ABSTRACT

Objective:

Breast cancer (BC) is the most prevalent female cancer globally and this is also true in Iranian women. Alteration in circulating microRNAs affects the fate of immune cells, affecting immunological response to neoplasia.

Materials and Methods:

We investigated the expression of miR-490-5p and miR-490-3p in peripheral blood mononuclear cells (PBMCs) and plasma of patients with BC. Moreover, the correlation of these microRNAs with the expression levels of CD3d, interleukin 2 (IL-2), IL-2 receptor chain alpha (IL-2RA), forkhead box O1 (FOXO1) and nuclear factor of activated T cells 5 (NFAT5) were investigated.

Conclusion:

These results suggest that down-regulation of the target genes by miR-490 may predispose and facilitate the production of Th17 lymphocytes and IL-17-producing Tregs. The variation in miR-490-5p/-3p and the investigated targets in the PBMCs of BC patients may be used as non-invasive diagnostic markers.

Results:

Two groups, including 42 patients with BC, aged 22–75 years with stage I, II, III disease without administration of immunosuppressive chemotherapy regimens/radiotherapy and 40 healthy controls aged 27–70 years, participated. Overexpression and higher circulation levels of miR-490-5p and miR-490-3p were found in the patients with consequent down-regulation of all targets investigated in PBMCs. Furthermore, there was a significant negative correlation between the overexpression of these microRNAs and a reduction in levels of CD3d, IL-2, and IL-2RA in patients with BC.

Key Points

• Overexpression and higher circulation of miR-490-5p and miR-490-3p were found in patients with stages I-III breast cancer.

• Furthermore, the expression of the targets of these microRNAs, including FOXO1, CD3d, NFAT5, IL-2, and IL-2RA were decreased in PBMCs of patients with breast cancer.

• These findings suggest a shift in lymphocyte population towards the production of Th17, Tregs, and IL-17-producing Tregs.

Introduction

Breast cancer (BC) is the most prevalent cancer amongst women worldwide, including in the Iranian population. Epigenetic factors play a crucial role in the initiation and progression of BC (1). One of these epigenetic factors is characterized by the variability in microRNAs in both tumoral tissues and circulation. These changes in microRNAs may act directly or indirectly to increase cancer cell proliferation or modification of the tumor microenvironment toward favorable tumor requirements, and drug resistance (2). MicroRNAs (miRNAs) are short (18, 19, 20, 21, 22, #*#ref23 #*#nucleotide), non-coding RNAs that regulate various complementary mRNAs post-transcriptionally. Secretory components, especially exosomes and microvesicles, have an important role in circulating and shuttling these regulatory factors throughout the body (3, 4).

Several studies have shown the effects of onco-microRNAs on the production of immune suppressor cells, including regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), M2 type macrophages, etc. (5, 6, 7, 8). Tregs are a subpopulation of T-lymphocytes that have been shown to play a role in BC (9). Soheilifar et al. (6) showed that shuttling or concomitant overexpression of some of these onco-microRNAs, such as miR-182-5p and miR-182-3p, can target some proteins such as nuclear factor of activated T cell (NFAT) proteins, the T-cell receptor/complementarity determining region 3 (TCR/CD3) complex, and the interleukin 2/interleukin 2 receptor A (IL-2/IL-2RA) pathway to induce Tregs. The same study demonstrated that concomitant targeting FOXP3 inducer transcription factor (Forkhead box O1; FOXO1), NFATs that inhibited FOXP3 transcription factor, activation of interleukin-6 (IL-6) signaling, and inhibition of IL-2 signaling by miR-182-5p/-3p could induce an increase in the population of Tregs, including FOXP3⁺ IL-17-producing Tregs and FOXP3⁺ Tregs in BC patients (6).

The Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway has a pivotal impact on modulation of immune cells (10, 11). Activation of IL-2/IL-2RA induces STAT5 phosphorylation and activation (12). STAT5 is a transcription factor that induces FOXP3 expression, in addition to other inducer transcription factors, such as FOXO1 (13). In contrast, activation of NFAT proteins can activate T cells and induce their differentiation toward Th1 and Th2 subpopulations as well as suppress FOXP3 expression and Treg formation (14). Furthermore, initiation of IL-6 signaling causes STAT3 phosphorylation and inhibition of FOXP3 expression, as well (15). Activation of different members of the NFAT family, such as NFATc1 and NFATc2, recruit NFSATc4 that play a pivotal role in the induction of IL-2 by activation of T cells and activation of TCR/CD3 complex signal transduction (16, 17, 18). The expression or shuttling of miR-182/miR-183-96 cluster to immune cells in a BC microenvironment inhibits IL-2 and IL-2RA expression by targeting various TCR/CD3–associated signal transduction proteins (6). Moreover, miR-182-3p and miR-183 can negatively affect IL-2 production via targeting NFATc4 but also by directly targeting IL-2RA to prevent IL-2/IL-2RA signaling initiation (16).

However, NFAT5, the other member of NFAT family, induces IL-2 pathway and enhances the IL-17 inducer genes (19). The data derived from Gene Expression Omnibus (GEO) used to identify some microRNAs such as miR-490, etc. were upregulated in tumor tissues and over circulated in sera of patients with BC (6). Yang et al. (20) showed that miR-490-3p could directly target FOXO1. Also, Yang et al. (21) showed that targeting miR-490-5p inhibits the suppressive function of Tregs. However, some studies showed that miR-490-5p and miR-490-3p had potential in the production of Tregs and their polarization to IL-17-producing cells. Also, they potentially target CD3d, IL-2, IL-2RA, FOXO1, and NFAT5 (22). Considering the role of miR-490 in IL-17 producing T cells formation and their potential in targeting FOXO1, CD3d, IL-2, IL-2RA, and NFAT5, after identification of miR-490-5p and miR-490-3p, we aimed to investigate their expression in peripheral blood mononuclear cells (PBMCs) and the plasma of the patients with BC. Finally, we evaluated the correlation of miR-490-5p and miR-490-3p with the expression of CD3d, IL-2, IL-2RA, FOXO1, and NFAT5.

Materials and Methods

Results

Discussion and Conclusion

Previously, microarray-derived meta-analytical findings demonstrated an increased level of several microRNAs in both tumor tissue and plasma of patients with BC. Also, the immunosuppressive roles of some of these microRNAs have been described (6). Modulation of proteins involved in TCR/CD3 complex, IL-2/IL-2RA interactions and some transcription factors, such as NFATs, may direct T cells towards different Treg phenotypes (6). Furthermore, a concomitant decrease in FOXO1 level and reduction of NFATs has been associated with the production of IL-17-producing Treg (6). FOXO1, a transcription factor, induces FOXP3 expression as the main step in directing T cells toward T regs, stimulated by STAT5 activation through some cytokine-related signaling pathways, such as the IL2/IL-2RA pathway (12, 13). In contrast, NFATs suppress the expression of FOXP3 and induce Th1 and Th2 activation in normal conditions, together with IL-2 expression, which stabilizes their functions (14, 15).

Several microRNAs have been reported to promote T-cell phenotype change from Th1 and Th2 toward Tregs or FOXP3⁺IL-17-producing Tregs, including miR-21, miR-182-5p, miR-182-3p, miR-183, miR-10a and Th17 cells are known to be involved in the progression of BC (5, 6, 23). Attenuation of TCR/CD3 signal transduction and reduction of NFATs, IL-2, and IL-2RA proteins may be effective in Treg production by activation of STAT5 (24). It has been shown that IL-2 expression is induced by NFATc1-4 and NFAT5 in different conditions (16, 17, 18, 25). However, Soheilifar et al. (6) showed that reduction of NFATs in patients with BC is associated with increased circulation of microRNAs, such as miR-182 and miR-183. Moreover, the negative effects of miR-490 on the expression of IL-2 have been confirmed by targeting NFAT5 (26).

The results of the current study showed a dramatic elevation of both miR-490-5p and miR-490-3p expression in PBMCs of BC patients, but only the plasma level of miR-490-5p was concomitantly increased while plasma levels of miR-490-3p were the same as in healthy controls. Therefore, increased expression of miR-490-5p in PBMCs and a concurrent high level in plasma means that these cells have both endogenous and exogenous sources of the microRNA whereas, because of the normal circulating levels of miR-490-3p, the only source potential pathogenic source for PBMCs is endogenous. Also, it was observed that the CD3d gene, coding for CD3d protein which is one of the CD3 complex proteins and a target for miR-490-3p, was significantly downregulated in PBMCs, and this reduction was associated with upregulation of miR-490-5p and miR-490-3p. So, both isoforms of miR-490-5p and miR-490-3p may be able to partially suppress the TCR/CD3 signal transduction cascade by downregulating CD3d expression, and may play a role in inhibition of T cell activation. In contrast, a significant reduction was observed in IL-2RA expression level. Moreover, this decrease in IL-2RA expression was associated with overexpression of both miR-490-5p and -3p isoforms. It is worth noting that a decrease in IL-2RA attenuates STAT5 phosphorylation and activation (12, 27). Such a decrease has been associated with increased levels of circulating onco-microRNAs, such as miR-182-3p, in sera of patients with BC (6). Furthermore, a significant reduction was found in FOXO1 expression in the current study, which is targeted by miR-490-3p. This relationship, like other onco-microRNAs in BC, such as miR-182-3p, miR-183, has been confirmed in different studies (20, 28). However, simultaneous reduction of FOXO1 and NFAT gene products may predispose to phenotype switch to IL-17-producing Tregs (6). Also, it has been shown that deficiency in the production of FOXO1 has a key role in directing macrophages toward the M2-phenotype, which plays a critical role in the development of different kinds of cancers, especially BC (29).

It seems that miR-490 plays a role in decreased IL-2 expression in PBMCs of patients with BC. Some studies have shown direct targeting by miR-490 of NFAT5, which induces IL-2 expression and production (5, 26). However, a significant reduction in IL-2 expression level was associated with miR-490-3p overexpression. Also, other studies have shown that IL-2 was expressed irregularly in higher stages and in metastatic BC, and it is in line with our results in the case of IL-2 (30, 31). Thus, this study provides evidence that miR-490-5p and miR-490-3p may tip the balance between Treg and Th17 towards a Th17 T-cell phenotype through targeting IL-2RA as well as reduction of IL-2 by targeting NFAT5.

In conclusion, the results suggest potential for miR-490 to modulate the activity of FOXO1, NFAT5, CD3d, and IL-2RA. Over expression of mir-490-5p/-3p may facilitate the production of some phenotypes of T cells, which play a role in the progression of BC, including Th17, and IL-17-producing Tregs. A similar function has been suggested for other onco-microRNAs. Furthermore, the overexpression of both miR-490-5p and miR-490-3p and consequent suppression of their targets in PBMCs of BC patients may suggest a role as minimally invasive diagnostic markers in patients with BC.

Patient Selection

The participants were as follows. The patient group consisted of 42 patients with BC, aged 22–75 years who were patients in stages I, II, III of the disease without administration of immunosuppressive chemotherapy regimens/radiotherapy. The control group comprised 40 healthy individuals aged 27–70 years who were referred to Shohada Hospital. Written informed consent was obtained from the participants before the study. Demographic characteristics of all participants, including age, and marriage status, and pathologic data were gathered using a questionnaire from the pathology department. Exclusion criteria were advanced and metastatic cancer, neoadjuvant, a significant clinical disorder, psychiatric drug use for the past 5 months.

5 mL peripheral blood was collected from all participants in tubes containing EDTA and centrifuged at 150 g at 4 ^oC for 2 min. Then, we separated plasma, and then an equal volume of phosphate-buffered saline (PBS) was added to each blood sample and diluted gently. Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density centrifugation was used to isolate the PBMCs and the buffy coat, that contained lymphocytes, was collected after centrifuging at 800 g at 4 ^oC for 15 min. This study was approved by the Ethics Committee of the Cancer Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran (ethics code: IR.SBMU.RETECH.REC.1397.562).

RNA Extraction and qRT‐PCR

Total RNA and circulating RNA were extracted from extracted PBMCs and 1 mL of plasma using the RiboEx LS reagent (Geneall, South Korea). Then, cDNA was synthesized for evaluation of the CD3d, IL-2, IL-2RA, FOXO1, and NFAT5 using a first-strand cDNA synthesis kit (Thermo Fisher Scientific) followed by PCR according to the manufacturer’s protocol. To assess microRNAs’ levels (RNA-derived from plasma), specific hairpin loop primers were used to synthesize cDNA of the microRNAs of interest. The expression and variation of microRNAs (miR-490-5p and miR-490-3p) and their targets were evaluated by SYBR Green master mix kit (Genaxxon kit, Germany) on a MIC qPCR instrument (BioMolecular Systems, Australia). The specific primers are listed in Table 1. Eventually, qRT‐PCR-derived data were analyzed by the 2^-ΔΔCT and 2^-ΔCT methods. Beta actin and GAPDH were utilized as housekeeping genes for comparison of the expression of target genes and RNU6 was used as the housekeeping gene for comparison of microRNAs.

Statistical Analysis

R-Studio 1.0.136 software was used to generate the correlation heatmap between miR-490-5p and miR-490-3p with their potential targets. In the current study, p<0.05 was considered statistically significant. Finally, multivariate analyses were performed to show the relationship between microRNAs and the expression level of their targets in the PBMCs of patients with BC. Comparison was made using the demographic and clinical characteristics of patients and controls. Statistical comparison was performed using SPSS, version 18 (IBM Inc., Armonk, NY, USA). Relative changes of microRNAs and their target genes in PBMCs of the patients with BC were assessed using student’s t-test. Also, receiver operator characteristic (ROC) curve analysis was performed for miR-490-5p and -3p besides their targets in PBMCs samples using SPSS, version 18 (IBM Inc., Armonk, NY, USA).

Pathologic Results

Pathology examinations showed that 23.8, 52.38, and 23.8 percent of the patients, respectively, were related to stages of I, II, and III. Almost 73.8% of patients expressed estrogen receptor (ER), and 64.2% were progesterone receptor (PR)-positive. Also, 21.42% of patients were human epidermal growth factor receptor 2 (HER-2)-positive. P53 mutation has been observed in 11.9 percent of these patients. Comparison between BC patients and control groups showed that patient groups had more abortions (nine abortions) even though the control group had more pregnancies (44 pregnancies) and longer time of breastfeeding (Table 2). However, there is not too much difference between other criteria between the two groups. Moreover, 36 of 40 people (90%) in the control group were married, and 36 of 42 people (86.7%) of patients were married. Of 42 patients, 35 were non-menopause (83.3%), while 87.5% of the control group were non- menopause (35 of 40 people). Moreover, there are no significant differences between these two groups in terms of menstruation (year). The pathological and clinical characteristics of patients with BC are summarized in Table 2.

Circulating miR-490-5p and miR-490-3p Were Increased in Plasma From Patients with BC

There was little published data on the expression of miR-490-3p in the plasma of the patients with BC. Thus miR-490-5p and miR-490-3p were evaluated separately. Expression analysis derived from real-time PCR showed that only miR-490-5p was significantly raised in the plasma of the patients with BC, whereas miR-490-3p in BC patients did not differ from healthy controls (Figure 1a). It was observed that miR-490-5p was increased 4.95-fold in patients with BC compared to controls (p<0.01).

MiR-490-5p and miR-490-3p Were Increased in PBMCs of Patients with BC

Both miR-490-5p and miR-490-3p were upregulated in PBMCs from patients with BC, by 15.6 times (p<0.001) and 13.14 times (p<0.001), respectively, compared to controls (Figure 1b).

Expression of Target Genes for miR-490-5p and miR-490-3p Were Decreased in PBMCs of Patients with BC

The expression levels of immune modulatory genes identified in the meta-analysis were compared between patients with BC and controls. This analysis showed significant down-regulation of the following genes: FOXO1 2.21 times (p<0.01), CD3d 3.9 times (p<0.01), NFAT5 4.8 times (p<0.01), IL-2 3.3 times (p<0.05), and IL-2RA 4.35 times (p<0.01) (Figure 1c).  Correlation analysis was performed to investigate the relationship between expression changes in miR-490-5p and miR-490-3p and their respective target genes in patients and controls. There was a negative correlation between miR-490-5p and CD3d (r = -0.658, p = 0.001) and IL-2RA (r = -0.670, p<0.001), whereas there was a strong correlation between miR-490-5p and miR-490-3p in PBMCs of BC patients (r = 0.823, p<0.001). There was also a negative correlation between miR-490-3p and CD3d (r = -0.698, p<0.001), IL-2 (r = -0.462, p = 0.03), and IL-2RA (r = -0.725, p<0.001). Moreover, there were a significant association between reduced expression of CD3d and FOXO1 (r = 0.41, p = 0.05) and CD3d and IL-2RA (r = 0.505, p = 0.014). A significant relationship was found between FOXO1 suppression and NFAT5 (r = 0.495, p = 0.016). Finally, the decrease in IL-2 expression was correlated with the decrease in IL-2RA (r = 0.601, p = 0.002) (Figure 2a).

ROC curve analysis was used to investigate the sensitivity and specificity of miR-490-5p, miR-490-3p, FOXO1, CD3d, NFAT5, IL-2, and IL-2RA expression levels in PBMCs of patients with BC compared to controls. The area under the curve (AUC) values for discrimination of BC patients from healthy individuals were 0.840 (p<0.001) for miR-490-5p and 0.747 (p = 0.009) for miR-490-3p, respectively (Figure 2b). In the case of miR-490-5p and -3p, 73.5% and 69.7% of the positive outcomes would be correctly identified by diagnostic tests as positive. Also, 22.6% and 36.4% of the negative would be incorrectly specified by diagnosis test as positive for miR-490-5p and -3p, respectively. Moreover, 71.4, 68.8, 63.5, 62.5 and 61.5 percent of the positive outcomes would be correctly identified by diagnostic tests as positive for IL-2, IL-2RA, CD3d, NFAT5 and FOXO1, respectively. Also, 33.3, 34.2, 28.6, 45 and 45.7 percent of the negative would be incorrectly specified by diagnosis test as positive for IL-2, IL-2RA, CD3d, NFAT5, and FOXO1, respectively. Similarly, the AUCs for BC patients compared to controls for expression levels of CD3d, IL-2, and IL-2RA were 0.680 (p = 0.014), 0.739 (p = 0.009), and 0.732 (p = 0.012), respectively. No significant difference was observed for NFAT5 (p = 0.088) and FOXO1 (p = 0.076) expression (Figure 2c). When the expression levels of miR490-5p and miR-490-3p and their target genes were examined in relation to clinical characteristics of the patients, no significant relationship was found.