Identification of MAMDC1 as a candidate susceptibility gene for systemic lupus erythematosus (SLE)

Abstract

Background: Systemic lupus erythematosus (SLE) is a complex autoimmune disorder with multiple susceptibility genes. We have previously reported suggestive linkage to the chromosomal region 14q21-q23 in Finnish SLE families.

Principal findings: Genetic fine mapping of this region in the same family material, together with a large collection of parent affected trios from UK and two independent case-control cohorts from Finland and Sweden, indicated that a novel uncharacterized gene, MAMDC1 (MAM domain containing glycosylphosphatidylinositol anchor 2, also known as MDGA2, MIM 611128), represents a putative susceptibility gene for SLE. In a combined analysis of the whole dataset, significant evidence of association was detected for the MAMDC1 intronic single nucleotide polymorphisms (SNP) rs961616 (P -value = 0.001, Odds Ratio (OR) = 1.292, 95% CI 1.103-1.513) and rs2297926 (P -value = 0.003, OR = 1.349, 95% CI 1.109-1.640). By Northern blot, real-time PCR (qRT-PCR) and immunohistochemical (IHC) analyses, we show that MAMDC1 is expressed in several tissues and cell types, and that the corresponding mRNA is up-regulated by the pro-inflammatory cytokines tumour necrosis factor alpha (TNF-alpha) and interferon gamma (IFN-gamma) in THP-1 monocytes. Based on its homology to known proteins with similar structure, MAMDC1 appears to be a novel member of the adhesion molecules of the immunoglobulin superfamily (IgCAM), which is involved in cell adhesion, migration, and recruitment to inflammatory sites. Remarkably, some IgCAMs have been shown to interact with ITGAM, the product of another SLE susceptibility gene recently discovered in two independent genome wide association (GWA) scans.

Significance: Further studies focused on MAMDC1 and other molecules involved in these pathways might thus provide new insight into the pathogenesis of SLE.

Figure 1. Identification of MAMDC1 as the SLE susceptibility locus in the 14q21-q23 linkage region.

Fine mapping towards the identification of MAMDC1 was first performed in the original Finnish family cohort by genotyping 19 MS markers and 17 SNPs, focusing on the two regions located on 14q11.2-q23 that showed haplotype sharing in the previous fine-mapping , . PDT and HPM analyses were used and regions providing significant results (p≤0.05) are shown in black and blue, respectively (top of the figure). The region between markers D14S1068 and rs1955810, containing the gene MAMDC1, was selected for further fine mapping (highlighted in blue). Twenty-four SNPs were subsequently genotyped in the whole sample material and as shown graphically in the figure, significant association for the SNPs rs961616 and rs2297926 could be identified using a combined analysis (bottom of the figure). The significance threshold were set to P = 0.0032 (see the statistics section under material and methods) and is represented by a red line.

Figure 2. Individual and pooled odds ratios for rs961616 and rs2297926.

The individual and pooled contribution of each sample population for rs961616 (A) and rs2297926 (B), shown as ORs.

Figure 3. Human MAMDC1 gene, mRNA and protein structure.

A) MAMDC1 genomic structure and exon-intron organization. Exons are reported as plain boxes with relative length (in bp) below. Intronic intervening sequences are also shown with relative length (in kb) above. Two alternative MAMDC1 mRNA isoforms are predicted to be transcribed from the MAMDC1 locus, corresponding to NCBI database entries NM_001113498 and NM_182830. Translation initiation codons (ATG) are indicated for both isoforms. B) Schematic representation of MAMDC1 predicted full-length protein (corresponding to mRNA isoform 1), and its structural domains with relative length (amino acid positions) below.

Figure 4. Comparison of homologs of the human MAMDC1 protein.

Alignment of MAMDC1 full-length polypeptide with other proteins containing identical structural domains (reported on the left: MAM, meprin/A5-protein/PTPmu; FNIII, fibronectin, type III-like fold, Ig-like, immunoglobulin-like; EGF, epidermal growth factor; MATH/TRAF, meprin and TRAF-C homology/TNF-receptor associated factor; FV/FVIII, coagulation factor V/VIII; CUB, complement C1r/C1s, Uegf, Bmp1). The branching diagram (cladogram) was generated by multiple sequence alignment of the protein sequences using ClustalW .

Figure 5. Analysis of MAMDC1 mRNA expression.

Northern blot analysis of MAMDC1 mRNA expression in different human tissues, showing several expressed splice variants. MAMDC1 mRNA transcript corresponding to the full-length isoform is indicated by an arrowhead on the right side. A β-Actin cDNA control probe was used for normalization (bottom).

Figure 6. Analysis of MAMDC1 protein expression.

IHC analysis of MAMDC1 protein expression in A) testis, showing positive staining in Leydig cells; B) testis, negative control; C) placenta, showing positive staining in syncytial trophoblasts (arrows); D) positive duodenal villi; E) kidney, positive staining observed in occasional glomerular neutrophils; F) kidney, negative control; G) SLE skin with positive staining in the upper dermis in the same region as elastic fibres; H) SLE skin stained with Weigert's Resorcin-Fuchsin, detecting elastic fibers: arrows in G and H mark corresponding regions I) Cutaneous squamous cell cancer, with positive staining in neutrophils. Scale bars: 5 µm (A, B, G, H), 2.5 µm (C, D, I), 1.6 µm (E, F).

Figure 7. Effect of selected cytokines on MAMDC1 mRNA expression in THP-1 monocytes.

THP-1 cells were treated for 24 h with LPS, TGF-β1, IL-1β, IFN-γ, TNF-α, or a combination of TNF-α and IFN-γ, and MAMDC1 mRNA expression was quantified by Real-Time PCR in triplicate experiments. The results are reported as fold changes relative to THP-1 cells grown in the absence of stimulation (control), with the smallest observation, lower quartile, median, upper quartile, and largest observation shown for each sample.