Olea europaea leaf extract alters microRNA expression in human glioblastoma cells

Abstract

Purpose

Glioblastoma multiforme (GBM) is the most common and the most lethal form of primary malignant tumors in the central nervous system. There is an increasing need for the development of more efficient therapeutic approaches for the treatment of these patients. One of the most attractive cancer therapy methods to date is the induction of tumor cell death by certain phytochemicals. Interestingly, bioactive compounds have been shown to alter micro RNA (miRNA) expression involved in several biological processes at the posttranscriptional level. The present study aimed to evaluate whether Olea europaea leaf extract (OLE) has an anticancer effect and modulates miRNA expression in GBM.

Materials and methods

Firstly, the anti-proliferative activity of OLE and the nature of the interaction with temozolomide (TMZ) of OLE were tested in human glioblastoma cell line T98G cells by trypan blue and WST-1 assays and than realized miRNA PCR array analysis. Potential mRNA targets were analyzed bioinformatically.

Results

OLE exhibited anti-proliferative effects on T98G cell lines. Cells were treated with temozolomide (TMZ) in the presence OLE, and changes to miRNA expression levels were identified by PCR array analysis. miRNA target genes are involved in cell cycle and apoptotic pathways. Specifically, miR-181b, miR-153, miR-145, miR-137, and let-7d were significantly upregulated after treatment with both TMZ and OLE.

Conclusion

Our results suggest that OLE modulates the expression of some miRNAs related to anticancer activity in GBM and the response to TMZ. Further studies and validations are needed, but we suggest that OLE might be used for in vivo studies and future medical drug studies.

Introduction

Glioblastoma multiforme (GBM) is the most frequently occurring and malignant brain tumor (Ohgaki and Kleihues 2005). Despite progress in surgical techniques, radiotherapy and chemotherapy, the prognosis remains poor, with a median survival of less than 1 year (Ohgaki and Kleihues 2005; Ohgaki et al. 2005). GBM is characterized by rapid, diffusely infiltrative growth and a high level of cellular heterogeneity associated with therapeutic resistance (Novakova et al. 2009). Considering advances in the molecular biology and genetics of GBM, there is currently no effective treatment or promising molecular targeting therapy for these tumors. For this reason, there is an increasing need for the development of more efficient therapeutic approaches for the treatment of patients with GBM. One of the most attractive cancer therapy methods to date is the induction of tumor cell death by certain phytochemicals derived from medicinal herbs and dietary plants (Mijatovic et al. 2011). More than half of the conventional drugs are established natural products or derived directly from them (Newman and Cragg 2007).

Olive tree (Olea europaea) leaves are widely used in traditional remedies in European and Mediterranean Countries, including Turkey. In these countries, they are frequently consumed as an extract or herbal tea. Olea europae leaf extract (OLE) contains many potentially bioactive compounds that may have antioxidant, antihypertensive, anti-atherogenic, anti-inflammatory, hypoglycemic, and hypocholesterolemic properties (Gilani et al. 2005). One of these potentially bioactive compounds is the secoiridoid oleuropein. Additional bioactive components found in olive leaves include related secoiridoids, triterpenes, and flavonoids (Tabera et al. 2004). Phenolic-type compounds are widely recognized for their antioxidant activity due to their ability to scavenge free radicals, thus saving the cell from a variety of diseases (Sreelatha and Padma 2009; Saito et al. 2008).

Only a few studies have associated cancer protection and treatment with Olea europaea. Abaza et al. (2007) determined that OLE has strong antitumor properties against human promyelocytic leukemia HL-60 cells. Reyes et al. (2006) demonstrated that OLE has anti-proliferative and pro-apoptotic effects in human colorectal adenocarcinoma HT29 and Caco-2 cell lines and suggested that apoptotic cell death was derived from activation of caspase-3 and DNA fragmentation. Fares et al. (2011) evaluated the anticancer and apoptotic effects of OLE on the human leukemic cell line Jurkat, and they found that OLE induces the apoptotic pathway via mechanisms independent of Bcl-2, Bax, and p53 proteins.

One of the novel approaches for the molecular characterization of tumors is based on expression profiling of microRNAs (miRNAs). miRNAs are small non-coding RNAs of 18–25 nucleotides in length that bind to complementary 3′UTR regions of target mRNAs, regulating the transcriptional activity of the target gene (Bartel 2009). Currently >1,000 human miRNA sequences are described (miRBase: http://www.mirbase.org), and it has been suggested that miRNAs may regulate 60 % of the human genome (Sayed and Abdellatif 2011). miRNAs have been implicated in a number of basic metabolic pathways and some biological processes, including cancer pathways (Arola-Arnal and Blade 2011). Therefore, alterations of miRNA expression and function appear to contribute to the initiation, maintenance, and progression of tumors as well as to invasiveness, metastasis, and even acquisition of drug resistance in cancer (Di Leva and Croce 2010; Gandellini et al. 2011). miRNAs have been shown to affect the pathophysiology of all types of human cancers. The aberrant expression of mature and/or precursor miRNA transcripts have been characterized in several different tumor tissues compared to corresponding normal tissues (Ferdin et al. 2010). Studies also demonstrate that there is distinct miRNA expression patterns associated with specific tumor types (Ferracin et al. 2011). Furthermore, not only have miRNAs been connected to several human diseases, there is also evidence that the modulation of miRNAs can provide therapeutic benefits (Chiba and Hijikata 2010; Qin and Zhang 2011; Farazi et al. 2011). Epidemiological observations might provide insight into the diet, lifestyle, and genetic factors that influence miRNA expression. Research into the effect of dietary factors is in its infancy. Interestingly, variable dietary factors, including micronutrients and non-nutrient dietary components, have been shown to alter miRNA gene expression (Chen and Xu 2010). For instance, dietary polyphenols such as soy isoflavones (Link et al. 2010) and the green tea polyphenol epigallocatechin gallate (Link et al. 2010; Tsang and Kwok 2010) have been shown to modulate miRNA expression in cancer cell lines. Similarly, curcumin (diferuloylmethane), a component of the spice Curcuma longa (or turmeric), has been found to modulate cancer signaling pathways, possibly through miRNA expression, in human pancreatic cancer cells (Reuter et al. 2011; Sun et al. 2008). It has been shown that curcumin inhibits tumor growth, invasion, and in vivo metastasis through inhibition of miR-21 transcriptional regulation (Mudduluru et al. 2011). Such observations for plant extracts have indicated that these molecules can modulate miRNAs targeting proliferation, apoptosis, invasion, and metastasis, depending on the cellular context (Sharon and Cindy 2011). Clearly, many aspects of this field should be clarified, including how plant extracts modulate miRNA expression, cell-specific responses to extracts, the timing of exposure and interactions between the bioactive component of extracts and miRNAs.

The first aim of this study was to analyze the anticancer effects of OLE on GBM cells. The second aim was to evaluate the nature of this interaction with temozolomide (TMZ) and OLE in human GBM cells. The third aim was to investigate whether OLE modulates miRNA expression in GBMs.

Materials and methods

OLE production

Olive (Olea europaea) leaves were collected from the Balikesir-Edremit region of Turkey in February 2010. Olive leaf samples were randomly hand-picked from pruned branches. Because Olea europaea sprouts branch in February, potential chemical substances such as pesticides and herbicides were eliminated after pruning the first branches. Leaves were washed, air-dried at room temperature in the dark for approximately 1 month, and homogenized with a Warren blender. The powder was soaked in 80 % ethanol with continuous shaking at room temperature. Standardized OLE (05.06.2007, 10-00014-00015-0) was kindly provided by Kale Naturel (Edremit-Balıkesir, Turkey). The extract was standardized to 15 % oleuropein content by the manufacturer, and stability and microbiological purity were confirmed. The OLE preparation was water soluble, a desirable physical property for a potential drug. It was dissolved in culture medium immediately before use.

Cell line maintenance

The T98G human GBM cell line was provided by the American Type Culture Collection (ATCC; Rockville, USA). Cells were grown in Dulbecco’s Modified Eagle’s Medium-F12 (DMEM-F12; HyClone^®, Utah, USA) containing l-glutamine supplemented with 10 % fetal bovine serum (FBS, BIOCHROME, Berlin, Germany), 1 mM sodium pyruvate, 100 μg/ml of streptomycin, and 100 U/ml of penicillin and were incubated in a 5 % CO₂ humidified incubator at 37 °C.

Cytotoxicity assay

After standard trypsinization, cells were seeded at 2 × 10⁴/well in 96-well plates for cytotoxicity tests. After 24 h of culture in normal growth medium, the cells were exposed to graded concentrations of OLE. The negative control was treated with 30 mM H₂O₂ and the positive control with normal growth medium over the 24 and 48 h incubation. T98G cells were treated with 10 different doses of OLE including 3 mg/ml, 2 mg/ml, 1 mg/ml, 500 μg/ml, 250 μg/ml, 100 μg/ml, 50 μg/ml, 30 μg/ml, 10 μg/ml, and 5 μg/ml, which was dissolved in sterile water immediately before use. Four cultures were prepared with the T98G cell sample from each dose. Cytotoxicity of ten different doses of OLE on T98G was assayed using thoma with 0.4 % trypan blue after at 24 and 48 h incubation. Cell viability was expressed as a percentage of the control value (untreated cells), which was arbitrarily set to 100 %.

Cell proliferation assay

Cell proliferation kit (WST-1, Roche Applied Sciences, Mannheim, Germany) was used to evaluate the viability of cells. After standard trypsinization, cells were seeded at 2 × 10⁴/well in 96-well plates for cytotoxicity tests. After 24 h of culture in normal growth medium, the cells were exposed to graded concentrations of OLE. For controls, 30 mM of H₂O₂, an inhibitor of proliferation was used as a negative control, and growth media were used as positive control over the 24 and 48 h incubation in a humidified incubator. T98G cells were treated with 10 different doses of OLE used for Cytotoxicity Assay, which was dissolved in sterile water immediately before use. Four cultures were prepared with the T98G cell sample from each dosage. This method is based on the ability of the mitochondria from metabolically active cells to convert tetrazolium salt into a dark red formazan product, which is measured colorimetrically using an ELISA microplate reader (Tecan Sunrise, Austria) at 450 nm with a reference wavelength at 620 nm. The results were expressed as a percentage of the control. The absorbance of the untreated control cells was set to 100 %, and the absorbance of OLE-treated cells was taken as a percentage of survival. The following formula was used to calculate the percent of inhibition:

Evaluation of viability in human peripheral blood lymphocytes

Human peripheral blood lymphocytes were used for in vitro viability assays. Heparinized total blood (5 ml) was obtained from two healthy, non-smoking volunteers, one male and one female of 30 and 40 years old, respectively, with their complete informed consent. Human mononuclear lymphocytes were isolated with density gradient centrifugation using Hitopaque^®-1077 (Sigma-Aldrich, Chemie gmbh, Steinheim, Germany) reagent and washed twice. Lymphocytes (3 × 10⁵/well) were added to 2 ml of medium containing 78 % RPMI 1640 (BIOCHROME, Berlin, Germany), 20 % inactivated fetal bovine serum (BIOCHROME, Berlin, Germany), antibiotics (penicillin and streptomycin), and 2 % phytohemagglutinin (PHA; Gibco-Invitrogen, Denmark) for stimulation in 6-well plates. Two cultures were prepared from each volunteer’s blood sample. In the viability assay, cells were exposed to two concentrations (2 and 1 mg/ml in culture medium) of the extract at 24 and 48 h after culture initiation at 37 °C in a 5 % CO₂ humidified incubator. An untreated culture was used as the negative control, and 30 mM H₂O₂ treated cells were used as the positive control.

Determination of synergism or antagonism

TMZ was kindly provided by Prof. Turkkan Evrensel (Uludag University of Turkey). After standard trypsinization, cells were seeded at 3 × 10⁵/well in 6-well plates to determine the synergism or antagonism of TMZ and OLE. In the proliferation assay, cells were exposed to two different concentrations of OLE that were effective in T98G cells, single graded concentrations of TMZ (300–500 μM TMZ), a combination of TMZ with OLE, or H₂O₂ at 24 or 48 h after culture initiation. Analysis was performed on dose–response curves of cell viability after treatment with OLE alone, cytostatic drugs alone or the combination for 24 h. The nature of the interaction between temozolomide and OLE was evaluated by WST analysis.

Real-time PCR-based miRNA expression profiling

miRNA expression profiling was performed to evaluate the molecular effect of OLE on GBM cells. After standard trypsinization, cells were seeded at 3 × 10⁵/well in 6-well plates to analyze expression levels of miRNA using real-time PCR. After 24 h of culture in normal growth medium, the cells were exposed to one of the following: two concentrations of OLE (2 or 1 mg/ml), two concentrations of temozolomide (325 or 450 μM TMZ), a combination of TMZ and OLE or growth medium as a positive control over 24 or 48 h incubation in a humidified incubator. Total RNA was extracted after 24 h of incubation using the miRNeasy Mini Kit (QIAGEN, Germantown, Maryland, USA) following the manufacturer’s protocol. Total RNA (5 ng) was reverse-transcribed using the RT² miRNA First Strand Kit (QIAGEN, Germantown, Maryland, USA). Samples were analyzed for the presence and differential expression of 40 miRNAs related to drug resistance and GBM formation using cancer RT² miRNA PCR arrays (RT² Profiler; SABiosciences, Frederick Md, USA) according to the manufacturer’s instructions. The accession numbers of the primers are shown in Table 1. Thermal cycling conditions for all assays were 95 °C for 10 min, 45 cycles at 95 °C for 15 s, and 60 °C for 30 s, followed by melting curve analysis in the LightCycler 480II (Roche Diagnostics, USA). RNA input was normalized to endogenous controls: SNORD 44, SNORD 47, and SNORD 48 for miRNAs and the TATA-binding protein for protein-encoding genes. The initial copy number in the samples and the threshold cycle (Ct) for miRNA expression were determined by the Light Cycler 480II’s software (Roche Diagnostics, USA). The average Ct value of up to three housekeeping genes from this assay was used as a baseline to normalize the PCR Array data. The miRNA Reverse Transcription Control Assay was used to test the efficiency of the miScript II Reverse Transcription Kit reaction using a primer set to detect a template synthesized from the kit’s built-in miRNA External RNA Control. Positive PCR control assays were used to test the efficiency of the polymerase chain reaction chemistry as well as the instrument using a predispensed artificial DNA sequence and a primer set designed to detect it. Two independent RT² miRNA PCR arrays were analyzed for each sample. The 2^(−ΔCt) method was used to calculate the fold change in miRNA expression between the tested samples (Livak and Schmittgen 2001). Data analysis was performed with a Web-based software package for the miRNA PCR array system (miScript miRNA PCR Array Data Analysis; http://www.sabiosciences.com/pcr/arrayanalysis.php).

Table 1.

Accession numbers of primers

miRNA Sanger ID	miRNA accession number	miRNA Sanger ID	miRNA accession number
hsa-miR-7	MIMAT0000252	hsa-miR-155	MIMAT0000646
hsa-miR-9	MIMAT0000441	hsa-miR-181a	MIMAT0000256
hsa-miR-10a	MIMAT0000253	hsa-miR-181b	MIMAT0000257
hsa-miR-10b	MIMAT0000254	hsa-miR-181c	MIMAT0000258
hsa-miR-15b	MIMAT0000417	hsa-miR-193b*	MIMAT0004767
hsa-miR-21	MIMAT0000076	hsa-miR-195	MIMAT0000461
hsa-miR-24	MIMAT0000080	hsa-miR-210	MIMAT0000267
hsa-miR-24-2*	MIMAT0004497	hsa-miR-219-1-3p	MIMAT0004567
hsa-miR-26a	MIMAT0000082	hsa-miR-221	MIMAT0000278
hsa-miR-101	MIMAT0000099	hsa-miR-222	MIMAT0000279
hsa-miR-124	MIMAT0000422	hsa-miR-296-5p	MIMAT0000690
hsa-miR-125b	MIMAT0000423	hsa-miR-320	MIMAT0000510
hsa-miR-128	MIMAT0000424	hsa-miR-330-3p	MIMAT0000751
hsa-miR-133a	MIMAT0000427	hsa-miR-451	MIMAT0001631
hsa-miR-133b	MIMAT0000770	hsa-miR-455-3p	MIMAT0004784
hsa-miR-135a	MIMAT0000428	hsa-miR-502-3p	MIMAT0004775
hsa-miR-137	MIMAT0000429	hsa-miR-584	MIMAT0003249
hsa-miR-145	MIMAT0000437	hsa-let-7a	MIMAT0000062
hsa-miR-146a	MIMAT0000449	hsa-let-7b	MIMAT0000063
hsa-miR-153	MIMAT0000439	hsa-let-7d	MIMAT0000065

miRNA target prediction

miRNA target genes were identified using the miRWalk online database (http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/). miRWalk provides information on published pathway targets from the KEGG (http://www.genome.jp/kegg/) and BioCarta (http://www.biocarta.com/) pathway databases. The functions of genes were obtained from KEGG and NCBI-Gene (http://www.ncbi.nlm.nih.gov).

Statistical analysis

The results are presented as the mean ± SD of triplicate observations from one representative of at least three experiments with similar results, unless otherwise indicated. Student’s t test was used to determine the statistical significance of differences. Values of P < 0.05 were considered to be statistically significant.

Results

Effects of OLE on cytotoxicity and proliferation

The cytotoxicity of olive plant leaf extract was studied using the T98G cell line. Cells were seeded at a density of 2 × 10⁴ cells/well in 96-well microtiter plates. After 24 h, cells were treated with different concentrations of OLE. A cell counting chamber assay using 0.4 % trypan blue and the WST assay were performed to study the proliferative and viability effects of OLE on T98G cells. The inhibitory concentration at which 50 % of the cells die was identified (IC₅₀). OLE decreased the viability of T98G cells. Optimal activity was observed at day 1. The percentage decrease in proliferation of T98G cells at 1 and 2 mg/ml ranged from 74 to 81 % at 24 h (Fig. 1). When T98G cells were treated with H₂O₂, the reduction in proliferation was determined to be 92 %.

Fig. 1.

Inhibition of cell viability at different OLE concentrations in 24 and 48 h.* P < 0.05

Very low cytotoxic effects were noted when the concentrations studied were tested on activated fresh human mononuclear lymphocytes, indicating that OLE preferentially inhibits T98G cells. When lymphocytes were treated with 2 and 1 mg/ml OLE at 24 h, the reduction in proliferation was determined to be 22 and 16 % compared with untreated cultures, whereas when lymphocytes were treated with H₂O₂, the reduction in proliferation was 87 %.

The role of OLE on the effectiveness of TMZ

Because plant extracts have often been used as supplements to chemotherapy, we evaluated potential interactions between OLE and the most commonly used cytostatic drug, TMZ. Cells were treated with a wide range of doses of temozolomide in the absence or presence of low-toxic concentrations of OLE. The WST assay was performed after 24 and 48 h of incubation, and drug interactions were evaluated by the reduction in proliferation (Fig. 1). There were no statistically significant differences between the effect of 24 and 48 h OLE treatment (P > 0.05). The IC₅₀ of OLE for T98G cells was 1 mg/ml. When T98G cells were treated with 450 and 325 μM TMZ at 24 h, the reduction in proliferation was determined to be 90.4 and 90.02 %, respectively, compared to an untreated culture. When T98G cells were treated with 325 μM TMZ in the presence of 1 mg/ml OLE at 24 h, the reduction in proliferation was determined to be 91.8 and 90.02 % compared to an untreated culture. The results presented in Fig. 2 demonstrate that the addition of OLE affected the toxicity of the applied TMZ in T98G cells. The responsiveness to TMZ was intensified in the presence of the extract. This result indicated that OLE has synergistic effects with the toxicity of TMZ.

Fig. 2.

Effect of OLE and TMZ concentration on cell viability

miRNA expression profiles in T98G cells treated with OLE or TMZ

To evaluate the capacity of OLE to modify miRNA expression, 40 miRNAs were screened in the human GBM cell line T98G. T98G cells were treated for 24 h with one of the following: 1 or 2 mg/ml of OLE, 325 or 450 μM of TMZ, or a combination of 325 μM TMZ and 1 mg/ml OLE. T98G cells from each treatment were analyzed by miRNA PCR arrays, and differentially expressed miRNAs were identified by using a raw P value filter based on statistical significance (P < 0.05) as calculated by the Web-based test (SABiosciences). The average of two independent analyses for each gene and sample was calculated and normalized to the endogenous reference control genes SNORD44, SNORD47, and SNORD48. TMZ and the two concentrations of extract modulated miRNA expression in T98G cells compared to the untreated control cells. OLE treatment differentially upregulated the expression of three miRNAs (miR-584, miR-210, miR-219-1-3b) and repressed the expression of one miRNA (miR-24), whereas TMZ treatment differentially repressed the expression of five miRNAs (miR-9, miR-24, miR-26a, miR-125b, miR-195) and upregulated the expression of nine others (miR-193b*, miR-584, miR-145, miR-155, miR-210, 181b, 135a, miR-133a, miR-219-1-3b). TMZ and OLE combinations altered the expression of twelve miRNAs (miR-193b, miR-584, miR-155, miR-210, miR-181b, miR-153, miR-145, miR-137, miR-135a, miR-133a, miR-219-1-3b, and let-7d). Notably, miR-181b, miR-153, miR-145, miR-137, and let-7d were significantly upregulated (Table 2). These results indicated that the compounds present in the extract specifically altered the expression of miRNAs, mentioned above, in comparison with pure TMZ (Fig. 3).

Table 2.

Differential expression of miRNAs in T98G cells treated with 325 mM TMZ in the presence or absence of 1 mg/ml OLE

	miR-181b	miR-153	miR-145	miR-137	let-7d
Untreated
2^(−Avg.(Delta(Ct)))	0,164368	0,055939	0,012387	0,007239	0,102593
1 mg/ml OLE
2^(−Avg.(Delta(Ct)))	0,303549	0,054788	0,014782	0,010273	0,146097
Fold change	1,8468	0,9794	1,1933	1,4191	1,4241
95 % CI	(0.16, 3.54)	(0.61, 1.35)	(0.00001, 2.47)	(0.24, 2.60)	(0.56, 2.29)
P value	0,177835	0,818583	0,350339	0,263866	0,414796
2 mg/ml OLE
2^(−Avg.(Delta(Ct)))	0,335256	0,073983	0,0201	0,013493	0,110593
Fold change	2,0397	1,3226	1,6226	1,8639	1,078
95 % CI	(1.51, 2.56)	(0.68, 1.97)	(1.46, 1.78)	(1.56, 2.17)	(0.43, 1.72)
P value	0,000517	0,288425	0,000025	0,000302	0,846384
1 mg/ml OLE + 325 μM TMZ
2^(−Avg.(Delta(Ct)))	95,339358	1	3,20428	1	1,003472
Fold change	580,0365	17,8766	258,6756	138,1412	9,7811
95 % CI	(375.17, 784.90)	(12.08, 23.67)	(132.47, 384.88)	(120.44, 155.84)	(3.94, 15.62)
P value	0,000233	0	0,00497	0	0
325 μM TMZ
2^(−Avg.(Delta(Ct)))	1,041262	0,030998	0,126599	0,03036	0,161918
Fold change	6,335	0,5541	10,2201	4,194	1,5783
95 % CI	(3.32, 9.35)	(0.36, 0.75)	(8.19, 12.25)	(3.29, 5.10)	(0.63, 2.52)
P value	0,005552	0,031683	0,00005	0,000142	0,22208
450 μM TMZ
2^(−Avg.(Delta(Ct)))	15,72513	0,264255	1,337928	0,222982	0,410371
Fold change	95,6704	4,724	108,0084	30,803	4
95 % CI	(35.99, 155.35)	(2.55, 6.90)	(58.91, 157.11)	(25.08, 36.53)	(1.35, 6.65)
P value	0,010478	0,003141	0,003715	0,000008	0,004205

Fig. 3.

Alterations in miRNA expression levels in T98G cells after OLE or temozolomide treatment

Identification of differentially expressed miRNA target genes using bioinformatic analysis

For further analysis of the results, bioinformatics studies were performed to identify the target genes of the differentially expressed miRNAs. We focused our attention on five miRNAs affected by OLE treatment. miRNA target genes were identified using the miRWalk online database. According to this database, several target genes related to gliomas were identified from the KEGG pathways database (http://www.genome.jp/kegg/) (Table 3). In addition to playing a role in cancer development, flavonoids are present at high levels in OLE and have a role in anti-proliferation and apoptosis. Therefore, we focused our attention on specific miRNAs’ target genes related to cell cycle and apoptotic pathways (Tables 4, 5). When these miRNAs were evaluated for cell cycle pathway involvement, the validated genes for these miRNAs were related to cyclins (CCNA2, CCND1, CCND2); cyclin-dependent kinases (CDK4, CDK6); cyclin-dependent kinases inhibitors (CDKN1A, CDKN1B, CDKN2A, CDKN2D) and cell cycle regulators (CDC25A, CDC25C, TP53, E2F1, E2F2, E2F3, SKP1, ATM, SMAD3, SMAD4, SMC1A, WEE1, TGFB1, TGFB3) (Table 4). All of these selected miRNAs were determined to be involved in apoptotic pathways (Table 5). The validated target genes for these miRNAs are related to cellular processes including anti- or pro-apoptotic regulation.

Table 3.

Validated target genes and related glioma pathway associated with each miRNA

miRNA	Target genea
hsa-miR-181b	TP53, CDKN2A, PTEN
hsa-miR-153	AKT1
hsa-miR-145	CDK4,CDKN1A, AKT1, EGF, TP53, EGFR, CCND1, E2F3, KRAS, FRAP1, PTEN, IGF1, PDGFB, IGF1R
hsa-miR-137	EGFR, CDK6, KRAS
hsa-let-7d	MAPK3, TP53, EGFR, NRAS, CDKN2A, PTEN, E2F1, CCND1, CDKN1A, KRAS, PIK3CA, CDK6, IGF1, AKT1, BRAF, E2F2, E2F3

^aFrom the miRWalk database (http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/) and the KEGG (http://www.genome.jp/kegg/)

Table 4.

Validated target genes and related cell cycle pathway associated with each miRNA

miRNA	Database namea	Target genesb	Function of the encoded proteinc
hsa-miR-181b	Biocarta	CDKN2A	Cyclin-dependent kinase inhibitor 2A
		CDKN1B	Cyclin-dependent kinase inhibitor 1B (p27, Kip1) controls the cell cycle progression at G1
	KEGG	ATM	Ataxia telangiectasia mutated. An important cell cycle checkpoint kinase
		CDKN2A	Cyclin-dependent kinase inhibitor 2A
		TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair
		CDKN1B	Cyclin-dependent kinase inhibitor 1B (p27, Kip1) controls the cell cycle progression at G1
		TGFB1	Transforming growth factor, beta 1. Regulates proliferation, differentiation, adhesion, migration
		SMAD4	SMAD family member 4
hsa-miR-145	Biocarta	CDKN1A	Cyclin-dependent kinase inhibitor 1A (p21, Cip). Regulator of cell cycle progression at G1
		CDK4	Cyclin-dependent kinase 4. Important for cell cycle G1 phase progression
		CCND1	Cyclin D1. Required for cell cycle G1/S transition
	KEGG	CDKN1A	Cyclin-dependent kinase inhibitor 1A (p21, Cip). Regulator of cell cycle progression at G1
		TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair
		SMC1A	Structural maintenance of chromosomes 1A. An important part of functional kinetochores. Interacts with BRCA1
		CCND1	Cyclin D1. Required for cell cycle G1/S transition
		CDK4	Cyclin-dependent kinase 4. Important for cell cycle G1 phase progression
		E2F3	E2F transcription factor 3. Plays a crucial role in the control of cell cycle
hsa-miR-137	Biocarta	CDK6	Cyclin-dependent kinase 6. Important regulator of cell cycle progression
	KEGG	CDK6	Cyclin-dependent kinase 6. Important regulator of cell cycle progression
hsa-let-7d	Biocarta	CCND1	Cyclin D1. Required for cell cycle G1/S transition
		CDKN2D	Cyclin-dependent kinase inhibitor 2D
		CDKN2A	Cyclin-dependent kinase inhibitor 2A
		E2F1	E2F transcription factor 1. Plays a crucial role in the control of cell cycle
		CCND2	Cyclin D2. Required for cell cycle G1/S transition
		CDK6	Cyclin-dependent kinase 6. Important regulator of cell cycle progression
		CDC25A	Cell division cycle 25 homolog A (S. pombe). Required for progression from G1 to S phase
		CDKN1A	Cyclin-dependent kinase inhibitor 1A (p21, Cip). Regulator of cell cycle progression at G1
	KEGG	CDKN2A	Cyclin-dependent kinase inhibitor 2A
		SMAD3	SMAD family member 3
		CCND1	Cyclin D1. Required for cell cycle G1/S transition
		E2F1	E2F transcription factor 1. Plays a crucial role in the control of cell cycle
		TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair
		CDKN2D	Cyclin-dependent kinase inhibitor 2D
		CDKN1A	Cyclin-dependent kinase inhibitor 1A (p21, Cip). Regulator of cell cycle progression at G1
		TGFB3	Transforming growth factor, beta 3. Regulates proliferation, differentiation, adhesion, migration
		CDK6	Cyclin-dependent kinase 6. Important regulator of cell cycle progression
		CDC25A	Cell division cycle 25 homolog A (S. pombe). Required for progression from G1 to S phase
		CDC25C	Cell division cycle 25 homolog C (S. pombe). Required for progression from G1 to S phase
		SKP1	S-phase kinase-associated protein 1. Includes regulators of cell cycle progression and development
		WEE1	WEE1 homolog (S. pombe)
		E2F3	E2F transcription factor 3. Plays a crucial role in the control of cell cycle
		CCNA2	Cyclin A2. Binds and activates CDC2 or CDK2 kinases, and thus promotes cell cycle G1/S and G2/M transitions
		TGFB1	Transforming growth factor, beta 1. Regulates proliferation, differentiation, adhesion, migration
		E2F2	E2F transcription factor 2. Plays a crucial role in the control of cell cycle
		CCND2	Cyclin D2. Required for cell cycle G1/S transition

^aFrom the KEGG (http://www.genome.jp/kegg/) and BioCarta (http://www.biocarta.com/) pathway databases

^bFrom the miRWalk database (http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/)

^cFrom the NCBI-Gene (http://www.ncbi.nlm.nih.gov) database

Table 5.

Validated target genes and apoptosis pathways (KEGG) associated with each miRNA

miRNA	Gene targetsa	Function of the encoded proteinb
hsa-miR-181b	BCL2	B-cell CLL/lymphoma 2. Blocks apoptotic death
	TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism
	ATM	Ataxia telangiectasia mutated. An important cell cycle checkpoint kinase
	NFKB1	Nuclear factor of kappa light polypeptide gene enhancer in B cells 1
hsa-miR-153	AKT1	V-akt murine thymoma viral oncogene homolog 1. A critical mediator of growth factor-induced neuronal survival
	BCL2	B-cell CLL/lymphoma 2. Blocks apoptotic death
hsa-miR-145	TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism
	AKT1	V-akt murine thymoma viral oncogene homolog 1. A critical mediator of growth factor-induced neuronal survival
	BCL2	B-cell CLL/lymphoma 2. Blocks apoptotic death
	DFFA	DNA fragmentation factor, 45 kDa, alpha polypeptide. Triggers DNA fragmentation during apoptosis
	TNFSF10	Tumor necrosis factor (ligand) superfamily, member 10. Induces apoptosis in transformed and tumor cells
	DFFB	DNA fragmentation factor, 40 kDa, beta polypeptide (caspase-activated DNase). Triggers both DNA fragmentation and chromatin condensation during apoptosis
	PPP3CA	Protein phosphatase 3, catalytic subunit, alpha isozyme
	BIRC2	Baculoviral IAP repeat-containing 2. Member of a family of proteins that inhibits apoptosis
	IRAK2	Interleukin-1 receptor-associated kinase 2. Participates in the IL1-induced upregulation of NF-kappaB
hsa-miR-137	BAX	BCL2-associated X protein. Act as anti- or pro-apoptotic regulator
hsa-let-7d	TP53	Tumor protein p53. Induces cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism
	PIK3CA	Protein phosphatase 3, catalytic subunit, alpha isozyme
	NFKBIA	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1
	BCL2	B-cell CLL/lymphoma 2. Blocks apoptotic death
	AKT1	V-akt murine thymoma viral oncogene homolog 1. A critical mediator of growth factor-induced neuronal survival
	FAS	TNF receptor superfamily, member 6. Plays a central role in the physiological regulation of programmed cell death
	NFKB1	Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1
	BCL2L1	BCL2-like 1(BCLX; BCL2L; BCLXL; BCLXS; Bcl-X; bcl-xL; bcl-xS; PPP1R52; BCL-XL/S)
	TNF	Tumor necrosis factor. Involved in cellular activities, including cell proliferation, differentiation, and apoptosis
	CYCS	Cytochrome c, somatic. Involved in initiation of apoptosis
	BIRC3	Baculoviral IAP repeats containing 3. Inhibit apoptosis by binding to tumor necrosis factor receptor-associated factors

^aFrom the miRWalk database (http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/)

^bFrom the NCBI-Gene (http://www.ncbi.nlm.nih.gov) database

Discussion

The long-term use of chemotherapeutic agents in oncological treatment can cause several serious problems for patients. Safer and more effective treatments are desperately needed, and some such treatments can be found in natural compounds such as phytochemicals (Sun et al. 2008). With established chemopreventive activity and preclinical antitumor effects, phytochemicals provide a novel therapeutic approach (Craig 1999; Setzer and Setzer 2003). Clinically tested plant-derived anticancer agents with proven benefits include paclitaxel, etoposide, teniposide, Vinca alkaloids, vinblastine, vincristine, and camptothecin derivatives (Craig 1999; Mans et al. 2000; Setzer and Setzer 2003; Cragg and Newman 2005). These compounds share common antitumor activities and modulate pathways related to the cell cycle and cell death (Singh et al. 2002). The present study is the first evaluation of the potential anticancer effects of OLE on human GBM cell line T98G. To date, only a few preclinical studies have been performed to validate the effectiveness of O. europaea as a cancer-fighting agent (Abaza et al. 2007; Juan et al. 2008; Fu et al. 2010; Omar 2010; Fares et al. 2011; Mijatovic et al. 2011; Reyes-Zurita et al. 2011). Investigators have reported significant anticancer activity of OLE in diverse cancer cell lines, including breast, colon, melanoma, and leukemia. These studies indicate that OLE has anti-proliferative activity in these cancer types. This study uniquely demonstrates that ethanol extracts from olive leaves can inhibit the proliferation of T98G cells. The IC₅₀ of OLE for T98G cells was 1 mg/ml at 24 and 48 h similarly. In general, the activities of flavonoids are known to be limited for only a few hours in the body (Kurisawa et al. 2003). In the present study, there were no statistically significant differences between the effect of 24 and 48 h OLE treatments. Therefore, 24-h OLE treatments were performed in the further analysis. OLE in the present study had a similar cytotoxic effect as that reported by Micol et al. (2005), who identified the OLE concentration corresponding to a 50 % cytotoxic effect to be approximately 1.25 mg/ml. Further investigation showed that the anti-proliferative effect of Olea europeae was due to the induction of apoptosis in cancer cell lines (Micol et al. 2005; Reyes et al. 2006; Fares et al. 2011). In addition, Olea europeae has been shown to possess antioxidant properties (Visioli and Galli 1994; Kohyama et al. 1997; Manna et al. 1997; Saija et al. 1998; Nousis et al. 2005). In conclusion, this plant possesses high anti-proliferative potency and strong pro-apoptotic effects with a low cytotoxicity. However, the molecular mechanisms of these effects are still not known. Fares et al. (2011) investigated the anti-proliferative activity of OLE on human leukemic cell line Jurkat to explain if cell death was due to apoptosis. They evaluated the expression level of apoptosis-related proteins (Bcl-2, Bax and p53) but did not observe any alteration in their expressions (Fares et al. 2011). Other mechanisms were suggested, such as induction of apoptosis through a receptor mechanism, activation of a MAP-kinase-mediated apoptotic pathway or through other unknown pathways. Therefore, further molecular investigation is required to clarify the molecular mechanism underlying the effect of OLE on tumor cells for possible medical applications.

Mijatovic et al. (2011) evaluated interactions between OLE and different chemotherapeutics in the mouse melanoma cell line B16 and suggested that OLE intensified the action of cisplatine and paclitaxel, but reduced the effectiveness of TMZ and doxorubicin. In the present study, we evaluated potential interactions between OLE and the most commonly used cytostatic drug, temozolomide, in GBM tumors. Our analysis suggests a reduction in the proliferative rate of cells when a combination of OLE and TMZ was used to treat T98G cells. The responsiveness to TMZ was intensified in the presence of the extract. These results revealed that addition of OLE is synergistic with TMZ toxicity. However, parallel treatment with OLE antagonized the action of TMZ in mouse melanoma B16 cell culture in experiments performed by Mijatovic et al. (2011). Due to these variable results, we suggest that constituents of OLE could either amplify or reduce the effectiveness of chemotherapy depending on the cell type.

To understand the molecular characterization of tumors in the last decade, miRNA expression profiling has become a frequently used tool. miRNAs are small non-coding RNAs regulating the transcriptional activity of target mRNAs (Bartel 2009). miRNAs have been implicated in the control of a number of basic metabolic pathways and various biological processes, including cancer pathways (Arola-Arnal and Blade 2011). Therefore, alterations of miRNA expression and function appear to contribute to tumor behaviors including invasiveness, metastasis, and drug resistance (Di Leva and Croce 2010; Gandellini et al. 2011). Furthermore, not only do miRNAs play a role in cancer development, there is also evidence that alteration of miRNA expression levels can provide therapeutic benefits (Arola-Arnal and Blade 2011). Epidemiological observations highlight different factors, such as diet, lifestyle, and genetic features, which can affect miRNA expression. Interestingly, these variables, including micronutrients and non-nutrient dietary components, have been shown to alter miRNA gene expression in cancer cell lines (Chen and Xu 2010; Link et al. 2010; Tsang and Kwok 2010). In this study, we were the first to investigate the effect of OLE on miRNA expression and to identify potential miRNA targets. To evaluate the capacity of OLE to modify miRNA expression, 40 selected miRNAs related to GBM formation and drug resistance were screened using T98G cells treated with OLE, TMZ or a combination of TMZ and OLE. This analysis showed that each treatment changed the expression profile of a subset of the miRNAs. Strikingly, low dose (325 μM) TMZ treatment in combination with OLE differentially affected the expression of five miRNAs (miR-181b, miR-153, miR-145, miR-137, and let-7d) compared to the same dose TMZ treatment alone in T98G cells (Table 2; Fig. 3). Our study found out that the compounds present in the extract specifically affected the expression of these miRNAs when compared to pure TMZ. Notably, a reduced expression of these miRNAs in GBM and other human cancers was associated with tumorigenesis, metastasis, and shortened postoperative survival period (Zhong et al. 2010; Chistiakov and Chekhonin 2012). Specifically, a brain-specific miRNA, miR-153, was found to be highly expressed in normal brain tissue but almost at undetectable levels in malignant gliomas, suggesting that miR-153 functions as a tumor suppressor (Sempere et al. 2004; Gaur et al. 2007). Xu et al. (2010) showed that upregulated expression of miR-153 increases apoptosis via targeting two anti-apoptosis family members, B-cell lymphoma 2 (Bcl-2), and myeloid cell leukemia sequence 1 (Mcl-1), linking this tumor suppressor to the apoptosis pathway in glioblastoma multiforme cell line. Xu et al. (2011) reported that miR-153 directly downregulates Irs-2 and consequently inhibits Akt phosphorylation. Moreover, miR-153 is upregulated to suppress Irs-2, Bcl-2, and Mcl-1 expressions thus downregulating the survival but upregulating the apoptotic pathways, implying that tumor suppressor miR-153 is dual life and death regulator (Xu et al. 2011). Xu et al. (2009) showed that miR-145 targets SOX2 and downregulates its expression in human embryonic stem cells. Fang et al. (2011) also confirmed that miR-145 also decreased SOX2 expression in GBM cells when they transfected LN229 GBM cell with miR-145. Schmitz et al. (2007) found that SOX2 is overexpressed in malignant glioma while displaying minimal expression in normal tissues. More recently, Gangemi et al. (2009) showed that silencing of the SOX2 in freshly derived glioblastoma tumor-initiating cells stopped proliferation and the resulting cells lost tumorigenicity in immunodeficient mice. Ikushima et al. (2009) showed that inhibition of TGF-beta signaling drastically deprived tumorigenicity of glioma-initiating cells by promoting their differentiation and that these effects were attenuated in these cells transduced with SOX2 or SOX4. Taking together, these data suggested that SOX2 is also a key gene in maintaining the stemness of glioma stem cells. Silber et al. (2008) identified that miR-137 was downregulated in both AAs and GBMs and transfection of this miRNA dramatically reduced the percentage of GFAP-positive cells in both CD133+ and CD133− GBM cell fractions. Their results indicate that miR-137 induces G0/G1 cell cycle arrest in GBM cells. So, they suggest that miR-137 may be a useful therapeutic agent for the treatment for GBMs. miR-181 family members can function as tumor suppressors in glioma cells that trigger growth inhibition, induce apoptosis, and inhibit invasion (Shi et al. 2008). In addition, reduced expression of let-7 miRNA families is associated with low responsiveness to a number of chemotherapeutic agents (Hummel et al. 2010). However, overexpression of let-7 miRNA creates a radiosensitive state of lung cancer cells (Lee et al. 2011). Lee et al. (2011) found that transfection of let-7 miRNA reduced expression of pan-RAS, N-RAS, and K-RAS in glioblastoma cells and also reduced the in vitro proliferation and migration of the cells and the sizes of the tumors produced after transplantation into nude mice. However, their results indicate that let-7 miRNA exerted no effect on the proliferation of normal human astrocytes. So, they speculated that let-7 miRNA has an anti-tumorigenic effect on glioblastoma cells and therefore is useful for treating glioblastoma. In the present study, our data indicate that OLE caused an upregulation in the expression levels of these miRNAs and the proliferative rate decreased when treated with a combination of OLE and TMZ in T98G cells. Therefore, the responsiveness to TMZ was intensified in the presence of the extract. Even though further miRNA transfection and protein analysis are necessary to validate our data, we think that OLE has potential utility as a therapeutic drug.

For further analysis of the results, bioinformatics studies were performed to identify the target genes of the differentially expressed miRNAs. We focused on five miRNAs affected by the OLE treatment in cultures treated with TMZ. In line with the results of previous studies, our results show that OLE plays a role in anti-proliferation and apoptosis. Therefore, we focused our attention on selected miRNAs’ target genes that are related to the cell cycle and apoptotic pathways. When the selected miRNAs were evaluated for cell cycle pathway, the validated target genes of four of these miRNAs were related to cyclins, cyclin-dependent kinases, cyclin-dependent kinase inhibitors, and cell cycle regulators. miR-181b, miR-145, miR-137, and let-7d were upregulated by OLE addition. The products of these genes control cell cycle progression at the G1 stage. A common validated target gene of miR-145 is E2F3, which plays a crucial role in cell cycle control and interacts with tumor suppressor proteins. A common validated target gene of miR-145 is E2F3, which plays a crucial role in cell cycle control and interacts with tumor suppressor proteins. A common target gene of miR-181b, miR-145, and let-7d is TP53, which can induce cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism. In addition, the validated target gene of miR-145 is CDK4 and of miR-137 and let-7d is CDK6, all important regulators of the cell progression. The upregulation of these miRNAs may lead to altered expression of their targets, therefore playing an important role in control of the cell cycle.

All of five significant miRNAs were identified in apoptotic pathways in the present study. The validated genes for these miRNAs are related to cellular processes, including anti- or pro-apoptotic regulation. For instance, let-7d target genes, FAS, TNF, and CYCS, induce apoptosis. The miR-145 target genes are DFFA and DFFB, and these proteins initiate both DNA fragmentation and chromatin condensation during apoptosis. The anti-proliferative effect induced by OLE might be derived from an apoptotic pathway. Similarly, Han et al. (2009) reported that oleuropein, the main component of OLE, remarkably decreases the number of MCF-7 cells by inhibiting the rate of cell proliferation and inducing apoptosis. These results support our observations that OLE-induced alteration of miRNA expression levels may be connected to cell cycle and apoptosis pathways.

Some studies have also shown that plant extracts modulate miRNAs that target proliferation, apoptosis, invasion, and metastasis (Sharon and Cindy 2011). For instance, different plant extracts, such as soy isoflavones (Link et al. 2010), the green tea polyphenol epigallocatechin gallate (Link et al. 2010; Tsang and Kwok 2010), curcumin (diferuloylmethane) (Reuter et al. 2011), and mistletoe lectin-1 (Li et al. 2011) have been shown to modulate miRNA expression in diverse cancer cell lines. However, our data are the first to demonstrate that OLE alters miRNA expression in human GBM cells. Our findings are also the first to indicate that OLE downregulated some miRNAs, contributing to its prominent anticancer activity and sensitizing GBM cells to TMZ treatment. Further studies and validation are needed, but we suggest that OLE might be used in therapeutic cancer drug studies. According to the clinical reports, herbal extracts can protect the non-cancerous cells and tissues in the body from possible damage caused by chemotherapy and can enhance the potency of chemotherapy. It is evident that much remains to be discovered with regard to OLE modulation of miRNA expression, including cell- or tissue-specific responses, the quantity of bioactive component needed to induce the desired biological effect, the timing of exposure, and other variables that can influence the response (including interactions with personal factors). Purification and identification of the active compounds of OLE are required for a better understanding of the protective mechanisms involved and for the possible future medical applications. Furthermore, interactions between OLE and drug resistance in GBM are open avenues for future investigation.

In conclusion, these observations suggest that modulation of miRNA expression may be an important mechanism underlying the biological effects of OLE. Therefore, our results provide evidence that OLE can influence miRNA expression, suggesting a new mechanism of action for this extract. To the best of our knowledge, this is the first time that the altered expression of miRNAs induced by OLE has been correlated with a response to TMZ. Finally, we showed that the miRNAs affected by OLE were significantly associated with the cell cycle and apoptosis. It is notable our data showed for the first time that OLE-induced miRNAs were correlated with a reduced resistance to TMZ, which may contribute to the development of a treatment for GBM.