Analysis of immune regulatory genes' copy number variants in Graves' disease

Abstract

Background: Copy number variants (CNVs) have recently been reported to be associated with several autoimmune conditions. Moreover, loci involved in immunity are enriched in CNVs. Therefore, we hypothesized that CNVs in immune genes associated with Graves' disease (GD) may contribute to the etiology of disease.

Methods: One hundred ninety-one North American Caucasian GD patients and 192 Caucasian controls were analyzed for CNVs in three major immune regulatory genes: CD40, PTPN22, and CTLA-4. Copy number was determined using quantitative-PCR (Q-PCR) assays specifically designed for determining copy numbers in genomic DNA. Additionally, a well-characterized CNV in the amylase gene was typed in a separate dataset of DNA samples that were derived from cell lines or blood.

Results: No CNVs could be confirmed in the CD40 and CTLA-4 genes, even though a CD40 CNV is cataloged in the Database of Genomic Variants. Only the PTPN22 CNV was confirmed in our cohort, but it was rare and appeared in only two individuals. A key finding was that the source of DNA has a significant effect on CNV typing. There was a statistically significant increase in amylase locus deletions in cell line-derived DNA compared to blood-derived DNA samples.

Conclusions: We conclude that CNV analysis should be performed only using blood-derived DNA Samples. Additionally, the CTLA-4, CD40, and PTPN22 loci do not harbor CNVs that play a role in the etiology of GD.

FIG. 1.

Comparison of copy number variants between blood- and cell line-derived DNA. To compare the differences between blood- and cell line-derived DNA, we analyzed a well-characterized CNV in the amylase gene. Calculated copy numbers of the AMY2A gene in blood-derived (left) and cell line-derived (right) DNA samples are shown. Each dot represents the copy number of each individual sample. The area between the horizontal black bars represents the range of values that constitutes two copies. Gray-shaded regions signify borderline values that cannot be clearly assigned, either a one or two copies (lower region) or a two or three copies (upper region). Samples that fell in these gray areas were excluded from analysis. DNA samples derived from cell lines showed a significant increase in the number of samples showing deletions compared to those derived from whole blood (p = 0.008).

FIG. 2.

Copy number variation analysis of CD40, PTPN22, and CTLA-4. TaqMan CNV assays were selected for CD40 and PTPN22 covering CNVs currently deposited in the Database of Genomic Variation. For CTLA-4, however, there was no cataloged CNV in the database, so a CNV assay encompassing the 5′UTR of the CTLA-4 gene was chosen. Each dot represents the copy number of each individual sample. The area between the horizontal black bars represents the range of values that constitutes two copies. Gray-shaded regions signify borderline values that cannot be clearly assigned, either a one or two copies (lower region) or a two or three copies (upper region). Samples that fell in these gray areas were excluded from analysis. Calculated copy numbers of CD40 (A), PTPN22 (B), and CTLA-4 (C) in control and Graves' disease (GD) patient samples are shown. CD40 and CTLA-4 showed no copy number variations in either the controls or the GD patients. PTPN22 showed no variation in the GD population, and only rare variation, one duplication and one deletion, in the control group.