Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigation of an adjuvant-enhanced model of murine arthritis and its therapeutic application Ratkay, Leslie Gabriel 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-894350.pdf [ 10.04MB ]
JSON: 831-1.0088039.json
JSON-LD: 831-1.0088039-ld.json
RDF/XML (Pretty): 831-1.0088039-rdf.xml
RDF/JSON: 831-1.0088039-rdf.json
Turtle: 831-1.0088039-turtle.txt
N-Triples: 831-1.0088039-rdf-ntriples.txt
Original Record: 831-1.0088039-source.json
Full Text

Full Text

INVESTIGATION OF AN ADJUVANT-ENHANCED MODEL OF MURINEARTHRITIS ANE ITS THERAPEUTIC APPLICATIONbyLESLIE GABRIEL RATKAYD.M.D., Semmeiweis University Medical School, Budapest, 1986A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDiES(Department of Oral Biology)We accept this thesis as conformingto the required standardTHE UNiVERSITY OF BRITISH COLUMBIAApril 1994© Leslie Gabriel Ratkay, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of O’eAL 2>IoL.orThe University of British ColumbiaVancouver, CanadaDate_______DE-6 (2/88)AbstmctA major difficulty in understanding the etiology and pathogenicmechanisms of rheumatoid arthritis has been the lack of a suitable animalmodel. Based on the IVfRL-lpr spontaneous arthritis model, a new murinearthritis model was developed, that is reliable and practical for therapeuticevaluations.The MRL-lpr mouse strain develops an early autoimmune disease. Afterthe characterization of our colony, it was concluded that while the strain did notexhibit differences in the lupus-like syndromes, the spontaneous arthritisbecame less severe than originally described.To enhance the spontaneous arthritis of MRL-lpr mice, the effect of completeFreund’s adjuvant (CFA) was investigated. In contrast to the low percentageobserved in control animals, 67-82% of mice showed clinical evidence of arthritis.Similarly, the histopathological and immunohistological analyses of the CFAinjected mice indicated a significantly higher frequency of inflammation,cartilage erosion and pannus formation, with marked infiltration of activatedinflammatory cells, including lymphocytes.The requirement of the lpr gene and background MRL genes was thenexamined. It was found that while both 7 month old MRL—+ and 3 month oldMRL—lpr mice developed arthritis, B6, B6—lpr, and 3 month old MRL—+ did notdisplay the condition after CFA—treatment. These observations suggest thatwhile the lpr gene causes a more severe early effect, background genes otherthan the lpr are also involved in the disease.11The effect of pregnancy, as another possible enhancing factor, was alsoinvestigated. Sixty-eight percent of female MRL-lpr mice developed a postpartum exacerbation of their mild spontaneous arthritis. Post-partum or postadjuvant-injection administration of estradiol prevented the enhancement ofarthritis.The effectiveness of a recently developed photodynamic therapy (PDT) wasalso assessed and compared with conventional experimental therapies in thetreatment of adjuvant arthritis. PDT inhibited the development of adjuvantenhanced arthritis with similar effectiveness as the conventional treatments, butwithout their negative side effects.These results illustrate that CFA-enhancement results in a reproduciblemodel of murine arthritis, which is useful in evaluating experimentaltherapeutic regimes, and demonstrating effectiveness of PDT arthritis therapy.111Table ofContentsAbstract iiTable of Contents ivList of tables viiList of figures viiiAcknowledgements xDedication xiGENERAL INTRODUCTION 1i.1 Systemic lupus erythematosus 2i.2 Rheumatoid arthritis 3i.2. 1 Histology of the joint 4i.2.2 Histopathology 579i.2.3 Pathogenesis of RAi.2.3.1 TCR repertoire in RAi.2.3.2 The arthritogenic antigen in RAi.2.3.3 The genetic control ofimmune responses by themajor histocompatibilitycomplex (MHC)i.2.3.4 Mechanism of tissuedestruction and boneresorption in RAi.2.3.5 Autoantibodies in RAi.2.4 Rheumatoid arthritis therapy910111517ivi.2.5. Animal models of rheumatoid arthritis 20i.2.5.1. Collagen arthritis 20i.2.5.2 Adjuvant arthritis 22i.2.5.3 Streptococcal cell wallarthritis in rats (SCW) 23i.2.5.3. Spontaneous arthritismodels 23i.3 Objective of the thesis 24MATERIALS & METHODS 25CHAPTER ONEThe MRL-lpr Murine Autoinnnune Model 47I. INTRODUCTION 48II. RESULTS 54A. Characterization of the systemic autoimmune diseasein our colony 54B. Antibodies to extracellular matrix proteins in the seraof MRL-lpr mice 66C. Spontaneous arthritis in MRL-lpr mice in our colony ... 78III. DISCUSSION 90A. Characterization of the systemic autoimmune diseasein our colony 90B. Antibodies to extracellular matrix proteins in the seraof MRL-lpr mice 92C. Spontaneous arthritis in MRL-lpr mice in our colony 95IV. SUMMARY 96CHAPTER TWOCFA enhancement of spontaneous arthritis in MRL-lpr mice 98I. INTRODUCTION 99II. RESULTS 101III. DISCUSSION 124IV. SUMMARY 129VCHAPTER THREELpr and MRL Background Gene Involvement in the Control of AdjuvantEnhanced Arthritis in MRL-Lpr Mice 130I. INTRODUCTION 131II. RESULTS 133III. DISCUSSION 149TV. SUMMARY 152CHAPTER FOUREvaluation of a Model for Post-partum Arthritis and the Role of Estrogen inPrevention of MRL-lpr Associated Rheumatic Conditions 153I. INTRODUCTION 154II. RESULTS 156III. DISCUSSION 169IV. SUMMARY 173CHAPTER FIVEPhotodynamic Therapy and its Comparison with Other ImmunomodulatoryTreatments of Adjuvant-Enhanced Arthritis in MRL-lpr Mice 174I. INTRODUCTION 175II. RESULTS 182A. Preventive photodynamic therapy of adjuvant enhancedarthritis in MRL-lpr mice 182B. Comparison of photodynamic therapy and otherimmunomodulatory treatments of adjuvantenhanced arthritis in MRL-lpr mice 192III. DISCUSSION 206IV. SUMMARY 211GENERAL DISCUSSION 212Future work 221BIBLIOGRAPHY 222viList ofTablesTable ii. 1 Assessment of reliability of measurements 35Table ii.2 Antibodies used for immunohistology 41Table 1.1 Analysis of the specificity of 4-5 month old MRL-lpr mousesera by extracellular matrix protein inhibition 77Table 2.1 Incidence of arthritis with different concentrations ofM. tuberculosis in CFA 101Table 2.2. Occurence of clinically detectable arthritis in CFA injectedMRL-lpr mice 103Table 2.3a. Histological evaluation of the CFA effect in MRL-lpr mice ... 110Table 2.3b Histological evaluation of the CFA enhancing effect inMRL-lpr mice 111Table 2.4 Serological assessment of the CFA effect in MRL-lpr mice .... 122Table 2.5 Systemic parameters of adjuvant injected MRL-lpr mice 123Table 3.1. Occurrence of clinically detectable arthritis in CFA injectedmice 134Table 3.2 Serological analysis of CFA injected mice 148Table 4.1 Systemic parameters in post-partum flare up of arthritis inMRL-lpr mice 165Table 4.2 Evaluation of the effect of estradiol on adjuvant enhancedarthritis in MRL-lpr mice 167Table 4.3 Evaluation of the effect of estradiol on systemic parameters inMRL-lpr mice 168Table 5.1 Histological evaluation of PDT in CFA-enhanced arthritis.... 186Table 5.2 Serological assessment of the effect of PDT on adjuvantenhanced arthritis in MRL-lpr mice 189Table iii. la Comparison of clinical arthritic manifestations 218Table Comparision of bistopathologic manifestations 219viiList ofFiguresFigure ii.1 The histological method ofjoint evaluation 36Figure ii.2 Custom made light-box for the photodynamic treatment ofmice 44Figure 1.1 Glomerulonephritis in MRL-lpr mice 56Figure 1.2 Characterization of the systemic autoimmune disease inMRL-lpr mice in our colony 59Figure 1.3 Kinetic response of MRL-lpr sera against collagens 67Figure 1.4 Kinetic response of MRL-lpr sera against extracellularmatrix proteins 70Figure 1.5 Titration curves for MRL-lpr sera against the listedextracellular matrix proteins 74Figure 1.6 Clinical parameters of the spontaneous arthritis 80Figure 1.7 Histopathological evaluation of the tarso-metatarsal joints oftwo to eight month old MRL-lpr mice 82Figure 1.8 Histopathological changes seen in the spontaneous arthritisof MRL-lpr mice 84Figure 1.9 Comparison of the articular histopathology of MRL-lpr micefrom two different colonies 88Figure 2.1 Swelling of the hindlegs of the MRL-lpr mice 104Figure 2.2 Clinical onset of arthritis after CFA injection of MRL-lprmice 107Figure 2.3 Histopathological changes observed in the tarsometatarsaljoints of four month old female MRL-lpr mice assessed 30 daysafter CFA injection 112Figure 2.4 Histological comparison of adjuvant enhanced andspontaneous arthritis in MRL-lpr mice 115Figure 2.5 Immunohistological analysis of the tarsometatarsal jointsof MRL-lpr mice assessed 30 days after CFA injection 118vu’Figure 3.1 Kinetic evaluation of the onset of swelling and erythema ofhindpaws in animals injected with CFA 135Figure 3.2 Histological evaluation of the joints of CFA injected mice .... 138Figure 3.3 Histological changes seen in the tarso-metatarsal joints of 8month old female MRL-÷ mice assessed 30 days after CFAinjection 140Figure 3.4 Histological evaluation of the effect of CFA injection afterthirty days (N=21) and 150 days (N=19) in seven months oldMRL-+ mice 145Figure 4.1 Post-partum flare up and the effect of estrogen on thearthritis of the MRL-lpr mice 157Figure 4.2 Histopathological evaluation of the joints 30 days afterdelivery 159Figure 4.3 Immunohistological characterization of post-partumarthritis in MRL-lpr mice 162Figure 5.1 Benzoporphyrin derivative - monoacid ring A (BPD) 178Figure 5.2 Effect of PDT treatment on the incidence of CFA-enhancedarthritis in MRL-.lpr mice 184Figure 5.3 The preventive effect of PDT treatment on the histopathologyof adjuvant-enhanced arthritis in MRL-lpr mice 187Figure 5.4 Mitogen response of MRL-lpr spleen cells after PDT 190Figure 5.5 Animals displaying swelling and erythema of hindpaws .... 194Figure 5.6 Bimaleolar ankle width 30 days after CFA injection 196Figure 5.7 Histopathological evaluation of the joints 30 days after CFAinjection 198Figure 5.8 Evaluation of systemic changes 30 days after CFA injection .. 201Figure 5.9 Histological evaluation of the kidneys 204ixAcknowledgementsI wish to thank my supervisors for their support throughout my studies:Dr. Joseph Tonzetich, who was always available whenever I needed advice,even after his retirement; Dr. Douglas Waterfield, who taught me withpatience and discreet guidance for the unwritten secrets of the scientificinvestigation, and made my Ph.D. study a pleasant experience. I would liketo express my appreciation to my thesis committee, especially Dr. Julia Levy,for their constructive participation.I am grateful to Ms. Barbara Tait, Drs. Donming Zhang, Anaklamaroon, Rubinah Chowdharry for their technical help in my experiments.I am also thankful to all the staff and students of the Department of OralBiology and the Quadralogic Technologies Inc. for their friendship andsupport. I am indebted for the advice from Dr. N. Quenville, Head of theAnatomical Pathology Department at Vancouver General Hospital, whoamazed me with his expertise in histopathology. I am also grateful for theenthusiasm of Dr. P. Reid, who taught me “patho-philosophy”.Most importantly I have to thank for my wife, Julia for her love andunderstanding; my daughters Sonja, for “still loving me”, and Nathalie, forthe happiness she showed seeing me at rare occasions during the writing ofthis thesis; and my parents, who were with me in spirit. Without their careand support I would not have completed my work.Finally, I would like to acknowledge the financial support from NSERC,UBC, and the B.C. Science Council.xDedicationFor Doug, and for my Familyfor all their care and love.xiIuopnpofrtujpiuaAutoimmurity can be loosely defined as an immune reaction againstself-antigens (Diamond, 1982). These immune reactions can be the result ofalterations in self-antigens (i.e. mutation, viruses), or tissue injury to areasnormally protected from the immune system (i.e. cornea, or cartilage).Autoimmunity can also arise from a dysfunction in the immune system,when immunological tolerance breaks down.Li Systemic lupus erythematosusSystemic lupus erythematosus (SLE) is an autoimmune disorder whosepathogenicity is ascribed to genetic, hormonal, and environmental factors.B cell hyperreactivity, which elicits autoantibody production, is characteristicof the disease. These autoantiboclies are the principal factors involved in thepathogenic mechanism. They can form immune complexes capable ofcausing glomerulonephritis, vasculitis, and neuropsychiatric (mental)manifestations (Inman, 1982). The autoantibodies may also influenceimmune regulation and directly react to antigens thereby causing severaltypes of hypersensitivity reactions, such as complement-dependent or-independent cytotoxicity or inflammation. The targets of the antibodiesinclude nuclear materials (deoxyribonucleic acid, extractable nuclear antigensuch as Sm, ribonucleoprotein RNP, histones), cell surface antigens, andimmunoglobulins (rheumatoid factors). Antibodies to nuclear material(ANA) is present in more than 95% of the patients.2The sex-hormone influence in the disease results in a ten to one femaleto male ratio. Patients with SLE have an abnormal estrogen metabolismwhich is reflected in a higher estrogenic activity. Pregnancy can trigger theonset or exacerbation of SLE pre-partum (estrogen effect), or post partum(prolactin effect) (Alarcon-Segovia, 1988).Arthritis, the most frequent clinical feature in SLE, differs from the onein rheumatoid arthritis (RA) in that it exhibits a low incidence of synovialproliferation and pannus formation. This nonerosive arthritis, however,causes severe inflammation of the ligaments and tendons.1.2 Rheumatoid arthritisRheumatoid arthritis, is another systemic autoimmune disorder ofunknown etiology, displaying both articular and extra-articular features. Itafflicts 300,000 Canadians, and world wide, about 8 million people (1% of thepopulation). It is manifested mainly by painful and ultimately disfiguringswelling of the small joints of the body. Progressive articular degenerationand ultimate destruction is the consequence of an autoimmune pathogenicsequence initiated and maintained by synovium-derived cells and cells of theimmune system.In 1988 the Arthritis Foundation published revised criteria for RAdiagnosis (Arnett et al., 1988). It cited seven criteria with the symptoms ofdiagnostic importance that are most frequently observed in the disease. Anindividual is diagnosed having RA if he/she exhibits four or more of thefollowing clinical symptoms or signs:3- morning stiffness- arthritis of three or more joints- arthritis of hand joints- symmetric arthritis- rheumatoid nodules- serum rheumatoid factor- radiographic changesi.2.1 Histology of thejointThe synovial membrane lines the joint cavity, with the exception of thecartilage (Soren, 1978). It is comprised of one continuous row of coveringsynoviocytes (synovial lining) and the subsynoviocytic tissue. There are twobasic types of synoviocytes (Barland et al., 1962): type A which is rich inlysosomes and is assumed to have phagocytic potential (macrophage-like)(Edwards et al., 1982b; Edwards et al., 1982a), and type B which contains largeamounts of endoplasmic reticulum, characteristic of elevated proteinsynthesis (fibroblast-like) (Soren, 1978). Mesothelium best describes themesodermal origin and epithelium-like character of the synoviocytes,although the synovial lining has no basement membrane (Soren, 1978).Directly beneath the synoviocytes lies a loose connective tissue, designated thesubsynovium, comprising of fibrocytes, blood vessels, a few histiocytes,macrophages, lymphocytes, all interspersed between collagen bundles inextracellular matrix consisting mainly of proteoglycans.41.2.2 HistopathologyThere are more than 100 different types of arthritis related to theunderlying disease or causative extrinsic and intrinsic factors such asphysical, chemical, metabolic, or infectious agents. The synovial membrane,a vascular connective tissue evokes an inflammatory response to injury in asimilar manner to other connective tissues. Both hyperaemia and oedemaare prominent. Increased vascularization in inflamed tissues provides easyaccess of blood mononuclear cells to the joint. The inflammatory cells may befocally or diffusely distributed in the infiltrates containing lymphocytes,plasmacytes, histiocytes, and sometimes polymorphonuclear neutrophilleucocytes. The proportion and number of each cell population depends on thetype and stage of arthritis development. Other manifestations of the diseaseinclude proliferation and/or degeneration of (sub)synovial cells, andfrequently proliferation of capillaries and connective tissue. Episodicprogression (recurrence) produces the coexistence of different stages of thedisease within the same tissue. Diagnosis can not rely solely onhistopathology. Although no histopathologic pattern has diagnosticsignificance for any single type of arthritis, the following are thecharacteristic features of RA histology (depending on activity, phase, andseverity). They include:- superficial and interstitial exudation of fluid and fibrin- infiltration of PMNs, lymphocytes, giant cells- proliferation of blood vessels and fibroblasts- synovial proliferation and hypertrophy5- villi of the synovial membrane- vasculitis- granulomata- cartilage and bone destruction- lipid deposits- fibrosisThe typical histological profile in the chronic phase of the diseaseportrays subsynovial hypercellularity with fibrocytes infiltrated by chronicinflammatory cells; palisading synoviocytes, and fibrin. In activeinflammation, exuded plasma fibrinogen may polymerize into fibrin on thesurface of the synoviocytes (Soren, 1978).Rheumatoid joint destruction occurs in three stages: inflammation,proliferation, and infiltration. In the initial inflammatory phase thepolymorphonuclear cells predominate in the synovial fluid exudate and bothtype A and B cells start to proliferate. However, histological examinationindicates that they only infiltrate the superficial synovial layer. Theunderlying supporting tissue exhibits focal accumulations of mononuclearCD4-positive T cells, plasma cells, and macrophages. If this inflammatorysynovitis persists, the synovial membrane thickens and villous formations areprojected into the articular cavity initiating the proliferative lesion. As thelesion progresses, the mesenchymal tissue (pannus) consisting mainly offibroblasts, infiltrates/erodes the surrounding tissue, replacing cartilage andbone at the periphery of the synovial reflection. This destruction eventuallyleads to decreased joint movement, then ankylosis.6Extraarticular features are also common in the advanced phase of RA.They include rheumatoid nodules (subcutaneous granuloma), scieritis,pleuritis, Sjogren’s syndrome, Felty’s syndrome (neutropenia,splenomegaly), and vasculitis.L2.3 Pathogenesis of RANumerous theories have been advanced on the pathogenic mechanismscontributing to RA. They included viral/bacterial antigen stimulation of thesynovial membrane (Sulitzeanu et al., 1988; Johnson et a?,., 1984), olygoclonalB-cell activation (Fox et a?,., 1986), T-cell regulatory defects (Poulter et al., 1985),and rheumatoid factor (RF) mediated autoimmunity (Swaak et al., 1989).There is now a generally accepted consensus that both systemic and articularmanifestations of the disease are mediated by T lymphocytes, monocytes,synoviocytes, and cytokines. However current theories differ mainly withrespect to the role, or importance that they allocate to the particular cellularcomponents of the disease mechanism. The main hypothesis advocates that ayet unknown antigen of bacterial, viral or self origin initiates theinflammation, and is persistently present and maintains chronicity (Panayiet al., 1992). The latest candidates for the etiology of the disease are thebacterial superantigens and heat-shock proteins (discussed below). The latterantigen(s) are presented to T-helper cells by antigen presenting cells (APCs)in a MHC class II restricted manner. This presentation is geneticallyassociated with HLA-DR subclasses (discussed below). Antigen-specific Tcells are then recruited by adherence to vascular endothelial cells and7transport to the joint. These T cells, and a large number of bystander T cellsaccumulate perivascularly in the (sub)synovium. As they proliferate, they inturn activate synoviocytes and infiltrating monocyte-macrophages, B cells,and other T cells, triggering both effector function and a pyrogenic cytokinecascade. This eventually results in both inflammation and joint destruction.The central role of T cells in RA is supported indirectly by several factors,such as HLA-DR gene association of RA, animal models (transfer of arthritisby T-cells), and anti-T cell immunotherapy (cyclosporin A, etc.), andamelioration by acquired immune deficiency syndrome (Ziff, 1990; Kingsleyet at., 1990). Ninety percent of CD4+ T cells from RA synovium werememory/helper-inducer subtype (CD4--/CD29) (Pitzalis et at., 1987), whichhave increased expression of adhesion molecules, thus can be recruited to thejoint more effectively, than the suppressor-inducer (CD8/CD29) T cells. Lessthan 5% of the T cells were proliferating, while 30% was in a post-activatedstate.The importance of this antigen-specific T cell reaction was recentlyquestioned, as only a small proportion of activated perivascular T cells werefound to express IL-2 receptors. However, animal experiments showed thatonly a few antigen-specific T cells are needed to initiate, or transfer animmune response (Panayi et at., 1992).Therefore it is possible, that a small number of activated antigen-specificT cells in the joint initiates an inflammatory cascade, which is amplified andmaintained by cytokines and other participating cells, such as macrophagesand synoviocytes (Panayi et at., 1992). TCR repertoire in BAThe examination of the T cell receptor repertoire in peripheral blood andarticular tissue is under way in human RA. In human RA, activated T-cellsare found in the inflamed synovium. Synovial fluid T cells showed, aheterogeneity of VB sequences expressed in CD4+ cells (Zouali et al., 1987).This may reflect the low number of antigen-specific (IL-2 R-’-) T cells inarthritic joints.One study in RA patients, however, showed the absence of Vj314 TCR inblood T cells and the simultaneous presence of this cell in synovial fluid. Asthis VB14 encodes a TCR for a bacterial superantigen, these antigens becamesuspect for initiating RA (Paliard et al., 1991).i.2.3.2 The arthritogenic antigen in BAThe inducing antigen in RA is still unknown. Bacterial, viral, andaltered self antigens have been implicated. Possible contribution torheumatoid arthritis of rubella virus, herpes simplex virus, parvovirus, andespecially Epstein-Barr virus and retroviruses (Gardner, 1992c).An association between cell-mediated immune responses in tuberculosis andin RA has been reexamined, as new evidence of T cell mediated cross reactionto cartilage antigens emerged (Gardner, 1992c). Recently the superantigensand heat-shock proteins have become the focus of interest. Heat shockproteins (hsp) are highly conserved superantigens, grouped into families by9their molecular weight (Gaston, 1991). In RA, hsp65 levels are enhanced inthe inflamed synovium. Increased proliferation of T cells to mycobacterialantigens (including heat shock proteins) has been shown (Holoshitz et al.,1986; Res et al., 1988; Res et al., 1988; Gaston et al., 1988b), however, it hasproven more difficult to implicate hsps in the mechanism of autoimmunediseases (Gaston, 1991). In murine SLE, idiotypic mimicry of hsps may berelated to the autoimmune responses, as there might be a functional linkbetween anti-idiotype antibodies reactive to DNA and antibodies to hsps.i.2.3.3 The genetic control of immune responses by the majorhistocompatibifity complex (MIIC)The MHC has a highly conserved transspecies homology. The humanMHC (called the Human Leucocyte Activation complex - HLA) resides onchromosome 6. Its polymorphism is controlled more by regulatory, thanstructural genes. The homology between murine and human MHC made itpossible to study the MHC functions in the mouse 1VIHC (H-2), and apply theresults to HLA. The MHC gene products include cell membrane associatedalloantigens, T-cell cytokines and complement factors. There is a significantassociation of certain diseases and HLA types. For this overview the mostimportant aspect is the association of DRW4 with rheumatoid arthritis, andDRW3, DW3 with systemic lupus erythematosus. Family studies have shownan association between the HLA-DR4 (Dw4 and Dw14 subtypes) and RA. Sixtyto eighty percent of Caucasian RA patients carry this gene. In otherpopulations the association is shown between HLA-DR1 and RA.10These associations do not have complete penetrance, because of non-HLAmodifying genes, environmental influence and heterogeneity of the disease.The DR (D related) locus antigens are selectively expressed on antigenpresenting cells, such as macrophages, B-lymphocytes, and on activated Tcells which are involved in cell-cell recognition. They are also homologes ofthe immune response associated MHC class II (Ta) antigens in mice.i.2.3.4 Mechrnim of tissue destruction and bone resorption in BAIn RA, initially the neutrophil leucocytes are the dominant cell speciesin the synovial fluid or synovium, followed by accumulation of inflammatorycells. These include T and B lymphocytes, macrophages, and occasionallymast cells. As the disease progresses proliferating synovial cells invade thecartilage, subchondral bone and tendon (creating the pannus), and in theprocess degrade the collagen and proteoglycan matrix of the cartilage andeventually the underlying bone (Snyderman et al., 1982).Local stimulation of mononuclear inflammatory cells, synoviocytes, andchondrocytes leads to the production and release of cytokines, growth factors,inflammatory mediators, antibodies and enzymes. These include IL-i, IL-6,tumor necrosis factor a (TNFoO, transforming growth factor B (TGFB),fibroblast growth factor, prostaglandine2(PGE2), leukotrienes, collagenases,elastase, other hydrolytic enzymes, and oxygen metabolites (Hayward et al.,1987). The synovial fluid of human arthritic joints varies in its cellularcontent but appears to have the common property of containing cells which11secrete high levels of cytokines that are postulated to be important in tissuedestruction, promotion of B cell growth and IgG production, andupregulating levels of class II MHC expression.These cytokines are promiscuous and their overall involvement in theinflammatory process depends on the target cells, and on the presence ofinhibitor molecules (Arend et at., 1990). Specific combinations of cytokinesmay also have synergistic, additive, subtractive, or suppressive effects.Unfortunately, we still do not comprehend the complexities of the cytokinecascade that operates in rheumatoid arthritis; our knowledge is still limitedto the detection of these cytokines and to some selected in vitro effects. Thisoverview will discuss in more detail the well characterized factors.Interleiilthi-1-a and B and TNF-aInterleukin 1-a and 3, are 17 kI) cytokines produced by macrophages, Band T cells, epithelial cells, and fibroblasts. They have very similar,occasionally synergistic effects as tumor necrosis factor-a (TNF-a), producedlargely by monocytes (Arend et at., 1990). It is generally accepted that IL-i isan important lymphokine in rheumatoid arthritis. Among its local effects,IL-i provides the signal from antigen presenting cells to T-cells to induceT-cell proliferation, IL-2 receptor expression, and endothelial cell activation.It promotes fibroblast and synoviocyte proliferation (Parrott et at., 1982;Schmidt et at., 1982), collagen and fibronectin synthesis, and IL-6 production.12IL-i was also found to be the osteoclast activating factor (Dewhirst et al., 1985)that is responsible for activation of osteoclasts and subsequent resorption ofbone. TNF-a also participates in bone resorption and fibroblast proliferation.Both IL-i and TNF-oc promote collagenase and PGE2 release by synoviocytesand chondrocytes (Dayer et al., 1985; Dayer et aL, 1986; Kingsley et al., 1990).IL-i also affects the proteoglycan synthesis of articular chondrocytes. It is ofparticular interest that elevated levels of both IL-i and TNF-(x have beenfound in synovial fluid of rheumatoid arthritis lesions (Harris, 1986; DiGiovine et aL, 1988; Saxne et al., 1988). Moreover, large doses of IL-i havebeen shown to enhance the severity of both collagen-induced arthritis and thespontaneous arthritis of MRL-lpr mice (Horn et al., 1990; Horn et al., 1988).Tnterleiikin-6IL-6 is a B-cell differentiation factor produced by macrophages, T and Bcells, synoviocytes, chondrocytes, and endothelial cells which suggests that itis involved in autoimmune disease pathology (Hirano et al., 1990). It iselevated in the synovial fluid and cell culture supernatant of RA patients(Hovdenes et aL, 1990). IL-6 production is upregulated by IL-i, LPS (inmonocytes), testosterone, estradiol, TNF-x, IFN-y, TGF-B (chondrocytes)(Guerne et al., 1990). Unlike IL-i, IL-6 does not stimulate collagenaseproduction or inhibit proteoglycan synthesis of chondrocytes (Kandel et al.,131990). However, via RF stimulation and stimulation of other cytokines, itmight mediate a loss of cartilage (Guerne et at., 1990; Arend et at., 1990).Interferon-yInterferon-y is produced by activated T-cells. This cytokine controlsantibody production (Johnson, 1983), regulates MHC class II expression andantigen presentation (Walsh et at., 1986), increases PMN leucocyte activity(Melby et at., 1982), and promotes synovial fibroblast proliferation(Brinckerhoff et at., 1985). IF’N-y inhibits PGE2mediated inflammatoryreactions, especially bone resorption (Boraschi et at., 1985; Gowen et at., 1986),and the Il-i induced collagenase production in chondrocytes (Andrews et at.,1990). However, the rheumatoid synovium was found to be IFN-y depleted(Veys et at., 1988). As IFN-’y level is not elevated in the joint, another factor,such as the granulocyte-monocyte colony stimulating factor (GM-CSF) mightbe responsible for MHC class II upregulation (Arend et at., 1990).EnzymesThe multicellular source of tissue degrading enzymes involves themacrophage, epithelial cells, fibroblasts, and leucocytes. Hydrolases,collagenase and neutral proteases are capable of destroying collagen,proteoglycan and other tissue constituents. Some of these enzymes also havecomplement and kininogen cleavage capability. Collagenase and gelatinase14are secreted in latent form, and can be activated by trypsin andorganomercurials (Mainardi, 1987). Gelatinase degrades denatured type I,II, III collagens (Harris et at., 1972), native type V (Hibbs et at., 1985), andtype XI collagens (Yu et at., 1990), laminin and fibronectin, and potentiatescollagenase activity (Hayken et at., 1979).ProstaglandinE2 (PGE2)Arachidonic acid metabolites, especially prostaglandins, generated bythe cyclooxygenase pathway, are of importance in regulation of fibroblastactivity. PGE2 is produced by macrophages, mast cells, leucocytes, andfibroblasts. PGE2inhibits DNA synthesis by about 50% (Ko et at., 1977). PGE2,PGI2,TXA2 are elevated in RA synovial fibroblast culture (Hamilton et at.,1985). The role of prostaglandins in bone resorption was shown in vitro byimmune complex mediated complement activation. The bone resorption isinhibited by indomethacin, a cyclooxygenase pathway inhibitor (Raisz et at.,1974). Finally, PGE2 regulates bone resorption through its effect onosteoclasts (Hayward et at., 1987). Autoantibodies in BAThe concept of immunopathogenesis of RA is changing continuously.Although the interest has shifted from the antibody-mediated process to the Tcell- macrophage-, cytokine-driven pathology, the importance ofautoantibodies in RA has not been completely eliminated.15The articular cartilage is the principal target of joint destruction in RA.Cartilage consists of collagen fibers, proteoglycans, and other organicmolecules, produced by cartilage chondrocytes. Seventy percent of the dryweight of cartilage is collagen. Eighty-five percent of the collagen is type IIcollagen, with minor amounts of type ]X, XI, and in trace amounts, type VIspecies. The surrounding bone and synovial tissue contains types I, III, Vcollagen, fibronectin, laminin, and other extracellular matrix molecules.Recently the role of collagen autoimmunity in the pathogenesis ofrheumatoid arthritis has been under intense investigation. Several studieshave reported the presence of antibodies to collagen in human sera andsynovial fluid in RA and other chronic inflammatory diseases (Stuart et al.,1983a; Petty et al., 1986; Singh et al., 1987; Gay et al., 1988; Hirsch et al., 1988;Morgan et al., 1988; Morgan et aL, 1989; Ofusu-Appiah et al., 1989; Morgan,1990). Anti-native type II collagen antibody was detected in approximately10% of patients with RA. Antibodies against the denatured type II collagenwere present in 25% and against minor cartilage collagens in 2% of thepatients. However, the findings in animal experiments have argued for apathogenic role for these antibodies. Type II collagen induced a chronicinflammatory arthritis in rats, mice and monkeys (Trentham et al., 1977a;Cathcart et al., 1986) and anti-type II collagen antibody was capable oftransferring arthritis to non-immunized animals (Stuart et al., 1983b) (seeAnimal models).However, anti-type II collagen autoantibodies in human are nonspecificto RA and display no correlation with activity, or severity of disease. Their16role as a primary agent is, therefore, questionable. It is possible that theautoantibodies are the secondary result of cartilage destruction and theimmune reaction against exposed collagen epitopes. The main IgG anti-typeII collagen subclasses produced in RA are the complement binding IgGi andIgG3 (Collins et at., 1988). Thus it is possible that these antibodies perpetuatethe disease by complement activation in the joint (Clague, 1989; Morgan,1990).i.2.4 Rheumatoid arthritis therapyThe primary goal of treatment of RA is to preserve joint function and torelieve pain. Until the causative agent(s) are determined only suppression,but not cure of the disease can be targeted (Lipsky, 1992). The drugs used incombating RA are the first line nonsteroidal anti-inflammatory drugs(NSAIDs, such as aspirin, which have gastrointestinal and hearing loss sideeffects), corticosteroids (side effect,s include osteoporosis, aseptic necrosis,infection, diabetes, hypertension, cataracts), and second-line agents. Thislatter group of disease-modifying antirheumatic drugs (DMARDs) includesulfasalazine, gold (aurothiomalate), D-penicillinamine, the antimalarialsand immunomodulators (side effects include skin rash, nephritis, bonemarrow depletion). Immunosuppressant and/or cytotoxic agents in use aremethotrexate, cyclophosphamide, and cyclosporin A (Kaufmann et al., 1984;Takagishi et at., 1986; Kaibara et at., 1983; Gardner, 1992b).17Since first line drugs (NSAIDs) fail at some point and second line drugsfrequently have significant side effects, there is a need for improved therapies.Experimental therapiesThe molecular basis of inflammation is becoming increasinglyunderstood and involves functional interactions between cells of themacrophage lineage and CD4-positive T cells which secrete cytokines that actin concert to initiate the inflammatory cascade. It is clear that the besttherapeutic strategy is one that targets the inflammatory response at theinitiation of the cascade. A number of experimental therapies have beensuccessful in the animal models of the disease. In collagen-induced arthritis,inhibition of B cell maturation in rats with anti-IgM serum has been shown toinhibit the onset of the disease, thus implicating humoral immunity in thepathogenic process (Helfgott et al., 1984). As the T cell is central to thepathogenesis of RA, anti-lymphocyte antisera has been used successfully intreatment of the disease. In both the adjuvant and collagen-induced arthritisexperimental systems the administration of anti-T cell antisera (Brahn et al.,1984) or anti-CD4 (Ranges et al., 1988) prevented induction of the disease.Moreover, antibodies against the relevant Ta antigens also decreased theincidence of arthritis and prolonged the average day of onset. The cellularimmune response has also been used to suppress collagen-induced arthritisin mice. T cell suppressor hybridomas have effectively decreased the severityof joint inflammation in treated mice; although the mechanism has not beenelucidated (Kresina, 1987).18Clinical trials have confirmed that anti-CD4 antibody could also evoke aremission of the polyarthritis (Herzog et al., 1987; Malawista et al., 1991; Choyet al., 1992). Other anti-T cell antisera, such as anti-CD7 and monoclonalantibody to the IL-2 receptor were less effective, they demonstrated a variableand short-lived clinical amelioration of the disease (Kirltham et al., 1988; Kyleet al., 1989; Lipsky, 1992). The disadvantage to these systemicimmunotherapies is a prolonged and marked depletion of helper/effector cellswhich results in immunosuppression, as it was clinically demonstrated withan anti-CD4 antibody, CAMPATH-IH. Another strategy uses the knowledgethat an MHC-peptide-TCR complex is necessary for T-cell activation. Whenpeptide competitors were used in rat adjuvant arthritis (AA), AA-associatedpeptides were found to prevent the disease (Lipsky, 1992).Another approach is the cytokine based immunotherapy, which includesanti-cytokine monoclonal antibodies, soluble cytokine receptors, and cytokineinhibitors. In preliminary trials anti-TNF reduced several indices of diseaseactivity, and it is the most promising anti-cytokine treatment up to date(Maini et al., 1993).Finally, another promising treatment modality that should be mentionedis photopheresis. The extracorporeal photochemoterapy has been usedsuccessfully in psoriasis treatment (see Chapter 5), and preliminary clinicalstudies showed short term clinical improvement in RA patients, with lessapparent toxicity than any of the other discussed therapies (Malawista et al.,1991).19i.2.5. Animal models of rheumatoid arthritisAnimal systems have offered an opportunity to study the pathogenicmechanisms in human inflammatory arthritis. Experimental systems inwhich disease has been studied are antigen induced arthritis (e.g.adjuvant-induced arthritis in rats (Pearson, 1956), collagen-induced arthritisin rats and mice (Wooley, 1988), streptococcal cell wall-induced arthritis(Cromartie et at., 1977), pristane-induced arthritis in mice (Wooley et at.,1989), antigen-induced arthritis in rabbit and mice (Lowther et at., 1978)IL-i-enhanced arthritis (Pettipher et at., 1988), proteoglycan-induced arthritisin mice (Glant et at., 1987)) and spontaneous models such as the MRL-tprmouse model (Hang et at., 1982).The most frequently studied animal models of chronic polyarthritis arediscussed below.i.2.5.1. Collagen arthritisA model of collagen-induced arthritis has been developed in which thedisease was initiated by immunization with type II collagen, and wasdependent on both a cellular and serological response to the immunogen(Stuart et at., 1984). Type II collagen arthritis is induced in mice (Courtenayet at., 1980), rats (Trentham et at., 1977b), and monkeys (Yoo et at., 1988;Cathcart et at., 1986) by an intradermal injection of either native heterologousor homologous type II collagen in complete Freund’s adjuvant (Holmdahl etat., 1986).20Multiple studies suggest that the developing arthritis is caused by cellularand humoral autoimmune responses to cartilage matrix components(Courtenay et at., 1980; Stuart et at., 1979; Trentham et at., 1978; Breedveld etat., 1987); although circulating immune complexes and sustained activationof the synovial macrophages have also been implicated (Breedveld et at., 1987).The arthritis is genetically restricted, similarly to RA, but by different MHCphenotypes (Wooley et at., 1987). Anti-type II collagen antibody is capable oftransferring arthritis to non-immunized animals (Stuart et at., 1983b), butantibodies only to certain epitopes are effective (Wooley et at., 1987).Arthritis develops in the distal joints of the extremities approximately 30days after injection (depending on species). Histologically, the initial changesconsist of activation of cells in the synovial lining, perivascular accumulationof mononuclear CD4-positive cells, and fibrin disposition on the surface of thesynovium and articular cartilage (Caulfield et at., 1982; Holmdahl et at., 1985).Granulation tissue is then formed consisting of pannus, polymorphonuclearleucocytes, and macrophages (Caulfield et at., 1982). The synovial pannusnext erodes the cartilage and subchondral bone. If the inflammationcontinues, the synovium becomes hyperplastic and periosteal new boneformation is initiated contributing to joint ankylosis. Extra-articularmanifestations involve the destruction of the elastic cartilage of the externalear and hydrops of the inner ear. Collagen arthritis has limited uses as amodel for understanding the pathogenesis of rheumatoid arthritis. There aremany differences between rheumatoid arthritis and collagen arthritis.Collagen arthritis does not occur spontaneously, it is self-limiting, does not21involve fluctuating remissions, and has antibody and cellular responses onlyto type II collagen (Trentham, 1982).i.2.5.2 Adjuvant arthritisAdjuvant arthritis has been established mainly in rats (Pearson, 1956).The disease is induced by injection of oil, containing desiccatedMycobacterium tuberculosis preparation. As in case of collagen arthritis thedisease involves an immune response to cartilage matrix componentsprobably initiated by T cell recognition of the exogenous antigens (a cartilagemimicking epitope is present on the mycobacterial 65 kD heat-shock protein)associated with Mycobacterium tuberculosis (Pearson, 1963; Quagliata et al.,1969). Both cellular and humoral immunity against type II collagen can bedemonstrated in the disease. However, the kinetics of the response vary, thecellular reactivity following the onset of the arthritis with the specificantibodies appearing later (Steffen et at., 1971; Trentham et at., 1980;Trentham et al., 1983). The joint disease shares many clinical andhistological features seen in collagen-induced arthritis, with the axialskeleton being more affected. Moreover, adjuvant arthritis displays morevaried extra-articular manifestations such as acute uveitis, urethritis, skinulcerations, granulomas in the liver, and perivascular nodules of the innerear (Breedveld et at., 1987).As with collagen-induced arthritis, there are important differencesbetween adjuvant-induced arthritis and human disease. Adjuvant arthritisdoes not occur spontaneously and there is no role for humoral immunereactions or elevated interleukin production by mononuclear cells in thepathogenesis of the disease. However, the histological changes resemble thehuman RA more closely than in other models.i.2.5.3 Streptococcal cell wall arthritis in rats (SCW)In susceptible rat strains, such as female Lewis rats, an erosivesynovitis can be induced upon intraperitoneal injection of streptococcal cellwall segments (Cromartie et at., 1977; Clark et at., 1979; Anderle et at., 1979).A chronic arthritis develops by 14 to 28 days post-injection. This arthropathycan be erosive and leads to joint deformity in some cases. The disease istransferrable by T cells (DeJoy et at., 1989). The major differences betweenSCW and RA include the lack of serological changes (RF, and anti-type IIcollagen autoantibodies), the lack of rheumatoid nodules, and the presence ofhepatic granulomas. There are also some differences in the T cellinvolvement, as the acute phase of SCW is described being T cell independentand T cells are involved only in the chronic phase.i.2.5.3. Spontaneous arthritis modelsRecently, several new spontaneous animal models have been reported inmice and goats of which the MRL-lpr mouse has been actively studied (Hanget at., 1982). This group also includes the male New Zealand black/RN mice(Nakamura et at., 1991), a high responder Biozzi substrain (Bouvet et at.,1990), and the male DBAI1 mice (Nordling et at., 1992).23The spontaneous arthritis model in MRL-lpr carries serologicalfeatures, which approximate the human condition, and develops an arthritis,whose dominant feature is synovial hyperplasia.1.3 Objective of the thesisA major difficulty in understanding the etiology and pathogeneticmechanisms of rheumatoid arthritis has been the lack of a suitable animalmodel. The main problem in using the existing animal models is that theyare experimentally induced, self-limiting, and lack many of the pathologicalfeatures seen in human disease. On the other hand, the existing spontaneousarthritis models are not suitable for therapeutic trials, because of the lowfrequency, and moderate pathology of their disease.The principal aim of this thesis was to develop a new animal model,based on a spontaneous arthritis that would be more reliable and practical fortesting therapeutic agents. Accordingly I selected to enhance the MRL-lprarthritic condition.Once the new model was established, the second aim was tocharacterize it to an extent, that will make it adequate for our plannedexperimental therapy.The final aim was to study the effectiveness of a recently developedphotodynanilc therapy in the treatment of arthritis in the described modelsystem.AnimalsMRLfMpJ-lprllpr (IVIRL—lpr), MRLIMpJ ÷1+ (MRL—i-), C57B1/6—lpr/lpr(B6-lpr), DBA/2, and C57B1/6 (B6) mice were obtained from a breeding colonymaintained in the animal facilities at the Department of Oral Biology. Thiscolony was established from stocks originally purchased from the JacksonLaboratories (Bar Harbor, ME) and bred for 2 to 9 generations. Histologicalanalysis of different generations of these mice demonstrated no differences intheir arthritic pathology of the spontaneous arthritis, in the post partum flareup, or in the ability of CFA to enhance the disease (data not shown). The micewere kept on a standard diet with water ad libitum. The colony was routinelyscreened for lvi. pulmonis, and M. arthritides , rodent coronaviruses(including hepatitis), and Sendai virus using Murine ImmunoComb test(Charles River Laboratories, MA). Only antibodies against coronaviruseswere detected. The protocol for the induction of arthritis was approved by theUniversity of British Columbia Animal Care Committee.Breeding protocol followed in Chapter Four experimentsTen weeks old virgin female MRL-lpr mice were mated. In the firstexperiment the females were separated from the males after mating, but notduring the second experiment. The litters were not weaned from themothers until the termination of the studies. At that time the female parentswere 17 weeks old.26Mitogen assaysSingle cell suspensions were prepared from spleens of 17 week old MRLlpr and MRL-+ or DBA/2 mice of a similar age. Following erythrocyte lysiswith 0.83% Tris-buffered ammonium chloride, pH 7.2, splenocytes werecultured at a cell density of 2 X 106 /ml (determined by trypan blue exclusionand manual counting of the cells in a haemocytometer) in RPMI 1640medium supplemented with L-Glutamine, 5% FBS, and 5 x 10-5 M 2-mercaptoethanol. Cells were seeded into 96-well plates at lOOp1 (2x105cells)/well and incubated with either concavalin A (Sigma Chemical Co, St.Louis, MO) (con A, at a final concentration of 2, 4, and 8 jig/ml) orlipopolysaccharide from E. coli 055B5 (LPS) (Sigma), at a final concentrationof 12.5, 25 and 50 p.g/ml) for 3 days at 37°C. Cell proliferation was determinedat the end of the incubation using the MTT colorimetric assay (Mosman, 1983;Alley, 1988).Extracellular matrix proteins (ECMPs)Type I acid-soluble rat-tail tendon collagen was prepared as described byJohnson et al (Johnson et al., 1985). An additional preparation of a similar,chromatographically purified type I collagen was purchased from EUREKAInc. (Sacramento, CA).Type II acid-soluble chicken collagen and human intervertebral discproteoglycan were kindly provided by Dr R.H. Pearce, Department ofPathology, UBC.27Type II human hyaline cartilage collagen was obtained fromCalbiochem Corp. (La Jolla, CA).Type I, III, IV, V acid-soluble human placental collagens, bovineplasma fibronectin, and ERS mouse sarcoma laminin were purchased fromSigma Chemical Co. (St. Louis, MO).Double stranded DNA (Sigma) was suspended in PBS buffer at 1 mg/mi,then sonicated to reduce the size of DNA to approximately 3000 bases. A 50tg/ml dilution was prepared and used in an ELISA assay.The purity of the proteins was confirmed by sodium dodecyl sulphatepolyacrylamide gel electrophoresis (SDS-PAGE), and/or chromatography.In the case of types I and II a comparative analysis of collagens obtained fromthe above sources yielded similar profiles.Serum collectionSera were obtained from both experimental and control groups of mice bytail bleeding, retroorbital bleeding, or by heart puncture when the mice weresacrificed 30 days post treatment. Sera used as a negative control werecollected from one year old, healthy C57BIJ6J mice. Eighteen mice from bothsexes, aged 2 to 36 weeks were bled for the kinetic study reported in ChapterlB. Sera were separated by centrifugation (10,000 rpm, Eppendorf centrifuge,Model 5412, Brinkmann Instruments Inc., Westbury, N.Y.) after coagulationof the blood and kept at - 20 °C until use.Antibodies to extracellular matrix proteins, rheumatoid factor, andserum immunoglobulin levels were evaluated in an ELISA assay.28Circulating immune complex (IC) determinitionImmune complexes were measured as described by Digeon et aL,(Digeon et al., 1977). At a low concentration of polyethylene-glycol (PEG) thesmall soluble ICs precipitate, while monomeric IgGs remain in thesupernatant. A 1J25 serum dilution was made with 0.1 M borate buffer,pH 8.4, then mixed 1/1 with 7% PEG. After overnight incubation at 4°C thepellet was washed with 3.5% PEG and dissolved in 0.1 M NaCI by sonication.The IC content was determined with an alkaline phosphatase labelled F(ab’)2rabbit anti-mouse antibody (1/3000 dilution; Serotec, Toronto, Canada), asdescribed in the ELISA assay.Semiquantitative enzyme-linked immunosorbent assay (ELISA)Acid-soluble collagen was dissolved in 0.5 M acetic acid, than dialysedagainst 50 mM Tris buffer (pH 8.3) overnight at 4°C. Fibronectin, lamininand proteoglycan were dissolved in the same TBS buffer. Ninety-six wellELISA plates (Falcon®, Becton Dickinson, Lincoln Park, NJ) were coatedovernight at 4°C with 15 .tg/ml of antigen solutions. After washing withphosphate-buffered saline (PBS) the nonsaturated binding sites of the plateswere blocked with 2% caseinfPBS during 2 hrs of incubation at 37°C, followedby three sequential washes with PBS. Fifty microliters of 1% Caseinl0.05%Tween 20/PBS buffer containing serum was applied in serial dilutions toduplicates of wells and incubated with continuous shaking overnight at 4°C.After washing three times with PBS/Tween 20, the plates were incubated29with 50 p.1 of alkaline phosphatase-conjugated goat anti-mouse IgG antibody(Calbiochem, San Diego, CA) in serum buffer (1:5000 dilution) for 1 hour at37°C. The plates were then washed again with PBSJTween 20 and developedby adding 50 p.]Jwell (0.5 mg/mi) ofp-nitrophenyl phosphate (Sigma, St. Louis,MO) in diethanolamine buffer. The optical densities were measured after 45min/37°C incubation at 405 nm by a Titertek Multiscan ELISA reader (EflabOy, Helsinki, Finland).Inhibition assaySerum at 1:50 dilution was preincubated with an equivalent volume of 0.1mg/mi antigen for 2 hr at room temperature. This mixture was reactedagainst the same type of antigen in ELISA as described above. Type I and IIcollagen inhibition was also assayed by the described ELISA method.The result was given asnoninhibited OD- inhibited OD%inhibition=______x 100noninhibited ODType I collagen immwiizationA suspension of collagen in PBS was mixed with an equal volume ofFreund’s complete adjuvant to a final concentration of 0.5 mg/mi antigen,then injected intraperitoneally (ip) and subcutaneously (0.5m1/mouse) into B630mice. Three consecutive booster injections (given i.p. in incomplete Freund’sadjuvant) were administered at 3 week intervals to develop a high titerantibody. This serum served as a positive control in anti-extracellular matrixautoantibody studies.Adjuvant injectionMice were randomly assigned into experimental groups. Thirteen tofourteen week old male and female mice were injected intradermally at twothoracic sites (selected for good lymph-drainage) with 0.05 ml completeFreund’s adjuvant supplemented to 0-10 mg/mi with heat inactivatedMycobacterium tuberculosis H37 RA (Difco, Detroit, MI) (CFA). The adjuvantwas prepared as a water in oil emulsion and administered with a 30 G 1J2needle. Control animals of the same strain and age were injected with PBS orincomplete Freund’s adjuvant.Clinical observationsA blind examination was performed on mice by the same observer everytwo days after injection for a period of 30 days, or in select cases 150 days. Thepresence of clinical disease (visual appearance of arthritis) was scored aspositive if erythema and swelling of a fore or hindpaw was observed.Bimaleolar ankle width measurements were also taken every 5 days using amicrometer. Previously the inter- and intra-examiner reproducibility and thesensitivity of the measurements were established (Table Disease31symptoms of background autoimmunity were assessed on day 30, at thetermination of the experiment, unless otherwise noted.Lymphadenopathy was evaluated by a subjective 0-4 non-invasive scoringsystem based on palpation of the lymph nodes, where a score of 0 denotes nodetectable palpable lymph nodes; 1, one palpable lymph node in one area;2, severely enlarged lymph node(s) in one area; 3, severely enlarged lymphnode(s) in both axillary and inguinal areas; and 4, massively enlarged lymphnodes in all areas.Skin lesions were evaluated by assessing the damage to the skin, theears and the tail, using a subjective 0-3 scoring system: 0, none; 1, minimal;2, moderate; 3, severe skin damage.Renal function analysisFor blood urea nitrogen (BUN) analysis, blood was collected byretroorbital sinus bleeding. The BUN content was semiquantitativelydetermined by using AZOstix (Miles, Ontario, Canada). The procedure usedfor determining the build up of urea in the blood involved a pH-dependentreaction: the sticks contain urease in the reagent area which generatesammonium hydroxide upon reaction with urea. The values were derived bycomparing the colour of the stick to a colour chart with the following ranges:1: 0.5 - 0.15 gIl; 2: 1.5 - 2.6 gIL; 3: 3.0 - 4.OgIl; 4: 5.0 - 8.0 gIl.Renal function was also assessed by semiquantitative determination ofthe urinary protein- sensitive to albumin - level in a colorimetric reaction32using Albustix sticks (Miles, Ontario, Canada). Protein was measured as g/Lwith concentrations above 0.3 gtL considered abnormal.Histological evaluationAfter the animals were sacrificed using CO2 asphyxiation, the hindpawswere removed and placed in buffered formalin for 2-7 days. Following removalof skin from the joints, they were decalcified in 10% formic acid for 48 h, thenprocessed for paraffin embedding. Serial sections were cut in a longitudinalplane to a thickness of 5 jim, and stained with Haernatoxylin and Eosin (H&E)(Figure ii.1).Histopathological alterations of the tarsal and metatarsal joints weregraded blind in consultation with two pathologists and assigned a score on aranking system modified from Horn et al. (1990). Briefly, the joints wereevaluated for the presence of the following features.1, subsynovial inflammation; 2, synovial hyperplasia; 3, cartilageerosion and pannus formation; and 4, bone destruction. Within eachparameter scores were assigned from 0 to 2 (0, no; 1, mild; and 2, severeinvolvement). The histological score was calculated by adding the values ofthe parameters, yielding a possible maximum combined score of 8 peranimal. This evaluation system was statistically determined acceptable afterassessing intra- and inter-examiner scoring (Table ii. ib). Micrographimages were obtained with a Zeiss Microscope using a transmitted lightmode.33Several animals with marked hindlimb arthritis in the tarsometatarsaljoints were also examined for the histology of other joints, namely fore limb,tail, and spine joints.Histological evaluation ofkidneysAfter sacrificing the animals, kidneys were dissected and fixed in 10%formalin. The tissues were then processed for paraffin embedding.Serial 3-5 jim thick sections were stained with haematoxylin and eosin.Kidney sections were examined for glomerular mesangial cell proliferation,crescent formation, interstitial cellular infiltration, and vasculitis.The histological changes were evaluated by a scoring system from 0 to 4,where 4 was considered severe, 3 as moderate disease, 2 corresponded to mildand 1 to minimal formed lesions (see Figure 1.1).34Table li.1 Assessment of reliabifity ofmeasurementsCriteria:x<0.4 poor0.4cx<0.75 fair1x>0.75 excellenta Bimaleolar ankle width measurement using micrometerintra examinermean of range=10.11±0.07reliability with Pearson correlation coefficient=0.678 fairreliability after ranking of measurements with Kendall coefficient ofconcordance=0.847 (excellent)inter examinerreliability with Pearson correlation coefficientraterlrater2 0.563-rater3 0.861 0.397-rater1 rater2 rater3reliability at p<O.Ol after ranking of measurements with Kendallcoefficient of concordance=0.784 (excellent)b Histological evaluation using the scoring system described earlierintra examinerreliability with Kendall coefficient of concordance=0.718 (fair)inter examinerreliability with Kendall coefficient of concordance=0.859 (excellent)35Figure ui The histological method ofjoint evaluationa The tarsometatarsal joints of a four month old MRL-lpr mice withspontaneous arthritis (H&E x 27.5)b The tarsometatarsal joints of a four month old MRL-lpr mice withCFA-enhanced arthritis (H&E x 27.5)c The plane of the histological sectioning is demonstrated on the skeletalanatomy of the mouse paw (Cook, 1965).36L€’Fi bulore(colconeum)Intermedium (ostrogolus)4th and 5th distal torsols TibioleCent role -— 1st distal tarsal3rd distal tarsal2nd distal tarsalMetatarsalClawPhalangesAdopted from Cook MJ: The anatomy of the laboratory mouse.London, Academic Press, 1965.38Tmmunohistological evaluationIn order to identify subpopulations of cells in the affected areas and anychanges during disease progression the tarso-metatarsal joints from mice ofboth control and experimental groups were fresh frozen and cryosectioned.The immunohistology was carried out applying commercially availablemonoclonal antibodies directed against mouse cell surface antigens (Tableii.2). The tissue blocks were immersed in Histo PrepTM (Fischer, Fair Lawn,NJ), snap frozen in liquid nitrogen prechilled isopentane and stored at -70°Cuntil cryosectioning, for a maximum of two weeks. Sections of 8-10 tmthickness were cut sagitally on a cryostat at -25°C. The sections were attachedto adhesive tape (Scotch Brand Tape 600, 3M Canada) (Van Noorden et aL,1986; Meacock et al., 1992), collected on glass slides, and stored at -20°C for amaximum of 1 week. Localization of antibody binding was established by abiotin-streptavidin-alkaline phosphatase method. The nonspecific bindingsites were saturated during 30 minutes of incubation with rabbit serum(Dako, Carpinteria, CA). After washing in Tris buffered saline (pH 7.6)(TBS), biotin labelled antibody was applied overnight at 4°C in 1:50 dilution inTBS containing 1% BSA. During incubation the slides were kept in a moistchamber to prevent them from drying. When non-biotinylated primaryantibody was used, after rinsing with TBS, an additional step was included: asecondary biotin labelled rabbit anti-rat Ig was added for 1 hour. Afterrinsing with TBS again the alkaline phosphatase conjugated streptavidin(Dako, Carpinteria, CA) was added for 30 minutes. After washing again inTBS the slides were treated with New Fuchsin containing TBS supplemented39with 0.1% Levamisol (Sigma, St. Louis, MO) prepared according to the factoryinstructions. The slides were developed under the microscope for about 5minutes until the desired intensity of red staining was achieved. The slideswere then counterstained with Mayer’s haematoxylin (Sigma, St. Louis, MO),covered with Crystal Mount (Biomeda, Foster City, CA) overnight at roomtemperature, and permanently mounted with Entellan (Merck, Darmstadt,Germany). They were then evaluated under a light microscope. For anegative control TBS was substituted for the primary antibody. A lymph nodeand the bone marrow within the test section from the same animal served aspositive controls. Alkaline phosphatase labelling was employed as peroxidasestaining is unreliable when the adhesive tape method is used. Endogenousstaining was reduced by 0.1 % Levamisol.40Table li.2 Antibodies used for immunohistologyMonoclonal antibody Clone Types of cells identified Source*Thy 1.2 53-2.1 Pan T lymphocytes PharmingenCD 3 YCD3-1 pan T lymphocytes Life Tech.V alpha 8 RT5O T cell receptor SerotecCD 4 R M-4-4 Helper T cells PharmingenCD 25 3C7 IL-2 receptor PharmingenCD 43 isoform IB11 activated T cells Dr. Ziltenerlak OX-6 MHC class II antigen Life Tech.F4/80 F4/80(CI:A3-1) mature macrophages SerotecAnti-rat Ig mouse Ig absorbed from rabbit Dakobiotinylated antibodies* Addresses: Dako, Carpinteria,CA; Life Technologies, Gaithersburg, MD; Pharmingen,San Diego, CA; Serotec, Kidlington, OX; Dr. H. Ziltener, Univ. of B.C., Vancouver, BC.41Estradiol adminictrationAnimals were randomly selected for estradiol injection. They received3.2 pg (0.08 mg/kg) B-estradiol-3-benzoate (Sigma, St Louis, MO) on day 2, 3, 9,15 and 21 after parturition. This dosage of estradiol was shown adequate toattain physiological serum estrogen levels that occur in pregnancy (Janssonet al., 1990).In a separate experiment, a pharmacological dose of 0.4 mg/kg/day inolive oil, was administered subcutaneously to non-mated, adjuvant injectedanimals starting day 0 for 14 days. Controls were injected with olive oil only.Treatment RegimensIndomethacin (Sigma, St. Louis, MO) was administered 1 mg/kg/days.c. Cyclosporin A (CsA) (a gift of Dr. J. F. Borel, Sandoz Ltd., Canada) at 40mg/kg/day was injected i.p. after it was dissolved in olive oil by heating in awaterbath at 65 °C. Whole body irradiation (WBI) with 3 Gy from a GOcobaltsource (Gammacell 220, Atomic Energy of Canada Limited, Ottawa Canada)was administered on day 1 of arthritis induction. Liposomally formulatedbenzoporphyrin derivative - monoacid ring A (BPD) at 2 mg/nil (Batch No.H92-902-017) was obtained from Quadra Logic Technologies (Vancouver, BC).Aliquots were prepared in 5% dextrose in water and approximately 0.5 mg/kgbody weight (20 ig/animal in 0.2 ml volume) BPD was injected i.v. via a tailvein. Animals were protected from light for 1 hour to allow for differentialbiodistribution of the sensitiser. They were then placed into individual,transparent, perforated plexiglass holders, and exposed over 1.5 hours to 8042JIcm2 (measured by an IL 1350 photometer, International Light Inc.)tungsten-halogen light filtered between 560-900 nm (Figure ii.2). Theintensity is equivalent to about 8 mW/cm2activating red light. To ensuremaximum irradiation of the hindlimbs the light was directed only at theventral side of the animals. After treatment, they were kept in the dark for 12hrs. Transcutaneous photodynamic therapy was administered on day 0, 10,and 20 after arthritis induction. The optimal concentration of BPD, light doseand timing to kill the target cells, while causing only minimal skinsensitivity, was previously determined (Richter et al., 1993b).43FIgure 11.2 Custom made light-box for the photodynamic treatment ofmiceSee specifications in the text.44Ar45-Statistical iuilysisData are expressed as mean ± standard error of the mean. In case oftwo groups, statistical differences between them were determined usingindependent t—test; multiple groups were compared by ANOVA and TukeyHSD, or Bonferroni multiple comparison tests.In Chapter one the serum dose-dependent curves were drawn as thefunction of logit (OD) = ln (ODI1-OD) versus in (dilution). The fitting of theregression lines was verified by Student t test and ANOVA (intralinevariations). The interline differences were analyzed by ANACOVA, Tukey,and Student-Newman-Kohl multiple comparison tests (Zar, 1984).46CHAPTER ONEThe MRL-lpr Murine Autoinunune Model47I. INTRODUCTIONThe MRL-MpJ-lprllpr (MRL-lpr) mouse strain develops an earlyautoimmune disease, sharing similarities with both human systemic lupuserythematosus (SLE) and rheumatoid arthritis (RA) (Andrews et al., 1978;Hang et al., 1982).The SLE-like disease presentation in these mice is largely manifested bya B cell hyperactivity. This is possibly caused by the overproduction of B-celldifferentiation factors, that in turn leads to a polyclonal expansion of self-specific B cells (Prud’homme et al., 1984). Serologic evaluation demonstratedelevated concentrations of serum immunoglobulin, antinuclear antibodies(anti-DNA, anti-Sm), and rheumatoid factors. There is a highly elevatedcryoglobulin and circulating immune-complex level, that results in animmune-complex glomerulonephritis (Theofilopoulos et aL, 1981;Theofflopoulos et al., 1985).A unique feature of the autoimmune disease in MRL-lpr mice is the lpr(lymphoproliferation) gene directed lymphadenopathy. This condition is theresult of CD3 TCR43Iow, CD4-, CD8-, (double negative) T lymphocytes, whichproliferate in the liver and migrate to the periphery (Ohteki et al., 1990).Recent results have shed light on the possible function of the lpr gene, and twoother genes, lprgc and gid, that reportedly cause similar diseases It wasdemonstrated that lpr encodes the Fas antigen (Watanabe-Fukunaga et al.,1992). Fas is a transmembrane polypeptide, which shows structuralhomology to the tumor necrosis factor receptor, nerve growth factor receptor,48and B-cell antigen CD 40. It is expressed in the thymus, liver, ovary, andheart. In other studies the structural gene of Fas was shown to berearranged by an insertion of an early retrotransposon of an endogenousretrovirus (Wu et at., 1993) in intron 2 and by a partial deletion. This resultedin nonfunctional mRNA transcripts, due to premature termination andsplicing errors (Watson et at., 1992; Adachi et at., 1993; Chu et at., 1993).The CBAJKiJmslprgc mouse strain carries a point mutation at the sameallele on chromosome 19, which is encoded by tpr (Watanabe et at., 1991), thusrendering the Fas gene nonfunctional (Watson et at., 1992). This straindisplays SLE-like symptoms similar to MRL-tpr (Matsuzawa et at., 1990;Kimura et at., 1991; Nemazee et at., 1991; Matsuzawa et at., 1992; Ogata et at.,1993). Another single gene autosomal recessive mutation is the generalizedlymphoproliferative disease (gtd) in CH3/[IeJ mice (Roths et at., 1984). Itdisplays similar manifestations as seen in MRL-lpr (Cohen et at., 1991).However, this gene is not allelic: gid is located on chromosome 1. It has beensuggested that gid and ipriFas are a ligand-receptor pair expressed bydifferent cells (Allen et at., 1990).The Fas antigen is possibly one of the apoptotic mediators responsible forprogrammed cell death of lymphocytes (Yonehara et at., 1989). It wassuggested that MRL-lpr lymphoproliferation could be the result of thedefective Fas gene’s involvement in clonal deletion of T cells in the thymus(Watanabe-Fukunaga et at., 1992). As part of the normal maturation processautoreactive T cells are destroyed by apoptosis in the thymus. The defect innegative selection of self-reactive T cells in the thyinus would enable49autoreactive thymocytes to escape and proliferate in the periphery(Matsumoto et at., 1991; Zhou et at., 1991; Zhou et at., 1993). These findingscould have important implications, as they point to the possibifity ofautoimmune diseases originating from a defect in lymphocyte apoptosis(Mountz et at., 1992).Fas was also shown to mediate apoptosis in the liver. An anti-Fasantibody had a lethal effect on normal, but not on Fas defective mice, due tosevere apoptotic liver damage (Ogasawara et at., 1993). As lpr double-negativelymphocytes were shown to proliferate in the liver (Ohteki et at., 1990), thissite might be more important than the thymus in the apoptotic failure seen inMRL-lpr mice.However, several reports have raised doubt about the above theory.These studies suggested that potential self-reactive (superantigen reactive) Tcells have been eliminated in MRL-lpr (Kotzin et at., 1988; Singer et at., 1989),the thymic events are relatively normal (Herron et at., 1993), and the doublenegative lymphocytes are derived by secondary loss of the CD4 or CD8 antigen(Singer et at., 1989).Another viable explanation is based on the finding that the Fas antigenis also expressed on activated and mature T-cells and B-cells (Yonehara et at.,1989). Thus it is possible that the Fas defect results in a failure in peripheraltolerance. Physiologically Fas could serve to eliminate the excess cells, whichare activated during normal immune function. An alternative route ofinactivation could be TCR, CD4, and CD8 downregulation on activated cells,resulting in the double-negative lymphocytes (Cohen et at., 1992).50Additionally, several experiments showed the involvement of B cells inthe SLE in MRL-lpr. Introduction of the xid gene (x linked B-cell maturationdefect) demonstrated that lpr alone is not sufficient to induce acceleratedautoimmunity (Steinberg et at., 1983). The xid gene markedly reduced 1gManti-nuclear antibodies, glomerulonephritis, and mortality.It has been shown that lpr B-cells manifest an intrinsic abnormality,resulting in high total IgG2a, antinuclear, and RF autoantibodies (Perkins etat., 1990). This autoantibody secretion requires T cells and both T and B cellsneed to be of lpr origin(Sobel et at., 1991; Sobel et at., 1993). The lpr geneaccelerates SLE, but does not cause the disease alone, since a mildermanifestation of the disease has been shown in other lpr congenic strains(Izui et at., 1984). These double negative cells have diminished lymphokineproduction and mitogen responses, and their function is questionable(Davignon et at., 1985). A new MRL-lpr substrain, with the presence of only15% of double negative cells, but with a decrease in IgG3 and IgG3-richcryoglobulins, lives almost twice as long with delayed glomerulonephritis(Fossati et at., 1993). Watson et at., identified glomerulonephritis modifyingloci on chromosomes 7 and 12 (Watson et at., 1992). These data indicate thatthe autoimmune condition in MRL-lpr mice is not the result of the tprmutation alone, but rather the consequence of complex genetic interactionswith important participation of background genes.MRL-tpr mice also spontaneously develop a rheumatoid arthritis-likedisease toward the end of their life span (Andrews et at., 1978; Hang et at.,1982). It was reported that the large and small joints of the hindlimbs of51female mice displayed a significant articular pathology. Histologicalexamination of the joints of the MRL-tpr mice demonstrated synovial cellproliferation, pannus formation, and articular cartilage erosion adjacent tothe proliferating synovial cells (Andrews et at., 1978; Theofilopoulos et at.,1981; Hang et at., 1982; O’Sullivan et al., 1985; Pataki et al., 1985; Tanaka etat., 1988). More severe manifestations of the arthropathy, such as destructionof the meniscal cartilage, formation of fibrocartilage, and bone destructionwere also observed in a certain proportion of the mice at 6-7 months of age(Hang et at., 1982; O’Sullivan et at., 1985; Tanaka et at., 1988). Investigationsof the immunological mechanisms in the localized disease have indicated thepresence of infiltrating T and B lymphocytes in the synovial tissue of theaffected mice. However, both their presence and contribution to the pathologyof the disease remains controversial (Theofilopoulos et at., 1981; Abe et at.,1981; Hang et at., 1982; O’Sullivan et at., 1985; Pataki et at., 1985; Edwards etat., 1986; Tarkowski et at., 1987). As with the human condition, the MRL-lprjoint pathology exhibits high levels of circulating 1gM and IgG rheumatoidfactors and antibodies to extracellular matrix proteins (Andrews et at., 1978;Eisenberg et at., 1979; Izui et at., 1980; Hang et at., 1982; Edwards et at., 1986;Gay et at., 1987).Inter-colony differences have been reported before in these mice(Edwards et at., 1986), likely resulting from the multiple background geneinfluences involved in disease manifestation, from genetic drift andenvironmental factors. Thus, it was important to characterize our colonybefore further experiments could take place.52This chapter discusses the characterization of the systemic autoimmunedisease in our colony, followed by the experiments related to the antiextracellular matrix protein autoantibodies in the sera of these mice. Finally,the spontaneous arthritis in MRL-lpr mice in our colony is described.53IL RESULTSA. Characterization of the systemic autoimmune disease in our colonyIn the SLE-like disease of the MRL-lpr mouse the main pathologicalfeature is immune complex-mediated glomerulonephritis (subacuteproliferative glomerulonephritis) (Figure 1.1), leading to a mortality rate of 50percent by 5.5 months of age (Hang et at., 1985; Wuthrich et at., 1989) (Figure1.2a). MRL-lpr mice are albino with approximately double the body weight ofother strains. By four months of age the average weight reaches about 50 g,then decreases with progression of illness (Figure 1.2b).As a consequence of glomerulonephritis, blood urea nitrogen (BUN)(Figure 1.2c) and proteinuria (Figure 1.2d) becomes significantly elevated byabout the fourth month of age.As a result of the autosomal, recessive lpr gene carried by this strain, amassive lymph node enlargement develops, caused by a proliferatingpopulation of CD3 CD4- CD8- (Thyl+ Lyl- Ly23-) T cells (Cohen et al., 1991).Palpable lymph node size increases to extreme levels by the time the animalsreach their fourth month (Figure 1.2e). The lymph node weight can increasea 100-fold upon aging (Figure 1.2f). Splenomegaly (Figure 1.2g), andenlargement of the thymus medulla (data not shown) also occur.A reduced responsiveness to mitogens by both T and B cells is describedas the animals age (Waterfield et at., 1987). However, about one-half of the Tcell concanavalin A and most of the B cell lipopolysaccharide response ispreserved in our colony in four month old mice (Figure 1.2h).54There is an abnormal surge in the amount of circulating immunecomplexes by the fourth month of age (Figure 1.2i). Deposition of immunecomplexes in microvasculature, particularly involving anti-dsDNAantibodies, results in damage to a number of essential organ systems andcauses lupus-like skin lesions (Figure 1.2j).Serologically, the disease is generally characterized by elevated levels ofpolyclonal Ig, circulating immune complexes, anti-nuclear antibodies(Figure 1.2k), rheumatoid factor production and antibodies against extracellular matrix proteins (Dixon, 1987), leucocytes and erythrocytes (Stim etal., 1984).55Figure 1.1 Glomerulonephritis in MRL-lpr miceKidney sections of four month old male mice were selected todemonstrate the characteristics of glomerulonephritis in MRL-lpr mice.a Minimal (focal) vasculitis, but otherwise normal histopathology (—+)(H&Ex220)b Severe interstitial cell proliferation (—+) (H&E x 440)c Moderate vasculitis (H&E x 220)d Severe mesangial cell proliferation (H&E x 440)(by, blood vessel; mcp, mesangial cell proliferation)565.7Figure 1.2 Characterization of the systemic autoimmune disease in in our colonyMeasurements were obtained as described in the Materials & Methodssection. Ten MRL-lpr mice per measurement were evaluated, and the meanof the values is displayed. Where error bars are shown, they represent thestandard error of the mean.a survivalb weightc blood urea nitrogend proteinuriae lymphadenopathyf peripheral lymph node weightg spleen weighth conA and LPS mitogen responses of MRL-lpr and DBA/2 micei circulating immune complexesj skin lesionsk anti-DNA antibody59a. Survival100806040200. I • I I •0 5 10 15 20 25Age (weeks)b. Weight6055,50454035,3025’20’• • • • •0 5 10 15 20 25Age (weeks)60c. Blood urea nitrogenI I0 5 10 15 20 25Age (weeks)d Proteinuria1080• • • •0 5 10 15 20 25Age (weeks)61e. Lymphadenopathy54C)00(I, 3-Ccu02C)0.E-J 10Age (weeks)t. Peripheral lymph node weight86.-.2’ ia)00c:.0-C>.0. .0E>‘ ._I C)C)0 20 weeks oldC57/Bl17 weeks old 24 weekso.MRL-lpr MRL-lpr62g. Spleen weight0)00)->.0C.00432020 weeks old 17 weeks old 24 weeks oldC571B1 MRL-lpr MRL-iprh. Comparison of DBA/2 andmitogen responsesMRL-lpr mice’s0•0CC0zE0)ConA 2 ConA 4 ConA 8 LPS 12.5 LPS 25 LPS 50Mitogen63i. Circulating immune complexes6O122Age (weeks)j. Skin lesions4.0• i •• i S. • •0 5 10 15 20 25Age (weeks)64k Anti-DNA antibody; 40z3.2 •• • •0 5 10 15 20 25Age (weeks)66B. Antibodies to extracellular matrix proteins in the sera ofMRL-lpr miceKinetics of antibody responses to interstitial coflagens in MLRL-lpr miceIn order to evaluate the production of antibodies to the interstitialcollagens, sera were collected from 18 male and female MRL-lpr mice ofvarious ages from two independently derived and housed colonies. The serawere then tested in the ELISA assay against native type I, II, III, and Vcollagens. It can be seen in Figure 1.3 that (at a serum dilution of 11100) MRLlpr mice were beginning to produce significant amounts of antibodies to allfour types of collagens between 17 to 20 weeks of age.Kinetics of antibody responses to fibronecthi, proteoglycan, and basementmembrane componentsThe presence of antibodies to fibronectin, proteoglycan, type IV collagen,and laminin was determined. These latter components all form part of thesynovial connective tissue. The sera of the above MRL-lpr mice were tested inan ELISA assay against these ECM proteins. Figure 1.4 shows that the miceproduce significant amounts of antibody to fibronectin after 20 weeks of age.However, the response to proteoglycan was significantly lower, appearing atabout 16 weeks and lasting for approximately 10 weeks before the responsesreturned to background levels. Antibodies against both type IV collagen andlaminin were also demonstrated between 17 and 20 weeks of age. However,these responses are significantly weaker than those to the interstitialcollagens.66Figure 1.3 Kinetic response of MRL-lpr sera against collagens: (a) type I; (b)type II; (c) type III; and (d) type V. The depicted profiles reflect the meanabsorbance values for a 1:100 dilution of serum from up to 40 week old MRLlpr mice. A 1:100 dilution of normal C57B]J6 sera is included for comparativepurposes (----).671.5a1.0>‘ci)-o0000.5 -.— .———-.——0.0 I • • I •0 10 20 30 40Age1.5-b1.0‘I)-o0000.50.0•• I • •0 10 20 30 40Age681.5C.1.0•>‘CoC0)-DCOC)0005 -.-.iJ.IJ.• I • I •0 10 20 30 40Age1.5-d1.0->‘Co0)-oCOC)000.0-0 10 20 30 40Age69Figure 1.4 Kinetic response ofMRL-lpr sera against extracellular matrixproteins: (a) proteoglycans; (b) fibronectin; (c) type TV collagen; and (d)laminin. The depicted proffles reflect the mean absorbance values for a 1:100dilution of serum from MRL-lpr mice. The responses are compared to thereactivity of a 1:100 dilution of normal C57 B116 sera (----).701.5a1.0•(0C0)as000O.52’“r- 40Age2.0b1.5->a)C0)-o1.00.50.0• I • I •0 10 20 30 40Age711.5C1.0•>CoCCo0C.)0Age1.5 -d1.0-U)Cci-oas00O304’OAge72Dilution dependency ofMRL-lpr sera against extracellular matrixcomponentsImmune complexes might interfere with the ELISA at low serumdilutions by trapping antibodies into large complexes. To avoid this problem aserum titration was conducted with a higher initial starting dilution. Pooledsera from 16- to 20-week-old and 30- to 36-week-old MRL-lpr mice were seriallydiluted and tested in the ELISA against the eight ECM components. Animmune serum derived from C57B1/6 mice immunized with type I collagen(+CI) has been included for comparison and tested against its inununogen(Figure 1.5). The analysis of the slopes, characteristic of the antibody affinity(Peterfy et al., 1983), showed similar affinities for the anti-ECM reactiveantibodies in the younger group, with a-CIT affinity being the lowest, butequal to the positive control, while in the older group a-CIT and a-CT had thehighest affinities. The reaction of the sera from the younger group wasgenerally lower than the positive control. In contrast, the older sera generallyyielded higher ELISA readings.When the ELISA data are considered as endpoint titres, a quantitativeindicator of antibody concentration (PeterfSr et al., 1983), the results confirm alower concentration of anti-ECM antibodies in the younger group sera. It isalso interesting that the concentration of antibody in the older group sera washigher than that of the positive control.73Figure 1.5 Titration curves for MRL-lpr sera against the listed extracellularmatrix proteinsData are expressed as logit versus the natural log of the sera dilution for(a) pooled sera from 4- to 5-month-old MRL-lpr mice and (b) pooled sera from7- to 8-month-old MRL-lpr mice. The significance level of the regressionallines is p<0.Ol (ANOVA). ANACOVA, SNK and Tukey analysis at the criticallevel c = 0.05 indicate the following affinities in descending order: (a) F P =CIII=CV=CICW=L=+CI=CIJand(b)CJJCJ=F+CJ=C1VPCIII = CV = L. Endpoint titrations were carried out to compare antibodyconcentrations which were found to be in the following descending order ofconcentration: (a) L, +CI, CIII, CI, F, P, CIV, CV, and CII; and (b) CV, CIII,CII, CI, CIV, L, ÷CI, and P.74-1.50 —0--— F• •____P+cI-2.00-7.50 -6.50 -5.50 -4.50Ln (dilution)2.0-1.0-Co0.0--o0-1.0E-7.50 -6.50 -5.50 -4.50Ln (dilution)75Specificity ofMHL-lpr pooled seraAliquots of the above pooled sera were preincubated with 0.1 mg/mi of thevarious extracellular matrix proteins and tested in the ELISA against thesame antigenic components. Table 1.1 indicates that the sera from 4- to 5-month-old mice could be inhibited by the specific collagen against which thesera was tested. The specific immune serum, included as a positive control,was also inhibited by the immunizing antigen.In order to evaluate the cross-reactivity, the pooled sera werepreincubated with 0.1 mg/mi type I or type II collagen prior to testing in theELISA assay against the eight ECM proteins. The results in Table 1.1demonstrate that the reactivity of the sera from the younger mice against allexamined ECM proteins is significantly reduced by inhibition with either typeI or type II collagen, suggesting extensive cross-reactivity of the antibodyresponses. As the pooled sera from the older mice exhibited similarinhibition characteristics, these data were not included. These inhibitionresults suggest that the response against the ECM molecules is highly cross-reactive; high CI inhibition of type I collagen immune sera when tested on thevarious ECM proteins similarly suggested the presence of shared antigenicepitopes.To avoid spurious readings in the ELISA assays, due to undetectedimpurities, or to the use of heteroantigen ECM proteins, we compared thereactions against different sources of the same types of collagens and obtainedsimilar results (data not shown).76Table 1.1 Analysis of the specificity of 4-5 month old MRJApr mouse sera byextracellular matrix protein inhibitionFour- to 5-month-old 1VIRL mice sera a CI immunized C57 miceAg CI CII CIInhibitionlnhibitionlnhibition InhibitionAntigen Uninhibited (%)b (%) (%) Uninhibited (%)C I 0.393 42 42 45 0.725 85C II 0.200 52 50 52 0.203 83C III 0.491 55 56 54 0.176 86C IV 0.310 37 54 57 0.089 96C V 0.327 3 42 34 0.087 82Laminin 0.552 42 36 0.184 24Proteoglycan 0.718 39 c c 0.040 83Fibronectin 0.377 34 37 38 0.425 61a Antibody sample was diluted to 11100 and OD recorded at 405 nm•b 0.05 mg/mi antigen (final concentration) was used for inhibition study.c Adverse effect on inhibition.77C. Spontaneous arthritis in MRL-lpr mice in our colonyIn the MRL-lpr mice clinical signs of arthritis were detected only bycareful, frequent examination. During the first three months of life we didnot discern swelling; the clinical signs developed during the second half oftheir life span. Thus, starting at 13 weeks of age mice were monitored for theappearance of swelling and erythema every second day. Their ankle widthwas measured every five days for one month (Figure 1.6). During this period18% (5/28) of the animals developed swelling of the hindleg. This swelllngwas both moderate and episodic. It usually did not occur in both hindlegs atthe same time and it was detectable only around the ankle and paw areas.The kinetic analysis of this arthritis was extended by evaluation of thehistopathology bimonthly starting at two months (Figure 1.7a-b). Thishistological analysis only showed minimal signs of arthritis in some of thetwo month old animals. By four months of age they developed a mildarthritis, characterized by synovial hyperplasia and subsynovialinflammation. By six months of age the animals displayed a markedarthritis. The prominent feature continued to be synovial hyperplasia, andbone erosion which was detected along with a moderate inflammation.Significant cartilage destruction and pannus formation only accompanied thedisease in the eight month group. However, monitoring the disease becamedifficult by six months of age, as the mortality rate reached 70 percent (Figure1.7b). Hence, the eight month old group might not be a representative sample,as only a few animals survived to this age. Figure 1.8 demonstrates thecharacteristic histological changes observed at different ages. In a limited78number of cases joints were also examined from the upper hindlimbs,forelimbs, spine and tail (data not shown). These joints generally displayedmuch milder pathological changes.The colony dependency of the arthritis was also examined (Figures 1.8and 1.9). The histopathology of six month old animals from our breedingcolony was compared with animals from Jackson Laboratory. The animalsfrom the two colonies showed no significant differences in the articularhistology.79Figure 1.6 Clinical parameters of the spontaneous arthritisa Animals displaying swelling and erythema of hindpawsb Bimaleolar ankle widthTwenty-eight animals were evaluated between 13 to 18 weeks of age.Error bars represent standard error of mean.80a.100—9080700600U50— 40302010•013.0 14.0 15.0 16.0 17.0Age (weeks)0.6-0.5-• 0.4-a)a)0.3a)E0200)..tzttta)IC) 0..00,13.0 14.0 15.0 16.0 17.0Age (weeks)81Figure 1.7 Histopathological evaluation of the tarso-metatarsaljoints of two toeight month old MRL-lpr micea Detailed analysis b Total histological scoreThe histological scores were determined as described in the Materialsand Methods. A given datapoint is based on the mean histological score of aminimum of five animals. In Figure 1.7b the mortality rates (%) of thetimepoints are also included.82a. Histological analysis- detailed2’• subsynovial inflammation• synovial hyperplasia—0—--- cartilage destructton & pannusD bone destructionAge (month)b. Histological analysis - total5,• total4,C)00C3,90%20%0 I I •2m 4m 6m 8mAge (month)83Figure 1.8 Ilistopathological changes seen in the spontaneous arthritis ofMRL4pr micea The ankle joint of a thirteen week old female mouse displaying normalhistology (H&E x 55).b The ankle joint of a seventeen week old female mouse exhibiting mildsynovial hyperplasia (H&E x 55).c The synovium of a tarsal joint of a seventeen week old male mouseexhibiting a mild hyperplasia of the synovial lining, with no synovialinflammation (H&E x 220).d The synovium of a tarsometatarsal joint of a seventeen week old malemouse exhibiting a severe hyperplasia of the synovial lining (H&E x 220).e The ankle joint of a twenty-six week old female mouse from JacksonLaboratory displaying a mild pannus formation, with no synovialinflammation (H&E x 110).f The ankle joint of a twenty-six week old male mouse from our colonyshowing severe subsynovial inflammation (H&E x 110).(as, articular surface; jc, joint cavity; p, pannus; sh, synovial hyperplasia; Si,subsynovial inflammation; sl, synovial lining; ss, subsynovium)84I-—.*aS.qpIp0-øe0-,61-F’L.F)SIC,00:I,.D-1&•f_s—:,9J......AFigure 1.9 Comparison of the articular histopathology ofMRJApr mice fromtwo different coloniesOur MRL-lpr colony was established eight generations ago from JacksonLaboratory (N=11). Another colony of MRL-lpr mice from Jackson Laboratorywas housed at our animal facilities since their second month of age (N=1O).At the time of evaluation the animals were six month old. A. Subsynovialinflammation, B. Synovial hyperplasia, C. Cartilage destruction andpannus formation, D. Bone destruction and E. Overall score for both groups.Statistical analysis showed no differences by independent t-test at p<O.O5 level.88C0 0 0 CD C, 0, Ci) C, a -I0mManifestationsQC)WI’) 01III. DISCUSSIONA. Characterization of the systemic autoimmune disease in our colonyInbred murine strains with spontaneously occurring autoimmunedisease have been successfully used since the 1950s for autoimmune studies.The MRL-lpr strain was developed in 1976 at the Jackson Laboratory. Duringthe last two decades this animal model has been widely studied and reportedin well over one-thousand related publications. In a 1985 reviewTheofilopoulos and Dixon published a thorough characterization of the model,including baseline data (Theofilopoulos et al., 1985). We established ourcolony’s baseline data regarding some of the systemic manifestations of theSLE disease and found them comparable to those described by Theofilopoulosand Dixon.The mortality rate rises from 0% at 12 weeks of age to about 20-24% at 17weeks of age. It reaches 50% by 22 weeks, 70% by 24 weeks, and over 90% by 30weeks of age. The body weight increases until about 5 months to 45-50 g, thendeclines.Glomerulonephritis becomes significant between 3-6 months of age, andis accompanied by a sharp increase in proteinuria and blood urea nitrogenlevels. The histopathology of the kidney is characteristic of a subacuteproliferative glomerulonephritis exhibiting monocyte infiltration, endothelialand mesangial cell proliferation, occasional crescent formation and basementmembrane thickening (Figures 1.1 and 5.9). Glomerular lesions are believedto be initiated by the deposition of circulating immune complexes. Extracted90renal immune complexes from MRL-lpr mice have been shown to containcationic anti-dsDNA IgG(2aand 2b) antibodies.Lymph node enlargement begins at 8 weeks and by 16-18 weeks ahundred-fold increase in size can occur. The lymphocytes hyperproliferate inthe liver and then infiltrate the periphery, especially the lymph nodes (Ohtekiet al., 1990). Although the absolute number of hyperproliferative lymph nodecells is high, only 5% of them are not in the resting phase. In the terminalstages of the disease, due to haemorrhaging and cystic necrosis, an actualdecrease in lymph node size can occur. Up to a seven-fold enlargement inspleen weight generally develops in MRL-lpr mice as they age.Excessive levels of circulating immune-complexes and autoantibodiesagainst a variety of macromolecules, especially nuclear components,collagens, neuronal components, and also to surface molecules in vascularand renal endothelia, erythrocytes, platelets and leucocytes are expressed as aresult of B cell activation, and can cause multisite thrombocytic events andinflammatory reactions. Complications have been documented in a spectrumof bodily functions including cutaneous abnormalities, nephritis, anddisorders of the circulatory and central nervous systems. These have beenpartly explained by the high cross reactivity of autoantibodies. The depositionof immune complexes causes skin lesions and can lead to loss of the outer earand tail.Abnormal lymphocyte function has been described, with low response toconA, and low IL-2 production.91MRL-lpr mice display large increases in serum immunoglobulinsespecially of the IgG and 1gM isotypes. Most of these are directed againstnuclear components such as ss and ds DNA, histones, and Sm antigen.This study has demonstrated that there are no noticeable changes in theseverity, onset, and symptoms of the SLE disease in our MRL-lpr colony. Itwas necessary to establish these parameters before conducting furtherexperiments. However, they were not unexpected, since breeding- especiallyat the Jackson Laboratory- has been carefully controlled to preserve these SLEattributes.B. Antibodies to extracellular matrix proteins in the sera ofMRL-lpr miceThe reported studies on anti-collagen reactivity of the sera from theMRL-lpr mice are inconsistent. Gay et al., found no reaction against collagentypes I and III. Anti-collagen type II responses appeared between 13 and 20weeks of age. Type IV responses were iow in older mice while type Vresponses were stronger (Gay et al., 1987). Anti-collagen type II responseswere confirmed by Tarkowski et al, and Bartlett et al, with minor differencesin the onset of the serological response (Tarkowski et al., 1986; Bartlett et al.,1988). Phadke and co-workers also demonstrated sex-independent anti-collagen type II responses, and reported an anti-collagen type I response aswell (Phadke et al., 1984).The joint contains most of the 17 types of collagen which are found in thecartilage or in the surrounding tissue (Gay et al., 1980; Mayne, 1989). Inaddition, the joint has other extracellular matrix components such as92vascular basement membrane proteins, proteoglycans, and fibronectin(Rosenberg et al., 1986; Carsons et aL, 1989). We reevaluated the serologicalreactivity of MRL-lpr mouse sera against the various extracellular matrixcomponents. These experiments reveal a much broader extracellular matrixantibody response than has been reported previously. Starting between 17 and20 weeks of age these mice display significant amounts of antibody productionto the interstitial collagens, types I, II, III, and V. Moreover, the sera alsocontained strong reactivity to fibronectin, a glycoprotein associated with cellsand interstitial collagen fibrils. The MRL-lpr sera demonstrated a lowerreactivity to proteoglycans. When tested against basement membrane-derivedtype IV collagen and laminin the sera displayed significantly weaker activitythan the responses to the interstitial collagens. These findings are indisagreement with Gay et al.,(1987) who showed little or no responses againstcollagens types I and III. Our differences could not be explained by differentdilutions of sera used in the ELISA assay or immune complex interference asinterline analysis of serially diluted )VFRL-lpr sera demonstrates thatserological responses of these mice are not significantly altered by titration ofthe autoimmune sera. Studies on the specificity of the antisera by antigeninhibition experiments suggest that the anti-collagen responses that wedemonstrate are highly cross-reactive. These findings are not surprising asimmunity is often limited to antigenic sequences found in the helical portionof the native molecules (Morgan, 1990) and these sequences can be shared bymany types of collagens. Thus, it seems that the responses to extracellularmatrix proteins cannot be explained simply by the production of a specific type93of antibody for each extracellular matrix protein. Hence, it is proposed thatthe diversity seen here is probably generated by a small group ofautoantibodies recognizing shared antigenic epitopes. Certainly there isevidence in the literature suggesting cross-reactivity between native anddenatured collagens (Morgan, 1990) as well as collagens and Clq (Antes et aL,1988). It was also demonstrated that sera from mice over 20 weeks of ageshow 20 to 40% extracellular matrix adsorbable cross-reactivity with RNP-Sm.Furthermore, it was shown that IgG rheumatoid factor from these mice hada 20%-40% cross-reactivity with anti-extracellular matrix antibodies (data notshown). These complex cross-reactivity patterns have been revealed betweenanti-DNA antibodies and proteoglycans (Faaber et al., 1984). In our studiesthere did not appear to be any correlation between the antibodies to theseextracellular matrix proteins and individual serum immunoglobulin levels,suggesting that the autoimmune response is oligoclonal in nature.It is apparent that the MRL-lpr mouse strain exhibits similar anti-collagen antibody profiles to those seen in the sera and synovial fluid ofpatients with rheumatoid arthritis. Anti-collagen type II antibodies havebeen found in 10% of rheumatoid arthritis patients (Morgan, 1990). Inaddition, increased levels of antibodies to collagens types I, III, 1V, and Vhave also been demonstrated in the disease, albeit with a lower frequency ofpatients being seropositive (Stuart et al., 1983a). In the MRL-lpr mousemodel, however, all mice tested that were older than 17-20 weeks exhibitedstrong reactions against the ECM proteins. Again this was predictable asthese mice are genetically identical, sharing both arthritogenic and immune94response genes. The human equivalent would be the subset of rheumatoidarthritis patients displaying genetic associations between HLA-DR genes andthe presence of anti-collagen type II autoantibodies.C. Spontaneous arthritis in 1WRJApr mice in our colonyIn 1982 MRL-lpr mice, were reported to spontaneously develop arheumatoid arthritis-like disease towards the end of their life span (Hang etal., 1982). At one month of age there were no signs ofjoint disease. By 3 to 4months early synovial pathology was recognizable in 45% of the animals. At 5to 6 months of age 75% of these mice had synovitis, periarticularinflammation, and vasculitis. In many instances erosion and destruction ofarticular cartilage were evident. It was also demonstrated that the pathologyin the joints correlated with the levels of circulating rheumatoid factor. Basedon related articles (Horn et al., 1990; Mountz et al., 1994) and on the above datafrom our colony, the disease observed in original animals has become lesssevere in succeeding generations. The animals did not develop a significantarthritis until six months of age; by that time they also exhibited a 70%mortality. Moreover, the differences between rheumatoid arthritis and theMRL-lpr spontaneous arthritis has become more evident. Severemanifestations of the disease (pannus formation and extensive cartilage andbone destruction) were rarely seen in our colony; a non-erosive arthritisremained the prominent feature for their practical life span (5.5 months).This makes the MRL-lpr spontaneous arthritis a less useful animal model forstudying both the pathogenesis and therapy of RA.95IV. SUMMARYThe MRL-lpr mouse strain develops an early autoimmune disease,sharing similarities with both human systemic lupus erythematosus andrheumatoid arthritis. This autoimmune condition is the result of complexgenetic interactions involving participation of both the ipriFas gene andbackground genes. Inter-colony differences have been reported before, andduring the last five years the arthritic component of the MRL-lprautoimmune disease has decreased. We characterized both the systemic andarthritic disease components in our colony to establish baseline data for ourexperiments. We also investigated the serology of these animals, andexpanded our study into the involvement of the anti-extracellular matrixprotein autoantibodies, to investigate the background of some conflictingprevious reports.Our results show that there are no noticeable changes in the severity,onset, and symptoms of the lupus-like disease in our MRL-lpr colony,compared to the original description by Theofilopoulos and Dixon. Mostimportantly the animals developed immune complex-mediatedglomerulonephritis, lymphoproliferation, and skin lesions, with a 50%mortality rate at 5.5 months. Serologically the disease can be characterizedby hypergamm aglobulinaemia, anti-nuclear antibodies, rheumatoid factorproduction and antibodies against extracellular matrix proteins. We showedthat in these mice, there is a highly cross-reactive, possibly oligoclonal antiextracellular matrix antibody production. It involves not only anti-type IIcollagen antibodies, but antibodies reacting with type I, III, IV, V collagens,96fibronectin, and less frequently laminin and proteoglycans. This points to amore generalized connective tissue disease, rather than arthritis suggestedpreviously by some authors.The spontaneous arthritis was less severe in our colony than originallydescribed. The animals did not develop a significant arthritis until sixmonths of age; by that time they also exhibited a 70% mortality. This was inagreement with recent reports, related to spontaneous arthritis in these mice.This makes the MRL-lpr mouse a less useful animal model for both studyingthe pathogenesis of RA and assessing therapeutical modalities.97CHAPTER TWOCFA enhancement ofspontaneous arthritis in MRL-lpr mice98I. INTRODUCTIONMRL-lpr mice develop a spontaneous polyarthritis at the latter part oftheir life span. As a result of their autoimmune condition, they acquire anarthritis of hindlimb by approximately six month of age (Hang et at., 1982;O’Sullivan et at., 1985; Bartlett et at., 1986). The severity of the SLE at this ageresults in a high mortality rate (Hang et at., 1985). Younger animals,however, display a milder state of arthritis. In addition, the severity andfrequency of joint disease found in these mice has decreased significantlyduring the past five years in many investigators colonies as well as in ourown. This is probably due to breeding programs that enhance the SLE-likecomponent of the disease. These factors diminish the usefulness of the strainfrom being a practical model both for human rheumatoid arthritis and forassessment of the efficacy of various treatment plans.Previously, Hom et at., have demonstrated that the spontaneous arthritisseen in MRL-lpr mice could be enhanced by the daily subcutaneous injectionof IL-lB (Hornet at., 1990). These authors report that the onset of the arthritisoccurred within one to two days after the initial injection with a maximalincidence at two to three days. The histopathological responses observedapproximated the arthropathy induced when IL-i was used to augment thearthritis in collagen immunized DBAJ1 mice (Horn et at., 1988). This arthriticresponse to IL-i was highly dose dependent; when rats were treated withsignificantly lower concentrations of IL-i, it reduced the severity and durationof the arthropathy (Jacobs et at., 1989). In a preliminary experiment MRLlpr mice were injected with similar doses of human recombinant IL-lB that99were used by Hoim et aL, It did not enhance the arthritis in our colony (datanot shown).Boissier et at., immunized MRL-lpr mice with native mouse type IIcollagen. They found that while collagen was antigenic, it did not elicitarthritogenic effects (Boissier et at., 1989). The preliminary experiment, inwhich the MRL-lpr mice were immunized with chicken type II collagen,yielded similar results (data not shown).Hang et at., showed that lipopolysaccharides were able to enhanceautoantibody responses and immunopathology of SLE in MRL-lpr (Hang etat., 1985). However , when we administered LPS derived from E. cdi, it didnot enhance the arthritis (data not shown).In continued attempt to accelerate the disease we injected CFAintradermally and found that CFA induces an earlier and more severearthritis in MRL-lpr mice.100II. RESULTSOptimal concentration ofM. tuberculosis in CFAThrough a series of preliminary experiments we established the optimalconcentration of M. tuberculosis (Table 2.1) by injecting intradermallycomplete Freund’s adjuvant containing M. tuberculosis concentrationsranging from 1-20 mg/mi. The lowest ratio that was used had no enhancingeffect on the spontaneous arthritis (12%; 2/16), while the 5 mg/miconcentration resulted in a moderate increase in incidence (43%; 3/7).The 10 mg/mi concentration was chosen as the lowest amount that resulted inhigh incidence (82%; 9/11). Using this concentration we avoided possibleundesirable side effects and facilitated the injection technique (highconcentrations of suspension tend to clog the needle).Table 2.1 Incidence ofarthritis with different concentrations ofM. tuberculosis in CFACFA Femalea Malea Totala %1 mg/mi 1/2 1114 2/16 125mg/mi 114 2/3 3/7 4310 mg/mi 6/6 3/5 9/11 8215 mg/mi 3/6 5/6 8/12 6720 mg/mi 3/4 8/12 11116 69a Ratio of clinically detectable arthritis / number of animals injected101Effect of CFAon the appearance of arthritis in I4JERL-lpr miceIn a series of six experiments, 67-82% of mice that received the adjuvantinjection showed evidence of arthritis characterized by swelling and erythemaof the hind legs (Table 2.2). Female mice appeared to be marginally moreaffected by the adjuvant injection as has been previously reported for thespontaneous model (Theofilopoulos et aL, 1981). In sharp contrast, only 6 outof 50 (12%) of the control mice (PBS- or IFA-injected) showed clinicalmanifestations of arthritis. The swelling in CFA injected animals wasusually moderate to severe (Figure However, extreme swelling was alsonoted in a few cases (Figure 2. ic). A kinetic study of this adjuvant-inducedenhancement showed that the swelling occurred as early as five days with themajority of animals exhibiting the onset between 14 and 24 days after theinjection (Figure 2.2a-b). Ankle width measurements were comparable tovisual observation results (Figure 5.2b).102Table 2.2. Occurrence ofclinicafly detectable arthritis in CFA injectedMRL-lpr miceIncidence of clinically detectable arthritisExperiment” Femalea Malea Totala Total%#1 6/6 3/5 9/11 82#2 5/6 11/15 16/21 76#3 3/8 5/6 8/12 67#4 9/12 11/16 20/28 71#5 11/15 7/11 18/25 69#6 4/5 7/10 11/15 73TotaIc 38/50 (76) 45/63 (71) 82/112 73Incidence of clinically detectable arthritisControl Femalea Malea Totala Total%#ld 0/5 1/7 1/12 8.3#2e 0/10- 0/10 0#3d 2/10 1/6 3/16 19#4d 1/7 1/5 2/12 17Total 3/32 3/18 6/50 12a Ratio of clinically detectable arthritis / number of animals.b Complete Freund’s adjuvant (CFA) injected MRL-Ipr mice.C Values in brackets denote percentages.d Phosphate buffered saline (PBS) injected MRL-Ipr mice.G Incomplete Freund’s adjuvant (IFA) injected MRL-Ipr mice.103Figure 2.1 Swelling of the hind1eg of the MRL-lpr micea Control, 17 week old mouse showing no visible effect of IFAinjectionb Littermate, injected with CFA, showing signs of inflammatoryoedema characterized by redness and swellingc Littermate, injected with CFA, showing extreme swelling both inthe paw and ankle area, which lasted throughout the experiment104SOFigure 2.2 Clinical onset of arthritis after CFA injection ofMRL-lpr micea Incidence of clinically detectable arthritisThe number of animals with erythema and swelling was recorded daily for 30days following CFA injection. Data represents a percentage incidence (N=57).b The occurrence of clinical arthritis in the females (N=22) and males(N=20) is presented. Animals were examined daily for 30 days after CFAinjection.107a100 —‘I,e00000rnrit__.__11 i1011 1213 1415 16 1718 1920 2122 232425 26272829 30Days after CFA injection80604020b0 5 10 15 20 25 30Days after CFA injectionO Male• Female80060012110123456789108Histological evaluation of the adjuvant enhanced arthritis in MRL.lpr miceThe histological features of the joints of animals displaying clinicaldisease were assessed 30 days after CFA injection. As demonstrated in Table2.3a, the adjuvant injected mice showed a significantly higher meanhistological score than the control group. CFA injection also significantlyincreased the percentages of mice developing a more severe degree of jointpathology. In order to correlate the histological with the clinical features ofthe joints, the mice receiving the injection were divided into two groups:Group 1: with joint swelling; Group 2: without joint swelling (Table 2.3b).The early histopathologic evidence of arthritis, such as subsynovialinflammation and synovial hyperplasia, were present in varying degrees inall the groups of animals at the time of the sacrifice. These changes,however, were only present to a mild degree in the control animals, but morepronounced in the injected animals (Figures 2.3a and 2.3b). The moreadvanced histopathologic evidence of arthritis, such as cartilage erosion andbone destruction, was rarely seen in the control animals. In contrast theCFA injected mice displayed a significantly higher frequency of advancedhistopathologic alterations (Figures 2.3c and 2.3d). Figure 2.4 compares thehistopathologic changes detected in CFA-injected animals to the different agegroups of spontaneous arthritis. The data demonstrate that CFA injection ofthree month old mice resulted in arthritis within 30 days, similar in severityto the spontaneous arthritis seen between seven to eight months of age.109Table 2.3a. Ilistological evaluation of the CFA effect in MRL-tpr miceAnimalsa CFA injected MRL4pr PBS injected MRL-lpr p1F M F Mnumber 48 51 99 16 15 31subsynovial 1.12 1.10 1.11(87) 0.56 0.47 0.52(48) 0.0000inflammationbsynovial 1.29 0.96 1.12(87) 0.69 0.47 0.58(55) 0.0000hyperplasiabcartilage erosion, 0.62 0.55 0.59(46) 0.19 0 0.10(10) 0.0002pannus formationbbone destructionb 0.54 0.57 0.56(43) 0.19 0.13 0.16(16) 0.0034histological 3.58 3.18 3.37(98) 1.62 1.07 1.35(77) 0.0001scorec±0.31 ±0.28 ±0.21 ±0.31 ±0.30 ±0.22a F=female M=male =Totalb mean histological score in the corresponding groupnumbers in brackets indicate % of the animals developing listed pathology° Mean ± s.e.m.d independent t-test results at p significance level between Z groups:ECFA injected vs 1PBS injected control110Table 23b Histological evaluation of the CFA enhancing effect in MRL-lprmiceCFA injected MRL-lpr CFAinjected MPeL-lprAnimalsa With sweflinu Without gwellinF M F Mnumber 11 17 5 9 14% subsynovial 100 94 96 100 56 71inflammationb%synovial 90 70 78 80 56 61hyperplasiab% cartilage erosion, 72 70 71 60 0 21pannus formationb% bone destructionb 73 41 11 14histological 5.18 3.29 4.04d 2.80 1.22 1.79dscorec ±0.68 ±0.43 ±0.41 ±0.58 ±0.28 ±0.33a F=female M=male Z=Totalb % of the animals in the corresponding groupc Mean ± standard error (SE)d Kruskal-Wallis and Tukey HSD test results at p<O.05 significance level between Zgroups111Figure 2.3 Histopathological changes observed in the tarsometatarsaljointsoffour month old female MBL-lpr mice assessed 30 days after CFA mjectiona The tendons and muscles surrounding the metatarsal joints areseverely affected by an inflammatory infiltrate (H&E x 55)b The expanded subsynovial and synovial area is invaded by a denseinflammatory infiltrate (H&E x 110)c Invading pannus (p) destroyed most of the cartilage and underlyingbone (H&E x 220)d Higher magnification of insert on C indicating the fibroblastic nature ofthe pannus (d) and mononuclear cell infiltrate (mc) in close proximityto a blood vessel (by) (H&E x 1100)(as, articular surface; b, bone; bm, bone marrow; by, blood vessel; jc, jointcavity; mc, mononuclear cells; p, pannus; si, synovial inflammation)112(JHkPFigure 2.4 Histological comparison of adjuvant enhanced and spontaneousarthritis in MRL-lpr miceHistological scores of the CFA injected group (4 month old, N=35; 20%mortality) are compared with different age groups of non-injected animals: 2month old (N=8; 0% mortality), 4 month old (N=31, 20%), 6 month old (N=11,70% mortality), and 8 month old (N=5, 90% mortality). Data are presented asmean ± s.e.m.11550I0U(00U00‘I,43210116Subsynovial Synovial Cartilage destruction Bone Totalinflammation hyperplasia & pannus formation destructionTmmwiohistological evaluationThe composition of immune cells present in the joints of control andCFA injected animals were examined 30 days after the initiation of theexperiment. Non-fixed tarso-metatarsal joints of animals displaying clinicalsigns of arthritis were cryosectioned and stained with the streptavidin-biotinalkaline-phosphatase system, as described in the materials and methodssection. Most of the synovial lining cells stained for the F4180 antigen, whichis detectable on mature macrophages (Figure 2.5a). The anti-MHC class IIantibody stained excessively in the subsynovium and in certain areas of thesynovial lining (Figure 2.5b-c). There was a moderate staining of CD3 antigen(pan T marker) bearing cells in the subsynovium and rare presence in theareas of the hypertrophic synovial lining (data not shown). In adjacentsections the anti-CD43 (activated T cell marker) antibody stained in a similarpattern, indicating that most of the infiltrating lymphocytes were activated(data not shown). CD4÷ cells were detected occasionally in the subsynovium(Figure 2.5d).117Figure 2.5 Immunobistological analysis of the tarsometatarsaljoints ofMRL-lpr mice assessed 30 days after CFA injectiona Most of the synovial lining cells stain for the mature macrophagemarker F4180 (—*) (New Fuchsin counterstained with Mayer’sHaematoxylin x 220)b MHC-class II surface marker bearing cells surround a large area ofbasophilic depositsThe adjacent synovial lining cells also express the MHC-class IIactivation marker (—*) (New Fuchsin counterstained with Mayer’shaematoxylin x 110)c The subsynovial tissue is infiltrated with cells staining for the MHCclass II activation marker (—*) (New Fuchsin counterstained withMayer’s haematoxylin x 110)d Anti-CD4 antibody staining cells are localized in the proximity ofsynovial lining cells (—+) (New Fuchsin counterstained with Mayer’shaematoxylin x 440)(as, articular surface; b, bone; bd, basophilic deposit; jc, joint cavity,ss, subsynovium)118‘4.rzoSerological assessment of the adjuvant enhanced arthritis in MRL-lpr miceChapter 1 revealed the presence of significant levels of antibodies tocollagens types I, II, III, IV, and V, fibronectin, laminin, and proteoglycansin untreated MRL-lpr mice starting between 17 and 20 weeks of age. Hence, itwas of interest to determine whether the serological profiles would change inthese mice after intradermal injection of CFA. Thirty days after CFAinjection the animals were bled and the sera tested against extracellularmatrix proteins. Table 2.4 demonstrates the serological responses of thecontrol and CFA injected groups of animals to the listed extracellular matrixproteins. The results show that the adjuvant injected mice demonstrated astatistically significant enhanced antibody production to Type I collagen, TypeII collagen, DNA, and the positive control M. tuberculosis. Moreover, serumimmunoglobulin levels were also significantly higher, but not the serum IgGlevels. The serological responses to chondroitin sulphate and 1gMrheumatoid factor were not significantly affected.121Table 2.4 Serological assessment of the CFA effect in MRL-lpr miceCFA injected MRL-lpr PBS injected MRL-lpr p(n1,n2)AntigenMean ODa SEM Mean GD SEMM.Iuberculosi,s 0.352C 0.026 0.208C 0.019 0.002 (48, 19)Type I collagen 0.548d 0.034 0.350d 0.039 0.002 (58, 22)Typeilcollagen 0.135e 0.015 0.081e 0.013 0.010 (16, 21)Chondroitin sulphate 0.196 0.025 0.126 0.042 0.146 (58, 22)1gM Rheumatoid factor 0.134 0.014 0.115 0.021 0.454 (57, 22)Serum Jg level 0.790f 0.035 0.639f 0.078 0.047 (68, 24)Serum IgG level 0.185 0.018 0.137 0.035 0.199 (68, 24)DNA 0.833g 0.036 0.64W 0.072 0.010 (59, 22)a OD represents Optical Density. Mean OD is calculated as the mean value of ELISA readings. ODis a mean value of triplicate measurements using 1:100 dilution of the serum under investigation.As a negative control, sera from age matched, uninjected C57B6 and MRL- + mice were used.(data not shown).For the anti type-I collagen assay polyclonal mouse antiserum developed against the antigen wasused as a positive control (see Materials and Methods)b SEM denotes ± standard error of mean.cdefg Indicates statistically significant difference (p<0.05) obtained with independent t-test between thecorresponding data.122Systemic effects of CFA hjectionThe possible systemic effects of CFA injection were evaluated. At the end ofthe experiment no significant changes were detected in mortality, weightgain, lymphadenopathy, proteinuria, and kidney histopathology due to theCFA injection (Table 2.5). However, there was a general (albeit not alwayssignificant) elevation of serum antibody levels, which is likely the result ofpolyclonal B-cell activation by the adjuvant (Table 2.4).Table 2.5 Systemic parameters of adjuvant hjected MRL-lpr miceAnimalsa Spontaneous CFA-enhancedarthritis arthritisMortality 28/38 26% 66/84 21%Weight- 1.6 ±1.42 (28) 1.8 ±0.74 (66)gain (g)Lymph- 2.68 ±0.23 (28) 2.63 ±0.15 (66)adenopathybProteinuriab 1.09 ±0.18 (28) 1.32 ±0.31 (66)Kidneyb 6.87 ±0.82 (16) 7.42 ±0.54 (45)histologyIndependent t-test showed no differences between the two groups at p<O.05 significance levela Data is presented as the mean± s.e.m. of the measurement at the end of the experiment.The number of animals are in parenthesis.b See Materials and Methods.123Ill. DISCUSSIONMRL-lpr lupus-prone mice have been described as the first inbred strainto demonstrate a spontaneous polyarthropathy that is useful as a model forthe study of human rheumatoid arthritis. Unfortunately, this model has notbeen useful for the study of therapeutic regimens as any significantmanifestations of the arthropathy appeared at the end of the animals lifespan.In the last decade CFA has been extensively used to induce anexperimental arthritis in rats, creating a model which has proven useful inthe elucidation of the mechanism(s) of tissue destruction (Pearson, 1956).Recently, a similar treatment has been successful in eliciting a murine formof the arthritis (Keitel et al., 1971; Geller et al., 1974; Knight et al., 1992). AsMRL-lpr mice are predisposed to develop a spontaneous rheumatoid arthritis-like disease it was of interest to determine whether CFA could affect theinflammatory sequelae of the rheumatoid condition found in young MRL-lprmice.One aspect of the study examined the short term effect of CFA injectionon the progression of the arthropathy seen in MRL-lpr mice. It was foundthat a single intradermal injection of CFA reproducibly enhances both thefrequency and the severity of the disease in young MRL-lpr mice. MRL-lprmice between thirteen and fourteen weeks of age were injected intradermallywith O.lnil of CFA supplemented to 10 mg/mi with M. tuberculosis. Seventy-three percent of the injected male and female animals demonstrated swellingand erythema of the hindlegs; the majority of these changes occurred between12414 and 24 days following the injection. The percentage of the above mice thatdeveloped arthritis after injection may be potentially higher as it was found inone of the experiments that upon examination of the injection site in one of theexperiments that the intradermal injections had been inappropriatelyadministered in three out of eighteen animals. In about 25% of the cases theswelling of the same joint remained severe throughout the whole experiment.In the rest of the animals the swelling was episodic.The histological scores of this hindlimb arthropathy in injected animalswere significantly higher than those of the control group of animals. Themore advanced histopathological evidence of arthritis, such as cartilageerosion and bone destruction, was more frequently found in injected animals(Figures 2.3c and 2.3d). Similarly, the percentages of animals exhibitingpathological effects within each ranking system was higher in CFA injectedmice than in the control mice. CFA injected animals demonstrated furtherelevation of the arthritic pathology when they developed swelling. However,the degree of swelling did not correlate with histopathology. Earlier studieshave demonstrated that in adjuvant induced arthritis in rats the animalselicited a cellular immune response to amino acids 180-188 in the sequence ofa Mycobacterium bovis BCG antigen. The etiology of that arthritis probablyresulted from a cross-reaction between mycobacterial antigen(s) and epitopespresent on the cartilage surface (Van Eden et aL, 1988; Van Eden et aL, 1989a;Van Eden et al., 1989b). Our findings indicate a significant serologic responseto M. tuberculosis in control mice occurring as a consequence of theprogression of the spontaneous disease, a situation similar to the cellular125responses seen to mycobacterial antigens in patients with rheumatoidarthritis (Holoshitz et al., 1986). In our model the cross reaction probablyoccurs between the cartilage immunogens and the mycobacterial antigens aswellThe results suggest that the induction of adjuvant arthritis in MRL-lprmight be T cell initiated as the administration of cyclosporin A, whichinterferes specifically with T cell activation, is effective in preventing swellingand histopathological changes in our mouse model (see Chapter 5). Antigen -M. tuberculosis -presentation to lymphocytes in the lymph nodes is likely toresult in their activation followed by their migration to the joints via thecirculatory system. An elevated expression of the MHC class II activationmarker was detected both on synovial lining and subsynovial cells. There wasa moderate presence of CD3, CD43, and occasionally CD4 surface antigenbearing T cells in the subsynovium as well. The presence of T cells, especiallythe activated T cells, confirms the T-cell involvement in the adjuvantenhancement. In mice showing no clinical signs of arthritis, jointinfiltration by lymphocytes was not observed. The tissue destruction is likelythe result of the large number of mature (F4/80 antigen bearing)macrophages that were detected in the arthritic joints.The serological evaluation of the CFA injected mice showed a significantincrease in the antibody response to Type I and Type II collagen, DNA, andthe positive control M. tuberculosis. However, the antisera levels tochondroitin sulphate and rheumatoid factor remained unchanged. It is alsoclear from the data that the overall serum immunoglobulin level is increased126in the CFA injected animals, possibly reflecting an increase in the antibodiesassociated with the disease state. This general elevation of serum antibodylevel is likely the result of some degree of polyclonal activation by the adjuvant.However, there were no other systemic changes related to CFA; CFA did notsignificantly modify lymphadenopathy, glomerulonephritis,or mortality.In spite of these findings the role of autoantibodies to the extracellularmatrix proteins in the pathogenesis of the disease in our model remainsunclear. It is possible that serological responses to homologous collagens andproteoglycans are not involved in the disease process and that another as yetundetermined effector mechanism is responsible for the pathology (Cremer etal., 1990).An important consideration in the analysis of our model is whether theCFA has enhanced the spontaneous arthritis that develops in these mice orwhether we have in fact demonstrated yet another example of a CFA-inducedmurine disease (Geller et al., 1974; Keitel et al., 1971; Knight et al.,, 1992). Thisis difficult to determine because of overlapping features seen in thehistological evaluations reported in both models. However, we have treatedMRL-+ (congenic mice that do not spontaneously develop a detectablearthropathy), C57B6-lpr, C57B6, and Balb/c mice with CFA using the sameexperimental protocol as described above and demonstrated that age matchedanimals do not develop any significant degree of clinical or histopathologicaldisease (Chapter 3). These findings suggest that we are indeed enhancing theclinical appearance and histopathology of the spontaneous IVIRL-lpr arthritis.127This study is the first report of adjuvant enhancing a spontaneousarthritis-like condition in mice. The fact that almost five times more CFAinjected than control animals develop a severe arthropathy with demonstrablecartilage erosion and pannus formation at seventeen to eighteen weeks of ageindicates that this model will be very useful for investigation of varioustherapeutic regimes.128IV. SUMMARYAn extensive investigation was carried out on the effect of completeFreund’s adjuvant (CFA) on the spontaneous arthritis of MRL-lpr mice. Thedevelopment of swelling and erythema was monitored for one month afterinjecting 13-14 week old mice intradermally with CFA. After one month thehistopathology, iminunohistology of the joints and serological responses toextracellular matrix proteins were investigated.In a series of six experiments, 67-82% of mice showed early clinicalevidence of arthritis in contrast to the low percentage observed in controlanimals. Similarly, the histopathological analyses of the CFA injected miceindicated a significantly higher frequency of advanced histopathologicalalterations, characterized by cartilage erosion and pannus formation. Theimmunohistological staining detected an elevated expression of the MHCclass II activation marker both in the synovial lining and in the subsynovium.There was a moderate presence of CD3 and CD43 surface antigen bearing Tcells in the subsynovium as well. The serological evaluation of the adjuvantinjected mice demonstrated a significantly enhanced antibody production totype I and type II collagens, DNA and the M. tuberculosis positive control.CFA injection did not cause change in the systemic pathological parameters,with the exception of generally elevated serum antibody responses.This reproducible adjuvant-enhanced model of murine arthritis will beextremely useful in evaluating experimental therapeutic regimes as thearthritis is initiated earlier and exhibits an enhanced frequency and severitycompared to the spontaneous arthritis seen in MRL-lpr mice.129CHAPTER THREELpr and MEL Background Gene Involvement in the Control ofAdjuvant-Enhanced Arthritis in MRL-lpr Mice130I. INTRODUCTIONAs previously discussed, MRL—lpr mice spontaneously develop a uniqueautoimmune condition clinically displaying vasculitis, proliferativeglomerulonephritis, and inflammatory arthritis (Andrews et al., 1978;Theofilopoulos et al., 1981; Hang et al., 1982; Steinberg, 1984; Theofflopoulos etal., 1985; O’Sullivan et al., 1985; Tanaka et al., 1988).The MRL mouse strains were developed by Murphy and Roths (Murphyet al., 1977; Murphy et al., 1979). These strains were derived from LG/J (75%),AKRJJ (12.6%, including the lak MHC gene), C3H!Di (12.1%), and C57BL/6J(0.3%). During inbreeding, a spontaneous autosomal recessive mutationarose in the MRL colony. This eventually led to the development of asubstrain, which shares 99.6% of its genome with the strain of origin, butdiffers at the lpr (lymphoproliferative) gene locus. The two strains have sincebeen designated MRL-+ and MRL-lpr. Recently, the lpr gene has beentransferred by selective mating onto different strains of mice (B6-lpr, AKRlpr, and C3H-lpr (Wofsy et al., 1981)). This has enabled investigators toevaluate the effect of the lpr gene, in the absence of the MRL backgroundgenes, on the immunoregulatory defects seen in these mice.It became clear that both the lupus and the rheumatoid arthritiscomponent of the unique autoimmune condition seen in the MRL—lpr miceare genetically determined. Studies using the autoimmune MRL—+ congenicpartner have suggested that the lpr gene considerably influences severallupus associated disease parameters such as lymphadenopathy,glomerulonephritis and serological responses to autoantigens (Hang et al.,1311985). rrhef it was of interest to determine the requirern for the iprgene and MRL backgron gene(s) in the developme of the CFA—enhancedarthritis seen in treated MRL—lpr mice.132II. RESULTSAnimds displaying swelling and erythema ofhindpaws after CFA injectionAs previously described, three or seven month old MRL—lpr, MRL—+,B6—lpr, and B6 mice were injected with CFA and examined over a period of 30days for the clinical appearance of arthritis. During this period treated threeor seven month old B6—lpr, and B6 mice failed to display any swelling orerythema of their hindpaws (Table 3.1).Fifteen percent of untreated and 74 percent of CFA injected 3 month oldMRL.—lpr mice developed clinical signs of arthritis. This is similar to theresults described in chapters one and two. Surprisingly two month oldMRL—lpr mice did not develop any significant signs of disease followingtreatment (8%, 2/24). Three month old MRL—+ mice also failed to developvisually detectable arthritis (3%, 2/75), while 7 month old injected animalsclearly demonstrated both swelling and erythema of their hindpaws (74%,42/57 MRL-lpr and 29/39 MRL-+). In these studies female MRL—lpr andMRL—+ mice appeared more severely affected by the adjuvant injection,although the difference were less noticeable in a larger pool of animals(Chapter 2). A kinetic study of the effect of adjuvant injection showed that theswelling in 3 month old MRL—lpr and 7 month old MRL—+ began atapproximately day eight, with the onset in most of animals appearingbetween 14 and 24 days after the injection (Figure 3.1).133Table 3.1. Occurrence of clinically detectable arthritis in CFA rnjected miceStrain Age Femalea Malea Totala %MRL-lprb 2 months 0/14 0/14 0/28 0(control) 3 months 3/22 3/18 6/40 157 months 1/8 118 12.5MRL-lprc 2 months 2/18 0/6 2 / 24 83 months 20/24 22/33 42/57 74MRL-+c 3 months 1/38 1/37 2/75 37 months 21/26 8/13 29/39 74B6lprc 3 months 0/12 0/13 0/25 07 months 0/7 0/6 0/13 0B6c 3 months 0/16 0/6 0 / 22 07 months 0/2 0/2 0/4 0a Ratio is calculated as number of animals with clinical signs of arthritis / totalnumber of animalsb Uninjected mice (control with spontaneously occurring arthritis)CFA injected mice134Figure 3.1 Kinetic evaluation of the onset of swelling and erythema ofhindpaws in animals injected with CFAThe number of animals with erythema and swelling was recorded everysecond day for 30 days following CFA injection. Data are presented aspercentage of the incidence. The groups of animals examined included 3month old MRL-lpr (uninjected control) (N=14), and CFA injected 3 month oldMRL-lpr (N=57), 7 month old MRL-+, (N=39) 3 or 7 month old B6-lpr (N=38),and 3 or 7 month old B6 mice (N=26).135100—a-—— MRL-lpr uninjected controlMRL-lpr80 —0-—-— MRL-+ [seven month old]—a——— B6-lpr and B660C’)CuE2 404200• I0 10 20 30Day (Post-injection)136Histological evaluation of the adjuvantr-euhanced arthritisMice were sacrificed 30 days after CFA injection and their jointsprocessed for histological evaluation. As indicated in Figure 3.2, the adjuvantinjected MRL—lpr mice displayed a notably higher mean histological scorethan the uninjected control group. Similarly, the MRL—+ mice also developeda significant arthropathy; the histological score approximated that found inthe untreated MRL—lpr mice. The joints of the MRL—+ mice showedsignificant subsynovial inflammation (Figure 3.3c) with associated myositis(Figure 3.3d). Basophilic deposits were found in the inflammatory areas ashas been reported for the MRL—lpr mouse (Figure 3.3d) (Edwards et at., 1986;Pataki et at., 1985). It is noteworthy that the average histological score forpannus formation in these mice in Figure 3.2 was higher than those found inuninjected MRL—lpr animals. An example of the fibrous pannus on thecartilage surface is shown in Figure 3.3e, while Figure 3.3f demonstrates abone and cartilage destruction of a highly invasive pannus. Moreover,although the average histological score for bone destruction was low, evidenceof bone destruction was discerned (Figures 3.3c and f).The lpr gene when present on the C57B1/6 background did not allow forthe development of any significant histopathology (Figure 3.2) confirming theclinical findings depicted in Table 2.1. The joints from non—autoinununecontrol mice (B6) also did not display any of the manifestations of theautoimmune condition.137Figure 3.2 Histological evaluation of the joints ofCFA injected miceThe experimental groups included 3 month old MRL-lpr (uninjectedcontrol) (N=31), and CFA injected 3 month old MRL-lpr (N=35), 7 month oldMRL-+, (N=39), 3 or 7 month old B6-lpr (N=38), and 3 or 7 month old B6 mice(N=22). The histological scores were established as described in the Materialsand Methods. A. Subsynovial inflammation, B. Synovial hyperplasia, C.Pannus formation and cartilage destruction, D. Bone destruction and E. Totalscore for each group. The numbers indicate Tukey HSD test results at psignificance levels between injected animals and the uninjected MRL-lprcontrol group. NC shows no histological changes. Error bars indicatestandard error of mean.138p=O.O1 5p=O.003p=O.852 Ep=O.000p=O.699p=O.579p=1.000 Dp=O.000p=O.923p=O.879p=0.142 Cp=0.005p=0.005p=O.O01p=1.000 Bp=0.000p=0.002p=O.000p=O.667 Ap=0.000Histoiogical Score0 1 2 3 4I • I • INCNCTflh-iNCNCIIIItII...—I_IIIIIIIIIIIIIII____________________NCINCIIIIIIIIIIIIIIIII-IEJ C57b1—I D C57b1.-Iprll1 MRL-++E MRL-Ipr• MRL-Ipr (control)139-1-IFigure 3.3 Histological changes seen in the tarso-metatarsal joints of 8 monthold female MRL-÷ mice assessed 30 days after CFA injectiona A fifteen month old uninjected MRL-+ mouse showed no signs of jointpathology (H&E x 110)b Accumulation of cells in the synovial cavity during the early stage of thechronic phase of the disease (day 21) (H&E x 1100)c Osteoclastic resorption of cortical bone and perivascular infiltration in theadjacent stroma (H&E x 440)d Association of myositis with aggregates of basophilic material (H&E x 440)e Widely spreading pannus with moderate destruction of the underlyingcartilage. The highly vascularized pannus consists mainly of type a and bsynoviocyte-like cells (H&E x 440)f A highly invasive pannus have destroyed most of the cartilage and the bonytissue 150 days after the CFA injection (H&E x 220)(as, articular surface; b, bone; bd, basophilic deposit; by, blood vessel; jc, jointcavity; mf, macrophage-like cell; oc. osteoclast-like cell; p, pannus; PC,plasma cell; si, synovial inflammation; sl, synovial lining; ss, subsynovium)140‘142..•“..:.Ia•.•—a.a’aaa--a. F•a.-a.•.,40.-...-,—-_:saI’bI.:‘‘3Histological evaluation of the short term and long term effect of CFA injectionin MRL-+ miceIn the previous section it was reported that the total histological scorewas significantly less in CFA injected MRL—+ mice compared to that of theirMRL—lpr counterparts. The damage approximated that found for uriinjectedMRL—lpr mice. It was possible that the time course for the development oftissue damage after CFA injection was different in MRL—+ mice. We injectedseven month old 1VIRL—i- mice with CFA to test this possibility. Theseanimals were sacrificed either thirty or 150 days after injection and the jointswere processed for histological evaluation as described previously. Figure 3.4shows that, in the short term, subsynovial inflammation is the moreprominent feature of the arthritis. In the long term, however, synovialhyperplasia was the significant feature. More advanced destruction;including pannus formation, cartilage destruction, and bone resorptionincreased with time as well.I144Figure 3.4 Histological evaluation of the effect of CFA injection after thirtydays (N=21) and 150 days (N=19) in seven month old MRL-+ miceThe histological scores were established as described in the Materialsand Methods. A. Subsynovial inflammation, B. Synovial hyperplasia, C.Pannus formation and cartilage destruction, D. Bone destruction and E. Totalscore for each group. The numbers show independent t-test results at psignificance levels between the two groups. Error bars indicate standarderror of mean. % depicts percent of animals that developed the listedpathology.145Histological ScoreP=°.365 D5%32%p=O.11414%58%p=O.045 B33%37%p=O.064I MRL/÷+ day 150I!1 MRL/++ day 30p=O.4610.0 1.076%E58%10%2.0j1146Serological evaluation of the adjuvant?.enhanced arthritisIn Chapter Two, it was demonstrated that CFA injected three month oldMRL-lpr mice displayed a statistically significant enhanced antibodyresponse to type I collagen, type II collagen, DNA, and the positive control M.tuberculosis. Moreover, serum immunoglobulin levels were also significantlyhigher. The serological responses to chondroitin sulfate and 1gM rheumatoidfactor, however, were not significantly affected. It was of interest todetermine whether the serological profiles would change in the various testmice after intradermal injection of CFA. Table 3.2 depicts the serumimmunoglobulin levels and serologic responses of the experimental andcontrol groups of animals. The responses to type II collagen and M.tuberculosis increased in all cases. The most pronounced increase occurredin MRL-+, and it was accompanied with an increase in serum Ig level.However, the increase did riot appear to correlate with the disease activityseen in the individual strains of mice.147Table 3.2 Serological analysis ofCFA injected miceStrain Age Term M. tuberculosis Ig CII #MRL-lpr 3 mo control 0.28±0.02 0.29±0.04 0.24±0.04 2130 day 0.34±0.02 0.24±0.02 0.40±0.02 457 mo control 0.46 0.22 0.34 1MRL-÷ 3 mo control 0.18±0.02 0.07±0.02 0.07±0.03 1230 day 0.33±0.03 0.26±0.02 0.44±0.03 707 mo control 0. 18±0.03 0. 16±0.03 0.28±0.07 1030 day 0.55±0.08 0.46±0.07 0.58±0.06 22150 day 0.92±0. 10 0.48±0.07 0.54±0.07 22B6 3 mo control 0.04±0.03 0.05±0.00 0.08±0.07 230 day 0.08±0.02 0.14±0.03 0.14±0.04 8150 day 0.31 0.09 0.27 17 mo control 0.13 0.09 0.07 130 day 0.32±0.08 0.17±0.07 0.15±0.05 3B6-lpr 3 mo control 0.04±0.01 0.04±0.01 0.06±0.01 1030 day 0.12±0.02 0.07±0.01 0.11±0.02 287mo control 0.03±0.01 0.11±0.05 0.10±0.06 330 day 0. 16±0.06 0.11±0.05 0. 19±0.08 4148III. DISCUSSIONMurine SLE is genetically determined and involves the interactions ofthe lpr gene with gene(s) of the MRL background strain (MRL—+ mice)(Theofilopoulos et al., 1981). The MRL—i- strain differs genetically by lessthan 1% from the MRL—lpr mice and develops a different pattern of SLEreflected both at the clinical and humoral/cellular level (Izui et at., 1984;Warren et at., 1984; Dixon, 1985; Alexander et at., 1985; Kelley et at., 1985).MRL-lpr mice develop an early-onset form of SLE with 50% mortality at 5.5months. The congenic MRL-+ mice develop a late-onset form of the diseasewith 50% mortality at 1.5 years in females, and 2 years in males (Hang et at.,1985).The mice also develop an arthropathy, which however, is very mild andoccurs only in mice older than 15 months (Theofilopoulos et at., 1981; Hang etat., 1985; Hewicker et at., 1988). The tpr gene was transferred to severalmouse strains, resulting in elevated autoantibody levels, and a milder degreeof lymphadenopathy in the recipient strains (Wofsy et at., 1981). However,these non-MRL strains, which carry the lpr gene, do not display arthritis.In our experiments MRL—lpr, MRL—f, B6—lpr, and B6 mice wereinjected with CFA intradermally. We detected significant clinicalpresentation of disease only in MRL—lpr and MRL—i- mice. The disease onsetoccurred between 14 and 24 days after the injection. There was a strong agerequirement, which might reflect a more advanced stage of the underlyingpathological mechanism. This mechanism in turn can be triggered by a CFAinjection. In MRL—i- animals the disease could only be induced in mice olderthan 6 months, while in MRL-tpr mice the optimal age was 3 months. These149findings were substantiated by the histological analyses of the joints. BothMRL—lpr and MRL—f- mice developed significant joint disease after CFAinjection, the MRL—i- pathology approximating that found in uninjectedMRL—lpr mice. Thus, the overall MRL—+ joint pathology found afteradjuvant treatment was not as severe as that demonstrated for the adjuvanttreated MRL—lpr mouse. The longer life span of MRL—+ mice made itpossible to examine the long term effects of CFA injection. The arthritis didbecome more severe with time following injection (150 days). In several cases,the animals had an immobilization of the tarso—metatarsal joints associatedwith severely destructive myositis and chronic inflammatory infiltrate.Clearly the lpr gene is not responsible for the arthropathy alone, as CFAinjected B6—lpr mice did not exhibit any significant histological changes inthe joints. These findings are in partial agreement with the work of Gilkesonet al., who showed that most MRL—lpr spontaneously developed synovialhypertrophy while B6—lpr failed to demonstrate any synovial changes(Gilkeson et al., 1989; Gilkeson et al., 1991).The evaluation of sera from the mice demonstrated a significantincrease in responses to type II collagen and M. tuberculosis. However, thesechanges did not correlate with the frequency or severity of the disease. Thisagain suggests that the antibody driven mechanisms play a less importantrole in this form of arthritis.Recently the lpr gene was identified as the defective Fas apoptotic gene((Watanabe et al., 1991; Watanabe-Fukunaga et al., 1992)- discussed in moredetail in Chapter 1). In the MRL—lpr mice the defect in the apoptotic150mechanism may cause a breakdown of the central and/or peripheraltolerance (Cohen et al., 1992). This results in failure of the apoptosisregulated elimination of the CD4/CD8 negative T cells, and in theiraccumulation in the secondary lymphoid organs. However, these cellsappear nonfunctional and it is unclear how they are implicated in theautoimmune process (Davignon et aL, 1985). Although it is clear from ourfindings that the lpr/Fas gene is involved in the generation of the CFAenhanced arthritis in MRL—lpr mice, it does not seem to have the primaryrole. Recently the genes moclifring the severity of the glomerulonephritis inthese mice were identified on chromosomes 7 and 12 (Watson et al., 1992);other than the lpr locus. There are probable arthritis modifying genes in theMRL genetic background and the arthritis might reflect an interaction at thelevel of the effector cells mediating the disease process and not on the Fassignal transduction pathway.1511V. SUMMARYThe requirement of the lpr gene and background MRL genes wasexamined on the CFA-enhanced murine arthritis. MRL—+, MRL-lpr, andB6—lpr mice (experimental) and B6 mice (control) were injected intradermallywith complete Freund’s adjuvant containing M. tuberculosis H37 RA. Thedevelopment of swelling and erythema was monitored for 1 month after theinjection, when the histopathology of the joints was investigated. It was foundthat while 74% of both 7 month old MRL—i- and 3 month old MRL—lpr micedisplayed clinically visible arthritis, B6, B6—lpr, and 3 month old MRL—÷ didnot develop the condition after CFA—treatment. In accordance with theclinical observations the histopathological changes were manifested only inolder MRL—i- and 3 month old MRL—lpr mice. One month after the CFAinjection, milder changes were observed in the MRL—+ than in the MRL—lprmice, with the MRL—+ mice developing a disease of similar severity touninjected MRL—lpr mice. In the long term (150 days) more severedestructive changes could be demonstrated in the cartilage and bone of theMRL—+ mice although the average histological scores did not show anystatistically significant differences from those found in the MRL—+ 30 daysafter injection. The serologic evaluation of the adjuvant-injected mice showedsignificantly enhanced antibody production to type II collagen and M.tuberculosis, but did not correlate with the disease activity. Theseobservations suggest that while the lpr gene causes a more severe early effect,background genes other than the lpr are more involved in the adjuvantenhanced arthritis afflicted mice.152CHAPTER FOUREvaluation ofa Model for Post.partumArthritis and the Role ofEstrogen in Prevention ofMRL-lpr Associated RheumaticConditions153I. INTRODUCTIONSystemic lupus erythematosus (SLE) is a progressive connective tissuedisorder accompanied by a wide spectrum of autoantibody production. Thetissues/organs commonly affected are the skin, kidney, brain, vasculature,heart, and diarthroidal joints. Articular disease is the most frequent earlymanifestation of SLE being present in more than 90 percent of cases. Thisarticular pathology is often transient, and generally less severe than that seenin rheumatoid arthritis (RA). As articular changes often precede otherfeatures of SLE, they are frequently diagnosed as RA. Notwithstanding, RAalso may coexist with SLE (Gardner, 1992a).SLE is a disease that most frequently afflicts women of childbearing age,and RA also occurs three times more frequently in women then men (Linos etal., 1979; Gardner, 1992b). Therefore the effect of pregnancy on these diseasesis of particular importance. It is known that estrogen levels increase duringpregnancy, and decline immediately after delivery. Estrogen has two effectson the immune system: it suppresses antigen-specific T cell-dependentimmune functions, characteristic of RA, and stimulates B-cell responses,characteristic of SLE (Holmdalil, 1989). The exacerbating effect of pregnancyin SLE is controversial (Petri et al., 1991; Urowitz et al., 1993), but itsameliorating effect on RA is well described (Ostensson et al., 1983). However,more than 90% of patients with RA remission during pregnancy develop apost-partum flare up (Rook et al.,). These findings suggest that estradiolmight influence these rheumatic conditions.154The MRL-tpr mouse strain has been successfully used as a model forboth SLE and RA (Murphy et at., 1977; Dixon et at., 1978; Andrews et at., 1978;Murphy, 1981; Theofilopoulos et at., 1981; Hang et at., 1982; Theofilopoulos etat., 1985). The strain develops a mild spontaneous arthritis which can beenhanced by an intradermal injection of Freund’s complete adjuvant(Chapter 2). The sex hormone involvement in SLE in MRL-tpr mice is wellcharacterized (Steinberg et at., 1980; Shear et at., 1983; Ansar Ahmed et at.,1985; Brick et at., 1988; McKenzie et at., 1988; Carlsten et at., 1989; Carlsten etat., 1990; Carlsten et at., 1991; Seggev et at., 1991; Caristen et at., 1992). Incomparison, less attention has been directed towards the effect of estrogen onarthritis in MRL-lpr mice(Carlsten, 1992), and its relation to pregnancy inthese mice has not been described.We noticed during routine breeding of MRL-lpr mice that a significantnumber of older female breeders developed post-partum joint erythema andswelling. Hence this study examines the possibility of whether the MRL-tprspontaneous arthritis could serve as a model for post-partum arthritis flareup in SLE or RA. We also investigated the effect of estradiol on the murinearthritis, using both the post-partum arthritis flare up model and theadjuvant enhanced arthritis model in MRL-tpr mice.155IL RESULThThe incidence of swelling and erythema in post-partum animals during30 days after delivery amounted to 76% (13/17) (Figure This finding wasconfirmed in a second experiment in which 60% of the animals demonstratedsimilar evidence of swelling and erythema (12/20) (data not shown). Theclinical onset of post-partum flare up was successfhlly prevented (21%, 3/14)and the time of the onset delayed by the administration of physiologicalamounts of estradiol. This frequency seen in estradiol injected animals wascomparable to the frequency of clinical signs of onset of arthritis in non-mated, age matched female MRL-lpr mice, which developed spontaneousarthritis (23%, 5/22). Similar patterns were shown by bimaleolar anklemeasurements (Figure 4. lb) which quantitate the severity of the swellingalthough there was a remission in swelling in some of the cases in all threegroups.Subsynovial inflammation, synovial hypertrophy, and the overallhistological score increased significantly post-partum (p<0.05) (Figure 4.2).However, there was no significant increase in cartilage erosion, pannusformation, or bone destruction. Administration of physiological doses ofestradiol significantly lowered the overall histological score (p=O.034), andprevented cartilage erosion, pannus formation, and bone destruction. It alsoslightly diminished the level of inflammation and synovial hypertrophy.156Figure 4.1 Postpartum flare up and the effect ofestrogen on the arthritis ofthe MRL-tpr micea Incidence of swelling and erythema b Changes in binialeolar ankle widthThe 30 days post-partum period had an exacerbating effect on the clinicalarthritis parameters (1). One group of animals after delivery ( o ) received0.08 mg/kg estracliol at days 2, 3, 9, 15, and 21 and showed an incidence,comparable to, and a severity, lower, than the control group with spontaneousarthritis ( •).157a100spontaneous arthritis• post partum arthritis—0-— post partum arthritis+estradiol60Et< 40Time (days)b8.0• postpartum arthritis• spontaneous arthrttis0 postpartum arthritis+estradiol< tt tb t 20t 30Time (days)158Figure 4.2 Histopathological evaluation of the joints 30 days after deliveryThe histological scores were determined as described in the Materialsand Methods. A. Subsynovial inflammation; B. Synovial hyperplasia: C.Pannus formation and cartilage erosion; D. Bone destruction; E. Overallhistological score. * the corresponding value is signicantly different fromthe post-partum arthritis group’s value (p<O.05 ANOVA and Boncerronimultiple comparison test).159C)10UU)C)00)00U)BConditionC D * E160Figure 4.3a demonstrates extensive subsynovial inflammation andinfiltration of MHC class II positive cells. Cells also expressed the MHC IIantigen at certain areas of the hyperplastic synovial lining (Figure 4.3b).There was a sporadic but significant presence of CD3 and CD43 expressingcells and occasionally of cells staining for the CD4 marker in thesubsynovium. Anti-IL-2R mAbs were rarely detected in the test or the positivecontrol sections (data not shown).Neither the post-partum period, nor the short term administration ofphysiological amounts of estradiol altered significantly any systemic diseasemanifestations recorded at the termination of the experiment. These micedeveloped a marked lymphadenopathy, significant proteinuria, andglomerulonephritis (Table 4.1).161Figure 4.3 Tmmunohistological characterization ofpost-partum arthritis inMPaLlpr micea The inflamed subsynovial tissue is infiltrated with anti-MHC class IIantibody stained cells (-+) (New Fuchsin counterstained with Mayer’shaematoxylin x 110)b Hyperplastic synovial lining cells express the MHC class II marker (—+)(New Fuchsin counterstained with Mayer’s haematoxylin x 220)c CD3 surface marker bearing cells infiltrate the subsynovium adjacent tothe synovial attachment area (—k) (New Fuchsin counterstained withMayer’s haematoxylin x 440)d Anti-CD4 antibody staining cells inifitrated the subsynovial tissue (—‘P)(New Fuchsin counterstained with Mayer’s haematoxylin x 110)(as, articular surface; jc, joint cavity; sh, synovial hyperplasia; ss,subsynovium)162Table 4.1 Systemic parameters in post-partum flare up ofartluitis inMRL-lpr miceAnimals a Spontaneous Post partum Post partum (p1, p2)arthritis arthritis arthritis + estradiolMortality 7/32 22% 5/21 24% 3/14 21%Weight (g) 40±0.94 (23) 43±0.89 (8) 42±1.07 (12) 0.210, 1.000Lymph- 3.13±0.24 (23) 2.75±0.37 (8) 2.75±0.35 (12) 1.000, 1.000adenopathybProteinuriab 1.23±0.23 (23) 1.40±0.41 (8) 1.02±0.35 (12) 1.000, 1.000Kidney 6.75±0.93 (12) 5.00±1.34 (5) N.E.histologya Data are presented as the mean ± s.e.m. of the measurement at the end of the experiment.The number of animals is listed in parenthesis.b See Materials and Methods.Pi ANOVA and Bonferroni multiple comparison test results at p significance level betweenspontaneous arthritis and post-partum arthritis groupsP2 ANOVA and Bonferroni multiple comparison test results at p significance level between postpartum arthritis with estradliol-injected and post-partum arthritis groupsNE. Not examined165The effect pharmacological amounts of estradiol (0.4 mg/kg) have on thedevelopment of adjuvant-enhanced arthritis in female JYi[RL-lpr mice was alsoinvestigated. Thirteen week old mice received this dose daily for two weeksstarting on the day of the CFA injection. This dose prevented the adjuvantenhancement of disease, as indicated by the similar frequency of clinicalarthritis in treated animals and control animals (Table 4.2). The severity ofswelling was significantly lower than in adjuvant-injected mice (p=0.012).Histological analysis revealed a significant degree of prevention in all thepathological parameters examined, including the overall score (p<0.05). Thisdose, however, increased the mortality level by 1/3 from the usual mortalitylevel of 22% which is characteristic of this strain at 4 months of age (Table4.3). In the survivors this was accompanied by a significant decrease oflymphadenopathy (pO.02), and a marked increase of proteinuria in someanimals, resulting a statistically non-significant (pO.183) elevation in themean proteinuria level.166Table 4.2 Evaluation of the effect ofestradiol on adjuvant enhanced arthritis inMRL-lpr miceAnimalsa Spontaneous CFA-enhanced CFA-enhanced (p1, p2)arthritis arthritis arthritis + estradiolSwelling 4126 15% 34/51 67% 4/19 21%Ankle widthb 0.10±0.04 (17) 0.30±0.05 (38) 0.08±0.05 (19) 1.000, 0.012increaseSubsynovial 0.70±0.13 (23) 1.16±0.09 (51) 0.47±0.16 (19) 0.789, 0.015inflammationSynovial 0.65±0.12 (23) 1.27±0.09 (51) 0.89±0.07 (19) 0.518, 0.000hyperplasiaPannus 0.22±0.09 (23) 0.63±0. 10 (51) 0.05±0.05 (19) 1.000, 0.019formation andcartilage erosionBone 0. 17±0.08 (23) 0.55±0. 10 (51) 0.26±0. 10 (19) 1.000, 0.041destructionOverall score 1.74±0.29 (23) 3.61±0.31 (51) 1.68±0.28 (19) 1.000, 0.000of histologya Data are presented as the mean ± s.e.m. of the measurement at the end of the experiment.The number of animals is listed in parenthesis.b Means maximum increase in bimaleolar ankle width during the experimentPi ANOVA and Bonferroni multiple comparison test results at p significance level between spontaneousarthritis and CFA-enhanced arthritis with estracliol treatment groupsP2 ANOVA and Bonferroni multiple comparison test results at p significance level between CFAenhanced arthritis and CFA-enhanced arthritis with estradiol treatment groups167Table 4.3 Evaluation of the effect ofesfradiol on systemic parameters in MPeL-prmiceAnimala Spontaneous CFA-enhanced CFA-enhanced (p1, p2)arthritis arthritis arthritis + estradiolMortality 7/32 22% 16/69 23% 10/29 34%Weight (g) 40±1.0 (23) 41±0.8 (39) 42±1.4 (19) 0.824, 1.000Lymph- 3.13±0.24 (23) 2.77±0.22 (39) 1.79±0.27 (19) 0.003, 0.020adenopathybProteinuriab 1.23±0.23 (23) 1.05±0.15 (39) 2.69±1.41 (19) 0.395, 0.183Kidney 6.75±0.93 (12) 8.38±0.75 (21) 7.27±0.95 (21) 1.000, 1.000histologyba Data are presented as the mean ± s.e.m. of the measurement at the end of the experiment.The number of animals is in parenthesis.b See Materials and Methods.Pi ANOVA and Bonferroni multiple comparison test results at p significance level betweenspontaneous arthritis and CFA-enhanced arthritis with estracliol treatment groupsP2 ANOVA and Bonferroni multiple comparison test results at p significance level betweenCPA-enhanced arthritis and CPA-enhanced arthritis with estradiol treatment groups168IlL DISCUSSIONThe hormonal changes associated with pregnancy affect even the healthymother, and can seriously alter the course of autoinirnune diseases.Pregnancy causes flare up in active SLE. While it ameliorates RA duringgestation, it exacerbates the disease after delivery (Hazes, 1991). Pregnancyand lactation also increase the dangers of the side effects that accompany thestandard treatment in these diseases. Therefore, it is deemed imperative todevelop animal models which allow the study of the mechanisms involved inthese complex situations and test treatment modalities which could be usedwithout endangering the health of both the mother and the fetus or breast-fednewborn.The most frequently used model for studying the effect of pregnancy inarthritis is collagen-induced arthritis. The pre-partum remission and postpartum exacerbation, as well as the acceleration of the onset of the disease bypregnancy was described in a study that was conducted in a small number ofDBAI1 mice with type II collagen induced arthritis (Waites et al., 1987).Recently it was reported that pregnancy prevented pristane induced arthritis(Thompson et al., 1992) and proteoglycan-induced arthritis in mice (Buzas etaL, 1993). As there are differences in both strain and in model in the reactionto pregnancy, it is important to investigate the effect of pregnancy inalternative model systems.The MRL-lpr mouse strain is the only described spontaneous SLE modelwhich also develops arthritis. SLE symptoms and some early169histopathological signs of arthritis are already present in two month oldanimals (O’Sullivan et al., 1985).The present study shows that this spontaneous form of arthritis inanimals mated at 10 weeks of age is significantly exacerbated during the postpartum period. The reason that this finding was not previously noticed ismost likely due to the fact that the MRL-lpr mice are generally bred at ayounger age to avoid the severity of the SLE, thereby dampening the intensityof the flare up of arthritis. This age dependence, reflecting the time requiredto reach a certain stage of underlying disease, was also noted in this strainduring the development of the adjuvant-enhanced arthritis model (Chapter 3).It is of interest that in both cases approximately 2/3 of the animals developedexacerbated arthritis. This indicates the presence of an underlyingcomponent of the disease at this age that can be accelerated by certain factors.The flare-up occurred as early as five days after delivery with a mean onset of13 days, and resulted in a massive subsynovial inflammatory infiltrate.There was, however, no considerable cartilage or bone destruction by day 30post-partum. We are planning to examine the longer-term effect of the flareup on the severity of arthritis histopathology in post-partum versus non-mated female mice.The ameliorating effect of pregnancy on RA has been attributed tophysiological factors, including estrogen, progesterone, prolactin (Berczi etal., 1982; Nagy et al., 1991), pregnancy associated proteins asimmunosuppressors (cz-2 glycoprotein), suppression of cell mediated170immunity, and alterations in the glycosylation of IgG (Rook et at., ) acting aspossible mediators (Griffin, 1990). In post-menopausal women estrogentreatment of active RA increased bone mineral density (van den Brink et at.,1993). The effect of estrogen containing contraceptives has been described aswell (Spector et at., 1990). However, there is no clear evidence whether the useof oral contraceptives provides protection against RA. This inconclusivenessis probably due to differences in patient selection, as European hospital-basedstudies contradicted US population-based findings.Estrogen is a probable factor in the controversial effect of the pill.Although it has been previously reported that female sex hormones have onlya minimal therapeutical effect on RA (Bijisma et at., 1992), the latestcontrolled study showed that adjunct therapy with estrogen, when applied topostmenopausal RA, can be effective if the resulting serum estrogen levelsare adequate (Hall et at., 1994) It is noteworthy that most of the pills containboth estrogen and progesterone (Hazes et at., 1991).Holmdahl et at. proposed that the remission of RA during pregnancy isprimarily due to estrogen, and that progesterone might potentiate this event.It is clear that progesterone alone had no effect in preventing type II collagen-induced arthritis in mice (Jansson et at., 1989). Mattsson et at. suggested thatpost-partuna flare-up is due to both the fall in the elevated pregnancy steroidlevels and the surge in prolactin release after delivery (Mattsson et at., 1991).Holmdahl et at. described the role of estrogen in RA prevention as model andstrain dependent (1989). Schlaghecke et at. reported the preventive effect ofestrogen on adjuvant arthritis in rats (1989).171The JvIRL-lpr arthritis models could serve as alternatives to the above models,since there is a need to study the described model and strain dependency ofthe estrogen effect. Here we report the preventive effect physiological doses ofestradiol have on the post-partum flare-up in the spontaneous arthritis andthe effect pharmacological doses have on the adjuvant-enhanced arthritis ofMRL-lpr mice.These data extend and support the concept of estrogen influence in RA andprovides a new spontaneous model for studying post-partum flare-up.Further investigations are needed to study the possible involvement of otherfactors, and to develop a strategy for safe therapeutic intervention.1721V. SUMMARYSixty-eight percent of female MRL-lpr mice developed a post-partumexacerbation of their mild spontaneous arthritis within thirty days ofparturition. The flare-up started between 5 and 15 days after delivery.Histologically it was characterized by a significant increase of subsynovialinflammation and synovial hyperplasia. However no changes in the level ofcartilage and bone erosion were observed. Immunohistologically, markedsubsynovial and frequent synovial staining of MHC class IT bearing cells wasnoted, along with the sporadic presence of TL-2 receptor bearing cells in thesubsynovium.Injection of physiological amounts (0.08 mg/kg) of estradiol on days 2, 3,9, 15 and 20 post-partum reduced the flare-up to 23% of the animals in whichthe onset was delayed.Administration of pharmacological amounts (0.4 mg/kg/day for 2 weeksafter complete Freund’s adjuvant injection) prevented adjuvant-enhancedarthritis, reducing the incidence from 67% to 21%.Deleterious changes in the underlying systemic lupus erythematosus(SLE), as demonstrated by proteinuria and mortality rate increases, werecaused only by estradiol in pharmacological amounts.These results indicate that the M.RL-lpr mice might serve as a suitablemodel for studying post-partum flare-up of arthritis in SLE and rheumatoidarthritis, provide an opportunity to study the complexity of the effects ofpregnancy on autoimmune diseases, and also provide further evidence for theinvolvement of estrogen in arthritis.173CHAPTER FiVEPhotodynamic Therapy and its Comparison with OtherImmunomodulatory Treatments ofAdjuvant-EnbancedArthritis inMRL-lpr Mice174I. INTRODUCTIONSLE has been treated or prevented in MRL-lpr mice by a number ofexperimental regimens: total lymphoid irradiation (Theofilopoulos et at.,1980), neonatal thymectomy (Steinberg et at., 1980), cyclosporin A (Berden etal., 1986; Mountz et at., 1987), and monoclonal antibody therapy against thepan T-cell marker Thyl (Seaman et aL, 1983; Wofsy et at., 1985), or the Thelper cell determinant CD4 (Santoro et at., 1987; Santoro et at., 1988;O’Sullivan et at., 1991; Gilkeson et at., 1992). However, similar treatments forthe spontaneous arthritis of these mice have been less-frequently studied(Rordorf-Adam et at., 1986; Rordorf et at., 1987) as the mice have a 50%mortality rate at 5.5 months and do not develop a severe form of the disease.In a new form of arthritis therapy Holoshitz et at., demonstrated thatinactivated T-cell clones are capable of ameliorating transfer-adjuvantarthritis (Holoshitz et at., 1983). A modification of this technique utilizingphotoinactivated autoimmune splenocytes has been reported by Berger et at.,(Berger et at., 1990). In this latter instance 8-methoxy psoralen andultraviolet A light (8 MOP-UVA or PUVA)- treated syngeneic lymphocytesprevented autoimmune disease after transfer to MRL-tpr mice.Photoactivated 8-MOP, as well as its effect of crosslinking DNA, is thought tofix the T-cell receptor on the cell surface making its idiotype moreimmunogenic. The host will mount an immunologic response to thephotomoclified T cells, thus the “fixed” T cell clones provide an effective“vaccine” that functions by controlling the effector clones responsible for thedisease (Edelson, 1988; Edelson, 1989).175Photodynamic (oxygen dependent) therapy (PDT) involves intravenousinjection of photosensitizing drugs, which have been shown to accumulateselectively in abnormal or neoplastic tissues (Proflo, 1990; Dougherty et at.,1992), as well as in immunologically activated cells in the circulation (Northet at., 1993). The photosensitizers at these sites can then be activated byirradiation with light of the appropriate excitation wavelength. Subsequentinteraction of the excited state photosensitizer with aqueous molecular oxygenresults in the generation of highly cytotoxic, short-lived oxidative species suchas singlet oxygen and oxygen free radicals. Cell death occurs predominantlyas a result of lipid peroxidation and the disruption of membranous structuresand related functions. Membrane associated and intracellular proteinfunction dependent on tertiary structure may also be lost as a result ofoxidation of individual amino acids at key sites (Henderson et at., 1992; Pass,1993).There are several groups of photosensitizers being clinically tested onvarious diseases. The most well-known is 8-methoxy psoralen (8-MOP), usedin the treatment of psoriasis. Psoralens can penetrate cellular membranesand react with nucleic acids, forming covalent cross-links between theopposite strands of DNA (Gasparro, 1988). Other photosensitizers includehypericin, phthalocyanines, merocyanine 540, and porphyrins. Their maintarget is the plasma membrane (North et at., 1993). The porphyrins include,among other molecules, hematophorphyrin derivative Photofrmn® andbenzoporphyrin derivative - monoacid ring A (BPD).176BPD (Figure is a relatively novel, highly hydrophobic, chiorin-likesecond generation photosensitiser (Richter et al., 1988; Richter et at., 1990). Itis eliminated rapidly from tissues (Richter et at., 1990). The rapiddisappearance reduces the duration of normal skin photosensitivity in mice toabout 24 hrs with the possibility of severe skin reactions in the case of highintensity light exposure after about 3 hrs (Richter et at., 1991). The maximumdetectable cutaneous photosensitivity is 3 to 5 days depending on BPD dose inhumans (Lui et at., 1993). This is a transient sensitivity compared to firstgeneration photosensitizers, such as Photofrin®, which cause prolonged (4-6weeks in mice, 6-8 weeks in humans) photosensitivity. It can develop as anacute inflammatory cascade reaction: interleukin 1 triggered mast celldegranulation, followed by vasodilation and chemotactic neutrophil andeosinophil infiltration (Kamide et at., 1984; Kupper et at., 1987; He et at., 1989).BPD’s formulation in fast release liposomes with a mean particle sizebelow 200 nm ensures its rapid transfer predominantly to plasmalipoproteins following injection (Allison et at., 1990). The low densitylipoproteins (LDL) are natural porphyrin-carriers in the circulation (Jori etat., 1984), and the role of the LDL cell surface receptor in porphyrin uptakehas been demonstrated in in vitro models (Mazière et at., 1991). Cells withhigher density and turnover of LDL receptors, such as rapidly proliferating(cancer) cells or the endothelium in tumour vasculature, are likelyresponsible for the preferential uptake of BPD in tissues with a high mitoticindex. Such tissues often involve a high degree of neovascularisation.177Figure 5.1 Benzoporphyrin derivative - monoacid ring A (BPD)a structure of BPDb absorption spectrum of BPD (), oxyhemoglobin ( .), and Photofrin® ( )178aH3ChCH3R1 = R2 = CO2Me__RR3 (CH)00MeR1 H C R(0H2)00H‘I13Cb4-— BPD—MAin et.hno1—— OxyhcmogiobinPholofrthIinPBS2-rcDit’Jl ‘•l‘V/ \M““‘ii.•.-—k__• 500 700ree1eagt.h (nm)179PDT causes the disruption of blood supply via microthrombosis andmicrovascular coagulation at these sites, resulting in the ischemic necrosis ofhyperproliferative tissues (Fingar et al., 1993).Studies in our collaborators’ laboratories have shown that in comparisonto normal murine splenocytes, mitogen-activated spleen cells accumulate upto 4 to 5-fold higher concentrations of BPD within 30 minutes of incubation(Richter et al., 1993a). Differential uptake of BPD by these cells represents abase for selective treatment, as it is known that subthreshold concentrationsof BPD do not cause irreparable damage to cells during light treatment. Thiswould suggest that immunologically activated lymphocytes (and possibly cellsin mitotic cycle vs resting cells) would be more likely to be eliminated uponexposure to light, and this has been established in in vitro cytotoxic studiesusing splenocytes (ongoing studies). Experiments dealing with viralinactivation of HIV infected blood have also shown that activated humanleucocytes expressing high levels of HLA-DR and IL-2R (interleukin-2receptor) can be selectively eradicated from a heterogeneous mixture bytreatment with BPD and light (North et al., 1993). PDT therefore tends toselectively destroy activated cells.BPD has a strong absorption band at 690 nm, a spectral region whichallows a minimal attenuation by haemoglobin, and 10 to 20 mm tissuepenetration. This is 30% deeper than the wavelength used for first generationphotosensitizers (North et al., 1992) (Figure Between 10 and 20% of thetotal blood volume is in the circulation through the cutaneousmicrovasculature at any given time and there is complete circulation every1804-6 minutes. This enables BPD associated with leucocytes in the circulation tobe photoactivated transcutaneously.A treatment window (i.e. conditions under which selective damage totarget cells can be achieved) for the exposure of animals to light haspreviously been established at one hour following intravenous administrationof BPD at doses up to 1.0 mg/kg. This allows sufficient time for differentialdistribution of the sensitiser between tissues, but skin photosensitivity isavoided as the accumulation of BPD in the skin is relatively slow (reachingmaximum levels 3-5 hours post injection (Richter et al., 1993a)), andinsignificant during the time frame in question (Richter et al., 1993b). Duringthis therapeutic window 75 to 85% (light-activated blood concentration vs darkcontrol) of the BPD in the vasculature can be activated by exposing shavedmice to light without causing measurable adverse reactions. Activation ofBPD results in photodegradation of the molecule, with the loss of itsfluorescence. Therefore the extent of photodegradation can be monitored bythe loss of serum fluorescence, and it is proportional to exposure to light (Jamet al., 1993). In unshaved animals this level of photobleaching is 50-60%(Chowdhary et al, submitted). This indicates that a significant PDT effect inthe vasculature can be achieved even with limited light penetration.BPD is currently being used in clinical and predinical trials employingphotodynamic therapy for treatment of malignant skin lesions (Liii et al.,1993), psoriasis (Hruza et al., 1993), atherosclerosis (Hsiang et aL, 1993), andleukaemia- with bone marrow purging (Jamieson et al., 1993).181TI. RESULTSA. Preventive photodyrnimic therapy of adjuvant enhanced arthritis in MRLlpr miceClinical signs of spontaneous arthritis in MRL-lpr mice as a result oftheir autoimmune condition developed in about 18% (5/28) of the animals overthe 30- day course of the experiment (Figure 5.2a). Around 70% (28/40) ofadjuvant-injected animals developed erythema and swelling of the hindllimbsduring the same period. Visual assessment of arthritis correlated withquantitative mean ankle-width measurements, which also showed severity tobe highest in the adjuvant-treated group that received no PDT (Figure 5.2b).This was confirmed by the severe histological manifestations of arthritisdisplayed by this experimental group (Table 5.1). Adjuvant-injected mice thatreceived photodynamic treatments at 10 day intervals throughout the courseof the experiment displayed considerably milder disease symptoms even incomparison to unmanipulated control mice. There was a reduced incidenceof visible swelling in animals receiving PDT, as well as a delay in the onset ofdisease. Furthermore, histological examination showed that although somesubsynovial infiltration remained along with considerable synovialhyperplasia, inflammation-mediated structural damage to cartilage and bonein the joint was prevented by periodic treatment with PDT (Figure 5.3). Whenhistological data was partitioned according to the sex of the experimentalanimals (Table 5.1), it revealed that in all three experimental groups, theoverall histological score was somewhat higher in females.182There were no statistically significant serological differences betweenthe three experimental groups at the termination of the experiment, howeverthe adjuvant-injected group that did not receive PDT showed an elevatedantibody reaction against M. Tuberculosis, type II collagen, and DNA, and ahigher total immunoglobulin level (Table 5.2).Similarly, the splenocyte response to mitogens concavalin A andlypopolysaccharides (LPS) was unchanged and low compared to the MRL-+control (Figure 5.4).183Figure 5.2 a Effect ofPDT treatment on the incidence of CFA-enhancedarthritis in MRL-lpr miceAnimals in the experimental group ( o ) received PDT at days 0, 10 and 20following adjuvant injection. Controls included a group of uninjectedanimals ( ), and an adjuvant-injected group that received no PDT (.).b Changes in bimaleolar rnk1e width in PDT treated ( o ) and untreated (.)adjuvant injected animals and non-injected control animals (•).ANOVA and Bonferroni multiple comparison test showed a significantdifference between the mean values of CFA-enhanced arthritis versusspontaneous and PDT-treated CFA-enhanced arthritis groups at p<0.O5 level,starting at day 5 of the experiment. The left and right foot measurementswere combined and the day 0 data were subtracted from the subsequentdatapoints for each animal.Number of animals/group: (.) N=28, ( • ) N=40, Co ) N=33. Error barsrepresent standard error of mean.184Increaseinbimaleolarwidth(mm)Animalsdisplayingswellinganderythema(%)1’0)000000-a0•-I-a-. 0-l r.3mF’)0-I C.)C.) 000F’)0•C.)0Table 5.1 Histological evaluation of PDT in CFA-enhanced arthritisAnimals Sex Spontaneous CFA-enhanced PDT treated CFA- P1, P2arthritis arthritis enhanced arthritisnumber F 14 15M 11 19 182 39 33Subsynovial F 0.79±0.19 1.30±0.15 0.67±0.13inflammationa M 0.82±0.26 1.32±0.13 0.50±0.122 0.80±0.15 (60) 1.31±0.10 (92) 0.58±0.09 (58) 0.006 0.000Synovial F 0.79±0.15 1.25±0.16 0.87±0.13hyperplasiaa M 0.73±0.19 0.84±0.11 0.89±0. 162 0.76±0. 12 (68) 1.05±0.10(82) 0.88±0. 10 (76) 0.208 0.724Cartilage erosion, F 0.21±0.11 0.60±0.13 0.07±0.07pannus formationa M 0.00±0.00 0.42±0. 14 0.00±0.002 0.12±0.07 (12) 0.51±0.10 (46) 0.03±0.03 (3) 0.002 0.000Bone destructiona F 0.21±0.11 0.50±0.15 0.00±0.00M 0. 18±0. 12 0.47±0. 18 0.06±0.062 0.20±0.08 (20) 0.49±0.12 (36) 0.03±0.03 (3) 0.094 0.001Overall histological F 2.00±0.38 3.65±0.48 1.73±0.29scorea M 1.73±0.45 3.05±0.39 1.33±0.261.88±0.29 (88) 3.36±0.31 (100) 1.52±0. 18 (88) 0.001 0.000a Mean histological score ± the corresponding group, based on data from 2 separate experiments.Numbers in parenthesis indicate % of the animals that developed listed pathologyPi ANOVA and Bonferroni multiple comparison test results at p significance level between spontaneous and CFAenhanced arthritis groupsp2 ANOVA and Bonferroni multiple comparison test results at p significance level between CFA-enhanced arthritisand PDT treated CFA-enhanced arthritis groups186Figure 5.3 The preventive eflèct ofPDT treatment on the histopathology ofadjuvantenhanced arthritis in MRL-lpr micea Severe inflammation and destruction of a heel bone in a CFA injectedmouse 30 days after injection (H&E, xllO)b PDT treatment prevented severe histopathological changesA tarsal area shows only mild hyperplasia of the synovial lining(H&E x55)(as, articular surface; b, bone; si, synovial lining)187•!••::.••‘p”,•‘n)••••‘Cl),;e1-•‘—;:•-Table 5.2 Serological assessment of the effect ofPDT on adjuvant-enhancedarthritis in MRJApr miceAntibody CFA injected CFAPDT treated UnmanipulatedAnti-M.Tuberculosis 0.316± 0.021 (40) 0.295± 0.041 (14) 0.274± 0.020 (21)Anti-chondroitin sulphate 0.064 ± 0.017 (25) 0.075 ± 0.032 (13) 0.054 ± 0.016 (16)Anti-type I collagen 0.110± 0.012 (24) 0.148 ± 0.020 (12) 0.093 ± 0.011 (16)Anti-type II collagen 0.101±0.019 (25) 0.084±0.019 (13) 0.089±0.016 (16)Anti-DNA 0.315 ± 0.027 (25) 0.249 ± 0.038 (13) 0.283 ± 0.043 (16)Total IgG 0.037 ± 0.006 (25) 0.055 ± 0.006 (13) 0. 040 ± 0.009 (16)Rheumatoid factor 0.065 ± 0.010 (25) 0.073 ± 0.018 (13) 0.064 ± 0.014 (16)Total Immunoglobulin 1.14-4 ± 0.132 (25) 0.765 ± 0.093 (13) 0.898 ± 0.134 (16)Values represent mean ± s.e.m. absorbance readings of the enzymatic product of alkalinephosphatase.Parentheses indicate the number of samples analyzed in each group.ANOVA and Bonferroni multiple comparison test results at p<0.05 level showed nosignificant differences between the CFA injected and the other two groups.189Figure 5.4 Mitogen response ofMRL-lpr spleen cells after PDTAt the termination of the experiment splenocytes pooled from tenanimals per group were activated with concavalin A (a) or LPS (b). Cellproliferation was assessed by the MTT assay. Experimental groups includedMRL-+ positive control (A), MRL-lpr spontaneous arthritis ( 0 ), CFAinjected MRL-lpr ( A), CFA injected and PDT treated ( •) animals.190Opticaldensity(595nm)0 0pOpticalDensity(595nm)p00Fo0 C 30•0 0 0 C 3I’,0 Co 0 0 (J1 00 a.0•f0) 0B. Comparison ofphotodynamic therapy and other immunomodulatorytreatments of adjuvant enhanced arthritis in MRL-lpr miceWe evaluated the effect of several treatment methods representingdifferent approaches. The NSAIDs were represented by indomethacin,general immunosuppression by whole body irradiation, and T cell specificimmunosuppression by cyclosporin A. The effect of these treatmentmodalities were compared to the experimental photodynamic therapy, whichis a new specific therapy directed against activated cells.Effect of treatments on clinical signs ofCFA-enhanced arthritisMRL-lpr mice, randomly assigned into a treatment group, weremonitored for 30 days for clinical signs of arthritis (Figure 5.5). Seventypercent (46/66) of CFA injected control animals developed erythema andswelling during the experiment. The incidence of swelling was reduced inthe treatment groups to the level found with the non-injected control animals(18%, 5/28), where the arthritis is the result of their autoimmune condition.In the indomethacin group a somewhat lower percentage of animalsdeveloped the swelling (8%, 1/12). Visual assessment of arthritis correlatedwith quantitative mean ankle width measurements: all treatment groups hada significantly reduced maximal increase in bimaleolar width (Figure 5.6).192Effect of treatments on histopathology of thejointsThirty days after adjuvant injection histological evaluation of the jointswas carried out for the various immunomodulatory treatments (Figure 5.7).The analysis showed that all treatment groups exhibited significantly lesssubsynovial inflammation than the CFA injected control. However,exacerbation of the synovial hyperplasia was notably prevented only by CsA.With the exception of indomethacin all treatments effectively preventedsignificant pannus formation and cartilage erosion, which is characteristic ofthe CFA-enhanced arthritis in MRL-lpr mice. Bone destruction wasprevented only by WBI and PDT.193Figure 5.5 Animals displaying swelling and erythema of hindpawsA, Spontaneous (n=28); B, CFA (n=66); C, CFA-i-Indomethacin (n=12);D, CFA-i-CsA (n=21); E, CFA-i-WBI (n=34); F, CFA+PDT (n=33).194Animals(%)F’)0000w C, m -I,000Figure 5.6 Bimaleolar ankle width 30 days after CFA uijectionA, Spontaneous (n=28); B, CFA (n=66); C, CFA+Tndomethacin (n=1O);D, CFA+CsA (n=12); E, CFA÷WBI (n=34); F, CFA-i-PDT (n=33).* means the corresponding value is significantly different from the CFAgroup’s value (p<O.Ol ANOVA and Bonferroni multiple comparison test).1960.4EE0.3Q0.2wa, 0.1•0.0A* B C D* E F*197Figure 5.7 Ilistopathological evaluation of thejoints 30 days after CFAiijectionThe histological scores were determined as described in the materials andmethods.A, Spontaneous (n=18); B, CFA (n=58); C, CFA+Tndomethacin (n=12);D, CFA+CsA (n=21); E, CFA-f-WBI (n=32); F, CFA÷PDT (n=27).* means the corresponding value is significantly different from the CFAgroup’s value (p<O.Ol ANOVA and Bonferroni multiple comparison test).198I00(0Cu00)00(0x4320SubsynovialinflammationA* B C* D* E* F* A* B C D* E F A* B C D* E* F* A B C D E* F* A* B C D* E* F*SynovialhyperplasiaPannus formation Bone destruction Overall scorecartilage erosion199The total histological scores summarize the overall prophylactic effectsof the described therapeutic modalities. All treatments, except indomethacin,significantly prevented the overall exacerbation of the disease; indomethacinexerted a significant effect only on subsynovial inflammation.Effect of treatments on systemic disease parametersAs spontaneous arthritis is an integral component of the underlyingSLE-like disease, it was of interest to evaluate the systemic effects of thetreatments. The survival of mice from the different treatment groups wascompared with the CFA-injected control (Figure 5.8a). The mortality rate ofthe groups was approximately 22% (50/230), which is characteristic of thismouse strain at 4 months of age (Dixon, 1987).The weight gain during the 30 days experimental period was slightlyreduced in the WBI group but the differences were not significant (Figure5.8b). Similarly, proteinuria measured at day 30 after CFA injection was notsignificantly different, although the value for the indomethacin group wasapparently elevated (Figure 5.8c). This might reflect deteriorating kidneycondition following indomethacin treatment that is not evident at thehistological level in individual animals.It was noteworthy that the lymphadenopathy was significantlyprevented only by WBI (Figure 5.8d).Histopathological evaluation of the kidneys indicated there were nosignificant changes between the various treated animals and the untreatedcontrols (Figure 5.9).200Figure 5.8 Evaluation ofsystemic changes 30 days after CFA injectiona Survival; b Weight-gain; c Proteinuria; d LymphadenopathyA, Spontaneous (n=28); B, CFA (n=66); C, CFA+Indomethacin (n=1O);D, CFA+CsA (n=12); E, CFA-i-WBI (n=34); F, CFA÷PDT (n=33).* means the corresponding value is significantly different from the CFAgroup’s (B) value (p<O.O1 ANOVA and Bonferroni multiple comparison test).201r’z0)0)0000000I•I•I•I•IWeight-gain(g)0Animals(%)wIIC) mwjLC) m-n-I-IArbitraryscore0.-LConcentration(gIL)o-Lo00)C,w C-) C m *-IlFigure 5.9 Histological evaluation of the kidneysThe histological scores were established as described in the Materials andMethods.A, Spontaneous (n=16); B, CFA (n=45); C, CFA+Indomethacin (n=1O);D, CFA+CsA (n=1O); E, CFA÷WBI (n=26); F, CFA+PDT (n=18).* means the corresponding value is significantly different from the CFAgroup’s (B) value (p<O.O5 ANOVA and Bonferroni multiple comparison test).CuL.0C)0CuC)0001086420Mesangial cell Crescent Interstitial cellular Vasculitis Overall scoreproliferation formation infiltrationIlL DISCUSSIONTherapeutic interventions in rheumatoid arthritis are primarily eitheranti-inflammatory or immunosuppressive; the latter designed to target theCD4-positive regulatory T cells that are central to the disease process.Representative pharmacologic and physical agents that have beensuccessfully used in either animal models or the human disorder areindomethacin, prednisolone (steroid), cyclophosphamide, cyclosporin A,thoracic duct drainage (Paulus et aL, 1977), lymphocytapheresis (Emery etal., 1986), and whole body irradiation (total lymphoid irradiation (Gaston etal., 1988a)). Unfortunately these treatment procedures have systemic effectson cell populations other than those directly involved in the pathogenicmechanism(s), or they cannot be employed because of the adverse effect onpatients. Monoclonal antibodies against pan T cell antigens, CD4, and T cellactivation antigens are under clinical trials. However, they show mostlytransient and partial effects on clinical arthritic manifestations, but depletethe system of the targeted lymphocyte subsets for a prolonged period,which isan unacceptable side effect. Furthermore, there are problems with an antiimmunoglobulin response against the administered mAbs.The conventional therapeutic modalities used in these experiments allexhibited beneficial effects on animals receiving adjuvant enhancement.Indomethacin (a nonsteroidal anti-inflammatory agent (NSAID) whichinhibits the synthesis of prostaglandins in inflamed tissues), while itsignificantly prevented swelling and subsynovial inflammation, increased theproteinuria and failed to prevent cartilage and bone destruction. CsA, a T cell206specific immunosuppressive anti-rheumatic drug, selectively inhibits Thelper/inducer function (Kahan, 1989). It has been effectively used in RAtherapy, but its possible side effects of nephrotoxicity, tumorogenecity,infections, as well as its expense preclude it from general use (McCune et al.,1991). Sublethal whole body irradiation, or the more frequently used totallymphoid irradiation therapy, has a nonspecific immunosuppressive effecteliminating dividing cells. Again its practical application in rheumatology islimited because of its side effects. CsA and WBI inhibited both the clinicalinduction of disease and the histopathology associated with the joints.Moreover, the overall toxicity of CsA and WBI as reflected by renal functionand histology, survival, and weight gain was minimal in our short termstudy. However, these latter two immunomodulators are highly dosedependent; a single dose of whole body irradiation (1.5 Gy) not only failed toprevent the arthritis (57%, 8/14 incidence of swelling), but in some cases,paradoxically, it exacerbated the disease, similar to a one-time response atday 0 treatment of CsA (67%, 14/21 incidence of swelling).Transdermal photodynamic therapy with BPD produced a significantalteration in arthritis development, in that it prevented the clinical andhistopathological signs of adjuvant-enhanced arthritis, being especiallyeffective in preventing the more severe stages of disease (paunus formation,cartilage erosion, and bone destruction). Furthermore the animals treated byPDT showed no signs of skin photosensitivity following treatment. Thetherapeutic dose of BPD had no effect on survival, weight gain, proteinuria,histology of the kidney, or general immune function (mitogen responses)207(Figures 5.2, 5.8 and 5.9). Treated animals were comparable to controlsduring a seven-day follow up of haematological evaluation (Chowdhary et al.,,submitted). The evaluation included the following parameters: liver andserum enzyme (alkaline phosphatase, aspartate transaminase, liver alaninetransaminase, and lactate dehydrogenase) levels - which would indicatefunction of main organs involved in systemic detoxification-, serum bloodurea nitrogen, and creatinine levels, splenocyte clonogenecity andhaematological parameters, such as haemoglobin level, haematocrit,erythrocyte and leucocyte counts, and erythrocyte membrane integrity tests.Histological examination of the major organs (liver, spleen, kidney), whichmight have received exposure to light, showed no detectable structuralchanges or changes in weight.As pharmacological studies did not show any significant changes innumbers of circulating leucocytes following PDT, it would appear that theleucocyte population targeted by this treatment is small. It is possible that theobserved preventive effect of transcutaneous PDT could be due to theelimination of the clones of activated M. Tuberculosis responsive lymphocytesin the circulation, which would prevent their localization and effectorfunction in the joints. As a less probable mechanism, localized exposure ofthe paws to light might also result in the destruction of leucocytes,particularly macrophages present in the potentially arthritic joint area. Thismight be important in limiting chemotactic attraction of further leucocytes tothe area which could prevent the unregulated inflammatory processesleading to structural damage to the joint.208Although this is the first animal model to show effectiveness oftranscutaneous PDT, it could have potential application in a number ofhuman autoimmune diseases which share a number of characteristics withthis species. Rheumatoid arthritis patients, for example, display similarhyporesponsiveness to antigens, anemia, vascular thrombi, etc., and benefitfrom the selective reduction of the elevated proportion of activated self-reactiveleucocytes in the circulation, without the general immunosuppressive sideeffects of therapies in common use. Some improvement has been seen inrheumatoid arthritis patients following the application of leukapheresiswhich involves physical, non-specific removal of leucocytes from thecirculation (Hahn et al., 1993). Similarly, whole body photopheresis usingpsoralens and uv light has been shown to be beneficial in a number ofautoimmune disorders including rheumatoid arthritis (Malawista et al.,1991). Photopheresis is an extracorporeal blood treatment with light via ashunt taking the blood from the circulation to a light emitting device andback. The mechanism of action is thought to involve replenishment of thecirculation with leucocytes of lower activation status and self reactivity.The non-invasive nature of the transcutaneous PDT, especially as itallows for the treatment of a larger volume of blood in a shorter period of time,could be ideal for multiple treatments, particularly at the deeper penetratinglong wavelength used for photoactivation of BPD. The lack of toxicity of bothBPD and red light and the possibility to locally treat areas of inflammationmakes PDT attractive in comparison to other described immunosuppressivetherapies in use for the control of autoimmune diseases.209These findings indicate that the photodynamic therapy can inhibit thedevelopment of adjuvant enhanced arthritis in MRL-lpr mice with similareffectiveness as conventional treatments, but without their deleterious sideeffects. The application of the photosensitizer BPD and light to eliminateactivated cells responsible for the inflammatory reaction at the arthritic siteor within the circulation may have significant clinical implications in thetreatment of human rheumatoid arthritis.210W. SUMMARYAlthough numerous experimental immunomodulatory regimens havebeen reported to be effective in the treatment of rheumatoid arthritis, they alsoproduce undesirable side effects. An alternative specific modality of localizedtreatment is photodynamic therapy (PDT). One group of CFA-injected MRLlpr mice received transcutaneous photodynamic therapy at days 0, 10, and 20,following the CFA injection. The other CPA-injected groups were treatedwith 1 mg/kg/day indomethacin, 40 mg/kg/day cyclosporin A (CsA), or treatedwith 3 Gy sublethal whole body irradiation (WBI). The development ofswelling was monitored for 1 month, at which time proteinuria,lymphadenopathy, and the histopathology of the joints and kidneys wereassessed. The results demonstrated that PDT and the conventionaltreatments significantly ameliorated swelling of the hindlimbs from 70% inthe untreated CFA injected animals to below the 19% level characteristic ofthe unmanipulated control. Histological examination showed a reduction inpannus formation, and cartilage and bone destruction. PDT did not affect thesurvival rate, lymphoproliferation, or proteinuria of the treated animals.However, indomethacin increased proteinuria, and was less effective inpreventing cartilage and bone destruction. Furthermore, lower doses of CsAand WBI exacerbated arthritis activity. These results indicate thatphotodynamic therapy can inhibit the development of adjuvant enhancedarthritis in MRL-lpr mice with similar effectiveness as the conventionaltreatments, but without their negative side effects.211ggUO!SSflOS!UTlLT9tT9A major difficulty in understanding the etiology and pathogeneticmechanisms of rheumatoid arthritis has been the lack of a suitable animalmodel. The main problem in using the existing animal models is that theyare experimentally induced, self-limiting, and lack many of the pathologicalfeatures seen in human disease. On the other hand, the spontaneousarthritis models are not suitable for therapeutic trials, because of the lowfrequency, and moderate pathology of their disease.Based on the MRL-lpr spontaneous arthritis model, in a series ofexperiments, a new murine arthritis model was developed, that is morereliable and practical for therapeutic evaluations.The MRL-lpr mouse strain develops an early autoimmune disease, thatshares similarities with both human systemic lupus erythematosus andrheumatoid arthritis. This autoimmune condition is the result of complexgenetic interactions involving participation of both the lpr/Fas gene andbackground genes. Inter-colony differences have been reported, and duringthe last five years the arthritic component of the MRL-lpr autoimmunedisease has decreased. We characterized both the systemic and arthriticdisease components in our colony and concluded that there are no noticeablechanges in the severity, onset, and symptoms of the lupus-like disease, butthe spontaneous arthritis is less severe than originally described. It isunlikely that this decrease is only the result of the inbreeding in our colony, asanimals, purchased directly from Jackson Laboratory (central supplier of thisstrain) showed similar histopathology. In addition several recent reportsdescribed the lack of arthritis before four month of age in these animals (see213Chapter 1). The decrease in arthritis is most likely the result of theinbreeding protocols, which give preference to maintenance of the lpr generelated lymphadenopathy and the SLE symptoms.In the second series of experiment an extensive investigation wascarried out on the effect of complete Freund’s adjuvant (CFA) on thespontaneous arthritis of MRL-lpr mice. 67-82% of mice showed early clinicalevidence of arthritis in contrast to the low percentage observed in controlanimals. Similarly, the histopathological analyses of the CFA injected miceindicated a significantly higher frequency of advanced histopathologicalalterations, characterized by cartilage erosion and pannus formation. Theimmunohistological staining detected an elevated expression of the MHCclass II activation marker both in the synovial lining and in the subsynovium,as well as a moderate presence of CD3 and CD43 surface antigen bearing Tcells in the subsynovium. This adjuvant-enhanced arthritis is initiatedearlier and exhibits an enhanced frequency and severity compared to thespontaneous arthritis seen m MRL-lpr imceIn the third series of experiments the requirement of the lpr gene andbackground MRL genes was examined on the CFA—enhanced murinearthritis. It was found that while 74% of both 7 month old MRL—+ and 3month old MRL—lpr mice displayed clinically visible arthritis, B6, B6—lpr,and 3 month old MRL—i- did not develop the condition after CFA—treatment.These observations suggest that while the lpr gene causes a more severe earlyeffect, background genes other than the lpr are more involved in the adjuvantenhanced arthritis afflicted mice.214Horn et at. (1990) reported that IL-lB enhanced the spontaneous arthritisin MRL-lpr. These findings could not be confirm in my experiments, as theinjection of human-recombinant IL-lB from a different supplier did not resultin any significant arthropathy. This discrepancy may be due to the differentpreparations of IL-i as their reported biological activity was based on differentassays. However, 10 times greater than the described dose also failed toimprove efficacy. The effect of IL-i is dose dependent as it was shown by theameliorating effect of lower doses on experimental arthritis. This can be theresult of the physiologically maintained balance in the different pro- and anti-arthritic effects of iL-i.Boissier et at. (1989) showed that type II collagen injection of MRL-lprmice did not have arthritic effect. They also found that CFA containing1 mg/mi M. tuberculosis had no arthritic effect. These reports are confirmedby our experiments: the optimal concentration of M. tuberculosis was 10mg/mi. This is likely the result of the requirement for a putative M.tuberculosis rheumatic antigen to provoke an inflammatory immuneresponse. M. tuberculosis acetone extract and D-wax component werereported being arthritogenic in rats in a much lower concentration. Thesefindings support the importance of the structural homology between M.tuberculosis and cartilage macromolecules in the induction of the disease,rather than the role of CFA as a general adjuvant.To summarize the characterization of this model, the comparison withRA, adjuvant arthritis (AA) and the spontaneous arthritis in MRL-lpr isdepicted in Table adjuvant-erihanced arthritis in MRL-lpr displays a hindlimbarthritis, with the main involvement of the tarso-metatarsal joints. Itsclinical manifestations, swelling and erythema, which occur five times morefrequently than in age matched non-injected MRL-lpr mice are prone toremissions and exacerbations during the course of the disease. The chronicarthritis progresses into movement impairment and deformity in someanimals. However, long term follow up observation is possible only in theMRL-+ strain, because of the high mortality rate of older MRL-lpr mice.Extraarticular features can not be easily assigned to the adjuvant effects, asthey are also present in non-injected mice causing SLE-related skin andgenital lesions. These features are also present in the AA rat model, butabsent in BA.MRL-lpr mice develop serological responses, resembling theautoantibody production in SLE and RA (including anti-nuclear, RF, anti-collagen antibodies). These features are retained after CFA-injection. Thereis a genetic component to the MRL-lpr arthritic condition as has been reportedfor RA and AA (although different genetic loci are involved).Histologically the disease is characterized by an extreme proliferation ofsynovial stromal fibroblast-like and mononuclear cells, and pronouncedthickening of the synovial lining and angiogenesis. Involvement of lakexpressing cells, and sporadic CD3+, and CD43+ T cell infiltration were alsonoted. lak expression correlates with the severity of the disease. Loss ofperiarticular bone can be seen due to elevated osteoclastic activity. Thepresence of inflammatory exudate in the joint cavity is less pronounced than216in RA. An erosive pannus develops in a significant portion of CFA-injectedmice, resulting in bone and cartilage destruction. Although these advanceddisease features are more pronounced compared to the spontaneous model,they are less severe than in AA. The movement impairment is the result oftendon and muscle destruction, rather than ankylosis. The fact that in ratsankylosis develops only six months to a year after adjuvant injection, and thatMRL-lpr mice have an early high mortality, prevents us from studying thelong term effects of CFA in this strain.To date the most important feature of this model is its reproducibility.Most of the induced models, such as adjuvant arthritis, collagen-arthritis,suffer from the problem of inconsistency in the arthritis incidence. For thelast two years in our laboratory, CFA injection resulted in 60 to 80% ofarthritis incidence, independently of the person administering the injection.We are eagerly awaiting to see whether the model can be established withsimilar effectiveness in other laboratories.217Table lila Comparison ofclinical arthritic manifestationsFeatures RA AA MRL-lpr MRL-1prICFACLINICALAcute and recurrent arthritisperipheral joints ++ ++ + ++Chronic deforming arthritis + + + + ÷1- +Skin lesions o + ++ ++Genital lesions o ++ + +Conjunctivitis or iritis o + + +Progressive joint disease +÷ ++ +1- +SEROLOGY + + +GENETIC BACKGROUND + + ++ ++(RA, rheumatoid arthritis; AA, adjuvant arthritis in rats, MRL-lpr,spontaneous arthritis; MRL-lpr/CFA, adjuvant-enhanced arthritis)After Pearson et al. (1962)218Table iiLlb Comparison ofhistopathologic manifestationsFeatures RA AA MRL-lpr MRL-1prICFAHISTOLOGICAcute or subacute synovitis + + + + + + +Periarthritis-t-+ ++ + ++Abscess formation o 01+ 0 +Primary mononuclear inflam. ++ ++T cell infiltratum + + + + -? +B cell infiltratum + + + + -?Inflam. exud. into joint fluid++ + +1- +Synoviocyte proliferation ++ +÷ ++ ++Synovial villus hypertrophy ++ ++ ++ ++Invasion of subchondral + + + + +1 +bone and joint space by pannusPeriostitis and osteitis + + + + + +Bursitis and (peri)tendinitis ++ ++ + ++Vasculitis ++ ÷+ ++ ++Fibrous and bony ankylosis + +- ?+(RA, rheumatoid arthritis; AA, adjuvant arthritis in rats, MRL-lpr,spontaneous arthritis; MRL-lpr/CFA, adjuvant-enhanced arthritis)After Pearson et al. (1962)219In the fourth series of experiments we investigated the enhancing effectsof pregnancy on the MRL-lpr spontaneous arthritis. Sixty-eight percent offemale MRL-lpr mice developed a post-partum exacerbation of their mildspontaneous arthritis within thirty days of parturition. The flare-up startedbetween 5 and 15 days after deliveiy. Histologically it was characterized by asignificant increase of subsynovial inflammation and synovial hyperplasia.However no changes in the level of cartilage and bone erosion were observed.Immunohistologically, marked subsynovial and frequent synovial staining ofMHC class II bearing cells was noted, along with the sporadic presence of IL-2 receptor bearing cells in the subsynovium. Administration of estradiol postpartum or post-adjuvant injection reduced the arthritis enhancement. Theseresults indicate that the MRL-lpr mice might serve as a suitable model forstudying post-partum flare-up of arthritis in SLE and rheumatoid arthritis.In the fifth series of experiments we assessed the effectiveness of arecently developed photodynamic therapy (PDT) and compared it withconventional therapies in the treatment of arthritis in the developed modelsystem. The results demonstrated that PDT can inhibit the development ofadjuvant enhanced arthritis in MRL-lpr mice with similar effectiveness asthe conventional treatments, but without their negative side effects.It is concluded that CFA-enhancement of the MRL-lpr spontaneousarthritic condition results in a reproducible model of murine arthritis, whichis useful in evaluating experimental therapeutic regimes, and PDT has apotential in arthritis therapy.220Future workThe development of a reliable and reproducible model is only thebeginning of its research life. Although the model was characterized, to anextent that it can be used for therapeutic trials, there is need for more work inthis regard.Although the irnmunohistological analysis showed the presence of Tcells in the inflamed synovium, only T cell transfer experiments will be able toclarify the extent of T cell involvement in the enhanced disease.The MRL-lpr background genes, responsible for the severity ofglomerulonephritis, were recently mapped. It would be of interest to map thegenes responsible for arthritis susceptibility in this mouse strain. The late-life spontaneous arthritis made it difficult to select for arthritis susceptibility.Having the adjuvant-enhancement, post-partum flare, and spontaneousmodels would allow us to carry out breeding experiments with a realisticchance to select out arthritis susceptible animals (about 70% of MRL-lprdevelop either adjuvant-enhanced, post-partum, or late-life spontaneousarthritis).As this model was developed to facilitate therapeutic trials, it will be ofutmost interest to continue with this application. Both systemic and localeffects of PDT in the prevention of adjuvant-enhanced arthritis in MRL-lprmice should be further investigated to determine the relative efficacy. It isalso of interest to determine whether mice displaying active arthritis wouldrespond in a similar manner to this treatment.221Bibliography222Abe C, Shiokawa Y. (1981), Shiokawa Y, Abe T, Yamauchi Y (eds):Spontaneous polyarthritis in MRL/l mice, in New horizons in rheumatoidarthritis. Excerpta Medica, International Congress Series no. 535, NY,pp.116- 131.Adachi, M., Watanabe-Fukunaga, R. and Nagata, S. (1993) Aberranttranscription caused by the insertion of an early transposable element in anintron of the Fas antigen gene of lpr mice. Proc. Nati. Acad. Sd. 90:1756-1760.Alarcon-Segovia D. (1988), Goodacre JA, Carson Dick W (eds): Systemic lupuserythematosus, in Immunopathogenetic mechanism of arthritis. MTP Press,Norwell, pp.198-210.Alexander, E.L., Moyer, C., Travlos, G.S., Roths, J.B. and Murphy, E.D.(1985) Two histopathologic types of inflammatory vascular disease in MRL/Mpautoimmune mice. Model for human vasculitis in connective tissue disease.Arthritis Rheum. 28:1146-1155.Allen, R.D., Marshall, J.D., Roths, J.B. and Sidman, C.L. (1990) Differencesdefined by bone marrow transplantation suggest that lpr and gid aremutations of genes encoding an interacting pair of molecules. J. exp. Med.172:1367-1375.Alley, M.C. (1988) Feasibility of drug screening with panels of human tumourcell lines using a microculture tetrazolium assay. Cancer Res. 48:586-601.Allison, B.A., Pritchard, H., Richter, A. and Levy, J.G. (1990) The plasmadistribution of benzoporphyrin derivative and the effects of plasmalipoproteins on its biodistribution. Photochem. Photobiol. 52:50 1-507.Anderle, S.K., Greenblatt, J.J., Cromartie, W.J., Clark, R. and Schwab, J.H.(1979) Modulation of the susceptibility of inbred and outbred rats to arthritisinduced by cell walls of group A streptococci. Infection & Immunity25:484-490.Andrews, B.S., Eisenberg, R.A., Theofilopoulos, A.N., Izui, S., Wilson, C.B.,McConahey, P.J., Murphy, E.D., Roths, J.B. and Dixon, F.J. (1978)Spontaneous murine lupus-like syndromes. Clinical andimmunopathological manifestations in several strains. J. exp. Med.148:1198-1215.Andrews, H.J., Bunning, R.A.D., Plumpton, T.A., Clark, I.M., Russell,R.G.G. and Cawston, T.E. (1990) Inhibition of interleukin-1-inducedcollagenase production in human articular chondrocytes in vitro byrecombinant human interferon-gamma. Arthritis Rheum. 33:1733-1738.223Ansar Ahmed, S., Dauphinee, M.J. and Talal, N. (1985) Effects of short-termadministration of sex hormones on normal and autoimmune mice. J.Immunol. 134:204-210.Antes, U., Heinz, H.-P. and Loos, M. (1988) Evidence for the presence ofautoantibodies to the collagen-like portion of Clq in systemic lupuserythematosus. Arthritis Rheum. 4:457-464.Arend, W.P. and Dayer, J.M. (1990) Cytokines and cytokine inhibitors orantagonists in rheumatoid arthritis. Arthritis Rheum. 33:305-315.Arnett, F.C., Edworthy, S.M., Bloch, D.E. (1988) The American RheumatismAssociation 1987 revised criteria for the classification of rheumatoid arthritis.Arthritis Rheum. 31:315-324.Barland, P., Novikoff, A.B. and Hamerman, D. (1962) Electron microscopy ofthe human synovial membrane. J. Cell. Biol 14:207-220.Bartlett, R.R., Popovic, S. and Raiss, R.X. (1988) Development ofautoimmunity in MRL/lpr mice and the effects of drugs on this murinedisease. Scand. J. Rheumatol. Suppi. 75:290-299.Bartlett RR, Raiss RX, Popovic 5, Schleyerbach R. (1986), Kuettner K (ed):Immunological, histologic, and ultrastructural studies in MRLI1 mice, inArticular cartilage biochemistry. Raven Press, N.Y., pp. 391-411.Berczi, I. and Nagy, E. (1982) A possible role of prolactin in adjuvant arthritis.Arthritis Rheum. 25:591-594.Berden, J.H., Faaber, P., Assmann, K.J. and Rijke, T.P. (1986) Effects ofcyclosporin A on autoimmune disease in MRL/l and BXSB mice. Scand. J.Immunol. 24:405-411.Berger, C.L., Perez, M., Laroche, L. and Edelson, R. (1990) Inhibition ofautoimmune disease in a murine model of systemic lupus erythematosusinduced by exposure to syngeneic photoinactivated lymphocytes. J. Invest.Dermat. 94:52-57.Bijisma, J.W. and van den Brink, H.R. (1992) Estrogens and rheumatoidarthritis. Am. J. Reprod. Immunol. 28:231-234.Boissier, M.C., Texier, B., Carlioz, A. and Fournier, C. (1989) Polyarthritis inMRL-lpr/lpr mice: mouse type II collagen is antigenic but not arthritogenic.Autoimmunity 4:31-41.224Boraschi, D., Censini, S., Bartalini, M. and Tagliabue, A. (1985) Regulation ofarachidonic acid metabolism in macrophages by immune and nonimmuneinterferons. J. Immunol. 135:502-505.Bouvet, J.P., Couderc, J., Bouthiffier, Y., Franc, B., Ducailar, A. andMouton, D. (1990) Spontaneous rheumatoid-like arthritis in a line of micesensitive to collagen-induced arthritis. Arthritis Rheum. 33:1716-1722.Brahn, E. and Trentham, D.E. (1984) Effect of antithymocyte serum oncollagen arthritis in rats: evidence that T cells are involved in itspathogenesis. Cell. Immunol. 86:421-428.Breedveld, F.C. and Trentham, D.E. (1987) Progress in the understanding ofinducible models of chronic arthritis. Rheum. Dis. Clin. North. Am.13:531-544.Brick, J.E., Walker, S.E. and Wise, K.S. (1988) Hormone control ofautoantibodies to calf thymus nuclear extract (CTE) and DNA in MRL-lpr andMRL-+/+ mice. Clin. Immunol. Immunopathol. 46:68-81.Brinckerhoff C.E. and Guyre, P.M. (1985) Increased proliferation of humansynovial fibroblasts treated with recombinant immune interferon. J.Immunol. 184:3142-3146.Buzas, E.I., Hollo, K., Rubliczky, L., Garzo, M., Nyirkos, P. and Giant, T.T.(1993) Effect of pregnancy on proteoglycan-induced progressive arthritis inBALB/c mice: remission of disease activity. Clin. exp. Immunol. 94:252-260.Carlsten, H. (1992) Murine SLE. Impact of estrogen on immune functions andprogression of disease. Scand. J. Rheumatol. Suppi. 93:22. (abstr.)Carlsten, H., Holmdahl, R., Tarkowski, A. and Nilsson, L.A. (1989)Oestradiol- and testosterone-mediated effects on the immune system innormal and autoimmune mice are genetically linked and inherited asdominant traits. Immunol. 68:209-214.Caristen, H., Nilsson, N., Jonsson, R., Backman, K., Holmdahl, R. andTarkowski, A. (1992) Estrogen accelerates immune complexglomerulonephritis but ameliorates T cell-mediated vasculitis andsialadenitis in autoimmune MRL lpr/lpr mice. Cell. Immunol. 144: 190-202.Carlsten, H., Nilsson, N., Jonsson, R. and Tarkowski, A. (1991) Differentialeffects of oestrogen in murine lupus: acceleration of glomerulonephritis andamelioration of T cell-mediated lesions. J. Autoimmunity 4:845-856.225Caristen, H., Tarkowski, A., Holmdahl, R. and Nilsson, L.A. (1990)Oestrogen is a potent disease accelerator in SLE-prone MEL lpr/lpr mice.Clin. exp. Immunol. 80:467-473.Carsons S, Lavietes BB, Diamond HS. (1989), Mosher DF (ed): Role offibronectin in rheumatic diseases, in Fibronectin. Academic Press, NewYork, pp. 327-361.Cathcart, E.S., Hayes, K.C., Gonnerman, W.A., Lazzari, A.A. andFranzblau, C. (1986) Experimental arthritis in a nonhuman primate. 1.Induction by bovine type II collagen. Lab. Invest. 54:26-31.Caulfield, J.P., Rein, A., Dynesius-Trentham, R. and Trentham, D.E. (1982)Morphologic demonstration of two stages in the development of type IIcollagen-induced arthritis. Lab. Invest. 46:321-343.Choy, E.H., Chikanza, I.C., Kingsley, G.H., Corrigall, V. and Panayi, G.S.(1992) Treatment of rheumatoid arthritis with single dose or weekly pulses ofchimaeric anti-CD4 monoclonal antibody. Scand. J. Immunol. 36:291-298.Chu, J.L., Drappa, J., Parnassa, A. and Elkon, K.B. (1993) The defect in FasmRNA expression in MRL/lpr mice is associated with insertion of theretrotransposon, ETn. J. exp. Med. 178:723-730.Clague, R.B. (1989) Autoantibodies to cartilage collagens in rheumatoidarthritis. Do they perpetuate the disease or are they irrelevant? Br. J.Rheumatol. 28:1-6.Clark, R.L., Cuttino, J.T.,Jr., Anderle, S.K., Cromartie, W.J. and Schwab,J.H. (1979) Radiologic analysis of arthritis in rats after systemic injection ofstreptococcal cell walls. Arthritis Rheum. 22:25-35.Cohen, P.L. and Eisenberg, R.A. (1991) Lpr and gid: single gene models ofsystemic autoimmunity and lymphoproliferative disease. Ann. Rev.Immunol. 9:243-269.Cohen, P.L. and Eisenberg, R.A. (1992) The lpr and gid genes in systemicautoimmunity: life and death in the Fas lane. Immunol. Today 13:427-428.Collins, I., Morgan, K., Clague, R.B., Brenchley, P.E.C. and Holt, P.J.L.(1988) IgG subclass distribution of antinative type II collagen andanti-denatured type II collagen antibodies in patients with rheumatoidarthritis. JRheumatol 15:770-774.Cook MJ (1965) The anatomy of the laboratory mouse. Academic Press,London,226Courtenay, J.S., Daliman, M.J., Dayan, A.D., Martin, A. and Mosedale, B.(1980) Immunisation against heterologous type II collagen induces arthritisin mice. Nature 283:666-668.Cremer, M.A., Townes, A.S. and Kang, A.H. (1990) Adjuvant-inducedarthritis in rats. Evidence that autoimmuruity to homologous collagens typesI, II, IX and XI is not involved in the pathogenesis of arthritis. Clin. exp.Immunol. 82:307-312.Cromartie, W.J., Craddock, J.G., Schwab, J.H., Anderle, S.K. and Yang,C.H. (1977) Arthritis in rats after systemic injection of streptococcal cells orcell walls. J. exp. Med. 146:1585-1602.Davignon, J.L., Budd, R.C., Ceredig, R., Piguet, P.F., MacDonald, H.R.,Cerottini, J.C., Vassalli, P. and Izui, S. (1985) Functional analysis of T cellsubsets from mice bearing the lpr gene. J. Immunol. 135:2423-2428.Dayer, J.M., Beutler, B. and Cerami, A. (1985) Cachectin/tumor necrosisfactor stimulates collagenase and prostaglandin E2 production by humansynovial cells and dermal fibroblasts. J. exp. Med. 162:2163-2168.Dayer, J.M., de Rochemontix, B. and Burrus, B. (1986) Human recombinantinterleukin 1 stimulates collagenase and prostaglandin E2 production byhuman synovial cells. J. Clin. Invest. 77:645-648.DeJoy, S.Q., Ferguson, K.M., Sapp, T.M., Zabriskie, J.B., Oronsky, A.L. andKerwar, S.S. (1989) Streptococcal cell wall arthritis. Passive transfer ofdisease with a T cell line and crossreactivity of streptococcal cell wall antigenswith Mycobacterium tuberculosis. J. exp. Med. 170:369-382.Dewhirst, F.L., Stashenko, P.P., Mole, J.E. and Tsurumachi, T. (1985)Purification and partial sequence of human osteoclast-activating factor:identity with interleukin 1cr. J. Immunol. 135:2562-2568.Di Giovine, F.S., Nuki, G. and Duff, G.W. (1988) Tumour necrosis factor insynovial exiidates. Ann. Rheum. Dis. 47:768-772.Diamond HS. (1982), Collagen autoinirnune diseases, in Handbook ofimmunology for students and house staff. Verlag Chemie International,Stuttgart, pp. 155-168.Digeon, M., Layer, M., Riza, J. and Bach, J.F. (1977) Detection of circulatingimmune complexes in human sera by simplified assays with polyethyleneglycol. J. Immunol. Methods 16: 165-183.Dixon, F.J. (1985) Murine lupus. Arthritis Rheum. 28:1081-1088.227Dixon, F.J. (1987) Basic elements of murine systemic lupus erythematosus. J.Rheumatol. suppi. 14:3-10.Dixon, F.J., Andrews, B.S., Eisenberg, R.A., McConahey, P.J.,Theofilopoulos, A.N. and Wilson, C.B. (1978) Etiology and pathogenesis of aspontaneous lupus-like syndrome in mice. Arthritis Rheum. 21:S64-S67.Dougherty, T.J. and Marcus, S.L. (1992) Photodynamic therapy. Euro. J.Cancer 28A 1734-42.Edelson, R. (1988) Light activated drugs. Scientific Amer. 259:68-75.Edelson, R.L. (1989) Update on photopheresis. Prog. in Dermatol. 28:1-7.Edwards, J.C., Cooke, A., Moore, A.R., Collins, C., Hay, F. and Willoughby,D.A. (1986) Connective tissue abnormalities in MRLI1 mice. Ann. Rheum.Dis. 45:512-518.Edwards, J.C.W., Sedgwick, A.D. and Willoughby, D.A. (1982a) Membraneproperties and esterase activity of synovial lining cells, further evidence for amononuclear phagocyte population. Ann. Rheum. Dis. 41:282-286.Edwards, J.C.W. and Willoughby, D.A. (1982b) Demonstration ofbone-marrow derived cells in synovial lining by means of giant intracellulargranulars as genetic markers. Ann. Rheum. Dis. 41:177-182.Eisenberg, R.A., Thor, L.T. and Dixon, F.J. (1979) Serum-serum interactionsin autoimmune mice. Arthritis Rheum. 22:1074-1081.Emery, P., Smith, G.N. and Panayi, G.S. (1986) Lymphocytapheresis- afeasible treatment for rheumatoid arthritis. Br. J. Rheumatol. 25:40-43.Faaber, P., Capel, P.J.A., Rijke, G.P.M., Vierwinden, G., Van De Putte,L.B.A. and Koene, R.A.P. (1984) Cross-reactivity of anti-DNA antibodies withproteoglycans. Clin. exp. Immunol. 55:502-508.Fingar, V.11., Siegel, K.A., Wieman, J. and Doak, K.W. (1993) The effect ofthromboxane inhibitors on the microvasculature and tumour response tophotodynamic therapy. Photochem. Photobiol. 58:393-399.Fossati, L., Takahashi, S., Merino, R., Iwamoto, M., Aubry, J.P., Nose, M.,Spach, C., Motta, R. and Izui, 5. (1993) An MRLJMpJ-lprllpr substrain with alimited expansion of lpr double-negative T cells and a reduced autoinununesyndrome. mt. Immunol. 5:525-532.228Fox, D.A. and Smith, B.R. (1986) Evidence for oligoclonal B cell expansion inthe peripheral blood of patients with rheumatoid arthritis. Ann. Rheum. Dis.45:991-995.Gardner DL. (1992a), Systemic lupus erythematosus, in Pathological basis ofthe connective tissue diseases. Lea & Febiger, Philadelphia, pp. 568-613.Gardner DL. (1992b), Rheumatoid arthritis: cell and tissue pathology, inPathological basis of the connective tissue diseases. Lea & Febiger,Philadelphia, pp. 444-526.Gardner DL. (1992c), Rheumatoid arthritis: pathogenesis, experimentalstudies and complications, in Pathological basis of the connective tissuediseases. Lea & Febiger, Philadelphia, pp. 526-567.Gasparro FP. (1988), Gasparro FP (ed): Psoralen-DNA interactions:thermodynamics and photochemistry, in Psoralen DNA Photobiology. Vol. I.CRC Press, Boca Raton, FL, pp. 5-36.Gaston, J.S. (1991) Heat shock proteins and autoimmunity. Sem. Immunol.3:35-42.Gaston, J.S., Strober, S. and Solovera, J.J. (1988a) Dissection of mechanismsof immune injury in rheumatoid arthritis using total lymphoid irradiation.Arthritis Rheum. 47:127-133.Gaston, J.S.H., Life, P.F., Bailey, L. and Bacon, P.A. (1988b) Synovial fluid Tcells and 65 kD heat-shock protein. Lancet ii:856.Gay, S., Fine, J.D. and Storer, J.S. (1988) Autoantibodies to extracellularmatrix components in epidermolysis bullosa and other bullous diseases.Arch. Dermatol. Res. 280:333-337.Gay, S., Gay, R.E. and lVliller, E.J. (1980) The collagens of the joint. ArthritisRheum. 23:937-941.Gay, S., O’Sullivan, F.X., Gay, R.E. and Koopman, W.J. (1987) Humoralsensitivity to native collagen types 1-VI in the arthritis of MRL/l mice. Clin.Immunol. Immunopathol. 45:63-69.Geller, V.G., Keitel, W. and Franke, A. (1974) Vergleichende histologie deradjuvans-arthritis bei ratten, mausen, und hamstern. Allerg. Immunol.20-21:251.Gilkeson, G.S., Ruiz, P., Grudier, J.P., Kurlander, R.J. and Pisetsky, D.S.(1989) Genetic control of inflammatory arthritis in congenic lpr mice. Clin.Immunol. Immunopathol. 53:460-474.229Gilkeson, G.S., Ruiz, P., Pritchard, A.J. and Pisetsky, D.S. (1991) Geneticcontrol of inflammatory arthritis and glomerulonephritis in congenic lprmice and their Fl hybrids. J. Autoimmunity 4:595-606.Gilkeson, G.S., Spurney, R., Coffman, T.M., Kuriander, R., Ruiz, P. andPisetsky, D.S. (1992) Effect of anti-CD4 antibody treatment on inflammatoryarthritis in MRL-lpr/lpr mice. Clin. Immunol. Immunopathol. 64:166-172.Giant, T.T., Mikecz, K., Arzoumanian, A. and Poole, A.R. (1987)Proteoglycan-induced arthritis in BALB/c mice. Clinical features andhistopathology. Arthritis Rheum. 30:201-212.Gowen, M. and Mundy, G.R. (1986) Actions of recombinant interieukin 1,interleukin 2, and interferon-gamma on bone resorption in vitro. J.Immunol. 136:2478-2482.Griffin J. (1990), Scott JS, Bird HA (eds): Rheumatoid arthritis: Biologicaleffects and management, in Pregnancy, autoimmunity, and connective tissuedisorders. Oxford University Press, New York, pp. 140-162.Guerne, P.A., Carson, D.A. and Lotz, M. (1990) FL-6 production by humanarticular chondrocytes. Modulation of its synthesis by citokines, growthfactors, and hormones in vitro. J. Immunol. 144:499-505.Hahn, G., Stuhimuller, B., Ham, N., Kalden, J.R., Pfizenmaier, H. andBurmester, G.R. (1993) Modulation of monocyte activation in patients withrheumatoid arthritis by leukapheresis therapy. J. Clin. Invest. 91:862-870.Hail, G.M., Daniels, M. and Spector, T.D. (1994) Estrogen replacement (ERT)as an adjunct therapy in postmenopausal rheumatoid arthritis (RA); arandomized study of 200 patients. Arthritis Rheum. 36 (Suppl.):S57-S50.(abstr.)Hamilton, J.A., Ciarris, B.J., Fraser, J.R.E. and Niall, M.C. (1985)Peripheral blood mononuclear cells stimulate prostacyclin levels of humansynovial fibroblast-like cells. Rheumatol. mt. 5:121-125.Hang, L., Theofliopoulos, A.N. and Dixon, F.J. (1982) A spontaneousrheumatoid arthritis-like disease in MRLI1 mice. J. exp. Med. 155:1690-1701.Hang, L.M., Aguado, M.T., Dixon, F.J. and Theofilopoulos, A.N. (1985)Induction of severe autoimmune disease in normal mice by simultaneousaction of multiple immunostimulators. J. exp. Med. 161:423-428.Harris, E.D. (1986) Synovial inflammation in rheumatoid arthritis. Hosp.Pract. 87:71-82.230Harris, E.D. and Krane, S.M. (1972) An endopeptidase from rheumatoidsynovial tissue culture. Biochem. Biophys. Acta 258:566-576.Hayken, G.D., Steinberg, M.E. and Schumacher, H.R.,Jr. (1979) Effects ofjoint immobilization on antigen-induced arthritis. Surgical Forum30:498-500.Hayward, M. and Fiedler-Nagy, C.H. (1987) Meeting report: Mechanism ofbone loss: rheumatoid arthritis, Periodontal disease and osteoporosis. Agents& Actions 22:251-254.Hazes, J.M.W (1991) Pregnancy and its effect on the risk of developingrheumatoid arthritis. Ann. Rheum. Dis. 50:71-72.Hazes, J.M.W. and van Zeben, D. (1991) Oral contraception and its possibleprotection against rheumatoid arthritis. Ann. Rheum. Dis. 50:72-74.He, D., Soter, N.A. and Lim, H.W. (1989) The late phase of hematoporphyrinderivative-induced phototoxicity in mice: release of histamine and histologicchanges. Photochem. Photobiol. 50:91-95.Helfgott, S.M., Bazin, H., Dessein, A. and Trentham, D.E. (1984) Suppressiveeffects of anti-mu serum on the development of collagen arthritis in rats.Clin. Immunol. Immunopathol. 31:403-411.Henderson, B. and Dougherty, T.J. (1992) How does photodynamic therapywork? Photochem. Photobiol. 55:144-157.Herron, L.R., Eisenberg, R.A., Roper, E., Kakkanaiah, V.N., Cohen, P.L. andKotzin, B.L. (1993) Selection of the T cell receptor repertoire in lpr mice. J.Immunol. 151:3450-3459.Herzog, C., Walker, C. and Pichier, W. (1987) Monoclonal anti-CD4 inarthritis. Lancet ii:1461-1462.Hewicker, M., Garces, L.H., Kromschroder, E. and Trautwein, G. (1988)Pathogenesis of spontaneous chronic polyarthritis in MRL mice. . Z.Rheumatol. 47:219-226.Hibbs, M.S., Hasty, K.A., Seyer, J.M., Kang, A.H. and Mainardi, C.L. (1985)Biochemical and immunological characterization of the secreted forms ofhuman neutrophil gelatinase. J. Biol. Chem 260:2493-2500.Hirano, T., Akira, S., Taga, T. and Kishimoto, T. (1990) Biological and clinicalaspects of interleukin 6. Immunol. Today 11:444-449.231Hirsch, H.Z., Tarkowski, A., Miller, E.J., Gay, S., Koopman, W.J. andMestecky, J. (1988) Autoimmunity to collagen in adult periodontal disease. J.Oral Pathol. 17:456-459.Holmdahl, R. (1989) Estrogen exaggerates lupus, but suppressesT-cell-dependent autoimmune disease. J. Autoimmunity 2:651-656.Holmdahl, R., Carlsten, H., Jansson, L. and Larsson, P. (1989) Oestrogen is apotent immunomodulator of murine experimental rheumatoid disease. Br. J.Rheumatol. 28 Suppi 1:54-58.Holmdahl, R., Jansson, L., Larsson, E., Rubin, K. and Kiareskog, L. (1986)Homologous type II collagen induces chronic and progressive arthritis inmice. Arthritis Rheum. 29:106.Holmdahl, R., Rubin, K., Kiareskog, L., Dencker, L., Gustafson, G. andLarsson, E. (1985) Appearance of different lymphoid cells in synovial tissueand in peripheral blood during the course of collagen II induced arthritis inrats. Scand. J. Immunol. 21:197-204.Holoshitz, J., Drucker, I., Yaretzky, A., Van Eden, W., Klajman, A., Lapidot,Z., Frenkel, A. and Cohen, I.R. (1986) T lymphocytes of rheumatoid arthritispatients show augmented reactivity to a fraction of mycobacteriacross-reactive with cartilage. Lancet ii:305-307.Holoshitz, J., Naparstek, Y., Ben-Nun, A. and Cohen, I.R. (1983) Lines of Tlymphocytes induce or vaccinate against autoimmune arthritis. Science219:56-58.Hom, J.T., Bendele, A.M. and Carison, D.G. (1988) In vivo administrationwith IL-i accelerates the development of collagen-induced arthritis in mice.J. Immunol. 141:834-841.Horn, J.T., Cole, H. and Bendele, A.M. (1990) Interleukin 1 enhances thedevelopment of spontaneous arthritis in MRL/lpr mice. Clin. Immunol.Immunopathol. 55:109-119.Hovdenes, J., Kvien, T.K. and Hovdenes, A.B. (1990) IL-6 in synovial fluids,plasma and supernatants from cultured cells of patients with rheumatoidarthritis and other inflammatory arthritides. Scand. J. Rheumatol.19:177-182.Hruza, L., Lui, H., Kollias, N., Wimberly, J., Dwyer, P., Gagel, M., Salvatori,V. and Anderson, R. (1993) Photodynamic therapy of psoriasis: preliminaryevaluation of benzoporphyrin derivative monoacid ring A (BPD-MA).Photochem. Photobiol. 57 (Suppl):50S. (abstr.)232Hsiang, Y.N., Crespo, M.T., Richter, A.M., Jam, A.K., Fragoso, M. andLevy, J.G. (1993) In vitro and in vivo uptake of benzoporphyrin derivative inhuman and miniswine atherosclerotic plaque. Photochem. Photobiol.57:670-674.Inman, R.D. (1982) Immune complexes in SLE. Clin. Rheum. Dis. 8:49-62.Izui, S. and Eisenberg, R.A. (1980) Circulating anti-DNA-rheumatoid factorcomplexes in MRL/1 mice. Clin. Immunol. Immunopathol. 15:536-551.Izui, S., Kelley, V.E., Masuda, K., Yoshida, H., Roths, J.B. and Murphy, E.D.(1984) Induction of various autoantibodies by mutant gene lpr in severalstrains of mice. J. Immunol. 133:227-233.Jacobs, C., Young, D., Tyler, S., Callis, G., Gillis, S. and Conlon, P.J. (1989)In vivo treatment with IL-i reduces the severity and duration ofantigen-induced arthritis in rats. J. Immunol. 141:2967.Jam, A.K., Richter, A., Neyndorff, H. and Levy, J.G. (1993) The potential ofbenzoporphyrin derivative (BPD) as a photosensitizer of blood-born targets.Life Science Advances: Oncology (in press).Jamieson, C.H., Richter, A. and Levy, J.G. (1993) Efficacy of benzoporphyrinderivative, a photosensitizer, in selective destruction of leukaemia cells usinga murine tumour model. Exptl. Haematol. 21:629-634.Jansson, L. and Holmdahl, R. (1989) Oestrogen induced suppression ofcollagen arthritis. IV. Progesterone alone does not affect the course ofarthritis but enhances the oestrogen-mediated therapeutic effect. J. Reprod.Immunol 15: 141-150.Jansson, L., Mattsson, A., Mattsson, R. and Holmdahl, R. (1990) Estrogeninduced suppression of collagen arthritis. V: Physiological level of estrogen inDBAI1 mice is therapeutic on established arthritis, suppresses anti-type IIcollagen T-cell dependent immunity and stimulates polyclonal B-cell activity.J. Autoimmunity 3:257-270.Johnson, H.M. (1983) Interferon and host defense systems. Adv. Exp. Med.Biol. 162:105-119.Johnson, P.M., Phua, K.K., Hart, C.A. and Bucknall, R.C. (1984) Antibody tostreptococcal cell wall peptidoglycan-polysaccharide polymers in seropositiveand seronegative rheumatic diseases. Clin. exp. Immunol. 55:115-124.Johnson, P.W., Tonzetich, J. and Pearce, R.H. (1985) Characterization ofhydrogen suiphide reaction with rat-tail tendon type I collagen in vitro. J.Periodontal. Res 20:403-410.233Jori, G., Beltramini, E., Reddi, B., Salvato, A., Pagnan, L., Ziron, L., Tomio,L. and Tsanov, T. (1984) Evidence for a major role of plasma lipoproteins ashematoporphyrin carriers in vivo. Cancer Lett. 24:291-297.Kahan, B.D. (1989) Cyclosporine. N. Engi. J. Med. 321:1725-1738.Kaibara, N., Hotokebuchi, T., Takagishi, K. and Katsuki, I. (1983)Paradoxical effects of cyclosporin A on collagen arthritis in rats. J. exp. Med.158:2007-2015.Kamide, R., Gigli, I. and Lim, H.W. (1984) Participation of mast cells andcomplement in the immediate phase of hematoporpyrin inducedphototoxicity. J. Invest. Dermat. 82:485-490.Kandel, R.A., Petelycky, M., Dinarello, C., Minden, M., Pritzker, K.P.H. andCruz, T.F. (1990) Comparison of the effect of interleukin 6 and interleukin 1 oncollagenase and proteoglycan production by chondrocytes. J. Rheumatol.17:953-957.Kaufmann, Y., Change, A.E., Robb, R.J. and Rosenberg, S.A. (1984)Mechanism of action of Cyclosporyn A: Inhibition of lymphokyne secretionstudied with antigen-stimulated T-cell hybridomas. J. Immunol. 133:3107.Keitel, W., Wille, R., Franke, A. and Ziegeler, J. (1971) Adjuvant arthritis inmice and hamsters. Acta Rheum. Scand. 17:31-34.Kelley, V.E. and Roths, J.B. (1985) Interaction of mutant lpr gene withbackground strain influences renal disease. Clin. Immunol. Immunopathol.37:220-229.Kimura, M., Ogata, Y., Shimada, K., Moriyama, T. and Matsuzawa, A.(1991) New mutant mice of autoimmunity, CBAJKiJms-lprcg/lprcg, that couldlink the lpr and gid genes . Autoimmunity 9:359-361.Kingsley, G.H., Pitzalis, C. and Panayi, G.S. (1990) Immunogenetic andcellular immune mechanisms in rheumatoid arthritis: relevance to newtherapeutic strategies. Br. J. Rheumatol. 29:58-64.Kirkham, B., Chikanza, I. and Pitzalis, C. (1988) Response to monoclonal CD7antibody in rheumatoid arthritis. Lancet ii:589.Kmght, B, Katz, D R, Isenberg, D A, Ibrahim, M A, Le Page, S andHutchings, P (1992) Indu.cton of adjuvant arthritis in mice Cltn expImmunol. 90:459-465.234Ko, S.D., Page, R.C. and Narayanan, A.S. (1977) Fibroblast heterogeneity andprostaglandin regulation of subpopulation. Proc. Natl. Acad. Sci.74:3429-3435.Kotzin, B.L., Babcock, S.K. and Herron, L.R. (1988) Deletion of potentiallyself-reactive T cell receptor specificities in L3T4-, Lyt-2- T cells of lpr mice. J.exp. Med. 168:2221-2229.Kresina, T.F. (1987) Tmmunotherapy of experimental arthritis. Analysis ofthe articular cartilage of mice suppressed for collagen-induced arthritis by aT-cell hybridoma. Am. J. Pathol. 129:257-266.Kupper, T.S., Chua, A.O., Flood, P., McGuire, J. and Gubler, V. (1987)Interleukin 1 gene expression in cultured human keratinocytes is augmentedby ultraviolet irradiation. J. Clin. Invest. 80:430-436.Kyle, V., Coughlan, R.J., Tighe, H., Waldmann, H. and Hazieman, B.L.(1989) Beneficial effect of monoclonal antibody to interleukin 2 receptor onactivated T cells in rheumatoid arthritis. Ann. Rheum. Dis. 48:428-429.Linos, A., Worthington, J.W., O’Fallon, M.W. and Kurland, L.T. (1979) Theepidemiology of rheumatoid arthritis in Rochester, Minnesota: a study ofincidence, prevalence, and mortality. Am. J. Epidem. 3:87-98.Lipsky, P.E. (1992) Immunopathogenesis and treatment of rheumatoidarthritis. Journal ofRheumatology 19 (Supplement 32):92-94.Lowther, D.A., Sandy, J.D., Santer, V.B. and Brown, H.L.G. (1978)Antigen-induced arthritis. Decreased proteoglycan content and inhibition ofproteoglycan synthesis in articular cartilage. Arthritis Rheum. 2 1:675-679.Lui, H., Hruza, L., Kollias, N., Wimberly, J., Salvatori, V. and Anderson,R.R. (1993) Photodynamic therapy of malignant skin tumors withBenzoporphyrin derivative-monoacid ring A (BPD-MA): Preliminaryobservations. SPIE 1876:147-151.Mainardi, C.L. (1987) Biochemical mechanisms of articular destruction.Rheum. Dis. Clin. North. Am. 13:215-233.Maini, R.N., Brennan, F.M., Williams, R., Chu, C.Q., Cope, A.P., Gibbons,D., Elliott, M. and Feldmann, M. (1993) TNF-alpha in rheumatoid arthritisand prospects of anti-TNF therapy. Clinical & Experimental Rheumatology11 Suppi 8:S173-175.Malawista, S.E., Trock, D.H. and Edelson, R.L. (1991) Treatment ofrheumatoid arthritis by extracorporeal photochemotherapy. ArthritisRheum. 34:646-654.235Matsumoto, K., Yoshikai, Y., Asano, T., Himeno, K., Iwasaki, A. andNomoto, K. (1991) Defect in negative selection in lpr donor-derived T cellsdifferentiating in non-lpr host thymus. J. exp. Med. 173:127-136.Matsuzawa, A., Moriyama, T., Kaneko, T., Tanaka, M., Kimura, M., Ikeda,H. and Katagiri, T. (1990) A new allele of the lpr locus, lprcg, thatcomplements the gid gene in induction of lymphadenopathy in the mouse. J.exp. Med. 171:519-531.Matsuzawa, A., Moriyama, P., Ogata, Y., Katagiri, T. and Kimura, M. (1992)A crucial role of the thymus in induction by the lprcg gene oflymphadenopathy with autoimmunity in the mouse. Immunol. 75:688-692.Mattsson, R., Mattsson, A., Holmdahl, R., Whyte, A. and Rook, G.A.W. (1991)Maintained pregnancy levels of oestrogen afford complete protection frompost-partum exacerbation of collagen-induced arthritis. Clin. exp. Immunol.85:41-47.Mayne, R. (1989) Cartilage collagens: What is their function, and are theyinvolved in articular disease? Arthritis Rheum. 32:241-246.Mazière, J.C., Morlière, P. and Santus, R. (1991) The role of the low densitylipoprotein receptor pathway in the delivery of lipophilic photosensitizers inthe photodynamic therapy of tumours. J. Photochem. Photobiol. 8:351-360.McCune, W.J. and Bayliss, G.E. (1991) Immunosuppressive therapy forrheumatic disease. Curr. Opin. Rheumatol. 3:355-362.McKenzie, W.N.,Jr., Sunderrajan, E.V., Kavanaugh, J.L., Braun, S.,Ansbacher, L. and Walker, S.E. (1988) Sex hormones modulate the responseof pulmonary perivascular inflammation to cyclophosphamide therapy inMRLJMpJ-lprllpr mice. Am. J. Pathol. 131:530-538.Meacock, S.C., Brandon, D.R., Brown, C.P. and Swann, B.P. (1992) A noveltechnique for immunohistoperoxidase staining of unfixed whole joints ofsmall animals. Histochem. J. 24:115-119.Melby, K., Midtredt, T. and Degre, M. (1982) Effect of human leukocyteinterferon on phagocytic activity of polymorphonuclear leukocytes. ActaPathol. Microbiol. Immunol. Scand. 90: 18 1-184.Morgan, K. (1990) What do anti-collagen antibodies mean? Ann. Rheum. Dis.49:62-65.236Morgan, K., Buckee, C., Collins, I., Ayad, S., Clague, R.B. and Holt, P.J.(1988) Antibodies to type II and XI collagens: Evidence for the formation ofantigen specific as well as cross reacting antibodies in patients withrheumatoid arthritis. Ann. Rheum. Dis. 47:1008-1013.Morgan, K., Clague, R.B., Collins, I., Ayad, S., Phinn, S.D. and Holt, P.J.(1989) A longitudinal study of cc-collagen antibodies in patients withrheumatoid arthritis. . Arthritis Rheum. 32:139-145.Mosman, P. (1983) Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J. Immunol. Methods65:55-63.Mountz, J.D. and Edwards, C.K. (1992) Murine models of autoimmunity:T-cell and B-cell defects. Curr. Opin. Rheumatol. 4:612-620.Mountz, J.D., Smith, H.R., Wilder, R.L., Reeves, J.P. and Steinberg, A.D.(1987) CS-A therapy in MRL-lpr/lpr mice: amelioration of immunopathologydespite autoantibody production. J. Immunol. 138:157-163.Mountz, J.D., Zhou, T., Long, R.E., Bluethmann, H., Koopman, W.J. andEdwards III, C.K. (1994) P cell influence on superantigen-induced arthritis inMRL-lpr/lpr mice. Arthritis Rheum. 37:113-124.Murphy ED. (1981), Gershwin ME, Merchant B (eds): Lymphoproliferation(lpr) and other single-locus models for murine lupus, in Immunologic defectsin laboratory animals. Plenum Press, New York, pp. 143-173.Murphy ED, Roths JB. (1979), Rose NR, Bigazzi PE, Warner NL (eds):Autoimmunity and lymphoproliferation: induction by mutant gene lpr, andacceleration by a male-associated factor in BXSB mice, in Genetic control ofautoimmune disease. Elsevier North Holland Inc., New York, pp. 207-220.Murphy, E.D. and Roths, J.R. (1977) New inbred strains. Mouse News Letter58:51-52.Nagy, E., Chalmers, I.M, Baragar, F.D., Friesen, H.G. and Berczi, I. (1991)Prolactin deficiency in rheumatoid arthritis. J. Rheumatol. 18:1662-1668.Nakamura, K., Kashiwazaki, S., Takagishi, K., Tsukamoto, Y., Morohoshi,Y., Nakano, T. and Kimura, M. (1991) Spontaneous degenerative polyarthritisin male New Zealand black/RN mice. Arthritis Rheum. 34:171-179.Nemazee, D., Guiet, C., Buerki, K. and Marshak-Rothstein, A. (1991) Blymphocytes from the autoimmune-prone mouse strain MLRI1pr manifest an237intrinsic defect in tetraparental MRL/lpr in equilibrium DBA/2 chimeras. J.Immunol. 147:2536-2539.Nordling, C., Karisson-Parra, A., Jansson, L., Holmdahl, R. and Kiareskog,L. (1992) Characterization of a spontaneously occurring arthritis in maleDBA/1 mice. Arthritis Rheum. 35:717-722.North, J., Neyndorff, H., King, D. and Levy, J.G. (1992) Viral inactivation inblood and red cell concentrates with benzoporphyrin derivative. Blood Cells18:129-140.North, J., Neyndorif, H. and Levy, J.G. (1993) Photosensitizers as virucidalagents. J. Photochem. Photobiol. B: Biol. 17:99-108.O’Sullivan, F.X., Fassbender, H.G., Gay, S. and Koopman, W.J. (1985)Etiopathogenesis of the rheumatoid arthritis-like disease in MRL/l mice. I.The histomorphologic basis of joint destruction. Arthritis Rheum. 28:529-536.O’Sullivan, F.X., Ray, C.J., Takeda, Y., Sharp, G.C. and Walker, S.E. (1991)Long-term anti-CD4 treatment of MRL/lpr mice amelioratesimmunopathology and lymphoproliferation but fails to suppress rheumatoidfactor production. Clin. Immunol. Immunopathol. 61:421-435.Ofusu-Appiah, W.A., Warrington, R.J. and Wilkins, J.A. (1989) Interleukin 2responsive T cell clones from rheumatoid and normal subjects: Proliferativeresponses to connective tissue elements. Clin. Immunol. Immunopathol.50:264-271.Ogasawara, J., Watanabe-Fukunaga, R., Adachi, M., Matsuzawa, A.,Kasugai, T., Kitamura, Y., Itoh, N., Suda, T. and Nagata, S. (1993) Lethaleffect of the anti-Fas antibody in mice. Nature 364:806-809.Ogata, Y., Kimura, M., Shimada, K., Wakabayashi, T., Onoda, H., Katagiri,T. and Matsuzawa, A. (1993) Distinctive expression of lprcg in theheterozygous state on different genetic backgrounds. Cell. Immunol.148:91-102.Ohteki, T., Seki, S., Abo, T. and Kumagai, K. (1990) Liver is a possible site forthe proliferation of abnormal CD3÷4-8- double-negative lymphocytes inautoimmune MRL-lpr/lpr mice. J. exp. Med. 172:7-12.Ostensson, M. and Husby, G. (1983) A prospective clinical study of the effect ofpregnancy on rheumatoid arthritis and ankylosing spondylitis. ArthritisRheum. 26:1155-1159.238Paliard, X., West, S.G., Lafferty, J.A., Clements, J.R., Kappler, J.W.,Marrack, P. and Kotzin, B.L. (1991) Evidence for the effects of a superantigenin rheumatoid arthritis. Science 253:325-329.Panayi, G.S., Lanchbury, J.S. and Kingsley, G.H. (1992) The importance ofthe T cell in initiating and maintaining the chronic synovitis of rheumatoidarthritis. Arthritis Rheum. 35:729-735.Parrott, D.P., Goldberg, R.L., Kaplan, S.R. and Fuller, G.C. (1982) The effectof lymphokines on proliferation and collagen synthesis of cultured humansynovial cells. Eur. J. Clin. Invest. 12:407-415.Pass, 11.1. (1993) Photodynaniic therapy in oncology: mechanisms and clinicaluse. J. Natl. Cancer Inst. 85:443-456.Pataki, A. and RordorfAdam, C. (1985) Polyarthritis in MRL lpr/lpr mice.Rheumatol. mt. 5:113-120.Paulus, H.E., Machieder, H.I., Levine, S., Yu, D.T.Y. and MacDonald, N.S.(1977) Lymphocyte involvement in rheumatoid arthritis - studies duringthoracic duct drainage. Arthritis Rheum. 20: 1249-62.Pearson, C.M. (1956) Development of arthritis, periarthritis, and periostitis inrats given adjuvant. Proc. Soc. Exp. Biol. Med. 91:95- 101.Pearson, C.M. (1963) Experimental joint disease. Observations onadjuvant-induced arthritis. J. chron. Dis. 16:863-870.Pearson, C.M. and Wood, F.D. (1962) Studies of arthritis and other lesionsinduced in rats by the injection of mycobacterial adjuvant. VII. Pathologicdetails of the arthritis and spondylitis. J. Pathol. 42:73-95.Perkins, D.L., Glaser, R.M., Mahon, C.A., Michaelson, J. andMarshak-Rothstein, A. (1990) Evidence for an intrinsic B cell defect in lpr/lprmice apparent in neonatal chimeras. J. Immunol. 145:549-555.PeterfSr, F., Kuusela, P. and Ma:kela:, 0. (1983) Affinity requirements forantibody assays mapped by monoclonal antibodies. J. Immunol.130:1809-1813.Petri, M., Howard, D. and Repke, J. (1991) Frequency of lupus flare inpregnancy: the Hopkins Lupus Pregnancy Center experience. ArthritisRheum. 34:1538-1545.Pettipher, E.R., Riggs, G.A. and Henderson, B. (1988) Studies oninterleukin- 1-induced arthritis. Develop. Biol. Standard. 69:111-114.239Petty, RE., Hunt, D.W.C. and Rosenberg, A.M. (1986) Antibodies to type IVcollagen in rheumatic diseases. J. Rheumatol. 13:246-253.Phadke, K., Fouts, R., Parrish, J. and Baker, R.S. (1984) Autoreactivity tocollagen in a murine lupus model. Arthritis Rheum. 27:313-319.Pitzalis, C., Kingsley, G., Murphy, J. and Panayl, G. (1987) Abnormaldistribution of the helper-inducer and suppressor-inducer T lymphocytesubsets in the rheumatoid joint. Clin. Immunol. Immunopathol. 45:252-258.Poulter, L.W., Duke, 0., Panayi, G.S., Hobbs, S., Raftery, M.J. and Janossy,G. (1985) Activated T lymphocytes of the synovial membrane in rheumatoidarthritis and other arthropaties. Scand. J. Immunol. 22:683-690.Proflo AE. (1990), Kessel D (ed): Fluorescence diagnosis and dosimetry usingporphyrins, in Photodynamic therapy of neoplastic disease. CRC Press, BocaRaton, Florida, pp. 77-89.Prud’homme, G.J., Fieser, T.M., Dixon, F.J. and Theofilopoulos, A.N. (1984)B-cell-tropic interleukins in murine systemic lupus erythematosus (SLE) 1.Immunol. Rev. 78: 159-183.Quagliata, F., Sanders, P.M. and Gardner, D.L. (1969) Suppression ofadjuvant arthritis by a new cytotoxic compound, rubidomycin. Ann. Rheum.Dis. 28:163-168.Raisz, L.G., Sandberg, A.L., Goodson, J.M., Simmons, H.A. andMergenhagen, S.E. (1974) Complement-dependent stimulation ofprostaglandin synthesis and bone resorption. Science 185:789-791.Ranges, G.E., Fortin, S., Barger, M.T., Sriram, S. and Cooper, S.M. (1988) Invivo modulation of murine collagen induced arthritis. International Reviewsof Immunology 4:83-90.Res, P.C.M., Schar, C.G. and Breedveld, F.C. (1988) Synovial fluid T cellreactivity against 65 lcD heat shock protein of mycobacteria in early chronicarthritis. Lancet ii:478-480.Richter, A.M., Cerruti-Sola, S., Sternberg, E.D., Dolphin, D. and Levy, J.G.(1990) Bioclistribution of tritiated benzoporphyrin derivative (3H-BPD-MA), anew potent photosensitizer, in normal and tumor bearing mice. J.Photochem. Photobiol. 5:23 1-244.Richter, A.M., Chowdhary, R., Ratkay, L.G., Jam, A.K., Canaan, A.J.,Meadows, H., Obochi, M., Waterfield, J.D. and Levy, J.G. (1993a)Non-oncologic potentials for photodynamic therapy. SPIE (in press).240Richter, A.M., Jam, A.K., Obochi, M., Meadows, H., Canaan, A.J. and Levy,J.G. (1993b) Activation of Benzoporphyrin derivative in the circulation of micewithout skin photosensitivity. Photochem. Photobiol. (in press).Richter, A.M., Sternberg, E.D., Waterfield, E., Dolphin, D. and Levy, J.G.(1988) Characterization of benzoporhyrin derivative, a new photosensitizer.SPIE 997:132-138.Richter, A.M., Waterfield, E., Jam, A.K., Sternberg, E.D., Dolphin, D. andLevy, J.G. (1990) In vitro evaluation of phototoxic properties of fourstructurally related benzoporphyrin derivatives. Photochem. Photobiol.52:495-500.Richter, A.M., Yip, S., Waterfleld, E., Logan, P.M., Slonecker, C.E. and Levy,J.G. (1991) Mouse skin photosensitization with benzophorphyrin derivativesand Photofrin: macroscopic and microscopic evaluation. Photochem.Photobiol. 53:281-286.Rook, G.A.W., Steele, J., Brealey, R., Whyte, A., Isenberg, D., Sumar, N.,Nelson, J.L., Bodman, KB., Young, A., Roitt, I.M., Williams, P., Scragg, I.,Edge, C.J., Arkwright, P.D., Ashford, D., Wormald, M., Rudd, P., Redman,C.W.G., Dwek, R.A. and Rademacher, T.W. (1991) Changes in IgG glycoformlevels are associated with remission of arthritis during pregnancy. J.Autoimmunity 4:779-794.Rordorf, C., Pataki, A., Nogues, V., Schiager, F., Feige, U. and Glatt, M.(1987) Arthritis in MRLI1pr mice and in collagen II sensitized DBA-1 miceand their use in pharmacology. mt. J. Tissue React. 9:341-347.Rordorf-Adam, C., Rordorf B., Serban, D. and Pataki, A. (1986) The effects ofanti-inflammatory agents on the serology and arthritis of the MRL lpr/lprmouse. Agents & Actions 19:309-310.Rosenberg LC, Buckwalter JA. (1986), Kuettner KE, Schleyerbach R, HascallVC (eds): Cartilage proteoglycans, in Articular cartilage biochemistry. RavenPress, New York, pp. 39-57.Roths, J.B., Murphy, E.D. and Eicher, E.M. (1984) A new mutation, gid, thatproduces lymphoproliferation and autoimmunity in C3H/HeJ mice. J. exp.Med. 159:1-20.Santoro, T.J., Lehmann, K.R., Batt, R.A. and Kotzin, B.L. (1987) The role ofL3T4+ cells in the pathogenesis of lupus in lpr-bearing mice. I. Defects in theproduction of interleukins 2 and 3. Europ. J. Immunol. 17:1131-1136.241Santoro, T.J., Portanova, J.P. and Kotzin, B.L. (1988) The contribution ofL3T4+ T cells to lymphoproliferation and autoantibody production inMRL-lprllpr mice. J. exp. Med. 167:1713-1718.Saxne, T., Palladino, M.A., Heinegard, D., Talal, N. and Wollheim, F.A.(1988) Detection of tumour necrosis factor c but not tumour necrosis factor 13in rheumatoid arthritis synovial fluid and serum. Arthritis Rheum.31:1041-1045.Schlaghecke, R., Lenhardt, E., Fortkamp, E., Juli, E. and Zeidiler, H. (1989)Preventive and therapeutic effect of estrogen administration in adjuvantarthritis. Z. Rheumatol. 48:313-316.Schmidt, J.A., Mizel, S.B., Cohen, D. and Gree, I. (1982) Interleukin 1: Apotential regulator of fibroblast proliferation. J. Immunol. 128:2177-2182.Seaman, W.E., Wofsy, D., Greenspan, J.S. and Ledbetter, J.A. (1983)Treatment of autoimmune MRL/Ipr mice with monoclonal antibody toThy-1.2: a single injection has sustained effects on lymphoproliferation andrenal disease. J. Immunol. 130:1713-1718.Seggev, J.S., Sunderrajan, E.V., Palomo, T., McKenzie, W.N., Braun, S.R.,O’Sullivan, F.X. and Walker, S.E. (1991) Pulmonary perivascular andinterstitial inflammation in MRLfMpJ-lprllpr mice. III. Modulation bycyclophosphamide and sex hormones in 4- and 6-month-old animals. Clin.Immunol. Immunopathol. 60:289-298.Shear, H.L., Wofsy, D. and Talal, N. (1983) Effects of castration and sexhormones on immune clearance and autoimmune disease inMRL/Mp-lpr/lpr and MRL/Mp-+/+ mice. Clin. Immunol. Immunopathol.26:361-369.Singer, P.A., Balderas, R.S., McEvilly, R.J., Bobardt, M. and Theofilopoulos,A.N. (1989) Tolerance-related V beta clonal deletions in normal CD4-8-,TCR-alphalbeta + and abnormal lpr and gid cell populations. J. exp. Med.170:1869-1877.Singh, B., Pandey, V.B., Joshi, V.K. and Gambhir, S.S. (1987)Anti-inflammatory studies on Polygonum glabrum. J. Ethnopharmac.19:255-267.Snyderman R, McCarthy GA. (1982), Genco RJ, Mergenhagen SE (eds):Analogous mechanisms of tissue destruction in rheumatoid arthritis andperiodontal research, in Host-parasite interactions in periodontal disease.American Society for Microbiology, Washington, D.C., pp. 354-362.242Sobel, E.S., Cohen, P.L. and Eisenberg, R.A. (1993) lpr T cells are necessaryfor autoantibody production in lpr mice. J. Immunol. 150:4160-4167.Sobel, E.S., Katagiri, T., Katagiri, K., Morris, S.C., Cohen, P.L. andEisenberg, R.A. (1991) An intrinsic B cell defect is required for the productionof autoantibodlies in the lpr model of murine systemic autoimmunity. J. exp.Med. 173:1441-1449.Soren A. (1978) Histodiagnosis and clinical correlation of rheumatoid andother synovitis. J.B. Lippincott Company, Toronto.Spector, T.D., Roman, E. and Silman, A.J. (1990) The pill, parity andrheumatoid arthritis. Arthritis Rheum. 33:782-789.Steffen, C. and Wick, G. (1971) Delayed hypersensitivity reactions to collagenin rats with adjuvant arthritis. Z. Immun. Forsch. Immunobiol. 141:169-174.Steinberg, A.D. (1984) Systemic lupus erythematosus: insights from animalmodels. Ann. Intern. Med. 100:714-726.Steinberg, A.D., Roths, J.B., Murphy, E.D., Steinberg, R.T. and Raveche, E.S.(1980) Effects of thymectomy or androgen administration upon theautoimmune disease of MRL/Mp-lprllpr mice. J. Immunol. 125:871-873.Steinberg, E.B., Santoro, T.J., Chused, T.M., Smathers, P.A. and Steinberg,A.D. (1983) Studies of congenic MRL-lprllpr.xid mice. J. Immunol.131:2789-2795.Stim, T.B., Persico, F.J. and Garrabrant, T.A. (1984) Erythrocytesedimentation rate quantitation in IVfRL/MpM-lpr/lpr mice.Immunopharmacol. 8:171-174.Stuart, J.M., Cremer, M.A., Kang, A.H. and Townes, A.S. (1979)Collagen-induced arthritis in rats: evaluation of early immunologic events.Arthritis Rheum. 22:1344-135 1.Stuart, J.M., Huffstuffer, E.H., Townes, A.S. and Kang, A.H. (1983a)Incidence and specificity of antibodies to types I, II, III, 1V, and V collagen inrheumatoid arthritis and other rheumatic diseases as measured by‘25-I-radioimmunoassay. Arthritis Rheum. 26:832-840.Stuart, J.M., Tomoda, K., Yoo, T.J., Townes, A.S. and Kang, A.H. (1983b)Serum transfer of collagen-induced arthritis. II. Identification andlocalisation of autoantibody to type II collagen in donor and recipient rats.Arthritis. Rheum. 26:1237-1244.243Stuart, J.M., Townes, A.S. and Kang, A.H. (1984) Collagen autoimmunearthritis. Ann. Rev. Immunol. 2:199-218.Sulitzeanu, D. and Anafi, M. (1988) EBV, molecular mimicry and rheumatoidarthritis. Ann. Rheum. Dis. 47:265-269.Swaak, A.J.G., Kleinveld, H.A., Kioster, J.F. and Hack, C.E. (1989) Possiblerole of free radical altered IgG in the etiopathogenesis of rheumatoidarthritis. Rheumatol. mt. 9:1-6.Takagishi, K., Kaibara, N., Hotokebuchi, T., Arita, C., Morinaga, M. andArai, K. (1986) Effects of Cyclosporin on collagen-induced arthritis in mice.Ann. Rheum. Dis. 45:339.Tanaka, A., O’Sullivan, F.X., Koopman, W.J. and Gay, S. (1988)Etiopathogenesis of rheumatoid arthritis-like disease in MRLI1 mice: II.Ultrastructural basis of joint destruction. J. Rheumatol. 15:10-16.Tarkowski, A., Holmdahl, R., Rubin, K., Kiareskog, L., Nilsson, L.A. andGunnarsson, K. (1986) Patterns of autoreactivity to collagen type II inautoimmune MRL/l mice. Clin. exp. Immunol. 63:441-449.Tarkowski, A., Jonsson, R., Holmdahl, R. and Kiareskog, L. (1987)Immunohistochemical characterization of synovial cells in arthriticMRL-lpr/lpr mice. Arthritis Rheum. 30:75-82.Theofilopoulos, A.N., Balderas, R., Shawler, D.L., Izui, S., Kotzin, B.L.,Strober, S. and Dixon, F.J. (1980) Inhibition of T cell proliferation and SLE-likesyndrome of MRL/1 mice by whole body or total lymphoid irradiation. J.Immunol. 125:2137-2142.Theofilopoulos, A.N. and Dixon, F.J. (1981) Etiopathogenesis of murine SLE.Immunol. Rev. 55:179-215.Theofilopoulos, A.N. and Dixon, F.J. (1985) Murine models of systemic lupuserythematosus. Adu. Immunol. 37:269-386.Thompson, S.J., Hitsumoto, Y., Zhang, Y.W., Rook, G.A.W. and Elson, C.J.(1992) Agalactosyl IgG in pristane-induced arthritis. Pregnancy affects theincidence and severity of arthritis and the glycosylation status of IgG. Clin.exp. Immunol. 89:434-438.Trentham, D.E. (1982) Collagen arthritis as a relevant model for rheumatoidarthritis. Evidence pro and con. Arthritis Rheum. 25:911-916.Trentham, D.E. and Dynesius-Trentham, R.A. (1983) Attenuation of anadjuvant arthritis by type II collagen. J. Immunol. 130:2689-2692.244Trentham, D.E., McCune, W.J., Susman, P. and David, J.R. (1980)Autoimmunity to collagen in adjuvant arthritis of rats. J. Clin. Invest.66:1109-1117.Trentham, D.E., Townes, A.S. and Kang, A.H. (1977a) Autoimmunity to typeII collagen: an experimental model of type II arthritis. J.exp. Med.146:857-867.Trentham, D.E., Townes, A.S. and Kang, A.H. (1977b) Autoimmunity to typeII collagen: an experimental model of arthritis. J. exp. Med. 146:857.Trentham, D.E., Townes, A.S., Kang, A.H. and David, J.R. (1978) Humoraland cellular sensitivity to collagen in type II collagen-induced arthritis inrats. J. Clin. Invest. 61:89-96.Urowitz, M.B., Gladman, D.D., Farewell, V.T., Stewart, J. and McDonald, J.(1993) Lupus and pregnancy studies. Arthritis Rheum. 36:1392-1397.van den Brink, H.R., Lems, W.F., Everdingen, A.A. and Bijlsma, J.W.J.(1993) Adjuvant oestrogen treatment increases bone mineral density inpostmenopausal women with rheumatoid arthritis. Ann. Rheum. Dis.52:302-305.Van Eden, W., Hogervorst, E.J., Hensen, E.J., van der Zee, R., van Embden,J.D. and Cohen, I.R. (1989a) A cartilage-mimicking T-cell epitope on a 65Kmycobacterial heat-shock protein: adjuvant arthritis as a model for humanrheumatoid arthritis. Curr. Top. Microbiol. Immunol. 145:27-43.Van Eden, W., Hogervorst, E.J., van der Zee, R., van Embden, J.D., Hensen,E.J. and Cohen, I.R. (1989b) The mycobacterial 65 kD heat-shock protein andautoimmune arthritis. Rheumatol. mt. 9:187-191.Van Eden, W., Thole, J.E., van der Zee, R., Noordzij, A., van Embden, J.D.,Hensen, E.J. and Cohen, I.R. (1988) Cloning of the mycobacterial epitoperecognized by T lymphocytes in adjuvant arthritis. Nature 331:171-173.Van Noorden, C.J. and Vogels, LM. (1986) Enzyme histochemical reactionsin unfixed and undecalcified cryostat sections of mouse knee joints withspecial reference to arthritic lesions. Histochem. 86:127-133.Veys, E.M., Mielants, H. and Vergruggen, G. (1988) Thymic hormones andinterferons in the treatments of rheumatoid arthritis. Scand. J. Rheum,atol.Suppi. 76:297-304.Waites, G.T. and Whyte, A. (1987) Effect of pregnancy on collagen-inducedarthritis in mice. Clin. exp. Immunol. 67:467-76.245Walsh, L.J., Seymour, G.J. and Poewell, R.N. (1986) Modulation of class II(DR and DQ) antigen expression on gingival Langerhans cells in vitro bygamma interferon and prostaglandin E2. I Oral Pathol. 15:347-351.Warren, R.W., Caster, S.A., Roths, J.B., Murphy, E.D. and Pisetsky, D.S.(1984) The influence of the lpr gene on B cell activation: differential antibodyexpression in lpr congenic mouse strains. Clin. Immunol. Immunopathol.31:65-77.Watanabe, T., Sakai, Y., Miyawaki, S., Shimizu, A., Koiwai, 0. and Ohno, K.(1991) A molecular genetic linkage map of mouse chromosome 19, includingthe lpr, Ly-44, and Tdt genes. Biochem. Genetics 29:325-335.Watanabe-Fukunaga, R., Brannan, C.I., Copeland, N.G., Jenkins, N.A. andNagata, S. (1992) Lymphoproliferation disorder in mice explained by defects inFas antigen that mediates apoptosis. Nature 356:314-317.Waterfield, J.D., Fairhurst, M., Chu, R. and Levy, J.G. (1987)Antigen-specific T-cell hyporesponsiveness in MRL congenic mice can beexplained by two independent cellular defects. Immunol. 61:173-178.Watson, M.L., Rao, J.K., Gilkeson, G.S., Ruiz, P., Eicher, E.M., Pisetsky,D.S., Matsuzawa, A., Rochelle, J.M. and Seiclin, M.F. (1992) Genetic analysisof MRL-lpr mice: relationship of the Fas apoptosis gene to diseasemanifestations and renal disease-modifring loci. J. exp. Med. 176:1645-1656.Wofsy, D., Ledbetter, J.A., Hendler, P.L. and Seaman, WE. (1985) Treatmentof murine lupus with monoclonal anti-T cell antibody. J. Immunol.134:852-857.Wofsy, D., Murphy, E.D., Roths, J.B., Dauphin:ee, M.J., Kipper, S.B. andTalal, N. (1981) Deficient interleukin 2 activity in MRL/Mp and C57BL/6J micebearing the lpr gene. J. exp. Med. 154:1671-1680.Wooley, P.R. (1988) Collagen-induced arthritis in the mouse. Method.Enzymol. 162:361-373.Wooley, P.R. and Chapedelaine, J.M. (1987) Immunogenetics ofcollagen-induced arthritis. CRC Crit.Rev. Immunol. 8:1-22.Wooley, P.R., Seibold, J.S., Whalen, J.D. and Chapdelaine, J.M. (1989)Pristane-induced arthritis: The immunologic and genetic features of anexperimental murine model of autoimmune disease. Arthritis Rheum.32:1022-1030.246Wu, J., Zhou, T., He, J. and Mountz, J.D. (1993) Autoimmune disease in micedue to integration of an endogenous retrovirus in an apoptosis gene. I exp.Med. 178:461-468.Wuthrich, R.P., Yui, M.A., Mazoujian, G., Nabavi, N., Glimcher, L.H. andKelley, V.E. (1989) Enhanced MHC class II expression in renal proximaltubules precedes loss of renal function in MRL/lpr mice with lupus nephritis.Am. J. Pathol. 134:45-51.Yonehara, S., Ishii, A. and Yonehara, M. (1989) A cell-killing monoclonalantibody (anti-Fas) to a cell surface antigen co-downmodulated with thereceptor of tumor necrosis factor. J. exp. Med. 169:1747-1756.Yoo, T.J., Kim, S.Y., Stuart, J.M., Floyd, R.A., Olson, G.A., Cremer, M.A.and Kang, A.H. (1988) Induction of arthritis in monkeys by immunizationwith type II collagen. J. exp. Med. 168:777-782.Yu, L.P.,Jr., Smith, G.N.,Jr., Brandt, K.D. and Kapello, W. (1990) Type XIcollagen-degrading activity in human osteoarthritic cartilage. ArthritisRheum. 33:1626-1633.Zar JH. (1984) Biostatistical analysis. Prentice-Hall, Englewood Cliffs, NJ,185-304.Zhou, T., Bluethmann, H., Eldridge, J., Berry, K. and Mountz, J.D. (1993)Origin of CD4-CD8-B220+ T cells in MRL-lpr/lpr mice. Clues from a T cellreceptor beta transgenic mouse J Immunol 150 365 1-3667Zhou, T, Bluethmann, H, Eldridge, J, Brockhaus, M, Berry, K andMountz, J D (1991) Abnormal thymocyte development and production ofautoreactive T cells in T cell receptor transgenic autoimmune mice JImmunol 147 466-474Ziff, M (1990) Rheumatoid arthntis - its present and future J Rheumatol171 27-33Zouali, M., Migliorini, P. and Stollar, D.B. (1987) Murine lupus anti-DNAantibodies cross-react with the hapten (4-hydroxy-5-iodo-3-nitrophenyl)acetyl,but immunization-induced anti-DNA antibodies do not. Europ. J. Immunol.17:509-513.247


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items