|Home | About | Journals | Submit | Contact Us | Français|
A genome analysis of mouse models may shed some light on the complex clinicopathological manifestations of systemic vasculitis. In the study of susceptibility loci to vasculitis in MRL mouse models, we found that systemic vasculitis developed through the cumulative effect of multiple gene loci, each of which by itself did not have a significant effect in inducing the related phenotype, thus indicating a polygenic system. The mice developed vasculitis in an additive manner with a hierarchical effect. Some of the susceptibility loci seemed to be common to those in other collagen diseases. Moreover, the loci controlling tissue specificity of vasculitis were present. One of the positional candidate genes for vasculitis showed an allelic polymorphism in the coding region, thus possibly causing a qualitative difference in its function. As a result, a particular combination of polygenes with such an allelic polymorphism may thus play a critical role in leading the cascade reaction to develop vasculitis, and also a regular variation of systemic vasculitis. This is designated as the polygene network in systemic vasculitis. (J Jpn Coll Angiol, 2009, 49: 11-16)
Systemic vasculitides are divided into primary and secondary types on the basis of their clinical features and pathological findings. The former includes vasculitis syndrome, which is regarded as a variation of collagen disease, and the latter includes vasculitides associated with other collagen diseases and those related to tumors, infections, and drug allergy. Vasculitis syndrome is clinically classified into various disease categories, but its actual clinical and pathological features are diverse, and not only its etiology but also pathogenesis is often unclear.
Concerning genetic factors of vasculitis, susceptibility genes have been analyzed by case-control studies of individual vasculitis categories from deductive viewpoints based on the conventional inflammatology/immunology using the data about polymorphisms and SNPs of candidate genes, particularly, about primary vasculitis. Genes targeted in such studies were, first, those of MHC, followed by inflammatory cytokines, chemokines, growth factors, their receptors, lectins, adhesion molecules, Fcγ receptors, and NO synthetase. The biological basis of various vasculitides is expected to be clarified by comprehensive analysis of these individual data. However, the allelic and genomic polymorphisms of each gene are extremely diverse, show racial variation, and are associated with a wide variety of environmental factors. Therefore, there are many difficulties in clarifying the genomic mechanism and pathogenesis of vasculitis from such information.
In this report, the results of a series of our studies simulating the genomic mechanism of vasculitis using a vasculitis model mouse MRL/Mp-lpr/lpr (MRL/lpr) are presented. Conclusions that we reached were: Systemic vasculitis included in collagen disease is a disease that can be genetically differentiated from other lesions of collagen disease, it develops under the control of multiple genes participating in an additive and hierarchical manner, and such genes can be understood as polygenes latently distributed in a population. The accumulation and combinations of susceptible alleles generated by genome crossing are considered to determine not only the occurrence and severity of vasculitis but also its tissue distribution and complication by other collagen diseases.
About 80% or more MRL/lpr mice develop systemic vasculitis at the age of 4–5 months. Vasculitis is remarkably observed in the arterial arches and interlobular arteries of the kidney, but it also occurs in small to medium-sized arteries of the pancreas, salivary glands, main aortic branches, tongue, and skeletal muscles of the limbs. Histopathologically, all lesions consist primarily of granulomatous arteritis. In its initial stage, systemic vasculitis begins with infiltration of CD4-positive lymphocytes around arteries but thereafter shows aggregates of activated macrophages and extends to destruction of the arterial external elastic laminae, degeneration of the media, and thickening of the intima.
This mouse was originally established by Murphy ED et al. of the Jackson Laboratory from an MRL strain of mice as a mutant mouse that develops systemic lymph node swelling and splenomegaly.1) This strain exhibits lymph node enlargement (lymphoproliferation, hence designated as lpr gene) inherited in an autosomal recessive pattern at the age of 3–4 months due to accumulation of CD4-CD8-B220+ T cells in lymph nodes. In addition, they spontaneously develop glomerulonephritis, arthritis, sialoadenitis, etc. as well as vasculitis. It also simultaneously exhibits various autoimmune phenomena. For this reason, this strain of mice has been studied widely as models of human polyarteritis nodosa, lupus nephritis, rheumatoid arthritis, and Sjögren syndrome.
When the MRL/lpr mouse was established, the series of manifestations of collagen disease was considered to be a single-gene disease due to the mutant gene lpr (a manifold effect of a single-gene mutation). In 1992, Nagata S’ group clarified that this lpr gene is a deletion mutant of Fas, which induces apoptosis. This is due to insertion of a transposon into the second intron.2) Fas molecules are expressed primarily in the activated T- and B-lymphocytes and macrophages in the immune system, and apoptosis of these cells is induced by binding of Fas ligand of activated T-lymphocytes. Therefore, Fas deletion mutation causes apoptotic insufficiency. This has been considered to lead to the appearance of auto-reactive lymphocytes, prevent the control of immune responses and reactions, permit sustained activation of these cells and prolongation of inflammation, and cause autoimmune diseases.
However, an important problem is that collagen disease does not occur with lpr alone. At least C3H/lpr and B6/lpr mice, which are other mouse strains created by transduction of lpr gene, rarely develop collagen disease including vasculitis while they both show lymph node enlargement and exhibit various phenomena of autoimmunity such as increases in immune complexes, anti-DNA antibodies, and rheumatoid factors in the circulation and cytokine abnormalities.3,4) Therefore, the phenotype of lpr only by itself is lymphoproliferation. On the other hand, MRL mice with no lpr gene manifest signs of collagen disease, though they are mild. Therefore, lpr gene can be understood as a “promoting factor” of collagen disease in MRL mice due to Fas-mediated apoptotic insufficiency, and the causative genes of collagen disease are considered to be background genes of MRL mice.3)
This concept has recently been confirmed by the accumulation of individual clinicopathological data obtained by forced expression or deletion of immunity-and inflammation-related genes. Thus, pathological traits of autoimmune diseases expressed by manipulation of a single gene differ according to the genetic background of the individual. For example, of the IL-1Ra deficient mice, BALB/c mice develop arthritis, and B6 mice develop arteritis.5,6) Among PD-1 deficient mice, B6 mice develop arthritis and glomerulonephritis, but BALB/c mice develop dilated cardiomyopathy, and the diabetic trait is promoted in NOD mice.7–9)
On the analysis of pathological traits of mice produced by backcrossing between MRL/lpr mice and C3H/lpr mice, which do not develop collagen disease (N2 generation), or brother-sister inbreeding (F2 generation) to reconstitute the background genes, individual features of collagen disease appear separately. Thus, in these generations, some animals exhibit glomerulonephritis, vasculitis, sialoadenitis, and arthritis alone or in various combinations.4) In addition, recombinant congenic McH5/lpr,10) which rarely develops nephritis but frequently develops severer vasculitis, was established by histopathologically selective brother-sister inbreeding at each generation after the backcrossed N2 generation. This means that vasculitis-susceptible alleles were isolated and fixed. Indeed, in these mice, all vasculitis-susceptibility loci mentioned below are occupied by susceptible alleles.
Recently, we have established 15 MXH/lpr strains as recombinant inbred (RI) strains between MRL/lpr and C3H/lpr mice (Fig. 1).11) In these strains, lines that develop glomerulonephritis, vasculitis, and arthritis randomly in various combinations were observed (Fig. 2). Moreover, autoantibodies that had been considered to be specific to particular pathological features were genetically dissociated in these RI strains, indicating that auto-antibodies are not necessarily involved in the pathogenic mechanisms of various diseases. Therefore, the collagen disease traits of MRL/lpr mice are considered to be determined by genetically isolatable background genes specific to individual clinical and pathologic features. It has recently become possible to identify autoantibodies that increase concurrently with the occurrence of vasculitis by exhaustive search of protein antigens corresponding to the autoantibodies in these 15 strains.
This series of phenomena suggests that the traits of a single-gene mutant generally differ according to the specific genetic background of the mouse, and MRL/lpr mice are considered to have indicated that many of the traits of knockout mice of various genes are not always specific to the genes.
In MRL/lpr x (MRL/lpr x C3H/lpr) F1 and (MRL/lpr × C3H/lpr) F2 mice, QTL analysis was performed according to pathomorphological severity of each collagen disease, using polymorphic microsatellite markers between the two strains, and then the susceptibility loci of each pathological feature were mapped (Table 1). At least 3 loci related to the development of renal vasculitis were found. Autoimmune renal vasculitis in MRL mice 1 (Arvm1) and Arvm2 were mapped as recessive susceptibility loci for MRL alleles, and Arvm3 located on D3Mit42 as a recessive suppressive locus.12)
An additive effect was observed in the pathogenesis of vasculitis among these loci (Fig. 3). The incidence of vasculitis was 90% when all 3 loci were occupied by susceptible alleles but was only 48–60% when only 1 locus was occupied by a susceptible allele. In addition, the incidence was 79% when both Arvm1 and Arvm2 were occupied by susceptible alleles, but no additive effect was noted when Arvm3 was combined with other loci, indicating the presence of hierarchy among these 3 loci.
In MRL/lpr mice, granulomatous arteritis occurs in not only renal arteries but also main branches of the aorta and limb arteries. By analysis of susceptibility loci for vasculitis concerning various arteries, they were mapped at sites different from those related to the kidney (Table 1).13) A vasculitis susceptibility locus of major branches of the aorta [designated as Autoimmune aortitis in MRL mice 1 (Aaom1)] was located very close to Arvm1, a susceptibility locus of renal vasculitis on chromosome 4, but Arvm2 and Arvm3, which are other susceptibility loci for renal vasculitis, showed no susceptibility. Also, the susceptibility loci for limb vasculitis Autoimmuni extremity vasculitis 1 (Aevm1) and Aevm2 completely differed from Aaom1 and were located on chromosome 8 and chromosome 5, respectively. This means the presence of genes that determine the tissue distribution of vasculitis.
QTL analysis in (MRL/lpr×C3H/lpr) F2 mice according to the morphological severity of glomerulonephritis suggested the presence of at least 3 loci related to the pathogenesis of glomerulonephritis (Table 1). We named them Autoimmune glomerulonephritis in MRL mice 1 (Agnm1), Agnm2 , and Agnm3.14) They showed recessive susceptibility, semidominant susceptibility, and semidominant susceptibility, respectively, with MRL alleles and additive effects as in vasculitis.
Of these loci, Agnm1 and Agnm2 are nearly identical with, or very closely to, the vasculitis susceptibility loci Arvm1 and Arvm2, respectively. Thus, the frequency of the concurrence of vasculitis and glomerulonephritis is considered to be high.
Cd72 is located on Arvm1, a renal vasculitis susceptibility locus, and Agnm1, a glomerulonephritis susceptibility locus. It is a type II transmembrane penetration protein expressed primarily in B cells, has an immunoreceptor tyrosine-based inhibition motif (ITIM), which induces negative signals against cell differentiation and activation in the intracellular region, and proliferation and differentiation of B cells under antigen stimulation. mRNA of this gene was found to be polymorphic between MRL and C3H mice, involving insertion of 3 bases in exon 7 and deletion of 21 bases in exon 8 were noted in MRLmice.12) This deletion in exon 8 was a result of aberrant splicing at the intron 7/exon 8 junction. Cd72 of MRL mice retained the ITIM, but the marked mutations at least at 2 sites in the extracellular region mentioned above may cause functional differences of CD72.
Generally, genetic factors of diseases can be classified into 2 types. In one, a disease occurs as a result of induction of particular functional abnormality due to deletion or mutation of a single gene. In the other, mutation may be involved but does not induce a disease alone, and the involvement of multiple genes including complementary genes and gene polymorphism is required for the development of the disease. The biological concept of polygenic inheritance (Mather, 1949) applies to the second type (polygenic diseases). Therefore, while a single gene consistently induces particular clinical and pathological conditions in the former without being affected by any genetic background of the individual, clinical and pathological features in the latter are affected by the genetic background of the individual and are presented primarily as continuous quantitative traits. Also, the latter is reported to be susceptible to the effects of environmental factors, and diverse clinical and pathological features may be observed depending on the combination of polygenes. At least on the basis of the results of our analysis in MRL model mice, genetic factors of vasculitis belong to the latter, and the combination of polygenes (polygene network) produced by genome crossing is also considered to determine whether vasculitis is complicated by other collagen diseases and the diversity of the tissue distribution of vasculitic lesions.
Masato Nose and the other authors have no conflict of interest.