No abstract

Previous epidemiologic studies showed four times increased risk of acute lymphoblastic leukemia (ALL) in children of women with multiple sclerosis (MS). MS shows a risk association with Human leukocyte antigens (HLA)-DRA single nucleotide polymorphism (SNP) rs3135388, which is a proxy marker for DRB1*1501. We examined the relevance of rs3135388 in childhood ALL risk along with two other HLA-DRA SNPs in two case-control groups: 114 cases and 388 controls from South Wales (UK) and 100 Mexican Mestizo cases and 253 controls. We first confirmed the correlation between rs3135388 and DRB1*1501 in HLA-typed reference cell lines. We noted a female-specific risk association in childhood ALL (pooled odds ratio (OR) = 2.6, 95% confidence interval (CI) = 1.5-4.5, Mantel-Haenszel P = 0.0009) similar to the stronger association of DRB1*1501 in females with MS. Examination of an HLA-C 5' flanking region SNP rs9264942, known to correlate with HLA-C expression, showed a protective association in girls (OR = 0.4, 95% CI = 0.2-0.7, Mantel-Haenszel P = 0.0003) similar to the protective HLA-Cw*05 association in MS. In a reference cell line panel, HLA-Cw5 homozygous samples (n = 8) were also homozygous for the minor allele of the SNP. Likewise, the male-specific protective association of interferon-gamma (IFNG) SNP rs2069727 in MS was replicated with the same sex specificity in childhood ALL (OR = 0.6, 95% CI = 0.4-1.0, Mantel-Haenszel P = 0.03). Two other SNPs in superkiller viralicidic activity 2-like and tenascin XB that are markers for systemic lupus erythematosus susceptibility showed female-specific associations but due to linkage disequilibrium with HLA-DRB1*15. Our observations supported the epidemiologic link between MS and childhood ALL and added the sex effect to this connection. It appears that only girls born to mothers with MS may have an increased risk of ALL. Investigating the mechanism of these sex-specific associations may help understand the pathogenesis of MS and ALL.

Two novel class II associations are described in Mexicans with MS: DRB1*0403 and DRB1*0802. Both alleles share with DRB1*1501, valine-86 and negatively charged amino acids, in the DRB1-anchoring motif of pocket 4.

Understanding the complex interaction between human leukocyte antigen (HLA) and killer immunoglobulin-like receptor (KIR) requires study of both HLA and KIR diversity in the same population. The presence of KIR genes 2DL1, 2, 3, 4, 5, KIR3DL1, 3DL2, 3DL3, KIR2DS1, 2DS2, 2DS3, 2DS4, 2DS5, KIR3DS1, KIR3DP1, KIR2DP1 was determined in 54 unrelated Mexican Mestizo donors. The PCR sequence-specific oligonucleotide probe One Lambda kit (Luminex) kindly given by J. Lee was used for typing. The software analyses the combination obtained for each of the five exons. Five controls (UCLA DNA exchange) were run as quality control. The gene frequency (GF) was calculated for the 16 KIR loci; the GF of individual genes was 100% for 2DL4, 3DL1, 3DL2, 3DL3, 3DP1. KIR2DL1 (76.43%), KIR2DL2 (37.64%), KIR2DL3 (76.43%), KIR2DL5 (29.29%), KIR3DS1 (23.02%), KIR2DS1 (21.83%), KIR2DS2 (37.64%), KIR2DS3 (50.93%), KIR2DS4 (86.93%), KIR2DS5 (29.29%), KIR2DP1 (86.39%). We observed similar frequencies with Caucasians and Mediterraneans, with exceptions: KIR3DL1 which was present in 100% Mexicans, ranged from 62% to 75% in Caucasians; 2DS3 (50.9%) vs 14-20% 2DS4 (86.39%) vs 65-79% and 2DS5 (29.29%) vs 11-18% in Caucasians. The finding of 23 phenotypes in 54 individuals accounting for both chromosomes, demonstrates the enormous diversity. We found 14 different combinations of stimulatory KIRs in the phenotypes; every subject had at least one stimulatory KIR; in all of them, 2DS4 existed except for one person who may have some new combination: 2DS2 2DS3. Extended family data will offer accurate and precise haplotypes to provide an insight on the significance of ethnic distribution and KIR repertoire.

No abstract