Differentially expressed genes in orbital adipose/connective tissue of thyroid-associated orbitopathy

Abstract

Background: Thyroid-associated orbitopathy (TAO) is a disease associated with autoimmune thyroid disorders and it can lead to proptosis, diplopia, and vision-threatening compressive optic neuropathy. To comprehensively understand the molecular mechanisms underlying orbital adipogenesis in TAO, we characterize the intrinsic molecular properties of orbital adipose/connective tissue from patients with TAO and control individuals.

Methods: RNA sequencing analysis (RNA-seq) was performed to measure the gene expression of orbital adipose/connective tissues of TAO patients. Differentially expressed genes (DEGs) were detected and analyzed through Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and Gene Set Enrichment Analysis (GSEA). The protein-protein interaction (PPI) network was constructed using the STRING database, and hub genes were identified by the Cytoscape plug-in, cytoHubba. We validated several top DEGs through quantitative real-time polymerase chain reaction (qRT-PCR).

Results: We identified 183 DEGs in adipose tissue between TAO patients (n = 3) and control patients (n = 3) through RNA sequencing, including 114 upregulated genes and 69 downregulated genes. The PPI network of these DEGs had 202 nodes and 743 edges. PCR-based validation results of orbital adipose tissue showed multiple top-ranked genes in TAO patients (n = 4) are immune and inflammatory response genes compared with the control individual (n = 4). They include ceruloplasmin isoform x3 (CP), alkaline tissue-nonspecific isozyme isoform x1 (ALPL), and angiotensinogen (AGT), which were overrepresented by 2.27- to 6.40-fold. Meanwhile, protein mab-21-like 1 (MAB21L1), phosphoinositide 3-kinase gamma-subunit (PIK3C2G), and clavesin-2 (CLVS2) decreased by 2.6% to 32.8%. R-spondin 1 (RSPO1), which is related to oogonia differentiation and developmental angiogenesis, was significantly downregulated in the orbital muscle tissues of patients with TAO compared with the control groups (P = 0.024).

Conclusions: Our results suggest that there are genetic differences in orbital adipose-connective tissues derived from TAO patients. The upregulation of the inflammatory response in orbital fat of TAO may be consistent with the clinical phenotype like eyelid edema, exophthalmos, and excess tearing. Downregulation of MAB21L1, PIK3C2G, and CLVS2 in TAO tissue demonstrates dysregulation of differentiation, oxidative stress, and developmental pathways.

Keywords: Differentially expressed genes; High-throughput sequencing; Inflammation; Thyroid Associated Ophthalmopathy; mRNA.

Figure 1. The differentially expressed genes were analyzed from RNA sequencing data.

(A) Volcano plot of different genes in control or TAO orbital fat. FC, fold change; DEGs, differentially expressed genes. (B) Hierarchical clustering heatmap showing gene expression differences.

Figure 2. Analysis of differential alternative splicing (AS) genes and distribution of the five main AS events.

(A) Schematic diagrams of the mechanisms of the five main AS events. (B) Venn diagram of the detected genes undergoing the five AS events and overlap of these genes. SE, exon skipping; RI, intron retention; A5SS, alternative 5′splice site; A3SS, alternative 3′splice site; MXE, mutually exclusive exons. (C) Distribution of differential AS events based on a threshold of P < 0.01.

Figure 3. The most significantly enriched GO terms and KEGG pathway analysis relevant to up- and downregulated genes.

(A) BP term of GO enrichment analysis, *p < 0.001. BP, biological process. (B) MF term of GO enrichment analysis, *p < 0.05. MF: the molecular function. (C) CC term of GO enrichment analysis, *p < 0.05. CC: cellular component. (D) KEGG pathway analysis showing pathways that are enriched in the TAO group. (E) Gene cluster enrichment analysis (GSEA) revealed a significant enrichment of the first five pathways in TAO patients.

Figure 4. The Venn diagram and the top hub genes identified in the protein–protein interaction (PPI) networks.

(A) The Venn diagram shows the differentially expressed gene identification in the two gene expression profile datasets. (B) PPI network of differentially expressed genes. (C) Identification of the top 10 hub genes.

Figure 5. Validation of the expression levels of mRNAs in the TAO groups and control groups.

(A and B) The mRNA expression levels in orbital adipose tissue as verified by qRT–PCR. (C and D) Expression levels of mRNAs in orbital muscle tissues as verified by qRT–PCR. The results are presented as the means ± SDs; n = 4, * p < 0.05, and **p < 0.01 for each pair of groups indicated.

Figure 6. Orbital adipose tissue inflammation in the TAO patients and control individuals.

(A) H & E staining in paraffin sections of orbital fat; scale bar, 25 µm. (B) Immunohistochemistry for CD45 (black arrows) and a hematoxylin nuclear counterstain (blue) was performed on orbital adipose tissue; scale bar, 25 µm. (C) Immunofluorescence detection of the macrophage-specific antigen F4/80 (green) in orbital adipose tissue from TAO patients and control individuals; scale bar, 25 µm.