Analysis of Interferon-Related Gene Expression in Thyroid Disease

Yihung Lai

Mentor: Drs. Mark Danielsen & Karen Ross, Georgetown University Medical Center

Date/Time: August 22nd, 2023 at 1pm.

Abstract: Graves’ Disease is an autoimmune disorder that affects the thyroid gland and often leads to hyperthyroidism. Previous work on a small number of samples suggested that genes involved in the interferon response were highly upregulated in patients with Graves’ Disease. Interferons play a pivotal role in orchestrating the immune response against viral infections and have been implicated in influencing autoimmune conditions. Remarkably, individuals undergoing interferon treatment have sometimes been observed to develop Graves’ disease. Thus, this project focuses on investigating the potential connection between Graves’ Disease and the interferon response by analyzing RNAseq and microarray gene expression data from monocytes collected from patients with thyroid disorders, including those with Graves’ disease, Hashimoto’s thyroiditis, thyroid cancer, and viral thyroiditis.

By consulting literature and databases, we compiled a list of 61 genes involved in the interferon response. Using RNAseq data from 39 patients with thyroid disease (55 samples), we constructed a heatmap of the expression of the 61 genes and performed hierarchical clustering on the samples. We found two distinct clusters within the heatmap. Notably, one of the clusters exhibited upregulation of previously identified interferon genes as well as some of the other 61 genes; we called this cluster Gi. This cluster contained 5 out of 6 Graves’ Disease samples while the remaining Graves’ sample was found in the other cluster (Gn). The Gi cluster also contained samples with Hashimoto’s thyroiditis and two samples with unknown diagnosis. We also looked at microarray expression data from 10 patients (20 samples). Here too, our analysis revealed the presence of the Gi cluster. Two out of the six Graves’ Disease samples were situated within this cluster. Thus, our analysis revealed an “interferon signature” in select samples, although it was not strictly correlated with the disease state.

Because monocytes are highly responsive to interferons, it was possible that the differences in expression of interferon-responsive genes was due to differing proportions of monocytes in the samples. Therefore, monocyte levels were quantified using Cibersortx analysis. Interestingly, samples in both groups contained a high proportion of monocytes (mean in G_i: 0.855; mean in G_n: 0.847) We used a two-tailed t-test to compare the means of two groups (G_i vs G_n) and found there was no significant difference (p = 0.611).

Despite observing an interferon signature in specific samples, the study highlights the complexity of the relationship between interferon response and disease, as not all Graves’ samples exhibit this pattern, and some patients with other thyroid disorders do exhibit the pattern. The investigation also underscores the need to comprehend the role of specific genes and pathways in distinguishing patient groups and disease states. In future work, we will examine gene expression in monocytes in patients with other infectious or autoimmune diseases to better understand what conditions induce the interferon response we observed in this study.