|UFDC Home||| Help|
This item is only available as the following downloads:
DETERMINATION OF FUNCTIONAL B CELL DEFECTS IN HIV 1 INFECTION, COMMON VARIABLE IMMUNODEFICIENCY AND SYSTEMIC LUPUS ERYTHEMATOSUS THROUGH THE FLOW CYTOMETRIC ANALYSIS OF NAVE AND MEMORY SUBSETS BY STEPHANIE HUDEY A Thesis Submitted to th e Division of Natural Sciences New College of Florida In partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr. Alfred Beulig Sarasota, Florida May, 2009
ii Acknowledgements I would first l ike to thank my family for their continued support of my education. I would also like to give a big thanks to Dr. Sleasman of the University of South Florida College of Medicine and All Childrens Hospital for allowing me this great opportunity, providing me with feedback on my drafts, and instilling in me a love of immunology and medicine in general. Also, thanks to Dr. Panida Sriaroon for all of her help and support. Thank you to Carla Duff and Matt Morrow for taking time out for me. I would also like t o thank Dr. Beulig for sponsoring this endeavor and serving as my academic advisor for the past two semesters, as well as Dr. Walstrom for serving as my advisor for my first three years at New College. Thanks to Dr. Clore, Dr. Sherman, and Dr. Scudder for serving on my committee. This research was conducted at the Childrens Research Institute at University of South Florida and All Childrens Hospital and was funded by the National Institute of Allergy and Infectious Diseases, part of the National Insti tutes of Health, under a two year exploratory/development grant. My personal expenses were partially funded by the Council of Academic Affairs at New College of Florida.
iii Table of Contents Acknowledgements .. . ii Table of Contents .. . ..iii List of Figures .. . v List of Tables .. . .vi Abstract .. ..vii Background 1.1 Introduction ........... 1 1.2 B Cell De velopment ...1 1.3 Immunological Tolerance Checkpoints . ............. 10 1.4 Expression of Surface Receptors ...... 11 1.5 Defects in B Cell Development . .......17 Systemic Lupus Erythematosus . .17 Common Variable Immunodeficiency .......21 Human Immunodeficiency Virus Type 1 ...........23 1.6 Principles of Flow Cytometry ......... 28 Materials and Methods 2.1 Screening and Recruitment of Subj ects .. ..32 2.2 Enrollment of Subjects . 34 2.3 Study Cohort . ... 36 2.4 Collection of Blood Samples . ..38 2.5 Whole Blood Processing and B Cell Staining . .. ..39 2.6 Data Acqu isition . . 42 2.7 Experimental Design . . .43 2.8 Statistical Analysis . . .48 Results 3.1 Frequency of CD19 + Lymphocytes . ..... 49 3.2 Frequency of CD27 + B Lymphocytes .. 50 3.3 Frequency of CD27 B220 + Nave and CD27 + B220 Memory B Lymphocytes Population s . . . 51 3.4 Frequency of Class Switched Nave and Memory B Lymphocyte Populations ... . 52 3.5 Frequency of CD21 + Nave and Memo ry
iv B Lymphocyte Populations . .. 54 3.6 Frequency of CD23 + Nave and Memory B Lymphocyte Populations 55 3.7 Frequency of CD80 + and/or CD86 + Nave and Memory B Lymphocyte Populations . .. 56 3.8 Frequency of CD27 + B220 Memory B C ells,IgM IgD Class S witched M emory B Cells and CD80 + CD86 + Memory B Cell Populations in Nave and Treated HIV infected Subjects ..58 Discussion 4.1 Frequency of CD19 + Lymphocytes 61 4.2 Frequency of CD27 + B Lymphocytes ... ... ..61 4.3 Frequency of CD27 B220 + Nave and CD27 + B220 Memory B Lymphocytes Population s .62 4.4 Frequency of Class Switched Nave and Memory B Ly mphocyte Populations . .62 4.5 Frequency of CD21 + Nave and Memory B Lymphocyte Populations 63 4.6 Frequency of CD23 + Nave and Memory B Lymphocyte Populations 64 4.7 Frequency of CD80 + and /or CD86 + Nave and Memory B Lymphocyte Populations ... 64 4.8 Frequency of CD27 + B220 Memory B C ells,IgM IgD Class S witched M emory B Cells and CD80 + CD86 + Memory B Cell Populations in Nave and Treated HIV infected Subjects ..65 4.9 In Summary ... ... .66 4.10 Future Direction of Research ..66 Appendix 1 5.1 Enrolled Patient Details .. ..68 Appendix 2 6.1 B Cell Staining Protocol Details . ..72 Appendi x 3 7.1 Certifications received by author to conduct this research .. 77 References ..78
v List of Figures Figure 1: Immunoglobulin structure 4 Figure 2: B cell development and trafficking in humans . 9 Figure 3: A simplified diagram of flow cytometry .. .29 Figure 4: Dot plot of side scatter versus forward scatter . .31 Figure 5: Strategy used to aliquot blood leukocytes . .. ..41 Figure 6: Density dot plot of side scatter versus forward scatter . .. ..45 Figure 7: Histogram and density dot plot of CD19 and CD27 ... ... ..46 Figure 8: Density dot plot of CD27 versus B220 ..... 46 Figure 9: Contour plots demonstrating class switching .. .47 Figure 10: Distribution of CD19 + B Cells ... .. ..49 Figure 11: Distribution of CD27 + B Cells ... 50 Figure 12: Distribution of CD27 B220 + nave B cells and CD27 + B220 memory B cells .. . .52 Figure 13: Contour plot demonstrating class switching . . .. ..53 Figure 14: Distribution of class switched nave and memory B cells ..... 54 Figure 15: Distribution of CD21 + nave and memory B cells .. ...55 Figure 16: Distribution of CD23 + nave and memory B cells . . 56 Figure 17: Distribution of CD80 + CD86 + nave and memory B cells . .. 57 Figure 18: Distribution of memory B cells among HIV nave and treated subjects .. . ..58 Figure 19: Distribution of class switched memory B cells among HIV nave and treated subjects . . . 59 Figure 20: Distribution of CD80 + CD86 + memory B cells among HIV nave and treated subjects . .... .60
vi List of Tables Table 1: Expression and function of selected sur face molecules .. ..15 Table 2: Differences in B cell defects among SLE, CVID, and HIV patients . ..27 Table 3: Patient demographics for all study populations 38 Table 4: Details pertaining to fluorochrome labeled ant ibodies ... ..42 Table 5: Autoimmune patient data ..68 Table 6: Normal ranges for laboratory tests ....70 Table 7: HIV Infected Patient Data . 70
vii DETERMINATION OF FUNCTIONAL B CELL DEFECTS IN HIV 1 INFECTION, COMMON VARIABLE IMMUNODEFICIENCY AND SYSTEMIC LUPUS ERYTHEMATOSUS THROUGH THE FLOW CYTOMETRIC ANALYSIS OF NAVE AND MEMORY SUBSETS Stephanie Hudey New College of Florida, 2009 ABSTRACT There is a common misconception th at HIV 1 infection affects only T cells, but the reality is that it has a drastic impact on both the cell mediated and humoral arms of the immune system. B cell defects have been well described in the literature for systemic lupus erythematosus and common variable immunodeficiency, two diseases that have similarities to HIV 1 infection in terms of B cell dysfunction, such as aberrant B cell activation and differentiation resulting in maturation arrest and lower memory B cell proportions. The purpose of this study was to examine the frequency of expression of activation and differentiation markers in the nave and memory B cell subsets in peripheral blood of the three patient populations. The results indicated that through the flow cytometric analysis of surf ace markers including CD19, B220, CD27, CD21, CD23, CD80, CD86, IgM, IgD and IgG, the functional B cell defects observed in HIV infection were found to be due to late stage defects in B cell development. This was evidenced by decreased late memory B cell f requencies (CD27 + B220 ), increased expression of activation and differentiation markers in this memory population (CD80 + CD86 + ) and
viii decreased expression of complement receptor CD21 in the memory population. The frequency of class switching, however, was ret ained in the memory B cells of HIV infected patients. ____________________________ Dr. Alfred Beulig Division of Natural Sciences
9 Background: 1.1 Introduction Looking closely at B cell development is an impor tant area of research in the field of immunology because defects in this process have been implicated in numerous diseases. This study will attempt to clarify some of the unknowns concerning the relationship between B cell development and the pathology of HIV 1 infection by comparing this patient population with two other diseases in which more detail is known. Systemic lupus erythematosus (SLE) and common variable immunodeficiency (CVID) are two conditions with well described defects in B cell development that can be used effectively to learn more about the B cell defects in HIV 1 infection. They have both similarities and differences to HIV 1 in terms of these defects that will be further explored in this research. It is important first to understand the d etailed process of B cell development in both health and disease. 1.2 B Cell Development There are two classifications for immune responses that defend the body against pathogens. The innate immune response is readily available to combat a wide range of antigens, but it does not lead to lasting immunity and does not have specificity for a particular pathogen. Adaptive immunity, however, often results in immunological memory, which confers lifelong protective immunity to re infection with the same pathogen (Murphy et al., 2007). It also has greater antigen specificity. B cells are critical to the proper functioning of the immune system because they are the means of providing one of the two essential components of the adaptive immune response. These two
10 comp onents are cell mediated immunity, which is mediated by T lymphocytes, macrophages and natural killer cells, and humoral immunity. Humoral immunity, the principal defense mechanism against extracellular microbes and their toxins, is mediated by membrane bo und and secreted antibodies that are produced by B cells at later stages in development. These antibodies found in multiple sites of the body act to bind toxins or viruses and prevent them from recognizing their receptor on a host cell. There are many imp ortant stages in B cell development, however, that precede their production of antibody. The pluripotent hematopoietic stem cells within the bone marrow give rise to all blood cells, including all the cells of the immune system (Akashi et al., 1999). Ther e are basic principles of cell differentiation that are followed from this precursor stem cell stage to the stage of branching into a B cell, T cell or NK cell. The eventual development of B cells is dependent upon the conditions of the microenvironment in the bone marrow and the signals that it provides in the switching on or off of particular genes. First, the hematopoietic stem cell differentiates into multipluripotent progenitor cells, which are no longer self renewing. The multipotent progenitors expre ss a cell surface receptor tyrosine kinase, FLT3 which binds to the FLT3 ligand on stromal cells (Kikushige et al., 2008). This allows for signaling that leads to the next stage of differentiation, the early lymphoid progenitor, which then progresses to th e common lymphoid progenitor. It is at this stage of development that the cells begin to express the interleukin 7 (IL 7) receptor, which is induced by the FLT3 signaling already discussed. Stromal cells begin to produce IL 7, which is a requirement for th e development of B lineage cells (Taguchi et al., 2007). There are other interactions that are essential in this stage, including the interaction
11 between VCAM 1 on stromal cells and the integrin VLA 4. Progenitor cells also interact through the binding of other cell adhesion molecules (CAMs). These many interactions allow for the progression of the progenitor cells to the earliest stage of B cell development, the early pro B cell stage. It is at the pro B cell stage that immunoglobulin gene rearrangement of the heavy chain begins. A cell progresses from the early to late pro B cell stage through interactions between stem cell factor (SCF), a cytokine on stromal cells, and tyrosine kinase Kit on precursor cells. A cell is considered to have a definitive B cel l fate with the induction of E2A, a B lineage specific transcription factor, as well as early B cell factor (EBF) (Murphy et al., 2008). In the early pro B cell stage, the D (diversity) and J (joining) gene segments of the immunoglobulin heavy chain rearr ange and their DNA is joined. In the late pro B cell stage, the V (variable) gene segment joins the combined DJ sequence forming the completed V H chain. As shown in Figure 1, the immunoglobulin structure has two heavy chains (blue) and two light chains (gr een). It also has a variable region (or F ab ), which is the region that binds antigen, and the constant region (or F c ), which is responsible for interacting with the effector functions of the immune system. There are five different immunoglobulin isotypes that define the characteristics of the constant portion of antibody. As the name implies, the variable region of each antibody is highly specific in its amino acid sequence, which allows them to target a wide variety of antigens and bind with high affinit y and specificity. The diversity of antibody binding sites is important for providing the best defense against many unique infectious agents (Abbas et al., 2007). The constant region has little amino acid variability among different antibodies, which is no t necessary because this is not a region that binds antigens.
12 The way in which this unique variability in amino acid sequence is achieved is through somatic recombination of separate gene segments. As previously mentioned, this begins in the pro B cell stage. The variable region of the antibody is encoded by more than one gene segment. The V L (variable light) chain is encoded by two segments, the V (variable) gene segment and the J (joining) gene segment. The V H chain is encoded by these two gene segmen ts as well, but between them is an additional J (joining) gene segment. There are multiple unique copies of each of these gene segments, and it is the multiple ways in which they can join to form the complete V region that generates diversity (Murphy et al ., 2008). This is referred to as combinatorial diversity. Another way in which diversity of the immunoglobulin variable region is generated is through junctional diversity, in which there is an addition or subtraction of particular nucleotides at the joint s between the different gene segments in the process of their recombination and junction (Meek, 1990). There is another form of combinatorial diversity which occurs when the heavy and light chains pair, because they can do so in many combinations to form t he complete antigen binding site of the immunoglobulin molecule. It is important to understand the huge number of unique antibodies that are Figure 1: Si mplified illustration of the immunoglobulin structure indicating the variable and constant regions, as well as the light chains (green) and heavy chains (blue)
13 ultimately produced. It is estimated that as many as 10 11 different receptors could make up the repertoire of rece ptors ultimately expressed on nave B cells (Murphy et al., 2008). Therefore, it is obvious that the need to regulate this process effectively is very important, which will be discussed later. At the end of the late pro B cell stage, the VDJ rearrangement of the heavy chains is complete, but has not yet begun for the light chain genes. The next stage in development is the large pre B cell stage, in which a pre B cell receptor can now be expressed because rearrangement of the V H chain has finished. This rec chain is only transiently expressed on the cell surface. The expression of this pre B cell receptor is necessary for the progression of development, and nearly 45% of cells are lost at this stage and fail to develop any further (Murphy et al., 2008). To reach completion of rearrangement in the immunoglobulin variable region, the next stage of development is concerned with VJ rearrangement of the light chain genes. This is called the small pre B ressed in the cytoplasm at this point. Upon successful assemblage of the light chain, the cell becomes an immature B cell that expresses a complete IgM molecule on its surface. These processes of gene rearrangement of both the heavy and light chain are dep endent upon a complex of enzymes termed the V(D)J recombinase, with two specific components called RAG 1 and RAG 2 (Grawunder et al., 1995). There are two other proteins expressed on the surface of a B cell alongside the pre B cell receptor as well as the B cell receptor in the purpose of transducing signals with interactions in the cytoplasm through their cytoplasmic tails (Martensson, Keenan & License, 2007). Their e xpression continues
14 through later stages of B cell development. It should be noted that these development steps discussed are taking place in the bone marrow. Many of the cells produced at this stage are not viable due to self reactivity, and are eliminate d at this stage, which will be discussed later in regards to tolerance checkpoints. The cells that undergo successful rearrangement can then progress from the bone marrow to the peripheral lymphoid organs, which include the spleen and lymph nodes. During the cells transit and upon arrival it is now in the transitional B cell stage. It can soon be termed a mature nave cell when it acquires surface expression of IgD, along with increased expression of surface IgM. Normally, the nave B cell will leave the peripheral lymphoid tissue and continue to recirculate via the lymph and blood, continually reentering the lymphoid tissue until it encounters antigen. When the mature nave cell finally encounters antigen, it will stop recirculating and again enter the p eripheral lymphoid tissue. In a T cell dependent antibody response, the B cell will internalize the antigen it has bound, process it and then present it as a peptide on its surface (Abbas et al., 2007). The B cell will then come into contact with a helper T cell that has appropriate antigen specificity. This contact occurs on the border between the T cell area and the primary lymphoid follicles occupied by B cells in the lymphoid tissues. It is with the help of the interaction between CD40 on B cells and CD 40 ligand on T cells that B cells become activated and begin differentiating (DOrlando et al., 2007). At this point, many of the B cells will migrate back to the primary follicles and a region will develop in some called the germinal center. The germinal center is a site of intense cell proliferation, and is also where B cells undergo several important modifications, including somatic hypermutation and class switching (Murphy et al., 2008). B cells that
15 progress to germinal centers are termed follicular B cells. Somatic hypermutation acts to further diversify the immunoglobulin receptors and leads to selection of Ig that binds antigen with higher affinity. Class switch recombination is the process by which gene rearrangement occurs in the constant region of the immunoglobulin and results in a change in the isotype of the antibody, acting to increase the functional diversity of the immunoglobulin repertoire without changing its antigen specificity (Maizels, 2005). There are five different isotypes of Ig with unique differences in their C H chain. They differ in their number of heavy chain domains, the number and location of disulfide bonds linking the chains, the lack or presence of a hinge region, and the distribution of carbohydrate groups (Murphy et al., 20 08). The constant region of immunoglobulin is important for recruiting other cells and molecules for help in the destruction and disposal of pathogens to which the variable region of the antibody has bound. They can also bind complement to initiate the com plement cascade and allow antibodies to reach different body compartments through active transport. The structural differences among the five antibody isotypes result in their unique effector functions. IgM and IgD appear first on the B cell surface in de velopment and therefore are termed nave B cell antigen receptors. IgM is responsible for complement activation (Abbas et al., 2007). IgD is unique in that it is always bound to the membrane and does not have a secreted form. It is through class switching that the IgG, IgA and IgE isotypes are generated. IgA functions in mucosal immunity, can form dimers in secreted form, and is most often found in the gastrointestinal and respiratory tracts (Woof & Kerr, 2006). IgE functions in defense against helminthic parasites and immediate hypersensitivity, and is found often in the skin. IgG functions in opsonization, complement activation, cell mediated cytotoxicity
16 that is antibody dependant, and feedback inhibition of B cells. The isotype that is generated differs depending on the antigen that was encountered by the B cell earlier, and often B cells in different anatomic sites switch to different isotypes (Abbas et al., 2007). Having these different antibody isotypes is extremely important in the human body for pro viding efficient defense against all pathogens, as is evident in individuals that have deficiencies in one particular class, leaving them more susceptible to infection. Before B cells can leave the germinal centers, they must encounter antigen again to se lect for survival of the high affinity cells only. In this process, follicular dendritic cells in the germinal centers display antigen. The cells that recognize antigen and bind with high affinity are selected to survive. Those that do not will undergo apo ptosis. It is at this point that the germinal center B cells, some of them class switched, can differentiate into two cell types: antibody secreting plasma cells or memory B cells. Plasma cells will become factories of constant antibody production and secr etion and will return to the bone marrow where they are long lived. These antibodies are useful at providing immediate protection against an antigen months or even years after it initially encountered it (Wrammert & Ahmed, 2008). These cells have little ex pression of surface immunoglobulin because it is mainly secreted. Also, they secrete primarily antibody of the IgM isotype, so have not undergone class switching. The second fate of a germinal center B cells is its development into a memory B cell. A memor y B cell is long lived and divides very slowly if at all. They express primarily surface immunoglobulin and secrete very little. These cells inherit the genetic changes that were made in the germinal centers, including somatic hypermutation and class switc hing. They continue to recirculate between blood, lymph and the peripheral lymphoid organs where they encounter antigen.
17 Figure 2: Illustration of B cell development and trafficking in humans. As described, an immat ure B cell leaves the bone marrow where it enters the bloodstream followed by the spleen or lymph nodes. After encountering antigen, it will home to the lymph node and begin further differentiation. The important process of class switching occurs in the ge rminal centers. Adapted from Warnatz & Schlesier 2008. Figure 2 summarizes all of the stages in B cell differentiation just discussed and depicts where the cells are trafficking. It shou ld be noted that there is another path in B cell development in which T cell interaction is not involved, called T cell independent antibody response. The antigens that B cells can recognize in the absence of T cells include nonprotein antigens such as pol ysaccharides and lipids. The antibodies that are produced following this response are typically of low affinity and consist mainly of the IgM isotype with very limited isotype switching to IgG. This type of a response can occur in the spleen, peritoneal ca vity,
18 mucosal sites, or even the bone marrow (Abbas et al., 2007). A distinct subset of B cells that are found in the marginal zones surrounding the lymphoid follicles in the spleen are called marginal zone B cells. They respond mainly to polysaccharides and differentiate into short lived plasma cells that produce antibody of the IgM isotype. Unlike in mice, they do recirculate rather than remain in the marginal zones in the spleen (Murphy et al., 2008). 1.3 Immunological Tolerance Checkpoints Toleranc e checkpoints are extremely important in the development of B cells. It is with positive and negative selection at several stages in the development process that B cells are tested for autoreactivity. In autoreactivity, the cell recognizes self tissue as a ntigen. Therefore, the sole purpose of these development checkpoints is to eliminate the B cells that are not tolerant to self antigens. The first tolerance checkpoint is called central tolerance, because it occurs in the bone marrow, the central lymphoid organ. At the immature B cell stage, when gene rearrangement of Ig heavy and light chains is completed and IgM is expressed on the surface of the cell, the first test for self tolerance occurs (Abbas et al., 2007). The B cell receptor which includes surfac will have the opportunity to react with its surrounding environment. If a cell is found to recognize self antigen, it has four possible fates: it can undergo cell death by apoptosis or clonal deletion, it can become anergic or unresponsi ve to antigen, it can undergo receptor editing, or it can remain immunologically ignorant because the antigen it encountered was of a very low concentration or bound with a weak affinity. Approximately 55% of newly arising B cells in the bone marrow are c onsidered autoreactive and potentially
19 harmful, but these are usually removed (Yurasov et al., 2005). The cells that are successful at self tolerance can then progress from the bone marrow to the spleen, a peripheral lymphoid organ, through the central sin us and venous blood supply. Peripheral tolerance is important for deleting cells responsive to self antigens not initially encountered in the bone marrow. The principal mechanisms of peripheral tolerance include anergy, deletion, and suppression. Regulato ry T cells are responsible for inducing the suppression of these B cells (Murphy et al., 2008). The first checkpoint in the periphery is said to occur immediately upon exit from the bone marrow, at the point between the transitional B cell stage and matur e nave B cells stage (Yurasov et al., 2005). The second checkpoint is said to occur between the mature nave stage and development into either a marginal zone cell or follicular B cell (Jacobi & Diamond, 2005). The germinal center is the third critical ch eckpoint in B cell development. During the process of somatic hypermutation and isotype switching, many cells will obtain new binding specificity and some of them can become potentially autoreactive. A mechanism of negative selection must remain intact fo r the proper elimination of B cells because those that survive will go on to join the pool of long lived plasma cells and memory cells. As will be seen, defects in these tolerance checkpoints during B cell development occur in autoimmune diseases, such as systemic lupus erythematosus, as well as other immune deficiency diseases. 1.4 Expression of Surface Receptors B cells at different stages of development acquire unique surface receptors that enable their interaction with other cells and components of t he immune system. They also
20 express different receptors depending upon the path they take in differentiation. These receptors are important for signaling and forming complexes with cells and proteins, but also allow for positive or negative selection of th e cell. There are many types of surface receptors of different classes or families and new ones are continually being discovered (see Table 1). Some of the families of receptors or proteins include the Ig superfamily, integrin family, TNF receptor family, selectin family and the C type lectin family. In immunology, many of these receptors are called clusters of differentiation or CD and are given a unique number. Advancements in flow cytometry have allowed for the analysis of many surface proteins at one time on a cell, which enables cellular phenotype to be identified based on the expression of CD proteins. The earliest surface expression begins with CD34 + hematopoietic stem cells. This marker is lost by the pre B cell stage, at which point B cells begi n to express CD19 + It is widely used as a B cell marker. When analyzing only B cells in flow cytometry, cells positive for this marker are selected. CD19 + is a member of the Ig superfamily and is expressed throughout all stages of B cell development until differentiation into a plasma cell, at which point it is lost (Yurasov et al., 2005). At the immature B cell stage, there is coexpression of other surface receptors including CD10 and CD20. As discussed earlier, this is the stage at which IgM is expressed as well. It is at the transitional and mature nave stages that B cells begin to express the CD21 surface receptor. CD21 is a receptor for the C3d complement component. It is considered an activation marker on B cells because it forms a coreceptor with CD 19 and CD81. These three receptors are referred to as the B cell coreceptor complex. This complex greatly enhances responsiveness to antigen through a particular signaling pathway. Through the activation of complement,
21 the components are deposited on the a ntigen of interest, causing the B cell receptor to then bind to the antigen. CD21 binds to the complement, and signals are generated through CD19 that activates a PI 3 kinase signaling pathway that in turn stimulates the B cell response. An intracellular s ignal results from this signaling pathway that leads to increased antibody production and differentiation of B cells. This also makes the cell more effective at eliciting T cell help (Murphy et al., 2008). CD23 is another surface receptor that is expressed early at the transitional B cell stage and usually loses expression earlier in differentiation. It is a low affinity IgE receptor of the c type lectin family. It may be implicated in enhancing IgE antibody levels in certain cases, but otherwise its functi on is not clearly understood (Abbas et al., 2007). Its expression during B cell development was investigated in more detail in this study. Some studies have indicated that marginal zone B cells can be differentiated from the major population of follicular B cells because they tend to express lower levels of CD23 and higher levels of CD21 (as reviewed by Morrow et al., 2008). Two other B cell surface markers that have increased expression, or upregulation during activation are CD80 and CD86. CD80, also call ed B7 1, is a member of the Ig superfamily and shows increased expression after a nave B cell has encountered antigen and becomes activated. This costimulatory molecule can help in the enlistment of helper T cells. CD86, or B7 2, is also of the Ig superfa mily and plays the same role. This upregulation of CD80 and CD86 often occurs in conjunction with other changes in the B cell, including increased expression of receptors for particular T cell derived cytokines and changes in expression of chemokine recept ors, both of which also allow the cell to
22 interact more easily with helper T cells. Some studies have shown that CD86 is upregulated first, and later CD80 expression ensues (Morrow et al., 2008). After antigen encounter and helper T cell interaction, a me mory B cell results, and it gains the expression of CD27. This receptor is a member of the TNF receptor family and has widely been used as a memory B cell marker. Therefore, nave B cells will have a CD19 + CD27 phenotype and memory B cells will have a CD19 + CD27 + phenotype. As already discussed, it is at the mature nave stage of development that surface IgM expression is joined by IgD expression. It is after class switching occurs in the germinal centers that memory cells can begin expressing IgA, IgG, and IgE. Often non class switched cells are indicated as IgM + IgD + and class switched cells are indicated as IgM IgD If a B cell takes the alternative route and differentiates into a plasma cell, this can be seen by expression of alternative surface recepto rs. Plasma cells are often differentiated by their expression of CD138. They retain the expression of CD27 but lose CD19, CD21, CD23, CD80 and CD86 expression. Recently, B cell phenotyping studies have been looking at development in terms of another marke r, called CD45R. It is commonly referred to as B220 and is one of the many isoforms of CD45. It is a transmembrane protein tyrosine phosphatase and is required for signal transduction through antigen receptors (as reviewed by Bleesing & Fleisher, 2003). U nlike expression in mice, there are stages in human B cell development in which B220 is downregulated, and this distinction has been used to look further at the subsets of nave and memory B cells. For instance, studies have examined naive cells in terms o f CD27 B220 + and CD27 B220 and they have found that B220 expression is
23 downregulated at some point after or during migration from follicular zones to germinal centers, but before reentry into circulation (Morrow et al., 2008). This is the stage just prior to a B cell becoming mature. Therefore, B220 expression can be used to pinpoint a particular stage in development more accurately. It is hypothesized that a CD27 B220 + phenotype indicates a very new nave mature B cell, and CD27 + B220 indicates a more es tablished memory B cell, with CD27 + B220 + as an intermediate stage (Bleesing & Fleisher, 2003). There has been great interest in looking at these B220 subsets in terms of CD27 expression in patients with diseases that affect B cell maturation, which is prec isely what this study entailed. Autoimmune disease patients, HIV infected patients and CVID patients all have particular B cell defects, with similarities and differences among them that were investigated in this study. Table 1: Expression and Function of Selected Surface Molecules in Humans Surface Protein Common Synonyms Molecular weight, structure; family Main Cellular Expression Known or proposed functions CD19 B4 95 kD; Ig superfamily B cell marker expressed on all B cells B cell activation; forms a coreceptor complex which delivers signals that synergize with signals from B cell antigen receptor complex. CD27 none 55 kD; TNF receptor memory B cell marker binds CD70; can function as a costimulator for T and B cells B220 CD45 isoform 220 kD; CD45 Isoform; protein tyrosine phosphatase family; fibronectin type III family B cells expressed off and on in B cell development plays critical role in regulating T and B cell antigen receptor mediated signaling
24 CD21 CR2; C3d receptor 14 5 kD; complement control protein family mature, activated B cells Activation marker; receptor for complement component C3d and Epstein Barr virus. Forms a coreceptor with CD19 and CD81, delivering activating signals to B cells. CD23 low affinity IgE receptor 45 kD; c type lectin mature, activated B cells regulation of B cell activation; adhesion molecule CD80 B7 1; BB1 60 kD; Ig superfamily activated B cells co stimulator for T cell activation; ligand for CD28 and CD 152 (CTLA 4) CD86 B7 2 80 kD; Ig superfamily activated B cells co stimulator for T cell activation; ligand for CD28 and CD152 (CTLA 4) IgM none 190 kD mature nave B cells Nave B cell antigen receptor, complement activation IgD none 184 kD mature nave B cells nave B cell antigen receptor; co expressed with IgM on cell surface; function unknown IgG none 4 subtypes (1,2,3, or 4), 146 165 kD memory B cells opsonization, complement activation, antibody dependent cell mediated cytotoxic ity, neonatal immunity, feedback inhibition of B cells Adapted from Murphy et al., 2007 & Abbas et al., 2008
25 1.5 Defects in B Cell Development There are several classifications for diseases that are caused by immune system dysfunctions. Autoimmune disea ses, such a system lupus erythematosus, involve defects in the tolerance checkpoints during B cell development that cause autoantibodies to be produced. HIV 1 is a virus that causes acquired, or secondary immune deficiency, and progressively leads to loss in function of both the cell mediated and the humoral arms of the immune system. One of the hallmarks of B cell dysfunction in HIV 1 is polyclonal activation of B cells with impaired responses (De Milito et al., 2001). According to the World Health Organiz ation, the prevalence of HIV infection worldwide has doubled in the past ten years, with 40 million people infected in 2006, compared with 20 million people infected worldwide in 1996. One million of these infections are in North America. Common variable immunodeficiency (CVID) is a primary immunodeficiency of a very heterogeneous nature, but patients have an intrinsic B cell defect that leads to a lack of antibody production in spite of normal B cell frequency in peripheral blood (Haymore et al., 2008). Primary immunodeficiencies affect approximately 250,000 people in the United States, and can often cause early death due to increased prevalence of cancer (Boyle & Buckley, 2007). System Lupus Erythematosus (SLE) Autoimmune diseases are a group of diso rders characterized by tissue damage in which immune responses are directed at autoantigens, or self tissues. Both genetic and environmental factors predispose or trigger autoimmunity. Some autoimmune diseases are systemic and others are tissue or organ sp ecific. Systemic lupus erythematosus is a
26 systemic disease that affects tissues as diverse as the brain, skin, kidneys and joints (Murphy et al., 2008). Interestingly, SLE is most commonly diagnosed in women, particularly of African American and African Ca ribbean descent (as reviewed by Lipsky, 2001). Some of the environmental factors that may trigger SLE are thought to include exposure to ultraviolet light, cigarette smoke, exposure to particular microbial agents and exposure to Epstein Barr virus (James e t al., 2001). However, little other evidence of environmental triggers has emerged, and genetic factors are considered to play a much greater role (Lipsky, 2001). There are known to be defects in the B cell tolerance checkpoints, but it has been diffic ult to pinpoint when tolerance is first broken, whether in the bone marrow or the periphery (Yurasov et al., 2005). This breakdown in tolerance allows escape into the circulation of B cells that are autoreactive, producing autoantibodies that often attack nuclear components of cells. The reason that SLE is often characterized by such diverse clinical abnormalities is due to the broad nature of the autoantibodies produced. Patients can experience many manifestations of SLE and must meet particular criteria f or diagnosis. Some of the common manifestations include the characteristic butterfly or malar rash, discoid rash, photosensitivity, oral ulcers, non erosive arthritis, pleuritis or pericarditis, nephritis, and neurological disorders including seizures and psychosis. The laboratory evidence includes hematological disorder, such as hemolytic anemia, leukopenia, lymphopenia and thrombocytopenia. There is most often evidence of a positive antinuclear antibody test, which detects titers of antibodies directed at the cell nucleus by immunofluorescence.
27 The titers of the specific autoantibodies are then looked at individually and they include anti dsDNA, anti ssDNA, anti Smith, anti histone, anti RNP, anti SS A/Ro, anti SS B/La, and anti topoisomerase (Scl 70). Anti dsDNA and anti ssDNA antibodies are simply those directly at double and single stranded DNA in the nucleus of cells. Anti Smith antibodies are usually unique to patients with SLE and are directed at an RNA binding nuclear protein of the Smith family. They are unique in their three dimensional structure of a ring assembly with an RNA oligonucleotide arranged as an arc in the center (He & Parker, 2000). Anti SS A/Ro and anti SS B/La antibodies were named after the first patients found to possess them, a long with anti Smith antibodies. SS A/Ro and SS B/La are also ribonucleoproteins. A positive test for these autoantibodies is usually associated with Sjogrens Syndrome, in which the immune system attacks the exocrine glands causing dry mouth and dry eyes (Dawson et al., 2006). Anti RNP antibodies are directed specifically at other ribonucleoproteins in the nucleus. Anti Scl 70 antibodies are directed at the enzyme topoisomerase, and are often associated with scleroderma (Cepeda & Reveille, 2004). There ar e also anti phospholipid antibodies that have been discovered including those directed at cell membrane components including cardiolipin, phosphatidylserine, 2 glycoprotein I. The failure to remove these self reactive antibodies during B cell tolerance checkpoints results in their progression to the periphery where they enter circulation and are sometimes targ eted to specific areas such as the joint tissue or kidneys. Individuals who develop SLE have evidence of high titers of autoantibodies in circulation long before they develop clinical symptoms of the disease (Arbuckle et al., 2003).
28 Treatment for SLE pat ients usually includes a combination of drugs. Two of the most common are prednisone, a corticosteroid that acts as an immunosuppressant, and hydroxychloroquine. Hydroxychloroquine, interestingly enough, is an anti malarial that acts to reduce inflammation in SLE patients and until recently, its mechanism of action was unknown. However, it has been discovered that this drug reduces the activation of dendritic cells by decreasing TLR signaling, thus mitigating the inflammatory process (Lafyatis et al., 2006) For patients with nephritis, cellcept, or mycophenolate mofetil, is often used as an immunosuppressive that blocks T cell and B cell activation to reduce the inflammation in the kidneys. Studies have indicated that there are several possible checkpoints at which B cell selection in SLE patients is aberrant. One such study has indicated that there may be a defect as early as the tolerance checkpoint between the bone marrow and periphery (as reviewed by Jacobi & Diamond, 2005). More commonly however, the s tages at which tolerance is probably defective in SLE occur at the peripheral checkpoint in the transition between a new emigrant cell, or transitional B cell, and a mature nave cell, as well as the stage in the germinal centers (Yurasov et al., 2005). Th is first defective checkpoint in the periphery subsequently results in the expression of large numbers of self reactive antibodies in the mature nave B cell compartment. The second faulty checkpoint in the germinal centers would subsequently result in exp ression of self reactive antibodies in the memory B cell subset. Numerous past studies have looked at the expression of particular cell surface receptors on different B cell subsets of SLE patients. Flow cytometric analysis of nave (CD27 ) and memory (CD2 7 + ) B cell subsets has identified that there is a relative
29 predominance of CD27 expressing peripheral blood B cells in patients with active SLE. This accumulation of immature B cells in the blood indicates disturbed maturation or premature exodus of immat ure B cells from the bone marrow. There also seems to be a decreased proportion of memory B cells. Several studies have indicated that activation markers CD80 and CD86 are both over expressed in patients with SLE. One study also determined that CD86 is inv olved in polyclonal antibody production (Nagafuchi et al., 2003). The increased expression of these markers is an indication of the polyclonal B cell activation that is a well described feature of SLE. Hypergammaglobulinemia, or increased antibody levels, is another feature of SLE that is described often in the literature. This is probably due to the increased frequency of plasmablasts, which secrete antibody at a high rate (Sato et al., 2004). To summarize, SLE patients have several abnormalities in the f requency of blood B cell subsets, such as expanded nave B cell and activated but diminished memory B cells. Common Variable Immunodeficiency (CVID) Common variable immunodeficiency is a very heterogeneous disorder defined by impaired immunoglobulin pr oduction. It is one of the most frequent primary immunodeficiencies found in adults of European origin (Piqueras et al., 2003). Primary immune deficiency diseases are a class of disorders characterized by an intrinsic defect in the immune system that has n ot been acquired, or is not secondary to infection with a virus, treatment with chemotherapy or any other external agent. Both sporadic and familial cases occur. Inheritance can occur with both autosomal dominant and recessive inheritance patterns (Abbas et al., 2007). It is the absence of plasma cells in those with
30 the disease which suggests a block in B cell differentiation to antibody producing cells. This lack of antibody production could be attributed to intrinsic B cell defects, deficient T cell help and excessive suppressor cell activity (Vodjgani et al., 2007). Clinical manifestations of this disorder include recurrent sinopulmonary infection by encapsulated bacteria, intestinal dysfunction, granulomatous or neoplastic disorders, splenomagaly, an d enlarged lymph nodes (Warnatz et al., 2002; Piqueras et al., 2003). Laboratory results indicate reduced serum levels of all class switched immunoglobulin isotypes (IgG, IgA, IgE), which is the reason patients suffer from frequent infection. Patients als o have an inability to respond to protein and polysaccharide antigen in immunization due to the antibody defects (Haymore et al., 2008). A majority (~80%) of CVID patients display normal numbers of T and B cells in peripheral blood, but 5 10% of patients have extremely low B cell counts, suggesting a defect in development at early development stages in the bone marrow. Another 5 10% of patients have a T cell deficiency (Warnartz et al., 2002). Therapy for CVID patients includes monthly treatments with int ravenous gammaglobulin, or IVIG, sometimes in combination with antibiotics. This IVIG treatment provides the patient with the immunoglobulin they need to prevent infections. This treatment is now also available subcutaneously, which is often preferred by m any patients (Garcia et al., 2007). Many studies have been looking more specifically at the CVID patients with normal numbers of B and T cells in peripheral blood. Flow cytometric analysis of their B cell subsets has attempted to distinguish the stage s at which B cell development is impaired. A particular study that looked at mature class switched memory B cells (CD27 + IgM IgD ) in CVID patients revealed that this subset is greatly diminished
31 (Vodjgani et al., 2007). However, these patients were found to possess normal numbers of mature nave B cells, indicating that the defect in development occurs in the periphery during late B cell differentiation. The production of memory B cells is linked to the formation of germinal centers in the peripheral lym phoid organs, as already discussed, so it has been suggested that a significant reduction in memory B cells in CVID is due to disturbed germinal center reactions (as reviewed by Warnatz et al., 2002). One suggestion is that this defect could be due to impa ired expression of activation markers like CD80 and CD86, which will be studied here. Unlike SLE in which CD27 + B cells are reduced but hyper activated to produce larger than normal amounts of antibody, CVID patients have impaired production of class swi tched antibodies. Therefore, the two conditions are alike in the sense that they both have abnormal B cell differentiation, but are different in that SLE patients possess hypergammaglobulinemia and polyclonal B cell activation and CVID patients do not. Human Immunodeficiency Virus Type 1 (HIV 1) Infection There is a common misconception that HIV 1 infection affects primarily T cells, but the reality is that it has a drastic impact on both the cell mediated and humoral arms of the immune system. HIV 1 i s a member of the lentivirus family of animal retroviruses. Lentiviruses are slowly progressive and are capable of long term latent infection of cells and short term cytopathic effects (Abbas et al., 2007). The virus particle consists of two strands of RNA and associated enzymes, including reverse transcriptase, integrase, and protease, packaged within a core of viral proteins. This is surrounded by a phospholipid bilayer that includes membrane proteins gp120 and gp41 which are bound
32 to the envelope and req uired for infection of cells. Through the binding of these membrane proteins to CD4 on T cells, the infection process begins. The virus can then fuse with the membrane of the cell and empty its viral genome into the cytoplasm. It is then able to use revers e transcriptase to convert its RNA to DNA, which is then incorporated into the CD4 T cells genome and the viruss genes are transcribed. New HIV proteins and its virion core can then be assembled and the replicated viruses expelled from the cell to contin ue the spread of infection. This process of virus production in CD4 + cells causes their numbers to diminish either directly by cell lysis or indirectly through functional inhibition. Other cells of the immune system also become infected with the virus, inc luding macrophages, dendritic cells and of course B cells. There are many clinical features that are associated with HIV, although periods of clinical latency are also possible in which little symptoms are present. Acute clinical features include fever, headaches, sore throat with pharyngitis, lymphadenopathy and rashes. When CD4 + T cell counts drop to a very low level and viral loads are persistently high, the state of AIDS is reached. Those with AIDS often fall victim to numerous types of opportunistic infections, tumors, encephalopathy and wasting syndrome. Depending on the individual, this state may not be reached for years or even decades. In recent years, it has been possible to keep the progression of the disease in better control with the use of ma ny types of antiretroviral drugs (ARVs) in particular combinations. The reason it is so difficult to combat HIV is due to its extremely high mutation rate which allows it to evade detection by antibodies, as well as T cells generated in response to viral p roteins. Many studies have concluded that HIV causes polyclonal activation of B cells, B cell hyperreactivity, increased cell turnover and hypergammaglobulinemia. HIV has also
33 been found to cause increased expression of activation markers CD80 and CD86 (De Milito et al., 2004). These B cell abnormalities are due to both direct and indirect interactions between B cells and the virus as HIV infection results in persistent viral replication. In direct interactions, HIV has been found to bind B cells in viv o via the complement receptor CD21 discussed earlier, although this is probably a minor pathway (Moir & Fauci, 2009). Additional possible receptors on B cells have been identified, including C type lectin receptors and surface immunoglobulins of the V H 3 f amily. The indirect interactions between HIV and B cells are largely unknown, but many ideas have been speculated. They are usually the result of HIV induced immune cell activation and CD4 + T cell depletion. Increased serum levels of IL 7 have been associ ated with increased B cell immaturity and decreased responses to antigens (as reviewed by Moir & Fauci, 2009). B cell abnormalities in HIV infection have also been associated with the induction of IFN stimulated genes and possibly increased levels of B ce ll activating factor (BAFF) and lipopolysaccharide (LPS) (as reviewed by Moir & Fauci, 2009). These are thought to be associated specifically with B cell hyperactivation. Two other HIV proteins, gp120 and Nef, have been proposed to act either directly or indirectly to activate B cells through mechanisms that remain unknown (Swingler et al., 2008). Many studies have looked specifically at the subsets of B cells that are increased or lost with HIV infection. Flow cytometric analysis has concluded that there is a loss of CD27 + memory B cells in HIV infection. This was found in both nave patients and patients undergoing ARV therapy (De Milito et al., 2001). An increase in the immature transitional B cells was also found. Not only this, but the population of s hort lived plasmablasts was also increased with HIV viraemia, which would be the reason for the
34 increased antibody production, as plasmablast have a very high rate of immunoglobulin production. There is also some evidence that defects in B cell development are present because the production of autoantibodies has been implicated in the disease (as reviewed by Moir & Fauci, 2009). It can be seen that B cell defects in HIV have many similarities to those in SLE and CVID. Aberrant B cell activation and differ entiation are common features shared by all three conditions. Polyclonal B cell activation and increased antibody levels have been observed in SLE and HIV while maturation arrest and lower memory B cell proportions have been described in all three conditio ns. The area of research in which little conclusive evidence has been found is determining the stage or stages in B cell development in HIV infection in which defects occur. As discussed, research has been done in SLE and CVID patients determining precisel y where the defects occur by looking more specifically at the B cell subpopulations by flow cytometry through analysis of their surface markers. The purpose of this study is to do just that in HIV, and to take a novel approach in doing so. The expression o f B220, the CD45 isoform discussed earlier, has been found recently to be helpful in pinpointing more precisely a true early nave B cell and a true late memory B cell by its expression or lack of expression (Bleesing et al., 2003). Therefore, in this study, an early nave B cell will be taken to have the phenotype CD27 B220 + and a late memory B cell will be phenotyped as CD27 + B220 This study will look at many activation and co stimulatory surface receptors on both nave and memory B cells in order to see where expression is up regulated or down regulated. All patient groups will be compared to a population of healthy controls. The primary hypothesis is that through analysis of surface markers including CD19, B220, CD27,
35 CD21, CD23, CD80, CD86, IgM, Ig D and IgG, the functional B cell defects observed in HIV infection will be found to be due to late stage defects in B cell development. This will potentially be evidenced by decreased late memory B cell frequencies, increased expression of activation and d ifferentiation markers in this population and decreased expression of class switched memory cells as compared with healthy controls. A second part to the hypothesis is that the expression of these receptors will be similar in both the HIV nave and treated subpopulations, as studies have indicated that therapy does not correct some B cell defects (as reviewed by Moir & Fauci, 2008). The phenotyping will also be conducted on SLE and CVID patients to examine where their similarities and differences lie in ter ms of B220. As past studies have only looked more broadly at nave and memory B cells, using the novel B220 surface receptor will result in more accurate and fine tuned data. Table 2: Differences in B cell defects among SLE, CVID and HIV patients Condi tion Systemic Lupus Erythematosus (SLE) Common Variable Immunodeficiency (CVID) HIV Infection Disease Category Autoimmunity Primary immune deficiency Acquired (secondary) immune deficiency Polyclonal B cell activation Yes N o Yes Antibody levels Increased Decreased Increased Autoantibodies Yes Yes Yes Abnormal B cell differentiation Yes Yes Yes B Cell Defects Immunization responses Normal Abnormal Abnormal
36 1.6 Principles of Flow Cytometry All of the phenotyping discussed earl ier that will be conducted in this study utilize a flow cytometer. In immunology, flow cytometry is useful for the purposes of determining tissue lineage, maturation stage and activation status of a particular cell type (Abbas et al., 2007). This informati on is obtained through the analysis of the cell surface or intracellular expression of different molecules. The expression of these surface molecules can be detected through the use of fluorescently labeled monoclonal antibodies specific to those molecules Flow cytometry functions on the basic principle of detecting and counting individual cells as they pass through a laser beam one at a time in a continuous stream of fluid. Cells of a mixed population isolated from whole blood are first incubated with th e monoclonal antibodies labeled with fluorochromes. These antibodies will be specific to the surface proteins of interest. These labeled cells are then taken up by the apparatus and combined with a large volume of saline where they are passed through a noz zle, creating a stream of liquid with cells individually spaced. The cells in the liquid stream encounter a laser beam one at a time, with each occurrence causing the laser light to scatter. If there is a fluorescently labeled antibody bound to the cell, i t will become excited and fluoresce. The scattered light, as well as the fluorescent emissions are detected by very sensitive photomultiplier tubes. The photons coming from the sample are first converted by the detectors into photoelectrons. The electric c urrents of these are subsequently converted into voltages (Snow, 2004). This conversion into electrical signals by the optical and electronics system results in an eventual output of the data by the computer system.
37 Flow cytometers have become more complex in terms of the number of lasers they use and the number of colors they can analyze. Currently, the most advanced cytometer uses four fixed alignment lasers and has the ability to detect up to 18 colors (BD Biosciences). This allows for multiple a ntibody labeling and more precise determination of the phenotypes of cells. The four common lasers emit light at 355 nm, 405 nm, 488 nm, and 633 nm. The fluorochromes that are used and the antibodies they are bound to are chosen very strategically, so know ledge of the wavelengths of the lasers is critical. It is also necessary to understand the complication that emission spectra of the fluorochromes will be overlapping, and compensation must be performed to account for Figure 3: A simplified diagram of flow cytometry with one laser shown. From Murphy et al. 2008
38 the area in which the emission spectra of two fluorochromes overlap. The fluorescence data from the flow cytometer can be displayed as a histogram of intensity versus cell numbers. Most frequently, however, the data will be looked at in the form of a two dimensional dot plot or contour plot, i n which the fluorescence of one fluorochrome labeled antibody is plotted against that of a second. The scattered light is detected as forward scatter and side scatter, both of which provide very useful information about the type of cell. Forward scatter p roperties reflect the size of the cell and side scatter properties reflect the cells internal complexity (Radcliff & Jaroszeski, 1998). On the basis of this alone, the white blood cells can be differentiated into granulocytes, monocytes and lymphocytes (s ee Figure 4). Lymphocytes are the smallest white blood cells and have little internal complexity. They make up 25 35% of the white blood cell population. Monocytes have a larger size and some granules in their cytoplasm, as well as a bilobate nucleus, refl ecting more internal complexity. They include only 5 10% of the white blood cell population. Granulocytes are slightly larger in size, have internal clumps of many granules in their cytoplasm, as well as a multi lobed, or segmented nucleus. They make up 60 70% of the leukocyte population. Often in the use of flow cytometry for immunology research, the plot of side scatter versus forward scatter is looked at first for the purpose of gating on the subpopulation of white blood cells of interest. As shown in Fi gure 4, the lymphocyte population will be closest to the x and y axes, and the granulocytes will be farthest, with the monocytes in between.
39 Flow cytometry has proven very advantageous because it allows for the analysis of thousands of cells in a matter of seconds (Radcliff & Jaroszeski, 1998). Data analysis is of paramount importance in this process and is quite visually oriented. The trained eye of the technician is just as important in acquiring accurate data as all other steps in the process. P attern recognition is important when two parameter dot or contour plots are used and gates must be made around populations to determine the positive and negative populations. This makes experience of the researcher vital in the process of flow cytomet ry. Figure 4: Dot plot of side scatter versus forward scatter for the purpose of distinguishing white blo od cells on the basis of size (FSC) and internal complexity (SSC). Adapted from http://www.cyto.purdue.edu/hmarchiv/Current/1373.htm
40 Materials and Methods: 2.1 Screening and Recruitment of Subjects Autoimmune and CVID patients were screened for enrollment in this study during weekly meetings with Carla Duff, CRNP, involving scanning the electronic charts of patients with sched uled appointments with four attending physicians at The University of South Florida and All Childrens Hospital in St Petersburg, Florida in the division of allergy, immunology, and rheumatology. These faculty physicians included Dr. Sleasman (chief of the Division of Allergy/Immunology/Rheumatology), Dr. Tang, Dr. Nickeson, and Dr. Sher. The doctors dictation notes and lab results were studied, focusing on specific elevated or lowered markers in their laboratory results, as well as notes concerning their manifestations of disease activity. Patients with autoimmune disease were required to be between the ages of 13 25 and meet the criteria for enrollment by having a history of at least three of the following clinical conditions: autoimmune nephritis (based on renal biopsy), arthritis, autoimmune thrombocytopenia, Coombs positive hemolytic anemia, discoid or malar rash, Sicca syndrome (Sjogrens syndrome), autoimmune parotitis, Raynauds phenomenon, polyserositis, photosensitivity, cerebritis/neuropathy/psych osis, leukopenia, immune mediated coagulopathy (stroke or presence of lupus anticoagulant), autoimmune mucositis, and autoimmune thyroid disease based on the presence of thyroid autoantibodies or abnormal T3/T4 and TSH. Laboratory criteria for inclusion in cluded the presence of at least one of the following: 1. positive antiphospholipid screen as defined by anticardiolipin antibodies, or detectable IgG or IgM antibodies to 2 glycoprotein I. 2. positive
41 antinuclear antibody > 1:320 with one of the following positive results: SS A/Ro, SS B/La, ssDNA, dsDNA, RNP, Sm proteins, or Scl 70. CVID patients of many types were recruited, and no age criteria was set for this population, as patients of all ages visit All Childrens Hospital to be treated by pediatric immunologists. As CVID refers to an immunodeficiency with several subtypes that can affect both the cellular and humoral components of the immune system, many different clinical manifestation are possible and included the following: hypogammaglobulinemia (low levels of IgG, IgA and IgM), poor titer levels in response to vaccination, polyarthritis, splenomegaly, enlarged lymph nodes, frequent recurrent infection including the ears, sinuses, nose, bronchi a nd lungs. The HIV infected patients were recruited from another ongoing study of HIV infection in children and adolescence entitled Biological Implications of HIV 1 Genetic Variability ( ACH IRB # 03 0706). Therefore, little screening of this population was required as most of them were already enrolled in an existing protocol. All of these subjects were known to be HIV infected based on viral antigen detection and positive HIV ELISA and Western blot results. Their viral load and CD4 count were obtained at t he time of draw. Healthy donors were recruited through a current study entitled Analysis of Lymphocytes from Normal Blood Donors (ACH IRB# 03 0881). These volunteer subjects were individuals who had no active medical conditions. There were no age requireme nts for this study population, however an attempt was made to age match them with already enrolled subjects from the other study populations if possible.
42 Exclusion criteria for all of the four study populations included refusal to sign consent or provide adequate blood samples, pregnancy, breast feeding, anemia (Hb< 7 g/dl), or active viral infection with cytomegalovirus (CMV), Hepatitis C virus (HCV), or Epstein Barr virus (EBV). Exclusion criteria for the autoimmune, CVID and healthy donor populations in cluded evidence of HIV infection based on HIV antigen detection. Exclusion criteria for the CVID study population included evidence of agammaglobulinemia (as B cells with surface immunoglobulin were being analyzed in this study, and therefore necessary). A ll subjects received standard medical care including appropriate diagnostic laboratory evaluations, immunosuppressive therapy, and prophylactic therapy. Antiretroviral therapy was permitted for the HIV population. All subjects continued to be monitored an d clinical, viral and immune data collected. 2.2 Enrollment of Subjects After initial screening of patients for enrollment, the prospective patients were approached at their upcoming doctors visits. The study was outlined for them, addressing the requir ements and possible risks of being a research subject and explaining whom this research will benefit. If the patient was under 18 years of age, the study was also outlined for the legal guardian. It was explained to the potential subjects that their blood specimen would be saved in a repository for further research in the future when more advanced technology is available. They were also notified that the only link between the patient and their specimen would be a patient identification number which is used by the laboratory, and the link between the subjects name and number is kept in a locked file only accessible to the research nurse who consents the patient. The risks of
43 taking blood were outlined, including bruising, bleeding and infection. It was also explained that while every attempt is made to keep a patients medical information confidential, there is always a small risk of an unauthorized person gaining access to information. No payment was provided for consenting to the study, but there was no cos t of being a part of it either. Patients were told that they had the option of voluntarily withdrawing their authorization of the use of their blood specimen at any time. In this case, their blood sample would be destroyed. At this time, informed consent f or the study approved by the Institutional Review Board at All Childrens Hospital was obtained by signature. Patients aged over 18 years gave this informed consent, and patients under 18 years provided assent while their parent or legal guardian gave cons ent. A copy of the informed consent paperwork was given to the patient with researcher contact information, and another copy placed in their medical records. The study in its entirety was conducted according to the Articles of Helsinki regarding research i nvolving human subjects. After the paperwork was completed, approximately 30 mL of blood was drawn from the patient by standard phlebotomy from a peripheral vein. The stipulations of the study included a second blood draw of the same volume approximately three months later. The patients chart was then held after doctors dictations were completed for collection of demographics, disease history (past and present symptoms), autoantibody laboratory studies, standard of care laboratory studies, HIV testing hi story, CD4 counts and viral loads, and history of other infections.
44 2.3 Study Cohort As some of the study populations were easier to recruit than others, the number of subjects in each group is variable. Enrollment of patients continues to be ongoing. S amples of whole blood and relevant clinical information were obtained for 12 healthy donors, 8 autoimmune disease patients, 17 HIV infected patients, and 11 CVID patients. The average age, sex ratio, and ethnicity for each of the study groups is presented in Table 3 Of the 8 subjects in the autoimmune disease population, 7 had systemic lupus erythematosus (SLE) and 1 had ITP (idiopat hic thrombocytopenia purpura) with history of autoantibodies. Three of these patients had evidence of disease activity within 6 months of the blood draw and 5 had their most recent disease activity greater than 6 months prior to the blood draw. Clinical manifestations were recorded for these patients in the form of past and current symptoms. Many of the patients had few if any c urrent symptoms, but their history was extensive and included many of the manifestations in the American College of Rheumatology criteria for SLE. All 8 patients had a history of arthritis (some mild), and some but not all subjects had a history of malar r ash, discoid rash, psychosis, leukopenia, pleuritis, nephritis, thrombocytopenia, photosensitivity, and alopecia. All 8 subjects had a history of a positive ANA, 2 subjects with a titer subjects with a titer = 1:640 and 2 subjects with a titer The ANA patterns were both speckled and homogeneous amongst the subjects. Looking more specifically at the antinuclear antibodies, 7 subjects had a history of very high anti dsDNA antibodies, 4 subjects had a history of high anti RNP antibodies, 6 subjects with high anti Sm
45 antibodies, 6 with high anti SSA/Ro antibodies and 4 had a history of high anti SSB/La antibodies. Only some of the patients had lab oratory testing performed for antiphospholi pid antibodies, but of those who did, 4 had a histo ry of anti cardiolipin IgM or IgG antibodies. Four subjects also had a history of low C3 or C4 complement levels. Four subjects had a history of a positive Coombs test. All autoimmune disease patients were being treated at the time of draw with the common drugs used for their conditions. Six patients were being treated with oral prednisone, 4 with hydroxychloroquine and 4 with cellcept (mycophenolate mofetil). The specific drugs of each patient as well as their clinical manifestations and lab results menti oned earlier, are listed in A ppendix 1 Of the 17 subjects in the HIV+ study population, 6 of them were nave (receiving no treatment), one of which was a newborn, and 11 were on therapy. Of the 11 subjects receiving therapy, 7 were categorized as VSIS (v iral success, immune success), 1 was VFIS (viral failure, immune success), and 3 were VFIF (viral failure, immune failure). CD4 + count and viral load for these 17 subjects at time of draw varied considerably (see Appendix 1). The CDC classification of the se patients was unknown. As mentioned earlier, the CVID patients were of many subtypes, with the only criteria being that they did not have agammaglobulinemia, as the existence of some B cells with surface immunoglobulin was necessary for this study. Immun oglobulin levels (IgG, IgA, IgM) from recent laboratory studies were collected for some of the subjects, but not all. No other clinical data were obtained for these subjects.
46 Table 3: Patient Demographics for All Study Populations Healthy Contro ls n= 12 Autoimmune disease subjects n= 8 HIV infected subjects n= 17 CVID subjects n=11 # of females: # of males 5 : 7 7: 1 9: 8 3: 8 Age, mean 22 17 17 33 Ethnicity: White Black Hispanic Asian 11 1 0 0 3 3 1 1 9 7 1 0 10 0 1 0 2.4 Collection of Blood Samples After informed consent was given by screened and recruited patients, their blood was drawn by standard phlebotomy from a peripheral vein by the research nurse. The blood was collected into 4 5 ACD (acid citrate dextrose) purple top tubes. An evacuated tube system was used with a winged infusion set. Gloves were worn by the research nurse at the time of draw and when handling the loose tubes. A total of ~30 ml was collected for each patient during a single visit. Each vacuum tube was labeled prior to collection with the identification number that was assigned to the patient, along with the collection date and time. The document linking the patient to their assigned laboratory research code was recor ded on a master spread sheet by the research nurse. The tubes of blood collected were inserted in a clear plastic bag and placed in a plastic container with a
47 biohazard warning. This process was repeated upon the patients second routine doctors appointme nt approximately 3 months later. The tubes of blood were transferred from All Childrens Hospital to the Childrens Research Center a short distance away. The specimen was assigned a control number by the laboratory, logged in the specimen book, and then placed on a rocker until further processing. 2.5 Whole Blood Processing and B Cell Staining It was required that the blood specimen be processed within 24 hours after the time of collection. Of the ~ 30 ml of blood collected from each subject, 2 ml from a full ACD tube were transferred to a conical vial and placed in the cell staining hood. The remaining blood was separated by a laboratory technician for experiments in other phases of the grant. The peripheral blood mononuclear cells (PBMCs) and plasma w ere isolated separately from the blood, aliquoted, and cryopreserved according to the ACTG Specimen Processing Guide. The cryopreserved PMBCs were not used for the B cell staining to be discussed, but only the fresh 2 ml of blood separated from the start. In healthy adults, B cells account for approximately 10% of all peripheral blood lymphocytes, and 20 40% of these cells phenotype as memory B cells (CD27 + ). Analysis of B cell subpopulations is most accurate using whole blood samples as some B cells are lo st in standard peripheral blood mononuclear cell (PBMC) isolation protocols. Therefore, all flow cytometric analysis of blood B cells in this experiment were done in real time (within 24 hours of blood draw) on whole blood specimens.
48 Protocol First, 1 m l of well mixed whole blood was poured into two 15 ml conical tubes. 10 ml of 1X D PBS was then added to each and mixed gently with inversion. The tubes were then centrifuged at 300 x g for 10 minutes. The plasma/D PBS layer was then aspirated without dist urbing the buffy coat (white blood cells layered on top of the red blood cell layer). Then, 10 ml of 1X Pharmlyse lysing buffer (room temperature) was added to each tube, mixing with gentle inversion. After sitting for 10 minutes at room temperature to all ow the RBCs to lyse, the tubes were centrifuged again 300 x g for 10 minutes. The supernatant was then aspirated without disturbing the cell pellet. A volume of 10 ml of 1X D PBS was added to each tube and the cell pellet was gently re suspended. The tubes were centrifuged again 300 x g for 10 minutes. During this last wash, the flow tubes were prepared. The first blood aliquot in tube 1 would soon be distributed into three separate tubes for analysis. The second blood aliquot would go only into one tube (s ee Figure 5). Plasma cells do not reside in the peripheral blood, however they transit through the peripheral blood en route to lymphoid tissues. Low frequencies (<0.01%) plasma cells may be detected only if large numbers of lymphocytes are analyzed (>100, 000). This is why all of the cells in tube one will stay in a single aliquot for analysis. After centrifugation, the supernatant was aspirated and the cell pellet re 3 in 1X D PBS). The cells we staining buffer was then added to the first three tubes to bring the total volume in all labeled antibodies were then added to the tubes as
49 describe d in Table 4. This was done with the hood light off so as to avoid excessive light because Cy7 dyes are particularly susceptible to light mediated degradation. All cell suspensions were then gently mixed and incubated at 4C for 30 minutes (in the dark). After incubation, 2 ml of 1X D PBS was added to each tube and they were centrifuged at 300 x g for 5 minutes. The supernatant was then decanted and 2 ml of 1X Pharmlyse lysing buffer (room temperature) was added to each tube and inverted to mix. After si tting for 10 minutes at room temperature to allow any last RBCs to lyse, the tubes were centrifuged again at 300 x g for 10 minutes. The supernatant was aspirated without disturbing the cell pellet. Then, 2 ml of 1X D PBS was added to each tube and centrif uged at 300 x g for 5 minutes. The supernant was then decanted and the cell pellet re suspended in 400 3 in 1X D PBS) or fixation buffer (4% paraformaldehyde in 1X D PBS). Figure 5: Strategy used to aliquot blood leukocytes for staining with fluorescent antibodies to specific surface recept ors
50 Tabl e 4: Details pertaining to the dose and type of fluorochrome labeled antibodies used in each of four tu bes containing blood leukocytes Tube Antigen Fluorochrome Manufacturer Dose (Volume) CD19 PE Cy7 BD l CD27 Qdot 655 Invitrogen l B220 (CD45R) PE Cy5.5 Invitrogen l IgM APC BD l IgD FITC BD l IgM/IgD/IgG IgG PE BD l CD19 PE Cy7 BD l CD27 Qdot 655 Invitrogen l B220 (CD45R) PE Cy5.5 Invitrogen 5 l CD23 PE BD l CD21/CD23 CD21 APC BD 20 l CD19 PE Cy7 BD l CD27 Qdot 655 Invitrogen l B220 (CD45R) PE Cy5.5 Invitrogen l CD80 PE BD l CD 80/CD86 CD86 APC BD l CD19 PE Cy7 BD l CD27 Qdot 655 Invitrogen l B220 (CD45R) PE Cy5.5 Invitrogen 5 l IgM APC BD l IgD FITC BD l IgG PE BD l PC (CD138) CD138 PE BD 20 l 2.6 Data Acquisition All data was collected using a BD LSR II benchtop flow cytometer and BD FACSDiva software. This apparatus was equipped with four fixed alignment lasers and
51 had the ability to detect up to 18 colors. Before beginning data collection, color compensation was performed if necessary. This process corrects the overlap of the emission of one fluorochrome into the detector designed to collect the emission of another flu orochrome. Each of the four tubes was then placed on the instrument for collection. As the granulocytes, monocytes and lymphocytes were displayed with forward scatter vs. side scatter, a gate was drawn around the lymphocyte and monocyte populations, exclud ing the granulocytes and RBC debris. Monocytes were retained so as not to lose any lymphocytes that blend in with them in the scatter. The storage gate was set on CD19 + cells, for analysis of B cells only. The event collection stopped when 5,000 CD19 + B ce lls were collected. The next tube, CD21/CD23, was placed on the instrument and the previous steps repeated. This tube was followed by the CD80/CD86 and PC (CD138) tubes. For the PC tube, the number of events to record was increased to 1,000,000. When col lection from all tubes was completed, a tube containing 2 ml of de ionized water was placed on the instrument. 2.7 Experimental Design As already outlined, the white blood cells isolated from the peripheral whole blood of each subject were analyzed by m ultiparameter flow cytometry. The lymphocytes were distinguished from granulocytes, monocytes, and cellular debris by the differences in their size and internal complexity in a density dot plot of side scatter versus forward scatter (see Figure 6). A gate was drawn around the lymphocyte population and it was examined in more detail. The CD19 + cells were selected for using a histogram that was
52 displayed with CD19 + B cells distinguished from CD19 T cells and NK cells (see Figure 7). The frequency of CD19 + c ells was recorded for each subject as a percentage of total lymphocytes. Then, in taking a novel approach to analysis, this CD19 + B cell population was subsequently looked at in terms of B220 versus CD27. A CD27 B220 + phenotype is reflective of an early n ave B cell in development. A CD27 + B220 phenotype is reflective of a late memory cell (see Figure 8). Therefore, this B220 marker is helpful in classifying these nave and memory B cell populations in more specific terms. From this dot plot, the frequenc ies of these two populations of interest were examined. Each of the four quadrants was then gated on individually to examine the expression of surface immunoglobulin and the other surface receptors. The first tube looked at these four quadrants in terms of IgM versus IgD, and IgM versus IgG. The purpose of looking at surface immunoglobulin in this way was to differentiate a non class switched B cell from a class switched B cell. Although data was collected for the expression of IgG, the only data that was necessary for this study was the percentage of IgM IgD B cells in the two populations CD27 B220 + and CD27 + B220 (see Figure 9). This distinguished the frequencies of nave and memory cells that are class switched. The second tube looked at these four q uadrants in terms of CD21 versus CD23 and recorded the frequency of CD21 + cells followed by CD23 + cells in the nave and memory populations. The third tube looked at the four quadrants in terms of CD86 versus CD80 and recorded the frequency of nave and me mory B cells that were CD80 + and/or CD86 + The purpose of looking at these four markers, CD21, CD23, CD80 and CD86 was to evaluate the activation state of the B cells of interest in the four study populations. Lastly, the fourth tube measured the frequency of plasma cells trafficking through the blood with the use of
53 a histogram that measured CD138 + cells only. These cells proved to be too infrequent in the blood for useful analysis to be possible. Therefore, the data from tube four was not examined in any detail. All of the measured frequencies discussed above were arranged in a database with each study subjects patient identification number linked to their results. Figure 6: Density dot plot of side scatter (SSC) versus forward scatter (FSC) for the purpose of identifying the blood lymphocyte population based on size and internal complexity. A blue gate is drawn around this population. Adapted from www.ab direct.com/uploads/gates1.jpg
54 Figure 7: A histogram distinguishing the CD19 + population from the CD19 population based on detected fluorescence by flow cytometry (left). A density dot plot of bloo d lymphocytes showing CD27 versus CD19, with quadrants I and IV displaying the CD19 + population of interest (right). Figure 8: A density dot plot of CD19 + B lymphocytes displaying B220 versus CD27. Quadrants II and IV are gated on individually to examine the nave and memory B cell populations in terms of activation markers.
55 The purpose of collecting these data concerning the frequency of particular cell populations in the patient groups was to define the defects that occur in B cell development. Because B cell defects are more clearly defined in SLE a nd CVID, the HIV infected study population was looked at in terms of these two diseases in the hopes of determining at what stage these malfunctions occur. All groups were compared to the healthy controls. In part two of this study, HIV infected subjects n ot on treatment were compared to those on treatment to see if any differences exist between these populations in terms of memory B cell frequency, frequency of class switching, or the frequency of activation markers CD80 and/or CD86. In the first part of this study, the independent variables were the patients clinical diagnoses as defined by the inclusion and exclusion criteria listed above. These variables included the healthy controls, autoimmune subjects, CVID subjects and HIV infected Nave CD27 B220+ 0 10 2 10 3 10 4 10 5 IgD 0 10 3 10 4 10 5 IgM 10.7 47.9 2.82 38.5 IgD IgM 0 10 2 10 3 10 4 10 5 IgD 0 10 3 10 4 10 5 IgM 0.86 63.3 30.5 5.43 Memory CD27+B220 Figure 9: Contour plots of IgM versus IgD in nave (left) and memory (right) B cell populations. The red boxes indicate the quadrant s of interest for examining class switching (IgM IgD ).
56 subjects. The de pendent variables were the measured frequencies of particular cell populations as defined by expression of the cell surface receptors discussed above. In the second part of this study, the independent variable was the status of current drug therapy receive d within the HIV infected study population. The dependent variables were again the measured frequencies of the particular cell populations of interest, including the CD27 + B220 memory B cell population, IgM IgD class switched population and the CD80 + an d/or CD86 + population. 2.8 Statistical Analysis The frequencies, or percentages, of the cell populations of interest were compared between the four study groups: healthy controls, autoimmune disease subjects, CVID subjects and HIV infected subjects. The frequency of CD19 + cells was compared first, followed by the frequency of CD27 + cells. Next were comparisons of the expression of B220 amongst the CD27 and CD27 + populations, and detailed looks at the surface expression of IgM, IgD, IgG, CD21, CD23, CD80, and CD86 receptors. Graphs of the frequencies for each cell population were made using Graphpad Prism software. Statistics were performed using this program as well. The nonparametric Kruskal Wallis one way ANOVA was used for comparison of the four group s, followed by post hoc pair wise comparisons of the autoimmune group to healthy controls, CVID group to healthy controls, and HIV group to healthy controls. In part two of the study, a nonparametric Mann Whitney t test was used for comparison of the HIV n ave population to the HIV treated population. Data was plotted in columns using a vertical box and whiskers style graph.
57 Results : 3.1 Frequency of CD19 + Lymphocytes A comparison of CD19 + B cell frequencies is shown in Figure 10. The purpose of this comparison was to determine if any significant differences existed in the frequency of B cells among the four populations in the context of total lymphocytes. They were compared using the Kruskal Wallis nonparametric 1 way ANOVA. The graph indicates that t he B cell population has a median value of around 10 15% of total lymphocytes. This median value is slightly larger in the HIV infected population, however a p value of 0.189 indicates that these results are not significant. Figure 10 : Distribution of CD19 + B cells among the four study populations. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages Statistics were con ducted using a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
58 3 .2 Frequency of CD27 + B Lymphocytes A comparison of CD27 + B cell frequencies is shown in Figure 11. The purpose of this comparison was to determine if any significant differences existed in the frequency of memory B cells among the four populations. They were compared using the Kruskal Wallis nonparametric 1 way ANOVA. The graph indicates that the CD27 + memory B cell population has a median value of around 15 20% of total B lymphocytes. A p value of 0.144 indicates that these results are not significant. Also, as can be seen from the whiskers on the graph, the range was considerable for the data in healthy controls and CVID patients. Therefore, nothing can be said about the relationship between CD27 + B cell frequency and disease state in this study. Next, these CD27 + and CD27 populations were looked at more specifically with the addition of the B220 marker to determine if differences exist in these populations when they are examined more precisely. Figure 11 : Distribution of CD27 + B cells among the four study populations. A box and whiskers style graph is use d with the whiskers containing the minimum and maximum percentages. Statistics were conducted using a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
59 3.3 Frequency of CD27 B220 + Nave B Lymphocyt e Population and CD27 + B220 Memory B Lymphocyte Population A comparison of nave and memory B cell populations is shown in Figure 12. The purpose of this comparison was to use a novel approach in further pinpointing these populations for the use of deter mining where defects occur during B cell development. More specifically, the purpose was to look at the memory B cell population in the HIV infected population in particular and determine if the functional B cell defects found in this disease are due to la te stage defects in development as predicted by the hypothesis. The graph of the frequencies of nave B cells had no significant differences between the four populations, with a p value of 0.485 when compared using the Kruskal Wallis nonparametric 1 way ANOVA. This indicates that in this study, there was no significant expansion of the nave B cell population in any of the diseases, however the graph does show slightly higher median values for the autoimmune, CVID and HIV infected populations. The graph of the frequencies of late memory B cells did show a significant reduction of this cell population for both the CVID and HIV infected study groups compared to healthy controls. A p value of 0.008 was obtained first in comparison of all four groups using t he Kruskal Wallis nonparametric 1 way ANOVA. This was followed by Dunns pair wise comparison of each infected group to healthy controls. There was a significant decrease in the frequency of memory B cells in the CVID population with p< 0.01. There was also a significant decrease in the frequency of memory B cells in the HIV infected population with p< 0.05.
60 3.4 Frequency of Class Switched Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations A comparison of class switching among the four study po pulations is shown for both the nave and memory B cell populations in Figure 14. The purpose of this comparison was to examine the possible defects that occur in this process for the three disease populations, particularly HIV, and compare them to the healthy controls. The graph of the frequencies of class switching in nave B cells displayed very low percentages for all populations, with the median consistently under 5%. The four populations were compared using the Kruskal Wallis nonparametric 1 way ANOVA. The differences among these populations were found to be statistically significant with p= 0.024. As the graph indicates, the median values for the CVID and HIV infected populations were slightly reduced. Figure 12 : Distribution of CD27 B220 + nave B cells and CD27 + B220 memory B cells among the four study populations. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were conducted using a Kruskal Wallis followed b y Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
61 The graph of the frequ encies of class switching in memory B cells displayed variable median percentages among the populations. The four populations were compared using the Kruskal Wallis nonparametric 1 way ANOVA. The differences among these populations were found to be statist ically significant with p= 0.0003. This was followed by Dunns pair wise comparison of each infected group to healthy controls. There was a significant reduction in the frequency of class switching in the CVID population with p < 0.05. Figure 1 3 : Contour plot demonstrating the increasing frequency of class switched (IgM IgD ) B cells in the shift from a nave to memory cell. Nave CD27 B220+ 0 10 2 10 3 10 4 10 5 IgD 0 10 3 10 4 10 5 IgM 10.7 47.9 2.82 38.5 Ig D IgM 0 10 2 10 3 10 4 10 5 IgD 0 10 3 10 4 10 5 IgM 0.86 63.3 30.5 5.43 Memory CD27+B220
62 3.5 Frequency of CD21 + Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations A comparison of the expression of CD21 among the four study populations was conducted on both na ve and memory B cells, as shown in Figure 15. The purpose of this was to examine and compare the activation in these cell populations. The graph of the frequencies of CD21 + nave B cells displayed similar medians among the study groups. However, the rang e of data for the autoimmune group was quite variable. When the four populations were analyzed using the Kruskal Wallis nonparametric 1 way ANOVA, no significance was found with p= 0.076. The graph of the frequencies of CD21 + memory B cells displayed stat istically significant results when analyzed with the Kruskal Wallis test, with a p value of 0.023. Figure 14 : Distribution of class switched (IgM IgD ) nave and memory B cells among the four study populations. A box and whiskers style graph is used with the whiskers containing the minimum and maxim um percentages. Statistics were conducted using a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
63 When this was followed by Dunns pair wise comparison of each infected group to healthy controls, both the autoimmune group and HIV group had significantly re duced frequencies of CD21 + cells with p< 0.05. 3.6 Frequency of CD23 + Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations A comparison of the expression of CD23 among the four study populations was conduc ted on both nave and memory B cells, as shown in Figure 16. The purpose of this was to examine and compare the activation in these cell populations. The graph of the frequencies of CD23 + nave B cells displayed differences in the medians among the study groups. When the four populations were analyzed using the Kruskal Wallis nonparametric 1 way ANOVA, they were found to be statistically Figure 1 5 : Distribution of CD21 + nave and memory B cells among the four stu dy populations. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were conducted using a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
64 significant with p= 0.015. When this was followed by Dunns pair wise comparison of each infected group to healthy contr ols, the HIV infected population had significantly reduced expression of CD23 with p < 0.01. The graph of the frequencies of CD23 + memory B cells did not display statistically significant results when analyzed with the Kruskal Wallis test, with a p valu e of 0.592. Therefore, there was no significant reduction in CD23 expression for any of the mature B cell study populations. 3.7 Frequency of CD80 + and/or CD86 + Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations A comparison of the expression of CD80 and/or CD86 among the four study populations was conducted on both nave and memory B cells, as shown in Figure 17. The purpose of this was to examine and compare the activation in these cell populations. Figure 16 : Distribution of CD23 + nave and memory B cells among the four study populations. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were conducted usi ng a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95%
65 The graph of the frequenc ies of CD80 + CD86 + nave B cells displayed small differences in the medians among the study groups. When the four populations were analyzed using the Kruskal Wallis nonparametric 1 way ANOVA, the differences were not found to be statistically significant wi th p= 0.489. This indicates that there was little to no upregulation of activation in the naive cell population amongst any of the study groups. The graph of the frequencies of CD80 + CD86 + memory B cells did display statistically significant results when analyzed with the Kruskal Wallis test, with a p value of 0.033. Therefore, there was a significant difference in CD80 + CD86 + expression among the four study populations. Specifically, the autoimmune disease and HIV infected study populations show increased expression compared to healthy controls. However, when a Dunns pair wise comparison of each infected group to healthy controls was conducted, no significance was found between the three disease populations and healthy controls. Figure 17 : Distribution of CD80 + CD86 + nave and memory B cells among the four study populations. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were conducted using a Kruskal Wallis followed by Dunns post hoc pairwise comparison of each group to healthy controls with 95% confidence intervals.
66 3.8 Frequency of C D27 + B220 Memory B C ells, IgM IgD Class S witched M emory B Cells and CD80 + CD86 + Memory B Cell Populations in Nave and Treated HIV infected Subjects The last section of this study looked more specifically at differences in expression of cell surface rece ptors in memory B cells of nave (untreated) HIV infected subjects compared with treated HIV infected subjects. The purpose of this was to determine if antiretroviral therapy can correct any of the defects in B cell function that are associated with HIV in fection. Figure 18 displays the frequencies of CD27 + B220 memory B cells among the two populations of HIV infected patients. They were analyzed statistically using the Mann Whitney nonparametric t test. The difference in frequency among the patient groups was not found to be significant, with p= 0.140. Figure 18 : Distribution of CD27 + B220 memory B cells among HIV nave and treated subjects. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were conducted using a Mann Whitney U nonparametric t test with 95% confidence interval.
67 Figure 19 : Distribution of class switched memory B cells among HIV nave and treated subjects. A box and whiskers style graph is used with the whiskers containing the minimum and maximum percentages. Statistics were condu cted using a Mann Whitney U nonparametric t test with 95% confidence interval. Figure 19 displays the frequency of class switched memory B cells among the two populations of HIV infected patients. They were analyzed statistically using the Mann Whitney nonparametric t te st. The difference in frequency among the patient groups was not found to be significant, with p= 0.958. Figure 20 displays the frequency of CD80 + CD86 + memory B cells among the two populations of HIV infected patients. They were analyzed statist ically using the Mann Whitney nonparametric t test. The difference in frequency among the patient groups was found to be significant with p= 0.023. The frequency of this cell population is therefore significantly reduced in HIV subjects on therapy compared with nave HIV subjects.
68 Figure 20 : Distribution of CD80 + CD86 + memory B cells among HIV nave and treated subjects. A box and whiskers style graph is used with the whiskers containing the minimum an d maximum percentages. Statistics were conducted using a Mann Whitney U nonparametric t test with 95% confidence interval.
69 Discussion: 4.1 Frequency of CD19 + Lymphocytes These results indicated that there was little to no expansion in the frequency of circulating B cells in the blood in any of the four study populations. Therefore, no ne of the patient groups had B cell defects that caused a large increase or decrease in the proportion of B cells as a whole, but rather any defects in development in these study populations must exist only within the more defined B cell subsets. Although there were no statistically significant results, it still appears as if the frequency of peripheral B cells was slightly elevated in the HIV infected population, which has been indicated in some past studies. The fact that the frequencies were still very s imilar among the four populations is not unexpected. 4.2 Frequency of CD27 + B Lymphocytes The data obtained in this part of the study indicated that there were no significant differences in the number of memory B cells among the four populations. These data are not in agreement with past studies that indicate a reduction in memory B cells in SLE, CVID, and HIV (Warnartz et al., 2002; De Milito et al., 2001). This disagreement is probably best explained by the large variability of percentages in the healt hy control and CVID populations. Also, it can be explained by the small sample sizes in some of the populations. This result is not crucial for the purpose of this experiment, as the goal is to further differentiate these populations using the B220 marker, which will be more revealing of defects in the developmental stages.
70 4.3 Frequency of CD27 B220 + Nave B Lymphocyte Population and CD27 + B220 Memory B Lymphocyte Population The results obtained in the true nave B cells indicated no significant differ ences between the population groups. However, when we look at the true memory B cell population we see a significant reduction in the frequency of cells in CVID and HIV infected patients. There is also a reduction in the frequency of cells in autoimmune pa tients although it is not significant. These results would indicate that the B cell defects that occur in HIV are similar to those in CVID in the sense that they both occur in the periphery during late B cell differentiation. Because differentiation appea rs to be normal in the nave population, there is no indication that defects occur in the bone marrow in HIV infection. Therefore, the most accurate prediction based on these data is that HIV causes defects in development of memory B cells at the level of the germinal centers. This is in agreement with the hypothesis stated earlier. 4.4 Frequency of Class Switched Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations The data presented on class switching in the study populations showed a s ignificant difference among the four populations for this process in the nave population. However, no groups had significant differences when compared pair wise to the healthy controls. We would not expect much class switching in nave B cells, as this pr ocess occurs later in the germinal centers as discussed in the introduction. However, because there is a slight increase in the frequency of this process in autoimmune disease, this
71 could be a reflection of a defect in the known checkpoint at the germinal centers in this disease. However in HIV, the population of interest, nothing significant can be said about this nave population. In the memory B cell population, there is a statistically significant difference in class switching among the four populatio ns. The CVID population has a significantly lower frequency, which is precisely as expected in this disease because it is characterized by decreased levels of IgG, IgA, and IgE. However, in the HIV population, there does not appear to be a reduction in cla ss switching but rather a preservation of this process, in contradiction to our hypothesis. Therefore, we can conclude that in the memory B cell population in HIV infection, there in no defect in the process of class switching. 4.5 Frequency of CD21 + Nav e (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations In examination of the first activation marker in the four study populations, no differences were observed in the nave population. However, we do see a downregulation of this marker in autoim mune disease and HIV infection in the memory B cell population. This was not predicted by the hypothesis, as we expected activation markers to be increased in HIV infection, representing hyperactivation of B cells previously described in the disease (De Mi lito et al., 2004). However, upon further examination of the literature, we find that a subset of CD21 + memory cells have been defined that are considered resting memory B cells, and HIV infection is associated with the loss of these cells. The loss of CD2 1 expression on peripheral blood B cells has even been described as a reliable marker of ongoing HIV replication and disease progression (Moir & Fauci,
72 2008). Therefore, we would expect patients with more chronic viral infection to have decreased expressio n of this marker. Further studies that categorize the heterogeneous HIV infected population will allow us to look at this marker more closely. 4.6 Frequency of CD23 + Nave (CD27 B220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations In examination of the second activation marker, we see a significant reduction in the frequency of CD23 expression in HIV infection in the nave B cell population. This would indicate that unlike the other diseases, the expression of this receptor is lost more quickly in H IV disease. Usually its expression is retained until the cell reaches the germinal center. This is reflective of a defect in B cell differentiation in HIV infection. In the memory B cell population, there was no significant upregulation or downregulation o f CD23 in any of the study populations. Further studies in HIV, as well as the other disease populations will be helpful in learning more about the expression of this marker throughout B cell development. 4.7 Frequency of CD80 + and/or CD86 + Nave (CD27 B 220 + ) and Memory (CD27 + B220 ) B Lymphocyte Populations The expression of the CD80 and CD86 activation markers indicated no upregulation of these receptors in the nave B cell population. However, there was significant upregulation in the memory B cell po pulations, specifically in autoimmune disease and HIV infection. This indicates that like autoimmune disease, HIV infection causes an increase in activation of memory B cells. The absence of increased expression
73 in nave cells indicates that this increased activation occurs in the periphery later in development. 4.8 Frequency of CD27 + B220 Memory B C ells, IgM IgD Class S witched M emory B Cells and CD80 + CD86 + Memory B Cell Populations in Nave and Treated HIV infected Subjects The purpose of examining the expression of cell surface receptors in subgroups of HIV infected subjects in to determine if antiretroviral therapy can correct some the established defects in B cell development. The data indicated that treated HIV infected subjects did not show a signi ficant difference in their frequency of memory B cells, demonstrating that therapy does not correct for defects in this subset. The data also indicated that treated HIV infected subjects did not show a significant difference in their frequency of class swi tched memory cells compared to nave subjects. However, we determined that HIV infection does not cause a significant defect in class switching to begin with, so we would not expect any difference among these two populations. Lastly, we looked at the incre ased expression of CD80 and CD86 in the HIV infected population to determine if therapy can correct for the increased activation that occurs. The statistically significant results indicated that antiretroviral treatment can in fact lower the increased expr ession of these activation markers, and correct some of the hyperactivation that occurs in this disease.
74 4.9 In Summary This research indicates that trends are evident in the data for some of the markers. The results point to the fact that it is helpfu l to look at the B cell defects in HIV infection in terms of what is known in the literature regarding SLE and CVID, and comparisons can be drawn in doing so. It is important to realize that in a pilot study such as this, the nature of the patients is lik ely to be diverse. However, we see this particularly in the HIV infected population, as this group included both nave and treated patients with a wide range of CD4 counts and viral loads (see Appendix 1). Because the results indicate important differences in particular B cell populations in HIV, this indicates that we should continue to study HIV patients in more detail, looking further at the precise relationships between B cell defects and viral load of CD4 counts. Also, some of the results in this study not found to support the original hypothesis may become significant upon the enrollment of more subjects and the splitting up of patient populations into defined groups. 4.10 Future Direction of Research This is an ongoing study at the Childrens Resear ch Institute at USF/All Childrens hospital, and patients are continually being enrolled. The number of subjects to be enrolled in each study population has recently been expanded as authorized by the Institutional Review Board. As this was a pilot study conducted with an uneven distribution of age, sex and ethnicity amongst the 4 populations, the future goal will be to fill in these gaps in the patient population. One of the difficulties has been the recruitment of adolescent healthy donors, particularly those of African American descent, so attempts
75 have been made to receive authorization to visit local high schools in St. Petersburg and reward students for donating blood with community service hours or a gift card. Another difficulty has been the enrollm ent of males in the SLE population, largely because the disease affects women in much greater frequencies. This is probably unavoidable and this imbalance will have to be accepted. As all of the patient populations are relatively small, further analysis on larger populations will be more revealing. When the population of autoimmune disease patients expands beyond the small number of 8, there will be an opportunity to look at the B cell subsets in terms of the severity of the disease using clinical manifesta tions and presence of autoantibodies as measures of this. This will provide for an interesting offshoot study. As far as expanding on this area of research more broadly, the fairly recent advancements in flow cytometry will allow a great amount of room fo r future phenotyping experiments such as this in which numerous receptors are examined simultaneously. Because B220 appeared to be helpful in further clarifying an early nave B cell and a late memory B cell, future studies should consider including this m arker in their analysis as opposed to using only CD27. Thinking more specifically about HIV infection, because this research demonstrated that antiretroviral therapy did not correct several of the B cell defects in the disease, alternative ways of correcti ng these defects will have to be discussed and examined.
76 Appendix 1 : 5.1 Enrolled Patient Details All of the values from laboratory studies reflect those done closest to the date of the blood draw that were out of the normal range, in most cases. Autoimmune Patient Data Patient ID Diagnosis Year of Birth Age at Blood Collection Gender Ethnicity Clinical Manifestations QAEA SLE Nephritis 1992 16 F B arthritis (mild), nephritis, leukopenia QAFA SLE Nephritis 1990 18 F W arthritis (mild), malar r ash, photosensitivity, nephritis, leukopenia QAGA SLE Nephritis 1992 16 F H arthritis, malar rash, discoid rash, nephritis QAIA SLE Nephritis 1994 14 F A arthritis, malar rash, nephritis QABA SLE 1990 18 F B arthritis, malar rash, psychosis, leukopen ia QACA SLE 1993 15 M W arthritis, malar rash, pleuritis, nephritis, thrombocytopenia, leukopenia QAHA SLE 1990 18 F B arthritis, malar rash, photosensitivity QADA ITP 1987 21 F W arthritis (mild), thrombocytopenia
77 Autoimmune Patient Data conti nued Patient ID ANA Titer and Pattern Specific DNA Antibodies (see bottom for ranges) Antiphosph olipid Antibodies Complement Levels Treatment QAEA 1:160/ Homog dsDNA 176 H, Sm 16 H, Ro 10 H, La <10 ( ) C4 3 L, C3 88 prednisone, cellcept QAFA 1: 640/ Homog dsDNA 132 H, RNP <1, Sm 3.32 H, Ro 4.34 H, La 7.76 H (+) C3 126, C4 15, CH50 14 L prednisone, hydroxychloroquine, cellcept, celebrex QAGA 1:640/ Speckled dsDNA 200 H, RNP 6 H, Sm 6 H, Ro 1.10 H (+) C3 138, C4 21, CH50 <13 L prednison e, cellcept, piroxicam, hydroxychloroquine QAIA 1:1280/ Homog dsDNA 200 H, RNP 6.0 H, Sm 4.79 H, Ro 6 H, La 1.86 H (+) C3 32 L, C4 3 L, CH50 <13 L hydroxychloroquine, prednisone, cellcept, aspirin QABA 736/ Not recorded dsDNA 342 H, RNP 736 H, Sm 161 H, Ro 20 H, La 24 H not tested C3 5 L, C4 35 L cogentin, prednisone, lexapro, azathioprine, lasix, alprazolam, diazepam QACA 1:320/ Speckled dsDNA 348 H, RNP 31 H, Sm 16 H, Ro 191 H, La 268 H ( ) C3 88, C4 18, CH50 75 H hydroxychloroquine QAHA 1:1280/ Homog dsDNA 200 H, RNP >5, Sm <10, SSA 4 H, SSB <3, (+) C3 92, C4 12 L, prednisone, azathioprine, methotrexate QADA 1:640/ Speckled dsDNA <30, RNP <1, Sm <1, Ro <1, La <1 (+) C3 98, C4 26, CH50 37 keflex, triamcinolone
78 NORMAL RANGES FOR LABORATORY TESTS ACH Labcorp Quest dsDNA IU/mL 0 29 dsDNA IU/mL 0 99 dsDNA IU/mL <30, <10 Sm IU/mL <10 Sm index <1 Sm index <1 RNP IU/mL <10 RNP index <1 RNP index <1 Ro IU/mL <10 Ro index <1 Ro index <1 La IU/mL <10 La index <1 La index <1 HIV Infected Patient Data Patient ID Year of Birth Age at Collection Gender Ethnicity CD4 + # / % Viral Load Treatment at ti me of draw ZBTA 2007 0.5 M B 3527/48 <50 Kaletra, Retrovir, Epivir ZARA 1993 15 M B 298/31 <400 Retrovir, Epivir, Efavirenz ZCGA 1996 12 M W 848/29 <400 Didanosine Stavudine, Nevirapine ZBIA 1989 19 F W 437/20 <50 Reyetaz, Norvir, Epzicom ZBPA 2004 4 M W 2293/32 105 Kaletra, Retrovir, Epivir, Ziagen ZBXA 1987 21 M W 589/18 <50 not available ZASA 1993 15 F B 1303/51 <50 Combivir, Kaletra ZACA 2003 5 F W 774/35 <50 Retrovir, Epivir, Kaletra
79 HIV Infected Patient Data continued ZBMA 1990 18 F W 231/20 <50 Atazanavir, Norvir, Epzicom ZBAA 1986 22 F B 230/20 44600 Reyataz, Norvir, Truvada ZCFA 1987 21 M W <20/1 94612 Kaletra, Epzicom, Trizivir ZBQA 1986 22 F B 116/9 18500 Epzicom, Lexiva, Epivir ZCLA 1986 22 M B 317/21 90125 Nave ZCHA 1983 25 F W 690/59 186 Nave ZCCA 1989 19 F H 463/36 73600 Nave ZBZA 1984 24 M B 742/31 1073 Nave ZCRA 1991 17 F W 447/24 85798 Nave
80 Appendix 2: 6.1 B C ell Staining Protocol Details Specimen Requirement s 1. Anticoagulated blood (ACD) 2. Less than 24 hours post draw 3. 2 mL minimum volume (ACD tubes must be completely full to ensure proper dilution of ACD solution) Reagents 1. 1X D PBS without Mg ++ and Ca ++ 2. Lysing Buffer Pharmlyse 10X concentrate (BD Biosciences, cat # 55589 9), diluted to 1X with water, store according to manufacturers directions. 3. Staining Buffer (0.1 % FBS + 0.05% NaN 3 in 1X D PBS, stored at 4C, expires in 6 months) 4. Fluorochrome tagged antibodies see table 4. 5. Fixation Buffer 4% paraformaldehyde in 1X D PBS (freeze in 10 ml aliquots at 20C, thaw prior to use, store unused solution at 4C for up to one week) Equipment/Supplies 1. Flow tubes (10 x 75 mm or 12 x 75 mm as appropriate) 2. 15 ml polypropylene conical tubes with screw top caps 3. Se ropipets (5 10 ml size or larger if needed) 4. Micropipette tips (20 200 1000 uL size) 5. Centrifuge
85 Appendix 3: 7.1 Certifications received by author to conduct this research: April 2008: CITI New College IRB Training for Research Using Vulnerable Populations June 2008: USF Biosafety Principles and Practices USF HIPAA Privacy Overview Course USF HIPAA Privacy in Research Course IRB approval was received from both New College of F lorida and All Childrens Hospital for conducting this research
86 References: Abbas, A., et al. 2007. Cellular and Molecular Immunology, 6 th ed. Saunders, Elsevier, Inc. Akashi, K., et al. 1999. Lymphoid development from stem cells and t he common lymphocyte progenitors. Cold Spring Harbor Symp. Quant. Biol. 64: 1 12 Arbuckle, M.R., McClain, M.T., et al. 2003. Development of autoantibodies before the clinical onset of system lupus erythematosus. N Eng J Med. 349: 1526 1533 Bleesing, J.J ., & Fleisher, T.A. 2003. Human B cells express a CD45 isoform that is similar to murine B220 and is downregulated with acquisition of the memory B cell marker CD27. Cytometry B Clin Cytom. 51: 1 8 Boyle, J.M., & Buckley, R.H. 2007. Population Prevalence of Diagnosed Primary Immunodeficiency Diseases in the United States. J Clin Immunol. 27: 497 502 Cepeda, E.J., & Reveille, J.D. 2004. Autoantibodies in systemic sclerosis and fibrosing syndromes: clinical indications and relevance. Curr Opin Rheumatol. 16 : 723 32 Dawson, L.J., Fox, P.C., & Smith, P.M. 2006. Sjogrens syndrome the non apoptotic model of glandular hypofunction. Rheumatology (Oxford). 45: 792 8
87 De Milito, A., et al. 2001. Loss of memory (CD27) B lymphocytes in HIV 1 infection. AIDS. 15: 957 64 De Milito, A. et al. 2004. Mechanisms of hypergammaglobulinemia and impaired antigen specific humoral immunity in HIV 1 infection. Blood. 103: 2180 86 DOrlando, et al. 2007. Outside inside signaling in CD40 mediated B cell activation. J Biol Regul Ho meost Agents. 21: 49 62 Garcia, J.M., Espanol, T.,et al. 2007. Update on the treatment of primary immunodeficiencies. Allergol Immunopathol (Madr). 35: 184 92 Grawunder, U., et al. 1995. Down regulation of RAG1 and RAG2 gene expression in preB cells afte r functional immunoglobulin heavy chain rearrangement. Immunity. 3: 601 8 Haymore, B.R., Mikita, C.P., & Tsokos, G.C. 2008. Common variable immune deficiency (CVID) presenting as an autoimmune disease: role of memory B cells. Autoimmune Rev. 7: 309 12 He W., & Parker, R. 2000. Functions of Lsm proteins in mRNA degradation and splicing. Current Opinion in Cell Biology. 12: 346 350
88 Jacobi, A. M ., & Diamond B. 2005. Balancing diversity and tolerance: lessons from patients with systemic lupus erythematosus J. Exp. Med 202 :341 344 James, H.A., Neas, B.R., Moser, K.L., et al. 2001. Systemic lupus erythematosus in adults is associated with previous Epstein Barr virus exposure. Arthritis Rheum. 44: 1122 6 Kikushige Y., et al. 2008. Human Flt3 is expressed a t the hematopoietic stem cell and the granulocyte/macrophage progenitor stages to maintain cell survival. J Immunol. 180: 7358 67 Lafyatis R., York, M., & Marshak Rothstein, A. 2006. Antimalarial agents: closing the gate on toll like receptors? Arthritis Rheum. 54: 3068 70 Lipsky, P.E. 2001. Systemic lupus erythematosus: an autoimmune disease of B cell hyperactivity. Nat Immunol. 2: 764 6 Maizels N. 2005. Immunoglobulin gene diversification. Annu Rev Genet. 39: 23 46 Martensson I.L., Keenan, R.A., & Lic ense, S. 2007. The pre B cell receptor. Curr Opin Immunol. 19: 137 42
89 Meek, K. 1990. Analysis of junctional diversity during B lymphocyte development. Science. 250: 820 3 Moire, S., & Fauci, A.S. 2008. Pathogenic mechanisms of B lymphocyte dysfunction in HIV disease. J Allergy Clin. Immunol. 122: 12 19 Moir, S., & Fauci, A.S. 2009. B cells in HIV infection and disease. Nature Reviews. 9: 235 45 Morrow, M., et al. 2008. A splenic marginal zone like peripheral blood CD27 + B220 B cell population is prefere ntially depleted in HIV type 1 infected individuals. AIDS Res Hum Retroviruses. 24: 621 33 Murphy, K., et al. 2008. Janeways Immunobiology, 7 th ed. Garland Science, Taylor & Francis Group, LLC Nagafuchi, H., Shimoyama, Y., et al. 2003. Preferential Expr ession of B7.2 (CD86), but not B7.1 (CD80), on B cells induced by CD40/CD40L interaction is essential for anti DNA autoantibody production in patients with systemic lupus erythematosus. Clin Exp Rheumatol. 21: 71 7
90 Piqueras B., Lavenu Bombled, C., et al. 2003. Common variable immunodeficiency patient classification based on impaired B cell memory differentiation correlates with clinical aspects. J Clin Immunol. 23: 385 99 Radcliff, G., Jaroszeski, M.J. 1998. Basics of flow cytometry. Methods Mol Biol. 9 1 : 1 24 Sato, S., Fujimoto, M., Hasegawa, M., Takehara, K. 2004. Altered blood B lymphocyte homeostasis in systemic sclerosis: expanded nave B cells and diminished but activated memory B cells. Arthritis Rheum. 50: 1918 27 Snow, C. 2004. Flow Cytometry Electronics. Cytometry A. 57 : 63 69 Swingler, S., et al. 2008. Evidence for a pathogenic determinant in HIV 1 Nef involved in B cell dysfunction in HIV/AIDS. Cell Host Microbe. 4: 63 76 Taguchi, T., et al. 2007. Interleukin 7 contributes to human pro B c ell development in mouse stromal cell dependent culture system. Exp Hematol. 35: 1398 407 Vodjgani, M., et al. 2007. Analysis of Class Switched Memory B Cells in Patients with Common Variable Immunodeficiency and Its Clinical Implications. J Investig Alle rgol Clin Immunol. 17: 321 328
91 Warnatz, K., Denz, A. et al. 2002. Severe deficiency of switched memory B cells (CD27 + IgM IgD ) in subgroups of patients with common variable immunodeficiency: a new approach to classify a heterogeneous disease. Blood. 99: 1544 51 Warnatz, K., & Schlesier. 2008. Flow Cytometric Phenotyping of Common Variable Immunodeficiency. Cytometry B Clin Cytom. 74: 261 71 Woof, J.M., & Kerr, M. A. 2006. The function of immunoglobulin A in immunity. J Pathol. 208: 270 82 Wrammert, J., & Ahmed, R. 2008. Maintenance of serological memory. Biol Chem. 389: 537 9. Yurasov, S.,Wardemann, H., et al. 2005. Defective B cell tolerance checkpoints in systemic lupus erythematosus. J Exp Med. 201: 703 11