The role of genome-scale leukocyte long noncoding RNA in identifying acute aortic dissection

2.1 Design and population

From September 1, 2016 to November 30, 2016, 25 AAD patients and 15 healthy volunteers were included in the prospective study. Inclusion criteria were: age $\ge$18 years, onset time $\le$2 weeks, and patients were diagnosed with AAD by CTA. Exclusion criteria were: pregnant women, patients with AAD caused by trauma, and AAD combined with infection or blood system, immune system or cancer and other diseases. There were 5 cases in the AAD group and 5 cases in the healthy group in initial experiments, and further experiments to verify role of differentially expressed lncRNAs were conducted with another independent group comprising of 20 cases in the AAD group and 10 cases in the healthy group.

Experimental groups included blank control group: without any intervention factors; LPS group: Lipopolysaccharide (LPS, 1 $\mu$g/mL); target SILNC-2-1 group: transfection of SILNC-2-1 + LPS (1 $\mu$g/mL); target SILNC-2-2 group: transfection of SILNC-2-2 + LPS (1 $\mu$g/mL); SINC control group: transfection of no-load siRNA + LPS (1 $\mu$g/mL).

2.2 Data and collection

Characteristics of patients and healthy volunteers were collected, including age, gender, vital signs, previous disease history, treatment plan, and drug information. The laboratory results of AAD patients and healthy volunteers were recorded, including levels of leukocytes, neutrophils, monocytes, interleukin-6 (IL-6), c-reactive protein (CRP), and D-dimer.

2.3 Collection of leukocytes and subtypes

Peripheral venous blood (6 mL) of AAD patients and healthy volunteers were collected and packed into anticoagulant ethylene diamine tetraacetic acid (EDTA) tubes (Invitrogen, CA, USA). The red blood cells were lysed and centrifuged to obtain the white plaque cell layer (white blood cells); after adding trizol reagent, it was frozen at −80 ${}^{\circ }$C for the subsequent isolation of leukocyte subtypes. The separation solution (Tianjin Haoyang biological products Technology Co., Ltd., Tianjin, China) was added to extract the neutrophils and monocytes. Finally, the monocytes were purified by differential attachment method. After the cells were collected, trizol reagent was added and the samples were frozen at −80 ${}^{\circ }$C.

2.4 Extraction, labeling, and hybridization of RNA

Total RNA was extracted from leukocytes and their subtypes by trizol reagent (Invitrogen, CA, USA) and purified by mirVana™ miRNA Isolation Kit (AM1561) according to the manufacturer’s instructions. The OD260/280 reading was determined using a spectrophotometer (NanoDrop nd-1000) to determine the purity and concentration of RNA. The integrity of RNA was determined by 1% formaldehyde denaturing gel electrophoresis. Samples were labeled and array hybridized according to the gene expression analysis specification of monochromatic microarray (Agilent technique).

2.5 Microarray imaging and data analysis

The purified RNA was screened using Agilent lncRNAs expression microarray V4.0 (Agilent, Santa Clara, CA, USA). The chip scanner was used to scan the cleaned chip and obtain the hybrid picture, and the Agilent feature extraction (V10.7) software (Agilent, Santa Clara, CA, USA) was used to analyze the hybrid images and extract the data. Agilent gene printing software (Agilent, Santa Clara, CA, USA) was used to normalize the data and analyze the differences between groups, and thereby identify the lncRNAs with statistical differences. The differentially expressed genes were identified by random variance model, and p-value was calculated by paired t test. Up-regulated and down-regulated genes with fold change $\ge$2.0 and p $\le$ 0.05 were considered as significant genes.

2.6 Functions and pathways analysis

GO (Gene Ontology) is an international standard classification system of gene function, including biological process, molecular function and cell composition. Through GOseq software (http://www.bioconductor.org/), according to the distribution relationship of differential genes in GO classification, enrichment analysis of differential genes was carried out. KEGG (Kyoto encyclopedia of genes and genomes) pathway analysis is a functional analysis that maps genes to KEGG pathway. According to KEGG pathway analysis (http://www.genome.jp/kegg/), the main pathway of the differential genes was determined. P $<$ 0.05, was considered to indicate statistical difference between GO and pathway enrichment; the lower the p-value, the more obvious the enrichment difference [17].

2.7 Quantitative multi-polymerase chain reaction (qRT-PCR)

Differential gene expression between two groups was verified using qRT-PCR. According to the high-throughput screening, the difference multiple is large, the original expression amount is more than 300, and its regulatory signal pathway can be used as a candidate target gene in bioinformatics analysis. The total RNA was separated with trizol reagent and then reverse transcribed into cDNA using RNeasy Mini kit17 (Takara, Shiga, Japan) according to the manufacturer’s protocol. Primer 5.0 was used to design the primers for each lncRNAs, and BLAST (NCBI) was used to ensure the uniqueness of the amplified products. The$\beta$-actin (CST, Danvers, MA, USA) gene was selected as the endogenous reference gene. In the same way, the lncRNAs and its target genes were amplified (Table 1).

2.8 Cell culture and transfection

THP-1 cells (purchased from the Institute of Biochemistry and Cell Biology, Shanghai Institute of Life Sciences, Chinese Academy of Sciences, Shanghai, China) were cultured and sub-cultured in the complete cell culture medium with 10% fetal bovine serum. THP-1 cells (8 $\times$ 10${}^5$) were inoculated into 24-well plates, and then the 1ug plasmid containing siRNA-lnc was transfected into the THP-1 cells by Life Technologies (Carlsbad, CA, USA) to inhibit the expression of our candidate lncRNA ENST00000593848.1. After transfection, lipopolysaccharide (LPS; 1 $\mu$g/mL; Sigma, St. Louis, MO, USA) was added to the cells, and the cell suspension was collected after 4 hours of culture for subsequent protein and mRNA detection.

2.9 Western blot

The total protein extracted from THP-1 cells was subjected to 10% sodium dodecyl sulfate-polyacrylamide Sigma-Aldrich, St. Louis, MO, USA) gel electrophoresis, and the separated proteins were transferred to PVDF (polyvinylidene fluoride) membrane (Minipore, Shanghai, China). The membrane was then incubated in 5% skim milk for 2 hours at room temperature, followed by incubation with primary antibody at 4 ${}^{\circ }$C overnight. The membrane was then washed in PBS and incubated with appropriate secondary antibody (1:3000) for 2 hours at room temperature. All protein bands were detected by enhanced chemiluminescence Kit (Pierce, Rockford, IL, USA), and $\beta$-actin was used as the internal reference protein for group analysis.

2.10 Enzyme-linked immunosorbent assay (ELISA)

The concentrations of IL-6, CRP, and matrix metalloprotein 9 (MMP-9) in the supernatant of THP-1 cell culture medium and intracellular concentrations of Ca${}^{2+}$ were measured according to the instructions of the ELISA kit (CUSABIO, Wuhan, China).

2.11 Data analysis

All data are expressed as mean $\pm$ standard deviation. The data comparison between two groups was carried out by an independent sample t test. Comparison of more than two groups as done by one-way analysis of variance. The comparison between two groups adopted the least significant difference method. The area under receiver operating characteristic (ROC) curve (AUC) was compared for different lncRNAs to judge their diagnostic efficacy for AAD. All data were analyzed using SPSS19.0. (IBM, Armonk, NY, USA) and a p $<$ 0.05 was considered to be statistically significant.

3.1 Characteristics of patients and healthy controls

The lncRNA expression profile was examined in 5 patients with AAD and 5 healthy controls. The average age of the patients was 56 years old, and 60% and 40% of the patients displayed type A and type B AAD, respectively. There was no difference in age, hypertension, diabetes, and smoking behavior between the patient and control groups (p $>$ 0.05, Table 2).

3.2 Expression profile of leukocyte lncRNAs

The expression of plasma lncRNAs and mRNA was examined in the patient and control groups using the human crystal core® lncRNA mRNA expression profile chip (the chip analysis was completed by Beijing Bo’ao company, Beijing, China). The microarray results showed that there was a significant difference in the expression profile of lncRNAs in leukocytes between the patient and control groups. A total of 26,235 lncRNAs were detected by microarray, including 540 up-regulated lncRNAs ($>$2-fold, p $<$ 0.05) and 1078 down-regulated lncRNAs. Cluster analysis was carried out for the differentially expressed lncRNAs between the two groups. The respective expression values in the patient and control groups are expressed as the median. The fold change of each lncRNA and the corresponding p-value were mapped and subjected to GO and KEGG analysis to analyze the biological functions of the lncRNAs in leukocytes of the AAD patients (Fig. 1).

Fig. 1. Differential expression of lncRNAs in patients with acute aortic dissection (AAD) and control individuals. Hierarchical clustering analysis of 1618 lncRNAs that were differentially expressed in the two groups. (A) Expression values are represented in red and green, indicating expression above and below the median expression value in AAD patients (sample21, sample22, sample23, sample24, sample25) or control individuals (sample11, sample12, sample13, sample14, sample15), respectively. (B) Scatter plot of differential lncRNAs expression. X-axis: N-U, Y-axis: AMI-U. (C) Volcano plot of differential lncRNAs expression. X-axis: log2 fold change; Y-axis: −1 $\times$ log10 (corrected p-value) for each probes. (D) Biological functions of lncRNAs in leukocytes of AAD patients by GO pathway enrichment analysis.

3.3 Network interpretation of the co-expression of lncRNAs and mRNA

To verify the results of co-expression between lncRNAs and mRNA, the co-expression network was analyzed. The data of different lncRNAs and mRNA genes were combined to calculate the correlation between each gene in the two sets of data, and the correlation coefficient and corresponding p-values were obtained. The p-value indicates the reliability of measuring the correlation between genes. Based on the similarity of gene expression, the possible interaction between the genes was analyzed, and the degree of interaction was quantified; degree = 10 represents that there are 10 genes related to the gene, and a larger degree reflects a higher number of genes interactions [18]. Based on the degree of each gene in the gene synergy network, the key genes in the network were obtained (Table 3).

3.4 Verification of chip results

Based on the above information, combined with the needs of clinical research, the possible biomarkers were selected from 174 up-regulated lncRNAs, and the conditions were defined as: average basic intensity $>$8, multiple of difference $>$2, p $\le$ 0.005, and lncRNA co-expressed with mRNA (AAD group: degree $\ge$10; control group: degree $\ge$8). After screening, only three lncRNAs met these criteria: ENST00000430893.1, ENST00000536517.1, and ENST00000593848.1. To further identify the origin of the three lncRNAs, neutrophils and monocytes from another group of independent population (20 patients with AAD and 10 controls) were isolated, and the expression of the three lncRNAs was examined by qRT-PCR. The expression of ENST00000430893.1, ENST0000536517.1, and ENST0000593848.1 in neutrophils of AAD patients was 0.98 times (p = 0.703), 0.60 times (p $<$ 0.001) and 1.58 times (p = 0.010) higher than that of the control group, respectively. The expression levels of ENST00000430893.1, ENST00000536517.1, and ENST00000593848.1 in mononuclear cells of AAD patients were 1.21 times (p = 0.501), 0.62 times (p = 0.001) and 1.32 times (p = 0.094) higher than those of the control group, respectively (Fig. 2).

Fig. 2. Real-time fluorescence quantitative multi-polymerase chain reaction to verify the chip results. AAD, acute aortic dissection; Control, healthy volunteers.

3.5 The diagnostic value of lncRNAs for AAD

In ROC curve analysis, the AUC of ENST00000430893.1, ENST00000536517.1, and ENST00000593848.1 in neutrophils was 0.440 (p = 0.598), 0.885 (p = 0.001), and 0.790 (p = 0.011); the AUC of ENST00000430893.1, ENST00000536517.1, and ENST00000593848.1 in monocytes was 0.581 (p = 0.481), 0.880 (p = 0.001), and 0.725 (p = 0.048), respectively (Fig. 3).

Fig. 3. The diagnostic value of lncRNAs of leukocytes for acute aortic dissection (AAD) by receiver operating characteristic (ROC) curve.

3.6 Expression of inflammatory factors

The expression of lnc2 was detected 48 hours after the transfection of SINC, SILNC2-1, and SILNC2-2. The results showed that both SILNC2-1 and SILNC2-2 could effectively reduce the expression of lnc2 in THP-1 cells compared with that in the SINC group (p $<$ 0.05). Compared with the SINC group, SILNC2-1 and SILNC2 groups showed increased Ca${}^{2+}$ concentration in THP-1 cells after LPS induction (p $<$ 0.05); SILNC2-2 group showed significantly increased IL-6 secreted by THP-1 cells after LPS induction (p $<$ 0.05) (Fig. 4). However, the SILNC2-1 and SILNC2-2 groups did not show significantly increased CRP secretion by THP-1 cells after LPS induction.

Fig. 4. The expression of lnc2 was detected 48 hours after the transfection of SINC, SILNC2-1, and SILNC2-2. Intracellular Ca${}^{2+}$ concentration and the concentration of IL-6, CRP, MMP9 of cell culture medium in each group were tested by ELISA kit. (A) The expression of LNC2 was detected at 48 hours after the transfection of SINC, SILNC2-1 and SILNC2-2. (B) Intracellular Ca${}^{2+}$ concentration of each group were tested by ELISA kit. (C, D and E). The concentration of IL-6, CRP, MMP9 in the supernatant of cell culture medium of each group were tested by ELISA kit. &p $<$ 0.05 compared with blank control group; #p $<$ 0.05 compared with group of LPS; *p $<$ 0.05 compared with SINC.

3.7 Target genes and possible signaling pathways

Compared with the control group, the LPS group showed a significantly down-regulated level of solute carrier family 8 member A1 (SLC8A1) mRNA in THP-1 cells (p $<$ 0.05) (Fig. 5). Compared with SINC group, both SILNC2-1 and SILNC2-2 groups showed significantly upregulated levels of SLC8A1 and ERK1/2 mRNA in THP-1 cells after LPS induction (p $<$ 0.05). Compared with SINC group, both SILNC2-1 and SILNC2-2 groups showed significant activation of the NF-$\kappa$B/p65 signaling pathway in THP-1 cells after LPS induction (p $<$ 0.05). Compared with the control group, the LPS group showed increased p-ERK1/2 activity in THP-1 cells (p $<$ 0.05).

Fig. 5. Relative SLC8A1 and ERK mRNA expression level and differential expression of target protein signaling pathway related proteins in each group. (A and B) relative SLC8A1 and ERK mRNA expression level were tested by real-time fluorescence quantitative multi-polymerase chain reaction. (C, D and E) expression of target protein signaling pathway related proteins in each group of cells by Western Blot. &p $<$ 0.05 compared with blank control group; #p $<$ 0.05 compared with group of LPS; *p $<$ 0.05 compared with SINC.