Identification of hsa_circ_0018905 as a New Potential Biomarker for Multiple Sclerosis

Abstract

Multiple sclerosis (MS) is a demyelinating autoimmune disease characterized by early onset, for which the interaction of genetic and environmental factors is crucial. Dysregulation of the immune system as well as myelinization-de-myelinization has been shown to correlate with changes in RNA, including non-coding RNAs. Recently, circular RNAs (circRNAs) have emerged as a key player in the complex network of gene dysregulation associated with MS. Despite several efforts, the mechanisms driving circRNA regulation and dysregulation in MS still need to be properly elucidated. Here, we explore the panorama of circRNA expression in PBMCs purified from five newly diagnosed MS patients and five healthy controls (HCs) using the Arraystar Human circRNAs microarray. Experimental validation was then carried out in a validation cohort, and a possible correlation with disease severity was tested. We identified 64 differentially expressed circRNAs, 53 of which were downregulated in PBMCs purified from MS compared to the HCs. The discovery dataset was subsequently validated using qRT-PCR with an independent cohort of 20 RRMS patients and 20 HCs. We validated seven circRNAs differentially expressed in the RRMS group versus the HC group. hsa_circ_0000518, hsa_circ_0000517, hsa_circ_0000514, and hsa_circ_0000511 were significantly upregulated in the MS group, while hsa_circ_0018905, hsa_circ_0048764, and hsa_circ_0003445 were significantly downregulated; Among them, the expression level of hsa_circ_0018905 was significantly decreased in patients showing a higher level of disability and in progressive forms of MS. We described the circRNAs expression profile of PBMCs in newly diagnosed MS patients and proposed hsa_circ_0018905 as potential MS biomarker.

Keywords: biomarkers; circular RNAs; immune regulation; multiple sclerosis; new diagnosis.

Figure 1

Differentially expressed circRNAs in MS patients versus HCs, circRNA array analysis. (A) Volcano plots, used to visualize up- and downregulated genes across MS samples as compared to HCs. The red (up) and green (down) dots in the plot represent the significative differentially expressed circRNAs. (B) Clustered heatmap of the differentially expressed circRNAs showing the relationships among the expression levels of samples. Upregulation is shown in red, and downregulation is in green. (C) Table showing the list of circRNAs differentially expressed, depicting the top 7 upregulated and 10 downregulated.

Figure 2

Characteristics of the circRNAs identified in PBMCs of MS patients versus HCs. (A) Distribution of significantly upregulated circRNAs according to the chromosomal location. (B) Class distribution of upregulated circRNAs based on the genomic origins. (C) Distribution of significantly downregulated circRNAs according to the chromosomal location. (D) Class distribution of downregulated circRNAs based on the genomic origins.

Figure 3

Validation of the circRNAs identified in PBMCs of MS patients versus HCs. Expression levels in PBMCs of five upregulated and four downregulated circRNAs (A) and the corresponding cognate linear mRNAs (B) were measured by qPCR analysis. The levels of circRNAs and mRNAs were normalized to GAPDH mRNA levels. Data are the means and standard deviation (+SD) from at least three independent experiments. ** p < 0.01, *** p < 0.001.

Figure 4

Validation of the circRNAs in serum of MS patients versus HCs. The levels in serum of five upregulated and four downregulated circRNAs (A) and the corresponding mRNAs (B) were measured by qPCR analysis. The levels of circRNAs and mRNAs were normalized to GAPDH mRNA levels. Data are the means and standard deviation (+SD) from at least three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

Identification of the miRNAs and RBP Targets. (A) Schematic representation of circRNAs with putative miRNA binding site (MRE) and RNA-binding protein binding site (RBP-bs). (B,C) Tables showing list of human circRNA identified from our studies and target miRNAs and interacting RNA-binding proteins as determined by analysis performed using miRanda and circInteractome, respectively.

Figure 6

Network of circRNA-miRNA-mRNA for MS-associated genes. (A) Network of upregulated circRNAs and (B) downregulated circRNAs. CircRNAs are represented as red or green diamonds, miRNAs as red or green circles, and mRNAs as light red or light green rectangles. Red represents network generated from upregulated circRNAs and green from downregulated circRNAs.

Figure 7

Validation of the circRNA expression in PBMCs and correlation with disease severity. (A) Expression levels in PBMCs of five upregulated and three downregulated circRNAs in MS with different disease severity measured by RT-qPCR analysis. The levels of circRNAs were normalized to GAPDH mRNA levels. (B) Receiver operating characteristic (ROC) curve of differentially expressed circRNAs in MS vs. HCs. Green line, hsa_circ_0003445 and blue line, hsa_circ_0018905. Data are represented as the means and standard deviation (+SD) from at least three independent experiments. ** p < 0.01, *** p < 0.001.