1 CHARACTERIZATION OF HYBRIDOMA ANTIBODIES TO IDENTIFY AUTOANTIGENS LIKE DISEASE By MAURO ALEJANDRO TUDARES HERNNDEZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF F LORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009
2 2009 Mauro Alejandro Tudares Hernndez
3 To myself
4 ACKNOWLEDGEMENTS I want to thank Dr. Ammon B. Peck for giving me the opportunity to carry out this project. I also want to thank members of my supervisory committee, Dr. Seunghee Cha and Dr. Shannon Wallet for giving me valuable guidance an d advice regarding my research project Marievic Bu lo san Janet Corneli us, and Dr. Coung Nguyen provided helpful technical recommendations. In addition, I thank my mom and family for providing support throughout the years.
5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................................................................................................. 4 LIST OF TABLES ................................................................................................................................ 6 LIST OF FIGURES .............................................................................................................................. 7 ABSTRACT .......................................................................................................................................... 8 CHAPTER 1 INTRODUCTION ....................................................................................................................... 10 Definition of Sjgrens Syndrome ............................................................................................. 10 Primary Sjgrens Syndrome .................................................................................................... 10 Secondary Sjgrens Syndrome. ................................................................................................ 10 Pathogenesis ................................................................................................................................ 11 Phases in the Pathogenesis of Sjgrens Syndrome like Disease. ........................................... 17 Role of Humoral Immunity in the Pathogenesis of Sjgrens Syndrome ............................... 18 Ro le of autoantibodies in the pathogenesis of Sjgrens syndrome ........................................ 18 Anti -Ro/SSA and Anti LA/SSB Autoantibodies ............................................................... 18 Anti M3R Recept or Autoantibodies .................................................................................. 20 2 MATERIALS AND METHODS ............................................................................................... 24 Animals ........................................................................................................................................ 24 Indir ect Immunohistochemistry ................................................................................................. 24 Western Blotting ......................................................................................................................... 24 Immunoprecipitation ................................................................................................................... 25 Protein Band Identification Using Mass Spectrometry ............................................................ 26 3 RESULTS .................................................................................................................................... 27 Immunohistochemistry ............................................................................................................... 27 Immunoprecipitation ................................................................................................................... 29 Protein Band Identification Using Band Sequencing Analysis ................................................ 31 4 DISCUS SION .............................................................................................................................. 42 REFERENCES ................................................................................................................................... 46 BIOGRAPHICAL SKETCH ............................................................................................................. 50
6 LIST OF TABLES Table page 3 1 Staining pattern of antibody -secreting hybridomas by Indirect Immunofluorescence ...... 33 3 2 Proteins of putative importance found in the gels using mass spectrometry ..................... 41
7 LIST OF FIGURES Figure page 3 1 Absolute distribution of supernatants according to staining patterns ................................. 32 3 2 Relative distribution of supernatants according to staining patterns .................................. 32 3 3 Indirect Immunofluorescence pattern of staining of Hep 2 cells ........................................ 34 3 4 Western Blot ........................................................................................................................... 36 3 5 Immunoprecipitation: gel 1 ................................................................................................... 37 3 6 Immu noprecipitation: gel 1 with bands picked for analysis ................................................ 38 3 7 Immunoprecipitation: gel 2 ................................................................................................... 39 3 8 Immunoprecipitation: gel 2 with b ands picked for analysis ................................................ 40
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHARACTERIZATION OF HYBRIDOMA ANTIBODIES TO IDENTIFY AUTOANTIGENS LIKE DISEASE By Mauro Alejandro Tudares Hernndez August 2009 Chair: Ammon B. Peck Major: Medical Sciences Sj grens syndrome is a chronic autoimmune disea se characterized by loss of secretion of tears and saliva from lachrymal and salivary glands respectively, lymphocytic infiltration of lachrymal and salivary glands, autoimmune attack against exocrine tissue and abnormal expre s sion of cytokines. Several in teracting factors may contribute to the onset of overt disease. One of the key events during the pathogenesis of the disease is the appearance of autoantibodies. They have a crucial role both as effectors and biomarkers of the disease. B cells from dr aining lymph nodes of the of C57BL/6.NOD -Ae c1Aec2 mice diagnosed with late -stage Sj grens syndrome like disease were fused with immortal murine myeloma cells in order to obtain antibody -secreting hybridomas that replicate the autoantibody response in the lymph nodes of this mouse model. Supernatants from a set of hybridomas were screened to determine their reactivities against cellular antigens using Hep 2 cells and immunohistochemistry. Western blotting was used to test the reactivity of representative supernatants to cellular antigens according to molecular weight, leading to potentially more precise results identification of autoantigens by immunoprecipitation and gel sequencing analyses using GC -mass spectrometry.
9 Ninety -one proteins were ide ntified representing several expected cellular autoantigens, i n cluding ribonuclear proteins, histones and structural proteins; however, definitive identification of any autoantigen among these ninety one proteins will require additional studies. Neverthele ss, these studies identified possible important proteins, constituting a novel opportunity to help elucidate the complex autoantibody responses that occur in the lymph nodes of the C57BL/6.NOD Aec1Aec2 mouse model of Sjgrens syndrome.
10 CHAPTER 1 INTRODUCTION Definition of Sjgrens Syndrome Sjgrens syndrome (SS) is a systemic disease characterized by a chronic autoimmune attack to various organs in the body, especially the exocrine tissues of the lacrimal and salivary gl ands, causing abnormalities in the quantity and quality of tears and saliva, respectively, leading to n u merous health problems. Additional target organs that might be affected include the thyroid gland, kidneys, liver, lungs, skin, peripheral nerves, joint s, and muscles causing numerous cond i tions that diminish the quality of life. Some patients may also develop life -threatening diseases such as malignant lymphoma. Key features of the autoimmune attack consist of lymphocytic i n filtration of target tissues, increased secretion of cytokines by T cells, and increased production of autoantibodies by B cells. Primary Sjgrens S yndrome. Primary Sjgrens syndrome (pSS) is characterized by a decrease in the function of exocrine glands without clinical or labo ratory evidence of any other autoimmune disease. Dysfunction of the lacrimal and salivary glands is early and progressive, as seen in mouse model of the disease (Nguyen et al 2007). Extraglandular tissues may be affected as well in pSS. Second ary Sjgren s S yndrome. Secondary Sjgrens syndrome (sSS) occurs in people who have been diagnosed previously with other autoimmune connective tissue diseases. The two entities most commonly associated with secondary Sjgrens syndrome are rheumatoid arthritis (RA) and systemic lupus erythema tosus (SLE) (Fox, R., 2005). In contrast, the non-obese diabetic (NOD) mouse develops type 1 diabetes (T1D) and SS. T1D is an autoimmune disease characterized by a T cell mediated d e -cells. Neve rtheless, SS in the NOD mouse is classified as sSS. In sSS
11 the degree of damage to the lacrimal and salivary glands can be less severe than in pSS as the disease in the two glands can occur separately (Nguyen et al 2007) Pathogenesis Development and onset of Sjgrens syndrome is the result of a combination of genetic, env i ronmental, and immunological factors not yet completely deciphered (Ng uyen, Singson et al 2006). Environmental factors have minimum impact on animal models of the disease, thus, e m phasis on genetic and immunological factors in mice will be discussed here. The similar clinical features between human and mice SS provides a practical mechanism to extrapolate findings from mice studies int o the human context of the disease. Up to now, genetic recombination studies in the NOD mice model of the disease have provided evidence as to how genetic predisposition may influence the development and onset of Sjgrens syndrome. Initially, these studie s aimed at analyzing the potential role of genes of the MHC, finding predisposition associated with markers from this genomic region (Fei et al 1991) Further studies focused on the potential role of insulin dependent diabetes susceptibility genes in glandular exocrine dysfunction and development of autoimmunity, based on the assumption that if these genes have an influence on the development and onset of diabetes in the NOD mice, then it would be possible that the same was true for SS (Brayer et al 2000). There are more than 21 chromosomal regions thus far known to be positive contributors to the development of diabetes in the NOD mice (Brayer et al 2000). Each chromosomal region is a genetic interval of precise distance and is called diabetes s usceptibility loci or idd To assess the potential role of each one of these twenty -one single idd over the development and onset of SS, they were indivi dually replaced from the NOD mice and inserted in the disease free C57BL/6 mice using genetic backcross techniques. Opposite to what it was expected, results from these studies showed no development of autoimmune exocrinopathy in the C57BL/6 mice when each idd loci
12 was indiv i dually replaced, leading to the conclusion that single genetic idd segments, with p roved positive influence over the development and onset of diabetes, do not have a role in the pathogenesis of Sjgren-like disease when recombined individually in the C57BL/6 mice (Brayer et al 2000). Moving past the effect of single idd alleles ov er the development of Sjgrenlike disease in the C57BL/6 mice, studies went on to evaluate the potential role of a combination of idd se g ments over the development of SjS like disease. Again, previous diabetes studies lead now to this assumption (Hill, N. J., et al, 2000). Not surprisingly, it was found that idd5 alleles from chromosome 1 and idd3 alleles from chromosome 3 of NOD mice together contribute additively to the development of autoimmune exocrinopathy when replaced in the C57BL/6 mice (Brayer et al 2000).The development of SjS like autoimmune response in these studies was evidenced by the presence of histological sialoadenitis lesions, appearance of serological markers of the disease such as antinuclear and cytoplasmic antibodies, and abnormal r esults from physiological tests such as measurement of saliva flow rates, protein concentration, amylase activity, and cysteine protease activity (Brayer et al 2000). Later on, idd5 genetic region on chromosome 1 was designated Aec2 and idd3 genetic reg ion on chromosome 3 was designated Aec1 (Aec = a u toimmune exocrinopathy) leading to the generation of the C57BL/6.NOD -Aec1Aec2 murine model for primary Sjgrens syndrome which is homozygous for both the idd3 and idd5 genetic intervals and does not exhibit many of the dysfunctional immune features of the NOD mice (Cha et al ., 2002). Both Aec1 and Aec2 loci have a different gene pool, length, and role in the pathogenesis of SjS -like disease (Nguyen Singson et al 2006). The length and limits of A ec1 a nd Aec2 are di fferent and were identified using microsatellite markers (Nguyen Singson et al 2006). Aec1 is
13 a 48.5 cM (centimorgan) region (Nguyen, Singson et al., 2006) that contain four diabetes suscept i bility alleles: Idd3, Idd10, Idd17, and Idd18, a s well as the gene for IL 2. A 0.15cM region from chromosome 3 has been implicated in the development of experimental autoimmune encephal o myelitis and diabetes (Encinas et al. 1999 and Sundvall et al 1995). Aec1 is associated with the immune and clinica l phase of the disease in the mice model. Aec2 is associated with the initial, preclinical or non immune phase of the disease in which autoimmunity events are not yet detectable, (Nguyen et al. 2007). Aec2 is a 68.4 cM genetic region comprising the gene locus for bcl 2, and genetic loci for caspase 8 and cytotoxic T lymphocyte associated antigen 4 (CTL 4) (Brayer et al 2000 and Nguyen et al 2006). CTLA 4 is a 33 kDa molecule that negatively regulates T -cell activation. It is also called CD152 an d it has been shown to play an important role in T -cell homeostasis and autoimmune diseases, since mice deficient in CTLA 4 develop lymphoproliferative disorders and die early in life due to tis sue destruction of multiple organs (Lundholm et al 2006) L ikewise, NOD activated T -cells show anomalous expression of CTLA 4. In this scenario, cytokines such as IL 2, can contribute to up regulated transcription of CTLA 4, especially upon T -cell activation through the CD3 complex in vitro; rendering CTLA 4 readi ly available on the cell surface (Lundholm et al ., 2006) It has been mentioned before about the synergistic role of Aec1 and Ae c2 genetic regions on the development of Sjgrens syndrome. Few studies so far have tried to understand the m o lecular mechanisms of such a genetic symbiosis in the C57BL/6.NOD -Aec1 Ae c2 mice. However, some other studies have been aimed at understanding more precisely how different type 1 d i abetes susceptibility genes located in distinct chromosomal regions may interplay a t early phases during type 1 diabetes initiation in the NOD mice. For instance, it has been shown
14 that Ctex and Idd3 downregulate Ctla -4 in the NOD mice (Lundholm et al ., 2006). Ctex is a 21 cM, telomeric novel locus identified by interval mapping methodology that positively contributes, along with Idd3, to the defective expression of Ctla -4 (Lundholm et al ., 2006). In this context, the molecular role of inadequately expressed Ctla -4 in susceptibility and initiation of diabetes is not well u n derstood to da te, but blockage of the protein using antibodies targeted against the full length structure of it increases the appearance of overt diabetes in the NOD mice probably because of activation of cytotoxic T -cells (L hder et al 1998 and Luhder et al ., 2000). Likewise, the li n kage between Ctla -4 down regulation and development of lymphoproliferative disorders and a u toimmunity represents a promising research topic that would help to better understand the prim a ry molecular processes that lead to the development o f these conditions. Interestingly, Idd5 contains the Bcl-2 locus as mentioned above. This fact deserves special consideration. Bcl-2 is an oncogene that promotes cell surviva l and proliferation (Petros et al 2004). It is part of the Bcl 2 family of genes that includes both cell survival ( Bcl-2, Bcl XL, Mcl 1, Ced -9, A1, Bfl -1 ) and cell death ( Bax, Bak Diva, Bcl-Xs, Bik, Bim, Bad, Bid, Egl-1) genes important for maintenance of cell homeostasis (Kaufmann et al 2004). Cancers, for i n stance, B cell lymphomas, arise when cell conditions favor gene expression of cell survival genes over cell death genes; stimulating proliferation and inhibiting apoptosis. Apoptosis is i m portant to maintain a lymphocyte population within the normal values. During infec tion states, active production of lymphocytes occurs and the number of these cells increases rapidly as a r e sponse to the presence of foreign pathogens. Accordingly, Bcl-2 expression increases to sustain and extend lymphocyte presence. Bcl -2 inhibits progr ammed cell death by binding to the mit o chondrial outer membrane and inhibiting the release of cytochrome -c into the cytoplasm. Ant i gen -receptor signaling plays an important role in the maintenance and survival of
15 lymphocytes as well. No evidence so far lin ks Bcl 2 over expression as an explanatory factor of B-cell lymph o cyte proliferation in SS. However, it would be important to address whether or not the presence of bcl -2 locus in the Aec -2 chromosomal region is an important co regulatory factor stimulatin g lymphocyte proliferation is SS. The Aec1 locus has been redefined more precisely to a region of just 20cM by generation of a recombinant inbred line of C57BL/6.NOD -Aec1 R1 Aec2 that shows a crossover region within chromosome 3 (Nguyen et al ., 2006). Under these genetic circumstances both male and females showed similar autoimmune phenotype of the parental C57BL/6.NOD Aec1Aec2 but females failed to develop lymphocytic infiltrates in the lachrymal glands and full blown SS disease, pe rhaps indicating that key genes associated with SS phenotypic gender differences are located within the removed genetic segment of approximately 30cM. Close location of the IL -2 gene to the area of the cross -over region in chromosome 3, might help to further understand the pathoge nesis of SS. Among other functions IL 2 is known for inducing secretion of antibodies by B lymphocytes and generation of Treg cells. In SS, increase production of autoantibodies occurs, especially in later phases of the disease. A direct role of IL 2 in triggering autoantibody produ c tion in SS remains speculative but needs to be studied. IL -7 gen is also located in the crossover region. IL 7 is a cytokine involved in B -cell proliferation and maturation as well. Expression of IL 7 could be implicated in the pathogenesis of SS via the inhibition of apoptosis of B cells by an indirect mechanism. Carbonic anhydrase and retroviruses genes are also found in the redefined A ec1R1 locus. However, their possible role in the pathogenesis of SS is logical but ha s not been studied yet. Recent studies have tried to understand more precisely the Aec2 region of chromosome 1 as well, and its influence on the development of pathogenesis of SS. To do so, recombinant
16 inbred lines of the C57BL/6.NOD Aec1Aec2 mo use have been generated in order to screen for a p pearance of crossover segments within this loci (Nguyen et al ., 2008). One new line, C57BL/6.NOD Aec1Aec2R ( n) is homozygous for a crossover region in the Aec2 loci and is the result of inbreeding the C57BL /6.NOD Aec1Aec2 mouse with the C57BL/6J mouse. Such an approach provided new clues about the presence of genes located within this region, which are valuable to study their potential involvement in the pathogenesis of the disease. It was found that the QTL -A th1 domain within this new delineated crossover Aec2 region contains genes of pa rticular importance such as Ox40L and Tnfsf6. FasL is the product of Tnfsf6 and is up regulated in the early phases of SS in mice, perhaps in association with increased gland ular cell apoptosis typical of this phase. Ox40L (Tnfsf4) encodes OX40L, an important protein found to be involved in preventing the maturation of Treg1 cells, and in the regulation of fatty acids metabolism. In summary on the importance of genetic r egulation in the development and onset of Sjgrens syndrome; in the absence of strong evidence regarding the association of class II MHC with SS, studies shifted direction to focus on the potential role of idd regions as key factors co n tributing to the pa thogenesis of SS. Evidence showed that NOD derived Aec1 and Aec2 chrom o somal regions both play an important role during initiation and progression of the disease, as confirmed by studies done in the C57BL/6.NOD -Aec1Aec2 murine model of pSS. Moreover, the f unctional relationship between these two chromosomal intervals is additive and synergistic, suggesting that the regulation of the expression of genes within Idd3 or Idd5 loci is not confined within each region separately but, instead, genes within Idd5 loc i have genetic regulation ele ments located in the idd3 chromosomal interval. For instance, IL 2 ( IL -2 is located close to Idd3) signaling is associated with mobilization of CTLA 4 ( CTLA -4 is located in Idd5)
17 from intracell u lar vesicles to the cellular memb rane in CD3 activated T cells. Contradictorily, low levels of CTLA 4 are required for proper T -cell activation, suggesting that CTLA 4 expression might be under different transcriptional control in nave and activated T -cells. In the other hand, regulation of Idd3 genes of importance for the development of Sjgrens syndrome has not been yet th o roughly approached in the primary murine model of the disease. However, notable advance has been made identifying potential candidate genes for development and onset of SS in the mouse model by creating crossover regions of both Idd3 and Idd5 In this context, the genetic intera c tion of Bcl-2 with elements of the idd3 or other locus, as a contributory factor in pathogenesis of SS remains to be elucidated. Phases in th e Pathogenesis of Sjgrens Syndrome -like D isease. Three phases have been described in the C57BL/6.NOD -Aec1Aec2 mouse model of the di s ease (Cuong et al 2006). The first phase is nonimmune and goes until about eight weeks of age. Important events in clude abnormal development of glands with altered expression of acinar antigens, and activation of apoptosis markers such as caspase 3 and caspase -11. The presence of elevated levels of INFy may reflect early involvement of viruses in this phase of the pat hogene sis of the disease. Second phase starts around eight to nine weeks. Main ongoing events are a consequence of alterations that occurred in the previous phase. Acinar cell apoptosis causes act i vation of an intense autoimmune response leading to leukocy te infiltration of target glands. E x posure of cellular antigens due to intracellular redistribution and apoptosis triggers the synthesis of autoantibodies that attack mainly nuclear antigens. Other autoantibodies are synthesized i n cluding the anti -M3R auto antibody. During the third phase starting at about 16 weeks, loss of se cretory function occurs. However, decreased saliva and tear flow can be observed as soon as 12 weeks of age in the C57BL/6.NOD Aec1Aec2 mouse.
18 Role of Humoral Immunity in the Patho genesis of Sjgrens Syndrome Hypergammaglobulinemia and presence of serum autoantibodies are key clinical findings in patients with SS. Their appearance is dependent upon the onset of autoimmunity, and also they mediate much of the inflammatory proce ss that leads to eventual dysfunction of salivary and lachrymal glands. Lately, much importance has been given to study the presence of autoantib o dies reactive with the muscarinic receptors on acinar tissues of the salivary and lachrymal glands. In suppor t of this concept, it was showed that congenic NOD B cell knockout mice (NOD.Ignull) develops exocrine gland lesions with small T cell lymphocytic infiltrates but do not show lac h rymal or salivary dysfunction or high glucose levels. The absence of insulit is, diabetes, and ex o crinopathy in the NOD.Ignull mice led to think that B lymphocytes play a key role in the onset of the pancreatic and glandular autoimmune attack. In addition, the composition of the lymphocytic infiltrates in salivary glands is mostly at e x pense of B lymphocytes in autoimmune and clinical phases of the disease. These B cells are highly reactive and are responsible for continuous secretion of autoantibodies and for the deve l opment of hypergammaglobulinemia (Cuong et al 2006). All taken together suggests that in Sj grens syndrome like disease, B cells have a protagonist participation in the pathogenesis of the disease by infiltration of target organs, secretion of autoantibodies, and in fewer cases, transformation of chr onically stimulated B lymphocytes into lymphomas (Cuong C. et al, 2006). Of course, B cells do not act alone. As in all autoimmune di s ease, participation and activation of T cells is required (Bennett JC 2008) Role of autoantibodies in the pathogenesis of Sjgrens syndrome Anti -Ro/SSA and Anti -LA/SSB Autoantibodies Presence of serum anti -Ro(SSA) and anti -La(SSB) autoantibodies is part of the international classification criteria for Sjgrens syndrome in human patients (Vitali et al ., 2002). Evidence of
19 serum levels of anti Ro (SSA) and anti -La (SSB) in SS patients are clinically associated with early manifestation of signs and symptoms, longer disease progression, and occurrence of parotid gland enlargement. Likewise, presence of these autoantibodies correlates histologically with more dramatic lymphocytic infiltration of minor salivary glands (Mavragani et al ., 2000) Although Ro (SSA) and La (SSB) are well known ribonuclear RNA -binding proteins and their physiological function has been studied exte nsively the role of anti -Ro (SSA) and anti La (SSB) autoantibodies in the pathogenesis of SS through recognition of their target antigens is not well understood yet (Tengnr et al 1998) So far, two Ro/SSA subunits have been identified: Ro 60 kD and R o 52 kD. They can be both localized in the nucleus and cytoplasm depending on conditions of cell stress and homeostasis. Such a dynamic localization makes it harder to d e scribe fluorescent patterns that can be universally accepted. While it is known that R o 60 kD binds to cytoplasmic RNA, the intracellular interactions of Ro 52 are not clear yet and some a rgue that is not a Ro protein (Boire et al 1995). La (SSB) a 48 kD localized both in the nucl eus and cytoplasm (de Wilde et al 1996). It has bind ing capacity especially to the poly -U tail of RNA polymerase III transcripts (Gottlieb et al .,1989) and can depict a cytoplasmic patter of staining specially when studied in conjunctival epithelium (Yannopoulos et al 1992). It is not known yet why both Ro (SSA) and La (SSB) are among the cellular targets of the autoimmune attack in SS, but cells secreting autoantibodies against Ro (SSA) and La (SSB) antigens have been locally identified to be in the affected sal i vary glands. In addition, anti -Ro (SSA) au toantibodies have been found on skin cells of patients with SLE (Golan et al 1992). Thus, in the context of autoimmune exocrinopathy, anti Ro (SSA) and anti -La (SSB) autoantibodies are thought to have a dual role both as markers of
20 the disease and mediat ors of the pathogenic events that lead to overt disease (Nguyen et al 2007). Anti M3R Receptor Autoantibodies The M3R receptor is a cell surface protein localized in exocrine and non -exocrine tissues across the body. It is responsible for the tra nsmission of neurological signaling messages from the extracellular to the intracellular space, acting as a mediator during the secretion mechanism of salivary and lachrymal glands. A functional M3R receptor is required for normal saliva and tear sec retion. Autoantibodies to the M3R receptor block post -ganglionic parasympathetic neurons innervating salivary and lac h rymal glands by impeding transmission of acetylcholine to acinar and ductal cells, thus media t ing glandular dysfunction and causing appea rance of clinical disease (Yamamoto et al ., 1996; Watermam et al ., 2000). In humans, anti -M3R antibodies have been found not only in pSS cases, but also in sSS ass o ciated with rheumatoid arthritis context (Watermam et al ., 2000). This study found pre sence of anti -M3R antibodies in 5 out of 9 patients (55%) diagnosed with pSS, and 6 out of 6 patients (100%) diagnosed with sSS. Indirect evidence of anti -M3R antibodies existence was based on the results of experiments showing the inhibitory effect of pat ients sera (n=11) on bladder smooth muscle contraction after in vitro stimulation of the M3R receptor with carbachol. Subs e quently, IgG fraction was purified from sera samples. Purified IgG fraction produced inhibited bladder smooth muscle carbachol -evoke d responses as well, consistent with results for sera alone. However, assuming no diagnostic errors at the time of enrollment and no experimental flaws; it cannot be concluded from this study that autoimmunity against M3R receptor is unive r sally present ac ross the study sample suffering from the disease, since sera from 44% of p a tients with pSS did not elicit significant inhibition of smooth muscle contraction.
21 Another finding that supports the participation of anti -M3R autoantibodies in the pathoge n e sis of SS is the persistence of normal secretory levels of tears and saliva in NOD.Ignull mice, a congenic strain lacking functional B lymphocytes (Robinson et al ., 1998). The structural role of the anti -M3R autoantibodies is perhaps to bind directly t o the M3R receptor. In this context, sy n thesis of anti -M3R autoantibodies and their interaction with the M3R receptor in pathogenesis of SS was demonstrated in vitro by initial incubation of mouse parotid gland membrane prepar a tions with the muscarinic rec eptor agonist [3H] quinuclidinyl benzilate ([3H] QNB), and subs e quent failure of the SS human IgG to immunoprecipitate the M3R receptor [H3] QNB complex; suggesting competition between SS human IgG and [H3] QNB for the same binding site at the M3R receptor Additionally, this experiment supports previous studies evidencing cross reactivity between mice M3R and human IgG, and rules out any functional or structural dysfun c tion of the M3R receptor in SS, although hypothetically, its density may be diminished a fter chronic stimulation by anti -M3R receptor autoantibodies. Desensitization of the M3R receptor is associated with chronic presence in sera of circulating auto antibodies against the membrane protein, as assessed by studies evaluating carbachol evok ed responses on smooth muscle strips isolated from NOD mice during the third phase of the disease (Cha et al ., 2006) In this study, anti -M3R autoantibodies initially improved cholinergic like M3R responses during a short period of time. However, receptor s responses to agonist m o lecules diminish ed after prolonged stimulation by anti M3R autoantibodies. Desensitization of the M3R receptor is a consequence of continuous binding by anti -M3R antibodies, leading to an increased threshold level to cholinergic s timulation and decreased exocrine function of salivary and lachrymal glands due to reduced intracellular calcium concentration. It is still unknown whether or not binding of anti -M3R autoantibodies to the receptor elicits either
22 conformational changes or a lters the expression of this membrane protein of the parasympathetic nervous system with consequent decrease of its density. Antibodies that inhibit M3R receptor have been reported not only in patients with SS but also in patients suffering from scle roderma (Cavill et al 2003); which is a chronic disease charact e rized by thickening of the skin along with vasculopathy and progressive fibrosis of multiple o r gans including lungs, kidneys, and gastrointestinal tract (Varga 2008). Although participatio n of the Fc fraction cannot be discarded as a mediator of autonomic contractile dysfunction in vivo; studies aimed at exploring the inhibitory mechanisms of autoantibodies found that both the Fab (monovalent) and F(ab)2 (divalent) fraction of patients Ig G functional autoantibodies are r e sponsible for producing inhibition of smooth muscle contraction on functional assays (Cavill et al 2003). On the contrary, n o n -SS patients IgG did not elicit the same inhibitory effect. Ho w ever, purified IgG from health y individuals (n=4) was able to neutralize functional anti -M3R a u toantibodies, but the mechanism involved, how this mechanism may differ from the one used by intravenous immunoglobulin (IVIG) antibodies, and in what circumstances it is activated are enigma s, have not been answered yet. IVIG is a purified pool of immunoglobulins that neutr a lizes autoantibodies inhibition of smooth muscle contraction in a concentration dependent ma n ner (Cavill et al 2003); therefore it has been used to treat dysautonomia in SS (Dupond et al 1999) In summary, then, autoantibodies play a major role in the pathogenesis of SS, a concept strongly supported by multiple studies carried out in mice models of SS. As a result, identifying the nature and reactivity of autoantibodies is an important endeavor in understanding SS. In the following studies, I have initiated an approach that is designed to identify tissue -specific autoa n tigens by defining the reactivity of autoantibodies in a global analysis using hybridoma
23 super n a tants generated from fusions of B lymphocytes isolated from the draining lymph nodes of di s eased C57BL/6.NOD Aec1Aec2 mice. The overall question is whether or not such an approach is feasible. To address this question, I established 2 specific aims: a) to carry out a general ch a racterization of supernatants reactivity using immunofluoresce nce and western blot analysis, b) to initiate a more precise identification of antibodies by carrying out i mmunoprecipitation and band sequencing analysis.
24 CHAPTER 2 MATERIALS AND METHODS Animals Male C57BL/6.NOD -Aec1Aec2 mice were maintained under specific pathogen -free cond i tions in the Department of Animal Care Services at the University of Florida, Gainesville, Flor i da. The animals were provided food and wate r ad libitum Twenty mice were euthanized at 20 weeks of age and their tissues harvested. Indirect Immunohistochemistry Hep 2 cells were grown in DMEM (with 10% FBS + antibiotic) at 37 oC until they reached 8090 % confluency. Subsequently, they were trypsinized and counted. Four thousand cells per well were seeded in 8 -well chamber slides and let grow overnight or until confluency reached 7080%. At this time, cells were washed twice with ice-cold PBS and fixed with 100% methanol for 30 minutes. Afte r two additional washes, 5 minute each with ice -cold PBS, 200 hundred L of undiluted primary antibody (hybridoma supernatant) was layered over the cells and incubated for 1 hour at 4 oC. Simultaneously, supernatants were diluted at 1/50 and 1/100 in PBS, layered over the cells and incubated for the same amount of time at 4 oC, too. After two more washes with ice -cold PBS, cells were covered with Alexa Fluor 594 goat anti -mouse IgG (H+L) diluted 1/50 and incubated for 1 hour at 4 oC. After two additional w ashes, cellular fluorescence was v i sualized by fluorescence microscopy at 100x magnification. Western B lot ting Hep 2 cells were grown in DMEM (with 10% FBS + antibiotic) at 37 oC until they reached 8090 % confluency. Cells were washed twice with PBS 2 ml of lysis buffer added (10mM Tris HCL, 0.4 M NaCl, 1 mM EDTA, 1% Triton X 100 (vol/vol); pH 7.5) supplemented with 20 L Protease inhibitor Cocktail Set I (Calbiochem, cat. # 539131), and scraped into a 1.5
25 microcentr i fuge tube. Then, cells were vor texed for 3 minutes, incubated for 20 minutes on ice, vortexed again for 3 minutes, and centrifuged at 14.000 rpm at 4 oC for 20 minutes. Supernatant was co l lected and protein concentration determined. Approximately 40 micrograms (~40 L) of whole ce ll protein lysate was denatured by boiling it in 4X SDS buffer and size -separated using a 4 15% Tris -HCL gel electrophoresis gel (Bio Rad, Hercules, Ca. Cat. # 161 1104) prior to transfer to PVDF membrane. Membrane was blocked overnight at 4 in blocking bu ffer (5% nonfat dry milk in TBS) with shaking. Next day, supernatants were diluted 1:10 in PBS and incubated separately with a piece of membrane at 4 oC with shaking. After overnight incubation, each piece of membrane was incubated with Goat Anti -M ouse Ig M+IgG+IgA (H+L) (SouthernBiotech, Birmingham, Alabama; USA) diluted 1:800 at 4 oC with shaking for 1 hour. After three 15 -minute washings with TBS, membrane was exposed to substrate for 1, 30 and 60 seconds and developed. Immunoprecipitation Hep 2 ce lls were grown in DMEM (with 10% FBS + antibiotic) at 37 oC until they were a c tively growing and reached 8090 % confluency. Cells were trypsinized, counted, pelleted, and washed twice with PBS. Nuclear and cytoplasmic lysate was separated using the PARIST M Kit (Applied Biosystems, Texas; USA). After removing the PBS wash, pellet was resuspended in 500L of ice -cold Cell Fractionation Buffer included in the kit. Following incubation on ice for 10 minutes, the suspension was centrifuged 5 minutes at 4 oC and 1000 rpm. The Supernatant (cytoplasmic fraction) was collected and 500 L of Cell Disruption Buffer added to the pellet followed by vigorous vortexing to lyse the nucleus. Protein concentration was determined in both cytoplasmic and nuclear fractions. Prior to the immunoprecipitation, 30 L of proteinG agarose (Upstate, Lake Placid; NY. Cat # 16266) was mixed overnight with 0.5 ml ice -cold PBS and 100 L of supernatant or mouse
26 serum at 4 oC with end over end shaking. Next day, following centrifuga tion for 30 seconds at 13000 rpm, supernatant was collected and saved. Antibody-conjugated beads in the bottom of the microcentrifuge tube were washed twice with ice -cold PBS and mixed with 10 L of 10% BSA plus 100 L of either cytoplasmic or nuclear Hep2 cells fractions for overnight incubation at 4 oC with end over end shaking. Next day, suspension was centrifuged 30 seconds at 13000 rpm, supernatant saved and pellet washed three times. Final pellet (~30 L) was mixed with an aliquot of SDS (gel 1) or 50 L of SDS and 50 L of PBS (gel 2), boiled for 5minutes and an a lyzed by one dimensional gel electrophoresis using a 4 15% Tris HCL gel (Bio Rad, Hercules, Ca. Cat. # 1611104). Gel was stained overnight with Coomassie blue staining solution and de -staine d with a de -staining solution (5% methanol, 10% glacial acetic acid). Protein Band I dentification Using Mass Spectrometry To identify proteins isolated by immunoprecipitat ion, a set of gel bands selected using mol e cular weight estimates were sent to the ICBR Protein Core Facility.
27 CHAPTER 3 RESULTS Immunohistochemistry Forty supernatants obtained from the ICBR Hybridoma Core were analyzed. None of these supernatants were derived from cloned hybridomas, but all had been tested for reactivity o n the Hep 2 cell line. As expected, all samples were positive for immunofluorescent staining of Hep2 cells. Staining patterns of the 40 supernatants on Hep2 cells were defined by six general pa t terns, and grouped: speckled (nuclear), nuclear membrane (nuclear rim), nuclear membrane line, nucleolar, cytoplasm (cytoskeletal), and cytoplasmic dot. Speckled pattern was observed in fi fteen samples (37%), nuclear membrane (nuclear rim) pattern was observed in seven samples (17%), nuclear membrane line pattern was observed in five samples (13%), nucleolar pattern was observed in one sample (3%), cytoplasm (cytoskeletal) pattern was observed in eleven samples (27%), and cytoplasmic dot pattern was observed in one sample (3%). These six staining pa t terns are pres ented in Figures 3 1 and 32. For further analysis, a prototype supernatant was ch o sen as a representative for each pattern. Supernatant 2E6 was chosen as the prototype sample for speckled pattern, 1G7 for nuclear membrane, 5A9 for nuclear membrane line, 5 A10 for nucle o lar, 4F12 for cytoplasm, and 5C1as cytoplasmic dot pattern (Figure 3 -3 ). Nuclear Speckled pattern as seen in clone 2E6 (Figure 3 3 C): Fine and diffuse speckling is seen in the nucleus leaving almost no space for black, dark dots. Mi totic cells are not clearly vi s ible in the frame possibly because of the strong staining of the cells, which is compatible with a La (SSB) pattern. The nucleoplasm is almost entirely stained and there are a few fluorescent dots visible in most of the nucle i of the cells and in the cytoplasm as well. This pattern of staining suggests that ribonuclear proteins such as Ro (SS/A), La (SS/B), or both might be the target anti gens.
28 Nuclear membrane line as seen in clone 5A9 (Figure 3 3 D): There are one o r two bands of intense staining surrounding the nucleus but it does not encircle it completely. The staining pa t tern of the nucleus is very diffuse and fluorescent dots are not visible. Target antigens might be part of the Golgi apparatus. Cytoplasmi c dot as seen in clone 5C1 (Figure 3 3 E): Numerous small but very intense dots are seen exclusively in the cytoplasm of the cells. However, some cells show a complete nuclear membrane ring and a single fluorescent dot within this nuclear peripheral str ucture. Interestingly, the cluster of cytoplasmic dots are seen very closely associated with the nuclear structure, thus it might be hypothesized that these numerous cytoplasmic dots are nuclear projections coming from the nucleus. Truly depiction of cytop lasmic dots could be explained by the reaction of the assay against lysosomes. In the other hand, as mentioned before, nuclear pore antigens could also produce this pattern. Nucleolar staining as seen exclusively in clone 5A10 (Figure 3 3 F): Brig ht fluorescent dots are seen associated with the nucleoplasm of the cell. The nucleus is covered with coarse like speckles which is consistent with the idea that nucleolar patterns are usually associated with nu c lear patterns. Target antigens might include Scl 70 protein, and other nucleolar proteins such as PM 1 and PL 6. Cytoplasmic (cytoskeletal) staining as seen in clone 4F12 (Figure 3 3 G ): Cytoplasm is uniformly stained and well delimited with respect of the surrounding area. Bright fluorescenc e intensity is seen in the cytoplasmic. Target cellular antigens representing this pattern might be a c tin, tubulin, and/or vimentin. Nuclear membrane (nuclear rim) staining as seen with clone 1G7 (Figure 3 3 H): A co m plete halo of fluorescence is seen around the nucleus, surrounding it completely. The rest of
29 the nucleus appears weekly and irregularly stained or not stained at all. In addition, Mitotic cells do not show staining of the chromosome region, ruling out a homogenous staining pattern. L amin and dsDNA are among the antigens that might depict this pattern of staining. Western Blot ting One representative supernatant for each staining pattern was picked for further analysis by Western blotting (Figure 3 4 ). Each supernatant tested sho wed different degrees of reactivity against whole cell fraction of Hep 2 cells transferred to a PVDF membrane. Samples 2E6, 5A9, and 5C1 reacted more favorable, while samples 5A10, 4F12, and 1G7 reacted very faintly or a l most showed no reactivity at all. S amples 2E6 and 5A9 showed various bands within the range of 25 to 75 kDa that looked like promising representatives of critical autoantibodies. Sample 5C1 showed two bands between 50 and 75 kDa that suggested further evaluation as well. Samples 4F12 and 1G 7 showed very weak reactivity, and sample 5A10 showed no reactivity at all, pe rhaps indicating that the blotting technique needed to be re -designed to meet special needs related to incubation conditions and dilution of the supernatants. Supernatants 4F12, 1G7, and 5A10 did not yield interesting blotting reactivity compatible with the results illustrated by the imm u nofluorescence staining. Immunoprecipitation Electrophoresis analysis of samples and controls showed different banding pattern on both im munoprecipitation assays (Figure 3 5 and 3 7 ). Differences were observed in number, size, and intensity of bands. However, similarities were also observable after in-depth comparison. Sample 2E6 showed interesting bands at approximately 100, 60, 45, and 20 kDa for antibodies exposed to cytoplasmic fractions of Hep2 cells and 100, 60, 54, 52, 45, 30 and 20 kDa for ant i bodies exposed to nuclear fractions (Figure 3 5 lanes 1 & 2). Sample 1G7 showed interesting bands at 100, 60, 54, 52, 45, and 20 kDa for a ntibodies exposed to cytoplasmic and
30 nuclear fra c tion of Hep2, however the intensity of the bands depicted in the nuclear fraction appeared to be less intense than their similar counterparts in the cytoplasmic fraction of the same sample, pe rhaps indicati ng differences in protein concentration among the two fractions (Figure 3 7 lanes 5 & 6). Analysis of the electrophoresis pattern of serum from C57BL/6.NOD Aec1Aec 2 mouse showed generally less reactivity than monoclonal supernatants (Figure 3 6 lane 3 an d Figure 3 7 lanes 3 & 4). Specifically, antibodies from sera of this mouse reacted with antigens that appeared to be at approximately 75, 54, 52, 45, and 30 kDa when exposed to nuclear fractions of Hep2 cells (Figure 3 6 lane 3). In the second immunopre cipitation assay, the band at 75 kDa appeared exclusively in mouse serum samples exposed to both cytoplasmic and nuclear extract of Hep 2 cells (Figure 3 7 lanes 3 & 4), but not in supernatant samples (Figure 3 7 lanes 1, 2, 5 & 6). This is consistent w ith results from the first immunoprecipitation assay. Bands at 45 and 30 kDa appeared both in serum and supernatant samples; however the intensity of the bands was weaker in serum samples, again, perhaps due to less protein concentration in serum compared to supe r natants Likewise, in gel 2 (Figure 3 7 ) bands at 54 and 52 kDa were present in all samples e x posed to both cytoplasmic and nuclear fractions, but the intensity is slightly higher in supernatant samples of both 2E6 and 1G7 than in serum samples. H owever, in gel 1 (Figure 3 5 ) bands at 52 and 54 kDa were visible only in hybridoma 2E6 and serum from mice exposed to nuclear cellular fractions (Figure 3 5 lanes 2 & 3). Bands at 20 kDa appeared strong and well defined in supe r natant samples of both 2E6 and 1G7, but not in whole serum samples. Electrophoresis pattern of samples representing washes showed efficient clearance of non -specific proteins at every step of the process during both immunoprecipitation assays (Figure 3 5 lanes 4 9, and Figure 3 7 lanes 5 8). Bands numbered 1, 2, 3, 4, 5, 6 and 7 from gel 1(Figure 3 6 ), and 8, 9 and 10 from gel two (Figure 3 8 ) were manually extracted for
31 sequencing analysis at the ICBR Protein Core Facility. Band molecular weight predictions were as follows: 100 kDa (band 1), 60 kDa (band 2), 54 kDa (band 3), 52 kDa (band 4), 45 kDa (band 5), 30 kDa (band 6), 75 kDa (band 7), 54 kDa (band 8), 52 kDa (band 9) and 20 kDa (band 10). Protein Band Identification Using Band Sequencing Analysis Protein sequencing of the 10 bands (where bands 8 and 9 were combined) by MS identified 91 unique proteins ranging from 6 proteins per band (band #3) to 29 proteins per band (band #5) (data not shown). Obviously, some proteins, especially keratins, were common to several (but not all) band sequences. Although several ways of analyzing the data were considered, e.g., m o lecular weights, presence in multiple bands, known autoantigens, and so forth, none proved to provide a reasonable and logical means to identify SSA/Ro, SSB/La, and anti M3R autoantigens. Nevertheless, several other proteins were found that might have been expected based on the known nuclear autoantigens of human SS and mice SS -like disease (table 3 -2 ). These included several histone and ribonuclear proteins, pl us ribosomal proteins. Similarly, the structural pr o tein units of keratin were also identified in both nuclear and cytoplasmic extracts by hybridoma supernatants, as well as whole sera from diseased C57BL/6.NOD -Aec1Aec2 mice. These data, while not identi fying precise autoantigens of known importance, establish a basis that should yield such identifications. Thus, the results provide strong support to this approach, even though any future approach will require modification.
32 Figure 3 1. Absolute distribution of supernatants according to staining patterns Figure 3 2 Relative distribution of supernatants according to staining patterns 37% 17% 13% 3% 27% 3% Speckled (nuclear) Nuclear membrane Nuclear membrane line Nucleolar Cytoplasm Cytoplasmic dot 15 7 5 1 11 1 Speckled (nuclear) Nuclear membrane Nuclear membrane line Nucleolar Cytoplasm Cytoplasmic dot
33 Table 3 1. Staining pattern of antibody -secr eting hybridomas by Indirect Immunofluorescence Hybridoma Staining pattern Titer Exposure Time 3B2 Speckled nuclear 1/50 2 seconds 4A4 Cytoplasmic/cytoskeletal 1:50 2 seconds 2E6 Speckled nuclear 1:100 2 seconds 3B2 Speckled nuclear 1:50 2 second s 5D9 Nuclear membrane line 1:100 3 seconds 1G7 Nuclear membrane/rim Undiluted 2 seconds 5F11 Cytoplasmic/cytoskeletal 1:100 2 seconds 2E2 Nuclear membrane/rim 1:50 2 seconds 5D4 Cytoplasmic/cytoskeletal 1:100 3 seconds 5A9 Nuclear membrane line 1: 50 2 seconds 4A4 Cytoplasmic/cytoskeletal Undiluted 2 seconds 1C4 Speckled nuclear 1:50 2 seconds 3C10 Cytoplasmic/cytoskeletal 1:50 3 seconds 4H7 Nuclear membrane line 1:100 2 seconds 5B4 Speckled nuclear 1:50 2 seconds 4E1 Speckled nuclear 1:50 2 seconds 2H10 Nuclear membrane/rim 1:50 2 seconds 5C1 Cytoplasmic dot 1:50 3 seconds 5H11 Speckled nuclear 1:50 2 seconds 3E2 Speckled nuclear 1:50 3 seconds 5E1 Speckled nuclear 1:50 3 seconds 4B1 Cytoplasmic/cytoskeletal 1:50 3 seconds IF9 Cytopla smic/cytoskeletal Undiluted 2 seconds 5G10 Nuclear membrane line 1:50 2 seconds 5F8 Cytoplasmic/cytoskeletal 1:50 2 seconds 4A5 Speckled nuclear 1:50 2 seconds 5D6 Nuclear membrane/rim 1:50 2 seconds 4F12 Cytoplasmic/cytoskeletal 1:100 2 seconds 5A 10 Nucleolar 1:50 2 seconds 1H9 Speckled nuclear 1:50 2 seconds 4C8 Cytoplasmic/cytoskeletal 1:50 2 seconds 5A5 Nuclear membrane/rim 1:100 2 seconds 5D5 Speckled nuclear 1:50 2 seconds 5D9 Speckled nuclear 1:50 2 seconds 4B10 Speckled nuclear 1:50 2 seconds 3H6 Nuclear membrane line 1:50 2 seconds 1H7 Nuclear membrane/rim 1:50 2 seconds 2A4 Cytoplasmic/cytoskeletal 1:50 2 seconds 2E11 Nuclear membrane/rim 1:50 2 seconds 2G1 Speckled nuclear 1:100 2 seconds
34 A B C D Figure 3 3. Indirect Immunofluorescence patter n of staining of H ep 2 cells. A) positive control; B) negative control; C ) speckled pattern (clone 2E6); D ) nuclear membrane line pa t tern (clone 5A9); E ) cytoplasmic dot pattern (clone 5C1); F ) nucleolar pattern (clone 5A10); G ) cytoplasmic staining pattern (clone 4F12); H ) nuclear membrane (rim) pa t tern (clone 1G7). Positive control consisted of human serum with antibody specific to nuclear antigens (Sigma Diagnostics, St. Louis, MO, Cat # 2022). All pictures taken at 100x magnification
35 E F G H Figure 3 3. Continued
36 Figure 3 4. Western Blot
37 F igure 3 5. Immunoprecipitation: g el 1
38 Figure 3 6. Immunoprecipitation: g el 1 with bands picked for analysis
39 Figure 3 7. Immunoprecipitation: g el 2
40 F igure 3 8. Immunoprecipitation: g el 2 with bands picked for analysis
41 Table 3 2 Proteins of putative importance found in the g els using mass spectrometry Gel 1 ______ ___________________________________ 2 _______________ Band 1 2 3 4 5 6 7 8 9 10 Pro tein ` kD Name 56 Atp5b X NR Pmp2 X 65 Lmna X 69 Ddx5 X 78 Hnrpm X 61 Hp1bp3 X 54 Vimentin X X X 63 Prosapip1 X X X X 207 Nes X X 116 Srebf1 X 84 Zfp184 X 42 Actb X 69 Rngtt X X 42 Rbmx X 35 Ebna1bp2 X 43 Ilf2 X 51 Eef1a1 X 46 Rpl3 X 84 Ctcf X 21 Hist1h1c X 27 Sfrs7 X 27 Rps3 X 23 Hist1h1b X 32 Rps2 X 34 Cdc2a X 20 H1fx X 30 Rps3a X 25 Sfrs2 X 30 Rps4x X 72 Cabc1 X X 65 Mettl3 X X 158 Ssh2 X
42 CHAPTER 4 DISCUSSION The overall goal of this project was to initiate a characterization of monoclonal antibodies (mAbs) secreted by a set of hybridomas generate d from fusions involving B lymphocytes present in the draining cervical lymph nodes of C57BL/6.NOD -Aec1Aec2 mice diagnosed with early stage Sj grens syndrome like disease. The specific issue in question was whether or not was possible to identify specifi c autoantigens being recognized during the development and/or onset of autoimmunity involving the salivary glands. The approach entailed three steps: 1) define the general staining pattern of each monoclonal antibodies -containing clone, 2) determine by wes t ern blot and immunoprecipitation if mAbs contained in the clones bind to unique proteins, and 3) carry out a preliminary analysis of proteins isolated by IP using GC -mass spec sequencing. Indirect immunohistochemistry and western blot analysis are t he methods of choice for ge n eral initial characterization of supernatants content. Immunoprecipitation and band sequencing analysis have the potential to provide more specific identification of supernatants content. Indi rect immunohistochemistry was used to detect patterns of staining of antinuclear autoantibodies, thus, providing initial clues regarding the localization of target antigens. Indirect immunohist o chemistry is a very common method to detect proteins in cells and/or tissues. A cell disruption and fixation method was required to allow for localization of cytoplasmic and nuclear antigens by immobilization of the proteins. Using this technique, it was shown that the spectrum of a u toantibody reaction to cellular antigens is wider than anticipated. Most likely, the pathological and clinical relevance of all putative autoantibodies secreted by hybridomas is not the same. However, since most auto antigens are potentially involved in important cellular functions, any single one that could be found and c haracterized in the supernatants represents a
43 new opportunity to attain a better understanding of the complicated pathogenesis of Sj grens syndrome -like di s ease in the mouse model. Ro/SSA and La/SSB are considered clinical markers of the disease in human. Staining patterns of Hep 2 cells obtained in this study do not exactly recapitulate the classical speckled pa t terns describe before for these autoantibodies. There might be some explanations for this; being the most notorious the fact that anti -SS/ A reacts with an epitope located in the 60 kD protein which is exclusive of humans (Cook L., 1998). Additional autoantigens identified in autoimmune diseases are ribonucleoproteins targeted by autoantibodies such as anti -sm or anti RNP, topo i somerase I pro teins targeted by antiScl 70 autoantibodies, and enzymes associated with tRNA metabolism targeted by anti Jo 1 autoantibodies. Staining patterns obtained might be associated with these auto antigens but further analysis was required. By using mAbs -containing supernatants on Western blots of whole crude lysate of Hep2 cells it was possible to characterize more specific reactivity patterns according to the molecular weight of the bands depicted on the blot. Reactivity differed from relatively high (s upernatants 2E6, 5A9, and 5C1) to low or nonexistent (supernatants 5A10, 4F12, and 1G7). Such a variable reaction among the samples might be explained by different degrees of binding specificity of the auto antibodies, or by assay related conditions that failed to meet minimum requirements for a n tigen antibody epitope recognition. However, the results were promising for those samples that showed some degree of reactivity and they led to new clues toward antigen identification. Both cytoplasmic and nuclear extracts from Hep 2 cells were proven to be an adequate source of antigen for immunoprecipitation protocols when using monoclonal supernatant from mouse as an antibody. An advantage of the immunoprecipitation is that isolates likely autoantibodies away
44 f rom nonrelevant proteins. Non relevant proteins were perhaps the cause of high background and low band definition in some of the samples tested (supernatants 5A10, 4F12, and 1G7). The high intensity of the bands of samples lanes is an indication of high concentration and/or abundance of the antigen in the cellular sub -fractions. The intensity was different among bands but was stronger that the banding pattern from mouse serum. This finding is indicative of higher concentration of antigen in tissues rathe r than in serum, and presence of autoantibody-secreting B-cell close by or within the target organs. Supernatants gave a fairly clean immunoprecipitation assay compared with the Western blot results. This finding is interesting; it might indicate not only that the antigen is abundant in the cells but also that the mAbs contained in the supernatant have high affinity for their antigens d e spite of the fact that usually their titer is lower compared to those of ascitis fluids. A clean assay with well d efined bands was crucial for band prediction and selection for further and more sp e cific analysis using GC -mass spec sequencing. GC -mass spec sequencing identified 91 diverse proteins, but whether or not these represent autoantigens remains questiona ble and will require further separation and characterization. Sequencing whole bands based on Coomassie Blue staining of one dimensional gels is apt to pr o vide multiple proteins. Two interesting observations are: 1) many of the proteins identified in thi s manner are not of the MWR expected, suggesting considerable breakdown, digestion and/or alternative splicing, and 2) several of the identified proteins fall into categories considered to be possible candidate autoantigens, for example, histone, ribonucle ar and ribosome associated proteins. Additional proteins of interest include vimentin, nestin and keratin, three molecules su g gested in the literature to be autoantigens under specific conditions. One unexpected result, but one that is clearly derived fr om the gel patterns, is the multiple banding that each supernatant
45 exhibits on immunohistochemical staining and Western blotting. This suggests that the hybr i domas contain more than one monoclonal antibody (mAb) and that each hybridoma requires cloning pr ior to further analysis.
46 REFERENCES Bennett J (2008). The role of T lymphocytes in rheumatoid arthritis and other autoimmune di s eases. Arthritis Rheum 58(2 Suppl):S53 7. Boire G, Gendron M, Monast N, Bastin B, Mnard H (1995). P urification of antigenically intact Ro ribonucleoproteins; biochemical and immunological evidence that the 52kD protein is not a Ro protein. Clin Exp Immunol 100(3):48998. Brayer J, Cha S, Nagashima H, Yasunari U, Lindberg A, Diggs S, Martinez J, Go a J, Hump h reys Beher M, Peck A (2000). IL 4 -dependent effector phase in autoimmune exocrinopathy as defined by the NOD.IL 4 -gene knockout mouse model of Sjgren's syndrome. Scand J Immunol 54(12):13340. Brayer J, Lowry J, Cha S, Robinson C, Yamachika S Peck A, Humphreys -Beher M (2000) A l leles from chromosomes 1 and 3 of NOD mice combine to influence Sjgren's syndrome -like autoimmune exo crinopathy. J Rheumatol 27(8):1896904. Carrillo J, Puertas M, Alba A, Ampudia R, Pastor X, Planas R, Riutort N, Alo nso N, Pujol Borrell R, Santamaria P, Vives Pi M, Verdaguer J (2005) Islet infiltrating B -cells in nonobese diabetic mice predominantly target nervous system elements. Diabetes 54(1):6977. Cavill D, Waterman S, Gordon T (2003). Antiidiotypic antibodies neutralize autoantibodies that inhibit cholinergic neurotransmission. Arthritis Rheum 48(12):3597602. Cha S, Nagashima H, Brown V, Peck A, Humphreys-Beher M (2002). Two NOD Idd associated intervals contribute synergistically to the development of autoimm une exocrinopathy (Sj gren's syndrome) on a healthy murine background. Arthritis Rheum 46(5):13908. Cha S, Singson E, Cornelius J, Yagna J, Knot H, Peck A (2006). Muscarinic acetylcholine type 3 receptor desensitization due to chronic exposure to Sj gren 's syndrome associated a u toantibodies. J Rheumatol 33(2):296306. Cook L. New Methods for Detection of Anti -nuclear Antibodies (1998). Clinical Immunology and Immunopathology: Academic Press p. 21120. de Wilde P, Kater L, Bodeutsch C, van den Hoogen F, van de Putte L, van Venrooij W (1996). Aberrant expression pattern of the SS B/La antigen in the labial salivary glands of patients with Sjgren's syndrome. Arthritis Rheum 39(5):78391. Dupond J, Gil H, Bouhaddi M, Magy N, Berthier S, Regnard J (1999) Acute dysautonomia secondary to autoimmune diseases: efficacy of intravenous immunoglobulin and correlation with a stimulation of plasma norepinephrine levels. Clin Exp Rheumatol 17(6):7336.
47 Encinas J, Wicker L, Peterson L, Mukasa A, Teuscher C, Sobel R, Weiner H, Seidman C, Seidman J, Kuchroo V (1999). QTL influencing autoimmune diabetes and encephalomyelitis map to a 0.15 -cM region containing Il2. Nat Genet 21(2):15860. Fei H, Kang H, Scharf S, Erlich H, Peebles C, Fox R (1991). Specific HLA -DQA and HLADRB1 alleles confer susceptibility to Sjgren's syndrome and autoantibody production. J Clin Lab Anal 5(6):38291. Fox R (2005) Sj gren's syndrome. Lancet 366(9482):32131. Gao J, Cha S, Jonsson R, Opalko J, Peck A (2004). Detection of anti type 3 muscarinic acety l choline receptor autoantibodies in the sera of Sj gren's syndrome patients by use of a transfected cell line assay. Arthritis Rheum 50(8):261521. Gao J, Killedar S, Cornelius J, Nguyen C, Cha S, Peck A (2006). Sj gren's syndrome in the NOD mouse model is an interleukin 4 time dependent, antibody isotype -specific a u toimmune disease. J Autoimmun 26(2):90103. Golan T, Elkon K, Gharavi A, Krueger J (1992). Enhanced membrane binding of autoantibodies to cultured keratinocytes of systemic lupus erythematosus patients after ultraviolet B/ultraviolet A irradiation. J Clin Invest 90(3):106776. Gottlieb E, Steitz J (1989). Function of the mammalian La protein: evidence for its action in transcription termination by RNA polymerase III. EMBO J 8 (3):85161. Hill N, Lyons P, Armitage N, Todd J, Wicker L, Peterson L (2000). NOD Idd5 locus controls insulitis and diabetes and overlaps the orthologous CTLA4/IDDM12 and NRAMP1 loci in hu mans. Diabetes 49(10):17447. Kaufmann T, Schinzel A, Borner C ( 2004). Bcl -w(edding) with mitochondria. Trends Cell Biol 14(1):812. Luhder F, Chambers C, Allison J, Benoist C, Mathis D (2000). Pinpointing when T cell costimu latory receptor CTLA 4 must be engaged to dampen diabetogenic T cells. Proc Natl Acad Sci U S A 97(22):122049. Lhder F, Hglund P, Allison J, Benoist C, Mathis D (1998). Cytotoxic T lymphocyte associated antigen 4 (CTLA 4) regulates the unfolding of autoimmune diabetes. J Exp Med 187(3):42732. Lundholm M, Motta V, L fgren Burstr m A, Duarte N, Bergman M, Mayans S, Holmberg D (2006). Defective induction of CTLA 4 in the NOD mouse is controlled by the NOD a l lele of Idd3/IL 2 and a novel locus (Ctex) telomeric on chromosome 1. Diabetes 55(2):53844.
48 Mavragani C, Tzioufas A, Moutsopoulos H (2000). Sjgren's syndrome: autoantibodies to cell u lar antigens. Clinical and molecular aspects. Int Arch Allergy Immunol 123(1):4657. Nguyen C, Cha S, Peck A (2007). Sj gren's syndrome (SjS) like disease of mice: the importance of B lymphocytes and autoantib odies. Front Biosci 12:176789. Nguyen C, Cornelius J, Cooper L, Neff J, Tao J, Lee B, Peck A (2008). Identification of possible candidate genes regulating Sj gren's syndrome associated autoimmunity: a potential role for TNFSF4 in autoimmune exocrinopathy Arthritis Res Ther 10(6):R137. Nguyen C, Kim H, Cornelius J, Peck A (2007). Development of Sjogren's syndrome in nonobese diabetic -derived autoimmune -prone C57BL/6.NOD -Aec1Aec2 mice is dependent on complement component 3. J Immunol 179(4):231829. Ngu yen C, Cornelius J, Singson E, Killedar S, Cha S, Peck A (2006). Role of complement and B lymphocytes in Sj gren's syndrome like autoimmune exocrinopathy of NOD.B10H2b mice. Mol Immunol 43(9):13329. Nguyen C, Singson E, Kim J, Cornelius J, Attia R, Doy le M, Bulosan M, Cha S, Peck A (2006). Sj gren's syndrome like disease of C57BL/6.NOD -Aec1 Aec2 mice: gender differences in keratoconjunctivitis sicca defined by a cross -over in the chromosome 3 Aec1 locus. Scand J Immunol 64(3):295307. Petros A, Olejniczak E, Fesik S (2004). Structural biology of the Bcl 2 family of proteins. Bio chim Biophys Acta 1644(23):8394. Robinson C, Brayer J, Yamachika S, Esch T, Peck A, Stewart C, Peen E, Jonsson R, Hump h reys Beher M (1998). Transfer of human serum IgG to non obese diabetic Igmu null mice reveals a role for autoantibodies in the loss of secretory function of exocrine tissues in Sj gren's syndrome. Proc Natl Acad Sci U S A 95(13):753843. Robinson C, Yamachika S, Bounous D, Brayer J, Jonsson R, Holmdahl R, Pec k A, Humphreys Beher M (1998). A novel NOD -derived murine model of primary Sj gren's syndrome. Arthritis Rheum 41(1):1506. Sundvall M, Jirholt J, Yang H, Jansson L, Engstrm A, Pettersson U, Holmdahl R (1995). Ident i fication of murine loci associated wi th susceptibility to chronic experimental autoimmune enc e phalomyelitis. Nat Genet1 10(3):3137. Tengn r P, Halse A, Haga H, Jonsson R, Wahren Herlenius M (1998). Detection of anti Ro/SSA and anti -La/SSB autoantibody producing cells in salivary glands from patients with Sj gren's syndrome. Arthritis Rheum 41(12):223848. Varga J (2008). Systemic sclerosis: an update Bull NYU Hosp Jt Dis 66(3):198202.
49 Vitali C, Bombardieri S, Jonsson R, Moutsopoulos H, Alexander E, Carsons S, Daniels T, Fox P, Fox R, Kassan S, Pillemer S, Talal N, Weisman M (2002). Classification criteria for Sj gren's syndrome: a revised version of the European criteria proposed by the Amer i can -European Consensus Group. Ann Rheum Dis 61(6):5548. Waterman S, Gordon T, Rischmueller M (2000 ). Inhibitory effects of muscarinic receptor a u toantibodies on parasympathetic neurotransmission in Sj gren's syndrome. Arthritis Rheum 43(7):164754. Yamamoto H, Sims N, Macauley S, Nguyen K, Nakagawa Y, Humphreys Beher M (1996). Alt e rations in the sec retory response of non -obese diabetic (NOD) mice to muscarinic rece p tor stimulation. Clin Immunol Immunopathol 78(3):24555. Yannopoulos D, Roncin S, Lamour A, Pennec Y, Moutsopoulos H, Youinou P (1992). Conjunc tival epithelial cells from patients with Sjgren's syndrome inappropriately express major hist o compatibility complex molecules, La(SSB) antigen, and heat -shock proteins. J Clin Immunol 12(4):25965.
50 BIOGRAPHICAL SKETCH The author was born in Maracaibo, Venezuela and raised in Maracaibo and Sant Cougat del Vall s, Spain. Back in Venezuela, he attended Colegio Claret, where he completed his middle and high school education. He continued his academic endeavors by completing a M.D. degree at Universidad del Zulia in his home city. For four years he worked as a phys ician in rural areas of the east and west coast of Venezuela. Then he moved to the United States to pursue his Master of Public Health and Master of Science degrees at the University of Florida. He aspires to follow studies leading to a Ph.D. degree.