecone treated groups and controls receiving a pellet without cchlordecone. (N=20 per gr oup). Elevated anti-chromatin autoantibody titers from NZB/NZW F1 mi ce at the end of an experiment are shown for comparison. (Sobel et al., in press)


23 Figure 1-5. Uterine hypertrophy in ovariecto mized NZB/NZW F1 mice 6 weeks after adminiatraion of pellets containing 0.05 mg estradiol and different doses of chlordecone. To compensate the mass of the mouse, uterine weight was expressed as a ratio to total body we ight (i.e., uterine mass/body mass 1,000). Estradiol significantly increased uterus weight. Ch lordecone at the dose of 36 mg also increased uterine wei ght as much as es tradiol treatment, while low doses of chlordecone (1.8 and 18 mg) only caused a moderate increase in uterine weig ht. (Sobel et al., 2005)


24 Figure 1-6. Potential pathogenesis of lupus de velopment. Although far from elucidation, several critical pathways listed in th e figure have been indicated to be responsible in lupus development. A combination of factors, including genetic, environmental, hormonal, and vi ral influences, can affect any of the pathways, and result in autoimmunity. Increased autoantigen burden 1. Increased rate of apoptosis 2. Defective clearance of apoptotic cells Example: c-mer deficiency 3. Decreased clearance of immune complexes Example FcR polymorphisms; C1q deficiency Altered signaling threshold 1. Central tolerance defects in selection of nave repertoire 2. Peripheral tolerance defects caused by increased stimulatory signals by T, B,or dendritic cells Example: Il-10 deficiency; overproduction of Type I interferons Resistance to apoptosis by cells of the immune system Examples: Fas deficiency; overexpression of Bcl-2 Loss of tolerance Increased autoantibody production Immunemediated end-organ damage Increased susceptibility to end-organ damage


25 CHAPTER 2 MATERIALS AND METHODS Experimental Animals Female NZB/NZW F1 mice (6-8 weeks ol d) were purchased from The Jackson Laboratories (Bar Harbor, ME). Mice were housed in temperature-, light-, and humiditycontrolled animal quarters in a specific pathog en-free (SPF) barrier facility. Mice were kept in polycarbonate cage s on corncob bedding, with fr ee access to food and water throughout the study. All procedures were appr oved by the Institutional Animal Care and Use Committee of the University of Florida. Test Materials and Treatments Chlordecone, described as 99.2% pure, wa s purchased from Crescent Chemical (Islandia, NY), and was formulated into 60day sustained-release pellets by Innovative Research of America (Sarasota, FL). The id entity of material as supplied was confirmed by gas chromatography-mass spectrometry. Each chlordecone pellet contained either 1 or 5 mg chlordecone. Pellets containing 0.05 mg 17-beta estradiol or matrix only (as controls) also obtained from Innovative Resear ch of America were used in the study for comparison purposes. At the beginning of the experiments, each mouse was surgically ovariectomized one week after they were received. The su rgery was performed under anesthesia with a mixture of ketamine (133 mg/kg) and xylazine (13 mg/kg) administer ed intraperitoneally. The mice were allowed to recover for two w eeks after the surgery. Following recovery, a pellet containing chlordecone, 17-beta estr adiol, or matrix only was implanted


26 subcutaneously between the shoulders unde r light methoxyfluorane anesthesia. Mice tolerated the ovariectomy and pellet implan tation procedures without complications. Mice were euthanized 5-6 weeks after pellet implantation, and tissues were harvested for study. Flow Cytometry Flow cytometry was performed with a CyAn ADP Analyzer (DakoCytomation, CA), and analysis was performed using FC S 3 Express (De Novo Software, Ontario, Canada). Single cell suspensions from spleen s were stained with mixtures of antibodies to the following surface markers: B 220, CD19, IgM, CD44, CD69, CXCR4, CXCR5, ICAM-1, VCAM-1, I-A(d), B7.2, GL7, Fc-gam ma-IIb, CD4, CD3, CD 8. Intracellular staining was performed as desc ribed by Sobel et al. (2002). Cells were incubated with rabbit anti-Bcl-2-PE or rabbit IgG-PE as isot ype control. All anti bodies were purchased from BD Biosciences Pharmingen (San Jose, California, USA). When the analysis was performed, the expression of some molecu les was not appeared as two distinctive populations. In this case, the positive populat ions were defined by setting that day’s control mouse value at 50%, and the percentage of positive populations from chlordeconeand estradiol-treated mice were calculated based on that cutoff. Apoptosis Assay B cells were enriched from splenocytes of chlordecone-, estr adioland control pellet-treated mice by negative selection with a mouse B cell isolation kit (Miltenyi Biotec, Auburn, CA) according to the manufactur er’s directions. Pu rified B cells (1 x 106/mL) were cultured in duplicate in comple te RPMI 1640 medium with or without 0.1 or 1 µ g/mL LPS at 37°C and 5% CO2 for 24 hours on 96-well flat plates (Corning Inc., Corning, NY). Cells were then stained with surface markers, and apoptosis was assessed


27 by flow cytometry with the annexin VPE and 7-Amino-actinomycin D (7-AAD) apoptosis detection kit I (BD Biosciences Ph armingen), according to the manufacturer’s instructions. CD4 T cells were enriched from splenocytes from chlordecone-, estradioland control pellet-treated mice by negative sele ction with a mouse CD4 T cell isolation kit (Miltenyi Biotec, Auburn, CA) acco rding to the manufacturer’s directions. Purified CD4 T cells (100,000 cells) were cultured in dupli cate in complete RPMI 1640 medium with total volume of 100 µ L at 37°C and 5% CO2 for 24 hours under the following three conditions: 1) with no stimulation and incuba ted on an uncoated 96-we ll plate; 2) with no stimulation and incubated on a 96-well T cell activation plate coated with anti-mouse CD3 (BD Biosciences, San Diego, CA); 3) with 2 µg/mL or 4 µg/mL anti-CD28 stimulation and incubated on a 96-well T cell activation plate coated with anti-mouse CD3 (BD Biosciences). Anti-CD3 crosslinks the T cell receptor and in the absence of anti-CD28 is a suboptimal stimulus for T cells . Crosslinking of CD28 provides a potent co-stimulatory signal and trigge rs a proliferative response. Cells were then stained with surface markers, and apoptosis was assessed by flow cytometry with the annexin V-PE and 7-AAD apoptosis detection kit I (BD Bi osciences Pharmingen) according to the manufacturer’s instru ctions. Apoptotic cells were iden tified as those cells excluded annexin V and 7-AAD double negative cells. Proliferation Assay CD4 T cells were purified by negative sele ction using a mouse CD4 T cell isolation kit (Miltenyi Biotec), and B ce lls were isolated as described above. Cells at (2 x 105/well) were cultured for 72 h in complete RPMI 1640 medium at 37°C and 5% CO2, with or without added stimuli. B cells were stimulated by 0.1 or 1 µ g/mL


28 lipopolysaccharide (LPS), and T ce lls were cultured either in an uncoated 96-well plate or an anti-mouse CD3-coated T-cell activation plate (Becton Dickinson Labware, MA), with or without 5 µ g/mL anti-CD28. Lymphocyte prolifer ation was assessed by testing [3H]TdR incorporation (ICN Biomedicals, Costa Mesa, CA) over the last 18 h of culture. Cells were collected by a cell harvester (C ambridge Technology Inc., Watertown, MA), and the thymidine incorporation was measured by a scintillation counter. Proliferation index was measured as the ratio of stimulat ed to unstimulated thymidine incorporation. CD4 T Cell Cytokine Determination CD4 T cells were enriched from splenocyt es by the mouse CD4 T cell isolation kit as described above. Purifed CD4 T cells (100,000 cells) were cultured for 24 h in 100 µ L of complete RPMI 1640 medium at 37°C and 5% CO2, either on a regular 96-well plate or an anti-mouse CD3 T-cell activat ion plate, with or without 5 µ g/mL CD28 stimulation, as described above. Assessments of cy tokine profiles were performed using a commercially available multiplexed kit (Beadlyte Mouse Multi-Cytokine Detection System 2; Upstate Biotechnology, Waltham, VA) and the Luminex (100) LabMAP System (Austin, TX). Simultaneous meas urement of 10 cytokines was performed: IL1 , IL-2, IL-4, IL-6, IL-8, IL-10, IL-12 (p70), tumor necrosis factor (TNF), -interferon (IFN), and granulocyte-macrophage colony-stimulating factor (GM-CSF). The assay was performed according to the manufacturer’s protocols. Fifty µ L of culture supernatant were used in the assay, and cytoki ne concentrations were determined utilizing SOFTmax PRO software (Molecula r Devices, Sunnyvale, CA) with four-parameter data analysis. cDNA Preparaion RNA from 107 splenic B cells isolated via magnetic beads was extracted by TRIZOL reagent (Invitrogen), and the c oncentration of RNA was measured by a


29 spectrophotometer. RNA (1 µg) was then tr eated with DNase I (Invitrogen) to remove genomic DNA and reverse transcribed to cDNA using Superscript II First-Strand Synthesis System (Invitrogen) for RT-PCR. Briefly, First strand cDNA was synthesized in 20 µl buffer containing 50 mM Tris -HCl (pH 8.3), 40 mM KCl, 6 mM MgCl2, 1 mM DTT, 200U/ µ L RNase inhibitor, 200 U/ µ L Superscript II Reverse transcriptase, 0.1 mM oligo (dT)18 primer and 1 µg total RNA. The reaction proceeded at 42 ºC for 50 min and was terminated by heating for 15 min at 70 ºC . Samples were then stored at 4 ºC. Real-time PCR Gene expression was determined by r eal-time PCR using SYBR green (Applied Biosystems, Foster City, CA). Primers were designed as shown in Table 2-1. One microliter of cDNA from the preparat ion as described above was added to a reaction mixture containing 3 mM MgCl2, 1 mM dNTP mixture, 0.025 U of Amplitaq Gold, SYBR Green dye (Applied Biosystems), optimized concentrations of specific forward and reverse primers, and DEPC-tre ated water in a final volume of 25 µL. Amplification conditions were as follows: 95° C (10 min), followed by 45 cycles of 94°C (15 s), 60°C (25 s), 72°C (25 s), with a final extension at 72°C for 8 min. Transcripts were quantified using the comparative (2 Ct) method. In some experiments, Mx1 and -actin expression levels were determined by real-time PCR with LUX primers and a platinum PCR supermix kit (Invitrogen). Am plification conditions were as follows: 50°C (2 min), 95°C (2 min), followed by 45 cycles of 95°C (15 s), 55°C (30 s), 72°C (30 s), and a final extension at 72°C for 8 min. ELISA for Autoantibody Titers. We measured autoantibody [IgG antidouble-strand DNA (anti-dsDNA)] titers in serum in some treatment groups using i ndirect ELISA (enzym e-linked immunosorbent


30 assay). Immulon 2 microtiter plates (Dynat ech Laboratories, Inc ., Chantilly, VA) were coated overnight at 4°C with a 1:10 (vo l/vol) dilution of pol y-l-lysine at 100 µ L/well. Between all steps, microplates were washed three times with bor ate-buffered saline (25 mM Na2B4O7, 75 mM NaCl, 100 mM H3BO3, pH 8.4) containing 0.05% Tween 20. After addition of calf thymus DNA (50 µ L/well at a concentration of 20 µ g/mL), the plate was blocked with 100 µ L/well of 3% bovine serum albumin (BSA) in phosphatebuffered saline (PBS). Dilutions (1: 200) of serum samples were prepared in PBS and incubated in the appropriate wells at room temperature for 1 hr. The secondary antibody (goat anti-mouse IgG peroxidase conj ugated Fc gamma specific; Jackson ImmunoResearch Laboratories, West Grove, PA ) was diluted to 1: 5,000 (v/v) in PBS and added at 50 µ L/well. The developing solution c onsisted of o-phenylene diamine at 0.4 mg/mL PC buffer (4.7 g/L citric acid and 13.8 g/L sodium phosphate dibasic heptahydrate, pH 5.1) with 0.01% H2O2. The substrate turnover was determined by the difference between the OD450 (optical density) and OD620 on a Molecular Devices (Sunnyvale, CA) microplate reader. The con centration of antigen-specific IgG is reported in equivalent dilution factors (EDFs) of standardized reference NZB NZW/F1 sera. This is defined by the formula EDF = (dilution of a standard reference sera that gives the equivalent OD of the test serum) 104. Determination of Serum Prolactin. Serum was obtained from final bleeding and frozen at -80 °C until analysis. Prolactin radioimmunoassay (RIA) used m ouse PRL (mPRL) reference preparation AFP6476C, mPRL AFP1077D for iodinati on, and anti-mPRL antiserum AFP131078 (provided by A.F. Parlow and the Nati onal Hormone and Peptide Program, Harbor– UCLA Medical Center, Torrance, California, USA). The assay was performed by the


31 National Hormone and Peptide Program. Briefl y, fifty µl of serum from each sample was used in the RIA, and approximately 20,000 cp m was added to each tube. Quantitative limits for the assays ranged from 3-6 ng/mL with intra and interassay coefficients of variation <15%. Samples were analyzed in duplicate at multiple dilutions. Producing and Labeling Apoptotic Cells Thymus cell suspension from female BALB/c mice was exposed to 10-6 uM of dexamethasone (Sigma) in complete RPMI media at 37°C with 5% CO2 for 18 hours. Thymus cells were then washed and staine d with 7-AAD (Sigma) for labeling. Briefly, cells were adjusted to 107 /mL in phosphate-buffered saline (PBS) followed by adding 7AAD into the cell suspension to a final concen tration of 20 µg/mL. Cells were then incubated at 4 ºC for 20 minutes followed by washing twice with PBS. Flow cytometry was performed to verify the apoptosis of ly mphocytes after the dexamethasone treatment, and more than 90% of the cells after staining were 7-AAD positive. Clearance of Apoptotic Cells Peritoneal cells from chlordecone-, estradiol-, and control-pellet treated mice were carefully harvested in PBS with 5% fetal bov ine serum (FBS) at pH 7.2 without red blood cell contamination. One million peritoneal cells were then stained with CD11b-APC, B220-APC-Cy7, CD19-PE-Cy7 and CD5-PE. After staining, cells were washed twice in PBS and then cultured in a 6-well plate with 2 mL complete DEME media at 37°C with 5% CO2 for 1 hour to allow cells recover. One million 7-AAD stained apoptotic cells were then added into the plate and mixed well with pre-cultured peritoneal cells. The mixed cells were incubated for one hour at 37°C with 5% CO2 before they were harvested using a scraper and collected for fl ow cytometry. Peritoneal macrophages that had engulfed apoptotic cells were defined as the 7-AADhighCD11bhigh and CD19low cell


32 population, while macrophages without apoptotic cells were 7-AADlowCD11bhigh and CD19low cells. B1 cells, also present in the peritoneal cavity a nd reported to have phagocytic capabilities (Borrello et al., 2001), were defined as CD19+CD5+ cells. Statistical Analysis . Statistical analyses were mainly conducte d using the software GraphPad Prism, version 4.00 (GraphPad, San Diego, CA). Si gnificance was determined using DunnettÂ’s procedure of the one-way ANOVA. All p < 0.05 were considered to be statistically significant. For many responses, control group da ta were visually not variable compared with the treatments, and this was confirme d using Levene's test for heterogeneity (Levene, 1960). Because of this, the control group data were excluded in the one-way analysis of variance model. Differences in treatment means were identified using Gabriel's comparison intervals method (Gabri el, 1978). If the Control group mean was outside the Gabriel comparison interval fo r any treatment, the treatment mean was considered to be significantly different fr om the control mean. Standard control-totreatment comparison methods, such as Dunnett' s test (Dunnett, 1980), could not be used here because typically the control group vari ance was one to two orders of magnitude smaller than any treatment group variance. Any treatment that caused significant increase compared with contro l group was marked with a on the top of the treatment. The groups with no statistically significant di fference were marked with a same letter (either a or b), while groups with significan t difference were marked with different letters.




34 CHAPTER 3 COMPARISON OF SPLENIC LYMPHOCY TE PHENOTYPIC CHANGES WITH TREATMENT OF CHLORDECONE AND ESTROGEN Introduction Systemic lupus erythematosus (SLE) is a chronic, multisystem autoimmune disease characterized by elevated levels of pa thogenic autoantibodies (Kotzin, 1996). B lymphocytes are active participants in humoral immune responses that lead to differentiation into antibody-s ecreting cells in either T cell-dependent or T cellindependent ways. Studying the phenotypic ch anges of lymphocytes provides direct cellular evidence for the disease proce ss and may reveal mechanisms for the pathogenesis. Although in murine models of lupus, DNA-r eactive B cells repres ent < 0.1% of the splenic cell population (Klinman et al., 1991), they are a fairly sensitive and very specific marker for the loss of humoral tolerance (Kotzin, 1996). Many m echanisms exist to maintain central B cell tolera nce in the bone marrow, incl uding clonal deletion (Nemazee et al., 1989; Hartley et al ., 1991), clonal anergy (Goodnow et al., 1988) and receptor editing (Tiegs et al., 1993; Radic et al ., 1993) of autoreactive receptors. In many autoimmune models, including NZB/NZW F 1, central tolerance is generally well preserved (Wellmann et al., 2001). However, because the process of somatic hypermutation occurring during a T cell-d ependent B cell response in secondary lymphoid tissue permits new autoreactive B cells to arise, especia lly in the germinal


35 center, additional tolerogenic mechanisms must also exist in the peri phery (Cornall et al., 1995; Paul et al., 2004). Germinal Centers In the past 15 years, an enormous amount of research ac tivity has been devoted to understanding the pathogenesis of autoantibodies. The vast majority of the data indicate that most IgG autoantibody responses are T cell-dependent, MHC-restricted, of high affinity and antigen-driven and essentially follow the same pathway as responses to T cell-dependent foreign antigens (Maddison, 1999) . The site where these responses occur is the germinal center. Germinal centers ar e microenvironments that form within follicles of secondary lymphoid tissue and are the site of rapid B cell expansion, somatic hypermutation, isotype switching, affinity matu ration, apoptosis, plasma cell commitment and memory cell formation. The initiation of germinal center formation and maturation requires several essential cell types, including dendritic cells, antigen-specific B cells, T helper (Th) cells and follicul ar dendritic cells (Cozine et al., 2005). In a conventional response to foreign antigens, dendritic cells at the site of exposure have taken up the antigen and processed the prot ein component into peptides which are displayed bound to MHC class I and class II molecules. In the cont ext of the damage at the site of the initial response, “danger” signals in the form of pathogen-associated molecular patterns (PAMPs) are delivered which cause the dendr itic cells to mature, upregulate potent costimulatory molecules and migrate to the draining lymph node. Soluble, blood-borne antigens are thought to cause the same type of stimulation to resident dendritic cells in the spleen. Within the secondary lymphoi d tissue, naïve CD4 T cells, which tend to circulate through the paracortical area, encount er the activated dendritic cells. Those T cells with a T cell receptor that can engage the peptide/MHC complex with sufficient


36 affinity are stimulated to proliferate and differentiate into arme d effector T cells. Subsequently, these antigen-sp ecific CD4 T cells engage antigen-specific naive B cells, and these activated B cells then migrate into nearby follicles, where they proliferate, differentiate, and undergo a stringent selec tion process. Based on its histological appearance, the germinal center can be divided into two distinct poles or zones, called the dark and light zones. B cells in the dark zone, called centroblasts, undergo rounds of rapid proliferation and somatic hypermutation of their antibody variable genes. The hypermutation process results in progeny B cells with different affinities for the stimulating antigen. The centroblasts then b ecome smaller, non-dividing centrocytes and undergo selection in the light zone based on th e affinity of their surface antibody for the inducing antigen, which is bound to follicular de ndritic cells (FDC), a population of cells distinct from bone marrow-derived dendritic cells. It is thou ght that further interactions with germinal center T cells provide impor tant proliferative and survival signals necessary to expand antigen-specific centrocytes in the light zone and form the germinal centers, despite their small percentage with in the germinal centers. Germinal center follicular dendritic cells also display imm une-complexed antigen on their surface, and enable the selection of high-affinity B cell clones (Cozine et al., 2005). Chemokine receptors CXCR4, CXCR5 and their ligands CXCL12, CXCL13 were reported to play a key role in germinal center organization based on CXCR4-/and CXCR5-/knock-out study (Allen et al., 2004). The dysregulation of mechanisms control ling normal or ectopic T-cell dependent germinal center reactions to exogenous or endogenous antigens may contribute to the emergence of SLE. Comparison between l upus prone mice, such as the NZB/NZW F1


37 mouse, lpr/lpr mouse, and non-autoimmune mice , showed that autor eactive B cells were able to form and/or enter splenic follicles in the lupus-prone mice, but were retained outside follicles in the T cell zone in non-auto immune mice. The failure to appropriately exclude autoreactive B cells from the germinal center may therefore lead to autoantibody formation. Non-autoreactive B cells may appropriately enter a germinal center response but then develop an autoreactive specificity th rough somatic hypermutation. Normally, these cells should be selected against at the cen trocyte stage, as th ey would not compete effectively for survival signals from follicular dendritic cells. The survival signals are delivered to those B cells with an immunogl obulin receptor specific for intact foreign antigen bound by the FDCs. Competition fo r FDC binding is strong, and most B cells fail to receive sufficient signals, die by apopt osis, and are efficiently cleared by tangiblebody macrophages. Therefore, defects in the cl earance of apoptotic cells in the germinal center could lead to escape of negative se lection by autoreactive B cells. Baumann et al. (2002) reported that in a sub-group of SLE patients, apoptotic cells were not properly cleared. Consequently, nuclear autoantigens bound to follicular dendritic cells and may thus have provided survival signals for autor eactive B cells. This action may override an important control mechanism for B cell development, resulting in the loss of tolerance for nuclear antigens. Marginal Zone B cells The marginal zone is a distinguishable concentric collection of cells surrounding splenic follicles and separates the white pulp from the red pulp. The marginal zone is primarily made up of marginal zone B cells, specialized macrophages, and reticular cells (Lopes-Carvalho et al., 2004). Whereas fo llicular B cells are freely recirculating


38 lymphocytes, marginal zone B cells are nonrecirculating cells. Marginal zone macrophages might play an important role in re taining marginal zone B cells within this area by expression of the ligand for the marginal zone B cells scavenger receptor MARCO (Kraal, 1992; Karlsson et al., 2003). Another pos sible mechanism is the interaction of integrins on B cells with integr in ligands that are induced in the marginal zone in stromal cells by signaling via the LT R (Lu et al., 2002). Marginal zone B cells have a distinctive phenotype. They expr ess high levels of IgM and very low levels of IgD and CD23. They also express higher levels of CD21, CD1d, CD38, CD9, and CD25 than follicular B cells (Gray et al., 1982; Oliver et al., 1997 and 1999; Waldschmidt et al., 1991; Hsu, 1985). There are two important factors involved in the development of the marginal zone B cell. The primary determinant is the specificity of the B cell receptor, which allo ws a developing B cell to choose between a follicular and a marginal zone B cell fate. The second major driving force is notch signaling. B cells are selected to mature into the marginal zone B cell only after they receive additional signals from Notch2 and NFB1 (Pillai et al., 2005). The most important functional feature of marginal zone B cells is their very early participation in immune responses. Their unique location permits them to respond very rapidly to blood-borne pathogens. Marginal zone B cells seem to have a lower threshold than recirculating or immature B cells for act ivation, proliferation a nd differentiation into antibody-secreting cells. Recent studies have re vealed the participati on of marginal zone B cells in T-independent as well as in Tdependent immune responses. Studies of marginal zone B cells also suggest that they can potentially play a role in autoimmunity. In the autoimmune NZB mouse model, there is an increased proportion of marginal zone


39 B cells with increased levels of co-stimu latory molecules as compared with the nonautoimmune mouse strains (Wither et al., 2000 ). Lupus-prone NZB/ NZW F1 mouse also showed expanded marginal zone B cell as early as 4 weeks of age (Wither et al., 2000), and the marginal zone B cells are responsib le for production of large amounts of antiDNA antibodies, as compared to follicular B ce lls (Zeng et al., 2000). The distribution of marginal zone and germinal center is show n in Figure 3-1, and the brief introduction of the phenotypic markers tested in this chapter was shown in Table 3-1. We hypothesized that autoimmunity c ould be accelerated by exposure to chlordecone in lupus-prone mice through e ffects on one or more of the checkpoints involved in maintaining peripheral tolerance. The experiments below were designed to compare the effects of chlordecone with those of estradiol at relatively early time points. Results Chlordecone Exposure Caused Only Slight Changes in Spleen, Uterus and Body Weight Compared to Estradiol In an attempt to examine early events in cells of the immune system following exposure to chlordecone, a cohort of two-m onth-old ovariectomized mice was implanted with control pellets, or pellets containing 1 mg of chlordecone, 5 mg of chlordecone, or 0.05 mg of estradiol. Although not a positive cont rol in the strictest sense, we compared the results with mice given 0.05 mg estradiol pellets, as th e effects of es tradiol on the immune system of NZB/NZW F1 mice have been well studied, even if all of the pathways affected by estradiol have not been fully delineated. The mice were evaluated five to six weeks later for evidence of incr eased immune activation, at which time point no significant increase of anti-dsDNA antibody t iter by chlordecone ha d been previously observed (Sobel et al., in press). Mice we re euthanized, and mouse body, spleen and


40 uterine weights were obtained. As can be seen in Figure 3-2, Panel A, all groups of mice had similar body weight before treatment. Five to six weeks after implantation, all groups of mice gained weight. The controltreated group gained the most weight, while the estradiol-treated group had the least weig ht gain. Mice treated with chlordecone did not show difference in body weight compared with the placebo-treated group (Figure 3-2, Panel B). In contrast, mice treated with 5 mg (but not 1 mg) chlordecone had a statistically significant increase in spleen weight relative to control mice. This was not as great an increase as was seen with estradiol (Figure 3-2, Panel C). Chlordecone treatment did not cause significant cha nge on uterus weight, while estradiol dramaticly increased the uterus weight about seven times compar ed with the control group (Figure 3-2, Panel D). Estradiol, but Not Chlordecone, Changed Lymphocyte Percentages Flow cytometric analysis showed no difference in the percent composition of CD4 and CD8 T and B cells in the spleen of chlord econe-treated mice. In contrast, estradioltreated mice showed a significant decrease in the percentage of bot h B cells (Figure 3-3, Panel A) and T cells (Figure 3-3, Panel B and C), suggesting that the splenomegaly in the estradiol-treated group was due to a relative increase in non-T and non-B cells. Although estradiol significantly reduced the lymphocyte percentage, th e ratio of CD4/CD8 T cells did not significantly change compared with control group (data not shown). Both Chlordecone and Estradiol Treatments Activated Splenic B Cells To further study chlordecone and estradio l effect on B lymphocytes, seven-color flow cytometric analysis was performed on splenocytes with antibodies to markers on B cells. The expression levels of MHC Class II, CD86 (B7.2), and CD44 did not result in distinct positive and negative peaks. Therefore, to control for any day-to-day fluctuations


41 in instrument setup, one mouse from each group was studied on each experimental day and the expression levels were normalized as the ratio of the geometric mean of the treated mice to that of that dayÂ’s control-treated mouse. Both chlordecone and estradiol exposure increased the expre ssion level of B cell activati on markers CD44 (Figure 3-4, Panel A), CD69 (Figure 3-4, Panel B) and th e B cell co-stimulation marker B7.2 (Figure 3-4, Panel C) based on the change on percentage of positive cells. There were systemic variations on the geometric mean of MHC Cl ass II molecule, which might come from the operation difference on the staining intensity from day to day. Since each day studied included one mouse from each treatment group, we decided it would be more accurate to normalize the results to that da yÂ’s control mouse. This nor malization process reduced the effects of day-to-day variations in stai ning conditions, and allowed the effects of chlordecone and estradiol to be seen more clearly. After normalization, both 5 mg chlordecone and estradiol significantly increa sed the level of MHC class II based on the mean fluorescence intensity (Figure 3-4, Panel D). Chlordecone Exposure Enhanced Germinal Center Reactions Since the germinal center is the site of T cell-dependent antibody responses, and the site of production of most high-affinit y, class-switched autoan tibodies, we looked more closely at germinal center B cells. As a percentage of tota l CD19+ B cells, the 5 mg chlordecone-treated group showed signifi cant increase in B cells staining for GL7, a marker of germinal center B cells (Han et al ., 1996), as shown in Figure 3-5. This also corresponded to an increase in the expression of CXCR5high and CXCR4high B cells (Figure 3-6, Panels A and C). When gated specifically on GL7+ germinal center B cells, CXCR4 and CXCR5 expression level was also increased in the chlordecone-treated group (Figures 3-6, Panels B and D). These two chemokine receptors are important in


42 the organization of germinal centers and the tr afficking of cells within them (Allen et al., 2004). Comparable increases in germinal cente r B cells were also seen in the estradioltreated group. As there are also CD4 T cells in the germinal center (Cozine et al., 2005), and CXCR5 is the specific marker for germinal center T cells (Kim et al., 2001), we also measured its expression level. Both chlordecone and estradiol increased the percentage of CXCR5high CD4 T cells, as shown in Figure 3-7. Examination of Expression Profile of Gene s Involved in Germinal Center Reactions Expression levels of the anti-apoptotic protein Bc1-2, cell adhesion molecules ICAM-1 and VCAM-1, and the inhibitory receptor Fc RIIb in B cells were measured by flow cytometry. Estradiol treatment was pr eviously been shown to upregulate Bcl-2 expression on splenic B cells of R4A-IgG 2b BALB/c transgenic mice (Bynoe et al., 2000), but has not been shown in non-transgenic s. We found that in the NZB/NZW F1 mouse, not only estradiol but also chlordeco ne treatment increased Bcl-2 expression on splenic B cells (Figure 3-8). It has been suggested that the signals produced by adhesion of LFA-1 to ICAM-1 and VLA-4 to VCAM-1 were involved in the interaction of follicular dendritic cells and B cells in germ inal center, and prevented germinal center B cell apoptosis (Koopman et al., 1994). Chlo rdecone treatment increased both ICAM-1 and VCAM-1 expression level of total CD19+ B cells (Figure 3-9, Panel A and C) as well as germinal center B cells (Panel B and D). This effect was similar to that seen in estradiol-treated mice. In contrast, neither chlordecone nor estradiol treatment affected the expression of inhibitory receptor Fc RIIb in germinal center (Figure 3-10).


43 Exposure to Chlordecone and Estradiol D ecreased the Rates of Germinal Center Apoptosis Without Affecting the Proliferation of Lymphocytes In order to characterize further the beha vior of B lymphocytes after chlordecone exposure, B cells were isolat ed by negative selection with magnetic beads (Miltenyi) and stimulated under a variety of conditions, in cluding LPS and anti-IgM. Apoptosis was measured by flow cytometry using 7-AAD and annexin-V staining, and the proliferation was tested by tritiated thymidine incorporati on as described in Chapter 2. After 24 hr culture, the B cell apoptosis was signifi cantly reduced in 5 mg chlordeconeand estradiol-treated groups, under both the conditions of “no st imulation” or “1 µg/mL LPS stimulation” (Figure 3-11, Panel A and B). B cell apoptosis showed large variation under anti-IgM stimulated conditions, and the result s did not reach statistical significance among groups (data not shown). As isolated germinal center B cells die within a few hours in culture, it is not po ssible to measure the apoptosis of this particular B cell population after 24 hours. Therefore, freshl y isolated B cells were stained with a combination of antibodies to GL7 and CD19 to identify germinal center B cells, and immediately tested for apoptosis by annexi n-V staining. Both 5 mg chlordecone and estradiol treatment decreased apoptosis of germ inal center B cells significantly, as shown in Figure 3-11, Panel C. On the other hand, B cell proliferation, as measured by proliferation index showed no difference in e ither chlordecone or es tradiol-treated group compared with the control treat ment (Figure 3-11, Panel D). In Contrast to Estradiol, Chlordecon e Treatment Did Not Alter Splenic B Cell Subsets Although chlordecone did not have a marked effect on total splenic B cells, it was possible that it altered the di stribution of B cell subsets, one of the better documented effects of estradiol, at least in R4A-IgG 2b BALB/c transgenic mice (Grimaldi et al.,


44 2001). We therefore examined the impact of sustained exposure to chlordecone or estradiol on the distribution of splenic B cell subsets in NZB/NZW F1 mice. Flow cytometric analysis revealed a signifi cant decrease in the percentage of B220+/CD24high immature B cells and an increased percentage of B220+/CD24low-int mature B cells in estradiol-treated mice (Figure 3-12, Panel A), wh ich is similar to the results reported in BALB/c transgenic mice (Grimald i et al., 2001). In contrast to estradiol, chlordecone treatment did not cause a decrease in the im mature B cell population. The reduction of immature B cells by estradiol was due mainly to a decrease in the percentage of CD21low/CD24high transitional type 1 (T1) B cells (Figure 3-12, Pa nel B), which represent the most immature sple nic B cell population and gives rise to the tran sitional type 2 (T2) B cells. The CD21high/CD24high transitional T2 population, which differentiates into the mature splenic B cell population, was not changed in estradiol-treated mice (Figure 3-12, Panel C). Chlordecone treatment had no demonstrable influence on the T1 or T2 population. The increased percen tage of mature B cells by estradiol were a result of a significant increase in CD21highCD24low marginal zone B cells (Figure 3-12, Panel E), with a statistically insignificant contribution by CD21lowCD24low follicular B cells (Panel D). In contrast, chlordecone treatment had no effect on marginal zone or follicular B cells. Estradiol, But not Chlordecone, Increased the Mature Plasma Cell Population Plasma cells are the major cell type th at secretes antibodies. Therefore we examined chlordecone and estradiolÂ’s effects on producing plasma cells in spleen. Only estradiol treatment significantly increased B220-CD138+ mature plasma cell population, while chlordecone treatment caused a slight increase which did not reach statistical


45 significance (Figure 3-13, Panel A). On the other hand, neither chlordecone nor estradiol changed the B220+CD138+ immature plasma cell populati on (Figure 3-13, Panel B). Estradiol, but Not Chlordecone, Enhanced Anti-dsDNA and Anti-ssDNA Autoantibody Titers in Serum As only estradiol treatment caused a signifi cant increase in plasma cells, the major source of circulating IgG anti bodies, we further measured the serum level of anti-dsDNA and anti-ssDNA autoantibodies in the differe nt groups. Each group included 15 mice. Chlordecone treatment did not increase the titers of either au toantibody specificity compared with the control group. However, th ere were 5 mice in the estradiol-treated group that clearly broke tolerance at this ear ly stage and had signifi cantly higher titers of both auto-antibodies (Figure 3-14). The mean titers of both autoantibody specificities in the estradiol group were signi ficantly higher than either the chlordecone or control groups. Estradiol, but Not Chlordecone, Activated Splenic T Cells The direct effects of chlordecone and estr adiol on T cells were measured by their influence on the expression of activation ma rkers CD69, CD44 and CD62L. Estradiol, but not chlordecone treatment significantly enhanced the expression of the early activation marker CD69 on both CD4 (Figure 3-15, Panel A) and CD8 T cells (Figure 315, Panel B). Furthermore, the effects of chlordecone and estradiol on CD4 T cell subsets were tested. Estradiol significantly increased activated T cell (Figure 3-16, Panel A) and memory T cell population (Figure 316, Panel B), while it reduced the naive T cell population (Figure 3-16, Panel C). Chlo rdecone treatment did not cause these effects.


46 Estradiol, but Not Chlordecone, Increa sed the Regulatory T Cell Population Although negative selection of autoreactive T cells in th e thymus and induction of anergy in the periphery are two main m echanisms to maintain T cell tolerance (Sakaguchi, 2000; Van Parijs et al., 1998), these two mechanis ms alone are insufficient to preserve the tolerant state (Polanczy et al., 2004). Recent studies have revealed a crucial role of CD4+CD25+ regulatory T cells in supp ression of responses to self-antigens in both mice and humans (Shevach, 2000; Maloy et al., 2001; Bach, 2005). Failures in the function of the regulatory T cell compartment can be responsible fo r the development of autoimmune disease. We therefore examined chlordecone and estradiolÂ’s effect on this specific T cell population. Estradiol, but not chlordecone sign ificantly increased CD4+CD25+ regulatory T cell population, as sh own in Figure 3-17, Panel A), which was confirmed by using another regulatory T cell marker glucocorticoid-induced TNF receptor (GITR, Figure 3-17, Panel B). Estradiol Treatment Enhanced the Expression of T Cell Receptor V , but Not V Chains As T cell receptor revision represents a nother important mechanism for induction of T cell tolerance, we examined chlordecone and estradiolÂ’s effects on the expression of V and V Chains of T cell receptor. Estradio l treatment increased the expression of V 2 and V 8.3 chains (Figure 3-18, Panel A and B), but did not change the expression of V 4 and V 8 chains (Figure 3-18, Panel C and D). In contrast, chlordecone treatment showed no influence on either t ype of T cell receptor chains.


47 Chlordecone and Estradiol Treatments Enha nced Bcl-2 Expression on CD4 T Cells The expression of anti-apoptotic molecule Bcl-2 in CD4 T cells was measured. Both chlordecone and estradiol treatment si gnificantly increased it s expression, as shown in Figure 3-19. Chlordecone and Estradiol Treatments Reduced CD4 T Cell Apoptosis Without Affecting Proliferation To characterize the behavior of T lymphocytes after chlordecone and estradiol exposure, CD4 T cells were isolated by negative selection with magnetic beads (Miltenyi) and stimulated under a variety of conditions, including anti-CD3 and -CD28. Apoptosis and proliferation of CD4 T cells were measured as described in Chapter 2. Both chlordecone and estradiol treatment significantly reduced CD4 T cell apoptosis either under “unstimulated” conditions (Figur e 3-20, Panel A), or after stimulation with anti-CD3 and 5 µg/mL anti-CD28 (Panel B). Based on the proliferation index, there is no significant change in CD4 T cell proliferati on within the different treatment groups, as shown in Figure 3-20, Panel C. Chlordecone Increased the Percentage of Macrophage Population, but Not Dendritic Cell Population in Spleen Although macrophage and dendritic cell on ly account for a small portion of the total splenic cells, it is still possible that the reduction of lymphocyte percentage in estradiol-treated group comes from a dramatic increase in macrophage or dendritic cell. Splenic macrophage was recognized as CD11b+CD11ccell population. The percentage of macrophage was significantly increased in 5 mg Chlordecone-treated group, as shown in Figure 3-21, Panel A. Estradiol treatment also increased macrophage population, but the result did not reach statis tical significance. On the other hand, neither chlordecone


48 nor estradiol affected the percentage of all three types of dendritic cells, as shown in Figure 3-21, Panel B. Discussion Although our previous experience has s hown that chlordecone can markedly accelerate autoimmunity in the NZB/NZW F1 mouse model, the mechanisms by which it mediates these effects remain unknown, and th e question whether chlo rdecone acts on the immune system as an estrogen mimic is stil l a puzzle. The objective of the study was to clarify chlordeconeÂ’s effects on several impor tant cellular events in the autoimmunity process. The dose of estradiol (0.05 mg pellet) in this test was decided by a series of dosedependent preliminary tests, since Walker et al. (1992) reported that a high dose of 17 estradiol can cause premature death with bladder outlet obstruction and serious hyperprolactinemia in the NZB/NZW F1 mous e. In our preliminary tests, we found similar effects. The starting dose was 1.8 mg per 60-day continuously released estradiol pellet, followed with a series reduced dos es of 1.5 mg, 1.0 mg and 0.05 mg, and there were 10 mice tested at each dose. Only when the estradiol dose wa s reduced to 0.05 mg per pellet, treated mice did not show serious genitourinary complications with clear loss of weight and early death. Subsequent tests showed that at this dose, estradiol clearly accelerated lupus development in this mouse mo del. The doses of chlordecone tested in this study were also based on studies of dose -response relationships. Sobel et al. (2005) reported that the lowest chlordecone dose per 60-day release pe llet found to produce a significant decrease in time to onset of renal diseas e was 0.5 mg (0.20 mg/kg/day), and the no observable effect level (NOEL) of ch lordecone in this m ouse model was 0.1 mg per pellet (0.04 mg/kg/day). Previous studies also showed that at 1 mg and 5 mg pellet


49 dose, chlordecone treated mice showed a s horter time in developing renal disease and mortality than sham-operated (ovary inta ct) mice. Follow-up study on serum IgG antidsDNA titers suggested 1 mg and 5 mg chlo rdecone pellet-treated mice had similar elevated level of autoantibodies as in the sham group, and immunofluorescence staining for IgG showed clear immune complex deposit ion in kidney in both 1 mg chlordeconetreated and 0.05 mg estradiol-tr eated mice. The staining in tensity was about the same between these two groups, which was stronger than in the age-controlled group (Sobel et al., 2005). Based on these results , we proposed that doses of chlordecone between 1 mg to 5 mg might affect the process of lu pus development at a similar level. At the beginning of the experiment, all mice weighed 28 ± 3 g before implantation which is a normal weight range for two-m onth old NZB/NZW F1 mi ce. Six weeks after implantation, all groups of mice gained wei ght, suggesting all mice tolerated the process and no treatment caused serious toxicity. Ho wever, the estradiol treated group showed a significantly smaller increase in body weight than the chlo rdecone-treated groups which is slightly lower than contro l. This indicates other adve rse effects of estradiol beyond those on the immune system might be occurri ng. After treatment, es tradiol significantly increased uterus weight about six times higher than the pl acebo-treated group, which is consistent with previous reports, while chlo rdecone-treated groups s howed a slight dosedependent increase. Statistical analysis s howed about 7% and 35% increase of uterine weight by 1 mg and 5 mg chlordecone pellet s respectively, and the increase in uterus weight is linear with dose. These data suggest that limited estrogeni c effects are seen in the reproductive organs at the doses of ch lordecone tested. Estradiol treatment significantly increased the spl een weight about 49% (p<0.01) compared with the control


50 group, and 5 mg chlordecone treated group show ed a weak estrogenic effect with a 13% (p<0.01) increase. Interestingly, 1 mg chlo rdecone treatment only caused about a 1% increase in spleen weight. Although this dos e did not increase the spleen weight, there are changes that clearly happened on the cellu lar levels that will be discussed later. Although chlordecone treatment slightly re duced the percentage of lymphocyte populations, the total decrease of 20% for both B and T cells in the estradiol group is very striking. We have tested the percentages of other cell types in cluding macrophages and dendritic cells (Figure 3-21), which did not show significant changes by estradiol treatment. One possibility is that estrad iol reduced the expression level of lymphocyte markers CD19, CD4 and CD8 instead of reduc ing the population. However, the answer to the question might not be that simple, as Masuzawa et al. (1994) have reported that ovariectomy enhanced B cell populations in bone marrow. Since we did not have a control group without ov ariectomy, it is hard to tell whet her there is an increase in the splenic B cell population caused purely by ovariectomy. Even had we included this group, it is likely that variation due to the estr us cycle would obscure effects. There are also reports from the studies of pregnant women and mice th at high level of estrogen can suppress B lymphopoiesis. Estradiol treatmen t might cause an increase in non-T and non-B cell populations, but the difference might not be as large as 20% observed here. Rijhsinghani et al. (1996) repor ted that estradiol treatment reduced CD4 and CD8 T cells in the thymus. Interestingly, Okasha et al. (2001) reported th at estradiol treatment could reduce the percentage of CD4+CD8+ (double positive) T cells and an increase in the percentage of CD4CD8 (double negative), CD4+ and CD8+ T cells, also in the thymus.


51 The difference between their experiment a nd ours is that they did a short term experiment, where cells were harvested on da ys 1, 3 and 7 after a single injection. Both chlordecone and estradiol enha nced B cell markers CD44, CD69, B7.2 and MHC class II molecule expression, suggesting an enhanced activation status of B cells. These markers reflect different aspects of B cell activation. CD44 is an integral, hyaluronan-binding receptor present on the su rface of lymphocytes and other cells of hematopoietic and nonhematopoietic origin (B ajorath, 2000). CD44 might play a role in the maturation of early lymphocyte precursors and in adaptive immune responses (Miyake et al., 1990). B cell maturation is re lated to the interac tion of CD44 with its stroma cell ligand fibronectin, as antibodies against CD44 impair the survival and differentiation of surface Ig-positive B cells and other hematopoietic progenitors in longterm murine bone marrow cultures (Miyake et al., 1990). Enhanced CD44 expression level on B cells might therefore be responsib le for speeded B cell maturation. Actually, estradiol treatment did significantly increase the percentage of matu re B cells, and reduce the immature B cell population (Figure 3-12). Increased expression of CD44 proteins is also the characteristic of T cell activation after encounter with its cognate antigen (DeGrendele et al., 1997). Another function of CD44 on lymphocytes is to traffick the lymphocytes to extravasate into sites of inflammation, by mediating the adhesion of lymphocytes to vascular endothe lial cells via binding of hya luronic acid (DeGrendele et al., 1997; Estess et al., 1998). We did not examine the CD44 expression in bone marrow cells, it is unknown whether chlo rdecone or estradiol had any effects on this function. CD69 is a leukocyte receptor tran siently induced after activati on, and it is usually used as an early lymphocyte activation marker. B7.2 is a B cell co-stimulation marker. The


52 traditional view of the role of B7.2 is in regulating T cell activation and tolerance with its pathway in the B7:CD28 family. Recent emer ging data indicate that B7.2 on B cells can signal bidirectionally. B7.2 stimulates CD28 on T cells and transduces positive signals into B cells that increase IgG1 and IgE pr oduction (Podojil et al., 2003, 2004). The class II molecule is very important in antigen pres entation to B cells and to T cell activation. Antigen recognized by a B cell's immunoglobulin is bound, internalize d, and presented in MHC Class II to a T cell receptor during B ce ll activation. Increased class II molecule expression mediates enhanced antigen presentin g signal. Grimaldi et al. (2001) reported that in an immunoglobulin transgenic BALB/c mouse model, estradiol did not increase the expression of MHC class II molecule. We found about 15% increase of MHC class II expression in both 5 mg chlordecone and estr adiol treated groups, and the increases are statistically significant. The difference mi ght be due to the difference between mouse strains. Both chlordecone and estradiol enhanced germinal center re actions, as shown by upregulated protein expre ssion of GL7, CXCR5, CXCR4, ICAM-1, VCAM. Although germinal center cells account for a small per centage of the total sp leen cell population, it is an important area where negative selection for autoreactive lymphocytes occurs. We also found both elevated Bcl-2 expression a nd reduced germinal center and total B cell apoptosis. Enhanced Bcl-2 expression helps to reduce apoptosis and increase survival of autoreactive cells (Kuo et al., 1999). These are consistent results. Taken together, chlordecone and estradiol might help au toreactive B escape ne gative selection by influencing one or more check points of the germinal center reactions. This is also


53 consistent with the finding of reduced germ inal center B cell apopt osis by chlordecone and estradiol treatments. The fate of B cells in the germinal cent er is associated with some important molecules. Previous reports showed the adhesion molecules ICAM-1 and VCAM-1 can mediate attachment of germinal center B cells to the follicular dendritic cells in vitro via the integrins LFA-1 and VLA-4. This association of B cells and follicular dendritic cells is deemed important for driving affinity ma turation (Hedman et al. 1992; Koopman et al. 1994). The enhanced expression of cell adhesion molecules ICAM-1 and VCAM by chlordecone and estradiol treatments might transfer a stronger signal from follicular dendritic cells to the B cells, which post pones the apoptosis of autoreactive B cells (Koopman et al., 1994). Chemokine receptor CXCR5 is an important marker for the germinal center. Forster et al. (1996) repo rted that in mice with targeted disruption of the CXCR5 gene, inguinal lymph nodes are absent and Peyer’s patches are either aborted or display severe morphological alterations with no defined B cell follicles. Also the spleen lacks follicles and exhibits impaired germinal center formation after immunization of these animals. Consistently, transferred lymphocytes from CXCR5-deficient mice fail to enter B cell follicles in the organs of wild-type anim als, demonstrating an essential role for CXCR5 in follicular homing. The chemokine receptor CXCR4 is expressed by all subsets of B cells throughout B cell ontogeny, and was proved to be importa nt in regulating homeostasis of B cell compartmentalization and humoral immunity (Nie et al., 2004). Allen et al. (2004) reported that chemokine CXCL12 and CXCL13 and their receptors CXCR4 and CXCR5 are crucial for the motility of germinal center cells – the germinal center organization


54 depends on sorting of centroblasts by CXCR4 into the dark zone. In contrast, CXCR5 helped direct cells to the light zone, and deficiency in CXCL13 was associated with aberrant light zone localization. The in creased expression of CXCR4 and CXCR5 by chlordecone and estradiol treatments might sp eed the movement of autoreactive cells and help them escape from negative se lection in the light zone. Chlordecone and estradiol treatments caused a reduction in total B cell apoptosis, which is consistent with enhanced expressi on of anti-apoptotic protein Bcl-2. Normally, reduced apoptosis is connected with increa sed number of cells. However, estradiol treatment caused a 7% decrease in splenic B cells, which seems inconsistent with reduced apoptosis. But, considering th e significant increase in spl een weight (48 %) by estradiol treatment, and presuming the number of total cells is proportion to the spleen weight, the total number of B cells was actually increased. Estradiol treatment has been reported to enhance marginal zone B cell populations significantly in a transgenic BALB/c mouse model, and the autoreactive B cells displayed a marginal zone phenotype (Grimaldi et al., 2001 ). We verified the finding of enhanced marginal zone B cell population by estradiol treatment in NZB/NZW F1 mouse model. However, chlordecone did not change the pe rcentage of marginal zone B cells, which might suggest chlordeconeÂ’s effects on au toimmunity is not through its effects on marginal zone B cells. However, it is st ill possible that chlo rdecone changed the composition of marginal zone B cells with an increase in autor eactive B cells without affecting the overall percentage. Unfort unately, there is no way to measure the autoreactive B cells in the conventional NZB/NZW F1 mous e model. This can be overcome by using either the R4A2b transgenic BALB/c mouse model or NZB/NZW


55 F1 transgene 3H9-µ or 3H9R mouse models. In these strains, there will be specific antibody to the transgene which can quantitati vely measure the autoreactive B cells. Estradiol but not chlordecone enhanced the percentage of mature plasma B cells in spleen, which is defined as the B220-CD138+ cell population. When immature plasma cells (B220+CD138+) were examined, neither chlordecone nor estradiol influenced this population. Chlordecone did not show enhanced presence of plasma cells, which might be due to the low dose we used in the experi ment and the short period time of treatment. Another possibility arises from the repor t that after antigen challenge and B cell activation, the emerging plasma cells undergo CXCL12-induced chemotaxis to the bone marrow, where they produce antibody and pers ists (Erickson et al. 2003). NZB/NZW F1 mice have a defect in migration of plasma cells to the bone marrow (Erickson et al., 2003), and it is possible that chlordecone partiall y reversed this effect . We did not stain for plasma cells in the bone marrow, but we be lieve this possibility is less likely. Plasma cells are also the maor source of secreted autoantibodies. We th erefore measured the serum level of two types of auto-antibodies, IgG anti-dsDNA and anti-ssDNA. Only the estradiol-treated group showed a significant increase in both anti body titers, which was consistent with the increase in plasma cell population. Chlordecone did not break tolerance at th e early time point when the mice were sacrificed, while 1/3 of the mice in estradio l-treated group broke the tolerance. We had chosen this early time point precisely to re duce the potential of identifying secondary effects, as our previous experience indi cated that chlordecone did not increase autoantibody titers 5-6 weeks af ter implantation of young mice (S obel et al, in press). However, this also might be an important reason why chlordecone showed weaker effects


56 than estradiol in many of the parameters ev aluated. We confirmed that chlordecone did not break the tolerance in ova riectomized mice. On the ot her hand, 0.05 mg estradiol did show some effects in increasing the tit ers of autoantibodies. Despite this, immunofluorescence staining s howed similar intensities of IgG immune-complex deposition at kidney in both chlordecone a nd estradiol treated gr oup eight weeks after implantation (Sobel et al., 2005). There are at least two possibilities for the difference: 1. NZB/NZW F1 mouse is a lupus-prone mouse model, and the disease process in this model is very sensitive to the time course . Chlordecone treatment did not break the tolerance at 6 weeks, but it might break the tolerance shortly thereafter. In unovariectomized NZB/NZW F1 mice, the titers of auto-antibodies were clearly higher at 8 weeks than at 6 weeks after implantion (Sobel et al., in press). 2. Although the absolute titers of serum antibodies were the same or even lower in chlordecone-treated group, chlordecone might have some subtle effects on the kidneys that cause more deposition in kidney than estradiol treatment. Although both chlordecone and estradiol activate B cells, it seems that only estradiol had a significant effect on T cells. Chlordecone had no influence on the early activation marker CD69 expression in both CD4 and CD8 T cells (Figure 3-15, Panel A and B), and caused a slight increase in activated CD4 T cell subset based on the expression level of CD44 and CD62L (Figure 3-16, Panel A). This might be due to the low dose of chlordecone in the experiment and early course of the disease. It is also possible that chlordecone might act on T ce lls secondarily. Based on CD44 and CD62L expression, estradiol but not chlordecone si gnificantly increased activated T cell and memory T cell percentage and reduced naive T cell percentage, sugge sting estradiol can


57 also influence T cell maturation, while chlordeco ne does not have this effect. In humans, Kamada et al. (2000) reported that ho rmone replacement therapy (HRT) caused a significant decrease in naive T cells and an increase in memory/activated T cells in women at late post-menopause than at early post-menopause. Okasha et al. (2001) also reported that estradiol treatment could in crease CD44 expression on thymocytes, which was accompanied with increased maturation of the thymocytes. Estradiol but not chlordecone in creased the percentage of CD25+CD4+ regulatory CD4 T cell population which has suppressive e ffects in the development of autoimmune disease (Maloy et al., 2001; Shevach, 2000). Th is seemed inconsistent with the finding of accelerated autoimmunity in estradioltreated mice. However, in considering estradiolÂ’s effect of down-regulating the per centage of CD4 T cell, the absolute number of total regulatory CD4 T cells was likely not significantly increased in estradiol-treated group. As we did not count the absolute number of CD4 T cells in the spleen, it is hard to reach a conclusion right now. A nother possibility is that, since the suppressive effects of CD25+CD4+ regulatory T cell requires direct cell-cel l interaction (Feher vari et al., 2004), estradiol might prevent this interaction by down-regulating some mo lecules on the target cells. Interestingly, Polanczyk et al. 2004 reported that estradiol treatment alone sufficiently expanded CD25+CD4+ regulatory T cell compartment in C57BL/6 mice, and the increase might be responsible for estrad iolÂ’s protection of mice from experimental autoimmune encephalomyelitis (EAE), a mous e model of human multiple sclerosis (Bebo et al., 2001). The role of CD25+CD4+ regulatory T cell in differe nt autoimmune diseases is far from being elucidated, and further stud ies are required to clarify its function in lupus development.


58 Estradiol treatment caused a small but statistically significant increase in the expression of T cell receptor V 2 and V 8.3 chains, but had no influence on V 4 and V 8 chains. It is unknown whet her this small increase in V chains is physiologically meaningful. In general, receptor editing allo ws T cell to move away from autoreactivity, and would induce tolerogenic mechanism. Fr om this point of view, the factor that chlordecone did not caused any receptor editi ng, was consistent with its capability to induce autoimmunity. Both chlordecone and estradiol reduced CD4 T cell apoptosis after culture for 24 hours in our experiments, this is consiste nt with the increased expression of antiapoptotic molecule Bcl-2. However, there were reports that estradio l can increase T cell apoptosis (McMurray et al., 2001 ; Do et al., 2002; Okasha et al., 2001; Jenkins et al., 2001). There are two major differences betw een their experiments and ours. First, McMurray et al. and Jenkins et al. used cell lines and treated with a significantly higher dose of estradiol; the physiologi cal connection between the do se and increased apoptosis is therefore not clear. Do et al. used a different strain of mouse. Second, in their experiment, they did a short-term exposure of estradiol either to cells or to mice, while we exposed estradiol to mice for six weeks, dur ing which estradiol might indirectly affect T cell apoptosis by its metabolites. It is al so possible that those pro-apoptotic T cells were already negatively selected during this time period, and only the T cells resistant to apoptosis were retained. The difference of estradiol on T cell a poptosis between shortterm and long-term treatment might be linked with different effect s on the kinetics of ERK phosphorylation and the length of tim e that phospho-ERKs are retained in the nucleus. Short time ERK phosphorylaton might be responsible for the anti-apoptotic


59 effects. Chen et al. (2005) reported that conversion of su stained ERK phosphorylation to transient, by means of cholera toxi n-induced activation of the adenylate cyclase/cAMP/protein kinase A pathway, can abrogate the pro-a poptotic effects of estradiol. Taken together, our results suggest that chlordecone alters susceptibility to autoimmunity through a number of subtle effects predominantly on B cells. The net effect of these perturbations is to enhan ce the germinal center reactions and reduce the autoreactive B and CD4 T cell apoptosis withou t affecting their pro liferation. On the other hand, estradiolÂ’s effects were more comp licated. Beyond the characters that shared with chlordecone, estradiol also changed B cell subsets with a significantly enhanced marginal zone population; it produced more pl asma cells and broke tolerance earlier than chlordecone; and it showed stronger e ffects on T cell activation and maturation.


60 Table 3-1. Introduction of phenotypic markers. Molecule name Cellular expression Main function CD69 Activated leukocyt e, natural killer cell, lymphocyte, platelets, Langerhans Early activation marker, signal tranmitting receptor CD44 Hematopoietic and nonhematopoietic cells, except platelets Activation marker, leukocyte and lymphocyte rolling, homing and aggregation CD86 (B7-2) Monocyte, activated B and T cell, dendritic cell, endothelial cell Costimulation of T cell activation and proliferation GL7 Germinal center B and T cell B and T cell activation antigen CD184 (CXCR4) T cell subset, B cell, natural killer cell, monocyte/macrophage, proliferating endothelial cell Cell migration, hemoto progenitor cell homing CD185 (CXCR5) T cell subset, B cell, cerebellar neurons Cell migration CD54 (ICAM-1) Endothelia cell, epithelia cell, monocyte, low on resting lymphocyte and upregualted upon acitvation Extravasation of leukocytes from blood vessels, regulation of T cell activation CD106 (VCAM-1) Activated endothelial cell, follicular dendritic cell, B cell Leukocyte adhesion, transmigration and costimulation of T cells CD24 B cell, granulocyt e, epithelia cell, monocyte Regulation of B cell proliferation and differentiation CD21 Mature B cell, follicular dendritic cell, T cell subset Signal tranduction CD138 Plasma cell, pre-B cell, epithelia cell, neutrophil Extracellular matrix receptor CD62L B cell, T cell subset, monocyte, granulocyte, natu ral killer cell, thymocyte Leukocyte and lymphocyte rolling and homing CD25 Activated T, B cell and monocyte, dendritic cell subs et, T regulatory cell IL-2 receptor chain, associate with and chains to form high affinity IL-2 receptor


61 Figure 3-1. Transverse section of spleen white pulp. The outside area of the white pulp is marginal zone, which sperates the white pulp from red pulp. It is primarily made of non-circulating marginal zone B cells, specialized macrophages, and reticular cells. Marginal zone B cells participate in early immune responses, and has lower threshold than recircula ting or immature B cells for activation, proliferation and differentiation into antibody-secreting cells. Germinal centers are mircroenvironment where rapid B-cell expansion, V(D)J somatic hypermutation (SHM), isotype switchi ng, affinity maturation, apoptosis, plasma cell commitment and memory cell formation occur. B cells, T helper cells and follicular dendritic cells are th e main cell types in germinal center.


62 A. Body weight before implantation B. Body weight after implantation C. Spleen weight after implantation D. Uterus weight after implantation Figure 3-2. Chlordecone and es tradiol effects on mice body, spl een and uterus weight. (A) The mean mouse body weight before implantation. (B) The mean mouse body weight six weeks after implantation. Mice-treated with estradiol gained significantly less body weight (p<0.01) , while chlordecone-treated groups were unchanged in body weight compared to the control grou p. (C) The mean mouse spleen weight six weeks after implantation. It was significantly increased by estradiol (p<0.01) and 5 mg chlordecone (p<0.01) treatment, while 1 mg chlordecone had no effect. (D) The mean mouse uterus weight six weeks after implantation. Estradiol treatment significantly increased the uterus weight about seven times (p< 0.01) compared with the control group, while chlordecone treatment did not show this effect.


63 A. CD19+ B cells B. CD4+ T cells C. CD8+ T cells Figure 3-3. Chlordecone and estradiol e ffects on splenic lymphocyte populations. Estradiol treatment significantly decreased the percentage of (A) B cells, (B) CD4 T cells and (C) CD8 T cells in th e spleen, and the decreases were all significant (p<0.01). While chlordecone did not show these effects.


64 A. CD44 expression on B cells B. CD69 expression on B cells C. B7.2 expression on B cells D. MHC Class II expression on B cells Figure 3-4. Analysis of chlordecone and es tradiol effects on B cell activation markers CD44 (A), CD69 (B), co-stimulation mark er B7.2 (C) and class II molecule IA(d) (D) expression on splenic B cells. Splenocytes were analyzed by flow cytometry using antibodies that recognize CD19, CD44, CD69, B7.2 and IA(d) (MHC Class II molecule). The do tted line represents the value of the control group in (A), (C) and (D). The methods for the flow cytometry analysis and statistical analysis were described in chapter 2. The expression of MHC Class II molecule was based on the mean fluorescence intensity, and was normalized to each dayÂ’s control. Estradiol and 5 mg chlordecone treatment significantly increased the e xpression of all these molecules on B cells compared with control group.


65 A. Control B. 5 mg chlordecone C. 0.05 mg Estradiol D. GL7 expression on B cells Figure 3-5. Enlarged germinal center in bot h chlordecone and estradiol-treated mice. Splenocytes were stained for the B cell marker CD19 and the germinal center marker GL7, and the level of GL7 expression on CD19+ B cells was determined by flow cytometry. Representative flow cytometry figures of GL7 expression by splenic B cells are shown here for a control-treated mouse (A), a 5 mg chlordecone-treated mouse (B) a nd an estradiol-treated mouse (C). The overall results are shown in Pane l D. Both 5 mg chlordecone and estradiol treatment significantly increased GL7 expression on B cells (p<0.01). CD19-PE-Cy7010110210310410 GL7-FITC010110210310410 3.64% 96.36% 0.00% 0.00% CD19-PE-Cy7010110210310410 GL7-FITC010110210310410 3.05% 96.95% 0.00% 0.00% 3.05% 96.95% 0.00% 0.00% CD19-PE-Cy7010110210310410 GL7-FITC010110210310410 1.53% 98.47% 0.00% 0.00%


66 A. CXCR5 expression on B cells B. CXCR5 expression on GC B cells C. CXCR4 expression on B cells D. CXCR4 expression on GC B cells Figure 3-6. Analysis of chlordecone and estrad iol effects on the expression of chemokine receptors CXCR4 and CXCR5 on total B cells and germinal center B cells from the spleen. The dotte d line represents the valu e of the control group in (B) and (D). Estradiol a nd 5 mg chlordecone treatm ent significantly increased the expression of CXCR5 on the total B cells (A) and the germinal center B cells (B). The expression level of CXCR4 on total B cells (C) and the germinal center B cells (D) were also increased significantly.


67 Figure 3-7. Analysis of chlordecone and estrad iol effects on the expression of chemokine receptors CXCR5 in CD4 T cells from the spleen. Both 5 mg chlordecone and estradiol treatment significantly increased CXCR5 expression in CD4 T cells.


68 Figure 3-8. Analysis of chlordecone and estrad iol effects on the expression of Bcl-2 in B cells. The dotted line represents the valu e of the control grou p. Both estradiol and chlordecone significan tly enhanced Bcl-2 expre ssion in B cells compared with control group. However, there wa s no statistically significant difference among chlordecone and estradiol treatments.


69 A. ICAM-1 expression on B cells B. ICAM-1 expression on GC B cells C. VCAM-1 expression on B cells D. VCAM -1 expression on GC B cells Figure 3-9. Analysis of chlordecone and estr adiol effects on the expression of the cell adhesion molecules ICAM-1 and VCAM-1 in the total B cells and the germinal center B cells. The dotted line represents the value of the control group. Both chlordecone and estradiol treatment significantly increased the expression of ICAM-1 in the total B cells (A) and the germinal center B cells (B) compared with control group. The e xpression of VCAM-1 in the total B cells (C) and the germinal center B cells (D) was also significantly increased compared with control group.


70 Figure 3-10. Analysis of chlordecone and es tradiol effects on the expression of the inhibitory receptor Fc IIb in the germinal center B cells. The dotted line represents the value of th e control group. Neither chlordecone nor estradiol treatment significantly changed the expression of Fc RIIb in germinal center B cells.


71 A. B cell apoptosis without stimulation B. B cell apoptosis with 1 g/mL LPS C. Germinal center B cell apoptosis D. B cell proliferation index with 1 g/mL LPS Figure 3-11. Analysis of B cell apoptosis and proliferation from chlordecone and estradiol-treated mice. The dotted line represents the value of the control group in (C). Both chlordecone and es tradiol treatment reduced total B cell apoptosis under conditions of no stimula tion (A) or stimulation with 1 g/mL LPS (B). Germinal center B cell apopto sis (C) was also reduced significantly by 5 mg chlordecone and es tradiol treatment compared with control group. However, there was no statistically significant difference among chlordecone and estradiol treated groups. For B cell proliferation, there were no differences in the proliferat ion index among the groups (D).


72 A. Immature B cell (B220+CD24+) B. T1 B cells (CD21-CD24+) C. T2 B cells (CD21+CD24+) D. Follicular B cells (CD21-CD24-) E. Marginal zone B cells (CD21+CD24-) Figure 3-12. Analysis of chlordecone and estradiol effects on B cell subsets. (A) Estradiol treatment significantly de creased the percentage of B220+CD24+ immature B cells (p<0.01). The decreas ed percentage of immature B cells came from a decrease in transitional type 1 B cells (B), but not the transitional type 2 B cells (C). The increased per centage of mature B cells in estradiol treatment was due to an increase in marginal zone B cells (E), but not follicular B cells (D).


73 A. Mature plasma cells (B220-CD138+) B. Immature plasma cells (B220+CD138+) Figure 3-13. Analysis of chlordecone and estr adiol effects on the percentage of mature plasma cells in spleen. (A) Estradio l treatment significantly enhanced the percentage of mature plasma cells (p <0.01). Chlordecone treatment did not show this effect. (B) Neither chlo rdecone nor estradiol affected the percentage of immature plasma cells.


74 A. dsDNA titers B. ssDNA titers Figure 3-14. Analysis of chlordecone and es tradiol effects on the serum titers of antidsDNA and anti-ssDNA antibodies. Six w eeks after implantation, five out of 15 mice in estradiol-treated group broke to lerance, and the mean values of the titers of anti-dsDNA (A) and anti-ssDNA antibodies (B) in estradiol-treated group were significantly higher than control group. While no mice in chlordecone-treated group broke tolerance.


75 A. CD69 expression on CD4 T cells B. CD69 expression on CD8 T cells Figure 3-15. Analysis of chlordecone and es tradiol effects on the expression of the activation marker CD69 on CD4 and CD 8 T cells. Estradiol, but not chlordecone significantly increased expression of CD69 in CD4 T cells (p<0.01) (A) and CD8 T cells (p<0.05) (B ). Estradiol, but not chlordecone significantly increased CD44+CD62L+ activated CD4 (p<0.05) (C) and CD44+CD62Lmemory CD4 T cells (p<0.01) (D), while decreasing CD44-CD62L+ naive CD4 T cells (p<0.01) (E).


76 A. Activated CD4 T cells (CD44+CD62L+) B. Memory CD4 T cells (CD44+CD62L-) C. Naive CD4 T cells (CD44-CD62L+) Figure 3-16. Analysis of chlordecone and estr adiol effects in the CD4 T cell subsets. Estradiol, but not chlordecone significantly increased CD44+CD62L+ activated CD4 (p<0.05) (A) and CD44+CD62Lmemory CD4 T cells (p<0.01) (B), while decreasing CD44-CD62L+ naive CD4 T cells (p<0.01) (C).


77 A. CD25 expression on CD4 T cells B. GITR expression on CD4 T cells Figure 3-17. Analysis of chlordecone and es tradiol effects in the regulatory CD4 T cell population. Estradiol, bu t not chlordecone, signif icantly increased the percentage of CD25+CD4+ T cell (A) and GITR+CD4+ T cell population (B).


78 A. TCR V 2 expression on CD4 T cells B. TCR V 8.3 expression on CD4 T cells C. TCR V 4 expression on CD4 T cells D. TCR V 8 expression on CD4 T cells Figure 3-18. Analysis of chlordecone and estrad iol effects on the T cell receptor revision. Estradiol, but not chlordecone, caused a statistically significant increase the percent of CD4 T cells expressing TCR V 2 (p<0.05) (A) or TCR V 8.3 (p<0.05) (B). Neither chlordecone nor estradiol affected the expression of TCR V 4 (C) or TCR V 4 (D) on CD4 T cells.


79 Figure 3-19. Analysis of chlordecone and estr adiol effects on the expression of Bcl-2 in CD4 T cells. The dotted line represents the value of the control group. Both estradiol and chlordecone significantly increased Bcl-2 expression in CD4 T cells compared with control group. However, there was no statistically significant difference among chlordec one and estradiol treatments.


80 A. CD4 T cell apoptosis (no stimulation) B. CD4 T cell apoptosis (CD3 and CD28) C. CD4 T cell proliferation index (CD3 and CD28) Figure 3-20. Analysis of chlordecone and es tradiol effects in CD 4 T cell apoptosis and proliferation. Both 5 mg chlordeconeand estradiol-treated groups showed a statistically significant reduction in CD 4 T cell apoptosis under both culture conditions (A and B). For CD4 T cell pr oliferation, there was no difference in thymidine incorporation under unstimulate d condition (data not shown). The proliferation index among groups was not changed either (C).


81 A. Macrophage population B. Dendritic cell population Figure 3-21. Analysis of chlordecone and es tradiol effects on macrophage and dendritic cell populations. Mice in the 5 mg ch lordecone treatment group showed a statistically significant in crease in the percentage of splenic macrophage (p<0.05) (A), while estradio l treatment did not show this effect. On the other hand, neither chlordecone nor estradio l affected the total dendritic cell populations (B).


82 CHAPTER 4 COMPARISION OF SPLENIC B CELL GENE EXPRESSION AND CD4 T CELL CYTOKINE SECRETION BY CHLORDECONE AND ESTROGEN TREATMENT Introduction B cell plays a significant role in the l upus process which produces autoantibodies and causes organ damage. Examining important genes expressed by B cells and known to be involved in lupus pathogenesis should be an effective way to clarify mechanisms and identify differences between chlordec one and estradiol treatment. Despite the genetic complexity of SLE a nd identification of more than 50 genes that can affect disease expression in the mouse, disease-asso ciated genes can generally be categorized into three groups. One group affects primarily clearance of apoptotic cells or immune complexes by macrophages, and is represented by targeted deletions of SAP, CRP, C1q and c Mer. Presumably, the inadequate clearance of apoptotic cells allows self-antigen to be presented in an immunogenic fashion. A s econd general category is that of defective regulation of apoptosis of cells of the imm une system, as represented by targeted or spontaneous deletions of Fas, Fas ligand, P TEN, and Bcl-x. With these defects, the ability of the immune system to clear auto reactive T and B cells is compromised. The third general pathway is represented by gene s whose deletion alters the threshold of signaling and includes CD19, Lyn, Fyn, Cr2, PtP1 c, CD22, Shp-1 and a host of others. Altering the signaling threshold allows au toreactive cells to escape tolerance mechanisms. Since splenic B cells are not di rectly involved in the pathway of apoptotic cell clearance, the firs t group of genes was not selected for testing. Several important


83 genes belonging to the second and third cate gories were selected and their expression level was tested by quantitative real-time PCR using mRNA extracted from B cells isolated using magenetic beads. Another important criterion for gene selection was that the regulation of these genes by estradiol has already been studied previously either by in vivo or in vitro tests, which sets a good control for comparing the effects between chlordecone and estradiol. Selected Genes The following genes, including Shp-1, Bcl-2, FAS, FAS ligand, IFN, Fc RIIb, TGF1, IL-2, IL-6, Ly5, TNF, were selected. Protein tyrosine phosphatase Shp-1 contains an SH2 doma in, and is utilized most widely in inhibitory receptor signaling pa thways (Plas et al., 1998), although it also associates with BCR, FcR, growth factor, comp lement and cytokine receptors (Bolland et al., 1999). It is activated after binding through its amino-terminal SH2 domains to phosphorylated immunoreceptor tyrosine-based inhibitory motifs (ITIMs) present on many receptors, and dephosphorylates the downstream proteins, and subsequently either terminates the activated signal or activates ot her pathways, such as apoptosis (Plas et al., 1998). Shp-1 deficient mice have reduced numbe rs of B cells but those that are present are activated (Shultz et al., 1997), and the mi ce develop severe autoimmune disease with immune complex deposition in skin, l ung and kidney (Shultz et al., 1987). Inhibitory receptor Fc R b is the only IgG Fc receptor on B cells and acts to set thresholds for B-cell activation upon co-cross linking with surface B ce ll antigen receptor (BCR), a mechanism whereby immune comple xs (ICs) can suppre ss the production of antibody (Jiang et al., 2000). Co-lig ation of the BCR complex with Fc R b promotes the induction of B cell growth ar rest in the G1 phase of the cell cycle, and this ultimately


84 results in commitment of such B cells to apoptosis. In addition to such feedback inhibition of ongoing B cell responses, ligation of Fc R b by immune complexes can induce B cell anergy and/or apoptosis, and he nce suppress aberrant B cell activation and potential autoimmunity (Pearse et al ., 1999). These characteristics of Fc R b might contribute to the maintenance of peripheral tolerance during the affinity maturation of B cell, by promoting the deletion of low-affinit y, self-reactive lymphocyt es (Dijstelbloem et al., 2001). Two strands of evid ence point to a role for Fc R b in the development of spontaneous autoimmune disease. First, the Fc R b knockout mouse had hyperactive B cells and developed SLE, dem onstrating that defects in Fc R b have potential to cause autoimmunity (Bolland et al., 2000). Second, a Fc R b defect occurs in all mouse models of SLE, such as NZB, BXSB, MRL (e xcept NZW) (Pritchard et al., 2003), with a deletion polymorphism in the Fcgr2b prom oter region among these mouse strains. Furthermore, Fc R b was reported to be down-regulat ed on the follicular germinal center B cell of NZB mouse strain, while it remained unchanged on the non-germinal center B cells (Jiang et al., 1999). This sp ecial down-regulation in germinal center B cells might be related to the escaping of autoreactive germinal center B cells from the negative selection (Jiang et al., 1999; Xiu et al., 2002). On the other hand, the NZW mouse has a normal expression of Fc R b on the germinal center B cell. The cross strain NZB/NZW F1 mouse shows an in termediate expression level of Fc R b between them. The Bcl-2 family of proteins has been shown to protect lymphocytes against apoptosis triggered by a variety of stimuli, in cluding growth factor removal, treatment with corticosteroids, and fas receptor activ ation (Cory et al., 1995). Bcl-2 has been


85 shown to increase survival of autoreactive B cells arising in the bone marrow, permitting them to populate peripheral lymphoid stru cture (Liang et al., 1997). In the R4A2b/bcl2 double-transgenic mouse, B cell specific ove rexpression of bcl-2 was associated with elevated serum titers of both highand low-affinity anti-ds DNA antibodies and rescue of autoreactive B cells that are normally eith er deleted or anergi zed (Kuo et al., 1999). Estrogen-treated mice showed up-regulation of Bcl-2 expression, not only in follicular B cells, but also germinal center B cells in spleens from immunohistochemical studies (Grimaldi et al., 2001). As a member of the TNF receptor family, FAS (CD95) is expressed at high levels on activated lymphocytes and other cells in several tissues, such as spleen, liver, lung, kidney and ovary (Itoh et al., 1991; Leithause r et al., 1993; Nagata et al., 1995). The binding of FAS to the cells expressing FA S ligand (a cell-surface member of the TNF family of proteins) transfers death signal and triggers apoptosis in the FAS-bearing cells (Wajant, 2002). Several lines of evidence suggest that FAS plays an important role in regulating the disposition of ac tivated, tolerant, autoreactive lymphocytes. FAS-deficient lpr/lpr and FAS ligand-deficient gld/gld mice s how defective apoptos is that leads to symptoms of lymphoproliferation and genera lized autoimmunity with high titers of autoantibodies (Izui et al., 1984; Watanabe-F ukunaga et al., 1992; Mountz et al., 1996; Allen et al., 1990; Lynch et al., 1994; Ramsdell et al., 1994; Takahashi et al., 1994). A specific role for FAS expressi on by B cells has been shown by in vivo adoptive transfer experiments (Sobel et al., 1991). Transforming growth factor 1 (TGF1) is the prototypical member of the super family of transforming growth factors (TGF) and regulates proliferation,

PAGE 100

86 differentiation, migration and a poptosis (Aoki et al., 2005). By increasing the expression of cell adhesion molecules, creating a chemotactic gradient and inducing proinflammatory cytokines, TGF1 promotes inflammation at early stage of the immune response (Aoki et al., 2005). It has a number of specific effects on B and T cells. TGF1 not only inhibits the prolif eration, differentiation and apopt osis of T cells, but it also enhances the growth of naïv e T cells and possibly promotes their differentiation into central memory T cells (Gorelik et al., 2002). In B cells, TGF1 also inhibits proliferation and immunoglobulin synthesis, and directs isotype switching to IgA and IgG2b in mice (Garcia et al., 1996; Aoki et al ., 2005). Estradiol was reported to increase TGFlevel (Salem, 2004). Cytokines Cytokines are small secreted proteins that mediate and regulate immunity, inflammation, and hematopoiesis. They provi de important signals which mediate the communication between antigen presenting cell s and lymphocytes. As CD4 T cells are one of the major cell types that secrete cytoki nes, we tested the le vels of a series of cytokines ex vivo by CD4 T cells isolated via magnetic-beads. Both Th1 and Th2 cytokines were tested since both types ar e reportedly involved in classical lupus pathogenesis, and a recent report suggests th at estrogen may affect the balance (Salem, 2004). Interleukin-6 (IL-6) is a Th2 pro-infla mmatory cytokine that promotes the differentiation and maturation of B cells a nd subsequent IgG production (Taga et al., 1997). Increased IL-6 has been linked to the formation of immune complex deposits detectable in SLE (Pelton et al., 1991). Major sources of IL-6 are monocyte/macrophages, fibroblas ts and endothelial cells, but B cells, mesangial cells and

PAGE 101

87 some other cell types can also produce IL-6 (Tackey et al., 2004). IL-6 administered in vivo to lupus-prone NZB/NZW F1 mice led to a worsening of disease and increased mortality (Finck et al., 1994; Ryffel et al., 1994 ). The role of estrogen in affecting IL-6 production is presently unclear. Estrogens ha ve been shown to be potent repressors of IL-6 production, which appears to be due to an interaction of the estrogen receptor with nuclear factor (NF)-IL6 and NF B, two transcription factors involved in IL-6 expression (Galien et al., 1997). However, estrogen al so induces fos and jun, two proto-oncogenes that together make up the act ivator protein (AP)-1 transc riptional complex that also suppresses the expression of IL-6 (Kic k et al., 1996; Hershko et al., 2002). Interferon(IFN) is a Th1 type cytokine main ly produced by activated natural killer cells and T cells (Ha ss et al., 1997). It plays an important role in acute inflammation, mainly by activati ng adhesive properties of en dothelial cells and induction of macrophage activatio n (Seery et al., 1997). In the subsequent antigen-specific phase of the immune response, IFNacts as a regulator of ly mphocyte proliferation and differentiation it can stimul ate immunoglobulin secretion by B cells and promote T-cell differentiation towards T helper 1 type (K rause et al., 2003; Billiau et al., 1998). Lymphocytes from female mice produced higher levels of IFNafter immune stimulation than did those of males both in vitro and in vivo (Huygen et al., 1984; McFarland et al., 1989). Administration of IFNaccelerated the rate of progression to glomerulonephritis in lupus-prone NZB/NZ W F1 mice (Engleman et al., 1981), while treatment with anti–IFNantibodies prevented glomerulone phritis (Jacob et al., 1987). Also transgenic mice that overexpressed IFNin the epidermis pr oduced antibodies to dsDNA (Seery et al., 1997). Elevated serum levels of IFNhave also been reported in

PAGE 102

88 patients with systemic lupus erythematosus (Kim et al., 1987). Several reports have shown that estradiol enhances IFNproduction by lymphocytes (Sarvetnick et al., 1990; Karpuzogle-Sahin et al., 2001). Tumor necrosis factor(TNF) is an important immunologic and proinflammatory cytokine in lupus (Theofil opoulos et al., 1999; Ar inger et al., 2003). Although Jacob et al. (1988) reported that the administration of TNFreduced the severity of lupus-like illness in NZB/NZ W F1 mice, this observation has not been replicated in several other lupus-prone strains (Edwards et al., 1996). mRNA levels of TNFwere reportedly higher in both MRL/lp r mice and NZB/NZW F1 mice with lupus nephritis, and the effect of TNFin NZB/NZW F1 mice depends on the dose and the age of the mice (Brennan et al., 1989). It is reported that levels of TNFwere raised in the serum of patients with SLE (Gabay et al., 1997 ), and that it has been detected in renal biopsies of patients with lupus nephritis (T akemura et al., 1994; Malide et al., 1995; Neale et al., 1995; Nakajima et al., 1999; Herrera-Esp arza et al., 1998). TNFpromoter polymorphisms have been reported in associ ation with SLE (Azizah et al., 2004; Correa et al., 2005). Although there is no doubt that TNFcan influence the immune system, it remains unknown whether the outcome of its effect s is beneficial or detrimental in human patients with SLE, as anti-TNFtherapy showed contradict ory results (Gomez et al., 2004; Lee et al., 1997; Jacob et al., 1990; Wilson et al., 1994). Also it is likely that TNFhas different target cells and molecules at different stages of lupus immunopathology (Mageed et al., 2002). Es tradiolÂ’s effects on TNFsecretion seem to be paradoxical, with some reports showing suppressive eff ects (Ito et al., 2001 and 2002; Tomaszewska et al., 2003; Wang et al., 2005) , and some showing enhanced tendency towards lupus

PAGE 103

89 (Baker et al., 2004). Also, Porter et al. ( 1991) reported that hormone replacement therapy resulted in higher levels of induced TNFin post-menopausal women. Granulocyte macrophage-col ony stimulating factor (GM-CSF) is the major proinflammatory regulator governing the f unctions of granulocyte and macrophage lineage populations at all stages of maturation (Ham ilton, 2002). Systemic or intraperitoneal GM-CSF injections in mice le d to increased number s of both circulating neutrophils and cycling perit oneal macrophages and dendritic cells (Metcalf et al., 1987). Campbell et al. 1997 and Bischof et al. 2000 reported that admi nistration of GM-CSF exacerbated arthritis in two murine arthriti s models. There was also a report that injection of GM-CSF into some patients exac erbated rheumatoid arthritis (Hazenberg et al., 1989). GM-CSF allows communication be tween hemopoietic cells and local tissue cells during inflammatory reactions (Hamilton et al., 2002). It can enhance host defenses against a broad spectrum of invading organi sms and may have a role as a vaccine adjuvant (Armitage, 1998). GM-CSF transgenic mice with overexpression of GM-CSF exhibited a syndrome of tissue damage, re sulting in progressive weight loss and premature death (Lang et al., 1987). GM-CSF has been shown to be secreted by several different cell types following stimulation, in cluding macrophages, monocytes, fibroblasts, endothelial cells, chondrocytes, smooth muscle cells and T lymphocytes (Zucali et al., 1986; Bagby et al., 1986; Leizer et al., 1990; Campbell et al., 1991; Filonzi et al., 1993; Hamilton, 1994). Estradiol was reported to be able to accelerate GM-CSF secretion (Kanda et al., 2004 and 2005; Suzuki et al., 2005), Interleukin-10 (IL-10) is a Th2 cytoki ne, and is a potent stimulator of B lymphocytes as well as a powerful inhibitor of antigen-presenting cells and T lymphocyte

PAGE 104

90 functions (Rousset et al., 1992; De Waal Malefyt et al., 1993). IL-10 has a major influence on autoantibody production in SLE. It may play a central role in the pathogenesis of SLE, in particular by i nducing autoantibody producti on (Llorente et al., 2003). Several studies have demonstrated th at IL-10 is strongly overproduced in SLE patients (Llorente et al., 1993 a nd 1994; Hagiwara et al., 1996; Horwitz et al., 1998). In the NZB/NZW F1 mouse model, Ishida et al. (1 994) showed that development of clinical features of murine lupus can be prevente d by administration of neutralizing anti-IL-10 antibodies. Moreover, by blocking IL-10 in vitro , it was possible to correct the impaired cellular immune responses in SLE patients (Miles et al., 1993 ; Lauwerys et al., 2000). Interleukin-4 (IL-4) is a Th2 cytokine with multiple functions in lupus-like autoimmunity (Singh, 2003). Previous literat ure shows conflicting results on the immune system, with some models s howing that IL-4 deficiency or neutralitions decreased autoantibody production and /or renal diseas e (Deocharan et al., 2003; Nakajima et al., 1997; Ochel et al., 1991), and ot hers showing IL-4 either pl ayed no role in autoantibody production (Kono et al., 1998 and 2000), or even an inhibitory role on autoantibody production (Santiago et al., 1997). This might be due to th e redundancy of its functions, different genetic backgrounds of the mouse models, and/or th e differential regulation of genes with opposing roles during the devel opment of lupus (Singh, 2003). Multiple roles for IL-4 in the development of lupus incl ude direct B cell stimulatory effects, Th2 promoting and Ig isotype switching effects, T cell suppress effect, and direct target organ damage (Singh et al., 2003). Increased IL-4 can rescue B cells from apoptosis and enhance their survival (Mori et al., 2000). Estradiol was reported to suppress IL-4 secretion (Elbourne et al., 1998).

PAGE 105

91 Interleukin-2 (IL-2) is a Th1 cytokine that functions predominantly as a potent T cell growth factor and can induce T cell ex pansion (Nelson, 2004). IL-2 plays an essential role in sensitizing T ce lls to activation-induced cell death in vitro , which limits the magnitude of T cell expansion and is me diated primarily by signals from FAS and TNF receptor (Green et al., 2003). IL-2-/mice showed autoimmunity, and it was speculated that this occurred through reduced AICD and impa ired peripheral tolerance (Nelson, 2004). Recent studies have also id entified a natural regulatory T cell population (Tregs) characterized by expression of CD25, th e high-affinity receptor subunit of IL-2. Loss of IL-2 in the murine system is associated with reduced munbers of Tregs and autoimmunity (Nelson, 2004). The regul ation of IL-2 by estradiol has been contradictory. Karpuzoglu-Sahin et al. (2001) reported that estradiol increased IL-2 mRNA in thymocytes, and sple nic lymphocytes. Selvaraj et al. (2005) showed the similar increasing effect on thymocytes. However, El bourne et al. (1998) and McMurray et al. (2001) reported a reduced IL-2 by estradiol treatment. Results Negative Selection by Magnetic Beads Result ed in Highly Purified Populations of B and CD4 T Cells In order to evaluate the effects of ch lordecone and estradiol on B cell gene expression and CD4 T cell cytokine secreti on, highly purified populations of cells were required. This need was only amplified by th e changes in subsets caused by estradiol. Negative selection using magnetic beads was appl ied to purify lymphocytes. This was to prevent any potential alteration in message levels in signal transduction by positive selection. After isolation, the purity of B a nd CD4 T cells were tested by flow cytometry using specific antibodies to B220 and CD4 respectively. Both lymphocytes were highly

PAGE 106

92 enriched with the purity higher than 90% (dat a not shown). There were few dead cells shown in the bottom of the flow cytometry figure, which could be one of the reasons for the variation among different treatment. Chlordecone Enhanced Bcl-2, Shp-1, and FAS Expression in B Cells Real-time PCR is a technique which allows for quantitative determinations of mRNA levels utilizing small samples (Freem an et al., 1999; Zimmerm ann et al., 1996). Results of real-time PCR for gene expression represent a relative level of each message compared to the level of the sample with th e lowest expression, corrected for the level of message for a common housekeeping gene (F reeman et al., 1999). To evaluate the potential effects of chlordecone and estrad iol exposure on transcri ptional levels of important B cell genes, we performed real-t ime PCR on freshly isolated B cells from young female NZB/NZW F1 mice exposed to sust ained-release estradiol or chlordecone over a five to six week period. Control mi ce received an implan ted pellet alone. Our results revealed that chlordecone treatmen t caused a significant increase (p<0.05) in Shp1 (Figure 4-1, Panel A), Bcl-2 (Panel B) a nd FAS (Figure 4-2, Panel A) expression in purified B cells. Five mg chlordecone cau sed an increase of 118%, 135%, and 78% in Shp-1, Bcl-2 and FAS expression, respectivel y. Estradiol treatment also caused an increase in the expression of these three genes, but none of them reached statistical significance. The corresponding increases caused by estradio l treatment were 57%, 64% and 58%, respectively. The f unctional expression of FAS li gand by B cells has been a matter of debate. However, as the FAS pathway is of great importance in the maintenance of tolerance, it was decided to examine FAS ligand expression in our highly purified B cells. Using primers based on th e mouse FAS ligand sequence, we detected

PAGE 107

93 signals by real-time PCR (Fi gure 4-2, Panel B). Howeve r, neither chlordecone nor estradiol had an appreciabl e effect on the expression. Estradiol Decreased IFNExpression in B Cells Inconsistent with previous reports (Sarve tnick et al., 1990; Kar puzogle-Sahin et al., 2001), estradiol treatment caused a significant down-regulation in IFNexpression in B cells. The mean value in estradiol-treated group was only one-fourth of the mean value shown in the control group. Chlordecone treatment decreased IFNexpression as well, but did not reach statistical signifi cance, as shown in Figure 4-3. Neither Chlordecone Nor Estradiol Changed the Expression of Fc RIIb, TNF, TGF, Ly5, IL-6 and IL-2 The real-time PCR results revealed that the expression levels of Fc RIIb (Figure 4-4, Panel A), TNF(Figure 4-4, Panel B), TGF(Figure 4-4, Panel C), Ly5 (Figure 4-4, Panel D), IL-2 (Figure 4-5, Panel A) and IL -6 (Figure 4-5, Panel B) appeared to be unaffected by either chlordecone or estradiol. CD3 and/or CD28 Stimulation Signific antly Increased Cytokine Secretion Since cytokines play very important roles in the autoimmunity process, we decided to test some of the important Th1 and Th2 cytokines secreted by magnetic-bead isolated CD4 T cells. As samples were limited, it was also decided to use a multiplex cytokine assay kit to simultaneously measure 10 cyt okines from a single well, utilizing the Luminex® 100™ system. A panel of ten cytokines was tested, including IL-1 , IL-2, IL4, IL-5, IL-6, IL-10, IL-12(p70), TNF , IFN and GM-CSF, utilizing sixteen to twenty samples from each group. We chose to test cultured cells for two reasons. First, preliminary test showed serum levels of cytokines were typically at or below the sensitivity of this method. Second, rheumatoid factor (i.e. autoantibodi es with specificity

PAGE 108

94 for the Fc portion of other autoantibodies) occu rs frequently in human and murine SLE, and would interfere with the assay (Haberman et al., 2003). To minimize artifacts due to in vitro conditions, we focused on short-term assays under conditions of optimal and suboptimal stimulation. Stimulation was provided by a combination of anti-CD3 and anti-CD28. CD3 is part of T cell recept or (TCR)/CD3 complex, and transduces the antigen binding signal outside the plasma membrane into ch emical signals by phosphorylation in the cytoplasm (Monks et al ., 1998). CD28 is a T cell co-stimulator, and co-ligation of TCR/CD3 and CD28 promotes T cell proliferation, cytokine secretion, and cell survival (G rakoui et al., 1999). Isolated CD4 T cells secreted undetecta ble levels of all ten cytokines in unstimulated conditions, indicating very low le vel of basal line cy tokine secretion by unactivated CD4 T cells, even in lupus-prone mice. After stimulation with 5 µg/mL antiCD28 and anti-CD3 which was pre-coated on the plate (Chapter 2), si x cytokine levels were significantly increased, including TNF(Figure 4-6, Panel A), IL-2 (Figure 4-6, Panel B), IFN(Figure 4-7, Panel A), GM-CSF (Fi gure 4-7, Panel B), IL-10 (Figure 48), and IL-4 (Figure 4-9). The other four cytokines IL-1 , IL-5, IL-6 and IL-12 remained undetectable (data not shown). As expected, there was a high degree of variability in absolute levels. They ra nged from 4000 to 8000 pg/mL for IL-2 and from 200 to 700 pg/mL for TNF. The other four cytokines show ed intermediate values: IL-4 (500 to 2000 pg/mL), IL-10 (250 to 700 pg/mL), IFN(400 to 2500 pg/mL), and GMCSF (250 to 1100 pg/mL). In all cases, these were above the lower limit of sensitivity, as provided by the manufacturer.

PAGE 109

95 Normalizing the Cytokine Levels Two plates were run on sequential days due to the large numbe r of samples, and there were differences in absolute values fo r cytokine levels between the two days that appeared to be systematic. The systematic differences might come from the differences in the standard curve. However, this would be expected to be small. A greater source of variability were the culture conditions themselv es, as cells from four mice at a time were cultured, one from each group. Since each day studied included one mouse from each treatment group, we decided it would more accurate to normalize the results to that dayÂ’s control mouse. This normalization process reduc ed the effects of day-to-day variations in culture conditions, and allowed th e effects of chlordecone and estradiol to be seen more clearly. Both Chlordecone and Estradiol Signif icantly Increased Secretion of the Proinflammatory Cytokines TNFand IL-2 Based on the normalized cytokine levels , secretion of the pro-inflammatory cytokines TNFand IL-2 were significantly increased in both 5 mg chlordeconeand estradiol-treated mice compared with the control group. TNFsecretion in the 1 mg chlordecone-treated group was also significantly higher than the control group. On the other hand, there was no statistica lly significant difference for TNFand IL-2 secretion among chlordecone and estradiol groups. Chlordecone exposure caused 22% and 36% increase of TNFlevels in 1 mg and 5 mg treated groups respectively, and estradiol treatment caused a 23% increase, as shown in Figure 4-6, Panel A. For IL-2 secretion, 5 mg chlordecone and estradio l treatment caused 19% and 27% increase respectively, as shown in Figure 4-6, Panel B.

PAGE 110

96 Chlordecone, but Not Estradiol, Increase d the Secretion of Pro-inflammatory Cytokines IFNand GM-CSF Chlordecone treatment significantly increas ed the secretion of pro-inflammatory cytokines IFNand GM-CSF compared with control group, and both chlordecone groups showed significantly higher level of GM-CSF, but not IFNcompared with estradiol group (Figure 4-7, Panel A and B). The levels of IFNand GM-CSF were increased 52% and 57% respectively in 1 mg group, and 67% and 59% respectively in 5 mg group. Estradiol treatment showed no effects on either IFNor GM-CSF levels compared with control group. Estradiol, but Not Chlordecone, Increased IL-10 Secretion In contrast, estradiol had a greater effect on IL-10 se cretion. After normalizing the raw data, estradiol-treated CD4+ T cells secr eted higher level of IL-10 compared with control group (Figure 4-8, Panel A). Ther e was a 32% increase in IL-10 level by estradiol treatment compared with control gr oup. Chlordecone treatment did not affect IL-10 secretion. There was no statistically significant difference among chlordecone and estradiol groups. Estradiol, but Not Chlordecone, Caused a Significant Decrease in IL-4 Level In contrast to the other cytokines, which were either up-regulated or not affected by estradiol, Th2 cytokine IL-4 level was significantly decreased in the estradiol group compared with control and chlordecone-tr eated groups. Chlordecone treatment had no effect on its secretion. Estrad iol caused a significant 35% decrease in IL-4 level, as shown in Figure 4-9.

PAGE 111

97 Discussion Gene expression investigation has arisen as an important method to reveal disease pathogenesis in recent years, and numerous techniques have been developed for this aspect. Traditional methods in molecular biol ogy generally work on a "one gene in one experiment" basis, which means that the throughput is very limited and the "whole picture" of gene function is hard to obtain. In the past several years, DNA microarray has emerged as a new method to monitor a group of genes, and even the whole genome, and has attracted tremendous interest among biologi sts. This technology provides researchers with a better picture of th e interactions among thousands of genes simultaneously. However, despite its popularity, experience accompanying the use of microarray has also shown inconsistency and false results from gene chips when compared with the data from traditional methods. It has frequently been emphasized that the data from microarray studies must be confirmed with more quantitativ e techniques before they can be accepted. Due to the financial cost and potential probl em of interpretation of the results from a gene array experiment, we chose the quantit ative real-time PCR technique to test the expression levels of a series of selected impor tant genes involved in lupus pathogenesis. The genes were carefully selected for their relevance to the differe nt pathways known to be affected by lupus, with a special emphasis on genes previously established by others to be affected by estradiol. The influence on gene expression is highly dependent on the doses of the compound and exposure time during the experiment. Because of the in vivo nature of this study, it was impossible to test all time points. For a number of reasons, we decided to carefully examine the effects of chlordecone 4-6 weeks after in itial exposure. First, at this time point, chlordecone did not break to lerance, so that we can analyze the gene

PAGE 112

98 expression on a relatively clean backgroup. S econd, based on the previous studies of the phenotypic markers, splenic B cells were activat ed by chlordecone at this time point, with a clear increase in germinal center responses. As the goa l of our experiments was to understand whether low doses of chlordecone influence the immune system through an estrogenic model, we chose 1 mg and 5 mg as the doses at which very little uterus hypertrophy effect was seen. On the other hand, the low dose might be responsible for the lack of effects on the gene expression in some cases. There were variations among samples in expression of some genes, this variation was sometimes as big as fifty times, but it is a normal phenomenon. Among the selected genes, both chlordecone and estradiol enha nced the expression of Bcl-2, Shp-1 and Fas in B cells, with 5 mg chlordecone having stronger effects than estradiol. On the other hand, they both decreased the expression of IFN, with estradiol having stronger effect. For the rest of the genes, including Fc RIIb, TNF, TGF, Ly5, IL-6 and IL-2, neither chlordecone nor estradiol cau sed statistically significant changes on their gene expression. Chlordecone showed stronger effects in up-regulating Shp-1 and Bcl-2 expression than estradiol. Enhanced Bcl-2 was speculate d to enhance the survival of autoreactive B cells that would normally be deleted at an immature state, and increased Shp-1 might raise the threshold for B cell receptor crosslin king that is required for the deletion of autoreactive B cells (Grimaldi et al., 2002) . Using mice with transgenic BALB/c background, estradiol (0 .18 mg/pellet) was reported to e nhance 20% of the protein level of Shp-1 and Bcl-2 by flow cytometric anal ysis (Grimaldi et al ., 2002). We found that the increase in gene expression level by estr adiol treatment was higher in the NZB/NZW

PAGE 113

99 F1 mouse model, even though their dose was 3.6 times higher than ours. The differences might come from potential post transcrip tion regulation which might lower the final protein level, or by the difference between m ouse strains. We have tested the protein level of Bcl-2 by flow cytometry, and the resu lts were shown in the previous chapter. Although both estradiol and chlo rdecone treatment caused a cl ear increase in protein level of Bcl-2, estradiol wh ich caused about 23% increase , had stronger effects than chlordecone which increased about 12% (Figure 38). This seemed consistent with a role for post transcriptional regulat ion. Although we found a clear increase in Bcl-2 and Shp1 expression in both chlordecone and estradiol group, but how this increase is directly related to accelerated autoimmunity is still not clear. We will discuss the flaw in the chapter 6 for potential methods of verify ing the importance of their changes. The expression of FAS gene on B cells was enhanced by both chlordecone and estradiol. This is difficult to interpret here, as increas ed FAS expression suggests the potential for stronger signali ng in activation-induced cell de ath, an important pathway for apoptosis. However, we have found clearly re duced B cell apoptosis in Chapter 3. It is possible that the protein levels of FAS are not changed as much as its gene level due to post-transcription regulation, since we did not have data on the protein level of FAS expression. It is also possibl e that the increased FAS expre ssion is a compensatory step for the down-regulation of FAS ligand. In our data, although the dow n-regulation of FAS ligand by chlordecone and estradiol did not r each statistical significance, it is possible that the small decrease is functionally important . The third possibility is that enhanced expression of Bcl-2 can overcome the effects caused by elevated FAS. The intrinsic apoptosis pathway through FAS involves the re lease of a number of proapoptotic factors

PAGE 114

100 by mitochondria, which results in caspase-9 ac tivity and in turn caspase-3 activity leading to apoptosis (Li et al., 1997). However, this mitochondrial apopt otic activity can be blocked by the expression of the Bcl-2 fam ily of antiapoptotic proteins, presumably by blocking the release of those proapoptotic factors within mitochondria (Green et al., 1998). Liu et al. (2003) reported that estrad iol increased expressi on of FAS protein and mRNA. Both chlordecone and estradiol caused a clear decrease in IFNlevel in B cells, with estradiol treatment reaching statistical significance. This finding was a surprise, as from several lines of evidence, the IFNexpression should be increased. First, previous studies reported that estradiol enhanced IFNproduction by lymphocytes (Sarvetnick et al., 1990; Karpuzogle-Sahin et al., 2001). Second, from the view of function, IFNacts as a regulator of lymphocy te proliferation and diffe rentiation and stimulates immunoglobulin secretion by B ce lls (Krause et al., 2003; Bil liau et al., 1998). Both chlordecone and estradiol treatment clearly ac tivated B cells and e nhanced auto-antibody secretion, theoretically B ce lls under their treatment should have an increased IFNlevel. Third, from our cytokine secreti on results, isolated CD4 T cells produced significantly higher level of IFNin both 1 mg and 5 mg chlordecone-treated groups. Estradiol treatment did not si gnificantly increase its secret ion, but it showed a slight increasing tendency. All these evidences seem to be against a down-regulation in IFNexpression. However, B cell is not the main resource to produce IFN, and since the gene expression of TNFin B cell is much lower than in T cell, the transcription regulation in different cell type s might be different. It is possible that there were some negative transcription factors up-regulated by chlordecone or estradiol treatment. As we

PAGE 115

101 did not test the protein level of IFN, it is hard to tell whether the decreased gene expression really reflected a down-regulation in the protein level. Further study using either western blot or flow cytometric anal ysis using intracellular staining should be conducted to verify this finding. Although the results did not reach statisti cal significance, chlordecone showed slight up-regulation of the cy tokines IL-2 and IL-6, this was different from the effects seen with estradiol treatment. This effect may be common to organochlorine pesticides. Tarraf et al. (2003) reported that dieldrin can up-regulat e IL-6 expression in CID-9 mammary cells, and Kim et al. (2005) reported that methoxychlor can increase macrophage secretion of IL-6. IL-6 was reported to promote the differentiation and maturation of B cells and subsequent IgG pr oduction (Taga et al., 1997) . The direct role of IL-6 in enhancing autoantibody production was demonstrated in the pristine induced model of lupus (Richard et al., 1998). Increased IL-6 has also been linked to the formation of immune complex deposits detect able in SLE (Pelton et al., 1991). Taken together, the increased tendency in IL-6 leve l caused by chlordecone is consistent with the previous findings and the the ability of chlordecone to accelerate autoimmunity. On the other hand, estradiol did not show a clear effect on the IL-6 le vel, although previous studies reported a suppressive effect due to the interaction of estrogen receptor and some transcription factors (Galien et al., 1997; Kick et al., 1996; Hershko et al., 2002). This was probably due to the low dose used in our experiment. Both chlordecone and estradiol effects on the IL-2 level in B cells were not very striking, although chlordecone caused a slight increase (not statisti cally significant), while estradiol did no t show any effects. On the other hand, CD4 T cells secreted

PAGE 116

102 significantly higher IL-2 in estradiol-treated group, this might be because IL-2 is more important to T cell function, as it is a potent T cell growth factor (G reen et al., 2003). Neither chlordecone nor estradiol showed influence on TNFgene expression in B cells, in contrast to the enhanced TNFsecretion by CD4 T cells. And neither chlordecone nor estradiol affected the expres sion of Ly5 and the inhibitory molecule Fc RIIb. From Dr. Betty DiamondÂ’s unpublished gene array result, Ly5 was upregulated in B cell under estradiol treatment, but the increase was weak (about 50%). And the dose they used was 3.6 times higher than ours. In our real-time PCR results, estradiol caused a 44% increase in the gene level, but because of the variation among samples, it is hard to tell whether the increa se was real. The overa ll gene expression of Fc RIIb in B cells was not affected by either ch lordecone or estradiol, combined with the unaffected protein expression on germinal cen ter B cells, we might conclude that the acceleration of autoimmunity by chlordecone and estradiol is not mediated through their effects on this inhibitory receptor. The expression of TGF1 gene was slightly increased by both chlordecone and estradiol treatment, but neither reached statistical significance. The increasing tendency by estradiol treatment is consistent with the result from gene array tests from Dr. Betty DiamondÂ’s lab (unpublished data). Increased TGF1 gene expression may be involved in reducing Fas-induced CD4 T cell apoptosis (G orelik et al., 2002). Chen et al. (2001) reported that TGF1 may protect T cells at multipl e sites in the death pathway, particularly by maintaining the essen tial integrity of mitochondria. TGFhas been shown to inhibit proliferation of several cell types through different mechanisms, such as upregulation of cell-cycle inhi bitors p15 (INK4B) (Hannon et al., 1994), p21 (Datto et al.,

PAGE 117

103 1995), or p27 (Polyak et al., 1994), as well as the down-regulation of a proto-oncogene cmyc that is crucial for cellula r proliferation (Coffey et al., 1988). In chapter 3, we found that neither chlordecone nor estradiol increas ed B cell proliferation, this increased TGFmight be partially responsible for that. The partial overlap of the changes in B cell gene expression was not surprising. Although we did not screen for global gene expression by microarrays, other researchers have done this type of test. For example, the effect of an estrogenic soy isoflavone genistein on thymocytes was compared to estr adiol using Affymetrix gene chips (Selvaraj et al., 2005). Among about 22,600 genes that we re investigated, only 807 genes (a small portion of the total genes) were significantly altered by expos ure to either genistein or estradiol, among which 538 genes were regu lated by estradiol, and 456 genes were regulated by genistein. There were 187 ge nes which account for 23% of the 807 genes, similarly regulated by both. This result shows that although ge nistein has a clear estrogenic effect, and it might share some im portant pathways with estradiol that are involved in modulation of the i mmune system, such as in re gulating several transcription factors and in controlling thymocyte apoptosis, a large portion of the genes did not appear to act through estrogenic effects. Estradio l and genistein have completely different structures, and genistein only weakly binds to the estrogen receptor (Selvaraj et al., 2005). Its effects on the immune system ar e complicated and the estrogenic pathway might only count for a small portion of the tota l effects. There were also reports that genistein acted through a non-estrogenic path way, although not on the immune system. For example, Fanti et al. (1998) found that genistein reduced both trabecular and compact bone loss on ovariectomized rat which were co mpletely different from estradiol, and it

PAGE 118

104 lacked the action to cause ut erus hypertrophy which is an im portant characteristics of an estrogenic effect. Chlordecone also has a very different structure compared to estradiol, and also weakly binds to the estrogen receptor. We infer that if a global gene expression test were conducted, chlordecone would act lik e genistein, showing some similarity, but also a large difference with estradiol in B cell gene expression. Cytokines are essential mo lecules involved in the di fferentiation, maturation and activation of cells, and have a signifi cant influence on the immunoinflammatory response. Helper T cells (Th) and macr ophages are the two pre dominant producers of cytokines. As a second part of this chapter, we tested ten cytokine levels, including both Th1 and Th2, secreted by isolated CD4 T cells. Four cytokines: IL-1 , IL-5, IL-6 and IL12, were below the sensitivity of our me thod. Among the six cytokines that were detectable under stimulation, TNF, IFN, GM-CSF, IL-2 are Th1 cytokines, while IL4 and IL-10 are Th2 cytokines. Although Chlordecone and estradiol treatment both increased TNFsecretion, they also showed clear differences in regulating the other cytokines. Chlordecone treatment seems to enhance Th1 cytokine secretion, while estradiol has stronger e ffects on Th2 cytokines. Chlordecone significantly enhanced GM-CSF and IFNsecretion, while estradiol had no influence on GM-CSF level, and slightly increased IFNsecretion which did not reach statistical significance. As GM-CSF is important in governing the functions of granulocyte and macrophage populations (Hamilton, 2002), increased GM-CSF by chlordecone treatment might pr ovide signals to macrophages a nd dendritic cells, increase their number, and further promote their matu ration. The maturation of antigen presenting cells under the condition of incr eased self-antigens might then break tolerance and lead to

PAGE 119

105 autoimmunity. We tested the macrophage popul ation in the spleen, showing that there was a significant increase of the percentage in the 5 mg chlordecone-treated group (Figure 3-18), which is consistent with chlo rdeconeÂ’s effects as an up-regulator of GMCSF secretion. Chlordecone significantly increased IFNsecretion, which is consistent with IFNÂ’s function in stimulating immunoglobulin secretion by B cells and accelerating autoimmunity (Krause et al., 2003). Although it did not reach statis tical significance, estradiol treatment also showed an increasing tendency in IFNsecretion, and is consistent with previous reports (Sarvetnic k et al., 1990; KarpuzogleSahin et al., 2001). However, we have discussed previo usly that the gene level of IFNin B cells was clearly down-regulated in chlordecone and estr adiol treatment. The reason still remains unknown. It is possible that th e decreased expression of IFNin B cells is related to estradiol and chlordecone effects on the half life of IFNmRNA. Estradiol and chlordecone treatments might increase mRNA le vel in the beginning of the treatment, but shorten the half life of mRNA. Both chlordecone and estradiol significantly enhanced TNFsecretion by CD4 T cells, however, the physiological function of th is increase was unknown, as there was still debate on TNFÂ’s suppressive or stimulating role in autoimmunity (Cope, 1998; Mageed et al., 2002). Another reason is that the main producers of TNFare macrophages, mast cells and natural killer cells. It is unknown whether th e increase of TNFby CD4 T cells is of physiological importance. The secretion of IL-2 was significant ly increased by 5 mg chlordecone and estradiol treatment. The enhanced secretion by estradiol treatment is different from the

PAGE 120

106 previous reports by Elbourne et al. (1998), in which they showed a decreased IL-2 secretion by estradiol treatment. However, th ey did not isolate CD4 T cells, and the IL-2 level they tested was from mixed lymphocytes , including B and T cells. McMurray et al. (2001) further reported that estradiol reduced IL-2 expression in the CD4+ Jurkat T cell line, and the suppression of IL-2 was associ ated with decreased nuclear binding of two important IL-2 promoter transcription factors: NF B and AP-1. However, KarpuzogluSahin et al. (2001) reported th at estradiol increased IL-2 mRNA in thymocytes, and splenic lymphocytes. Selvaraj et al. (2005) al so reported that estrad iol and the estrogenic soy isoflavone genistein increased IL-2 gene expression level in mouse CD4 thymocytes by quantitative real-time PCR. The difference between in vivo and in vitro treatments can not be overlooked, as the regulation of a number of transcription factors, such NFAT, AP1, CREB and NF B, which directly cont rol the production of IL-2 by T cells, might be different between in vivo and in vitro tests. In in vivo experiments, estradiol might produce effects indirectly thr ough its metabolites. Selvaraj et al. s uggested that the upregulation of IL-2 by estradiol treatment is a homeostatic mechanism to compensate for reduced IL-2 signaling in thymocytes. Th e more significantly increased secretion of IL-2 in the estradiol-treated group than th e chlordecone-treated group might explain the differences in T cell activation. The previous chapter showed that estradiol, but not chlordecone treatment, activates T cells. Th e activation by estradiol might be potentially mediated by enhanced IL-2, as it is a poten t T cell growth factor and can induce T cell expansion and activation. A lthough unactivated, CD4 T cells in the chlordecone-treated group still showed stronger effects on the secr etion of some pro-infl ammatory cytokines, this might be because the genes of those cy tokines were significantly up-regulated under

PAGE 121

107 chlordecone treatment, as the regulation of different genes under the same treatment can be very different. We did not measure the ge ne expression of these cytokines, so it is hard to conclude here. Estradiol significantly enhanced IL-10 secretion. Chlordec one treatment also showed a slight increase in its level which did not reach statistical significance. The increased secretion of IL-10 is consistent with its ro le in enhancing autoantibody production (Llorente et al., 2003). Estradiol significantly reduced IL-4 secretion by CD4 T cells, which is consistent with previous reports (Elbourne et al., 1998) , while chlordecone treatment did not show this effect. IL-4 has a T cell suppressor e ffect (Singh, 2003). The reduced IL-4 secretion in the estradiol treated group might partially explain the act ivation of T cells shown by the expansion of activation marker s, as shown in Figure 3-13.

PAGE 122

108 A. Shp-1 expression in B cells B. Bcl-2 expression in B cells Figure 4-1. Increased gene expression of S hp-1 and Bcl-2 by chlordecone treatment in purified splenic B cells. B cells were enriched by negativ e isolation with magnetic-beads. By real-time PCR, there was a statistically significant increase in Shp-1 (A) and Bcl-2 (B) (p <0.05) gene expression in mice treated with 5 mg chlordecone. Mice treated with estradiol did not show these effects. House-keeping gene -actin was used to normalize the gene expression.

PAGE 123

109 A. FAS expression in B cells B. FAS ligand expression in B cells Figure 4-2. Analysis of gene expressi on of FAS and FAS ligand by chlordecone treatment in purified splenic B cells. (A) Splenic B cells from mice treated with 5 mg chlordecone had a statistica lly significant increase of FAS gene expression (p<0.05). Estradiol-treated mice had an increase of approximately the same magnitude, but did not reach statistical significance. (B) Neither chlordecone nor estradiol showed eff ects on the FAS ligand expression by B cells.

PAGE 124

110 Figure 4-3. Comparison of chlordecone and es tradiol effects on the expression of IFNon splenic B cells. Estradiol treatment significantly decreased IFNexpression (p<0.05), while chlordecone tr eatment did not show this effect.

PAGE 125

111 A. Fc RIIb expression in B cells B. TNFexpression in B cells C. TGFexpression in B cells D. Ly5 expression in B cells Figure 4-4. Comparison of chlordecone and estradiol effects on the expression of Fc RIIb, TNF, TGFand Ly5 in splenic B cells. Neither chlordecone nor estradiol caused a significant change in expression levels of any of these genes.

PAGE 126

112 A. IL-2 expression in B cells B. IL-6 expression in B cells Figure 4-5. Comparison of chlordecone and es tradiol effects on the expression of IL-2 and IL-6 in splenic B cells. Neithe r chlordecone nor estradiol caused a significant change in expression levels of these two genes.

PAGE 127

113 A. TNFsecretion by CD4 T cells B. IL-2 secretion by CD4 T cells Figure 4-6. Analysis of the secretion of the pro-inflammatory cytokine TNFand IL-2 by CD4 T cells. The bar in each treatment represents the relative cytokine level that was normalized to the cytokine level of each dayÂ’s control mouse. The dotted line represents the level of control mice. All treatments significantly increased the secretions of TNFcompared with the normalized control value that was one, while there was no difference among the three treatments (A). IL-2 level was increased in 5 mg chlordecone and estradiol treatment compared with control gr oup, and there was no difference among three treatments (B).

PAGE 128

114 A. IFNsecretion by CD4 T cells B. GM-CSF secretion by CD4 T cells Figure 4-7. Comparison of chlordecone and estr adiol effects on the secretion of cytokines IFNand GM-CSF by the CD4 T cells. The dotted line represents the level of control mice. Cultured CD4+ T cells, exposed to 1 or 5 mg of chlordecone in vivo, had statistically significant increase in IFN(A) and GM-CSF (B) secretion compared with contro l group. On the other hand, IFNlevel was not significantly different among chlordec one and estradiol tr eatments, while GM-CSF levels in 1 and 5 mg chlordec one treatment was significantly higher than estradiol group.

PAGE 129

115 Figure 4-8. Comparison of chlordecone and estr adiol effects on the secretion of cytokine IL-10 by the CD4 T cells. The dotted line re presents the level of control mice. Estradiol, but not chlordecone exposur e, caused a statistically significant increase of IL-2 compared with cont rol group by cultured and stimulated CD4 T cells, but there was no differen ce among chlordecone and estradiol treatments.

PAGE 130

116 Figure 4-9. Comparison of chlordecone and estr adiol effects on the secretion of cytokine IL-4 by cultured CD4 T cells. The dotted line represents the level of control mice. Estradiol, but not chlordecone treatment, caused a statistically significant decrease in IL-4 secretion by CD4 T cells compared with control and chlordecone treatments.

PAGE 131

117 CHAPTER 5 COMPARISON OF CHLORDECONE AND ESTROGEN EFFECTS ON MACROPHAGE FUNCTIONS AND PEPTIDE HORMONE PROLACTIN SECRETION Introduction We have focused mainly on chlordecone and estradiol effects on splenic lymphocytes so far. However, lupus is a sy stemic autoimmune disease that affects the whole body, and there are many other factors invo lved in the disease pathogenesis. In this chapter, we will try to identify some other mechanisms that might potentially induce autoimmunity beyond direct effects on lymphocytes. Macrophage and Autoimmunity Macrophages are generally the first cells to encounter a foreign substance in the body. They nonspecifically engulf such materi als, as well as scav enge normal cellular debris, and degrade it using pow erful hydrolytic enzymes and oxi dative attack. Peptides from the degraded proteins are then ca rried to the macrophage cell surface bound to MHC class II where they can be recognized by T lymphocytes. The macrophage clearance of apoptotic cells is very effective, and generally results in no inflammatory response (Savill et al., 1993). However, recen t experimental systems have shown that when clearance mechanisms are overloaded by ei ther an abnormal number of apoptotic cells (Mevorach et al., 1998) or by deficient clearance mechanisms, autoimmunity can result (Cohen et al., 2002; Sc ott et al., 2001). In addition, in vitro differentiated macrophages from a subgroup of SLE pa tients showed significantly reduced phagocytosis of apoptotic cells (Baumann et al., 2002).

PAGE 132

118 Toxic Effect on Macrophage There is evidence that other organochlori ne pesticides, toxaphene, dieldrin, p,pDDT and DDE, can have inhibitory effects on macrophages. C57BL/ 6 (B6) mice given a single dose of dieldrin intrap eritoneally demonstrated a dose-dependent decrease in macrophage antigen processing 10 days later (K rzystyniak et al., 1989) . An effect could be seen at doses as low as 9 mg/kg, while inhibition of phagocytosis was seen at 36 mg/kg. Toxaphene given to Swiss-Webster mice at concentrations of 10, 100, and 200 ppm in diet for eight weeks also inhibited macrophage func tion (Allen et al., 1983). In other experiments, offspring of mice exposed to these concentrations of toxaphene in the diet were tested. Even in mice exposed to 10 ppm in diet through gestation and weaning, there was a 50% reduction in phagocytosis (Allen et al., 1983). Finally, DDT and DDE have been found to have similar effects. In vitro, exposure of the murine macrophage line J774A.1 to either p,p’-DDT or p,p’-DDE at 2.5 µg/mL resulted in more than 50% inhibition of macrophage function by a numbe r of parameters (N unez G et al., 2002). There is no direct evidence for chlordec one’s effect on macrophage function, but Carmines et al. (1979) have proven that chlordecone produced a dose-dependent inhibition of cellular proliferation and a bim odal alteration of phagocytic activity in the P388D1 ‘macrophage-like’ cell in tissue culture. In contrast, it has been reported that in guinea pigs estrogen can e nhance splenic macrophage Fc R-dependent clearance of IgGcoated erythrocytes, but cannot affect th e C3-dependent clearance of IgM-coated erythrocytes by hepatic macropha ges (Schreiber et al., 1988). Prolactin and Autoimmunity Prolactin (PRL) is a single-chain pol ypeptide hormone consisting of 200 amino acids, which is synthesized and secreted primar ily by the anterior pitu itary gland, but also

PAGE 133

119 in many extrapituitary sites, including cells of the immune system (Ben-Jonathan et al., 1996; Bole-Feysot et al., 1998; Hiestand et al., 1986). Montgomery et al. (1990 and 1992) identified prolacti n-like proteins that are produced and secreted by lymphocytes. It remains unclear whether the am ount of prolactin secreted by lymphocytes is sufficient to affect the prolactin level in the serum. While the main functi on of prolactin is to regulate the growth and differentiation of the mammary gland and the ovary (B ole-Feysot et al., 1998), it also acts as an important connection between the endocrine and immune system (Peeva et al., 2005). Prolactin has also been classified as a cytoki ne and signals through specific membrane prolactin receptors, which ar e also members of the cytokine receptor superfamily and lack intrinsic tyrosine ki nase activity (Leonard et al., 1998). The prolactin receptor has been found on monocyt es, and on B and T lymphocytes (Matera et al., 1988; Pellegrini et al., 1992; Matera et al., 1997; Gagnerau lt et al., 1993; Russel et al., 1984; Matera et al., 2000; M ontgomery et al., 1992). Prol actin ligand binds to the receptors, causing the dimerization of recepto rs and the recruitment of cytoplasmic molecules, such as the tyrosine kinase Janus kinase 2 (JAK2) and the STAT family members, to bind to prolactin receptor and me diate the signal transduction (Leonard et al., 1998). Through its receptors, prolactin modulates immune system function by stimulating both cell pro liferation and survival (Bole-Feysot et al., 1998). In mouse, four isoforms of prolactin recept ors have been identified, with one long isoform and three short isoforms. The short isoforms only di ffer by a few amino acids in the C-terminal end (Davis et al., 1989; Clarke et al., 1993). About 15 – 25% of SLE patients show ed increased prolactin le vels (Vidaller et al., 1986; McMurray, 1996; Lavalle et al., 1993; Walk er et al., 1998; Mcmurray et al., 1995;

PAGE 134

120 Allen et al., 1996; Jara et al., 1992). It has been demonstrated that both nonstimulated and mitogen-stimulated lymphocytes from lupus patients secrete more prolactin than control lymphocytes (Gutierrez et al., 1995; Larrea et al., 1997) . Prolactin affects B cell development and maturation, causing a decrease in the percentage of immature B cells in the spleen (Peeva et al., 2003). As B cells mature from T1 to T2 stage, they are susceptible to negative selection. Normally there are more T1 than T2 B cells in the spleen, reflecting the loss of autoreactive ce lls occurring at the T1 to T2 transition. Prolactin treatment inverts the T1/T2 ratio w ith more T2 than T1 cells, suggesting that prolactin diminishes negative selection of immature B cells (Peeva et al., 2003). Prolactin also upregulates expression of Bcl-2 and CD40 in B cells, causing a decrease in apoptosis and an increased susceptibility to co stimulation (Peeva et al., 2003), both of which may contribute to the survival and rescue of B cells that are signaled by antigen to undergo negative se lection. Results Both Chlordecone and Estradiol Reduced Peritoneal Macrophage Clearance of Apoptotic Cells To test chlordecone and estradiol ef fects on macrophage function in engulfing apoptotic bodies, peritoneal ce lls from mice receiving differe nt treatments were used in an in vitro clearance experiment. Since macropha ges are not the only cell type among peritoneal cells, flow cytometric tests were conducted to identify the change in the percentages of different popul ations. Neither chlordecone nor estradiol treatment affected the percentage of CD11bhighB220low cells (data not shown), which were the traditional macrophages. More recently, it ha s been shown that B1 cells, a subpopulation of B cells identified by co-e xpression of CD5 and presen t in high numbers in the

PAGE 135

121 peritoneal cavity, are also pha gocytic (Borrello et al., 2001). Moreover, B1 cells can coexpress the classic monocyte/macrophage mark er CD11b (Borrello et al., 2001). To rule out the influence caused by B1 cells, the per itoneal cells were stained with different markers that could differentiate B1 cells from macrophages, and only the traditional CD11bhighB220low macrophages were gated for these e xperiments. Both chlordecone and estradiol reduced clearance by macrophages (F igure 5-1, Panel A). Chlordecone showed a dose-dependent decrease and the 5 mg per pe llet group reached statis tical significance. We also measured the clearance of B1 cells, and the result is shown in Figure 5-1, Panel B. Neither chlordecone nor estradiol affected B1 cell clearance of apoptotic cells. Chlordecone Treatment Did Not Affect the Proliferation of RAW 267.4 Cells To test chlordecone and es tradiol effects on macrophage survival and proliferation, RAW 267.4 macrophage-like cell line was used in the experiments. A series of concentrations, including 0.01 µM, 0.1 µM, 1 µM, 10 µM, 50 µM and 100 µM for chlordecone, and 0.01 µM, 0.1 µM, 1 µM, 10 µM for estradiol were used to treat 1 million RAW cells for 24 hours in complete DMEM media. Chlordecone showed clear toxic effects on RAW cell survival at 50 µM and 100 µM, while the survival appeared not to be affected at 10 µM and lower concentra tions (data not shown). On the other hand, estradiol treatment did not affect the surv ival at any concentration tested in our experiment. The proliferation test was conducted using RAW cells pre-stained with carboxyfluorescein diacetate succinimidyl es ter (CFSE). They were treated with different concentrations of chlordecone (including 0.1 µM, 1 µM and 10 µM) and estradiol (0.1 µM, 1 µM) for up to four days, and the CFSE staining intensity was measured every 24 hours by flow cytometry. Th e results of control, 10 µM chlordecone, 1 µM estradiol treatment on day 0, 1, 2 and 4 were shown in Figure 5-2, Panel A. The

PAGE 136

122 intensity peaks from different treatment on day 2 were merged in Figure 5-2, Panel B, and they showed a complete overlap, suggesti ng neither chlordecone nor estradiol in our experiment affected RAW cell proliferation. Both Clordecone and Estradiol Enhanced IL-10 Secretion in RAW 267.4 Cells To test chlordecone and estr adiol effect on macrophage IL -10 secretion, a series of concentrations including 0.01 µM, 0.1 µM, 1 µM, and 10 µM chlordecone and estradiol were tested. The cytokine concentration in the supernatant was measured by ELISA. Both chlordecone and estradiol treatment showed a concentration-dependent increase in IL-10 secretion (Figure 5-3). Treatment w ith 0.01 µM chlordecone did not change the IL-10 level, while 1 uM chlordecone tr eatment caused more than 100% increase compared with the vehicle control. At the highest concentration 10 µM, chlordecone showed a clear decrease in the IL-10 level co mpared to the 1 µM treatment, but it was still higher than control levels. Estradiol also caused a concentra tion-dependent increase with almost 100% increase in 10 µM estradiol treatment. Chlordecone Exposure Decreased Prolactin Secretion in Marked Contrast to the Effects of Estradiol Recent studies have emphasized an immunosti mulatory role for prolactin in the effects of estrogens (Elbourne et al., 1998; Grimaldi et al., 20 05). It has been shown that estrogen, in the absence of its effects of prolactin, is im munosuppressive (McMurray, 2001). Our previous data suggested that chlo rdecone was weakly es trogenic at the doses used based on the uterus hypertrophy assay (C hapter 3, Figure 3-2, Pa nel D). As another measurement of the estrogenic effects of chlordecone, we decided to measure serum levels of prolactin from mice tr eated with chlordecone, estradio l or control. To evaluate likely steady-state measurements and to correl ate with our other data , sera were collected

PAGE 137

123 from ovariectomized mice 5-6 weeks post-implan tation and sent to a reference laboratory (National Hormone and Peptide Program, Harbor–UCLA Medical Center, Torrance, California, USA). A radioimmunoassay (RIA) te st was used to test the prolactin level and results were shown in Figure 5-4, Panel A. As expected, estradiol treatment caused a dramatic 10to 20-fold increase in prolactin levels, with an average concentration of 500 ng/mL. In marked contrast and unexpect edly, chlordecone-treated mice showed a significant and consistent dose-dependent decreas e in prolactin levels. These differences were statistically significant. Because of these unexpected findings, we decided to also look at prolactin levels in serum sample s obtained from a cohort of unovariectomized BALB/c mice that had been tested for the de velopment of autoimmunity when exposed to chlordecone or estradiol. Long-term trea tment did not result in the development of autoimmunity in this non-autoimmune-prone st rain (Sobel et al., in press). Because these mice were not ovariectomized, there was no es tradiol-treated group. In ovary-intact BALB/c mice, chlordecone e xposure showed a slight tendency to decrease serum prolactin levels, but this did not achieve statistical significan ce (Figure 5-4, Panel B). Estradiol, but Not Chlordecone, Signific antly Enhanced Prolactin Receptor Gene Expression in B Cells and CD4 T Cells As prolactin’s effects on lymphocytes ar e mainly through the prolactin receptor (Leonard et al., 1998), we tested the prolac tin receptor mRNA level in magnetic-bead isolated splenic B and CD4 T cells by r eal-time PCR. Estradiol exposure in vivo produced an almost three-fold increase in prol actin receptor in B ce lls and more than twofold increase in CD4 T cells compared with the placebo group, while the 5 mg chlordecone treatment showed no influence on prol actin receptor levels on either type of lymphocytes (Figure 5-5).

PAGE 138

124 Discussion As defective clearance of apoptotic cells can contribute to break of peripheral tolerance (Mok et al., 2003) and is an impor tant pathway in lupus pathogenesis, the impairment of phagocytic function of m acrophages by chlordecone and estradiol was important to assess. The defective clearance of apoptotic cells leads to the accumulation of apoptotic debris, and further raises th e amount of autoantigens. The increased apoptotic bodies which in a normal situa tion should be eliminated without any inflammatory response, can now lead to inflammatory reactions. The presumable mechanism of defective clearan ce to break tolerance is that those increased autoantigens might activate antigen presenting cells (APCs) , such as dendritic cells and B cells, and those APCs further provide survival signa l to autoreactive B cells which should be otherwise deleted. Although es tradiol treatment has been reported to increase the clearance of IgG-coated eryt hrocytes by enhancing macrophage receptor affinity for the Fc portion of immunoglobulin G (Friedman et al. 1985), there are clear differences between apoptotic cells and immune complexes. In our experiment, we observed reduced clearance of apoptotic cells by peritoneal macrophages in 5 mg chlordeconeor 0.05 mg estradiol-treated groups, and also decreased apoptosis of B cells which indicated increased amount of autoreactive B cells. A partially redundant and promiscuous system of receptors, such as integrins, scavenge r receptors, CR3 and CR4, calreticulin, CD14, a phosphatidylserine (PS) – specific receptor a nd Mer receptor ensure s an efficient and rapid uptake of apoptotic cel ls (Pittoni et al., 2002; Fadok et al., 2000; and Fadok et al., 2001). The effects of chlordecone and estr adiol on these receptors are unknown. We speculated that it is possible that chlordec one and estradiol might dysregulate some the receptors and therefore affect clearance function. In our experiment, B1 cells showed

PAGE 139

125 less capability in engulfing a poptotic cells than macrophage s, and neither chlordecone nor estradiol affected this function. The mechanism and physiological meaning of B1 cell clearance of apoptotic cells is still not clear. Both chlordecone and estradiol enhan ced RAW cell secretion of IL-10, although the highest dose of chlordecone at 10 µM s howed a decrease, but the secreted level was still higher than the control. This decrease may be due to some adverse effects that affected cytokine secretion at high concentr ations. The previous survival test and proliferation test using CFSE showed th at 10 µM chlordecone was the upper limit concentration resulting in no reduction in RA W cell survival and pr oliferation, but it is still possible that this concentration of chlord econe can affect the secr etion of cytokines. Although it is an in vitro experiment, it might bridge th e connection between chlordecone treatment and increased anti-dsDNA titers seen in vivo . Llorente et al. (1995) showed that after transferring periphe ral blood monocuclear cells fr om lupus patients to SCID mice, high titers of anti-dsDNA IgG were induc ed in serum. However the level of IgG was decreased when the mice were treated wi th anti-IL-10 antibody at the same time. Furthermore, Levy et al. 1994 has reported th at IL-10 can prevent spontaneous death of germinal center B cells by induction of Bcl-2 expression. We found both reduced germinal center B cell apoptosis and increase d Bcl-2 levels in chlordecone and estradiol treated groups, which were consistent with this report. Estradiol can stimulate prolactin secret ion, which has been shown in both the mouse model and humans (McMurray, 2001; Frantz et al., 1978; Franks et al., 1983). In our experiment, estradiol significantly incr eased serum prolactin levels, which is consistent with previous reports. Furthermor e, prolactin receptor expression on splenic B

PAGE 140

126 cells and CD4 T cells was also significan tly increased, suggesting the communication between lymphocytes and circul ating prolactin was enhanced. Since high prolactin and low estradiol could increase autoimmunity, wh ile high estradiol and low prolactin did not accelerate autoimmunity (Elbourne et al. 1998), it would appear that in mice, estradiol affects autoimmunity predominantly through the prolactin pathway. Because of the reported estrogenic effects of chlordecone (Hodges et al., 2000; Okubo et al., 2004) and the modest but reproducible increase in uterin e weight that we saw, we had anticipated that chlordecone would cause a modest to m oderate increase in serum prolactin. We were thus very surprised to see that chlo rdecone caused a dose-dependent decrease in serum prolactin levels. Although serum prolactin levels were decr eased, it was still possible that the prolactin pathway was being induced in cells of the immune system through chlordecone, perhaps through upregulation of the prolactin receptor and lo cal secretion of prolactin. For this reason, we also tested the leve l of expression of prolactin receptor on lymphocytes by real-time PCR. The primer th at we designed for th e experiment covered a fragment of gene between exon 6 and e xon 7, which encodes all four isoforms of prolactin receptor protein. Chlordecone showed no influence on B cell and CD4 T cell overall prolactin receptor expression at the mRNA level. Unfortunatly we did not test the gene expression of different prolactin recepto r isoforms. However, we think it is less likely that chlordecone would upregulate on e isoform while down-regulating another one, and make the overall outcome unchanged. There is little literature to date discussing the effects of different prolactin receptor isoforms on the immune system. Taken together, these data strongly suggest that the prolactin pathway did not play an important role in

PAGE 141

127 acceleration of autoimmunity by ch lordecone. In fact, our data would have predicted that chlordecone would have inhibited autoimmunity. This is the most st riking difference so far between chlordecone and estradiol and suggests that chlordecone must function through potent non-estrogenic pathways. Th e prolactin receptor expression in the estradiol-treated group was signi ficantly increased in both B and CD4 T cells, this is consistent with enhanced signaling transduc tion in the prolactin pathway. Previous studies have reported that estradiol can incr ease prolactin receptor expression in both mice and humans (Mizoguchi et al., 1997; Tsen g et al., 1998; Leondire s et al., 2002), but in different cell types. We found the increased prolactin re ceptor expression by estradiol on lymphocytes for the first time. However, whether the reduction of prolac tin by chlordecone only happens in the NZB/NZW F1 mouse model, and whether ovar iectomy affects the process is not yet clear. We have tested chlordecone’s effects on serum prolactin level in the unovariecotmized BALB/c mouse model. In this test, placebo, 1 mg and 5 mg chlordecone were used, and each group in cluded 9 – 10 mice. The results were inconsistent (Figure 5-4, Panel B) . There are at least two possi bilities for these reults: 1) chlordecone did not affect prolactin leve ls in the BALB/c mouse model, and the reduction caused by chlordecone was not signi ficant and might come from variations among samples. Under this possibility, if the number of the mice were large enough, there should be still no difference. 2) chlo rdecone did reduce the serum prolactin level, and the fact that the reduction effect was not so clear comes from the endogenous estradiol influenced by the estrus cycle. Un fortunately, we did not have overiectomized BALB/c mice treated by chlordecone, making it impossible to distinguish between these

PAGE 142

128 two possibilities. We have also te sted overiectomized C57B/6 mice, and Sle1.Sle2.Sle3 B6 triple congenic mice (data not shown). Mi ce treated with 5 mg chlordecone showed a slightly lower prolactin level compared with the control group, but there were only four mice in each group with big variations among sa mples, so we can not reach conclusions on these mouse models. There are several interpreta tions for chlordeconeÂ’s reduc tion of prolactin level in the NZB/NZW F1 mouse model of lupus. It is possible that chlordecone produced an antagonist effect on the pituitary gland by stimulating dopamine receptors, as the prolactin inhibitor bromocriptin e does (Peeva et al., 2005). It is interesting that Ho et al. 1981 reported that chlordecone affected the dopaminergic pa thway and its interaction with other neurotransmitter systems. This might provide some clues to chlordeconeÂ’s effects on the serum prolactin level. It is also possible that chlordecone reduced the prolactin level by indirectly affecting transcri ption factors, such as pituitary-1 (Pit-1), which plays a critical role in prolactin a nd growth hormone secretion by the pituitary gland. Chlordecone might also increase the de gradation process of prolactin, which in a normal situation, represents a physiological mechanism to prevent over-accumulation of prolactin. This may be a concentration at which chlordecone shows equivalent estrogenicity with estradiol. Although there is no doubt that chlordecone caused a dose-dependent reduction of serum prolactin levels, how this reduction affects the immune system remains unknown. Elbourne et al. 1998 reported that mice with high estradiol and low prolactin level showed suppressed immune response, shown as lower cumulative albuminuria, anti-DNA antibody and higher survival compared with control mice with hormonal manipulations

PAGE 143

129 designed to produce normal estrogen and norma l prolactin. An earlier study by Nagy et al. (1978) showed that hypophysectomized rats (pituitary depletion) present symptoms of immunosuppression, such as lymphopenia and impaired humoral and cellular antigen responses. However, it is also interesting that Bouchard et al. (1999) reported that immune system development and function proc eed normally in the absence of prolactinmediated signaling using a prolactin recepto r knockout mouse model, and that prolactin pathways are not essential for immunomodulation in vivo . However, it is known that lymphocytes can produce prolactin or prol actin-like proteins by themselves (BenJonathan et al., 1996; Bole-Feysot et al., 1998 ; Hiestand et al., 1986). Prolactin receptor knockout shuts down the communication of lymp hocytes with outside prolactin, but the endogenous prolactin produced inside the lymphocytes, although in a small amount, might be enough for lymphocyte development and function. Although there is still argument as to wh ether reduced prolactin will suppress the immune system or not, there is no doubt that the reduction will not stimulate the immune response, which means that there must be other potential mechanisms for chlordecone to accelerate autoimmunity different from estradiol.

PAGE 144

130 A. Macrophage clearance of apoptotic cells B. B1 cell clearance of apoptotic cells Figure 5-1. Chlordecone and es tradiol treatments impaired the clearance of apoptotic cells by peritoneal macr ophages but not B1 cells ex vivo. Both 5 mg chlordecone and estradiol caused a st atistically significant decrease in macrophage clearance of 7-AAD stained apopt otic cells (p<0.05) (A). Neither chlordecone nor estradiol affected the clearance of B1 cells (B).

PAGE 145

131 day 4 day 2 day 1 day 0 A. RAW cell proliferation test B. Merged proliferation result on day 2 Figure 5-2. Chlordecone or estr adiol treatment did not affect the proliferation of RAW 267.4 cell line. One million RAW cells, wh ich were pre-stained with CFSE, were treated with vehicle control, 0. 1 µM, 1 µM and 10 µM chlordecone or 0.1 µM and 1 µM estradiol for up to four days. The CFSE staining intensity was measured by flow cytometry on day 0, 1, 2 and 4. The results from vehicle control, 10 µM chlordecone and 1 µM estradiol treatment were shown in (A). The merged intensity peaks on day 2 was shown in (B), and they showed a complete merge, indicating no difference in proliferation. CFSE010110210310410 Count140 130 120 110 100 90 80 70 60 50 40 30 20 10 0 Chlordecone Vehicle Control Estradiol Chlordecone Vehicle Control Estradiol 010110210310410 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 Chlordecone 010110210310410 130 120 110 100 90 80 70 60 50 40 30 20 10 0 Estradiol 010110210310410 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0 Vehicle Control

PAGE 146

132 0 20 40 60 80 100 120 140 160 180 10 uM Chlordecone 1 uM Chlordecone 0.1 uM Chlordecone 0.01 uM Chlordecone 10 uM estradiol 1 uM estradiol 0.1 uM estradiol 0.01 uM estradiol EtOH control pg/ml Figure 5-3. Comparison of chlordecone and estr adiol effects on the secretion of cytokine IL-10 by the RAW 267.4 cell line. The re sult was the average of duplicated experiments. Both chlordecone and estradiol treatment caused a concentration-dependent increase in IL10 secretion compared with controls. The highest concentration of chlordec one, 10 µM caused a decrease in IL-10 secretion compared with the secretion level of cells treated with 1 µM chlordecone, but was still hi gher than the controls.

PAGE 147

133 A. Serum prolactin level in ovariectomized NZB/NZW F1 mice B. Serum prolactin level in unovar iectomized BALB/c mice Figure 5-4. Comparison of chlordecone and estr adiol effects on serum prolactin level in ovariectomized NZB/NZW F1 mice a nd unovariectomized BALB/c mice. (A) In ovariectomized NZB/NZW F1 mi ce, estradiol treatment caused a remarkable increase in serum prolactin level; while chlordecone treatment decreased prolactin level in a dose-depe ndent way. (B) In unovariectomized BALB/c mice, both 1 mg and 5 mg ch lordecone treatment decreased serum prolactin level compared with plac ebo treatment, but neither reached statistical significance.

PAGE 148

134 A. Prolactin receptor expression on B cells B. Prolactin receptor expression on CD4 T cells Figure 5-5. Comparison of chlordecone and estr adiol effects on the gene expression level of prolactin receptor on splenic B a nd CD4 T cells. Estradiol, but not chlordecone treatment, significantly increased prolactin receptor gene expression on purified B cells (p< 0.01) and CD4 T cells (p<0.05).

PAGE 149

135 CHAPTER 6 GENERAL DISCUSSION AND PERSPECTIVE Autoimmune disorders are the result of a co mbination of factors including genetic, environmental, virus, and hormonal influen ces (D'Cruz 2000; Roubinian et al. 1978). Recently, there has been considerable intere st in determining whether exposure to chemicals in the environment is an important factor influencing the incidence or severity of autoimmune diseases (He ss, 2002; Mayes, 1999; Van Love ren et al., 2001). Mercury, trichloroethylene, and silica are among the e nvironmental agents that have been linked with autoimmune disease, through epidemio logical studies, experiments with animal autoimmune models, or both (Hess, 2002; Park s et al., 2002; Pollard et al., 1999; Via et al., 2002). Other chemical exposures, such as to the isoprenoid alkane pristane (2,6,10,14-tetramethylpentadecane), can cause autoimmunity in virtually any mouse strain (Satoh et al., 1994). A common th eme amongst these agents is that they may create novel epitopes on self antig ens (mercury and trichloroethylene) or they may act as an adjuvant (silica and pristane) (Hess, 2002). It has long been hypothesized that environmental factors influence the onset and course of autoimmune diseases. Despite this , the number of chemicals clearly shown to influence autoimmunity is relatively small. Although a number of potential mechanisms can be postulated, they are thought to fall into three ge neral categories (Rao and Richardson 1999; Sobel et al., 2005 ). The first category is one in which the chemical alters self antigen such that it appears fo reign to the immune system. This can occur when small molecules act as haptens or if exposure can cause novel cleavage fragments

PAGE 150

136 to which the immune system was ignorant. Heavy metals, such as mercury, may be an example of this pathway. Mercury may modi fy self proteins by interacting with the sulfhydral groups on indigenous proteins. Th e mercuric-thiol interaction results in modification of the molecule and antigenic properties of the self proteins, which is normally ignored by the immune system (He ss et al., 2002). The second category is one in which the chemical prevents the central to lerance of autoreactive T or B cells and is represented by procainamide hydroxylamin e (PAHA). Most patients treated with prolonged procainamide therapy develop anti-denatured DNA (dDNA) (Blomgren et al., 1972) and anti-histone antibodies (Fritzler et al., 1978). Kretz-Rommel et al. (1998) also proved that a single injection of PAHA into the thymus of (C57BL/6 x DBA/2)F1 mice resulted in the rapid appearance of IgM an ti-dDNA and anti-histone antibodies, and after a second intrathymic inje ction, chromatin-specific T cells were detected in the spleen, and IgG anti-chromatin antibodies rose to levels typically seen in spontaneous models of murine lupus. The third category involves alte ration of gene expression. Because many of the OCPs have been shown to have es trogenic effects, we decided to test representative estrogenic OCPs in the well-characterized NZB/NZW F1 model of murine lupus. We originally proposed the novel hypot hesis that organochlorine pesticide chlordecone may accelerate SLE through their estrogenic effects, as chlordecone has been shown to have estrogenic effects (Shelby et al., 1996, Sobel et al., 2005), and estrogen has long been proposed to be a risk factor for SLE in humans (Buyon et al., 2005). However, at doses that can clearly accelerate autoimmunity, chlordecone showed only minor estrogenic effects on the reproducti ve system, as measured by wet uterine

PAGE 151

137 weight. These data strongly suggested that chlordecone functioned through nonestrogenic pathways, but the possibility remain ed that cells of th e immune system may respond to the estrogenic effects of chlordec one differently than cells of the reproductive system. Therefore, the overall objective of this dissertation was to better characterize chlordecone effects on cellular and gene expression in immunocytes, as well as characterize some other importa nt checkpoints in tolerance. A second objective was to further clarify the mechanisms of chlordec one on the accelerati on of SLE by directly comparing chlordecone effects w ith those of estradiol. From the information gathered by these studies, we seek to establish a model to test other pesticides which have also shown the potential to accelerate autoimmunity, and fi nally to see if there is a general pattern that would point to a mechanistic basis for our observations. To achieve these goals, low doses of ch lordecone and estradiol were delivered subcutaneously to overiectomized lupus prone NZB/NZW F1 mice for six weeks by continuous-release pellets. C ontrol mice received pellets with vehicle alone. Mice were then euthanized, and multiple experiments we re conducted. Part of the comparison of effects between chlordecone and estradiol was focused on the splenic cells, as it is one of the most important secondary lymphoid organs in the immune system and has been wellstudied in murine lupus. Phenotypic changes on immunocytes (measured mostly by flow cytometric technique), important B cell gene expression (tested by quantitative real-time PCR), and cytokine levels (s ecreted by purified CD4 T cells) were reported in chapter 3 and chapter 4. The other part of the co mparison focused on the effects on macrophage functions and a peptide hormone prolactin, due to their importa nce in lupus pathogenesis.

PAGE 152

138 Those experiments were shown in chapter 5. Table 6-1 includes a summary of the overall results in this dissertation. The outcome from these experiments showed a partial overlap in effects between chlordecone and estradiol. On one hand, th ey both expanded the number of germinal center B cells and reduced lymphocytes apopt osis, which might be responsible for the increased survival of auto reactive lymphocyte. Furthe r studies on gene expression revealed that the escape of presumably auto reactive lymphocytes from negative selection might be mediated by enhancing Bcl-2 and Shp-1 expression (Grimaldi et al., 2002). Although the findings have been published, ther e are clear limitations as to how far these data can be interpreted. For example, Grimaldi et al did not establ ish a direct connection between up-regulation of Bcl-2 and Shp-1 and the loss of tolerance. To prove this directly, one method is to use transgenic mouse models with Bcl-2 and/or Shp-1 overexpression. Kuo et al. (1999) proved that Bcl-2 overexpression led to the expression of anti-DNA B cells. However, there is st ill no transgenic mouse model with Shp-1 overexpression. Another way to test this hypot hesis is to use RNAi techniques to knock down the expression of Bcl-2 and/or Shp-1 b ack to the expression levels of control groups. If these two molecules are directly responsible for the br eaking of tolerance, after knocking down their expression, the tolerance should be restored. Both chlordecone and estadiol reduced macrophage clearance of apoptotic cells, which is an important function of macropha ges and may provide one of the critical pathways for chlordecone and estradiol in breaking tolerance. Although we have found a reduction in clearance, the mechanisms behind it have not yet been elucidated. A number of possibilities exist. One possibility is that reduction might be due to down-regulation

PAGE 153

139 expression of a receptor or rece potors involved in clearance. We tried to test the protein expression of c-mer, a member of the Axl/Mer /Tyro 3 receptor tyrosine kinases family, on peritoneal macrophages. Mer knock-out mice with a cytoplasmic truncation of Mer had macrophages deficient in the clearance of apoptotic thymocytes (Cohen et al., 2002) and a lupus-like autoimmunity, and the phagocytic deficiency was restricted to apoptotic cells and was independent of Fc receptor-m ediated phagocytosis or ingestion of other particles (Scott et al., 2001). Unfortuna tely, a monoclonal anti-mouse mer antibody marketed for flow cytometric applications was taken off the market, apparently because of specificity problems. One way to test this is to purify peritoneal macrophages by magnetic-beads, and then test the protein ex pression of c-mer by we stern blot, and the gene expression by quantitative re al-time PCR. However, to make this worthwhile would require a more systematic approach in sc reening for a variety of genes involved in clearance mechanisms. On the other hand, despite the similarities shared by chlordecone and estradiol, we feel that there was clear evid ence pointing to different pathways, as some of the most striking phenotypic changes induced by estradio l were not shared by chlordecone. One of these was the marked increase in marginal zone B cells caused by estradiol treatment (Figure 3-12, Panel E). Interestingly, an in creased proportion of ma rginal zone B cells was observed in the lupus-prone NZB mouse model (Wither et al., 2000; Theofilopoulos et al., 1985) and NZB/NZW F1 mouse model (Wither et al., 2000), suggesting some connection of increased marginal zone B ce lls with autoimmunity. Zeng et al. (2000) reported that in NZB/NZW F1 mouse model, marginal zone B cells are responsible for

PAGE 154

140 production of large amounts of anti-DNA antibodi es, as compared to follicular B cells. This built a direct connection of enhanced ma rginal zone with loss of tolerance. Another example is the different influen ce of chlordecone and estradiol on CD4 T cells. There were several effects on CD4 T ce ll by estradiol that were not shared with chlordecone: 1) Estradiol had a stronger im pact on the expression of T cell activation markers; 2) Estradiol changed T cell subset s by increasing the percentage of activated and memory CD4 T cell and reducing the naiv e T cell population; 3) Estradiol increased the percentage of CD25+CD4+ regulatory T ce lls, which might be predicted to maintain tolerance; and 4) Estradiol i nduced the expression of TCR V chain. Chlordecone and estradiol’s effects on CD4 T ce ll cytokine secretion were al so different – although they both enhanced TNFand IL-2 secretion, chlordecone significantly increased proimflammatory cytokines GM-CSF and IFNsecretion, while estr adiol significantly increased the secretion of IL-10, but reduced IL-4 level. The most striking difference, however, wa s that chlordecone and estradiol caused opposite effects on prolactin s ecretion. In contrast to the 10-20 fold increase in circulating prolactin in estr adiol-treated ovariectomized mice, we found a 70% reduction in prolactin secretion with 5 mg chlordec one group. As it has been published that estradiol-treated mice did not have accelerated autoimmunity when prolactin was suppressed, these results strongl y suggest that there must be alternative pathways for chlordecone, distinct from estradiol, that mediate the acceleration of autoimmunity. Our experiments were designed to determ ine whether chlordecone was mediating its effects through estrogenic pa thways. One outcome was to further elucidate the effects of estradiol on the immune system of NZ B/NZW F1 mice, by pathways which were

PAGE 155

141 novel. For example, estradiol treatment caused a significant decrease in the percentage of lymphocytes, a clear enhanced germinal cen ter reactions, increased expression of TCR V 2 and V 8.3, and decreased IFNgene expression on B cells. Although there were previous reports of estradiol’s effects on up-regulating prolactin receptor expression (Mizoguchi et al., 1997; Tseng et al., 1998; Leondires et al ., 2002), we found for the first time that estradiol can also increase the prolactin receptor expression on lymphocytes. Taken together, these data suggest that estradiol might mediat e its effects through enhanced germinal center reactions, increased marginal zone B cells, as well as activated T cells. Most importantly, it significantly activated the prolactin pathway that was thought to be the critical factor in estradio l effects in accelerating autoimmunity. It is very interesting that chlordecone enhanced the germinal center response in NZB/NZW F1 mice. However, this mouse mo del has many limitations. First, the onset of autoimmunity is variable so that a rela tively large number of mi ce are needed to be certain of the effect. Sec ond, it is difficult to identify autoimmune B cells specifically and thereby study the important checkpoints that maintain to lerance in germinal center responses. These problems can be at least partially addressed by using Ig transgenic models of NZB/NZW F1 mice, a few of wh ich have been developed by Dr. Tony Marion. These include NZB/NZW F1 mice transgenic for the conventional µ heavy chain 3H9 and the VH3H9R transgene inserted as a site-directed JH knock-in. The particular interest in these two mouse models is that despite the ability of both transgenes to encode antidsDNA, the NZB/NZW F1 mice transgenic fo r the conventional c onstruct maintained tolerance, while the knock-in did not (Steev es et al., 2004). This would provide the possibility to test whether chlordecone can cau se a break in toleran ce (conventional) or

PAGE 156

142 accelerate autoimmunity (knock-in). Using these transgenic mouse models, several checkpoints in the germinal center responses, such as the central tolerance, follicular exclusion, T cell receptor revision, decreased cl earance of apoptotic cells and decreased apoptosis, could be further examined followi ng chlordecone and estradiol treatment. Quantitative measurements of B cell gene expression in chlordeconeand estradioltreated mice also revealed a complex relationship. Chlordecone and estradiol are structurally unrelated, and th e proposed connection for thei r biological effects came from their ability to bind to estr ogen receptors. However, the affinity of chlordecone for estrogen receptors is very low (Hammond et al., 1979), which might indicate that there are potential differences in down stream pa thways. Although a microarray screen for gene expression was not used in our experime nt, results of comparisons of global gene expression between another estrogenic compoun d genistein and estradiol on thymocytes, showed only a partial overlap in gene re gulation (Selvaraj et al. 2005). That study revealed the complexity of how estrogeni c compounds may affect the immune system differently from estradiol. Our real-time PCR results comparing chlordecone and estradiolÂ’s regulation on the expression of selected gene s on splenic B cells likewise showed a partial overlap. Major similariti es included increased Bcl-2, Shp-1, and Fas expression, with reduced IFNexpression. On the other ha nd, slight differences were seen in the effects on expression of IL-2 and IL-6. There were also several genes whose expression was unaffected by either chlordecone or estradiol. This might be due to the low dose or timing of our experiments. An im portant distinction that we showed in this work was that estradiol and chlordecone had significantly diffe rent effects on the

PAGE 157

143 prolactin pathway. Global gene expression scre ens could lead to the discovery of new, non-estrogenic pathways being mediated through chlordecone. We found that both chlordecone and estradio l activated splenic B cells, but whether these activated B cells were pa thogenically important and were the direct cause of disease was not directly determined. One method to test this would be to transfer purified B cells from control-, chlordeconeor estradiol-treated donor NZB/ NZW F1 mice into untreated young NZB/NZW F1 mice. The serum levels of antibody and autoantibody of the hosts could be measured at various time points after injection. If the auto antibody levels were increased and accompanied by accelerated glom erulonephritis, this would indicate that the transferred B cells were pathogenic. One pr oblem for this experiment is that there is no way to trace the injected B cells by us ing the conventional NZB/NZW F1 mouse. However, this could be overcome by using the transgenic 3H9R NZB/NZW F1 mouse as a donor. The B cells of this mouse model can be distinguished by using a specific monoclonal antibody. Another important area that needs to be further clarifie d is the direct effects of chlordecone on the kidney. Previous histol ogical study showed that both chlordecone and estradiol accelerated the development of glomerulonephritis, which is the most serious manifestation of lupus in th e NZB/NZW F1 mouse model, and the immunohistofluorescence staini ng revealed similar intensity of IgG immune-complex deposition 8 weeks after implantation (Sobel et al., 2005). However, the pathways behind the same outcome might be different. Estradiol might increase titers of antoautibodies significantly, while chlordecone treatment might cause changes in the kidney so that autoantibody deposition is e nhanced. It was previously shown that

PAGE 158

144 chlordecone exposure did not affect renal f unction in normal BALB/c mice (Sobel et al., in press). However, since these mice did not break tolerance, it is formally possible that differences would appear only under stressed co nditions. One way to test this hypothesis would be to inject a specific glomerulopath ic antibody intravenously into mice which had been previously implanted for a short time with chlordecone or estradiol. One week after injection, the mice would be s acrificed and the kidneys w ould be stained for immune complex deposits. If chlordecone caused more antibody deposition, mice in this group should have stronger staining than in the estradiol group. Overall, our results suggest a complicated role for the effects of chlordecone on autoimmune disease in NZ B/NZW F1 mice. Despite in vivo studies showing acceleration of disease comparable to estrogen, there was only partial overlap in some of the specific phenotypic changes in T and B cells known to be associated with estrogen exposure, and some of these were attenuated. Other effect s of estrogen, such as the marked increase in marginal zone B cells, were not seen at all with chlordecone exposure. Moreover, unlike estradiol, the effects seen with chlordecone were unlikely mediated by prolactin. Taken together, our data suggest an important role for non-estrogenic pathways in the acceleration of autoimm unity with exposure to chlordecone.

PAGE 159

145 Table 6-1. Summary of the results in the studies. Chlordecone treatment Estradiol treatment Result from this study Reports from literature Body weight No change Consistent Uterus weight Slightly Consistent Spleen weight Consistent Splenic B cell percentage No change Novel Splenic CD4+ T cell percentage No change Consistent Splenic CD8+ T cell percentage No change Consistent CD69 Novel CD44 Consistent B7.2 Consistent I-A(d) Novel GL-7 Novel CXCR4 Consistent CXCR5 Novel ICAM-1 Novel VCAM-1 Consistent Bcl-2 Consistent Immature B cell (CD24+) No change Consistent T1 B cell (CD21-CD24+) No change Consistent T2 B cell (CD21+CD24+) No change No change Consistent Follicular B cell (CD21-CD24-) No change No change Consistent Marginal Zone B cell (CD21+CD24-) No change Consistent Mature plasma cell (CD138+B220-) No change Consistent Immature plasma cell (CD138+B220+) No change No change Novel Splenic B cells Apoptosis (no stimulation) Consistent

PAGE 160

146 Table 6-1. Continued Chlordecone treatment Estradiol treatment Result from this study Reports from literature Apoptosis (1 µg/mL LPS) Consistent Proliferation index (1 µg/mL LPS) No change No change Inconsistent Shp-1 (gene) No change Inconsistent Bcl-2 (gene) No change Inconsistent Fas (gene) No change Inconsistent Fas ligand (gene) No change No change Inconsistent IFN(gene) No change Inconsistent Fc RIIb (gene) No change No change Novel Ly5 (gene) No change No change Inconsistent TNF(gene) No change No change Paradoxical IL-2 (gene) No change No change Paradoxical IL-6 (gene) No change No change Inconsistent Splenic B cells Prolactin receptor (gene) No change Consistent CXCR5 Novel CXCR4 Novel ICAM-1 Novel VCAM-1 Consistent Fc RIIb No change No change Novel Germinal center B cells (GL7+CD19+) Apoptosis Consistent CD69 No change Consistent CD44 No change Consistent Bcl-2 Consistent TCR V 2 No change Novel TCR V 8.3 No change Novel TCR V 4 No change No change Novel TCR V 8 No change No change Novel Prolactin receptor No change Consistent Apoptosis (no stimulation) Inconsistent Splenic CD4+ T cells Apoptosis (CD3 and 2 µg/mL CD28) No change Inconsistent

PAGE 161

147 Table 6-1. Continued Chlordecone treatment Estradiol treatment Result from this study Reports from literature Apoptosis (CD3 and 5 µg/mL CD28 No change Inconsistent Proliferation index (CD3 and 2 µg/mL CD28) No change No change Inconsistent Proliferation index (CD3 and 5 µg/mL CD28) No change No change Inconsistent IFNNo change Inconsistent TNFNo change Paradoxical GM-CSF No change Consistent IL-4 No change Consistent IL-2 No change Paradoxical Splenic CD4+ T cells IL-10 No change Paradoxical Clearance of apoptotic cells Inconsistent Macrophage IL-10 secretion Paradoxical Serum prolactin (in NZB/NZW F1 mice) Consistent Serum prolactin (in ovary-intact BALB/c mice) No change

PAGE 162

148 LIST OF REFERENCES Allen AL, Koller LD, Pollock GA. (1983) E ffect of toxaphene exposure on immune responses in mice. J Toxico l Environ Health. 11(1): 61-69. Allen CD, Ansel KM, Low C, Lesley R, Tamamura H, Fujii N, Cyster JG. (2004) Germinal center dark and light zone organization is mediated by CXCR4 and CXCR5. Nat Immunol. 5(9):943-952. Allen RD, Marshall JD, Roths JB, Sidman CL. (1990) Differen ces defined by bone marrow transplantation suggest that lpr and gld are mutations of genes encoding an interacting pair of mol ecules. J Exp Med. 172(5):1367-1375. Allen SH, Sharp GC, Wang G, Conley C, Takeda Y, Conroy SE, Walker SE. (1996) Prolactin levels and antinuclear antibody profiles in women tested for connective tissue disease. Lupus. 5(1):30-37. Ansar Ahmed S, Dauphinee MJ, Talal N. (1985) Effects of short-term administration of sex hormones on normal and autoimm une mice. J Immunol. 134(1):204-210. Ansell SM, Bedhesi S, Ruff B, Mahomed AG, Richards G, Mer M, Feldman C. (1996) Study of critically ill patien ts with systemic lupus erythematosus. Crit Care Med. 24(6):981-984. Aoki CA, Borchers AT, Li M, Flavell RA , Bowlus CL, Ansari AA, Gershwin ME. (2005) Transforming growth factor be ta (TGF-beta) and autoimmunity. Autoimmun Rev. 4(7):450-459. Aringer M, Smolen JS. (2003) SLE Complex cytokine effects in a complex autoimmune disease: tumor necrosis factor in syst emic lupus erythematosus. Arthritis Res Ther. 5(4):172-177. Armitage JO. (1998) Emerging applicati ons of recombinant human granulocytemacrophage colony-stimulati ng factor. Blood. 92(12):4491-4508. Azizah MR, Kuak SH, Ainol SS, Rahim MN, Normaznah Y, Norella K. (2004) Association of the tumor necrosis fa ctor alpha gene polymorphism with susceptibility and clinical-immunologi cal findings of systemic lupus erythematosus. Asian Pac J A llergy Immunol. 22(2-3):159-163. Bach JF. (1995) Organ-specific auto immunity. Immunol Today. 16(7):353-355.

PAGE 163

149 Bagby GC Jr, Dinarello CA, Wallace P, Wagne r C, Hefeneider S, McCall E. (1986) Interleukin 1 stimulates granulocyte macrophage colony-stimulating activity release by vascular endothelial ce lls. J Clin Invest. 78(5):1316-1323. Bajorath J. (2000) Molecular organizati on, structural featur es, and ligand binding characteristics of CD44, a highly variable cell surface glycoprotein with multiple functions. Proteins. 39(2):103-111. Barrett C, Neylon N, Snaith ML. (19 86) Oestrogen-induced systemic lupus erythematosus. Br J Rheumatol. 25:300-301. Baumann I, Kolowos W, Voll RE, Manger B, Gaipl U, Neuhuber WL, Kirchner T, Kalden JR, Herrmann M. (2002) Impaired uptake of apoptotic cells into tingible body macrophages in germinal centers of patients with systemic lupus erythematosus. Arthritis Rheum. 46(1):191-201. Bebo BF Jr, Fyfe-Johnson A, Adlard K, Beam AG, Vandenbark AA, Offner H. (2001) Low-dose estrogen therapy ameliora tes experimental autoimmune encephalomyelitis in two different inbred mouse strains. J Immunol. 166(3):20802089. Ben-Jonathan N, Mershon JL, Allen DL, Stei nmetz RW. (1996) Extrapituitary prolactin: distribution, regulation, functions, and clinical aspects. Endocr Rev. 17:639–669. Billiau A, Heremans H, Vermeire K, Matthys P. (1998) Immunomodulatory properties of interferon-gamma. An update. Ann N Y Acad Sci. 856:22-32. Bischof RJ, Zafiropoulos D, Hamilton JA, Campbell IK. (2000) Exacerbation of acute inflammatory arthritis by the colony-s timulating factors CSF-1 and granulocyte macrophage (GM)-CSF: evidence of m acrophage infiltration and local proliferation. Clin Exp Immunol. 119(2):361-367. Blomgren SE, Condemi JJ, Vaughan JH . (1972) Procainamide-induced lupus erythematosus. Clinical and laborat ory observations. Am J Med. 52(3):338-348. Bole-Feysot C, Goffin V, Edery M, Binart N, Kelly PA. (1998) Prolactin (PRL) and its receptor: actions, signal transduction pa thways and phenotypes observed in PRL receptor knockout mice. Endocr Rev. 19(3):225-268. Bolland S, Ravetch JV. (1999) Inhibitory pathways triggered by ITIM-containing receptors. Adv Immunol. 72: 149-177. Bolland S, Ravetch JV. (2000) Spontaneous autoimmune disease in Fc(gamma)RIIBdeficient mice results from strain-speci fic epistasis. Immunity. 13(2):277-285. Borrello MA, Palis J, Phipps RP. (2001) Th e relationship of CD5+ B lymphocytes to macrophages: insights from normal biphe notypic B/macrophage cells. Int Rev Immunol. 20(1):137-155.

PAGE 164

150 Bouchard B, Ormandy CJ, Di Santo JP, Kelly PA. (1999) Immune system development and function in prolactin receptor-def icient mice. J Immunol. 163(2):576-582. Brennan DC, Yui MA, Wuthrich RP, Kelley VE . (1989) Tumor necrosis factor and IL-1 in New Zealand Black/White mice. Enhan ced gene expression and acceleration of renal injury. J Immunol. 143(11):3470-3475. Brown KS, Blair D, Reid SD, Nichols on EK, Harnett MM. (2004) FcgammaRIIbmediated negative regulation of BCR signalli ng is associated with the recruitment of the MAPkinase-phosphatase, Pac-1, and the 3'-inositol phosphatase, PTEN. Cell Signal. 16(1):71-80. Buyon JP, Petri MA, Kim MY, Kalunian KC, Grossman J, Hahn BH, Merrill JT, Sammaritano L, Locksin M, Alarcon GS , Manzi S, Belmont HM, Askanase AD, Sigler L, Dooley MA, Von Feldt J, Mc Cune WJ, Friedman A, Wachs J, Cronin M, Hearth-Holmes M, Tan M, Licciardi F. (2005) The effect of combined estrogen and progesterone hormone replacement therapy on disease activity in systemic lupus erythematosus: A randomized trial. Ann Int Med. 142:953-962. Bynoe MS, Grimaldi CM, Diamond B. (2000) Estrogen up-regulates Bcl-2 and blocks tolerance induction of naive B cells. Proc Natl Acad Sci U S A. 97(6):2703-2708. Cambier JC, Fong D, Tamir I. (1999) The unexpected complexity of Fc gamma RIIB signal transduction. Curr T op Microbiol Immunol. 244:43-55. Campbell IK, Bendele A, Smith DA, Ham ilton JA. (1997) Granulocyte-macrophage colony stimulating factor exacerbates co llagen induced arthritis in mice. Ann Rheum Dis. 56(6):364-368. Campbell IK, Novak U, Cebon J, Layton JE , Hamilton JA. (1991) Human articular cartilage and chondrocytes produce hemopoi etic colony-stimulating factors in culture in response to IL -1. J Immunol. 147(4):1238-1246. Carmines EL, Carchman RA, Borzelleca JF . (1979) Kepone: cellu lar sites of action. Toxicol Appl Pharmacol. 49(3): 543-550. Carruba G, D'Agostino P, Miele M, Calabro M, Barbera C, Bella GD, Milano S, Ferlazzo V, Caruso R, Rosa ML, Cocciadiferro L, Campisi I, Castagnetta L, Cillari E. (2003) Estrogen regulates cytokine production and apoptosis in PMAdifferentiated, macrophage-l ike U937 cells. J Cell Biochem. 90(1):187-196. Carter RH, Tuveson DA, Park DJ, Rhee SG , Fearon DT. (1991) The CD19 complex of B lymphocytes. Activation of phospholipas e C by a protein tyrosine kinasedependent pathway that can be enha nced by the membrane IgM complex. J Immunol. 147(11):3663-3671.

PAGE 165

151 Chen JR, Plotkin LI, Aguirre JI, Han L, Jilka RL, Kousteni S, Bellido T, Manolagas SC. (2005) Transient versus sustained phos phorylation and nuclear accumulation of ERKs underlie anti-versus pro-apoptotic effects of estrogens. J Biol Chem. 280(6):4632-4638. Chen W, Jin W, Tian H, Sicurello P, Fra nk M, Orenstein JM, Wahl SM. Requirement for transforming growth factor beta1 in controlling T cell apoptosis. J Exp Med. 194(4):439-453. Clynes R, Maizes JS, Guinamard R, Ono M, Takai T, Ravetch JV. (1999) Modulation of immune complex-induced inflammation in vivo by the coordina te expression of activation and inhibitory Fc receptors. J Exp Med. 189(1):179-185. Coggeshall KM. (2000) Positive and negativ e signaling in B lymphocytes. Curr Top Microbiol Immunol. 245(1):213-260. Cohen JH, Danel L, Cordier G, Saez S, Revillard JP. (1983) Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cel ls. J Immunol. 131(6):2767-2771. Cornall RJ, Goodnow CC, Cyster JG. (1995) Th e regulation of self -reactive B cells. Curr Opin Immunol. 7(6):804-811. Cozine CL, Wolniak KL, Waldschmidt TJ. ( 2005) The primary germinal center response in mice. Curr Opin Immunol. 17(3):298-302. Coffey RJ Jr, Bascom CC, Sipes NJ, Grav es-Deal R, Weissman BE, Moses HL. (1988) Selective inhibition of grow th-related gene expression in murine keratinocytes by transforming growth factor be ta. Mol Cell Biol. 8(8):3088-3093. Cohen PL, Caricchio R, Abraham V, Cameni sch TD, Jennette JC, Roubey RA, Earp HS, Matsushima G, Reap EA. (2002) Delayed apoptotic cell clearance and lupus-like autoimmunity in mice lacking the c-mer membrane tyrosine kinase. J Exp Med. 196(1):135-140. Cope AP. (1998) Regulation of autoimmunity by proinflammatory cytokines. Curr Opin Immunol. 10(6):669-676. Correa PA, Gomez LM, Cadena J, Anaya JM. (2005) Autoimmunity and tuberculosis. Opposite association with TNF polym orphism. J Rheumatol. 32(2):219-224. Cory S. (1995) Regulation of lymphocyte su rvival by the bcl-2 gene family. Annu Rev Immunol. 13: 513-543. Cutolo M, Sulli A, Barone A, Seriolo B, Accardo S. (1993) Macrophages, synovial tissue and rheumatoid arthritis. Clin Exp Rheumatol. 11(3):331-339.

PAGE 166

152 Cutolo M, Sulli A, Seriolo B, Accardo S, Masi AT. (1995) Estrogens, the immune response and autoimmunity. Clin Exp Rheumatol. 13(2):217-226. Danel L, Souweine G, Monier JC, Saez S. (1983) Specific estrogen binding sites in human lymphoid cells and thymic cells . J Steroid Bioche m. 18(5):559-563. Das SK, Taylor JA, Korach KS, Paria BC, Dey SK, Lubahn DB. (1997) Estrogenic responses in estrogen recept or-alpha deficient mice reve al a distinct estrogen signaling pathway. Proc Natl A cad Sci U S A. 94(24): 12786-12791. Daston GP, Gooch JW, Breslin WJ, Shuey DL , Nikiforov AI, Fico TA, Gorsuch JW. (1997) Environmental estrogens and re productive health: a discussion of the human and environmental data. Reprod Toxicol. 11(4): 465-481. Datto MB, Li Y, Panus JF, Howe DJ, Xiong Y, Wang XF. (1995) Transforming growth factor beta induces the cyclin-dependent kinase inhibitor p21 through a p53independent mechanism. Proc Natl Acad Sci U S A. 92(12):5545-5549. D'Cruz D. (2000) Autoimmune diseases a ssociated with drugs, chemicals and environmental factors. T oxicol Lett. 112-113:421-432. DeGrendele HC, Estess P, Siegelman MH. ( 1997) Requirement for CD44 in activated T cell extravasation into an inflammatory site. Science. 278(5338):672-675. DeGrendele HC, Kosfiszer M, Estess P, Siegelman MH. (1997) CD44 activation and associated primary adhesion is inducib le via T cell receptor stimulation. J Immunol. 159(6):2549-2553. Deocharan B, Marambio P, Edelman M, Pu tterman C. (2003) Differential effects of interleukin-4 in peptide induced auto immunity. Clin Immunol. 108(2):80-88. Deshpande R, Khalili H, Pergolizzi RG, Mi chael SD, Chang MD. (1997) Estradiol downregulates LPS-induced cytokine produc tion and NFkB activation in murine macrophages. Am J Reprod Immunol. 38(1): 46-54. de Waal Malefyt R, Yssel H, de Vries JE. (1993) Direct effects of IL-10 on subsets of human CD4+ T cell clones and resting T cells. Specific inhibition of IL-2 production and proliferati on. J Immunol. 150(11):4754-4765. Dijstelbloem HM, van de Winkel JG, Kallenberg CG. (2001) Inflammation in autoimmunity: receptors for IgG revi sited. Trends Immunol. 22(9): 510-516. Do Y, Ryu S, Nagarkatti M, Nagarkatti PS. (2002) Role of death receptor pathway in estradiol-induced T-cell apoptosis in vivo. Toxicol Sci. 70(1):63-72. Doody GM, Justement LB, Delibrias CC, Matthews RJ, Lin J, Thomas ML, Fearon DT. (1995) A role in B cell activation for CD 22 and the protein tyrosine phosphatase SHP. Science. 269(5221):242-244.

PAGE 167

153 Dunnett CW. (1980) Pairwise multiple comparisons in the homogeneous variance, unequal sample size case. J Am Stat Asso. 75:789 – 795. Edwards CK 3rd, Zhou T, Zhang J, Baker TJ, De M, Long RE, Borcherding DR, Bowlin TL, Bluethmann H, Mountz JD. (1996) Inhibition of superantigen-induced proinflammatory cytokine production and inflammatory arthr itis in MRL-lpr/lpr mice by a transcriptional inhibitor of TNF-alpha. J Immunol. 157(4):1758-1772. Elbourne KB, Keisler D, McMurray RW. ( 1998) Differential effects of estrogen and prolactin on autoimmune disease in the NZB/NZW F1 mouse model of systemic lupus erythematosus. Lupus. 7(6):420-427. Emlen W, Niebur J, Kadera R. (1994) Accel erated in vitro apoptosis of lymphocytes from patients with systemic lupus er ythematosus. J Immunol. 152(7):3685-3692. End DW, Carchman RA, Dewey WL. (1981) Ne urochemical correlates of chlordecone neurotoxicity. J Toxicol E nviron Health. 8(5-6):707-718. Engelman MW, Hodge SJ, Owen LG. (1979) Cutaneous immunofluorescence in systemic disease. J Ky Med Assoc. 75(6):273-278. Enmark E, Gustafsson JA. (1996) Orphan nucle ar receptors--the firs t eight years. Mol Endocrinol. 10(11): 1293-307. Erickson LD, Lin LL, Duan B, Morel L, Noel le RJ. (2003) A genetic lesion that arrests plasma cell homing to the bone marrow. Proc Natl Acad Sci U S A. 100(22):12905-12910. Estess P, DeGrendele HC, Pascual V, Sieg elman MH. (1998) Functional activation of lymphocyte CD44 in peripheral blood is a marker of autoimmune disease activity. J Clin Invest. 102(6):1173-1182. Fadok VA, Bratton DL, Henson PM. (2001) Ph agocyte receptors for apoptotic cells: recognition, uptake, and conseque nces. J Clin Invest. 108:957–962. Fadok VA, Bratton DL, Rose DM, Pearson A, Ezekewitz RA, Henson PM. (2000) A receptor for phosphatidylserine-specific clea rance of apoptotic cells. Nature. 405:85–90. Fanti P, Monier-Faugere MC, Geng Z, Schm idt J, Morris PE, Cohen D, Malluche HH. (1998) The phytoestrogen genistein reduces bone loss in short-term ovariectomized rats. Osteopor os Int. 8(3):274-281. Fehervari Z, Sakaguchi S. (2004) CD4+ Tr egs and immune control. J Clin Invest. 114(9):1209-1217.

PAGE 168

154 Filonzi EL, Zoellner H, Stanton H, Ha milton JA. (1993) Cyt okine regulation of granulocyte-macrophage colony stimulat ing factor and macrophage colonystimulating factor production in hum an arterial smooth muscle cells. Atherosclerosis. 99(2):241-252. Finck BK, Chan B, Wofsy D. (1994) Interleu kin 6 promotes murine lupus in NZB/NZW F1 mice. J Clin Invest. 94(2): 585-591. Flynn A. (1986) Expression of Ia and the production of interleukin 1 by peritoneal exudate macrophages activated in vivo by steroids. Life Sci. 38(26):2455-2460. Forster R, Mattis AE, Kremmer E, Wolf E, Brem G, Lipp M. (1996). A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen. Cell 87:10371047. Franks S. (1983) Regulation of prolactin secretion by oestroge ns: physiological and pathological significance. Clin Sci (Lond). 65(5):457-462. Frantz AG. (1978) Prolactin. N Engl J Med. 298(4):201-207. Friedman D, Netti F, Schreiber AD. (1985) E ffect of estradiol and steroid analogues on the clearance of immunoglobulin G-coated erythrocytes. J Clin Invest. 75(1):162167. Fritzler MJ, Tan EM. (1978) Antibodies to hist ones in drug-induced and idiopathic lupus erythematosus. J Clin Invest. 62(3):560-567. Fujimori K, Nabeshima T, Ho IK, Mehendale HM. (1982) Effects of oral administration of chlordecone and mirex on brain biogenic amines in mice. Neurotoxicology. 3(2):143-148. Gabay C, Cakir N, Moral F, Roux-Lombard P, Meyer O, Dayer JM, Vischer T, Yazici H, Guerne PA. (1997) Circulating levels of tu mor necrosis factor soluble receptors in systemic lupus erythematosus are signif icantly higher than in other rheumatic diseases and correlate with diseas e activity. J Rheumatol. 24(2):303-308. Gagnerault MC, Touraine P, Savino W, Kelly PA, Dardenne M. (1993) Expression of prolactin receptors in murine lympho id cells in normal and autoimmune situations. J Immunol . 150(12):5673-5681. Galien R, Garcia T. (1997) Estrogen receptor impairs interleukin-6 expression by preventing protein binding on the NF-k appaB site. Nucleic Acids Res. 25(12):2424-2429. Gabriel KR (1978) A simple method of multiple comparisons of means. J Am Stat Asso. 73: 364.

PAGE 169

155 Garcia B, Rodriguez R, Angulo I, H eath AW, Howard MC, Subiza JL. (1996) Differential effects of transforming gr owth factor-beta 1 on IgA vs. IgG2b production by lipopolysaccharide-stimulated lymph node B cells: a comparative study with spleen B cells. Eur J Immunol. 26(10):2364-2370. Gellert RJ, Wilson C. (1979) Reproductive f unction in rats exposed prenatally to pesticides and polychlorinated biphe nyls (PCB). Environ Res. 18(2):437-443. Gomez D, Correa PA, Gomez LM, Cadena J, Molina JF, Anaya JM. (2004) Th1/Th2 cytokines in patients with systemic lupus erythematosus: is tumor necrosis factor alpha protective? Semin Arthritis Rheum. 33(6):404-413. Goodnow CC, Crosbie J, Adelstein S, Lavoi e TB, Smith-Gill SJ, Brink RA, PritchardBriscoe H, Wotherspoon JS, Loblay RH, Ra phael K, Trent RJ, Basten A. (1988) Altered immunoglobulin expression and f unctional silencing of self-reactive B lymphocytes in transgenic mice. Nature. 334(6184):676-682. Gordon CJ, Rowsey PJ. (1999) Are circulat ing cytokines interleukin-6 and tumor necrosis factor alpha involved in ch lorpyrifos-induced fever? Toxicology. 134(1):9-17. Gorelik L, Flavell RA. (2002) Transforming growth factor-beta in T-cell biology. Nat Rev Immunol. 2(1):46-53. Grakoui A, Bromley SK, Sumen C, Davis MM, Shaw AS, Allen PM, Dustin ML. (1999) The immunological synapse: a molecula r machine controlling T cell activation. Science. 285(5425):221-227. Gray D, MacLennan IC, Bazin H, Khan M. (1982) Migrant mu+ de lta+ and static mu+ deltaB lymphocyte subsets. Eur. J. Immunol. 12:564 569. Green DR, Droin N, Pinkoski M. (2003) Ac tivation-induced cell death in T cells. Immunol Rev. 193:70-81. Green DR, Reed JC. (1998) Mitochondria and apoptosis. Scien ce. 281(5381):1309-1312. Greenwald RJ, Freeman GJ, Sharpe AH. ( 2005) The B7 family revisited. Annu Rev Immunol.23:515-548. Grimaldi CM, Cleary J, Dagtas AS, Mou ssai D, Diamond B. ( 2002) Estrogen alters thresholds for B cell apoptosis and ac tivation. J Clin Invest. 109(12):1625-1633. Grimaldi CM, Hill L, Xu X, Peeva E, ( 2005) Diamond B. Hormonal modulation of B cell development and repertoire selec tion. Mol Immunol. 42(7):811-820. Grimaldi CM, Michael DJ, Diamond B. (2001) Cutting edge: expansion and activation of a population of autoreactive marginal zone B cells in a model of estrogen-induced lupus. J Immunol. 167(4):1886-1890.

PAGE 170

156 Grondal G, Gunnarsson I, Ronnelid J, Rogbe rg S, Klareskog L, Lundberg I. (2000) Cytokine production, serum levels and disease activity in systemic lupus erythematosus. Clin Exp Rheumatol. 18(5):565-570. Grondal G, Traustadottir KH, Kristjansdottir H, Lundberg I, Klareskog L, Erlendsson K, Steinsson K. (2002) Increased T-lym phocyte apoptosis/necrosis and IL-10 producing cells in patients and their spouses in Icelandic systemic lupus erythematosus multicase families. Lupus. 11(7):435-442. Gutierrez MA, Molina JF, Jara LJ, Cuellar ML, Garcia C, Gutierrez-Urena S, Gharavi A, Espinoza LR. (1995) Prolactin and system ic lupus erythematosus: prolactin secretion by SLE lymphocytes and prolif erative (autocrine) activity. Lupus. 4(5):348-352. Haberman AM, William J, Euler C, Shlomchik MJ. (2003) Rheumatoid factors in health and disease: structure, function, induction and re gulation. Curr Dir Autoimmun. 6:169-195. Hagiwara E, Gourley MF, Lee S, Klinman DK . (1996) Disease severity in patients with systemic lupus erythematosus correlates with an increased ra tio of interleukin10:interferon-gamma-secreting cells in the peripheral blood. Arthritis Rheum. 39(3):379-385. Hamilton JA. (1994) Coordinate and noncoordina te colony stimulating factor formation by human monocytes. J Le ukoc Biol. 55(3):355-361. Hamilton JA. (2002) GM-CSF in inflammati on and autoimmunity. Trends Immunol. 23(8):403-408. Hammond B, Katzenellenbogen BS, Krauthamme r N, McConnell J. (1979) Estrogenic activity of the inse cticide chlordecone (Kepone) a nd interaction with uterine estrogen receptors. Proc Natl Acad Sci U S A. 76(12):6641-6645. Han S, Zheng B, Schatz DG, Spanopoulou E, Kelsoe G. (1996) Neoteny in lymphocytes: Rag1 and Rag2 expression in germinal center B cells. Science. 274(5295):20942097. Hannon GJ, Beach D. (1994) p15INK4B is a poten tial effector of TG F-beta-induced cell cycle arrest. Natu re. 371(6494):257-261. Hartley SB, Crosbie J, Brink R, Kantor AB , Basten A, Goodnow CC. (1991) Elimination from peripheral lymphoid tissues of self-reactive B lymphocytes recognizing membrane-bound antigens. Nature. 353(6346):765-769. Hazenberg BP, Van Leeuwen MA, Van Rijs wijk MH, Stern AC, Vellenga E. (1989) Correction of granulocytopenia in Felty 's syndrome by granulocyte-macrophage colony-stimulating factor. Simultaneous i nduction of interleukin-6 release and flare-up of the arthri tis. Blood. 74(8):2769-2770.

PAGE 171

157 Hedman H, Lundgren E. (1992) Regulati on of LFA-1 avidity in human B cells: requirements for dephosphorylation events for high avidity ICAM-1 binding. J Immunol. 149:2295-2299. Herrera-Esparza R, Barbosa-Cisneros O, Villalobos-Hurtado R, Avalos-Diaz E. (1998) Renal expression of IL-6 and TNFgenes in lupus nephritis. Lupus. 7(3):154158. Hess EV. (2002) Environmental chemicals and autoimmune disease: cause and effect. Toxicology. 181-182:65-70. Hiestand PC, Mekler P, Nordmann R, Grieder A, Permmongkol C. (1986) Prolactin as a modulator of lymphocyte responsiveness provides a possible mechanism of action for cyclosporine. Proc Natl Acad Sci U S A. 83(8):2599-2603. Ho IK, Fujimori K, Huang TP, Chang-Tusi H. (1981) Neurochemical evaluation of chlordecone toxicity in the mouse. J Toxicol Environ Health. 8(5-6):701-706. Hodges LC, Bergerson JS, Hunter DS, Wa lker CL. (2000) Estrogenic effects of organochlorine pesticides on uterine leio myoma cells in vitro. Toxicol Sci. 54(2):355-364. Hong JS, Ali SF. (1982) Chlordecone (Kepone) e xposure in the neonate selectively alters brain and pituitary endorphin levels in prepuberal and adult rats. Neurotoxicology. 3(2):111-118. Horwitz DA, Gray JD, Behrendsen SC, Kubin M, Rengaraju M, Ohtsuka K, Trinchieri G. (1998) Decreased production of interleuki n-12 and other Th1-type cytokines in patients with recent-onset systemic lupus erythematosus. Arthritis Rheum. 41(5):838-844. Hoskins B, Ho IK. (1982) Chlordecone-induced alterations in cont ent and subcellular distribution of calcium in mouse brain. J Toxicol Environ Health. 9(4):535-544. Hsu SM. (1985) Phenotypic expression of B lym phocytes. III. Marginal zone B cells in the spleen are characterized by the expre ssion of Tac and alkaline phosphatase. J. Immunol. 135:123 130. Hu SK, Mitcho YL, Rath NC. (1988) Effect of estradiol on interle ukin 1 synthesis by macrophages. Int J Immunopharmacol. 10(3):247-252. Hurst NP, Nuki G, Wallington T. (1984) Ev idence for intrinsic cellular defects of 'complement' receptor-mediated phagocyto sis in patients with systemic lupus erythematosus (SLE). Clin Exp Immunol. 55(2):303-312. Huygen K, Palfliet K. (1984) Strain variation in interferon gamma production of BCGsensitized mice challenged with PPD II. Importance of one ma jor autosomal locus and additional sexual influences . Cell Immunol. 85(1): 75-81.

PAGE 172

158 Ikegwuonu FI, Mehendale HM. (1991) Biochemical assessment of the genotoxicity of the in vitro interaction between chlordec one and carbon tetrachloride in rat hepatocytes. J Appl Toxicol. 11(4):303-310. Ishida H, Muchamuel T, Sakaguchi S, Andrade S, Menon S, Howard M. (1994) Continuous administration of anti-interl eukin 10 antibodies delays onset of autoimmunity in NZB/W F1 mice. J Exp Med. 179(1):305-310. Ito A, Bebo BF Jr, Matejuk A, Zamora A, Silverman M, Fyfe-Johnson A, Offner H. (2001) Estrogen treatment down-regulate s TNF-alpha production and reduces the severity of experimental autoimmune encephalomyelitis in cytokine knockout mice. J Immunol. 167(1):542-552. Ito A, Buenafe AC, Matejuk A, Zamora A, Silverman M, Dwyer J, Vandenbark AA, Offner H. (2002) Estrogen inhibits syst emic T cell expression of TNF-alpha and recruitment of TNF-alpha(+) T cells and macrophages into the CNS of mice developing experimental encephalomy elitis. Clin Immunol. 102(3):275-282. Izui S, Kelley VE, Masuda K, Yoshida H, Roths JB, Murphy ED. (1984) Induction of various autoantibodies by mutant gene lpr in several strains of mice. J Immunol. 133(1):227-233. Jacob CO, Fronek Z, Lewis GD, Koo M, Ha nsen JA, McDevitt HO. (1990) Heritable major histocompatibility complex class II -associated differences in production of tumor necrosis factor alpha: relevance to genetic predisposition to systemic lupus erythematosus. Proc Natl A cad Sci U S A. 87(3):1233-1237. Jacob CO, McDevitt HO. (1988) Tumour necrosis factorin murine autoimmune lupus nephritis. Nature. 331 : 356-358. Jacob CO, van der Meide PH, McDevitt HO. (1987) In vivo treatment of (NZB X NZW)F1 lupus-like nephritis with m onoclonal antibody to gamma interferon. J Exp Med. 166(3): 798-803. Jara LJ, Gomez-Sanchez C, Silveira LH, Martinez-Osuna P, Vasey FB, Espinoza LR. (1992) Hyperprolactinemia in systemic lupus erythematosus: association with disease activity. Am J Med Sci. 303(4):222-226. Jenkins JK, Suwannaroj S, Elbourne KB, Nd ebele K, McMurray RW. (2001) 17-betaestradiol alters Jurkat lymphocyte cell cyc ling and induces apoptosis through suppression of Bcl-2 and cyclin A. Int Immunopharmacol. 1(11):1897-1911. Jiang Y, Hirose S, Abe M, Sanokawa-Akakura R, Ohtsuji M, Mi X, Li N, Xiu Y, Zhang D, Shirai J, Hamano Y, Fujii H, Shirai T. (2000) Polymorphisms in IgG Fc receptor IIB regulatory regions associat ed with autoimmune susceptibility. Immunogenetics. 51(6): 429-435.

PAGE 173

159 Jiang Y, Hirose S, Sanokawa-Akakura R, Abe M, Mi X, Li N, Miura Y, Shirai J, Zhang D, Hamano Y, Shirai T. (1999) Genetica lly determined aberrant down-regulation of FcgammaRIIB1 in germinal center B cells associated with hyper-IgG and IgG autoantibodies in murine systemic lupus erythematosus. Int Immunol. 11(10):1685-1691. Jordan JE, Grice T, Mishra SK, Desaiah D. (1 981) Acute chlordecone toxicity in rats: a relationship between tremor and ATPase activities. Neurotoxicology. 2(2):355364. Kamada M, Irahara M, Maegawa M, Yasui T, Takeji T, Yamada M, Tezuka M, Kasai Y, Aono T. (2000) Effect of hormone replacement therapy on post-menopausal changes of lymphocytes and T cell subs ets. J Endocrinol Invest. 23(6):376-382. Kanda N, Watanabe S. (2004) 17beta-estradiol enhances the production of granulocytemacrophage colony-stimulating factor in human keratinocytes. J Invest Dermatol. 123(2):329-337. Kanda N, Watanabe S. (2005) Regulatory ro les of sex hormones in cutaneous biology and immunology. J Dermatol Sci. 38(1):1-7. Karlsson MC, Guinamard R, Bolland S, Sanka la M, Steinman RM, Ravetch JV. (2003) Macrophages control the retention and traffi cking of B lymphocytes in the splenic marginal zone. J Exp Med. 198:333 340. Karpuzoglu-Sahin E, Zhi-Jun Y, Lengi A, Sriranganathan N, Ansar Ahmed S. (2001) Effects of long-term estrogen treatmen t on IFN-gamma, IL-2 and IL-4 gene expression and protein synthe sis in spleen and thymus of normal C57BL/6 mice. Cytokine. 14(4):208-217. Kick G, Messer G, Plewig G, Kind P, Goet z AE. (1996) Strong and prolonged induction of c-jun and c-fos proto-oncogenes by photodynamic therapy. Br J Cancer. 74(1): 30-36. Kim CH, Rott LS, Clark-Lewis I, Cam pbell DJ, Wu L, Butcher EC. (2001) Subspecialization of CXCR5+ T cells: B help er activity is focused in a germinal center-localized subset of CXCR5 + T cells. J Exp Med. 193(12):1373-1381. Kim JY, Oh KN, Han EH, Kim DH, Jeong TC , Lee ES, Jeong HG. (2005) Methoxychlorinduced inducible nitric oxi de synthase and proinflamma tory cytokines expression in macrophages via NF-kappaB, ERK, and p38 mitogen-activated protein kinases. Biochem Biophys Res Commun. 333(4):1234-1240. Kim SJ, Gershov D, Ma X, Brot N, Elkon KB. (2003) Opsonization of apoptotic cells and its effect on macrophage and T cell imm une responses. Ann N Y Acad Sci. 987:68-78.

PAGE 174

160 Kim T, Kanayama Y, Negoro N. (1987) Seru m levels of interferons in patients with systemic lupus erythematosus. Clin Exp Immunol. 70(3):562-569. Klinman DM, Shirai A, Ishigatsubo Y, Conove r J, Steinberg AD. (1991) Quantitation of IgMand IgG-secreting B cells in the peripheral blood of patients with systemic lupus erythematosus. Arthritis Rheum. 34(11):1404-1410. Kodavanti PR, Cameron JA, Yallapragada PR, De saiah D. (1990) Effect of chlordecone (Kepone) on calcium transport mechanisms in rat heart sarcoplasmic reticulum. Pharmacol Toxicol. 67(3):227-234. Kono DH, Balomenos D, Park MS, Theofilopoul os AN. (2000) Development of lupus in BXSB mice is independent of IL-4. J Immunol. 164(1):38-42. Kono DH, Balomenos D, Pearson DL, Park MS , Hildebrandt B, Hultman P, Pollard KM. (1998) The prototypic Th2 autoimmunity induced by mercury is dependent on IFN-gamma and not Th1/Th2 imbalance. J Immunol. 161(1):234-240. Koopman G, Keehnen RM, Lindhout E, Newman W, Shimizu Y, van Seventer GA, de Groot C, Pals ST. (1994) Adhesion through the LFA-1 (CD11a/CD18)-ICAM-1 (CD54) and the VLA-4 (CD49d)-VCAM-1 (C D106) pathways prevents apoptosis of germinal center B cells . J Immunol. 152(8):3760-3767. Kotzin BL. (1996) Systemic lupus erythematosus. Cell. 85(3):303-306. Kraal G. (1992) Cells in the marginal z one of the spleen. Int Rev Cytol. 132:31 74. Kramer PR, Kramer SF, Guan G. (2004) 17 be ta-estradiol regulates cytokine release through modulation of CD16 expression in monocytes and monocyte-derived macrophages. Arthritis Rheum. 50(6):1967-1975. Krause I, Valesini G, Scrivo R, Shoenfeld Y. (2003) Autoimmune aspects of cytokine and anticytokine therapies. Am J Med. 115(5): 390-397. Kretz-Rommel A, Duncan SR, Rubin RL. ( 1997) Autoimmunity caused by disruption of central T cell tolerance. A murine mode l of drug-induced lupus. J Clin Invest. 99(8):1888-1896. Krzystyniak K, Flipo D, Mansour S, F ournier M. (1989) Suppression of avidin processing and presentation by mouse macrophages after sublethal exposure to dieldrin. Immunopharmacology. 18(3): 157-166. Kuo P, Bynoe MS, Wang C, Diamond B. (1999) Bcl-2 leads to expression of anti-DNA B cells but no nephritis: a model for a clinical subset. Eu r J Immunol. 29(10): 3168-3178. Lahita RG. (1996) The connective tissue diseas es and the overall influence of gender. Int J Fertil Menopausal Stud. 41(2):156-165.

PAGE 175

161 Lahita RG. (1999) The role of sex hormones in systemic lupus erythematosus. Curr Opin Rheumatol. 11(5):352-356. Lang J, Arnold B, Hammerling G, Harris AW , Korsmeyer S, Russell D, Strasser A, Nemazee D. (1997) Enforced Bcl-2 expre ssion inhibits antigen-mediated clonal elimination of peripheral B cells in an antigen dose-dependent manner and promotes receptor editing in autore active, immature B cells. J Exp Med. 186(9):1513-1522. Larrea F, Martinez-Castillo A, Cabrera V, Alcocer-Varela J, Queipo G, Carino C, Alarcon-Segovia D. (1997) A bioactive 60-kilodalt on prolactin species is preferentially secreted in cultures of mitogen-stimulated and nonstimulated peripheral blood mononuclear cells fro m subjects with systemic lupus erythematosus. J Clin En docrinol Metab. 82(11):3664-3669. Lauwerys BR, Garot N, Renauld JC, Houssi au FA. (2000) Interleukin-10 blockade corrects impaired in vitr o cellular immune responses of systemic lupus erythematosus patients. Arthritis Rheum. 43(9):1976-1981. Lavalle C, Graef A, Baca V, Ramirez-Laca yo M, Blanco-Favela F, Ortiz O. (1993) Prolactin and gonadal hormones: a key relationship that may have clinical, monitoring and therapeutic implications in systemic lupus erythematosus. Lupus. 2(2):71-75. Lee BC, Kamata M, Akatsuka Y, Takeda M, Ohno K, Kamei T, Magara Y. (2004) Effects of chlorine on the decrease of estrogenic chemicals. Water Res. 38(3): 733-739. Lee SH, Park SH, Min JK, Kim SI, Yoo WH , Hong YS, Park JH, Cho CS, Kim TG, Han H, Kim HY. (1997) Decreased tumour n ecrosis factor-beta production in TNFB*2 homozygote: an important predisposing f actor of lupus nephritis in Koreans. Lupus. 6(7):603-609. Leizer T, Cebon J, Layton JE, Hamilton JA. (1990) Cytokine regulation of colonystimulating factor production in cultured human synovial fibroblasts: I. Induction of GM-CSF and G-CSF production by interl eukin-1 and tumor necrosis factor. Blood. 76(10):1989-1996. Leonard WJ, O'Shea JJ. (1998) Jaks and STATs: biological imp lications. Annu Rev Immunol. 16:293-322. Leondires MP, Hu ZZ, Dong J, Tsai-Morris CH , Dufau ML. (2002) Estradiol stimulates expression of two human prolactin receptor isoforms with alternative exons-1 in T47D breast cancer cells. J Steroi d Biochem Mol Biol. 82(2-3):263-268. Levene H. (1960) Robust tests for the equality of variance. I. Olkin, ed., Contributions to probability and statistics, 278--292, Palo Alto, CA: Stanford University Press.

PAGE 176

162 Li P, Nijhawan D, Budihardjo I, Srin ivasula SM, Ahmad M, Alnemri ES, Wang X. (1997) Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic pr otease cascade. Cell. 91(4):479-489. Linder RE, Scotti TM, McElroy WK, Lask ey JW, Strader LF, Powell K. (1983) Spermotoxicity and tissue accumulation of chlordecone (Kepone) in male rats. J Toxicol Environ Health. 12(2-3):183-192. Linker-Israeli M, Bakke AC, Kitridou RC, Ge ndler S, Gillis S, Horwitz DA. (1983) Defective production of interleukin 1 and in terleukin 2 in patien ts with systemic lupus erythematosus (SLE). J Immunol. 130(6):2651-2655. Lang RA, Metcalf D, Cuthbertson RA, Lyons I, Stanley E, Kelso A, Kannourakis G, Williamson DJ, Klintworth GK, Gonda TJ, Dunn AR. (1987) Transgenic mice expressing a hemopoietic growth factor gene (GM-CSF) develop accumulations of macrophages, blindness, and a fa tal syndrome of tissue damage. Cell. 51(4):675-686. Larson PS, Egle JL Jr, Hennigar GR, Lane RW , Borzelleca JF. (1979) Acute, subchronic, and chronic toxicity of chlordecone. T oxicol Appl Pharmacol. 48(1 Pt 1):29-41. Linder RE, Scotti TM, McElroy WK, Lask ey JW, Strader LF, Powell K. (1983) Spermotoxicity and tissue accumulation of chlordecone (Kepone) in male rats. J Toxicol Environ Health. 12(2-3):183-192. Liu H, Lee ES, Gajdos C, Pearce ST, Chen B, Osipo C, Loweth J, McKian K, De Los Reyes A, Wing L, Jordan VC. (2003) Apopt otic action of 17beta-estradiol in raloxifene-resistant MCF-7 cells in vitro and in vivo. J Natl Cancer Inst. 95(21):1586-1597. Llorente L, Richaud-Patin Y. (1993) The ro le of interleukin-10 in systemic lupus erythematosus. J Autoimmun. 20(4):287-289. Llorente L, Richaud-Patin Y, Fior R, Al cocer-Varela J, Wijdenes J, Fourrier BM, Galanaud P, Emilie D. (1994) In vivo production of interleukin-10 by non-T cells in rheumatoid arthritis, Sjogren's syndrom e, and systemic lupus erythematosus. A potential mechanism of B lymphocyte hype ractivity and autoimmunity. Arthritis Rheum.37(11):1647-1655. Llorente L, Richaud-Patin Y, Wijdenes J, Alcocer-Varela J, Maillot MC, DurandGasselin I, Fourrier BM, Galanaud P, Emilie D. (1993) Spontaneous production of interleukin-10 by B lymphocytes and monocytes in systemic lupus erythematosus. Eur Cytokine Netw. 4(6):421-427. Llorente L, Zou W, Levy Y, Richaud-Patin Y, Wijdenes J, Alcocer-Varela J, MorelFourrier B, Brouet JC, Alarcon-Segovia D, Galanaud P, Emilie D. (1995) Role of interleukin 10 in the B lymphocyte hype ractivity and auto antibody production of human systemic lupus erythema tosus. J Exp Med. 181(3):839-844.

PAGE 177

163 Lopes-Carvalho T, Kearney JF. (2004) Develo pment and selection of marginal zone B cells. Immunol Rev. 197:192-205. Lorenz HM, Grunke M, Hieronymus T, Herrm ann M, Kuhnel A, Manger B, Kalden JR. (1997) In vitro apoptosis and expression of apoptosis -related molecules in lymphocytes from patients with systemic lupus erythematosus and other autoimmune diseases. Arthritis Rheum. 40(2):306-317. Lu TT, Cyster JG. (2002) Integrin-mediated long-term B cell retention in the splenic marginal zone. Science 297:409 412. Lynch DH, Watson ML, Alderson MR, Baum PR, Miller RE, Tough T, Gibson M, Davis-Smith T, Smith CA, Hunter K, et al. (1994) The mouse Fas-ligand gene is mutated in gld mice and is part of a TN F family gene cluster. Immunity. (2):131136. Maddison PJ. (1999) Autoantibodies in SLE. Disease associations. Adv Exp Med Biol. 455:141-145. Mageed RA, Isenberg DA. (2002) Tumour n ecrosis factor alpha in systemic lupus erythematosus and anti-DNA autoanti body production. Lupus. 11(12):850-855. Malide D, Russo P, Bendayan M. (1995) Pres ence of tumor necrosis factor alpha and interleukin-6 in renal mesangial cells of lupus nephritis patients. Hum Pathol. 26(5):558-564. Manolagas SC, Bellido T, Jilka RL. (1995) Se x steroids, cytokines and the bone marrow: new concepts on the pathogenesis of osteoporosis. Ciba Found Symp. 191:187202. Masuzawa T, Miyaura C, Onoe Y, Kusano K, Ohta H, Nozawa S, Suda T. (1994) Estrogen deficiency stimulates B lym phopoiesis in mouse bone marrow. J Clin Invest. 94(3):1090-1097. Matera L, Cutufia M, Geuna M, Contarini M, Buttiglieri S, Galin S, Fazzari A, Cavaliere C. (1997) Prolactin is an au tocrine growth factor for the Jurkat human T-leukemic cell line. J Neuroimmunol. 79(1):12-21. Matera L, Geuna M, Pastore C, Buttiglie ri S, Gaidano G, Savarino A, Marengo S, Vonderhaar BK. (2000) Expression of pr olactin and prolactin receptors by nonHodgkin's lymphoma cells. Int J Cancer. 85(1):124-130. Matera L, Muccioli G, Cesano A, Bellussi G, Genazzani E. (1988) Prolactin receptors on large granular lymphocytes: dual regu lation by cyclosporin A. Brain Behav Immun. 2(1):1-10. Mayes MD. (1999) Epidemiologic studies of environmental agents and systemic autoimmune diseases. Environ H ealth Perspect. 107 Suppl 5:743-748.

PAGE 178

164 Maloy KJ, Powrie F. (2001) Regulatory T cel ls in the control of immune pathology. Nat Immunol. 2(9):816-822. McFarland HI, Bigley NJ. ( 1989) Sex-dependent, early cy tokine production by NK-like spleen cells following infection with the D variant of encephalomyocarditis virus (EMCV-D). Viral Immunol . 2(3): 205-214. McMurray RW. (1996) Prolactin and systemic lupus erythematosus. Ann Med Interne (Paris). 147(4):253-258. McMurray RW. (2001) Estrogen, prolactin, and autoimmunity: actions and interactions. Int Immunopharmacol. (6):995-1008. McMurray RW, Ndebele K, Hardy KJ, Jenkins JK. (2001) 17-beta-estradiol suppresses IL-2 and IL-2 receptor. Cytokine. 14(6):324-333. McMurray RW, Suwannaroj S, Ndebele K, Jenk ins JK. (2001) Differential effects of sex steroids on T and B cells: modulation of cell cycle phase distribution, apoptosis and bcl-2 protein levels . Pathobiology. 69(1):44-58. McMurray RW, Weidensaul D, Allen SH, Walk er SE. (1995) Efficacy of bromocriptine in an open label therapeutic trial for systemic lupus erythematosus. J Rheumatol. 22(11):2084-2091. Medina KL, Kincade PW. (1994) Pregnancy-re lated steroids are potential negative regulators of B lymphopoiesis. Proc Natl Acad Sci U S A. 91(12):5382-5386. Mevorach D, Zhou JL, Song X, Elkon KB. ( 1998) Systemic exposure to irradiated apoptotic cells induces autoanti body production. J Exp Med. 188(2):387-392. Miles S, Isenberg D. (1993) A review of serological abnormalities in relatives of SLE patients. Lupus. 2(3):145-150. Miyake K, Medina KL, Hayashi S, Ono S, Hamaoka T, Kincade PW. (1990) Monoclonal antibodies to Pgp-1/CD44 block lymphohemopoiesis in long-term bone marrow cultures. J Exp Med. 171(2):477-488. Mizoguchi Y, Kim JY, Enami J, Sakai S. (1997 ) The regulation of th e prolactin receptor gene expression in the mammary gland of early pregnant mouse. Endocr J. 44(1):53-58. Mo R, Chen J, Grolleau-Julius A, Murphy HS, Richardson BC, Yung RL. (2005) Estrogen regulates CCR gene expressi on and function in T lymphocytes. J Immunol 15;174(10):6023-6029. Mohan C, Adams S, Stanik V, Datta SK . (1993) Nucleosome: a major immunogen for pathogenic autoantibody-inducing T cells of lupus. J Exp Med. 177(5):1367-1381.

PAGE 179

165 Mok CC, Lau CS. (2003) Pathogenesis of system ic lupus erythematosus. J Clin Pathol. 56(7):481-490. Monks CR, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. (1998) Three-dimensional segregation of supramolecular activa tion clusters in T cells. Nature. 395(6697):82-86. Montgomery DW, LeFevre JA, Ulrich ED, Adamson CR, Zukoski CF. (1990) Identification of prolactin-like prot eins synthesized by normal murine lymphocytes. Endocrinology. 127(5):2601-2603. Montgomery DW, Shen GK, Ulrich ED, Stei ner LL, Parrish PR, Zukoski CF. (1992) Human thymocytes express a prolactin -like messenger ribonucleic acid and synthesize bioactive prolactin-like pr oteins. Endocrinology. 131(6):3019-3026. Morel L, Wakeland EK. (1998) Susceptibility to lupus nephritis in the NZB/W model system. Curr Opin Immunol. 10(6):718-725. Mori M, Morris SC, Orekhova T, Marinaro M, Giannini E, Finkelman FD. (2000) IL-4 promotes the migration of circulating B ce lls to the spleen and increases splenic B cell survival. J I mmunol. 164(11):5704-5712. Mountz JD, Zhou T, Su X, Cheng J, Pierson M, Bluethmann H, Edwards CK 3rd. (1996) Autoimmune disease results from multip le interactive defects in apoptosis induction molecules and signaling pathwa ys. Behring Inst Mitt. (97):200-219. Nagy E, I Berczi. (1978) Immunodeficiency in hypophysectomized rats. Acta Endocrinol. 89:530-537. Nakajima A, Hirose S, Yagita H, Okumura K. (1997) Roles of IL-4 and IL-12 in the development of lupus in NZB/W F1 mice. J Immunol. 158(3):1466-1472. Nakajima M, Kawahara S, Sakagami Y, Takagawa K, Akazawa H, Kamitsuji H, Yoshioka A. (1999) Immunogold labelling of cytokines in glomeruli in children with various renal diseas es. Nephron. 83(2):132-138. Neale TJ, Ruger BM, Macaulay H, Dunbar PR , Hasan Q, Bourke A, Murray-McIntosh RP, Kitching AR. (1995) Tumor necrosis f actor-alpha is expr essed by glomerular visceral epithelia l cells in human membranous nephropathy. Am J Pathol. 146(6):1444-1454. Nelson BH. (2004) IL-2, regulatory T cells, and tolerance. J I mmunol. 172(7):3983-3988. Nemazee DA, Burki K. (1989) Clonal deletion of B lymphocytes in a transgenic mouse bearing anti-MHC class I anti body genes. Nature. 337(6207):562-566.

PAGE 180

166 Nie Y, Waite J, Brewer F, Sunshine MJ, Littman DR, Zou YR. (2004) The role of CXCR4 in maintaining peripheral B cell compartments and humoral immunity. J Exp Med. 200(9):1145-1156. Novotny EA, Raveche ES, Sharrow S, Ottinger M, Steinberg AD. (1983) Analysis of thymocyte subpopulations following treatmen t with sex hormones. Clin Immunol Immunopathol. 28(2):205-217. Nunez G MA, Estrada I, Calderon-Aranda ES. (2002) DDT inhibits the functional activation of murine macrophages and d ecreases resistance to infection by Mycobacterium microti. Toxicology. 174(3): 201-210. Ochel M, Vohr HW, Pfeiffer C, Gleichmann E. (1991) IL-4 is required for the IgE and IgG1 increase and IgG1 autoantibody formation in mice treated with mercuric chloride. J Immunol. 146(9):3006-3011. Ohashi PS, DeFranco AL. (2002) Making a nd breaking tolerance. Curr Opin Immunol. 14(6):744-759. Okasha SA, Ryu S, Do Y, McKallip RJ, Nagarkatti M, Nagarkatti PS. (2001) Evidence for estradiol-induced apoptosis and dysre gulated T cell maturation in the thymus. Toxicology. 163(1):49-62. Okubo T, Yokoyama Y, Kano K, Soya Y, Kano I. (2004) Es timation of estrogenic and antiestrogenic activities of selected pest icides by MCF-7 cell proliferation assay. Arch Environ Contam Toxicol. 46(4):445-453. Oliver AM, Martin F, Gartland GL, Carter RH , Kearney JF. (1997) Marginal zone B cells exhibit unique activation, pr oliferative and immunoglobu lin secretory responses. Eur. J. Immunol. 27:2366 2374. Oliver AM, Martin F, Kearney JF. (1999) IgMhighCD21high lymphocytes enriched in the splenic marginal zone generate effector cells more rapidly than the bulk of follicular B cells. J. Immunol. 162:7198 7207. O'Rourke LM, Tooze R, Turner M, Sandova l DM, Carter RH, Tybulewicz VL, Fearon DT. (1998) CD19 as a membrane-anchored adaptor protein of B lymphocytes: costimulation of lipid and protein ki nases by recruitment of Vav. Immunity. 8(5):635-645. Paavonen T, Andersson LC, Adlercreutz H. ( 1981) Sex hormone regulation of in vitro immune response. Estradiol enhances hum an B cell maturation via inhibition of suppressor T cells in pokeweed mit ogen-stimulated cultures. J Exp Med. 154(6):1935-1945. Pani G, Kozlowski M, Cambier JC, Mills GB, Siminovitch KA. (1995) Identification of the tyrosine phosphatase PTP1C as a B ce ll antigen receptor-associated protein involved in the regulation of B ce ll signaling. J Exp Med. 181(6):2077-2084.

PAGE 181

167 Parks CG, Cooper GS, Nylander-French LA , Sanderson WT, Dement JM, Cohen PL, Dooley MA, Treadwell EL, St Clair EW, Gilkeson GS, Hoppin JA, Savitz DA. (2002) Occupational exposure to crystal line silica and risk of systemic lupus erythematosus: a population-based, case-con trol study in the southeastern United States. Arthritis Rheum. 46(7):1840-1850. Paul E, Lutz J, Erikson J, Carroll MC. (2004) Germinal center checkpoints in B cell tolerance in 3H9 transgenic mi ce. Int Immunol. 16(2):377-384. Pearse RN, Kawabe T, Bolland S, Guinamard R, Kurosaki T, Ravetch JV. (1999) SHIP recruitment attenuates Fc gamma RIIB-induced B cell apoptosis. Immunity. 10(6):753-760. Peeva E, Grimaldi C, Spatz L, Diamond B. (2000) Bromocriptine restores tolerance in estrogen-treated mice. J C lin Invest. 106(11):1373-1379. Pellegrini I, Lebrun JJ, Ali S, Kelly PA. (1992) Expression of prolactin and its receptor in human lymphoid cells. Mol Endocrinol. 6(7):1023-1031. Pelton BK, Speckmaier M, Hylton W, Farra nt J, Denman AM. (1991) Cytokineindependent progression of immuno globulin production in vitro by B lymphocytes from patients with systemic lupus erythematosus. Clin Exp Immunol. 83(2): 274-279. Phillips DE, Eroschenko VP. (1985) An elect ron microscopic study of alterations in mouse peripheral nerve and skeletal muscle after chlordecone exposure. Neurotoxicology. 6(1):141-150. Pillai S, Cariappa A, Moran ST. (2005) Marginal zone B cells. Annu Rev Immunol. 23:161-196. Pittoni V, Valesini G. (2002) The clearance of apoptotic cells: implications for autoimmunity. Autoimmun Rev. 1(3):154-161. Plas DR, Thomas ML. (1998) Negative regu lation of antigen receptor signaling in lymphocytes. J Mol Med. 76(8): 589-595. Podojil JR, Kin NW, Sanders VM. (2004) CD86 and 2-adrenergic receptor signaling pathways, respectively, increase Oct-2 and OCA-B Expression and binding to the 3'-IgH enhancer in B cells. J. Biol. Chem. 279:23394 23404. Podojil JR, Sanders VM. (2003) Selective regu lation of mature IgG1 transcription by CD86 and 2-adrenergic receptor st imulation. J. Immunol. 170:5143 5151. Pollard KM, Pearson DL, Hultman P, Hildebrandt B, Kono DH. (1999) Lupus-prone mice as models to study xenobiotic -induced acceleration of systemic autoimmunity. Environ Health Perspect. 107 Suppl 5:729-735.

PAGE 182

168 Polyak K, Kato JY, Solomon MJ, Sherr CJ, Massague J, Roberts JM, Koff A. (1994) p27Kip1, a cyclin-Cdk inhibitor, links transforming growth factor-beta and contact inhibition to cell cycl e arrest. Genes Dev. 8(1):9-22. Porter VR, Greendale GA, Schocken M, Zhu X, Effros RB. (2001) Immune effects of hormone replacement therapy in post-menopausal women. Exp Gerontol. 36(2):311-326. Pritchard NR, Smith KG. (2003) B cell i nhibitory receptors and autoimmunity. Immunology. 108(3): 263-273. Radic MZ, Erikson J, Litwin S, Weigert M. (1993) B lymphocytes may escape tolerance by revising their antigen recep tors. J Exp Med. 177(4):1165-1173. Ramsdell F, Seaman MS, Miller RE, Tough TW, Alderson MR, Lynch DH. (1994) gld/gld mice are unable to express a f unctional ligand for Fas. Eur J Immunol. (4):928-933. Ray P, Ghosh SK, Zhang DH, Ray A. (1997) Re pression of interleukin-6 gene expression by 17 beta-estradiol: inhibition of the DNA-binding activity of the transcription factors NF-IL6 and NF-kappa B by the es trogen receptor. FEBS Lett. 409(1):7985. Reuber MD. (1979) The carcinogeni city kepone. J Environ Pathol Toxicol. 2(3):671-686. Richards HB, Satoh M, Shaw M, Libert C, Poli V, Reeves WH. (1998) Interleukin 6 dependence of anti-DNA antibody production: evidence for two pathways of autoantibody formation in pristane-i nduced lupus. J Exp Med. 188(5):985-990. Rijhsinghani AG, Thompson K, Bhatia SK, Waldschmidt TJ. (1996) Estrogen blocks early T cell development in the thymus . Am J Reprod Immunol. 36(5):269-277. Rood MJ, Van Der Velde EA, Ten Cate R, Breedveld FC, Huizinga TW. (1998) Female sex hormones at the onset of systemic l upus erythematosus affect survival. Br J Rheumatol. 37(9): 1008-1010. Roubinian JR, Talal N, Greenspan JS, Goodman JR, Siiteri PK. (1978) Effect of castration and sex hormone treatment on survival, anti-nucleic acid antibodies, and glomerulonephritis in NZB/NZW F1 mice. J Exp Med. 147(6):1568-1583. Roubinian JR, Talal N, Greenspan JS, Goodman JR, Siiteri PK. (1979) Delayed androgen treatment prolongs survival in murine lupus. J Clin Invest 63: 902-911. Rousset F, Garcia E, Defrance T, Peronne C, Vezzio N, Hsu DH, Kastelein R, Moore KW, Banchereau J. (1992) Interleukin 10 is a potent growth and differentiation factor for activated human B lymphocyte s. Proc Natl Acad Sci U S A. 1992 Mar 1;89(5):1890-1893.

PAGE 183

169 Russell DH, Matrisian L, Kibler R, Larson DF, Poulos B, Magun BE. (1984) Prolactin receptors on human lymphocytes and th eir modulation by cyclosporine. Biochem Biophys Res Commun. 121(3):899-906. Ryffel B, Car BD, Gunn H, Roman D, Hies tand P, Mihatsch MJ. (1994) Interleukin-6 exacerbates glomerulonephritis in (NZB x NZW)F1 mice. Am J Pathol. 144(5): 927-937. Sakaguchi S. (2000) Regulatory T cells: key controllers of immunologic self-tolerance. Cell. 101(5):455-458. Salem ML. (2004) Estrogen, a double-edged sword: modulation of TH1and TH2mediated inflammations by differentia l regulation of TH1/TH2 cytokine production. Curr Drug Targets Infl amm Allergy. 3(1):97-104. Salmon JE, Kimberly RP, Gibofsky A, Fo tino M. (1984) Defective mononuclear phagocyte function in systemic lupus eryt hematosus: dissociation of Fc receptorligand binding and internaliza tion. J Immunol. 133(5):2525-2531. Sanchez-Guerrero J, Karlson EW, Liang MH, Hunter DJ, Speizer FE, Colditz GA. (1997) Past use of oral contraceptives and the risk of developing systemic lupus erythematosus. Arthritis Rheum. 40(5):804-808. Santiago ML, Fossati L, Jacquet C, Muller W, Iz ui S, Reininger L. (1997) Interleukin-4 protects against a genetically linked lupus-like autoimmune syndrome. J Exp Med. 185(1):65-70. Sarvetnick N, Fox HS. (1990) Interfer on-gamma and the sexual dimorphism of autoimmunity. Mol Biol Med. 7(4):323-331. Satoh M, Reeves WH. (1994) Induction of l upus-associated autoantibodies in BALB/c mice by intraperitoneal injection of pristane. J Exp Med. 180(6):2341-2346. Savill J, Fadok V, Henson P, Haslett C. ( 1993) Phagocyte recognition of cells undergoing apoptosis. Immunol Today. 14(3):131-6. Savill J, Dransfield I, Gregory C, Haslett C. (2002) A blast from the past: clearance of apoptotic cells regulates immune res ponses. Nat Rev Immunol. 2(12):965-975. Schreiber AD, Nettl FM, Sanders MC, King M, Szabolcs P, Friedman D, Gomez F. (1988) Effect of endogenous and synthetic sex steroids on the clearance of antibody-coated cells. J Immunol. 141(9): 2959-2966. Scott RS, McMahon EJ, Pop SM, Reap EA, Caricchio R, Cohen PL, Earp HS, Matsushima GK. (2001) Phagocytosis and cl earance of apoptotic cells is mediated by MER. Nature. 411(6834):207-211.

PAGE 184

170 Seery JP, Carroll JM, Cattell V, Watt FM. ( 1997) Antinuclear autoantibodies and lupus nephritis in transgenic mice expressing in terferon gamma in the epidermis. J Exp Med 186(9):1451-1459. Selvaraj V, Bunick D, Finnigan-Bunick C, Johnson RW, Wang H, Liu L, Cooke PS. (2005) Gene expression profiling of 17beta -estradiol and genistein effects on mouse thymus.Toxicol Sci. 87(1):97-112. Shelby MD, Newbold RR, Tully DB, Chae K, Davis VL. (1996) Assessing environmental chemicals for estrogenicity using a combination of in vitro and in vivo assays. Environ Health Perspect. 104(12):1296-1300. Shevach EM. (2000) Regulatory T cells in autoimmmunity. Annu Rev Immunol. 18:423449. Shultz LD, Coman DR, Bailey CL, Beamer WG , Sidman CL. (1984) "Viable motheaten," a new allele at the motheaten locus. I. Pathology. Am J Pathol. 116(2): 179-192. Shultz LD, Rajan TV, Greiner DL. (1997) Seve re defects in immunity and hematopoiesis caused by SHP-1 protein-tyrosine-phospha tase deficiency. Trends Biotechnol. 15(8):302-307. Shultz LD, Sidman CL. (1987) Genetica lly determined murine models of immunodeficiency. Annu Rev Immunol. 5:367-403. Singh RR. ( 2003) IL-4 and many roads to lupuslike autoimmunity. Clin Immunol. 108(2):73-79. Singh RR, Saxena V, Zang S, Li L, Fi nkelman FD, Witte DP, Jacob CO. (2003) Differential contribution of IL-4 and STAT6 vs STAT4 to the development of lupus nephritis. J Im munol. 170(9):4818-4825. Sirica AE, Wilkerson CS, Wu LL, Fitzge rald R, Blanke RV, Guzelian PS. (1989) Evaluation of chlordecone in a two-st age model of hepatocarcinogenesis: a significant sex difference in the he patocellular carci noma incidence. Carcinogenesis. 10(6):1047-1054. Smithson G, Beamer WG, Shultz KL, Christ ianson SW, Shultz LD, Kincade PW. (1994) Increased B lymphopoiesis in genetically sex steroid-deficient hypogonadal (hpg) mice. J Exp Med. 180(2):717-720. Sobel ES, Gianini J, Butfiloski EJ, Croker BP, Schiffenbauer J, Roberts SM. (2005) Acceleration of autoimmunity by organochl orine pesticides in (NZB x NZW)F1 mice. Environ Health Perspect. 113(3):323-328. Sobel ES, Katagiri T, Katagiri K, Morri s SC, Cohen PL, Eisenberg RA. (1991) An intrinsic B cell defect is required for th e production of autoantibodies in the lpr model of murine systemic auto immunity. J Exp Med. 173(6):1441-1449.

PAGE 185

171 Sobel ES, Satoh M, Chen Y, Wakeland EK, Morel L. (2002) The major murine systemic lupus erythematosus susceptib ility locus Sle1 results in abnormal functions of both B and T cells. J Immunol. 169(5):2694-2700. Sobel ES, Wang F, Butfiloski E, Croker B, Roberts SM. (in press) Comparison of chlordecone effects on autoimmunity in (NZBxNZW) F(1) and BALB/c mice. Toxicology. Soto AM, Chung KL, Sonnenschein C. (1994) Th e pesticides endosulfan, toxaphene, and dieldrin have estrogenic e ffects on human estrogen-sensi tive cells. Environ Health Perspect. 102(4):380-383. Steeves MA, Marion TN. (2004) Tolerance to DNA in (NZB x NZW)F1 mice that inherit an anti-DNA V(H) as a conventional micr o H chain transgene but not as a V(H) knock-in transgene. J Immunol. 172(11):6568-6577. Suzuki T, Sullivan DA. (2005) Estrogen stim ulation of proinflammatory cytokine and matrix metalloproteinase gene expressi on in human corneal epithelial cells. Cornea. 24(8):1004-1009. Svensson B. (1980) Occurrence of deficient m onocyte yeast cell phagocytosis in presence of rheumatic sera. Scand J Rheumatol Suppl. 31:21-27. Swartz WJ, Mall GM. (1989) Chlordecone-induced follicular toxicity in mouse ovaries. Reprod Toxicol. 1989;3(3):203-206. Takahashi T, Tanaka M, Brannan CI, Jenki ns NA, Copeland NG, Suda T, Nagata S. (1994) Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell. 76(6):969-976. Takemura T, Yoshioka K, Murakami K, Akano N, Okada M, Aya N, Maki S. (1994) Cellular localization of inflammatory cytokines in human glomerulonephritis. Virchows Arch. 424(5):459-464. Tarraf C, El-Sabban M, Bassam R, Beyr outhy M, Chamoun J, Talhouk R. (2003) Functional consequence of exposure to dieldrin on mammary development and function. Food Addit Contam. 20(9):819-828. Taylor JR. (1985) Neurological manifestatio ns in humans exposed to chlordecone: follow-up results. Neurotoxicology. 6(1):231-236. Theofilopoulos AN, Dixon FJ. (1985) Murine mo dels of systemic lupus erythematosus. Adv Immunol. 37:269-390. Theofilopoulos AN, Lawson BR. (1999) Tumour necrosis factor and other cytokines in murine lupus. Ann Rheum Dis. 58 (Suppl 1):49-55.

PAGE 186

172 Tiegs SL, Russell DM, Nemazee D. (1993) Recept or editing in self-reactive bone marrow B cells. J Exp Med. 177(4):1009-1020. Tomaszewska A, Guevara I, Wilczok T, De mbinska-Kiec A. (2003) 17beta-estradioland lipopolysaccharide-induced cha nges in nitric oxide, tumor necrosis factor-alpha and vascular endothelial growth fact or release from RA W 264.7 macrophages. Gynecol Obstet Invest. 56(3):152-159. Tseng L, Zhu HH. (1998) Progest in, estrogen, and insu lin-like growth factor-I stimulate the prolactin receptor mRNA in human endometrial stromal cells. J Soc Gynecol Investig. 5(3):149-155. Uphouse L. (1986) Single injection with chlo rdecone reduces behavioral receptivity and fertility of adult rats. Neurobeha v Toxicol Teratol. 8(2):121-126. Uphouse L, Mason G, Hunter V. (1984) Pers istent vaginal estrus and serum hormones after chlordecone (kepone) treatment of adult female rats. Toxicol Appl Pharmacol. 72(2):177-186. van Eijk M, de Groot C. (1999) Germinal center B cell apoptosis requires both caspase and cathepsin activity. J Immunol. 163(5):2478-2482. van Loveren H, Vos JG, Germolec D, Simeonova PP, Eijkemanns G, McMichael AJ. (2001) Epidemiologic associations be tween occupational and environmental exposures and autoimmune disease: re port of a meeting to explore current evidence and identify research needs. Int J Hyg Environ Health. 203(5-6):483495. Van Parijs L, Abbas AK. (1998) Homeostasi s and self-tolerance in the immune system: turning lymphocytes off. Science. 280(5361):243-248. Vidaller A, Llorente L, Larrea F, Mendez JP, Alcocer-Varela J, Alarcon-Segovia D. (1986) T-cell dysregulation in patients with hyperprolactinemia: effect of bromocriptine treatment. Clin Immunol Immunopathol. 38(3):337-343. von Muhlen CA, Tan EM. (1995) Autoantibodies in the diagnosis of systemic rheumatic diseases. Semin Arthritis Rheum. 24(5):323-358. Wakeland EK, Wandstrat AE, Liu K, Morel L. (1999) Genetic disse ction of systemic lupus erythematosus. Curr Opin Immunol. 11(6):701-707. Walker SE, McMurray RW, Besch-Williford CL, Keisler DH. (1992) Premature death with bladder outlet obstruction and hyperp rolactinemia in New Zealand black X New Zealand white mice treated with ethi nyl estradiol and 17 beta-estradiol. Arthritis Rheum.35(11):1387-1392.

PAGE 187

173 Walker SE, McMurray RW, Houri JM, Allen SH, Keisler D, Sharp GC, Schlechte JA. (1998) Effects of prolactin in stimulat ing disease activity in systemic lupus erythematosus. Ann N Y Acad Sci. 840:762-772. Wang TP, Ho IK, Mehendale HM. (1981) Correlation between neurotoxicity and chlordecone (Kepone) levels in brain a nd plasma in the mouse. Neurotoxicology. 2(2):373-381. Wang Y, Wang L, Zhao J, Qiao Z. (2005) Es trogen, but not testos terone, down-regulates cytokine production in nicotine-induced murine macrophage. Methods Find Exp Clin Pharmacol. 27(5):311-316. Watanabe-Fukunaga R, Brannan CI, Copela nd NG, Jenkins NA, Nagata S. (1992) Lymphoproliferation disorder in mice e xplained by defects in Fas antigen that mediates apoptosis. Nature. 356(6367):314-317. Wellmann U, Werner A, Winkler TH. ( 2001) Altered selection processes of B lymphocytes in autoimmune NZB/W mice, de spite intact central tolerance against DNA. Eur J Immunol. 31(9):2800-2810. Wilder RL. (1998) Hormones, pregnancy, and autoimmune diseases. Ann N Y Acad Sci. 840:45-50. Wilson AG, Gordon C, di Giovine FS, de Vrie s N, van de Putte LB, Emery P, Duff GW. (1994) A genetic associati on between systemic lupus erythematosus and tumor necrosis factor alpha. Eu r J Immunol. 24(1):191-195. Wither JE, Paterson AD, Vukusic B. (2000) Ge netic dissection of B cell traits in New Zealand black mice. The expanded populati on of B cells expressing up-regulated costimulatory molecules shows linkage to Nba2. Eur J Immunol. 30(2):356-365. Wither JE, Roy V, Brennan LA. (2000) Activat ed B cells express increased levels of costimulatory molecules in young auto immune NZB and (NZB x NZW)F(1) mice. Clin Immunol. 94(1):51-63. Zeng D, Lee MK, Tung J, Brendolan A, Strober S. (2000) Cutting edge: a role for CD1 in the pathogenesis of lupus in NZ B/NZW mice. J Imm unol. 164(10):5000-5004. Zucali JR, Dinarello CA, Oblon DJ, Gro ss MA, Anderson L, Weiner RS. (1986) Interleukin 1 stimulates fibroblasts to produce granulocyte-macrophage colonystimulating activity and prostaglandi n E2. J Clin Invest. 77(6):1857-1863.

PAGE 188

174 BIOGRAPHICAL SKETCH Fei Wang was born on April 22, 1978, in Shanghai, P. R. China, to Xiaoxian Wang and Rongde Wang. He attended Shanghai Ji ao Tong University, China, where he received his Bachelor of Science degree in 2000. Fei Wang joined the University of FloridaÂ’s interdisciplinary program in bi omedical sciences in August 2001, where he pursued doctoral studies under th e guidance of Dr. Stephen M. Roberts in the Department of Pharmacology and Therapeutics, and Dr. Eric S. Sobel in th e Department of Medicine.