MicroRNA-184 inhibits neuroblastoma cell survival through targeting the serine/threonine kinase AKT2

Neuroblastoma tumour samples were obtained from patients at Our Lady's Hospital for Sick Children in Crumlin, Ireland or through the Children's Oncology Group (USA) and have been previously described in aCGH [24], mRNA [25] and miRNA [3] profiling studies.

Kelly and SK-N-AS cell lines were purchased from the European Collection of Animal Cell Cultures (Porton Down, United Kingdom). SHEP-TET21 cells were obtained from Dr. Louis Chesler with permission of Prof. Manfred Schwab [26]. Kelly cells and SHEP-TET21 cells were grown in RPMI 1640 supplemented with 10% fetal bovine serum, 2 mM Glutamine and 2 mM penicillin and streptomycin (GIBCO). SK-N-AS cells were cultured in EMEM (GIBCO) supplemented with 10% fetal bovine serum, glutamine and penicillin and streptomycin.

Pre-miR™ and Anti-miR™ to miR-184 and negative control 1 (a scrambled oligonucleotide) were obtained from Ambion (Austin, Texas). Short interfering (si)RNAs targeting AKT2 were obtained from Applied Biosystems (Foster City, CA). Three different siRNAs against AKT2 were chosen (s1215 sense CAACUUCUCCGUAGCAGAAtt, anti-sense UUCUGCUACGGAGAAGUUGtt, s1217 sense strand UGACUUCGACUA UCUCAAAtt and anti sense strand UUUGAGAUAGUCGAAGUCAtt) (s228853 sense strand ACAACUUCUCCGUAGCAGAtt and anti sense strand UCUGCUACGGAGAAGUUGUtt).

The Pre-miR™ and Anti-miR™ to miR-184, negative control 1 and the siRNAs to AKT2 were introduced into the cells by reverse transfection using the transfection agent siPORT™ Neo FX™(Ambion). Cell culture media was changed after 8 hours to remove the transfection reagent in an attempt to avoid toxicity which may be caused by NeoFX™. Total RNA/miRNA was extracted 24, 48 and 72 hours after transfection using RNeasy Kit/mirNeasy^© kit (Qiagen, UK).

Reverse transcription was carried out using 50 ng of total RNA with the primer specific for miR-184 and the TaqMan microRNA reverse transcription kit (Applied biosystems). qPCR was carried out on the 7900 HT Fast Realtime System (Applied Biosystems). RNU66, a small RNA encoded in the intron of RPL5 (chr1:93,018,360-93,018,429; 1p22.1), was used for normalization in miRNA studies and RPLPO ribosomal protein was used for normalization in gene expression studies (chr12: 119,118,300-119,124; 12q24.2). A relative fold change in expression of the target miRNA/gene transcript was determined using the comparative cycle threshold method (2^-ΔΔCT).

The significance of miRNA differential expression over tumour sub-types was evaluated by assigning P-values based on the non-parametric Mann-Whitney test.

The stem loop precursor sequence of miR184 was cloned into the pcDNA6.2-GW/EmGFP expression vector (BLOCK-iT Pol II miR RNAi Expression Vector kit, Invitrogen). The following oligonucleotides were designed which encode the sense and antisense strands of the pre-miR184 sequence. These oligonucleotides include the appropriate 5' and 3' overhangs to facilitate cloning into the linearised pcDNA6.2-GW/EmGFP vector (supplied within the BLOCK-iT kit, Invitrogen).

Pre-miR184 Sense strand:

Pre-miR184 Antisense strand:

CCTG TCAATCACCTACCCTTATCAGTTCTCCTGCCAACACTTACAGTCACAAAGCTGGCTGGAAAAGTGATAAGGGGACGTGACTGGC

The pcDNA6.2-GW/EmGFP-miRNA-184 construct or the control construct (pcDNA6.2-GW/EmGFP-miRnegative control, Invitrogen) was transfected into Kelly and SK-N-AS cells using lipofectamine 2000 (Invitrogen, Carslbad) according to manufacturers instructions. Quantitative real-time PCR and fluorescent microscopy were carried out to determine efficient transfection and transcription of the vector.

The expression vector pcDNA 3 containing AKT2 was obtained from Prof. Joe Testa (Fox Chase Cancer Centre, Philadelphia) [27]. 1 μg of the vector or the control empty vector was transfected into Kelly and SK-N-AS cells using Lipofectamine 2000.

Apoptosis was demonstrated by annexin-V staining and propidium iodide (PI) exclusion using the FITC Annexin-V Apoptosis Detection Kit I (BD Pharmingen, San Diego, CA, USA). Briefly, adherent and supernatant Kelly cells were collected, washed twice in cold PBS, and resuspended in 1× Annexin-V binding buffer at a concentration of 1 × 10⁶ cells/ml. An aliquot of 100 μl of suspension (1 × 10⁵ cells) was stained with 5 μl Annexin-V-FITC and 5 μl PI, and incubated for 15 minutes at room temperature in the dark. Binding buffer (400 ul) was added and cells acquired (10,000 cells) immediately using a BD LSR II flow cytometer (Becton Dickinson, San Jose, CA, USA) and analysed using BD FACSDiva 4.0 Software. Experiments were performed in multiples to qualify apoptosis by phosphatidylserine (PS) externalization.

Cell Death was also evaluated using the 3/7 Caspase detection kit from Promega (Madison, WI). Neuroblastoma experimental cells were plated in quadruplicate in 96-well plates. 72 hours after transfection, 10 ul of caspase 3/7 was added to each well. Samples were read after 1 hr of incubation with the caspase substrate on a Viktor Microplate luminometer (Molecular Devices, Sunnyvale, CA).

For cell number assays, cells were set up in triplicate in 6 well plates. Cells were seeded at equal densities of 3 × 10⁴ cells per well. When carrying out transfections using the microRNA mimics or anti-miRs (as described above) each time point was set up with a non-transfected (with transfection reagent) and a scrambled oligonucleotide control (negative 1). Cells were trypsinised from 6 well plates at 24, 48 and 72 hour time points, and re-suspended in 1 ml of media. A haemocytometer was used to count cell numbers. Counts from triplicate wells were averaged.

Total protein was isolated from cells using a radioimmunoprecipitation assay (RIPA) lysis buffer (Sigma). Protein concentration was measured using the BCA assay from Millipore. Proteins were fractionated on 10% polyacrylamide gels, and blotted onto nitrocellulose membrane. The membrane was probed with the Anti-AKT2 Antibody (Millipore) or anti-MYCN (Abcam), anti β-Actin from (Abcam) or anti GAPDH (abcam) (used for loading controls). Signal was detected using Immoblion Western (Millipore).

A 76nt long region of the 3'UTR of AKT2 containing the predicted miR-184 binding site (underlined) was ligated into the pMiR-Reporter vector (Ambion) 3' of the luciferase gene:5'CTAGTCCTCTGTGTGCGATGTTGTTATCTGACAGTTCTCCGTCC CTACTGGCCTTTCTCCTCGTCTTCGCTCAGC A 3'

As a negative control, three mutations (lower case) were introduced into the seed region of miR-184 binding site of this sequence:

5'CTAGTCCTCTGTGTGCGATGTTGTTATCTGACAGTTCTtCaaCC CTACTGGCCTTTCTCCTCGTCTTCGCTCAGC A 3'

KELLY cells were plated at 8 × 10⁴ in 12 well format. After 24 hrs the pMir-Reporter containing the AKT2 binding site for miR-184 or the mutated AKT2 binding site were co-transfected with either the pre-miR-184 mimic or a scrambled negative control sequence using Lipofectamie 2000. All experiments were also co-transfected with the pmiR-Report β-galactosidase vector as a control for transfection efficiency and normalization. Luciferase activity was measured by One-Glo luciferase assay (Promega) according to manufactures instructions after 24 hours on the Viktor Plate Reader.

In order to experimentally determine that MYCN levels influence quantities of miR-184, we examined miR-184 expression in the SH-EP TET21 neuroblastoma cell line containing a MYCN construct which is repressible by doxycycline. As demonstrated by qPCR and Western blotting, addition of doxycycline to the cell culture caused a dramatic reduction in both MYCN mRNA and protein levels (Additional File 1a and 1b). MYCN depleted SH-EP cells had an 8 fold increase in MiR-184 levels (Additional File 1c), indicating that MYCN either directly or indirectly suppresses miR-184 transcription, consistent with our earlier expression profiling studies of primary tumors [3].

It was previously reported that transfection of miR-184 mimics into neuroblastoma cell lines causes a decrease in cell numbers and increase in apoptosis in neuroblastoma cell lines [3]. Here, we first assessed whether the reciprocal experiment of knocking down endogenous miR-184 in Kelly and SK-N-AS cell lines following transfection with an anti-miR-184 also has a detectable biological effect. As illustrated in Figure 1a and 1b, the inhibition of miR-184 results in a reproducible and statistically significant increase in cell numbers (~1.6 fold in Kelly; p < 0.0001; and 1.3 fold in SK-N-AS; p < 0.0001) (Figure 1a and 1b). In addition, we also determined that ectopic over-expression of miR-184 at physiological levels using an expression plasmid has similar biological effects on Kelly and SK-N-AS cells as the mature miR-184 mimics which were introduced at supra-physiological levels (Additional File 2a-g). The molecular mechanism by which miR-184 exerts effects on cell numbers and apoptosis of neuroblastoma cells, however, is unknown.

An examination of the Sanger microcosm database http://​microrna.​sanger.​ac.​uk/​sequences/​ indicated that miR-184 has a very large number of computationally predicted mRNA targets. Among the top 3% (n = 30) of miR-184 predicted targets was the 3'UTR of AKT2, which had a high level of sequence identity with the miR-184 seed region, a 13 base pair match (Figure 2a). We focused our studies on AKT2 as a potential miR-184 target given that this was the only gene in the top 3% whose function might account for the apoptotic phenotype induced by miR-184. AKT2 is a well documented pro-survival proteinFor an initial assessment of whether AKT2 mRNA levels and miR-184 levels might be inversely related, qPCR analysis of AKT2 mRNA was carried out on 10 tumors with low miR-184 and 10 with high levels. As illustrated in Figure 2b, AKT2 mRNA levels are significantly lower in tumors with higher miR-184 (P < 0.002). As one might expect, AKT2 was expressed at higher levels in MNA tumors relative to non-MNA tumors (P < 0.035), as MYCN suppresses miR-184 transcript quantities (Figure 2c). An inverse relationship between miR-184 and AKT2 mRNA levels was also determined to exist in neuroblastoma cell lines (Figure 3a and 3b). Levels of AKT2 protein corresponded to the levels of mRNA in the cell lines (Figure 3c).

To confirm that AKT2 was indeed regulated by miR-184 in neuroblastoma, the miR-184 mimic and the anti-miR-184 were individually transfected into Kelly and SK-N-AS cell lines. A significant decrease of AKT2 mRNA was observed over three time points (24, 48 and 72 hrs) following transfection of miR-184 mimics into Kelly (P = 0.004)(Figure 4a) and SK-N-AS (P = 0.003)(Figure 4b), and conversely, the suppression of miR-184 using the anti-mir-184 caused an increase in mRNA for AKT2 across all time points in both cell lines (P = 0.009 and 0.01 respectively). These results were also observed at protein level (Figure 4c). MiR-184 is not predicted to target the 3'UTRs of related AKT family members, AKT1 nor AKT3. Consistent with expectations, AKT1 mRNA levels remained constant when miR-184 was transfected into Kelly and SK-N-AS (Additional File 3a), while AKT3 mRNA was undetectable in both cell lines by TaqMan qPCR.

The effect of miR-184 on AKT2 levels appears to be a direct effect of miR-184 targeting AKT2 mRNA, since co-transfection of a pMir-Reporter containing the AKT2 binding site for miR-184 and mature miR-184 mimics significantly (p < 0.003) diminished luciferase activity while co-transfection of the reporter with a negative control oligonucleotide had no effect (Figure 4d). A three base pair mutation introduced into the seed region of the miR-184 binding site in the AKT2 3' UTR completely abolished the ability of mature miR-184 mimics to affect luciferase activity.

MiR-184 has many computationally predicted targets, so in order to determine if the anti-proliferative effects of miR-184 could be attributable to targeting AKT2, we transfected both Kelly and SK-N-AS neuroblastoma cell lines with three different siRNAs to AKT2 and examined the effects on the rate of accumulation of cell numbers. AKT2 mRNA knockdown ranged from 68 to 98% by 48 hrs, depending on the siRNA (Additional File 4a and 4c), with AKT2 protein being proportionally reduced (Additional File 4b and 4d). Both cell lines exhibited a marked decline in cell numbers for each siRNA relative to the negative control (Figure 5a and 5b), along with a highly significant increase (3.9 fold; P < 0.0001) in the late apoptotic cell fraction assessed by FACs analysis of an annexin V staining assay (Figure 6a and 6b). A statistically significant increase in caspase 3/7 activation in both Kelly and SK-N-AS cells following siRNA mediated AKT2 knock down also occurred (Figure 6c and 6d). Consistent with expectations, AKT1 mRNA levels remained constant when AKT2 siRNA was transfected into the cell lines (Additional File 3b).

To demonstrate that the increase in cell numbers that occurred following miR-184 knockdown resulted specifically from AKT2 up-regulation, we transfected the pcDNA3-AKT2 plasmid into Kelly neuroblastoma cells. This resulted in a 5 to 22 fold increase in AKT2 mRNA levels, and a 30% increase in cell numbers by the 72 hr time point relative to the negative control (p = 0.006). (Figure 5c). Since the pcDNA3-AKT2 construct lacks the miR-184 binding site in the 3' UTR, we also co-transfected pcDNA3-AKT2 along with the miR-184 mimics to determine if ectopic up-regulation of AKT2 could rescue Kelly cells from the anti-proliferative effects of miR-184. As illustrated in Figure 5c, the numbers of cells accumulated over 72 hours for Kelly cells co-transfected with pcDNA3-AKT2 and miR-184 was not statistically different to that of Kelly cells transfected with a negative control oligonucleotide and the pcDNA3 empty vector. However, this co-transfection with pcDNA3-AKT2 and miR-184 mimics yielded a cell accumulation rate that was significantly higher then cells transfected with miR-184 mimics (p <0.003) or miR-184 mimics and pc-DNA3.1 empty vector (p <0.003), indicating that ectopic AKT2 lacking a miR-184 binding site can rescue the cells from ectopic miR-184 up-regulation (Figure 5c). As illustrated in Additional File 5, RT-qPCR analysis of AKT2 mRNA indicated that there were statistically significant (p < 0.01) differences in AKT2 mRNA levels in each of the transfected cell populations at each time point, consistent with expectations. From all of the above experiments, we conclude that the phenotypic effects of miR-184, at least to a large extent, can be attributed to the targeting and reduction of AKT2.