Genetic Dissection of Lupus Nephritis in Murine ... - Springer Link

Journal of Clinical Immunology, Vol. 17, No. 4, 1997

Special Article

Genetic Dissection of Lupus Nephritis in Murine Models of SLE EDWARD K. WAKELAND,1,2 LAURENCE MOREL,1,2 CHANDRA MOHAN,1 and MARY YUI1

Accepted: April 16, 1997

hypothesis that SLE susceptibility is inherited as a multifactorial genetic disease (1). However, the complex and non-organ-specific nature of SLE has made it difficult for researchers to unravel the pathogenic mechanisms underlying this disease. Although the effector mechanisms responsible for the clinical features of SLE have been studied extensively, the immunogenetic mechanisms responsible for the initiation of disease pathogenesis remain poorly understood. Over the past 30 years, several mouse models of lupus bearing differing sets of phenotypes and genotypes have collectively contributed a great deal toward our understanding of the disease (reviewed in Refs. 2 and 3). In addition, recent advances in molecular genetics have provided new technologies that allow a detailed analysis of the inheritance of multifactorial traits such as autoimmune disease (4). We and others have utilized these methodologies to identify the locations of genes mediating disease susceptibility in several murine SLE models. These studies have provided some important new insights into the genetic basis of susceptibility to autoimmunity and are beginning to reveal the functional properties of individual susceptibility loci. Here we briefly summarize our current understanding of the genetics of lupus susceptibility in murine models of SLE and characterize the properties of genes that potentiate autoimmunity.

KEY WORDS: Lupus nephritis; systemic lupus erythematosus; murine models; genetics.

INTRODUCTION

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with extremely heterogeneous clinical features, varying in pathogenicity from mild forms of the disease to those that advance relentlessly leading to progressive end organ damage. The underlying autoimmune disorder can express as a variety of immunologic abnormalities but generally results in the production of autoantibodies primarily directed against a spectrum of nuclear antigens. Much of the pathogenicity of SLE results from inflammatory processes initiated as a consequence of either the deposition of immune complexes or the targeting of autoantibodies to various anatomic sites. The most serious clinical consequences of SLE result from immune complex deposition in the kidney, resulting in lupus nephritis and culminating in kidney failure. Although the factors responsible for the development of SLE are poorly understood, a wealth of data clearly indicates that genetic predisposition is a key element in susceptibility. The high concordance rates of SLE among monozygotic twins (57%) and relatively low concordance rates in dizygotic twins (5%) and first-degree relatives (5-12%) of SLE probands strongly support the

SUSCEPTIBILITY GENES FOR HUMORAL AUTOIMMUNITY AND LUPUS NEPHRITIS

Center for Mammalian Genetics, College of Medicine, University of Florida, Gainesville, Florida 32610. 2 Department of Immunology, Pathology, and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, Florida 32610-0275.

Technological advances in molecular genetics, driven in part by the human genome project, have led to the 272 0271-9142/97/0700-0272$ 12.50/0 ® 1997 Plenum Publishing Corporation

273

GENETICS OF SUSCEPTIBILITY TO LUPUS

description of over 7000 polymorphic microsatellite marker loci for linkage analysis in the mouse (5, 6) and the development of sophisticated computer programs for the generation of linkage maps and the analysis of polygenic inheritance (4, 7, 8). These technologies have provided the means to determine the number and locations of genes mediating susceptibility to autoimmunity via the analysis of test crosses between lupus-prone and -resistant mouse strains. Several groups have utilized these tools to assess the inheritance of susceptibility to lupus nephritis and a variety of lupus-associated autoimmune phenotypes such as humoral autoimmunity to nuclear antigens and splenomegaly. Thus far, linkage analysis has been reported for eight separate crosses with various mouse models of lupus, including the Fasdefective MRL/Lpr mouse crossed with CAST/Ei (9); F2 and BC1 crosses between NZB and NZW (10-13), the NZB/W-related NZM2410 lupus-prone strain crossed with C57BL/6 (14); NZB crossed with a collection of RI strains (15); and NZW crossed with BALB/c (16). These studies have identified a total of 22 genomic intervals containing genes that are associated with susceptibility to lupus nephritis or a lupus-related autoimmune phenotype. Figure 1 presents the positions of these susceptibility genes in the mouse genome. For completeness, we have included all intervals reported with statistical levels of P < 0.01, although recent statistical studies of linkage analysis with total genome scans have suggested more stringent levels (17). In this regard, of the 22 intervals listed, 8 do not achieve the recommended level for "suggestive" linkage suggested by Lander and Kruglyak, and await further confirmation of their linkage to lupus susceptibility. Several interesting insights into the genetic basis for susceptibility to humoral autoimmunity and lupus nephritis are revealed by this compilation. First, the number of genes detected illustrates the complexity of the genetics of this disease. In the strain combinations analyzed thus far, from three to eight susceptibility genes were found to segregate in association with lupus or lupusassociated autoimmune traits. Each strain combination tested identified one or more unique genetic intervals associated with autoimmunity, indicating that susceptibility to lupus nephritis is mediated by distinct genetic pathways in the separate mouse models tested. However, some intervals, most notably, telomeric chromosome 1, middle chromosome 4, and centromeric chromosome 7, were associated with susceptibility and/or humoral autoimmunity in multiple crosses, suggesting that these susceptibility genes may be common to the autoimmune pathogenesis in multiple lupus-prone strains.


Within each cross, some genomic segments were associated with a single phenotype, consistent with the idea that individual genes contribute specific component phenotypes to disease pathogenesis. For example, intervals on chromosomes 3 and 19 were associated only with the production of specific autoantibody (12; Morel et al., submitted), while chromosomes 4 (middle) and 10 (11, 18; Morel et al., submitted) were associated only with susceptibility to glomerulonephritis. However, stringent comparisons between phenotypes associated with most intervals cannot be made since individual research groups scored slightly different lupus-associated traits, with proteinuria or glomerulonephritis being the most common. In addition, we have found (19) that the phenotype associated with a single susceptibility locus may be different from that detected during linkage analysis in a test cross where multiple loci are segregating. Finally, regions such as telomeric chromosome 1, centromeric chromosome 7, and heterozygosity for MHC genes on chromosome 17, which were associated with disease in several different crosses, strongly affected both humoral autoimmunity and lupus nephritis. Figure 1 also tabulates the positions of genes controlling quantitative variations in humoral immune responses elicited by immunization with exogenous protein antigens. These intervals were detected by QTL analyses of variations in antibody liters elicited by immunizations with exogenous antigen in test crosses of Biozzi's high and low responder strains (20) and an F2 intercross of A.SW and SJL (21). As shown in Fig. 1, 6 of the 10 genomic intervals containing these immune response genes affecting antibody titers colocalize with intervals containing lupus susceptibility genes. This result is consistent with the notion that some of the genes associated with humoral responsiveness to exogenous antigens may also influence antibody-mediated autoimmune diseases. This possibility is not unreasonable, since factors enhancing immune responsiveness would be predicted to aggravate the severity of autoimmunity. MULTIFACTORIAL INHERITANCE OF LUPUS NEPHRITIS Lupus nephritis is a complex trait that is inherited in a multifactorial fashion in which several susceptibility genes potentiate expression, but expression is dependent upon a triggering event. Given the multitude of genes contributing to lupus susceptibility in mouse models, this complexity is not surprising. A key concept in complex trait expression is the hypolhesized interactions of genetic effects and triggering events. Essentially, the notion is that genes associated with complex traits predispose

274

WAKELAND, MOREL, MOHAN, AND YUI

Fig. 1. Locations of susceptibility intervals associated with autoimmune diseases by linkage analyses on the 19 murine autosomes. None of these loci has been found yet to be X-linked. Each susceptibility locus is indicated by a vertical bar centered on the SSR marker, indicated on the left, that showed maximum linkage. The name of the susceptibility locus, it any, is indicated to the right. The length of each interval has been arbitrarily set at 20 cM, which typically corresponds to the support interval obtained for each



individuals to the development of a phenotype, rather than directly cause it. To illustrate the difference, consider the inheritance of a simple mutation (relatively speaking) that is directly causal for the albino phenotype in mice. Mice with an albino phenotype (white coat color, for example, BALB/c or NZW), lack a functional tyrosinase gene and, as a result, are unable to synthesize the pigment melanin. Mice carrying an albino mutation, which is encoded by the tyr locus (formerly c) on chromosome 7, will always fail to synthesize melanin and consequently express the albino phenotype. As a result, the phenotype (albino) is directly and invariably linked with a specific genotype (nonfunctional tyr gene) in a given test cross. This direct relationship between phenotype and genotype does not hold for genes associated with complex traits such as autoimmune disease. To illustrate this, consider the expression of lupus nephritis in the lupusprone NZM2410 strain. Although the NZM2410 strain is fully inbred, and consequently all the mice express an identical array of susceptibility genes, the proportion of animals (penetrance) developing lupus nephritis increases as a function of age, initiating with a penetrance of about 25% at 5 months and increasing to about 85% at 12 months of age. Thus, 15% of mice with a lupussusceptible genotype fail to develop lupus at any time in their lives and the penetrance of lupus increases as a function of age. This variation in penetrance is a common feature of autoimmune phenotypes in inbred mouse models and is a standard characteristic of many complex traits. The age-related variation in expression is thought to indicate that the expression of complex traits requires a triggering event (or events) and that the probability of this event occurring increases as a function of time. In some instances, the phenotype may never be triggered, thus resulting in a proportion of animals that fail to develop lupus despite carrying all the requisite genes. As a result, the association of genotype with phenotype is incomplete, even when the entire repertoire of susceptibility genes is present. The "triggering event" for lupus would represent the immunologic interactions which lead the immune system to initiate a pathogenic autoimmune response. The nature of this event or, more likely, series of events is currently unknown. Theoretically, the initiation of autoimmunity could be triggered by an event external to the animal

275

(i.e., infection or environmental stimulus) or could simply result from a specific stochastic event occurring during T or B cell ontogeny or the normal internal functioning of the immune system. The important concept is that lupus susceptibility genes act to potentiate this triggering event or accelerate and exacerbate the consequences, rather than directly to initiate autoimmune disease. As a result, an individual's genotype is uncoupled from a causal relationship with autoimmunity and is only indirectly related to the development of disease. The second important feature of multifactorial inheritance for lupus nephritis is that multiple genes are required for the development of the disease, but no gene is individually necessary or sufficient for the development of autoimmunity (14). Thus, lupus nephritis is a polygenic disease in which several susceptibility genes contribute to susceptibility via the expression of a component or intermediate phenotype. The expression of each of these component phenotypes will then increase the probability that the triggering event will initiate the development of lupus nephritis. This mode of inheritance for multifactorial diseases, which is a common feature of many human diseases and developmental malformations, is commonly described as a threshold of genetic liability (22). The threshold liability model, which was originally developed by Sewell Wright to explain the inheritance of polydactyl hind feet in guinea pigs (23, 24), postulates that when several genes contribute incrementally to the overall susceptibility of individuals to a specific disease or malformation, the penetrance of the disease will increase as a function of the susceptibility gene content of individuals. We first demonstrated that this relationship held for the inheritance of susceptibility to lupus nephritis in our analysis of (NZM2410 X C57BL/6)F, X NZM2410 BC, progeny (14). Subsequent studies by other investigators using other models of lupus (15), diabetes (25), and experimental allergic encephalomyelitis (26, 27) have reached similar conclusions, indicating that this mode of inheritance is a common feature in autoimmune disease. Figure 2 provides an illustration of the threshold liability model as it applies to the inheritance of susceptibility to lupus nephritis in a test cross of NZM2410 and C57BL/6. In this illustration, which was adapted from Sewall Wright's original publication (23), the X axis

locus in the analysis of experimental crosses presented here (100 to 150 progeny typed with about 100 polymorphic SSR markers). The component phenotypes associated with each locus (GN, autoantibodies, or splenomegaly) are indicated by different patterns. Side-by-side bars are shown when several phenotypes are associated with the same locus either in the same cross or in different studies. Black bars indicate intervals associated with humoral immunity to exogenous antigens (IR genes).


276

Fig. 2. Threshold liability model applied to the inheritance of lupus nephritis in a test cross between NZM2410, a lupus-prone mouse strain, and C57BL/6, a resistant strain. The X axis represents increasing disease liability, and the Y axis the proportion of animals involved. The position of the disease threshold is indicated by the heavy dashed lines. In the upper panel, bell-shape curves represent the distribution of the two strains and their F, hybrid relative to the disease liability. The lower panel illustrates the distribution of disease liability in an (NZM2410 X C57BL/6)F, X NZM2410 backcross progeny based on the number of susceptibility genes (one to four) segregating. [Adapted from Ref. 23 according to the data obtained by Morel el al. (14).J

represents increasing liability to the development of disease, with disease being expressed in all individuals beyond the "threshold" level, marked by the heavy vertical dashed line. The upper panel models the hypothetical relationships of the parental strains (C57BL/6, NZM2410, and their F1 hybrid) to the liability threshold for lupus nephritis. For each of these strains, their content of susceptibility genes for lupus places them at a specific position on the liability axis in relation to the threshold. Their genotype would define an intrinsic level of susceptibility to lupus, with the probability of the triggering event leading to a normal distribution in the liability of individuals with this genotype around this specific point. Thus, the lupus-prone strain NZM2410, which has a mean penetrance of 85% for lupus nephritis, is positioned such that 85% of all NZM2410 mice are beyond the liability threshold. C57BL/6 and their F, hybrid, which do not have detectable lupus nephritis, are all below the threshold. The lower panel in Fig. 2 models the inheritance of susceptibility in a population of (NZM2410 X C57BL/6)F, X NZM2410 backcross progeny in which four susceptibility alleles are segregat-


ing. In this test cross, the penetrance of lupus nephritis increased as a function of the number of homozygous susceptibility genes present in individual progeny. The inheritance of lupus as a threshold liability in this cross is depicted by the movement of populations of these individuals down the liability axis on the basis of their genotypes for susceptibility alleles. Thus, the mean penetrance of lupus among individuals homozygous was 7% for one susceptibility allele, 30% for two, 80% for three, and 90% for four (14). In the threshold liability model, this variation in penetrance results from a combination of their intrinsic genetic liability and the probability of the triggering event causing the initiation of disease. Disease susceptibilities that are inherited as a multifactorial threshold liability would be postulated to have several interesting genetic features. Alleles from many genes could contribute with varying degrees of impact to the susceptibility of an individual via a multitude of pathogenic pathways. As a result, in outbred human populations, different families with high frequencies of the disease may segregate different subsets of these genes and consequently appear to exhibit different modes of inheritance and penetrance. Furthermore, detectable linkage relationships will differ depending upon which genes are segregating within specific families. Another complication will be that several different combinations of susceptibility alleles may lead to a highly susceptible phenotype, and as a consequence, not all affected individuals will necessarily share the same susceptibility alleles. Given this degree of complexity, it is not surprising that linkage analysis of multifactorial traits and diseases in human populations requires extremely large sample populations to obtain significant results. GENETIC DISSECTION OF LUPUS PATHOGENESIS INTO COMPONENT PHENOTYPES The complex, polygenic inheritance of susceptibility to lupus has complicated the process of identifying and functionally characterizing individual susceptibility genes. The major problems encountered in this complex system are (a) determining the precise locations of individual susceptibility genes in the genome and (b) determining the component phenotype contributed by each susceptibility gene. The data in Fig. 1 clearly demonstrate that the positions of lupus susceptibility genes are generally localized only to intervals of about 20 cM, even when hundreds of test-cross animals are analyzed. Since 20 cM represents a little over 1% of the mouse genome, each of these intervals would be predicted to contain more than 500 genes! The statistical


277


Fig. 3. Strategy used to analyze the genetics of a complex trait such as lupus nephritis.

problems associated with determining precisely the positions of individual genes contributing to complex traits such as lupus have been recently discussed in detail (17). The major difficulties are that no single gene is required for the expression of lupus and, as discussed above, the expression of lupus is uncoupled from the genotype by an undefined triggering event that is required for the initiation of disease. Similarly, it is difficult to associate specific phenotypes with individual genes when several genes are segregating simultaneously in a test cross. Thus, although these genetic studies have provided important insights into the number and locations of genes responsible for autoimmunity in lupus-prone strains, they have provided little information concerning the nature of the component phenotype that each gene contributes in lupus susceptibility. How, then, can we progress from knowing the approximate locations of lupus susceptibility genes to identifying these genes and understanding their functional properties? The approach that our laboratory has followed to determine the genetic basis for lupus susceptibility in the lupus-prone NZM2410 strain is outlined in Fig. 3. In the first stage of this strategy, we identified the positions of genomic intervals containing lupus susceptibility genes in NZM2410 via the analysis of a test cross between NZM2410 and C57BL/6 (14). This study identified four potent loci effecting susceptibility to lupus nephritis on chromosomes 1, 4, 7, and 17 (H2) in NZM2410. In the second stage, we dissected the genetic complexity of this model system into its component elements via the production of a collection of interval-specific congenic strains on the C57BL/6 genome. Thus far, we have produced four B6-congenic strains, each carrying one of the NZM2410-derived genomic intervals associated with


lupus susceptibility (28). In the third stage, the component phenotypes contributed by each susceptibility gene have been determined via detailed analysis of the immunologic properties of each congenic strain in comparison to standard C57BL/6 mice (19, 29). These analyses, which are still ongoing, have already provided important new insights into the phenotypes contributed by each of these genes to lupus pathogenesis. The primary goal of this strategy is to dissect the complex pathogenesis exhibited by lupus-prone strains such as NZM2410 into its component elements via the production of congenic strains. Once these strains are produced and functionally characterized, then a variety of experimental approaches can be utilized to determine the genetic mechanisms responsible for lupus. As illustrated in Fig. 3, two separate research thrusts can be initiated. In one strategy, the exact position of each susceptibility gene can be determined via the production and analysis of congenic recombinants. Essentially, the component phenotype expressed by each susceptibility gene can be used to mark the position of this gene on recombinant chromosomes produced by crosses with C57BL/6 mice, thus allowing the fine mapping and ultimately positional cloning of each gene. In the second strategy, a detailed functional analysis of the component phenotype of each susceptibility gene can be pursued via comparison with the parental NZM2410 strain and the C57BL/6 congenic partner. As described briefly below, this approach has provided some interesting data concerning the component phenotypes contributed by Slel and Sle2. Finally, interactions between the individual susceptibility genes can be assessed by intercrossing the individual congenic strains to produce bi- and polycongenic strains. The goal of these studies is to progressively reconstruct the lupus phenotype by combining the susceptibility genes in various combinations. In summary, this strategy, which owes its origins to the Nobel Prizewinning studies of histocompatibility by George Snell in the 1940s, is designed to dissect a complex trait into its component genetic elements, identify and characterize each gene, and then progressively reconstruct the complex pathogenesis exhibited in the original lupus-prone strain. COMPONENT PHENOTYPES OF LUPUS NEPHRITIS

Based on our current understanding of the genetics of this disease, it is clear that individual susceptibility genes for lupus express component phenotypes that potentiate the initiation of pathogenesis. By understanding precisely what these component phenotypes are, it should be possible to understand the event or events that "trigger"

278

the initiation of disease pathogenesis. We have recently completed the first stage of our analysis of the component phenotypes associated with three of the genes that are responsible for lupus susceptibility in the NZM2410 strain. These studies have involved an analysis of the immunologic phenotypes expressed in our interval specific congenic strains carrying individual lupus susceptibility genes. The results of these studies provide some interesting insights into the function of these susceptibility genes in disease pathogenesis. Slel: Targeting Nuclear Autoantigens for Immune Recognition. One of the hallmarks of lupus is the loss of tolerance to nuclear autoantigens. We have recently demonstrated that Slel, a potent lupus susceptibility gene located in the telomeric region of chromosome 1, causes the loss of tolerance to chromatin autoantigens when expressed on the B6 background in our interval-specific congenic strain, B6.NZMcl (19). Slel potentiates a strong, spontaneous humoral response reacting primarily with H2A/H2B/DNA subnucleosomes (Mohan et al., submitted). This targeted immune response peaks at 7 to 9 months of age, affects both sexes with equally high penetrance (>75%), and interestingly, does not "spread" to other subnucleosomal components on chromatin. In addition, despite the presence of high titers of autoantibody, these animals experience little or no kidney pathology. Slel leads to an expanded pool of histonereactive T cells, which may play a role in driving the anti-H2A/H2B/DNA B cells. However, these mice do not exhibit generalized immunological defects or quantitative aberrations in lymphocyte apoptosis. We hypothesize that Slel mediates the presentation of chromatin in an immunogenic fashion or directly impacts the tolerance of the chromatin-specific B cells. Sle2: B-Cell Hyperactivity. Numerous investigations have demonstrated polyclonal seroreactivity, intrinsic B-cell hyperresponsiveness to LPS and conventional antigens, and heightened, spontaneous B-cell proliferation and IgM secretion in vitro as consistent features in both murine and human lupus (30-46). We have recently demonstrated that interval specific congenic mice carrying Sle2 on chromosome 4 (B6.NZMc4) exhibit a variety of immunophenotypes affecting their B cells (19, 29). They have an early, but transient, expansion of splenic CD2310 B-cells. Thereafter, their B cells show signs of spontaneous activation by surface phenotype and functional criteria, paralleled by elevated serum levels of polyreactive/polyclonal IgM. Importantly, Sle2 leads to a heightened B-cell responsiveness to in vitro stimuli and to in vivo antigenic challenge. Finally, they exhibit increased levels of peritoneal and splenic Bl cells. These results indicate that Sle2 harbors a gene that


leads to spontaneous B-cell hyperactivity and elevated Bl-cell formation. However, Sle2 by itself on the normal B6 background is insufficient to generate IgG autoantibody or nephritis. These results suggest that Sle2 may amplify ongoing autoimmune responses by reducing the B-cell signaling threshold and consequently heightening autoimmune responsiveness. Sle3: Amplification of Helper T Cells for Humoral Autoimmunity. The oligospecificity and structural features of pathogenic autoantibodies provide strong evidence that some additional factors, extrinsic to polyclonal B-cell activation, are crucial in driving the expansion of autoimmune B cells. The concomitant presence of antihistone and antihistone/DNA antibodies in patients with anti-DNA antibodies, specific for conformational DNA-protein epitopes dependent on the quaternary structure of chromatin, suggests that chromatin may be the culprit immunogen in this disease (47). Antichromatin autoantibodies in human and murine lupus evolve in a stepwise manner, being directed initially against conformational epitopes on native chromatin and the (H2A-H2B)2-DNA subnucleosome, with the subsequent appearance of antihistone and anti-dsDNA specificities (47-50). The kinetics of appearance, the IgG2a/ IgG2b subclass preference, and the increased expression of arginine replacement mutations in the antibody VH and VL CDR regions of these ANAs (reviewed in Ref. 51) are all consistent with being the outcome of Tdependent immunization with chromatin. We have recently demonstrated that interval-specific congenic mice carrying Sle3 on chromosome 7 (B6.NZMc7) exhibit a variety of immune phenotypes indicative of a dysregulation in the T-cell compartment. These mice spontaneously produce relatively low titers of autoantibodies against a variety of nuclear autoantigens and, although kidney function is not severely impaired, develop significant levels of glomerulonephritis by 12 months of age (19). In addition, we have recently demonstrated a spontaneous expansion of CD4 T cells in the periphery of these mice, as well as indications that B6.NZMc7 mice respond significantly better than B6 mice when challenged with exogenous antigen (Yu, Mohan, Morel, and Wakeland, unpublished observations). These results indicate that Sle3 potentiates T-cell responsiveness in general and can initiate spontaneous autoimmunity to a broad spectrum of nuclear autoantigens. We hypothesize that Sle3 plays an important role in potentiating the antigen-dependent expansion of helper T cells that can drive lupus B cells to make autoantibodies.



Fig. 4. Model of SLE pathogenesis as a progression of three successive stages, and the position of the Sle loci in this progression. Sle loci can confer SLE susceptibility through a variety of pathogenic mechanisms and may act at a precise stage of the disease. Sle2 and Sle3 both contribute to a generalized immune hyperresponsiveness, while Slel results in a specific response to nuclear antigens. Finally, Sle3 may predispose to end organ damage, either by modifying the end organ itself, or by increasing the pathogenicity of the immune processes developed in the preceding stages. (Adapted from Ref. 19.)

A MODEL OF GENETIC INTERACTIONS IN THE PATHOGENESIS OF LUPUS NEPHRITIS Our current understanding of the component phenotypes associated with individual lupus susceptibility genes has led to some insights into the genetic mechanisms responsible for lupus susceptibility. As illustrated in Fig. 4, three separate, functionally distinct types of genes appear to impact lupus susceptibility. First, genes that mediate generalized immune hyperresponsiveness, such as Sle2 and Sle3, potentiate autoimmunity by lowering the activation threshold of immune reactive cells and increasing the levels of effector functions (19, 29). Genes in this class can affect either T (Sle3) or B (Sle2) cells and, based on our analysis of their component phenotypes when bred onto a lupus-resistant genome, typically will enhance immune responsiveness to exogenous antigens. Interestingly, 6 of 10 genes influencing immune responsiveness to exogenous antigens mapped into the same genomic regions as lupus susceptibility genes (see Fig. 1), raising the possibility that several other lupus susceptibility genes may also be of this type. Taken together, these findings suggest that genetic polymorphisms that potentiate high immune responsiveness to exogenous antigens may, when expressed in combination with other disease-specific genes, become deleterious via potentiating the expansion of autoimmune B cell clones. This type of polymorphic immune response gene probably persists in natural populations as a. consequence of two antagonistic selective pressures that operate on genes whose functions impact immune responsiveness.


279

On the one hand, the extensive antigenic heterogeneity and rapid evolution of pathogenic microorganisms will favor individuals with genomes that potentiate high immune responsiveness. At the same time, the deleterious effects of autoimmunity and autoaggressive reactions will favor individuals with genomes that potentiate lower immune responsiveness. Given the complexity of hostpathogen interactions and the antagonistic nature of these selective pressures, no single genetic solution will be favored in all situations. As a result, allelic polymorphisms in genes that functionally diversify immune responsiveness will be favored in natural populations. It is tempting to speculate that the prevalence of autoimmune diseases may be due in part to the deleterious impact of multiple alleles mediating hyper-immune responsiveness in a single genome. The second class of lupus susceptibility genes, represented by Sle1 on chromosome 1, is genes that mediate targeting of nuclear autoantigens. The precise mechanism by which Sle1 causes a loss of tolerance to nuclear antigens remains a mystery at present, however, this strain does not appear to have a generalized dysregulation of the immune system. Instead, Sle1 mediates the initiation of a humoral autoimmune response predominantly against the H2A/H2B/DNA subnucleosome. Identifying precisely the component phenotype of Sle1 should provide important insights into the immunologic mechanisms responsible for the exposure of nuclear antigens to the immune system during the initiation of humoral autoimmunity. Interestingly, Tsao and coworkers have recently reported strong linkage between SLE susceptibility and an interval of the human genome that is syntenic with the telomeric region of chromosome 1 in the mouse (52). Thus, the human homologue of Sle1 may play an important role in lupus susceptibility in humans, although further work will be needed to establish firmly that the same gene is involved. The final class of lupus susceptibility genes, represented by Sle3, is genes that potentiate end organ damage. We propose that this class of gene must exist on the basis of our findings with B6.NZMcl mice, which express high titers of antinuclear autoantibodies but fail to develop significant levels of glomerulonephritis (19). Theoretically, end organ damage could be mediated via genes in several distinct pathways. For example, genes that enhance effector functions of humoral immune responses could potentiate the severity of end organ damage. Deficiencies in complement factor H, a soluble protein involved in the C3b regulation pathway, have been shown to result in a lupus-like glomerulonephritis in humans (53). Alternatively, genes that facilitate the development of pathogenic autoantibodies by increasing

280

the antigen-binding site diversification of B-cell clones during an autoimmune response could also facilitate end organ damage by driving the development of higher affinity, pathogenic autoantibodies. Finally, genes that modify the end organ itself could potentiate the severity of humoral autoimmunity by increasing the sensitivity of the organ to damage by antibody deposition. The resolution of which pathway is responsible for the enhanced end organ damage mediated by Sle3 will await future analyses. In conclusion, several research groups are actively pursuing the genetic basis for lupus susceptibility in several mouse models of SLE. These studies have made significant progress and provided important new insights into the inheritance of susceptibility. However, the precise immunologic events responsible for the initiation of humoral autoimmunity are still unknown, although it is now clear that several distinct types of genes can potentiate lupus susceptibility. A thorough understanding of the component phenotypes contributed by individual lupus susceptibility genes may provide important clues as to the nature of the triggering event that initiates autoimmunity.

REFERENCES 1. Winchester R: Genetic susceptibility to systemic lupus erythematosus. In Systemic Lupus Erythematosus, RG Lahita (ed). New York, Churchill Livingstone, 1992, pp 65-85 2. Theofilopoulos AN: The basis of autoimmunity. I. Mechanisms of aberrant self-recognition. Immunol Today 16:90-98, 1995 3. Mohan C, Datta SK: Lupus: Key pathogenic mechanisms and contributing factors. Clin Immunol Immunopathol 77:209-220, 1995 4. Lander ES, Schork NJ: Genetic dissection of complex traits. Science 265:2037-2048, 1995 5. Dietrich W, Katz H, Lincoln SE, Shin H-S, Friedman J, Dracopoli N, Lander ES: A genetic map of the mouse suitable for typing intraspecific crosses. Genetics 131:423-447, 1992 6. Dietrich WE, Miller JC, Steen RG, Merchant M, Damron D, Nahf R, Gross A, Joyce DC, Wessel M, Dredge RD, Marquis A, Stein LD, Goodman N, Page DC, Lander ES: A genetic map of the mouse with 4,006 simple sequence length polymorphisms. Nature Genet 7:220-245, 1994 7. Lander ES, Green P, Abrahamson J, Barlow A, Daly MJ, Lincoln SE, Newburg L: MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174-181, 1987 8. Lander ES, Botstein D: Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121:185199, 1989 9. Watson ML, Rao JK, Gilkeson GS, Ruiz P, Eicher EM, Pisetsky DS, Matsuzawa A, Rochelle JM, Seldin MF: Genetic analysis of MRL-lpr mice: Relationship of the fas apoptosis gene to disease manifestations and renal disease-modifying loci. J Exp Med 176:1645-1656, 1992


10. Kono DH, Burlingame RW, Owens DC, Kuramochi A, Balderas RS, Balomenos D, Theofilopoulos AN: Lupus susceptibility loci in New Zealand mice. Proc Natl Acad Sci USA 91:10168-10172, 1994 11. Drake CG, Babcok SK, Palmer E, Kotzin BK: Genetic analysis of the NZB contribution to lupus-like autoimmune disease. Proc Natl Acad Sci USA 91:4062-4065, 1994 12. Mohan C, Adams S, Stannik V, Datta SK: A major immunogen for pathogenic autoantibody-inducing T cells of lupus. J Exp Med 177:1367-1381, 1993 13. Hirose S, Tsurui H, Nishimura H, Jiang Y, Shirai T: Mapping of a gene for hypergammaglobulinemia to the distal region on chromosome 4 in NZB mice and its contribution to systemic lupus erythematosus in (NZB X NZW)FI mice. Int Immunol 6:18571864, 1994 14. Morel L, Rudofsky UH, Longmate JA, Schilfenbauer J, Wakeland EK: Polygenelic control of susceptibility to murine systemic lupus erythematosus. Immunology l(3):219-229, 1994 15. Drake CG, Roz/o SJ, Hirschfeld HE, Smarnworawong NP. Palmer E, Kotzin BK: Analysis of the NZB contribution to lupus-like renal disease: Multiple genes that operate in a threshold manner. J Immunol 154:2441-2447, 1995 16. Vyse TJ, Morel L, Tanner FJ, Wakeland EK, Kotzin BL: Backcross analysis of genes linked to autoantibody production in New Zealand white mice. J Immunol 157:2719-2727, 1996 17. Lander ES, Kruglyak L: Genetic dissection of complex traits: Guidelines for interpreting and reporting linkage results. Nature Genet 11:241-247, 1995 18. Klinman DM: Polyclonal B cell activation in lupus-prone mice precedes and predicts the development of autoimmune disease. J Clin Invest 86:1249, 1990 19. Morel L, Mohan C, Croker BP, Tian X-H, Wakeland EK: Functional dissection of systemic lupus erythematosus using congenie mouse strains. J Immunol 1997 (in press) 20. Puel A, Groot PC, Lathrop MG, Demant P, Mouton D: Mapping of genes controlling qualitative antibody production in Biozzi's mice. J Immunol 154:5799-5805, 1995 21. Wu JM, Longmate JA, Adamus G, Hargrave PA, Wakeland EK: Interval mapping of quantitative trait loci controlling humoral immunity to exogenous antigens: Evidence that non-MHC immune response genes may also influence susceptibility to autoimmunity. J Immunol 157:2498-2505, 1996 22. Gelehrter TD, Collins FS: Principles of Medical Genetics. Baltimore, MD: Williams and Wilkins, 1990 23. Wright S: An analysis of variability in number of digits in an inbred strain of guinea pigs. Genetics 19:506-536, 1933 24. Wright S: The results of crosses between inbred strains of guinea pigs differing in number of digits. Genetics 19:537-551 25. McAleer MA, Reifsnyder P, Palmer SM, Prochazka M, Love JM, Copeman JB, Powell EE, Rodrigues NR, Prins J-P, Serreze DV, DeLarato NH, Wicker LS, Peterson LB, Schork NJ, Todd JA, Leiter EH: Crosses of NOD mice with the related NON strain: A polygenic model for type I diabetes. Diabetes 44:1186-1195, 1999 26. Sundvall M, Jirholt J, Yang HT, Jansson L, Engstrom A, Petterson U, Holmdahl R: Identification of murine loci associated with susceptibility to chronic autoimmune encephalomyelitis. Nature Genet 10:313-317. 1995 27. Encinas JA, Lees MB, Sobel RA, Symonowycz C, Greer JM, Shovlin CL, Weiner HL, Seidman CE, Seidman JG, Kuchroo VJ: Genetic analysis of susceptibility to experimental autoimmune



28.

29.

30.

31. 32. 33.

34.

35.

36.

37.

38.

39.

40.

41.

encephalomyelitis in a cross between SJL/J and B10.S mice. J I m u n o l 157:2186-2192, 1996 Morel L, Yu Y, Blenman KR, Caldwell RA, Wakeland EK: Production of congenic mouse strains carrying SLE-susceptibility genes derived from the SLE-prone NZM/Aeg2410 strain. Mammal Genome 7:335-339, 1996 Mohan C, Morel L, Yang P, Wakeland EK: Genetic dissection of SLE pathogenesis: Sle2 on murine chromosome 4 leads to B-cell hyperactivity. J Immunol 1997 (in press) Klinman DM, Steinberg AD: Systemic autoimmune disease arises from polyclonal B cell activation. J Exp Med 165:1755-1760, 1987 Staples PJ, Talal N: Relative inability to induce tolerance in adult NZB and NZB/NZW .Fl mice. J Exp Med 129:123-139, 1969 Cohen PL, Ziff M: Abnormal polyclonal B cell activation in NZB/NZW Fl mice. J Immunol 119:1534-1537, 1977 Moutsopoulos HM, Boehm M, Truitt S, Kassan S, Chused TM: Demonstration of activation of B lymphocytes in New Zealand black mice at birth by an immunoradiometric assay for murine IgM. J Immunol 119:1639-1644, 1977 Izui S, McConahey PJ, Dixon FJ: Increased spontaneous polyclonal activation of B lymphocytes in mice with spontaneous autoimmune disease. J Immunol 121:2213-2219, 1978 Merino R, Fossati L. Lancour M, Izui S: Selective autoantibody production by Yaa+ B cells in autoimmune Yaa+Yaa-bone marrow chimeric mice. J Exp Med 174:1023-1029, 1991 Eastcott JW, Schwartz RS, Datta SK: Genetic analysis of the inheritance of B cell hyperactivity in relation to the development of autoantibodies and glomerulonephritis in NZB X SWR crosses. J Immunol 131:2232-2239, 1983 Jyonouchi HP, Kincade PW, Good RA, Gershwin ME: B lymphocyte lineage cells in newborn and very young NZB mice: Evidence for regulatory disorders affecting B cell formation. J Immunol 131:2219-2225, 1983 Perkins DL, Glaser RM, Mahon CA, Michaelson J, RothsteinMarshak A: Evidence for an intrinsic B cell defect in Ipr/lpr mice apparent in noenatal chimeras. J Immunol 145:549-555, 1990 Sobel ES, Katagiri T, Katagiri K, Morris SC, Cohen PL, Eisenberg RA: An intrinsic B cell defect is required for the production of autoantibodies in the Ipr model of murine systemic autoimmunity. J Exp Med 173:1441-1449, 1991 Nemazee D, Guiet C, Buerki K, Rothstein-Marshak A: B lymphocytes from the autoimmune-prone mouse strain MLR/lpr manifest an intrinsic defect in tetraparental MRL/lpr in equilibrium DBA/2 chimeras. J Immunol 147:2536-2539, 1991 Reininger L, Winkler TH, Kalberer CP, Jourdan M, Melchers F, Relink AG: Intrinsic B cell defects in NZB and NZW mice contribute to systemic lupus erythematosus in (NZB X NZW)F1 mice. J Exp Med 184:853-861, 1996


281

42. Budman DR, Merchant EB, Steinberg AD, Doft B, Gershwin ME, Lizzio E, Reeves JP: Increased spontaneous activity of antibodyforming cells in the peripheral blood of patients with active SLE. Arth Rheum 20:829-833, 1977 43. Reininger L, Radaszkiewicz T, Kosco M, Melchers F, Rolink AG: Development of autoimmune disease in SCID mice populated with long-term "in vitro" proliferating (NZB X NZW)F1 pre-B cells. J Exp Med 176:1343-1353, 1992 44. Fauci AS, Moutsopoulos HM: Polyclonally triggered B cells in the peripheral blood and bone marrow of normal individuals and in patients with systemic lupus erythematosus and primary Sjogren's syndrome. Arth Rheum 24:577-583, 1981 45. Sakane T, Suzuki N, Takada S, Ueda Y, Murakawa Y, Tsuchida T, Yamauchi Y, Kishimoto T: B cell hyperactivity and its relation to distinct clinical features and the degree of disease activity in patients with systemic lupus erythematosus. Arth Rheum 31:8087, 1988 46. Klinman DM, Shirai A, Ishigatsubo Y, Conover J, Steinberg AD: Quantiation of IgM- and IgG-secreting B cells in the peripheral blood of patients with systemic lupus erythematosus. Arth Rheum 34:1404-1410, 1991 47. Monestier M, Kotzin BL: Antibodies to histones in systemic lupus erythematosus and drug-induced lupus syndromes. Clin Rheum Dis 18:415-436, 1992 48. Burlingame RW, Rubin RL, Balderas RS, Theofilopoulos AN: Genesis and evolution of anti-chromatin autoantibodies in murine lupus implicates T-dependent immunization with self-antigen. J Clin Invest 91:1687-1696, 1993 49. Burlingame RW, Boey ML, Starkebaum G, Rubin RL: The central role of chromatin in autoimmune responses to histones and DNA in systemic lupus erythematosus. J Clin Invest 94:184-192, 1994 50. Amoura Z, Chabre H, Koutouzov S, Lotton C, Cabrespines A, Bach JF, Jacob L: Nucleosome-restricted antibodies are detected before anti-dsDNA and/or antihistone antibodies in serum of MRL-Mp Ipr/lpr and +/+ mice, and are present in kidney eluates of lupus mice with proteinuria. Arth Rheum 37:1684-1688, 1994 51. Marion TN, Tillman DM, Jou NT, Hill RJ: Selection of immunoglobulin variable regions in autoimmunity to DNA. Immunol Rev 128:123-149, 1992 52. Tsao BP, Cantor RM, Kalunian C, Chen C-J, Badsha H, Singh R, Wallace DJ, Kitridou RC, Chen S-I, Shen N, Song YW, Isenberg DA, Yu C-L, Hahn BH, Rotter JI: Evidence for linkage of a candidate chromosome 1 region to human systemic lupus erythematosus. J Clin Invest 99:725-731, 1997 53. Cook L, Agnello V: Complement deficiency and systemic lupus erythematosus. In Systemic Lupus Erythematosus, RG Lahita (ed). New York, Churchill Livingstone, 1992, pp 569-586