Allelic spectrum in lupus

Overall, ANAs are much more common than lupus. ANAs may represent a general tendency towards autoimmunity. Almost all lupus patients have ANAs, but ANA alone has a low positive predictive value for the diagnosis of lupus. On the basis of evidence from mouse and human studies, the ANA phenotype appears to have a large genetic footprint. Furthermore, many autoimmunity alleles are associated with more than one autoimmune disease (e.g. PTPN22: RA, SLE, T1D autoimmune thyroid disease [Bottini et al 2004]; FCRL3: RA, SLE, thyroiditis [Kochi et al 2005], Table 1), and therefore cannot easily account for either the relative specificity of the autoantibody spectrum, or the end-organ pathology that distinguishes lupus from other autoantibody-mediated diseases (e.g. myasthenia gravis, Goodpasture's syndrome).

Wakeland's threshold liability model suggests that the SLE phenotype results from crossing a critical threshold number of active SLE-associated genes (Subramanian & Wakeland 2005). This model is based on observations of the effects of individual murine loci on mouse lupus development. It predicts that 4% of all individuals that simply present with ANAs but no other lupus manifestations have relatively common alleles that promote loss of tolerance to nuclear antigens. In NZ mice these would be represented by Slela, Slelb, Slelc, etc. A proportion of these individuals may also have alleles that result in dysregulation of the immune system (i.e. Sle2, Sle3, Sle5, Fas), and may have some manifestations of lymphoid hyperactivity but still do not present with full blown disease. Alleles that result in end-organ targeting (i.e. FCGR3A, Sleld) are necessary for pathogenic autoim-munity, typically with familial aggregation and characteristic end-organ manifestations (nephritis, neurological disorders, arthritis, vasculitis, etc.).

What remains to be determined is whether overall, the allelic spectrum of the general tendency towards autoimmunity, including the production of ANAs, is similar to that of specific end-organ pathology phenotypes that are identified as discrete autoimmune diseases (Fig. 1). In other words, is lupus the result of the cumulative effect of multiple qualitatively similar alleles, or does the added presence of rare alleles explain the pathophysiological pathway that distinguishes lupus from other autoimmune disease? Since weak autoimmunity alleles, which affect thresholds for lymphocyte activation, may confer some selective advantage in the context of pathogenic threat, they might obtain higher population frequencies (Ueda et al 2003). By contrast the complete lupus phenotype is rare and is likely to be subject to purifying selection by virtue of its presentation in women in childbearing years and effects on fecundity. It is then plausible that general autoimmunity alleles and autoimmune disease-specific alleles will follow different distributions.

There are arguments against alleles with strong phenotypic effects in lupus. First, very few single genes in humans have so far been found to confer strong

TABLE 1 Lupus-associated alleles

Gene

Human Replicated OR

Allele

PTPN22 lp'13

1.37

W620; C1858T

C1Q FCRL3

lp36.3 No lq2'l Yes

FCGR2A lq22-23 Yes

Stop codon

-169C

FCGR2B lq23

232T

FCGR3A lq23

F'158

FASLG CRP

CTLA4 PDCD1

lq23 Yes lq23.2 Yes

2q33 No 2q35-37 Yes

(GT intronic repeat) C-318T; A49G;

Mechanism

Other diseases

References

T cell activation RA, T1D, Graves'

disease

Handling of apoptotic Infection debris

Altered expression and RA, AITD

NF-kB binding IC handling; reduced IgG2 binding

IC handling; B cell signalling theshold; IgG2 binding; plasma cell differentiation IC handling; reduced

IgG'1/3 binding Tolerance ALPS

Handling of apoptotic Vascular disease debris

T cell activation T'lD; RA; MG; CD

threshold; tolerance T cell signalling T'lD; RA; Allergy theshold tolerance

Bottini et al 2004, Orozco et al 2005, Reddy et al 2005 Manderson et al 2004

Kochi et al 2005

Kyogoku et al 2002, Jonsen et al 2004, Magnusson et al 2004 Kono'et al 2005

Karassa et al 2003,

Magnusson et al 2004 Wu etal 1996 Russell et al 2004

Ueda et al 2003 Prokunina et al 2002

SPP1

4q22

1.57 +1239

Macrophage and B cell activation threshold

RA

Forton et al 2002

HLA DR3,

6p11-22

Yes

2.8

HLA DR3-DQ2-

Antigen presentation

Graham et al 2002,

DR2

C4AQ0

Schur et al 1982

C2

6p21

Yes

A25, B18, Drw2, BFS, C2Q0, C4A4B2 (type I)

IC handling; B cell activation threshold

HSP; polymyositis

Manderson et al 2004, Meyer et al 1985

C4A

6p21

Yes

AQ0

IC handling

T1D; bacterial meningitis

Meyer et al 1985

TNF

6p21.3

Yes

1.6

-308A

End-organ damage

Cerebral malaria susceptibility; mucocutaneous leishmaniasis; scarring trachomata; lepromatous leprosy; possible RA, asthma, AS

Sigurdsson et al 2005

IRF5

7q31

No

1.6

rs2004640 intronic SNP

End-organ damage: IFNa response

Sigurdsson et al 2005

MBL2

10q21

Yes

1.406

54A 57A LX

IC handling

Susceptibility to infection

Lee et al 2005, Sullivan et al 1996

FAS

10q24

No

RR 5.0

297C; 416G

Tolerance

ALPS

Horiuchi et al 1999

TYK2

19p13

No

1.6

S684

End-organ damage:

RA

Sigurdsson et al 2005

AITD, autoimmune thyroid disease; ALPS, autoimmune lymphoproliferative syndrome; AS, ankylosing spondylitis; CD, celiac disease; HSP, Henoch-Schonlein purpura; MG, myasthenia gravis; RA, rheumatoid arthritis; T1D, type 1 diabetes.

AITD, autoimmune thyroid disease; ALPS, autoimmune lymphoproliferative syndrome; AS, ankylosing spondylitis; CD, celiac disease; HSP, Henoch-Schonlein purpura; MG, myasthenia gravis; RA, rheumatoid arthritis; T1D, type 1 diabetes.

SNPs in genome

Number of genetic variants

Whole genome

Common shift

Positive selection

Rare shift

Non-synonymous protein coding mutations Purifying selection

Minor allele frequency

FIG. 1. Scenarios to describe the allelic architecture of lupus. (A,B) Distribution of pheno-typic effects of lupus and ANA associated alleles relative to the expected distribution of all SNPs (grey line). Lupus may be the outcome of the composite effect of numerous alleles that each cause a weak tendency towards autoimmunity (ANA production) (filled circles). Alternatively, an additional distinct set of alleles may account for the discrete lupus phenotype (open circles). A conservative estimate is that lupus-associated alleles will follow the overall QTL distribution (A and B), with the majority associated with weak phenotypic effects. (C,D) The allelic spectrum of ANA and lupus genes relative to variation across the genome (dark grey line). ANA is a common phenotype, and associated alleles may offer a weak selective advantage, resulting in a common shift (C) whereas lupus-associated alleles are likely to be rare and subject to purifying selection (D).

risk for SLE development, although this may be a consequence of the methods used to identify these genes. Second, many genetically-manipulated mouse models that identified single genes leading to murine SLE have now been questioned due to confounding genetic background or strain-dependent effects. For example, the effects of Fas, Fcgr2b, C1qa, Apcs (SAP) and Pdcdl turn out to reflect, at least in part, passively transferred lupus susceptibility genes from the 129/Sv ES cells used to generate the knockouts (Bygrave et al 2004). In the case of Apcs and Fcgr2b, the 129 interval surrounding these genes is sufficient to generate ANAs (including dsDNA on a BL/6 background).

On the other hand, evidence for genetic heterogeneity in lupus includes identification of linkage regions after stratification of pedigrees according to specific clinical manifestations, which was not previously evident. Examples include thrombocytopenia (where the LOD score at 1q22-23 increased from 2.75 to 3.71), the presence of a relative with self-reported rheumatoid arthritis (RA) (which revealed a novel linkage to 5p 15.3) (Namjoy et al 2002), neuropsychiatry disease (Nath et al 2002), anti-dsDNA antibodies (Namjou et al 2002), haemolytic anaemia, nucleolar ANA, renal disease (Quintero-Del-Rio et al 2002), vitiligo and discoid lupus erythematosus (Nath et al 2001). Finally, C1q represents an important precedent for a rare allele with strong effect resulting in the complex lupus phenotype. Multiple rare mutations that result in C1q deficiency cause severe, early onset lupus in more than 80% of affected individuals, and the penetrance of lupus characterized by ANA, dsDNA, skin, kidney and CNS disease is >90%, with a sex ratio of close to one (Meyer et al 1985, Manderson et al 2004).

Arthritis Relief Now

Arthritis Relief Now

When you hear the word arthritis, images of painful hands and joints comes into play. Few people fully understand arthritis and this guide is dedicated to anyone suffering with this chronic condition and wants relief now.

Get My Free Ebook


Post a comment