Identification of differentially expressed circular RNAs in human monocyte derived macrophages response to Mycobacterium tuberculosis infection

Macrophages act as the first line of host immune defense against Mycobacterium tuberculosis (Mtb). Recent studies have demonstrated circular RNAs (circRNAs) are implicated in a variety of physiological and pathological processes; however, the role of circRNAs in macrophages response to Mtb infection remain unknown. To address this issue, here we characterized circRNAs expression profiles in human monocyte derived macrophages (MDMs) response to Mtb infection using microarray assay. Our results revealed that many circRNAs were differentially expressed in human MDMs after Mtb infection; of these, 32 circRNAs were up-regulated and 110 were down-regulated. Real time PCR results were generally consistent with the microarray data. Furthermore, we found that hsa_circ_0043497 and hsa_circ_0001204 may be effective diagnostic biomarkers for TB. This study provides the first evidence that circRNAs alterations are involved in human MDMs response to TB infection and reveal potential targets for diagnostics and the treatment of TB.

Bacterial strains and culture. Mtb H37Rv strain 27294 was purchased from the American Type Culture Collection (ATCC). They were grown in Middlebrook 7H9 broth containing 10% oleic acid-albumin-dextrose-catalase (OADC) and 0.05% Tween-80 at 37 °C. They were washed with PBS and passed through a syringe, to disrupt the clumps, before infecting human MDMs.
Cell infection. The human MDMs were plated in 12-well plates, and incubated overnight. The cells were infected at a multiplicity of infection (MOI) of 5. Uninfected cells which received only PBS served as controls. After 4-6 h incubation at 37 °C, non-phagocytosed bacteria were washed off using PBS. MDMs were replenished with fresh RPMI 1640 and incubated for 24 h at 37 °C with 5% CO 2 . After 24 h, cells were harvested and RNA was isolated.
RNA isolation. Total RNA was extracted using TRIzol reagent (Invitrogen, USA) according to the manufacturer's protocol. The integrity of the RNA was assessed by electrophoresis on a denaturing agarose gel. A NanoDrop ND-1000 spectrophotometer was used for the accurate measurement of RNA concentration.
CircRNAs microarray and computational analysis. Human CircRNA Array V2.0 (8 × 15 K) is manufactured by Arraystar Technologies (Rockville, MD, USA). Six MDMs samples, including three infected samples and three uninfected samples, were sent to KangChen Bio-tech (Shanghai, China) for the Arraystar circRNA microarray analysis. Microarray hybridization were performed according to the protocols of Arraystar. The scatter plot is a visualization method used for assessing the circRNA expression variation. Differentially expressed circRNAs with statistical significance (fold changes >2.0 and P < 0.05) between groups were identified using fold change cut-off or volcano plot filtering, respectively. The circRNAs/microRNAs (miRNAs) interaction was predicted using Arraystar's home-made miRNA target prediction software based on TargetScan and miRanda. Quantitative real-time PCR. Total RNA (2 μg) was reversely transcribed into cDNA using the Reverse Transcription System Kit (Takara, Dalian, China). The expression levels of circRNAs and miRNAs were determined by quantitative real-time PCR (RT-qPCR) using SYBR Master Mix (Applied Biosystems, Foster City, CA, USA) and mirVanaTM RT-qPCR miRNA Detection Kit (Ambion, Austin, USA) on Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, Foster City, USA), respectively. Primers used in this study were listed in Supplementary Table 1. The PCR primers of U6 and miRNAs were purchased from RIBOBIO. All RT-qPCR experiments were performed in triplicate. The relative expression levels of circRNAs and miRNAs were calculated by the 2 −ΔΔCt method. The the relative expression levels of circRNAs were normalized to GAPDH and the relative expression levels of miRNAs were normalized to U6 10 . PCR product was examined by agarose gel electrophoresis using 2% (w/v) LE agarose (Seakem) stained with ethidium bromide.

Statistics.
Numerical data were shown as the mean ± standard error of the mean (SEM). A one-way ANOVA test, Mann-Whitney test or Student t-test was used for statistical analysis. Receiver operating characteristic (ROC) analysis was used to evaluate the power of candidate circRNAs. All statistical tests were performed with GraphPad Prism 5.0 (GraphPad Software, San Diego, USA). P < 0.05 was considered statistically significant.

Results
Overview of circRNA profiles. To explore the potential circRNAs involved in human MDMs response to TB infection, we examined the circRNAs expression profiles through microarray analysis. In total, 142 cir-cRNAs were identified with differential expression between Mtb-infected group and uninfected group (fold   Table 2). Specifically, the most up-regulated circRNAs were: hsa_circ_0003528, hsa_ circ_0001417, hsa_circ_0043497 and hsa_circ_0038929, of which hsa_circ_0003528 was the highest. The most highly down-regulated were: hsa_circ_0068784, hsa_circ_0057090, hsa_circ_0056247 and hsa_circ_0001204, of which hsa_circ_0068784 showed the largest down-regulation. In present study, the top 40 up-and down-regulated circRNAs are listed in Table 1 by fold change. We summarized the classification of dysregulated circRNAs (Fig. 1). Among the up-regulated circRNAs, there were 29 exonic, 2 intronic and 1 sense overlapping. Among the down-regulated circRNAs, there were 101 exonic, 6 intronic and 3 sense overlapping.
Validation of circRNAs expression. To validate microarray analysis findings, we randomly selected 5 circRNAs from the differentially expressed circRNAs with fold change >3.0 and analyzed their expression by RT-qPCR in expanded MDMs samples. The results showed that these circRNAs were significantly differentially expressed. Particularly, hsa_circ_0030045, hsa_circ_0001417 and hsa_circ_0043497 are significantly up-regulated in Mtb-infected group, while hsa_circ_0030569 and hsa_circ_0001204 are significantly down-regulated in Mtb-infected group (Fig. 2). Our data confirmed the findings of the microarray analysis.

Diagnostic value of hsa_circ_0043497 and hsa_circ_0001204 expression levels in TB.
Recently, many studies have demonstrated that circRNAs can serve as novel diagnostic markers for diseases. Thus, we sought to determine whether the circRNAs identified in our microarray analysis could potentially serve as diagnostic biomarkers for TB. The levels of 5 circRNAs (hsa_circ_0030045, hsa_circ_0001417, hsa_circ_0043497, hsa_circ_0030569 and hsa_circ_0001204) were measured in PBMCs samples from 96 patients with active pulmonary TB and 85 healthy controls using RT-qPCR. The demographic characteristics of the participants are showed in Table 2. Of the five circRNAs studied, the levels of hsa_circ_0043497 were significantly elevated and hsa_ circ_0001204 were significantly reduced in PBMCs of TB patients as compared to healthy controls. There were no significant differences in the level of hsa_circ_0030045, hsa_circ_0001417 and hsa_circ_0030569 between TB patients and controls (Fig. 3). To compare the diagnostic value of hsa_circ_0043497 and hsa_circ_0001204 as candidate biomarkers of TB, we performed ROC curve analysis for hsa_circ_0043497 and hsa_circ_0001204. ROC curve analysis exhibited a significant distinguishing efficiency with an area under the curve (AUC) value of 0.860 for hsa_circ_0043497 and 0.848 for hsa_circ_0001204 (Fig. 4). These results suggest that altered expression of hsa_circ_0043497 and hsa_circ_0001204 in PBMCs are potential novel biomarkers for the diagnosis of TB.
Effects of anti-TB therapy on level of hsa_circ_0043497 and hsa_circ_0001204. Furthermore, we compared the expression of hsa_circ_0043497 and hsa_circ_0001204 in 12 patients before and after TB therapy. As compared to their pre-treatment, the levels of hsa_circ_0043497 decreased after anti-TB treatment. As compared to controls, the mean levels of hsa_circ_0043497 which was higher in TB infected group came to near normal after therapy. The mean hsa_circ_0043497 levels did not differ significantly between healthy control and TB treated group. As compared to healthy controls, the mean levels of hsa_circ_0001204 were significantly lower in TB infected group; the levels increase post therapy. The mean hsa_circ_0001204 levels did not differ significantly between healthy control and TB treated group (Fig. 5).

Target gene prediction.
To evaluate circRNAs potential functions, we investigated potential miRNAs binding with circRNAs using Arraystar's home-made miRNA target prediction software. We selected 18 differentially expressed circRNAs with the highest fold-change (fold-change > 3.0; P < 0.05) to predict their miRNA response elements (MREs), including 9 up-regulated circRNAs and 9 down-regulated circRNAs. Five MREs with good mirSVR scores for each circRNA are shown in Table 3. In the study, two confirmed circRNAs (hsa_ circ_0043497 and hsa_circ_0001204) and their target miRNAs were focused (Fig. 6). Then, the expression levels of hsa_circ_0043497-target miRNAs and hsa_circ_0001204-target miRNAs were validated by RT-qPCR in ten additional human MDMs after 24 h of infection with Mtb. Of the ten miRNAs studied, the levels of miR-186-5p were significantly reduced and the levels of miR-377-3p were significantly elevated in Mtb-infected MDMs (Fig. 7).  Relative levels of hsa_circ_0043497 and hsa_circ_0001204 in healthy controls (n = 32), pulmonary TB naive patients (n = 36) and TB treated group (n = 12). The value between the healthy control and TB treated group is not significantly different. ***P < 0.001.

Discussion
CircRNAs are a new class of long non-coding RNAs which do not have 5′ or 3′ ends but are covalently linked to form a closed circular structure. Recent evidence has suggested that circRNAs play essential roles in various physiological and pathological processes 15 . The dysregulation of circRNAs have also been implicated in the occurrence and progression of many human diseases 16 . However, the relationship between circRNAs and TB infection has not been reported.
In this study, we comprehensively profiled the circRNAs expression of human MDMs response to Mtb infection in order to improve our understanding of the pathogenesis of TB infection as well as identify potential circRNAs biomarkers for TB. We demonstrated that the expression of 142 circRNAs was significantly different in the Mtb-infected group compared with uninfected group. Among these, 32 circRNAs were up-regulated, while 110 circRNAs were down-regulated. Most of differentially expressed circRNAs originate from the exons. Some are from introns, while, a few are other sources. We then randomly selected five circRNAs from the differentially expressed cricRNAs with a fold change >3.0 and performed a quantitative real time PCR to examine these circR-NAs' expression levels in Mtb-infected MDMs. RT-qPCR results were generally consistent with the microarray data. To our knowledge, this is the first report on the expression of circRNAs in the human MDMs response to TB infection.
Early diagnosis of TB infection is essential for controlling the spread of the disease and providing early therapy for TB epidemic 17 . Several groups have reported the potential use of circRNAs as promising biomarkers for detection of human diseases 18,19 . For example, recently, Cui et al. 20 showed that hsa_circRNA_103636 in PBMCs can be used as a novel biomarker for the diagnosis and treatment of major depressive disorder. Chen et al. 21 reported that hsa_circ_0000190 yields a diagnostic characteristic with a sensitivity of 71.2% and a specificity of 75.0% for detection of gastric cancer. However, there are no reports regarding the use of circRNAs as biomarkers for  Table 3. Annotation for differentially expressed circRNAs/miRNAs interaction. Note: We selected 18 differentially expressed circRNAs with the highest fold-change (fold-change >3.0; P < 0.05) to predict their miRNAs response elements (MREs), including 9 up-regulated circRNAs and 9 down-regulated circRNAs. The circRNAs/miRNAs interaction was predicted with Arraystar's home-made miRNA target prediction software based on TargetScan and miRanda.  TB. In this study, there were many differentially expressed circRNAs induced by Mtb infection, which indicated that circRNAs have the potential to be novel biomarkers for TB. To identify the clinically applicable biomarkers, 5 significant difierentially expressed circRNAs were selected for further analysis. We found the expression of hsa_circ_0043497 and hsa_circ_0001204 were significantly elevated in TB infected patients compared to healthy subjects. In ROC plots, the AUC was 0.856 for hsa_circ_0043497 and 0.882 for hsa_circ_0001204. Next, we investigated the effect of anti-TB drug treatment on hsa_circ_0043497 and hsa_circ_0001204 expression. In the present study, we observed that the levels of hsa_circ_0043497 and hsa_circ_0001204 which were altered in naive TB patients reverted to normal after completion of anti-TB therapy and proven to be acid-fast bacillius (AFB) negative in sputum. However, a larger sample size is needed to confirm our results. Some studies have revealed that circRNAs could function as miRNAs sponges, modulate alternative splicing or transcription, and regulate the expression of parental genes 7,8 . To evaluate hsa_circ_0043497 and hsa_ circ_0001204 potential functions, we investigated potential miRNAs binding with hsa_circ_0043497 and hsa_circ_0001204. The potential miRNAs targets of hsa_circ_0043497 include miR-335-3p, miR-186-5p, miR-380-5p, miR-296-3p and miR-522-3p. For hsa_circ_0001204, the potential miRNAs targets include miR-612, miR-657, miR-362-3p, miR-377-3p and miR-136-5p. These potential target miRNAs were verified in MDMs with or without Mtb infection using RT-qPCR. And the data of RT-qPCR showed that the expression levels of miR-377-3p were markedly elevated and the expression levels of miR-186-5p were significantly reduced in Mtb-infected MDMs. However, due to the limited known function of circRNAs and miRNAs, a lot of circRNAs/ miRNAs interactions should be analyzed in the future.
In conclusion, in the study, we described differentially expressed circRNAs in human MDMs response to TB infection. Furthermore, hsa_circ_0043497 and hsa_circ_0001204 were identified as potential non-invasive hsa_circ_0043497. The circRNAs/miRNAs interaction was predicted with Arraystar's home-made miRNA target prediction software based on TargetScan and miRanda. Binding sites of conserved miRNAs with good mirSVR scores are represented. molecular markers for rapid diagnosis of TB. Our findings shed a novel light on our understandings of the pathogenesis of TB infection. To our knowledge, this is the first research addressing circRNAs expression profiles in macrophages response to TB infection. Further studies should focus on the function of circRNAs involved in TB infection, which may lead to new theories for TB pathogenesis and give new potentially therapeutic targets in active TB. Figure 7. Expression levels of hsa_circ_0043497 and hsa_circ_0001204 and their target miRNAs in Mtbinfected human MDMs. RT-qPCR was performed to evaluate the expression levels of hsa_circ_0043497 (A) and its top 5 predicted miRNA targets (miR-335-3p, miR-186-5p, miR-380-5p, miR-296-3p and miR-522-3p) (B), hsa_circ_0001204 (C) and its top 5 predicted miRNA targets (miR-612, miR-657, miR-362-3p, miR-377-3p and miR-136-5p) (D) in ten human MDMs after 24 h of infection with H37Rv. Data are expressed as the means ± SEM. The relative expression levels of circRNAs were normalized to levels of GAPDH and the relative expression levels of miRNAs were normalized to levels of U6. **P < 0.01, ***P < 0.001 compared to the control group.