Introduction

Lung cancer is the leading reason of cancer deaths in the world (Laskin and Sandler 2005). In recent years, some therapeutic drugs have been developed to greatly improve both survival times and quality of life in patients with lung cancer, such as cisplatinum-based combination chemotherapy and targeted drugs. However, lung cancer exerts a low, 5-year survival rate of less than 15 % after initial diagnosis. Therefore, innovative and reliable diagnostic or prognostic biomarkers and more efficient therapeutic methods are urgently needed to treat lung cancer.

MicroRNAs (miRNAs) are short non-coding RNAs (18–22 nt) and can inhibit the expression of genes. Mature miRNAs are produced by the RNase III enzymes Drosha and Dicer, then incorporated into the RNA-induced silencing complex (RISC), and finally binds to the 3′-untranslated region (3′-UTR) of their target gene mRNA and inhibits their expression (Bartel 2009; Filipowicz et al. 2008). miRNAs have been demonstrated to have various physiological and pathological functions, and accumulating evidences have demonstrated that there is a robust association between miRNAs and carcinogenesis (Kim et al. 2009; Carthew and Sontheimer 2009). Previous literatures have reported that miRNAs begin to emerge as novel biomarkers for various cancers. miR-22 injection has been reported to be an efficient therapy to suppress tumor growth and metastasis (Xu et al. 2011). Many proteins are the targets of miR-22, such as estrogen receptor a (ERa), c-Myc-binding protein (MYCBP), Myc-associated factor X (MAX), and PTEN.

The ErbB family of 4 closely related type 1 transmembrane tyrosine kinase receptors contains the epidermal growth factor receptor (EGFR/ErbB1), related family members ErbB2, ErbB3, and ErbB4. ErbB3 has no kinase activity, but can heterodimerize with and is phosphorylated by other ErbB receptors. ErbB3 is a key signal molecule for cancer progression and seems to be an important target for cancer therapy. For example, ErbB3 has been demonstrated to be involved in the proliferation of ErbB2-driven breast cancer cells (Holbro et al. 2003). Inhibition of ErbB3 signaling plays an essential role in gefitinib-induced apoptosis in gefitinib-sensitive lung cancer cells (Engelman et al. 2005).

The aim of the present study is to investigate the therapeutic effect of miR-22 on the lung cancer. Furthermore, detailed molecular mechanism was discussed, and the effect of miR-22 on ErbB3 was focused on. These results will be helpful for deeper understanding of anti-cancer mechanism of miR-22.

Materials and methods

Cell lines and human tissues

A549 and H1299, two human non-small cell lung cancer lines, were obtained from the Cell Bank of Type Culture Collection of the Chinese Academy of Science (Shanghai, China). All of them were grown in RPMI 1640 containing 10 % fetal bovine serum. Human lung cancer specimens (n = 11) and paired non-cancerous normal lung specimens (n = 11) were obtained from the patients at Daping Hospital Research Institute of Surgery Third Military Medical University with documented informed consent in each case. Patients undergoing surgery for lung cancer provided the written consent to donate tissue for analysis.

Quantitative real-time PCR

For quantitative expression analysis of miRNAs, total RNA was reverse-transcribed by multiscribe RT and miRNA-specific miRNA primers (ABI), and quantitative real-time PCR (qRT-PCR) was performed by using a Tapman microRNA assay kit (ABI). The comparative cycle threshold (Ct) method was applied to quantify the expression levels through calculating the 2−∆∆Ct method. U48 small nuclear RNA was used as an internal standard.

Cell counting kit-8 (CCK8) assay

A549 and H1299 cells were trypsinized and seeded at 3 × 103 cells/well in 96-well plates. After 24 h, the 20 μl of CCK8 (Sigma, USA) solution (5 g/l) in phosphate-buffered saline (PBS) was added. Plates were incubated for another 3 h. The optical density for each well was measured using a microculture plate reader (BioTek, USA) at a wavelength of 490 nm.

Annexin V/7-AAD analysis

Flow cytometric analysis was performed to identify and quantify the apoptotic cells by using Annexin V/7-AAD apoptosis detection kit (BD Bioscience) according to the manufacturer’s instruction.

In vitro invasion assay

The invasion ability of cells was measured in transwell chambers coated with 5 μg of Matrigel (8 μm pore; BD Biosciences, Franklin Lakes, NJ). The bottom chamber was filled with DMEM containing 10 % FBS. For the invasion assay, tumor cells (1 × 105 cells in a total volume of 100 μl) were placed on the chamber and incubated at 37 °C in 5 % CO2 humidified air. After 16 h, non-invading cells on the upper surface of the membrane were removed, and the cells that invaded to the underside of the polycarbonate membrane were fixed with ethanol and stained by crystal violet for 5 min. The number of invasive cells was then determined for five independent fields under a microscope. The mean of triplicate assays for each experimental condition was used for analysis.

Transfection

miRNA-22 expression plasmid was constructed using expression plasmid pcDNA3.1. miRNA-22 was amplified using the following primers: 5′- CGCGGATCCTCAGCGAGGTTAACAGCTTC-3′ (forward primer), 5′- CCGGAATTCTGCTGTTCGGACCTTCCCAG-3′ (reverse primer). Luciferase reporter plasmid with ERBB3 3′UTR and mutated 3′UTR was also constructed using pmirGLO. The primers for ERBB3 3′UTR were as followed: 5′-AAAC CTCCTGCTCCCTGTGGCACTCAGGGAGCATTTAATGGCAGCTAGTGCCTTTAGAGT-3′ (forward primer), 5′-CTAGA CTCTAAAGGCACTAGCTGCCATTAAATGCTCCCTGAGTGCCACAGGGAGCAGGAG GTTT-3′ (reverse primer); two kinds of ErbB3 stabling expression plasmids (one contains coding domains (CDs) and the other contains CDs and 3′-UTR of ErbB3) were constructed using pEGFP-N1 expression plasmid. The primers were as followed: for the former, the primers are 5′- CTAGCTAGCGTCATGAGGGCGAACGACGC-3′ (forward primer) and 5′-CGGGGTACCTTACGTTCTCTGGGCATTAGC-3′ (reverse primer); for the latter, the primers are 5′-CTAGCTAGCGTCATGAGGGCGAACGACGC-3′ (forward primer) and 5′-CGGGGTACCCTCTAAAGGCACTAGCTG-3′ (reverse primer).

Luciferase assays

Cells were analyzed for luciferase activity using the Dual-Glo® Luciferase Assay System (Cat.# E2920) and a MicroLumatPlus LB96V luminometer (Berthold).

Western blotting

Cells were lysed in 0.4 ml of lysis buffer (50 mM Tris pH 8.0, 150 mM NaCl, 10 mM EDTA, 1 % NP40, 20 mM NaF, 1 mM orthovanadate and protease inhibitor cocktail). Lysates were separated by electrophoresis, blotted to membrane, and reacted with specific antibodies. The primary antibodies and appropriate secondary antibodies were from Santa Cruz.

In vivo proliferation assay

BABL/c nude mice (5 weeks old) were purchased from the Animal Center of Chinese Academy of Science (Shanghai, China). All animal studies were conducted in accordance with National Institutes of Health animal use guidelines and the current Chinese regulations and standards on the use of laboratory animals. A total of 1 × 107 A549 cells stably expressing miR-22 were injected subcutaneously into nude mice. 60 days after injection, the mice were killed. The tumor volume was evaluated using the following formula: Tumor formula = 4π/3 × (width/2)2 × (length/2). The tumors were weighted.

Statistical analysis

Data were expressed as the mean ± SD. Statistical comparisons were made between two groups with the t test. A value of p < 0.05 was considered significant.

Results

Reduced miR-22 expression in lung cancer tissues and cell lines

To compare the miR-22 expression in normal lung tissues and lung cancer tissues, Q-PCR methods were employed to quantify the miR-22 level 11 pairs of matched human lung cancer tissues. As shown in Fig. 1, the expression of miR-22 is much lower in lung cancer specimens than that in normal specimens. Additionally, the expression of miR-22 in A549 and H1299 cells was also shown to be lower than normal tissues.

Fig. 1
figure 1

Reduced miR-22 expression in lung cancer tissues and cell lines. The expression levels of miR-22 were normalized against the expression level of U48

Full size image

Over-expression of miR-22 inhibits the increased cell numbers and invasion of lung cancer cell lines

Because miR-22 was down-regulated in lung cancer tissues and cell lines, we asked whether miR-22 played a tumor-suppressive role in lung cancer development. Therefore, the impact of miR-22 on cellular numbers was evaluated. Compared with untreated group and pcDNA3.1 control group, transfection of miR-22 expression plasmid significantly increased the expression level of miR-22 for A549 (p < 0.05, Fig. 2a) and H1229 (p < 0.05, Fig. 2d). At 48 h after miR-22 transfection, the over-expression of miR-22 could significantly inhibit the cellular numbers by 70 % (Fig. 2b) for A549 and 60 % for H1229 (Fig. 2e), respectively. The apoptosis is an important cause for the reduction of cell number (Fig. 1S). Additionally, the invasion capacity of A549 and H1229 was also inhibited by over-expression of miR-22 (Fig. 2c, f).

Fig. 2
figure 2

Over-expression of miR-22 significantly inhibits the proliferation and invasion of A549 and H1299. a Transfection of miR-22 expression plasmid to A549 increases the expression of miR-22. b The inhibitory effect of miR-22 toward the proliferation of A549. c The inhibitory effect of miR-22 toward the invasion of A549. d Transfection of miR-22 expression plasmid to H1299 increases the expression of miR-22. e The inhibitory effect of miR-22 toward the proliferation of H1299. f The inhibitory effect of miR-22 toward the invasion of H1299. *p < 0.05, compared with untreated group; # p < 0.05, compared with pcDNA3.1 control group

Full size image

ErbB3 is a direct target of miR-22

Transfection of pre-miR-22 into A549 significantly decreased the expression of ErbB3 (Fig. 3a). Human ErbB3 3′-UTR has a potential binding site for miR-22 (Fig. 3c). To investigate the mechanism of which miR-22 regulates ErbB3, we constructed a luciferase reporter vector containing a fragment of the 3′-UTR or mutated 3′-UTR with a miR-22 binding site. A549 cells were co-transfected with this reporter vector and pre-miR-22 or control. The results showed that the over-expression of miR-22 could decrease the luciferase activity. However, miR-22 does not influence the expression of luciferase with a mutated miR-22 binding elements (Fig. 3b). All these results indicated that miR-22 exert inhibitory effects on ErbB3 expression via interaction with the 3′-UTR of ErbB3. Additionally, over-expression of ErbB3 in A549 and H1299 cells significantly prevents the inhibitory effects of miR-22 on proliferation and invasion (Fig. 4).

Fig. 3
figure 3

miR-22 inhibits the expression of ErbB3 by binding to 3′-UTR. a The expression of ErbB3 significantly decreased after transfection of miR-22 in A549 cells. *p < 0.05, compared with untreated group; # p < 0.05, compared with pcDNA3.1 control group. b The relative luciferase activity in A549 cells was determined after the ErbB3 3′-UTR or mutant plasmids were co-transfected with miR-22. c The structure of wild-type (WT) ErbB3 3′-UTR and mutated (MUT) 3′-UTR

Full size image
Fig. 4
figure 4

The over-expression of ErbB3 in A549 and H1299 cells protects the cells from the inhibition of proliferation and invasion induced by over-expression of miR-22. a Transfection of ErbB3 expression plasmid increases the expression of ErbB3 in A549 cells. b Over-expression of ErbB3 in A549 cells significantly prevents the proliferation inhibition effect of miR-22. c Over-expression of ErbB3 in A549 cells significantly prevents the invasion inhibition effect of miR-22. d Transfection of ErbB3 expression plasmid increases the expression of ErbB3 in H1299 cells. e Over-expression of ErbB3 in H1299 cells significantly prevents the proliferation inhibition effect of miR-22. f Over-expression of ErbB3 in H1299 cells significantly prevents the invasion inhibition effect of miR-22

Full size image

miR-22 inhibits the growth of lung cancer in vivo

To further investigate the contribution of miR-22 in vivo, we established a miR-22 expression stable cell line, which was named A549-miR-22. This cell line was injected subcutaneously, and tumor progression was studied over time. At 60 days post-implantation, the mice were killed, and the tumors were removed. As shown in Fig. 5, the volume and weight of tumors obtained from injection of A549-miR-22 cells were significantly less than that in untreated group and pcDNA3.1 control group.

Fig. 5
figure 5

In vivo anti-lung caner effect of miR-22. a Representative figures of tumors in untreated, pcDNA3.1 control, and miR-22-treated groups. b Determination of tumor volumes at different time points at different groups. c miR-22 significantly decreased the weight of tumors. *p < 0.05, compared with untreated group; # p < 0.05, compared with pcDNA3.1 control group

Full size image

Discussion

Lung cancer remains to be the leading cancer type and the leading cause of cancer deaths in males globally (Jemal et al. 2011). miRNAs have been demonstrated to have close relationship with lung cancer. miR-22, originally identified in HeLa cells, has been found to be over-expressed in prostate cancer, but down-regulated in breast cancer, cholangiocarcinoma, multiple myeloma, and hepatocellular carcinoma (Zhang et al. 2010). In the present study, the miR-22 levels were compared between normal tissues and lung cancer tissues, and the results showed that the level of miR-22 was lower in the lung cancer tissues.

Furthermore, the influence of miR-22 on proliferation and invasion of lung cancer cells was investigated. Two representative kinds of non-small cell lung carcinoma cell lines (A549 and H1299) were employed in this study. The results showed that the over-expression of miR-22 could significantly prevent the proliferation and invasion, which was consistent with previous report that some other miRNAs could inhibit the progression of cancer cells, such as miR-342 (Wang et al. 2011).

Individual miRNA can modulate the expression of many genes. The molecular targets of miR-22 have been drawing much attention of researchers. To date, a variety of targets of miR-22 have been reported in previous literatures. In breast cancer cell lines, miR-22 has been demonstrated to target estrogen receptor a (ERa) and repress estrogen signaling (Pandey and Picard 2009; Xiong et al. 2010a). Xiong et al. (2010b) found that miR-22 could act as a tumor repressor through direct inhibition of c-Myc-binding protein MYCBP expression and subsequent reduction of oncogenic c-Myc activities. MiR-22 has also been reported to be a novel regulatory molecule in the PTEN/AKT pathway (Bar and Dikstein 2010). HIF-1a was a target of miR-22 in colon cancer (Yamakuchi and Yagi 2011). In the present study, ErbB3 was demonstrated to a new target of miR-22. ErbB3 has not drawn much attention in the past years. However, many recent studies showed that this receptor is very important in cancer. ErbB3 has been regarded to cause cell proliferation of cancer cells (Sheng et al. 2010), and some ErbB3-targeting drugs are being developed in their clinical phases, such as the ErbB3 antibody MM-121 (Schoeberl et al. 2010). Therefore, miR-22 will be a potential therapeutic drug to target ErbB3 for treatment of lung cancer. This therapeutic effect has been demonstrated in vivo in the present study.

In conclusion, our results reveal that miR-22 exhibits inhibitory effects toward lung cancer through down-regulating the expression of ErbB3. This newly identified miR-22/ErbB3 link provides a new, potential therapeutic target to treat lung cancer.