PAX6, a novel target of miR-335, inhibits cell proliferation and invasion in glioma cells

Following the publication of this paper, it was drawn to the Editors' attention by a concerned reader that the western blotting data shown in Fig. 1B were strikingly similar to data appearing in different form in other articles by different authors. Owing to the fact that the contentious data in the above article had already been published elsewhere, or were already under consideration for publication, prior to its submission to Molecular Medicine Reports, the Editor has decided that this paper should be retracted from the Journal. After having been in contact with the authors, they agreed with the decision to retract the paper. The Editor apologizes to the readership for any inconvenience caused. [the original article was published in Molecular Medicine Reports 10: 399‑404, 2014; DOI: 10.3892/mmr.2014.2150].


Introduction
Paired box 6 (PAX6) is a highly conserved transcription regulatory factor involved in multiple biological processes, including embryogenesis, cell differentiation, biosynthesis and cell motility (1,2). Accumulating evidence has demonstrated that PAX6 is crucial in the development of the central nervous system (3). Previously, several studies have suggested that PAX6 acts as a tumor suppressor in malignant glioma (4,5). As one of the most common types of primary brain cancer, the prognosis of malignant glioma is very poor, mainly due to its resistance to radiotherapy, chemotherapy or adjuvant therapies (6)(7)(8). The median survival rate of glioma has not improved over the last decade, despite the development of therapeutic strategies for various types of human cancer (9). Since PAX6 has been suggested to act as a suppressive regulator in malignant glioma, it may be a promising therapeutic target.
MicroRNAs (miRNAs) are a type of endogenous non-coding RNA, which are able to bind to the 3'-untranslated region (3'-UTR) of its target mRNAs, eventually causing mRNA degradation or translational repression (10). In previous decades, accumulating studies have demonstrated that miRNAs act as key regulators in the development and progression of various types of cancer, including malignant glioma (11). For instance, miR-155 was reported to regulate cell proliferation and invasion in glioma (12). miR-128 was demonstrated to promote cell-cell adhesion in glioma U87 cells (13). Therefore, miRNAs are potential therapeutic targets or candidates for the treatment of glioma.
miR-335 has been demonstrated to be implicated in multiple types of malignant tumor, including small cell lung cancer, osteosarcoma, ovarian cancer, prostate cancer, hepatocellular carcinoma, meningioma and gastric cancer (14)(15)(16)(17)(18)(19)(20)(21)(22). Recently, Jiang et al have suggested that miR-335 upregulation is associated with advanced tumor progression in glioma and is an independent marker for the predication of the clinical outcome of patients with glioma (23). However, the exact regulatory roles of miR-335 and PAX6, as well as their association with glioma, have never been studied.
The present study investigated the expression of PAX6 and miR-335 in glioma tissues, as well as their effects on glioma cell proliferation, colony formation and invasion in vitro. Furthermore, the present study also investigated the involved molecular mechanisms and unveiled their target association.

Materials and methods
Tissues and cell lines. All protocols in the present study were approved by the Ethical Committee of Central South University (Changsha, Hunan, China). In addition, all informed consent forms for patients participating in the present study were obtained. In total, 24 glioma tissues (six cases of WHO I, six cases of WHO II, six cases of WHO III and six cases of WHO IV), as well as six cases of normal brain tissues were obtained from the Department of Neurosurgery, Xiangya Hospital of Central South University (Changsha, Hunan, China). Human glioma U251 and U87 cell lines were obtained from the Cell Bank of Central South University (Changsha, Hunan, China). All cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 100 IU/ml of penicillin and 100 µg/ml of streptomycin sulfate at 37˚C in a humidified incubator containing 5% CO 2 .
RNA extraction and real-time RT-PCR analysis. According to the manufacturer's instructions, total RNA was extracted with TRIzol (Invitrogen Life Technologies, Carlsbad, CA, USA). For the analysis of miR-335 expression, 10 ng of RNA was converted to cDNA using miR-335-specific primers and an ABI miRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA). Following that, real-time PCR was performed on an ABI 7500 thermocycler (Applied Biosystems). The U6 gene was used as an internal reference. For the analysis of PAX6 expression, a TaqMan Reverse Transcription kit and Power SYBR-Green kit (Thermo Fisher Scientific, Waltham, MA, USA) were used to perform real-time RT-PCR, and β-actin was used as an internal reference. The relative expression was analyzed by the 2 -ΔΔCt method. The primers used were as follows: PAX6, forward 5'-AACGATAACATACCAAGCGTGT-3' and reverse 5'-GGTCTGCCCGTTCAACATC-3'; β-actin, forward 5'-AGGGGCCGGACTCGTCATACT-3' and reverse 5'-GGCGGCACCACCATGTACCCT-3'.
Western blot analysis. Tissues or cells were solubilized in cold RIPA lysis buffer. Proteins were separated with 10% SDS-PAGE and then transferred onto a polyvinylidene difluoride (PVDF) membrane, which was then incubated with TBST containing 5% skimmed milk at room temperature for 4 h, and then with the primary antibodies of anti-PAX6, anti-MMP-2, anti-MMP-9 and anti-β-actin (Abcam, Cambridge, UK), respectively, at room temperature for 3 h. Following being washed by PBST for 15 min, the PVDF membrane was incubated for 40 min with the corresponding secondary antibodies. An ECL kit (Pierce Biotechnology, Inc., Rockford, IL, USA) was used to perform chemiluminescent detection. Image-Pro plus software 6.0 was used to analyze the relative protein expression, presented as the density ratio versus β-actin. Dual luciferase reporter assays. A QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA) was used to generate a wild type and a mutant type 3'-UTR of PAX6, which was then inserted into the multiple cloning sites in the psiCHECK™2 vector (Promega, Madison, WI, USA), respectively. U87 and U251 cells were cultured to ~50-60% confluence in a 24-well plate and then co-transfected with psiCHECK™2-PAX6 -3'-UTR or psiCHECK™2-mut PAX6-3'-UTR vector, with or without 50 nM of miR-335, using cellfectin II reagent (Invitrogen Life Technologies). The luciferase activity for each group was determined 48 h after cotransfection using the dual-luciferase reporter assay system (Promega) and a Beckman Coulter LD 400 luminometer (Beckman Coulter, Fullerton, CA, USA). Renilla luciferase activity was normalized to firefly luciferase activity.
Cell proliferation assay. An MTT assay was used to measure cell proliferation. At 48 h post-transfection, the transfection medium in each well was replaced with 100 µl of fresh serum-free medium with 0.5 g/l of MTT. Following incubation at 37˚C for 4 h, the MTT medium was removed by aspiration and 50 µl of DMSO was added to each well. Following incubation at 37˚C for a further 10 min, the optical density at 570 nm was measured using a Bio-Tek ELx800 type ELISA reader (Bio-Tek Instruments, Inc., Winooski, VT, USA). This experiment was repeated three times.
Invasion assay. The invasive assay was performed in 24-well transwell chambers (Chemicon, Temecula, CA, USA) with a layer of matrigel. According to the manufacturer's instructions, 200 µl of cell suspension (1x10 6 cells/ml) was added in triplicate wells for each group, which was then incubated in a humidified atmosphere of 5% CO 2 at 37˚C for 24 h. The cells that had migrated through the filter were stained with gentian violet. Following being washed with PBS for 15 min, five fields were randomly selected under the microscope and the cell number was counted.
Statistical analysis. Statistical analysis was performed using SPSS 17.0 software. The results are expressed as the mean ± SD of three independent experiments. For each experiment, the statistical analysis was repeated three times. Statistical analysis of differences was performed by one-way analysis of variance (ANOVA) or Student's t-test. * P<0.05 and ** P<0.01 was considered to indicate a statistically significant difference.

Results
Expression of PAX6 and miR-335 in glioma tissues. The expression of PAX6 was examined in human glioma tissues and normal brain tissues by real-time RT-PCR. As shown in Fig. 1A, the mRNA expression level of PAX6 in glioma tissues was reduced compared with normal brain tissues. Furthermore, the glioma tissues with different grades demonstrated different expression levels of PAX6, the expression of which was the lowest in grade WHO IV, and highest in grade WHO I, which was further confirmed by the data from western blotting. The expression of miR-335 in human glioma tissues and normal brain tissues was then examined. As shown in Fig. 1B, the level of miR-335 in glioma tissues gradually increased, when compared with those in normal brain tissues and its expression was positively associated with the malignancy of gliomas.
PAX6 is a novel target of miR-335. As shown in Fig. 2A, bioinformatical analysis results demonstrated that the putative binding site for miR-335 at the 3'UTR of PAX6 is conserved and the mutant type 3'-UTR of PAX6, which was constructed, was also shown. The results of the luciferase reporter assay

R E T R A C T E D
demonstrated that the renilla/firefly value of luciferase activity was markedly decreased only following co-transfection with the wild type 3'UTR of PAX6 and miR-335 in U87 and U251 cells (P<0.01), whereas the renilla/firefly value of luciferase activity in other groups demonstrated no difference (Fig. 2B). Thus, the present study suggests that PAX6 is a direct target of miR-335 in glioma U87 and U251 cells.
Effect of miR-335 on the protein expression of PAX6. The effect of miR-335 on the protein expression of PAX6 in U87 and U251 cells was further investigated. Following transfection of U87 and U251 cells with an miR-335 mimic or inhibitor, respectively, the miR-335 expression level was examined using real-time RT-PCR. As shown in Fig. 3A, the expression level of miR-335 was significantly upregulated following transfection with the miR-335 mimic, however, it was downregulated following transfection with the miR-335 inhibitor, when compared with that in the control group (P<0.01). Thus, the transfection efficiency was satisfactory. The protein expression of PAX6 was then determined using western blot analysis and the PAX6 protein level was revealed to be decreased in miR-335-overexpressed cells, however, it was increased in miR-335-downregulated cells (Fig. 3B). According to these data, the present study suggests that miR-335 negatively regulates the expression of PAX6 at a post-transcriptional level in glioma U87 and U251 cells.
Effects of PAX6 and miR-335 on the regulation of glioma cell proliferation. In order to investigate the functions of PAX6 and miR-335 in glioma cells, U251 cells were transfected with PAX6-pCMV-NEO-BAM, the miR-335 mimic or the two combined, respectively. As shown in Fig. 4A, the protein expression of PAX6 was markedly upregulated in U251 cells transfected with PAX6-pCMVp-NEO-BAM. Furthermore, the miR-335 mimic inhibited the PAX6 protein expression, however, in the co-transfection group, the PAX6 protein level was restored. Following that, the cell proliferation rate was markedly reduced in PAX6-overexpressed glioma cells compared with control groups, respectively. However, in miR-335-overexpressed U251 cells, the cell proliferation was significantly upregulated, which was effectively reversed by co-transfection with PAX6-pCMV-NEO-BAM and the miR-335 mimic. Taken, together, these results indicate that PAX6 has an inhibitory effect on U251 and U87 cell proliferation, while miR-335 is able to promote glioma cell proliferation.

Effects of PAX6 and miR-335 on the regulation of colony formation in glioma cells.
The effects of PAX6 and miR-335 overexpression on colony formation ability in U251 cells were further studied. As shown in Fig. 5, the PAX6 overexpressed U251 cells demonstrated lower colony formation efficiency compared with the control group. However, miR-335 upregulation promoted colony formation in U251 cells, which was reversed by co-transfection with PAX6 and the miR-335 mimic. Accordingly, our findings suggest that PAX6, negatively regulated by miR-335, inhibits the malignant characteristics of glioma cells.   Fig. 6A, overexpression of PAX6 significantly inhibited U251 cell invasion (P<0.05). However, miR-335 overexpression notably promoted cell invasion (P<0.05), which was effectively reversed by PAX6 overexpression. These data indicate that PAX6 has an inhibitory effect on glioma cell invasion, while miR-335 is able to promote it.
MMP-2 and MMP-9 are crucial in the regulation of cancer cell invasion. Thus, the alterations of MMP-2 and MMP-9 expression were further investigated, following transfection of U251 cells with PAX6-pCMV-NEO-BAM, the miR-335 mimic or the two combined, respectively. As shown in Fig. 6B, the protein expression of MMP-2 and MMP-9 were decreased

R E T R A C T E D
in PAX6-overexpressed U251 cells, when compared with that in the control group (P<0.01). However, their protein levels were upregulated following transfection with the miR-335 mimic (P<0.01), which was reversed by the overexpression of PAX6.
Taken together, our findings suggest that PAX6, negatively regulated by miR-335, is able to inhibit MMP-2 and MMP-9 expression, which is possibly responsible for the suppressive effect of PAX6 on glioma cell invasion.

Discussion
As one of the most common types of primary brain cancer, the tumorigenesis of glioma is regulated by a complex cascade of molecular events monitoring cell proliferation and motility (8). Thus, understanding the molecular regulatory mechanisms has vital significance. The present study demonstrated that the expression of PAX6 was reduced parallel to the upregulation of miR-335. More importantly, the present study for the first time, to the best of our knowledge, identified PAX6 as a direct target of miR-335. Functional analysis revealed that PAX6, negatively regulated by miR-335, effectively inhibited cell proliferation, colony formation and invasion in glioma cells.
As a key transcription factor, PAX6 has been demonstrated to be involved in the development of the eye, central nervous system and pancreas (3,24). Previously, PAX6 was revealed to be involved in the regulation of malignant glioma (4,25). Zhou et al demonstrated a correlation between a low expression level of PAX6 and unfavorable patient outcomes in patients with malignant astrocytic gliomas (5). The present study also demonstrated that the expression of PAX6 was downregulated in glioma tissues compared with normal brain tissues. The authors further demonstrated that PAX6 acted as a tumor suppressor in glioblastoma cells and the overexpression of PAX6 was able to effectively inhibit the growth of glioblastoma cells in vitro and in vivo (26), which is partially consistent with our findings. Despite the suppressive effect of PAX6 on glioma being revealed, the regulatory mechanisms remain largely unknown.
Based on bioinformatical predication, a luciferase reporter assay was performed and PAX6 was identified as a novel target of miR-335, which has been demonstrated to be implicated in various types of cancer (14)(15)(16)(17)(18)(19)(20)(21)(22). However, the detailed role of miR-335 in the development and progression of various types of cancer remains controversial. For instance, it has been suggested that miR-335 may function as a tumor suppressor in osteosarcoma (14), however, it may act as an oncogene in meningiomas (21). These findings suggest that the exact regulatory function of miR-335 in tumorigenesis is tissue specific, since different microenvironments in various types of cancer may have different effects on the expression and activity of miR-335.
Previously, miR-335 was revealed to be involved in several types of malignant tumor in the central nervous system (23,27,28). Shu et al reported that miR-335 acted as a promoter in growth and invasion in malignant astrocytoma, partially at least through directly targeting Daam1 (27). In addition, the authors further demonstrated that miR-335 was required for the differentiation of malignant glioma cells induced by the activation of the cAMP/protein kinase A pathway (28). Jiang et al demonstrated that the upregulated expression of miR-335 was associated with advanced tumor progression as well as a poorer survival time in patients with malignant glioma (23), indicating that miR-335 may become an independent marker for the predication of the clinical outcome of patients with gliomas. Consistently with their findings, the present study also revealed that a high expression of miR-335 was significantly associated with a higher WHO grade of glioma.
Since the present study for the first time, to the best of our knowledge, identified PAX6 as a target of miR-335, their association in glioma cells was further determined. As expected, miR-335 overexpression significantly downregulated the protein expression of PAX6, while inhibition of miR-335 promoted the protein level of PAX6 in U87 and U251 cells, which are WHO IV grade. In fact, several other miRNAs have also been reported to directly target PAX6, including miR-290-295, miR-365-3p and miR-328 (29)(30)(31). Huang et al demonstrated that miR-223 promoted the growth and invasion of glioblastoma cells by targeting PAX6 (32). Since our findings revealed a novel regulatory pathway involved in PAX6 and miR-335, the present study enriches the understanding of the molecular mechanisms by which PAX6 is regulated in glioma.
In addition, the present study demonstrated that the protein levels of MMP-2 and MMP-9 were decreased in the PAX6-overexpressed glioma cells, however, were upregulated in the miR-335-overexpressed glioma cells. MMP-2 and MMP-9 are two key proteases secreted by tumor and microenvironmental cells and are able to lead to extracellular matrix degradation, which directly links to cancer cell invasion and metastasis (33). Furthermore, PAX6 has been reported to suppress the transcription of the MMP-2 gene (4). Based on these findings and ours, the present study suggests that miR-335 promotes glioma cell invasion, possibly through directly inhibiting PAX6 and thus upregulating MMP-2 and MMP-9.
In conclusion, the present study demonstrated that PAX6, as a novel target of miR-335, functions as a tumor suppressor in malignant glioma cells. Our findings suggest that PAX6 and miR-335 are promising therapeutic targets for glioma.