HLA‐DPB1 is associated with ANCA ‐associated vasculitis in children

OBJECTIVE

Antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is a rare, life-threatening inflammation of blood vessels that can affect both adults and children. Compared to adult-onset disease, AAV is especially rare in children, with an annual prevalence of 0.5-6.4 cases per million children. The etiology of AAV remains largely unknown, and both environmental and genetic factors are likely involved. The present study was undertaken to explore the genetic susceptibility factors recently identified in adult patients, including HLA-DP and HLA-DQ, in pediatric patients.

METHODS

We performed a genome-wide association study of pediatric AAV in patients of European ancestry (n = 63 AAV cases, n = 315 population-matched controls).

RESULTS

We identified a significant genetic association between pediatric AAV and the HLA-DPB1*04:01 allele (P = 1.5 × 10^-8 , odds ratio [OR] 3.5), with a stronger association observed in children with proteinase 3-ANCA positivity than in children with myeloperoxidase-ANCA positivity. Among the HLA alleles, the HLA-DPB1*04:01 allele was the most highly associated with AAV, although not significantly, in a follow-up adult AAV cohort (P = 2.6 × 10^-4 , OR 0.4). T cell receptor and interferon signaling pathways were also shown to be enriched in the pediatric AAV cohort.

CONCLUSION

The HLA-DPB1 locus showed an association with pediatric AAV, as similarly shown previously in adult AAV. Despite the difference in the age of onset, these findings suggest that childhood- and adult-onset vasculitis share a common genetic predisposition. The identification of genetic variants contributing to AAV is an important step to improved classification tools and treatment strategies.

THE IMMUNOLOGICAL SPECIFICITY OF CHEMICALLY ALTERED PROTEINS : HALOGENATED AND NITRATED PROTEINS

The serological properties of iodoproteins prepared by a method which involves less drastic treatment of the protein than the methods previously used for this purpose confirm the findings of Obermayer and Pick (1) and later authors, that iodination of proteins results in a more or less complete loss of species specificity and that a new specificity characteristic for iodoproteins is produced. A serological investigation of brominated proteins has been made for the first time. These preparations are only slightly different from iodized proteins in their serological properties. Evidence is submitted which indicates that the radical in iodoproteins which is responsible for the specificity is not iodine but the 3:5 dihalogenated tyrosine grouping. Thus marked inhibition of the iodoprotein (or bromoprotein) precipitin reactions is effected by 3:5 dihalogenated tyrosine, not by iodophenol or potassium iodide. A reinvestigation has been made of the serological properties of nitrated and diazotized proteins. Proteins nitrated by nitric acid, or by a method which does not appear to have been used for proteins hitherto, namely nitration with tetranitromethane in neutral or slightly alkaline solution, acquire a new common serological specificity. The nitrated proteins and diazotized proteins show, in confirmation of the results of Landsteiner and Prášek (22) and in contrast to the findings of Obermayer and Pick, very little difference in their reactions. Thus diazotized proteins and proteins nitrated by either of the two methods above mentioned react equally well with any nitroprotein antiserum. This interaction exists in spite of the difference in the substituents, either because the substitution with the nitro- or diazo-group occurs in the same position in the aromatic nucleus, possibly in the ortho position to the hydroxyl group, or because of some other structural similarity. In the last connection it is suggested that both compounds may have a quinoid structure as has been assumed for ortho-nitrophenols. Whilst this assumption could account for the marked serological difference of nitrated and halogenated proteins it should also be mentioned that iodination (and bromination) lead to a disubstitution of halogen in the two ortho positions relative to the hydroxyl group of the tyrosine whereas nitration of proteins probably results in the formation of mononitrotyrosine and substitution in the tryptophane group as well (19, 36, 20-a). It is probably impossible therefore, to draw a strict analogy between nitration (or diazotization) and halogenation of proteins since a comparison of their immunological properties is not exactly a comparison of the effect of substituting a different group in the same position. Accordingly it would appear that as yet no definite conclusions can be drawn as to the serological effect of differences in the chemical nature of various substituents in the aromatic nucleus although some influence is likely for general reasons. All of the chemically altered proteins still retain a small amount of the original species specificity, and the antisera always react to a slightly greater extent with the homologous antigen than with similarly treated antigens prepared from heterologous sera. This difference occurs even when the possibility of some unaltered protein being present in the antigen can be practically excluded.