- Email: [email protected]
Autoimmunity and the kidney
Kidney International, Vol. 35 (1989), pp. 923—928
Autoimmunity and the kidney DAVID B.G. OLIVEIRA and D. KEITH PETERS Department of Medicine, Addenbrooke's Hospital, Cambridge, England, United Kingdom
The pathogenesis of autoimmune disease can be considered encoded molecule, and T cell receptor—confer the essential under two broad headings: firstly, the failure of the mechanisms specificity on the immune (or autoimmune) response. Since by which tolerance to self antigens is maintained, and secondly tolerance requires the specific inactivation of T cells, it seems the processes on the effector side of the immune system that likely on first principles that it would also be restricted by MHC
lead to tissue damage once tolerance is broken. In this article molecules, and there is indeed experimental support for this we propose to highlight recent developments, with particular [13, 14]. What is known about these three critical elements will reference to renal autoimmunity, and then to consider some of be discussed with reference to renal autoimmunity. the therapeutic implications of our improved understanding of pathogenesis.
There are three classical mechanisms of tolerance: 1) the
Autoantigens Current knowledge about a number of autoantigens important in renal autoimmune disease is summarized in Table 1. All these
deletion or inactivation of autoreactive clones of lymphocytes; 2) the active suppression of such autoreactive cells; and 3) the autoantigens are defined (or are being defined) by reactivity sequestration or masking of autoantigens so that they are not with autoantibodies. They therefore represent the epitopes (that "visible" to the immune system. Recent experiments have is, the portion of an antigen reacting with antibody or T cell confirmed the existence and advanced our understanding of a T receptor) recognized by B cells, not T cells, which as indicated cell clonal deletion/inactivation mechanism [1—3], whereas evi- above are probably of more fundamental importance. The dence for suppressor systems in the maintenance of tolerance epitopes recognized by B and T cells are likely to be sited on the is, as yet, lacking [4, 5]. However, as considered below, it is too same molecule, but need not be, as inter-molecular intrasoon to conclude that suppression has no part to play in self structural linkage (that is, T cell recognizing one cell component tolerance. helping the B cell response to a separate component of the same A substantial body of work indicates a crucial role for T cells cell) is well documented [21]. in the maintenance of self tolerance. While it seems clear that Progress in defining the targets of autoantibodies can be some degree of tolerance is imposed on B cells [61, self-reactive expected to continue, particularly with the addition of techB cells can readily be demonstrated in normal individuals [7] niques such as the screening of expression eDNA libraries, and may indeed be vital for the correct development of the which has lately been used to identify the targets of autoantiimmune system [81. Although tolerance at the B cell level may bodies in other autoimmune diseases such as primary biliary have some part to play, to understand tolerance it is necessary cirrhosis [22]. to consider the manner in which T cells interact with self (and foreign) antigens. A major difference between the B cell and its MHC
receptor (antibody) on the one hand and the T cell and its receptor on the other, is that the former can interact with free antigen, whereas the latter can only interact with antigen on cell surfaces, bound to products of the major histocompatibility complex (MHC). Understanding of this binding between antigen and MHC products has lately advanced rapidly. It is now clear that the immune response (Ir) gene phenomenon, in which it can be demonstrated that possession of a particular MHC gene allows an immune response to a particular antigen, is due largely to the specific molecular interactions between the two [9]. The features of an antigen that enable this association to occur are emerging, both for interaction with MHC molecules in general [10, II] and for specific MHC molecules [12]. The combination of antigen and MHC molecule is then recognized by the T cell receptor. These three elements—antigen, MHC
© 1989 by the International Society of Nephrology
Two general considerations follow from the strong MHC linkage of most autoimmune diseases (examples in nephrology are the association of DR2 with antiglomerular basement mem-
brane (GBM) disease [23] and of DR3 with membranous nephropathy [24]). Firstly, as it is the T cell that recognizes antigen in the context of the MHC, and not the B cell, this linkage is yet further evidence for the key role of T cells in the pathogenesis of these diseases. Secondly, the MHC association suggests that tolerance to self is only broken to a very limited
number of epitopes, possibly only one. The reason for this deduction is the converse of the explanation for the rarity of Ir
gene effects in the immune response to pathogens in man. Because such complex structures as viruses and bacteria posses
a multitude of different epitopes, it is most unlikely that an individual will be unable to bind some of these with their MHC molecules. It is therefore very rare to be unable to respond, and it is correspondingly difficult to identify which MHC molecules are involved in the response to such complex antigens. The fact
923
924
Oliveira and Peters: Autoimmunity and the kidney
Table 1. Characteristics of some autoantigens involved in renal
disease
Disease
Autoantigen
Anti-glomerular M2 component of basement non-collagenous domain of type IV membrane disease collagen [151 Systemic Neutrophil cytoplasm antigen ? alk. vasculitis phosphatase [17] ? myeloperoxidase
How defined Antibodies
Restricted B cell response Yes 1161
genes does not advance understanding of autoimmunity and renal disease. Antibodies
Yes [17]
3M-I antigen [19]
T cell
Little is known about autoreactive T cells in man as the
[181
Anti-tubular basement membrane disease Membranous nephropathy
is approximately 50% [30], indicating that environmental influences are important. In addition other genetic factors such as Gm allotypes are also involved, and it is therefore unreasonable to suppose that the type of MHC analysis outlined above will be revealing in every case. Conversely, it would be surprising if the application of RFLP analysis and direct sequencing of MHC
Antibodies
Yes [191
technical problems involved in their isolation and growth are formidable. The difficulties are closely related to the problems in identifying autoantigens seen by T cells, as the availability of
one would facilitate the isolation and study of the other. Renal glycoprotein
Antibodies
?
[201
However, such T cells have been defined in myasthenia gravis [311 and autoimmune thyroid disease [32], and the next few
years should see progress in this area. This will reflect the availability of new techniques, for example the growth of autoreactive T cells in the absence of antigen [33] and the use of
that by contrast MHC associations are readily identifiable in many autoimmune diseases indicates the much simpler structure of the corresponding autoantigen, as seen by the autoreactive T cell. Since this aspect (MHC linkage) is central to understanding autoimmunity, it is worth considering how this area has advanced. The analysis of individual MHC disease associations has progressed by a number of steps of increasing sophistication. The first is the serological definition of MHC products, which still provides the bulk of the information. However, a single serological specificity can cover a number of different, albeit related, MHC gene products. More precision is achieved
by the use of restriction fragment length polymorphisms (RFLP). This exploits the fact that different alleles at an MHC locus, which may produce products with the same serological
specificity, can differ in their possession of target sites for restriction enzymes. When DNA is cut by such enzymes and the fragments labelled with probes for the desired locus, the different alleles produce fragments of different sizes which may be identified in a suitable gel separation system. This approach has been used, for example, to identify a particular allele of DQw3 that is strongly associated with rheumatoid arthritis [25]. To our knowledge, this technique has only been applied in the field of renal autoimmunity to membranous nephropathy [26] and failed to produce any evidence for a disease specific allele within the DR3 specificity. However, the existence of such an allele can only be confirmed or excluded by use of the most sophisticated technique, namely the direct sequencing of the DNA of the alleles involved. The feasibility of this has improved dramatically with the development of rapid methods of sequencing [27]. These have not yet been applied to renal autoimmune disease, but have been informative in other cases. Thus there is evidence for a positive association between a disease specific allele at the DQJ3 locus and the autoimmune skin disease pemphigus vulgaris [281. The negative association
(that is, presence of an allele linked with protection from disease) between other such DQp alleles and insulin-dependent diabetes mellitus (IDDM) [29] is considered in the section on suppression. The concordance rate in studies of identical twins with IDDM
such T cells to screen cDNA expression libraries for T cell epitopes [34].
The same techniques that are used for the analysis of the MHC and autoimmunity can in principle be applied to the autoreactive T cell receptor. There is as yet no information on serologically-defined T cell receptor alleles in autoimmunity. These would be analogous to the well defined Gm allotypes on immunoglobulin molecules, such as those associated with antiGBM disease [35]. However, RFLP analysis has demonstrated
that a particular T cell receptor beta chain constant region polymorphism is associated with membranous nephropathy [36]. Direct sequencing of the T cell receptor gene will require cloned autoreactive T cells, which are not as yet available in the renal autoimmune diseases. Active suppression of autoreactivity Little is known about the putative immunoregulatory disorders underlying human renal autoimmunity, but it is an area of great importance—its understanding offers the best chance of developing specific immunotherapies.
Experimental work has identified two major systems involved in immunoregulation: the idiotypic network and the suppressor cell system. Although usually associated with the humoral and cellular components of the immune system, respectively, there is probably a close connection between the two mechanisms. Thus perturbation of the normal development
of B cells profoundly affects the T suppressor cell repertoire [37], and, of relevance to renal autoimmunity, in an experimental model of interstitial nephritis a dominant idiotype is shared
between anti-tubular basement membrane (TBM) antibodies and suppressor cells [38]. Idiotypic regulation Although established in a number of experimental systems [reviewed in 39], evidence for this mechanism in autoimmune disease in man is sparse. The only convincing case in which anti-idiotypic antibodies have been shown to be involved in the regulation of autoimmunity is the control of anti-factor VIII:c autoantibodies [40]. However, other specific attempts to demonstrate anti-idiotypic antibodies in both renal [41] and non-
Oliveira and Peters: Autoimmunity and the kidney
925
renal autoimmunity [42] have failed, despite earlier claims in the
system specialized for antigen presentation (such as macrophages, dendritic cells, and B cells). These molecules act as restriction elements for helper and inducer T cells, which play some form of control via the idiotypic network; once tolerance a key role in the immune response by helping both B cells and is broken at the T cell level there is in principle no reason why cytotoxic T cells. It was observed that tissues such as thyroid the autoreactive T cell should not help a range of B cells specific gland undergoing an autoimmune attack expressed class II for other linked epitopes. That this does not happen in many molecules, which the normal tissue does not. This led to the autoimmune diseases could have explanations other than idio- proposal that such aberrant class II expression allowed T typic control, such as tolerance imposed at the B cell level [6]. helper/inducer cells to recognize previously "invisible" autoanIt is a difficult but important experimental challenge to deter- tigens on the surface of cells. This aberrant expression could be mine the relative importance of the mechanisms responsible for caused by cytokines such as gamma interferon, released as the this restricted response, which, whatever its cause, has thera- result of a viral infection. The resulting autoreactive T cells would then initiate the autoimmune process. It now seems peutic implications. likely that this abnormal expression is not sufficient in itself to Suppression initiate the autoimmune process [54], but may well be important There are many published studies on defects of non-specific in its perpetuation. We are not aware of any data on class II suppression in human autoimmune disease [45]. By contrast, expression in renal autoimmunity, but it would be surprising if the only two well documented examples of antigen-specific the release of mediators such as gamma interferon during the defects that we are aware of involve autoimmune thyroid [46] autoimmune process did not lead to the upregulation of class II and liver disease [47]. However, the association between par- expression. However, the significance of this is difficult to
latter case [43, 44]. However, the very restricted B cell response seen in a number of renal diseases (Table 1) hints at
ticular DQ alleles, namely those possessing an aspartate residue
interpret, as in a rat model of renal transplantation, rejected and
at position 57 of the f3 chain, and resistance to IDDM [29] tolerated grafts had similar increases in class II expression [55]. mentioned previously may be relevant here. There are now a Effector mechanisms in autoimmunity number of examples in which products of the DQ locus are involved in suppressing the immune response to exogenous There are a vast array of effector mechanisms available to the antigens [reviewed in 48], that is, the locus is acting as an immune system, which this paper will not comprehensively immune suppression (Is) gene locus. An attractive explanation cover. One recent idea is that a group of autoimmune diseases for this association is that possession of a particular DQ allele, typified by systemic lupus erythematosus (SLE) is caused acting as an Is gene, leads to suppression of the autoimmune primarily by abnormalities in the effector side of the immune response to the islet cells, and the lack of this gene leads to response, specifically in the handling of immune complexes. susceptibility to autoimmunity. Using this model as a guide, a Clearance of immune complexes re-examination of the MHC associations of autoimmune disAbnormal handling of immune complexes may well be a key eases, searching for similar negative associations, may provide hints as to which diseases are worth investigating for evidence regulatory defect in diseases such as SLE. This hypothesis has been outlined in greater detail elsewhere [56] and will only be of antigen-specific suppression. There has been progress in dissecting the suppressor cell summarized here. In normal individuals, the bulk of immune systems involved in animal models of interstitial autoimmune complexes, solubilized by the complement system, are probarenal disease [49, 50]. An intriguing model is the nephritis found bly carried via the CR1 receptor on red blood cells to the liver in kdkd mice, which is inherited in an autosomal recessive for proteolytic disposal. In the presence of abnormalities of the manner [51]. Analysis has revealed the presence of contra- complement system or receptor, either congenital or acquired, suppressor cells, which act, as their name implies, by reducing complexes are not cleared as efficiently by this route and can the action of suppressor cells, thus facilitating the development deposit in target organs such as the kidney. Here they set up an of autoimmunity. Analogies, both histological and genetic, inflammatory response, which by the release of cytokines and between this animal model and the human medullary cystic autoantigens drives the production of autoantibodies. This disease/juvenile nephronopthisis complex have suggested that leads to the formation of further immune complexes and the similar regulatory abnormalities may underlie the human dis- perpetuation of the cycle. In this type of disease, therefore, ease [52]. there may not be the fundamental breakdown of self tolerance discussed earlier. Instead there is over production of what may Sequestration or masking of autoantigens well be physiological autoantibodies, normally involved in the In its simplest form this long standing concept of tolerance is safe disposal of effete cells and their breakdown products. It is based on the notion that certain bodily structures, for anatom- a defect on the effector side of the immune system that results ical reasons, are not exposed to immuno-competent cells. The in disease. latest variant of this idea involves control of MHC molecule This type of mechanism may well be important in the expression [53]. MHC molecules fall into two classes, class I perpetuation of other autoimmune diseases even if it is not and class II. Class I molecules are expressed on essentially all critically involved in their initiation. For example the polymeric nucleated cells, and act as restriction elements for cytotoxic T IgA and IgA rheumatoid factor found in IgA nephropathy can cells. Their principal role is probably the control of viral inhibit the ability of complement to solubilize immune cominfections, allowing cytotoxic T cells to kill virus-infected host plexes [57]. Although this does not explain the initial abnormalcells. Class II molecules have a more restricted distribution and ities in IgA, it is easy to see how it may contribute to the are normally only expressed on components of the immune resulting tissue damage.
926
Oliveira and Peters: Autoimmunity and the kidney
Cellular mechanisms
role of T cells is again emphasized, but clearly there is some
Although the pathogenic role of autoantibodies in renal [58] and non-renal [59] autoimmune disease has been clear for some time, it is only relatively recently that the importance of a direct role for T cells (as opposed to their indirect effect in helping
form of (presumably) idiotypic connection between the humoral
shown clearly in animal models of renal autoimmunity [50, 60,
1). One possibility is that pooled human gamma globulin may provide a convenient source of anti-idiotypic antibody. This
and cellular systems, Further work is required to define this, but in any event there are important therapeutic possibilities.
By analogy these are most likely to be realized in human
autoantibody formation) has been accepted. This has been diseases showing a similarly focused antibody response (Table 61], and there is also indirect evidence in man [62]. These T cells can cause damage directly via cellular cytotoxic mechanisms, and can also recruit in macrophages which contribute substantially to the increased glomerular cellularity and crescent formation in acute crescentic nephritis [63]. The macrophages and possibly the endogenous glomerular cells can release procoagulant substances which lead to fibrin deposition and damage to the glomerular microcirculation [64). However, the picture is complex, and much more work is required to determine the relative contribution of the various cell types.
material has certainly been successful in the treatment of immune thrombocytopenic purpura [74] and in down-regulating anti-factor VIII antibodies [75].
As an extension of this approach, autoreactive T cells in a number of animal models can be used as "vaccines" to prevent the development of the associated autoimmune disease, possibly via the development of idiotype-specific cytotoxic cells as shown in a rat model of autoimmune encephalomyelitis [76]. Once analogous cells are available in man, a similar form of treatment can be considered. However, given that such cells
would have to be derived anew from each individual, it is Therapeutic implications unlikely that this would ever be of more than theoretical Because of the multiple components involved in the autoiminterest, except perhaps in intractable cases of established mune process, therapy is likely to involve the use of a number autoimmune disease. of different modalities. We will briefly examine some of the Activation of autoantigen-specific suppressor mechanisms is possibilities that arise from considerations of pathogenesis. an attractive therapeutic approach. It has been used to great
Anti-T cell therapy There are a number of different points at which the autoreactive T cell can be attacked. Thus monoclonal antibodies to the mouse CD4 molecule, a marker for the T helper/inducer cell
subset, can reverse advanced lupus in a mouse model [65].
effect in animal models, again in anti-TBM nephritis [77]. The method used is to couple the TBM antigen used to initiate the
autoimmune process to syngeneic lymphocytes, and inject these back into the animal. Antigen specific suppressor cells are induced. These are able to attenuate the renal injury even when
induced at up to six weeks after the onset of the autoimmune process. Because of the use of syngeneic cells, such an approach is in principle feasible in man, although is presumably to interact with class II molecules, it is not surprising that dependent on the use of the correct antigen. anti-class II antibodies can have a similar effect in ameliorating autoimmune disease, including murine lupus [66]. Unfortu"Clearance" therapy nately, attempts to adopt this approach in primates have so far Given the fundamental role of class II molecules in the autoimmune process, and that the role of the CD4 molecule is probably
been limited by severe side effects [66]. Cyclosporin A is another treatment with powerful effects on T cells and which is
clearly effective in a number of animal models [67, 68] and possibly also in human nephrotic syndrome resistant to other modalities of treatment [69]. Prostaglandin E1 also has immunosuppressive properties on T cells [70, 71], and it is possibly via the production of prostaglandins with attenuated inflammatory action that a diet rich in omega-3 fatty acids (as found in fish oils) dramatically reduces the severity of murine lupus [72].
The concepts outlined in Clearance of immune complexes, whereby abnormalities of the complement system cause or exacerbate autoimmunity, have a number of therapeutic conse-
quences. The rational approach to complement deficiency or depletion is replacement therapy. If the antibodies involved in immune complex formation are of a class or subclass that does not fix complement, then their solubilization and clearance will
be inhibited. In this case, immunotherapy would be directed Lastly, interleukin 2 (1L2) is a key molecule involved in the towards inducing the appropriate class switch. proliferation of T cells. High affinity receptors for 1L2 are only expressed on activated T cells, and antibodies directed against these receptors are showing considerable promise in the control of unwanted immune responses (Strom and Kelley, this issue).
Immunoregulatory therapy The aim here is to initiate or boost the mechanisms discussed above that may normally control the immune response. In an experimental model of anti-TBM nephritis mentioned earlier (Table 1) treatment with an anti-idiotype against the major idiotype present on anti-TBM antibodies prevents renal injury [73]. The treatment is also effective if administered two weeks into the disease. T cell responses are attenuated, but the anti-TBM antibody titer is only very slightly lowered. The key
Concluding remarks
There have been a number of developments in our understanding of the mechanisms underlying the establishment of self
tolerance. We now have a clearer picture of the interaction between antigen and MHC gene products that is the basis for the initiation of the immune (or autoimmune) response, and the tools of molecular biology are allowing detailed characterization of autoantigens and autoimmune disease-associated MHC
molecules. Progress on the afferent side of autoimmunity is being matched by the unravelling of the multiple processes that result in subsequent tissue damage. This understanding can be expected to have consequences for therapy.
927
Oliveira and Peters: Autoimmunity and the kidney
Acknowledgments D.B.G. Oliveira is the holder of a Lister Institute Research Fellowship.
anti-tubular basement membrane antibody-associated interstitial nephritis. J Clin invest 77:1143—1 147, 1986 20. NILES J, COLLINS B, BAIRD L, ERIK5ON M, BRADFORD D, PAN G, HSIUNG CK, SCHNEEBERGER B, BHAN A, MCCLUSKEY R: Anti-
bodies reactive with a renal glycoprotein and with deposits in Reprint requests to Dr. D.B.G. Oliveira, Department of Medicine, Addenbrooke's Hospital, Hills Road, Cambridge, CB2 2QQ England, United Kingdom.
membranous nephritis. (abstract) Kidney mt 3 1:338, 1987 21. LAKE P, MITCHI50N NA: Associative control of the immune response to cell surface antigens. Immunol Comm 5:795—805, 1976 22. GERSHWIN ME, MACKAY IR, STURGESS A, COPPEL RL:
Identification and specificity of a cDNA encoding the 70 kD
References
mitochondrial antigen recognised in primary biliary cirrhosis. J I,nmunol 138:3525—3531, 1987
1. KAPPLER JW, ROEHM N, MARRACK P: T cell tolerance by clonal elimination in the thymus. Cell 49:273—280, 1987 2. KAPPLER JW, STAERZ U, WHITE J, MARRACK PC: Self-tolerance
23. REES AJ, PETERS DK, COMPSTON DAS, BATCHELOR JR: Strong
eliminates T cells specific for Mis-modified products of the major histocompatibility complex. Nature 322:35—40, 1988
24. KLOUDA PT, MANOS J, ACHESON EJ, DYER PA, GOLDBY FS, HARRIS R, LAWLER W, MALLICK NP, WILLIAMS 0: Strong asso-
3. MACDONALD HR, SCHNEIDER R, LEES RK, HOWE RC, ACHAORBA H, FESTENSTEIN H, ZINKERNAGEL RM, HENGARTNER H:
ciation between idiopathic membranous nephropathy and HLADRw3. Lancet 11:770—771, 1979
T-cell receptor Vf3 use predicts reactivity and tolerance to Mlsa encoded antigens. Nature 332:40—45, 1988
25. SINGAL DP, D'SOUZA M, REID B, BENSEN WG, KASSAM YB,
4. GAMMON G, DUNN K, SHASTRI N, OKI A, WILBUR 5, SERCARZ
EE: Neonatai T-cell tolerance to minimal immunogenic peptides is caused by clonal inactivation. Nature 319:4l3—415, 1986 5. GAMMON GM, OKI A, SHASTRI N, SERCARZ EE: Induction of tolerance to one determinant on a synthetic peptide does not affect the response to a second linked determinant. Implications for the mechanism of neonatal tolerance. J Exp Med 164:667—672, 1986 6. GOODNOW CC, CRO5BIE J, ADELSTEIN 5, L.avoi TB, SMITH-GILL SJ, BRINK RA, PRITCHARD-BRISCOE H, WOTHERSPOON JS, Lo-
BLAY RH, RAPHAEL K, TRENT Ri, BASTEN A: Altered immuno-
globulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature 334:676—682, 1988 7. MACKAY IR: Natural autoantibodies to the fore—forbidden clones to the rear? Immunol Today 4:340—342, 1983 8. VAKIL M, KEARNEY JF: Functional characterisation of monoclonal auto-anti-idiotype antibodies isolated from the early B cell reper-
toire of BALB/c mice. Eurflmmunol 16:1151—I 158,
1986
9. Buus 5, SETTE A, COLON SM, MILES C, GREY HM: The relation between major histocompatibility complex (MHC) restriction and the capacity of Ia to bind immunogenic peptides. Science 235:1353—
association between HLA DRw2 and antibody mediated Goodpasture's syndrome. Lancet 1:966—968, 1978
ADACHI JD: HLA-DQ beta-chain polymorphism in HLA-DR4 haplotypes associated with rheumatoid arthritis. Lancet II: 1118— 1120, 1987
26. SACKS SH, BUSt-tELL A, RUST NA, KARAGIANNIS JA, JEWELL DP,
LEIMNGHAM JG, WOOD KJ, MCM!CHAEL AJ: Functional and biochemical subtypes of the haplotype HLA-DR3 in patients with celiac disease or idiopathic membranous nephropathy. Hum Immunol 20:175—187, 1987 27. SAIKI RK, BUGAWAN TL, HORN QT, MuLLI5 KB, ERLICK HA: Analysis of enzymatically amplified /3-globin and HLA-DQ-a DNA
with allele-specific oligonucleotide probes. Nature
324:163—166,
1986
28. SINHA AA, BRAUTBAR C, SZAFER F, FRIEDMANN A, TZFONI B,
TODD JA, STEINMAN L, MCDEVITT HO: A newly characterised HLA DQ-/3 allele associated with pemphigus vulgaris. Science 239: 1026—1029, 1988
29. TODD JA, BELL JI, MCDEVITT HO: HLA-DQ-/3 gene contributes to
susceptibility and resistance to insulin-dependent diabetes mellitus.
Nature 329:599—604, 1987 30. TATTERSALL RB, PYKE DA: Diabetes in identical twins. Lancet
II:
1120—1125, 1972
1358, 1987 10. MARGALIT H, SPOUGE JL, CORNETTE JL, CEASE KB, DELI5I C, BER.ZOFSKY JA: Prediction of immunodominant helper T cell anti-
31. HOHLFELD R, TOYKA KY, HEININGER K, GROSSE-WILUE H, KALIES I: Autoimmune human T lymphocytes specific for acetyl-
genic Sites from the primary sequence. J Immunol 138:2213—2229,
32. LONDEI M, BOTTAZZO GF, FELDMANN M: Human T-cell clones from autoimmune thyroid glands: specific recognition of autologous
1987 11. ROTHBARD J, TOWNSEND A, EDWARDS M, TAYLOR W: Pattern
choline receptor. Nature thyroid cells. Science
310:244—246, 1984
228:85—89, 1985
recognition among T cell epitopes. Hamatol Bluttransfus 31:324—
33. PADULA 5, POLLARD MK, LUIGENHELD EG, CLARK RB: Mainte-
331, 1987 12. SETTE A, BUUS S, COLON S, MILES C, GREY HM: I-A<'-binding
nance of antigen specificity by human interleukin-2-dependent T cell lines. Use of antigen-presenting cells and OKT3 antibody in the absence of antigen. J Clin Invest 74:788—797, 1985
peptides derived from unrelated protein antigens share a common structural motif. J Immunol 141:45-48, 1988
34. GRIFFITHS J, MITCHISON NA, OLIvEIRA DBG, SALGAME P: New
13. MATZINGER P, ZAMOYSKA R, WALDMANN H: Self tolerance is
developments in epitope analysis, in Progress in Vaccinology I,
H-2-restricted. Nature 308:738—741, 1984 14. RAMMENSEE H-G, BEVAN MJ: Evidence from in vitro studies that
tolerance to self antigens is MHC-restricted. Nature
308:741—744,
1984 15. WIESLANDER J, KATAJA M, HUDSON BG: Characterisation of the human Goodpasture antigen. C/in Exp Immunol 69:332—340, 1987 16. PRESSEY A, PUSEY CD, DASH A, PETERS DK, LOCKWOOD CM:
Production of a monoclonal antibody to autoantigenic components of human glomerular basement membrane. C/in Exp Immuno/ 54: 178—184,
1983
17. LOCKWOOD CM, BAKES D, JONES S, WHITAKER KB, Moss DW, SAVAGE COS: Association of alkaline phosphatase with an autoan-
tigen recognised by circulating anti-neutrophil antibodies in systemic vasculitis. Lancet 1:716—719, 1987 18. FALK RJ, JENNETTE JC: Anti-neutrophil cytoplasmic autoantibod-
ies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotising and crescentic glomerulonephritis. NEnglJMed3l8:1651—1657, 1988 19. CLAYMAN MD, MICHAUD L, BRENTJENS J, ANDRES GA, KEFA-
LIDES NA, NEILSON EG: Isolation of the target antigen of human
edited by TALWAR GP, New York, Springer-Verlag, 1988; pp. 359— 367
35. REES AJ, DEMAINE AG, WELSH K!: The association between immunoglobulin allotypes with auto-antibodies to glomerular basement membrane and their titre. Hum Immunol 10:213—221, 1984 36. DEMAINE AG, VAUGHAN RW, TAUBE DH, WELSH KI: Association
of membranous nephropathy with T-cell receptor constant beta chain and immunoglobulin heavy chain switch region polymorphisms. Immunogenetics 27:19-23, 1988 37. HAYGLASS KT, BENACERRAF B, SY M-S: T cell development in B
cell-deficient mice. V. Stopping anti-p treatment results in Ighrestricted expansion of the T suppressor cell repertoire concomitant with the development of normal immunogiobulin levels. J Exp Med 164:36—49, 1986
38. NEILSON EG, MCCAFFERTY E, MANN R, MICHAUD L, CLAYMAN
M: Tubular antigen-derivatized cells induce a disease-protective, antigen-specific, and idiotype-specific suppressor T cell network restricted by I-J and 1gb-V in mice with experimental interstitial nephritis. J Exp Med 162:215—230, 1985 39. PUSEY CD, YENNING MC, PETERS DK: Immunopathology of
928
Oliveira and Peters: Autoimmunity and the kidney
glomerular and interstitial disease, in Diseases of the Kidney (4th ed), edited by SCHRIER RW, GOTTSCHALK CW, Boston/Toronto, Little, Brown and Company, 1988, pp. 1827—1883 40. SULTAN Y, Rossl F, KAZATCHKINE MD: Recovery from anti-VILI:
c (anti-hemophilic factor) autoimmune disease is dependant on generation of anti-idiotypes against anti-V1II:c autoantibodies. Proc Nat! Acad Sci USA 84:828—831, 1987 41. SAVAGE COS: Regulation of autoantibody production in man. The role of idiotype-antiidiotype network interactions in anti-glomerular
basement membrane antibody mediated disease. PhD Thesis, University of London, 1986 42. VINCENT AC: Are spontaneous anti-idiotypic antibodies against
anti-acetylcholine receptor antibodies present in myasthenia gravis? J Autoimmunity 1:131—142, 1988 43. DWYER DS, BRADLEY CK, URQUHART CK, KEARNEY iF: Natu-
rally occurring anti-idiotypic antibodies in myasthenia gravis patients. Nature 301:611—614, 1983 44. LEFVERT AK, PIRSKANEN R, SVANBORG E: Anti-idiotypic antibod-
ies, acetylcholine receptor antibodies and disturbed neuromuscular function in healthy relatives to patients with myasthenia gravis. J Neuroimmunol 9:41—53, 1985
45. MILLER KB, SCHWARTZ RS: Autoimmunity and suppressor T lymphocytes. Adv Intern Med 27:281—313, 1982
46. OKITA N, ROW VW, VOLPE R: Suppressor T lymphocyte defi-
ciency in Graves' disease and Hashimoto's thyroiditis. J C/in Endocrinol Metab 52:528-533, 1981 47. VENTO 5, O'BRIEN CJ, MCFARLANE 10, WILLIAMS R, EDDLESTON
59. TOYKA Ky, DRACHMAN DB, PESTRONK A, KAO I: Myasthenia gravis: Passive transfer from man to mouse. Science 190:397—399, 1975
60. BANNISTER KM, ULICH TR, WILSON CB: Induction, characteriza-
tion, and cell transfer of autoimmune tubulointestinal nephritis. Kidney mt 32:642-651, 1987 61. MIHARA M, OHSUGI Y, Sn'o K, MIYAI T, TOGASH M, ONO S, MURAKAMI S, DOBASH K, HIRAYAMA F, HAMAOKA T: Immuno-
logic abnormality in NZB/NZW F1 mice: Thymus-independent occurrence of B cell abnormality and requirement for T cells in the development of autoimmune disease, as evidenced by an analysis of the athymic nude individuals. J Immunol 141:85—90, 1988 62. NEALE Ti, TIPING P0, CARSON SD, HOLDSWORTH SR: Participa-
tion of cell-mediated immunity in deposition of fibrin in glomerulonephritis. Lancet 11:421—424, 1988 63. ATKINS RC, HOLDSWORTH SR, HANCOCK WH, THOMSON NM, GLASGOW EF: Cellular immune mechanisms in human glomerulo-
nephritis: The role of mononuclear leucocytes. Springer Semin Immunopathol 5:269—296, 1982
64. BRENTJENS JR: Glomerular procoagulant activity and glomerulone-
phritis. Lab Invest 57:107—111, 1987 65. WOFSEY D, SEAMAN WE: Reversal of advanced murine lupus in
NZB/NZW Fl mice by treatment with monoclonal antibody to L3T4. J Immunol 138:3247—3253, 1987 66. MCDEVITT HO, PERRY R, STEINMAN LA: Monoclonal anti-Ia
antibody therapy in animal models of autoimmune disease, in Autoimmunity and Autoimmune Disease, New York, Wiley, Chichester (Ciba Foundation Symposium 129), 1987, pp. 184—193
ALWF: T-cell inducers of suppressor lymphocytes control liver-
67. MOUNTZ JD, SMITH HR, WILDER RL, REEVES JP, STEINBERG AD:
directed autoimmunity. Lancet 1:886—888, 1987 48. OLIVEIRA DBG, MITCHISON NA: Immune suppression genes. Gun Exp Immuno! (in press)
CS-A therapy in MLR-lpr/lpr mice: Amelioration of immunopathology despite autoantibody production. J Immunol 138:157—163, 1987 68. GIMENEZ A, LEYVA-COBIAN F, FIERRO C, RIO M, BRIcIO T, MAMPASO F: Effect of cyclosporin A on autoimmune tubulointerstitial nephritis in the brown Norway rat. C/in Exp Immunol 69:550— 556, 1987 69. MEYRIER A: Treatment with cyclosporin of patients with idiopathic nephrotic syndrome. Springer Semin Immunopathol 9:441—450,
49. UEDA S, WAKASHIN M, WAKASHIN Y, MORI T, YOSHIDA H, MORI
Y, IESATO K, OGAWA M, AZEMOTO R, KATO I, OKUDA K:
Suppressor system in murine intestitial nephritis. Analysis of tubular basement membrane (TBM)-specific suppressor T cells and their soluble factor in C57BL/6 mice using a syngeneic system. C/in Immunol Immunopathol 45:78—91, 1987
1987
70. HINGLAIS N, PELLETIER L, VIAL MC, SAPIN C, BASCOU C, SUN MJ, NEILSON EG: Effector T cell differentiation in experimenBARIETY J, KELLEY V, DRUET P: Effect of prostaglandin El in tal interstitial nephritis. I. The development and modulation of brown Norway rats with mercury-induced autoimmune disease.
50. MANN R, KELLY CJ, HINES WH, CLAYMAN MD, BLANCHARD N,
effector lymphocyte maturation by I-J+ regulatory T cells. J Immunol 138:4200—4208, 1987
51. KELLY CJ, NEILSON EG: Contrasuppression in autoimmunity. Abnormal contrasuppression facilitates expression of nephritogenic effector T cells and interstitial nephritis in kdkd mice. J Exp Med 165:107—123, 1987
52. KELLY CJ, NEILSON EG: Medullary cystic disease. An inherited form of autoimmune interstitial nephritis? Am J Kidney Dis 10:389— 395, 1987 53. Bormzzo GF, PUJOL-BORRELL R, HANAFUSA T, FELDMANN M:
Role of aberrant HLA-DR expression and antigen presentation in
the induction of endocrine autoimmunity. Lancet 11:1115—1119, 1983
54. FELDMANN M: Probing human autoimmune diseases. Immunology (in press) 55. Woot KJ, HOPLEY A, DALLMAN MJ, MORRIS PJ: Lack of corre-
lation between the induction of donor class I and class II major histocompatibility complex antigens and graft rejection. Transplantation 45:759—767, 1988
C/in Immunol Immunopatho/ 40:401—409, 1986 71. KELLY Ci, ZURIER RB, KRAKAUER KA, BLANCHARD N, NEILSON
EG: Prostaglandin El inhibits effector T cell induction and tissue damage in experimental murine interstitial nephritis. J C/in In vest 79:782—779, 1987 72. ALEXANDER NJ, SMYTHE NL, JOKINEN MP: The type of dietary fat
affects the severity of autoimmune disease in NZB/NZW mice. Am JPatho! 127:106—121, 1987
73. CLAYMAN MD, SUN MJ, MICHAUD L, BRILL-DASHOFF J, RIBLET
R, NEILSON EG: Clonotypic heterogeneity in experimental interstitial nephritis, Restricted specificity of the anti-tubular basement membrane B cell repertoire is associated with a disease-modifying crossreactive idiotype. J Exp Med 167:1296—1312, 1988 74. IMBACH P, BARANDUN S, D'APUZO V, BAUMGARTNER C, HIRT A, MORELL A, ROSSI E, SCHONI M, VEST M, WAGNER HP: High-dose
intravenous gammaglobulin for idiopathic thrombocytopenic purpura in childhood. Lancet 1:1228—1231, 1981 75. SULTAN Y, KAZATCHKINE MD, MAISONNEUVE P. NYDEGGER UE:
Anti-idiotypic suppression of autoantibodies to factor VIII (anti-
56. LACHMANN PJ: Complement—friend or foe? Br J Rheumato! 26:
haemophilic factor) by high-dose intravenous gammaglobulin. Lan-
409—415, 1987 57. SCHENA PF, PASTORE A, SINICO RA, MONTINARO V, FORNASIERI
76. SUN D, QIN Y, CHLUBA J, EPPLEN JT, WEKERLE H: Suppression
A: Polymeric IgA and IgA rheumatoid factor decrease the capacity of serum to solubilize circulating immune complexes in patients with primary IgA nephropathy. J Immunol 141:125—130, 1988 58. LERNER RA, GLASSOCK RJ, DIXON FJ: The role of anti-glomerular
basement membrane antibody in the pathogenesis of human gbmerulonephritis. J Exp Med 126:989—1004, 1967
cet
11:765—768, 1984
of experimentally induced autoimmune encephalomyelitis by cytolytic T-T cell interactions. Nature 332:843—845, 1988
77. MANN R, NEILSON EG: Murine interstitial nephritis. V. The auto-induction of antigen-specific Lyt-2 suppressor T cells diminishes the expression of interstitial nephritis in mice with antitubular basement membrane disease. J Immuno! 136:908—912, 1986
Autoimmunity and the kidney DAVID B.G. OLIVEIRA and D. KEITH PETERS Department of Medicine, Addenbrooke's Hospital, Cambridge, England, United Kingdom
The pathogenesis of autoimmune disease can be considered encoded molecule, and T cell receptor—confer the essential under two broad headings: firstly, the failure of the mechanisms specificity on the immune (or autoimmune) response. Since by which tolerance to self antigens is maintained, and secondly tolerance requires the specific inactivation of T cells, it seems the processes on the effector side of the immune system that likely on first principles that it would also be restricted by MHC
lead to tissue damage once tolerance is broken. In this article molecules, and there is indeed experimental support for this we propose to highlight recent developments, with particular [13, 14]. What is known about these three critical elements will reference to renal autoimmunity, and then to consider some of be discussed with reference to renal autoimmunity. the therapeutic implications of our improved understanding of pathogenesis.
There are three classical mechanisms of tolerance: 1) the
Autoantigens Current knowledge about a number of autoantigens important in renal autoimmune disease is summarized in Table 1. All these
deletion or inactivation of autoreactive clones of lymphocytes; 2) the active suppression of such autoreactive cells; and 3) the autoantigens are defined (or are being defined) by reactivity sequestration or masking of autoantigens so that they are not with autoantibodies. They therefore represent the epitopes (that "visible" to the immune system. Recent experiments have is, the portion of an antigen reacting with antibody or T cell confirmed the existence and advanced our understanding of a T receptor) recognized by B cells, not T cells, which as indicated cell clonal deletion/inactivation mechanism [1—3], whereas evi- above are probably of more fundamental importance. The dence for suppressor systems in the maintenance of tolerance epitopes recognized by B and T cells are likely to be sited on the is, as yet, lacking [4, 5]. However, as considered below, it is too same molecule, but need not be, as inter-molecular intrasoon to conclude that suppression has no part to play in self structural linkage (that is, T cell recognizing one cell component tolerance. helping the B cell response to a separate component of the same A substantial body of work indicates a crucial role for T cells cell) is well documented [21]. in the maintenance of self tolerance. While it seems clear that Progress in defining the targets of autoantibodies can be some degree of tolerance is imposed on B cells [61, self-reactive expected to continue, particularly with the addition of techB cells can readily be demonstrated in normal individuals [7] niques such as the screening of expression eDNA libraries, and may indeed be vital for the correct development of the which has lately been used to identify the targets of autoantiimmune system [81. Although tolerance at the B cell level may bodies in other autoimmune diseases such as primary biliary have some part to play, to understand tolerance it is necessary cirrhosis [22]. to consider the manner in which T cells interact with self (and foreign) antigens. A major difference between the B cell and its MHC
receptor (antibody) on the one hand and the T cell and its receptor on the other, is that the former can interact with free antigen, whereas the latter can only interact with antigen on cell surfaces, bound to products of the major histocompatibility complex (MHC). Understanding of this binding between antigen and MHC products has lately advanced rapidly. It is now clear that the immune response (Ir) gene phenomenon, in which it can be demonstrated that possession of a particular MHC gene allows an immune response to a particular antigen, is due largely to the specific molecular interactions between the two [9]. The features of an antigen that enable this association to occur are emerging, both for interaction with MHC molecules in general [10, II] and for specific MHC molecules [12]. The combination of antigen and MHC molecule is then recognized by the T cell receptor. These three elements—antigen, MHC
© 1989 by the International Society of Nephrology
Two general considerations follow from the strong MHC linkage of most autoimmune diseases (examples in nephrology are the association of DR2 with antiglomerular basement mem-
brane (GBM) disease [23] and of DR3 with membranous nephropathy [24]). Firstly, as it is the T cell that recognizes antigen in the context of the MHC, and not the B cell, this linkage is yet further evidence for the key role of T cells in the pathogenesis of these diseases. Secondly, the MHC association suggests that tolerance to self is only broken to a very limited
number of epitopes, possibly only one. The reason for this deduction is the converse of the explanation for the rarity of Ir
gene effects in the immune response to pathogens in man. Because such complex structures as viruses and bacteria posses
a multitude of different epitopes, it is most unlikely that an individual will be unable to bind some of these with their MHC molecules. It is therefore very rare to be unable to respond, and it is correspondingly difficult to identify which MHC molecules are involved in the response to such complex antigens. The fact
923
924
Oliveira and Peters: Autoimmunity and the kidney
Table 1. Characteristics of some autoantigens involved in renal
disease
Disease
Autoantigen
Anti-glomerular M2 component of basement non-collagenous domain of type IV membrane disease collagen [151 Systemic Neutrophil cytoplasm antigen ? alk. vasculitis phosphatase [17] ? myeloperoxidase
How defined Antibodies
Restricted B cell response Yes 1161
genes does not advance understanding of autoimmunity and renal disease. Antibodies
Yes [17]
3M-I antigen [19]
T cell
Little is known about autoreactive T cells in man as the
[181
Anti-tubular basement membrane disease Membranous nephropathy
is approximately 50% [30], indicating that environmental influences are important. In addition other genetic factors such as Gm allotypes are also involved, and it is therefore unreasonable to suppose that the type of MHC analysis outlined above will be revealing in every case. Conversely, it would be surprising if the application of RFLP analysis and direct sequencing of MHC
Antibodies
Yes [191
technical problems involved in their isolation and growth are formidable. The difficulties are closely related to the problems in identifying autoantigens seen by T cells, as the availability of
one would facilitate the isolation and study of the other. Renal glycoprotein
Antibodies
?
[201
However, such T cells have been defined in myasthenia gravis [311 and autoimmune thyroid disease [32], and the next few
years should see progress in this area. This will reflect the availability of new techniques, for example the growth of autoreactive T cells in the absence of antigen [33] and the use of
that by contrast MHC associations are readily identifiable in many autoimmune diseases indicates the much simpler structure of the corresponding autoantigen, as seen by the autoreactive T cell. Since this aspect (MHC linkage) is central to understanding autoimmunity, it is worth considering how this area has advanced. The analysis of individual MHC disease associations has progressed by a number of steps of increasing sophistication. The first is the serological definition of MHC products, which still provides the bulk of the information. However, a single serological specificity can cover a number of different, albeit related, MHC gene products. More precision is achieved
by the use of restriction fragment length polymorphisms (RFLP). This exploits the fact that different alleles at an MHC locus, which may produce products with the same serological
specificity, can differ in their possession of target sites for restriction enzymes. When DNA is cut by such enzymes and the fragments labelled with probes for the desired locus, the different alleles produce fragments of different sizes which may be identified in a suitable gel separation system. This approach has been used, for example, to identify a particular allele of DQw3 that is strongly associated with rheumatoid arthritis [25]. To our knowledge, this technique has only been applied in the field of renal autoimmunity to membranous nephropathy [26] and failed to produce any evidence for a disease specific allele within the DR3 specificity. However, the existence of such an allele can only be confirmed or excluded by use of the most sophisticated technique, namely the direct sequencing of the DNA of the alleles involved. The feasibility of this has improved dramatically with the development of rapid methods of sequencing [27]. These have not yet been applied to renal autoimmune disease, but have been informative in other cases. Thus there is evidence for a positive association between a disease specific allele at the DQJ3 locus and the autoimmune skin disease pemphigus vulgaris [281. The negative association
(that is, presence of an allele linked with protection from disease) between other such DQp alleles and insulin-dependent diabetes mellitus (IDDM) [29] is considered in the section on suppression. The concordance rate in studies of identical twins with IDDM
such T cells to screen cDNA expression libraries for T cell epitopes [34].
The same techniques that are used for the analysis of the MHC and autoimmunity can in principle be applied to the autoreactive T cell receptor. There is as yet no information on serologically-defined T cell receptor alleles in autoimmunity. These would be analogous to the well defined Gm allotypes on immunoglobulin molecules, such as those associated with antiGBM disease [35]. However, RFLP analysis has demonstrated
that a particular T cell receptor beta chain constant region polymorphism is associated with membranous nephropathy [36]. Direct sequencing of the T cell receptor gene will require cloned autoreactive T cells, which are not as yet available in the renal autoimmune diseases. Active suppression of autoreactivity Little is known about the putative immunoregulatory disorders underlying human renal autoimmunity, but it is an area of great importance—its understanding offers the best chance of developing specific immunotherapies.
Experimental work has identified two major systems involved in immunoregulation: the idiotypic network and the suppressor cell system. Although usually associated with the humoral and cellular components of the immune system, respectively, there is probably a close connection between the two mechanisms. Thus perturbation of the normal development
of B cells profoundly affects the T suppressor cell repertoire [37], and, of relevance to renal autoimmunity, in an experimental model of interstitial nephritis a dominant idiotype is shared
between anti-tubular basement membrane (TBM) antibodies and suppressor cells [38]. Idiotypic regulation Although established in a number of experimental systems [reviewed in 39], evidence for this mechanism in autoimmune disease in man is sparse. The only convincing case in which anti-idiotypic antibodies have been shown to be involved in the regulation of autoimmunity is the control of anti-factor VIII:c autoantibodies [40]. However, other specific attempts to demonstrate anti-idiotypic antibodies in both renal [41] and non-
Oliveira and Peters: Autoimmunity and the kidney
925
renal autoimmunity [42] have failed, despite earlier claims in the
system specialized for antigen presentation (such as macrophages, dendritic cells, and B cells). These molecules act as restriction elements for helper and inducer T cells, which play some form of control via the idiotypic network; once tolerance a key role in the immune response by helping both B cells and is broken at the T cell level there is in principle no reason why cytotoxic T cells. It was observed that tissues such as thyroid the autoreactive T cell should not help a range of B cells specific gland undergoing an autoimmune attack expressed class II for other linked epitopes. That this does not happen in many molecules, which the normal tissue does not. This led to the autoimmune diseases could have explanations other than idio- proposal that such aberrant class II expression allowed T typic control, such as tolerance imposed at the B cell level [6]. helper/inducer cells to recognize previously "invisible" autoanIt is a difficult but important experimental challenge to deter- tigens on the surface of cells. This aberrant expression could be mine the relative importance of the mechanisms responsible for caused by cytokines such as gamma interferon, released as the this restricted response, which, whatever its cause, has thera- result of a viral infection. The resulting autoreactive T cells would then initiate the autoimmune process. It now seems peutic implications. likely that this abnormal expression is not sufficient in itself to Suppression initiate the autoimmune process [54], but may well be important There are many published studies on defects of non-specific in its perpetuation. We are not aware of any data on class II suppression in human autoimmune disease [45]. By contrast, expression in renal autoimmunity, but it would be surprising if the only two well documented examples of antigen-specific the release of mediators such as gamma interferon during the defects that we are aware of involve autoimmune thyroid [46] autoimmune process did not lead to the upregulation of class II and liver disease [47]. However, the association between par- expression. However, the significance of this is difficult to
latter case [43, 44]. However, the very restricted B cell response seen in a number of renal diseases (Table 1) hints at
ticular DQ alleles, namely those possessing an aspartate residue
interpret, as in a rat model of renal transplantation, rejected and
at position 57 of the f3 chain, and resistance to IDDM [29] tolerated grafts had similar increases in class II expression [55]. mentioned previously may be relevant here. There are now a Effector mechanisms in autoimmunity number of examples in which products of the DQ locus are involved in suppressing the immune response to exogenous There are a vast array of effector mechanisms available to the antigens [reviewed in 48], that is, the locus is acting as an immune system, which this paper will not comprehensively immune suppression (Is) gene locus. An attractive explanation cover. One recent idea is that a group of autoimmune diseases for this association is that possession of a particular DQ allele, typified by systemic lupus erythematosus (SLE) is caused acting as an Is gene, leads to suppression of the autoimmune primarily by abnormalities in the effector side of the immune response to the islet cells, and the lack of this gene leads to response, specifically in the handling of immune complexes. susceptibility to autoimmunity. Using this model as a guide, a Clearance of immune complexes re-examination of the MHC associations of autoimmune disAbnormal handling of immune complexes may well be a key eases, searching for similar negative associations, may provide hints as to which diseases are worth investigating for evidence regulatory defect in diseases such as SLE. This hypothesis has been outlined in greater detail elsewhere [56] and will only be of antigen-specific suppression. There has been progress in dissecting the suppressor cell summarized here. In normal individuals, the bulk of immune systems involved in animal models of interstitial autoimmune complexes, solubilized by the complement system, are probarenal disease [49, 50]. An intriguing model is the nephritis found bly carried via the CR1 receptor on red blood cells to the liver in kdkd mice, which is inherited in an autosomal recessive for proteolytic disposal. In the presence of abnormalities of the manner [51]. Analysis has revealed the presence of contra- complement system or receptor, either congenital or acquired, suppressor cells, which act, as their name implies, by reducing complexes are not cleared as efficiently by this route and can the action of suppressor cells, thus facilitating the development deposit in target organs such as the kidney. Here they set up an of autoimmunity. Analogies, both histological and genetic, inflammatory response, which by the release of cytokines and between this animal model and the human medullary cystic autoantigens drives the production of autoantibodies. This disease/juvenile nephronopthisis complex have suggested that leads to the formation of further immune complexes and the similar regulatory abnormalities may underlie the human dis- perpetuation of the cycle. In this type of disease, therefore, ease [52]. there may not be the fundamental breakdown of self tolerance discussed earlier. Instead there is over production of what may Sequestration or masking of autoantigens well be physiological autoantibodies, normally involved in the In its simplest form this long standing concept of tolerance is safe disposal of effete cells and their breakdown products. It is based on the notion that certain bodily structures, for anatom- a defect on the effector side of the immune system that results ical reasons, are not exposed to immuno-competent cells. The in disease. latest variant of this idea involves control of MHC molecule This type of mechanism may well be important in the expression [53]. MHC molecules fall into two classes, class I perpetuation of other autoimmune diseases even if it is not and class II. Class I molecules are expressed on essentially all critically involved in their initiation. For example the polymeric nucleated cells, and act as restriction elements for cytotoxic T IgA and IgA rheumatoid factor found in IgA nephropathy can cells. Their principal role is probably the control of viral inhibit the ability of complement to solubilize immune cominfections, allowing cytotoxic T cells to kill virus-infected host plexes [57]. Although this does not explain the initial abnormalcells. Class II molecules have a more restricted distribution and ities in IgA, it is easy to see how it may contribute to the are normally only expressed on components of the immune resulting tissue damage.
926
Oliveira and Peters: Autoimmunity and the kidney
Cellular mechanisms
role of T cells is again emphasized, but clearly there is some
Although the pathogenic role of autoantibodies in renal [58] and non-renal [59] autoimmune disease has been clear for some time, it is only relatively recently that the importance of a direct role for T cells (as opposed to their indirect effect in helping
form of (presumably) idiotypic connection between the humoral
shown clearly in animal models of renal autoimmunity [50, 60,
1). One possibility is that pooled human gamma globulin may provide a convenient source of anti-idiotypic antibody. This
and cellular systems, Further work is required to define this, but in any event there are important therapeutic possibilities.
By analogy these are most likely to be realized in human
autoantibody formation) has been accepted. This has been diseases showing a similarly focused antibody response (Table 61], and there is also indirect evidence in man [62]. These T cells can cause damage directly via cellular cytotoxic mechanisms, and can also recruit in macrophages which contribute substantially to the increased glomerular cellularity and crescent formation in acute crescentic nephritis [63]. The macrophages and possibly the endogenous glomerular cells can release procoagulant substances which lead to fibrin deposition and damage to the glomerular microcirculation [64). However, the picture is complex, and much more work is required to determine the relative contribution of the various cell types.
material has certainly been successful in the treatment of immune thrombocytopenic purpura [74] and in down-regulating anti-factor VIII antibodies [75].
As an extension of this approach, autoreactive T cells in a number of animal models can be used as "vaccines" to prevent the development of the associated autoimmune disease, possibly via the development of idiotype-specific cytotoxic cells as shown in a rat model of autoimmune encephalomyelitis [76]. Once analogous cells are available in man, a similar form of treatment can be considered. However, given that such cells
would have to be derived anew from each individual, it is Therapeutic implications unlikely that this would ever be of more than theoretical Because of the multiple components involved in the autoiminterest, except perhaps in intractable cases of established mune process, therapy is likely to involve the use of a number autoimmune disease. of different modalities. We will briefly examine some of the Activation of autoantigen-specific suppressor mechanisms is possibilities that arise from considerations of pathogenesis. an attractive therapeutic approach. It has been used to great
Anti-T cell therapy There are a number of different points at which the autoreactive T cell can be attacked. Thus monoclonal antibodies to the mouse CD4 molecule, a marker for the T helper/inducer cell
subset, can reverse advanced lupus in a mouse model [65].
effect in animal models, again in anti-TBM nephritis [77]. The method used is to couple the TBM antigen used to initiate the
autoimmune process to syngeneic lymphocytes, and inject these back into the animal. Antigen specific suppressor cells are induced. These are able to attenuate the renal injury even when
induced at up to six weeks after the onset of the autoimmune process. Because of the use of syngeneic cells, such an approach is in principle feasible in man, although is presumably to interact with class II molecules, it is not surprising that dependent on the use of the correct antigen. anti-class II antibodies can have a similar effect in ameliorating autoimmune disease, including murine lupus [66]. Unfortu"Clearance" therapy nately, attempts to adopt this approach in primates have so far Given the fundamental role of class II molecules in the autoimmune process, and that the role of the CD4 molecule is probably
been limited by severe side effects [66]. Cyclosporin A is another treatment with powerful effects on T cells and which is
clearly effective in a number of animal models [67, 68] and possibly also in human nephrotic syndrome resistant to other modalities of treatment [69]. Prostaglandin E1 also has immunosuppressive properties on T cells [70, 71], and it is possibly via the production of prostaglandins with attenuated inflammatory action that a diet rich in omega-3 fatty acids (as found in fish oils) dramatically reduces the severity of murine lupus [72].
The concepts outlined in Clearance of immune complexes, whereby abnormalities of the complement system cause or exacerbate autoimmunity, have a number of therapeutic conse-
quences. The rational approach to complement deficiency or depletion is replacement therapy. If the antibodies involved in immune complex formation are of a class or subclass that does not fix complement, then their solubilization and clearance will
be inhibited. In this case, immunotherapy would be directed Lastly, interleukin 2 (1L2) is a key molecule involved in the towards inducing the appropriate class switch. proliferation of T cells. High affinity receptors for 1L2 are only expressed on activated T cells, and antibodies directed against these receptors are showing considerable promise in the control of unwanted immune responses (Strom and Kelley, this issue).
Immunoregulatory therapy The aim here is to initiate or boost the mechanisms discussed above that may normally control the immune response. In an experimental model of anti-TBM nephritis mentioned earlier (Table 1) treatment with an anti-idiotype against the major idiotype present on anti-TBM antibodies prevents renal injury [73]. The treatment is also effective if administered two weeks into the disease. T cell responses are attenuated, but the anti-TBM antibody titer is only very slightly lowered. The key
Concluding remarks
There have been a number of developments in our understanding of the mechanisms underlying the establishment of self
tolerance. We now have a clearer picture of the interaction between antigen and MHC gene products that is the basis for the initiation of the immune (or autoimmune) response, and the tools of molecular biology are allowing detailed characterization of autoantigens and autoimmune disease-associated MHC
molecules. Progress on the afferent side of autoimmunity is being matched by the unravelling of the multiple processes that result in subsequent tissue damage. This understanding can be expected to have consequences for therapy.
927
Oliveira and Peters: Autoimmunity and the kidney
Acknowledgments D.B.G. Oliveira is the holder of a Lister Institute Research Fellowship.
anti-tubular basement membrane antibody-associated interstitial nephritis. J Clin invest 77:1143—1 147, 1986 20. NILES J, COLLINS B, BAIRD L, ERIK5ON M, BRADFORD D, PAN G, HSIUNG CK, SCHNEEBERGER B, BHAN A, MCCLUSKEY R: Anti-
bodies reactive with a renal glycoprotein and with deposits in Reprint requests to Dr. D.B.G. Oliveira, Department of Medicine, Addenbrooke's Hospital, Hills Road, Cambridge, CB2 2QQ England, United Kingdom.
membranous nephritis. (abstract) Kidney mt 3 1:338, 1987 21. LAKE P, MITCHI50N NA: Associative control of the immune response to cell surface antigens. Immunol Comm 5:795—805, 1976 22. GERSHWIN ME, MACKAY IR, STURGESS A, COPPEL RL:
Identification and specificity of a cDNA encoding the 70 kD
References
mitochondrial antigen recognised in primary biliary cirrhosis. J I,nmunol 138:3525—3531, 1987
1. KAPPLER JW, ROEHM N, MARRACK P: T cell tolerance by clonal elimination in the thymus. Cell 49:273—280, 1987 2. KAPPLER JW, STAERZ U, WHITE J, MARRACK PC: Self-tolerance
23. REES AJ, PETERS DK, COMPSTON DAS, BATCHELOR JR: Strong
eliminates T cells specific for Mis-modified products of the major histocompatibility complex. Nature 322:35—40, 1988
24. KLOUDA PT, MANOS J, ACHESON EJ, DYER PA, GOLDBY FS, HARRIS R, LAWLER W, MALLICK NP, WILLIAMS 0: Strong asso-
3. MACDONALD HR, SCHNEIDER R, LEES RK, HOWE RC, ACHAORBA H, FESTENSTEIN H, ZINKERNAGEL RM, HENGARTNER H:
ciation between idiopathic membranous nephropathy and HLADRw3. Lancet 11:770—771, 1979
T-cell receptor Vf3 use predicts reactivity and tolerance to Mlsa encoded antigens. Nature 332:40—45, 1988
25. SINGAL DP, D'SOUZA M, REID B, BENSEN WG, KASSAM YB,
4. GAMMON G, DUNN K, SHASTRI N, OKI A, WILBUR 5, SERCARZ
EE: Neonatai T-cell tolerance to minimal immunogenic peptides is caused by clonal inactivation. Nature 319:4l3—415, 1986 5. GAMMON GM, OKI A, SHASTRI N, SERCARZ EE: Induction of tolerance to one determinant on a synthetic peptide does not affect the response to a second linked determinant. Implications for the mechanism of neonatal tolerance. J Exp Med 164:667—672, 1986 6. GOODNOW CC, CRO5BIE J, ADELSTEIN 5, L.avoi TB, SMITH-GILL SJ, BRINK RA, PRITCHARD-BRISCOE H, WOTHERSPOON JS, Lo-
BLAY RH, RAPHAEL K, TRENT Ri, BASTEN A: Altered immuno-
globulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature 334:676—682, 1988 7. MACKAY IR: Natural autoantibodies to the fore—forbidden clones to the rear? Immunol Today 4:340—342, 1983 8. VAKIL M, KEARNEY JF: Functional characterisation of monoclonal auto-anti-idiotype antibodies isolated from the early B cell reper-
toire of BALB/c mice. Eurflmmunol 16:1151—I 158,
1986
9. Buus 5, SETTE A, COLON SM, MILES C, GREY HM: The relation between major histocompatibility complex (MHC) restriction and the capacity of Ia to bind immunogenic peptides. Science 235:1353—
association between HLA DRw2 and antibody mediated Goodpasture's syndrome. Lancet 1:966—968, 1978
ADACHI JD: HLA-DQ beta-chain polymorphism in HLA-DR4 haplotypes associated with rheumatoid arthritis. Lancet II: 1118— 1120, 1987
26. SACKS SH, BUSt-tELL A, RUST NA, KARAGIANNIS JA, JEWELL DP,
LEIMNGHAM JG, WOOD KJ, MCM!CHAEL AJ: Functional and biochemical subtypes of the haplotype HLA-DR3 in patients with celiac disease or idiopathic membranous nephropathy. Hum Immunol 20:175—187, 1987 27. SAIKI RK, BUGAWAN TL, HORN QT, MuLLI5 KB, ERLICK HA: Analysis of enzymatically amplified /3-globin and HLA-DQ-a DNA
with allele-specific oligonucleotide probes. Nature
324:163—166,
1986
28. SINHA AA, BRAUTBAR C, SZAFER F, FRIEDMANN A, TZFONI B,
TODD JA, STEINMAN L, MCDEVITT HO: A newly characterised HLA DQ-/3 allele associated with pemphigus vulgaris. Science 239: 1026—1029, 1988
29. TODD JA, BELL JI, MCDEVITT HO: HLA-DQ-/3 gene contributes to
susceptibility and resistance to insulin-dependent diabetes mellitus.
Nature 329:599—604, 1987 30. TATTERSALL RB, PYKE DA: Diabetes in identical twins. Lancet
II:
1120—1125, 1972
1358, 1987 10. MARGALIT H, SPOUGE JL, CORNETTE JL, CEASE KB, DELI5I C, BER.ZOFSKY JA: Prediction of immunodominant helper T cell anti-
31. HOHLFELD R, TOYKA KY, HEININGER K, GROSSE-WILUE H, KALIES I: Autoimmune human T lymphocytes specific for acetyl-
genic Sites from the primary sequence. J Immunol 138:2213—2229,
32. LONDEI M, BOTTAZZO GF, FELDMANN M: Human T-cell clones from autoimmune thyroid glands: specific recognition of autologous
1987 11. ROTHBARD J, TOWNSEND A, EDWARDS M, TAYLOR W: Pattern
choline receptor. Nature thyroid cells. Science
310:244—246, 1984
228:85—89, 1985
recognition among T cell epitopes. Hamatol Bluttransfus 31:324—
33. PADULA 5, POLLARD MK, LUIGENHELD EG, CLARK RB: Mainte-
331, 1987 12. SETTE A, BUUS S, COLON S, MILES C, GREY HM: I-A<'-binding
nance of antigen specificity by human interleukin-2-dependent T cell lines. Use of antigen-presenting cells and OKT3 antibody in the absence of antigen. J Clin Invest 74:788—797, 1985
peptides derived from unrelated protein antigens share a common structural motif. J Immunol 141:45-48, 1988
34. GRIFFITHS J, MITCHISON NA, OLIvEIRA DBG, SALGAME P: New
13. MATZINGER P, ZAMOYSKA R, WALDMANN H: Self tolerance is
developments in epitope analysis, in Progress in Vaccinology I,
H-2-restricted. Nature 308:738—741, 1984 14. RAMMENSEE H-G, BEVAN MJ: Evidence from in vitro studies that
tolerance to self antigens is MHC-restricted. Nature
308:741—744,
1984 15. WIESLANDER J, KATAJA M, HUDSON BG: Characterisation of the human Goodpasture antigen. C/in Exp Immunol 69:332—340, 1987 16. PRESSEY A, PUSEY CD, DASH A, PETERS DK, LOCKWOOD CM:
Production of a monoclonal antibody to autoantigenic components of human glomerular basement membrane. C/in Exp Immuno/ 54: 178—184,
1983
17. LOCKWOOD CM, BAKES D, JONES S, WHITAKER KB, Moss DW, SAVAGE COS: Association of alkaline phosphatase with an autoan-
tigen recognised by circulating anti-neutrophil antibodies in systemic vasculitis. Lancet 1:716—719, 1987 18. FALK RJ, JENNETTE JC: Anti-neutrophil cytoplasmic autoantibod-
ies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotising and crescentic glomerulonephritis. NEnglJMed3l8:1651—1657, 1988 19. CLAYMAN MD, MICHAUD L, BRENTJENS J, ANDRES GA, KEFA-
LIDES NA, NEILSON EG: Isolation of the target antigen of human
edited by TALWAR GP, New York, Springer-Verlag, 1988; pp. 359— 367
35. REES AJ, DEMAINE AG, WELSH K!: The association between immunoglobulin allotypes with auto-antibodies to glomerular basement membrane and their titre. Hum Immunol 10:213—221, 1984 36. DEMAINE AG, VAUGHAN RW, TAUBE DH, WELSH KI: Association
of membranous nephropathy with T-cell receptor constant beta chain and immunoglobulin heavy chain switch region polymorphisms. Immunogenetics 27:19-23, 1988 37. HAYGLASS KT, BENACERRAF B, SY M-S: T cell development in B
cell-deficient mice. V. Stopping anti-p treatment results in Ighrestricted expansion of the T suppressor cell repertoire concomitant with the development of normal immunogiobulin levels. J Exp Med 164:36—49, 1986
38. NEILSON EG, MCCAFFERTY E, MANN R, MICHAUD L, CLAYMAN
M: Tubular antigen-derivatized cells induce a disease-protective, antigen-specific, and idiotype-specific suppressor T cell network restricted by I-J and 1gb-V in mice with experimental interstitial nephritis. J Exp Med 162:215—230, 1985 39. PUSEY CD, YENNING MC, PETERS DK: Immunopathology of
928
Oliveira and Peters: Autoimmunity and the kidney
glomerular and interstitial disease, in Diseases of the Kidney (4th ed), edited by SCHRIER RW, GOTTSCHALK CW, Boston/Toronto, Little, Brown and Company, 1988, pp. 1827—1883 40. SULTAN Y, Rossl F, KAZATCHKINE MD: Recovery from anti-VILI:
c (anti-hemophilic factor) autoimmune disease is dependant on generation of anti-idiotypes against anti-V1II:c autoantibodies. Proc Nat! Acad Sci USA 84:828—831, 1987 41. SAVAGE COS: Regulation of autoantibody production in man. The role of idiotype-antiidiotype network interactions in anti-glomerular
basement membrane antibody mediated disease. PhD Thesis, University of London, 1986 42. VINCENT AC: Are spontaneous anti-idiotypic antibodies against
anti-acetylcholine receptor antibodies present in myasthenia gravis? J Autoimmunity 1:131—142, 1988 43. DWYER DS, BRADLEY CK, URQUHART CK, KEARNEY iF: Natu-
rally occurring anti-idiotypic antibodies in myasthenia gravis patients. Nature 301:611—614, 1983 44. LEFVERT AK, PIRSKANEN R, SVANBORG E: Anti-idiotypic antibod-
ies, acetylcholine receptor antibodies and disturbed neuromuscular function in healthy relatives to patients with myasthenia gravis. J Neuroimmunol 9:41—53, 1985
45. MILLER KB, SCHWARTZ RS: Autoimmunity and suppressor T lymphocytes. Adv Intern Med 27:281—313, 1982
46. OKITA N, ROW VW, VOLPE R: Suppressor T lymphocyte defi-
ciency in Graves' disease and Hashimoto's thyroiditis. J C/in Endocrinol Metab 52:528-533, 1981 47. VENTO 5, O'BRIEN CJ, MCFARLANE 10, WILLIAMS R, EDDLESTON
59. TOYKA Ky, DRACHMAN DB, PESTRONK A, KAO I: Myasthenia gravis: Passive transfer from man to mouse. Science 190:397—399, 1975
60. BANNISTER KM, ULICH TR, WILSON CB: Induction, characteriza-
tion, and cell transfer of autoimmune tubulointestinal nephritis. Kidney mt 32:642-651, 1987 61. MIHARA M, OHSUGI Y, Sn'o K, MIYAI T, TOGASH M, ONO S, MURAKAMI S, DOBASH K, HIRAYAMA F, HAMAOKA T: Immuno-
logic abnormality in NZB/NZW F1 mice: Thymus-independent occurrence of B cell abnormality and requirement for T cells in the development of autoimmune disease, as evidenced by an analysis of the athymic nude individuals. J Immunol 141:85—90, 1988 62. NEALE Ti, TIPING P0, CARSON SD, HOLDSWORTH SR: Participa-
tion of cell-mediated immunity in deposition of fibrin in glomerulonephritis. Lancet 11:421—424, 1988 63. ATKINS RC, HOLDSWORTH SR, HANCOCK WH, THOMSON NM, GLASGOW EF: Cellular immune mechanisms in human glomerulo-
nephritis: The role of mononuclear leucocytes. Springer Semin Immunopathol 5:269—296, 1982
64. BRENTJENS JR: Glomerular procoagulant activity and glomerulone-
phritis. Lab Invest 57:107—111, 1987 65. WOFSEY D, SEAMAN WE: Reversal of advanced murine lupus in
NZB/NZW Fl mice by treatment with monoclonal antibody to L3T4. J Immunol 138:3247—3253, 1987 66. MCDEVITT HO, PERRY R, STEINMAN LA: Monoclonal anti-Ia
antibody therapy in animal models of autoimmune disease, in Autoimmunity and Autoimmune Disease, New York, Wiley, Chichester (Ciba Foundation Symposium 129), 1987, pp. 184—193
ALWF: T-cell inducers of suppressor lymphocytes control liver-
67. MOUNTZ JD, SMITH HR, WILDER RL, REEVES JP, STEINBERG AD:
directed autoimmunity. Lancet 1:886—888, 1987 48. OLIVEIRA DBG, MITCHISON NA: Immune suppression genes. Gun Exp Immuno! (in press)
CS-A therapy in MLR-lpr/lpr mice: Amelioration of immunopathology despite autoantibody production. J Immunol 138:157—163, 1987 68. GIMENEZ A, LEYVA-COBIAN F, FIERRO C, RIO M, BRIcIO T, MAMPASO F: Effect of cyclosporin A on autoimmune tubulointerstitial nephritis in the brown Norway rat. C/in Exp Immunol 69:550— 556, 1987 69. MEYRIER A: Treatment with cyclosporin of patients with idiopathic nephrotic syndrome. Springer Semin Immunopathol 9:441—450,
49. UEDA S, WAKASHIN M, WAKASHIN Y, MORI T, YOSHIDA H, MORI
Y, IESATO K, OGAWA M, AZEMOTO R, KATO I, OKUDA K:
Suppressor system in murine intestitial nephritis. Analysis of tubular basement membrane (TBM)-specific suppressor T cells and their soluble factor in C57BL/6 mice using a syngeneic system. C/in Immunol Immunopathol 45:78—91, 1987
1987
70. HINGLAIS N, PELLETIER L, VIAL MC, SAPIN C, BASCOU C, SUN MJ, NEILSON EG: Effector T cell differentiation in experimenBARIETY J, KELLEY V, DRUET P: Effect of prostaglandin El in tal interstitial nephritis. I. The development and modulation of brown Norway rats with mercury-induced autoimmune disease.
50. MANN R, KELLY CJ, HINES WH, CLAYMAN MD, BLANCHARD N,
effector lymphocyte maturation by I-J+ regulatory T cells. J Immunol 138:4200—4208, 1987
51. KELLY CJ, NEILSON EG: Contrasuppression in autoimmunity. Abnormal contrasuppression facilitates expression of nephritogenic effector T cells and interstitial nephritis in kdkd mice. J Exp Med 165:107—123, 1987
52. KELLY CJ, NEILSON EG: Medullary cystic disease. An inherited form of autoimmune interstitial nephritis? Am J Kidney Dis 10:389— 395, 1987 53. Bormzzo GF, PUJOL-BORRELL R, HANAFUSA T, FELDMANN M:
Role of aberrant HLA-DR expression and antigen presentation in
the induction of endocrine autoimmunity. Lancet 11:1115—1119, 1983
54. FELDMANN M: Probing human autoimmune diseases. Immunology (in press) 55. Woot KJ, HOPLEY A, DALLMAN MJ, MORRIS PJ: Lack of corre-
lation between the induction of donor class I and class II major histocompatibility complex antigens and graft rejection. Transplantation 45:759—767, 1988
C/in Immunol Immunopatho/ 40:401—409, 1986 71. KELLY Ci, ZURIER RB, KRAKAUER KA, BLANCHARD N, NEILSON
EG: Prostaglandin El inhibits effector T cell induction and tissue damage in experimental murine interstitial nephritis. J C/in In vest 79:782—779, 1987 72. ALEXANDER NJ, SMYTHE NL, JOKINEN MP: The type of dietary fat
affects the severity of autoimmune disease in NZB/NZW mice. Am JPatho! 127:106—121, 1987
73. CLAYMAN MD, SUN MJ, MICHAUD L, BRILL-DASHOFF J, RIBLET
R, NEILSON EG: Clonotypic heterogeneity in experimental interstitial nephritis, Restricted specificity of the anti-tubular basement membrane B cell repertoire is associated with a disease-modifying crossreactive idiotype. J Exp Med 167:1296—1312, 1988 74. IMBACH P, BARANDUN S, D'APUZO V, BAUMGARTNER C, HIRT A, MORELL A, ROSSI E, SCHONI M, VEST M, WAGNER HP: High-dose
intravenous gammaglobulin for idiopathic thrombocytopenic purpura in childhood. Lancet 1:1228—1231, 1981 75. SULTAN Y, KAZATCHKINE MD, MAISONNEUVE P. NYDEGGER UE:
Anti-idiotypic suppression of autoantibodies to factor VIII (anti-
56. LACHMANN PJ: Complement—friend or foe? Br J Rheumato! 26:
haemophilic factor) by high-dose intravenous gammaglobulin. Lan-
409—415, 1987 57. SCHENA PF, PASTORE A, SINICO RA, MONTINARO V, FORNASIERI
76. SUN D, QIN Y, CHLUBA J, EPPLEN JT, WEKERLE H: Suppression
A: Polymeric IgA and IgA rheumatoid factor decrease the capacity of serum to solubilize circulating immune complexes in patients with primary IgA nephropathy. J Immunol 141:125—130, 1988 58. LERNER RA, GLASSOCK RJ, DIXON FJ: The role of anti-glomerular
basement membrane antibody in the pathogenesis of human gbmerulonephritis. J Exp Med 126:989—1004, 1967
cet
11:765—768, 1984
of experimentally induced autoimmune encephalomyelitis by cytolytic T-T cell interactions. Nature 332:843—845, 1988
77. MANN R, NEILSON EG: Murine interstitial nephritis. V. The auto-induction of antigen-specific Lyt-2 suppressor T cells diminishes the expression of interstitial nephritis in mice with antitubular basement membrane disease. J Immuno! 136:908—912, 1986